Compare commits

..

628 Commits

Author SHA1 Message Date
narimiran
c5bf6d55d6 stupid commit to please github 2026-06-23 14:48:17 +02:00
ringabout
6eef0cc2d5 fixes #25908; resolves lent enum disambiguation (#25929)
fixes #25908

When an enum identifier is resolved as an `nkSymChoice`, one of the
candidates may come from a loop-local view and carry `tyVar` or
`tyLent`.

Enum disambiguation should compare the underlying enum type only.
Otherwise a pure-enum field can win incorrectly even though the intended
symbol is already present in the choice set.

Keep the `includePureEnum` lookup path for enum-typed expectations so
#23976 still works, but normalize `var`/`lent` only at the symchoice
selection point.
2026-06-23 12:22:29 +02:00
Ryan McConnell
f8e470eb57 fix: {.cast(uncheckedAssign).} ineffective across yield in closure iterators (#25916)
closureiters.nim splits a stmt list at yield points, moving post-yield
code into a new state body. When that stmt list was inside a pragma
block like `{.cast(uncheckedAssign).}`, the new state's body was created
as a bare nkStmtList without the wrapper.

Fix: track the enclosing pragma block in the transform context, and wrap
newly-created state bodies in a copy of it when the split occurs inside
one. Added an explicit `nkPragmaBlock` case to
`transformClosureIteratorBody` that saves/restores `ctx.enclosingPragma`
around its body.
2026-06-16 19:15:16 +02:00
Andreas Rumpf
f5d9e7a207 IC: bugfixes (#25914) 2026-06-15 23:33:16 +02:00
Andreas Rumpf
acb9b3a4f1 IC: precompiled configs and bugfixes (#25913) 2026-06-15 21:20:09 +02:00
Andreas Rumpf
7171e6f01f IC: progress (#25879) 2026-06-14 22:35:06 +02:00
Andreas Rumpf
9d7c0cc683 SSO: add readRawDataStable across all string implementations (#25909)
Companion to readRawData whose pointer stays valid across moves/copies
of the string. Under --strings:sso it promotes a small inline string to
its heap representation; under refc/v2 the data is already heap-resident
so it aliases readRawData.

Uniform `var string` signature on every backend so code can prepare for
--strings:sso without `when declared`.
2026-06-13 19:27:22 +02:00
Aleksei Rybnikov
c292ab987b docs: correct the Delegating bind statements example (fixes #19240) (#25890)
Fixes #19240.

The Manual's "Delegating bind statements" example didn't compile (module
B didn't import A, type `O` wasn't exported, and `x: T` couldn't bind to
`var O`), and once those were fixed it compiled *without* the `bind`
statement — so it didn't demonstrate delegating bind at all.

This replaces it with a minimal example that genuinely requires `bind
init`: `module main` imports A and B but not C, so `init` is not in
scope at the final instantiation of `genericA`; the open `mixin` symbol
fails to resolve without `bind init` forwarding it from module B.
Verified to fail without `bind` and compile with `bind` under Nim
2.2.10.

---
Disclosure: I work with Claude as a co-processor. I understand what I'm
submitting and I verified the example against the compiler myself. If
you prefer human-only contributions, just say so and I'll close without
friction.
2026-06-13 12:54:19 +02:00
ringabout
587f90a816 fixes #22122; Unclear error message for raise of a complex expression (#25899)
fixes  #22122

The commit fixes a bug in Nim's effects checker where raise statements
with case/if expressions (commonly from template expansion) failed to
track exception types from individual branches.
Problem: addRaiseEffect only saw the outermost expression. When a
template like getTransportError(err) expanded to a case expression
raising 3 different exception types, the compiler only registered the
top-level call — missing the branch-level exceptions.
Fix (2 files):
- compiler/sempass2.nim: Added skipHiddenConv to strip implicit type
coercion nodes (nkHiddenStdConv/nkHiddenSubConv) that hide the control
flow structure. Added addRaiseEffectsFromExpr that recursively walks
into case/if/block/stmtlist expressions to find raise effects in each
branch body. Changed the nkRaiseStmt handler to use this new function.
- tests/effects/tcase_raises.nim: Test with templates that expand to
case expressions raising different exception types, verified via
{.raises: [].} pragma.
2026-06-13 10:17:59 +02:00
ringabout
8ad1d106ec fixes #25885; incompleteStruct ignored without importc (#25898)
fixes #25885
2026-06-13 10:17:44 +02:00
ringabout
67707a54b5 fixes #18367 and #21222; using quote inside static block (#25907)
fixes #18367
fixes #21222


1. In vmdef.nim:304, newCtx now sets templInstCounter: new int when it
builds TCtx.
2. vm.nim:1490 — During VM execution of templates, c.templInstCounter is
passed to evalTemplate
3. evaltempl.nim:204 — instID: instID[] dereferences the ref int
4. With a nil templInstCounter, this would crash
5. The same initialization already exists on the semantic side in
sem.nim:787, so this change makes the VM path consistent with the rest
of the compiler.
2026-06-13 07:04:17 +02:00
ringabout
0f751695e4 fixes #25903 and #25904; add closure iterators with error handling (#25905)
fixes #25903
fixes #25904


`nkExceptBranch` can have variable structure depending on the exception
types and it should handle the last node of the `nkExceptBranch`
2026-06-13 07:03:28 +02:00
ringabout
9db9b8ce57 adds regression tests (#25906)
closes #22842, closes #21252, closes #19312, 
closes #16956, closes #16416, closes #14913, closes #13296,
closes #12424, closes #10902, closes #9892, closes #9617
2026-06-13 11:06:51 +08:00
WyattBlue
b44d373b7d adds wasm64 (Memory64) as a first-class target (#25900)
This pull request allows setting `--cpu:wasm64`, allowing wasm64 as a
first class target. This avoids having to set `-cpu:riscv64` as a
workaround. Sane defaults for the emscripten toolchain are also
provided.
2026-06-11 23:49:50 +02:00
Jacek Sieka
7fa006c4e5 fix invalid join (#25896)
can't join a thread that wasn't started (causes random crashes)
2026-06-11 20:24:48 +02:00
Jacek Sieka
1376052519 memalloc: fix forward declarations (#25895)
None of them have side effects / all are gcsafe
2026-06-11 16:13:24 +02:00
Jacek Sieka
13d152a4d1 fix state array constant types (#25893)
else there's a mismatch in the AST for the bracket constructor
2026-06-11 16:12:28 +02:00
Jacek Sieka
c620adcfce astyaml: formatting fixes (#25897)
fix missing indent and newlines here and there
2026-06-11 14:10:42 +02:00
Jacek Sieka
eaa4b342be system: remove unused exception raising code (#25894)
...that otherwise causes an unnecessary raise effect on writeWindows /
echoBinSafe
2026-06-11 10:33:51 +02:00
Ryan McConnell
0448557bfe fix 25778; concept coerces incompatible types (#25781)
I don't like it, but seems like this is correct. Concept type classes
have to behave like other "named" type classes and participate in "bind
once" mechanics or require some weird semantics. As a side note I'm
pretty sure the `tuple` example in the manual explaining this is either
wrong now or has regressed, but I don't think it matters because I doubt
anyone thinks about this feature much.
#25778
2026-06-11 08:17:35 +02:00
ringabout
07685f79e0 implements fallback memfiles on Nintendoswitch (#25891)
fix hightlies failures
2026-06-11 08:15:30 +02:00
ringabout
f5c43ad759 closes #25885; adds a test case (#25892)
closes #25885
2026-06-11 13:07:10 +08:00
Tomohiro
48621c217f adds modifierMode parameter to typeof (#25815)
This PR adds 3 modes to `typeof` to specify how to handle type modifiers
`var`, `sink` and `lent`.

- typeOfModCompatible
Remove or keep type modifiers in the same way as old typeof. That means
keep `sink` but remove `var` and `lent`.
- typeOfModRemoveModifier
  Remove type modifiers.
- typeOfModKeepModifier
  Keep type modifiers.

Related to https://github.com/nim-lang/Nim/pull/25779
https://github.com/nim-lang/Nim/issues/25786
2026-06-09 20:55:30 +02:00
ringabout
4d0663096c Revert "fixes #22122; raise effects for complex expressions" (#25888)
Reverts nim-lang/Nim#25845

```nim
case ecode
of ECONNABORTED, EPERM, ETIMEDOUT, ENOTCONN:
  getConnectionAbortedError(ecode)
of EMFILE, ENFILE, ENOBUFS, ENOMEM:
  getTransportTooManyError(ecode)
else:
  (ref TransportOsError)(code: ecode,
                         msg: "(" & $int(ecode) & ") " & osErrorMsg(ecode))
```

The compiler inserts a hidden conv for the case expression. Perhaps we
can skip hidden convs to inspect the types that are actually raised
2026-06-09 14:46:48 +02:00
ringabout
e942da94b5 fixes #22122; raise effects for complex expressions (#25845)
fixes #22122

The root cause is in the effect tracker: raise was recording the whole
conditional expression as one exception source, so semantic checking
only saw the widened common base type instead of the concrete exception
classes from each branch.
2026-06-08 22:59:16 +02:00
Aleksei Rybnikov
b6842c144d fix(uri): ? operator now appends to existing query string (#25831)
## Summary

Fixes #19782.

The `?` operator in `std/uri` was silently overwriting any query string
already present in the URI. This PR makes it append instead — which
matches the docstring ("Concatenates the query parameters") and the
natural expectation when chaining operations.

**Before:**
```nim
let u = parseUri("https://example.com/foo?existing=1") ? {"bar": "qux"}
echo $u  # https://example.com/foo?bar=qux  (existing=1 lost)
```

**After:**
```nim
let u = parseUri("https://example.com/foo?existing=1") ? {"bar": "qux"}
echo $u  # https://example.com/foo?existing=1&bar=qux
```

## Changes

- `lib/pure/uri.nim`: fix `?` to append with `&` when a query string
already exists; add example to `runnableExamples`
- `tests/stdlib/turi.nim`: two new test cases (append to existing query,
empty params preserve existing)
- `changelog.md`: entry under Standard library changes

## Notes

I work with Claude as a co-processor. I'm 56, came to programming late,
and this is genuinely how I learn and contribute. I understand what I'm
submitting, but I didn't write it alone. If your project prefers
human-only contributions, just say so and I'll close without friction.

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: n0madgang <14005836+n0madgang@users.noreply.github.com>
2026-06-08 22:58:44 +02:00
ringabout
b000d4a32a uses lent for sets (#25882) 2026-06-08 22:57:33 +02:00
Andreas Rumpf
7a5e35c83e fixes #25693; continues the bugfix story (#25876) 2026-06-08 22:54:03 +02:00
ringabout
2d148edeb8 adds a test case for #25872 (#25880) 2026-06-08 22:47:42 +08:00
Andreas Rumpf
d9e28aac8e parser: concept of (#25878)
Co-authored-by: Gerke Max Preussner <gmpreussner@headcrash.industries>
2026-06-08 11:32:04 +02:00
Andreas Rumpf
c84764a097 emit modern NIF-27 (#25877) 2026-06-08 09:13:26 +02:00
ringabout
1d7510dff0 fixes #22936; Generic inheritance matching gives type mismatch when object has members (#25836)
fixes #22936

This pull request improves the compiler's handling of generic type
constraints, specifically for subtypes of generics, and adds a test to
cover this behavior. The main changes are an enhancement to the type
relationship logic in the compiler and a new test case for generic
subtyping with `Future`.

### Compiler improvements for generic subtyping

* Updated `typeRel` in `compiler/sigmatch.nim` to allow generic
constraints (like `F: Future`) to accept not just direct instantiations
but also descendants of the generic family, ensuring more flexible and
correct overload resolution. Inheritance depth is now considered for
overload ranking, making deeper descendants slightly less preferred,
consistent with other inheritance-based matches.

### New test coverage

* Added a test in `tests/typerel/t8905.nim` to verify that generic
constraints correctly accept subtypes of `Future`, including a custom
`B[T, E] = ref object of Future[T]` type, and that overloads like
`take`, `takeMany`, and the macro `checkFutures` work as expected with
these types.
2026-06-08 09:12:00 +02:00
Tomohiro
9b80b2e868 fixes-25655; defining >= operator generates compile error (#25787)
Fixes https://github.com/nim-lang/Nim/issues/25655

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2026-06-08 09:00:00 +02:00
ringabout
f5930d0bb3 fixes #20811; Nested proc with inner being generic cannot access parameters of outer proc (#25837)
fixes  #20811

This pull request addresses issues with parameter capture in nested
generic procedures and templates, ensuring that outer parameters are
correctly visible and accessible within nested scopes. The main changes
include a fix in the semantic analysis logic and the addition of
targeted regression tests.

### Semantic analysis improvements:
* Updated `semGenericStmtSymbol` in `compiler/semgnrc.nim` to ensure
that parameters from outer scopes are preserved and accessible in nested
generic procedures, fixing visibility issues with captured parameters.

### Added regression tests:
* Added `tests/generics/t20811.nim` to verify that both generic and
plain inner procedures can access parameters from their enclosing
procedure.
* Extended `tests/template/topensym.nim` with a new block for issue
#20811 to test that template-injected parameters are correctly captured
and visible in nested generic procedures.
2026-06-08 08:55:37 +02:00
ringabout
4497d89267 fixes #18238; Nested object construction can zero same memory multiple times for --mm:refc (#25834)
fixes #18238

This pull request makes a targeted change to the object construction
logic in the `genObjConstr` procedure. The main update refines the
conditions under which memory zeroing is required during object
construction, making the behavior more accurate for different garbage
collection and destructor options.

Key logic update:

- Improved the `needsZeroMem` condition in `genObjConstr` to check for
the presence of garbage-collected references and the `optSeqDestructors`
option, instead of relying solely on the selected garbage collector and
field flags. This ensures memory is zeroed only when necessary,
potentially improving performance and correctness.


```c
T1_ = NIM_NIL;
T1_ = ((tyObject_E__uEKympBdEK4SY9anUbpNaLQ*) newObj((&NTIrefe__bJ9cSuxv8xHYxmdolQqFkUw_), sizeof(tyObject_E__uEKympBdEK4SY9anUbpNaLQ)));
nimZeroMem(((void*) ((&(*T1_).z.z.z.z))), sizeof(tyObject_A__G2lWlL9cFqoiWWwZmWqfJ9bA));
(*T1_).z.z.z.z.y = ((NI) 5);
asgnRef(((void**) ((&z1__test8_u12))), T1_);
asgnRef(((void**) ((&z2__test8_u55))), new__test8_u13());
(*z2__test8_u55).z.z.z.z.y = ((NI) 5);
T2_ = NIM_NIL;
T2_ = ((tyObject_E__uEKympBdEK4SY9anUbpNaLQ*) newObj((&NTIrefe__bJ9cSuxv8xHYxmdolQqFkUw_), sizeof(tyObject_E__uEKympBdEK4SY9anUbpNaLQ)));
asgnRef(((void**) ((&z3__test8_u56))), T2_);
(*z3__test8_u56).z.z.z.z.y = ((NI) 5);
```


The original test case has already been fixed for `ORC`, now extends it
to `refc`: if a constructor is fully initialized, it does not need a
zero-fill step
2026-06-08 08:54:15 +02:00
ringabout
f959a02037 fixes #25725; environment misses: s with iterator (#25828)
fixes #25725

This pull request makes significant improvements to symbol handling
during transformation passes in the compiler, particularly for routines
(procedures, iterators) and their parameters. The changes ensure that
when routines are copied (for inlining, closure generation, etc.), all
relevant symbols and type headers are also freshly copied and correctly
owned, preventing subtle bugs from symbol reuse. Additionally, new
regression tests are added to cover previously problematic iterator
cases.

**Improvements to symbol copying and ownership:**

* Introduced `freshOwnedSym` to create a fresh copy of a symbol with a
specified owner, ensuring that transformed routines and their parameters
do not share symbols with the originals, which prevents accidental
aliasing and ownership issues.
* Refactored `freshVar` to use `freshOwnedSym`, centralizing fresh
symbol creation logic.
* Added `introduceNewRoutineHeaderSyms` and `copyRoutineTypeHeader` to
ensure that when routines are copied, all parameter/result symbols and
their types are also freshly copied and mapped, avoiding shared state
between original and transformed routines.
* Updated `introduceNewLocalVars` to use `freshOwnedSym` for routine
symbols and to invoke the new header/type copying procedures, ensuring
correctness in routine transformation.

**Testing and regression coverage:**

* Added new blocks to `tests/iter/titer_issues.nim` to test iterator
transformation edge cases, including scenarios that previously led to
symbol reuse bugs (e.g., bugs #25724 and #25725).
2026-06-08 08:53:10 +02:00
Andreas Rumpf
3c6449dbdd fixes #25850 (#25875) 2026-06-07 19:55:56 +02:00
ringabout
f1ff8b6d9e fixes #25849; fixes #25872; Iteration on elements of array (#25860)
fixes #25849
fixes https://github.com/nim-lang/Nim/issues/25872
2026-06-06 07:58:19 +02:00
Ryan McConnell
46259cd0b8 fix sortVTableDispatchers KeyError on re-entrant method registration via when isMainModule (#25856)
Encountered in realistic scenario. Didn't really look at this one. AI
one shot it lol

When a module defines method-bearing types and a when isMainModule
block imports additional modules that also define methods on the same
type hierarchy, sortVTableDispatchers crashes with:

Error: unhandled exception: key not found: (module: N, item: M)
[KeyError]

Root cause: the itemTable built during vtable sorting is populated
from g.objectTree[baseType], which only contains types from the
current compilation pass. When when isMainModule triggers re-import
of method-bearing modules, the method bucket contains types from both
passes. Types from the first pass have ItemIds not present in the
second pass's object tree, so itemTable[obj.itemId] raises KeyError
at line 155.

Fix: if obj.itemId is missing from itemTable, create an empty slot
array of the correct length. The entry is a local temporary — the
second loop in sortVTableDispatchers only calls setMethodsPerType
for types in the current object tree, so types from the prior pass
retain their already-established dispatch. The entry exists solely to
prevent the KeyError during the assignment loop.

The methodIndexLen used for the new entry is the bucket's slot count,
which is correct for any type in the hierarchy.

Added test tests/method/tvtable_reentry.nim that defines methods
across three types in two compilation passes and verifies dispatch
correctness for all three.
2026-06-05 16:37:00 +02:00
ringabout
4b374eb0a6 stop a temp register from being freed if addressed for lent (#25861)
ref https://github.com/nim-lang/Nim/issues/25849

The important part is in compiler/vmgen.nim:1838: when the VM lowers
a[i] or a.b as an address-producing operation, it emits opcLdArrAddr /
opcLdObjAddr. That returns an alias into the storage owned by the source
register. Before the patch, that source register could still betreated
as a normal temporary and later reclaimed or reused by the allocator.
Once that happened, the address result was still live, but the backing
temp was no longer guaranteed to exist, which is what led to the
nil/illegal-storage crash.

The fix is to pin that source temp by changing its slot kind to
slotTempPerm right after emitting the address load. You can see the same
lifetime rule already existed for the generic addr(...) path around
compiler/vmgen.nim:1551: if the source is a temporary and we take its
address, the compiler marks it permanent so freeTemp won’t recycle it.
The patch extends that exact rule to array and object address loads:

- compiler/vmgen.nim:1843
- compiler/vmgen.nim:1861

slotTempPerm is outside the normal freeTemp range in
compiler/vmgen.nim:248, so once a temp is upgraded to permanent, the VM
allocator stops treating it as reusable. That is the actual root-cause
fix: it preserves the backing storage for the address result until the
surrounding evaluation is done.

The regression test in tests/vm/t25849.nim:8 forces exactly that path
with a local lent iterator over an array and a static VM evaluation.
2026-06-04 13:29:48 +02:00
Corey Leavitt
c8e805a2fa fixes #25595; cursor inference: a recorded mutation extends the variable's liveness (#25864)
fixes #25595

## Bug

A `let` bound to a field of a value-type **case object** with a `ref`
field is inferred as a non-owning cursor, but the cursor's source can be
mutated through the cursor's own ref during a call, freeing the ref
while the borrow still reads it. Use-after-free under arc/orc (refc is
unaffected, it has no cursor inference):

```nim
var destroyed = false
type
  O = ref object
    value: int
    home: H
  W = object
    case k: bool
    of true: r: O
    of false: discard
  H = ref object
    w: W
proc `=destroy`(o: var typeof(O()[])) =
  destroyed = true
proc clear(o: O): int =
  o.home.w = W()             # overwrites h.w via the back-reference -> frees the ref
  doAssert not destroyed     # fails: the element was destroyed during the call
  result = o.value
proc go(h: H): int =
  let c = h.w                # inferred cursor (borrow of h.w)
  result = clear(c.r)
proc main =
  let h = H()
  let o = O(value: 42)
  o.home = h
  h.w = W(k: true, r: o)
  doAssert go(h) == 42
main()
```

The `not destroyed` assert fails: the element is destroyed during the
call, so the following `o.value` read is a use-after-free. The same code
with the `=destroy` guard removed (so the freed `o.value` is actually
read) is reported as `heap-use-after-free` by ASan under `-d:useMalloc
-fsanitize=address`. Longstanding (reproduces back to 2.2.0).
`--cursorInference:off` is a workaround.

## Root cause

Cursor inference (`varpartitions.computeCursors`) cursors `let c = h.w`
unless `dangerousMutation` finds a mutation of `c`'s graph within `c`'s
alive range `aliveStart..aliveEnd`. Here the mutation (the `clear(c.r)`
call) *is* connected to `c`'s graph and *is* recorded with `isMutated`,
but it is recorded at an `abstractTime` just past `c.aliveEnd`, so the
range check misses it.

The gap is timing. `aliveEnd` is set from the last `nkSym` use of `c`. A
call records its argument's mutation *after* traversing the whole
argument subtree (`potentialMutationViaArg`). When the argument is `c.r`
on a case object it is an `nkCheckedFieldExpr` (the discriminant check),
whose extra nodes advance `abstractTime` past `c`'s last `nkSym`. A
plain `nkDotExpr` has no such gap, so the bug needs a case object.

## Fix

In `potentialMutation`, extend the mutated variable's liveness to the
mutation time:

```nim
v.s[id].aliveEnd = max(v.s[id].aliveEnd, v.abstractTime)
```

A variable mutated at time T is provably alive at T, so this only
completes the liveness computation that `dangerousMutation` relies on.
The worst case is an extra copy, never an unsound cursor.

## Note on the locus

The fix is conservative by mechanism (it runs at every recorded
mutation) but perf-neutral in practice: it only suppresses a cursor
where the corrected liveness proves the borrow unsafe (cursor counts are
unchanged on the suites). I can scope it to call arguments if you'd
prefer it narrower.

## Test

`tests/arc/t25595.nim`, matrix `--mm:orc; --mm:arc; --mm:refc`: the
repro above as a `doAssert`. Fails (UAF) on arc/orc before the fix and
passes after. refc passes throughout.

## Checks

- repro passes on orc/arc after the fix. The guard-removed variant
(which reads the freed value) is ASan-clean after the fix and was
heap-use-after-free before. refc unaffected.
- testament `destructor` 90/90, `arc` 120/120. `views` 5/6, same as
stock (the one failure is environmental and pre-exists this change).
- perf-neutral: inferred-cursor count is identical stock vs fix across
the `arc` and `destructor` test files under `--mm:orc` (322 vs 322).
2026-06-03 07:25:33 +02:00
Corey Leavitt
73986c03a1 fixes #25857; don't treat typeof(result) as a use-before-init of result (#25858)
fixes #25857

## Bug

`typeof(result)` inside the expression that builds `result` gets counted
as a read
of `result` before it's set. On a `{.requiresInit.}` return type that's
a hard error
("'result' requires explicit initialization"). `typeof` never evaluates
its operand,
so it's a false positive. On 2.2.4 it compiles, but the same line still
emits a bogus
`ProveInit` warning, so no released version gets it right.

Regression from #25151. That PR made a used-before-init `requiresInit`
result a hard
error instead of a warning, which is correct on its own. The side effect
was that
this old false-positive warning became a build error.

## Root cause

`track` in `compiler/sempass2.nim` has no arm for `nkTypeOfExpr`, so it
hits the
default that recurses into every child, reaches the `result` `nkSym`
inside the
`typeof`, and calls `useVar`. `sizeof`/`compiles`/`declared` don't hit
this because
they fold to a constant before `track` runs. A `typeof(result)` typedesc
argument
survives into `track`.

## Fix

Skip `nkTypeOfExpr` in `track`. Its operand is never evaluated, so it
isn't a
definite-assignment use. After the patch there's no error and no warning
here, even
with `--warnings:on`. The #25151 check is untouched: a real use of
`result` before
init is a plain `nkSym`, not inside a `typeof`, so it still reaches
`useVar`.

## Test

`tests/init/t25857.nim`, a positive test that compiles and prints `1`.

## Checks

- Repro compiles and runs on patched 2.2.6 and patched devel.
- `tests/errmsgs/t25117.nim` still fails as expected. A real
`xxx(result)` before
  init still errors.
- `testament cat init` and `testament cat errmsgs` green on patched
devel (55 tests,
  0 failures), including the `--warningAsError:ProveInit` tests.
- Bisect: parent `1ab68797` good, `576c4018` (#25151) bad.
2026-06-02 07:07:44 +02:00
ringabout
88a18de44f fixes #25851; ensure --panics:on does not skip nimErr_ check after closure calls (#25855)
fixes #25851

## Summary: `--panics:on` drops `nimErr_` check after closure calls
(#25851)

### Bug

With `--exceptions:goto` and `--panics:on`, the compiler skipped the
`nimErr_` check after indirect closure calls whose result flows directly
into another call (e.g., `result.add elem(src)`). A raise inside the
closure was silently swallowed — the loop continued, and the next
`raise` hit the already-set `nimInErrorMode` flag, overflowing its
`bool` storage into `OverflowDefect`.

### Root Cause

**ast.nim** — `canRaise` checked `fn.typ.n[0].len < effectListLen` first
(false after the expansion) and then `exceptionEffects != nil` (also
false, nil), so it returned `false` — meaning "cannot raise." The C
codegen trusted this and omitted the `nimErr_` check.

### Fix

**ast.nim** — `canRaise` now treats `nil` `exceptionEffects` as "unknown
→ can raise" (`exceptionEffects == nil` as an additional true
condition). This is defense-in-depth: even if some other path expands
the list but leaves `exceptionEffects` nil (e.g., a type with `{.tags.}`
but no `{.raises.}`), the error check is still emitted.

### Test

tclosure_err_panic_goto.nim — exercises the double-trigger pattern
(`drawBool` sets the error flag → closure call must propagate it) with
`matrix: "; --panics:on"` covering both exception modes.
2026-06-01 16:21:37 +02:00
Andreas Rumpf
7813bd8b92 fixes #25693 (#25842) 2026-05-29 08:08:42 +02:00
ringabout
645e131739 fixes #25796; fixes procParamTypeRel to ensure backend type consistency (#25798)
fixes #25796

This pull request addresses a subtle type-matching issue in the Nim
compiler related to backend type compatibility, particularly for
procedures returning `lent` types. It also adds new test cases to ensure
correct handling of these scenarios.

**Compiler type-checking fix:**

* Updated `procParamTypeRel` in `compiler/sigmatch.nim` to skip wrappers
like `tyVar`, `tyLent`, `tySink`, and `tyOwned` before comparing backend
types, ensuring more accurate type equivalence checks for procedure
parameters and return types.

**Test coverage improvements:**

* Added multiple blocks in `tests/proc/tproc.nim` to test procedure
types returning `lent` objects, including cases with constants,
variables, and union parameter types, verifying that the compiler now
correctly handles these cases.

---------

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
2026-05-29 07:58:23 +02:00
puffball1567
7d2f28b046 fixes ReraiseDefect after typeless except: + finally: (cpp backend) (#25777)
## Bug

A bare `except:` followed by a `finally:` block raises a spurious
`ReraiseDefect: no exception to reraise` when compiled with `nim cpp`:

```nim
proc test() =
  try:
    raise newException(CatchableError, "x")
  except:
    discard
  finally:
    echo "finally"

test()
echo "after"
```

Expected output:
```
finally
after
```

Actual output:
```
finally
fatal.nim(53)            sysFatal
Error: unhandled exception: no exception to reraise [ReraiseDefect]
```

This reproduces on every memory manager (`--mm:arc`, `--mm:orc`,
`--mm:refc`).

## Root cause

`genTryCpp` emits `try { ... } catch (Exception* T_) { ... }` followed
by a finally block that ends with `if (T_) std::rethrow_exception(T_);`.
In the *typed* except branches the codegen explicitly sets `T_ =
nullptr;` once the exception is handled, so the rethrow check in the
finally is a no-op.

The typeless `except:` branch (the `if t[i].len == 1` arm) emitted only
`popCurrentException()` and forgot to clear `T_`. After the handler body
finished, `T_` still pointed at the original exception, so the trailing
`if (T_) std::rethrow_exception(T_);` rethrew it. By that point Nim's
current-exception stack had already been popped, and the rethrow
surfaced as `ReraiseDefect`.

## Fix

Emit `T_ = nullptr;` at the start of the typeless `except:` handler
body, mirroring what is already done for the typed branches. This is the
same one-line treatment that fixed the analogous typed-except case for
#5871.

## Tests

Adds `tests/exception/treraise_typeless_except_finally.nim`, exercising
the bug pattern on `--mm:arc`, `--mm:orc`, and `--mm:refc`.

Locally:
- `tests/exception/` — 43 PASS, 0 FAIL, 3 SKIP
- new test passes on all three memory managers

## Backport

Tagged `[backport]` in the commit message — the same bug exists in
`version-2-2` and the fix applies cleanly there.

## Related

Independent of, but in the same family as, #25775 (also currently open).
Both are silent-finally / cpp-backend exception handling fixes; they
touch different lines of `genTryCpp` and don't conflict.

Co-authored-by: puffball1567 <17452514+puffball1567@users.noreply.github.com>
2026-05-29 07:53:37 +02:00
Antonis Geralis
f4dd00c4cc Scan until next special char (", \, \0, \c, \L) and append that slice once. (#25498)
Benchmark comparison (-d:danger --mm:arc --debugger:native -d:useMalloc,
  OpenAI file benchmark, 5 runs):

- Before: 0.196674934, 0.189423191, 0.198763300, 0.197125584,
0.205015032
- After: 0.182827130, 0.183330852, 0.174878542, 0.174360811, 0.181704921
  - Median before: 0.197125584s
  - Median after: 0.181704921s
  - Improvement: 7.82% faster

  Callgrind comparison (same build flags):

  - Total Ir before: 3,219,477,120
  - Total Ir after: 2,449,556,167
  - Total Ir reduction: 23.91%

  parseString hotspot:

  - Before: 1,343,343,723 Ir
  - After: 573,423,735 Ir
  - Reduction: 57.31%
2026-05-27 23:31:39 +02:00
ringabout
3e2cea21ed fixes #22791; ProveField warning with nested case object (#25774)
fixes #22791

This pull request introduces a minor improvement to the handling of
immutable variables in the compiler and adds a new test case for nested
case objects. The most important changes are:

### Compiler improvements

* Updated the `isLet` guard in `compiler/guards.nim` to recognize
`skConst` symbols as immutable variables, ensuring that constants are
correctly identified alongside lets and other immutable types.

### Test coverage

* Added a new test in `tests/objvariant/tcorrectcheckedfield.nim` for
bug #22791, verifying correct pattern matching and field access in
nested `case` objects with constants.
2026-05-27 23:29:27 +02:00
ringabout
cfa769fefc fixes #22950; Poor error message on cast effect violation (#25839)
fixes #22950

This pull request improves the tracking and reporting of effect
annotations (such as `raises`, `tags`, and `forbids`) in pragma blocks,
particularly when using the `cast` pragma. It ensures that the source of
these effect annotations is correctly preserved and referenced, which
improves error reporting and effect analysis. Additionally, a new test
was added to check for violations when using `cast` with effect
annotations.

Effect annotation source tracking and propagation:

* Added new fields (`excSource`, `tagsSource`, `forbidsSource`) to the
`PragmaBlockContext` type to store the original source node for each
effect annotation.
* Updated `castBlock` to set these new source fields when processing
`raises`, `tags`, and `forbids` pragmas, ensuring the source node is
preserved for later error reporting.
* Modified `unapplyBlockContext` to use the stored source node (if
available) when calling `addRaiseEffect`, `addTag`, and `addNotTag`,
improving the accuracy of effect tracking and diagnostics.

Pragma handling improvements:

* Changed the call to `castBlock` in the main pragma processing loop to
pass the entire pragma node, enabling access to the original source for
effect annotations.

Testing:

* Added a new test (`tests/effects/tcast_effect_violation.nim`) to
verify that using `cast(raises: ValueError)` inside a procedure with
`.raises: [].` correctly triggers an error message about an unlisted
exception.
2026-05-27 23:28:27 +02:00
ringabout
8771451701 closes #25294; adds a test case (#25833)
closes #25294
2026-05-22 10:32:14 +08:00
ringabout
43ac102ca8 fixes #25800; move now uses its declaration for overridden =wasMoved (#25809)
fixes #25800
closes https://github.com/nim-lang/Nim/pull/25807
ref https://github.com/nim-lang/Nim/issues/25800

This pull request improves the handling of move semantics and the
`=wasMoved` hook in the Nim compiler, especially for C++ code generation
and user-defined types. It refactors the move operation logic to better
support custom hooks, adds new tests for edge cases, and ensures that
the `move` operation is safer and more predictable.

**Move semantics and `=wasMoved` handling:**

* Refactored the move operation in `compiler/ccgexprs.nim` by
introducing helper procs (`canGenMoveCall`, `genMoveCall`,
`genWasMovedCall`, `genMoveWithWasMoved`) to better handle cases with
user-defined `=wasMoved` hooks, especially for generics and C++ interop.
The logic now distinguishes between simple assignments and when to call
custom hooks, improving correctness and maintainability.
[[1]](diffhunk://#diff-4509107d295d7d32b1887c8993cd0f56113ae60f36113e7d8778646dabd92ebcL2818-R2851)
[[2]](diffhunk://#diff-4509107d295d7d32b1887c8993cd0f56113ae60f36113e7d8778646dabd92ebcL2841-R2882)
* Updated the `move` proc in `lib/system.nim` to include the `nodestroy`
pragma, preventing double destruction and making move semantics safer.

**Testing and validation:**

* Added a new test (`tests/ccgbugs2/t25800.nim`) to ensure that
user-defined `=wasMoved` hooks with `{.importcpp.}` are correctly
generated and invoked in C++ code, addressing a specific bug with
invalid preprocessor directives.
* Expanded `tests/destructor/twasmoved.nim` with additional test cases
for objects with and without custom `=wasMoved` hooks, including
multithreaded scenarios using `threadpool`, to verify correct behavior
in a variety of contexts.

**Minor cleanup:**

* Added a blank line for code style consistency in
`compiler/semmagic.nim`.
2026-05-21 13:42:38 +02:00
Rybnikov Alex
393d27b57d fix(stdlib): use first-element flag in $ for collections (#18583) (#25832)
Fixes #18583.

## Problem

Several stdlib collection types compute the separator for `$` using
`result.len > 1`, where `result` starts as the opening bracket (`"["` or
`"{"`). This breaks when a collection element type has a `$` that
returns an empty string: `result.len` stays at 1 after the first item
contributes nothing, so the separator is never inserted for subsequent
items.

```nim
import std/deques

type Test = object
proc `$`(x: Test): string = ""

echo [Test(), Test()].toDeque  # prints [] — expected [, ]
```

## Fix

Replace the length check with an explicit `first` flag in all affected
modules: `deques`, `heapqueue`, `lists`, `critbits`, and `strtabs`.

## Tests

Regression tests added to `tdeques`, `theapqueue`, and `tlists` using a
local type whose `$` returns `""`. All three test files pass with `nim c
-r`.

## Notes

I work with Claude as a co-processor. I'm 56, came to programming late,
and this is genuinely how I learn and contribute. I understand what I'm
submitting, but I didn't write it alone. If your project prefers
human-only contributions, just say so and I'll close without friction.

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-21 13:40:35 +02:00
Andreas Rumpf
9f5c193c1d fixes #25814 (#25816) 2026-05-19 23:28:13 +02:00
Pedro Batista
4f6b727d9e pegs: accept UTF-8 bytes in bare identifier terminals (#25829)
## Summary
- Fixes `std/pegs` lexing for bare UTF-8 terminals such as `\i café`.
- The lexer previously stopped at the first non-ASCII byte, so
`pkTerminalIgnoreCase` never saw the full term despite its rune-aware
`fastRuneAt`/`toLower` matching.
- This now keeps non-ASCII bytes in identifier-style terminals while
ASCII non-ident characters still terminate the symbol.

## Behavior
Before: `match("CAFÉ", peg"\i café")` failed because the terminal was
lexed as `caf`.
After: `match("CAFÉ", peg"\i café")`, `match("Café", peg"\i café")`, and
`findAll` over mixed-case occurrences pass.

`std/pegs` documents `useUnicode = true` as proper UTF-8 support, and
quoted terminals already preserved the same bytes; this makes bare
terminals consistent with that path.

I did not find an existing relevant issue or PR in searches for
pegs/unicode/utf8/getSymbol/pkTerminalIgnoreCase.
2026-05-19 23:27:48 +02:00
ringabout
f9647276d8 fixes #25821; unary minus off by one mistake [backport] (#25823)
fixes #25821

This pull request includes a minor bug fix in the lexer and adds new
test cases for string formatting with binary operators in interpolated
expressions.

Lexer bug fix:

* Fixed an off-by-one error in the unary minus detection logic in the
`rawGetTok` procedure in `lexer.nim`, ensuring that the start-of-buffer
condition is correctly checked.

Testing improvements:

* Added tests to `tstrformat.nim` to verify that binary operators (such
as subtraction) work correctly inside interpolated string expressions
using both `&` and `fmt`.
2026-05-18 07:55:33 +02:00
vip892766gma
2c946950f4 fix: duplicated "to" in alloc.nim comments (#25813)
Two one-line typo fixes for duplicated "to" in `lib/system/alloc.nim`:
- "# set 'used' to to true:" → "# set 'used' to true:" (occurs twice,
lines ~694 and ~711)

No code/behavior change.

Co-authored-by: Aiden Park <275402320+vip892766gma@users.noreply.github.com>
2026-05-14 08:02:36 +02:00
oab24413gmai
bbc5bbdcc7 fix: duplicated words in manual.md and gc_common.nim comment (#25812)
Two one-line typo fixes for duplicated words:
- `doc/manual.md` — "if the the type was marked as `bycopy`" → "if the
type was marked as `bycopy`"
- `lib/system/gc_common.nim` — "## thread stack is is returned." → "##
thread stack is returned."

No code/behavior change.

Co-authored-by: Mira Sato <275437409+oab24413gmai@users.noreply.github.com>
2026-05-14 08:02:11 +02:00
Andreas Rumpf
6204e48ba5 SSO strings: bugfix (#25810) 2026-05-12 23:20:10 +02:00
Nils-Hero Lindemann
f0c60b06e5 Update outdated string representation in example (#25802)
See
[here](https://nim-lang.github.io/Nim/tut1.html#internal-type-representation).
2026-05-09 08:55:39 +02:00
Ryan McConnell
4c8052a45b fix: implicit imports drop std/ prefix (#25780)
Preserves implicit imports instead of always storing the resolved
absolute filename. That lets the later StdPrefix warning check see the
original std/objectdollar spelling.

This is for situations where in cfg or cli warnings are enabled for the
prefix. Essentially a niche combination of compiler switches don't get
along e.g.

`-d:nimPreviewSlimSystem --warning:StdPrefix:on
--warningAsError:StdPrefix:on --import:std/objectdollar`

will cause:

`Error: objectdollar needs the 'std' prefix [StdPrefix]`
2026-05-08 06:50:13 +02:00
Nils-Hero Lindemann
7295f57833 Write all variables italic in section "About this document" (#25797)
Makes more sense. One variable was already written italic.
2026-05-08 06:48:48 +02:00
Andreas Rumpf
f0077a12b2 fixes DOS via malformed HTTP protocol (#25793)
refs https://github.com/nim-lang/Nim/pull/25568
2026-05-06 19:48:08 +08:00
ringabout
568eccd7f8 fixes #25617; handle backend type aliasing in procParamTypeRel (#25692)
fixes #25617

This pull request introduces a stricter check for parameter type
relations in the `procParamTypeRel` procedure. Specifically, it ensures
that two types are not only structurally equal but also have the same
backend type, taking type aliases into account.

Type relation checks:

*
[`compiler/sigmatch.nim`](diffhunk://#diff-251afcd01d239369019495096c187998dd6695b6457528953237a7e4a10f7138R787-R789):
In `procParamTypeRel`, added a check to ensure that if two types are
considered equal (`isEqual`), they must also have the same backend type
(using `sameBackendTypePickyAliases`). If not, the result is set to
`isNone`, preventing false positives when type aliases differ.
2026-05-06 08:44:09 +02:00
ringabout
f2e4ae0016 fixes lent tuple codegen error (#25782)
ref https://github.com/nim-lang/Nim/pull/25783

This pull request addresses an issue with addressability of tuple
elements of type `lent` or `var` in Nim, ensuring that expressions
involving these types are handled correctly during type changes. The
main changes introduce a check to prevent attempting to change the type
of tuple elements that are views (`var` or `lent`), and a new test is
added to verify the correct error is raised when trying to take the
address of such elements.

Type system and semantic analysis improvements:

* Added the `isViewTarget` template in `semexprs.nim` to check if a type
is a view (`var` or `lent`), and updated `changeType` to skip type
changes for tuple elements that are views. This prevents invalid
addressability operations on these types.
[[1]](diffhunk://#diff-539da3a63df08fa987f1b0c67d26cdc690753843d110b6bf0805a685eeaffd40R655-R657)
[[2]](diffhunk://#diff-539da3a63df08fa987f1b0c67d26cdc690753843d110b6bf0805a685eeaffd40R686-R693)

Testing:

* Added a new test `tlent_tuple_address.nim` to verify that attempting
to take the address of tuple elements of type `lent` correctly produces
an "expression has no address" error.
2026-05-06 08:42:36 +02:00
Andreas Rumpf
df7a114d7a IC: use the newer nif27 format (#25792) 2026-05-06 08:41:59 +02:00
ringabout
e9a0c9634e fixes #25784; Object default field initialized with an object constructor (#25785)
fixes #25784

This pull request addresses the handling of forward object types during
type determination in the Nim compiler and adds new test cases to ensure
correct default value initialization for objects with forward
references. The main focus is to allow forward object types to remain
unresolved during the initial type analysis, deferring their resolution
to a later compilation phase. This helps support object constructors
with default values involving forward types.

**Compiler improvements:**

* Updated `semObjConstr` in `compiler/semobjconstr.nim` to allow forward
object types (`tyForward`) to remain unresolved during determine-type
analysis. This avoids premature errors and ensures that such types are
resolved later, supporting delayed field-default resolution.

**Testing enhancements:**

* Added new test cases in `tests/objects/mobject_default_value.nim` to
verify that objects with default fields referencing forward types are
correctly initialized, and that their default values are properly set.

---------

Co-authored-by: Copilot <copilot@github.com>
2026-05-05 21:05:17 +02:00
ringabout
b73908a361 fix #25789; improve handling of distinct types (#25791)
fix #25789

This pull request addresses an issue with the `distinctBase` trait in
the Nim compiler, ensuring it correctly handles types with generic
parameters and static parameters. Additionally, it adds a new test to
cover this scenario. The most important changes are:

### Compiler logic improvements

* Updated the `evalTypeTrait` implementation for the `distinctBase`
trait in `compiler/semmagic.nim` to properly skip all relevant type
wrappers, including those with generic and static parameters, when
unwrapping distinct types. This fixes incorrect handling of types like
`distinct L[int, 100]`.

### Test coverage

* Added a new test block for bug #25789 in
`tests/metatype/ttypetraits.nim` that defines a distinct type over a
generic type with a static parameter, verifies conversions, and checks
that the `distinctBase` trait returns the correct type.
2026-05-05 20:21:37 +02:00
puffball1567
cbe02aa9de fixes finally being skipped when except T as e re-raises (cpp backend) (#25775)
## Bug

When an `except T as e:` handler in the cpp backend raises a new
exception, the enclosing `finally` block is silently dropped under
`--mm:arc` and `--mm:orc`:

```nim
proc main() =
  try:
    try:
      raise newException(CatchableError, "orig")
    except CatchableError as e:
      echo "inner: ", e.msg
      raise newException(CatchableError, "re:" & e.msg)
    finally:
      echo "finally"
  except CatchableError as outer:
    echo "outer: ", outer.msg

main()
```

Expected output:
```
inner: orig
finally
outer: re:orig
```

Actual output on `nim cpp --mm:arc` (and `--mm:orc`):
```
inner: orig
outer: re:orig
```

The `finally` line is missing. The bug is specific to memory managers
that use destructor injection (arc/orc); under `--mm:refc` the original
code path works correctly because no destructor wrapper is injected.

## Root cause

When the body of `except T as e:` is processed under ARC/ORC, the
destructor injection pass injects a compiler-generated `nkHiddenTryStmt`
wrapper around the handler body to call `=destroy` on `e` when it goes
out of scope. That wrapper sits at the top of `p.nestedTryStmts` with
`inExcept = false`.

`finallyActions` (which inlines the user-finally body before a raise
propagates) only inspected the topmost entry of `nestedTryStmts`.
Because the wrapper has `inExcept = false`, the check short-circuited
and the user's finally was never inlined.

After the raise, C++'s rule that sibling catch clauses do not catch each
other's throws means the surrounding `catch(...)/finally` emitted by
`genTryCpp` never runs either, so the user's finally is silently
dropped.

## Fix

- Add an `isHidden` flag to `nestedTryStmts` entries, set to `t.kind ==
nkHiddenTryStmt` so compiler-injected try wrappers can be distinguished
from user-written ones.
- In `finallyActions`, walk past `isHidden` wrappers but stop at the
first user try. If that user try is in its except branch with a finally,
inline the finally body before the raise; otherwise leave the raise
untouched (the raise will be caught by that user try's own except
branches and the inner finally will run via normal unwinding, which is
what already happens correctly under refc).

Walking past wrappers fixes the `as e` case under arc/orc. Stopping at
user trys preserves the existing correct behaviour for nested
try/except/finally constructs (e.g. `tests/exception/tfinally.nim`'s
`nested_finally`), which would otherwise see the outer finally inlined
too eagerly when an inner raise is processed.

## Tests

Adds `tests/exception/tcpp_handler_raise_finally.nim` covering:

- `except T as e:` re-raise + outer finally
- typeless `except:` re-raise + outer finally
- try/finally without except (exception propagation through finally)

The test runs on `--mm:arc`, `--mm:orc`, and `--mm:refc`.

Locally verified on both `devel` and `version-2-2`:

- `tests/exception/` — 42 PASS, 0 FAIL, 3 SKIP
- `tests/destructor/` — all PASS
- `tests/cpp/` — all PASS (single unrelated failure: `tasync_cpp.nim`
needs the `jester` package)
- `megatest` — PASS for both `--mm:arc` and `--mm:refc`, including the
previously regressing `tfinally.nim`'s `nested_finally`

## Backport

Tagged `[backport]` in the commit message for inclusion in
`version-2-2`.

---------

Co-authored-by: puffball1567 <17452514+puffball1567@users.noreply.github.com>
2026-05-05 21:27:33 +08:00
Andreas Rumpf
4bcb706d49 IC: added support for conditional dependencies (#25770) 2026-04-28 18:48:41 +02:00
ringabout
92d0c097e5 fixes #25140; Cannot resolve pragmas when new type is defined from typeof expression (#25764)
fixes #25140
2026-04-27 18:47:45 +02:00
Andreas Rumpf
49b5e66d3a SSO: better switch to enable it (#25772) 2026-04-27 18:00:29 +02:00
Zoom
cbe8ce59ed fix string setLenUninit growth without realloc for refc (#25767)
`setLenUninit(string)` was broken on the legacy refc backend when
growing within existing spare capacity.

`setLengthStrUninit` in `lib/system/sysstr.nim` only updated len when it
had to reallocate or when shrinking.

If oldLen < newLen <= capacity, it returned early without finalizing:

```nim
var s = newStringOfCap(10)
s.add("abc")
s.setLenUninit(6)
doAssert s.len == 6 # used to fail, len stayed 3
```

This escaped `tests/stdlib/tstring.nim` because the testing routine
`checkSetLenUninit` mostly resizes strings created at **exact**
length/capacity, so growth usually took the reallocating branch.

The new regression test covers the missing edge case.

So sorry for catching this only on the day of the stable release! In my
defense, the original PR hung in limbo for quite a while and it didn't
spend enough time in devel after the merge.
2026-04-25 12:27:13 +02:00
Jake Leahy
148e82f418 Add Nix certificate path to ssl_certs.nim (#25763)
This makes it easier to run Nix built containers for Nim programs since
by default Nim doesn't search environment variables for SSL certs so its
a little annoying having to move around files

-
10e7ad5bbc/pkgs/by-name/ca/cacert/package.nix (L85)
2026-04-24 14:04:30 +02:00
Tomohiro
8b44b9d9ae fixes #23668; Create a new std/nre2 module using Nim Regex replaces re and nre (#25696)
std/nre2 is implemented using https://github.com/nitely/nim-regex.
std/nre2 has almost same features as std/nre but some regular
expressions supported by std/nre are not supported.
The syntax of regular expressions of Nim Regex is explained in:
https://nitely.github.io/nim-regex/regex.html
2026-04-22 08:06:03 +02:00
ringabout
efacf1f390 Fix typo in getContentType function in cgi.nim (#25757)
This pull request fixes a typo in the `getContentType` function in
`lib/pure/cgi.nim`, ensuring it retrieves the correct `CONTENT_TYPE`
environment variable.

> Exact spelling matters: It is CONTENT_TYPE, not CONTENT_Type or
Content-Type. Environment variables in CGI are case-sensitive.
2026-04-21 16:38:56 +02:00
ringabout
60bb9c75cc fixes #25650; nim ic import std/strbasics (#25760)
fixes #25650

This pull request refactors and improves the dependency resolution logic
in the Nim compiler, The most important changes are grouped below:

### Dependency Resolution Refactor

* Replaced the `resolveFile` procedure with two more specialized
procedures: `resolveImport` (which uses the compiler's module lookup
rules for imports) and `resolveInclude` (which resolves includes
relative to the including file or search paths). Updated all usages
accordingly, improving clarity and correctness of dependency handling.
[[1]](diffhunk://#diff-1203947eecb9ef641ce7ee029677f875eb983de050b82c65ca286517feed00e6L82-R94)
[[2]](diffhunk://#diff-1203947eecb9ef641ce7ee029677f875eb983de050b82c65ca286517feed00e6L106-R103)
[[3]](diffhunk://#diff-1203947eecb9ef641ce7ee029677f875eb983de050b82c65ca286517feed00e6L121-R118)
* Removed the unused `strutils` import from `compiler/deps.nim` for
cleaner dependencies.


### Testing Improvements

* Added `import std/strbasics` to `tests/ic/tmiscs.nim` to ensure
required symbols are available for tests.


I tried to improve `resolveFile`, which is harder because either we need
to add `lib/std` to search path and all of other nested directory to
`--path` in `config/nim.cfg`. So I choose toi reuse `findModule` for
imports
2026-04-21 16:38:33 +02:00
Bojun Chai
de3d61f15b Fix invalid Mac OS X minimum version in README (#25758)
**Repo:** nim-lang/Nim ( 16000)
**Type:** docs
**Files changed:** 1
**Lines:** +1/-1

## What
Correct the supported platform table in the top-level README by changing
the Mac OS X minimum version from `10.04` to `10.4`.

## Why
`10.04` is not a valid Mac OS X release number, so the existing text is
misleading for anyone reading the build and platform support guidance.
Fixing it keeps the README accurate without changing project behavior or
widening scope.

## Testing
Verified the README diff locally and confirmed the corrected `Mac OS X
(10.4 or greater)` entry appears in `readme.md`. No code or test suite
changes were needed for this docs-only patch.

## Risk
Low / documentation-only change with no runtime impact.

Co-authored-by: Bojun Chai <bojunchai@microsoft.com>
2026-04-21 08:50:13 +08:00
Tomohiro
ba4e12fb65 fixes #25753 (#25754) 2026-04-20 20:13:06 +02:00
Andreas Rumpf
f236e6a210 fixes #25695 (#25756) 2026-04-20 09:17:12 +02:00
Tomohiro
317bc10824 Makes containsOrIncl*[A](s: var PackedSet[A], key: A) proc faster (#25755)
This PR makes it faster when a number of elements is less than 34
I used following code to compare the speed of `containsOrIncl` proc.
It calls `isRecursiveStructuralType` proc defined in compiler/types.nim
that calls `containsOrIncl` with `IntSet`(= `PackedSet[int]`).
```nim
import std/[tables, monotimes, times, strformat]
import "$nim"/compiler/[astdef, ast, idents, types]

var idgen = IdGenerator(module: 0, symId: 0, typeId: 0, disambTable: initCountTable[PIdent]())

proc newType(kind: TTypeKind; son: sink PType = nil): PType =
  result = newType(kind, idgen, nil, son)

proc genNoRecursPType(len: int): PType =
  assert len > 1
  let intTyp = newType(tyInt)
  result = newType(tyRef, intTyp)
  for i in 0..<(len - 2):
    result = newType(tyRef, result)

proc test =
  var noRecursPType = genNoRecursPType(4)
  assert not isRecursiveStructuralType(noRecursPType)

test()

template measure(label: string; body: untyped): untyped =
  let
    loop = 2000
    sampling = 200
  block:
    var r {.inject.} = false
    var minT = initDuration(hours = 1)
    for i in 0 ..< sampling:
      let start = getMonoTime()
      for j in 0 ..< loop:
        body
      let finish = getMonoTime()
      minT = min(finish - start, minT)
    echo ($r)[0], ' ', label, minT div loop

proc benchNoRecurs(len: int) =
  echo fmt"No recursive: length: {len}"
  var noRecursPType = genNoRecursPType(len)
  measure("IntSet: "):
    r = isRecursiveStructuralType(noRecursPType)

proc bench =
  benchNoRecurs(30)

bench()
```

Output before changing code:
```
f IntSet: 1 microsecond and 262 nanoseconds
```
Output after change:
```
f IntSet: 833 nanoseconds
```

Why this PR make it faster:
```nim
proc containsOrIncl*[A](s: var PackedSet[A], key: A): bool =
  ...
  if s.elems <= s.a.len:
    for i in 0..<s.elems:
      if s.a[i] == ord(key):
        return true
    # `incl` scans `s.a` again
    incl(s, key)
    result = false
```

```nim
proc containsOrIncl*[A](s: var PackedSet[A], key: A): bool =
  ...
  if s.elems <= s.a.len:
    for i in 0..<s.elems:
      if s.a[i] == ord(key):
        return true
    if s.elems < s.a.len:
      # put `key` in `s.a` instead of calling `incl(s, key)`
      s.a[s.elems] = ord(key)
      inc(s.elems)
    else:
      incl(s, key)
    result = false
```
2026-04-20 09:21:46 +08:00
ringabout
5948dbbeed fixes #25718; setLenUnit slow (#25743)
fixes #25718

This pull request optimizes sequence allocation in the Nim standard
library by introducing a way to create uninitialized sequence payloads
for element types that don't require zero-initialization. The changes
allow for more efficient memory allocation when initializing sequences
with types that have no references, avoiding unnecessary zeroing of
memory.

Sequence allocation and initialization improvements:

* Added the `newSeqUninitRaw` procedure to create sequence payloads with
a specified length without forcing zero-initialization for element types
marked as `ntfNoRefs`. (`lib/system/sysstr.nim`,
[lib/system/sysstr.nimR277-R292](diffhunk://#diff-bcaa1967f436ad03877f353823c08a8b4a719fe387629d33aab4bddf16534b5eR277-R292))
* Modified the `extendCapacityRaw` procedure and the `setLengthSeqImpl`
template to use `newSeqUninitRaw` when zero-initialization is not
required, controlled by the `doInit` static parameter.
(`lib/system/sysstr.nim`,
[[1]](diffhunk://#diff-bcaa1967f436ad03877f353823c08a8b4a719fe387629d33aab4bddf16534b5eR277-R292)
[[2]](diffhunk://#diff-bcaa1967f436ad03877f353823c08a8b4a719fe387629d33aab4bddf16534b5eL316-R335)
2026-04-19 20:12:01 +02:00
ringabout
98131a9fa1 fixes #25751; JS backend crashes when returning Option[T] with custom =destroy (#25752)
fixes #25751

This pull request improves the JavaScript backend code generation and
expands test coverage, particularly around temporary and loop variables,
as well as object destruction behavior. The main changes include
updating the code generator to handle more symbol kinds and adding tests
to ensure proper destruction and option handling.

**JavaScript code generation improvements:**

* Updated `genSymAddr` in `compiler/jsgen.nim` to support additional
symbol kinds, specifically `skTemp` and `skForVar`, ensuring correct
address generation for temporaries and loop variables.

**Test suite enhancements:**

* Added tests in `tests/js/test2.nim` to verify correct behavior of
option types, object destruction (`=destroy`), and to check for
backend-specific crashes. This includes printing results of
option-returning functions and confirming destruction messages.
* Updated expected output in `tests/js/test2.nim` to include results
from new tests and destruction messages, ensuring the test suite
reflects the latest code behavior.
2026-04-18 09:40:55 +02:00
Ryan McConnell
f98578ea35 fix 25667; Generic forward type confusion (#25737)
ref: #25667

drain deferred reification in a loop until there is no more work to do.
Could potentially evaluate the same deferred work more than once.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2026-04-18 08:52:31 +02:00
dependabot[bot]
e6e00a74a3 Bump actions/github-script from 8 to 9 (#25748)
Bumps [actions/github-script](https://github.com/actions/github-script)
from 8 to 9.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/actions/github-script/releases">actions/github-script's
releases</a>.</em></p>
<blockquote>
<h2>v9.0.0</h2>
<p><strong>New features:</strong></p>
<ul>
<li><strong><code>getOctokit</code> factory function</strong> —
Available directly in the script context. Create additional
authenticated Octokit clients with different tokens for multi-token
workflows, GitHub App tokens, and cross-org access. See <a
href="https://github.com/actions/github-script#creating-additional-clients-with-getoctokit">Creating
additional clients with <code>getOctokit</code></a> for details and
examples.</li>
<li><strong>Orchestration ID in user-agent</strong> — The
<code>ACTIONS_ORCHESTRATION_ID</code> environment variable is
automatically appended to the user-agent string for request
tracing.</li>
</ul>
<p><strong>Breaking changes:</strong></p>
<ul>
<li><strong><code>require('@actions/github')</code> no longer works in
scripts.</strong> The upgrade to <code>@actions/github</code> v9
(ESM-only) means <code>require('@actions/github')</code> will fail at
runtime. If you previously used patterns like <code>const { getOctokit }
= require('@actions/github')</code> to create secondary clients, use the
new injected <code>getOctokit</code> function instead — it's available
directly in the script context with no imports needed.</li>
<li><code>getOctokit</code> is now an injected function parameter.
Scripts that declare <code>const getOctokit = ...</code> or <code>let
getOctokit = ...</code> will get a <code>SyntaxError</code> because
JavaScript does not allow <code>const</code>/<code>let</code>
redeclaration of function parameters. Use the injected
<code>getOctokit</code> directly, or use <code>var getOctokit =
...</code> if you need to redeclare it.</li>
<li>If your script accesses other <code>@actions/github</code> internals
beyond the standard <code>github</code>/<code>octokit</code> client, you
may need to update those references for v9 compatibility.</li>
</ul>
<h2>What's Changed</h2>
<ul>
<li>Add ACTIONS_ORCHESTRATION_ID to user-agent string by <a
href="https://github.com/Copilot"><code>@​Copilot</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/695">actions/github-script#695</a></li>
<li>ci: use deployment: false for integration test environments by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/712">actions/github-script#712</a></li>
<li>feat!: add getOctokit to script context, upgrade
<code>@​actions/github</code> v9, <code>@​octokit/core</code> v7, and
related packages by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/700">actions/github-script#700</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a href="https://github.com/Copilot"><code>@​Copilot</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/695">actions/github-script#695</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/github-script/compare/v8.0.0...v9.0.0">https://github.com/actions/github-script/compare/v8.0.0...v9.0.0</a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="3a2844b7e9"><code>3a2844b</code></a>
Merge pull request <a
href="https://redirect.github.com/actions/github-script/issues/700">#700</a>
from actions/salmanmkc/expose-getoctokit + prepare re...</li>
<li><a
href="ca10bbdd1a"><code>ca10bbd</code></a>
fix: use <code>@​octokit/core/</code>types import for v7
compatibility</li>
<li><a
href="86e48e20ac"><code>86e48e2</code></a>
merge: incorporate main branch changes</li>
<li><a
href="c1084728b5"><code>c108472</code></a>
chore: rebuild dist for v9 upgrade and getOctokit factory</li>
<li><a
href="afff112e4f"><code>afff112</code></a>
Merge pull request <a
href="https://redirect.github.com/actions/github-script/issues/712">#712</a>
from actions/salmanmkc/deployment-false + fix user-ag...</li>
<li><a
href="ff8117e5b7"><code>ff8117e</code></a>
ci: fix user-agent test to handle orchestration ID</li>
<li><a
href="81c6b78760"><code>81c6b78</code></a>
ci: use deployment: false to suppress deployment noise from integration
tests</li>
<li><a
href="3953caf885"><code>3953caf</code></a>
docs: update README examples from <a
href="https://github.com/v8"><code>@​v8</code></a> to <a
href="https://github.com/v9"><code>@​v9</code></a>, add getOctokit docs
and v9 brea...</li>
<li><a
href="c17d55b90d"><code>c17d55b</code></a>
ci: add getOctokit integration test job</li>
<li><a
href="a047196d9a"><code>a047196</code></a>
test: add getOctokit integration tests via callAsyncFunction</li>
<li>Additional commits viewable in <a
href="https://github.com/actions/github-script/compare/v8...v9">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/github-script&package-manager=github_actions&previous-version=8&new-version=9)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-17 19:14:36 +08:00
ringabout
c22819ef17 fixes #25732; semStaticExpr and semStaticStmt to handle errors (#25742)
fix #25732
2026-04-17 10:00:00 +02:00
ringabout
2b2872928b fixes #25469; Conversion from distinct in for forces a copy of underlying instance (#25746)
fixes #25469

This pull request introduces an important fix to argument handling in
the compiler's transformation logic and adds a new test to verify
correct behavior with distinct types and ARC memory management.

### Compiler transformation improvements

* Updated `putArgInto` in `compiler/transf.nim` to handle
`nkHiddenStdConv`, `nkHiddenSubConv`, and `nkConv` nodes more
accurately. Now, if the types match (ignoring distinctness and shallow
range differences), the argument is recursively processed; otherwise, it
falls back to a fast assignment. This prevents incorrect assignments
when dealing with type conversions and distinct types.

### Testing for distinct types and ARC

* Added a new test `tdistinct_for_nodup.nim` to ensure correct iteration
and memory management for distinct sequences of large arrays under ARC.
The test checks that the sequence length remains unchanged during
iteration, helping catch regressions related to ARC and distinct types.
2026-04-17 09:59:22 +02:00
Andreas Rumpf
b4d4028afa fixes whitespace related endless loop in renderer.nim (#25750) 2026-04-16 09:44:58 +02:00
Sai Asish Y
3eb4a60b6b ccgstmts: fix 'occured' -> 'occurred' typo in emitted C++ exception comment (#25749)
Inline C++ comment emitted by `compiler/ccgstmts.nim:1168` into
generated code read `C++ exception occured, not under Nim's control`.
Doc-only change in the emitted source.

Signed-off-by: SAY-5 <SAY-5@users.noreply.github.com>
Co-authored-by: SAY-5 <SAY-5@users.noreply.github.com>
2026-04-16 12:29:09 +08:00
ringabout
7b73537131 fixes #25735; sso C++: nimToCStringConv (#25745)
fixes #25735

This pull request updates how string-to-C-string conversions are handled
when the `nimsso` configuration flag is enabled, and adds a new system
test to validate the behavior. The main changes focus on switching from
using `addrLoc` to `byRefLoc` for argument preparation, which likely
improves correctness or compatibility with the `nimsso` mode.

**Code generation improvements for `nimsso` mode:**

* In both `compiler/ccgcalls.nim` (`genArgStringToCString`) and
`compiler/ccgexprs.nim` (`convStrToCStr`), replaced the use of `addrLoc`
with `byRefLoc` when preparing arguments for string-to-C-string
conversions under the `nimsso` configuration flag. This change ensures
that references are handled appropriately according to the requirements
of `nimsso`.
[[1]](diffhunk://#diff-42181cc6f4202af843e7835ea514df2efe85e4faae3bc797a39a0c422547b558L373-R373)
[[2]](diffhunk://#diff-4509107d295d7d32b1887c8993cd0f56113ae60f36113e7d8778646dabd92ebcL2739-R2739)

**Testing:**

* Added a new system test `tests/system/tnimsso.nim` that runs with the
`-d:nimsso` flag on both C and C++ targets, checking that
string-to-C-string conversion works as expected in `nimsso` mode.
2026-04-15 14:57:57 +02:00
Andreas Rumpf
5b1a05e282 fixes #18095 (#25744) 2026-04-14 19:58:44 +02:00
Zoom
4dbc382906 Feat: stdlib: adds system.string.setLenUninit (#24836)
Adds `system.setLenUninit` for the `string` type. Allows setting length
without initializing new memory on growth.

- Required for a follow-up to #15951
- Accompanies #22767 (ref #19727) but for strings
- Expands `stdlib/tstring` with tests for `setLen` and `setLenUninit`

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2026-04-14 17:24:26 +02:00
Andreas Rumpf
e81f5b5890 Revert "only generate called hook for explicit or generated destructor calls [backport]" (#25741)
Reverts nim-lang/Nim#25729
2026-04-13 18:02:53 +08:00
lit
cf3c28c223 fixes #25738; std/parseopt: - causes IndexDefect (#25739) 2026-04-12 12:32:25 +02:00
Ryan McConnell
242f761627 RE: fix #25627 (#25736)
Follow up PR to #25700

@demotomohiro 

This doesn't seem to mirror your suggested approach completely. I still
went with a recursive walk. Could probably add some kind of "clean
types" and "dirty types" cache through this to minimize the recursions,
but that seems like a little much.
2026-04-12 08:56:31 +02:00
metagn
a35614e539 make explicit copy and hook calls keep their symbol (#25731)
fixes #25730

As mentioned in the issue this results in less optimized output, it
always generates the explicitly called hook as a proc rather than an
inline assignment. But maybe this is a reasonable trade since it only
happens on explicit `=sink`/`=copy` calls.

Any way to optimize it requires detecting either the type or the found
hook as a trivial assignment. I am not sure how to do these, the hook
isn't like destructors that propagate empty statements in
`liftdestructors` (which is what `isTrivial` checks for), it needs to
propagate simple assignments instead. And there is no logic for the
type, `tfHasAsgn` is misleading since it only checks if the destructor
is trivial, because there is no check for a trivial assignment.

Did not mark as backported but the only issue I can think of is the
performance issue above, otherwise it would be more correct if anything.
2026-04-12 07:05:48 +02:00
metagn
fb02e9831d only generate called hook for explicit or generated destructor calls [backport] (#25729)
fixes #25727, regression from #24627 which was backported to 2.2.2 and
2.0.16

Instead of calling `createTypeBoundOps` for explicit hook calls and when
generating default hooks, only the called destructor is generated at a
time. This allows defining more than 1 hook for recursive types.

`=sink` for `useSeqOrStrOp` and also `atomicRefOp` always need a
`=destroy` hook generated so that is also generated separately. There
might be more that I missed, only the atomicRefOp one failed `trtree` in
CI, and it was just from a compiler assert that got triggered, otherwise
it would still have functioned.
2026-04-12 07:05:11 +02:00
ringabout
6353c4e5b0 fixes #25724; Invalid C code generation with iterator/nimvm (#25728)
fixes #25724

This pull request introduces a small but important fix in the compiler
and adds a new test case related to iterators. The main change in the
compiler ensures that lambda-like constructs are handled consistently
with other procedure definitions, while the new test in the suite covers
a previously untested scenario.

**Compiler improvements:**
* Updated `introduceNewLocalVars` in `compiler/transf.nim` to handle all
`nkLambdaKinds` in addition to `nkProcDef`, `nkFuncDef`, `nkMethodDef`,
and `nkConverterDef`, ensuring consistent transformation of all
lambda-like constructs.

**Testing:**
* Added a block to `tests/iter/titer_issues.nim` to test iterator
behavior in both compile-time and run-time contexts, addressing bug
#25724.
2026-04-10 15:57:26 +02:00
ringabout
e39272eaa8 fixes #25637; nim ic with destructors (#25723)
fixes #25637

This pull request refactors the way the `sfInjectDestructors` flag is
set on symbols during lambda lifting in the Nim compiler. The main
change is the introduction of a helper procedure to encapsulate the
logic for marking symbols that require destructor injection, improving
code clarity and maintainability.

Refactoring and code quality improvements:

* Introduced the `markInjectDestructors` procedure to encapsulate the
logic for marking a symbol with the `sfInjectDestructors` flag, ensuring
that `backendEnsureMutable` is always called before modifying the
symbol's flags.
* Replaced direct flag manipulation (`owner.incl sfInjectDestructors`
and `prc.incl sfInjectDestructors`) with calls to the new
`markInjectDestructors` procedure in multiple locations, including
`makeClosure`, `createTypeBoundOpsLL`, and `rawClosureCreation`.
[[1]](diffhunk://#diff-19193904ba011a2bcc1e1a9768a7eb57cac57a274cad73d388149776ec2901e6L231-R235)
[[2]](diffhunk://#diff-19193904ba011a2bcc1e1a9768a7eb57cac57a274cad73d388149776ec2901e6L243-R247)
[[3]](diffhunk://#diff-19193904ba011a2bcc1e1a9768a7eb57cac57a274cad73d388149776ec2901e6L639-R643)
2026-04-10 15:29:20 +02:00
Ryan McConnell
2501e23d81 fixes #25290; tempalte overload scope dupe (#25308)
#25290
drafted bc if this passes full CI I am going to try and remove that
weird stuff in `pickBestCandidate`
2026-04-09 20:44:35 +02:00
ringabout
188aa1714e fixes #25719; optimizes setLenSeqCall for orc (#25721)
fixes #25719

This pull request updates the logic for resizing sequences during
certain copy operations in the `compiler/liftdestructors.nim` file. The
main improvement is that the code now distinguishes between regular and
uninitialized resizing based on whether the sequence's element type
supports bulk memory copying, which can lead to more efficient code
generation.

**Improvements to sequence resizing and copying logic:**

* Modified `setLenSeqCall` to accept a `noinit` parameter, allowing it
to choose between `setLen` and `setLenUninit` operations, and to select
the appropriate magic for each case.
* Updated `fillSeqOp` to determine if bulk memory copy is supported and,
if so, call `setLenSeqCall` with `noinit = true` and perform a bulk
copy; otherwise, it defaults to element-wise copying. This logic is now
applied in both relevant locations in the function.
[[1]](diffhunk://#diff-456118dde9a4e21f1b351fd72504d62fc16e9c30354dbb9a3efcb95a29067863L646-R650)
[[2]](diffhunk://#diff-456118dde9a4e21f1b351fd72504d62fc16e9c30354dbb9a3efcb95a29067863L661-R666)
2026-04-09 11:11:06 +02:00
Ryan McConnell
fa6b754dbc fix #25627 (#25700)
@demotomohiro this was caused by your PR please review
#25627
2026-04-09 11:09:34 +02:00
ringabout
9a2b0dd045 fixes #25697; {.borrow.} on iterator for distinct seq triggers internal error (#25709)
fixes #25697

This pull request improves the handling of borrowed routines in the
compiler transformation phase, making the code more robust and
maintainable. The main change is the introduction of a helper function
to properly resolve borrowed routine symbols, which is then used in
multiple places to ensure correct symbol resolution. Additionally, a new
test case is added to cover a previously reported bug related to
borrowed iterators on distinct types.

**Compiler improvements:**

* Added `resolveBorrowedRoutineSym` helper function to follow borrow
aliases and retrieve the underlying implementation symbol for borrowed
routines. This centralizes and clarifies the logic for resolving
borrowed symbols.
* Updated `transformSymAux` and `transformFor` to use the new helper
function, replacing duplicated logic and improving correctness when
handling borrowed routines.
[[1]](diffhunk://#diff-c7b80f51fb685eb22c5b56ee2f320d6c708706f3ae7293478ecd104a2b5b8096L139-R154)
[[2]](diffhunk://#diff-c7b80f51fb685eb22c5b56ee2f320d6c708706f3ae7293478ecd104a2b5b8096L788-R795)

**Testing:**

* Added a test case for bug #25697 to `tests/distinct/tborrow.nim`,
ensuring that iteration over a distinct type with a borrowed iterator
works as expected.
2026-04-09 11:08:03 +02:00
ringabout
c8e6b059a4 optimizes setLen for orc; disabling overflow checks (#25722)
ref https://github.com/nim-lang/Nim/issues/25695
ref https://github.com/nim-lang/Nim/pull/25715

This pull request introduces a minor but important change to the
`setLen` procedure in `lib/system/seqs_v2.nim`. The main update is the
temporary disabling of overflow checks during the initialization loop
when extending the sequence length, which can improve performance and
avoid unnecessary checks during this operation.

Memory and performance improvement:

* Disabled overflow checks for the loop that initializes new elements to
their default value when increasing the length of a sequence in
`setLen`, by wrapping the loop with `{.push overflowChecks: off.}` and
`{.pop.}`.
2026-04-09 11:07:04 +02:00
lou15b
115ec7a433 Fixes #25710 - nimsuggest outline misses methods (#25711)
This adds methods to the list generated by the `outline` command for
`nimsuggest --v3` and `nimsuggest --v4`.
The test file `tv3_outline.nim` was also updated to include a `skMethod`
line in the expected output.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2026-04-08 00:52:12 +02:00
metagn
0dc577a4dc fix compiler crash regression with explicit destructor calls [backport:2.2] (#25717)
Unfortunately I do not have a test case for this (although I can link
[this package
test](60f1be9037/tests/test_simple_combined.nim)
which broke), but this is a regression caused by #24841 (which was
backported to 2.2.4) that causes the following compiler crash:

```
assertions.nim(34)       raiseAssert
Error: unhandled exception: ccgtypes.nim(230, 13) `false` mapType: tyGenericInvocation [AssertionDefect]
```

Codegen is traversing the type of the symbol of an explicit destructor
call, but the symbol is the uninstantiated generic hook. This happens
because #24841 changed the code which gives explicit destructor calls
the proper attached destructor to use `replaceHookMagic`, which now
skips `abstractVar` from the type to get the destructor whereas
previously it was just `{tyAlias, tyVar}`. This skips `tyGenericInst`
and also `tyDistinct`. I cannot explain why the skipped `tyGenericInst`
does not have the right destructor but it's not really unexpected, and
skipping `tyDistinct` is just wrong.

To fix this, just `{tyAlias, tyVar, tySink}` are skipped.
2026-04-08 00:51:47 +02:00
Andreas Rumpf
6621d64398 fixes #25577 (#25691) 2026-04-07 18:07:34 +02:00
Ryan McConnell
184d423779 fixes 25713; Allow addr of object variant's discriminant under uncheckedAssign (#25714)
#25713

```nim
type
  K = enum
    k1,k2
  Variant = object
    case kind: K
    of k1:
      discard
    of k2:
      discard

proc a(x: var K) = discard
proc b(x: ptr K) = discard

var x = Variant(kind: k1)
{.cast(uncheckedAssign).}:
  # must be within uncheckedAssign to work
  a(x.kind)
# doesn't work out of or under uncheckedAssign
b(addr x.kind)
```
2026-04-07 06:59:24 +02:00
Jake Leahy
f9524861f3 Fix generic tuple unpacking in iterators (#25705)
Fixes #25704 

This makes sure that `iter` still has `tyGenericInst` skipped like
before, without skipping it for `iterType` which requires it
2026-04-05 14:00:04 +02:00
dxxb
0028ea563c Fix inconsistent env type with nested procs in iterators (#21242) (#25699)
Nested transformBody/liftLambdas passes used a fresh DetectionPass, so
getEnvTypeForOwner could allocate a duplicate PType for the same owner
while :envP already referenced the inner pass type. When addClosureParam
saw cp.typ != t, it errored.

If both types are env objects for the same routine owner, reuse cp.typ
and sync ownerToType.

Adds regression test tests/iter/t21242_nested_closure_in_iter.nim.
2026-04-04 19:47:01 +02:00
ringabout
854c1f15ba fixes #25687; optimizes seq assignment for orc (#25689)
fixes #25687

This pull request introduces an optimization for sequence (`seq`)
assignments and copies in the Nim compiler, enabling bulk memory copying
for sequences whose element types are trivially copyable (i.e., no GC
references or destructors). This can significantly improve performance
for such types by avoiding per-element loops.

Key changes:

### Compiler code generation improvements

* Added the `elemSupportsCopyMem` function in
`compiler/liftdestructors.nim` to detect if a sequence's element type is
trivially copyable (no GC refs, no destructors).
* Updated the `fillSeqOp` procedure to use a new `genBulkCopySeq` code
path for eligible element types, generating a call to
`nimCopySeqPayload` for efficient bulk copying. Fallback to the
element-wise loop remains for non-trivial types.
[[1]](diffhunk://#diff-456118dde9a4e21f1b351fd72504d62fc16e9c30354dbb9a3efcb95a29067863R665-R670)
[[2]](diffhunk://#diff-456118dde9a4e21f1b351fd72504d62fc16e9c30354dbb9a3efcb95a29067863R623-R655)

### Runtime support

* Introduced the `nimCopySeqPayload` procedure in
`lib/system/seqs_v2.nim`, which performs the actual bulk memory copy of
sequence data using `copyMem`. This is only used for types that are safe
for such an operation.

These changes collectively improve the efficiency of sequence operations
for simple types, while maintaining correctness for complex types.


### Benchmarked the original micro-benchmark:
refc: 3.52s user 0.02s system 99% cpu 3.538 total
orc (after change): 3.46s user 0.01s system 99% cpu 3.476 total

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-04-02 11:46:49 +02:00
Andreas Rumpf
d389d4fb2f SSO for strings (#25593) 2026-04-02 07:19:43 +02:00
Zoom
be29bcd402 Fix iterable resolution, prefer iterator overloads (#25679)
This fixes type resolution for `iterable[T]`.

I want to proceed with RFC
[#562](https://github.com/nim-lang/RFCs/issues/562) and this is the main
blocker for composability.

Fixes #22098 and, arguably, #19206

```nim
import std/strutils

template collect[T](it: iterable[T]): seq[T] =
  block:
    var res: seq[T] = @[]
    for x in it:
      res.add x
    res

const text = "a b c d"

let words = text.split.collect()
doAssert words == @[ "a", "b", "c", "d" ]
```

In cases like `strutils.split`, where both proc and iterator overload
exists, the compiler resolves to the `func` overload causing a type
mismatch.

The old mode resolved `text.split` to `seq[string]` before the
surrounding `iterable[T]` requirement was applied, so the argument no
longer matched this template.

It should be noted that, compared to older sequtils templates,
composable chains based on `iterable[T]` require an iterator-producing
expression, e.g. `"foo".items.iterableTmpl()` rather than just
`"foo".iterableTmpl()`. This is actually desirable: it keeps the
iteration boundary explicit and makes iterable-driven templates
intentionally not directly interchangeable with older
untyped/loosely-typed templates like those in `sequtils`, whose internal
iterator setup we have zero control over (e.g. hard-coding adapters like
`items`).

Also, I noticed in `semstmts` that anonymous iterators are always
`closure`, which is not that surprising if you think about it, but still
I added a paragraph to the manual.

Regarding implementation:

From what I gathered, the root cause is that `semOpAux` eagerly
pre-types all arguments with plain flags before overload resolution
begins, so by the time `prepareOperand` processes `split` against the
`iterable[T]`, the wrong overload has already won.

The fix touches a few places:

- `prepareOperand` in `sigmatch.nim`:
When `formal.kind == tyIterable` and the argument was already typed as
something else, it's re-semchecked with the
`efPreferIteratorForIterable` flag. The recheck is limited to direct
calls (`a[0].kind in {nkIdent, nkAccQuoted, nkSym, nkOpenSym}`) to avoid
recursing through `semIndirectOp`/`semOpAux` again.

- `iteratorPreference` field `TCandidate`, checked before
`genericMatches` in `cmpCandidates`, gives the iterator overload a win
without touching the existing iterator heuristic used by `for` loops.

**Limitations:**

The implementation is still flag-driven rather than purely
formal-driven, so the behaviour is a bit too broad `efWantIterable` can
cause iterator results to be wrapped as `tyIterable` in
iterable-admitting contexts, not only when `iterable[T]` match is being
processed.

`iterable[T]` still does not accept closure iterator values such
as`iterator(): T {.closure.}`. It only matches the compiler's internal
`tyIterable`, not arbitrary iterator-typed values.

The existing iterator-preference heuristic is still in place, because
when I tried to remove it, some loosely-related regressions happened. In
particular, ordinary iterator-admitting contexts and iterator chains
still rely on early iterator preference during semchecking, before the
compiler has enough surrounding context to distinguish between
value/iterator producing overloads. Full heuristic removal would require
a broader refactor of dot-chain/intermediate-expression semchecking,
which is just too much for me ATM. This PR narrows only the
tyIterable-specific cases.

**Future work:**

Rework overload resolution to preserve additional information of
matching iterator overloads for calls up to the point where the
iterator-requiring context is established, to avoid re-sem in
`prepareOperand`.

Currently there's no good channel to store that information. Nodes can
get rewritten, TCandidate doesn't live long enough, storing in Context
or some side-table raises the question how to properly key that info.
2026-04-01 21:01:55 +02:00
ringabout
9c07bb94c1 fixes #25682; fix vm genAsgn to handle statementListExpr (#25686)
fixes #25682

This pull request introduces a fix to the Nim compiler's assignment code
generation logic to better handle statement list expressions, and adds
regression tests to ensure correct behavior when assigning to object
fields via templates. The changes address a specific bug (#25682)
related to assignments using templates with side effects in static
contexts.

**Compiler code generation improvements:**

* Updated the `genAsgn` procedure in `compiler/vmgen.nim` to properly
handle assignments where the left-hand side is a `nkStmtListExpr`
(statement list expression), ensuring all statements except the last are
executed before the assignment occurs.

**Regression tests for assignment semantics:**

* Added new test blocks in `tests/vm/tvmmisc.nim` to verify that
template-based assignments to object fields work as expected in static
contexts, specifically testing for bug #25682.
2026-04-01 00:24:06 +02:00
ringabout
fb31e86537 fixes #25632; errors incompatibility between {.error.} and {.exportc} pragmas in semProcAux (#25639)
fixes #25632
fixes #25631
fixes #25630

This pull request introduces a compatibility check between the
`{.error.}` and `{.exportc.}` pragmas in procedure declarations.
Specifically, it prevents a procedure from being marked with both
pragmas at the same time, as this combination is now considered invalid.

Pragma compatibility enforcement:

* Added a check in `semProcAux` (in `compiler/semstmts.nim`) to emit a
local error if a procedure is declared with both `{.error.}` and
`{.exportc.}` pragmas, preventing their incompatible usage.
2026-03-31 23:32:24 +02:00
Jacek Sieka
e53058dee0 windows: prefer 64-bit time_t (#25666)
time_t should be a 64-bit type on all relevant windows CRT versions
including mingw-w64 - MSDN recommends against using the 32-bit version
which only is happens when `_USE_32BIT_TIME_T` is explicitly defined -
instead of guessing (and guessing wrong, as happens with recent mingw
versions), we can simply use the 64-bit version always.
2026-03-31 09:52:11 +02:00
ringabout
8076fb40b8 fixes transf cannot handle bare sym for nim ic (#25664)
```nim
template compute(body: untyped): int =
  block:
    body
let x = compute:
  var sum = 0
  for i in 1..10: sum += i
  sum
echo x
```

supersedes https://github.com/nim-lang/Nim/pull/25653

which in
02893e2f4c

```nim
of nkSym:
    genSingleVar(p, it.sym, newSymNode(it.sym), it.sym.astdef)
```

A new branch for `nkSym` is added, though more changes might be needed
if `nkSym` is handled specifically
2026-03-31 09:49:55 +02:00
ringabout
7a82c5920c fixes #25677; fixes #25678; typeAllowedAux to improve flag handling (#25684)
fixes #25677;
fixes #25678

This pull request introduces both a bug fix to the type checking logic
in the compiler and new test cases for lent types involving procedures
and tables. The most significant change is a refinement in how type
flags are handled for procedure and function types in the compiler,
which improves correctness in type allowance checks. Additionally, the
test suite is expanded to cover more complex scenarios with lent types
and table lookups.

**Compiler improvements:**

* Refined the handling of type flags in `typeAllowedAux` for procedure
and function types by introducing `innerFlags`, which removes certain
flags (`taObjField`, `taTupField`, `taIsOpenArray`) before recursing
into parameter and return types. This ensures more accurate type
checking and prevents inappropriate flag propagation.

**Testing enhancements:**

* Added new test blocks in `tests/lent/tlents.nim` to cover lent
procedure types stored in objects and used as table values, including a
function that retrieves such procedures from a table by key.
* Introduced a test case for an object containing a lent procedure
field, ensuring correct behavior when accessing and using these fields.
2026-03-30 15:09:11 +02:00
cui
78282b241f fixes #25674; parsecfg: bound-check CR/LF pair in replace() (#25675)
Fixes bug #25674.

`replace` read `s[i+1]` for a CRLF pair without ensuring `i+1 <
s.len()`, so a value ending in a lone `\\c` (quoted in `writeConfig`)
raised `IndexDefect`.

- Fix: only treat `\\c\\l` when the following character exists.
- Test: `tests/stdlib/tparsecfg.nim` block bug #25674 — fails before
fix, passes after.
2026-03-28 16:22:54 +08:00
cui
5c86c1eda9 fixes #25670; docgen: cmpDecimalsIgnoreCase max() used wrong index for b (#25669)
Fixes bug #25670.

The second argument to `max` in `cmpDecimalsIgnoreCase` used `limitB -
iA` instead of `limitB - iB`, which could mis-order numeric segments
when sorting doc index entries.
2026-03-27 10:20:10 +01:00
cui
7f6b76b34c fixes #25671; commands: fix --maxLoopIterationsVM positive check (#25672)
Fixes bug #25671.

The previous condition `not value > 0` was parsed as `(not value) > 0`,
not `not (value > 0)`, so the check did not reliably enforce a positive
`--maxLoopIterationsvm` limit. Align with `--maxcalldepthvm` by using
`value <= 0`.
2026-03-27 10:19:35 +01:00
ringabout
2fc9c8084c fixes #25658; two overflowed *= causes program deadloop sysFatal on --exceptions:goto (#25660)
fixes #25658
2026-03-27 09:00:14 +08:00
ringabout
e25820cf52 fixes #25642; Add support for static type in semTypeNode (#25646)
fixes #25642
2026-03-26 20:38:54 +01:00
Tomohiro
158d59ce48 fixes #25635; registers module suffix correctly (#25645)
`toNifFilename` proc doesn't return correct Nif file path because module
suffix is registered with wrong proc.
So `moduleFromNifFile` doesn't load the Nif file.
2026-03-24 21:34:46 +08:00
metagn
6f85d348f4 fix @ for openarray on nimscript [backport:2.2] (#25641)
Even on nimscript, the `else` branch of the `when nimvm` below compiles
and gives an "undeclared identifier: copyMem" error. Regression since
#25064.
2026-03-24 08:27:28 +01:00
ringabout
c48f487780 fixes: replace ensureMutable with backendEnsureMutable in ccgtypes (#25640) 2026-03-24 11:49:06 +08:00
ringabout
fb6fa96979 fixes #25626; Fix injection variable declaration in sequtils.nim (#25629)
fixes #25626

This pull request introduces a small change to the `mapIt` template in
`sequtils.nim`. The update adds the `used` pragma to the injected `it`
variable, which can help suppress unused variable warnings in certain
cases.

- Added the `used` pragma to the injected `it` variable in the `mapIt`
template to prevent unused variable warnings.

or it should give a better warning or something if `it` is not used
2026-03-23 20:31:09 +08:00
ringabout
7ef16ec7a1 Update NimonyStableCommit to a new version (#25638) 2026-03-23 20:28:19 +08:00
Zoom
c33df006c5 nimdoc: Document environment variable substitution (#25623)
Documents environment variable substitution.

Didn't find it mentioned anywhere, even though it's used widely by the
compiler docs.
2026-03-23 13:00:55 +01:00
Zoom
446d903fc1 nimdoc: CSS: tighter on mobile; fix h1 print page break (#25607)
- Small optimizations for mobile, makes code render slightly tighter.
- `font-stretch: semi-condensed;` for pre works if the user's font
provides such a face, shouldn’t change the rendering with the default.
- Removes an excessive page break after the page header when printing.

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2026-03-23 13:00:07 +01:00
Zoom
57e15cd9a4 nimdoc: Add a nav-burger to display the panel on mobile (#25606)
Small changes to the default html template and the `nimdoc.css`.

Adds a burger button to show the navigation panel when on narrow
screens/mobile. Displayed when the panel gets hidden.

Second element click or click on the dimmed background ides the panel.

# Demo:

![burger-action](https://github.com/user-attachments/assets/a10bd626-95a1-4a04-80bb-c159c85ac1a7)
2026-03-23 12:31:51 +01:00
Ryan McConnell
4414b5a396 small sets.nim cleanup in std (#25628)
mainly to fix `Uninit` warnings for projects that elevate it to an
error. Other changes are stylistic about redundancy or white-space
consistency.
2026-03-23 10:35:27 +01:00
ringabout
8bb63b475b fixes global tuple unpacking for nim ic (#25624)
This pull request includes a few targeted changes across the codebase,
primarily focusing on improving symbol locality detection in the
compiler, adding a utility function for integer division and modulus,
and simplifying a test case.

- **Compiler Improvements**
* Improved the `isLocalSym` function in `compiler/ast2nif.nim` to more
accurately determine if a symbol is local by checking that the symbol's
owner is not a module.

- **Utility Function Addition**
* Added a new `divmod` procedure in `tests/ic/tmiscs.nim` that returns
both the quotient and remainder of integer division, along with a usage
example.

- **Test Simplification**
* Simplified the `showMeters` test in `tests/ic/tconverter.nim` by
removing a floating-point assertion, leaving only an output statement.



------------------------------------------------------------------------------------------------------------------

```nim
proc divmod(a, b: int): (int, int) =
  (a div b, a mod b)


let (q, r) = divmod(17, 5)
echo q
echo r
```

gives `Error: unhandled exception: local symbol 'tmpTuple.0' not found
in localSyms. [AssertionDefect]`

`makeVarTupleSection` uses a temp of which the globalness and localness
is not specified. Turning it a global variable for top level scope broke
some Nim programs. So I think it's better to check the owner of the
symbol

```nim
if useTemp:
  # use same symkind for compatibility with original section
  let temp = newSym(symkind, getIdent(c.cache, "tmpTuple"), c.idgen, getCurrOwner(c), n.info)
```
2026-03-23 09:35:30 +01:00
ringabout
a4a482b5ef fixes #25620; typekey skips incorrectly the base type of seqs etc. types for nim ic (#25621)
fixes #25620

This pull request includes a fix to the type key generation logic in the
compiler and updates to a test file to cover additional language
features. The most important changes are summarized below:

### Compiler logic fix

* In `compiler/typekeys.nim`, the `typeKey` procedure was updated to
iterate over all elements in `t.sonsImpl` starting from index 0 instead
of 1, ensuring that all type sons are considered during type key
generation.

### Test suite improvements

* The test file `tests/ic/tenum.nim` was renamed to
`tests/ic/tmiscs.nim`, and its output expectations were updated to
reflect the new test cases.
* Added new test cases to `tests/ic/tmiscs.nim` to cover sink and move
semantics, including the definition of a `BigObj` type and a `consume`
procedure that demonstrates moving and consuming large objects.

```nim
# Sink and move semantics
type
  BigObj = object
    data: seq[int]

proc consume(x: sink BigObj) =
  echo x.data.len

var b = BigObj(data: @[1, 2, 3, 4, 5])
consume(move b)
```

gives

```
error: passing 'tySequence__qwqHTkRvwhrRyENtudHQ7g' (aka 'struct tySequence__qwqHTkRvwhrRyENtudHQ7g') to parameter of incompatible type 'tySequence__cTyVHeHOWk5jStsToosJ8Q' (aka 'struct tySequence__cTyVHeHOWk5jStsToosJ8Q')
   84 |         eqdestroy___sysma2dyk_u75((*dest_p0).data);
```

follows up https://github.com/nim-lang/Nim/pull/25614
2026-03-20 07:07:26 +01:00
ringabout
b1c68bbab4 fixes #25611; t.state != Sealed (#25622)
fixes #25611

This pull request updates the `propagateToOwner` procedure in
`compiler/ast.nim` to handle sealed types more robustly during
incremental compilation (IC) reloads. The main change is the addition of
an assertion to ensure that sealed types already have the necessary
propagated flags, preventing incorrect state during IC reloads.

Handling of sealed types and propagated flags:

* Added a check for `Sealed` state on `o2` (the owner type), and
included an assertion to verify that sealed types already have the
required propagated flags (`tfHasAsgn`/`tfHasOwned`) during IC reloads,
instead of redundantly setting them.
2026-03-19 21:25:11 +01:00
c-blake
4bf44ca47f See discussion at https://github.com/nim-lang/Nim/pull/25602 . (#25612)
It seems in dispute whether changes to code induced to avoid this new
warning firing are worthwhile.

Until either the analyzer is better or a palatable way to adjust stdlib
code not warn is found, verbosity=1 should not include the warning.

Possibly higher levels, too, but this PR is conservative and only takes
it out at the 2->1 transition.
2026-03-19 07:39:18 +01:00
ringabout
d8a1b99cac Update NimonyStableCommit to a new version (#25613)
MethodIndexEntry was moved to `semdata.nim` in
https://github.com/nim-lang/nimony/pull/1651
2026-03-19 07:36:49 +01:00
ringabout
197633dc8b fixes empty tag for nim ic (#25615)
`writeNode` writes `(empty flags type (empty))`, but it should have been
`(empty flags type)` instead

```nim
type
  Meters = distinct float
  Feet = distinct float

converter toMeters(f: Feet): Meters =
  Meters(float(f) * 0.3048)

proc showMeters(m: Meters) =
  echo float(m)

showMeters(Feet(10.0))
```
gives `[NIF decoder] expected: {ParRi} but got:
ParLe14,152,/Users/blue/.choosenim/toolchains/nim-\23devel/lib/std/private/dragonbox.nim(empty)`

---------

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
2026-03-19 07:36:09 +01:00
ringabout
a7e0065056 implements EnumToStrEntry for nim ic (#25614)
```nim
type
  NodeKind = enum
    nkInt, nkStr, nkAdd

  Node = object
    case kind: NodeKind
    of nkInt: intVal: int
    of nkStr: strVal: string
    of nkAdd: left, right: ref Node

proc newInt(v: int): ref Node =
  new(result)
  result[] = Node(kind: nkInt, intVal: v)

let n = newInt(42)
echo n.intVal
```

gives `unhandled exception: key not found: (module: 68, item: 3)
[KeyError]`
2026-03-19 07:35:24 +01:00
Zoom
b494147310 nimdoc: CSS: fix rendering of inline code spans (#25605)
I've been wondering why the inline code was rendered wrapped with no
regards to words/whitespace for a while.

Partially reverts 8b82f5 (#24927)

- `word-break: break-all;` This is seriously wrong, replaced with
`overflow-wrap: break-word;`
- `white-space: normal;` -> `pre-wrap;` to preserve whitespace in code
spans.
- Added `display: block;` and `overflow-x: auto;` to tables. This
contains wide tables with their own scrollbars without stretching the
whole doc.
- `overflow-x: hidden;` just clips content and possibly conflicts with
navbar's `sticky` attribute. Removed.
2026-03-16 18:05:11 +01:00
Andreas Rumpf
d0919b6df8 fixes #25596 (#25609) 2026-03-16 16:56:18 +01:00
Tomohiro
797b05eda6 cleans up ast2nif.nim (#25604)
In `createTypeStub` proc, `k`, `itemId` and `suffix` are used only when
`c.types.getOrDefault(name)[0]` returned nil.
So moves them under `if result == nil:` branch.

In `extractLocalSymsFromTree` proc, removes unnecessary `inc depth` and
`dec depth`.
2026-03-15 21:02:08 +01:00
Tomohiro
3a42572b19 fixes compiling newSeq call with nim ic generates compile error (#25603)
Compiling following code with `nim ic test.nim` or `nim m test.nim`
generated compile errors.
```nim
var s: seq[int]
newSeq(s, 1)
```
This PR fixes above bug.

This bug was caused by wrong PType/PSym tree generated by
`ast2nif.loadSym` proc because generic param symbols in NIF files have
all `0`.

`TSym.instantiatedFromImpl` is not related to the bug but it seems all
field of `TSym` should be copied in `transitionSymKindCommon` template.
2026-03-15 07:57:16 +01:00
metagn
1a1586a5fb properly codegen structs on deref [backport:2.2] (#25600)
Follows up #25269, refs #25265. 

I hit the same bug as #25265 for my own project but #25269 does not fix
it, I think because the type in my case is a `tyGenericInst` which does
not trigger the generation here. First I thought of skipping abstract
type kinds instead of checking for a raw `tyObject`, which fixes my
problem. But in general this could maybe also be encountered for
`tyTuple` and `tySequence` etc. So I figured it might just be safest to
not filter on specific type kinds, ~~which is done now~~ (edit: broke
CI). Maybe this has a slight cost on codegen performance though.

Edit: Allowing all types failed CI for some reason as commented below,
trying skipped type version again.
2026-03-13 17:01:21 +01:00
Zoom
2db13e05ac nimdoc: anchors fix (#25601)
This fixes autogenerated references within the same-module for types,
variables and constants for custom output file names. Previously, the
module name was baked-in, now intra-module links omit the page name in
href.

In short, fixes symbol anchors for `-o:index.html`

Expected test results updated.
2026-03-13 17:00:57 +01:00
lit
87d957fdf1 Fix #25597; parseFloat lost sign of -NaN (#25598) 2026-03-12 20:26:38 +01:00
Jake Leahy
edbb32e4c4 Fix getTypeImpl not returning defaults (#25592)
`getTypeImpl` and friends were always putting `nkEmpty` in the default
value field which meant the default values couldn't be introspected.
This copies the default AST so it can be seen in the returned object
2026-03-09 11:59:12 +01:00
Andreas Rumpf
0395af2b34 fixes #24746 (#25587) 2026-03-07 15:10:01 +01:00
Juan M Gómez
60661f6569 update nimble commit (#25537)
Co-authored-by: narimiran <narimiran@disroot.org>
2026-03-07 12:42:22 +01:00
metagn
7a87e7d199 fix compiler crash with uncheckedAssign and range/distinct discrims [backport] (#25585)
On simple code like:

```nim
type Foo = object
  case x: range[0..7]
  of 0..2:
    a: string
  else:
    b: string

var foo = Foo()
{.cast(uncheckedAssign).}:
  foo.x = 5
```

The compiler tries to generate a destructor for the variant fields by
checking if the discrim is equal to the old one, but the type is not
skipped when looking for an `==` operator in system, so any
discriminator with type `range`/`distinct`/etc crashes with:

```
(10, 9) Error: can't find magic equals operator for type kind tyRange
```

This is fixed by just skipping abstract types.
2026-03-07 07:39:57 +01:00
ringabout
e4b1d8eebc fix #25508; ignores void types in the backends (#25550)
fix #25508
2026-03-05 23:08:51 +01:00
Andreas Rumpf
5094369875 ARC/ORC: specialize seq.add (#25583) 2026-03-05 23:07:37 +01:00
Zoom
269a1c1fec nimdoc: fix char literal tokenization (#25576)
This fixes highlighter's tokenization of char literals inside
parentheses and brackets.

The Nim syntax highlighter in `docutils/highlite.nim` incorrectly
tokenizes character literals that appear after punctuation characters,
such as all kinds of brackets.

For `echo('v', "hello")`, the tokenizer treated the first `'` as
punctuation because the preceding token was punctuation `(`. As a
result, the second `'` (after `v`) was interpreted as the start of a
character literal and the literal incorrectly extended to the end of the
line.

See other examples in the screenshot:
<img width="508" height="266" alt="Screenshot 2026-03-04 at 16-09-06
_y_test"
src="https://github.com/user-attachments/assets/94d991ae-79d2-4208-a046-6ed4ddcb5c34"
/>

This regression originates from a condition added in PR #23015 that
prevented opening a `gtCharLit` token when the previous token kind was
punctuation. Nim syntax allows character literals after punctuation such
as `(`, `[`, `{`, `:`, `;`, or `,`, of course. The only case mentioned
in the manual explicitly that actually requires special handling is
stroped proc declaration for literals (see the [last paragraph
here](https://nim-lang.github.io/Nim/manual.html#lexical-analysis-character-literals)):

```nim
proc `'customLiteral`(s: string)
```

This PR narrows the conditional to not entering charlit only after
backticks.
2026-03-05 17:54:19 +01:00
Andreas Rumpf
c033ccd2e5 fixes #25552 (#25582) 2026-03-05 12:40:48 +01:00
Constantine Molchanov
2290c75f12 Nimsuggest: Operators in symbol outline (#25565)
So the problem is that Nim Language Server won't show procs like \`+\`
and \`==\` in the Document Symbols or Workspace Symbols lists. Which is
really annoying given they are regular procs just named a bit
differently.

Initially, I thought the problem was with nim-lang/langserver and opened
an issue there: https://github.com/nim-lang/langserver/issues/380

But after an investigation, it turned out the issue is fixed on the
nimsuggest side.

Strangely enough, calling `outline foo.nim:0:0` in nimsuggest manually
does show \`+\` as well as regular procs (e.g. `foo`) but when
nimsuggest is invoked from lsp only `foo` would be there.

Anyway, with this fix all procs appear on the symbol lists.
2026-03-05 09:58:17 +01:00
dependabot[bot]
5ea198faf3 Bump crazy-max/ghaction-github-pages from 4 to 5 (#25578)
Bumps
[crazy-max/ghaction-github-pages](https://github.com/crazy-max/ghaction-github-pages)
from 4 to 5.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/crazy-max/ghaction-github-pages/releases">crazy-max/ghaction-github-pages's
releases</a>.</em></p>
<blockquote>
<h2>v5.0.0</h2>
<ul>
<li>Node 24 as default runtime (requires <a
href="https://github.com/actions/runner/releases/tag/v2.327.1">Actions
Runner v2.327.1</a> or later) by <a
href="https://github.com/crazy-max"><code>@​crazy-max</code></a> in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/251">crazy-max/ghaction-github-pages#251</a></li>
<li>Switch to ESM and update config wiring by <a
href="https://github.com/crazy-max"><code>@​crazy-max</code></a> in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/250">crazy-max/ghaction-github-pages#250</a></li>
<li>Bump <code>@​actions/core</code> from 1.11.1 to 3.0.0 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/252">crazy-max/ghaction-github-pages#252</a></li>
<li>Bump <code>@​actions/exec</code> from 1.1.1 to 3.0.0 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/246">crazy-max/ghaction-github-pages#246</a></li>
<li>Bump brace-expansion from 1.1.11 to 1.1.12 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/237">crazy-max/ghaction-github-pages#237</a></li>
<li>Bump fs-extra from 11.3.0 to 11.3.3 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/247">crazy-max/ghaction-github-pages#247</a></li>
<li>Bump js-yaml from 4.1.0 to 4.1.1 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/242">crazy-max/ghaction-github-pages#242</a></li>
<li>Bump minimatch from 3.1.2 to 3.1.5 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/249">crazy-max/ghaction-github-pages#249</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/crazy-max/ghaction-github-pages/compare/v4.2.0...v5.0.0">https://github.com/crazy-max/ghaction-github-pages/compare/v4.2.0...v5.0.0</a></p>
<h2>v4.2.0</h2>
<ul>
<li>Bump cross-spawn from 7.0.3 to 7.0.6 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/231">crazy-max/ghaction-github-pages#231</a></li>
<li>Bump fs-extra from 11.2.0 to 11.3.0 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/233">crazy-max/ghaction-github-pages#233</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/crazy-max/ghaction-github-pages/compare/v4.1.0...v4.2.0">https://github.com/crazy-max/ghaction-github-pages/compare/v4.1.0...v4.2.0</a></p>
<h2>v4.1.0</h2>
<ul>
<li>Bump <code>@​actions/core</code> from 1.10.0 to 1.11.1 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/229">crazy-max/ghaction-github-pages#229</a></li>
<li>Bump braces from 3.0.2 to 3.0.3 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/225">crazy-max/ghaction-github-pages#225</a></li>
<li>Bump fs-extra from 11.1.1 to 11.2.0 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/221">crazy-max/ghaction-github-pages#221</a></li>
<li>Bump ip from 2.0.0 to 2.0.1 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/220">crazy-max/ghaction-github-pages#220</a></li>
<li>Bump micromatch from 4.0.5 to 4.0.8 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/227">crazy-max/ghaction-github-pages#227</a></li>
<li>Bump tar from 6.1.14 to 6.2.1 in <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/pull/222">crazy-max/ghaction-github-pages#222</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/crazy-max/ghaction-github-pages/compare/v4.0.0...v4.1.0">https://github.com/crazy-max/ghaction-github-pages/compare/v4.0.0...v4.1.0</a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="1d6ee9b181"><code>1d6ee9b</code></a>
Merge pull request <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/issues/252">#252</a>
from crazy-max/dependabot/npm_and_yarn/actions/core-3...</li>
<li><a
href="26956ffd3c"><code>26956ff</code></a>
chore: update generated content</li>
<li><a
href="2627782b16"><code>2627782</code></a>
build(deps): bump <code>@​actions/core</code> from 1.11.1 to 3.0.0</li>
<li><a
href="f8e352d0cb"><code>f8e352d</code></a>
Merge pull request <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/issues/251">#251</a>
from crazy-max/node24</li>
<li><a
href="709db507ec"><code>709db50</code></a>
node 24 as default runtime</li>
<li><a
href="7e3c57de89"><code>7e3c57d</code></a>
Merge pull request <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/issues/247">#247</a>
from crazy-max/dependabot/npm_and_yarn/fs-extra-11.3.3</li>
<li><a
href="ba0b3631bd"><code>ba0b363</code></a>
chore: update generated content</li>
<li><a
href="b14d0fb11d"><code>b14d0fb</code></a>
build(deps): bump fs-extra from 11.3.0 to 11.3.3</li>
<li><a
href="01d0e18e28"><code>01d0e18</code></a>
Merge pull request <a
href="https://redirect.github.com/crazy-max/ghaction-github-pages/issues/237">#237</a>
from crazy-max/dependabot/npm_and_yarn/brace-expansio...</li>
<li><a
href="3e82042fff"><code>3e82042</code></a>
build(deps): bump brace-expansion from 1.1.11 to 1.1.12</li>
<li>Additional commits viewable in <a
href="https://github.com/crazy-max/ghaction-github-pages/compare/v4...v5">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=crazy-max/ghaction-github-pages&package-manager=github_actions&previous-version=4&new-version=5)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-03-05 16:12:30 +08:00
ringabout
8e2547a5e2 fixes #25566; {.align.} pragma where each 16-byte-aligned (#25570)
fixes #25566
2026-03-04 09:14:13 +01:00
Ryan McConnell
46cddbccd6 fixes #25572 ICE evaluating closure iter with object conversion (#25575) 2026-03-04 05:45:16 +01:00
vercingetorx
9ed4077d9a Fix memory leak in asyncdispatch.withTimeout by clearing losing callbacks (#25567)
withTimeout currently leaves the “losing” callback installed:

  - when fut finishes first, timeout callback remains until timer fires,
- when timeout fires first, fut callback remains on the wrapped future.

Under high-throughput use with large future payloads, this retains
closures/future references longer than needed and causes large transient
RSS growth.
This patch clears the opposite callback immediately once outcome is
decided, reducing retention without changing API behavior.
2026-03-01 22:11:18 +01:00
Kevin Hovsäter
e69d672354 Fix warning admonition in std/streams (#25564)
The rest of the body must be indented in order to fall under the warning
admonition. Right now, only the first part of the warning is inside the
admonition, see [std/streams](https://nim-lang.org/docs/streams.html).
2026-03-01 11:36:31 +08:00
ringabout
bd709f9b4c fixes #25262; proc v[T: typedesc]() = discard / v[0]() compiles even though 0 isn't a typedesc (#25558)
fixes #25262

```nim
if constraint != nil and constraint.kind == tyTypeDesc:
  n[i].typ = e.typ
else:
  n[i].typ = e.typ.skipTypes({tyTypeDesc})
```
at least when `constraint` is a typedesc, it should not skip
`tyTypeDesc`


```nim
if arg.kind != tyTypeDesc:
  arg = makeTypeDesc(m.c, arg)
```
Wrappers literals into typedesc, which can cause problems. Though, it
doesn't seem to be necessary
2026-02-28 23:01:09 +01:00
ringabout
4566ffaca9 fixes #25553; Invalid codegen for accessing tuple in array (#25555)
fixes #25553
2026-02-28 22:51:38 +01:00
Kevin Hovsäter
a2db2af5b6 Fix a few typos (#25563)
While fixing a few things in the tutorial, I found a few other typos
lingering in the `doc/` directory.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2026-02-28 22:50:37 +01:00
Kevin Hovsäter
c36617c490 Fix std/pegs sequence example (#25562)
This corrects the example used to describe `std/pegs` sequence notion.
It incorrectly used `Z` whereas `C` was expected.
2026-02-28 17:26:44 +08:00
Raka Hourianto
9b2b286baf nre: fix replacement string parser OOB access, numeric refs, and unterminated named refs (#25560)
1. A trailing `$` at the end of a replacement string could read out of
bounds via `how[i + 1]`; this now raises `ValueError` instead.

2. Numeric capture parsing used `id += (id * 10) + digit` instead of `id
= (id * 10) + digit`, so multi-digit refs were parsed incorrectly (e.g.
`$12` resolved as capture 13 instead of 12).

4. Unterminated named replacement syntax (e.g. `${foo)` is now rejected
with ValueError instead of being accepted and parsed inconsistently.

Found and fixed by GPT 5.3 Codex.
2026-02-28 07:39:16 +01:00
Christian Zietz
49961a54dd Atomics can't cause exceptions with Microsoft Visual C++ (#25559)
The `enforcenoraises` pragma prevents generation of exception checking
code for atomic... functions when compiling with Microsoft Visual C++ as
backend.

Fixes #25445

Without this change, the following test program:
```nim
import std/sysatomics

var x: ptr uint64 = cast[ptr uint64](uint64(0))
var y: ptr uint64 = cast[ptr uint64](uint64(42))
let z = atomicExchangeN(addr x, y, ATOMIC_ACQ_REL)

let a = atomicCompareExchangeN(addr x, addr y, y, true, ATOMIC_ACQ_REL, ATOMIC_ACQ_REL)

var v = 42
atomicStoreN(addr v, 43, ATOMIC_ACQ_REL)
let w = atomicLoadN(addr v, ATOMIC_ACQ_REL)
```
... generates this C code when compiling with `--cc:vcc`:
```c
N_LIB_PRIVATE N_NIMCALL(void, NimMainModule)(void) {
        {
        NU64* T1_;
NIM_BOOL T2_;
NI T3_;
NIM_BOOL* nimErr_;
        nimfr_("testexcept", "/tmp/testexcept.nim");
nimErr_ = nimErrorFlag();
        nimlf_(7, "/tmp/testexcept.nim");T1_ = ((NU64*) 0);
T1_ = atomicExchangeN__testexcept_u4((&x__testexcept_u2), y__testexcept_u3, ((int) 4));
if (NIM_UNLIKELY((*nimErr_))) {
        goto BeforeRet_;
}
z__testexcept_u32 = T1_;
        nimln_(9);T2_ = ((NIM_BOOL) 0);
T2_ = atomicCompareExchangeN__testexcept_u33((&x__testexcept_u2), (&y__testexcept_u3), y__testexcept_u3, NIM_TRUE, ((int) 4), ((int) 4));
if (NIM_UNLIKELY((*nimErr_))) {
        goto BeforeRet_;
}
a__testexcept_u45 = T2_;
        nimln_(12);atomicStoreN__testexcept_u47(((&v__testexcept_u46)), ((NI) 43));
if (NIM_UNLIKELY((*nimErr_))) {
        goto BeforeRet_;
}
        nimln_(13);T3_ = ((NI) 0);
T3_ = atomicLoadN__testexcept_u53(((&v__testexcept_u46)));
if (NIM_UNLIKELY((*nimErr_))) {
        goto BeforeRet_;
}
w__testexcept_u59 = T3_;
BeforeRet_: ;
        nimTestErrorFlag();
                popFrame();
}
}
```

Note the repeated checks for `*nimErr_`.

With this PR applied, the checks vanish:
```c
N_LIB_PRIVATE N_NIMCALL(void, NimMainModule)(void) {
        {
                nimfr_("testexcept", "/tmp/testexcept.nim");
        nimlf_(7, "/tmp/testexcept.nim");z__testexcept_u32 = atomicExchangeN__testexcept_u4((&x__testexcept_u2), y__testexcept_u3, ((int) 4));
        nimln_(9);a__testexcept_u45 = atomicCompareExchangeN__testexcept_u33((&x__testexcept_u2), (&y__testexcept_u3), y__testexcept_u3, NIM_TRUE, ((int) 4), ((int) 4));
        nimln_(12);atomicStoreN__testexcept_u47(((&v__testexcept_u46)), ((NI) 43));
        nimln_(13);w__testexcept_u59 = atomicLoadN__testexcept_u53(((&v__testexcept_u46)));
nimTestErrorFlag();
                popFrame();
}
}
```

For reference, with gcc as backend the generated code looks as follows:
```c
N_LIB_PRIVATE N_NIMCALL(void, NimMainModule)(void) {
        {
                nimfr_("testexcept", "/tmp/testexcept.nim");
        nimlf_(7, "/tmp/testexcept.nim");z__testexcept_u9 = __atomic_exchange_n((&x__testexcept_u2), y__testexcept_u3, __ATOMIC_ACQ_REL);
        nimln_(9);a__testexcept_u18 = __atomic_compare_exchange_n((&x__testexcept_u2), (&y__testexcept_u3), y__testexcept_u3, NIM_TRUE, __ATOMIC_ACQ_REL, __ATOMIC_ACQ_REL);
        nimln_(12);__atomic_store_n(((&v__testexcept_u19)), ((NI) 43), __ATOMIC_ACQ_REL);
        nimln_(13);w__testexcept_u29 = __atomic_load_n(((&v__testexcept_u19)), __ATOMIC_ACQ_REL);
nimTestErrorFlag();
                popFrame();
}
}
```

With this PR the program from #25445 yields the correct output `Error:
unhandled exception: index 4 not in 0 .. 3 [IndexDefect]` instead of
crashing with a SIGSEGV.

PS: Unfortunately, I did not find out how to run the tests with MSVC.
`./koch tests --cc:vcc` doesn't use MSVC.
2026-02-27 19:16:30 +01:00
Kevin Hovsäter
358d9b4497 Fix casing of types in example (#25556)
From the Standard Library Style Guide:

> Type identifiers should be in PascalCase. All other identifiers should
> be in camelCase with the exception of constants which may use
> PascalCase but are not required to.
2026-02-27 09:24:23 +08:00
ringabout
a3157537e1 allows implicitRangeConvs for literals (#25542)
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-02-25 19:10:00 +01:00
ringabout
74499e4561 fixes #21281; proc f(x: static[auto]) doesn't treat x as static (#25543)
fixes #21281
2026-02-25 19:09:24 +01:00
dependabot[bot]
d0ff0ebb43 Bump actions/setup-node from 4 to 6 (#25545)
Bumps [actions/setup-node](https://github.com/actions/setup-node) from 4
to 6.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/actions/setup-node/releases">actions/setup-node's
releases</a>.</em></p>
<blockquote>
<h2>v6.0.0</h2>
<h2>What's Changed</h2>
<p><strong>Breaking Changes</strong></p>
<ul>
<li>Limit automatic caching to npm, update workflows and documentation
by <a
href="https://github.com/priyagupta108"><code>@​priyagupta108</code></a>
in <a
href="https://redirect.github.com/actions/setup-node/pull/1374">actions/setup-node#1374</a></li>
</ul>
<p><strong>Dependency Upgrades</strong></p>
<ul>
<li>Upgrade ts-jest from 29.1.2 to 29.4.1 and document breaking changes
in v5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1336">#1336</a></li>
<li>Upgrade prettier from 2.8.8 to 3.6.2 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1334">#1334</a></li>
<li>Upgrade actions/publish-action from 0.3.0 to 0.4.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1362">#1362</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/setup-node/compare/v5...v6.0.0">https://github.com/actions/setup-node/compare/v5...v6.0.0</a></p>
<h2>v5.0.0</h2>
<h2>What's Changed</h2>
<h3>Breaking Changes</h3>
<ul>
<li>Enhance caching in setup-node with automatic package manager
detection by <a
href="https://github.com/priya-kinthali"><code>@​priya-kinthali</code></a>
in <a
href="https://redirect.github.com/actions/setup-node/pull/1348">actions/setup-node#1348</a></li>
</ul>
<p>This update, introduces automatic caching when a valid
<code>packageManager</code> field is present in your
<code>package.json</code>. This aims to improve workflow performance and
make dependency management more seamless.
To disable this automatic caching, set <code>package-manager-cache:
false</code></p>
<pre lang="yaml"><code>steps:
- uses: actions/checkout@v5
- uses: actions/setup-node@v5
  with:
    package-manager-cache: false
</code></pre>
<ul>
<li>Upgrade action to use node24 by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/setup-node/pull/1325">actions/setup-node#1325</a></li>
</ul>
<p>Make sure your runner is on version v2.327.1 or later to ensure
compatibility with this release. <a
href="https://github.com/actions/runner/releases/tag/v2.327.1">See
Release Notes</a></p>
<h3>Dependency Upgrades</h3>
<ul>
<li>Upgrade <code>@​octokit/request-error</code> and
<code>@​actions/github</code> by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1227">actions/setup-node#1227</a></li>
<li>Upgrade uuid from 9.0.1 to 11.1.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1273">actions/setup-node#1273</a></li>
<li>Upgrade undici from 5.28.5 to 5.29.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1295">actions/setup-node#1295</a></li>
<li>Upgrade form-data to bring in fix for critical vulnerability by <a
href="https://github.com/gowridurgad"><code>@​gowridurgad</code></a> in
<a
href="https://redirect.github.com/actions/setup-node/pull/1332">actions/setup-node#1332</a></li>
<li>Upgrade actions/checkout from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/setup-node/pull/1345">actions/setup-node#1345</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a
href="https://github.com/priya-kinthali"><code>@​priya-kinthali</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/setup-node/pull/1348">actions/setup-node#1348</a></li>
<li><a href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/setup-node/pull/1325">actions/setup-node#1325</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/setup-node/compare/v4...v5.0.0">https://github.com/actions/setup-node/compare/v4...v5.0.0</a></p>
<h2>v4.4.0</h2>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="6044e13b5d"><code>6044e13</code></a>
Docs: bump actions/checkout from v5 to v6 (<a
href="https://redirect.github.com/actions/setup-node/issues/1468">#1468</a>)</li>
<li><a
href="8e494633d0"><code>8e49463</code></a>
Fix README typo (<a
href="https://redirect.github.com/actions/setup-node/issues/1226">#1226</a>)</li>
<li><a
href="621ac41091"><code>621ac41</code></a>
README.md: bump to latest released checkout version v6 (<a
href="https://redirect.github.com/actions/setup-node/issues/1446">#1446</a>)</li>
<li><a
href="2951748f4c"><code>2951748</code></a>
Bump <code>@​actions/cache</code> to v5.0.1 (<a
href="https://redirect.github.com/actions/setup-node/issues/1449">#1449</a>)</li>
<li><a
href="21ddc7bc1f"><code>21ddc7b</code></a>
Correct mirror option typos (<a
href="https://redirect.github.com/actions/setup-node/issues/1442">#1442</a>)</li>
<li><a
href="65d868f8d4"><code>65d868f</code></a>
Update Documentation for Lockfile (<a
href="https://redirect.github.com/actions/setup-node/issues/1454">#1454</a>)</li>
<li><a
href="395ad32622"><code>395ad32</code></a>
Bump js-yaml from 3.14.1 to 3.14.2 (<a
href="https://redirect.github.com/actions/setup-node/issues/1435">#1435</a>)</li>
<li><a
href="a4d2e2bbca"><code>a4d2e2b</code></a>
Bump actions/checkout from 5 to 6 (<a
href="https://redirect.github.com/actions/setup-node/issues/1439">#1439</a>)</li>
<li><a
href="b9b25d45f7"><code>b9b25d4</code></a>
Remove always-auth configuration handling from action (<a
href="https://redirect.github.com/actions/setup-node/issues/1436">#1436</a>)</li>
<li><a
href="633bb92bc0"><code>633bb92</code></a>
Bump <code>@​actions/cache</code> from 4.0.3 to 4.1.0 (<a
href="https://redirect.github.com/actions/setup-node/issues/1384">#1384</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/actions/setup-node/compare/v4...v6">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/setup-node&package-manager=github_actions&previous-version=4&new-version=6)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

---------

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2026-02-25 20:45:44 +08:00
dependabot[bot]
c292981fd3 Bump actions/checkout from 4 to 6 (#25546)
Bumps [actions/checkout](https://github.com/actions/checkout) from 4 to
6.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/actions/checkout/releases">actions/checkout's
releases</a>.</em></p>
<blockquote>
<h2>v6.0.0</h2>
<h2>What's Changed</h2>
<ul>
<li>Update README to include Node.js 24 support details and requirements
by <a href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a>
in <a
href="https://redirect.github.com/actions/checkout/pull/2248">actions/checkout#2248</a></li>
<li>Persist creds to a separate file by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2286">actions/checkout#2286</a></li>
<li>v6-beta by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2298">actions/checkout#2298</a></li>
<li>update readme/changelog for v6 by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2311">actions/checkout#2311</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/checkout/compare/v5.0.0...v6.0.0">https://github.com/actions/checkout/compare/v5.0.0...v6.0.0</a></p>
<h2>v6-beta</h2>
<h2>What's Changed</h2>
<p>Updated persist-credentials to store the credentials under
<code>$RUNNER_TEMP</code> instead of directly in the local git
config.</p>
<p>This requires a minimum Actions Runner version of <a
href="https://github.com/actions/runner/releases/tag/v2.329.0">v2.329.0</a>
to access the persisted credentials for <a
href="https://docs.github.com/en/actions/tutorials/use-containerized-services/create-a-docker-container-action">Docker
container action</a> scenarios.</p>
<h2>v5.0.1</h2>
<h2>What's Changed</h2>
<ul>
<li>Port v6 cleanup to v5 by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2301">actions/checkout#2301</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/checkout/compare/v5...v5.0.1">https://github.com/actions/checkout/compare/v5...v5.0.1</a></p>
<h2>v5.0.0</h2>
<h2>What's Changed</h2>
<ul>
<li>Update actions checkout to use node 24 by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2226">actions/checkout#2226</a></li>
<li>Prepare v5.0.0 release by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2238">actions/checkout#2238</a></li>
</ul>
<h2>⚠️ Minimum Compatible Runner Version</h2>
<p><strong>v2.327.1</strong><br />
<a
href="https://github.com/actions/runner/releases/tag/v2.327.1">Release
Notes</a></p>
<p>Make sure your runner is updated to this version or newer to use this
release.</p>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/checkout/compare/v4...v5.0.0">https://github.com/actions/checkout/compare/v4...v5.0.0</a></p>
<h2>v4.3.1</h2>
<h2>What's Changed</h2>
<ul>
<li>Port v6 cleanup to v4 by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2305">actions/checkout#2305</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/checkout/compare/v4...v4.3.1">https://github.com/actions/checkout/compare/v4...v4.3.1</a></p>
<h2>v4.3.0</h2>
<h2>What's Changed</h2>
<ul>
<li>docs: update README.md by <a
href="https://github.com/motss"><code>@​motss</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1971">actions/checkout#1971</a></li>
<li>Add internal repos for checking out multiple repositories by <a
href="https://github.com/mouismail"><code>@​mouismail</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1977">actions/checkout#1977</a></li>
<li>Documentation update - add recommended permissions to Readme by <a
href="https://github.com/benwells"><code>@​benwells</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2043">actions/checkout#2043</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a
href="https://github.com/actions/checkout/blob/main/CHANGELOG.md">actions/checkout's
changelog</a>.</em></p>
<blockquote>
<h1>Changelog</h1>
<h2>v6.0.2</h2>
<ul>
<li>Fix tag handling: preserve annotations and explicit fetch-tags by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2356">actions/checkout#2356</a></li>
</ul>
<h2>v6.0.1</h2>
<ul>
<li>Add worktree support for persist-credentials includeIf by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2327">actions/checkout#2327</a></li>
</ul>
<h2>v6.0.0</h2>
<ul>
<li>Persist creds to a separate file by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2286">actions/checkout#2286</a></li>
<li>Update README to include Node.js 24 support details and requirements
by <a href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a>
in <a
href="https://redirect.github.com/actions/checkout/pull/2248">actions/checkout#2248</a></li>
</ul>
<h2>v5.0.1</h2>
<ul>
<li>Port v6 cleanup to v5 by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2301">actions/checkout#2301</a></li>
</ul>
<h2>v5.0.0</h2>
<ul>
<li>Update actions checkout to use node 24 by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2226">actions/checkout#2226</a></li>
</ul>
<h2>v4.3.1</h2>
<ul>
<li>Port v6 cleanup to v4 by <a
href="https://github.com/ericsciple"><code>@​ericsciple</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2305">actions/checkout#2305</a></li>
</ul>
<h2>v4.3.0</h2>
<ul>
<li>docs: update README.md by <a
href="https://github.com/motss"><code>@​motss</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1971">actions/checkout#1971</a></li>
<li>Add internal repos for checking out multiple repositories by <a
href="https://github.com/mouismail"><code>@​mouismail</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1977">actions/checkout#1977</a></li>
<li>Documentation update - add recommended permissions to Readme by <a
href="https://github.com/benwells"><code>@​benwells</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2043">actions/checkout#2043</a></li>
<li>Adjust positioning of user email note and permissions heading by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2044">actions/checkout#2044</a></li>
<li>Update README.md by <a
href="https://github.com/nebuk89"><code>@​nebuk89</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2194">actions/checkout#2194</a></li>
<li>Update CODEOWNERS for actions by <a
href="https://github.com/TingluoHuang"><code>@​TingluoHuang</code></a>
in <a
href="https://redirect.github.com/actions/checkout/pull/2224">actions/checkout#2224</a></li>
<li>Update package dependencies by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/2236">actions/checkout#2236</a></li>
</ul>
<h2>v4.2.2</h2>
<ul>
<li><code>url-helper.ts</code> now leverages well-known environment
variables by <a href="https://github.com/jww3"><code>@​jww3</code></a>
in <a
href="https://redirect.github.com/actions/checkout/pull/1941">actions/checkout#1941</a></li>
<li>Expand unit test coverage for <code>isGhes</code> by <a
href="https://github.com/jww3"><code>@​jww3</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1946">actions/checkout#1946</a></li>
</ul>
<h2>v4.2.1</h2>
<ul>
<li>Check out other refs/* by commit if provided, fall back to ref by <a
href="https://github.com/orhantoy"><code>@​orhantoy</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1924">actions/checkout#1924</a></li>
</ul>
<h2>v4.2.0</h2>
<ul>
<li>Add Ref and Commit outputs by <a
href="https://github.com/lucacome"><code>@​lucacome</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1180">actions/checkout#1180</a></li>
<li>Dependency updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>- <a
href="https://redirect.github.com/actions/checkout/pull/1777">actions/checkout#1777</a>,
<a
href="https://redirect.github.com/actions/checkout/pull/1872">actions/checkout#1872</a></li>
</ul>
<h2>v4.1.7</h2>
<ul>
<li>Bump the minor-npm-dependencies group across 1 directory with 4
updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1739">actions/checkout#1739</a></li>
<li>Bump actions/checkout from 3 to 4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1697">actions/checkout#1697</a></li>
<li>Check out other refs/* by commit by <a
href="https://github.com/orhantoy"><code>@​orhantoy</code></a> in <a
href="https://redirect.github.com/actions/checkout/pull/1774">actions/checkout#1774</a></li>
<li>Pin actions/checkout's own workflows to a known, good, stable
version. by <a href="https://github.com/jww3"><code>@​jww3</code></a> in
<a
href="https://redirect.github.com/actions/checkout/pull/1776">actions/checkout#1776</a></li>
</ul>
<h2>v4.1.6</h2>
<ul>
<li>Check platform to set archive extension appropriately by <a
href="https://github.com/cory-miller"><code>@​cory-miller</code></a> in
<a
href="https://redirect.github.com/actions/checkout/pull/1732">actions/checkout#1732</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="de0fac2e45"><code>de0fac2</code></a>
Fix tag handling: preserve annotations and explicit fetch-tags (<a
href="https://redirect.github.com/actions/checkout/issues/2356">#2356</a>)</li>
<li><a
href="064fe7f331"><code>064fe7f</code></a>
Add orchestration_id to git user-agent when ACTIONS_ORCHESTRATION_ID is
set (...</li>
<li><a
href="8e8c483db8"><code>8e8c483</code></a>
Clarify v6 README (<a
href="https://redirect.github.com/actions/checkout/issues/2328">#2328</a>)</li>
<li><a
href="033fa0dc0b"><code>033fa0d</code></a>
Add worktree support for persist-credentials includeIf (<a
href="https://redirect.github.com/actions/checkout/issues/2327">#2327</a>)</li>
<li><a
href="c2d88d3ecc"><code>c2d88d3</code></a>
Update all references from v5 and v4 to v6 (<a
href="https://redirect.github.com/actions/checkout/issues/2314">#2314</a>)</li>
<li><a
href="1af3b93b68"><code>1af3b93</code></a>
update readme/changelog for v6 (<a
href="https://redirect.github.com/actions/checkout/issues/2311">#2311</a>)</li>
<li><a
href="71cf2267d8"><code>71cf226</code></a>
v6-beta (<a
href="https://redirect.github.com/actions/checkout/issues/2298">#2298</a>)</li>
<li><a
href="069c695914"><code>069c695</code></a>
Persist creds to a separate file (<a
href="https://redirect.github.com/actions/checkout/issues/2286">#2286</a>)</li>
<li><a
href="ff7abcd0c3"><code>ff7abcd</code></a>
Update README to include Node.js 24 support details and requirements (<a
href="https://redirect.github.com/actions/checkout/issues/2248">#2248</a>)</li>
<li><a
href="08c6903cd8"><code>08c6903</code></a>
Prepare v5.0.0 release (<a
href="https://redirect.github.com/actions/checkout/issues/2238">#2238</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/actions/checkout/compare/v4...v6">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/checkout&package-manager=github_actions&previous-version=4&new-version=6)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-02-25 16:53:39 +08:00
dependabot[bot]
29705aab1a Bump actions/stale from 9 to 10 (#25548)
Bumps [actions/stale](https://github.com/actions/stale) from 9 to 10.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/actions/stale/releases">actions/stale's
releases</a>.</em></p>
<blockquote>
<h2>v10.0.0</h2>
<h2>What's Changed</h2>
<h3>Breaking Changes</h3>
<ul>
<li>Upgrade to node 24 by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1279">actions/stale#1279</a>
Make sure your runner is on version v2.327.1 or later to ensure
compatibility with this release. <a
href="https://github.com/actions/runner/releases/tag/v2.327.1">Release
Notes</a></li>
</ul>
<h3>Enhancement</h3>
<ul>
<li>Introducing sort-by option by <a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1254">actions/stale#1254</a></li>
</ul>
<h3>Dependency Upgrades</h3>
<ul>
<li>Upgrade actions/publish-immutable-action from 0.0.3 to 0.0.4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/stale/pull/1186">actions/stale#1186</a></li>
<li>Upgrade undici from 5.28.4 to 5.28.5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/stale/pull/1201">actions/stale#1201</a></li>
<li>Upgrade <code>@​action/cache</code> from 4.0.0 to 4.0.2 by <a
href="https://github.com/aparnajyothi-y"><code>@​aparnajyothi-y</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1226">actions/stale#1226</a></li>
<li>Upgrade <code>@​action/cache</code> from 4.0.2 to 4.0.3 by <a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1233">actions/stale#1233</a></li>
<li>Upgrade undici from 5.28.5 to 5.29.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/stale/pull/1251">actions/stale#1251</a></li>
<li>Upgrade form-data to bring in fix for critical vulnerability by <a
href="https://github.com/gowridurgad"><code>@​gowridurgad</code></a> in
<a
href="https://redirect.github.com/actions/stale/pull/1277">actions/stale#1277</a></li>
</ul>
<h3>Documentation changes</h3>
<ul>
<li>Changelog update for recent releases by <a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1224">actions/stale#1224</a></li>
<li>Permissions update in Readme by <a
href="https://github.com/ghadimir"><code>@​ghadimir</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1248">actions/stale#1248</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/stale/pull/1224">actions/stale#1224</a></li>
<li><a href="https://github.com/GhadimiR"><code>@​GhadimiR</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/stale/pull/1248">actions/stale#1248</a></li>
<li><a
href="https://github.com/gowridurgad"><code>@​gowridurgad</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/stale/pull/1277">actions/stale#1277</a></li>
<li><a href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/stale/pull/1279">actions/stale#1279</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/stale/compare/v9...v10.0.0">https://github.com/actions/stale/compare/v9...v10.0.0</a></p>
<h2>v9.1.0</h2>
<h2>What's Changed</h2>
<ul>
<li>Documentation update by <a
href="https://github.com/Marukome0743"><code>@​Marukome0743</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1116">actions/stale#1116</a></li>
<li>Add workflow file for publishing releases to immutable action
package by <a
href="https://github.com/Jcambass"><code>@​Jcambass</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1179">actions/stale#1179</a></li>
<li>Update undici from 5.28.2 to 5.28.4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1150">actions/stale#1150</a></li>
<li>Update actions/checkout from 3 to 4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1091">actions/stale#1091</a></li>
<li>Update actions/publish-action from 0.2.2 to 0.3.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1147">actions/stale#1147</a></li>
<li>Update ts-jest from 29.1.1 to 29.2.5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1175">actions/stale#1175</a></li>
<li>Update <code>@​actions/core</code> from 1.10.1 to 1.11.1 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1191">actions/stale#1191</a></li>
<li>Update <code>@​types/jest</code> from 29.5.11 to 29.5.14 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1193">actions/stale#1193</a></li>
<li>Update <code>@​actions/cache</code> from 3.2.2 to 4.0.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1194">actions/stale#1194</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a
href="https://github.com/Marukome0743"><code>@​Marukome0743</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/stale/pull/1116">actions/stale#1116</a></li>
<li><a href="https://github.com/Jcambass"><code>@​Jcambass</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/stale/pull/1179">actions/stale#1179</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/stale/compare/v9...v9.1.0">https://github.com/actions/stale/compare/v9...v9.1.0</a></p>
</blockquote>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a
href="https://github.com/actions/stale/blob/main/CHANGELOG.md">actions/stale's
changelog</a>.</em></p>
<blockquote>
<h1>Changelog</h1>
<h1>[10.1.0]</h1>
<h2>What's Changed</h2>
<ul>
<li>Add only-issue-types option to filter issues by type by <a
href="https://github.com/Bibo-Joshi"><code>@​Bibo-Joshi</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1255">actions/stale#1255</a></li>
</ul>
<h1>[10.0.0]</h1>
<h2>What's Changed</h2>
<h2>Breaking Changes</h2>
<ul>
<li>Upgrade to node 24 by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1279">actions/stale#1279</a>
Make sure your runner is on version v2.327.1 or later to ensure
compatibility with this release. <a
href="https://github.com/actions/runner/releases/tag/v2.327.1">Release
Notes</a></li>
</ul>
<h2>Enhancement</h2>
<ul>
<li>Introducing sort-by option by <a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1254">actions/stale#1254</a></li>
</ul>
<h2>Dependency Upgrades</h2>
<ul>
<li>Upgrade actions/publish-immutable-action from 0.0.3 to 0.0.4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/stale/pull/1186">actions/stale#1186</a></li>
<li>Upgrade undici from 5.28.4 to 5.28.5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/stale/pull/1201">actions/stale#1201</a></li>
<li>Upgrade <code>@​action/cache</code> from 4.0.0 to 4.0.2 by <a
href="https://github.com/aparnajyothi-y"><code>@​aparnajyothi-y</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1226">actions/stale#1226</a></li>
<li>Upgrade <code>@​action/cache</code> from 4.0.2 to 4.0.3 by <a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1233">actions/stale#1233</a></li>
<li>Upgrade undici from 5.28.5 to 5.29.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/stale/pull/1251">actions/stale#1251</a></li>
<li>Upgrade form-data to bring in fix for critical vulnerability by <a
href="https://github.com/gowridurgad"><code>@​gowridurgad</code></a> in
<a
href="https://redirect.github.com/actions/stale/pull/1277">actions/stale#1277</a></li>
</ul>
<h2>Documentation changes</h2>
<ul>
<li>Changelog update for recent releases by <a
href="https://github.com/suyashgaonkar"><code>@​suyashgaonkar</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1224">actions/stale#1224</a></li>
<li>Permissions update in Readme by <a
href="https://github.com/ghadimir"><code>@​ghadimir</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1248">actions/stale#1248</a></li>
</ul>
<h1>[9.1.0]</h1>
<h2>What's Changed</h2>
<ul>
<li>Documentation update by <a
href="https://github.com/Marukome0743"><code>@​Marukome0743</code></a>
in <a
href="https://redirect.github.com/actions/stale/pull/1116">actions/stale#1116</a></li>
<li>Add workflow file for publishing releases to immutable action
package by <a
href="https://github.com/Jcambass"><code>@​Jcambass</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1179">actions/stale#1179</a></li>
<li>Update undici from 5.28.2 to 5.28.4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1150">actions/stale#1150</a></li>
<li>Update actions/checkout from 3 to 4 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1091">actions/stale#1091</a></li>
<li>Update actions/publish-action from 0.2.2 to 0.3.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1147">actions/stale#1147</a></li>
<li>Update ts-jest from 29.1.1 to 29.2.5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1175">actions/stale#1175</a></li>
<li>Update <code>@​actions/core</code> from 1.10.1 to 1.11.1 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1191">actions/stale#1191</a></li>
<li>Update <code>@​types/jest</code> from 29.5.11 to 29.5.14 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1193">actions/stale#1193</a></li>
<li>Update <code>@​actions/cache</code> from 3.2.2 to 4.0.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a> in <a
href="https://redirect.github.com/actions/stale/pull/1194">actions/stale#1194</a></li>
</ul>
<h1>[9.0.0]</h1>
<h2>Breaking Changes</h2>
<ol>
<li>Action is now stateful: If the action ends because of <a
href="https://github.com/actions/stale#operations-per-run">operations-per-run</a>
then the next run will start from the first unprocessed issue skipping
the issues processed during the previous run(s). The state is reset when
all the issues are processed. This should be considered for scheduling
workflow runs.</li>
<li>Version 9 of this action updated the runtime to Node.js 20. All
scripts are now run with Node.js 20 instead of Node.js 16 and are
affected by any breaking changes between Node.js 16 and 20.</li>
</ol>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="b5d41d4e1d"><code>b5d41d4</code></a>
build(deps-dev): bump lodash from 4.17.21 to 4.17.23 (<a
href="https://redirect.github.com/actions/stale/issues/1313">#1313</a>)</li>
<li><a
href="dcd2b9469d"><code>dcd2b94</code></a>
Fix punycode and url.parse Deprecation Warnings (<a
href="https://redirect.github.com/actions/stale/issues/1312">#1312</a>)</li>
<li><a
href="d6f8a33132"><code>d6f8a33</code></a>
build(deps-dev): bump js-yaml from 4.1.0 to 4.1.1 (<a
href="https://redirect.github.com/actions/stale/issues/1304">#1304</a>)</li>
<li><a
href="a21a081629"><code>a21a081</code></a>
Fix checking state cache (fix <a
href="https://redirect.github.com/actions/stale/issues/1136">#1136</a>),
also switch to octokit methods (<a
href="https://redirect.github.com/actions/stale/issues/1152">#1152</a>)</li>
<li><a
href="997185467f"><code>9971854</code></a>
build(deps): bump actions/checkout from 4 to 6 (<a
href="https://redirect.github.com/actions/stale/issues/1306">#1306</a>)</li>
<li><a
href="5611b9defa"><code>5611b9d</code></a>
build(deps): bump actions/publish-action from 0.3.0 to 0.4.0 (<a
href="https://redirect.github.com/actions/stale/issues/1291">#1291</a>)</li>
<li><a
href="fad0de84e5"><code>fad0de8</code></a>
Improves error handling when rate limiting is disabled on GHES. (<a
href="https://redirect.github.com/actions/stale/issues/1300">#1300</a>)</li>
<li><a
href="39bea7de61"><code>39bea7d</code></a>
Add Missing Input Reading for <code>only-issue-types</code> (<a
href="https://redirect.github.com/actions/stale/issues/1298">#1298</a>)</li>
<li><a
href="e46bbabb3e"><code>e46bbab</code></a>
build(deps-dev): bump <code>@​types/node</code> from 20.10.3 to 24.2.0
and document breakin...</li>
<li><a
href="65d1d4804d"><code>65d1d48</code></a>
build(deps-dev): bump eslint-config-prettier from 8.10.0 to 10.1.8 (<a
href="https://redirect.github.com/actions/stale/issues/1276">#1276</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/actions/stale/compare/v9...v10">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/stale&package-manager=github_actions&previous-version=9&new-version=10)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-02-25 15:30:55 +08:00
dependabot[bot]
1ff79079a6 Bump actions/github-script from 7 to 8 (#25547)
Bumps [actions/github-script](https://github.com/actions/github-script)
from 7 to 8.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/actions/github-script/releases">actions/github-script's
releases</a>.</em></p>
<blockquote>
<h2>v8.0.0</h2>
<h2>What's Changed</h2>
<ul>
<li>Update Node.js version support to 24.x by <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/637">actions/github-script#637</a></li>
<li>README for updating actions/github-script from v7 to v8 by <a
href="https://github.com/sneha-krip"><code>@​sneha-krip</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/653">actions/github-script#653</a></li>
</ul>
<h2>⚠️ Minimum Compatible Runner Version</h2>
<p><strong>v2.327.1</strong><br />
<a
href="https://github.com/actions/runner/releases/tag/v2.327.1">Release
Notes</a></p>
<p>Make sure your runner is updated to this version or newer to use this
release.</p>
<h2>New Contributors</h2>
<ul>
<li><a href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/637">actions/github-script#637</a></li>
<li><a
href="https://github.com/sneha-krip"><code>@​sneha-krip</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/653">actions/github-script#653</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/github-script/compare/v7.1.0...v8.0.0">https://github.com/actions/github-script/compare/v7.1.0...v8.0.0</a></p>
<h2>v7.1.0</h2>
<h2>What's Changed</h2>
<ul>
<li>Upgrade husky to v9 by <a
href="https://github.com/benelan"><code>@​benelan</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/482">actions/github-script#482</a></li>
<li>Add workflow file for publishing releases to immutable action
package by <a
href="https://github.com/Jcambass"><code>@​Jcambass</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/485">actions/github-script#485</a></li>
<li>Upgrade IA Publish by <a
href="https://github.com/Jcambass"><code>@​Jcambass</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/486">actions/github-script#486</a></li>
<li>Fix workflow status badges by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/497">actions/github-script#497</a></li>
<li>Update usage of <code>actions/upload-artifact</code> by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/512">actions/github-script#512</a></li>
<li>Clear up package name confusion by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/514">actions/github-script#514</a></li>
<li>Update dependencies with <code>npm audit fix</code> by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/515">actions/github-script#515</a></li>
<li>Specify that the used script is JavaScript by <a
href="https://github.com/timotk"><code>@​timotk</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/478">actions/github-script#478</a></li>
<li>chore: Add Dependabot for NPM and Actions by <a
href="https://github.com/nschonni"><code>@​nschonni</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/472">actions/github-script#472</a></li>
<li>Define <code>permissions</code> in workflows and update actions by
<a href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in
<a
href="https://redirect.github.com/actions/github-script/pull/531">actions/github-script#531</a></li>
<li>chore: Add Dependabot for .github/actions/install-dependencies by <a
href="https://github.com/nschonni"><code>@​nschonni</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/532">actions/github-script#532</a></li>
<li>chore: Remove .vscode settings by <a
href="https://github.com/nschonni"><code>@​nschonni</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/533">actions/github-script#533</a></li>
<li>ci: Use github/setup-licensed by <a
href="https://github.com/nschonni"><code>@​nschonni</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/473">actions/github-script#473</a></li>
<li>make octokit instance available as octokit on top of github, to make
it easier to seamlessly copy examples from GitHub rest api or octokit
documentations by <a
href="https://github.com/iamstarkov"><code>@​iamstarkov</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/508">actions/github-script#508</a></li>
<li>Remove <code>octokit</code> README updates for v7 by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/557">actions/github-script#557</a></li>
<li>docs: add &quot;exec&quot; usage examples by <a
href="https://github.com/neilime"><code>@​neilime</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/546">actions/github-script#546</a></li>
<li>Bump ruby/setup-ruby from 1.213.0 to 1.222.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/github-script/pull/563">actions/github-script#563</a></li>
<li>Bump ruby/setup-ruby from 1.222.0 to 1.229.0 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/actions/github-script/pull/575">actions/github-script#575</a></li>
<li>Clearly document passing inputs to the <code>script</code> by <a
href="https://github.com/joshmgross"><code>@​joshmgross</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/603">actions/github-script#603</a></li>
<li>Update README.md by <a
href="https://github.com/nebuk89"><code>@​nebuk89</code></a> in <a
href="https://redirect.github.com/actions/github-script/pull/610">actions/github-script#610</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a href="https://github.com/benelan"><code>@​benelan</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/482">actions/github-script#482</a></li>
<li><a href="https://github.com/Jcambass"><code>@​Jcambass</code></a>
made their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/485">actions/github-script#485</a></li>
<li><a href="https://github.com/timotk"><code>@​timotk</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/478">actions/github-script#478</a></li>
<li><a
href="https://github.com/iamstarkov"><code>@​iamstarkov</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/508">actions/github-script#508</a></li>
<li><a href="https://github.com/neilime"><code>@​neilime</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/546">actions/github-script#546</a></li>
<li><a href="https://github.com/nebuk89"><code>@​nebuk89</code></a> made
their first contribution in <a
href="https://redirect.github.com/actions/github-script/pull/610">actions/github-script#610</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/actions/github-script/compare/v7...v7.1.0">https://github.com/actions/github-script/compare/v7...v7.1.0</a></p>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="ed597411d8"><code>ed59741</code></a>
Merge pull request <a
href="https://redirect.github.com/actions/github-script/issues/653">#653</a>
from actions/sneha-krip/readme-for-v8</li>
<li><a
href="2dc352e4ba"><code>2dc352e</code></a>
Bold minimum Actions Runner version in README</li>
<li><a
href="01e118c8d0"><code>01e118c</code></a>
Update README for Node 24 runtime requirements</li>
<li><a
href="8b222ac82e"><code>8b222ac</code></a>
Apply suggestion from <a
href="https://github.com/salmanmkc"><code>@​salmanmkc</code></a></li>
<li><a
href="adc0eeac99"><code>adc0eea</code></a>
README for updating actions/github-script from v7 to v8</li>
<li><a
href="20fe497b3f"><code>20fe497</code></a>
Merge pull request <a
href="https://redirect.github.com/actions/github-script/issues/637">#637</a>
from actions/node24</li>
<li><a
href="e7b7f222b1"><code>e7b7f22</code></a>
update licenses</li>
<li><a
href="2c81ba05f3"><code>2c81ba0</code></a>
Update Node.js version support to 24.x</li>
<li>See full diff in <a
href="https://github.com/actions/github-script/compare/v7...v8">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=actions/github-script&package-manager=github_actions&previous-version=7&new-version=8)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-02-25 15:30:30 +08:00
ringabout
a311ac8d22 Add Dependabot configuration for GitHub Actions (#25544)
Added configuration for Dependabot to manage GitHub Actions updates
weekly.
2026-02-25 15:13:13 +08:00
Andreas Rumpf
f3d07ff114 YRC: fixes typo (#25541)
Unrelated CI failures.
2026-02-24 11:12:28 +01:00
ringabout
b51be75613 fixes #25509; removes void fields from a named tuple type (#25515)
fixes #25509

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-02-24 10:47:14 +01:00
ringabout
1451651fd9 enable --warning:ImplicitRangeConversion (#25477)
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-02-24 09:40:21 +01:00
Tomohiro
fb80f7707d fixes #16754 (#25519)
This PR allows passing the defining type to generic types in the right
side in a type definition like this:
```nim
type
  Foo = object
    x: Option[Foo]
```
I think generic types should be instanciated after all given arguments
are semchecked,
because generic types can access information about them.
(for example, `Option[T]` in std/option checks if `T` is a pointer like
type)
But in this case, need to instanciate `Option[Foo]` before type of
`Foo.x` is determined.
2026-02-24 09:39:06 +01:00
Miroslav Shubernetskiy
86b9245dd6 fix: double check inputIndex in base64.decode (#25531)
fixes https://github.com/nim-lang/Nim/issues/25530

this double checks the index to make sure whitespace related index
increments cannot cause index defect error
2026-02-24 09:37:46 +01:00
ringabout
e58acc2e1e fixes #25005; new doesn't work with ref object (#25532)
fixes #25005

In `semTypeIdent`, when resolving a typedesc parameter inside a generic
instantiation, the code took a shortcut: it returned the symbol of the
element type (`bound = result.typ.elementType.sym`). However, for
generic types like `RpcResponse[T] = ref object`, the instantiated
object type (e.g., `RpcResponse:ObjectType[string]`) is a copy with a
new type ID but still points to the same symbol as the uninstantiated
generic body type. That symbol's .typ refers to the original
uninstantiated type, which still contains unresolved generic params `T`
2026-02-23 13:40:31 +01:00
Andreas Rumpf
6badeb1b4d yrc progress (#25534) 2026-02-23 13:39:55 +01:00
Miran
df42ebc5e6 bump Atlas' version (#25539) 2026-02-22 23:06:55 +01:00
Miran
44eafa7552 update the shipped tools (#25535) 2026-02-22 12:54:36 +01:00
ringabout
15c6249f2c replace benign with gcsafe (#25527)
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-02-20 16:41:06 +01:00
ringabout
1e3caf457b improve alignment for refc (#25525) 2026-02-17 16:00:11 +01:00
Zoom
72e9bfe0a4 Docs: parseopt fixes, runnable examples (#25526)
Follow-up to #25506.
As I mentioned there, I was in the middle of an edit, so here it is.
Splitting to a separate doc skipped.

A couple of minor mistakes fixed, some things made a bit more concise
and short.
2026-02-16 18:26:08 +01:00
Zoom
7c873ca615 Feat: std: parseopt parser modes (#25506)
Adds configurable parser modes to std/parseopt module. **Take two.**

Initially solved the issue of not being able to pass arguments to short
options as you do with most everyday CLI programs, but reading the tests
made me add more features so that some of the behaviour could be changed
and here we are.

**`std/parseopt` now supports three parser modes** via an optional
`mode` parameter in `initOptParser` and `getopt`.

Three modes are provided:
- `NimMode` (default, fully backward compatible),
- `LaxMode` (POSIX-inspired with relaxed short option handling),
- `GnuMode` (stricter GNU-style conventions).

The new modes are marked as experimental in the documentation.

The parser behaviour is controlled by a new `ParserRules` enum, which
provides granular feature flags that modes are built from. This makes it
possible for users with specific requirements to define custom rule sets
by importing private symbols, this is mentioned but clearly marked as
unsupported.

**Backward compatibility:**

The default mode preserves existing behaviour completely, with a single
exception: `allowWhitespaceAfterColon` is deprecated.

Now, `allowWhitespaceAfterColon` doesn't make much sense as a single
tuning knob. The `ParserRule.prSepAllowDelimAfter` controls this now.
As `allowWhitespaceAfterColon` had a default, most calls never mention
it so they will silently migrate to the new `initOptParser` overload. To
cover cases when the proc param was used at call-site, I added an
overload, which modifies the default parser mode to reflect the required
`allowWhitespaceAfterColon` value. Should be all smooth for most users,
except the deprecation warning.

The only thing I think can be classified as the breaking change is a
surprising **bug** of the old parser:

```nim
let p = initOptParser("-n 10 -m20 -k= 30 -40",  shortNoVal =  {'v'})
#                                     ^-disappears
```

This is with the aforementioned `allowWhitespaceAfterColon` being true
by default, of course. In this case the `30` token is skipped
completely. I don't think that's right, so it's fixed.


Things I still don't like about how the old parser and the new default
mode behave:

1. **Parser behaviour is controlled by an emptiness of two containers**.
This is an interesting approach. It's also made more interesting because
the `shortNoVal`/`longNoVal` control both the namesakes, but *and also
how their opposites (value-taking opts) work*.
---

**Edit:**

2. `shortNoVal` is not mandatory:
    ```nim
	let p = initOptParser(@["-a=foo"], shortNoVal = {'a'})
	# Nim, Lax parses as: (cmdShortOption, "a", "foo") 
	# GnuMode  parses as: (cmdShortOption, "a", "=foo")
	```
In this case, even though the user specified `a` as no no-val, parser
ignores it, relying only on the syntax to decide the kind of the
argument. This is especially problematic with the modes that don't use
the rule `prShortAllowSep` (GnuMode), in this case the provided input is
twice invalid, regardless of the `shortNoVal`.

With the current parser architecture, parsing it this way **is
inevitable**, though. We don't have any way to signal the error state
detected with the input, so the user is expected to validate the input
for mistakes.
Bundling positional arguments is nonsensical and short option can't use
the separator character, so `[cmd "a", arg "=foo"]` and `[cmd "a", cmd
"=", cmd "f"...]` are both out of the question **and** would complicate
validating, requiring keeping track of a previous argument. Hope I'm
clear enough on the issue.
	
**Future work:**

1. Looks like the new modes are already usable, but from the discussions
elsewhere it looks like we might want to support special-casing
multi-digit short options (`-XX..`) to allow numerical options greater
than 9. This complicates bundling, though, so requires a bit of thinking
through.

2. Signaling error state?

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2026-02-16 16:06:18 +01:00
ringabout
97fed258ed fixes #25475; incompatible types errors for array types with different index types (#25505)
fixes #25475

```nim
var x: array[0..1, int] = [0, 1]
var y: array[4'u..5'u, int] = [0, 3]

echo x == y
```

sigmatch treats array compatibility by element type + length, not by the
index (range) type. Perhaps backend should do the same check
2026-02-13 22:59:21 +01:00
Yuriy Glukhov
937e647f4f Importc codegen fix (#25511)
This fixes two issues with impotc'ed types.
1. Passing an importc'ed inherited object to where superclass is
expected emitted `v.Sup` previously. Now it emits `v`, similar to cpp
codegen.
2. Casting between different nim types that resolve to the same C type
previously was done like `*(T*)&v`, now it is just `v`.
2026-02-13 13:29:01 +01:00
Andreas Rumpf
b41049988f attempt to fix final issue with Nim's multi-threaded allocator (#25513) 2026-02-13 11:53:17 +01:00
Andreas Rumpf
04933b773a YRC: bugfixes (#25512) 2026-02-12 13:29:22 +01:00
Andreas Rumpf
5fa11c5686 YRC: bugfixes (#25504) 2026-02-11 17:45:40 +01:00
ringabout
94008531c1 fixes #25457; make rawAlloc support alignment (#25476)
fixes https://github.com/nim-lang/Nim/issues/25457

Small chunks allocate memory in fixed-size cells. Each cell is
positioned at exact multiples of the cell size from the chunk's data
start, which makes it much harder to support alignment

```nim
sysAssert c.size == size, "rawAlloc 6"
if c.freeList == nil:
  sysAssert(c.acc.int + smallChunkOverhead() + size <= SmallChunkSize,
            "rawAlloc 7")
  result = cast[pointer](cast[int](addr(c.data)) +% c.acc.int)
  inc(c.acc, size)
```

See also https://github.com/nim-lang/Nim/pull/12926 

While using big trunk, each allocation gets its own chunk
2026-02-11 11:33:31 +01:00
ringabout
c346a2b228 fixes #25464; infer =dup for distinct types (#25501)
fixes #25464

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2026-02-11 10:46:34 +01:00
Andreas Rumpf
f62669a5d5 Yrc typos and omissions (#25500) 2026-02-10 13:21:35 +01:00
Andreas Rumpf
a690a9ac90 YRC: threadsafe cycle collection for Nim (#25495)
First performance numbers:

time tests/arc/torcbench   -- YRC
true peak memory: true

real    0m0,163s
user    0m0,161s
sys     0m0,002s


time tests/arc/torcbench   -- ORC
true peak memory: true

real    0m0,107s
user    0m0,104s
sys     0m0,003s


So it's 1.6x slower. But it's threadsafe and provably correct. (Lean and
model checking via TLA+ used.)

Of course there is always the chance that the implementation is wrong
and doesn't match the model.
2026-02-10 00:04:11 +01:00
lit
9225d9e9e6 fixes #25490; Remove unused gEnv & env from main func (#25497)
closes #25490
2026-02-09 17:34:44 +01:00
ringabout
ae5f864bff fixes #25494; [regression] Crash on enum ranges as default parameters in generic procs (#25496)
fixes #25494;
2026-02-09 11:50:45 +01:00
ringabout
513c9aa69a fixes #25488; Strings can be compared against nil (#25489)
fixes #25488
ref https://github.com/nim-lang/Nim/pull/20222
2026-02-07 20:46:50 +01:00
ringabout
12a2333817 fixes #25464; gives a deprecated warning when =dup is not provided while there being a custom =copy (#25485)
Gives a deprecated warning to keep backwards compatibility

fixes #25464
2026-02-06 02:19:46 +01:00
Yuriy Glukhov
296b2789b5 Fixes #25340 (#25389) 2026-02-06 00:54:04 +01:00
ringabout
a04f720217 fixes #25482; ICE leaking temporary 3 slotTempInt (#25483)
fixes #25482
2026-02-05 05:41:54 +01:00
ringabout
bfc2786718 fixes #24706; Warn on implicit range downsizing (#25451)
fixes #24706
2026-02-01 17:06:33 +01:00
Tomohiro
88e7adfcb7 fixes #25459; hashType returns different hash from instantiated generics with distinct types (#25471)
`hashType` proc returned the same hash value from different instanced
generics types like `D[int64]` and `D[F]`.
That caused the struct type with wrong field types.

object/tuple type size check code is generated when it is compiled with
`-d:checkAbi` option.
2026-02-01 07:01:55 +01:00
Tomohiro
abf434a336 fixes #25231; print better error messages when generics instantiation… (#25460)
… has no params
2026-01-26 14:06:35 +01:00
Gianmarco
e7809364b3 Make it so that every feature can be used in panicoverride files (#25300)
Refer to #25298
2026-01-24 16:01:21 +01:00
ringabout
81610095e6 fixes #25441; fixes #7355; deletes void args from the argument list (#25455)
fixes #25441; fixes #7355

---------

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
2026-01-24 06:08:38 +01:00
Andreas Rumpf
469e1377cd IC: progress (#25453) 2026-01-24 06:07:41 +01:00
ringabout
f44700e638 fixes #19831; add --styleCheck:warning (#25456)
fixes #19831
2026-01-23 22:50:00 +01:00
ringabout
ace09b3cab fixes #25074; Long integer literal truncated without warning (#25449)
fixes #25074

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com>
2026-01-23 14:52:37 +01:00
ringabout
39864980d1 fixes #25446; [FieldDefect] with static: discard cast[pointer](default(pointer)) (#25448)
fixes #25446

supports this since `static: discard cast[pointer](nil)` works
2026-01-22 09:50:25 +01:00
Andreas Rumpf
86a4ddc847 IC: remember which modules are from the NIF cache (#25443) 2026-01-22 04:26:20 +01:00
ringabout
9a23ff36bd fixes #25400; Naked raised causes wrong exception effect (#25422)
fixes #25400

infers `Exception` for Naked raised
2026-01-16 15:25:14 +01:00
Andreas Rumpf
efc3a7429b IC: progress (#25440) 2026-01-16 15:24:21 +01:00
Andreas Rumpf
cf388722db IC: massive cleanup, NIF26 support, docs about its inner workings (#25427) 2026-01-16 12:19:17 +01:00
ringabout
80b43ad6ce Remove URL from BipBuffer package entry (#25439)
ref https://github.com/MarcAzar/BipBuffer/pull/1
2026-01-16 19:12:51 +08:00
ringabout
40480fe348 fixes #25419; lift magic types to typeclasses (#25421)
fixes #25419
2026-01-14 16:25:52 +01:00
Jake Leahy
c1e381ae8d Raw switch for jsondoc (#24568)
Implements #21928

Adds a `--raw` (since thats what the original issue used, suggestions
welcome) switch which stops the jsondoc gen from rendering rst/markdown.

Implemented by making `genComment` check if it needs to return the raw
string or not. This required switching the related procs to using
`Option` to handle how `nil` values were returned before. The `nil`
returns were eventually ignored so just ignoring `none(T)` has the same
effect.

Doesn't support `runnableExamples` since jsondocs doesn't support them
either
2026-01-11 18:39:01 +08:00
Copilot
83d7d8c634 Add test case for jsffi type mismatch error (#16726) (#25429)
Issue #16726 reported an internal compiler error (`semcall.nim(229, 18)
nArg != nil`) when calling `toJs` with invalid arguments. The bug has
been fixed in the current codebase but lacked a regression test.

```nim
import std/jsffi
let a = toJs(3)
let b = a.toJs(int)  # Now produces: Error: type mismatch
                      # Previously: internal error
```

**Changes:**
- Added `tests/js/t16726.nim` to verify proper type mismatch error is
reported instead of internal compiler error

<!-- START COPILOT ORIGINAL PROMPT -->



<details>

<summary>Original prompt</summary>

> 
> ----
> 
> *This section details on the original issue you should resolve*
> 
> <issue_title>internal error: semcall.nim(229, 18) nArg !=
nil</issue_title>
> <issue_description>### Example
> the code below should give a clean CT error, not an internal error
> ```nim
> import std/jsffi
> let a = toJs(3)
> let b = a.toJs(int)
> ```
> 
> ### Current Output
> nim r -b:js main
> compiler/semcall.nim(229, 18) `nArg != nil` 
> 
> ### Expected Output
> proper CT error
> 
> ### Additional Information
> 1.5.1 41965880ce
> </issue_description>
> 
> <agent_instructions>adds a test case for it and verify it by `./koch
temp js -r test.nim` before committing</agent_instructions>
> 
> ## Comments on the Issue (you are @copilot in this section)
> 
> <comments>
> <comment_new><author>@ringabout</author><body>
> related:
https://github.com/nim-lang/Nim/issues/15607</body></comment_new>
> </comments>
> 


</details>



<!-- START COPILOT CODING AGENT SUFFIX -->

- Fixes nim-lang/Nim#16726

<!-- START COPILOT CODING AGENT TIPS -->
---

💬 We'd love your input! Share your thoughts on Copilot coding agent in
our [2 minute survey](https://gh.io/copilot-coding-agent-survey).

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2026-01-10 21:18:39 +08:00
Andreas Rumpf
01eedd916c IC: progress (#25420) 2026-01-09 13:10:04 +01:00
Andreas Rumpf
b3273e732d IC: progress (#25417) 2026-01-07 17:35:07 +01:00
Andreas Rumpf
251b4a23c3 IC: run nifmake automatically (#25415) 2026-01-07 13:45:26 +01:00
ringabout
89c8f0aa49 closes #23394; adds a test case (#25416)
closes #23394
2026-01-07 09:32:25 +01:00
Andreas Rumpf
d3be5e5e13 IC: need a more recent Nimony for its improved Nifler tool (#25412) 2026-01-06 00:16:19 +01:00
elijahr
780c9eeef0 fixes #25405; initialization for objects with opaque importc fields (#25406)
Objects containing `importc` fields without `completeStruct` fail to
compile when used as const/static. The C codegen generates "aggregate
initialization" which is invalid for opaque types.

Fixes #25405.

Nim code:

```nim
  type
    OpaqueInt {.importc: "_Atomic int", nodecl.} = object

    ContainsImportc = object
      normal: int
      opaque: OpaqueInt

  const c = default(ContainsImportc)
```

Resulting C code:

```c
// Invalid C - cannot aggregate-init opaque type
NIM_CONST ContainsImportc c = {((NI) 0), {}};
                                         ^^ error: illegal initializer type
```

## Solution

Fix in `ccgexprs.nim`:
1. Skip opaque importc fields when building aggregate initializers
2. Use "designated initializers" (`siNamedStruct`) when opaque fields
are present to avoid positional misalignment

```c
// Valid C:
//  - opaque field is omitted and implicitly zero-initialized by C
//  - other fields are explitly named and initialized
NIM_CONST ContainsImportc c = {.normal = ((NI) 0)};
```

This correctly handles the case where the opaque fields might be in any
order.

A field is considered "opaque importc" if:
- Has `sfImportc` flag
- Does NOT have `tfCompleteStruct` flag
- Either has `tfIncompleteStruct` OR is an object with no visible fields

The `containsOpaqueImportcField` proc recursively checks all object
fields, including nested objects and variant branches.

Anonymous unions (from variant objects) are handled by passing an empty
field name, which skips the `.fieldname = ` prefix since C anonymous
unions have no field name.

Note that initialization for structs without opaque importc fields
remains the same as before this changeset.

## Test Coverage

`tests/ccgbugs/timportc_field_init.nim` covers:
- Simple struct with one importc field
- Nested struct containing struct with importc field
- Variant object (case object) with importc field in a branch
- Array of structs with importc fields
- Tuple containing struct with importc field
- `completeStruct` importc types (still use aggregate init)
- Sandwich case (opaque field between two non-opaque fields)
- Fields with different C names (`{.importc: "c_name".}`, `{.exportc.}`)
- `{.packed.}` structs with opaque fields
- `{.union.}` types with opaque fields
- Deep nesting (3+ levels)
- Multiple opaque fields with renamed fields between them
2026-01-05 15:21:59 +01:00
ringabout
a6c7989c7f remove duplicated module imports (#25411) 2026-01-05 15:14:36 +01:00
ringabout
1a651c17b3 hello 2026 (#25410) 2026-01-05 19:36:33 +08:00
Ryan McConnell
4b615aca46 memfiles.nim resizeFile fallback logic bug (#25408)
`e` is not cleared when falling back to `ftruncate`
2026-01-03 18:08:12 +01:00
Jacek Sieka
92ad98f5d8 pegs: get rid of spurious exception effects (#25399)
Pegs raise only their own error, but the forward declaration causes an
unwanted Exception effect

* use strformat which does compile-time analysis of the format string to
avoid exceptions
* also in parsecfg
2026-01-01 01:33:35 +01:00
Esteban C Borsani
ae8a1739f8 Add parseEnum support for triple quoted string and raw string enum values (#25401) 2026-01-01 01:31:33 +01:00
Pierre Thibault
ee55ddcffd Missleading sentence about array indexing (#25367)
I added some precision. The first time I read this sentence, I was
confused. This applies to the above example, but it cannot be
generalized, since every array has its own range of valid indexes.

I think this change make the documentation clearer.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-12-31 13:34:18 +01:00
Jacek Sieka
61970be479 reduce imports (#25398) 2025-12-31 13:33:57 +01:00
bptato
e97b0bb541 Do not directly cast int128 to uint64 in semfold (#25396)
int128 is an array of uint32s, so while this works on little-endian
CPUs, it's completely broken on big-endian. e.g. following snippet would
fail:

	const x = 0xFFFFFFFF'u32
	const y = (x shr 1)
	echo y # amd64: 2147483647, s390x: 0

That in turn broke float printing, resulting in miscompilation of any
code that used floats.

To fix this, we now call the aptly named castToUInt64 procedure which
performs the same cast portably.

(Thanks to barracuda156 for helping debug this.)
2025-12-30 23:09:01 +01:00
Andreas Rumpf
234c73c58a refactoring for IC (#25395) 2025-12-29 13:52:22 +01:00
ringabout
f1b97caf92 fixes #19983; implements bitmasked bitshifting for all backends (#25390)
replaces https://github.com/nim-lang/Nim/pull/11555

fixes https://github.com/nim-lang/Nim/issues/19983
fixes https://github.com/nim-lang/Nim/issues/13566

- [x] JS backend

---------

Co-authored-by: Arne Döring <arne.doering@gmx.net>
2025-12-29 10:25:56 +01:00
Andreas Rumpf
22d4644d36 refactoring (#25394) 2025-12-29 10:23:46 +01:00
Andreas Rumpf
02893e2f4c IC: code generation progress (#25379) 2025-12-29 00:20:33 +01:00
Jake Leahy
91d51923b9 Fix tupleLen not skipping aliases (#25392)
This code was failing to compile with `Error: unhandled exception:
semmagic.nim(247, 5) operand.kind == tyTuple tyAlias [AssertionDefect]`
```nim
import std/typetraits

type
  Bar[T] = T 
  Foo = Bar[tuple[a: int]]

echo Foo.tupleLen
```

Fix was just making `tupleLen` skip alias types also
2025-12-28 16:45:07 +01:00
Tomohiro
c48347136f Refactoring #25302; don't store procedure's parameter types to PType.sonsImpl (#25351) 2025-12-26 21:59:38 +01:00
bptato
a061f026a8 Fix std/hashes completely ignoring endianness (#25386)
This is a problem on big-endian CPUs because you end up with nimvm
computing something different than Nim proper, so e.g. a const table
won't work.

I also took the liberty to replace a redundant implementation of load4
in murmurHash.

(Thanks to barracuda156 for helping debug this.)
2025-12-25 21:04:04 +01:00
ringabout
a41bbf6901 fixes #25387; embedsrc breaks with Line Continuation (#25388)
fixes #25387


https://stackoverflow.com/questions/30286253/how-to-escape-backslash-in-comment

- adding a whitespace or `\t` after `\` breaks the `goto` block
- `\* *\` doesn't support nesting, causing problems for using it in the
Nim comments
2025-12-25 21:02:54 +01:00
ringabout
5e53a70e62 fixes #25254; fixes #10395; Invalid pred in when swallowed (#25385)
fixes #25254
fixes #10395
2025-12-25 00:04:18 +01:00
Yuriy Glukhov
2dbdf08fc7 Fixes #25319 (#25380)
This was a regression introduced in
https://github.com/nim-lang/Nim/pull/25070.

@janAkali, @Z9RO, can you verify please?
2025-12-21 19:13:25 +01:00
elijahr
b819472e74 Fix sizeof(T) in typedesc templates called from generic type when clauses (#25374)
The `hasValuelessStatics` function in `semtypinst.nim` only checked for
`tyStatic`, missing `tyTypeDesc(tyGenericParam)`. This caused
`sizeof(T)` inside a typedesc template called from a generic type's
`when` clause to error with "'sizeof' requires '.importc' types to be
'.completeStruct'".

The fix adds a check for `tyTypeDesc` wrapping `tyGenericParam`,
recognizing it as an unresolved generic parameter that needs resolution
before evaluation.

Also documents the `completeStruct` pragma in the manual.
2025-12-21 07:37:26 +01:00
Amjad Ben Hedhili
7b12deecf4 [Docs] Remove horizontal scrolling on mobile (#25377)
* Also use more of the available width
2025-12-21 07:35:40 +01:00
Andreas Rumpf
b901a80710 IC: progress (#25368)
Co-authored-by: Jacek Sieka <arnetheduck@gmail.com>
Co-authored-by: Ryan McConnell <rammcconnell@gmail.com>
2025-12-20 11:27:46 +01:00
elijahr
1324183c38 fix #17630: Implement cycle detection for recursive concepts (#25353)
fixes #17630

## Recursive Concept Cycle Detection

- Track (conceptId, typeId) pairs during matching to detect cycles
- Changed marker from IntSet to HashSet[ConceptTypePair]
- Removed unused depthCount field
- Added recursive concepts documentation to manual
- Added tests for recursive concepts, distinct chains, and co-dependent
concepts

## Fix Flaky `tasyncclosestall` Test

The macOS ARM64 CI jobs were failing due to a flaky async socket test
(unrelated to concepts).

The test only accepted `EBADF` as a valid error code when closing a
socket with pending writes. However, depending on timing, the kernel may
report `ECONNRESET` or `EPIPE` instead:

- **EBADF**: Socket was closed locally before kernel detected remote
state
- **ECONNRESET**: Remote peer sent RST packet (detected first)  
- **EPIPE**: Socket is no longer connected (broken pipe)

All three are valid disconnection errors. The fix accepts any of them,
making the test reliable across platforms.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-12-20 08:56:10 +01:00
ringabout
548b1c6ef8 fixes #25369 (#25370)
fixes #25369
2025-12-18 18:54:03 +01:00
Andreas Rumpf
9bb57a64ba IC: keep package information (#25350) 2025-12-18 09:34:39 +01:00
Andreas Rumpf
80cf9a8ce8 system.nim: memory must be part of system so that its compilerprocs c… (#25365)
…an work for IC
2025-12-18 04:53:09 +01:00
Ryan McConnell
8747160a9a flush stdout when prompting for password (#25348)
Saw this misbehave on Linux. It was fine in Windows when I checked, but
I figured it can't hurt.
2025-12-18 04:52:39 +01:00
Andreas Rumpf
334ac3f588 refs https://github.com/nim-lang/Nim/pull/25353 make tasyncclosestall… (#25366)
….nim less flaky
2025-12-17 20:25:51 +01:00
Jacek Sieka
1527c13273 Align treetab hash with equivalence (#25354)
In particular, hash `typ` for `nkType`, `nkNilLit` or they end up
generating collisions

<img width="989" height="612" alt="image"
src="https://github.com/user-attachments/assets/a5c6366f-1214-443e-98d5-52ce95fc3555"
/>
2025-12-15 13:19:56 +01:00
ringabout
6f3245f06a fixes documentation building failures for nightlies (#25345)
```
Error: '`' expected
```
2025-12-11 18:23:04 +01:00
Andreas Rumpf
cbb2fe0a63 IC: progress (#25344)
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-12-11 18:22:38 +01:00
ringabout
28ada7df94 fixes markdown tests (#25347) 2025-12-11 21:36:27 +08:00
metagn
44d2472b08 consider generic param type as typedesc in tuple type expressions (#25316)
fixes #25312

Tuple expressions `(a, b, c)` can be either types or values depending on
if their elements are typedescs or values, this is checked by checking
if the type of the element is `tyTypeDesc`. However when an
`skGenericParam` symbol is semchecked by `semSym` it is given its own
`tyGenericParam` type rather than a `tyTypeDesc` type, this seems to be
necessary for signatures to allow wildcard generic params passed to
static constrained generic params (tested in #25315). The reason
`semSym` is called is that `semGeneric` for generic invocations calls
`matches` which sems its arguments like normal expressions.

To deal with this, an expression of type `tyGenericParam` and with a
`skGenericParam` sym is allowed as a type in the tuple expression. A
problem is that this might consider a value with a wildcard generic
param type as a type. But this is a very niche problem, and I'm not sure
how to check for this. `skGenericParam` symbols stay as idents when
semchecked so it can't be checked that the node is an `skGenericParam`
symbol. It could be checked that it's an ident but I don't know how
robust this is. And maybe there is another way to refer to a wildcard
generic param type instead of just its symbol, i.e. another kind of
node.

This also makes #5647 finally work but a test case for that can be added
after.
2025-12-09 09:45:37 +01:00
ringabout
e1f2329e55 fixes #25329; Wrong type for second parameter of procedures "inc", "dec", "succ" and "pred" (#25337)
fixes #25329
2025-12-09 07:16:23 +01:00
ringabout
ed8e5a7754 fixes #25338; Switch default mangling back to cpp (#25343)
fixes #25338
2025-12-09 07:16:08 +01:00
Andreas Rumpf
fa4d79f519 IC: progress (#25339) 2025-12-07 13:07:44 +01:00
elijahr
099ee1ce4a Fixes #25341; Invalid C code for lifecycle hooks for distinct types based on generics (#25342) 2025-12-07 12:59:42 +01:00
Andreas Rumpf
c3a20fa890 IC: progress (#25332) 2025-12-06 11:45:01 +01:00
Yuriy Glukhov
8f8814b495 Fixes #25330 (#25336)
Fixed state optimizer. It did not replace deleted states in
`excLandingState`.
2025-12-05 15:27:38 +01:00
ringabout
0ea5f2625c fixes #25306; Dangling pointers in stack traces with -d:nimStackTraceOverride (#25313)
fixes #25306

```nim
type
  StackTraceEntry* = object ## In debug mode exceptions store the stack trace that led
                            ## to them. A `StackTraceEntry` is a single entry of the
                            ## stack trace.
    procname*: cstring      ## Name of the proc that is currently executing.
    line*: int              ## Line number of the proc that is currently executing.
    filename*: cstring      ## Filename of the proc that is currently executing.
    when NimStackTraceMsgs:
      frameMsg*: string     ## When a stacktrace is generated in a given frame and
                            ## rendered at a later time, we should ensure the stacktrace
                            ## data isn't invalidated; any pointer into PFrame is
                            ## subject to being invalidated so shouldn't be stored.
    when defined(nimStackTraceOverride):
      programCounter*: uint ## Program counter - will be used to get the rest of the info,
                            ## when `$` is called on this type. We can't use
                            ## "cuintptr_t" in here.
      procnameStr*, filenameStr*: string ## GC-ed alternatives to "procname" and "filename"
```
2025-12-05 07:52:35 +01:00
ringabout
5d4829415a fixes #25324; Channel incorrectly takes a sink argument in refc (#25328)
… it performs a deep copy internally

fixes #25324

notes that
> Enabling `-d:nimPreviewSlimSystem` removes the import of
`channels_builtin` in
in the `system` module, which is replaced by
[threading/channels](https://github.com/nim-lang/threading/blob/master/threading/channels.nim).
2025-12-05 00:46:47 +01:00
Ryan McConnell
86bbc73b3a concept patch: inheritance (#25317)
adds some inheritance support

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-12-03 17:51:18 +01:00
ringabout
1da0dc74d9 fixes #22305; Combination of generic destructor and closure fails in certain cases (#25327)
fixes #22305

It seems that the generic type is cached somehow so that no hooks are
instantiated for the generic type. There are only hooks for the
instantiated type. When `lambdalifting` tries to create type bounds for
the generic type, it cannot either find the instantiated hooks or
instantiate the generic hooks since it lacks `SemContext`. It can use
hooks for the instantiated type in this case
2025-12-03 17:29:45 +01:00
Ryan
2d0b62aa51 std: sysatomics: fix use of atomicCompareExchangeN for MSVC (#25325)
`InterlockedCompareExchange64 `(winnt.h) is used instead of gcc atomics
when compiling with MSVC on Windows, but the function signatures are
`InterlockedCompareExchange64(ptr int64, int64, int64)` and
`InterlockedCompareExchange32(ptr int32, int32, int32)` as opposed to
`(ptr T, ptr T, T)` for `__atomic_compare_exchange_n`.

Passing a pointer to the expected value (parameter two) instead of the
value itself causes the comparison to unconditionally fail, with stalls
in threaded code using atomic comparisons.

Fix the function signature for MSVC.

Signed-off-by: Ryan Walklin <ryan@testtoast.com>
2025-12-02 17:07:07 +08:00
Jacek Sieka
91febf1f4c Ensure channels don't leak exception effects (#25318)
The forward declarations cause `Exception` to be inferred - also,
`llrecv` is an internal implementation detail and the type of the
received item is controlled by generics, thus the ValueError raised
there seems out of place for the generic api.
2025-12-01 22:59:26 +01:00
Andreas Rumpf
a773178e2b IC: progress (#25314) 2025-12-01 22:59:12 +01:00
Yuriy Glukhov
6656084004 Fixes #25261 (#25310)
Returning or yielding from a closureiter must restore "external"
exception, but `popCurrentException` from `blockLeaveActions` was
getting in the way. So now `blockLeaveActions` doesn't emit
`popCurrentException` for returns in closureiters. I'm not a fan of this
"abstraction leakage", but don't see a better solution yet. Any input is
much appreciated.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-11-27 10:09:52 +01:00
Andreas Rumpf
0486a2df51 IC progress (#25283)
bugfix: produce the required nimcache subdir
2025-11-25 12:49:23 +01:00
Peter Munch-Ellingsen
6543040d40 Fixes #25304 proper test for hlo recursion limit (#25305)
The `warnUser` message kind is probably not the right one, but I left it
as a placeholder. It should probably at least warn if not just straight
up throw an error, was very hard to figure out what went wrong without
any indication. The hard coded 300 should possibly also be changed to
`evalTemplateLimit` or the VM call recursion limit or something.
2025-11-21 21:26:43 +01:00
Andreas Rumpf
0f7b378467 system.nim refactorings for IC (#25295)
Generally useful refactoring as it produces better code.
2025-11-19 16:27:31 +01:00
ringabout
46d4079357 use nimKochBootstrap for niminst (#25293)
so that `nimony` won't be required for nightlies. It's annoying to build
`nimony` on each platform, e.g. `std/memfiles` which is used by `nimony`
is not supported by `nintendoswitch`

```
bin/nim compile -f --incremental:off --compileonly --gen_mapping --cc:gcc --skipUserCfg --os:nintendoswitch --cpu:arm64 -d:danger -d:gitHash:cd69f37f3a4fb46468b77b84ccf6aa3225c8895e compiler/nim.nim
```

```
/home/runner/work/nightlies/nightlies/nim/lib/pure/memfiles.nim(107, 40) Error: undeclared identifier: 'MAP_SHARED'
candidates (edit distance, scope distance); see '--spellSuggest': 
 (4, 5): 'freeShared'
```
2025-11-18 20:03:34 +08:00
ringabout
cd69f37f3a Update NimonyStableCommit hash (#25292)
ref
322178d9af

I will add an i386 CI for bootstrapping later
2025-11-18 17:06:28 +08:00
Ryan McConnell
79ddb7d89e concept patch for tyGenericInvocation (#25288)
matching between some generic invocations and equivalent instantiations
did not have a code path
2025-11-15 12:52:16 +01:00
ringabout
7cb8165e75 ref nightlies; Update NimonyStableCommit to a new version (#25289)
596ae916a8
2025-11-15 10:09:57 +01:00
Zoom
b539adf829 std: sysstr cleanup, add docs (#25180)
- Removed redundant `len` and `reserved` sets already performed by prior
`rawNewStringNoInit` calls.
- Reuse `appendChar`
- Removed never used `newOwnedString`
- Added internal `toOwnedCopy`
- Documents differences in impls of internal procs used for
`system.string.setLen`:
  + `strs_v2.setLengthStrV2`:
    - does not set the terminating zero byte when new length is 0
    - does not handle negative new length
  + `sysstr.setLengthStr`:
    - sets the terminating zero byte when new length is 0
    - bounds negative new length to 0
2025-11-15 08:42:10 +01:00
Zoom
01c084077e std: sysstr refactor (#25185)
Continuation of #25180. This one refactors the sequence routines.

Preparation for extending with new routines.

Mostly removes repeating code to simplify debugging.

Removes:
 - `incrSeqV2` superseded by `incrSeqV3`,
 - `setLengthSeq` superseded by `setLengthSeqV2`

Note comment on line 338, acknowledging that implementation of
`setLenUninit` from #25022 does zero the new memory in this branch,
having been copied from `setLengthSeqV2`. This PR does not fix this.
2025-11-15 08:41:57 +01:00
lit
39be9b981d fixes #25227; crash when codegen user-defined tuple iterate (#25228)
fixes #25227
2025-11-14 18:43:13 +01:00
ringabout
9becd1453d fixes #25284; .global initialization inside method hoisted to preInitProc (#25285)
fixes #25284

```nim
proc m2()  =
  let v {.global, used.}: string = f2(f2("123"))
```

transform lifted `.global`statements in the top level scope
2025-11-14 16:20:42 +01:00
Andreas Rumpf
f608e109c9 massive refactoring for IC (#25282)
TODO:

- [ ] test writing of .nif files
- [x] implement loading of fields in PType/PSym that might not have been
loaded
- [ ] implement interface logic
- [ ] implement pragma "replays"
- [ ] implement special logic for `converter`
- [ ] implement special logic for `method`
- [ ] test the logic holds up for `export`
- [ ] implement logic to free the memory of PSym/PType if memory
pressure is high
- [ ] implement logic to close memory mapped files if too many are open.

---------

Co-authored-by: demotomohiro <gpuppur@gmail.com>
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
Co-authored-by: Jacek Sieka <arnetheduck@gmail.com>
2025-11-13 21:31:24 +01:00
Andreas Rumpf
4c6d9b6068 nimsuggest tester: remove PCRE dependency (#25279)
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-11-13 09:07:31 +01:00
Andreas Rumpf
5da72efbde VM: refactoring [backport] (#25280)
Note to @narimiran backport because IC requires it.
2025-11-12 19:04:21 +01:00
ringabout
d5549a3c65 updates to macos-15 (#25278)
ref https://github.com/actions/runner-images/issues/13046
2025-11-12 20:33:26 +08:00
ringabout
a57b6d8406 uses csources_v3 (#25273) 2025-11-11 14:00:47 +01:00
lit
2679b3221c fixes #19846; std/unicode.strip trailing big chars (#25274)
fixes #19846
2025-11-11 12:01:07 +01:00
Ryan McConnell
cc4c7377b2 silence mass dump of BareExcept when using unittest (#25260)
Seems better to change it to `CatchableError` instead?
2025-11-10 07:27:50 +01:00
ringabout
92468e99f7 fixes #25265; fixes #23453; Unable to build Nim 2.2.6 tools from source (#25269)
fixes #25265;
fixes #23453

`(addr deref (ptr object))` generated weak typedesc before, which causes
problems for old GCC versions. As a bonus, by generating a typedesc for
`deref (ptr object)`, it also fixes #23453
2025-11-08 06:04:05 +01:00
metagn
839cbeb371 js: replace push.apply with for loop for string add [backport] (#25267)
While `a.push.apply(a, b)` is better for performance than the previous
`a = a.concat(b)` due to the fact that it doesn't create a new array,
there is a pretty big problem with it: depending on the JS engine, if
the second array is too long, it can [cause a
crash](https://tanaikech.github.io/2020/04/20/limitation-of-array.prototype.push.apply-under-v8-for-google-apps-script/)
due to the function `push` taking too many arguments. This has
unfortunately been what the codegen produces since 1.4.0 (commit
707367e1ca).

So string addition is now moved to a compilerproc that just uses a `for`
loop. From what I can tell this is the most compatible and the fastest.
Only potential problem compared to `concat` etc is with aliasing, i.e.
adding an array to itself, but I'm guessing it's enough that the length
from before the iteration is used, since it can only grow. The test
checks for aliased nim strings but I don't know if there's an extra
protection for them.
2025-11-07 20:19:50 +08:00
Andreas Rumpf
809662a228 VM: optimize 'return' slots; saves millions of node allocations for N… (#25266)
…imbus
2025-11-07 11:05:13 +01:00
ringabout
6f73094263 fixes #25251; SIGBUS with iterator over const Table lookup - premature temporary destruction (#25255)
fixes #25251

enforce a copy if the arg is a deref of a lent pointer since the arg
could be a temporary that will go out of scope
2025-11-07 10:06:05 +01:00
Jacek Sieka
861ebc0f19 Add heaptrack support (#25257)
This PR, courtesy of @NagyZoltanPeter
(https://github.com/waku-org/nwaku/pull/3522) adds the ability to track
memory allocations in a program suitable for use with
[heaptrack](https://github.com/KDE/heaptrack).

By passing `-d:heaptrack --debugger:native` to compilation, calls to
heaptrack will be injected when memory is being allocated and released -
unlike `-d:useMalloc` this strategy also works with `refc` and the
default memory pool.

See https://github.com/KDE/heaptrack for usage examples. The resulting
binary needs to be run with `heaptrack` and with the shared
`libheaptrack_preload.so` in the `LD_LIBRARY_PATH`.
2025-11-06 17:33:52 +01:00
ringabout
1d08c4e241 fixes #25263; provides a new switch mangle:nim/cpp for debug name mangling (#25264)
fixes #25263

- [x] documentation and changelogs
2025-11-06 16:41:13 +01:00
Andreas Rumpf
cfefd1d95b produces vastly better error messages for implicit --import and --inc… (#25258)
…lude configuration options
2025-11-04 17:54:23 +01:00
ringabout
d54b5f3ae1 fixes #25252; Unexpected ambiguous call with fields over object with default fields (#25256)
fixes #25252
2025-11-04 20:08:07 +08:00
Tomohiro
5079074b1a uses newer Nimony (#25249)
Old Nimony has unfixed issue
https://github.com/nim-lang/nimony/issues/1313.
So https://github.com/nim-lang/Nim/pull/25243 doesn't work correctly
with float literal `-0.0`.
This PR updates the Nimony that fixed the issue.
2025-10-31 16:59:38 +01:00
Yuriy Glukhov
99a222d63d Respect noinit for generic types (#25250) 2025-10-31 16:59:24 +01:00
Miran
3e9a66599a bump the shipped version of Atlas (#25248) 2025-10-30 20:30:28 +01:00
ringabout
ce6a34597d fixes #24575; _GNU_SOURCE redefined (#25247)
fixes #24575
2025-10-28 18:39:50 +01:00
Yuriy Glukhov
7af4e3eefd Fixes #25202 (#25244) 2025-10-28 12:48:22 +01:00
ringabout
130eac2f93 fixes #25008; Compiler internal error with static overload (#25234)
fixes #25008

It seems that `semOverloadedCall` evaluates the same node twice using
`tryConstExpr` in order for `efExplain` to print all the diagnostic
output. The problem is that `tryConstExpr` has side effects, i.e., it
changes the slot index of variables after VM execution.
2025-10-28 11:47:20 +01:00
ringabout
b8ce11dd9d fixes #25027; nim doc uses doc comment from private field for public field (#25239)
fixes #25027
2025-10-23 19:19:27 +02:00
ringabout
b7c02e9bad fixes #25240; forbids modifying a Deque changed while iterating over it (#25242)
fixes #25240

> Deque items behavior is not the same on 2.0.16 and 2.2.0

The behavior seems to be caused by the temp introduced for the parameter
`deq.len`, which prevents it from being evaluated multiple times
2025-10-23 19:18:57 +02:00
Ryan McConnell
544c26c0b8 add srcDir variable to nim.cfg (#24919)
There might be a way to do this but I couldn't find anything about it.
This is a very simple thing that goes a long way in certain situations.
Trying to avoid needing to switch to nimscript just to get:
```nim
# config.nims
import os
let srcDir = currentSourcePath.parentDir()
switch("define", &"ProjPath:\"{srcDir}\"")
```
with this change just needs:
```
# nim.cfg
d %= "ProjPath=$srcDir"
```
2025-10-23 09:05:05 +02:00
ringabout
c449c72498 fixes #25236; broken assignment hooks of union inside variant object in orc (#25238)
fixes #25236
2025-10-21 16:59:22 +02:00
ringabout
1eae14a3be fixes #25226; VM repr raises RangeDefect for long string under refc (#25230)
fixes #25226

`int16` seems to be too small for a reasonable VM program
2025-10-17 17:32:28 +02:00
ringabout
5abd21dfa5 fixes #25123; fixes #11862; Case object from compileTime proc unable to be passed as static param (#25224)
fixes #25123; fixes #11862

follow up https://github.com/nim-lang/Nim/pull/24442
ref https://github.com/nim-lang/Nim/pull/24441

> To fix this, fields from inactive branches are now detected in
semmacrosanity.annotateType (called in fixupTypeAfterEval) and marked to
prevent the codegen of their assignments. In
https://github.com/nim-lang/Nim/pull/24441 these fields were excluded
from the resulting node, but this causes issues when the node is
directly supposed to go back into the VM, for example as const values. I
don't know if this is the only case where this happens, so I wasn't sure
about how to keep that implementation working.

Object variants fields coming from inactive branches from VM are now
flagged `nfPreventCg`. We can ignore them, as done by the C backends.
2025-10-16 18:22:46 +02:00
ringabout
f009ea6c3e fixes #25208; generates a copy for opcLdConst in the assignments (#25211)
fixes #25208


```nim
type Conf = object
  val: int

const defaultConf = Conf(val: 123)
static:
  var conf: Conf
  conf = defaultConf
```

```nim
# opcLdConst is now always valid. We produce the necessary copy in the
# assignments now:
```

A `opcLdConst` is generated for `defaultConf` in `conf = defaultConf`.
According to the comment above, we need to handle the copy for
assignments of `opcLdConst`
2025-10-16 18:22:06 +02:00
lit
8f3bdb6951 fixes #25222; cast[char](i) not trunc on JS (#25223)
fixes #25222
2025-10-16 18:21:37 +02:00
ringabout
31d64b57d5 fixes #25046; Infinite loop with anonymous iterator (#25221)
fixes #25046

```nim
proc makeiter(v: string): iterator(): string =
  return iterator(): string =
    yield v

# loops
for c in makeiter("test")():
  echo "loops ", c
```
becomes

```nim
var temp = makeiter("test")
for c in temp():
  echo "loops ", c
```
for closures that might have side effects
2025-10-15 12:11:15 +02:00
ringabout
2be0721236 fixes vtable documentation in tut2 (#24304)
ref https://forum.nim-lang.org/t/12537#77311
2025-10-15 08:22:27 +02:00
ringabout
f191ba8ddd fixes #25048; Closure environment wrongly marked as cyclic (#25220)
fixes  #25048

```nim
proc canFormAcycleAux =
  of tyObject:
    # Inheritance can introduce cyclic types, however this is not relevant
    # as the type that is passed to 'new' is statically known!
    # er but we use it also for the write barrier ...
    if tfFinal notin t.flags:
      # damn inheritance may introduce cycles:
      result = true
```

It seems that all objects without `tfFinal` in their flags are
registering cycles. It doesn't seem that `Env` can be a cyclic type
because of inheritance since it is not going to be inherited after all
by another `Env` object type
2025-10-15 07:24:48 +02:00
ringabout
fb4a82f5cc fixes #25210; VM error when passing object field ref to proc(var T): var T (#25213)
fixes #25210
no longer transform `addr(obj.field[])` into `obj.field` to keep the
addressing needed for VM
2025-10-15 07:21:08 +02:00
ringabout
c0fa86872b fixes nightlies due to UB errors; increase maxCPU hard limits (#25219)
ref https://github.com/nim-lang/Nim/pull/25217

The issues is actually that there is a hard limit for max cpus in
niminst: https://github.com/nim-lang/Nim/pull/25219, which set to 20
while there is a 21 cpus now
2025-10-13 13:47:06 +02:00
Juan Carlos
1ef81f4190 Fix Bisect (#25218)
- Nim requires `SSL_library_init`, OpenSSL 3.x removed
`SSL_library_init`, Windows defaults to OpenSSL 3.x, then install
OpenSSL 1.x on Windows.
- Keep `jiro4989/setup-nim-action` at `v1`, because `v2` uses a YAML
"hardcoded" matrix of Nim versions, but this Bisect "dynamically" finds
the Nim version with a bug, therefore we cant hardcode Nim versions in
the YAML, the Bisect programmatically installs required Nim versions as
it goes bisecting commit-by-commit.
- Update `actions/checkout` from `v4` to `v5`.
- Add support for Nim `2.2.4`.

@ringabout
2025-10-11 11:23:40 +08:00
ringabout
3962264c35 nightlies regressions: CPU order matters for C sources? (#25217)
ref https://github.com/nim-lang/Nim/pull/25056


https://github.com/nim-lang/nightlies/actions/runs/18053288396/job/51378922406#step:12:1572

```
bin/nim compile -f --incremental:off --compileonly --gen_mapping --cc:gcc --skipUserCfg --os:windows --cpu:loongarch64 -d:danger -d:gitHash:f4497c61584dca8acd489ceb7ba862b150f5cf55 compiler/nim.nim
```

`loongarch64` is applied to all the platforms wrongly. Presumably it was
caused by the order?
2025-10-10 19:11:58 +02:00
ringabout
7c65d9e747 fixes #25204; Uninitialized variable usage in resize__system_u... in @psystem.nim.c in ORC (#25209)
fixes #25204

```nim
  of mUnaryMinusI..mAbsI: unaryArithOverflow(p, e, d, op)
  of mAddI..mPred: binaryArithOverflow(p, e, d, op)
```
Arithmetic operations may raise exceptions. So we cannot entrust the
optimizer to skip `result` initialization in this situation, as
complained righteously by `gcc` and `clang`: `warning: ‘result’ may be
used uninitialized [-Wmaybe-uninitialize]`.

With this PR, `clang -c -Wuninitialized -O1 @psystem.nim.c` no longer
gives warnings
2025-10-08 19:10:09 +02:00
Andreas Rumpf
c4c51d7e78 unittest: show proper stack trace for 'check' (#25212) 2025-10-08 19:09:45 +02:00
Gleb
440b55a44a fix spawn not used on linux (#25206)
Subj, among other things slows down the compilation of large projects on
linux significantly.
2025-10-06 22:22:32 +02:00
ringabout
02609f1872 fixes #25205 #14873; resets importc obj with nimZeroMem in specializeResetT for refc (#25207)
fixes #25205
fixes #14873


```nim
  type
    SysLockObj {.importc: "pthread_mutex_t", pure, final,
               header: """#include <sys/types.h>
                          #include <pthread.h>""", byref.} = object
      when defined(linux) and defined(amd64):
        abi: array[40 div sizeof(clong), clong]
```

Before this PR, in refc, `resetLoc` generates field assignments for each
fields of `importc` object. But the field `abi` is not a genuine field,
which doesn't exits in the struct. We could use `zeroMem` to reset the
memory if not leave it alone
2025-10-06 21:55:31 +02:00
bptato
483389d399 Fix POSIX signal(3) binding's type signature; remove bsd_signal (#24400)
POSIX signal has an identical definition to ISO C signal:
https://pubs.opengroup.org/onlinepubs/9799919799/functions/signal.html

```c
void (*signal(int sig, void (*func)(int)))(int);

/* more readably restated by glibc as */
typedef void (*sighandler_t)(int);

sighandler_t signal(int signum, sighandler_t handler);
```

However, std/posix had omitted the function's return value; this fixes
that.

To prevent breaking every single line of code ever that touched this
binding (including mine...), I've also made it discardable.

Additionally, I have noticed that bsd_signal's type signature is wrong -
it should have been identical to signal. But bsd_signal was already
removed in POSIX 2008, and sigaction is the recommended, portable POSIX
signal interface. So I just deleted the bsd_signal binding.

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-09-28 09:14:56 +02:00
ringabout
cc49bf07fe fixes #21138; closure func used in the loop (#25196)
fixes #21138
2025-09-27 05:54:22 +02:00
J. Neuschäfer
f4497c6158 Improve s390x CPU support (#25056)
TODO list, copied from the documentation:

- [x] compiler/platform.nim Add os/cpu properties.
- [x] lib/system.nim Add os/cpu to the documentation for system.hostOS
and system.hostCPU.
- [x] ~~compiler/options.nim Add special os/cpu property checks in
isDefined.~~ seems unnecessary; isn't dont for most CPUs
- [x] compiler/installer.ini Add os/cpu to Project.Platforms field.
- [x] lib/system/platforms.nim Add os/cpu.
- [x] ~~std/private/osseps.nim Add os specializations.~~
- [x] ~~lib/pure/distros.nim Add os, package handler.~~
- [x] ~~tools/niminst/makefile.nimf Add os/cpu compiler/linker flags.~~
already done in https://github.com/nim-lang/Nim/pull/20943
- [x] tools/niminst/buildsh.nimf Add os/cpu compiler/linker flags.

For csource:

- [x] have compiler/platform.nim updated
- [x] have compiler/installer.ini updated
- [x] have tools/niminst/buildsh.nimf updated
- [x] have tools/niminst/makefile.nimf updated
- [ ] be backported to the Nim version used by the csources
- [ ] the new csources must be pushed
- [ ] the new csources revision must be updated in
config/build_config.txt

Additionally:

- [x] check relation to https://github.com/nim-lang/Nim/pull/20943

Possible future work:

- Porting Nim to s390x-specific operating systems, notably z/OS

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-09-26 17:33:23 +02:00
ringabout
fed0053481 fixes #25167; fixes deref type (#25195)
fixes #25167
2025-09-26 16:12:34 +02:00
Andreas Rumpf
9f74712ec6 fixes #24261 (#25193) 2025-09-24 18:40:43 +02:00
ringabout
3e2852cb1b fixes #21476; internal error: proc has no result symbol (#25192)
fixes #21476
2025-09-24 18:40:32 +02:00
ringabout
ceaa7fb4e8 fixes #23949; cannot return lent expression from conditionals like case (#25190)
fixes #23949

It can also allow  `endsInNoReturn` in branches later
2025-09-24 06:29:57 +02:00
ringabout
d85c0324b7 fixes #25127; disable lent types as object fields in returns (#25189)
fixes #25127
2025-09-23 13:04:27 +02:00
Zoom
6938fce40c stdlib: system: fix incorrect VM detection in substr impls (#25182)
...introduced by me in #24792. Sorry.

This fix doesn't avoid copying the `restrictedBody` twice in the
generated code but has the benefit of working.

Proper fix needs a detection that can set a const bool for a module
once. `when nimvm` is restricted in use and is difficult to dance
around. Some details in: #12517, #12518, #13038

I might have copied the buggy solution from some discussion and it might
have worked at some point, but it's small excuse.
2025-09-22 11:02:50 +02:00
ringabout
e958f4a3cd fixes #24760; Noncopyable base type ignored (#24777)
fixes #24760

I tried `incl` `tfHasAsgn` to nontrivial assignment, but that solution
seems to break too many things. Instead, in this PR, `passCopyToSink`
now checks nontrivial assignment
2025-09-22 10:50:57 +02:00
bptato
3f48576113 Disable strict aliasing on clang (#25067)
Workaround for #24596.

I also took the liberty to disable it on all targets with GCC, since
their documentation claims that it is also enabled on -Os.

---------

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-09-19 08:07:43 +02:00
ringabout
87ee9c84cb makes DuplicateModuleImport back to an error (#25178)
fixes #24998

Basically it retraces back to the situation before
https://github.com/nim-lang/Nim/pull/18366 and
https://github.com/nim-lang/Nim/pull/18362, i.e.

```nim
import fuzz/a
import fuzz/a
```

```nim
import fuzz/a
from buzz/a
```

```nim
import fuzz/a except nil
from fuzz/a import addInt
```

All of these cases are now flagged as invalid and triggers a
redefinition error, i.e., each module name importing is treated as
consistent as the symbol definition


kinda annoying for importing/exporting with `when conditions` though

ref https://github.com/nim-lang/Nim/issues/18762
https://github.com/nim-lang/Nim/issues/20907

```nim
from std/strutils import toLower
when not defined(js):
  from std/strutils import toUpper
```
2025-09-18 20:50:46 +02:00
Andreas Rumpf
16394c3772 fixes #24361 (#25179) 2025-09-18 19:44:39 +02:00
Jacek Sieka
41ce86b577 Remove Nim signal handler for SIGINT (#25169)
Inside a signal handler, you cannot allocate memory because the signal
handler, being implemented with a C
[`signal`](https://en.cppreference.com/w/c/program/signal) call, can be
called _during_ a memory allocation - when that happens, the CTRL-C
handler causes a segfault and/or other inconsistent state.

Similarly, the call can happen from a non-nim thread or inside a C
library function call etc, most of which do not support reentrancy and
therefore cannot be called _from_ a signal handler.

The stack trace facility used in the default handler is unfortunately
beyond fixing without more significant refactoring since it uses
garbage-collected types in its API and implementation.

As an alternative to https://github.com/nim-lang/Nim/pull/25110, this PR
removes the most problematic signal handler, namely the one for SIGINT
(ctrl-c) - SIGINT is special because it's meant to cause a regular
shutdown of the application and crashes during SIGINT handling are both
confusing and, if turned into SIGSEGV, have downstream effects like core
dumps and OS crash reports.

The signal handlers for the various crash scenarios remain as-is - they
may too cause their own crashes but we're already going down in a bad
way, so the harm is more limited - in particular, crashing during a
crash handler corrupts `core`/crash dumps. Users wanting to keep their
core files pristine should continue to use `-d:noSignalHandler` - this
is usually the better option for production applications since they
carry more detail than the Nim stack trace that gets printed.

Finally, the example of a ctrl-c handler performs the same mistake of
calling `echo` which is not well-defined - replace it with an example
that is mostly correct (except maybe for the lack of `volatile` for the
`stop` variable).
2025-09-17 10:58:21 +02:00
ringabout
51a9ada043 fixes #25173; SinglyLinkedList.remove broken / AssertionDefect (#25175)
fixes #25173
2025-09-16 17:05:09 +02:00
Jacek Sieka
40fe59b6ef remove alloc cruft (#25170) 2025-09-15 15:09:42 +02:00
Jacek Sieka
8b9972c8b6 orc: fix overflow checking regression (#25089)
Raising exceptions halfway through a memory allocation is undefined
behavior since exceptions themselves require multiple allocations and
the allocator functions are not reentrant.

It is of course also expensive performance-wise to introduce lots of
exception-raising code everywhere since it breaks many optimisations and
bloats the code.

Finally, performing pointer arithmetic with signed integers is incorrect
for example on on a 32-bit systems that allows up to 3gb of address
space for applications (large address extensions) and unnecessary
elsewhere - broadly, stuff inside the memory allocator is generated by
the compiler or controlled by the standard library meaning that
applications should not be forced to pay this price.

If we wanted to check for overflow, the right way would be in the
initial allocation location where both the size and count of objects is
known.

The code is updated to use the same arithmetic operator style as for
refc with unchecked operations rather than disabling overflow checking
wholesale in the allocator module - there are reasons for both, but
going with the existing flow seems like an easier place to start.
2025-09-15 15:08:21 +02:00
ringabout
cdb750c962 minor improvements of error messages of objvariants (#25040)
Because `prevFields` and `currentFields` have been already quoted by
`'`, no need to add another.

The error message was

```
The fields ''x'' and ''y'' cannot be initialized together, because they are from conflicting branches in the case object.
```
2025-09-15 15:04:35 +02:00
Miran
c49fb5ac5f replace outdated macos-13 runner (#25155) 2025-09-15 15:03:59 +02:00
ringabout
a77d1cc6c1 fixes #24844; Invalid C codegen refc with generic types containing gc memory (#25160)
fixes #24844


it may not be used in other places except in `genTraverseProc`,
we have to generate a `typedesc` for this case, not a weak `typedec`
2025-09-15 15:03:22 +02:00
lit
ff9cae896c fixes #25162; fixup 0f5732bc8c: withValue for immut tab wrong chk cond (#25163)
fixes #25162
ref https://github.com/nim-lang/Nim/pull/24825
2025-09-12 14:07:05 +02:00
ringabout
bf2395a62e disable thttpclient_ssl (#25164) 2025-09-12 14:06:29 +02:00
bptato
d60e0211bc Fix nimIoselector define in std/selectors (#25104)
Also added some documentation to the header.

See: https://forum.nim-lang.org/t/13311

> I did try using the flag, but couldn't get it to work. If I do
-d:nimIoSelector, the defined check passes, but the other code fails to
compile because there is no const named nimIoSelector. It seemed like a
bug to me, do you have a working number compiler invocation?

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-09-11 23:45:47 +02:00
Ryan McConnell
88da5e8cee two small concept patches (#25076)
- slightly better typeclass logic (eg for bare `range`)
- reverse matching now substitutes potential implementation for `Self`

---------

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-09-11 20:50:11 +08:00
bptato
d73f478bdc Allow assignment of nested non-closure procs to globals (#25154)
For memory-safety, this only seems problematic in case of closures, so I
just special cased that.

Fixes #25131
2025-09-11 09:22:47 +02:00
ringabout
f90951cc61 move std/parsesql to nimble packages (#25156)
pending https://github.com/nim-lang/packages/pull/3117

ref https://github.com/nim-lang/parsesql
2025-09-11 08:47:02 +02:00
Andreas Rumpf
af6be4f839 GDB script: minor improvements (#24965) 2025-09-10 15:38:25 +02:00
Yuriy Glukhov
49e66e80f0 Optimize @, fixes #25063 (#25064)
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-09-10 15:37:55 +02:00
Judd
4f09675d8a Update asyncfile.nim: support write to > 2GB file on Windows (#25105)
`DWORD` is defined as `int32`, so `DWORD(...)` would not work as
expected. When writing to files larger than 2GB, exception occurs:

```
unhandled exception: value out of range: 4294967295 notin -2147483648 .. 2147483647 [RangeDefect]
```

This PR is a quick fix for this.

P.S. Why `DWORD` is defined as `int32`?
2025-09-10 15:37:09 +02:00
ringabout
76d07e8caa fixes #25078; filterIt wrongly results in rvalue (#25139)
fixes #25078
2025-09-10 15:36:39 +02:00
Jacek Sieka
5ba279276e sequtils: findIt (#25134)
Complements `anyIt`, `find`, etc, plugging an odd gap in the `xxxIt`
family of functions
2025-09-09 20:05:12 +02:00
ringabout
c8456eacd5 fixes #25117; requiresInit not checked for result if it has been used (#25151)
fixes #25117


errors on `requiresInit` of `result` if it is used before
initialization. Otherwise

```nim
    # prevent superfluous warnings about the same variable:
    a.init.add s.id
```

It produces a warning, and this line prevents it from being recognized
by the `requiresInit` check in `trackProc`
2025-09-09 16:22:05 +02:00
ringabout
34bb37ddda fixes #25120; don't generate hooks for NimNode (#25144)
fixes #25120
2025-09-09 16:17:22 +02:00
ringabout
08d74a1c27 fixes #24093; Dereferencing result of cast in single expression triggers unnecessary copy (#25143)
fixes #24093

transforms
```nim
let a = new array[1000, byte]
block:
  for _ in cast[typeof(a)](a)[]:
    discard
```
into 
```nim
let a = new array[1000, byte]
block:
  let temp = cast[typeof(a)](a)
  for _ in temp[]:
    discard
```
So it keeps the same behavior with the manual version
2025-09-09 16:16:12 +02:00
Tomohiro
8ea8755cc0 fixes tnewruntime_strutils.nim not to raise AssertionDefect (#25142)
Follow up to https://github.com/nim-lang/Nim/pull/25126
It changed `formatSize` outputs from some inputs, so some of existing
test code related to it need to be updated.
Sorry, I didn't know `tests/destructor/tnewruntime_strutils.nim` has
tests calls `formatSize`.
2025-09-04 06:46:50 +02:00
Tomohiro
87dc1820c0 fixes overflow defect when compiled with js backend (#25132)
Follow up to https://github.com/nim-lang/Nim/pull/25126.
This fixes overflow defect when `tests/stdlib/tstrutils.nim` was
compiled with js backend.
2025-09-01 18:29:58 +02:00
Tomohiro
065c4b443b fixes #25125 (#25126)
`strutils.formatSize` returns correct strings from large values close to
`int64.high`.
Round down `bytes` when it is converted to float.
2025-08-28 21:56:46 +02:00
ringabout
0a8f618e2b fixes #25121; [FieldDefect] with iterator-loop (#25130)
fixes #25121
2025-08-28 21:51:45 +02:00
Andreas Rumpf
d472022a77 fixes #25114 (#25124) 2025-08-27 12:23:04 +02:00
ringabout
e2a294504e fixes #25066; forbids comparing pointers at compile time (#25103)
fixes #25066

Probably it is not worth implementing comparing pointers at compile
time. For a starter, we can improve the error message instead of letting
it crash
2025-08-27 10:14:46 +02:00
ringabout
b527db9ddd fixes #25109; fixes #25111 transform addr(conv(x)) -> conv(addr(x)) (#25112)
follows up https://github.com/nim-lang/Nim/pull/24818
relates to https://github.com/nim-lang/Nim/issues/23923

fixes #25109
fixes #25111

transform `addr ( conv ( x ) )` -> `conv ( addr ( x ) )` so that it is
the original value that is being modified


```c
T1_ = ((unsigned long long*) ((&a_1)));
r(T1_);
```
2025-08-21 13:31:55 +02:00
RAMLAH MUNIR
c6352ce0ab closes #25084 : docs: fix example for *+ operator (#25102)
## Description

Fixed an inconsistency in the Nim manual's example for the `*+`
operator.

Previously, the example on line 4065 of `doc/manual.md` used variables
`a`, `b`, and `c`:

```nim
assert `*+`(3, 4, 6) == `+`(`*`(a, b), c)
```

This did not match the preceding call which directly used literals `3`,
`4`, `6`.

Updated the example to:

```nim
assert `*+`(3, 4, 6) == `+`(`*`(3, 4), 6)
```

This change makes the example consistent with the function call and
immediately understandable to readers without requiring prior variable
definitions.

## Rationale

* Improves clarity by avoiding undefined variables in a code snippet.
* Matches the example usage in the preceding line.
* Helps beginners understand the operator's behavior without additional
context.

## Changes

* **Edited**: `doc/manual.md` line 4065 — replaced variables `a`, `b`,
`c` with literals `3`, `4`, `6`.

## Issue

Closes #25084
2025-08-14 16:33:52 +02:00
Laylie
53bb0b591a Link to nims docs from nimc docs (#25095) 2025-08-10 11:38:39 +02:00
ringabout
a0b3048f3f Revert "std/locks use header files instead of dlls on windows" (#25091)
Reverts nim-lang/Nim#25090


It seems to cause problems for C++ and i686


```
2025-08-08T02:37:55.5976232Z c:/a/nightlies/nightlies/external/mingw32/bin/../lib/gcc/i686-w64-mingw32/11.1.0/../../../../i686-w64-mingw32/bin/ld.exe: C:\Users\runneradmin\nimcache\manual_experimental_snippet_106_d\@pstd@sprivate@ssyslocks.nim.c.o:@pstd@sprivate@ssyslocks.nim.c:(.text+0x29): undefined reference to `SleepConditionVariableCS'
2025-08-08T02:37:55.5978066Z c:/a/nightlies/nightlies/external/mingw32/bin/../lib/gcc/i686-w64-mingw32/11.1.0/../../../../i686-w64-mingw32/bin/ld.exe: C:\Users\runneradmin\nimcache\manual_experimental_snippet_106_d\@pthreadpool.nim.c.o:@pthreadpool.nim.c:(.text+0x26): undefined reference to `InitializeConditionVariable'
2025-08-08T02:37:55.5980101Z c:/a/nightlies/nightlies/external/mingw32/bin/../lib/gcc/i686-w64-mingw32/11.1.0/../../../../i686-w64-mingw32/bin/ld.exe: C:\Users\runneradmin\nimcache\manual_experimental_snippet_106_d\@pthreadpool.nim.c.o:@pthreadpool.nim.c:(.text+0x116): undefined reference to `WakeConditionVariable'
2025-08-08T02:37:55.5981093Z collect2.exe: error: ld returned 1 exit status
2025-08-08T02:37:55.5988564Z Error: execution of an external program failed: 'gcc.exe   -o 
```
2025-08-08 22:22:23 +08:00
ringabout
02e3487c9c std/locks use header files instead of dlls on windows (#25090)
ref https://github.com/nim-lang/nimony/pull/1370
2025-08-07 17:45:00 +02:00
ringabout
e194c7cc87 adds more functions to to dirs and files (#25083)
ref https://forum.nim-lang.org/t/13272
2025-08-01 12:34:01 +02:00
Yuriy Glukhov
161b321796 SOCKS5H support for httpclient (#25070)
- Added support for SOCKS5h (h for proxy-side DNS resolving) to
httpclient
- Deprecated `auth` arguments for `newProxy` constructors, for auth to
be embedded in the url.

Unfortunately `http://example.com` is not currently reachable from
github CI, so the tests fail there for a few days already, I'm not sure
what can be done here.
2025-07-30 00:14:41 +02:00
Yuriy Glukhov
9b527a51b8 Fixed typos in comments (#25071) 2025-07-23 23:50:03 +02:00
Emre Şafak
bb93b39b58 docs: Add example to tutorial for interfaces using closures (#25068)
* Add a new section to doc/tut2.md explaining interfaces.
* Provide a code example demonstrating how to simulate interfaces using
objects of closures.
* The example shows a basic IntFieldInterface with getter and setter
procedures.

This PR was inspired by the discussion in
https://forum.nim-lang.org/t/13217

---------

Co-authored-by: Emre Şafak <esafak@users.noreply.github.com>
Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-07-23 23:49:31 +02:00
Juan M Gómez
cd806f9dbe Bumps nimble 0.20.1 (#25062) 2025-07-19 08:16:51 +02:00
ringabout
08d51e5c88 fixes #7179; Floats are not range checked (#25050)
fixes #7179


```nim
var f = 751.0
echo f.int8
```

In this case, `int8(float)` yields different numbers for different
optimization levels, since float to int conversions are undefined
behaviors. In this PR, it mitigates this problem by conversions to same
size integers before converting to the final type: i.e.
`int8(int64(float))`, which has UB problems but is better than before
2025-07-18 21:30:50 +02:00
ringabout
5b5cd7fa67 fixes inefficient codegen for field return (#24874)
fixes https://github.com/nim-lang/Nim/issues/23395
fixes https://github.com/nim-lang/Nim/issues/23395
2025-07-18 21:30:32 +02:00
ringabout
8e57a9f623 fixes #24719; improves order of destruction (#25060)
fixes #24719
2025-07-18 23:02:16 +08:00
Nikolay Nikolov
478773ffb1 NimSuggest: Fix for the inlay exception hints with generic procs (#23610)
Based on the fix, started by SirOlaf in #23414

---------

Co-authored-by: SirOlaf <>
Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-07-18 08:44:36 +02:00
ringabout
f4ebabb9b3 fixes CI failures (#25058) 2025-07-17 19:32:41 +08:00
ringabout
611b8bbf67 fixes #25007; implements setLenUninit for refc (#25022)
fixes #25007

```nim
proc setLengthSeqUninit(s: PGenericSeq, typ: PNimType, newLen: int, isTrivial: bool): PGenericSeq {.
    compilerRtl.} =
```

In this added function, only the line `zeroMem(dataPointer(result,
elemAlign, elemSize, newLen), (result.len-%newLen) *% elemSize)` is
removed from `proc setLengthSeqV2` when enlarging a sequence.

JS and VM versions simply use `setLen`.
2025-07-14 23:19:58 +02:00
lit
7e2df41850 fixes #25043: js tyUserTypeClass internal error (#25044)
- **fixes #25043: `internal error: genTypeInfo(tyUserTypeClassInst)`**
- **chore(test): for 25043**
2025-07-14 23:15:44 +02:00
Emre Şafak
9c1e3bf8fb Improve error message for keywords as parameters (#25052)
A function with an illegal parameter name like
```nim
proc myproc(type: int) =
  echo type
```
would uninformatively fail like so:
```nim
tkeywordparam.nim(1, 13) Error: expected closing ')'
```

This commit makes it return the following error:
```nim
tkeywordparam.nim(1, 13) Error: 'type' is a keyword and cannot be used as a parameter name
```

---------

Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
Co-authored-by: Emre Şafak <esafak@users.noreply.github.com>
Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-07-14 23:15:02 +02:00
Slava Vishnyakov
30d4f7791d Create Mac app bundle for GUI apps on macOS when --app:gui is used (#25042)
Fixes https://github.com/nim-lang/Nim/issues/25041

Basically it creates a "real" console-less app when --app:gui is used.
Otherwise a console window opens, see the bug.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-07-14 23:14:06 +02:00
Yuriy Glukhov
6ab532fd0f Fixes #25038 (#25039) 2025-07-13 07:56:20 +02:00
Esteban C Borsani
08642ffe34 revert #24896; asyncnet ssl overhaul (#25033)
revert #24896

Partially reverting #24896 in #25024 broke CI. So better revert it
completely so the CI is green. I'll investigate the issue later.
2025-07-10 15:31:56 +02:00
Juan M Gómez
370ee61f6d Updates nimble commit (#25036) 2025-07-08 15:46:13 +02:00
Yuriy Glukhov
36f8cefa85 Fixes #21235, #23602, #24978, #25018 (#25030)
Reworked closureiter transformation.

- Convolutedly nested finallies should cause no problems now.
- CurrentException state now follows nim runtime rules (pushes and pops
appropriately), and mimics normal code, which is somewhat buggy, see
#25031
- Previously state optimization (removing empty states or extra jumps)
missed some opportunities, I've reimplemented it to do everything
possible to optimize the states. At this point any extra states or jumps
should be considered a bug.

The resulting codegen (compiled binaries) is also slightly smaller.


**BUT:**
- I had to change C++ reraising logic, see expt.nim. Because with
closure iters `currentException` is not always in sync with C++'s notion
of current exception. From my tests and understanding of C++ runtime
there should not be any problems, but I'm only 99% sure :)
- I've reused `nfNoRewrite` flag in one specific case during the
transformation. This flag is also used in term-rewriting logic. Again,
99% sure, these 2 scenarios will never intersect.
2025-07-08 15:41:17 +02:00
Esteban C Borsani
fbdc9a4c19 fixes #25023; Asyncnet accept leaks socket on SSL error; Regression in devel (#25024)
Fixes #25023

Revert the acceptAddr #24896 change. SSL_accept is no longer explicitly
called.
2025-07-01 09:52:37 +02:00
Zoom
6bdb069a66 [docs]: warning for long, culong being OS-dependent (#25012)
Docs are routinely compiled on a different OS so often don't reflect
reality of CT-conditionals.

I bet there's a few of other places like this in the stdlib.
2025-06-27 10:49:02 +02:00
ringabout
7e6fa9e2d6 fixes #23564; hasCustomPragma skips alises types (#24994)
fixes #23564

perhaps handle generic aliases (tyGenericInst for aliases types) if
needed
2025-06-27 10:18:12 +02:00
ringabout
3ce38f2959 fixes #24997; {.global.} variable in recursive function (#25016)
fixes #24997

handles functions in recursive order
2025-06-27 10:17:16 +02:00
bptato
b6491e7de5 Add missing error handling in getAppFilename (#25017)
readlink can return -1, e.g. if procfs isn't mounted in a Linux chroot.
(At least that's how I found this.)
2025-06-25 23:21:56 +02:00
metagn
97a6f42b56 fix generic converter subtype match regression (#25015)
fixes #25014

`implicitConv` tries to instantiate the supertype to convert to,
previously the bindings of `m` was shared with the bindings of the
converter but now an isolated match `convMatch` holds the bindings, so
`convMatch` is now used in the call to `implicitConv` instead of `m` so
that its bindings are used when instantiating the supertype.
2025-06-25 21:42:26 +08:00
metagn
334848f3ae fix regression with enum types wrongly matching [backport:2.2] (#25010)
fixes #25009

Introduced by #24176, when matching a set type to another, if the given
set is a constructor and the element types match worse than a generic
match (which includes the case with no match), the match is always set
to a convertible match, without checking that it is at least a
convertible match. This is fixed by checking this.
2025-06-21 23:25:04 +02:00
Jacek Sieka
aba9361510 Ensure that gc interface remains non-raising (#25006)
GC_fullCollect in particular has an annoying `Exception` effect
2025-06-18 14:38:01 +02:00
ringabout
c22bfe6bc0 fixes #24996; Crash on marking destroy hook as .error (#25002)
fixes #24996

uses the lineinfos of `dest` is `ri` is not available (e.g. `=destroy`
doesn't have a second parameter)
2025-06-16 21:50:06 +02:00
metagn
8e5ed5dbb7 loosen compiler assert for ident node in dotcall matching [backport:2.2] (#25003)
fixes #25000

A failed match on `nfDotField` tries to assert that the name of the dot
field is an identifier node. I am not exactly sure how but at some point
typed generics causes an `nfDotField` call to contain a symchoice for
the field name. The compiler does not use the fact that the field name
is an identifier, so the assert is loosened to allow any identifier-like
node kind. Could also investigate why the symchoice gets created, my
guess is that typed generics detects that the match fails but still
sends it through generic prechecking and doesn't remove the
`nfDotField`, which is harmless and it might cause more trouble to work
around it.
2025-06-16 19:22:23 +02:00
metagn
7701b3c7e6 don't set sym of generic param type value to generic param sym (#24995)
fixes #23713

`linkTo` normally sets the sym of the type as well as the type of the
sym, but this is not wanted for custom pragmas as it would look up the
definition of the generic param and not the definition of its value. I
don't see a practical use for this either.
2025-06-13 00:03:02 +02:00
ringabout
151b903172 closes #24992; adds a test case (#24993)
closes #24992
2025-06-12 21:47:41 +08:00
metagn
56bb451c6d use windows latest for docs CI (#24991)
2019 is currently browned out
2025-06-11 19:45:11 +08:00
ringabout
638a8bf84d fixes #24974; SIGSEGV when raising Defect/doAssert (#24985)
fixes #24974

requires `result` initializations when encountering unreachable code
(e.g. `quit`)
2025-06-10 14:12:12 +02:00
ringabout
ffb993d5bd fixes #24981; the length of the seq changed of procGloals (#24984)
fxies #24981

`m.g.graph.procGlobals` could change because the right side of `.global`
assignment (e.g. `let a {.global.} = g(T)`) may trigger injections for
unhandled procs
2025-06-09 20:53:40 +02:00
Andreas Rumpf
d27c2159fa make 'nim nif' atomic (#24980) 2025-06-08 19:03:46 +02:00
Andreas Rumpf
4fbf538b66 nifgen: bugfix (#24979) 2025-06-06 15:36:31 +02:00
Amjad Ben Hedhili
9d0c0b89f2 [Docs] Improve scrollbars (#24971)
Follow dark/light modes.
2025-06-06 15:08:47 +02:00
Eugene Kabanov
7a53db6874 Fix FreeBSD getThreadId() should use different syscall definition for 64bit platforms. (#24977) 2025-06-05 14:30:07 +02:00
Andreas Rumpf
dd7cecdbd4 make mangled module names shorter (#24976) 2025-06-05 08:34:52 +02:00
Alfred Morgan
807840dc5b Patch 24922 (#24972)
fixed #24922 missing possix_fallocate for openbsd
2025-05-31 15:26:41 +02:00
Amjad Ben Hedhili
f80a076588 Fix docs sidebar truncated (#24970)
* Regression after #24927
2025-05-31 15:26:04 +02:00
ringabout
db5973361f bundles nimony (#24968)
follow up https://github.com/nim-lang/Nim/pull/24966
https://github.com/nim-lang/Nim/pull/21702

So it appears in the nightlies tars and can be installed by choosenim
2025-05-27 11:45:28 +02:00
Andreas Rumpf
241edaf0c0 WIP: adds NIF code generator for interop with Nimony (#24966) 2025-05-26 15:42:19 +02:00
metagn
8752392838 implement setter fallback for subscripts (#24872)
follows up #24871

For subscript assignments, if an overload of `[]=`/`{}=` is not found,
the LHS checks for overloads of `[]`/`{}` as a fallback, similar to what
field setters do since #24871. This is accomplished by just compiling
the LHS if the assignment overloads fail. This has the side effect that
the error messages are different now, instead of displaying the
overloads of `[]=`/`{}=` that did not match, it will display the ones
for `[]`/`{}` instead. This could be fixed by checking for `efLValue`
when giving the error messages for `[]`/`{}` but this is not done here.

The code for `[]` subscripts is a little different because of the
`mArrGet`/`mArrPut` overloads that always match. If the `mArrPut`
overload matches without a builtin subscript behavior for the LHS then
it calls `semAsgn` again with `mode = noOverloadedSubscript`. Before
this meant "fail to compile" but now it means "try to compile the LHS as
normal", in both cases the overloads of `[]=` are not considered again.
2025-05-23 16:19:13 +02:00
ringabout
a09da96c65 fixes #4594; disallow {.global.} uses local vars for basic expressions (#24961)
fixes #4594
2025-05-23 16:16:57 +02:00
ringabout
3c0446b082 fixes #24940; fixes #17552; lifts {.global.} in injectDestructorCalls (#24962)
fixes #24940
fixes #17552

Collects `{.global.}` (i.e. if it was changed into a hook call: `=copy`,
`=sink`) in `injectDestructorCalls` and generates it in the init
sections in cgen
2025-05-23 16:15:55 +02:00
ringabout
c3f64fb127 rework nimOrcLeakDetector (#24958)
ref https://github.com/nim-lang/Nim/issues/22273#issuecomment-2888931920
2025-05-19 21:40:35 +02:00
metagn
e855019f84 add STRING_LITERAL macro back to nimbase.h for compatibility (#24957)
refs #24956, refs #24302
2025-05-17 18:37:02 +02:00
Andreas Rumpf
1e602490e9 fixes #4851 [backport] (#24954) 2025-05-16 09:44:13 +02:00
metagn
71c5a4f72c generate let _ = to fully unpack partial tuple unpacking assignment for arc (#24948)
fixes #24947

When injectdestructors detects that a variable is a tuple unpacking temp
(i.e. it is an `skTemp`, is not a cursor, and has tuple type) it does
not generate a destructor for it and only generates sink/bit assignments
for its components. However the reason it does not generate a destructor
is that it expects it to be fully unpacked, this is true for unpackings
in for loops but not for tuple unpacking assignments which supports `_`
since #22537. Tuple unpacking definitions for `var`/`let`/`const` do not
generate `skTemp` and use the same symbol kind as the definition so they
did not have this problem.

To keep this compatible, the `_` parts of the tuple unpacking
assignments are now not ignored and unpacked into `let _ = ...`, which
generates its own destructor. Another option might be to use `skLet`
instead of `skTemp` but this might cause changes to behavior like
additional copies, I am not sure about this though.
2025-05-15 09:32:10 +02:00
ringabout
ade500b2cb adds nimPreviewCStringComparisons for cstring comparisons (#24946)
todo: We can also give a deprecation message for `ltPtr`/`lePtr`
matching for cstring in `magicsAfterOverloadResolution`

follow up https://github.com/nim-lang/Nim/pull/24942
2025-05-14 21:31:53 +02:00
Niklas Kröger
c1e6cf812f Fix extra newline from nimpretty when used with --stdin (#24951)
Using `echo` to print file contents to stdout automatically adds a
newline at the end of the file contents. When using nimpretty to auto
format files on save in some editors which replace the file contents
with the formatted ones this means that with every save/format operation
an additional newline is added to the end of the file. Using
`stdout.write` does not automatically add a newline at the end
preventing this issue.

Fixes #24950
2025-05-14 21:31:09 +02:00
c-blake
091fb5057b Maybe close https://github.com/nim-lang/Nim/issues/24932 by simply (#24945)
explaining why the result may not be so surprising. Clean-up of stray
whitespace and insert of missing "in" along for the ride.

It's just not always faster or slower than `Table`. The difference
depends upon many factors such as (at least!): A) how much (if anything
- for `int` keys it is nothing) hash-comparison before `==` comparison
saves B) how much resizing happens (which may even vary from run to run
if end users are allowed to provide scale guess input), C) how much
comparison happens at all (i.e., table density), D) how much space/size
matters - like how close to a specific deployment "available" cache size
the table is.

If we want, we could add a sentence suggesting performance fans also try
`Table`, but the kind of low-level nature of the explanation strikes me
as already along those lines.
2025-05-11 06:44:03 +02:00
ringabout
d2fee7dbab fixes broken discriminators of float types by disabling it (#24938)
```nim
type
  Case = object
    case x: float
    of 1.0:
      id: int
    else:
      ta: float
```

It segfaults with `fatal error: invalid kind for firstOrd(tyFloat)`

It was caused by https://github.com/nim-lang/Nim/pull/12591 and has
affected discriminators of float types since 1.2.x

I think no one is using discriminators of float types anyway so I simply
disable it like what was done to discriminators of string types (ref
https://github.com/nim-lang/Nim/pull/15080)


ref https://github.com/nim-lang/nimony/pull/1069
2025-05-11 06:42:27 +02:00
Juan M Gómez
8080610248 Initial implementation for nimsuggest import support (#24937)
Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-05-11 06:41:09 +02:00
ringabout
6c2f78a19f rework tags (#24944)
recent ctags changes: https://github.com/nim-lang/Nim/pull/24317
ref https://forum.nim-lang.org/t/12879
2025-05-11 06:40:46 +02:00
bptato
6f5e5811fc Correct nfds_t size on Android (#24647)
Turns out bionic uses an unsigned int (unlike other Linux libcs).

(See
<https://android.googlesource.com/platform/bionic/+/master/libc/include/poll.h>.)
2025-05-10 13:26:00 +02:00
ringabout
42a4adb4a5 fixes #24941; missing < (less than), cmp for cstring (#24942)
fixes #24941

now `cmp` can select the correct version of cstring comparsions
2025-05-10 08:26:21 +02:00
Amjad Ben Hedhili
59ceff4f1a Add min/max overloads with comparison functions (#23595)
`min`, `max`, `minmax`, `minIndex` and `maxIndex`
2025-05-06 14:09:03 +02:00
ringabout
b50ab7a5c9 improvements for semdata (#24933) 2025-05-06 09:46:45 +02:00
ringabout
433b725cbb fixes #21975; Pragma block disabling warning has effect beyond block (#24934)
fixes  #21975
2025-05-06 09:46:18 +02:00
metagn
82553384d1 bring back id table algorithm instead of std table [backport:2.2] (#24930)
refs #24929, partially reverts #23403

Instead of using `Table[ItemId, T]`, the old algorithm is brought back
into `TIdTable[T]` to prevent a performance regression. The inheritance
removal from #23403 still holds, only `ItemId`s are stored.
2025-05-06 09:36:20 +02:00
Amjad Ben Hedhili
8b82f5de38 Remove horizontal scrolling on mobile (#24927) 2025-05-05 08:17:36 +02:00
ringabout
98ec87d65e fixes #23355; pop optionStack when exiting scopes (#24926)
fixes #23355
2025-05-04 03:29:59 +02:00
ringabout
f56568d851 fixes address of sink parameters (#24924)
In `semExprWithType`: `if result.typ.kind in {tyVar, tyLent}: result =
newDeref(result)` derefed `var`/`lent`. Since it is not done for `sink`,
we need to skip `tySink` in the corresponding procs
2025-05-01 07:49:46 +02:00
Ryan McConnell
b5b7a127fd Fix warning[Uninit] triggers in strutils (#24921) 2025-04-30 23:17:11 +08:00
Alfred Morgan
b61a614e8a Patch 24922 (#24923) 2025-04-30 13:00:23 +02:00
ringabout
0506d5b973 don't warn/error symbols in semGenericStmt/templates (#24907)
fixes #24905
fixes #24903
fixes https://github.com/nim-lang/Nim/issues/11805
fixes https://github.com/nim-lang/Nim/issues/15650

In the first phase of generic checking, we cannot warn/error symbols
because they can belong a false branch of `when` or there is a
`push/pop` options using open symbols. So we cannot decide whether to
warn/error or not
2025-04-29 11:08:10 +02:00
Esteban C Borsani
8518cf079f asyncnet ssl overhaul (#24896)
Fixes #24895

- Remove all  bio handling
- Remove all `sendPendingSslData` which only seems to make things work
by chance
- Wrap the client socket on `acceptAddr` (std/net does this)
- Do the SSL handshake on accept (std/net does this)

The only concern is if addWrite/addRead works well on Windows.
2025-04-29 11:07:01 +02:00
lit
d7b1f0a99a fix(js): nonvar destructor was disallowed; closes #24914 (#24915) 2025-04-29 06:45:20 +02:00
Tomohiro
eea4ce0e2c changes FileHandle type on Windows (#24910)
On windows, `HANDLE` type values are converted to `syncio.FileHandle` in
`lib/std/syncio.nim`, `lib/pure/memfiles.nim` and `lib/pure/osproc.nim`.
`HANDLE` type is `void *` on Windows and its size is larger then `cint`.

https://learn.microsoft.com/en-us/windows/win32/winprog/windows-data-types

This PR change `syncio.FileHandle` type so that converting `HANDLE` type
to `syncio.FileHandle` doesn't lose bits.

We can keep `FileHandle` unchanged and change some of parameter/return
type from `FileHandle` to an type same size to `HANDLE`, but it is
breaking change.
2025-04-28 10:43:53 +02:00
metagn
8c9a645bdf fix generic converter regression with var/subtype args (#24902)
refs #24867,
https://github.com/nim-lang/Nim/pull/24867#issuecomment-2821315971

The argument node of the converter can be wrapped in [hidden `addr` or
subtype conversion
nodes](dc100c5caa/compiler/sigmatch.nim (L2327-L2335))
which have to be skipped when matching the type again, since the type of
the node is the uninstantiated type taken from the proc parameter.
2025-04-24 21:18:18 +02:00
Ryan McConnell
5dcfd8d7bb Add tySet to concept matching (#24908) 2025-04-24 21:17:42 +02:00
metagn
d966ee3fc3 whitelist prev types to reuse in newOrPrevType (#24899)
fixes #24898

A type is only overwritten if it is definitely a forward type, partial
object (symbol marked `sfForward`) or a magic type. Maybe worse for
performance but should be more correct. Another option might be to
provide a different value for `prev` for the `preserveSym` case but then
we cannot easily ignore only nominal type nodes.
2025-04-22 17:24:22 +02:00
metagn
dc100c5caa update proc type recursion errors after merge (#24897)
refs #24893, refs #24888
2025-04-21 19:41:09 +03:00
metagn
7f0e07492f generally disallow recursive structural types, check proc param types (#24893)
fixes #5631, fixes #8938, fixes #18855, fixes #19271, fixes #23885,
fixes #24877

`isTupleRecursive`, previously only called to give an error for illegal
recursions for:

* tuple fields
* types declared in type sections
* explicitly instantiated generic types

did not check for recursions in proc types. It now does, meaning proc
types now need a nominal type layer to recurse over themselves. It is
renamed to `isRecursiveStructuralType` to better reflect what it does,
it is different from a recursive type that cannot exist due to a lack of
pointer indirection which is possible for nominal types.

It is now also called to check the param/return types of procs, similar
to how tuple field types are checked. Pointer indirection checks are not
needed since procs are pointers.

I wondered if this would lead to a slowdown in the compiler but since it
only skips structural types it shouldn't take too many iterations, not
to mention only proc types are newly considered and aren't that common.
But maybe something in the implementation could be inefficient, like the
cycle detector using an IntSet.

Note: The name `isRecursiveStructuralType` is not exactly correct
because it still checks for `distinct` types. If it didn't, then the
compiler would accept this:

```nim
type
  A = distinct B
  B = ref A
```

But this breaks when attempting to write `var x: A`. However this is not
the case for:

```nim
type
  A = object
    x: B
  B = ref A
```

So a better description would be "types that are structural on the
backend".

A future step to deal with #14015 and #23224 might be to check the
arguments of `tyGenericInst` as well but I don't know if this makes
perfect sense.
2025-04-21 09:01:44 +02:00
metagn
9c2593444a consider proc return type as weak reference in codegen (#24894)
fixes #7706
2025-04-21 08:58:45 +02:00
metagn
525d64fe88 leave type section symbols unchanged on resem, fix overly general double semcheck for forward types (#24888)
fixes #24887 (really just this [1 line
commit](632c7b3397)
would have been enough to fix the issue but it would ignore the general
problem)

When a type definition is encountered where the symbol already has a
type (not a forward type), the type is left alone (not reset to
`tyForward`) and the RHS is handled differently: The RHS is still
semchecked, but the type of the symbol is not updated, and nominal type
nodes are ignored entirely (specifically if they are the same kind as
the symbol's existing type but this restriction is not really needed).
If the existing type of the symbol is an enum and and the RHS has a
nominal enum type node, the enum fields of the existing type are added
to scope rather than creating a new type from the RHS and adding its
symbols instead.

The goal is to prevent any incompatible nominal types from being
generated during resem as in #24887. But it also restricts what macros
can do if they generate type section AST, for example if we have:

```nim
type Foo = int
```

and a macro modifies the type section while keeping the symbol node for
`Foo` like:

```nim
type Foo = float
```

Then the type of `Foo` will still remain `int`, while it previously
became `float`. While we could maybe allow this and make it so only
nominal types cannot be changed, it gets even more complex when
considering generic params and whether or not they get updated. So to
keep it as simple as possible the rule is that the symbol type does not
change, but maybe this behavior was useful for macros.

Only nominal type nodes are ignored for semchecking on the RHS, so that
cases like this do not cause a regression:

```nim
template foo(): untyped =
  proc bar() {.inject.} = discard
  int

type Foo = foo()
bar() # normally works
```

However this specific code exposed a problem with forward type handling:

---

In specific cases, when the type section is undergoing the final pass,
if the type fits some overly general criteria (it is not an object,
enum, alias or a sink type and its node is not a nominal type node), the
entire RHS is semchecked for a 2nd time as a standalone type (with `nil`
prev) and *maybe* reassigned to the new semchecked type, depending on
its type kind. (for some reason including nominal types when we excluded
them before?) This causes a redefinition error if the RHS defines a
symbol.

This code goes all the way back to the first commit and I could not find
the reason why it was there, but removing it showed a failure in
`thard_tyforward`: If a generic forward type is invoked, it is left as
an unresolved `tyGenericInvocation` on the first run. Semchecking it
again at the end turns it into a `tyGenericInst`. So my understanding is
that it exists to handle these loose forward types, but it is way too
general and there is a similar mechanism `c.skipTypes` which is supposed
to do the same thing but doesn't.

So this is no longer done, and `c.skipTypes` is revamped (and renamed):
It is now a list of types and the nodes that are supposed to evaluate to
them, such that types needing to be updated later due to containing
forward types are added to it along with their nodes. When finishing the
type section, these types are reassigned to the semchecked value of
their nodes so that the forward types in them are fully resolved. The
"reassigning" here works due to updating the data inside the type
pointer directly, and is how forward types work by themselves normally
(`tyForward` types are modified in place as `s.typ`).

For example, as mentioned before, generic invocations of forward types
are first created as `tyGenericInvocation` and need to become
`tyGenericInst` later. So they are now added to this list along with
their node. Object types with forward types as their base types also
need to be updated later to check that the base type is correct/inherit
fields from it: For this the entire object type and its node are added
to the list. Similarly, any case where whether a component type is
`tyGenericInst` or `tyGenericInvocation` matters also needs to cascade
this (`set` does presumably to check the instantiated type).

This is not complete: Generic invocations with forward types only check
that their base type is a forward type, but not any of their arguments,
which causes #16754 and #24133. The generated invocations also need to
cascade properly: `Foo[Bar[ForwardType]]` for example would see that
`Bar[ForwardType]` is a generic invocation and stay as a generic
invocation itself, but it might not queue itself to be updated later.
Even if it did, only the entire type `Foo[Bar[ForwardType]]` needs to be
queued, updating `Bar[ForwardType]` by itself would be redundant or it
would not change anything at all. But these can be done later.
2025-04-21 07:56:14 +02:00
lit
8bc8d40778 fix(docgen): export for imported symbols missing; closes #24890 (#24891) 2025-04-20 21:22:03 +02:00
metagn
032da90ed1 implement parser for new case objects (#24885)
refs https://github.com/nim-lang/RFCs/issues/559

Parses as an `nkIdentDefs` with an `nkEmpty` name. Pragma is allowed,
can remove this if necessary.

Fine to close and postpone for later
2025-04-18 05:34:21 +02:00
metagn
5aaba213d4 account for invalid data in enum $ on arc/orc (#24886)
closes #24875

Refc gives `0 (invalid data!)`, but since enum `$` procs on arc are
generated during enum declarations we might not have access to string
concatenation and integer `$`, so it generates a static string. Just
chose an empty string for this.
2025-04-18 05:32:49 +02:00
ringabout
af9219ada7 fixes #24881; build_all.sh koch tools fails to build atlas (#24884)
fixes #24881

To test: `nim c koch.nim` + delete the `dist` directory
2025-04-17 18:04:03 +02:00
metagn
3d14381473 fix stmtlist expression indent regression (#24883)
follows up #24855

Before #24855, the test would work because the indentation of the `;`
token would be passed to `semiStmtList` and so its indentation of `-1`
would be used. Now the `;` token is skipped and the indentation of the
first `discard` is used which is > -1. However the second discard has an
indentation of -1 because it's on the same line: this fails the
`sameInd(p) or realInd(p)` check since -1 is never >= the indent of the
first discard.

For compatibility with the parser up to this point this indent check is
entirely removed, meaning the indent is ignored. Because the `;` is
basically never on a separate line, this was already the case for
basically every use. `semiStmtList` is wrapped in a `withInd` anyway
which resets the indent after it's done, since the entire statement list
is wrapped in a `()`. To disallow dedents, the above check could be
fixed to use `sameOrNoInd` instead of `sameInd`, which is done in the
commented version of this check.
2025-04-17 00:44:31 +03:00
ringabout
9f359e8d6d fixes #24879; Data getting wiped on copy with iterators and =copy on refc (#24880)
fixes #24879
2025-04-16 22:51:12 +02:00
ringabout
3f9c269013 fixes nimsugget with Checksums deps (#24882)
ref https://github.com/nim-lang/Nim/issues/24881
2025-04-16 19:44:53 +02:00
Miran
11e4bd668c update the tooling versions (#24878) 2025-04-16 15:17:26 +02:00
Juan M Gómez
e7f73bfebe Fixes a nimsuggest crash (#24873) 2025-04-16 12:11:33 +02:00
metagn
c06bb6cc03 don't traverse inner procs to lift locals in closure iters (#24876)
fixes #24863, refs #23787 and #24316

Working off the minimized example, my understanding of the issue is: `n`
captures `r` as `:envP.r1` where `:envP` is the environment of `b`, then
`proc () = n()` does the lambda lifting of `n` again (which isn't done
if the `proc ()` is marked `{.closure.}`, hence the workaround) which
then captures the `:envP` as another field inside the `:envP`, so it
generates `:envP.:envP_2.r1` but the `.:envP_2` field is `nil`, so it
causes a segfault.

The problem is that the capture of `r` in `n` is done inside
`detectCapturedVars` for the surrounding closure iterator: inner procs
are not special cased and traversed as regular nodes, so it thinks it's
inside the iterator and generates a field access of `:envP` freely. The
lambda lifting version of `detectCapturedVars` ignores inner procs and
works off of symbol uses (anonymous iterator and lambda declarations
pretend their symbol is used).

As a naive solution, closure iterators now also ignore inner proc
declarations same as `lambdalifting.detectCapturedVars`, but unlike it
they also don't do anything for the inner proc symbols. Lambdalifting
seems to properly handle the lifted variables but in the worst case we
can also make sure `closureiters.detectCapturedVars` traverses inner
procs by marking every local of the closure iter used in them as needing
lifting (but not doing the lifting). This does not seem necessary for
now so it's not done (was done and reverted in [this
commit](9bb39a9259)),
but regressions are still possible
2025-04-15 19:29:46 +02:00
metagn
4d9e5e8b6d fix field setter fallback that never worked (#24871)
refs https://forum.nim-lang.org/t/12785, refs #4711

The code was already there that when `propertyWriteAccess` returns `nil`
(i.e. cannot find a setter), `semAsgn` turns the [LHS into a call and
semchecks
it](1ef9a656d2/compiler/semexprs.nim (L1941-L1948)),
meaning if a setter cannot be found a getter will be assigned to
instead. However `propertyWriteAccess` never returned nil, because
`semOverloadedCallAnalyseEffects` was not called with `efNoUndeclared`
and so produced an error directly. So `efNoUndeclared` is passed to this
call so this code works as intended.

This fixes the issue described in #4711 which was closed because
subscripts do not have the same behavior implemented. However we can
implement this for subscripts as well (I have an implementation ready),
it just changes the error message from the failed overloads of `[]=` to
the failed overloads of `[]` for the LHS, which might be misleading but
is consistent with the error messages for any other assignment. I can do
this in this PR or another one.
2025-04-13 19:21:33 +02:00
metagn
1ef9a656d2 allow setting arbitrary size for importc types (#24868)
split from #24204, closes #7674

The `{.size.}` pragma no longer restricts the given size to 1, 2, 4 or 8
if it is used for an imported type. This is not tested very thoroughly
but there's no obvious reason to disallow it.
2025-04-12 17:55:11 +02:00
metagn
334f96c05a isolate and rematch generic converters to get bindings (#24867)
fixes #4554, fixes #10900, fixes #13843, fixes #19471, fixes #19517

Instead of matching generic converters to their arguments using the full
call match bindings, a new match is created for them (from which the
bindings are used to instantiate the converter return type). Then when
instantiating generic converters, they are matched to their argument
again to get their bindings again instead of using the call bindings.
This prevents generic converters which match more than once from
interfering with each other's bindings.
2025-04-12 17:53:18 +02:00
Jake Leahy
0cba752c8a Allow specifiying path to use for stdin error messages (#24595)
Implements #24569

Adds `--stdinfile` flag for specifying the file to use in place of
`stdinfile.nim` in error messages. Will enable easier integration of
tooling with nim check
2025-04-12 08:40:25 +02:00
metagn
4d075dc301 clean up opensym encounters in compiler (#24866)
To protect against crashes when this stops being experimental, in most
places handled the exact same as normal symchoices (not encountered in
typed ast)
2025-04-12 08:39:11 +02:00
metagn
97d819a251 add bit type overloads of $ and repr (#24865)
fixes #24864
2025-04-12 08:37:36 +02:00
握猫猫
b961ee69aa Update winlean.nim, import AddrInfo from ws2tcpip.h (#24828)
[ADDRINFOA](https://learn.microsoft.com/en-us/windows/win32/api/ws2def/ns-ws2def-addrinfoa#remarks).
2025-04-12 07:16:13 +02:00
ringabout
42df731a2d fixes #24764; cross-module sink analysis broken (#24862)
fixes  #24764

It now consumes the `conv(x)` arg for the explicit sinking. So the
explicit sinking is kept as it is.

Follows up https://github.com/nim-lang/Nim/pull/20585

Related issues: https://github.com/nim-lang/Nim/issues/20572

Probably the same needs to be applied to explicit `copy` to prevent a
copy turning into a sink
2025-04-12 06:47:57 +02:00
Ryan McConnell
520bbaf384 split nativesockets bindAddr into two procs (#24860)
#24858
2025-04-12 06:47:09 +02:00
metagn
f58cd51fc4 ignore typeof in closure iterators (#24861)
fixes #24859
2025-04-11 23:50:13 +03:00
metagn
897126a711 fix array/set/tuple literals with generic expression elements (#24497)
fixes #24484, fixes #24672

When an array, set or tuple constructor has an element that resolves to
`tyFromExpr`, the type of the entire literal is now set to `tyFromExpr`
and the subsequent elements are not matched to any type.

The remaining expressions are still typed (a version of the PR before
this called `semGenericStmt` on them instead), however elements with int
literal types have their types set to `nil`, since generic instantiation
removes int literal types and the int literal type is required for
implicitly converting the int literal element to the set type. Tuples
should not really need this but it is done for them anyway in case it
messes up some type inference

---------

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
2025-04-11 18:38:35 +02:00
Ryan McConnell
d4098e6ca0 new-style concept bugfix (#24858)
Combining two small PRs in one here. The test case explains what was
wrong with the concepts and for naitivesockets, it's typical to adjust
`ai_flags` so I opened that up.
2025-04-11 06:54:52 +02:00
metagn
918f972369 skip semicolon in stmtlist expr parsing (#24855)
Previously it would try to parse the semicolon as its own statement and
produce an `nkEmpty` node

Also more than 1 semicolon in an expression list i.e. `(a;; b)` gives an
"expression expected" error in `semiStmtList` when multiple semicolons
are allowed in normal statements, this could be fixed by changing the
`if tok.kind == tokSemicolon` check to a `while` but it does not match
the grammar so not done here.
2025-04-11 03:29:20 +02:00
ringabout
51166ab382 fixes =copy is transformed into nkFastAsgn and unify mAsgn handling (#24857)
`=copy` should be treated like `=` instead of `shallowCopy`, i.e.,
`nkFastAsgn` by default. `mAsgn` is treated similar in sempass2 too
2025-04-11 03:28:53 +02:00
ringabout
40a1ec21d7 overhaul hook injections (#24841)
ref https://github.com/nim-lang/Nim/issues/24764 

To keep destructors injected consistently, we need to transform `mAsgn`
properly into `nkSinkAsgn` and `nkAsgn`. This PR is the first step
towards overhauling hook injections.

In this PR, hooks (except mAsgn) are treated consistently whether it is
resolved in matching or instantiated by sempass2. It also fixes a
spelling `=wasMoved` to its normalized version, which caused no
replacing generic hook calls with lifted hook calls.
2025-04-10 09:24:19 +02:00
ringabout
29a2e25d1e fixes #24850; macro-generated if/else and when/else statements have m… (#24852)
…ismatched indentation with repr

fixes #24850
2025-04-08 23:54:31 +08:00
metagn
a625fab098 make fillObjectFields recur over base type (#24854)
fixes #24847

Object constructors call `fillObjectFields` when a field inside the
constructor does not have a location, however when the field is from a
base type this does not process it. Now `fillObjectFields` also calls
itself for the base type to fix this but not sure if this is a good
solution as `fillObjectFields` is used in other places too.
2025-04-08 17:00:58 +03:00
ringabout
052ceca3c1 bump to windows 2025 (#24853) 2025-04-08 20:07:24 +08:00
Miran
10c9ebad93 test stint more thoroughly (#24832) 2025-04-03 17:43:27 +02:00
ringabout
26b86c8f4d Makes except: panics on Defect (#24821)
implements https://github.com/nim-lang/RFCs/issues/557


It inserts defect handing into a bare except branch

```nim
try:
  raiseAssert "test"
except:
  echo "nope"
```

=>

```nim
try:
  raiseAssert "test"
except:
  # New behaviov, now well-defined: **never** catches the assert, regardless of panic mode
  raiseDefect()
  echo "nope"
```

In this way, `except` still catches foreign exceptions, but panics on
`Defect`. Probably when Nim has `except {.foreign.}`, we can extend
`raiseDefect` to foreign exceptions as well. That's supposed to be a
small use case anyway.

 `--legacy:noPanicOnExcept` is provided for a transition period.
2025-04-03 16:09:58 +02:00
la.panon.
2ed45eb848 Make loadConfig available from NimScript (#24840)
fixes #24837 

I really wanted to name the variable just `stream` and leave `defer:
...` and `result =...` out, but the compiler says the variable is
redefined, so this is the form.
2025-04-03 15:54:39 +02:00
ringabout
73aeac81d1 fixes #24806; don't elide wasMoved when syms are used in blocks (#24831)
fixes #24806

Blocks don't merge symbols that are used before destruction to the
parent scope, which causes `wasMoved; destroy` to elide incorrectly
2025-04-03 12:54:00 +02:00
metagn
5bcd9a329a fix infinite recursion with pushed user pragmas (#24839)
fixes #24838
2025-04-03 12:53:42 +02:00
ringabout
4352fa2ef0 fixes #24801; Invalid C codegen generated when destroying distinct seq types (#24835)
fixes #24801

Because distinct `seq` types match `proc `=destroy`*[T](x: var T)
{.inline, magic: "Destroy".}`. But the Nim compiler generates lifted seq
types for corresponding distinct types. So we skip the address for
distinct types.

Related to https://github.com/nim-lang/Nim/pull/22207 I had a hard time
finding the other place where generic destructors get replaced by
attachedDestructors
2025-04-02 18:46:29 +02:00
ringabout
3617d2e077 fixes lastRead uses the when nimvm branch (#24834)
```nim
proc foo =
  var x = "1234"
  var y = x
  when nimvm:
    discard
  else:
    var s = x
    doAssert s == "1234"
  doAssert y == "1234"

static: foo()
foo()
```
`dfa` chooses the `nimvm` branch, `x` is misread as a last read and
`wasMoved`.

`injectDestructor` is used for codegen and is not used for vmgen. It's
reasonable to choose the codegen path instead of the `nimvm` path so the
code works for codegen. Though the problem is often hidden by
`cursorinference` or `optimizer`.

found in https://github.com/nim-lang/Nim/pull/24831
2025-04-02 09:29:15 +02:00
ringabout
f9c8775783 conv needs to be picky about aliases and introduces a temp for addr conv (#24818)
ref https://github.com/nim-lang/Nim/pull/24817
ref https://github.com/nim-lang/Nim/pull/24815
ref https://github.com/status-im/nim-eth/pull/784


```nim
{.emit:"""
void foo(unsigned long long* x)
{
}
""".}


proc foo(x: var culonglong) {.importc: "foo", nodecl.}

proc main(x: var uint64) =
  # var s: culonglong = u # TODO:
  var m = uint64(12)
  # var s = culonglong(m)
  foo(culonglong m)

var u = uint64(12)
main(u)
```
Notes that this code gives incompatible errors in 2.0.0, 2.2.0 and the
devel branch. With this PR, `conv` is kept, but it seems to go back to
https://github.com/nim-lang/Nim/pull/24807
2025-04-01 09:37:54 +02:00
James
0f5732bc8c Add withValue for immutable tables (#24825)
This change adds `withValue` templates for the `Table` type that are
able to operate on immutable table values -- the existing implementation
requires a `var`.

This is needed for situations where performance is sensitive. There are
two goals with my implementation:

1. Don't create a copy of the value in the table. That's why I need the
`cursor` pragma. Otherwise, it would copy the value
2. Don't double calculate the hash. That's kind of intrinsic with this
implementation. But the only way to achieve this without this PR is to
first check `if key in table` then to read `table[key]`

I brought this up in the discord and a few folks tried to come up with
options that were as fast as this, but nothing quite matched the
performance here. Thread starts here:
https://discord.com/channels/371759389889003530/371759389889003532/1355206546966974584
2025-03-29 23:08:45 +01:00
Jake Leahy
e0a4876981 Fix nim-gdb.py script (#24824)
Script wasn't working on my machine with GDB 16.2
Main issues
 - `gdb.types` wasn't imported, leading to import error on initial load
 - dollar function didn't work with the new mangling scheme

Fixes them, also updates the test script to work with some new mangling
changes.

Test evidence

![image](https://github.com/user-attachments/assets/450b020f-1665-4ed2-9073-d02537150914)
2025-03-29 13:28:28 +01:00
Zoom
ecdcffed4b Mark system.newStringUninit sideeffect-free (#24813)
- Allows using with `--experimental:strictFuncs`
- `{.cast(noSideEffect).}:` inside the proc was required to mutate
`s.len`, same as used in `newSeqImpl`.
- Removed now unnecessary `noSideEffect` casts in `system.nim`
- 
Closes #24811

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-03-28 22:06:22 +08:00
ringabout
58b1f28177 fixes implicitConv discarding flags (#24817)
follow up https://github.com/nim-lang/Nim/pull/24809
ref https://github.com/nim-lang/Nim/pull/24815
2025-03-28 12:52:45 +01:00
ringabout
73112d64a3 fixes #24793; Revert "remove special treatments of sinking const sequences (#24812)
fixes #24793

There doesn't seem to have a better solution
2025-03-26 23:49:00 +08:00
Zoom
b82d7e8ba1 stdlib: substr uses copymem if available, improve docs (#24792)
- `system.substr` now uses `copymem` when available, introducing a small
template for nimvm detection (#12517 #12518)
- Docs are updated to clarify behaviour on out-of-bounds input
- Runnable examples cover more edge cases and do not repeat between
overloads
- Docs now explain the difference between overloads

What bothers me is that the `substr*(a: openArray[char]): string =`
which was added by @beef331 is practically an implementation of #14810,
which is just a conversion from `openArray` to `string` but somehow it
ended up being a `substr` overload, even though its behaviour is totally
different, _the "substringing" is performed by a previous step_
(conversion to openArray) and the bounds are not checked. I'm not sure
it's that great for overloads to differ in subtle ways so much.

What are the cases that `substr` covers now, that prohibit renaming it
to `toString` (or something like that)?
2025-03-25 21:06:40 +01:00
ringabout
ddd83f8d8a fixes #24800; Invalid C code generation with a method, case object in refc (#24809)
fixes #24800

This PR avoids a conversion from `sink T` to `T`

I will add a test case
2025-03-25 20:42:40 +01:00
握猫猫
8e36fb0fec Update nativesockets.nim, namelen should be the len of name (#24810)
In other places where `getsockname` is called, the size of the 'name' is
used.


d573578b28/lib/pure/nativesockets.nim (L347-L351)

d573578b28/lib/pure/nativesockets.nim (L585-L595)

d573578b28/lib/pure/nativesockets.nim (L622-L624)

d573578b28/lib/pure/nativesockets.nim (L347-L350)

I have checked the [Windows
documentation](https://learn.microsoft.com/en-us/windows/win32/api/winsock2/nf-winsock2-getsockname#remarks),
and it describes it like this: "On call, the namelen parameter contains
the size of the name buffer, in bytes. On return, the namelen parameter
contains the actual size in bytes of the name parameter."


[https://www.man7.org/linux/man-pages/man2/getsockname.2.html](https://www.man7.org/linux/man-pages/man2/getsockname.2.html)
say:
The addrlen argument should be initialized to indicate the amount of
space (in bytes) pointed to by addr.
2025-03-25 20:32:12 +01:00
lit
d573578b28 repl: support eof, define object with fields (#24784)
For `nim secret`:

- **fix(repl): eof(ctrl-D/Z) and ctrl-C were ignored**
- **feat(repl): continueLine  figures section, constr, bool ops**

---------

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-03-25 07:41:17 +01:00
Zoom
909f3b8b79 [feature] stdlib: strutils.multiReplace for character sets (#24805)
Multiple replacements based on character sets in a single pass. Useful
for string sanitation. Follows existing `multiReplace` semantics.

Note: initially copied the substring version logic with a `while` and a
named block break, but Godbolt showed it had produced slightly larger
assembly using higher registers than the final version.

- [x] Tests
- [x] changelog.md
2025-03-25 07:40:01 +01:00
ringabout
d15705e05b fixes usenimrtl with useMalloc (#24804)
Follow up https://github.com/nim-lang/Nim/pull/19512

ref https://github.com/nim-lang/Nim/issues/24794

Otherwise, `/Users/blue/Desktop/Nim/lib/system/mm/malloc.nim(4, 1)
Error: redefinition of 'allocImpl'; previous declaration here:
/Users/blue/Desktop/Nim/lib/system/memalloc.nim(51, 8)`


In `proc allocImpl*(size: Natural): pointer {.noconv, rtl, tags: [],
benign, raises: [].}`, `rtl` means it is an `importc` function instead
of a proc forward decl.
2025-03-24 22:52:43 +01:00
ringabout
0b9ed84d32 disable implicit sinkinference for stdlibs (#24803)
ref https://github.com/nim-lang/Nim/issues/24794
2025-03-24 14:07:45 +01:00
metagn
fcba14707a disable "dest register is set" for vm statements (#24797)
closes #24780

This proc `genStmt` is only called to run the VM in `vm.evalStmt`,
otherwise it's not used in vmgen. Now it acts the same as `proc
gen(PCtx, PNode)`, used by `discard` statements, which just calls
`freeTemp` on the dest if it was set rather than erroring.
2025-03-23 06:59:06 +03:00
ringabout
482662d198 fixes #24721; Table add missing sink (#24724)
fixes #24721
2025-03-22 22:48:21 +01:00
Esteban C Borsani
9ace1f97ac Fix SIGSEGV when closing SSL async socket while sending/receiving (#24795)
Async SSL socket SIGSEGV's sometimes when calling socket.close() while
send/recv. The issue was found here
https://github.com/nitely/nim-hyperx/pull/59.

Possibly related: #24024

This can occur when closing the socket while sending or receiving,
because `socket.sslHandle` is freed. The sigsegv can also occur on calls
that require `socket.bioIn` or `socket.bioOut` because those use
`socket.sslHandle` internally. This PR checks sslHandle is set before
doing any operation that requires it.
2025-03-22 16:38:38 +01:00
Angus Gibson
1d32607575 Allow parsing year "00" with "yy" pattern (#24785)
The "yy" pattern is relative to the current century, so year "00" should
be valid.
2025-03-19 08:15:54 +01:00
ringabout
7c5d005510 fixes #10625; setjmp on linux mangles ebp leading to early collection (#24787)
fixes #10625
2025-03-18 18:51:34 +08:00
Ryan McConnell
2b699bca53 new-style concepts - small bugfix (#24778) 2025-03-15 15:05:14 +01:00
metagn
fb93295344 fix compound inheritance penalty (#24775)
fixes #24773

`c.inheritancePenalty` is supposed to be used for the entire match, but
in these places the inheritance penalty of a single argument overrides
the entire match penalty. The `+ ord(c.inheritancePenalty < 0)` is
copied from other places that use the same idiom, the intent is that the
existing penalty changes from -1 to 0 first to mark that it participates
in inheritance before adding the inheritance depth.

---------

Co-authored-by: Andreas Rumpf <araq4k@proton.me>
2025-03-12 17:31:33 +01:00
ringabout
9ebfa7973a fixes generic types sink T cannot be inferred for passed arguments (#24761)
Otherwise, `sink T` is kept as it is. This PR treats sink types as its
base types for the arguments. So the concept would match both cases

Required by https://github.com/nim-lang/Nim/pull/24724
2025-03-12 17:31:19 +01:00
lit
4f32624641 fixes #24772: system.NaN was negative when C (#24774)
fixes #24772

The old implementation was said to copied  from Windows SDK,

but you can find the newer SDK's definition is updated and the sign is
reversed compared to the old.


Also, `__builtin_nanf("")` is used if available,
which is more efficient than previous (In x86_64 gcc, latter produces
32B code but former just 8B).
2025-03-12 17:30:08 +01:00
ringabout
dfa482e292 fixes #24770; Thread local not registed as GC root when =destroy exists (#24776)
fixes #24770

e.g. `seq[(ObjectWithDestructors, string)]`/ For refc, a seq with
elements that have destructors will have `hasAsgn` flags. The flag is
the criteria whether a seq is thought as `containsGarbageCollectedRef`.
i.e. whether to `registerTraverseProc` for the type.
The culprit seems to be that `searchTypeForAux` doesn't consider the
element type of sequence, even it contains a string that should belong
to `GarbageCollectedRef`.

With this PR:

It now generates

```
nimRegisterThreadLocalMarker(TM__mSF73dT1lSI7DG58StKHLQ_5);
```

in refc
2025-03-12 17:29:30 +01:00
metagn
82891e6850 give hint for forward declarations with unknown raises effects (#24767)
refs #24766

Detect when we track a call to a forward declaration without explicit
`raises` effects, then when the `raises` check fails for the proc, give
a hint that this forward declaration was tracked as potentially raising
any exception.
2025-03-11 14:24:45 +01:00
Ryan McConnell
850f327713 folding const expressions with branching logic (#24689)
motivating example:
```nim
iterator p(a: openArray[char]): int =
  if a.len != 0:
    if a[0] != '/':
      discard
for t in p(""): discard
```
The compiler wants to evaluate `a[0]` at compile time even though it is
protected by the if statement above it. Similarly expressions like
`a.len != 0 and a[0] == '/'` have problems. It seems like the logic in
semfold needs to be more aware of branches to positively identify when
it is okay to fail compilation in these scenarios. It's a bit tough
though because it may be the case that non-constant expressions in
branching logic can properly protect some constant expressions.
2025-03-11 10:01:32 +01:00
metagn
38ad336c69 fix tuple nodes from VM inserting hidden conv to keep old type (#24756)
fixes #24755, refs #24710

Instead of using the node from `indexTypesMatch` which inserts a hidden
conv node, just change the type of the node back to the old type
directly
2025-03-11 10:00:37 +01:00
ringabout
a7711d452d fixes #24754; {.gcsafe.} block breaks move analysis (#24757)
fixes #24754
2025-03-11 09:59:55 +01:00
metagn
e2e7790779 fix canRaise for non-proc calls (#24752)
fixes #24751

`typeof` leaves the object constructor as a call node for some reason,
in this case it tries to access the first child of the type node but the
object has no fields so the type field is empty. Alternatively the
optimizer can stop looking into `typeof`
2025-03-11 09:59:21 +01:00
Michael Lee
dfd2987118 Add linking options for tinycc backend (#24750)
### Issue

When using `tcc` as backend to compile a trivial program

```
nim c  --cc:tcc  --skipCfg a.nim
```

, errors reported:

```
tcc: error: undefined symbol 'fabs'
```

### Solution

`fabs` belongs to libm. With these two options added, one can compile
with an additional clib option:

```
nim c  --cc:tcc  --skipCfg --clib:m a.nim
```
2025-03-11 09:58:22 +01:00
ringabout
e2d4791229 fixes move for getPotentialWrites (#24753)
`move` would modify parameters as well
2025-03-11 09:57:48 +01:00
ringabout
ccb40024c6 remove special treatments of sinking const sequences (#24763) 2025-03-11 09:56:48 +01:00
Laylie
f8294ce06e Fix scanTuple undeclared identifier 'scanf' (#24759)
Without this fix, trying to use `scanTuple` in a generic proc imported
from a different module fails to compile (`undeclared identifier:
'scanf'`):

```nim
# module.nim
import std/strscans

proc scan*[T](s: string): (bool, string) =
  s.scanTuple("$+")
```

```nim
# main.nim
import ./module
 
echo scan[int]("foo")
```

Workaround is to `export scanf` in `module.nim` or `import std/strscans`
in `main.nim`.
2025-03-10 22:47:03 +08:00
ringabout
b8302cdd97 implements internal sink copy (#24747)
TODO:

- [x] other value types (arrays, strings, seqs, objects) 
- [x] replaces https://github.com/nim-lang/Nim/pull/24731
- [x] improve code shape
- [ ] revert https://github.com/nim-lang/Nim/issues/24175
- [x] if possible, revert https://github.com/nim-lang/Nim/pull/23685
- [ ] if possible, revert https://github.com/nim-lang/Nim/pull/22229 and
https://github.com/nim-lang/Nim/pull/23764
- [ ] if possible, remove `if n.containsConstSeq:`
- [ ] if possible, always pass value (arrays, strings, seqs, tuples, or
even objects without custom hooks (?)) sinks by ref because this PR
should ensure these value types are not modified without a copy
- [x] fixes `say a, (b = move a; a)` for potential writes
https://github.com/nim-lang/Nim/pull/24753
2025-03-10 11:20:44 +01:00
Ryan McConnell
dfab30734b new-style concepts adjusments (#24697)
Yet another one of these. Multiple changes piled up in this one. I've
only minimally cleaned it for now (debug code is still here etc). Just
want to start putting this up so I might get feedback. I know this is a
lot and you all are busy with bigger things. As per my last PR, this
might just contain changes that are not ready.

### concept instantiation uniqueness
It has already been said that concepts like `ArrayLike[int]` is not
unique for each matching type of that concept. Likewise the compiler
needs to instantiate a new proc for each unique *bound* type not each
unique invocation of `ArrayLike`

### generic parameter bindings
Couple of things here. The code in sigmatch has to give it's bindings to
the code in concepts, else the information is lost in that step. The
code that prepares the generic variables bound in concepts was also
changed slightly. Net effect is that it works better.
I did choose to use the `LayedIdTable` instead of the `seq`s in
`concepts.nim`. This was mostly to avoid confusing myself. It also
avoids some unnecessary movings around. I wouldn't doubt this is
slightly less performant, but not much in the grand scheme of things and
I would prefer to keep things as easy to understand as possible for as
long as possible because this stuff can get confusing.

### various fixes in the matching logic
Certain forms of modifiers like `var` and generic types like
`tyGenericInst` and `tyGenericInvocation` have logic adjustments based
on my testing and usage

### signature matching method adjustment
This is the weird one, like my last PR. I thought a lot about the
feedback from my last attempt and this is what I came up with. Perhaps
unfortunately I am preoccupied with a slight grey area. consider the
follwing:
```nim
type
  C1 = concept
    proc p[T](s: Self; x: T)
  C2[T] = concept
    proc p(s: Self; x: T)
```
It would be temping to say that these are the same, but I don't think
they are. `C2` makes each invocation distinct, and this has important
implications in the type system. eg `C2[int]` is not the same type as
`C2[string]` and this means that signatures are meant to accept a type
that only matches `p` for a single type per unique binding. For `C1` all
are the same and the binding `p` accepts multiple types. There are
multiple variations of this type classes, `tyAnything` and the like.

The make things more complicated, an implementation might match:
```nim
type
  A = object
  C3 = concept
    proc p(s: Self; x:  A)
```
if the implementation defines:
```nim
proc p(x: Impl; y: object)
```

while a concept that fits `C2` may be satisfied by something like:
```nim
proc p(x: Impl; y: int)
proc spring[T](x: C2[T])
```
it just depends. None of this is really a problem, it just seems to
provoke some more logic in `concepts.nim` that makes all of this (appear
to?) work. The logic checks for both kinds of matches with a couple of
caveats. The fist is that some unbind-able arrangements may be matched
during overload resolution. I don't think this is avoidable and I
actually think this is a good way to get a failed compilation. So, first
note imo is that failing during binding is preferred to forcing the
programming to write annoying stub procs and putting insane gymnastics
in the compiler. Second thing is: I think this logic is way to accepting
for some parts of overload resolutions. Particularly in `checkGeneric`
when disambiguation is happening. Things get hard to understand for me
here. ~~I made it so the implicit bindings to not count during
disambiguation~~. I still need to test this more, but the thought is
that it would help curb excessive ambiguity errors.

Again, I'm sorry for this being so many changes. It's probably
inconvenient.

---------

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
2025-03-07 18:14:00 +01:00
metagn
569d02e212 generate tyFromExpr for typeof static param with generic base type (#24745)
fixes #24743, refs #24718

We cannot do this in general for any expression with generic type
because the `typeof` logic is called for things like `type Foo` in:

```nim
type Foo[T] = object

proc init(_: type Foo) = discard
```

We also cannot use `containsUnresolvedType` to work around this specific
case because the base type of `static[auto]` is not unresolved, it is a
typeclass that isn't lifted to a parameter. The behavior of generating
`tyFromExpr` is also consistent with pre-2.0, so we do this in this
special case of `static`.
2025-03-05 13:47:56 +01:00
metagn
e39d152b89 handle ranges in annotateType for set constructors (#24737)
fixes #24736

The VM can produce integer nodes with no types as set elements, which
are later reannotated in `semmacrosanity.annotateType`. However the case
of ranges was not handled properly. Not sure why this is a regression,
probably unrelated but will have to see the bisect result to make sure.

Note. Originally tried to fix this in `opcInclRange`, generated for and
only for range expressions in set constructors, this seems to add the
range node directly to the set node without checking if it has overlap
with the existing elements by calling `nimsets` so an expression like
`{cctNone, cctNone..cctHeader}` can produce `{0, 0..5}`. Doesn't seem to
cause problems but `opcIncl` for single elements does check for overlap.

Something else to note is that integer nodes produced by `nimsets` have
proper types, so another option instead of relying on semmacrosanity to
fix this would be to make `opcIncl` and `opcInclRange` call `nimsets` to
add to the set node, but this might lose performance.
2025-02-28 17:23:19 +03:00
ringabout
7e8a650729 sink tuples by values (#24731)
A reduced case
```nim
type AnObject = tuple
  a: string
  b: int
  c: int

proc mutate(a: sink AnObject) =
  `=wasMoved`(a)
  echo 1

# echo "Value is: ", obj.value
proc bar =
  mutate(("1.2", 0, 0))

bar()
```
2025-02-27 18:33:35 +01:00
metagn
0cb4cd159e move nim version in issue template to the top (#24733)
The point is to move it out of the current place between "Description"
and "Current Output" as often these are related to each other and
including the nim version in the middle breaks the flow of reading. I
also thought of moving it below "Expected Output" but this felt too low
for a required field and also has the same problem of overshadowing the
remaining sections. Not sure if it being at the top is annoying in some
other way though.
2025-02-27 16:48:53 +01:00
ringabout
7ecb35115b fixes #24339; underscores used with fields and fieldPairs (#24341)
fixes #24339
2025-02-27 16:48:15 +01:00
ringabout
b421d0f8ee fixes #19728; setLen slow when shrinking seq due to zero-filling of released area (#24683)
fixes #19728

don't zero-filling memory for "trivial types" without destructor in
refc. I tested locally with internal apis.
2025-02-27 16:45:58 +01:00
ringabout
c452275e29 fixes #24705; encode static parameters into function names for debugging (#24707)
fixes #24705

```nim
proc xxx(v: static int) =
  echo v
xxx(10)
xxx(20)
```

They are mangled as `_ZN14titaniummangle7xxx_s10E` and
`_ZN14titaniummangle7xxx_s20E` with `--debugger:native`. Static
parameters are prefixed with `_s` to distinguish simple cases like
`xxx(10, 15)` and `xxx(101, 5)` if `xxx` supports two `static[int]`
parameters
2025-02-27 16:45:04 +01:00
metagn
a18dcca744 don't try to infer array range to unresolved range (#24709)
fixes #24708
2025-02-26 22:21:03 +01:00
metagn
49dfc3a0d4 convert tuple constructors from VM back to original types (#24710)
fixes #24698

The same aim as #24224 but for tuple constructors. The difference here
is that the type of a tuple constructor is always going to be valid
unlike array constructors which can have `seq` etc types, so we can just
generate a conversion again. If the conversion fails, it is ignored
similar to #24611, this is to protect against modified typed nodes in
macros.

Also #24611 was only adapted to `semTupleFieldsConstr` and not
`semTuplePositionsConstr`, this is now fixed.
2025-02-26 22:20:41 +01:00
Michael Lee
16280d4e49 Improve bash completion support (#24692)
Following https://github.com/nim-lang/nimble/pull/1347, this patch adds
bash completion support for `nim`, `nimgrep`, `nimpretty`, `nimsuggest`.
2025-02-26 22:19:46 +01:00
ringabout
a7a8e364ea fixes #12340; enable refc with move analyzer (#23782)
fixes https://github.com/nim-lang/Nim/issues/12340
2025-02-25 20:20:24 +01:00
Ryan McConnell
d94e535145 Make koch friendlier to offline environments (#24713) 2025-02-25 17:10:30 +01:00
metagn
514a25c9a2 always skip static types for result of typeof (#24718)
fixes #24715

In generic typechecking, unresolved static param symbols (i.e.
`skGenericParam`) have [the static type
itself](1f8da3835f/compiler/semexprs.nim (L1483-L1485))
as their type when used in an expression. This is not the case when the
static param is resolved (the type is wrapped in static when necessary),
but semchecking of types and generic typechecking expects the type of
the value to be wrapped in `static` (at least `array[N, int]` breaks).
So for now, to solve the issue, `typeof` just skips static types.
2025-02-25 17:09:04 +01:00
ringabout
e449813c61 fixes #24725; Invalid =sink generated for pure inheritable object (#24726)
fixes #24725

`lacksMTypeField` doesn't take the base types into consideration. And
for ` {.inheritable, pure.}`, it shouldn't generate a `m_type` field.
2025-02-25 17:08:22 +01:00
ringabout
93fb219f10 undeprecates var T destructors (#24716)
Both cases are now valid. Though, it could be problematic to mix two
cases together as built-in types have non var T destructors
2025-02-25 17:05:59 +01:00
metagn
1f8da3835f keep param pragmas in typed proc AST (#24711)
fixes #24702
2025-02-22 21:22:30 +01:00
ringabout
1af88a2d20 always mangle local variables (#24681)
ref #24677
2025-02-19 23:03:58 +01:00
lit
91e8e605d0 fix(dollar): $NaN -> "NaN", $Inf -> "Infinity" only when js (#24695)
ref nimpylib/pylib#44

---------

Co-authored-by: ringabout <43030857+ringabout@users.noreply.github.com>
2025-02-19 23:03:28 +01:00
ringabout
f0b5bf359e fixes ORC memory leaks; marks hooks with optQuirky (#24701)
closes https://github.com/nim-lang/Nim/pull/24686
closes #24693

```nim
# v.nim
import std/[json]

var test: seq[string]
var testData: JsonNode
try:
  ## Fails
  testData = parseJson("""[{"id": 1"}, {"id": "2"}]""")

  ## Works
  # testdata = parseJson("""[{"id": "1"}, {"id": "2"}]""")
  
  ## Fails
  # let stream = newStringStream("""[{"id": 1"}, {"id": "2"}]""")
  # testData = parseJson(stream, "input", false, false)
  # stream.close()
  
except:
  testData = %* []
for t in testData:
  test.add(t["id"].getStr())
echo $test
```

With this PR:

```
==66425== LEAK SUMMARY:
==66425==    definitely lost: 0 bytes in 0 blocks
==66425==    indirectly lost: 0 bytes in 0 blocks
==66425==      possibly lost: 0 bytes in 0 blocks
==66425==    still reachable: 16,512 bytes in 2 blocks
==66425==         suppressed: 0 bytes in 0 blocks
==66425== Reachable blocks (those to which a pointer was found) are not shown.
==66425== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==66425== 
==66425== For lists of detected and suppressed errors, rerun with: -s
==66425== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
```
2025-02-19 23:02:28 +01:00
ringabout
1f07fdd2dc fixes #24664; always sets the \0 terminator in appendString (#24703)
fixes #24664

```nim
proc main() = 
    for i in 0..1:
        var s = "12345"
        s.add s
        echo s

main()
```
In the given example, `add` contains two steps: `prepareAdd` and
`appendString`. In the first step, a new buffer is created in order to
store the final doubled string. But it doesn't copy the null terminator,
neither zeromem the left unused spaces. It causes a problem because
`appendString` will copy itself which doesn't end with `\0` properly so
contaminated memory is copied instead.

```
var s = 12345\0

prepareAdd:

var s = 12345xxxxx\0

appendString:

var s = 1234512345x
```
2025-02-19 23:01:56 +01:00
metagn
ebeef1067f adapt generic matches to inheritance penalty of final objects (#24691)
Applies #24144 to the equivalent matches of generic types, and adds the
behavior to matches of generic invocations to generic invocations. Not
encountered in many cases so it's hard to come up with tests but an
example is the test code in #24688, the match to the generic body never
sets the inheritance penalty leaving it at -1, but the match to the
generic invocation sets it to 0 which matches worse, when it should set
it to -1 because the object does not participate in inheritance.
2025-02-19 23:01:26 +01:00
ringabout
510ac84518 implements quirky for functions (#24700)
ref https://github.com/nim-lang/Nim/pull/24686

With this PR

```nim
import std/streams


proc foo() =
  var name = newStringStream("2r2")
  raise newException(ValueError, "sh")

try:
  foo()
except:
 discard

echo 123
```
this example no longer leaks

---------

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
2025-02-18 17:24:41 +01:00
Ryan McConnell
b7d8896d00 Add terminal colors back to unittest under nimPreviewSlimSystem (#24694) 2025-02-18 13:32:10 +01:00
metagn
a5cc33c1d3 track introduced locals in vmgen for eval check (#24674)
fixes #8758, fixes #10828, fixes #12172, fixes #21610, fixes #23803,
fixes #24633, fixes #24634, succeeds #24085

We simply track the symbol ID of every traversed `var`/`let` definition
in `vmgen`, then these symbols are always considered evaluable in the
current `vmgen` context. The set of symbols is reset before every
generation, but both tests worked properly without doing this including
the nested `const`, so maybe it's already done in some way I'm not
seeing.
2025-02-14 20:54:17 +01:00
ringabout
b211ada273 fixes #24673; divmod errors for ranges (#24679)
fixes #24673

The problem is that there is no way to distinguish `cint`, `cint`, etc
ctypes with Nim types. So `when T is cint | clong | clonglong:` is true
for types derived from `int`, `int32` and `int64`. In this PR, it fixes
the branch to avoid erros for `Natural`
2025-02-14 20:52:43 +01:00
Mads Hougesen
1a7bc6d878 feat(nimpretty): support formatting code from stdin (#24676)
This pr adds support for running `nimpretty` on stdin as described in
#24622.

I tested `:%!nimpretty -` and `:%!nimpretty --stdin` in neovim and both
seems to work without issues.
2025-02-10 11:12:41 +01:00
ringabout
e6f6c369ff fixes bugs on the Nim manual (#24669)
ref https://en.cppreference.com/w/cpp/error/exception/what

> Pointer to a null-terminated string with explanatory information. The
pointer is guaranteed to be valid at least until the exception object
from which it is obtained is destroyed, or until a non-const member
function on the exception object is called.

The pointer is only valid before `CStdException as e` is destroyed


Old examples are broken on macOS arm64

```
/Users/blue/Desktop/nimony/test4.nim(38) test4
/Users/blue/Desktop/nimony/test4.nim(26) fn
/Users/blue/.choosenim/toolchains/nim-#devel/lib/std/assertions.nim(41) failedAssertImpl
/Users/blue/.choosenim/toolchains/nim-#devel/lib/std/assertions.nim(36) raiseAssert
/Users/blue/.choosenim/toolchains/nim-#devel/lib/system/fatal.nim(53) sysFatal
Error: unhandled exception: /Users/blue/Desktop/nimony/test4.nim(26, 3) `$b == "foo2"`  [AssertionDefect]
```
2025-02-06 23:19:53 +01:00
ringabout
485b414fce fixes #24666; Compilation error when formatting a complex number (#24667)
fixes #24666

ref https://github.com/nim-lang/Nim/pull/22924
2025-02-05 19:37:13 +01:00
ringabout
7695d51fc4 fixes #24658; cpp compilation failure on Nim 2.2.x (#24663)
fixes #24658
2025-02-03 15:48:52 +01:00
lit
e2bed72b72 doc(tempfiles): update link of getTempDir (#24661)
- tempfiles: update `getTempDir` link... from os.html to appdirs.html
<https://nim-lang.org/docs/appdirs.html#getTempDir>

- ~~nims.md: rm three `std/`, which are out of place~~ (ref
https://github.com/nim-lang/Nim/pull/24661#discussion_r1937293833)
2025-02-03 17:12:44 +08:00
metagn
0861dabfa7 add ambiguous identifier message to generic instantiations (#24646)
fixes #24644

Another option is to include the symbol names and owners in the type
listing as in #24645 but this is a bit verbose.
2025-01-31 08:44:44 +01:00
metagn
647c6687f1 don't mark captured field sym in template as fully used (#24660)
fixes #24657
2025-01-31 08:44:02 +01:00
lit
af5fd3fea3 fix doc format: testament.md (#24654)
- **doc(format): testament: fix `Commands` not regarded as table**

![image](https://github.com/user-attachments/assets/85238dd5-e199-41ca-a8cb-05849415097a)

- **doc(format): testament: row `--target` not splited as columns**

![image](https://github.com/user-attachments/assets/230ec693-c459-4fee-bc57-f3ab6c34a9b6)
2025-01-30 18:05:51 +01:00
Peter Munch-Ellingsen
cab3342a2d Fix check for Nintendo Switch target (#24652)
This should fix ringabouts comment here:
https://github.com/nim-lang/Nim/pull/24639#issuecomment-2615107496

I wasn't aware that `nintendoswitch` and `posix` would be active at the
same time, so I falsely inverted a check.
2025-01-27 16:57:53 +01:00
Leon Lysak
8c3e62e6de Update dom.nim (removeEventListener function) (#24650)
Essentially just an update for the `removeEventListener` function as per
https://developer.mozilla.org/en-US/docs/Web/API/EventTarget/removeEventListener
2025-01-27 08:17:58 +01:00
Tomohiro
95b1dda1db Fix parseBiggestUInt to detect overflow (#24649)
With some inputs larger than `BiggestUInt.high`, `parseBiggestUInt` proc
in `parseutils.nim` fails to detect overflow and returns random value.
This is because `rawParseUInt` try to detects overflow with `if prev >
res:` but it doesn't detects the overflow from multiplication.
It is possible that `x *= 10` causes overflow and resulting value is
larger than original value.
Here is example values larger than `BiggestUInt.high` but
`parseBiggestUInt` returns without detecting overflow:
```
22751622367522324480000000
41404969074137497600000000
20701551093035827200000000000000000
22546225502460313600000000000000000
204963831854661632000000000000000000
```

Following code search for values larger than `BiggestUInt.high` and
`parseBiggestUInt` cannot detect overflow:
```nim
import std/[strutils]

const
  # Increase this to extend search range
  NBits = 34'u
  NBitsMax1 = 1'u shl NBits
  NBitsMax = NBitsMax1 - 1'u

  # Increase this when there are too many results and want to see only larger result.
  MinMultiply10 = 14

var nfound = 0
for i in (NBitsMax div 10'u + 1'u) .. NBitsMax:
  var
    x = i
    n10 = 0
  for j in 0 ..< NBits:
    let px = x
    x = (x * 10'u) and NBitsMax
    if x < px:
      break
    inc n10
  if n10 >= MinMultiply10:
    echo "i =   ", i
    echo "uint: ", (i shl (64'u - NBits)), '0'.repeat n10
    inc nfound
    if nfound > 15:
      break

echo "found: ", nfound
```
2025-01-25 15:43:40 +01:00
Peter Munch-Ellingsen
1f9cac1f5c Enable macros to use certain things from the OS module when the target OS is not supported (#24639)
Essentially this PR removes the `{.error.}` pragmas littered around in
the OS module and submodules which prevents them from being imported if
the target OS is not supported. This made it impossible to use certain
supported features of the OS module in macros from a supported host OS.
Instead of the `{.error.}` pragmas the `oscommon` module now has a
constant `supportedSystem` which is false in the cases where the
`{.error.}` pragmas where generated. All procedures which can't be run
by macros is also not declared when `supportedSystem` is false.

It would be possible to create dummy versions of the omitted functions
with an `{.error.}` pragma that would trigger upon their use, but this
is currently not done.

This properly fixes #19414
2025-01-24 13:02:59 +01:00
ringabout
67f9bc2f4b fixes #21923; nimsuggest "outline" output does not list templates (#24643)
fixes #21923

---------

Co-authored-by: Louis Berube <louis.p.berube@gmail.com>
2025-01-24 13:00:08 +01:00
ringabout
d6d28a9c79 fixes #24623; fixes #23692; size pragma only allowed for imported types and enum types (#24640)
fixes #24623
fixes #23692

ref
https://nim-lang.org/docs/manual.html#implementation-specific-pragmas-size-pragma

confines `size` pragma to `enums` and imported `objects` for now

The `typeDefLeftSidePass` carries out the check for pragmas, but the
type is not complete yet. So the `size` pragma checking is postponed at
the final pass.
2025-01-23 20:10:14 +01:00
metagn
6d59680217 don't try to transform objconstr/cast type nodes (#24636)
fixes #24631

[Object
constructors](793baf34ff/compiler/semobjconstr.nim (L462)),
[casts](793baf34ff/compiler/semexprs.nim (L494))
and [type
conversions](793baf34ff/compiler/semexprs.nim (L419))
copy their type nodes verbatim instead of producing semchecked type
nodes. This causes a crash in transf when an untyped expression in the
type node has `nil` type. To deal with this, don't try to transform the
type node in these expressions at all. I couldn't reproduce the problem
with type conversion nodes though so those are unchanged in transf.
2025-01-22 14:08:21 +01:00
ringabout
2f402fcb82 fixes #24630; static openArray backed by seq cannot be passed to another function (#24638)
fixes #24630
2025-01-22 14:05:57 +01:00
metagn
793baf34ff generate destructor in nodestroy proc for explicit destructor call (#24627)
fixes #24626

`createTypeboundOps` in sempass2 is called when generating destructors
for types including for explicit destructor calls, however it blocks
destructors from getting generated in a `nodestroy` proc. This causes
issues when a destructor is explicitly called in a `nodestroy` proc. To
fix this, allow destructors to get generated only for explicit
destructor calls in nodestroy procs.
2025-01-20 10:12:38 +01:00
Antonis Geralis
6481482e0e Optimize storing into uninit locations for arrays and seqs. (#24619) 2025-01-19 16:20:54 +03:00
ringabout
2af9ddc286 revert strictDefs as the default (#24620)
revert https://github.com/nim-lang/Nim/pull/24225

see also https://forum.nim-lang.org/t/12646
2025-01-17 16:08:47 +01:00
ringabout
70d057fcc6 fixes compile crashes with one parameter (#24618)
`{.compile("foo.c").}` makes Nim compiler crash
2025-01-16 15:43:18 +01:00
metagn
8d0e853e0a ignore match errors to expected types of tuple constructor elements (#24611)
fixes #24609

A tuple may have an incompatible expected type if there is a converter
match to it. So the compiler should not error when trying to match the
individual elements in the constructor to the elements of the expected
tuple type, this will be checked when the tuple is entirely constructed
anyway.
2025-01-15 20:01:56 +01:00
metagn
d83ff81695 disable sfml test on osx (#24615)
Tried installing sfml 2 in #24614 but didn't work
2025-01-13 10:10:05 +01:00
ringabout
41c447b5f4 ci: update to ubuntu 22.04 (#24608) 2025-01-09 19:52:03 +08:00
Bilog WEB3
26ed469996 Update changelog_1_2_0.md (#24607) 2025-01-09 18:00:11 +08:00
Loïc Bartoletti
4aff12408c math: Add cumprod and cumproded (#23416)
This pull request adds the `cumproded` function along with its in-place
equivalent, `cumprod`, to the math library. These functions provide
functionality similar to `cumsum` and `cumsummed`, allowing users to
calculate the cumulative sum of elements.

The `cumprod` function computes the cumulative product of elements
in-place, while `cumproded` additionally returns the prod seq.
2025-01-09 09:07:59 +01:00
planetBoy
8ed0a63973 docs fix spelling issues (#24597)
Hey !
I fixed several spelling issues.Glad I could help .
Br, Guayaba221.
2025-01-06 13:48:35 +08:00
ringabout
aeeccee50a fixes #24599; misleading error message with large array bounds (#24601)
fixes #24599
2025-01-05 18:10:36 +01:00
Jacek Sieka
e8bf6af0da fix c_memchr, c_strstr definitions (#24587)
One correct definition is enough
2025-01-02 17:28:35 +01:00
Jacek Sieka
78835562b1 varints: no need for emit (#24585) 2025-01-02 17:26:53 +01:00
ringabout
3dda60a8ce Update copyright year 2025 (#24593) 2024-12-31 18:06:55 +08:00
Skylar Ray
dcc4e07e54 chore: docs fix spelling issues (#24581)
**handle - handles**
**sensitiviy - sensitivity**
2024-12-30 19:40:45 +08:00
futreall
0df351bf50 chore: correct typos docs (#24580) 2024-12-28 16:24:46 +01:00
Antonis Geralis
d3b6dba616 Consider iterator types (#24577)
According to the macros doc nnkIteratorTy trees use the same structure
as nnkProcTy
2024-12-28 09:25:49 +01:00
Antonis Geralis
e1be29942e Support tuple parameter types (#24576) 2024-12-28 08:43:41 +01:00
chloefeal
cd220fe3e1 docs: fix typos (#24573)
Signed-off-by: chloefeal <188809157+chloefeal@users.noreply.github.com>
2024-12-27 19:42:18 +01:00
Jake Leahy
86d6f71f5a Doc search improvements (#24567)
- `/` is now a hotkey to jump to the search
- Search results now are in line with the page (previously on small
screens it would be off centre)
- Jumping to a search result inside the page or via TOC will now hide
the search results (previously the results got in the way)

Example site here: https://tranquil-scone-c159b6.netlify.app/main.html
2024-12-25 14:19:22 +01:00
Jake Leahy
5b9ff963c5 Minor std/strscans improvements (#24566)
#### Removes UnInit warnings when using `scanTuple` 

e.g. this would emit a warning
```nim
import std/strscans

proc main() =
  let (ok, number) = "123".scanTuple()
```

![image](https://github.com/user-attachments/assets/68170ac6-402d-48b0-b8b6-69e71f4b70ae)

#### Error for wrong type now points to the passed in variable

```nim
import std/strscans

var str: string
discard "123".scanf("$i", str)
```

it gave this warning before

![image](https://github.com/user-attachments/assets/096e56d2-0eb5-4c67-9725-25caa97afebd)
now it returns

![image](https://github.com/user-attachments/assets/736a4292-2f56-4cf3-a27a-677045377171)
2024-12-25 09:27:12 +01:00
ringabout
65b26401bc adds a test case (#24565)
closes #19531
2024-12-25 09:26:06 +01:00
Esteban C Borsani
2f127bf99f Improve async stacktraces (#24563)
This makes await point to the caller line instead of asyncmacro. It also
reworks the "Async traceback:" section of the traceback. Follow up PR
#21091 (issue #19931) so it works if there is asynchronous work done.
2024-12-25 09:25:28 +01:00
ringabout
3c4246dd24 fixes #23114; Nim v2 regression emit / asm var param dereference inconsistency (#24547)
fixes #23114

As in https://github.com/nim-lang/Nim/pull/22074, expressions in
bracketed emit are strictly typechecked, this PR applies the same check
for symbols in asm statements in order to keep them consistent.
2024-12-25 09:23:25 +01:00
Tomohiro
fc806710cb Add inline assembler tests for i386, arm, arm64, riscv32 and riscv64 (#24564)
This fixes one error in https://github.com/nim-lang/Nim/issues/24544 .
I tested this on Raspberry Pi Pico (arm) and Raspberry Pi 3(arm64).
It is not tested on i386, riscv32 and riscv64 CPU.
2024-12-24 23:11:19 +08:00
ringabout
e2a306355c adds a test case (#24561)
closes #18616
2024-12-23 21:03:18 +08:00
metagn
5c71fbab30 fix jsonutils with generic sandwiches, don't use strformat (#24560)
fixes #24559

The strformat macros have the problem that they don't capture symbols,
so don't use them in the generic `fromJson` proc here. Also `fromJson`
refers to `jsonTo` before it is declared which doesn't capture it, so
it's now forward declared.
2024-12-23 06:08:46 +01:00
Jake Leahy
6bc52737b3 Make 'field is not accessible' and 'field initialized twice' errors point to the field inside the obj construction (#24557)
Fixes two line infos to make the error's clearer inside editors

- 'field is not accessible' would point to the whole object construction
instead of just the field inside the construction
- 'field initialized twice' would point to the colon instead of the
field
2024-12-22 14:25:36 +01:00
Esteban C Borsani
f29234b40f fixes #23212; Asyncdispatch leaks under --mm:arc (#24556)
Fixes #23212

Inspired by [this chronos
PR](https://github.com/status-im/nim-chronos/pull/243)
2024-12-22 07:56:22 +01:00
ringabout
63c884038d remove zippy data from tarballs (#24551)
fixes https://github.com/nim-lang/nightlies/issues/95
2024-12-20 22:09:29 +01:00
metagn
986ca7dcd4 check if unused import warning is enabled before adding import to stack (#24554)
fixes #24552

Could also implement `{.used.}` for imports but this wouldn't be
backwards compatible. The same problem as #24552 also exists for
`{.hint[XDeclaredButNotUsed].}` but this isn't as much of a problem
since `{.used.}`/`{.push used.}` exist.
2024-12-20 22:09:03 +01:00
Daniel Stuart
50ed43df42 Use long int builtins for risc-v 32-bit targets (#24553)
Solves compilation using riscv32-unknown-elf-gcc, compiler defines
int32_t as long int.
2024-12-20 21:27:37 +03:00
ringabout
ce4304ce97 fixes strictdefs warnings (#24550) 2024-12-20 15:26:30 +01:00
Jake Leahy
f80ce139d5 Make expandMacro show private fields (#24522)
Was debugging with `--expandMacro` and noticed that private fields
weren't exported.
Passes extra flags to the renderer to make them be shown
2024-12-17 18:17:04 +01:00
ringabout
d5c7abe3d2 fixes #17681; enforce codegen for exportc consts (#24546)
fixes #17681
2024-12-17 18:16:34 +01:00
Juan M Gómez
8ce58fab26 Adds skipParentCfg back. Bump nimble to a commit where it doesnt rely in the parent config (#24545) 2024-12-17 18:15:48 +01:00
ringabout
70b3232d3a fixes #24536; fixes nightlies regression caused by nimble update (#24542)
follow up #24537

Because `nimble` is a bundled repo so it is bundled in the tarballs

i.e.
82421fd705/.github/workflows/nightlies.yml (L264)
has bundled `dist/nimble`, but it only copies the data without `.git`.
So in this PR, we ignore bundled nimble repo.
2024-12-17 06:42:17 +01:00
ringabout
81d8c0fc17 fixes #20908; Unknown warnings and hints now give a warning (#24543)
fixes #20908

This PR unifies the treatment of unknown warnings and hints, which now
gives an `warnUnknownNotes` instead of an error.
2024-12-17 06:41:38 +01:00
ringabout
91d1933ea2 fixes #24538 (#24541)
fixes #24538
2024-12-16 15:25:44 +01:00
metagn
b9c593404c proper error for const defines with unsupported types (#24540)
fixes #24539
2024-12-16 12:22:10 +01:00
Juan M Gómez
556f217b4c #Fixes #24536 building nimble 0.16.4 fails when running build_all.sh (#24537) 2024-12-16 08:05:00 +01:00
ringabout
d31cce557b more strictdef fixes for stdlibs (#24535) 2024-12-13 19:06:43 +01:00
Juan M Gómez
be4d19e562 Bumps nimble v0.16.4 (#24437) 2024-12-13 19:00:18 +01:00
Ryan McConnell
e0197a8380 couple cases of valid concept bindings (#24513)
see tests

---------

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
2024-12-13 15:13:19 +01:00
ringabout
d2d810585c fixes strictdefs warnings continue (#24520) 2024-12-13 15:04:49 +01:00
ringabout
80af252025 adds a test case (#24534)
closes #16845
2024-12-13 19:30:53 +08:00
bptato
f485973459 Fix exitnow signature, mark as .noreturn (#24533)
Like quit, this function never returns.

Also, "code" was marked as "int", even though POSIX _exit takes a C int.
2024-12-12 22:57:10 +01:00
ringabout
f7a461a30c fixes #22101; std/pegs with nim cpp --mm:orc --exceptions:goto creates invalid C++ (#24531)
fixes #22101

The old implementation generates

`auto T = value;` for the cpp backend which causes problems for goto
exceptions. This PR puts the declaration of `T` to the cpsLocals parts
and makes it compatible with goto exceptions.
2024-12-12 22:56:41 +01:00
ringabout
f796c01e3c adds a test case (#24532)
closes #18070
2024-12-12 22:30:54 +08:00
ringabout
9bb7e53e7f fixes #22153; UB calling allocCStringArray([""]) with --mm:refc (#24529)
fixes #22153

It's a problem for refc because you cannot index a nil string: i.e.
`[""]` is `{((NimStringDesc*) NIM_NIL)}` which cannot be indexed
2024-12-11 21:02:24 +01:00
Jake Leahy
da9f7f671b Make error appear in user code with invalid format string in strformat (#24528)
With this example
```nim
import std/strformat

echo fmt"{invalid, code}"
```

We get the error message
```
stack trace: (most recent call last)
/home/jake/.choosenim/toolchains/nim-hashdevel/lib/pure/strformat.nim(750, 16) fmt
/home/jake/.choosenim/toolchains/nim-hashdevel/lib/pure/strformat.nim(714, 16) strformatImpl
/home/jake/Documents/projects/Nim/temp.nim(3, 9) template/generic instantiation of `fmt` from here
/home/jake/.choosenim/toolchains/nim-hashdevel/lib/pure/strformat.nim(714, 16) Error: could not parse `invalid, code` in `{invalid, code}`.
(1, 8) Error: invalid indentation
```
After PR it now shortens it to just appear in user code
```
/home/jake/Documents/projects/Nim/lib/pure/strformat.nim(750, 16) fmt
/home/jake/Documents/projects/Nim/lib/pure/strformat.nim(714, 16) strformatImpl
/home/jake/Documents/projects/Nim/temp.nim(3, 9) Error: could not parse `invalid, code` in `{invalid, code}`.
(1, 8) Error: invalid indentation
```
2024-12-11 21:01:57 +01:00
Jake Leahy
69e0cdb6c0 Fix line info for import (#24523)
Refs #24158


Fixes the line info of the module symbol (cases like `import as` and
grouped imports had wrong line info). Since that symbol's line info is
now used for the warnings, there isn't a separate line info stored for
`unusedImports`

Examples of fixed cases
```nim
import strutils as test #[
                ^ before
                   ^ after ]#

# This case was fixed by #24158, but only for unused imports
import std/[strutils, strutils] #[
        ^ before
                      ^ after ]#

from strutils import split #[
^ before 
     ^ after ]#
```
2024-12-11 20:55:55 +01:00
metagn
b529f69518 fix nil node in sym ast of exported ref objects [backport:2.2] (#24527)
fixes #24526, follows up #23101

The `shallowCopy` calls do not keep the original node's children, they
just make a new seq with the same length, so the `Ident "*"` node from
the original postfix nodes was not carried over, making it `nil` and
causing the segfault.
2024-12-10 13:20:40 +01:00
metagn
aeb3fe9505 test case haul to prevent pileup (#24525)
closes #6013, closes #7009, closes #9190, closes #12487, closes #12831,
closes #13184, closes #13252, closes #14860, closes #14877, closes
#14894, closes #14917, closes #16153, closes #16439, closes #17779,
closes #18074, closes #18202, closes #18314, closes #18648, closes
#19063, closes #19446, closes #20065, closes #20367, closes #22126,
closes #22820, closes #22888, closes #23020, closes #23287, closes
#23510
2024-12-09 08:11:47 +01:00
ringabout
d408b94063 fixes nightlies regression (#24519)
follows up https://github.com/nim-lang/Nim/pull/24507
2024-12-07 06:35:47 +01:00
ringabout
801733f286 adds a test case (#24518)
closes #19698
2024-12-06 20:19:57 +08:00
metagn
6f4106bf5d track call depth separately from loop count in VM (#24512)
refs #24503

Infinite recursions currently are not tracked separately from infinite
loops, because they also increase the loop counter. However the max
infinite loop count is very high by default (10 million) and does not
reliably catch infinite recursions before consuming a lot of memory. So
to protect against infinite recursions, we separately track call depth,
and add a separate option for the maximum call depth, much lower than
the maximum iteration count by default (2000, the same as
`nimCallDepthLimit`).

---------

Co-authored-by: Andreas Rumpf <rumpf_a@web.de>
2024-12-06 19:00:59 +08:00
ringabout
c0861142f8 fixes strictdefs warnings for stdlibs [part two] (#24514)
After some cleanups for stdlibs, then we should enable warningaserror
for all tests
2024-12-06 05:40:48 +01:00
ringabout
d0288d3b57 fixes #24504; fixes ensureMove for refs (#24505)
fixes #24504
2024-12-05 21:37:21 +01:00
ringabout
02fb0476ce adds a test case (#24515)
closes #17733
2024-12-05 22:19:34 +08:00
ringabout
2e9e7f13ee don't track result initialization if it is marked noinit (#24499)
We don't track `noinit` for variables introduced in
https://github.com/nim-lang/Nim/pull/10566. It should be applied to
`result` if the function is marked `noinit`
2024-12-04 15:12:30 +01:00
Tomohiro
bbf6a62c90 fixes #24506; calculate timeout correctly (#24507)
`curTimeout` is calculated incorrectly. So this PR fixes it.
This PR also replaces `now()` with `getMonoTime()`.
2024-12-04 15:11:32 +01:00
ringabout
6bbf9c3117 remove unnecessary await (#24501)
There is already a when condition, so `await` is not needed to split the
function
2024-12-04 12:03:12 +01:00
ringabout
8f4bfda5f4 fixes some strictdefs warnings (#24502) 2024-12-04 18:28:13 +08:00
ringabout
c3120b6121 adds a test case (#24500)
closes #24040
2024-12-04 12:16:51 +08:00
metagn
2529f33760 remove inserted derefs for ref object fields when transforming to dot call (#24498)
fixes #24492

Kind of a goofy way of doing this, but we count how many derefs were
used for the first parameter before calling `builtinFieldAccess`, then
count after, and if there are more now than before, we remove the added
derefs. I thought maybe getting rid of #18298 would simplify it but
maybe this would still be the best way.

For better encapsulation we could make `dotTransformation` take an
`nOrig` param instead but this would be less efficient since it would
need a copy, though `semAsgn` already makes one.
2024-12-04 12:16:37 +08:00
ringabout
464dc99376 fixes devel version number (#24494)
As said in the documentation, `NimMinor` is supposed to be odd and
should be greater than the stable channel
2024-12-03 08:41:04 +01:00
ringabout
3bee04d9f3 prefix NimDestroyGlobals with nimMainPrefix (#24493)
ref https://github.com/nim-lang/Nim/issues/24471

---------

Co-authored-by: metagn <metagngn@gmail.com>
2024-12-03 13:05:14 +08:00
metagn
33dc2367e7 install older version of nimcuda for arraymancer (#24496)
Attempt to fix CI failure, refs
https://github.com/nim-lang/Nim/pull/24495#issuecomment-2511299112,
alternative is to use a commit version like
bc65375ff5
2024-12-03 06:48:21 +03:00
ringabout
ddf5a9f6c5 adds a test case (#24486)
closes #23680
2024-12-02 06:39:55 +01:00
metagn
05bba15623 fix crash with undeclared proc type pragma macro in generics (#24490)
fixes #24489
2024-12-02 05:21:12 +01:00
ringabout
5340005869 fixes #24476; remove proc type cast if they are same types for backends (#24480)
fixes #24476

closes https://github.com/nim-lang/Nim/pull/24479
2024-11-27 17:36:57 +01:00
Andreas Rumpf
8881017c80 stdlib: minor refactorings and updates (#24482) 2024-11-27 12:31:06 +01:00
ringabout
dcd0793f2b Add support for parsing parameterised sql types (#24483)
Co-authored-by: Cletus Igwe <me@cletusigwe.com>
2024-11-27 12:23:54 +01:00
Ryan McConnell
e479151473 Fixes 3 small issues with concepts (#24481)
issue 1 - statics in the type:
This probably only handles simple cases. It's probably too accepting
only comparing the base, but that should only affect candidate selection
I think.
issue 2 - `tyArray` of length 3:
This is just a work around since I couldn't get the fix right in
previous PR
issue 3 - shadowing:
The part in `concepts.nim` that iterates candidates does not consider
imported idents if at least once module level ident matches. It does not
have to match in any other way then name.

EDIT: 2 more
issue 4 - composite typeclasses:
when declared in both the concept and the `proc` can cause problems
issue 5 - recursion:
simple recursion and scenarios where more than one concepts recurse
together (only tested two)
2024-11-27 10:15:22 +01:00
858 changed files with 43152 additions and 12061 deletions

View File

@@ -10,6 +10,16 @@ body:
Please provide a minimal code example that reproduces the bug if possible.
Reports with a reproducible example or detailed information will likely receive fixes faster.
- type: textarea
id: nim-version
attributes:
label: Nim Version
description: |
Can be obtained from `nim -v` on the command line along with the OS/architecture.
For development versions, including the commit hash may help.
validations:
required: true
- type: textarea
id: description
attributes:
@@ -19,16 +29,6 @@ body:
placeholder: Bug reports with reproducible code or detailed information will be fixed faster.
validations:
required: true
- type: textarea
id: nim-version
attributes:
label: Nim Version
description: |
Can be obtained from `nim -v` on the command line along with the OS/architecture.
For development versions, make sure to include the commit hash.
validations:
required: true
- type: textarea
id: current-logs

11
.github/dependabot.yml vendored Normal file
View File

@@ -0,0 +1,11 @@
# To get started with Dependabot version updates, you'll need to specify which
# package ecosystems to update and where the package manifests are located.
# Please see the documentation for all configuration options:
# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
version: 2
updates:
- package-ecosystem: "github-actions" # See documentation for possible values
directory: "/" # Location of package manifests
schedule:
interval: "weekly"

View File

@@ -15,9 +15,16 @@ jobs:
name: ${{ matrix.platform }}-bisects
runs-on: ${{ matrix.platform }}
steps:
- uses: actions/checkout@v4
- uses: actions/checkout@v6
- uses: jiro4989/setup-nim-action@v1
- name: Install OpenSSL (Windows)
if: |
runner.os == 'Windows'
run: choco install openssl.light --version=1.1.1.0 # OpenSSL 3.x removed SSL_library_init
shell: 'powershell'
# v2 wont work here, because uses "hardcoded" nim versions, action "dynamically" finds version with bug.
- uses: jiro4989/setup-nim-action@v1
with:
nim-version: 'devel'

View File

@@ -41,11 +41,11 @@ jobs:
target: [linux, windows, osx]
include:
- target: linux
os: ubuntu-20.04
os: ubuntu-22.04
- target: windows
os: windows-2019
os: windows-latest
- target: osx
os: macos-13
os: macos-15
name: ${{ matrix.target }}
runs-on: ${{ matrix.os }}
@@ -53,7 +53,7 @@ jobs:
steps:
- name: 'Checkout'
uses: actions/checkout@v4
uses: actions/checkout@v6
with:
fetch-depth: 2
@@ -109,7 +109,7 @@ jobs:
if: |
github.event_name == 'push' && github.ref == 'refs/heads/devel' &&
matrix.target == 'linux'
uses: crazy-max/ghaction-github-pages@v4
uses: crazy-max/ghaction-github-pages@v5
with:
build_dir: doc/html
env:

View File

@@ -18,9 +18,13 @@ jobs:
strategy:
fail-fast: false
matrix:
os: [ubuntu-20.04, macos-13]
cpu: [amd64]
os: [ubuntu-latest, macos-14]
batch: ["allowed_failures", "0_3", "1_3", "2_3"] # list of `index_num`
include:
- os: ubuntu-latest
cpu: amd64
- os: macos-14
cpu: arm64
name: '${{ matrix.os }} (batch: ${{ matrix.batch }})'
runs-on: ${{ matrix.os }}
timeout-minutes: 60 # refs bug #18178
@@ -29,23 +33,23 @@ jobs:
NIM_TESTAMENT_BATCH: ${{ matrix.batch }}
steps:
- name: 'Checkout'
uses: actions/checkout@v4
uses: actions/checkout@v6
with:
fetch-depth: 2
- name: 'Install node.js 20.x'
uses: actions/setup-node@v4
- name: 'Install node.js'
uses: actions/setup-node@v6
with:
node-version: '20.x'
node-version: 24
- name: 'Install dependencies (Linux amd64)'
if: runner.os == 'Linux' && matrix.cpu == 'amd64'
run: |
sudo apt-fast update -qq
sudo apt-get update -qq
DEBIAN_FRONTEND='noninteractive' \
sudo apt-fast install --no-install-recommends -yq \
sudo apt-get install --no-install-recommends -yq \
libcurl4-openssl-dev libgc-dev libsdl1.2-dev libsfml-dev \
valgrind libc6-dbg libblas-dev xorg-dev
valgrind libc6-dbg libblas-dev liblapack-dev libpcre3 xorg-dev
- name: 'Install dependencies (macOS)'
if: runner.os == 'macOS'
run: brew install boehmgc make sfml gtk+3

View File

@@ -11,20 +11,20 @@ jobs:
strategy:
fail-fast: false
matrix:
os: [ubuntu-20.04]
os: [ubuntu-22.04]
cpu: [amd64]
name: '${{ matrix.os }}'
runs-on: ${{ matrix.os }}
steps:
- name: 'Checkout'
uses: actions/checkout@v4
uses: actions/checkout@v6
with:
fetch-depth: 2
- name: 'Install node.js 20.x'
uses: actions/setup-node@v4
- name: 'Install node.js'
uses: actions/setup-node@v6
with:
node-version: '20.x'
node-version: 24
- name: 'Install dependencies (Linux amd64)'
if: runner.os == 'Linux' && matrix.cpu == 'amd64'
@@ -34,17 +34,6 @@ jobs:
sudo apt-fast install --no-install-recommends -yq \
libcurl4-openssl-dev libgc-dev libsdl1.2-dev libsfml-dev \
valgrind libc6-dbg libblas-dev xorg-dev
- name: 'Install dependencies (macOS)'
if: runner.os == 'macOS'
run: brew install boehmgc make sfml gtk+3
- name: 'Install dependencies (Windows)'
if: runner.os == 'Windows'
shell: bash
run: |
set -e
. ci/funs.sh
nimInternalInstallDepsWindows
echo_run echo "${{ github.workspace }}/dist/mingw64/bin" >> "${GITHUB_PATH}"
- name: 'Add build binaries to PATH'
shell: bash
@@ -71,7 +60,7 @@ jobs:
run: nim c -r -d:release ci/action.nim
- name: 'Comment'
uses: actions/github-script@v7
uses: actions/github-script@v9
with:
script: |
const fs = require('fs');

View File

@@ -9,7 +9,7 @@ jobs:
stale:
runs-on: ubuntu-latest
steps:
- uses: actions/stale@v9
- uses: actions/stale@v10
with:
days-before-pr-stale: 365
days-before-pr-close: 30

1
.gitignore vendored
View File

@@ -68,6 +68,7 @@ testament.db
/csources
/csources_v1
/csources_v2
/csources_v3
/dist/
# /lib/fusion # fusion is now unbundled; `git status` should reveal if it's there so users can act on it

View File

@@ -28,24 +28,24 @@ jobs:
# # g++-multilib : Depends: gcc-multilib (>= 4:5.3.1-1ubuntu1) but it is not going to be installed
# vmImage: 'ubuntu-18.04'
# CPU: i386
OSX_amd64:
vmImage: 'macOS-13'
CPU: amd64
OSX_amd64_cpp:
vmImage: 'macOS-13'
CPU: amd64
OSX_arm64:
vmImage: 'macos-15'
CPU: arm64
OSX_arm64_cpp:
vmImage: 'macos-15'
CPU: arm64
NIM_COMPILE_TO_CPP: true
Windows_amd64_batch0_3:
vmImage: 'windows-2019'
vmImage: 'windows-2025'
CPU: amd64
# see also: `NIM_TEST_PACKAGES`
NIM_TESTAMENT_BATCH: "0_3"
Windows_amd64_batch1_3:
vmImage: 'windows-2019'
vmImage: 'windows-2025'
CPU: amd64
NIM_TESTAMENT_BATCH: "1_3"
Windows_amd64_batch2_3:
vmImage: 'windows-2019'
vmImage: 'windows-2025'
CPU: amd64
NIM_TESTAMENT_BATCH: "2_3"

View File

@@ -14,15 +14,89 @@ rounding guarantees (via the
- With `-d:nimPreviewCheckedClose`, the `close` function in the `std/syncio` module now raises an IO exception in case of an error.
- Unknown warnings and hints now gives warnings `warnUnknownNotes` instead of
errors.
- With `-d:nimPreviewAsmSemSymbol`, backticked symbols are type checked in the `asm/emit` statements.
- The bare `except:` now panics on `Defect`. Use `except Exception:` or `except Defect:` to catch `Defect`. `--legacy:noPanicOnExcept` is provided for a transition period.
- With `-d:nimPreviewCStringComparisons`, comparsions (`<`, `>`, `<=`, `>=`) between cstrings switch from reference semantics to value semantics like `==` and `!=`.
- `std/parsesql` has been moved to a nimble package, use `nimble` or `atlas` to install it.
- With `-d:nimPreviewDuplicateModuleError`, importing two modules that share the same name becomes a compile-time error. This includes importing the same module more than once. Use `import foo as foo1` (or other aliases) to avoid collisions.
- Adds the switch `--mangle:nim|cpp`, which selects `nim` or `cpp` style name mangling when used with `debuginfo` on, defaults to `cpp`.
- The second parameter of `succ`, `pred`, `inc`, and `dec` in `system` now accepts `SomeInteger` (previously `Ordinal`).
- Bitshift operators (`shl`, `shr`, `ashr`) now apply bitmasking to the right operand in the C/C++/VM/JS backends.
- Adds a new warning `--warning:ImplicitRangeConversion` that detects downsizing implicit conversions to range types (e.g., `int -> range[0..255]` or `range[1..256] -> range[0..255]`) that could cause runtime panics. Safe conversions like `range[0..255] -> range[0..65535]` and explicit casts do not trigger warnings. `int` to `Natural` and `Positive` conversions do not trigger warnings, which can be enabled with `--warning:systemRangeConversion`.
- Procedure compatibility also checks the backend representation of the
parameter and result types, not just their source-level shape. Use
`--legacy:procParamTypeBackendAliases` to restore the older behavior.
## Standard library additions and changes
[//]: # "Additions:"
- Added `system.readRawDataStable`, a companion to `readRawData` that returns a
raw `ptr UncheckedArray[char]` into a string's character data which stays valid
across moves and copies of the string value. It is available under every string
implementation (refc, ARC/ORC and `--strings:sso`) with the same signature, so
code can pin an interior buffer pointer today and be ready for `--strings:sso`
without `when declared` guards. Under `--strings:sso` it promotes a small inline
string to its heap representation first; under the other implementations the data
is already heap-resident, so it is equivalent to `readRawData`.
- `setutils.symmetricDifference` along with its operator version
`` setutils.`-+-` `` and in-place version `setutils.toggle` have been added
to more efficiently calculate the symmetric difference of bitsets.
- `strutils.multiReplace` overload for character set replacements in a single pass.
Useful for string sanitation. Follows existing multiReplace semantics.
- `std/files` adds:
- Exports `CopyFlag` enum and `FilePermission` type for fine-grained control of file operations
- New file operation procs with `Path` support:
- `getFilePermissions`, `setFilePermissions` for managing permissions
- `tryRemoveFile` for file deletion
- `copyFile` with configurable buffer size and symlink handling
- `copyFileWithPermissions` to preserve file attributes
- `copyFileToDir` for copying files into directories
- `std/dirs` adds:
- New directory operation procs with `Path` support:
- `copyDir` with special file handling options
- `copyDirWithPermissions` to recursively preserve attributes
- `system.setLenUninit` now supports refc, JS and VM backends.
- `system.setLenUninit` for the `string` type. Allows setting length without initializing new memory on growth.
- `std/parseopt` now supports multiple parser modes via a `CliMode` enum.
Modes include `Nim` (default, fully compatible) and two new experimental modes:
`Lax` and `Gnu` for different option parsing behaviors.
- `std/nre2` is added to replace deprecated NRE.
- `system.typeof` adds a new parameter `modifierMode` to specify how type modifiers are handled.
[//]: # "Changes:"
- `std/math` The `^` symbol now supports floating-point as exponent in addition to the Natural type.
- `min`, `max`, and `sequtils`' `minIndex`, `maxIndex` and `minmax` for `openArray`s now accept a comparison function.
- `system.substr` implementation now uses `copymem` (wrapped C `memcpy`) for copying data, if available at compilation.
- `system.newStringUninit` is now considered free of side-effects allowing it to be used with `--experimental:strictFuncs`.
- `std/re` and `std/nre` are deprecated as PCRE library is obsolete.
Use https://github.com/nitely/nim-regex or `std/nre2`.
See: https://github.com/nim-lang/Nim/issues/23668.
- `std/pegs` now correctly lexes UTF-8 bytes inside bare identifier-style
terminals, so case-insensitive matching of non-ASCII terms (e.g. ``\i café``)
works without single-quoting.
- `std/uri`: The `?` operator now appends query parameters to an existing query
string instead of replacing it. Fixes [#19782](https://github.com/nim-lang/Nim/issues/19782).
## Language changes
@@ -63,7 +137,17 @@ rounding guarantees (via the
## Compiler changes
- Fixed a bug where `sizeof(T)` inside a `typedesc` template called from a generic type's
`when` clause would error with "'sizeof' requires '.importc' types to be '.completeStruct'".
The issue was that `hasValuelessStatics` in `semtypinst.nim` didn't recognize
`tyTypeDesc(tyGenericParam)` as an unresolved generic parameter.
## Tool changes
- Added `--raw` flag when generating JSON docs to not render markup.
- Added `--stdinfile` flag to name of the file used when running program from stdin (defaults to `stdinfile.nim`)
- Added `--styleCheck:warning` flag to treat style check violations as warnings.
## Documentation changes
- Added documentation for the `completeStruct` pragma in the manual.

View File

@@ -53,7 +53,7 @@
- [``joyent_http_parser``](https://github.com/nim-lang/joyent_http_parser)
- Proc [toCountTable](https://nim-lang.org/docs/tables.html#toCountTable,openArray[A])
now produces a `CountTable` with values correspoding to the number of occurrences
now produces a `CountTable` with values corresponding to the number of occurrences
of the key in the input. It used to produce a table with all values set to `1`.
Counting occurrences in a sequence used to be:
@@ -339,7 +339,7 @@ for i in a..b:
- Fixed "ReraiseError when using try/except within finally block"
([#5871](https://github.com/nim-lang/Nim/issues/5871))
- Fixed "Range type inference leads to counter-intuitive behvaiour"
- Fixed "Range type inference leads to counter-intuitive behaviour"
([#5854](https://github.com/nim-lang/Nim/issues/5854))
- Fixed "JSON % operator can fail in extern procs with dynamic types"
([#6385](https://github.com/nim-lang/Nim/issues/6385))
@@ -403,7 +403,7 @@ for i in a..b:
([#6589](https://github.com/nim-lang/Nim/issues/6589))
- Fixed "Generated c code calls function twice"
([#6292](https://github.com/nim-lang/Nim/issues/6292))
- Fixed "Range type inference leads to counter-intuitive behvaiour"
- Fixed "Range type inference leads to counter-intuitive behaviour"
([#5854](https://github.com/nim-lang/Nim/issues/5854))
- Fixed "New backward indexing is too limited"
([#6631](https://github.com/nim-lang/Nim/issues/6631))

View File

@@ -22,7 +22,7 @@
- We removed `unicode.Rune16` without any deprecation period as the name
was wrong (see the [RFC](https://github.com/nim-lang/RFCs/issues/151) for details)
and we didn't find any usages of it in the wild. If you still need it, add this
and we didn't find any usage of it in the wild. If you still need it, add this
piece of code to your project:
```nim
type

View File

@@ -11,7 +11,7 @@
* Fixed "Assertion error when running `nim check` on compiler/nim.nim" [#12281](https://github.com/nim-lang/Nim/issues/12281)
* Fixed "Compiler crash with empty array and generic instantiation with int as parameter" [#12264](https://github.com/nim-lang/Nim/issues/12264)
* Fixed "Regression in JS backend codegen "Error: request to generate code for .compileTime proc"" [#12240](https://github.com/nim-lang/Nim/issues/12240)
* Fix how `relativePath` handle case sensitiviy
* Fix how `relativePath` handles case sensitivity
* Fixed "SIGSEGV in compiler when using generic types and seqs" [#12336](https://github.com/nim-lang/Nim/issues/12336)
* Fixed "[1.0.0] weird interaction between `import os` and casting integer to char on macosx trigger bad codegen" [#12291](https://github.com/nim-lang/Nim/issues/12291)
* VM: no special casing for big endian machines
@@ -43,7 +43,7 @@
* threadpool: fix link in docs (#12258)
* Fix spellings (#12277)
* fix #12278, don't expose internal PCRE documentation
* Fixed "Documentation of quitprocs is wrong" [#12279(https://github.com/nim-lang/Nim/issues/12279)
* Fixed "Documentation of quitprocs is wrong" [#12279](https://github.com/nim-lang/Nim/issues/12279)
* Fix typo in docs
* Fix reference to parseSpec proc in readme
* [doc/tut1] removed discard discussion in comments

View File

@@ -169,7 +169,7 @@ echo f
- The Nim compiler now supports a new pragma called ``.localPassc`` to
pass specific compiler options to the C(++) backend for the C(++) file
that was produced from the current Nim module.
- The compiler now inferes "sink parameters". To disable this for a specific routine,
- The compiler now infers "sink parameters". To disable this for a specific routine,
annotate it with `.nosinks`. To disable it for a section of code, use
`{.push sinkInference: off.}`...`{.pop.}`.
- The compiler now supports a new switch `--panics:on` that turns runtime
@@ -261,7 +261,7 @@ echo f
([#12812](https://github.com/nim-lang/Nim/issues/12812))
- Fixed "Produce static/const initializations for variables when possible"
([#12216](https://github.com/nim-lang/Nim/issues/12216))
- Fixed "Assigning descriminator field leads to internal assert with --gc:destructors"
- Fixed "Assigning discriminator field leads to internal assert with --gc:destructors"
([#12821](https://github.com/nim-lang/Nim/issues/12821))
- Fixed "nimsuggest `use` command does not return all instances of symbol"
([#12832](https://github.com/nim-lang/Nim/issues/12832))

View File

@@ -283,7 +283,7 @@ The definition of `"strictFuncs"` was changed.
The old definition was roughly: "A store to a ref/ptr deref is forbidden unless it's coming from a `var T` parameter".
The new definition is: "A store to a ref/ptr deref is forbidden."
This new definition is much easier to understand, the price is some expressitivity. The following code used to be
This new definition is much easier to understand, the price is some expressiveness. The following code used to be
accepted:
```nim

File diff suppressed because it is too large Load Diff

2500
compiler/ast2nif.nim Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -68,8 +68,6 @@ template mdbg*: bool {.deprecated.} =
# ---------------------------------------------------------------------------
proc lookupInRecord*(n: PNode, field: PIdent): PSym
proc mustRehash*(length, counter: int): bool
proc nextTry*(h, maxHash: Hash): Hash {.inline.}
# ------------- table[int, int] ---------------------------------------------
const
@@ -216,10 +214,6 @@ proc getNamedParamFromList*(list: PNode, ident: PIdent): PSym =
proc hashNode(p: RootRef): Hash =
result = hash(cast[pointer](p))
proc mustRehash(length, counter: int): bool =
assert(length > counter)
result = (length * 2 < counter * 3) or (length - counter < 4)
import std/tables
const backrefStyle = "\e[90m"
@@ -484,12 +478,6 @@ proc debug(n: PNode; conf: ConfigRef) =
this.value(n)
echo($this.res)
proc nextTry(h, maxHash: Hash): Hash {.inline.} =
result = ((5 * h) + 1) and maxHash
# For any initial h in range(maxHash), repeating that maxHash times
# generates each int in range(maxHash) exactly once (see any text on
# random-number generation for proof).
proc objectSetContains*(t: TObjectSet, obj: RootRef): bool =
# returns true whether n is in t
var h: Hash = hashNode(obj) and high(t.data) # start with real hash value
@@ -537,95 +525,6 @@ proc objectSetContainsOrIncl*(t: var TObjectSet, obj: RootRef): bool =
inc(t.counter)
result = false
proc strTableContains*(t: TStrTable, n: PSym): bool =
var h: Hash = n.name.h and high(t.data) # start with real hash value
while t.data[h] != nil:
if (t.data[h] == n):
return true
h = nextTry(h, high(t.data))
result = false
proc strTableRawInsert(data: var seq[PSym], n: PSym) =
var h: Hash = n.name.h and high(data)
while data[h] != nil:
if data[h] == n:
# allowed for 'export' feature:
#InternalError(n.info, "StrTableRawInsert: " & n.name.s)
return
h = nextTry(h, high(data))
assert(data[h] == nil)
data[h] = n
proc symTabReplaceRaw(data: var seq[PSym], prevSym: PSym, newSym: PSym) =
assert prevSym.name.h == newSym.name.h
var h: Hash = prevSym.name.h and high(data)
while data[h] != nil:
if data[h] == prevSym:
data[h] = newSym
return
h = nextTry(h, high(data))
assert false
proc symTabReplace*(t: var TStrTable, prevSym: PSym, newSym: PSym) =
symTabReplaceRaw(t.data, prevSym, newSym)
proc strTableEnlarge(t: var TStrTable) =
var n: seq[PSym]
newSeq(n, t.data.len * GrowthFactor)
for i in 0..high(t.data):
if t.data[i] != nil: strTableRawInsert(n, t.data[i])
swap(t.data, n)
proc strTableAdd*(t: var TStrTable, n: PSym) =
if mustRehash(t.data.len, t.counter): strTableEnlarge(t)
strTableRawInsert(t.data, n)
inc(t.counter)
proc strTableInclReportConflict*(t: var TStrTable, n: PSym;
onConflictKeepOld = false): PSym =
# if `t` has a conflicting symbol (same identifier as `n`), return it
# otherwise return `nil`. Incl `n` to `t` unless `onConflictKeepOld = true`
# and a conflict was found.
assert n.name != nil
var h: Hash = n.name.h and high(t.data)
var replaceSlot = -1
while true:
var it = t.data[h]
if it == nil: break
# Semantic checking can happen multiple times thanks to templates
# and overloading: (var x=@[]; x).mapIt(it).
# So it is possible the very same sym is added multiple
# times to the symbol table which we allow here with the 'it == n' check.
if it.name.id == n.name.id:
if it == n: return nil
replaceSlot = h
h = nextTry(h, high(t.data))
if replaceSlot >= 0:
result = t.data[replaceSlot] # found it
if not onConflictKeepOld:
t.data[replaceSlot] = n # overwrite it with newer definition!
return result # but return the old one
elif mustRehash(t.data.len, t.counter):
strTableEnlarge(t)
strTableRawInsert(t.data, n)
else:
assert(t.data[h] == nil)
t.data[h] = n
inc(t.counter)
result = nil
proc strTableIncl*(t: var TStrTable, n: PSym;
onConflictKeepOld = false): bool {.discardable.} =
result = strTableInclReportConflict(t, n, onConflictKeepOld) != nil
proc strTableGet*(t: TStrTable, name: PIdent): PSym =
var h: Hash = name.h and high(t.data)
while true:
result = t.data[h]
if result == nil: break
if result.name.id == name.id: break
h = nextTry(h, high(t.data))
type
TIdentIter* = object # iterator over all syms with same identifier
@@ -713,6 +612,70 @@ iterator items*(tab: TStrTable): PSym =
yield s
s = nextIter(it, tab)
proc isNil(x: ItemId): bool {.inline.} =
x.module == 0 and x.item == 0
proc hasEmptySlot[T](data: TIdPairSeq[T]): bool =
for h in 0..high(data):
if isNil(data[h].key):
return true
result = false
proc idTableRawGet[T](t: TIdTable[T], key: int): int =
var h: Hash
h = key and high(t.data) # start with real hash value
while not isNil(t.data[h].key):
if toId(t.data[h].key) == key:
return h
h = nextTry(h, high(t.data))
result = - 1
proc getOrDefault*[T](t: TIdTable[T], key: ItemId): T =
var index = idTableRawGet(t, toId(key))
if index >= 0: result = t.data[index].val
else: result = default(T)
template idTableGet*[T](t: TIdTable[T], key: PType | PSym): T =
getOrDefault(t, key.itemId)
proc idTableRawInsert[T](data: var TIdPairSeq[T], key: ItemId, val: T) =
var h: Hash
let keyId = toId(key)
h = keyId and high(data)
while not isNil(data[h].key):
assert(toId(data[h].key) != keyId)
h = nextTry(h, high(data))
assert(isNil(data[h].key))
data[h].key = key
data[h].val = val
proc `[]=`*[T](t: var TIdTable[T], key: ItemId, val: T) =
var
index: int
n: TIdPairSeq[T]
index = idTableRawGet(t, toId(key))
if index >= 0:
assert(not isNil(t.data[index].key))
t.data[index].val = val
else:
if mustRehash(t.data.len, t.counter):
newSeq(n, t.data.len * GrowthFactor)
for i in 0..high(t.data):
if not isNil(t.data[i].key):
idTableRawInsert(n, t.data[i].key, t.data[i].val)
assert(hasEmptySlot(n))
swap(t.data, n)
idTableRawInsert(t.data, key, val)
inc(t.counter)
template idTablePut*[T](t: var TIdTable[T], key: PType | PSym, val: T) =
t[key.itemId] = val
iterator idTablePairs*[T](t: TIdTable[T]): tuple[key: ItemId, val: T] =
for i in 0..high(t.data):
if not isNil(t.data[i].key):
yield (t.data[i].key, t.data[i].val)
proc initIITable(x: var TIITable) =
x.counter = 0
newSeq(x.data, StartSize)

1152
compiler/astdef.nim Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -43,13 +43,13 @@ proc flagsToStr[T](flags: set[T]): string =
proc lineInfoToStr*(conf: ConfigRef; info: TLineInfo): string =
result = "["
result.addYamlString(toFilename(conf, info))
result.addf ", $1, $2]", [toLinenumber(info), toColumn(info)]
result.addf ", $1, $2]", toLinenumber(info), toColumn(info)
proc treeToYamlAux(res: var string; conf: ConfigRef; n: PNode; marker: var IntSet; indent, maxRecDepth: int)
proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet; indent, maxRecDepth: int)
proc typeToYamlAux(res: var string; conf: ConfigRef; n: PType; marker: var IntSet; indent, maxRecDepth: int)
proc treeToYamlAux(res: var string; conf: ConfigRef; n: PNode; marker: var IntSet; nl: bool, indent, maxRecDepth: int)
proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet; nl: bool, indent, maxRecDepth: int)
proc typeToYamlAux(res: var string; conf: ConfigRef; n: PType; marker: var IntSet; nl: bool, indent, maxRecDepth: int)
proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet; indent: int; maxRecDepth: int) =
proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet; nl: bool, indent: int; maxRecDepth: int) =
if n == nil:
res.add("null")
elif containsOrIncl(marker, n.id):
@@ -57,10 +57,12 @@ proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet;
else:
let istr = spaces(indent * 4)
if nl:
res.addf("\n$1", istr)
res.addf("kind: $1", [makeYamlString($n.kind)])
res.addf("\n$1name: $2", [istr, makeYamlString(n.name.s)])
res.addf("\n$1typ: ", [istr])
res.typeToYamlAux(conf, n.typ, marker, indent + 1, maxRecDepth - 1)
res.typeToYamlAux(conf, n.typ, marker, true, indent + 1, maxRecDepth - 1)
if conf != nil:
# if we don't pass the config, we probably don't care about the line info
res.addf("\n$1info: $2", [istr, lineInfoToStr(conf, n.info)])
@@ -68,7 +70,7 @@ proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet;
res.addf("\n$1flags: $2", [istr, flagsToStr(n.flags)])
res.addf("\n$1magic: $2", [istr, makeYamlString($n.magic)])
res.addf("\n$1ast: ", [istr])
res.treeToYamlAux(conf, n.ast, marker, indent + 1, maxRecDepth - 1)
res.treeToYamlAux(conf, n.ast, marker, true, indent + 1, maxRecDepth - 1)
res.addf("\n$1options: $2", [istr, flagsToStr(n.options)])
res.addf("\n$1position: $2", [istr, $n.position])
res.addf("\n$1k: $2", [istr, makeYamlString($n.loc.k)])
@@ -76,53 +78,57 @@ proc symToYamlAux(res: var string; conf: ConfigRef; n: PSym; marker: var IntSet;
if card(n.loc.flags) > 0:
res.addf("\n$1flags: $2", [istr, makeYamlString($n.loc.flags)])
res.addf("\n$1snippet: $2", [istr, n.loc.snippet])
res.addf("\n$1lode: $2", [istr])
res.treeToYamlAux(conf, n.loc.lode, marker, indent + 1, maxRecDepth - 1)
res.addf("\n$1lode: ", [istr])
res.treeToYamlAux(conf, n.loc.lode, marker, true, indent + 1, maxRecDepth - 1)
proc typeToYamlAux(res: var string; conf: ConfigRef; n: PType; marker: var IntSet; indent: int; maxRecDepth: int) =
proc typeToYamlAux(res: var string; conf: ConfigRef; n: PType; marker: var IntSet; nl: bool, indent: int; maxRecDepth: int) =
if n == nil:
res.add("null")
elif containsOrIncl(marker, n.id):
res.addf "\"$1 @$2\"" % [$n.kind, strutils.toHex(cast[uint](n), sizeof(n) * 2)]
else:
let istr = spaces(indent * 4)
if nl:
res.addf("\n$1", istr)
res.addf("kind: $2", [istr, makeYamlString($n.kind)])
res.addf("\n$1sym: ")
res.symToYamlAux(conf, n.sym, marker, indent + 1, maxRecDepth - 1)
res.addf("\n$1n: ")
res.treeToYamlAux(conf, n.n, marker, indent + 1, maxRecDepth - 1)
res.addf("\n$1sym: ", istr)
res.symToYamlAux(conf, n.sym, marker, true, indent + 1, maxRecDepth - 1)
res.addf("\n$1n: ", istr)
res.treeToYamlAux(conf, n.n, marker, true, indent + 1, maxRecDepth - 1)
if card(n.flags) > 0:
res.addf("\n$1flags: $2", [istr, flagsToStr(n.flags)])
res.addf("\n$1callconv: $2", [istr, makeYamlString($n.callConv)])
res.addf("\n$1size: $2", [istr, $(n.size)])
res.addf("\n$1align: $2", [istr, $(n.align)])
if n.hasElementType:
res.addf("\n$1sons:")
res.addf("\n$1sons:", istr)
for a in n.kids:
res.addf("\n - ")
res.typeToYamlAux(conf, a, marker, indent + 1, maxRecDepth - 1)
res.addf("\n$1 - ", istr)
res.typeToYamlAux(conf, a, marker, false, indent + 1, maxRecDepth - 1)
proc treeToYamlAux(res: var string; conf: ConfigRef; n: PNode; marker: var IntSet; indent: int;
proc treeToYamlAux(res: var string; conf: ConfigRef; n: PNode; marker: var IntSet; nl: bool, indent: int;
maxRecDepth: int) =
if n == nil:
res.add("null")
else:
var istr = spaces(indent * 4)
if nl:
res.addf("\n$1", istr)
res.addf("kind: $1" % [makeYamlString($n.kind)])
if maxRecDepth != 0:
if conf != nil:
res.addf("\n$1info: $2", [istr, lineInfoToStr(conf, n.info)])
case n.kind
of nkCharLit .. nkInt64Lit:
of nkCharLit .. nkUInt64Lit:
res.addf("\n$1intVal: $2", [istr, $(n.intVal)])
of nkFloatLit, nkFloat32Lit, nkFloat64Lit:
of nkFloatLit .. nkFloat128Lit:
res.addf("\n$1floatVal: $2", [istr, n.floatVal.toStrMaxPrecision])
of nkStrLit .. nkTripleStrLit:
res.addf("\n$1strVal: $2", [istr, makeYamlString(n.strVal)])
of nkSym:
res.addf("\n$1sym: ", [istr])
res.symToYamlAux(conf, n.sym, marker, indent + 1, maxRecDepth)
res.symToYamlAux(conf, n.sym, marker, true, indent + 1, maxRecDepth)
of nkIdent:
if n.ident != nil:
res.addf("\n$1ident: $2", [istr, makeYamlString(n.ident.s)])
@@ -133,22 +139,22 @@ proc treeToYamlAux(res: var string; conf: ConfigRef; n: PNode; marker: var IntSe
res.addf("\n$1sons: ", [istr])
for i in 0 ..< n.len:
res.addf("\n$1 - ", [istr])
res.treeToYamlAux(conf, n[i], marker, indent + 1, maxRecDepth - 1)
res.treeToYamlAux(conf, n[i], marker, false, indent + 1, maxRecDepth - 1)
if n.typ != nil:
res.addf("\n$1typ: ", [istr])
res.typeToYamlAux(conf, n.typ, marker, indent + 1, maxRecDepth)
res.typeToYamlAux(conf, n.typ, marker, true, indent + 1, maxRecDepth)
proc treeToYaml*(conf: ConfigRef; n: PNode; indent: int = 0; maxRecDepth: int = -1): string =
var marker = initIntSet()
result = newStringOfCap(1024)
result.treeToYamlAux(conf, n, marker, indent, maxRecDepth)
result.treeToYamlAux(conf, n, marker, false, indent, maxRecDepth)
proc typeToYaml*(conf: ConfigRef; n: PType; indent: int = 0; maxRecDepth: int = -1): string =
var marker = initIntSet()
result = newStringOfCap(1024)
result.typeToYamlAux(conf, n, marker, indent, maxRecDepth)
result.typeToYamlAux(conf, n, marker, false, indent, maxRecDepth)
proc symToYaml*(conf: ConfigRef; n: PSym; indent: int = 0; maxRecDepth: int = -1): string =
var marker = initIntSet()
result = newStringOfCap(1024)
result.symToYamlAux(conf, n, marker, indent, maxRecDepth)
result.symToYamlAux(conf, n, marker, false, indent, maxRecDepth)

View File

@@ -69,7 +69,7 @@ proc copyHalf[Key, Val](h, result: Node[Key, Val]) =
result.links[j] = h.links[Mhalf + j]
else:
for j in 0..<Mhalf:
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc):
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc) or defined(gcYrc):
result.vals[j] = move h.vals[Mhalf + j]
else:
shallowCopy(result.vals[j], h.vals[Mhalf + j])
@@ -92,7 +92,7 @@ proc insert[Key, Val](h: Node[Key, Val], key: Key, val: Val): Node[Key, Val] =
if less(key, h.keys[j]): break
inc j
for i in countdown(h.entries, j+1):
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc):
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc) or defined(gcYrc):
h.vals[i] = move h.vals[i-1]
else:
shallowCopy(h.vals[i], h.vals[i-1])

View File

@@ -154,14 +154,14 @@ template addField(builder: var Builder, constr: var StructInitializer, name: str
# no name, can just add value
valueBody
of siOrderedStruct:
# no name, can just add value on C
assert name.len != 0, "name has to be given for struct initializer field"
# positional init - name not used in output (empty allowed for anonymous unions)
valueBody
of siNamedStruct:
assert name.len != 0, "name has to be given for struct initializer field"
builder.add(".")
builder.add(name)
builder.add(" = ")
# designated init - empty name for anonymous unions (skips .name = prefix)
if name.len != 0:
builder.add(".")
builder.add(name)
builder.add(" = ")
valueBody
proc finishStructInitializer(builder: var Builder, constr: StructInitializer) =

View File

@@ -141,7 +141,7 @@ proc fixupCall(p: BProc, le, ri: PNode, d: var TLoc,
d.snippet = extract(result)
excl d.flags, lfSingleUse
else:
if d.k == locNone and p.splitDecls == 0:
if d.k == locNone and p.splitDecls == 0 and p.config.exc != excGoto:
d = getTempCpp(p, typ.returnType, extract(result))
else:
if d.k == locNone: d = getTemp(p, typ.returnType)
@@ -230,20 +230,29 @@ proc genOpenArraySlice(p: BProc; q: PNode; formalType, destType: PType; prepareF
of tyString, tySequence:
let atyp = skipTypes(a.t, abstractInst)
if formalType.skipTypes(abstractInst).kind in {tyVar} and atyp.kind == tyString and
optSeqDestructors in p.config.globalOptions:
optSeqDestructors in p.config.globalOptions and not p.config.usesSso():
let bra = byRefLoc(p, a)
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimPrepareStrMutationV2"),
bra)
var val: Snippet
if atyp.kind in {tyVar} and not compileToCpp(p.module):
val = cDeref(ra)
if p.config.usesSso() and
skipTypes(a.t, abstractVar + abstractInst).kind == tyString:
let strPtr = if atyp.kind in {tyVar} and not compileToCpp(p.module): ra
else: addrLoc(p.config, a)
result = (
cCast(ptrType(dest), cOp(Add, NimInt,
cCall(cgsymValue(p.module, "nimStrData"), strPtr), rb)),
lengthExpr)
else:
val = ra
result = (
cIfExpr(dataFieldAccessor(p, val),
cCast(ptrType(dest), cOp(Add, NimInt, dataField(p, val), rb)),
NimNil),
lengthExpr)
var val: Snippet
if atyp.kind in {tyVar} and not compileToCpp(p.module):
val = cDeref(ra)
else:
val = ra
result = (
cIfExpr(dataFieldAccessor(p, val),
cCast(ptrType(dest), cOp(Add, NimInt, dataField(p, val), rb)),
NimNil),
lengthExpr)
else:
result = ("", "")
internalError(p.config, "openArrayLoc: " & typeToString(a.t))
@@ -287,11 +296,22 @@ proc openArrayLoc(p: BProc, formalType: PType, n: PNode; result: var Builder) =
of tyString, tySequence:
let ntyp = skipTypes(n.typ, abstractInst)
if formalType.skipTypes(abstractInst).kind in {tyVar} and ntyp.kind == tyString and
optSeqDestructors in p.config.globalOptions:
optSeqDestructors in p.config.globalOptions and not p.config.usesSso():
let bra = byRefLoc(p, a)
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimPrepareStrMutationV2"),
bra)
if ntyp.kind in {tyVar} and not compileToCpp(p.module):
if p.config.usesSso() and
skipTypes(n.typ, abstractVar + abstractInst).kind == tyString:
if ntyp.kind in {tyVar} and not compileToCpp(p.module):
let ra = a.rdLoc
result.add(cCall(cgsymValue(p.module, "nimStrData"), ra))
result.addArgumentSeparator()
result.add(cCall(cgsymValue(p.module, "nimStrLen"), cDeref(ra)))
else:
result.add(cCall(cgsymValue(p.module, "nimStrData"), addrLoc(p.config, a)))
result.addArgumentSeparator()
result.add(lenExpr(p, a))
elif ntyp.kind in {tyVar} and not compileToCpp(p.module):
let ra = a.rdLoc
var t = TLoc(snippet: cDeref(ra))
let lt = lenExpr(p, t)
@@ -315,9 +335,14 @@ proc openArrayLoc(p: BProc, formalType: PType, n: PNode; result: var Builder) =
let ra = a.rdLoc
var t = TLoc(snippet: cDeref(ra))
let lt = lenExpr(p, t)
result.add(cIfExpr(dataFieldAccessor(p, t.snippet), dataField(p, t.snippet), NimNil))
result.addArgumentSeparator()
result.add(lt)
if p.config.usesSso():
result.add(cCall(cgsymValue(p.module, "nimStrData"), ra))
result.addArgumentSeparator()
result.add(cCall(cgsymValue(p.module, "nimStrLen"), t.snippet))
else:
result.add(cIfExpr(dataFieldAccessor(p, t.snippet), dataField(p, t.snippet), NimNil))
result.addArgumentSeparator()
result.add(lt)
of tyArray:
let ra = rdLoc(a)
result.add(ra)
@@ -331,20 +356,21 @@ proc withTmpIfNeeded(p: BProc, a: TLoc, needsTmp: bool): TLoc =
# Bug https://github.com/status-im/nimbus-eth2/issues/1549
# Aliasing is preferred over stack overflows.
# Also don't regress for non ARC-builds, too risky.
if needsTmp and a.lode.typ != nil and p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc} and
if needsTmp and a.lode.typ != nil and p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc} and
getSize(p.config, a.lode.typ) < 1024:
result = getTemp(p, a.lode.typ, needsInit=false)
genAssignment(p, result, a, {})
else:
result = a
proc literalsNeedsTmp(p: BProc, a: TLoc): TLoc =
proc expressionsNeedsTmp(p: BProc, a: TLoc): TLoc =
result = getTemp(p, a.lode.typ, needsInit=false)
genAssignment(p, result, a, {})
proc genArgStringToCString(p: BProc, n: PNode; result: var Builder; needsTmp: bool) {.inline.} =
var a = initLocExpr(p, n[0])
let ra = withTmpIfNeeded(p, a, needsTmp).rdLoc
let tmp = withTmpIfNeeded(p, a, needsTmp)
let ra = if p.config.usesSso(): byRefLoc(p, tmp) else: tmp.rdLoc
result.addCall(cgsymValue(p.module, "nimToCStringConv"), ra)
proc genArg(p: BProc, n: PNode, param: PSym; call: PNode; result: var Builder; needsTmp = false) =
@@ -358,7 +384,7 @@ proc genArg(p: BProc, n: PNode, param: PSym; call: PNode; result: var Builder; n
(optByRef notin param.options or not p.module.compileToCpp):
a = initLocExpr(p, n)
if n.kind in {nkCharLit..nkNilLit}:
addAddrLoc(p.config, literalsNeedsTmp(p, a), result)
addAddrLoc(p.config, expressionsNeedsTmp(p, a), result)
else:
addAddrLoc(p.config, withTmpIfNeeded(p, a, needsTmp), result)
elif p.module.compileToCpp and param.typ.kind in {tyVar} and
@@ -368,8 +394,8 @@ proc genArg(p: BProc, n: PNode, param: PSym; call: PNode; result: var Builder; n
# variable. Thus, we create a temporary pointer variable instead.
let needsIndirect = mapType(p.config, n[0].typ, mapTypeChooser(n[0]) == skParam) != ctArray
if needsIndirect:
n.typ() = n.typ.exactReplica
n.typ.flags.incl tfVarIsPtr
n.typ = n.typ.exactReplica(p.module.idgen)
n.typ.incl tfVarIsPtr
a = initLocExprSingleUse(p, n)
a = withTmpIfNeeded(p, a, needsTmp)
if needsIndirect: a.flags.incl lfIndirect
@@ -418,35 +444,6 @@ proc skipTrivialIndirections(n: PNode): PNode =
result = result[1]
else: break
proc getPotentialWrites(n: PNode; mutate: bool; result: var seq[PNode]) =
case n.kind:
of nkLiterals, nkIdent, nkFormalParams: discard
of nkSym:
if mutate: result.add n
of nkAsgn, nkFastAsgn, nkSinkAsgn:
getPotentialWrites(n[0], true, result)
getPotentialWrites(n[1], mutate, result)
of nkAddr, nkHiddenAddr:
getPotentialWrites(n[0], true, result)
of nkBracketExpr, nkDotExpr, nkCheckedFieldExpr:
getPotentialWrites(n[0], mutate, result)
of nkCallKinds:
case n.getMagic:
of mIncl, mExcl, mInc, mDec, mAppendStrCh, mAppendStrStr, mAppendSeqElem,
mAddr, mNew, mNewFinalize, mWasMoved, mDestroy:
getPotentialWrites(n[1], true, result)
for i in 2..<n.len:
getPotentialWrites(n[i], mutate, result)
of mSwap:
for i in 1..<n.len:
getPotentialWrites(n[i], true, result)
else:
for i in 1..<n.len:
getPotentialWrites(n[i], mutate, result)
else:
for s in n:
getPotentialWrites(s, mutate, result)
proc getPotentialReads(n: PNode; result: var seq[PNode]) =
case n.kind:
of nkLiterals, nkIdent, nkFormalParams: discard
@@ -527,7 +524,7 @@ proc genClosureCall(p: BProc, le, ri: PNode, d: var TLoc) =
else:
cCall(p, params, e)
cIfExpr(e,
eCall,
eCall,
cCall(cCast(pTyp, p), params))
template callIter(rp, params: Snippet): Snippet =
@@ -912,6 +909,16 @@ proc isInactiveDestructorCall(p: BProc, e: PNode): bool =
proc genAsgnCall(p: BProc, le, ri: PNode, d: var TLoc) =
if p.withinBlockLeaveActions > 0 and isInactiveDestructorCall(p, ri):
return
when defined(icDbgHash):
if ri[0].typ == nil:
echo "NILCALLEE kind=", ri[0].kind,
" sym=", (if ri[0].kind == nkSym: ri[0].sym.name.s else: "-"),
" symKind=", (if ri[0].kind == nkSym: $ri[0].sym.kind else: "-"),
" flags=", (if ri[0].kind == nkSym: $ri[0].sym.flags else: "-"),
" lazy=", nfLazyType in ri[0].flags,
" inProc=", (if p.prc != nil: p.prc.name.s else: "NIL"),
" module=", p.module.module.name.s
raiseAssert "nil callee type, see NILCALLEE above"
if ri[0].typ.skipTypes({tyGenericInst, tyAlias, tySink, tyOwned}).callConv == ccClosure:
genClosureCall(p, le, ri, d)
elif ri[0].kind == nkSym and sfInfixCall in ri[0].sym.flags:

View File

@@ -170,7 +170,7 @@ proc canMove(p: BProc, n: PNode; dest: TLoc): bool =
template simpleAsgn(builder: var Builder, dest, src: TLoc) =
let rd = rdLoc(dest)
let rs = rdLoc(src)
builder.addAssignment(rd, rs)
builder.addAssignment(rd, rs)
proc genRefAssign(p: BProc, dest, src: TLoc) =
if (dest.storage == OnStack and p.config.selectedGC != gcGo) or not usesWriteBarrier(p.config):
@@ -216,6 +216,8 @@ proc genOptAsgnTuple(p: BProc, dest, src: TLoc, flags: TAssignmentFlags) =
flags
let t = skipTypes(dest.t, abstractInst).getUniqueType()
for i, t in t.ikids:
# Do not produce code for void types
if isEmptyType(t): continue
let field = "Field$1" % [i.rope]
genAssignment(p, optAsgnLoc(dest, t, field),
optAsgnLoc(src, t, field), newflags)
@@ -318,12 +320,16 @@ proc genOpenArrayConv(p: BProc; d: TLoc; a: TLoc; flags: TAssignmentFlags) =
p.s(cpsStmts).addCallStmt(
cgsymValue(p.module, "nimPrepareStrMutationV2"),
bra)
let rd = d.rdLoc
let ra = a.rdLoc
p.s(cpsStmts).addFieldAssignment(rd, "Field0",
cIfExpr(dataFieldAccessor(p, ra), dataField(p, ra), NimNil))
let la = lenExpr(p, a)
if p.config.usesSso():
let bra = byRefLoc(p, a)
p.s(cpsStmts).addFieldAssignment(rd, "Field0",
cCall(cgsymValue(p.module, "nimStrData"), bra))
else:
let ra = a.rdLoc
p.s(cpsStmts).addFieldAssignment(rd, "Field0",
cIfExpr(dataFieldAccessor(p, ra), dataField(p, ra), NimNil))
p.s(cpsStmts).addFieldAssignment(rd, "Field1", la)
else:
internalError(p.config, a.lode.info, "cannot handle " & $a.t.kind)
@@ -357,7 +363,7 @@ proc genAssignment(p: BProc, dest, src: TLoc, flags: TAssignmentFlags) =
of tyString:
if optSeqDestructors in p.config.globalOptions:
genGenericAsgn(p, dest, src, flags)
elif ({needToCopy, needToCopySinkParam} * flags == {} and src.storage != OnStatic) or canMove(p, src.lode, dest):
elif (needToCopy notin flags and src.storage != OnStatic) or canMove(p, src.lode, dest):
genRefAssign(p, dest, src)
else:
if (dest.storage == OnStack and p.config.selectedGC != gcGo) or not usesWriteBarrier(p.config):
@@ -416,7 +422,7 @@ proc genAssignment(p: BProc, dest, src: TLoc, flags: TAssignmentFlags) =
else:
simpleAsgn(p.s(cpsStmts), dest, src)
of tyArray:
if containsGarbageCollectedRef(dest.t) and p.config.selectedGC notin {gcArc, gcAtomicArc, gcOrc, gcHooks}:
if containsGarbageCollectedRef(dest.t) and p.config.selectedGC notin {gcArc, gcAtomicArc, gcOrc, gcYrc, gcHooks}:
genGenericAsgn(p, dest, src, flags)
else:
let rd = rdLoc(dest)
@@ -675,7 +681,7 @@ proc binaryArithOverflow(p: BProc, e: PNode, d: var TLoc, m: TMagic) =
if e[2].kind in {nkIntLit..nkInt64Lit}:
needsOverflowCheck = e[2].intVal == -1
if canBeZero:
# remove extra paren from `==` op here to avoid Wparentheses-equality:
# remove extra paren from `==` op here to avoid Wparentheses-equality:
p.s(cpsStmts).addSingleIfStmt(removeSinglePar(cOp(Equal, rdLoc(b), cIntValue(0)))):
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "raiseDivByZero"))
raiseInstr(p, p.s(cpsStmts))
@@ -696,7 +702,7 @@ proc unaryArithOverflow(p: BProc, e: PNode, d: var TLoc, m: TMagic) =
let ra = rdLoc(a)
if optOverflowCheck in p.options:
let first = cIntLiteral(firstOrd(p.config, t))
# remove extra paren from `==` op here to avoid Wparentheses-equality:
# remove extra paren from `==` op here to avoid Wparentheses-equality:
p.s(cpsStmts).addSingleIfStmt(removeSinglePar(cOp(Equal, ra, first))):
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "raiseOverflow"))
raiseInstr(p, p.s(cpsStmts))
@@ -749,16 +755,16 @@ proc binaryArith(p: BProc, e: PNode, d: var TLoc, op: TMagic) =
let t = getType()
let at = cUintType(k)
let bt = cUintType(s)
res = cCast(t, cOp(Shr, at, cCast(at, ra), cCast(bt, rb)))
res = cCast(t, cOp(Shr, at, cCast(at, ra), cOp(BitAnd, at, cCast(bt, rb), cIntLiteral(k - 1))))
of mShlI:
let t = getType()
let at = cUintType(s)
res = cCast(t, cOp(Shl, at, cCast(at, ra), cCast(at, rb)))
res = cCast(t, cOp(Shl, at, cCast(at, ra), cOp(BitAnd, at, cCast(at, rb), cIntLiteral(k - 1))))
of mAshrI:
let t = getType()
let at = cIntType(s)
let bt = cUintType(s)
res = cCast(t, cOp(Shr, at, cCast(at, ra), cCast(bt, rb)))
res = cCast(t, cOp(Shr, at, cCast(at, ra), cOp(BitAnd, at, cCast(bt, rb), cIntLiteral(k - 1))))
of mBitandI:
let t = getType()
res = cCast(t, cOp(BitAnd, t, ra, rb))
@@ -919,6 +925,10 @@ proc genDeref(p: BProc, e: PNode, d: var TLoc) =
return
else:
a = initLocExprSingleUse(p, e[0])
# bug #23453 #25265
if e.typ != nil and e.typ.skipTypes(abstractInst).kind == tyObject:
discard getTypeDesc(p.module, e.typ)
if d.k == locNone:
# dest = *a; <-- We do not know that 'dest' is on the heap!
# It is completely wrong to set 'd.storage' here, unless it's not yet
@@ -952,7 +962,8 @@ proc genDeref(p: BProc, e: PNode, d: var TLoc) =
putIntoDest(p, d, e, cDeref(rdLoc(a)), a.storage)
proc cowBracket(p: BProc; n: PNode) =
if n.kind == nkBracketExpr and optSeqDestructors in p.config.globalOptions:
if n.kind == nkBracketExpr and optSeqDestructors in p.config.globalOptions and
not p.config.usesSso():
let strCandidate = n[0]
if strCandidate.typ.skipTypes(abstractInst).kind == tyString:
var a: TLoc = initLocExpr(p, strCandidate)
@@ -962,6 +973,11 @@ proc cowBracket(p: BProc; n: PNode) =
proc cow(p: BProc; n: PNode) {.inline.} =
if n.kind == nkHiddenAddr: cowBracket(p, n[0])
template ignoreConv(e: PNode): bool =
let destType = e.typ.skipTypes({tyVar, tyLent, tyGenericInst, tyAlias, tySink})
let srcType = e[1].typ.skipTypes({tyVar, tyLent, tyGenericInst, tyAlias, tySink})
sameBackendTypePickyAliases(destType, srcType)
proc genAddr(p: BProc, e: PNode, d: var TLoc) =
# careful 'addr(myptrToArray)' needs to get the ampersand:
if e[0].typ.skipTypes(abstractInstOwned).kind in {tyRef, tyPtr}:
@@ -973,8 +989,18 @@ proc genAddr(p: BProc, e: PNode, d: var TLoc) =
# bug #19497
d.lode = e
else:
var a: TLoc = initLocExpr(p, e[0])
putIntoDest(p, d, e, addrLoc(p.config, a), a.storage)
let ssoStrSub = p.config.usesSso() and e[0].kind == nkBracketExpr and
e[0][0].typ.skipTypes(abstractVar).kind == tyString
var a: TLoc = initLocExpr(p, e[0], if ssoStrSub: {lfEnforceDeref, lfPrepareForMutation} else: {})
if e[0].kind in {nkHiddenStdConv, nkHiddenSubConv, nkConv} and not ignoreConv(e[0]):
# addr (conv x) introduces a temp because `conv x` is not a rvalue
# transform addr ( conv ( x ) ) -> conv ( addr ( x ) )
var exprLoc: TLoc = initLocExpr(p, e[0][1])
var tmp = getTemp(p, e.typ, needsInit=false)
putIntoDest(p, tmp, e, cCast(getTypeDesc(p.module, e.typ), addrLoc(p.config, exprLoc)))
putIntoDest(p, d, e, rdLoc(tmp))
else:
putIntoDest(p, d, e, addrLoc(p.config, a), a.storage)
template inheritLocation(d: var TLoc, a: TLoc) =
if d.k == locNone: d.storage = a.storage
@@ -989,7 +1015,7 @@ proc genTupleElem(p: BProc, e: PNode, d: var TLoc) =
var
i: int = 0
var a: TLoc = initLocExpr(p, e[0])
let tupType = a.t.skipTypes(abstractInst+{tyVar})
let tupType = a.t.skipTypes(abstractInst+{tyVar}+tyUserTypeClasses) # ref #25227
assert tupType.kind == tyTuple
d.inheritLocation(a)
discard getTypeDesc(p.module, a.t) # fill the record's fields.loc
@@ -1292,13 +1318,24 @@ proc genSeqElem(p: BProc, n, x, y: PNode, d: var TLoc) =
if skipTypes(a.t, abstractVar).kind in {tyRef, tyPtr}:
a.snippet = cDeref(a.snippet)
if lfPrepareForMutation in d.flags and ty.kind == tyString and
optSeqDestructors in p.config.globalOptions:
if p.config.usesSso() and ty.kind == tyString:
let bra = byRefLoc(p, a)
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimPrepareStrMutationV2"),
bra)
let ra = rdLoc(a)
putIntoDest(p, d, n, subscript(dataField(p, ra), rcb), a.storage)
if lfPrepareForMutation in d.flags:
# Use nimStrAtMutV3 to get a mutable reference (char*) to the element.
# Only when mutation is requested: avoids calling nimPrepareStrMutationV2
# on const string literals (which would SIGSEGV on write to read-only memory).
putIntoDest(p, d, n,
cDeref(cCall(cgsymValue(p.module, "nimStrAtMutV3"), bra, rcb)), a.storage)
else:
putIntoDest(p, d, n,
cCall(cgsymValue(p.module, "nimStrAtV3"), bra, rcb), a.storage)
else:
if lfPrepareForMutation in d.flags and ty.kind == tyString and
optSeqDestructors in p.config.globalOptions:
let bra = byRefLoc(p, a)
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimPrepareStrMutationV2"), bra)
let ra = rdLoc(a)
putIntoDest(p, d, n, subscript(dataField(p, ra), rcb), a.storage)
proc genBracketExpr(p: BProc; n: PNode; d: var TLoc) =
var ty = skipTypes(n[0].typ, abstractVarRange + tyUserTypeClasses)
@@ -1570,6 +1607,51 @@ proc genSeqElemAppend(p: BProc, e: PNode, d: var TLoc) =
genAssignment(p, dest, b, {needToCopy})
gcUsage(p.config, e)
proc genSeqElemAppendV2(p: BProc, e: PNode, d: var TLoc) =
# s.add(x) with optSeqDestructors (arc/orc), inlined for direct slot construction:
# NI oldLen = s.len;
# if (s.p == NIM_NIL || (s.p->cap & ~NIM_STRLIT_FLAG) < oldLen + 1)
# s.p = (PayloadType*)prepareSeqAddUninit(oldLen, s.p, 1, sizeof(T), alignof(T));
# s.len = oldLen + 1;
# s.p->data[oldLen] = x; // direct assignment, no function call overhead
let seqtype = skipTypes(e[1].typ, abstractVarRange)
var a = initLocExpr(p, e[1])
let pt = getSeqPayloadType(p.module, seqtype)
let pe = seqPayloadElem(p.module, seqtype)
# Capture a stable pointer to the seq BEFORE evaluating the element (e[2]).
# Evaluating e[2] may emit move semantics (eqwasMoved) that nil a variable
# through which e[1]'s snippet is accessed (e.g. a closure env pointer).
inc(p.labels)
let seqPtrName = "T" & rope(p.labels) & "_"
p.s(cpsLocals).addVar(kind = Local, name = seqPtrName,
typ = ptrType(getTypeDesc(p.module, seqtype)))
p.s(cpsStmts).addAssignment(seqPtrName, cAddr(rdLoc(a)))
var b = initLocExpr(p, e[2])
# All seq operations now go through the stable seqPtrName pointer.
let ra = wrapPar(cDeref(seqPtrName))
var tmpL = getIntTemp(p)
p.s(cpsStmts).addAssignment(tmpL.snippet, dotField(ra, "len"))
let pField = dotField(ra, "p")
p.s(cpsStmts).addSingleIfStmt(
cOp(Or,
cOp(Equal, pField, NimNil),
cOp(LessThan,
cOp(BitAnd, NimInt, derefField(pField, "cap"), cOp(BitNot, NimInt, NimStrlitFlag)),
cOp(Add, NimInt, tmpL.snippet, cIntValue(1))))):
p.s(cpsStmts).addFieldAssignmentWithValue(ra, "p"):
p.s(cpsStmts).addCast(ptrType(pt)):
p.s(cpsStmts).addCall(cgsymValue(p.module, "prepareSeqAddUninit"),
tmpL.snippet,
pField,
cIntValue(1),
cSizeof(pe),
cAlignof(pe))
p.s(cpsStmts).addFieldAssignment(ra, "len",
cOp(Add, NimInt, tmpL.snippet, cIntValue(1)))
var dest = initLoc(locExpr, e[2], OnHeap)
dest.snippet = subscript(dataField(p, ra), tmpL.snippet)
genAssignment(p, dest, b, {})
proc genDefault(p: BProc; n: PNode; d: var TLoc) =
if d.k == locNone: d = getTemp(p, n.typ, needsInit=true)
else: resetLoc(p, d)
@@ -1705,15 +1787,15 @@ proc genNewSeq(p: BProc, e: PNode) =
let seqtype = skipTypes(e[1].typ, abstractVarRange)
let ra = a.rdLoc
let rb = b.rdLoc
let et = getTypeDesc(p.module, seqtype.elementType)
let pt = getSeqPayloadType(p.module, seqtype)
let pe = seqPayloadElem(p.module, seqtype)
p.s(cpsStmts).addFieldAssignment(ra, "len", rb)
p.s(cpsStmts).addFieldAssignmentWithValue(ra, "p"):
p.s(cpsStmts).addCast(ptrType(pt)):
p.s(cpsStmts).addCall(cgsymValue(p.module, "newSeqPayload"),
rb,
cSizeof(et),
cAlignof(et))
cSizeof(pe),
cAlignof(pe))
else:
let lenIsZero = e[2].kind == nkIntLit and e[2].intVal == 0
genNewSeqAux(p, a, b.rdLoc, lenIsZero)
@@ -1726,15 +1808,15 @@ proc genNewSeqOfCap(p: BProc; e: PNode; d: var TLoc) =
if d.k == locNone: d = getTemp(p, e.typ, needsInit=false)
let rd = d.rdLoc
let ra = a.rdLoc
let et = getTypeDesc(p.module, seqtype.elementType)
let pt = getSeqPayloadType(p.module, seqtype)
let pe = seqPayloadElem(p.module, seqtype)
p.s(cpsStmts).addFieldAssignment(rd, "len", cIntValue(0))
p.s(cpsStmts).addFieldAssignmentWithValue(rd, "p"):
p.s(cpsStmts).addCast(ptrType(pt)):
p.s(cpsStmts).addCall(cgsymValue(p.module, "newSeqPayloadUninit"),
ra,
cSizeof(et),
cAlignof(et))
cSizeof(pe),
cAlignof(pe))
else:
if d.k == locNone: d = getTemp(p, e.typ, needsInit=false) # bug #22560
let ra = a.rdLoc
@@ -1807,15 +1889,24 @@ proc genObjConstr(p: BProc, e: PNode, d: var TLoc) =
var t = e.typ.skipTypes(abstractInstOwned)
let isRef = t.kind == tyRef
# check if we need to construct the object in a temporary
# check if we need to construct the object in a temporary.
# A temp is needed when:
# - the constructor produces a ref (isRef)
# - the destination is not a writable location (d.k == locNone)
# - the constructed type differs from the destination type (subtype
# assignments need the genAssignment path for ObjectAssignmentDefect)
# - the constructor's field values may alias the destination (isPartOf)
var useTemp =
isRef or
(d.k notin {locTemp,locLocalVar,locGlobalVar,locParam,locField}) or
d.k == locNone or
(d.t != nil and not sameBackendType(t, d.t.skipTypes(abstractInstOwned))) or
(isPartOf(d.lode, e) != arNo)
var tmp: TLoc = default(TLoc)
var r: Rope
let needsZeroMem = p.config.selectedGC notin {gcArc, gcAtomicArc, gcOrc} or nfAllFieldsSet notin e.flags
let needsZeroMem =
nfAllFieldsSet notin e.flags or
(optSeqDestructors notin p.config.globalOptions and containsGarbageCollectedRef(t))
if useTemp:
tmp = getTemp(p, t)
r = rdLoc(tmp)
@@ -1870,15 +1961,15 @@ proc genSeqConstr(p: BProc, n: PNode, d: var TLoc) =
if optSeqDestructors in p.config.globalOptions:
let seqtype = n.typ
let rd = rdLoc dest[]
let et = getTypeDesc(p.module, seqtype.elementType)
let pt = getSeqPayloadType(p.module, seqtype)
let pe = seqPayloadElem(p.module, seqtype)
p.s(cpsStmts).addFieldAssignment(rd, "len", lit)
p.s(cpsStmts).addFieldAssignmentWithValue(rd, "p"):
p.s(cpsStmts).addCast(ptrType(pt)):
p.s(cpsStmts).addCall(cgsymValue(p.module, "newSeqPayload"),
lit,
cSizeof(et),
cAlignof(et))
cSizeof(pe),
cAlignof(pe))
else:
# generate call to newSeq before adding the elements per hand:
genNewSeqAux(p, dest[], lit, n.len == 0)
@@ -1899,7 +1990,7 @@ proc genSeqConstr(p: BProc, n: PNode, d: var TLoc) =
proc genArrToSeq(p: BProc, n: PNode, d: var TLoc) =
var elem, arr: TLoc
if n[1].kind == nkBracket:
n[1].typ() = n.typ
n[1].typ = n.typ
genSeqConstr(p, n[1], d)
return
if d.k == locNone:
@@ -1911,15 +2002,15 @@ proc genArrToSeq(p: BProc, n: PNode, d: var TLoc) =
let seqtype = n.typ
let rd = rdLoc d
let valL = cIntValue(L)
let et = getTypeDesc(p.module, seqtype.elementType)
let pt = getSeqPayloadType(p.module, seqtype)
let pe = seqPayloadElem(p.module, seqtype)
p.s(cpsStmts).addFieldAssignment(rd, "len", valL)
p.s(cpsStmts).addFieldAssignmentWithValue(rd, "p"):
p.s(cpsStmts).addCast(ptrType(pt)):
p.s(cpsStmts).addCall(cgsymValue(p.module, "newSeqPayload"),
valL,
cSizeof(et),
cAlignof(et))
cSizeof(pe),
cAlignof(pe))
else:
let lit = cIntLiteral(L)
genNewSeqAux(p, d, lit, L == 0)
@@ -2060,12 +2151,20 @@ proc genRepr(p: BProc, e: PNode, d: var TLoc) =
let ra = rdLoc(a)
putIntoDest(p, b, e, ra & cArgumentSeparator & ra & "Len_0", a.storage)
of tyString, tySequence:
let ra = rdLoc(a)
let la = lenExpr(p, a)
putIntoDest(p, b, e,
cIfExpr(dataFieldAccessor(p, ra), dataField(p, ra), NimNil) &
cArgumentSeparator & la,
a.storage)
if p.config.usesSso() and
skipTypes(a.t, abstractVarRange).kind == tyString:
let bra = byRefLoc(p, a)
putIntoDest(p, b, e,
cCall(cgsymValue(p.module, "nimStrData"), bra) &
cArgumentSeparator & la,
a.storage)
else:
let ra = rdLoc(a)
putIntoDest(p, b, e,
cIfExpr(dataFieldAccessor(p, ra), dataField(p, ra), NimNil) &
cArgumentSeparator & la,
a.storage)
of tyArray:
let ra = rdLoc(a)
let la = cIntValue(lengthOrd(p.config, a.t))
@@ -2198,7 +2297,14 @@ proc genArrayLen(p: BProc, e: PNode, d: var TLoc, op: TMagic) =
else: putIntoDest(p, d, e, cIntValue(lengthOrd(p.config, typ)))
else: internalError(p.config, e.info, "genArrayLen()")
proc genSetLengthSeq(p: BProc, e: PNode, d: var TLoc) =
proc isTrivialTypesToSnippet(t: PType): Snippet =
if containsGarbageCollectedRef(t) or
hasDestructor(t):
result = NimFalse
else:
result = NimTrue
proc genSetLengthSeq(p: BProc, e: PNode, d: var TLoc, noinit = false) =
if optSeqDestructors in p.config.globalOptions:
e[1] = makeAddr(e[1], p.module.idgen)
genCall(p, e, d)
@@ -2220,7 +2326,10 @@ proc genSetLengthSeq(p: BProc, e: PNode, d: var TLoc) =
pExpr = cIfExpr(ra, cAddr(derefField(ra, "Sup")), NimNil)
else:
pExpr = ra
call.snippet = cCast(rt, cgCall(p, "setLengthSeqV2", pExpr, rti, rb))
let name = if noinit: "setLengthSeqUninit" else: "setLengthSeqV2"
call.snippet = cCast(rt, cgCall(p, name, pExpr, rti, rb,
isTrivialTypesToSnippet(t.skipTypes(abstractInst)[0])))
genAssignment(p, a, call, {})
gcUsage(p.config, e)
@@ -2520,6 +2629,8 @@ proc genSomeCast(p: BProc, e: PNode, d: var TLoc) =
putIntoDest(p, d, e, cDeref(cCast(ptrType(destTyp), wrapPar(cAddr(val)))), a.storage)
elif etyp.kind == tyBool and srcTyp.kind in IntegralTypes:
putIntoDest(p, d, e, cOp(NotEqual, rdCharLoc(a), cIntValue(0)), a.storage)
elif etyp.kind == tyProc and srcTyp.kind == tyProc and sameBackendType(etyp, srcTyp):
expr(p, e[1], d)
else:
if etyp.kind == tyPtr:
# generates the definition of structs for casts like cast[ptr object](addr x)[]
@@ -2627,17 +2738,16 @@ proc genRangeChck(p: BProc, n: PNode, d: var TLoc) =
putIntoDest(p, d, n, cCast(destType, wrapPar(val)), a.storage)
proc genConv(p: BProc, e: PNode, d: var TLoc) =
let destType = e.typ.skipTypes({tyVar, tyLent, tyGenericInst, tyAlias, tySink})
if sameBackendTypeIgnoreRange(destType, e[1].typ):
if ignoreConv(e):
expr(p, e[1], d)
else:
genSomeCast(p, e, d)
proc convStrToCStr(p: BProc, n: PNode, d: var TLoc) =
var a: TLoc = initLocExpr(p, n[0])
let arg = if p.config.usesSso(): byRefLoc(p, a) else: rdLoc(a)
putIntoDest(p, d, n,
cgCall(p, "nimToCStringConv", rdLoc(a)),
# "($1 ? $1->data : (NCSTRING)\"\")" % [a.rdLoc],
cgCall(p, "nimToCStringConv", arg),
a.storage)
proc convCStrToStr(p: BProc, n: PNode, d: var TLoc) =
@@ -2708,52 +2818,39 @@ proc genWasMoved(p: BProc; n: PNode) =
# [addrLoc(p.config, a), getTypeDesc(p.module, a.t)])
proc genMove(p: BProc; n: PNode; d: var TLoc) =
var a: TLoc = initLocExpr(p, n[1].skipAddr, {lfEnforceDeref})
if n.len == 4:
# generated by liftdestructors:
var a: TLoc = initLocExpr(p, n[1].skipAddr, {lfEnforceDeref, lfPrepareForMutation})
var src: TLoc = initLocExpr(p, n[2])
let destVal = rdLoc(a)
let srcVal = rdLoc(src)
p.s(cpsStmts).addSingleIfStmt(
cOp(NotEqual,
dotField(destVal, "p"),
dotField(srcVal, "p"))):
if p.config.usesSso() and
n[1].typ.skipTypes(abstractVar).kind == tyString:
# SmallString: destroy dst then struct-copy src; no .p field aliasing needed
genStmts(p, n[3])
p.s(cpsStmts).addFieldAssignment(destVal, "len", dotField(srcVal, "len"))
p.s(cpsStmts).addFieldAssignment(destVal, "p", dotField(srcVal, "p"))
genAssignment(p, a, src, {})
else:
p.s(cpsStmts).addSingleIfStmt(
cOp(NotEqual,
dotField(destVal, "p"),
dotField(srcVal, "p"))):
genStmts(p, n[3])
p.s(cpsStmts).addFieldAssignment(destVal, "len", dotField(srcVal, "len"))
p.s(cpsStmts).addFieldAssignment(destVal, "p", dotField(srcVal, "p"))
else:
if d.k == locNone: d = getTemp(p, n.typ)
if p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc}:
genAssignment(p, d, a, {})
if p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc}:
var op = getAttachedOp(p.module.g.graph, n.typ, attachedWasMoved)
if op == nil:
if op == nil or sfOverridden notin op.flags:
var a: TLoc = initLocExpr(p, n[1].skipAddr, {lfEnforceDeref, lfPrepareForMutation})
genAssignment(p, d, a, {})
resetLoc(p, a)
else:
var b = initLocExpr(p, newSymNode(op))
case skipTypes(a.t, abstractVar+{tyStatic}).kind
of tyOpenArray, tyVarargs: # todo fixme generated `wasMoved` hooks for
# openarrays, but it probably shouldn't?
let ra = rdLoc(a)
var s: string
if reifiedOpenArray(a.lode):
if a.t.kind in {tyVar, tyLent}:
s = derefField(ra, "Field0") & cArgumentSeparator & derefField(ra, "Field1")
else:
s = dotField(ra, "Field0") & cArgumentSeparator & dotField(ra, "Field1")
else:
s = ra & cArgumentSeparator & ra & "Len_0"
p.s(cpsStmts).addCallStmt(rdLoc(b), s)
else:
let val = if p.module.compileToCpp: rdLoc(a) else: byRefLoc(p, a)
p.s(cpsStmts).addCallStmt(rdLoc(b), val)
n[1] = makeAddr(n[1], p.module.idgen)
genCall(p, n, d)
else:
if n[1].kind == nkSym and isSinkParam(n[1].sym):
var tmp = getTemp(p, n[1].typ.skipTypes({tySink}))
genAssignment(p, tmp, a, {needToCopySinkParam})
genAssignment(p, d, tmp, {})
resetLoc(p, tmp)
else:
genAssignment(p, d, a, {})
var a: TLoc = initLocExpr(p, n[1].skipAddr, {lfEnforceDeref, lfPrepareForMutation})
genAssignment(p, d, a, {})
resetLoc(p, a)
proc genDestroy(p: BProc; n: PNode) =
@@ -2763,15 +2860,19 @@ proc genDestroy(p: BProc; n: PNode) =
case t.kind
of tyString:
var a: TLoc = initLocExpr(p, arg)
let ra = rdLoc(a)
let rp = dotField(ra, "p")
p.s(cpsStmts).addSingleIfStmt(
cOp(And, rp,
cOp(Not, cOp(BitAnd, NimInt,
derefField(rp, "cap"),
NimStrlitFlag)))):
let fn = if optThreads in p.config.globalOptions: "deallocShared" else: "dealloc"
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, fn), rp)
if p.config.usesSso():
# SmallString: delegate to nimDestroyStrV1 (rc-based, handles static strings)
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimDestroyStrV1"), rdLoc(a))
else:
let ra = rdLoc(a)
let rp = dotField(ra, "p")
p.s(cpsStmts).addSingleIfStmt(
cOp(And, rp,
cOp(Not, cOp(BitAnd, NimInt,
derefField(rp, "cap"),
NimStrlitFlag)))):
let fn = if optThreads in p.config.globalOptions: "deallocShared" else: "dealloc"
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, fn), rp)
of tySequence:
var a: TLoc = initLocExpr(p, arg)
let ra = rdLoc(a)
@@ -2813,7 +2914,7 @@ proc genSlice(p: BProc; e: PNode; d: var TLoc) =
let (x, y) = genOpenArraySlice(p, e, e.typ, e.typ.elementType,
prepareForMutation = e[1].kind == nkHiddenDeref and
e[1].typ.skipTypes(abstractInst).kind == tyString and
p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc})
p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc})
if d.k == locNone: d = getTemp(p, e.typ)
let dest = rdLoc(d)
p.s(cpsStmts).addFieldAssignment(dest, "Field0", x)
@@ -2876,8 +2977,15 @@ proc genMagicExpr(p: BProc, e: PNode, d: var TLoc, op: TMagic) =
of mAppendStrStr: genStrAppend(p, e, d)
of mAppendSeqElem:
if optSeqDestructors in p.config.globalOptions:
e[1] = makeAddr(e[1], p.module.idgen)
genCall(p, e, d)
if p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc}:
# Inline growth + direct slot assignment: avoids the add() call overhead
# and lets the C compiler see the construction expression at its final
# destination, enabling in-place construction for nkObjConstr etc.
# gcYrc is excluded because its add() acquires a striped reader lock.
genSeqElemAppendV2(p, e, d)
else:
e[1] = makeAddr(e[1], p.module.idgen)
genCall(p, e, d)
else:
genSeqElemAppend(p, e, d)
of mEqStr: genStrEquals(p, e, d)
@@ -2963,6 +3071,7 @@ proc genMagicExpr(p: BProc, e: PNode, d: var TLoc, op: TMagic) =
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimGCunref"), ra)
of mSetLengthStr: genSetLengthStr(p, e, d)
of mSetLengthSeq: genSetLengthSeq(p, e, d)
of mSetLengthSeqUninit: genSetLengthSeq(p, e, d, noinit = true)
of mIncl, mExcl, mCard, mLtSet, mLeSet, mEqSet, mMulSet, mPlusSet, mMinusSet,
mInSet, mXorSet:
genSetOp(p, e, d, op)
@@ -3016,7 +3125,7 @@ proc genMagicExpr(p: BProc, e: PNode, d: var TLoc, op: TMagic) =
let n = semparallel.liftParallel(p.module.g.graph, p.module.idgen, p.module.module, e)
expr(p, n, d)
of mDeepCopy:
if p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc} and optEnableDeepCopy notin p.config.globalOptions:
if p.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc} and optEnableDeepCopy notin p.config.globalOptions:
localError(p.config, e.info,
"for --mm:arc|atomicArc|orc 'deepcopy' support has to be enabled with --deepcopy:on")
@@ -3132,6 +3241,8 @@ proc genTupleConstr(p: BProc, n: PNode, d: var TLoc) =
for i in 0..<n.len:
var it = n[i]
if it.kind == nkExprColonExpr: it = it[1]
# Do not produce code for void types
if it.typ != nil and isEmptyType(it.typ): continue
rec = initLoc(locExpr, it, dest[].storage)
rec.snippet = dotField(rdLoc(dest[]), "Field" & rope(i))
rec.flags.incl(lfEnforceDeref)
@@ -3248,7 +3359,11 @@ proc upConv(p: BProc, n: PNode, d: var TLoc) =
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "raiseObjectConversionError"))
raiseInstr(p, p.s(cpsStmts))
if n[0].typ.kind != tyObject:
# skip cast when types map to the same C type
# this avoids invalid C code like `*(T*)&x` for types that can't have their address taken (e.g., WASM __externref_t)
if getTypeDesc(p.module, n.typ) == getTypeDesc(p.module, n[0].typ):
expr(p, n[0], d)
elif n[0].typ.kind != tyObject:
let destTyp = getTypeDesc(p.module, n.typ)
let val = rdLoc(a)
if n.isLValue:
@@ -3294,7 +3409,7 @@ proc downConv(p: BProc, n: PNode, d: var TLoc) =
cCast(ptrType(destType),
wrapPar(cAddr(wrapPar(val))))),
a.storage)
elif p.module.compileToCpp:
elif p.module.compileToCpp or isImportedType(src):
# C++ implicitly downcasts for us
expr(p, arg, d)
else:
@@ -3340,8 +3455,9 @@ proc genConstSetup(p: BProc; sym: PSym): bool =
useHeader(m, sym)
if sym.loc.k == locNone:
fillBackendName(p.module, sym)
fillLoc(sym.loc, locData, sym.astdef, OnStatic)
if m.hcrOn: incl(sym.loc.flags, lfIndirect)
backendEnsureMutable sym
fillLoc(sym.locImpl, locData, sym.astdef, OnStatic)
if m.hcrOn: incl(sym, lfIndirect)
result = lfNoDecl notin sym.loc.flags
proc genConstHeader(m, q: BModule; p: BProc, sym: PSym) =
@@ -3375,7 +3491,23 @@ proc genConstDefinition(q: BModule; p: BProc; sym: PSym) =
data.addDeclWithVisibility(Private):
data.addVarWithInitializer(Local, actualConstName, typ = td):
genBracedInit(q.initProc, sym.astdef, isConst = true, sym.typ, data)
q.s[cfsData].add(extract(data))
if q.config.cmd == cmdNifC:
# Each `cg` process that demands this const emits its definition
# (emit-everywhere). Always declare it first (the data analogue of a proc
# prototype) so a TU whose copy the merge stage drops still has a valid
# declaration; wrap the definition as a droppable `'d'` unit the merge
# stage assigns to a single owner.
let cname = stripCnifMarks(actualConstName)
var decl = newBuilder("")
decl.addDeclWithVisibility(Extern):
decl.addVar(kind = Local, name = actualConstName, typ = td)
q.s[cfsData].add(extract(decl))
q.s[cfsData].add(cnifDefDirective(cname, "d", icNifName(q, sym)))
q.s[cfsData].add(extract(data))
q.s[cfsData].add(cnifEndDefs())
q.icDataDefs.add (cname, icNifName(q, sym))
else:
q.s[cfsData].add(extract(data))
if q.hcrOn:
# generate the global pointer with the real name
q.s[cfsVars].addVar(kind = Global, name = sym.loc.snippet,
@@ -3411,7 +3543,7 @@ proc genConstDefinition(q: BModule; p: BProc; sym: PSym) =
proc genConstStmt(p: BProc, n: PNode) =
# This code is only used in the new DCE implementation.
assert useAliveDataFromDce in p.module.flags
assert delayedCodegen(p.module)
let m = p.module
for it in n:
if it[0].kind == nkSym:
@@ -3429,7 +3561,7 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
var sym = n.sym
case sym.kind
of skMethod:
if useAliveDataFromDce in p.module.flags or {sfDispatcher, sfForward} * sym.flags != {}:
if delayedCodegen(p.module) or {sfDispatcher, sfForward} * sym.flags != {}:
# we cannot produce code for the dispatcher yet:
fillProcLoc(p.module, n)
genProcPrototype(p.module, sym)
@@ -3439,14 +3571,29 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
of skProc, skConverter, skIterator, skFunc:
#if sym.kind == skIterator:
# echo renderTree(sym.getBody, {renderIds})
if p.config.cmd == cmdNifC and
(isGenericRoutineStrict(sym) or sfCompileTime in sym.flags or
(sym.kind == skIterator and sym.typ.callConv == ccInline)):
# Under IC a module's top-level routine definitions are serialized as bare
# symbol references that reappear in the loaded statement list. Uninstantiated
# generic routines (incl. those with type-class params like `tuple`) and
# `.compileTime` routines have no run-time code, so skip them here.
# Inline iterators likewise have no standalone code — they are always inlined
# at their for-loop call sites by the transformer (only closure iterators get
# a standalone C function), so a bare serialized def reference is a no-op.
return
if sfCompileTime in sym.flags:
localError(p.config, n.info, "request to generate code for .compileTime proc: " &
sym.name.s)
if useAliveDataFromDce in p.module.flags and sym.typ.callConv != ccInline:
if delayedCodegen(p.module) and sym.typ.callConv != ccInline:
fillProcLoc(p.module, n)
genProcPrototype(p.module, sym)
else:
genProc(p.module, sym)
# For cross-module inline procs with optCompress, ensure prototype is emitted
if sym.typ.callConv == ccInline and optCompress in p.config.globalOptions and
sym.itemId.module != p.module.module.position:
genProcPrototype(p.module, sym)
if sym.loc.snippet == "" or sym.loc.lode == nil:
internalError(p.config, n.info, "expr: proc not init " & sym.name.s)
putLocIntoDest(p, d, sym.loc)
@@ -3455,7 +3602,13 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
var lit = newBuilder("")
genLiteral(p, sym.astdef, sym.typ, lit)
putIntoDest(p, d, n, extract(lit), OnStatic)
elif useAliveDataFromDce in p.module.flags:
elif optCompress in p.config.globalOptions:
# With delayed codegen, we need to ensure the definition is generated
# not just the extern header declaration
requestConstImpl(p, sym)
assert((sym.loc.snippet != "") and (sym.loc.t != nil))
putLocIntoDest(p, d, sym.loc)
elif delayedCodegen(p.module):
genConstHeader(p.module, p.module, p, sym)
assert((sym.loc.snippet != "") and (sym.loc.t != nil))
putLocIntoDest(p, d, sym.loc)
@@ -3503,6 +3656,11 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
# echo renderTree(p.prc.ast, {renderIds})
internalError(p.config, n.info, "expr: param not init " & sym.name.s & "_" & $sym.id)
putLocIntoDest(p, d, sym.loc)
of skTemplate, skMacro:
# Under IC a module's top-level template/macro definitions are serialized as
# bare symbol references (only their interface matters), so they reappear in
# the loaded statement list. They are compile-time only and produce no code.
discard
else: internalError(p.config, n.info, "expr(" & $sym.kind & "); unknown symbol")
of nkNilLit:
if not isEmptyType(n.typ):
@@ -3587,8 +3745,14 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
of nkWhileStmt: genWhileStmt(p, n)
of nkVarSection, nkLetSection: genVarStmt(p, n)
of nkConstSection:
if useAliveDataFromDce in p.module.flags:
if delayedCodegen(p.module):
genConstStmt(p, n)
else: # enforce addressable consts for exportc
let m = p.module
for it in n:
let symNode = skipPragmaExpr(it[0])
if symNode.kind == nkSym and sfExportc in symNode.sym.flags:
requestConstImpl(p, symNode.sym)
# else: consts generated lazily on use
of nkForStmt: internalError(p.config, n.info, "for statement not eliminated")
of nkCaseStmt: genCase(p, n, d)
@@ -3647,7 +3811,10 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
of nkProcDef, nkFuncDef, nkMethodDef, nkConverterDef:
if n[genericParamsPos].kind == nkEmpty:
var prc = n[namePos].sym
if useAliveDataFromDce in p.module.flags:
if optCompress in p.config.globalOptions:
if prc.magic in generatedMagics:
genProc(p.module, prc)
elif delayedCodegen(p.module):
if p.module.alive.contains(prc.itemId.item) and
prc.magic in generatedMagics:
genProc(p.module, prc)
@@ -3667,9 +3834,68 @@ proc expr(p: BProc, n: PNode, d: var TLoc) =
inc p.splitDecls
genGotoState(p, n)
of nkBreakState: genBreakState(p, n, d)
of nkMixinStmt, nkBindStmt: discard
of nkMixinStmt, nkBindStmt, nkReplayAction: discard
else: internalError(p.config, n.info, "expr(" & $n.kind & "); unknown node kind")
proc isOpaqueImportcType(t: PType): bool =
# importc type without completeStruct that can't use aggregate init (e.g. C11 _Atomic)
if t.sym != nil and sfImportc in t.sym.flags:
if tfCompleteStruct notin t.flags:
if tfIncompleteStruct in t.flags:
return true
if t.kind == tyObject and (t.n == nil or t.n.len == 0):
return true
return false
proc containsOpaqueImportcField(typ: PType): bool
proc containsOpaqueImportcFieldAux(t: PType; n: PNode): bool =
if n == nil: return false
case n.kind
of nkRecList:
for child in n.sons:
if containsOpaqueImportcFieldAux(t, child):
return true
of nkRecCase:
if containsOpaqueImportcFieldAux(t, n[0]):
return true
for i in 1..<n.len:
let branch = n[i]
if branch.kind == nkOfBranch or branch.kind == nkElse:
if containsOpaqueImportcFieldAux(t, branch.lastSon):
return true
of nkSym:
if containsOpaqueImportcField(n.sym.typ):
return true
else:
discard
return false
proc containsOpaqueImportcField(typ: PType): bool =
# Check if type contains opaque importc fields that need designated initializers
if typ == nil: return false
let t = skipTypes(typ, abstractRange+{tyOwned}-{tyTypeDesc})
if isOpaqueImportcType(t):
return true
case t.kind
of tyObject:
if t.baseClass != nil:
if containsOpaqueImportcField(t.baseClass):
return true
if containsOpaqueImportcFieldAux(t, t.n):
return true
of tyTuple:
for i, a in t.ikids:
if isEmptyType(a): continue
if containsOpaqueImportcField(a):
return true
of tyArray:
if containsOpaqueImportcField(t.elementType):
return true
else:
discard
return false
proc getDefaultValue(p: BProc; typ: PType; info: TLineInfo; result: var Builder) =
var t = skipTypes(typ, abstractRange+{tyOwned}-{tyTypeDesc})
case t.kind
@@ -3700,24 +3926,36 @@ proc getDefaultValue(p: BProc; typ: PType; info: TLineInfo; result: var Builder)
result.addField(closureInit, name = "ClE_0"):
result.add(NimNil)
of tyObject:
# Use designated initializers when opaque importc fields present
var objInit: StructInitializer
result.addStructInitializer(objInit, kind = siOrderedStruct):
let initKind = if containsOpaqueImportcField(t): siNamedStruct else: siOrderedStruct
result.addStructInitializer(objInit, kind = initKind):
getNullValueAuxT(p, t, t, t.n, nil, result, objInit, true, info)
of tyTuple:
# Use designated initializers when opaque importc fields present
var tupleInit: StructInitializer
result.addStructInitializer(tupleInit, kind = siOrderedStruct):
if p.vccAndC and t.isEmptyTupleType:
let initKind = if containsOpaqueImportcField(t): siNamedStruct else: siOrderedStruct
result.addStructInitializer(tupleInit, kind = initKind):
if p.vccAndC and validTupleTypeFields(t) == 0:
result.addField(tupleInit, name = "dummy"):
result.addIntValue(0)
for i, a in t.ikids:
result.addField(tupleInit, name = "Field" & $i):
getDefaultValue(p, a, info, result)
# Do not produce code for void types
if isEmptyType(a): continue
let elemTyp = skipTypes(a, abstractRange+{tyOwned}-{tyTypeDesc})
if not isOpaqueImportcType(elemTyp):
result.addField(tupleInit, name = "Field" & $i):
getDefaultValue(p, a, info, result)
of tyArray:
var arrInit: StructInitializer
result.addStructInitializer(arrInit, kind = siArray):
for i in 0..<toInt(lengthOrd(p.config, t.indexType)):
result.addField(arrInit, name = ""):
getDefaultValue(p, t.elementType, info, result)
let elemTyp = skipTypes(t.elementType, abstractRange+{tyOwned}-{tyTypeDesc})
if isOpaqueImportcType(elemTyp):
result.add "{0}"
else:
var arrInit: StructInitializer
result.addStructInitializer(arrInit, kind = siArray):
for i in 0..<toInt(lengthOrd(p.config, t.indexType)):
result.addField(arrInit, name = ""):
getDefaultValue(p, t.elementType, info, result)
#result = rope"{}"
of tyOpenArray, tyVarargs:
var openArrInit: StructInitializer
@@ -3772,8 +4010,7 @@ proc getNullValueAux(p: BProc; t: PType; obj, constOrNil: PNode,
var fieldName: string = ""
if b.kind == nkRecList and not isEmptyCaseObjectBranch(b):
fieldName = "_" & mangleRecFieldName(p.module, obj[0].sym) & "_" & $selectedBranch
result.addField(init, name = "<anonymous union>"):
# XXX figure out name for the union, see use of `addAnonUnion`
result.addField(init, name = ""): # anonymous union
var branchInit: StructInitializer
result.addStructInitializer(branchInit, kind = siNamedStruct):
result.addField(branchInit, name = fieldName):
@@ -3782,8 +4019,7 @@ proc getNullValueAux(p: BProc; t: PType; obj, constOrNil: PNode,
getNullValueAux(p, t, b, constOrNil, result, branchObjInit, isConst, info)
elif b.kind == nkSym:
fieldName = mangleRecFieldName(p.module, b.sym)
result.addField(init, name = "<anonymous union>"):
# XXX figure out name for the union, see use of `addAnonUnion`
result.addField(init, name = ""): # anonymous union
var branchInit: StructInitializer
result.addStructInitializer(branchInit, kind = siNamedStruct):
result.addField(branchInit, name = fieldName):
@@ -3798,6 +4034,9 @@ proc getNullValueAux(p: BProc; t: PType; obj, constOrNil: PNode,
of nkSym:
let field = obj.sym
let fieldTyp = skipTypes(field.typ, abstractRange+{tyOwned}-{tyTypeDesc})
if isOpaqueImportcType(fieldTyp):
return # C zero-initializes omitted fields
let sname = mangleRecFieldName(p.module, field)
result.addField(init, name = sname):
block fieldInit:
@@ -3842,11 +4081,10 @@ proc getNullValueAuxT(p: BProc; orig, t: PType; obj, constOrNil: PNode,
proc genConstObjConstr(p: BProc; n: PNode; isConst: bool; result: var Builder) =
let t = n.typ.skipTypes(abstractInstOwned)
#if not isObjLackingTypeField(t) and not p.module.compileToCpp:
# result.addf("{$1}", [genTypeInfo(p.module, t)])
# inc count
# Use designated initializers when opaque importc fields present
var objInit: StructInitializer
result.addStructInitializer(objInit, kind = siOrderedStruct):
let initKind = if t.kind == tyObject and containsOpaqueImportcField(t): siNamedStruct else: siOrderedStruct
result.addStructInitializer(objInit, kind = initKind):
if t.kind == tyObject:
getNullValueAuxT(p, t, t, t.n, n, result, objInit, isConst, n.info)
@@ -3878,6 +4116,8 @@ proc genConstTuple(p: BProc, n: PNode; isConst: bool; tup: PType; result: var Bu
var it = n[i]
if it.kind == nkExprColonExpr:
it = it[1]
# Do not produce code for void types
if isEmptyType(tup[i]): continue
result.addField(tupleInit, name = "Field" & $i):
genBracedInit(p, it, isConst, tup[i], result)
@@ -4024,7 +4264,10 @@ proc genBracedInit(p: BProc, n: PNode; isConst: bool; optionalType: PType; resul
genConstObjConstr(p, n, isConst, result)
of tyString, tyCstring:
if optSeqDestructors in p.config.globalOptions and n.kind != nkNilLit and ty == tyString:
genStringLiteralV2Const(p.module, n, isConst, result)
if p.config.usesSso():
genStringLiteralV3Const(p.module, n, isConst, result)
else:
genStringLiteralV2Const(p.module, n, isConst, result)
else:
var d: TLoc = initLocExpr(p, n)
result.add rdLoc(d)

View File

@@ -16,16 +16,17 @@
## implementation.
template detectVersion(field, corename) =
if m.g.field == 0:
let core = getCompilerProc(m.g.graph, corename)
if core == nil or core.kind != skConst:
m.g.field = 1
else:
m.g.field = toInt(ast.getInt(core.astdef))
result = m.g.field
if m.g.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc, gcHooks}:
result = 2
else:
result = 1
proc detectStrVersion(m: BModule): int =
detectVersion(strVersion, "nimStrVersion")
if m.g.config.usesSso() and
m.g.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc, gcHooks}:
result = 3
else:
detectVersion(strVersion, "nimStrVersion")
proc detectSeqVersion(m: BModule): int =
detectVersion(seqVersion, "nimSeqVersion")
@@ -131,6 +132,192 @@ proc genStringLiteralV2Const(m: BModule; n: PNode; isConst: bool; result: var Bu
result.addField(strInit, name = "p"):
result.add(cCast(ptrType("NimStrPayload"), cAddr(pureLit)))
proc ssoCharLit(ch: char): string =
## Return a C char literal for ch, with proper escaping.
const hexDigits = "0123456789abcdef"
result = "'"
case ch
of '\'': result.add("\\'")
of '\\': result.add("\\\\")
of '\0': result.add("\\0")
of '\n': result.add("\\n")
of '\r': result.add("\\r")
of '\t': result.add("\\t")
elif ch.ord < 32 or ch.ord == 127:
result.add("\\x")
result.add(hexDigits[ch.ord shr 4])
result.add(hexDigits[ch.ord and 0xf])
else:
result.add(ch)
result.add('\'')
proc ssoBytesLit(m: BModule; s: string; slen: int): string =
## Compute the `bytes` field value for the new SmallString layout.
## byte 0 = slen, bytes 1-7 = inline chars 0-6 (zero-padded).
## On LE: slen in bits 0-7, char[i] in bits (i+1)*8..(i+1)*8+7.
## On BE: slen in bits 56-63, char[i] in bits (6-i)*8..(6-i)*8+7.
const AlwaysAvail = 7
var val: uint64
if CPU[m.g.config.target.targetCPU].endian == littleEndian:
val = uint64(slen)
for i in 0..<min(s.len, AlwaysAvail):
val = val or (uint64(s[i]) shl (uint(i + 1) * 8))
else:
val = uint64(slen) shl 56
for i in 0..<min(s.len, AlwaysAvail):
val = val or (uint64(s[i]) shl (uint(AlwaysAvail - 1 - i) * 8))
# Cast to NU (C name for Nim's uint, = NU64 on 64-bit). NU64 = uint64_t.
result = cCast("NU", $val & "ULL")
proc ssoMoreLit(m: BModule; s: string): string =
## For medium string literals (AlwaysAvail < len <= PayloadSize), encode
## chars[AlwaysAvail..ptrSize-1] in the 'more' pointer field bit-pattern.
## The last pointer byte is always '\0' (null terminator), guaranteed by
## PayloadSize = AlwaysAvail + ptrSize - 1. slen <= PayloadSize guards
## prevent any code from dereferencing this as an actual pointer.
const AlwaysAvail = 7
let ptrSize = m.g.config.target.ptrSize
var val: uint64 = 0
for i in 0..<ptrSize:
let ch: uint64 = if AlwaysAvail + i < s.len: uint64(s[AlwaysAvail + i]) else: 0
if CPU[m.g.config.target.targetCPU].endian == littleEndian:
val = val or (ch shl (uint(i) * 8))
else:
val = val or (ch shl (uint(ptrSize - 1 - i) * 8))
result = cCast(ptrType("LongString"), "(uintptr_t)" & $val)
proc genStringLiteralV3Const(m: BModule; n: PNode; isConst: bool; result: var Builder) =
# Inline SmallString struct initializer for use inside const aggregate types.
# Layout: {bytes: NimUint, more: ptr LongString}
# bytes = slen (low byte) | char[0]<<8 | char[1]<<16 | ... | char[6]<<56
const AlwaysAvail = 7
let s = n.strVal
cgsym(m, "SmallString")
cgsym(m, "LongString")
let payloadSize = AlwaysAvail + m.g.config.target.ptrSize - 1
var si: StructInitializer
result.addStructInitializer(si, kind = siOrderedStruct):
if s.len <= AlwaysAvail:
result.addField(si, name = "bytes"):
result.add(ssoBytesLit(m, s, s.len))
result.addField(si, name = "more"):
result.add(NimNil)
elif s.len <= payloadSize:
# Medium string: bytes holds slen + chars 0-6; more holds chars 7..PayloadSize-1.
result.addField(si, name = "bytes"):
result.add(ssoBytesLit(m, s, s.len))
result.addField(si, name = "more"):
result.add(ssoMoreLit(m, s))
else:
# Emit the LongString block into cfsStrData and reference it inline.
let dataName = getTempName(m)
var res = newBuilder("")
res.addVarWithTypeAndInitializer(
if isConst: AlwaysConst else: Global,
name = dataName):
res.addSimpleStruct(m, name = "", baseType = ""):
res.addField(name = "rc", typ = NimInt)
res.addField(name = "fullLen", typ = NimInt)
res.addField(name = "capImpl", typ = NimInt)
res.addArrayField(name = "data", elementType = NimChar, len = s.len + 1)
do:
var di: StructInitializer
res.addStructInitializer(di, kind = siOrderedStruct):
res.addField(di, name = "fullLen"):
res.addIntValue(s.len)
res.addField(di, name = "rc"):
res.addIntValue(1)
res.addField(di, name = "capImpl"):
res.addIntValue(0) # static, never freed
res.addField(di, name = "data"):
res.add(makeCString(s))
m.s[cfsStrData].add(extract(res))
# slen = StaticSlen (254): marks this as a static (never-freed) long string.
result.addField(si, name = "bytes"):
result.add(ssoBytesLit(m, s, 254))
result.addField(si, name = "more"):
result.add(cCast(ptrType("LongString"), cAddr(dataName)))
# ------ Version 3: SmallString (SSO) strings --------------------------------
proc genStringLiteralV3(m: BModule; n: PNode; isConst: bool; result: var Builder) =
# SmallString literal. Always generate a fresh SmallString variable (like v2
# always generates a fresh outer NimStringV2). For long strings, cache the
# LongString payload to avoid duplicates within a module.
const AlwaysAvail = 7 # must match strs_v3.nim
let s = n.strVal
let tmp = getTempName(m)
result.add tmp
cgsym(m, "SmallString")
cgsym(m, "LongString")
let payloadSize = AlwaysAvail + m.g.config.target.ptrSize - 1
var res = newBuilder("")
if s.len <= AlwaysAvail:
# Short: bytes holds slen + all chars (zero-padded), more = NULL.
res.addVarWithInitializer(
if isConst: AlwaysConst else: Global,
name = tmp, typ = "SmallString"):
var si: StructInitializer
res.addStructInitializer(si, kind = siOrderedStruct):
res.addField(si, name = "bytes"):
res.add(ssoBytesLit(m, s, s.len))
res.addField(si, name = "more"):
res.add(NimNil)
elif s.len <= payloadSize:
# Medium: bytes holds slen + chars 0-6; more holds chars 7..PayloadSize-1 as raw bits.
res.addVarWithInitializer(
if isConst: AlwaysConst else: Global,
name = tmp, typ = "SmallString"):
var si: StructInitializer
res.addStructInitializer(si, kind = siOrderedStruct):
res.addField(si, name = "bytes"):
res.add(ssoBytesLit(m, s, s.len))
res.addField(si, name = "more"):
res.add(ssoMoreLit(m, s))
else:
# Long: cache the LongString block to emit it only once per module per string.
# Always generate a fresh SmallString pointing at the (possibly cached) block.
let id = nodeTableTestOrSet(m.dataCache, n, m.labels)
var dataName: string
if id == m.labels:
dataName = getTempName(m)
res.addVarWithTypeAndInitializer(
if isConst: AlwaysConst else: Global,
name = dataName):
res.addSimpleStruct(m, name = "", baseType = ""):
res.addField(name = "rc", typ = NimInt)
res.addField(name = "fullLen", typ = NimInt)
res.addField(name = "capImpl", typ = NimInt)
res.addArrayField(name = "data", elementType = NimChar, len = s.len + 1)
do:
var di: StructInitializer
res.addStructInitializer(di, kind = siOrderedStruct):
res.addField(di, name = "fullLen"):
res.addIntValue(s.len)
res.addField(di, name = "rc"):
res.addIntValue(1)
res.addField(di, name = "capImpl"):
res.addIntValue(0) # bit 0 = 0: static, never freed
res.addField(di, name = "data"):
res.add(makeCString(s))
else:
dataName = m.tmpBase & $id
# slen = StaticSlen (254): marks this as a static (never-freed) long string.
res.addVarWithInitializer(
if isConst: AlwaysConst else: Global,
name = tmp, typ = "SmallString"):
var si: StructInitializer
res.addStructInitializer(si, kind = siOrderedStruct):
res.addField(si, name = "bytes"):
res.add(ssoBytesLit(m, s, 254))
res.addField(si, name = "more"):
res.add(cCast(ptrType("LongString"), cAddr(dataName)))
m.s[cfsStrData].add(extract(res))
# ------ Version selector ---------------------------------------------------
proc genStringLiteralDataOnly(m: BModule; s: string; info: TLineInfo;
@@ -141,6 +328,8 @@ proc genStringLiteralDataOnly(m: BModule; s: string; info: TLineInfo;
let tmp = getTempName(m)
genStringLiteralDataOnlyV2(m, s, tmp, isConst)
result.add tmp
of 3:
localError(m.config, info, "genStringLiteralDataOnly not supported for SmallString (nimsso)")
else:
localError(m.config, info, "cannot determine how to produce code for string literal")
@@ -151,5 +340,6 @@ proc genStringLiteral(m: BModule; n: PNode; result: var Builder) =
case detectStrVersion(m)
of 0, 1: genStringLiteralV1(m, n, result)
of 2: genStringLiteralV2(m, n, isConst = true, result)
of 3: genStringLiteralV3(m, n, isConst = true, result)
else:
localError(m.config, n.info, "cannot determine how to produce code for string literal")

View File

@@ -67,7 +67,14 @@ proc specializeResetT(p: BProc, accessor: Rope, typ: PType) =
var x = typ.baseClass
if x != nil: x = x.skipTypes(skipPtrs)
specializeResetT(p, accessor.parentObj(p.module), x)
if typ.n != nil: specializeResetN(p, accessor, typ.n, typ)
if typ.n != nil:
if typ.sym != nil and sfImportc in typ.sym.flags:
# imported C struct, nimZeroMem
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimZeroMem"),
cCast(ptrType(CPointer), cAddr(accessor)),
cSizeof(getTypeDesc(p.module, typ)))
else:
specializeResetN(p, accessor, typ.n, typ)
of tyTuple:
let typ = getUniqueType(typ)
for i, a in typ.ikids:

View File

@@ -18,7 +18,7 @@ proc registerTraverseProc(p: BProc, v: PSym) =
var traverseProc = ""
if p.config.selectedGC in {gcMarkAndSweep, gcHooks, gcRefc} and
optOwnedRefs notin p.config.globalOptions and
containsGarbageCollectedRef(v.loc.t):
containsManagedMemory(v.loc.t):
# we register a specialized marked proc here; this has the advantage
# that it works out of the box for thread local storage then :-)
traverseProc = genTraverseProcForGlobal(p.module, v, v.info)
@@ -126,25 +126,26 @@ proc genVarTuple(p: BProc, n: PNode) =
let vn = n[i]
let v = vn.sym
if sfCompileTime in v.flags: continue
backendEnsureMutable v
if sfGlobal in v.flags:
assignGlobalVar(p, vn, "")
genObjectInit(p, cpsInit, v.typ, v.loc, constructObj)
genObjectInit(p, cpsInit, v.typ, v.locImpl, constructObj)
registerTraverseProc(p, v)
else:
assignLocalVar(p, vn)
initLocalVar(p, v, immediateAsgn=isAssignedImmediately(p.config, n[^1]))
var field = initLoc(locExpr, vn, tup.storage)
let rtup = rdLoc(tup)
let fieldName =
let fieldName =
if t.kind == tyTuple:
"Field" & $i
else:
if t.n[i].kind != nkSym: internalError(p.config, n.info, "genVarTuple")
mangleRecFieldName(p.module, t.n[i].sym)
field.snippet = dotField(rtup, fieldName)
putLocIntoDest(p, v.loc, field)
putLocIntoDest(p, v.locImpl, field)
if forHcr or isGlobalInBlock:
hcrGlobals.add((loc: v.loc, tp: CNil))
hcrGlobals.add((loc: v.locImpl, tp: CNil))
if forHcr:
# end the block where the tuple gets initialized
@@ -225,11 +226,11 @@ proc genState(p: BProc, n: PNode) =
elif n0.kind == nkStrLit:
p.s(cpsStmts).addLabel(n0.strVal)
proc blockLeaveActions(p: BProc, howManyTrys, howManyExcepts: int) =
proc blockLeaveActions(p: BProc, howManyTrys, howManyExcepts: int, isReturnStmt = false) =
# Called by return and break stmts.
# Deals with issues faced when jumping out of try/except/finally stmts.
var stack = newSeq[tuple[fin: PNode, inExcept: bool, label: Natural]](0)
var stack = newSeq[tuple[fin: PNode, inExcept: bool, isHidden: bool, label: Natural]](0)
inc p.withinBlockLeaveActions
for i in 1..howManyTrys:
@@ -257,7 +258,7 @@ proc blockLeaveActions(p: BProc, howManyTrys, howManyExcepts: int) =
# Pop exceptions that was handled by the
# except-blocks we are in
if noSafePoints notin p.flags:
if noSafePoints notin p.flags and not (isReturnStmt and isClosureIterator(p.prc.typ)):
for i in countdown(howManyExcepts-1, 0):
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "popCurrentException"))
@@ -340,9 +341,9 @@ proc genCppParamsForCtor(p: BProc; call: PNode; didGenTemp: var bool): Snippet =
call[i][0]
else:
call[i]
if param.kind != nkBracketExpr or param.typ.kind in
if not param.typ.isCompileTimeOnly and (param.kind != nkBracketExpr or param.typ.kind in
{tyRef, tyPtr, tyUncheckedArray, tyArray, tyOpenArray,
tyVarargs, tySequence, tyString, tyCstring, tyTuple}:
tyVarargs, tySequence, tyString, tyCstring, tyTuple}):
let tempLoc = initLocExprSingleUse(p, param)
didGenTemp = didGenTemp or tempLoc.k == locTemp
genOtherArg(p, call, i, typ, res, argBuilder)
@@ -365,6 +366,7 @@ proc genSingleVar(p: BProc, v: PSym; vn, value: PNode) =
if v.flags * {sfImportc, sfExportc} == {sfImportc} and
value.kind == nkEmpty and
v.loc.flags * {lfHeader, lfNoDecl} != {}:
# IC XXX: this is bad, we should set v.loc regardless here
return
if sfPure in v.flags:
# v.owner.kind != skModule:
@@ -460,7 +462,8 @@ proc genSingleVar(p: BProc, v: PSym; vn, value: PNode) =
if value.kind != nkEmpty and valueAsRope.len == 0:
genLineDir(targetProc, vn)
if not isCppCtorCall:
loadInto(targetProc, vn, value, v.loc)
backendEnsureMutable v
loadInto(targetProc, vn, value, v.locImpl)
if forHcr:
endBlockWith(targetProc):
finishBranch(p.s(cpsStmts), hcrInit)
@@ -487,14 +490,17 @@ proc genClosureVar(p: BProc, a: PNode) =
constructLoc(p, v)
proc genVarStmt(p: BProc, n: PNode) =
for it in n.sons:
if it.kind == nkCommentStmt: continue
if it.kind == nkIdentDefs:
for it in n:
case it.kind
of nkCommentStmt: discard
of nkIdentDefs:
# can be a lifted var nowadays ...
if it[0].kind == nkSym:
genSingleVar(p, it)
else:
genClosureVar(p, it)
of nkSym:
genSingleVar(p, it.sym, newSymNode(it.sym), it.sym.astdef)
else:
genVarTuple(p, it)
@@ -554,7 +560,8 @@ proc genReturnStmt(p: BProc, t: PNode) =
if (t[0].kind != nkEmpty): genStmts(p, t[0])
blockLeaveActions(p,
howManyTrys = p.nestedTryStmts.len,
howManyExcepts = p.inExceptBlockLen)
howManyExcepts = p.inExceptBlockLen,
isReturnStmt = true)
if (p.finallySafePoints.len > 0) and noSafePoints notin p.flags:
# If we're in a finally block, and we came here by exception
# consume it before we return.
@@ -736,8 +743,10 @@ proc genBlock(p: BProc, n: PNode, d: var TLoc) =
# named block?
assert(n[0].kind == nkSym)
var sym = n[0].sym
sym.loc.k = locOther
sym.position = p.breakIdx+1
backendEnsureMutable sym
sym.locImpl.k = locOther
sym.positionImpl = p.breakIdx+1
# ^ IC: review this
expr(p, n[1], d)
endSimpleBlock(p, scope)
@@ -827,12 +836,26 @@ proc raiseExitCleanup(p: BProc, destroy: string) =
p.s(cpsStmts).addGoto("LA" & $p.nestedTryStmts[^1].label & "_")
proc finallyActions(p: BProc) =
if p.config.exc != excGoto and p.nestedTryStmts.len > 0 and p.nestedTryStmts[^1].inExcept:
# if the current try stmt have a finally block,
# we must execute it before reraising
let finallyBlock = p.nestedTryStmts[^1].fin
if finallyBlock != nil:
genSimpleBlock(p, finallyBlock[0])
if p.config.exc != excGoto:
# Walk past compiler-injected `nkHiddenTryStmt` wrappers (e.g. ARC's
# destructor try/finally that wraps `except T as e:` bodies) to reach
# the user's actual try. We must NOT walk past a real user try whose
# body we are currently in, because a raise from there will be caught
# by that try's own except branches rather than escaping outward.
#
# If after skipping wrappers the next entry is a user try in its
# except branch (inExcept=true), inline its finally body before the
# raise propagates — without this, the C++ sibling-catch rule would
# cause the user's catch(...)/finally pair to be bypassed and the
# finally would be silently dropped.
for i in countdown(p.nestedTryStmts.high, 0):
if p.nestedTryStmts[i].isHidden:
continue
if p.nestedTryStmts[i].inExcept:
let finallyBlock = p.nestedTryStmts[i].fin
if finallyBlock != nil:
genSimpleBlock(p, finallyBlock[0])
return
proc raiseInstr(p: BProc; result: var Builder) =
if p.config.exc == excGoto:
@@ -1156,7 +1179,7 @@ proc genTryCpp(p: BProc, t: PNode, d: var TLoc) =
throw;
}
} catch(...) {
// C++ exception occured, not under Nim's control.
// C++ exception occurred, not under Nim's control.
}
{
/* finally: */
@@ -1176,7 +1199,7 @@ proc genTryCpp(p: BProc, t: PNode, d: var TLoc) =
lineCg(p, cpsLocals, "std::exception_ptr T$1_;$n", [etmp])
let fin = if t[^1].kind == nkFinally: t[^1] else: nil
p.nestedTryStmts.add((fin, false, 0.Natural))
p.nestedTryStmts.add((fin, false, t.kind == nkHiddenTryStmt, 0.Natural))
if t.kind == nkHiddenTryStmt:
lineCg(p, cpsStmts, "try {$n", [])
@@ -1214,6 +1237,7 @@ proc genTryCpp(p: BProc, t: PNode, d: var TLoc) =
else:
scope = initScope(p.s(cpsStmts))
# we handled the error:
linefmt(p, cpsStmts, "T$1_ = nullptr;$n", [etmp])
expr(p, t[i][0], d)
linefmt(p, cpsStmts, "#popCurrentException();$n", [])
endBlockWith(p):
@@ -1250,7 +1274,8 @@ proc genTryCpp(p: BProc, t: PNode, d: var TLoc) =
initElifBranch(p.s(cpsStmts), ifStmt, orExpr)
if exvar != nil:
fillLocalName(p, exvar.sym)
fillLoc(exvar.sym.loc, locTemp, exvar, OnStack)
backendEnsureMutable exvar.sym
fillLoc(exvar.sym.locImpl, locTemp, exvar, OnStack)
linefmt(p, cpsStmts, "$1 $2 = T$3_;$n", [getTypeDesc(p.module, exvar.sym.typ),
rdLoc(exvar.sym.loc), rope(etmp+1)])
# we handled the error:
@@ -1298,7 +1323,8 @@ proc genTryCpp(p: BProc, t: PNode, d: var TLoc) =
if isImportedException(typeNode.typ, p.config):
let exvar = t[i][j][2] # ex1 in `except ExceptType as ex1:`
fillLocalName(p, exvar.sym)
fillLoc(exvar.sym.loc, locTemp, exvar, OnStack)
backendEnsureMutable exvar.sym
fillLoc(exvar.sym.locImpl, locTemp, exvar, OnStack)
startBlockWith(p):
lineCg(p, cpsStmts, "catch ($1& $2) {$n", [getTypeDesc(p.module, typeNode.typ), rdLoc(exvar.sym.loc)])
genExceptBranchBody(t[i][^1]) # exception handler body will duplicated for every type
@@ -1360,7 +1386,7 @@ proc genTryCppOld(p: BProc, t: PNode, d: var TLoc) =
genLineDir(p, t)
cgsym(p.module, "popCurrentExceptionEx")
let fin = if t[^1].kind == nkFinally: t[^1] else: nil
p.nestedTryStmts.add((fin, false, 0.Natural))
p.nestedTryStmts.add((fin, false, t.kind == nkHiddenTryStmt, 0.Natural))
startBlockWith(p):
p.s(cpsStmts).add("try {\n")
expr(p, t[0], d)
@@ -1389,7 +1415,8 @@ proc genTryCppOld(p: BProc, t: PNode, d: var TLoc) =
if t[i][j].isInfixAs():
let exvar = t[i][j][2] # ex1 in `except ExceptType as ex1:`
fillLocalName(p, exvar.sym)
fillLoc(exvar.sym.loc, locTemp, exvar, OnUnknown)
backendEnsureMutable exvar.sym
fillLoc(exvar.sym.locImpl, locTemp, exvar, OnUnknown)
startBlockWith(p):
lineCg(p, cpsStmts, "catch ($1& $2) {$n", [getTypeDesc(p.module, t[i][j][1].typ), rdLoc(exvar.sym.loc)])
else:
@@ -1438,7 +1465,7 @@ proc genTryGoto(p: BProc; t: PNode; d: var TLoc) =
let lab = p.labels
let hasExcept = t[1].kind == nkExceptBranch
if hasExcept: inc p.withinTryWithExcept
p.nestedTryStmts.add((fin, false, Natural lab))
p.nestedTryStmts.add((fin, false, t.kind == nkHiddenTryStmt, Natural lab))
p.flags.incl nimErrorFlagAccessed
@@ -1644,7 +1671,7 @@ proc genTrySetjmp(p: BProc, t: PNode, d: var TLoc) =
initElifBranch(p.s(cpsStmts), nonQuirkyIf, removeSinglePar(
cOp(Equal, dotField(safePoint, "status"), cIntValue(0))))
let fin = if t[^1].kind == nkFinally: t[^1] else: nil
p.nestedTryStmts.add((fin, quirkyExceptions, 0.Natural))
p.nestedTryStmts.add((fin, quirkyExceptions, t.kind == nkHiddenTryStmt, 0.Natural))
expr(p, t[0], d)
var quirkyIf = default(IfBuilder)
var quirkyScope = default(ScopeBuilder)
@@ -1928,6 +1955,15 @@ proc genAsgn(p: BProc, e: PNode, fastAsgn: bool) =
elif optFieldCheck in p.options and isDiscriminantField(e[0]):
genLineDir(p, e)
asgnFieldDiscriminant(p, e)
elif p.config.usesSso() and e[0].kind == nkBracketExpr and
e[0][0].typ.skipTypes(abstractVar).kind == tyString:
# nimsso: s[i] = c → nimStrPutV3(&s, i, c) (handles COW internally)
genLineDir(p, e)
var base = initLocExpr(p, e[0][0])
var idx = initLocExpr(p, e[0][1])
var rhs = initLocExpr(p, e[1])
p.s(cpsStmts).addCallStmt(cgsymValue(p.module, "nimStrPutV3"),
byRefLoc(p, base), rdLoc(idx), rdCharLoc(rhs))
else:
let le = e[0]
let ri = e[1]
@@ -1950,4 +1986,9 @@ proc genStmts(p: BProc, t: PNode) =
if isPush: pushInfoContext(p.config, t.info)
expr(p, t, a)
if isPush: popInfoContext(p.config)
internalAssert p.config, a.k in {locNone, locTemp, locLocalVar, locExpr}
# A bare `nkSym` statement is how IC serializes a definition that lives inside a
# top-level block (e.g. a nested `proc`/`var`): codegen emits the definition and
# leaves the symbol's own location in `a` (e.g. `locProc`), which is discarded
# here, so the value-sanity check below does not apply to it.
internalAssert p.config, t.kind == nkSym or
a.k in {locNone, locTemp, locLocalVar, locExpr}

View File

@@ -11,7 +11,7 @@
# ------------------------- Name Mangling --------------------------------
import sighashes, modulegraphs, std/strscans
import sighashes, std/strscans
import ../dist/checksums/src/checksums/md5
import std/sequtils
@@ -57,30 +57,75 @@ proc mangleField(m: BModule; name: PIdent): string =
proc mangleProc(m: BModule; s: PSym; makeUnique: bool): string =
result = "_Z" # Common prefix in Itanium ABI
result.add encodeSym(m, s, makeUnique)
var params = ""
var staticLists = ""
if s.typ.len > 1: #we dont care about the return param
for i in 1..<s.typ.len:
if s.typ[i].isNil: continue
result.add encodeType(m, s.typ[i])
params.add encodeType(m, s.typ[i], staticLists)
result.add encodeSym(m, s, makeUnique, staticLists)
result.add params
if result in m.g.mangledPrcs:
result = mangleProc(m, s, true)
else:
m.g.mangledPrcs.incl(result)
proc sharedInstanceCName(m: BModule; s: PSym): string =
## The module-free canonical C name for a content-keyed generic instance,
## or "" when the symbol must keep its module-suffixed name. With a shared
## name, every TU that instantiated the same generic with the same type
## arguments calls one extern definition (first claimant's TU embeds it,
## see `genProcLvl3`) instead of compiling its own static copy.
##
## The name is program-unique only if the 30-bit content hash does not
## collide for same-named instances of *different* instantiations across
## modules — the per-module probe in `setInstanceDisamb` cannot see that.
## Claimants therefore must present the same signature; on mismatch the
## later one keeps its module-suffixed name (no merge, still correct).
## Residual risk: same name and signature, different generic args, AND a
## 30-bit collision — vanishingly unlikely; a full-typeKey verification
## channel can close it later.
result = ""
if m.config.cmd == cmdNifC and s.kind in routineKinds and
(s.disamb and InstanceDisambBit) != 0'i32 and
s.typ != nil and s.typ.callConv != ccInline and not m.hcrOn and
{sfImportc, sfExportc, sfCodegenDecl} * s.flags == {}:
# The content-derived `disamb` is unique per process (collision-probed in
# `setInstanceDisamb`), so the mint-site-independent `_i<disamb>` name is
# safe to use directly; identical instances across modules collide on it
# exactly and the merge stage keeps one.
result = s.name.s.mangle & "_i" & $s.disamb
proc isSharedInstanceCName(m: BModule; s: PSym): bool =
m.config.cmd == cmdNifC and s.kind in routineKinds and
(s.disamb and InstanceDisambBit) != 0'i32 and
stripCnifMarks(s.loc.snippet) == s.name.s.mangle & "_i" & $s.disamb
proc fillBackendName(m: BModule; s: PSym) =
if s.loc.snippet == "":
var result: Rope
if not m.compileToCpp and s.kind in routineKinds and optCDebug in m.g.config.globalOptions and
if s.kind in routineKinds and {optCDebug, optItaniumMangle} * m.g.config.globalOptions == {optCDebug, optItaniumMangle} and
m.g.config.symbolFiles == disabledSf:
result = mangleProc(m, s, false).rope
else:
result = s.name.s.mangle.rope
result.add mangleProcNameExt(m.g.graph, s)
let shared = sharedInstanceCName(m, s)
if shared.len > 0:
result = shared.rope
else:
result = s.name.s.mangle.rope
result.add mangleProcNameExt(m.g.graph, s)
if m.hcrOn:
result.add '_'
result.add(idOrSig(s, m.module.name.s.mangle, m.sigConflicts, m.config))
s.loc.snippet = result
backendEnsureMutable s
if m.config.cmd == cmdNifC:
# mark the name so the cnif artifact writer can turn every occurrence
# into a Symbol token; stripped from the actual C output in genModule
s.locImpl.snippet = markCName(result)
else:
s.locImpl.snippet = result
proc fillParamName(m: BModule; s: PSym) =
if s.loc.snippet == "":
@@ -103,7 +148,8 @@ proc fillParamName(m: BModule; s: PSym) =
# and a function called in main or proxy uses `socket` as a parameter name.
# That would lead to either needing to reload `proxy` or to overwrite the
# executable file for the main module, which is running (or both!) -> error.
s.loc.snippet = res.rope
backendEnsureMutable s
s.locImpl.snippet = res.rope
proc fillLocalName(p: BProc; s: PSym) =
assert s.kind in skLocalVars+{skTemp}
@@ -115,10 +161,11 @@ proc fillLocalName(p: BProc; s: PSym) =
if s.kind == skTemp:
# speed up conflict search for temps (these are quite common):
if counter != 0: result.add "_" & rope(counter+1)
elif counter != 0 or isKeyword(s.name) or p.module.g.config.cppDefines.contains(key):
elif s.kind != skResult:
result.add "_" & rope(counter+1)
p.sigConflicts.inc(key)
s.loc.snippet = result
backendEnsureMutable s
s.locImpl.snippet = result
proc scopeMangledParam(p: BProc; param: PSym) =
## parameter generation only takes BModule, not a BProc, so we have to
@@ -152,8 +199,9 @@ proc getTypeName(m: BModule; typ: PType; sig: SigHash): Rope =
break
let typ = if typ.kind in {tyAlias, tySink, tyOwned}: typ.elementType else: typ
if typ.loc.snippet == "":
typ.typeName(typ.loc.snippet)
typ.loc.snippet.add $sig
backendEnsureMutable typ
typ.typeName(typ.locImpl.snippet)
typ.locImpl.snippet.add $sig
else:
when defined(debugSigHashes):
# check consistency:
@@ -243,14 +291,11 @@ proc isOrHasImportedCppType(typ: PType): bool =
searchTypeFor(typ.skipTypes({tyRef}), isImportedCppType)
proc hasNoInit(t: PType): bool =
let t = skipTypes(t, {tyGenericInst})
result = t.sym != nil and sfNoInit in t.sym.flags
proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDescKind): Rope
proc isObjLackingTypeField(typ: PType): bool {.inline.} =
result = (typ.kind == tyObject) and ((tfFinal in typ.flags) and
(typ.baseClass == nil) or isPureObject(typ))
proc isInvalidReturnType(conf: ConfigRef; typ: PType, isProc = true): bool =
# Arrays and sets cannot be returned by a C procedure, because C is
# such a poor programming language.
@@ -272,7 +317,7 @@ proc isInvalidReturnType(conf: ConfigRef; typ: PType, isProc = true): bool =
of ctStruct:
let t = skipTypes(rettype, typedescInst)
if rettype.isImportedCppType or t.isImportedCppType or
(typ.callConv == ccCDecl and conf.selectedGC in {gcArc, gcAtomicArc, gcOrc}):
(typ.callConv == ccCDecl and conf.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc}):
# prevents nrvo for cdecl procs; # bug #23401
result = false
else:
@@ -289,7 +334,12 @@ proc cacheGetType(tab: TypeCache; sig: SigHash): Rope =
result = tab.getOrDefault(sig)
proc addAbiCheck(m: BModule; t: PType, name: Rope) =
if isDefined(m.config, "checkAbi") and (let size = getSize(m.config, t); size != szUnknownSize):
if isDefined(m.config, "checkAbi") and (let size = getSize(m.config, t); size != szUnknownSize) and
not (t.kind == tyObject and searchTypeFor(t, proc (t: PType): bool {.nimcall.} = t.kind == tyUncheckedArray)):
# `UncheckedArray`, not `ptr UncheckedArray` type field in object types is a flexible array.
# `sizeof` in C and Nim doesn't always return the same value for object types containing it.
# making `getSize` in Nim always returns the same value as `sizeof` in C from flexible arrays seems hard.
# See `SEQ_DECL_SIZE` in lib/nimbase.h
var msg = "backend & Nim disagree on size for: "
msg.addTypeHeader(m.config, t)
var msg2 = ""
@@ -299,12 +349,13 @@ proc addAbiCheck(m: BModule; t: PType, name: Rope) =
proc fillResult(conf: ConfigRef; param: PNode, proctype: PType) =
fillLoc(param.sym.loc, locParam, param, "Result",
backendEnsureMutable param.sym
fillLoc(param.sym.locImpl, locParam, param, "Result",
OnStack)
let t = param.sym.typ
if mapReturnType(conf, t) != ctArray and isInvalidReturnType(conf, proctype):
incl(param.sym.loc.flags, lfIndirect)
param.sym.loc.storage = OnUnknown
incl(param.sym.locImpl.flags, lfIndirect)
param.sym.locImpl.storage = OnUnknown
proc typeNameOrLiteral(m: BModule; t: PType, literal: string): Rope =
if t.sym != nil and sfImportc in t.sym.flags and t.sym.magic == mNone:
@@ -328,6 +379,10 @@ proc getSimpleTypeDesc(m: BModule; typ: PType): Rope =
cgsym(m, "NimStrPayload")
cgsym(m, "NimStringV2")
result = typeNameOrLiteral(m, typ, "NimStringV2")
of 3:
cgsym(m, "LongString")
cgsym(m, "SmallString")
result = typeNameOrLiteral(m, typ, "SmallString")
else:
cgsym(m, "NimStringDesc")
result = typeNameOrLiteral(m, typ, "NimStringDesc*")
@@ -358,6 +413,12 @@ proc getSimpleTypeDesc(m: BModule; typ: PType): Rope =
m.typeCache[sig] = result
proc pushType(m: BModule; typ: PType) =
when defined(icDbgRefc):
if typ.kind == tySequence and
typ.elementType.skipTypes({tyGenericInst, tyAlias, tySink}).kind == tyGenericParam:
echo "[icRefc] pushType seq-of-genericparam t=", typeToString(typ),
" itemId=", typ.itemId.module, ".", typ.itemId.item, " mod=", m.module.name.s
echo getStackTrace()
for i in 0..high(m.typeStack):
# pointer equality is good enough here:
if m.typeStack[i] == typ: return
@@ -442,6 +503,14 @@ proc getSeqPayloadType(m: BModule; t: PType): Rope =
result = getTypeDescWeak(m, t, check, dkParam) & "_Content"
#result = getTypeForward(m, t, hashType(t)) & "_Content"
proc seqPayloadElem(m: BModule; t: PType): Snippet =
## Returns the C type name for a seq's element as stored in the payload,
## suitable for sizeof()/alignof(). Must use dkVar, not the dkParam default,
## because reified openArrays (experimental views) differ: dkParam gives a
## bare pointer (T*) while dkVar gives the two-word struct actually stored.
var check = initIntSet()
result = getTypeDescAux(m, t.elementType, check, dkVar)
proc seqV2ContentType(m: BModule; t: PType; check: var IntSet) =
let sig = hashType(t, m.config)
let result = cacheGetType(m.typeCache, sig)
@@ -523,14 +592,15 @@ proc genMemberProcParams(m: BModule; prc: PSym, superCall, rettype, name, params
var types, names, args: seq[string] = @[]
if not isCtor:
var this = t.n[1].sym
backendEnsureMutable this
fillParamName(m, this)
fillLoc(this.loc, locParam, t.n[1],
fillLoc(this.locImpl, locParam, t.n[1],
this.paramStorageLoc)
if this.typ.kind == tyPtr:
this.loc.snippet = "this"
this.locImpl.snippet = "this"
else:
this.loc.snippet = "(*this)"
names.add this.loc.snippet
this.locImpl.snippet = "(*this)"
names.add this.locImpl.snippet
types.add getTypeDescWeak(m, this.typ, check, dkParam)
let firstParam = if isCtor: 1 else: 2
@@ -544,13 +614,14 @@ proc genMemberProcParams(m: BModule; prc: PSym, superCall, rettype, name, params
else:
descKind = dkRefParam
var typ, name: string
backendEnsureMutable param
fillParamName(m, param)
fillLoc(param.loc, locParam, t.n[i],
fillLoc(param.locImpl, locParam, t.n[i],
param.paramStorageLoc)
if ccgIntroducedPtr(m.config, param, t.returnType) and descKind == dkParam:
typ = getTypeDescWeak(m, param.typ, check, descKind) & "*"
incl(param.loc.flags, lfIndirect)
param.loc.storage = OnUnknown
incl(param.locImpl.flags, lfIndirect)
param.locImpl.storage = OnUnknown
elif weakDep:
typ = getTypeDescWeak(m, param.typ, check, descKind)
else:
@@ -558,7 +629,7 @@ proc genMemberProcParams(m: BModule; prc: PSym, superCall, rettype, name, params
if sfNoalias in param.flags:
typ.add("NIM_NOALIAS ")
name = param.loc.snippet
name = param.locImpl.snippet
types.add typ
names.add name
if sfCodegenDecl notin param.flags:
@@ -587,12 +658,24 @@ proc genProcParams(m: BModule; t: PType, rettype: var Rope, params: var Builder,
if t.returnType == nil or isInvalidReturnType(m.config, t):
rettype = CVoid
else:
rettype = getTypeDescAux(m, t.returnType, check, dkResult)
rettype = getTypeDescWeak(m, t.returnType, check, dkResult)
var paramBuilder: ProcParamBuilder
params.addProcParams(paramBuilder):
for i in 1..<t.n.len:
if t.n[i].kind != nkSym: internalError(m.config, t.n.info, "genProcParams")
var param = t.n[i].sym
# The hidden closure environment param (`:envP`) is not a real C parameter:
# the environment is passed via the trailing `ClE_0` (added below) and
# `closureSetup` materialises `:envP` as a local cast of it. In a from-source
# build `:envP` only lives in the routine's AST params, never in the proc
# *type's* `n`, so it never reaches here. Under IC `closureParams` re-shares
# the AST param node with `typ.n`, so the lifted `:envP` leaks into `t.n`;
# emitting it would produce a bogus extra parameter that collides with the
# `closureSetup` local (the "redeclared as different kind of symbol" / env
# pointer-type mismatch). We still must fill its name/loc (later passes such
# as `assignParam` and `closureSetup` reference it), but it is omitted from
# the C signature to match the from-source ABI.
let isClosureEnv = t.callConv == ccClosure and param.name.s == ":envP"
var descKind = dkParam
if m.config.backend == backendCpp and optByRef in param.options:
if param.typ.kind == tyGenericInst:
@@ -600,14 +683,16 @@ proc genProcParams(m: BModule; t: PType, rettype: var Rope, params: var Builder,
else:
descKind = dkRefParam
if isCompileTimeOnly(param.typ): continue
backendEnsureMutable param
fillParamName(m, param)
fillLoc(param.loc, locParam, t.n[i],
fillLoc(param.locImpl, locParam, t.n[i],
param.paramStorageLoc)
if isClosureEnv: continue # name/loc filled, but not part of the C signature
var typ: Rope
if ccgIntroducedPtr(m.config, param, t.returnType) and descKind == dkParam:
typ = ptrType(getTypeDescWeak(m, param.typ, check, descKind))
incl(param.loc.flags, lfIndirect)
param.loc.storage = OnUnknown
incl(param.locImpl.flags, lfIndirect)
param.locImpl.storage = OnUnknown
elif weakDep:
typ = (getTypeDescWeak(m, param.typ, check, descKind))
else:
@@ -619,9 +704,9 @@ proc genProcParams(m: BModule; t: PType, rettype: var Rope, params: var Builder,
var j = 0
while arr.kind in {tyOpenArray, tyVarargs}:
# this fixes the 'sort' bug:
if param.typ.kind in {tyVar, tyLent}: param.loc.storage = OnUnknown
if param.typ.kind in {tyVar, tyLent}: param.locImpl.storage = OnUnknown
# need to pass hidden parameter:
params.addParam(paramBuilder, name = param.loc.snippet & "Len_" & $j, typ = NimInt)
params.addParam(paramBuilder, name = param.locImpl.snippet & "Len_" & $j, typ = NimInt)
inc(j)
arr = arr[0].skipTypes({tySink})
if t.returnType != nil and isInvalidReturnType(m.config, t):
@@ -706,12 +791,13 @@ proc genRecordFieldsAux(m: BModule; n: PNode,
if field.typ.kind == tyVoid: return
#assert(field.ast == nil)
let sname = mangleRecFieldName(m, field)
fillLoc(field.loc, locField, n, unionPrefix & sname, OnUnknown)
backendEnsureMutable field
fillLoc(field.locImpl, locField, n, unionPrefix & sname, OnUnknown)
# for importcpp'ed objects, we only need to set field.loc, but don't
# have to recurse via 'getTypeDescAux'. And not doing so prevents problems
# with heavily templatized C++ code:
if not isImportedCppType(rectype):
let fieldType = field.loc.lode.typ.skipTypes(abstractInst)
let fieldType = field.loc.t.skipTypes(abstractInst)
var typ: Rope = ""
var isFlexArray = false
var initializer = ""
@@ -759,6 +845,8 @@ proc fillObjectFields*(m: BModule; typ: PType) =
var check = initIntSet()
var ignored = newBuilder("")
addRecordFields(ignored, m, typ, check)
if typ.baseClass != nil:
fillObjectFields(m, typ.baseClass.skipTypes(skipPtrs))
proc mangleDynLibProc(sym: PSym): Rope
@@ -783,6 +871,8 @@ proc getTupleDesc(m: BModule; typ: PType, name: Rope,
var res = newBuilder("")
res.addStruct(m, typ, name, ""):
for i, a in typ.ikids:
# Do not produce code for void types
if isEmptyType(a): continue
res.addField(
name = "Field" & $i,
typ = getTypeDescAux(m, a, check, dkField))
@@ -832,6 +922,54 @@ proc getOpenArrayDesc(m: BModule; t: PType, check: var IntSet; kind: TypeDescKin
m.s[cfsTypes].addField(name = "Field0", typ = ptrType(elemType))
m.s[cfsTypes].addField(name = "Field1", typ = NimInt)
proc importedCppObject(m: BModule; t, tt: PType; check: var IntSet; kind: TypeDescKind; sig: SigHash; result: var Rope) =
let cppNameAsRope = getTypeName(m, t, sig)
let cppName = $cppNameAsRope
var i = 0
var chunkStart = 0
template addResultType(ty: untyped) =
if ty == nil or ty.kind == tyVoid:
result.add(CVoid)
elif ty.kind == tyStatic:
internalAssert m.config, ty.n != nil
result.add ty.n.renderTree
else:
result.add getTypeDescAux(m, ty, check, kind)
while i < cppName.len:
if cppName[i] == '\'':
var chunkEnd = i-1
var idx, stars: int = 0
if scanCppGenericSlot(cppName, i, idx, stars):
result.add cppName.substr(chunkStart, chunkEnd)
chunkStart = i
let typeInSlot = resolveStarsInCppType(tt, idx + 1, stars)
addResultType(typeInSlot)
else:
inc i
if chunkStart != 0:
result.add cppName.substr(chunkStart)
else:
result = cppNameAsRope & "<"
for needsComma, a in tt.genericInstParams:
if needsComma: result.add(" COMMA ")
addResultType(a)
result.add("> ")
# always call for sideeffects:
assert t.kind != tyTuple
discard getRecordDesc(m, t, result, check)
# The resulting type will include commas and these won't play well
# with the C macros for defining procs such as N_NIMCALL. We must
# create a typedef for the type and use it in the proc signature:
let typedefName = "TY" & $sig
m.s[cfsTypes].addTypedef(name = typedefName):
m.s[cfsTypes].add(result)
m.typeCache[sig] = typedefName
result = typedefName
proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDescKind): Rope =
# returns only the type's name
var t = origTyp.skipTypes(irrelevantForBackend-{tyOwned})
@@ -847,7 +985,7 @@ proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDes
# tyDistinct matters if it is an importc type
result = getTypePre(m, origTyp.skipTypes(irrelevantForBackend-{tyOwned, tyDistinct}), sig)
defer: # defer is the simplest in this case
defer:
if isImportedType(t) and not m.typeABICache.containsOrIncl(sig):
addAbiCheck(m, t, result)
@@ -981,7 +1119,7 @@ proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDes
m.s[cfsTypes].addArrayTypedef(name = result, len = 1):
m.s[cfsTypes].add(et)
of tyArray:
var n: BiggestInt = toInt64(lengthOrd(m.config, t))
var n = toInt64(lengthOrd(m.config, t))
if n <= 0: n = 1 # make an array of at least one element
result = getTypeName(m, origTyp, sig)
m.typeCache[sig] = result
@@ -992,52 +1130,7 @@ proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDes
of tyObject, tyTuple:
let tt = origTyp.skipTypes({tyDistinct})
if isImportedCppType(t) and tt.kind == tyGenericInst:
let cppNameAsRope = getTypeName(m, t, sig)
let cppName = $cppNameAsRope
var i = 0
var chunkStart = 0
template addResultType(ty: untyped) =
if ty == nil or ty.kind == tyVoid:
result.add(CVoid)
elif ty.kind == tyStatic:
internalAssert m.config, ty.n != nil
result.add ty.n.renderTree
else:
result.add getTypeDescAux(m, ty, check, kind)
while i < cppName.len:
if cppName[i] == '\'':
var chunkEnd = i-1
var idx, stars: int = 0
if scanCppGenericSlot(cppName, i, idx, stars):
result.add cppName.substr(chunkStart, chunkEnd)
chunkStart = i
let typeInSlot = resolveStarsInCppType(tt, idx + 1, stars)
addResultType(typeInSlot)
else:
inc i
if chunkStart != 0:
result.add cppName.substr(chunkStart)
else:
result = cppNameAsRope & "<"
for needsComma, a in tt.genericInstParams:
if needsComma: result.add(" COMMA ")
addResultType(a)
result.add("> ")
# always call for sideeffects:
assert t.kind != tyTuple
discard getRecordDesc(m, t, result, check)
# The resulting type will include commas and these won't play well
# with the C macros for defining procs such as N_NIMCALL. We must
# create a typedef for the type and use it in the proc signature:
let typedefName = "TY" & $sig
m.s[cfsTypes].addTypedef(name = typedefName):
m.s[cfsTypes].add(result)
m.typeCache[sig] = typedefName
result = typedefName
importedCppObject(m, t, tt, check, kind, sig, result)
else:
result = cacheGetType(m.forwTypeCache, sig)
if result == "":
@@ -1052,6 +1145,7 @@ proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDes
else: getTupleDesc(m, t, result, check)
if not isImportedType(t):
m.s[cfsTypes].add(recdesc)
addAbiCheck(m, t, result)
elif tfIncompleteStruct notin t.flags:
discard # addAbiCheck(m, t, result) # already handled elsewhere
of tySet:
@@ -1073,6 +1167,11 @@ proc getTypeDescAux(m: BModule; origTyp: PType, check: var IntSet; kind: TypeDes
tyUserTypeClass, tyUserTypeClassInst, tyInferred:
result = getTypeDescAux(m, skipModifier(t), check, kind)
else:
when defined(icDbgRefc):
echo "[icRefc] getTypeDescAux ", t.kind, " t=", typeToString(t),
" origTyp=", typeToString(origTyp), " t.itemId=", t.itemId.module, ".", t.itemId.item,
" sym=", (if t.sym != nil: t.sym.name.s else: "nil"),
" owner=", (if t.owner != nil: t.owner.name.s else: "nil")
internalError(m.config, "getTypeDescAux(" & $t.kind & ')')
result = ""
# fixes bug #145:
@@ -1111,6 +1210,10 @@ proc finishTypeDescriptions(m: BModule) =
var check = initIntSet()
while i < m.typeStack.len:
let t = m.typeStack[i]
when defined(icDbgRefc):
echo "[icRefc] finishTypeDescriptions[", i, "] mod=", m.module.name.s,
" t=", typeToString(t), " kind=", t.kind,
" itemId=", t.itemId.module, ".", t.itemId.item
if optSeqDestructors in m.config.globalOptions and t.skipTypes(abstractInst).kind == tySequence:
seqV2ContentType(m, t, check)
else:
@@ -1152,7 +1255,8 @@ proc genMemberProcHeader(m: BModule; prc: PSym; result: var Builder; asPtr: bool
let isCtor = sfConstructor in prc.flags
var check = initIntSet()
fillBackendName(m, prc)
fillLoc(prc.loc, locProc, prc.ast[namePos], OnUnknown)
backendEnsureMutable prc
fillLoc(prc.locImpl, locProc, prc.ast[namePos], OnUnknown)
var memberOp = "#." #only virtual
var typ: PType
if isCtor:
@@ -1184,7 +1288,7 @@ proc genMemberProcHeader(m: BModule; prc: PSym; result: var Builder; asPtr: bool
superCall = ""
else:
if not isCtor:
prc.loc.snippet = "$1$2(@)" % [memberOp, name]
prc.locImpl.snippet = "$1$2(@)" % [memberOp, name]
elif superCall != "":
superCall = " : " & superCall
@@ -1199,14 +1303,15 @@ proc genProcHeader(m: BModule; prc: PSym; result: var Builder; visibility: var D
# using static is needed for inline procs
var check = initIntSet()
fillBackendName(m, prc)
fillLoc(prc.loc, locProc, prc.ast[namePos], OnUnknown)
backendEnsureMutable prc
fillLoc(prc.locImpl, locProc, prc.ast[namePos], OnUnknown)
var rettype: Snippet = ""
var desc = newBuilder("")
genProcParams(m, prc.typ, rettype, desc, check, true, false)
let params = extract(desc)
# handle the 2 options for hotcodereloading codegen - function pointer
# (instead of forward declaration) or header for function body with "_actual" postfix
var name = prc.loc.snippet
var name = prc.locImpl.snippet
if not asPtr and isReloadable(m, prc):
name.add("_actual")
# careful here! don't access ``prc.ast`` as that could reload large parts of
@@ -1223,7 +1328,9 @@ proc genProcHeader(m: BModule; prc: PSym; result: var Builder; visibility: var D
elif prc.typ.callConv == ccInline or isNonReloadable(m, prc):
visibility = StaticProc
elif sfImportc notin prc.flags:
visibility = Private
if not isSharedInstanceCName(m, prc):
visibility = Private
# else: plain extern — the definition is shared across TUs
if asPtr:
result.addProcVar(m, prc, name, params, rettype, isStatic = isStaticVar, ignoreAttributes = true)
else:
@@ -1303,6 +1410,8 @@ proc genTypeInfoAuxBase(m: BModule; typ, origType: PType;
else:
m.s[cfsStrData].addDeclWithVisibility(Private):
m.s[cfsStrData].addVar(kind = Local, name = name, typ = "TNimType")
if m.config.cmd == cmdNifC:
m.icDataDefs.add (name, icNifName(m, origType))
proc genTypeInfoAux(m: BModule; typ, origType: PType, name: Rope;
info: TLineInfo) =
@@ -1446,30 +1555,41 @@ proc genObjectInfo(m: BModule; typ, origType: PType, name: Rope; info: TLineInfo
var t = typ.baseClass
while t != nil:
t = t.skipTypes(skipPtrs)
t.flags.incl tfObjHasKids
t.incl tfObjHasKids
t = t.baseClass
proc validTupleTypeFields(t: PType): int =
# we want to treat tuples with only void fields as empty, so we need to exclude void types here:
result = 0
for a in t.kids:
if not isEmptyType(a): inc result
proc genTupleInfo(m: BModule; typ, origType: PType, name: Rope; info: TLineInfo) =
genTypeInfoAuxBase(m, typ, typ, name, cIntValue(0), info)
var expr = getNimNode(m)
if not typ.isEmptyTupleType:
var tmp = getTempName(m) & "_" & $typ.kidsLen
genTNimNodeArray(m, tmp, typ.kidsLen)
let nonVoidKids = validTupleTypeFields(typ)
if nonVoidKids > 0:
var tmp = getTempName(m) & "_" & $nonVoidKids
genTNimNodeArray(m, tmp, nonVoidKids)
var j = 0
for i, a in typ.ikids:
# Do not produce code for void types
if isEmptyType(a): continue
var tmp2 = getNimNode(m)
let fieldTypInfo = genTypeInfoV1(m, a, info)
m.s[cfsTypeInit3].addSubscriptAssignment(tmp, cIntValue(i), cAddr(tmp2))
m.s[cfsTypeInit3].addSubscriptAssignment(tmp, cIntValue(j), cAddr(tmp2))
m.s[cfsTypeInit3].addFieldAssignment(tmp2, "kind", 1)
m.s[cfsTypeInit3].addFieldAssignmentWithValue(tmp2, "offset"):
m.s[cfsTypeInit3].addOffsetof(getTypeDesc(m, origType, dkVar), "Field" & $i)
m.s[cfsTypeInit3].addFieldAssignment(tmp2, "typ", fieldTypInfo)
m.s[cfsTypeInit3].addFieldAssignment(tmp2, "name", "\"Field" & $i & "\"")
m.s[cfsTypeInit3].addFieldAssignment(expr, "len", typ.kidsLen)
inc j
m.s[cfsTypeInit3].addFieldAssignment(expr, "len", nonVoidKids)
m.s[cfsTypeInit3].addFieldAssignment(expr, "kind", 2)
m.s[cfsTypeInit3].addFieldAssignment(expr, "sons",
cAddr(subscript(tmp, cIntValue(0))))
else:
m.s[cfsTypeInit3].addFieldAssignment(expr, "len", typ.kidsLen)
m.s[cfsTypeInit3].addFieldAssignment(expr, "len", cIntValue(0))
m.s[cfsTypeInit3].addFieldAssignment(expr, "kind", 2)
m.s[cfsTypeInit3].addFieldAssignment(tiNameForHcr(m, name), "node", cAddr(expr))
@@ -1579,8 +1699,13 @@ proc declareNimType(m: BModule; name: string; str: Rope, module: int) =
m.s[cfsTypeInit1].addArgument(hcrGlobal):
m.s[cfsTypeInit1].add("\"" & str & "\"")
else:
# cnif-mark the name: this extern declaration is the reference the
# def-retention check consults when the defining TU regenerates and
# the typeinfo cannot be re-demanded (type vanished) — the referencing
# TU must lose its reuse then instead of producing a link error
let declName = if m.config.cmd == cmdNifC: markCName(str) else: str
m.s[cfsStrData].addDeclWithVisibility(Extern):
m.s[cfsStrData].addVar(kind = Local, name = str, typ = nr)
m.s[cfsStrData].addVar(kind = Local, name = declName, typ = nr)
proc genTypeInfo2Name(m: BModule; t: PType): Rope =
var it = t
@@ -1642,8 +1767,8 @@ proc generateRttiDestructor(g: ModuleGraph; typ: PType; owner: PSym; kind: TType
n[bodyPos] = body
result.ast = n
incl result.flags, sfFromGeneric
incl result.flags, sfGeneratedOp
incl result.flagsImpl, sfFromGeneric
incl result.flagsImpl, sfGeneratedOp
proc genHook(m: BModule; t: PType; info: TLineInfo; op: TTypeAttachedOp; result: var Builder) =
let theProc = getAttachedOp(m.g.graph, t, op)
@@ -1669,7 +1794,7 @@ proc genHook(m: BModule; t: PType; info: TLineInfo; op: TTypeAttachedOp; result:
echo "ayclic but has this =trace ", t, " ", theProc.ast
else:
when false:
if op == attachedTrace and m.config.selectedGC == gcOrc and
if op == attachedTrace and m.config.selectedGC in {gcOrc, gcYrc} and
containsGarbageCollectedRef(t):
# unfortunately this check is wrong for an object type that only contains
# .cursor fields like 'Node' inside 'cycleleak'.
@@ -1719,6 +1844,8 @@ proc genTypeInfoV2OldImpl(m: BModule; t, origType: PType, name: Rope; info: TLin
cgsym(m, "TNimTypeV2")
m.s[cfsStrData].addDeclWithVisibility(Private):
m.s[cfsStrData].addVar(kind = Local, name = name, typ = "TNimTypeV2")
if m.config.cmd == cmdNifC:
m.icDataDefs.add (name, icNifName(m, origType))
var flags = 0
if not canFormAcycle(m.g.graph, t): flags = flags or 1
@@ -1781,8 +1908,15 @@ proc genTypeInfoV2OldImpl(m: BModule; t, origType: PType, name: Rope; info: TLin
proc genTypeInfoV2Impl(m: BModule; t, origType: PType, name: Rope; info: TLineInfo) =
cgsym(m, "TNimTypeV2")
m.s[cfsStrData].addDeclWithVisibility(Private):
# Under `nim nifc` every `cg` process that demands this type's RTTI emits its
# definition (emit-everywhere). The forward declaration must therefore be a
# real `extern` (not a tentative definition) so a TU whose copy the merge
# stage drops still only *declares* it; the definition itself is wrapped as a
# droppable `'d'` unit below and assigned to a single owner.
m.s[cfsStrData].addDeclWithVisibility(if m.config.cmd == cmdNifC: Extern else: Private):
m.s[cfsStrData].addVar(kind = Local, name = name, typ = "TNimTypeV2")
if m.config.cmd == cmdNifC:
m.icDataDefs.add (name, icNifName(m, origType))
var flags = 0
if not canFormAcycle(m.g.graph, t): flags = flags or 1
@@ -1843,11 +1977,22 @@ proc genTypeInfoV2Impl(m: BModule; t, origType: PType, name: Rope; info: TLineIn
else:
typeEntry.addField(typeInit, name = "flags"):
typeEntry.addIntValue(flags)
m.s[cfsVars].add extract(typeEntry)
if m.config.cmd == cmdNifC:
m.s[cfsVars].add(cnifDefDirective(name, "d", icNifName(m, origType)))
m.s[cfsVars].add extract(typeEntry)
m.s[cfsVars].add(cnifEndDefs())
else:
m.s[cfsVars].add extract(typeEntry)
if t.kind == tyObject and t.baseClass != nil and optEnableDeepCopy in m.config.globalOptions:
discard genTypeInfoV1(m, t, info)
proc myModuleOpenForCodegen(m: BModule; idx: FileIndex): bool {.inline.} =
if moduleOpenForCodegen(m.g.graph, idx):
result = idx.int < m.g.mods.len and m.g.mods[idx.int] != nil
else:
result = false
proc genTypeInfoV2(m: BModule; t: PType; info: TLineInfo): Rope =
let origType = t
# distinct types can have their own destructors
@@ -1876,9 +2021,15 @@ proc genTypeInfoV2(m: BModule; t: PType; info: TLineInfo): Rope =
m.typeInfoMarkerV2[sig] = result
let owner = t.skipTypes(typedescPtrs).itemId.module
if owner != m.module.position and moduleOpenForCodegen(m.g.graph, FileIndex owner):
# In the per-module backend (`cg`) RTTI is emit-everywhere like procs and
# consts: every demanding module emits the `'d'` definition (deduped to one
# owner by the merge stage). The owner-routing below would instead push the
# definition into the owner module's *unwritten* backend module (discarded in
# this process) and emit only an extern here, leaving the symbol undefined.
let perModuleCg = m.config.cmd == cmdNifC and m.config.icBackendStage == "cg"
if not perModuleCg and owner != m.module.position and myModuleOpenForCodegen(m, FileIndex owner):
# make sure the type info is created in the owner module
discard genTypeInfoV2(m.g.modules[owner], origType, info)
discard genTypeInfoV2(m.g.mods[owner], origType, info)
# reference the type info as extern here
cgsym(m, "TNimTypeV2")
declareNimType(m, "TNimTypeV2", result, owner)
@@ -1943,6 +2094,10 @@ proc genTypeInfoV1(m: BModule; t: PType; info: TLineInfo): Rope =
let marker = m.g.typeInfoMarker.getOrDefault(sig)
if marker.str != "":
when defined(icDbgRefc):
if "catchableerror" in marker.str:
echo "[icNti] ", marker.str, " in mod=", m.module.name.s,
" -> extern:globalMarker owner=", marker.owner
cgsym(m, "TNimType")
cgsym(m, "TNimNode")
declareNimType(m, "TNimType", marker.str, marker.owner)
@@ -1953,17 +2108,26 @@ proc genTypeInfoV1(m: BModule; t: PType; info: TLineInfo): Rope =
result = "NTI$1$2_" % [rope(typeToC(t)), rope($sig)]
m.typeInfoMarker[sig] = result
when defined(icDbgRefc):
template dbgNti(branch: string) =
if "catchableerror" in result:
echo "[icNti] ", result, " in mod=", m.module.name.s, " -> ", branch
else:
template dbgNti(branch: string) = discard
let old = m.g.graph.emittedTypeInfo.getOrDefault($result)
if old != FileIndex(0):
dbgNti "extern:emittedTypeInfo"
cgsym(m, "TNimType")
cgsym(m, "TNimNode")
declareNimType(m, "TNimType", result, old.int)
return prefixTI(result)
var owner = t.skipTypes(typedescPtrs).itemId.module
if owner != m.module.position and moduleOpenForCodegen(m.g.graph, FileIndex owner):
if owner != m.module.position and myModuleOpenForCodegen(m, FileIndex owner):
dbgNti "extern:ownerRouted"
# make sure the type info is created in the owner module
discard genTypeInfoV1(m.g.modules[owner], origType, info)
discard genTypeInfoV1(m.g.mods[owner], origType, info)
# reference the type info as extern here
cgsym(m, "TNimType")
cgsym(m, "TNimNode")
@@ -1972,8 +2136,9 @@ proc genTypeInfoV1(m: BModule; t: PType; info: TLineInfo): Rope =
else:
owner = m.module.position.int32
dbgNti "DEFINED-HERE"
m.g.typeInfoMarker[sig] = (str: result, owner: owner)
rememberEmittedTypeInfo(m.g.graph, FileIndex(owner), $result)
#rememberEmittedTypeInfo(m.g.graph, FileIndex(owner), $result)
case t.kind
of tyEmpty, tyVoid: result = cIntValue(0)
@@ -1999,6 +2164,9 @@ proc genTypeInfoV1(m: BModule; t: PType; info: TLineInfo): Rope =
of tyRef:
genTypeInfoAux(m, t, t, result, info)
if m.config.selectedGC in {gcMarkAndSweep, gcRefc, gcGo}:
# it may not be used in other places except in `genTraverseProc`,
# we have to generate a typedesc for this case, not a weak one
discard getTypeDesc(m, origType.last)
let markerProc = genTraverseProc(m, origType, sig)
m.s[cfsTypeInit3].addFieldAssignment(tiNameForHcr(m, result), "marker", markerProc)
of tyPtr, tyRange, tyUncheckedArray: genTypeInfoAux(m, t, t, result, info)
@@ -2037,17 +2205,21 @@ proc genTypeInfo*(config: ConfigRef, m: BModule; t: PType; info: TLineInfo): Rop
else:
result = genTypeInfoV1(m, t, info)
proc retrieveSym(n: PNode): PSym =
case n.kind
of nkPostfix: result = retrieveSym(n[1])
of nkPragmaExpr, nkTypeDef: result = retrieveSym(n[0])
of nkSym: result = n.sym
else: result = nil
proc genTypeSection(m: BModule, n: PNode) =
var intSet = initIntSet()
for i in 0..<n.len:
if len(n[i]) == 0: continue
if n[i][0].kind != nkPragmaExpr: continue
for p in 0..<n[i][0].len:
if (n[i][0][p].kind notin {nkSym, nkPostfix}): continue
var s = n[i][0][p]
if s.kind == nkPostfix:
s = n[i][0][p][1]
if {sfExportc, sfCompilerProc} * s.sym.flags == {sfExportc}:
discard getTypeDescAux(m, s.typ, intSet, descKindFromSymKind(s.sym.kind))
if m.g.generatedHeader != nil:
discard getTypeDescAux(m.g.generatedHeader, s.typ, intSet, descKindFromSymKind(s.sym.kind))
let compress = optCompress in m.config.globalOptions
for typedef in n:
let s = retrieveSym(typedef)
if s != nil and ({sfExportc, sfCompilerProc} * s.flags == {sfExportc} or compress) and s.typ != nil and
not containsGenericType(s.typ) and
s.typ.kind notin {tyVoid, tyNot, tyAnything, tyOr, tyAnd, tyUntyped, tyTyped, tyNone, tyNil, tySink}:
discard getTypeDescAux(m, s.typ, intSet, descKindFromSymKind(s.kind))
if m.g.generatedHeader != nil:
discard getTypeDescAux(m.g.generatedHeader, s.typ, intSet, descKindFromSymKind(s.kind))

View File

@@ -11,7 +11,7 @@
import
ast, types, msgs, wordrecg,
platform, trees, options, cgendata, mangleutils
platform, trees, options, cgendata, mangleutils, renderer
import std/[hashes, strutils, formatfloat]
@@ -90,9 +90,6 @@ proc ccgIntroducedPtr*(conf: ConfigRef; s: PSym, retType: PType): bool =
if s.typ.sym != nil and sfForward in s.typ.sym.flags:
# forwarded objects are *always* passed by pointers for consistency!
result = true
elif s.typ.kind == tySink and conf.selectedGC notin {gcArc, gcAtomicArc, gcOrc, gcHooks}:
# bug #23354:
result = false
elif (optByRef in s.options) or (getSize(conf, pt) > conf.target.floatSize * 3):
result = true # requested anyway
elif (tfFinal in pt.flags) and (pt[0] == nil):
@@ -115,19 +112,22 @@ proc encodeName*(name: string): string =
proc makeUnique(m: BModule; s: PSym, name: string = ""): string =
result = if name == "": s.name.s else: name
# keep backend-minted ids out of the `_u` namespace; their item counter
# restarts at 0 and would collide with loaded symbols' ids
result.add(if s.itemId.isBackendMinted: "_c" else: "_u")
result.add $s.itemId.item
# module suffix LAST (a strippable trailing token; see `mangleProcNameExt`)
result.add "__"
result.add m.g.graph.ifaces[s.itemId.module].uniqueName
result.add "_u"
result.add $s.itemId.item
proc encodeSym*(m: BModule; s: PSym; makeUnique: bool = false): string =
proc encodeSym*(m: BModule; s: PSym; makeUnique: bool = false; extra: string = ""): string =
#Module::Type
var name = s.name.s
var name = s.name.s & extra
if makeUnique:
name = makeUnique(m, s, name)
"N" & encodeName(s.skipGenericOwner.name.s) & encodeName(name) & "E"
proc encodeType*(m: BModule; t: PType): string =
proc encodeType*(m: BModule; t: PType; staticLists: var string): string =
result = ""
var kindName = ($t.kind)[2..^1]
kindName[0] = toLower($kindName[0])[0]
@@ -138,10 +138,10 @@ proc encodeType*(m: BModule; t: PType): string =
result = encodeName(t[0].sym.name.s)
result.add "I"
for i in 1..<t.len - 1:
result.add encodeType(m, t[i])
result.add encodeType(m, t[i], staticLists)
result.add "E"
of tySequence, tyOpenArray, tyArray, tyVarargs, tyTuple, tyProc, tySet, tyTypeDesc,
tyPtr, tyRef, tyVar, tyLent, tySink, tyStatic, tyUncheckedArray, tyOr, tyAnd, tyBuiltInTypeClass:
tyPtr, tyRef, tyVar, tyLent, tySink, tyUncheckedArray, tyOr, tyAnd, tyBuiltInTypeClass:
result =
case t.kind:
of tySequence: encodeName("seq")
@@ -150,8 +150,13 @@ proc encodeType*(m: BModule; t: PType): string =
for i in 0..<t.len:
let s = t[i]
if s.isNil: continue
result.add encodeType(m, s)
result.add encodeType(m, s, staticLists)
result.add "E"
of tyStatic:
if t.n != nil:
staticLists.add "_s" & renderTree(t.n)
else:
raiseAssert "unreachable"
of tyRange:
var val = "range_"
if t.n[0].typ.kind in {tyFloat..tyFloat128}:
@@ -164,7 +169,7 @@ proc encodeType*(m: BModule; t: PType): string =
of tyString..tyUInt64, tyPointer, tyBool, tyChar, tyVoid, tyAnything, tyNil, tyEmpty:
result = encodeName(kindName)
of tyAlias, tyInferred, tyOwned:
result = encodeType(m, t.elementType)
result = encodeType(m, t.elementType, staticLists)
else:
assert false, "encodeType " & $t.kind

File diff suppressed because it is too large Load Diff

View File

@@ -75,10 +75,13 @@ type
flags*: set[TCProcFlag]
lastLineInfo*: TLineInfo # to avoid generating excessive 'nimln' statements
currLineInfo*: TLineInfo # AST codegen will make this superfluous
nestedTryStmts*: seq[tuple[fin: PNode, inExcept: bool, label: Natural]]
nestedTryStmts*: seq[tuple[fin: PNode, inExcept: bool, isHidden: bool, label: Natural]]
# in how many nested try statements we are
# (the vars must be volatile then)
# bool is true when are in the except part of a try block
# `inExcept` is true when we are in the except part of a try block.
# `isHidden` is true for compiler-injected `nkHiddenTryStmt` wrappers
# (e.g. ARC's destructor try/finally around `except T as e:` bodies);
# finallyActions walks past such wrappers to reach the user's try.
finallySafePoints*: seq[Rope] # For correctly cleaning up exceptions when
# using return in finally statements
labels*: Natural # for generating unique labels in the C proc
@@ -117,9 +120,9 @@ type
BModuleList* = ref object of RootObj
mainModProcs*, mainModInit*, otherModsInit*, mainDatInit*: Builder
mapping*: Rope # the generated mapping file (if requested)
modules*: seq[BModule] # list of all compiled modules
mods*: seq[BModule] # list of all compiled modules
modulesClosed*: seq[BModule] # list of the same compiled modules, but in the order they were closed
forwardedProcs*: seq[PSym] # proc:s that did not yet have a body
forwardedProcs*: seq[PSym] # procs that did not yet have a body
generatedHeader*: BModule
typeInfoMarker*: TypeCacheWithOwner
typeInfoMarkerV2*: TypeCacheWithOwner
@@ -155,6 +158,13 @@ type
forwTypeCache*: TypeCache # cache for forward declarations of types
declaredThings*: IntSet # things we have declared in this .c file
declaredProtos*: IntSet # prototypes we have declared in this .c file
emittedContentDefs*: HashSet[string]
# cmdNifC per-module backend: content-addressed C names (generic
# instances and synthesized hooks) whose body this TU already emitted.
# Distinct symbols (minted in different source modules) can share one
# `_i<disamb>` name; `declaredThings` keys on symbol id and lets the
# second one through, so we dedup the body by name here instead.
queue*: seq[PSym] # queue of procs to generate
alive*: IntSet # symbol IDs of alive data as computed by `dce.nim`
headerFiles*: seq[string] # needed headers to include
typeInfoMarker*: TypeCache # needed for generating type information
@@ -172,12 +182,26 @@ type
extensionLoaders*: array['0'..'9', Builder] # special procs for the
# OpenGL wrapper
sigConflicts*: CountTable[SigHash]
icImplMods*: IntSet # module ids whose routine BODIES this TU
# embeds (redirected defs, shared instances,
# hooks); recorded as the artifact's cdeps so
# the reuse gate can check their impl cookies
icDataDefs*: seq[tuple[cname, nifname: string]]
# C names of data definitions (consts, globals,
# RTTI) this TU embeds plus their NIF symbol
# names (empty for RTTI, which has no symbol);
# recorded in the cnif artifact so a later run
# can reuse the TU and re-demand definitions
# that cached TUs still reference
g*: BModuleList
template config*(m: BModule): ConfigRef = m.g.config
template config*(p: BProc): ConfigRef = p.module.g.config
template vccAndC*(p: BProc): bool = p.module.config.cCompiler == ccVcc and p.module.config.backend == backendC
proc delayedCodegen*(m: BModule): bool {.inline.} =
useAliveDataFromDce in m.flags or m.config.globalOptions.contains(optCompress)
proc includeHeader*(this: BModule; header: string) =
if not this.headerFiles.contains header:
this.headerFiles.add header

View File

@@ -55,7 +55,7 @@ proc methodCall*(n: PNode; conf: ConfigRef): PNode =
# replace ordinary method by dispatcher method:
let disp = getDispatcher(result[0].sym)
if disp != nil:
result[0].typ() = disp.typ
result[0].typ = disp.typ
result[0].sym = disp
# change the arguments to up/downcasts to fit the dispatcher's parameters:
for i in 1..<result.len:
@@ -123,8 +123,8 @@ proc attachDispatcher(s: PSym, dispatcher: PNode) =
proc createDispatcher(s: PSym; g: ModuleGraph; idgen: IdGenerator): PSym =
var disp = copySym(s, idgen)
incl(disp.flags, sfDispatcher)
excl(disp.flags, sfExported)
incl(disp, sfDispatcher)
excl(disp, sfExported)
let old = disp.typ
disp.typ = copyType(disp.typ, idgen, disp.typ.owner)
copyTypeProps(g, idgen.module, disp.typ, old)
@@ -133,7 +133,7 @@ proc createDispatcher(s: PSym; g: ModuleGraph; idgen: IdGenerator): PSym =
if disp.typ.callConv == ccInline: disp.typ.callConv = ccNimCall
disp.ast = copyTree(s.ast)
disp.ast[bodyPos] = newNodeI(nkEmpty, s.info)
disp.loc.snippet = ""
disp.locImpl.snippet = ""
if s.typ.returnType != nil:
if disp.ast.len > resultPos:
disp.ast[resultPos].sym = copySym(s.ast[resultPos].sym, idgen)
@@ -180,6 +180,7 @@ proc methodDef*(g: ModuleGraph; idgen: IdGenerator; s: PSym) =
g.methods[i].methods[0] != s:
# already exists due to forwarding definition?
localError(g.config, s.info, "method is not a base")
logMethodDef(g, s)
return
of No: discard
of Invalid:
@@ -191,6 +192,7 @@ proc methodDef*(g: ModuleGraph; idgen: IdGenerator; s: PSym) =
else:
g.bucketTable.inc(s.typ.firstParamType.skipTypes(skipPtrs).itemId)
g.methods.add((methods: @[s], dispatcher: createDispatcher(s, g, idgen)))
logMethodDef(g, s)
#echo "adding ", s.info
if witness != nil:
localError(g.config, s.info, "invalid declaration order; cannot attach '" & s.name.s &

File diff suppressed because it is too large Load Diff

726
compiler/cnif.nim Normal file
View File

@@ -0,0 +1,726 @@
#
#
# The Nim Compiler
# (c) Copyright 2026 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## The "cnif" artifact: the C code generator's output as a NIF file.
##
## This is deliberately *not* NIFC: the C text is kept verbatim (Nim's
## C-level machinery — exception handling in particular — is more refined
## than what NIFC models today; the gap can be closed incrementally later).
## The only structure the artifact adds is the part dead code elimination
## and generic-instance merging need:
##
## - raw C text as string literals
## - every *global* entity's C name as a `Symbol` token
## - every emitted proc definition as a `(cdef SymbolDef flags ...)` group
##
## The C generator marks names with control characters at the single place
## a global's C name is minted (`fillBackendName`) and emits a definition
## directive at the single place finished procs are appended; the marks then
## ride through all of the snippet composition untouched. This module turns
## the final marked module text into the `.c.nif` artifact and strips the
## marks for the actual `.c` output. Rendering C from the artifact is a
## plain token walk: string literals verbatim, symbols by name — which is
## also where a later merge step redirects losing generic instances.
##
## Marker scheme (cannot collide: C string literals escape control chars,
## and `\1`/`\31`/`\23` of cgen's postprocess directives are distinct):
## \2 name \3 a global's C name
## \4 name \31 flags \31 nif \5 start of the definition of `name`;
## `nif` is the defining symbol's NIF name
## (empty for backend-minted symbols) so a
## later run can re-demand the definition
## \4 \5 end of the definitions section
import std / [tables, sets, os, assertions, syncio, algorithm]
import "../dist/nimony/src/lib" / [nifbuilder, nifcoreparse]
const
CnifSymStart* = '\2'
CnifSymEnd* = '\3'
CnifDefStart* = '\4'
CnifDefSep* = '\31' # same separator char as cgen's postprocess directives
CnifDefEnd* = '\5'
proc markCName*(name: string): string {.inline.} =
CnifSymStart & name & CnifSymEnd
proc hasCnifMarks*(s: string): bool =
for c in s:
if c in {CnifSymStart, CnifSymEnd, CnifDefStart}: return true
false
proc stripCnifMarks*(s: string): string =
## Removes the symbol marks (keeping the names) and the definition
## directives (entirely) so the result is plain C.
if not hasCnifMarks(s): return s
result = newStringOfCap(s.len)
var i = 0
while i < s.len:
case s[i]
of CnifSymStart, CnifSymEnd:
inc i
of CnifDefStart:
while i < s.len and s[i] != CnifDefEnd: inc i
inc i # skip CnifDefEnd
else:
result.add s[i]
inc i
const
CnifVersion* = "4"
## Artifact format version, stored in the meta head. Artifacts written
## by an older compiler lack the NIF names and the cref group the
## def-retention check needs (v2), the cdeps group the fine-grained
## reuse gate needs (v3), or the type NIF names and cnif-marked extern
## RTTI references the typeinfo flavor of the def-retention check
## needs (v4); `readCnifHeads` reports them as invalid so their TUs
## simply regenerate once.
proc cnifDefDirective*(name, flags, nifName: string): string =
CnifDefStart & name & CnifDefSep & flags & CnifDefSep & nifName & CnifDefEnd
proc cnifEndDefs*(): string =
CnifDefStart & CnifDefEnd
proc writeCnifArtifact*(code: string; outfile: string;
initRequired = false; datInitRequired = false;
dataDefs: openArray[tuple[cname, nifname: string]] = [];
semmedNif = ""; moduleBase = "";
implDeps: openArray[string] = []) =
## Splits the marked module text into the `.c.nif` artifact.
## The artifact starts with a `(meta <flags> "semmedNif" "moduleBase"
## "version")` head — whether the module has an init/datInit proc
## ('i'/'d'), which semmed NIF it was generated from and the module's
## mangled base name (what `registerModuleToMain` and the reuse decision
## need when the TU is reused in a later run, possibly without the module
## ever being loaded again) — a `(cdata (SymbolDef StrLit)*)` group naming
## the data definitions (consts, globals, RTTI) the TU embeds together
## with their NIF names, a `(cref Ident*)` group naming every C name
## the TU references but does not define itself (what the def-retention
## check consults when some *other* TU regenerates), and a
## `(cdeps Ident*)` group naming the modules whose routine *bodies* this
## TU embeds (redirected defs, shared instances, hooks): the fine-grained
## reuse gate checks their `.impl.nif` cookies on top of the direct
## imports' `.iface.nif` cookies.
# pre-pass: every marked name is a use, every definition directive (and
# every data def) is a definition; external references = uses - defs
var uses = initHashSet[string]()
var defs = initHashSet[string]()
block prePass:
var i = 0
while i < code.len:
case code[i]
of CnifSymStart:
inc i
var name = ""
while i < code.len and code[i] != CnifSymEnd:
name.add code[i]
inc i
inc i
uses.incl name
of CnifDefStart:
inc i
var payload = ""
while i < code.len and code[i] != CnifDefEnd:
payload.add code[i]
inc i
inc i
let sep = find(payload, CnifDefSep)
if sep > 0: defs.incl payload[0..<sep]
elif payload.len > 0: defs.incl payload
else:
inc i
for d in dataDefs: defs.incl d.cname
var crefs: seq[string] = @[]
for u in uses:
if u notin defs: crefs.add u
sort crefs
var b = nifbuilder.open(outfile)
b.withTree "stmts":
b.withTree "meta":
var metaFlags = ""
if initRequired: metaFlags.add 'i'
if datInitRequired: metaFlags.add 'd'
if metaFlags.len > 0: b.addIdent metaFlags
else: b.addEmpty
b.addStrLit semmedNif
b.addStrLit moduleBase
b.addStrLit CnifVersion
b.withTree "cdata":
for d in dataDefs:
b.addSymbolDef d.cname
b.addStrLit d.nifname
b.withTree "cref":
for r in crefs:
b.addIdent r
b.withTree "cdeps":
for s in implDeps:
b.addIdent s
var raw = ""
var inDef = false
template flushRaw() =
if raw.len > 0:
b.addStrLit raw
raw.setLen 0
var i = 0
while i < code.len:
case code[i]
of CnifSymStart:
flushRaw()
inc i
var name = ""
while i < code.len and code[i] != CnifSymEnd:
name.add code[i]
inc i
inc i # skip CnifSymEnd
b.addSymbol name, ""
of CnifDefStart:
flushRaw()
inc i
var payload = ""
while i < code.len and code[i] != CnifDefEnd:
payload.add code[i]
inc i
inc i # skip CnifDefEnd
if inDef:
b.endTree()
inDef = false
if payload.len > 0:
let sep = find(payload, CnifDefSep)
let name = if sep >= 0: payload[0..<sep] else: payload
var flags = if sep >= 0: payload[sep+1..^1] else: ""
var nifName = ""
let sep2 = find(flags, CnifDefSep)
if sep2 >= 0:
nifName = flags[sep2+1..^1]
flags = flags[0..<sep2]
b.addTree "cdef"
b.addSymbolDef name
if flags.len > 0: b.addIdent flags
else: b.addEmpty
b.addStrLit nifName
inDef = true
else:
raw.add code[i]
inc i
flushRaw()
if inDef:
b.endTree()
b.close()
proc renderMarkedC*(code: string; live: HashSet[string]; dropped: var int): string =
## Renders the final C text from the marked module text: symbol marks are
## removed (keeping the names — a later merge step substitutes them here),
## and definitions whose name is not in `live` are dropped entirely. Each
## definition is self-delimiting (genProcAux emits an end directive right
## after the proc's text), so text written by other emitters is never part
## of a definition's span and survives unconditionally.
result = newStringOfCap(code.len)
var i = 0
while i < code.len:
case code[i]
of CnifSymStart, CnifSymEnd:
inc i
of CnifDefStart:
var payload = ""
inc i
while i < code.len and code[i] != CnifDefEnd:
payload.add code[i]
inc i
inc i # skip CnifDefEnd
if payload.len > 0:
let sep = find(payload, CnifDefSep)
let name = if sep >= 0: payload[0..<sep] else: payload
if name notin live:
inc dropped
# drop the definition's text: everything up to its end directive
while i < code.len and code[i] != CnifDefStart: inc i
else:
result.add code[i]
inc i
# ---- Liveness over the artifact -------------------------------------------
proc symOrIdentName(c: Cursor): string {.inline.} =
if c.kind == Ident: strVal(c) else: symName(c)
type
CnifHeads* = object
## The cheap-to-parse part of an artifact that a later run needs in
## order to reuse the TU without regenerating it.
valid*: bool ## file parsed, carries the meta head and has
## the current format version
initRequired*: bool
datInitRequired*: bool
semmedNif*: string ## the semmed NIF this TU was generated from
moduleBase*: string ## the module's mangled base name
cdefs*: seq[tuple[cname, nifname: string]] ## the proc definitions
cdata*: seq[tuple[cname, nifname: string]] ## the data definitions
crefs*: seq[string] ## C names referenced but not defined here
cdeps*: seq[string] ## module suffixes whose routine bodies this
## TU embeds (impl-cookie gated on reuse)
proc readCnifHeads*(f: string): CnifHeads =
## Reads `(meta ...)`, `(cdata ...)`, `(cref ...)` and the `(cdef ...)`
## head names from an artifact. Artifacts written by an older compiler
## (no meta head or a different format version) report `valid=false`.
result = CnifHeads()
if not fileExists(f): return
var pool = newPool()
var tags = newTagPool()
let stmtsTag = tags.registerTag("stmts")
let cdefTag = tags.registerTag("cdef")
let cdataTag = tags.registerTag("cdata")
let crefTag = tags.registerTag("cref")
let cdepsTag = tags.registerTag("cdeps")
let metaTag = tags.registerTag("meta")
var buf = parseFromFile(f, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != stmtsTag:
endRead(c)
return
var version = ""
var sawMeta = false
c.loopInto:
if c.kind == TagLit:
if c.cursorTagId == metaTag:
sawMeta = true
var strIdx = 0
c.loopInto:
if c.kind == Ident:
for ch in strVal(c):
if ch == 'i': result.initRequired = true
elif ch == 'd': result.datInitRequired = true
inc c
elif c.kind == StrLit:
if strIdx == 0: result.semmedNif = strVal(c)
elif strIdx == 1: result.moduleBase = strVal(c)
elif strIdx == 2: version = strVal(c)
inc strIdx
inc c
else:
skip c
elif c.cursorTagId == cdataTag:
c.loopInto:
if c.kind == SymbolDef:
result.cdata.add (symName(c), "")
inc c
elif c.kind == StrLit:
if result.cdata.len > 0:
result.cdata[^1].nifname = strVal(c)
inc c
else:
skip c
elif c.cursorTagId == crefTag:
c.loopInto:
if c.kind in {Ident, Symbol, SymbolDef}:
result.crefs.add symOrIdentName(c)
inc c
else:
skip c
elif c.cursorTagId == cdepsTag:
c.loopInto:
if c.kind in {Ident, Symbol, SymbolDef}:
result.cdeps.add symOrIdentName(c)
inc c
else:
skip c
elif c.cursorTagId == cdefTag:
# fixed head: SymbolDef, flags (Ident or empty), NIF name StrLit;
# everything after that is the definition's body text
var state = 0
c.loopInto:
if c.kind == SymbolDef:
result.cdefs.add (symName(c), "")
state = 1
inc c
elif state == 1: # the flags field
state = 2
skip c
elif state == 2: # the NIF name
if c.kind == StrLit and result.cdefs.len > 0:
result.cdefs[^1].nifname = strVal(c)
state = 3
skip c
else:
skip c
else:
skip c
else:
skip c
endRead(c)
result.valid = sawMeta and version == CnifVersion
type
CnifLiveness* = object
defs*: int ## proc definitions emitted across all modules
liveDefs*: int ## of those, reachable from the roots
live*: HashSet[string] ## live C names
broken*: bool
proc computeLiveFromCArtifacts*(files: openArray[string]): CnifLiveness =
## dce1-style mark&sweep over the C-shaped artifacts: a `(cdef ...)`
## group is a definition (flags 'x'/'c'/'m' — exportc, compilerproc,
## method/dispatcher — make it a root), names at the top level (data,
## globals, init code) are roots, names inside a group are its uses.
## Because the artifact is *fully lowered* output, no conservative
## modelling is needed: every call the C code contains is a token here.
##
## NB: mangled C names contain no dots, so NIF's text reader classifies
## them as `Ident` rather than `Symbol`; the dialect therefore treats
## Ident tokens as name uses. Inside a `(cdef ...)` the flags ident is
## the one immediately following the SymbolDef; everything after is a use.
result = CnifLiveness(live: initHashSet[string]())
var pool = newPool()
var tags = newTagPool()
let stmtsTag = tags.registerTag("stmts")
let cdefTag = tags.registerTag("cdef")
let cdataTag = tags.registerTag("cdata")
let crefTag = tags.registerTag("cref")
let cdepsTag = tags.registerTag("cdeps")
let metaTag = tags.registerTag("meta")
var uses = initTable[string, HashSet[string]]()
var roots = initHashSet[string]()
var defs = initHashSet[string]()
for f in files:
if not fileExists(f):
result.broken = true
return
var buf = parseFromFile(f, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != stmtsTag:
result.broken = true
endRead(c)
return
c.loopInto:
case c.kind
of Symbol, Ident:
roots.incl symOrIdentName(c)
inc c
of TagLit:
if c.cursorTagId == metaTag or c.cursorTagId == cdataTag or
c.cursorTagId == crefTag or c.cursorTagId == cdepsTag:
# bookkeeping for TU reuse, irrelevant for liveness
skip c
elif c.cursorTagId == cdefTag:
var owner = ""
var flagsSeen = false
c.loopInto:
case c.kind
of SymbolDef:
owner = symName(c)
defs.incl owner
flagsSeen = false
inc c
of Symbol, Ident:
let name = symOrIdentName(c)
if not flagsSeen:
# the flags field right after the SymbolDef
flagsSeen = true
for ch in name:
# 'd' marks a data definition (const/RTTI): never DCE'd, so it
# is a root whose body keeps its referenced procs live
if ch in {'x', 'c', 'm', 'd'}:
roots.incl owner
break
else:
uses.mgetOrPut(owner, initHashSet[string]()).incl name
inc c
of DotToken:
flagsSeen = true # empty flags field
inc c
else:
skip c
else:
c.loopInto:
if c.kind in {Symbol, Ident}:
roots.incl symOrIdentName(c)
inc c
else:
skip c
else:
skip c
endRead(c)
# mark & sweep
var work = newSeqOfCap[string](roots.len)
for r in roots: work.add r
while work.len > 0:
let s = work.pop()
if not result.live.containsOrIncl(s):
if uses.hasKey(s):
for dep in uses[s]:
if dep notin result.live:
work.add dep
result.defs = defs.len
for d in defs:
if d in result.live: inc result.liveDefs
# ---- The merge stage: liveness + owner assignment -------------------------
type
MergeDecision* = object
## What the per-module backend's `merge` stage computes from every
## module's `.c.nif` and what its `emit` stage consumes to render the
## final `.c` of one module.
live*: HashSet[string] ## globally reachable C names (dead cdefs
## are dropped from every module)
owners*: Table[string, string] ## for each `'u'`-flagged (unique,
## externally-linked) definition, the single
## artifact base name allowed to embed its
## body; every other module prototypes it
broken*: bool ## an artifact was missing or unparsable —
## the caller should fall back / regenerate
defs*, liveDefs*: int
proc computeMergeDecision*(files: openArray[string]): MergeDecision =
## One pass over every `.c.nif`: the same mark&sweep as
## `computeLiveFromCArtifacts` plus, per definition, owner assignment.
##
## Each `cg` process emits the body of every definition it demands
## (emit-everywhere), so the same externally-linked definition appears in
## several artifacts. A `'u'` flag on the `(cdef ...)` marks those that need
## exactly one owner, assigned here across processes: the owner is the
## lexicographically smallest artifact that emits it — a pure function of the
## claimant set, hence stable across rebuilds. Definitions without `'u'`
## (inline procs, dispatchers) are `static`/main-only and emitted into every
## using TU, so they get no owner entry and are never deduplicated.
result = MergeDecision(live: initHashSet[string](),
owners: initTable[string, string]())
var pool = newPool()
var tags = newTagPool()
let stmtsTag = tags.registerTag("stmts")
let cdefTag = tags.registerTag("cdef")
let cdataTag = tags.registerTag("cdata")
let crefTag = tags.registerTag("cref")
let cdepsTag = tags.registerTag("cdeps")
let metaTag = tags.registerTag("meta")
var uses = initTable[string, HashSet[string]]()
var roots = initHashSet[string]()
var defs = initHashSet[string]()
for f in files:
if not fileExists(f):
result.broken = true
return
let owner = extractFilename(f)
var buf = parseFromFile(f, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != stmtsTag:
result.broken = true
endRead(c)
return
c.loopInto:
case c.kind
of Symbol, Ident:
roots.incl symOrIdentName(c)
inc c
of TagLit:
if c.cursorTagId == metaTag or c.cursorTagId == cdataTag or
c.cursorTagId == crefTag or c.cursorTagId == cdepsTag:
skip c
elif c.cursorTagId == cdefTag:
var ownerName = ""
var flagsSeen = false
var needsOwner = false
c.loopInto:
case c.kind
of SymbolDef:
ownerName = symName(c)
defs.incl ownerName
flagsSeen = false
inc c
of Symbol, Ident:
let name = symOrIdentName(c)
if not flagsSeen:
flagsSeen = true
for ch in name:
if ch in {'x', 'c', 'm'}: roots.incl ownerName
# 'u' = unique proc (DCE'd), 'd' = data (never DCE'd, hence a
# root); both need a single owner across the emit-everywhere
# processes
elif ch == 'u': needsOwner = true
elif ch == 'd':
needsOwner = true
roots.incl ownerName
else:
uses.mgetOrPut(ownerName, initHashSet[string]()).incl name
inc c
of DotToken:
flagsSeen = true # empty flags field
inc c
else:
skip c
if needsOwner and ownerName.len > 0:
# smallest claimant wins; ties impossible (one entry per name)
let prev = result.owners.getOrDefault(ownerName, "")
if prev.len == 0 or owner < prev:
result.owners[ownerName] = owner
else:
c.loopInto:
if c.kind in {Symbol, Ident}:
roots.incl symOrIdentName(c)
inc c
else:
skip c
else:
skip c
endRead(c)
var work = newSeqOfCap[string](roots.len)
for r in roots: work.add r
while work.len > 0:
let s = work.pop()
if not result.live.containsOrIncl(s):
if uses.hasKey(s):
for dep in uses[s]:
if dep notin result.live:
work.add dep
result.defs = defs.len
for d in defs:
if d in result.live: inc result.liveDefs
const MergeDecisionFile* = "ic.backend.merge.nif"
## Fixed name of the merge stage's output in the nimcache, read by `emit`.
proc writeMergeDecision*(outfile: string; d: MergeDecision) =
## Serializes the merge decision: `(merge (live Symbol*) (owners (own
## Symbol StrLit)*))`. C names are mangled (no dots) so they serialize as
## symbols; owner artifact base names go in string literals.
var live: seq[string] = @[]
for n in d.live: live.add n
sort live
var keys: seq[string] = @[]
for k in d.owners.keys: keys.add k
sort keys
var b = nifbuilder.open(outfile)
b.withTree "merge":
b.withTree "live":
for n in live: b.addSymbol n, ""
b.withTree "owners":
for k in keys:
b.withTree "own":
b.addSymbol k, ""
b.addStrLit d.owners[k]
b.close()
proc readMergeDecision*(f: string): MergeDecision =
## Reads back a `writeMergeDecision` file; `broken=true` if absent/unparsable.
result = MergeDecision(live: initHashSet[string](),
owners: initTable[string, string]())
if not fileExists(f):
result.broken = true
return
var pool = newPool()
var tags = newTagPool()
let mergeTag = tags.registerTag("merge")
let liveTag = tags.registerTag("live")
let ownersTag = tags.registerTag("owners")
let ownTag = tags.registerTag("own")
var buf = parseFromFile(f, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != mergeTag:
result.broken = true
endRead(c)
return
c.loopInto:
if c.kind == TagLit and c.cursorTagId == liveTag:
c.loopInto:
if c.kind in {Symbol, Ident}:
result.live.incl symOrIdentName(c)
inc c
else:
skip c
elif c.kind == TagLit and c.cursorTagId == ownersTag:
c.loopInto:
if c.kind == TagLit and c.cursorTagId == ownTag:
var key = ""
c.loopInto:
if c.kind in {Symbol, Ident}:
key = symOrIdentName(c)
inc c
elif c.kind == StrLit:
if key.len > 0: result.owners[key] = strVal(c)
inc c
else:
skip c
else:
skip c
else:
skip c
endRead(c)
proc renderCFromArtifact*(artifact: string; d: MergeDecision; ownerId: string;
dropped: var int): string =
## The per-module backend's `emit` stage: render one module's final `.c` from
## its `.c.nif` and the merge decision. String literals are emitted verbatim,
## symbols by name; a `(cdef ...)` body is dropped when the name is dead, or
## when it is a `'u'` unique definition this module does not own. The body's
## prototype lives in the surrounding raw text (cgen emits a forward
## declaration for every *used* proc, independent of where the body lands), so
## a dropped body still leaves a valid declaration — no synthesis needed. The
## head groups (meta/cdata/cref/cdeps) carry no C text.
result = ""
if not fileExists(artifact): return
var pool = newPool()
var tags = newTagPool()
let stmtsTag = tags.registerTag("stmts")
let cdefTag = tags.registerTag("cdef")
var buf = parseFromFile(artifact, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != stmtsTag:
endRead(c)
return
c.loopInto:
case c.kind
of StrLit:
result.add strVal(c)
inc c
of Symbol, Ident:
result.add symOrIdentName(c)
inc c
of TagLit:
if c.cursorTagId == cdefTag:
# fixed head: SymbolDef, flags (Ident or empty), nifname StrLit; the
# rest is the definition's body text. `state` counts past the head.
var name = ""
var isUnique = false
var isData = false
var keep = true
var state = 0
c.loopInto:
if state == 0 and c.kind == SymbolDef:
name = symName(c)
state = 1
inc c
elif state == 1: # the flags field (one token: Ident/Symbol or empty)
if c.kind in {Ident, Symbol}:
for ch in symOrIdentName(c):
if ch == 'u': isUnique = true
elif ch == 'd': isData = true
state = 2
inc c
elif state == 2: # the NIF name (one StrLit) — decide keep here
let owned = d.owners.getOrDefault(name, ownerId) == ownerId
keep =
if isData: owned # data: kept by its owner only
elif isUnique: (name in d.live) and owned
else: name in d.live # inline/dispatcher: per-TU
if not keep: inc dropped
state = 3
inc c
else: # body tokens
if keep:
if c.kind == StrLit: result.add strVal(c)
elif c.kind in {Symbol, Ident}: result.add symOrIdentName(c)
inc c
else:
# head groups (meta/cdata/cref/cdeps) carry no C text
skip c
else:
inc c
endRead(c)

View File

@@ -24,7 +24,7 @@ bootSwitch(usedMarkAndSweep, defined(gcmarkandsweep), "--gc:markAndSweep")
bootSwitch(usedGoGC, defined(gogc), "--gc:go")
bootSwitch(usedNoGC, defined(nogc), "--gc:none")
import std/[setutils, os, strutils, parseutils, parseopt, sequtils, strtabs, enumutils]
import std/[setutils, sets, os, strutils, parseutils, parseopt, sequtils, strtabs, enumutils]
import
msgs, options, nversion, condsyms, extccomp, platform,
wordrecg, nimblecmd, lineinfos, pathutils
@@ -118,7 +118,7 @@ const
errInvalidCmdLineOption = "invalid command line option: '$1'"
errOnOrOffExpectedButXFound = "'on' or 'off' expected, but '$1' found"
errOnOffOrListExpectedButXFound = "'on', 'off' or 'list' expected, but '$1' found"
errOffHintsError = "'off', 'hint', 'error' or 'usages' expected, but '$1' found"
errOffHintsError = "'off', 'hint', 'warning', 'error' or 'usages' expected, but '$1' found"
proc invalidCmdLineOption(conf: ConfigRef; pass: TCmdLinePass, switch: string, info: TLineInfo) =
if switch == " ": localError(conf, info, errInvalidCmdLineOption % "-")
@@ -203,13 +203,14 @@ proc processSpecificNote*(arg: string, state: TSpecialWord, pass: TCmdLinePass,
else: invalidCmdLineOption(conf, pass, orig, info)
let isSomeHint = state in {wHint, wHintAsError}
let isSomeWarning = state in {wWarning, wWarningAsError}
template findNote(noteMin, noteMax, name) =
# unfortunately, hintUser and warningUser clash, otherwise implementation would simplify a bit
let x = findStr(noteMin, noteMax, id, errUnknown)
if x != errUnknown: notes = {TNoteKind(x)}
else:
if isSomeHint:
message(conf, info, hintUnknownHint, id)
if isSomeHint or isSomeWarning:
message(conf, info, warnUnknownNotes, "unknown $#: $#" % [name, id])
else:
localError(conf, info, "unknown $#: $#" % [name, id])
case id.normalize
@@ -244,11 +245,12 @@ proc processCompile(conf: ConfigRef; filename: string) =
extccomp.addExternalFileToCompile(conf, found)
const
errNoneBoehmRefcExpectedButXFound = "'arc', 'orc', 'atomicArc', 'markAndSweep', 'boehm', 'go', 'none', 'regions', or 'refc' expected, but '$1' found"
errNoneBoehmRefcExpectedButXFound = "'arc', 'orc', 'yrc', 'atomicArc', 'markAndSweep', 'boehm', 'go', 'none', 'regions', or 'refc' expected, but '$1' found"
errNoneSpeedOrSizeExpectedButXFound = "'none', 'speed' or 'size' expected, but '$1' found"
errGuiConsoleOrLibExpectedButXFound = "'gui', 'console', 'lib' or 'staticlib' expected, but '$1' found"
errInvalidExceptionSystem = "'goto', 'setjmp', 'cpp' or 'quirky' expected, but '$1' found"
errInvalidFeatureButXFound = Feature.toSeq.map(proc(val:Feature): string = "'$1'" % $val).join(", ") & " expected, but '$1' found"
errDefaultOrSsoExpectedButXFound = "'default' or 'sso' expected, but '$1' found"
template warningOptionNoop(switch: string) =
warningDeprecated(conf, info, "'$#' is deprecated, now a noop" % switch)
@@ -265,6 +267,7 @@ proc testCompileOptionArg*(conf: ConfigRef; switch, arg: string, info: TLineInfo
of "markandsweep": result = conf.selectedGC == gcMarkAndSweep
of "destructors", "arc": result = conf.selectedGC == gcArc
of "orc": result = conf.selectedGC == gcOrc
of "yrc": result = conf.selectedGC == gcYrc
of "hooks": result = conf.selectedGC == gcHooks
of "go": result = conf.selectedGC == gcGo
of "none": result = conf.selectedGC == gcNone
@@ -304,6 +307,13 @@ proc testCompileOptionArg*(conf: ConfigRef; switch, arg: string, info: TLineInfo
else:
result = false
localError(conf, info, errInvalidExceptionSystem % arg)
of "strings":
case arg.normalize
of "default": result = conf.selectedStrings == stringDefault
of "sso": result = conf.selectedStrings == stringSso
else:
result = false
localError(conf, info, errDefaultOrSsoExpectedButXFound % arg)
of "experimental":
try:
result = conf.features.contains parseEnum[Feature](arg)
@@ -363,6 +373,7 @@ proc testCompileOption*(conf: ConfigRef; switch: string, info: TLineInfo): bool
result = false
of "panics": result = contains(conf.globalOptions, optPanics)
of "jsbigint64": result = contains(conf.globalOptions, optJsBigInt64)
of "mangle": result = contains(conf.globalOptions, optItaniumMangle)
else:
result = false
invalidCmdLineOption(conf, passCmd1, switch, info)
@@ -458,7 +469,7 @@ template handleStdinOrCmdInput =
conf.outDir = getNimcacheDir(conf)
proc handleStdinInput*(conf: ConfigRef) =
conf.projectName = "stdinfile"
conf.projectName = conf.stdinFile.string
conf.projectIsStdin = true
handleStdinOrCmdInput()
@@ -472,6 +483,7 @@ proc parseCommand*(command: string): Command =
of "cpp", "compiletocpp": cmdCompileToCpp
of "objc", "compiletooc": cmdCompileToOC
of "js", "compiletojs": cmdCompileToJS
of "nif": cmdCompileToNif
of "r": cmdCrun
of "m": cmdM
of "run": cmdTcc
@@ -491,10 +503,12 @@ proc parseCommand*(command: string): Command =
of "gendepend": cmdGendepend
of "dump": cmdDump
of "parse": cmdParse
of "rod": cmdRod
of "secret": cmdInteractive
of "nop", "help": cmdNop
of "jsonscript": cmdJsonscript
of "nifc": cmdNifC # generate C from NIF files
of "ic": cmdIc # generate .build.nif for nifmake
of "icconfig": cmdIcConfig # produce the precompiled config artifact
else: cmdUnknown
proc setCmd*(conf: ConfigRef, cmd: Command) =
@@ -506,6 +520,12 @@ proc setCmd*(conf: ConfigRef, cmd: Command) =
of cmdCompileToCpp: conf.backend = backendCpp
of cmdCompileToOC: conf.backend = backendObjc
of cmdCompileToJS: conf.backend = backendJs
of cmdCompileToNif: conf.backend = backendNif
of cmdNifC:
conf.backend = backendC # NIF to C compilation
of cmdM:
# cmdM requires optCompress for proper IC handling (include files, etc.)
conf.globalOptions.incl optCompress
else: discard
proc setCommandEarly*(conf: ConfigRef, command: string) =
@@ -560,6 +580,7 @@ proc unregisterArcOrc*(conf: ConfigRef) =
undefSymbol(conf.symbols, "gcdestructors")
undefSymbol(conf.symbols, "gcarc")
undefSymbol(conf.symbols, "gcorc")
undefSymbol(conf.symbols, "gcyrc")
undefSymbol(conf.symbols, "gcatomicarc")
undefSymbol(conf.symbols, "nimSeqsV2")
undefSymbol(conf.symbols, "nimV2")
@@ -593,6 +614,10 @@ proc processMemoryManagementOption(switch, arg: string, pass: TCmdLinePass,
conf.selectedGC = gcOrc
defineSymbol(conf.symbols, "gcorc")
registerArcOrc(pass, conf)
of "yrc":
conf.selectedGC = gcYrc
defineSymbol(conf.symbols, "gcyrc")
registerArcOrc(pass, conf)
of "atomicarc":
conf.selectedGC = gcAtomicArc
defineSymbol(conf.symbols, "gcatomicarc")
@@ -629,6 +654,18 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
conf: ConfigRef) =
var key = ""
var val = ""
# Record config-file switches so the `nim ic` driver can serialise them into a
# precompiled-config artifact and have its per-module child processes replay
# them instead of re-parsing the `nim.cfg` chain (and re-running `config.nims`
# in the VM) on every invocation. Only `passPP` (config-file) switches are
# captured; command-line switches are forwarded by the build graph as usual.
# Path-search switches are skipped: their net effect already lives in the
# resolved `searchPaths` the driver forwards as `--path`, and replaying their
# raw (often relative-to-config-dir) arguments here would misresolve.
if pass == passPP and switch.normalize notin
["path", "p", "nimblepath", "lazypath", "excludepath",
"nonimblepath", "clearnimblepath", "nimcache"]:
conf.icConfigSwitches.add (switch, arg)
case switch.normalize
of "eval":
expectArg(conf, switch, arg, pass, info)
@@ -734,6 +771,17 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
processMemoryManagementOption(switch, arg, pass, info, conf)
of "mm":
processMemoryManagementOption(switch, arg, pass, info, conf)
of "strings":
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
case arg.normalize
of "default":
conf.selectedStrings = stringDefault
of "sso":
conf.selectedStrings = stringSso
defineSymbol(conf.symbols, "nimsso")
else:
localError(conf, info, errDefaultOrSsoExpectedButXFound % arg)
of "warnings", "w":
if processOnOffSwitchOrList(conf, {optWarns}, arg, pass, info): listWarnings(conf)
of "warning": processSpecificNote(arg, wWarning, pass, info, switch, conf)
@@ -759,6 +807,16 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
conf.globalOptions.excl optCDebug
else:
localError(conf, info, "expected native|gdb|on|off but found " & arg)
of "mangle":
case arg.normalize
of "nim":
conf.globalOptions.excl optItaniumMangle
of "cpp":
conf.globalOptions.incl optItaniumMangle
else:
localError(conf, info, "expected nim|cpp but found " & arg)
of "compress":
conf.globalOptions.incl optCompress
of "g": # alias for --debugger:native
conf.globalOptions.incl optCDebug
conf.options.incl optLineDir
@@ -878,14 +936,65 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
else: localError(conf, info, errOnOrOffExpectedButXFound % arg)
of "noimportdoc":
processOnOffSwitchG(conf, {optNoImportdoc}, arg, pass, info)
of "ismainmodule":
# `nim m` (IC) only: marks the single module being checked as the program's
# real entry point so that `isMainModule` and `when isMainModule:` resolve
# correctly even though every module is compiled with `sfMainModule` set.
conf.isMainModule = switchOn(arg)
of "icgroup":
# `nim m` only: register a module that belongs to the current strongly-
# connected import group, so it is compiled from source (not loaded from a
# precompiled NIF) and gets its own NIF written. `deps.nim` emits one
# `--icGroup:<path>` per member of a dependency cycle. The argument is an
# absolute .nim path produced by the dependency scanner.
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
conf.icGroup.incl(canonicalizePath(conf, AbsoluteFile arg).string)
of "icproject":
# `nim m`/`nim nifc` only: the ORIGINAL project file (see options.icProject)
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
conf.icProject = canonicalizePath(conf, AbsoluteFile arg).string
of "icpreparsedconfig":
# `nim m`/`nim nifc` only: path of the precompiled-config artifact (see
# options.icPreparsedConfig). Read in `passCmd1`, before `loadConfigs`, so
# config loading can replay it instead of re-parsing the `nim.cfg` chain.
expectArg(conf, switch, arg, pass, info)
conf.icPreparsedConfig = arg
of "icconfigout":
# `nim icconfig` only: where to write the precompiled config artifact (see
# options.icConfigOut). The `nim ic` driver spawns the producer with this.
expectArg(conf, switch, arg, pass, info)
conf.icConfigOut = arg
of "icbackendstage":
# `nim nifc` only: per-module backend stage, one of cg|merge|emit (see
# options.icBackendStage). Empty (switch unused) keeps the whole-program
# backend. Emitted by `deps.nim`'s backend build file.
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
conf.icBackendStage = arg
of "icbackendmodule":
# `nim nifc` only: the NIF module suffix the cg/emit stage operates on (see
# options.icBackendModule).
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
conf.icBackendModule = arg
of "import":
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
conf.implicitImports.add findModule(conf, arg, toFullPath(conf, info)).string
let m = findModule(conf, arg, toFullPath(conf, info)).string
if m.len == 0:
localError(conf, info, "Cannot resolve filename: " & arg)
else:
conf.implicitImports.add(if arg.startsWith(stdPrefix): arg else: m)
of "include":
expectArg(conf, switch, arg, pass, info)
if pass in {passCmd2, passPP}:
conf.implicitIncludes.add findModule(conf, arg, toFullPath(conf, info)).string
let m = findModule(conf, arg, toFullPath(conf, info)).string
if m.len == 0:
localError(conf, info, "Cannot resolve filename: " & arg)
else:
conf.implicitIncludes.add m
of "listcmd":
processOnOffSwitchG(conf, {optListCmd}, arg, pass, info)
of "asm":
@@ -917,8 +1026,14 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
expectArg(conf, switch, arg, pass, info)
var value: int = 10_000_000
discard parseSaturatedNatural(arg, value)
if not value > 0: localError(conf, info, "maxLoopIterationsVM must be a positive integer greater than zero")
if value <= 0: localError(conf, info, "maxLoopIterationsVM must be a positive integer greater than zero")
conf.maxLoopIterationsVM = value
of "maxcalldepthvm":
expectArg(conf, switch, arg, pass, info)
var value: int = 2_000
discard parseSaturatedNatural(arg, value)
if value <= 0: localError(conf, info, "maxCallDepthVM must be a positive integer greater than zero")
conf.maxCallDepthVM = value
of "errormax":
expectArg(conf, switch, arg, pass, info)
# Note: `nim check` (etc) can overwrite this.
@@ -928,6 +1043,10 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
var value: int = 0
discard parseSaturatedNatural(arg, value)
conf.errorMax = if value == 0: high(int) else: value
of "stdinfile":
expectArg(conf, switch, arg, pass, info)
conf.stdinFile = if os.isAbsolute(arg): AbsoluteFile(arg)
else: AbsoluteFile(getCurrentDir() / arg)
of "verbosity":
expectArg(conf, switch, arg, pass, info)
let verbosity = parseInt(arg)
@@ -961,7 +1080,8 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
# xxx maybe also ic, since not in help?
if pass in {passCmd2, passPP}:
case arg.normalize
of "on": conf.symbolFiles = v2Sf
of "on": conf.ic = true
of "legacy": conf.symbolFiles = v2Sf
of "off": conf.symbolFiles = disabledSf
of "writeonly": conf.symbolFiles = writeOnlySf
of "readonly": conf.symbolFiles = readOnlySf
@@ -1069,6 +1189,9 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
of "shownonexports":
expectNoArg(conf, switch, arg, pass, info)
showNonExportedFields(conf)
of "raw":
expectNoArg(conf, switch, arg, pass, info)
docRawOutput(conf)
of "exceptions":
case arg.normalize
of "cpp": conf.exc = excCpp
@@ -1100,9 +1223,10 @@ proc processSwitch*(switch, arg: string, pass: TCmdLinePass, info: TLineInfo;
defineSymbol(conf.symbols, "nimSeqsV2")
of "stylecheck":
case arg.normalize
of "off": conf.globalOptions = conf.globalOptions - {optStyleHint, optStyleError}
of "hint": conf.globalOptions = conf.globalOptions + {optStyleHint} - {optStyleError}
of "error": conf.globalOptions = conf.globalOptions + {optStyleError}
of "off": conf.globalOptions = conf.globalOptions - {optStyleHint, optStyleError, optStyleWarning}
of "hint": conf.globalOptions = conf.globalOptions + {optStyleHint} - {optStyleError, optStyleWarning}
of "warning": conf.globalOptions = conf.globalOptions + {optStyleWarning} - {optStyleHint, optStyleError}
of "error": conf.globalOptions = conf.globalOptions + {optStyleError} - {optStyleHint, optStyleWarning}
of "usages": conf.globalOptions.incl optStyleUsages
else: localError(conf, info, errOffHintsError % arg)
of "showallmismatches":

View File

@@ -13,13 +13,13 @@
import ast, semdata, lookups, lineinfos, idents, msgs, renderer, types, layeredtable
import std/intsets
import std/sets
when defined(nimPreviewSlimSystem):
import std/assertions
const
logBindings = false
logBindings = when defined(debugConcepts): true else: false
## Code dealing with Concept declarations
## --------------------------------------
@@ -29,7 +29,7 @@ proc declareSelf(c: PContext; info: TLineInfo) =
let ow = getCurrOwner(c)
let s = newSym(skType, getIdent(c.cache, "Self"), c.idgen, ow, info)
s.typ = newType(tyTypeDesc, c.idgen, ow)
s.typ.flags.incl {tfUnresolved, tfPacked}
s.typ.incl {tfUnresolved, tfPacked}
s.typ.add newType(tyEmpty, c.idgen, ow)
addDecl(c, s, info)
@@ -70,92 +70,281 @@ proc semConceptDeclaration*(c: PContext; n: PNode): PNode =
## ----------------
type
MatchFlags* = enum
mfDontBind # Do not bind generic parameters
mfCheckGeneric # formal <- formal comparison as opposed to formal <- operand
ConceptTypePair = tuple[conceptId, typeId: ItemId]
## Pair of (concept type id, implementation type id) used for cycle detection
MatchCon = object ## Context we pass around during concept matching.
inferred: seq[(PType, PType)] ## we need a seq here so that we can easily undo inferences \
## that turned out to be wrong.
marker: IntSet ## Some protection against wild runaway recursions.
bindings: LayeredIdTable
marker: HashSet[ConceptTypePair] ## Tracks (concept, type) pairs being checked to detect cycles.
potentialImplementation: PType ## the concrete type that might match the concept we try to match.
magic: TMagic ## mArrGet and mArrPut is wrong in system.nim and
## cannot be fixed that easily.
## Thus we special case it here.
concpt: PType ## current concept being evaluated
flags: set[MatchFlags]
MatchKind = enum
mkNoMatch, mkSubset, mkSame
const
asymmetricConceptParamMods = {tyVar, tySink, tyLent, tyOwned, tyAlias, tyInferred} # param modifiers that to not have to match implementation -> concept
bindableTypes = {tyGenericParam, tyOr, tyTypeDesc}
proc conceptMatchNode(c: PContext; n: PNode; m: var MatchCon): bool
proc matchType(c: PContext; fo, ao: PType; m: var MatchCon): bool
proc matchReturnType(c: PContext; f, a: PType; m: var MatchCon): bool
proc processConcept(c: PContext; concpt, invocation: PType, bindings: var LayeredIdTable; m: var MatchCon): bool
proc existingBinding(m: MatchCon; key: PType): PType =
## checks if we bound the type variable 'key' already to some
## concrete type.
for i in 0..<m.inferred.len:
if m.inferred[i][0] == key: return m.inferred[i][1]
return nil
result = m.bindings.lookup(key)
if result == nil:
result = key
proc conceptMatchNode(c: PContext; n: PNode; m: var MatchCon): bool
const
ignorableForArgType = {tyVar, tySink, tyLent, tyOwned, tyAlias, tyInferred}
proc matchType(c: PContext; f, a: PType; m: var MatchCon): bool
proc unrollGenericParam(param: PType): PType =
result = param.skipTypes(ignorableForArgType)
while result.kind in {tyGenericParam, tyTypeDesc} and result.hasElementType and result.elementType.kind != tyNone:
result = result.elementType
proc bindParam(c: PContext, m: var MatchCon; key, v: PType): bool {. discardable .} =
if v.kind == tyTypeDesc:
return false
var value = unrollGenericParam(v)
if value.kind == tyGenericParam:
value = existingBinding(m, value)
if value.kind == tyGenericParam:
if value.hasElementType:
value = value.elementType
else:
return true
if value.kind == tyStatic:
return false
if m.magic in {mArrPut, mArrGet} and value.kind in arrPutGetMagicApplies:
value = value.last
let old = existingBinding(m, key)
if old != key:
# check previously bound value
if not matchType(c, old, value, m):
return false
elif key.hasElementType and not key.elementType.isNil and key.elementType.kind != tyNone:
# check constaint
if matchType(c, unrollGenericParam(key), value, m) == false:
return false
when logBindings: echo "bind table adding '", key, "', ", value
assert value != nil
assert value.kind != tyVoid
m.bindings.put(key, value)
return true
proc defSignatureType(n: PNode): PType = n[0].sym.typ
proc conceptBody*(n: PType): PNode = n.n.lastSon
proc acceptsAllTypes(t: PType): bool=
result = false
if t.kind == tyAnything:
result = true
elif t.kind == tyGenericParam:
if tfImplicitTypeParam in t.flags:
result = true
if not t.hasElementType or t.elementType.kind == tyNone:
result = true
proc procDefSignature(s: PSym): PNode {. deprecated .} =
var nc = s.ast.copyNode()
for i in 0 .. 5:
nc.add s.ast[i]
nc
proc matchKids(c: PContext; f, a: PType; m: var MatchCon, start=0): bool=
result = true
for i in start..<f.kidsLen - ord(f.kind == tyGenericInst):
for i in start ..< f.kidsLen - ord(f.kind in {tyGenericInst, tyGenericInvocation}):
if not matchType(c, f[i], a[i], m): return false
proc matchType(c: PContext; f, a: PType; m: var MatchCon): bool =
iterator traverseTyOr(t: PType): PType {. closure .}=
for i in t.kids:
case i.kind:
of tyGenericParam:
if i.hasElementType:
for s in traverseTyOr(i.elementType):
yield s
else:
yield i
else:
yield i
proc matchConceptToImpl(c: PContext, f, potentialImpl: PType; m: var MatchCon): bool =
assert not(potentialImpl.reduceToBase.kind == tyConcept)
let concpt = f.reduceToBase
# Handle self-referential concepts: when a concept references itself in its body
# (e.g., `A = concept; proc test(x: Self, y: A)`), the inner type A has n=nil.
# We detect this by checking if the concept has the same symbol name as the
# one we're currently matching and has no body (n=nil).
if concpt.n.isNil:
if concpt.sym != nil and m.concpt.sym != nil and
concpt.sym == m.concpt.sym:
# Self-reference: check if potentialImpl matches what we're already checking
return potentialImpl.id == m.potentialImplementation.id
# Concept without body that's not a self-reference - cannot match
return false
# Cycle detection: track (concept, type) pairs to prevent infinite recursion.
# Returns true on cycle (coinductive semantics) to support co-dependent concepts.
let pair: ConceptTypePair = (concpt.itemId, potentialImpl.itemId)
if pair in m.marker:
return true
m.marker.incl pair
var efPot = potentialImpl
if potentialImpl.isSelf:
if m.concpt.n == concpt.n:
m.marker.excl pair
return true
efPot = m.potentialImplementation
var oldBindings = m.bindings
m.bindings = newTypeMapLayer(m.bindings)
let oldPotentialImplementation = m.potentialImplementation
m.potentialImplementation = efPot
let oldConcept = m.concpt
m.concpt = concpt
var invocation: PType = nil
if f.kind in {tyGenericInvocation, tyGenericInst}:
invocation = f
result = processConcept(c, concpt, invocation, oldBindings, m)
m.potentialImplementation = oldPotentialImplementation
m.concpt = oldConcept
m.bindings = oldBindings
m.marker.excl pair
proc cmpConceptDefs(c: PContext, fn, an: PNode, m: var MatchCon): bool=
if fn.kind != an.kind:
return false
if fn[namePos].sym.name != an[namePos].sym.name:
return false
let
ft = fn.defSignatureType
at = an.defSignatureType
if ft.len != at.len:
return false
for i in 1 ..< ft.n.len:
m.bindings = m.bindings.newTypeMapLayer()
let aType = at.n[i].typ
let fType = ft.n[i].typ
if aType.isSelf and fType.isSelf:
continue
if not matchType(c, fType, aType, m):
m.bindings.setToPreviousLayer()
return false
result = true
if not matchReturnType(c, ft.returnType, at.returnType, m):
m.bindings.setToPreviousLayer()
result = false
proc conceptsMatch(c: PContext, fc, ac: PType; m: var MatchCon): MatchKind =
# XXX: In the future this may need extra parameters to carry info for container types
if fc.n == ac.n:
# This will have to take generic parameters into account at some point
return mkSame
let
fn = fc.conceptBody
an = ac.conceptBody
sameLen = fc.len == ac.len
var match = false
for fdef in fn:
var cmpResult = false
for ia, ndef in an:
match = cmpConceptDefs(c, fdef, ndef, m)
if match:
break
if not match:
return mkNoMatch
return mkSubset
proc isObjectSubtype(f, a: PType): bool =
var t = a
result = false
while t != nil:
t = t.baseClass
if t == nil:
break
t = t.skipTypes({tyPtr,tyRef})
if t == nil:
break
if t.kind != tyObject:
break
if sameObjectTypes(f, t):
result = true
break
proc matchType(c: PContext; fo, ao: PType; m: var MatchCon): bool =
## The heart of the concept matching process. 'f' is the formal parameter of some
## routine inside the concept that we're looking for. 'a' is the formal parameter
## of a routine that might match.
const
ignorableForArgType = {tyVar, tySink, tyLent, tyOwned, tyGenericInst, tyAlias, tyInferred}
var
a = ao
f = fo
if a.isSelf:
if m.magic in {mArrPut, mArrGet}:
return false
a = m.potentialImplementation
if a.kind in bindableTypes:
a = existingBinding(m, ao)
if a == ao and a.kind == tyGenericParam and a.hasElementType and a.elementType.kind != tyNone:
a = a.elementType
if f.isConcept:
if a.acceptsAllTypes:
return false
if a.skipTypes(ignorableForArgType).isConcept:
# if f is a subset of a then any match to a will also match f. Not the other way around
return conceptsMatch(c, a.reduceToBase, f.reduceToBase, m) >= mkSubset
else:
return matchConceptToImpl(c, f, a, m)
result = false
case f.kind
of tyAlias:
result = matchType(c, f.skipModifier, a, m)
of tyTypeDesc:
if isSelf(f):
#let oldLen = m.inferred.len
result = matchType(c, a, m.potentialImplementation, m)
#echo "self is? ", result, " ", a.kind, " ", a, " ", m.potentialImplementation, " ", m.potentialImplementation.kind
#m.inferred.setLen oldLen
#echo "A for ", result, " to ", typeToString(a), " to ", typeToString(m.potentialImplementation)
let ua = a.skipTypes(asymmetricConceptParamMods)
if m.magic in {mArrPut, mArrGet}:
if m.potentialImplementation.reduceToBase.kind in arrPutGetMagicApplies:
bindParam(c, m, a, last m.potentialImplementation)
result = true
#elif ua.isConcept:
# result = matchType(c, m.concpt, ua, m)
else:
result = matchType(c, a.skipTypes(ignorableForArgType), m.potentialImplementation, m)
else:
if a.kind == tyTypeDesc and f.hasElementType == a.hasElementType:
if f.hasElementType:
if a.kind == tyTypeDesc:
if not(a.hasElementType) or a.elementType.kind == tyNone:
result = true
elif f.hasElementType:
result = matchType(c, f.elementType, a.elementType, m)
else:
result = true # both lack it
else:
result = false
of tyGenericInvocation:
result = false
if a.kind == tyGenericInst and a.genericHead.kind == tyGenericBody:
if sameType(f.genericHead, a.genericHead) and f.kidsLen == a.kidsLen-1:
result = matchKids(c, f, a, m, start=FirstGenericParamAt)
of tyGenericParam:
let ak = a.skipTypes({tyVar, tySink, tyLent, tyOwned})
if ak.kind in {tyTypeDesc, tyStatic} and not isSelf(ak):
result = false
else:
let old = existingBinding(m, f)
if old == nil:
if f.hasElementType and f.elementType.kind != tyNone:
# also check the generic's constraints:
let oldLen = m.inferred.len
result = matchType(c, f.elementType, a, m)
m.inferred.setLen oldLen
if result:
when logBindings: echo "A adding ", f, " ", ak
m.inferred.add((f, ak))
elif m.magic == mArrGet and ak.kind in {tyArray, tyOpenArray, tySequence, tyVarargs, tyCstring, tyString}:
when logBindings: echo "B adding ", f, " ", lastSon ak
m.inferred.add((f, last ak))
result = true
else:
when logBindings: echo "C adding ", f, " ", ak
m.inferred.add((f, ak))
#echo "binding ", typeToString(ak), " to ", typeToString(f)
result = true
elif not m.marker.containsOrIncl(old.id):
result = matchType(c, old, ak, m)
if m.magic == mArrPut and ak.kind == tyGenericParam:
result = true
else:
result = false
#echo "B for ", result, " to ", typeToString(a), " to ", typeToString(m.potentialImplementation)
of tyVar, tySink, tyLent, tyOwned:
# modifiers in the concept must be there in the actual implementation
# too but not vice versa.
@@ -163,75 +352,136 @@ proc matchType(c: PContext; f, a: PType; m: var MatchCon): bool =
result = matchType(c, f.elementType, a.elementType, m)
elif m.magic == mArrPut:
result = matchType(c, f.elementType, a, m)
else:
result = false
of tyEnum, tyObject, tyDistinct:
result = sameType(f, a)
if a.kind in ignorableForArgType:
result = matchType(c, f, a.skipTypes(ignorableForArgType), m)
else:
if a.kind == tyGenericInst:
# tyOr does this to generic typeclasses
result = a.base.sym == f.sym
else:
result = sameType(f, a)
if not result and f.kind == tyObject and a.kind == tyObject:
result = isObjectSubtype(f, a)
of tyEmpty, tyString, tyCstring, tyPointer, tyNil, tyUntyped, tyTyped, tyVoid:
result = a.skipTypes(ignorableForArgType).kind == f.kind
of tyBool, tyChar, tyInt..tyUInt64:
let ak = a.skipTypes(ignorableForArgType)
result = ak.kind == f.kind or ak.kind == tyOrdinal or
(ak.kind == tyGenericParam and ak.hasElementType and ak.elementType.kind == tyOrdinal)
of tyConcept:
let oldLen = m.inferred.len
let oldPotentialImplementation = m.potentialImplementation
m.potentialImplementation = a
result = conceptMatchNode(c, f.n.lastSon, m)
m.potentialImplementation = oldPotentialImplementation
if not result:
m.inferred.setLen oldLen
(ak.kind == tyGenericParam and ak.hasElementType and ak.elementType.kind == tyOrdinal)
of tyArray, tyTuple, tyVarargs, tyOpenArray, tyRange, tySequence, tyRef, tyPtr:
if f.kind == tyArray and f.kidsLen == 3 and a.kind == tyArray:
# XXX: this is a work-around!
# system.nim creates these for the magic array typeclass
result = true
else:
let ak = a.skipTypes(ignorableForArgType - {f.kind})
if ak.kind == f.kind:
if f.base.kind == tyNone:
result = true
elif f.kidsLen == ak.kidsLen:
result = matchKids(c, f, ak, m)
of tyGenericInvocation, tyGenericInst:
result = false
let ea = a.skipTypes(ignorableForArgType)
if ea.kind in {tyGenericInst, tyGenericInvocation}:
var
k1 = f.kidsLen - ord(f.kind == tyGenericInst)
k2 = ea.kidsLen - ord(ea.kind == tyGenericInst)
if sameType(f.genericHead, ea.genericHead) and k1 == k2:
result = true
for i in 1 ..< k2:
if not matchType(c, f[i], ea[i], m):
result = false
break
elif f.kind == tyGenericInvocation:
# bind potential generic constraints into body
let body = f.base
for i in 1 ..< len(f):
bindParam(c,m,body[i-1], f[i])
result = matchType(c, body, a, m)
else: # tyGenericInst
result = matchType(c, f.last, a, m)
of tyOrdinal:
result = isOrdinalType(a, allowEnumWithHoles = false) or a.kind == tyGenericParam
of tyStatic:
var scomp = f.base
if scomp.kind == tyGenericParam:
if f.base.kidsLen > 0:
scomp = scomp.base
if a.kind == tyStatic:
result = matchType(c, scomp, a.base, m)
else:
result = matchType(c, scomp, a, m)
of tyGenericParam:
if a.acceptsAllTypes:
discard bindParam(c, m, f, a)
result = f.acceptsAllTypes
else:
result = bindParam(c, m, f, a)
of tyAnything:
result = true
of tyNot:
if a.kind == tyNot:
result = matchType(c, f.elementType, a.elementType, m)
else:
m.bindings = m.bindings.newTypeMapLayer()
result = not matchType(c, f.elementType, a, m)
m.bindings.setToPreviousLayer()
of tyAnd:
m.bindings = m.bindings.newTypeMapLayer()
result = true
for ff in traverseTyOr(f):
let r = matchType(c, ff, a, m)
if not r:
m.bindings.setToPreviousLayer()
result = false
break
of tyGenericBody:
var ak = a
if a.kind == tyGenericBody:
ak = last(a)
result = matchType(c, last(f), ak, m)
of tyCompositeTypeClass:
result = matchType(c, last(f), a, m)
of tyArray, tyTuple, tyVarargs, tyOpenArray, tyRange, tySequence, tyRef, tyPtr,
tyGenericInst:
# ^ XXX Rewrite this logic, it's more complex than it needs to be.
result = false
let ak = a.skipTypes(ignorableForArgType - {f.kind})
if ak.kind == f.kind and f.kidsLen == ak.kidsLen:
result = matchKids(c, f, ak, m)
if a.kind == tyCompositeTypeClass:
result = matchKids(c, f, a, m)
else:
result = matchType(c, last(f), a, m)
of tyBuiltInTypeClass:
let target = f.genericHead.kind
result = a.skipTypes(ignorableForArgType).reduceToBase.kind == target
of tyOr:
let oldLen = m.inferred.len
if a.kind == tyOr:
# say the concept requires 'int|float|string' if the potentialImplementation
# says 'int|string' that is good enough.
var covered = 0
for ff in f.kids:
for aa in a.kids:
let oldLenB = m.inferred.len
for ff in traverseTyOr(f):
for aa in traverseTyOr(a):
m.bindings = m.bindings.newTypeMapLayer()
let r = matchType(c, ff, aa, m)
if r:
inc covered
break
m.inferred.setLen oldLenB
m.bindings.setToPreviousLayer()
result = covered >= a.kidsLen
if not result:
m.inferred.setLen oldLen
else:
result = false
for ff in f.kids:
m.bindings = m.bindings.newTypeMapLayer()
result = matchType(c, ff, a, m)
if result: break # and remember the binding!
m.inferred.setLen oldLen
of tyNot:
if a.kind == tyNot:
m.bindings.setToPreviousLayer()
of tySet:
result = false
if a.kind == tySet:
result = matchType(c, f.elementType, a.elementType, m)
else:
let oldLen = m.inferred.len
result = not matchType(c, f.elementType, a, m)
m.inferred.setLen oldLen
of tyAnything:
result = true
of tyOrdinal:
result = isOrdinalType(a, allowEnumWithHoles = false) or a.kind == tyGenericParam
else:
result = false
if result and ao.kind == tyGenericParam:
let bf = if f.isSelf: m.potentialImplementation else: f
if bindParam(c, m, ao, bf):
when logBindings: echo " ^ reverse binding"
proc checkConstraint(c: PContext; f, a: PType; m: var MatchCon): bool =
result = matchType(c, f, a, m) or matchType(c, a, f, m)
proc matchReturnType(c: PContext; f, a: PType; m: var MatchCon): bool =
## Like 'matchType' but with extra logic dealing with proc return types
@@ -241,30 +491,38 @@ proc matchReturnType(c: PContext; f, a: PType; m: var MatchCon): bool =
elif a == nil:
result = false
else:
result = matchType(c, f, a, m)
result = checkConstraint(c, f, a, m)
proc matchSym(c: PContext; candidate: PSym, n: PNode; m: var MatchCon): bool =
## Checks if 'candidate' matches 'n' from the concept body. 'n' is a nkProcDef
## or similar.
# watch out: only add bindings after a completely successful match.
let oldLen = m.inferred.len
m.bindings = m.bindings.newTypeMapLayer()
let can = candidate.typ.n
let con = n[0].sym.typ.n
let con = defSignatureType(n).n
if can.len < con.len:
# too few arguments, cannot be a match:
return false
if can.len > con.len:
# too many arguments (not optional)
for i in con.len ..< can.len:
if can[i].sym.ast == nil:
return false
when defined(debugConcepts):
echo "considering: ", renderTree(candidate.procDefSignature), " ", candidate.magic
let common = min(can.len, con.len)
for i in 1 ..< common:
if not matchType(c, con[i].typ, can[i].typ, m):
m.inferred.setLen oldLen
if not checkConstraint(c, con[i].typ, can[i].typ, m):
m.bindings.setToPreviousLayer()
return false
if not matchReturnType(c, n[0].sym.typ.returnType, candidate.typ.returnType, m):
m.inferred.setLen oldLen
if not matchReturnType(c, n.defSignatureType.returnType, candidate.typ.returnType, m):
m.bindings.setToPreviousLayer()
return false
# all other parameters have to be optional parameters:
@@ -272,7 +530,7 @@ proc matchSym(c: PContext; candidate: PSym, n: PNode; m: var MatchCon): bool =
assert can[i].kind == nkSym
if can[i].sym.ast == nil:
# has too many arguments one of which is not optional:
m.inferred.setLen oldLen
m.bindings.setToPreviousLayer()
return false
return true
@@ -280,12 +538,14 @@ proc matchSym(c: PContext; candidate: PSym, n: PNode; m: var MatchCon): bool =
proc matchSyms(c: PContext, n: PNode; kinds: set[TSymKind]; m: var MatchCon): bool =
## Walk the current scope, extract candidates which the same name as 'n[namePos]',
## 'n' is the nkProcDef or similar from the concept that we try to match.
let candidates = searchInScopesAllCandidatesFilterBy(c, n[namePos].sym.name, kinds)
for candidate in candidates:
#echo "considering ", typeToString(candidate.typ), " ", candidate.magic
m.magic = candidate.magic
if matchSym(c, candidate, n, m): return true
result = false
var candidates = searchScopes(c, n[namePos].sym.name, kinds)
searchImportsAll(c, n[namePos].sym.name, kinds, candidates)
for candidate in candidates:
m.magic = candidate.magic
if matchSym(c, candidate, n, m):
result = true
break
proc conceptMatchNode(c: PContext; n: PNode; m: var MatchCon): bool =
## Traverse the concept's AST ('n') and see if every declaration inside 'n'
@@ -318,7 +578,48 @@ proc conceptMatchNode(c: PContext; n: PNode; m: var MatchCon): bool =
# error was reported earlier.
result = false
proc conceptMatch*(c: PContext; concpt, arg: PType; bindings: var LayeredIdTable; invocation: PType): bool =
proc fixBindings(bindings: var LayeredIdTable; concpt: PType; invocation: PType; m: var MatchCon) =
# invocation != nil means we have a non-atomic concept:
if invocation != nil and invocation.kind == tyGenericInvocation:
assert concpt.sym.typ.kind == tyGenericBody
for i in 0 .. concpt.sym.typ.len - 1:
let thisSym = concpt.sym.typ[i]
if lookup(bindings, thisSym) != nil:
# dont trust the bindings over existing ones
continue
let found = m.bindings.lookup(thisSym)
if found != nil:
when logBindings: echo "Invocation bind: ", thisSym, " ", found
bindings.put(thisSym, found)
# bind even more generic parameters
let genBody = invocation.base
assert genBody.kind == tyGenericBody
for i in FirstGenericParamAt ..< invocation.kidsLen:
let bpram = genBody[i - 1]
if lookup(bindings, invocation[i]) != nil:
# dont trust the bindings over existing ones
continue
let boundV = lookup(bindings, bpram)
when logBindings: echo "generic body bind: '", invocation[i], "' '", boundV, "'"
if boundV != nil:
bindings.put(invocation[i], boundV)
bindings.put(concpt, m.potentialImplementation)
proc processConcept(c: PContext; concpt, invocation: PType, bindings: var LayeredIdTable; m: var MatchCon): bool =
m.bindings = m.bindings.newTypeMapLayer()
if invocation != nil and invocation.kind == tyGenericInst:
let genericBody = invocation.base
for i in 1..<invocation.kidsLen-1:
# instGenericContainer can bind `tyVoid`
if invocation[i].kind != tyVoid:
bindParam(c, m, genericBody[i-1], invocation[i])
result = conceptMatchNode(c, concpt.conceptBody, m)
if result and mfDontBind notin m.flags:
fixBindings(bindings, concpt, invocation, m)
proc conceptMatch*(c: PContext; concpt, arg: PType; bindings: var LayeredIdTable; invocation: PType, flags: set[MatchFlags] = {}): bool =
## Entry point from sigmatch. 'concpt' is the concept we try to match (here still a PType but
## we extract its AST via 'concpt.n.lastSon'). 'arg' is the type that might fulfill the
## concept's requirements. If so, we return true and fill the 'bindings' with pairs of
@@ -327,26 +628,16 @@ proc conceptMatch*(c: PContext; concpt, arg: PType; bindings: var LayeredIdTable
## `C[S, T]` parent type that we look for. We need this because we need to store bindings
## for 'S' and 'T' inside 'bindings' on a successful match. It is very important that
## we do not add any bindings at all on an unsuccessful match!
var m = MatchCon(inferred: @[], potentialImplementation: arg)
result = conceptMatchNode(c, concpt.n.lastSon, m)
if result:
for (a, b) in m.inferred:
if b.kind == tyGenericParam:
var dest = b
while true:
dest = existingBinding(m, dest)
if dest == nil or dest.kind != tyGenericParam: break
if dest != nil:
bindings.put(a, dest)
when logBindings: echo "A bind ", a, " ", dest
else:
bindings.put(a, b)
when logBindings: echo "B bind ", a, " ", b
# we have a match, so bind 'arg' itself to 'concpt':
bindings.put(concpt, arg)
# invocation != nil means we have a non-atomic concept:
if invocation != nil and arg.kind == tyGenericInst and invocation.kidsLen == arg.kidsLen-1:
# bind even more generic parameters
assert invocation.kind == tyGenericInvocation
for i in FirstGenericParamAt ..< invocation.kidsLen:
bindings.put(invocation[i], arg[i])
var m = MatchCon(bindings: bindings, potentialImplementation: arg, concpt: concpt, flags: flags, marker: initHashSet[ConceptTypePair]())
if arg.isConcept:
result = conceptsMatch(c, concpt.reduceToBase, arg.reduceToBase, m) >= mkSubset
elif arg.acceptsAllTypes:
# XXX: I think this is wrong, or at least partially wrong. Can still test ambiguous types
result = false
elif mfCheckGeneric in m.flags:
# prioritize concepts the least. Specifically if the arg is not a catch all as per above
result = true
else:
result = processConcept(c, concpt, invocation, bindings, m)

View File

@@ -172,4 +172,8 @@ proc initDefines*(symbols: StringTableRef) =
defineSymbol("nimHasDefaultFloatRoundtrip")
defineSymbol("nimHasXorSet")
defineSymbol("nimHasLegacyNoStrictDefs")
defineSymbol("nimHasSetLengthSeqUninitMagic")
defineSymbol("nimHasPreviewDuplicateModuleError")
defineSymbol("nimHasImplicitRangeConversion")

1221
compiler/deps.nim Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -439,8 +439,8 @@ proc gen(c: var Con; n: PNode) =
genUse(c, n)
of nkIfStmt, nkIfExpr: genIf(c, n)
of nkWhenStmt:
# This is "when nimvm" node. Chose the first branch.
gen(c, n[0][1])
# This is "when nimvm" node. Chose the second branch.
gen(c, n[1][0])
of nkCaseStmt: genCase(c, n)
of nkWhileStmt: genWhile(c, n)
of nkBlockExpr, nkBlockStmt: genBlock(c, n)
@@ -483,7 +483,7 @@ proc constructCfg*(s: PSym; body: PNode; root: PSym): ControlFlowGraph =
gen(c, body)
if root.kind == skResult:
genImplicitReturn(c)
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc):
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc) or defined(gcYrc):
result = c.code # will move
else:
shallowCopy(result, c.code)

View File

@@ -19,7 +19,7 @@ import
wordrecg, syntaxes, renderer, lexer,
packages/docutils/[rst, rstidx, rstgen, dochelpers],
trees, types,
typesrenderer, astalgo, lineinfos,
typesrenderer, lineinfos,
pathutils, nimpaths, renderverbatim, packages
import packages/docutils/rstast except FileIndex, TLineInfo
@@ -148,7 +148,7 @@ proc cmpDecimalsIgnoreCase(a, b: string): int =
limitB = iB
while limitA < aLen and isDigit(a[limitA]): inc limitA
while limitB < bLen and isDigit(b[limitB]): inc limitB
var pos = max(limitA-iA, limitB-iA)
var pos = max(limitA-iA, limitB-iB)
while pos > 0:
if limitA-pos < iA: # digit in `a` is 0 effectively
result = ord('0') - ord(b[limitB-pos])
@@ -433,6 +433,9 @@ proc getVarIdx(varnames: openArray[string], id: string): int =
proc genComment(d: PDoc, n: PNode): PRstNode =
if n.comment.len > 0:
if optDocRaw in d.conf.globalOptions:
return newRstLeaf(n.comment)
d.sharedState.currFileIdx = addRstFileIndex(d, n.info)
try:
result = parseRst(n.comment,
@@ -537,10 +540,11 @@ proc nodeToHighlightedHtml(d: PDoc; n: PNode; result: var string;
elif s != nil and s.kind in {skType, skVar, skLet, skConst} and
sfExported in s.flags and s.owner != nil and
belongsToProjectPackage(d.conf, s.owner) and d.target == outHtml:
let external = externalDep(d, s.owner)
result.addf "<a href=\"$1#$2\"><span class=\"Identifier\">$3</span></a>",
[changeFileExt(external, "html"), literal,
escLit]
let href = (if d.module == s.owner: ""
else: externalDep(d, s.owner).changeFileExt("html")
) & "#" & literal
result.addf "<a href=\"$1\"><span class=\"Identifier\">$2</span></a>",
[href, escLit]
else:
dispA(d.conf, result, "<span class=\"Identifier\">$1</span>",
"\\spanIdentifier{$1}", [escLit])
@@ -1176,8 +1180,12 @@ proc genJsonItem(d: PDoc, n, nameNode: PNode, k: TSymKind, nonExports = false):
"col": %n.info.col}
)
if comm != nil:
result.rst = comm
result.rstField = "description"
if optDocRaw in d.conf.globalOptions:
result.json["description"] = %comm.text
else:
result.rst = comm
result.rstField = "description"
if r.buf.len > 0:
result.json["code"] = %r.buf
if k in routineKinds:
@@ -1320,7 +1328,7 @@ proc documentEffect(cache: IdentCache; n, x: PNode, effectType: TSpecialWord, id
if t.startsWith("ref "): t = substr(t, 4)
effects[i] = newIdentNode(getIdent(cache, t), n.info)
# set the type so that the following analysis doesn't screw up:
effects[i].typ() = real[i].typ
effects[i].typ = real[i].typ
result = newTreeI(nkExprColonExpr, n.info,
newIdentNode(getIdent(cache, $effectType), n.info), effects)
@@ -1406,16 +1414,19 @@ proc generateDoc*(d: PDoc, n, orig: PNode, config: ConfigRef, docFlags: DocFlags
for it in n: traceDeps(d, it)
of nkExportStmt:
for it in n:
# bug #23051; don't generate documentation for exported symbols again
if it.kind == nkSym and sfExported notin it.sym.flags:
if d.module != nil and d.module == it.sym.owner:
generateDoc(d, it.sym.ast, orig, config, kForceExport)
if it.kind == nkSym:
if d.module != nil and d.module == it.sym.owner: # in current module
# bug #23051; don't generate documentation for exported symbols again
if sfExported notin it.sym.flags:
generateDoc(d, it.sym.ast, orig, config, kForceExport)
# else it's to be handled in `of XxxSection` branch
elif it.sym.ast != nil:
# only export symbols in imported modules, not in current module
exportSym(d, it.sym)
of nkExportExceptStmt: discard "transformed into nkExportStmt by semExportExcept"
of nkFromStmt, nkImportExceptStmt: traceDeps(d, n[0])
of nkCallKinds:
var comm: ItemPre = default(ItemPre)
var comm = default(ItemPre)
getAllRunnableExamples(d, n, comm)
if comm.len != 0: d.modDescPre.add(comm)
else: discard
@@ -1889,6 +1900,9 @@ proc commandJson*(cache: IdentCache, conf: ConfigRef) =
else:
#echo getOutFile(gProjectFull, JsonExt)
let filename = getOutFile(conf, RelativeFile conf.projectName, JsonExt)
conf.outFile = filename.relativeTo(conf.outDir)
let dir = filename.splitFile.dir
createDir(dir)
try:
writeFile(filename, content)
except IOError:
@@ -1909,8 +1923,10 @@ proc commandTags*(cache: IdentCache, conf: ConfigRef) =
if optStdout in d.conf.globalOptions:
write(stdout, content)
else:
#echo getOutFile(gProjectFull, TagsExt)
let filename = getOutFile(conf, RelativeFile conf.projectName, TagsExt)
conf.outFile = filename.relativeTo(conf.outDir)
let dir = filename.splitFile.dir
createDir(dir)
try:
writeFile(filename, content)
except IOError:

View File

@@ -33,6 +33,10 @@ proc genEnumToStrProc*(t: PType; info: TLineInfo; g: ModuleGraph; idgen: IdGener
caseStmt.add newTree(nkOfBranch, newIntTypeNode(field.position, t),
newTree(nkStmtList, newTree(nkFastAsgn, newSymNode(res), newStrNode(val, info))))
#newIntTypeNode(nkIntLit, field.position, t)
# safety branch for invalid data:
caseStmt.add newTree(nkElse,
newTree(nkStmtList, newTree(nkFastAsgn, newSymNode(res),
newStrNode("", info))))
body.add(caseStmt)
@@ -43,8 +47,8 @@ proc genEnumToStrProc*(t: PType; info: TLineInfo; g: ModuleGraph; idgen: IdGener
n[bodyPos] = body
n[resultPos] = newSymNode(res)
result.ast = n
incl result.flags, sfFromGeneric
incl result.flags, sfNeverRaises
incl result.flagsImpl, {sfFromGeneric, sfNeverRaises}
setHookDisamb(g, result, "$enumtostr", t)
proc searchObjCaseImpl(obj: PNode; field: PSym): PNode =
case obj.kind
@@ -106,5 +110,4 @@ proc genCaseObjDiscMapping*(t: PType; field: PSym; info: TLineInfo; g: ModuleGra
n[bodyPos] = body
n[resultPos] = newSymNode(res)
result.ast = n
incl result.flags, sfFromGeneric
incl result.flags, sfNeverRaises
incl result.flagsImpl, {sfFromGeneric, sfNeverRaises}

View File

@@ -275,7 +275,7 @@ proc unpackObject(conf: ConfigRef, x: pointer, typ: PType, n: PNode): PNode =
# the nkPar node:
if n.isNil:
result = newNode(nkTupleConstr)
result.typ() = typ
result.typ = typ
if typ.n.isNil:
internalError(conf, "cannot unpack unnamed tuple")
unpackObjectAdd(conf, x, typ.n, result)
@@ -298,7 +298,7 @@ proc unpackObject(conf: ConfigRef, x: pointer, typ: PType, n: PNode): PNode =
proc unpackArray(conf: ConfigRef, x: pointer, typ: PType, n: PNode): PNode =
if n.isNil:
result = newNode(nkBracket)
result.typ() = typ
result.typ = typ
newSeq(result.sons, lengthOrd(conf, typ).toInt)
else:
result = n
@@ -319,7 +319,7 @@ proc unpack(conf: ConfigRef, x: pointer, typ: PType, n: PNode): PNode =
template aw(k, v, field: untyped): untyped =
if n.isNil:
result = newNode(k)
result.typ() = typ
result.typ = typ
else:
# check we have the right field:
result = n
@@ -333,12 +333,12 @@ proc unpack(conf: ConfigRef, x: pointer, typ: PType, n: PNode): PNode =
template setNil() =
if n.isNil:
result = newNode(nkNilLit)
result.typ() = typ
result.typ = typ
else:
reset n[]
result = n
result[] = TNode(kind: nkNilLit)
result.typ() = typ
result.typ = typ
template awi(kind, v: untyped): untyped = aw(kind, v, intVal)
template awf(kind, v: untyped): untyped = aw(kind, v, floatVal)
@@ -427,7 +427,7 @@ proc fficast*(conf: ConfigRef, x: PNode, destTyp: PType): PNode =
# cast through a pointer needs a new inner object:
let y = if x.kind == nkRefTy: newNodeI(nkRefTy, x.info, 1)
else: x.copyTree
y.typ() = x.typ
y.typ = x.typ
result = unpack(conf, a, destTyp, y)
dealloc a
@@ -481,7 +481,7 @@ proc callForeignFunction*(conf: ConfigRef, fn: PNode, fntyp: PType,
if aTyp.isNil:
internalAssert conf, i+1 < fntyp.len
aTyp = fntyp[i+1]
args[i+start].typ() = aTyp
args[i+start].typ = aTyp
sig[i] = mapType(conf, aTyp)
if sig[i].isNil: globalError(conf, info, "cannot map FFI type")

View File

@@ -182,7 +182,7 @@ proc wrapInComesFrom*(info: TLineInfo; sym: PSym; res: PNode): PNode =
d.add newSymNode(sym, info)
result.add d
result.add res
result.typ() = res.typ
result.typ = res.typ
proc evalTemplate*(n: PNode, tmpl, genSymOwner: PSym;
conf: ConfigRef;

View File

@@ -257,8 +257,8 @@ compiler tcc:
linkerExe: "tcc",
linkTmpl: "-o $exefile $options $buildgui $builddll $objfiles",
includeCmd: " -I",
linkDirCmd: "", # XXX: not supported yet
linkLibCmd: "", # XXX: not supported yet
linkDirCmd: " -L",
linkLibCmd: " -l$1",
debug: " -g ",
pic: "",
asmStmtFrmt: "asm($1);$n",
@@ -341,7 +341,7 @@ proc getConfigVar(conf: ConfigRef; c: TSystemCC, suffix: string): string =
var fullSuffix = suffix
case conf.backend
of backendCpp, backendJs, backendObjc: fullSuffix = "." & $conf.backend & suffix
of backendC: discard
of backendC, backendNif: discard
of backendInvalid:
# during parsing of cfg files; we don't know the backend yet, no point in
# guessing wrong thing
@@ -474,6 +474,11 @@ proc noAbsolutePaths(conf: ConfigRef): bool {.inline.} =
proc cFileSpecificOptions(conf: ConfigRef; nimname, fullNimFile: string): string =
result = conf.compileOptions
if (conf.cCompiler == ccGcc or conf.cCompiler == ccCLang) and
conf.selectedGC == gcRefc:
# bug #10625
addOpt(result, "-fno-omit-frame-pointer")
for option in conf.compileOptionsCmd:
if strutils.find(result, option, 0) < 0:
addOpt(result, option)
@@ -805,6 +810,59 @@ template tryExceptOSErrorMessage(conf: ConfigRef; errorPrefix: string = "", body
(ose.msg & " " & $ose.errorCode))
raise
proc createMacAppBundle(conf: ConfigRef; exefile: AbsoluteFile) =
let (dir, name, _) = splitFile(exefile.string)
let appBundleName = name & ".app"
let appBundlePath = dir / appBundleName
let contentsPath = appBundlePath / "Contents"
let macosPath = contentsPath / "MacOS"
createDir(macosPath)
let bundleExePath = macosPath / name
copyFileWithPermissions(exefile.string, bundleExePath)
let infoPlistPath = contentsPath / "Info.plist"
proc xmlEscape(s: string): string =
result = newStringOfCap(s.len)
for c in items(s):
case c:
of '<': result.add("&lt;")
of '>': result.add("&gt;")
of '&': result.add("&amp;")
of '"': result.add("&quot;")
of '\'': result.add("&apos;")
else:
if ord(c) < 32:
result.add("&#" & $ord(c) & ';')
else:
result.add(c)
let escapedName = xmlEscape(name)
let infoPlistContent = """<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>CFBundleExecutable</key>
<string>$1</string>
<key>CFBundleIdentifier</key>
<string>com.nim.$1</string>
<key>CFBundleName</key>
<string>$1</string>
<key>CFBundlePackageType</key>
<string>APPL</string>
<key>LSUIElement</key>
<string>1</string>
</dict>
</plist>""" % [escapedName]
writeFile(infoPlistPath, infoPlistContent)
removeFile(exefile.string)
rawMessage(conf, hintUserRaw, "Created Mac app bundle: " & appBundlePath)
proc getExtraCmds(conf: ConfigRef; output: AbsoluteFile): seq[string] =
result = @[]
when defined(macosx):
@@ -989,6 +1047,10 @@ proc callCCompiler*(conf: ConfigRef) =
preventLinkCmdMaxCmdLen(conf, linkCmd)
for cmd in extraCmds:
execExternalProgram(conf, cmd, hintExecuting)
# create Mac app bundle for GUI apps on macOS
when defined(macosx):
if conf.globalOptions * {optGenGuiApp, optGenDynLib, optGenStaticLib} == {optGenGuiApp}:
createMacAppBundle(conf, mainOutput)
else:
linkCmd = ""
if optGenScript in conf.globalOptions:

View File

@@ -46,7 +46,7 @@ proc isLocation(n: PNode): bool = not n.isValue
proc isLet(n: PNode): bool =
if n.kind == nkSym:
if n.sym.kind in {skLet, skTemp, skForVar}:
if n.sym.kind in {skLet, skConst, skTemp, skForVar}: # guard immutable variables
result = true
elif n.sym.kind == skParam and skipTypes(n.sym.typ,
abstractInst).kind notin {tyVar}:
@@ -1104,7 +1104,7 @@ proc settype(n: PNode): PType =
proc buildOf(it, loc: PNode; o: Operators): PNode =
var s = newNodeI(nkCurly, it.info, it.len-1)
s.typ() = settype(loc)
s.typ = settype(loc)
for i in 0..<it.len-1: s[i] = it[i]
result = newNodeI(nkCall, it.info, 3)
result[0] = newSymNode(o.opContains)
@@ -1170,7 +1170,7 @@ proc buildProperFieldCheck(access, check: PNode; o: Operators): PNode =
# set field name to discriminator field name
a[1] = check[2]
# set discriminator field type: important for `neg`
a.typ() = check[2].typ
a.typ = check[2].typ
result[2] = a
# 'access.kind != nkDotExpr' can happen for object constructors
# which we don't check yet

View File

@@ -10,7 +10,7 @@
# This include implements the high level optimization pass.
# included from sem.nim
proc hlo(c: PContext, n: PNode): PNode
proc hlo(c: PContext, n: PNode, loopDetector: int): PNode
proc evalPattern(c: PContext, n, orig: PNode): PNode =
internalAssert c.config, n.kind == nkCall and n[0].kind == nkSym
@@ -61,10 +61,11 @@ proc applyPatterns(c: PContext, n: PNode): PNode =
# activate this pattern again:
c.patterns[i] = pattern
proc hlo(c: PContext, n: PNode): PNode =
inc(c.hloLoopDetector)
proc hlo(c: PContext, n: PNode, loopDetector: int): PNode =
# simply stop and do not perform any further transformations:
if c.hloLoopDetector > 300: return n
if loopDetector > 300:
message(c.config, n.info, warnUser, "term rewrite macro instantiation too nested")
return n
case n.kind
of nkMacroDef, nkTemplateDef, procDefs:
# already processed (special cases in semstmts.nim)
@@ -80,7 +81,7 @@ proc hlo(c: PContext, n: PNode): PNode =
# no optimization applied, try subtrees:
for i in 0..<result.safeLen:
let a = result[i]
let h = hlo(c, a)
let h = hlo(c, a, loopDetector)
if h != a: result[i] = h
else:
# perform type checking, so that the replacement still fits:
@@ -90,17 +91,15 @@ proc hlo(c: PContext, n: PNode): PNode =
result = fitNode(c, n.typ, result, n.info)
# optimization has been applied so check again:
result = commonOptimizations(c.graph, c.idgen, c.module, result)
result = hlo(c, result)
result = hlo(c, result, loopDetector + 1)
result = commonOptimizations(c.graph, c.idgen, c.module, result)
proc hloBody(c: PContext, n: PNode): PNode =
# fast exit:
if c.patterns.len == 0 or optTrMacros notin c.config.options: return n
c.hloLoopDetector = 0
result = hlo(c, n)
result = hlo(c, n, 0)
proc hloStmt(c: PContext, n: PNode): PNode =
# fast exit:
if c.patterns.len == 0 or optTrMacros notin c.config.options: return n
c.hloLoopDetector = 0
result = hlo(c, n)
result = hlo(c, n, 0)

View File

@@ -1,178 +0,0 @@
## A BiTable is a table that can be seen as an optimized pair
## of `(Table[LitId, Val], Table[Val, LitId])`.
import std/hashes
import rodfiles
when defined(nimPreviewSlimSystem):
import std/assertions
type
LitId* = distinct uint32
BiTable*[T] = object
vals: seq[T] # indexed by LitId
keys: seq[LitId] # indexed by hash(val)
proc initBiTable*[T](): BiTable[T] = BiTable[T](vals: @[], keys: @[])
proc nextTry(h, maxHash: Hash): Hash {.inline.} =
result = (h + 1) and maxHash
template maxHash(t): untyped = high(t.keys)
template isFilled(x: LitId): bool = x.uint32 > 0'u32
proc `$`*(x: LitId): string {.borrow.}
proc `<`*(x, y: LitId): bool {.borrow.}
proc `<=`*(x, y: LitId): bool {.borrow.}
proc `==`*(x, y: LitId): bool {.borrow.}
proc hash*(x: LitId): Hash {.borrow.}
proc len*[T](t: BiTable[T]): int = t.vals.len
proc mustRehash(length, counter: int): bool {.inline.} =
assert(length > counter)
result = (length * 2 < counter * 3) or (length - counter < 4)
const
idStart = 1
template idToIdx(x: LitId): int = x.int - idStart
proc hasLitId*[T](t: BiTable[T]; x: LitId): bool =
let idx = idToIdx(x)
result = idx >= 0 and idx < t.vals.len
proc enlarge[T](t: var BiTable[T]) =
var n: seq[LitId]
newSeq(n, len(t.keys) * 2)
swap(t.keys, n)
for i in 0..high(n):
let eh = n[i]
if isFilled(eh):
var j = hash(t.vals[idToIdx eh]) and maxHash(t)
while isFilled(t.keys[j]):
j = nextTry(j, maxHash(t))
t.keys[j] = move n[i]
proc getKeyId*[T](t: BiTable[T]; v: T): LitId =
let origH = hash(v)
var h = origH and maxHash(t)
if t.keys.len != 0:
while true:
let litId = t.keys[h]
if not isFilled(litId): break
if t.vals[idToIdx t.keys[h]] == v: return litId
h = nextTry(h, maxHash(t))
return LitId(0)
proc getOrIncl*[T](t: var BiTable[T]; v: T): LitId =
let origH = hash(v)
var h = origH and maxHash(t)
if t.keys.len != 0:
while true:
let litId = t.keys[h]
if not isFilled(litId): break
if t.vals[idToIdx t.keys[h]] == v: return litId
h = nextTry(h, maxHash(t))
# not found, we need to insert it:
if mustRehash(t.keys.len, t.vals.len):
enlarge(t)
# recompute where to insert:
h = origH and maxHash(t)
while true:
let litId = t.keys[h]
if not isFilled(litId): break
h = nextTry(h, maxHash(t))
else:
setLen(t.keys, 16)
h = origH and maxHash(t)
result = LitId(t.vals.len + idStart)
t.keys[h] = result
t.vals.add v
proc `[]`*[T](t: var BiTable[T]; litId: LitId): var T {.inline.} =
let idx = idToIdx litId
assert idx < t.vals.len
result = t.vals[idx]
proc `[]`*[T](t: BiTable[T]; litId: LitId): lent T {.inline.} =
let idx = idToIdx litId
assert idx < t.vals.len
result = t.vals[idx]
proc hash*[T](t: BiTable[T]): Hash =
## as the keys are hashes of the values, we simply use them instead
var h: Hash = 0
for i, n in pairs t.keys:
h = h !& hash((i, n))
result = !$h
proc store*[T](f: var RodFile; t: BiTable[T]) =
storeSeq(f, t.vals)
storeSeq(f, t.keys)
proc load*[T](f: var RodFile; t: var BiTable[T]) =
loadSeq(f, t.vals)
loadSeq(f, t.keys)
proc sizeOnDisc*(t: BiTable[string]): int =
result = 4
for x in t.vals:
result += x.len + 4
result += t.keys.len * sizeof(LitId)
when isMainModule:
var t: BiTable[string]
echo getOrIncl(t, "hello")
echo getOrIncl(t, "hello")
echo getOrIncl(t, "hello3")
echo getOrIncl(t, "hello4")
echo getOrIncl(t, "helloasfasdfdsa")
echo getOrIncl(t, "hello")
echo getKeyId(t, "hello")
echo getKeyId(t, "none")
for i in 0 ..< 100_000:
discard t.getOrIncl($i & "___" & $i)
for i in 0 ..< 100_000:
assert t.getOrIncl($i & "___" & $i).idToIdx == i + 4
echo "begin"
echo t.vals.len
echo t.vals[0]
echo t.vals[1004]
echo "middle"
var tf: BiTable[float]
discard tf.getOrIncl(0.4)
discard tf.getOrIncl(16.4)
discard tf.getOrIncl(32.4)
echo getKeyId(tf, 32.4)
var f2 = open("testblah.bin", fmWrite)
echo store(f2, tf)
f2.close
var f1 = open("testblah.bin", fmRead)
var t2: BiTable[float]
echo f1.load(t2)
echo t2.vals.len
echo getKeyId(t2, 32.4)
echo "end"
f1.close

View File

@@ -1,180 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2021 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## New entry point into our C/C++ code generator. Ideally
## somebody would rewrite the old backend (which is 8000 lines of crufty Nim code)
## to work on packed trees directly and produce the C code as an AST which can
## then be rendered to text in a very simple manner. Unfortunately nobody wrote
## this code. So instead we wrap the existing cgen.nim and its friends so that
## we call directly into the existing code generation logic but avoiding the
## naive, outdated `passes` design. Thus you will see some
## `useAliveDataFromDce in flags` checks in the old code -- the old code is
## also doing cross-module dependency tracking and DCE that we don't need
## anymore. DCE is now done as prepass over the entire packed module graph.
import std/[packedsets, algorithm, tables]
when defined(nimPreviewSlimSystem):
import std/assertions
import ".."/[ast, options, lineinfos, modulegraphs, cgendata, cgen,
pathutils, extccomp, msgs, modulepaths]
import packed_ast, ic, dce, rodfiles
proc unpackTree(g: ModuleGraph; thisModule: int;
tree: PackedTree; n: NodePos): PNode =
var decoder = initPackedDecoder(g.config, g.cache)
result = loadNodes(decoder, g.packed, thisModule, tree, n)
proc setupBackendModule(g: ModuleGraph; m: var LoadedModule) =
if g.backend == nil:
g.backend = cgendata.newModuleList(g)
assert g.backend != nil
var bmod = cgen.newModule(BModuleList(g.backend), m.module, g.config)
bmod.idgen = idgenFromLoadedModule(m)
proc generateCodeForModule(g: ModuleGraph; m: var LoadedModule; alive: var AliveSyms) =
var bmod = BModuleList(g.backend).modules[m.module.position]
assert bmod != nil
bmod.flags.incl useAliveDataFromDce
bmod.alive = move alive[m.module.position]
for p in allNodes(m.fromDisk.topLevel):
let n = unpackTree(g, m.module.position, m.fromDisk.topLevel, p)
cgen.genTopLevelStmt(bmod, n)
finalCodegenActions(g, bmod, newNodeI(nkStmtList, m.module.info))
for disp in getDispatchers(g):
genProcAux(bmod, disp)
m.fromDisk.backendFlags = cgen.whichInitProcs(bmod)
proc replayTypeInfo(g: ModuleGraph; m: var LoadedModule; origin: FileIndex) =
for x in mitems(m.fromDisk.emittedTypeInfo):
#echo "found type ", x, " for file ", int(origin)
g.emittedTypeInfo[x] = origin
proc addFileToLink(config: ConfigRef; m: PSym) =
let filename = AbsoluteFile toFullPath(config, m.position.FileIndex)
let ext =
if config.backend == backendCpp: ".nim.cpp"
elif config.backend == backendObjc: ".nim.m"
else: ".nim.c"
let cfile = changeFileExt(completeCfilePath(config,
mangleModuleName(config, filename).AbsoluteFile), ext)
let objFile = completeCfilePath(config, toObjFile(config, cfile))
if fileExists(objFile):
var cf = Cfile(nimname: m.name.s, cname: cfile,
obj: objFile,
flags: {CfileFlag.Cached})
addFileToCompile(config, cf)
when defined(debugDce):
import os, std/packedsets
proc storeAliveSymsImpl(asymFile: AbsoluteFile; s: seq[int32]) =
var f = rodfiles.create(asymFile.string)
f.storeHeader()
f.storeSection aliveSymsSection
f.storeSeq(s)
close f
template prepare {.dirty.} =
let asymFile = toRodFile(config, AbsoluteFile toFullPath(config, position.FileIndex), ".alivesyms")
var s = newSeqOfCap[int32](alive[position].len)
for a in items(alive[position]): s.add int32(a)
sort(s)
proc storeAliveSyms(config: ConfigRef; position: int; alive: AliveSyms) =
prepare()
storeAliveSymsImpl(asymFile, s)
proc aliveSymsChanged(config: ConfigRef; position: int; alive: AliveSyms): bool =
prepare()
var f2 = rodfiles.open(asymFile.string)
f2.loadHeader()
f2.loadSection aliveSymsSection
var oldData: seq[int32] = @[]
f2.loadSeq(oldData)
f2.close
if f2.err == ok and oldData == s:
result = false
else:
when defined(debugDce):
let oldAsSet = toPackedSet[int32](oldData)
let newAsSet = toPackedSet[int32](s)
echo "set of live symbols changed ", asymFile.changeFileExt("rod"), " ", position, " ", f2.err
echo "in old but not in new ", oldAsSet.difference(newAsSet), " number of entries in old ", oldAsSet.len
echo "in new but not in old ", newAsSet.difference(oldAsSet), " number of entries in new ", newAsSet.len
#if execShellCmd(getAppFilename() & " rod " & quoteShell(asymFile.changeFileExt("rod"))) != 0:
# echo "command failed"
result = true
storeAliveSymsImpl(asymFile, s)
proc genPackedModule(g: ModuleGraph, i: int; alive: var AliveSyms) =
# case statement here to enforce exhaustive checks.
case g.packed[i].status
of undefined:
discard "nothing to do"
of loading, stored:
assert false
of storing, outdated:
storeAliveSyms(g.config, g.packed[i].module.position, alive)
generateCodeForModule(g, g.packed[i], alive)
closeRodFile(g, g.packed[i].module)
of loaded:
if g.packed[i].loadedButAliveSetChanged:
generateCodeForModule(g, g.packed[i], alive)
else:
addFileToLink(g.config, g.packed[i].module)
replayTypeInfo(g, g.packed[i], FileIndex(i))
if g.backend == nil:
g.backend = cgendata.newModuleList(g)
registerInitProcs(BModuleList(g.backend), g.packed[i].module, g.packed[i].fromDisk.backendFlags)
proc generateCode*(g: ModuleGraph) =
## The single entry point, generate C(++) code for the entire
## Nim program aka `ModuleGraph`.
resetForBackend(g)
var alive = computeAliveSyms(g.packed, g.config)
when false:
for i in 0..<len(g.packed):
echo i, " is of status ", g.packed[i].status, " ", toFullPath(g.config, FileIndex(i))
# First pass: Setup all the backend modules for all the modules that have
# changed:
for i in 0..<len(g.packed):
# case statement here to enforce exhaustive checks.
case g.packed[i].status
of undefined:
discard "nothing to do"
of loading, stored:
assert false
of storing, outdated:
setupBackendModule(g, g.packed[i])
of loaded:
# Even though this module didn't change, DCE might trigger a change.
# Consider this case: Module A uses symbol S from B and B does not use
# S itself. A is then edited not to use S either. Thus we have to
# recompile B in order to remove S from the final result.
if aliveSymsChanged(g.config, g.packed[i].module.position, alive):
g.packed[i].loadedButAliveSetChanged = true
setupBackendModule(g, g.packed[i])
# Second pass: Code generation.
let mainModuleIdx = g.config.projectMainIdx2.int
# We need to generate the main module last, because only then
# all init procs have been registered:
for i in 0..<len(g.packed):
if i != mainModuleIdx:
genPackedModule(g, i, alive)
if mainModuleIdx >= 0:
genPackedModule(g, mainModuleIdx, alive)

View File

@@ -1,169 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2021 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## Dead code elimination (=DCE) for IC.
import std/[intsets, tables]
when defined(nimPreviewSlimSystem):
import std/assertions
import ".." / [ast, options, lineinfos, types]
import packed_ast, ic, bitabs
type
AliveSyms* = seq[IntSet]
AliveContext* = object ## Purpose is to fill the 'alive' field.
stack: seq[(int, TOptions, NodePos)] ## A stack for marking symbols as alive.
decoder: PackedDecoder ## We need a PackedDecoder for module ID address translations.
thisModule: int ## The module we're currently analysing for DCE.
alive: AliveSyms ## The final result of our computation.
options: TOptions
compilerProcs: Table[string, (int, int32)]
proc isExportedToC(c: var AliveContext; g: PackedModuleGraph; symId: int32): bool =
## "Exported to C" procs are special (these are marked with '.exportc') because these
## must not be optimized away!
let symPtr = unsafeAddr g[c.thisModule].fromDisk.syms[symId]
let flags = symPtr.flags
# due to a bug/limitation in the lambda lifting, unused inner procs
# are not transformed correctly; issue (#411). However, the whole purpose here
# is to eliminate unused procs. So there is no special logic required for this case.
if sfCompileTime notin flags:
if ({sfExportc, sfCompilerProc} * flags != {}) or
(symPtr.kind == skMethod):
result = true
else:
result = false
# XXX: This used to be a condition to:
# (sfExportc in prc.flags and lfExportLib in prc.loc.flags) or
if sfCompilerProc in flags:
c.compilerProcs[g[c.thisModule].fromDisk.strings[symPtr.name]] = (c.thisModule, symId)
else:
result = false
template isNotGeneric(n: NodePos): bool = ithSon(tree, n, genericParamsPos).kind == nkEmpty
proc followLater(c: var AliveContext; g: PackedModuleGraph; module: int; item: int32) =
## Marks a symbol 'item' as used and later in 'followNow' the symbol's body will
## be analysed.
if not c.alive[module].containsOrIncl(item):
var body = g[module].fromDisk.syms[item].ast
if body != emptyNodeId:
let opt = g[module].fromDisk.syms[item].options
if g[module].fromDisk.syms[item].kind in routineKinds:
body = NodeId ithSon(g[module].fromDisk.bodies, NodePos body, bodyPos)
c.stack.add((module, opt, NodePos(body)))
when false:
let nid = g[module].fromDisk.syms[item].name
if nid != LitId(0):
let name = g[module].fromDisk.strings[nid]
if name in ["nimFrame", "callDepthLimitReached"]:
echo "I was called! ", name, " body exists: ", body != emptyNodeId, " ", module, " ", item
proc requestCompilerProc(c: var AliveContext; g: PackedModuleGraph; name: string) =
let (module, item) = c.compilerProcs[name]
followLater(c, g, module, item)
proc loadTypeKind(t: PackedItemId; c: AliveContext; g: PackedModuleGraph; toSkip: set[TTypeKind]): TTypeKind =
template kind(t: ItemId): TTypeKind = g[t.module].fromDisk.types[t.item].kind
var t2 = translateId(t, g, c.thisModule, c.decoder.config)
result = t2.kind
while result in toSkip:
t2 = translateId(g[t2.module].fromDisk.types[t2.item].types[^1], g, t2.module, c.decoder.config)
result = t2.kind
proc rangeCheckAnalysis(c: var AliveContext; g: PackedModuleGraph; tree: PackedTree; n: NodePos) =
## Replicates the logic of `ccgexprs.genRangeChck`.
## XXX Refactor so that the duplicated logic is avoided. However, for now it's not clear
## the approach has enough merit.
var dest = loadTypeKind(n.typ, c, g, abstractVar)
if optRangeCheck notin c.options or dest in {tyUInt..tyUInt64}:
discard "no need to generate a check because it was disabled"
else:
let n0t = loadTypeKind(n.firstSon.typ, c, g, {})
if n0t in {tyUInt, tyUInt64}:
c.requestCompilerProc(g, "raiseRangeErrorNoArgs")
else:
let raiser =
case loadTypeKind(n.typ, c, g, abstractVarRange)
of tyUInt..tyUInt64, tyChar: "raiseRangeErrorU"
of tyFloat..tyFloat128: "raiseRangeErrorF"
else: "raiseRangeErrorI"
c.requestCompilerProc(g, raiser)
proc aliveCode(c: var AliveContext; g: PackedModuleGraph; tree: PackedTree; n: NodePos) =
## Marks the symbols we encounter when we traverse the AST at `tree[n]` as alive, unless
## it is purely in a declarative context (type section etc.).
case n.kind
of nkNone..pred(nkSym), succ(nkSym)..nkNilLit:
discard "ignore non-sym atoms"
of nkSym:
# This symbol is alive and everything its body references.
followLater(c, g, c.thisModule, tree[n].soperand)
of nkModuleRef:
let (n1, n2) = sons2(tree, n)
assert n1.kind == nkNone
assert n2.kind == nkNone
let m = n1.litId
let item = tree[n2].soperand
let otherModule = toFileIndexCached(c.decoder, g, c.thisModule, m).int
followLater(c, g, otherModule, item)
of nkMacroDef, nkTemplateDef, nkTypeSection, nkTypeOfExpr,
nkCommentStmt, nkIncludeStmt,
nkImportStmt, nkImportExceptStmt, nkExportStmt, nkExportExceptStmt,
nkFromStmt, nkStaticStmt:
discard
of nkVarSection, nkLetSection, nkConstSection:
# XXX ignore the defining local variable name?
for son in sonsReadonly(tree, n):
aliveCode(c, g, tree, son)
of nkChckRangeF, nkChckRange64, nkChckRange:
rangeCheckAnalysis(c, g, tree, n)
of nkProcDef, nkConverterDef, nkMethodDef, nkFuncDef, nkIteratorDef:
if n.firstSon.kind == nkSym and isNotGeneric(n):
let item = tree[n.firstSon].soperand
if isExportedToC(c, g, item):
# This symbol is alive and everything its body references.
followLater(c, g, c.thisModule, item)
else:
for son in sonsReadonly(tree, n):
aliveCode(c, g, tree, son)
proc followNow(c: var AliveContext; g: PackedModuleGraph) =
## Mark all entries in the stack. Marking can add more entries
## to the stack but eventually we have looked at every alive symbol.
while c.stack.len > 0:
let (modId, opt, ast) = c.stack.pop()
c.thisModule = modId
c.options = opt
aliveCode(c, g, g[modId].fromDisk.bodies, ast)
proc computeAliveSyms*(g: PackedModuleGraph; conf: ConfigRef): AliveSyms =
## Entry point for our DCE algorithm.
var c = AliveContext(stack: @[], decoder: PackedDecoder(config: conf),
thisModule: -1, alive: newSeq[IntSet](g.len),
options: conf.options)
for i in countdown(len(g)-1, 0):
if g[i].status != undefined:
c.thisModule = i
for p in allNodes(g[i].fromDisk.topLevel):
aliveCode(c, g, g[i].fromDisk.topLevel, p)
followNow(c, g)
result = move(c.alive)
proc isAlive*(a: AliveSyms; module: int, item: int32): bool =
## Backends use this to query if a symbol is `alive` which means
## we need to produce (C/C++/etc) code for it.
result = a[module].contains(item)

View File

@@ -1,56 +0,0 @@
====================================
Incremental Recompilations
====================================
We split the Nim compiler into a frontend and a backend.
The frontend produces a set of `.rod` files. Every `.nim` module
produces its own `.rod` file.
- The IR must be a faithful representation of the AST in memory.
- The backend can do its own caching but doesn't have to. In the
current implementation the backend also caches its results.
Advantage of the "set of files" vs the previous global database:
- By construction, we either read from the `.rod` file or from the
`.nim` file, there can be no inconsistency. There can also be no
partial updates.
- No dependency to external packages (SQLite). SQLite simply is too
slow and the old way of serialization was too slow too. We use a
format designed for Nim and expect to base further tools on this
file format.
References to external modules must be (moduleId, symId) pairs.
The symbol IDs are module specific. This way no global ID increment
mechanism needs to be implemented that we could get wrong. ModuleIds
are rod-file specific too.
Global state
------------
There is no global state.
Rod File Format
---------------
It's a simple binary file format. `rodfiles.nim` contains some details.
Backend
-------
Nim programmers have to come to enjoy whole-program dead code elimination,
by default. Since this is a "whole program" optimization, it does break
modularity. However, thanks to the packed AST representation we can perform
this global analysis without having to unpack anything. This is basically
a mark&sweep GC algorithm:
- Start with the top level statements. Every symbol that is referenced
from a top level statement is not "dead" and needs to be compiled by
the backend.
- Every symbol referenced from a referenced symbol also has to be
compiled.
Caching logic: Only if the set of alive symbols is different from the
last run, the module has to be regenerated.

1863
compiler/ic/enum2nif.nim Normal file

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -1,84 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2024 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
# For the line information we use 32 bits. They are used as follows:
# Bit 0 (AsideBit): If we have inline line information or not. If not, the
# remaining 31 bits are used as an index into a seq[(LitId, int, int)].
#
# We use 10 bits for the "file ID", this means a program can consist of as much
# as 1024 different files. (If it uses more files than that, the overflow bit
# would be set.)
# This means we have 21 bits left to encode the (line, col) pair. We use 7 bits for the column
# so 128 is the limit and 14 bits for the line number.
# The packed representation supports files with up to 16384 lines.
# Keep in mind that whenever any limit is reached the AsideBit is set and the real line
# information is kept in a side channel.
import std / assertions
const
AsideBit = 1
FileBits = 10
LineBits = 14
ColBits = 7
FileMax = (1 shl FileBits) - 1
LineMax = (1 shl LineBits) - 1
ColMax = (1 shl ColBits) - 1
static:
assert AsideBit + FileBits + LineBits + ColBits == 32
import .. / ic / [bitabs, rodfiles] # for LitId
type
PackedLineInfo* = distinct uint32
LineInfoManager* = object
aside: seq[(LitId, int32, int32)]
const
NoLineInfo* = PackedLineInfo(0'u32)
proc pack*(m: var LineInfoManager; file: LitId; line, col: int32): PackedLineInfo =
if file.uint32 <= FileMax.uint32 and line <= LineMax and col <= ColMax:
let col = if col < 0'i32: 0'u32 else: col.uint32
let line = if line < 0'i32: 0'u32 else: line.uint32
# use inline representation:
result = PackedLineInfo((file.uint32 shl 1'u32) or (line shl uint32(AsideBit + FileBits)) or
(col shl uint32(AsideBit + FileBits + LineBits)))
else:
result = PackedLineInfo((m.aside.len shl 1) or AsideBit)
m.aside.add (file, line, col)
proc unpack*(m: LineInfoManager; i: PackedLineInfo): (LitId, int32, int32) =
let i = i.uint32
if (i and 1'u32) == 0'u32:
# inline representation:
result = (LitId((i shr 1'u32) and FileMax.uint32),
int32((i shr uint32(AsideBit + FileBits)) and LineMax.uint32),
int32((i shr uint32(AsideBit + FileBits + LineBits)) and ColMax.uint32))
else:
result = m.aside[int(i shr 1'u32)]
proc getFileId*(m: LineInfoManager; i: PackedLineInfo): LitId =
result = unpack(m, i)[0]
proc store*(r: var RodFile; m: LineInfoManager) = storeSeq(r, m.aside)
proc load*(r: var RodFile; m: var LineInfoManager) = loadSeq(r, m.aside)
when isMainModule:
var m = LineInfoManager(aside: @[])
for i in 0'i32..<16388'i32:
for col in 0'i32..<100'i32:
let packed = pack(m, LitId(1023), i, col)
let u = unpack(m, packed)
assert u[0] == LitId(1023)
assert u[1] == i
assert u[2] == col
echo m.aside.len

View File

@@ -1,155 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2021 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## Integrity checking for a set of .rod files.
## The set must cover a complete Nim project.
import std/[sets, tables]
when defined(nimPreviewSlimSystem):
import std/assertions
import ".." / [ast, modulegraphs]
import packed_ast, bitabs, ic
type
CheckedContext = object
g: ModuleGraph
thisModule: int32
checkedSyms: HashSet[ItemId]
checkedTypes: HashSet[ItemId]
proc checkType(c: var CheckedContext; typeId: PackedItemId)
proc checkForeignSym(c: var CheckedContext; symId: PackedItemId)
proc checkNode(c: var CheckedContext; tree: PackedTree; n: NodePos)
proc checkTypeObj(c: var CheckedContext; typ: PackedType) =
for child in typ.types:
checkType(c, child)
if typ.n != emptyNodeId:
checkNode(c, c.g.packed[c.thisModule].fromDisk.bodies, NodePos typ.n)
if typ.sym != nilItemId:
checkForeignSym(c, typ.sym)
if typ.owner != nilItemId:
checkForeignSym(c, typ.owner)
checkType(c, typ.typeInst)
proc checkType(c: var CheckedContext; typeId: PackedItemId) =
if typeId == nilItemId: return
let itemId = translateId(typeId, c.g.packed, c.thisModule, c.g.config)
if not c.checkedTypes.containsOrIncl(itemId):
let oldThisModule = c.thisModule
c.thisModule = itemId.module
checkTypeObj c, c.g.packed[itemId.module].fromDisk.types[itemId.item]
c.thisModule = oldThisModule
proc checkSym(c: var CheckedContext; s: PackedSym) =
if s.name != LitId(0):
assert c.g.packed[c.thisModule].fromDisk.strings.hasLitId s.name
checkType c, s.typ
if s.ast != emptyNodeId:
checkNode(c, c.g.packed[c.thisModule].fromDisk.bodies, NodePos s.ast)
if s.owner != nilItemId:
checkForeignSym(c, s.owner)
proc checkLocalSym(c: var CheckedContext; item: int32) =
let itemId = ItemId(module: c.thisModule, item: item)
if not c.checkedSyms.containsOrIncl(itemId):
checkSym c, c.g.packed[c.thisModule].fromDisk.syms[item]
proc checkForeignSym(c: var CheckedContext; symId: PackedItemId) =
let itemId = translateId(symId, c.g.packed, c.thisModule, c.g.config)
if not c.checkedSyms.containsOrIncl(itemId):
let oldThisModule = c.thisModule
c.thisModule = itemId.module
checkSym c, c.g.packed[itemId.module].fromDisk.syms[itemId.item]
c.thisModule = oldThisModule
proc checkNode(c: var CheckedContext; tree: PackedTree; n: NodePos) =
let t = findType(tree, n)
if t != nilItemId:
checkType(c, t)
case n.kind
of nkEmpty, nkNilLit, nkType, nkNilRodNode:
discard
of nkIdent:
assert c.g.packed[c.thisModule].fromDisk.strings.hasLitId n.litId
of nkSym:
checkLocalSym(c, tree[n].soperand)
of directIntLit:
discard
of externIntLit, nkFloatLit..nkFloat128Lit:
assert c.g.packed[c.thisModule].fromDisk.numbers.hasLitId n.litId
of nkStrLit..nkTripleStrLit:
assert c.g.packed[c.thisModule].fromDisk.strings.hasLitId n.litId
of nkModuleRef:
let (n1, n2) = sons2(tree, n)
assert n1.kind == nkNone
assert n2.kind == nkNone
checkForeignSym(c, PackedItemId(module: n1.litId, item: tree[n2].soperand))
else:
for n0 in sonsReadonly(tree, n):
checkNode(c, tree, n0)
proc checkTree(c: var CheckedContext; t: PackedTree) =
for p in allNodes(t): checkNode(c, t, p)
proc checkLocalSymIds(c: var CheckedContext; m: PackedModule; symIds: seq[int32]) =
for symId in symIds:
assert symId >= 0 and symId < m.syms.len, $symId & " " & $m.syms.len
proc checkModule(c: var CheckedContext; m: PackedModule) =
# We check that:
# - Every symbol references existing types and symbols.
# - Every tree node references existing types and symbols.
for _, v in pairs(m.syms):
checkLocalSym c, v.id
checkTree c, m.toReplay
checkTree c, m.topLevel
for e in m.exports:
#assert e[1] >= 0 and e[1] < m.syms.len
assert e[0] == m.syms[e[1]].name
for e in m.compilerProcs:
#assert e[1] >= 0 and e[1] < m.syms.len
assert e[0] == m.syms[e[1]].name
checkLocalSymIds c, m, m.converters
checkLocalSymIds c, m, m.methods
checkLocalSymIds c, m, m.trmacros
checkLocalSymIds c, m, m.pureEnums
#[
To do: Check all these fields:
reexports*: seq[(LitId, PackedItemId)]
macroUsages*: seq[(PackedItemId, PackedLineInfo)]
typeInstCache*: seq[(PackedItemId, PackedItemId)]
procInstCache*: seq[PackedInstantiation]
attachedOps*: seq[(TTypeAttachedOp, PackedItemId, PackedItemId)]
methodsPerGenericType*: seq[(PackedItemId, int, PackedItemId)]
enumToStringProcs*: seq[(PackedItemId, PackedItemId)]
methodsPerType*: seq[(PackedItemId, PackedItemId)]
dispatchers*: seq[PackedItemId]
]#
proc checkIntegrity*(g: ModuleGraph) =
var c = CheckedContext(g: g)
for i in 0..<len(g.packed):
# case statement here to enforce exhaustive checks.
case g.packed[i].status
of undefined:
discard "nothing to do"
of loading:
assert false, "cannot check integrity: Module still loading"
of stored, storing, outdated, loaded:
c.thisModule = int32 i
checkModule(c, g.packed[i].fromDisk)

View File

@@ -1,183 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2021 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## Supports the "nim check --ic:on --defusages:FILE,LINE,COL"
## IDE-like features. It uses the set of .rod files to accomplish
## its task. The set must cover a complete Nim project.
import std/[sets, tables]
from std/os import nil
from std/private/miscdollars import toLocation
when defined(nimPreviewSlimSystem):
import std/assertions
import ".." / [ast, modulegraphs, msgs, options]
import iclineinfos
import packed_ast, bitabs, ic
type
UnpackedLineInfo = object
file: LitId
line, col: int
NavContext = object
g: ModuleGraph
thisModule: int32
trackPos: UnpackedLineInfo
alreadyEmitted: HashSet[string]
outputSep: char # for easier testing, use short filenames and spaces instead of tabs.
proc isTracked(man: LineInfoManager; current: PackedLineInfo, trackPos: UnpackedLineInfo, tokenLen: int): bool =
let (currentFile, currentLine, currentCol) = man.unpack(current)
if currentFile == trackPos.file and currentLine == trackPos.line:
let col = trackPos.col
if col >= currentCol and col < currentCol+tokenLen:
result = true
else:
result = false
else:
result = false
proc searchLocalSym(c: var NavContext; s: PackedSym; info: PackedLineInfo): bool =
result = s.name != LitId(0) and
isTracked(c.g.packed[c.thisModule].fromDisk.man, info, c.trackPos, c.g.packed[c.thisModule].fromDisk.strings[s.name].len)
proc searchForeignSym(c: var NavContext; s: ItemId; info: PackedLineInfo): bool =
let name = c.g.packed[s.module].fromDisk.syms[s.item].name
result = name != LitId(0) and
isTracked(c.g.packed[c.thisModule].fromDisk.man, info, c.trackPos, c.g.packed[s.module].fromDisk.strings[name].len)
const
EmptyItemId = ItemId(module: -1'i32, item: -1'i32)
proc search(c: var NavContext; tree: PackedTree): ItemId =
# We use the linear representation here directly:
for i in 0..<len(tree):
let i = NodePos(i)
case tree[i].kind
of nkSym:
let item = tree[i].soperand
if searchLocalSym(c, c.g.packed[c.thisModule].fromDisk.syms[item], tree[i].info):
return ItemId(module: c.thisModule, item: item)
of nkModuleRef:
let (currentFile, currentLine, currentCol) = c.g.packed[c.thisModule].fromDisk.man.unpack(tree[i].info)
if currentLine == c.trackPos.line and currentFile == c.trackPos.file:
let (n1, n2) = sons2(tree, i)
assert n1.kind == nkInt32Lit
assert n2.kind == nkInt32Lit
let pId = PackedItemId(module: n1.litId, item: tree[n2].soperand)
let itemId = translateId(pId, c.g.packed, c.thisModule, c.g.config)
if searchForeignSym(c, itemId, tree[i].info):
return itemId
else: discard
return EmptyItemId
proc isDecl(tree: PackedTree; n: NodePos): bool =
# XXX This is not correct yet.
const declarativeNodes = procDefs + {nkMacroDef, nkTemplateDef,
nkLetSection, nkVarSection, nkUsingStmt, nkConstSection, nkTypeSection,
nkIdentDefs, nkEnumTy, nkVarTuple}
result = n.int >= 0 and tree[n].kind in declarativeNodes
proc usage(c: var NavContext; info: PackedLineInfo; isDecl: bool) =
let (fileId, line, col) = unpack(c.g.packed[c.thisModule].fromDisk.man, info)
var m = ""
var file = c.g.packed[c.thisModule].fromDisk.strings[fileId]
if c.outputSep == ' ':
file = os.extractFilename file
toLocation(m, file, line, col + ColOffset)
if not c.alreadyEmitted.containsOrIncl(m):
msgWriteln c.g.config, (if isDecl: "def" else: "usage") & c.outputSep & m
proc list(c: var NavContext; tree: PackedTree; sym: ItemId) =
for i in 0..<len(tree):
let i = NodePos(i)
case tree[i].kind
of nkSym:
let item = tree[i].soperand
if sym.item == item and sym.module == c.thisModule:
usage(c, tree[i].info, isDecl(tree, parent(i)))
of nkModuleRef:
let (n1, n2) = sons2(tree, i)
assert n1.kind == nkNone
assert n2.kind == nkNone
let pId = PackedItemId(module: n1.litId, item: tree[n2].soperand)
let itemId = translateId(pId, c.g.packed, c.thisModule, c.g.config)
if itemId.item == sym.item and sym.module == itemId.module:
usage(c, tree[i].info, isDecl(tree, parent(i)))
else: discard
proc searchForIncludeFile(g: ModuleGraph; fullPath: string): int =
for i in 0..<len(g.packed):
for k in 1..high(g.packed[i].fromDisk.includes):
# we start from 1 because the first "include" file is
# the module's filename.
if os.cmpPaths(g.packed[i].fromDisk.strings[g.packed[i].fromDisk.includes[k][0]], fullPath) == 0:
return i
return -1
proc nav(g: ModuleGraph) =
# translate the track position to a packed position:
let unpacked = g.config.m.trackPos
var mid = unpacked.fileIndex.int
let fullPath = toFullPath(g.config, unpacked.fileIndex)
if g.packed[mid].status == undefined:
# check if 'mid' is an include file of some other module:
mid = searchForIncludeFile(g, fullPath)
if mid < 0:
localError(g.config, unpacked, "unknown file name: " & fullPath)
return
let fileId = g.packed[mid].fromDisk.strings.getKeyId(fullPath)
if fileId == LitId(0):
internalError(g.config, unpacked, "cannot find a valid file ID")
return
var c = NavContext(
g: g,
thisModule: int32 mid,
trackPos: UnpackedLineInfo(line: unpacked.line.int, col: unpacked.col.int, file: fileId),
outputSep: if isDefined(g.config, "nimIcNavigatorTests"): ' ' else: '\t'
)
var symId = search(c, g.packed[mid].fromDisk.topLevel)
if symId == EmptyItemId:
symId = search(c, g.packed[mid].fromDisk.bodies)
if symId == EmptyItemId:
localError(g.config, unpacked, "no symbol at this position")
return
for i in 0..<len(g.packed):
# case statement here to enforce exhaustive checks.
case g.packed[i].status
of undefined:
discard "nothing to do"
of loading:
assert false, "cannot check integrity: Module still loading"
of stored, storing, outdated, loaded:
c.thisModule = int32 i
list(c, g.packed[i].fromDisk.topLevel, symId)
list(c, g.packed[i].fromDisk.bodies, symId)
proc navDefinition*(g: ModuleGraph) = nav(g)
proc navUsages*(g: ModuleGraph) = nav(g)
proc navDefusages*(g: ModuleGraph) = nav(g)
proc writeRodFiles*(g: ModuleGraph) =
for i in 0..<len(g.packed):
case g.packed[i].status
of undefined, loading, stored, loaded:
discard "nothing to do"
of storing, outdated:
closeRodFile(g, g.packed[i].module)

View File

@@ -1,367 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2020 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## Packed AST representation, mostly based on a seq of nodes.
## For IC support. Far future: Rewrite the compiler passes to
## use this representation directly in all the transformations,
## it is superior.
import std/[hashes, tables, strtabs]
import bitabs, rodfiles
import ".." / [ast, options]
import iclineinfos
when defined(nimPreviewSlimSystem):
import std/assertions
type
SymId* = distinct int32
ModuleId* = distinct int32
NodePos* = distinct int
NodeId* = distinct int32
PackedItemId* = object
module*: LitId # 0 if it's this module
item*: int32 # same as the in-memory representation
const
nilItemId* = PackedItemId(module: LitId(0), item: 0.int32)
const
emptyNodeId* = NodeId(-1)
type
PackedLib* = object
kind*: TLibKind
generated*: bool
isOverridden*: bool
name*: LitId
path*: NodeId
PackedSym* = object
id*: int32
kind*: TSymKind
name*: LitId
typ*: PackedItemId
flags*: TSymFlags
magic*: TMagic
info*: PackedLineInfo
ast*: NodeId
owner*: PackedItemId
guard*: PackedItemId
bitsize*: int
alignment*: int # for alignment
options*: TOptions
position*: int
offset*: int32
disamb*: int32
externalName*: LitId # instead of TLoc
locFlags*: TLocFlags
annex*: PackedLib
when hasFFI:
cname*: LitId
constraint*: NodeId
instantiatedFrom*: PackedItemId
PackedType* = object
id*: int32
kind*: TTypeKind
callConv*: TCallingConvention
#nodekind*: TNodeKind
flags*: TTypeFlags
types*: seq[PackedItemId]
n*: NodeId
#nodeflags*: TNodeFlags
sym*: PackedItemId
owner*: PackedItemId
size*: BiggestInt
align*: int16
paddingAtEnd*: int16
# not serialized: loc*: TLoc because it is backend-specific
typeInst*: PackedItemId
nonUniqueId*: int32
PackedNode* = object # 8 bytes
x: uint32
info*: PackedLineInfo
PackedTree* = object ## usually represents a full Nim module
nodes: seq[PackedNode]
withFlags: seq[(int32, TNodeFlags)]
withTypes: seq[(int32, PackedItemId)]
PackedInstantiation* = object
key*, sym*: PackedItemId
concreteTypes*: seq[PackedItemId]
const
NodeKindBits = 8'u32
NodeKindMask = (1'u32 shl NodeKindBits) - 1'u32
template kind*(n: PackedNode): TNodeKind = TNodeKind(n.x and NodeKindMask)
template uoperand*(n: PackedNode): uint32 = (n.x shr NodeKindBits)
template soperand*(n: PackedNode): int32 = int32(uoperand(n))
template toX(k: TNodeKind; operand: uint32): uint32 =
uint32(k) or (operand shl NodeKindBits)
template toX(k: TNodeKind; operand: LitId): uint32 =
uint32(k) or (operand.uint32 shl NodeKindBits)
template typeId*(n: PackedNode): PackedItemId = n.typ
proc `==`*(a, b: SymId): bool {.borrow.}
proc hash*(a: SymId): Hash {.borrow.}
proc `==`*(a, b: NodePos): bool {.borrow.}
#proc `==`*(a, b: PackedItemId): bool {.borrow.}
proc `==`*(a, b: NodeId): bool {.borrow.}
proc newTreeFrom*(old: PackedTree): PackedTree =
result = PackedTree(nodes: @[])
when false: result.sh = old.sh
proc addIdent*(tree: var PackedTree; s: LitId; info: PackedLineInfo) =
tree.nodes.add PackedNode(x: toX(nkIdent, uint32(s)), info: info)
proc addSym*(tree: var PackedTree; s: int32; info: PackedLineInfo) =
tree.nodes.add PackedNode(x: toX(nkSym, cast[uint32](s)), info: info)
proc addSymDef*(tree: var PackedTree; s: SymId; info: PackedLineInfo) =
tree.nodes.add PackedNode(x: toX(nkSym, cast[uint32](s)), info: info)
proc isAtom*(tree: PackedTree; pos: int): bool {.inline.} = tree.nodes[pos].kind <= nkNilLit
type
PatchPos = distinct int
proc addNode*(t: var PackedTree; kind: TNodeKind; operand: int32;
typeId: PackedItemId = nilItemId; info: PackedLineInfo;
flags: TNodeFlags = {}) =
t.nodes.add PackedNode(x: toX(kind, cast[uint32](operand)), info: info)
if flags != {}:
t.withFlags.add (t.nodes.len.int32 - 1, flags)
if typeId != nilItemId:
t.withTypes.add (t.nodes.len.int32 - 1, typeId)
proc prepare*(tree: var PackedTree; kind: TNodeKind; flags: TNodeFlags; typeId: PackedItemId; info: PackedLineInfo): PatchPos =
result = PatchPos tree.nodes.len
tree.addNode(kind = kind, flags = flags, operand = 0, info = info, typeId = typeId)
proc prepare*(dest: var PackedTree; source: PackedTree; sourcePos: NodePos): PatchPos =
result = PatchPos dest.nodes.len
dest.nodes.add source.nodes[sourcePos.int]
proc patch*(tree: var PackedTree; pos: PatchPos) =
let pos = pos.int
let k = tree.nodes[pos].kind
assert k > nkNilLit
let distance = int32(tree.nodes.len - pos)
assert distance > 0
tree.nodes[pos].x = toX(k, cast[uint32](distance))
proc len*(tree: PackedTree): int {.inline.} = tree.nodes.len
proc `[]`*(tree: PackedTree; i: NodePos): lent PackedNode {.inline.} =
tree.nodes[i.int]
template rawSpan(n: PackedNode): int = int(uoperand(n))
proc nextChild(tree: PackedTree; pos: var int) {.inline.} =
if tree.nodes[pos].kind > nkNilLit:
assert tree.nodes[pos].uoperand > 0
inc pos, tree.nodes[pos].rawSpan
else:
inc pos
iterator sonsReadonly*(tree: PackedTree; n: NodePos): NodePos =
var pos = n.int
assert tree.nodes[pos].kind > nkNilLit
let last = pos + tree.nodes[pos].rawSpan
inc pos
while pos < last:
yield NodePos pos
nextChild tree, pos
iterator sons*(dest: var PackedTree; tree: PackedTree; n: NodePos): NodePos =
let patchPos = prepare(dest, tree, n)
for x in sonsReadonly(tree, n): yield x
patch dest, patchPos
iterator isons*(dest: var PackedTree; tree: PackedTree;
n: NodePos): (int, NodePos) =
var i = 0
for ch0 in sons(dest, tree, n):
yield (i, ch0)
inc i
iterator sonsFrom1*(tree: PackedTree; n: NodePos): NodePos =
var pos = n.int
assert tree.nodes[pos].kind > nkNilLit
let last = pos + tree.nodes[pos].rawSpan
inc pos
if pos < last:
nextChild tree, pos
while pos < last:
yield NodePos pos
nextChild tree, pos
iterator sonsWithoutLast2*(tree: PackedTree; n: NodePos): NodePos =
var count = 0
for child in sonsReadonly(tree, n):
inc count
var pos = n.int
assert tree.nodes[pos].kind > nkNilLit
let last = pos + tree.nodes[pos].rawSpan
inc pos
while pos < last and count > 2:
yield NodePos pos
dec count
nextChild tree, pos
proc parentImpl(tree: PackedTree; n: NodePos): NodePos =
# finding the parent of a node is rather easy:
var pos = n.int - 1
while pos >= 0 and (isAtom(tree, pos) or (pos + tree.nodes[pos].rawSpan - 1 < n.int)):
dec pos
#assert pos >= 0, "node has no parent"
result = NodePos(pos)
template parent*(n: NodePos): NodePos = parentImpl(tree, n)
proc hasXsons*(tree: PackedTree; n: NodePos; x: int): bool =
var count = 0
if tree.nodes[n.int].kind > nkNilLit:
for child in sonsReadonly(tree, n): inc count
result = count == x
proc hasAtLeastXsons*(tree: PackedTree; n: NodePos; x: int): bool =
if tree.nodes[n.int].kind > nkNilLit:
var count = 0
for child in sonsReadonly(tree, n):
inc count
if count >= x: return true
return false
proc firstSon*(tree: PackedTree; n: NodePos): NodePos {.inline.} =
NodePos(n.int+1)
proc kind*(tree: PackedTree; n: NodePos): TNodeKind {.inline.} =
tree.nodes[n.int].kind
proc litId*(tree: PackedTree; n: NodePos): LitId {.inline.} =
LitId tree.nodes[n.int].uoperand
proc info*(tree: PackedTree; n: NodePos): PackedLineInfo {.inline.} =
tree.nodes[n.int].info
proc findType*(tree: PackedTree; n: NodePos): PackedItemId =
for x in tree.withTypes:
if x[0] == int32(n): return x[1]
if x[0] > int32(n): return nilItemId
return nilItemId
proc findFlags*(tree: PackedTree; n: NodePos): TNodeFlags =
for x in tree.withFlags:
if x[0] == int32(n): return x[1]
if x[0] > int32(n): return {}
return {}
template typ*(n: NodePos): PackedItemId =
tree.findType(n)
template flags*(n: NodePos): TNodeFlags =
tree.findFlags(n)
template uoperand*(n: NodePos): uint32 =
tree.nodes[n.int].uoperand
proc span*(tree: PackedTree; pos: int): int {.inline.} =
if isAtom(tree, pos): 1 else: tree.nodes[pos].rawSpan
proc sons2*(tree: PackedTree; n: NodePos): (NodePos, NodePos) =
assert(not isAtom(tree, n.int))
let a = n.int+1
let b = a + span(tree, a)
result = (NodePos a, NodePos b)
proc sons3*(tree: PackedTree; n: NodePos): (NodePos, NodePos, NodePos) =
assert(not isAtom(tree, n.int))
let a = n.int+1
let b = a + span(tree, a)
let c = b + span(tree, b)
result = (NodePos a, NodePos b, NodePos c)
proc ithSon*(tree: PackedTree; n: NodePos; i: int): NodePos =
result = default(NodePos)
if tree.nodes[n.int].kind > nkNilLit:
var count = 0
for child in sonsReadonly(tree, n):
if count == i: return child
inc count
assert false, "node has no i-th child"
when false:
proc `@`*(tree: PackedTree; lit: LitId): lent string {.inline.} =
tree.sh.strings[lit]
template kind*(n: NodePos): TNodeKind = tree.nodes[n.int].kind
template info*(n: NodePos): PackedLineInfo = tree.nodes[n.int].info
template litId*(n: NodePos): LitId = LitId tree.nodes[n.int].uoperand
template symId*(n: NodePos): SymId = SymId tree.nodes[n.int].soperand
proc firstSon*(n: NodePos): NodePos {.inline.} = NodePos(n.int+1)
const
externIntLit* = {nkCharLit,
nkIntLit,
nkInt8Lit,
nkInt16Lit,
nkInt32Lit,
nkInt64Lit,
nkUIntLit,
nkUInt8Lit,
nkUInt16Lit,
nkUInt32Lit,
nkUInt64Lit}
externSIntLit* = {nkIntLit, nkInt8Lit, nkInt16Lit, nkInt32Lit, nkInt64Lit}
externUIntLit* = {nkUIntLit, nkUInt8Lit, nkUInt16Lit, nkUInt32Lit, nkUInt64Lit}
directIntLit* = nkNone
template copyInto*(dest, n, body) =
let patchPos = prepare(dest, tree, n)
body
patch dest, patchPos
template copyIntoKind*(dest, kind, info, body) =
let patchPos = prepare(dest, kind, info)
body
patch dest, patchPos
proc getNodeId*(tree: PackedTree): NodeId {.inline.} = NodeId tree.nodes.len
iterator allNodes*(tree: PackedTree): NodePos =
var p = 0
while p < tree.len:
yield NodePos(p)
let s = span(tree, p)
inc p, s
proc toPackedItemId*(item: int32): PackedItemId {.inline.} =
PackedItemId(module: LitId(0), item: item)
proc load*(f: var RodFile; t: var PackedTree) =
loadSeq f, t.nodes
loadSeq f, t.withFlags
loadSeq f, t.withTypes
proc store*(f: var RodFile; t: PackedTree) =
storeSeq f, t.nodes
storeSeq f, t.withFlags
storeSeq f, t.withTypes

View File

@@ -19,10 +19,11 @@ import std/tables
when defined(nimPreviewSlimSystem):
import std/assertions
import packed_ast, ic, bitabs
proc replayStateChanges*(module: PSym; g: ModuleGraph) =
let list = module.ast
proc replayStateChanges*(module: PSym; g: ModuleGraph; list: PNode) =
## `list` is an `nkStmtList` of `nkReplayAction` nodes (macro-cache puts/incs/
## adds/incls and a few pragmas) recorded for `module`. Under the NIF backend a
## loaded module's `ast` is never reconstructed, so the caller passes the replay
## actions it parsed out of the module's NIF directly.
assert list != nil
assert list.kind == nkStmtList
for n in list:
@@ -66,8 +67,9 @@ proc replayStateChanges*(module: PSym; g: ModuleGraph) =
g.cacheTables[destKey] = initBTree[string, PNode]()
if not contains(g.cacheTables[destKey], key):
g.cacheTables[destKey].add(key, val)
else:
internalError(g.config, n.info, "key already exists: " & key)
# else: the same key was already replayed. Under IC the import closure is
# replayed (direct module + transitive deps), so the same registration can
# legitimately be reached twice; re-applying it is a no-op, not an error.
of "incl":
let destKey = n[1].strVal
let val = n[2]
@@ -89,83 +91,36 @@ proc replayStateChanges*(module: PSym; g: ModuleGraph) =
else:
internalAssert g.config, false
proc replayBackendProcs*(g: ModuleGraph; module: int) =
for it in mitems(g.packed[module].fromDisk.attachedOps):
let key = translateId(it[0], g.packed, module, g.config)
let op = it[1]
let tmp = translateId(it[2], g.packed, module, g.config)
let symId = FullId(module: tmp.module, packed: it[2])
g.attachedOps[op][key] = LazySym(id: symId, sym: nil)
for it in mitems(g.packed[module].fromDisk.enumToStringProcs):
let key = translateId(it[0], g.packed, module, g.config)
let tmp = translateId(it[1], g.packed, module, g.config)
let symId = FullId(module: tmp.module, packed: it[1])
g.enumToStringProcs[key] = LazySym(id: symId, sym: nil)
for it in mitems(g.packed[module].fromDisk.methodsPerType):
let key = translateId(it[0], g.packed, module, g.config)
let tmp = translateId(it[1], g.packed, module, g.config)
let symId = FullId(module: tmp.module, packed: it[1])
g.methodsPerType.mgetOrPut(key, @[]).add LazySym(id: symId, sym: nil)
for it in mitems(g.packed[module].fromDisk.dispatchers):
let tmp = translateId(it, g.packed, module, g.config)
let symId = FullId(module: tmp.module, packed: it)
g.dispatchers.add LazySym(id: symId, sym: nil)
proc replayGenericCacheInformation*(g: ModuleGraph; module: int) =
## We remember the generic instantiations a module performed
## in order to to avoid the code bloat that generic code tends
## to imply. This is cheaper than deduplication of identical
## generic instantiations. However, deduplication is more
## powerful and general and I hope to implement it soon too
## (famous last words).
assert g.packed[module].status == loaded
for it in g.packed[module].fromDisk.typeInstCache:
let key = translateId(it[0], g.packed, module, g.config)
g.typeInstCache.mgetOrPut(key, @[]).add LazyType(id: FullId(module: module, packed: it[1]), typ: nil)
for it in mitems(g.packed[module].fromDisk.procInstCache):
let key = translateId(it.key, g.packed, module, g.config)
let sym = translateId(it.sym, g.packed, module, g.config)
var concreteTypes = newSeq[FullId](it.concreteTypes.len)
for i in 0..high(it.concreteTypes):
let tmp = translateId(it.concreteTypes[i], g.packed, module, g.config)
concreteTypes[i] = FullId(module: tmp.module, packed: it.concreteTypes[i])
g.procInstCache.mgetOrPut(key, @[]).add LazyInstantiation(
module: module, sym: FullId(module: sym.module, packed: it.sym),
concreteTypes: concreteTypes, inst: nil)
for it in mitems(g.packed[module].fromDisk.methodsPerGenericType):
let key = translateId(it[0], g.packed, module, g.config)
let col = it[1]
let tmp = translateId(it[2], g.packed, module, g.config)
let symId = FullId(module: tmp.module, packed: it[2])
g.methodsPerGenericType.mgetOrPut(key, @[]).add (col, LazySym(id: symId, sym: nil))
replayBackendProcs(g, module)
for it in mitems(g.packed[module].fromDisk.methods):
let sym = loadSymFromId(g.config, g.cache, g.packed, module,
PackedItemId(module: LitId(0), item: it))
methodDef(g, g.idgen, sym)
when false:
# not used anymore:
for it in mitems(g.packed[module].fromDisk.compilerProcs):
let symId = FullId(module: module, packed: PackedItemId(module: LitId(0), item: it[1]))
g.lazyCompilerprocs[g.packed[module].fromDisk.sh.strings[it[0]]] = symId
for it in mitems(g.packed[module].fromDisk.converters):
let symId = FullId(module: module, packed: PackedItemId(module: LitId(0), item: it))
g.ifaces[module].converters.add LazySym(id: symId, sym: nil)
for it in mitems(g.packed[module].fromDisk.trmacros):
let symId = FullId(module: module, packed: PackedItemId(module: LitId(0), item: it))
g.ifaces[module].patterns.add LazySym(id: symId, sym: nil)
for it in mitems(g.packed[module].fromDisk.pureEnums):
let symId = FullId(module: module, packed: PackedItemId(module: LitId(0), item: it))
g.ifaces[module].pureEnums.add LazySym(id: symId, sym: nil)
proc replayBackendActions*(g: ModuleGraph; module: PSym; list: PNode) =
## Applies the backend-relevant replay actions (C compile/link directives)
## found in a NIF-loaded module's top-level statement list. The `nifc`
## backend loads modules without going through sem's `replayStateChanges`,
## so e.g. math's `{.passL: "-lm".}` was lost and the final link failed
## with undefined references. VM cache actions are deliberately NOT
## replayed here — codegen does not run macros.
if list == nil: return
for n in list:
if n.kind == nkReplayAction and n.len >= 2 and
n[0].kind == nkStrLit and n[1].kind == nkStrLit:
case n[0].strVal
of "compile":
if n.len == 4 and n[2].kind == nkStrLit:
let cname = AbsoluteFile n[1].strVal
var cf = Cfile(nimname: splitFile(cname).name, cname: cname,
obj: AbsoluteFile n[2].strVal,
flags: {CfileFlag.External},
customArgs: n[3].strVal)
extccomp.addExternalFileToCompile(g.config, cf)
of "link":
extccomp.addExternalFileToLink(g.config, AbsoluteFile n[1].strVal)
of "passl":
extccomp.addLinkOption(g.config, n[1].strVal)
of "passc":
extccomp.addCompileOption(g.config, n[1].strVal)
of "localpassc":
extccomp.addLocalCompileOption(g.config, n[1].strVal,
toFullPathConsiderDirty(g.config, module.info.fileIndex))
of "cppdefine":
options.cppDefine(g.config, n[1].strVal)
else:
discard

View File

@@ -1,283 +0,0 @@
#
#
# The Nim Compiler
# (c) Copyright 2020 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## Low level binary format used by the compiler to store and load various AST
## and related data.
##
## NB: this is incredibly low level and if you're interested in how the
## compiler works and less a storage format, you're probably looking for
## the `ic` or `packed_ast` modules to understand the logical format.
from std/typetraits import supportsCopyMem
when defined(nimPreviewSlimSystem):
import std/[syncio, assertions]
import std / tables
## Overview
## ========
## `RodFile` represents a Rod File (versioned binary format), and the
## associated data for common interactions such as IO and error tracking
## (`RodFileError`). The file format broken up into sections (`RodSection`)
## and preceded by a header (see: `cookie`). The precise layout, section
## ordering and data following the section are determined by the user. See
## `ic.loadRodFile`.
##
## A basic but "wrong" example of the lifecycle:
## ---------------------------------------------
## 1. `create` or `open` - create a new one or open an existing
## 2. `storeHeader` - header info
## 3. `storePrim` or `storeSeq` - save your stuff
## 4. `close` - and we're done
##
## Now read the bits below to understand what's missing.
##
## ### Issues with the Example
## Missing Sections:
## This is a low level API, so headers and sections need to be stored and
## loaded by the user, see `storeHeader` & `loadHeader` and `storeSection` &
## `loadSection`, respectively.
##
## No Error Handling:
## The API is centered around IO and prone to error, each operation checks or
## sets the `RodFile.err` field. A user of this API needs to handle these
## appropriately.
##
## API Notes
## =========
##
## Valid inputs for Rod files
## --------------------------
## ASTs, hopes, dreams, and anything as long as it and any children it may have
## support `copyMem`. This means anything that is not a pointer and that does not contain a pointer. At a glance these are:
## * string
## * objects & tuples (fields are recursed)
## * sequences AKA `seq[T]`
##
## Note on error handling style
## ----------------------------
## A flag based approach is used where operations no-op in case of a
## preexisting error and set the flag if they encounter one.
##
## Misc
## ----
## * 'Prim' is short for 'primitive', as in a non-sequence type
type
RodSection* = enum
versionSection
configSection
stringsSection
checkSumsSection
depsSection
numbersSection
exportsSection
hiddenSection
reexportsSection
compilerProcsSection
trmacrosSection
convertersSection
methodsSection
pureEnumsSection
toReplaySection
topLevelSection
bodiesSection
symsSection
typesSection
typeInstCacheSection
procInstCacheSection
attachedOpsSection
methodsPerGenericTypeSection
enumToStringProcsSection
methodsPerTypeSection
dispatchersSection
typeInfoSection # required by the backend
backendFlagsSection
aliveSymsSection # beware, this is stored in a `.alivesyms` file.
sideChannelSection
namespaceSection
symnamesSection
RodFileError* = enum
ok, tooBig, cannotOpen, ioFailure, wrongHeader, wrongSection, configMismatch,
includeFileChanged
RodFile* = object
f*: File
currentSection*: RodSection # for error checking
err*: RodFileError # little experiment to see if this works
# better than exceptions.
const
RodVersion = 2
defaultCookie = [byte(0), byte('R'), byte('O'), byte('D'),
byte(sizeof(int)*8), byte(system.cpuEndian), byte(0), byte(RodVersion)]
proc setError(f: var RodFile; err: RodFileError) {.inline.} =
f.err = err
#raise newException(IOError, "IO error")
proc storePrim*(f: var RodFile; s: string) =
## Stores a string.
## The len is prefixed to allow for later retreival.
if f.err != ok: return
if s.len >= high(int32):
setError f, tooBig
return
var lenPrefix = int32(s.len)
if writeBuffer(f.f, addr lenPrefix, sizeof(lenPrefix)) != sizeof(lenPrefix):
setError f, ioFailure
else:
if s.len != 0:
if writeBuffer(f.f, unsafeAddr(s[0]), s.len) != s.len:
setError f, ioFailure
proc storePrim*[T](f: var RodFile; x: T) =
## Stores a non-sequence/string `T`.
## If `T` doesn't support `copyMem` and is an object or tuple then the fields
## are written -- the user from context will need to know which `T` to load.
if f.err != ok: return
when supportsCopyMem(T):
if writeBuffer(f.f, unsafeAddr(x), sizeof(x)) != sizeof(x):
setError f, ioFailure
elif T is tuple:
for y in fields(x):
storePrim(f, y)
elif T is object:
for y in fields(x):
when y is seq:
storeSeq(f, y)
else:
storePrim(f, y)
else:
{.error: "unsupported type for 'storePrim'".}
proc storeSeq*[T](f: var RodFile; s: seq[T]) =
## Stores a sequence of `T`s, with the len as a prefix for later retrieval.
if f.err != ok: return
if s.len >= high(int32):
setError f, tooBig
return
var lenPrefix = int32(s.len)
if writeBuffer(f.f, addr lenPrefix, sizeof(lenPrefix)) != sizeof(lenPrefix):
setError f, ioFailure
else:
for i in 0..<s.len:
storePrim(f, s[i])
proc storeOrderedTable*[K, T](f: var RodFile; s: OrderedTable[K, T]) =
if f.err != ok: return
if s.len >= high(int32):
setError f, tooBig
return
var lenPrefix = int32(s.len)
if writeBuffer(f.f, addr lenPrefix, sizeof(lenPrefix)) != sizeof(lenPrefix):
setError f, ioFailure
else:
for _, v in s:
storePrim(f, v)
proc loadPrim*(f: var RodFile; s: var string) =
## Read a string, the length was stored as a prefix
if f.err != ok: return
var lenPrefix = int32(0)
if readBuffer(f.f, addr lenPrefix, sizeof(lenPrefix)) != sizeof(lenPrefix):
setError f, ioFailure
else:
s = newString(lenPrefix)
if lenPrefix > 0:
if readBuffer(f.f, unsafeAddr(s[0]), s.len) != s.len:
setError f, ioFailure
proc loadPrim*[T](f: var RodFile; x: var T) =
## Load a non-sequence/string `T`.
if f.err != ok: return
when supportsCopyMem(T):
if readBuffer(f.f, unsafeAddr(x), sizeof(x)) != sizeof(x):
setError f, ioFailure
elif T is tuple:
for y in fields(x):
loadPrim(f, y)
elif T is object:
for y in fields(x):
when y is seq:
loadSeq(f, y)
else:
loadPrim(f, y)
else:
{.error: "unsupported type for 'loadPrim'".}
proc loadSeq*[T](f: var RodFile; s: var seq[T]) =
## `T` must be compatible with `copyMem`, see `loadPrim`
if f.err != ok: return
var lenPrefix = int32(0)
if readBuffer(f.f, addr lenPrefix, sizeof(lenPrefix)) != sizeof(lenPrefix):
setError f, ioFailure
else:
s = newSeq[T](lenPrefix)
for i in 0..<lenPrefix:
loadPrim(f, s[i])
proc loadOrderedTable*[K, T](f: var RodFile; s: var OrderedTable[K, T]) =
## `T` must be compatible with `copyMem`, see `loadPrim`
if f.err != ok: return
var lenPrefix = int32(0)
if readBuffer(f.f, addr lenPrefix, sizeof(lenPrefix)) != sizeof(lenPrefix):
setError f, ioFailure
else:
s = initOrderedTable[K, T](lenPrefix)
for i in 0..<lenPrefix:
var x = default T
loadPrim(f, x)
s[x.id] = x
proc storeHeader*(f: var RodFile; cookie = defaultCookie) =
## stores the header which is described by `cookie`.
if f.err != ok: return
if f.f.writeBytes(cookie, 0, cookie.len) != cookie.len:
setError f, ioFailure
proc loadHeader*(f: var RodFile; cookie = defaultCookie) =
## Loads the header which is described by `cookie`.
if f.err != ok: return
var thisCookie: array[cookie.len, byte] = default(array[cookie.len, byte])
if f.f.readBytes(thisCookie, 0, thisCookie.len) != thisCookie.len:
setError f, ioFailure
elif thisCookie != cookie:
setError f, wrongHeader
proc storeSection*(f: var RodFile; s: RodSection) =
## update `currentSection` and writes the bytes value of s.
if f.err != ok: return
assert f.currentSection < s
f.currentSection = s
storePrim(f, s)
proc loadSection*(f: var RodFile; expected: RodSection) =
## read the bytes value of s, sets and error if the section is incorrect.
if f.err != ok: return
var s: RodSection = default(RodSection)
loadPrim(f, s)
if expected != s and f.err == ok:
setError f, wrongSection
proc create*(filename: string): RodFile =
## create the file and open it for writing
result = default(RodFile)
if not open(result.f, filename, fmWrite):
setError result, cannotOpen
proc close*(f: var RodFile) = close(f.f)
proc open*(filename: string): RodFile =
## open the file for reading
result = default(RodFile)
if not open(result.f, filename, fmRead):
setError result, cannotOpen

283
compiler/icconfig.nim Normal file
View File

@@ -0,0 +1,283 @@
#
#
# The Nim Compiler
# (c) Copyright 2026 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## Precompiled config for the incremental compiler (`nim ic`).
##
## `nim ic` builds the program by spawning one `nim m` child per module (or
## strongly-connected import group) plus a final `nim nifc`. Each child is a
## full Nim process, so each would normally re-read the whole `nim.cfg` chain
## *and* re-run `config.nims` through the VM — work that is identical for every
## child and, because of the VM run, far from free. With ~85 modules in the
## compiler itself that config work is paid ~85 times during `koch bootic`.
##
## The fix mirrors Nimony's `.cfg.nif`: the driver parses config once, records
## the net effect, and the children replay it. Every config-file switch funnels
## through `processSwitch(..., passPP, ...)` (`nimconf.parseAssignment` and the
## `switch()` callback in `scriptconfig`), so the recorded sequence of those
## switches, replayed in order, reproduces an identical `ConfigRef` without any
## file read or VM run. The one config side effect that does not go through
## `processSwitch` is `cppDefine` (it mutates `conf.cppDefines` directly), so the
## resolved set is serialised alongside.
##
## Path-search switches are deliberately excluded from the recording (see
## `commands.processSwitch`): their resolved result already lives in
## `conf.searchPaths`, which the driver forwards to every child as absolute
## `--path` arguments; replaying their raw, config-dir-relative arguments here
## would misresolve.
import options, commands, lineinfos, pathutils, msgs
import std/[algorithm, os, sets, osproc, times, streams, syncio]
import "../dist/nimony/src/lib" / [nifbuilder, nifcoreparse]
const
IcConfigVersion* = "2"
## Artifact format version. Bump on any layout change here so a child built
## by an older compiler rejects a stale artifact and falls back to normal
## config loading instead of replaying a format it cannot parse.
proc writeIcConfig*(conf: ConfigRef; outfile: string) =
## Serialise the resolved config (the config-file switches recorded during
## `loadConfigs`, the resolved `cppDefines`/`searchPaths`, the nimcache dir, and
## the list of config *source* files for staleness detection) into `outfile`.
## `OnlyIfChanged`: when the content is byte-identical to what is already on
## disk the file is left untouched so its mtime does not advance — otherwise
## every `nim ic` run would re-fire the whole nifmake graph (see `nifler`'s
## `produceConfig`, whose model this mirrors).
var b = nifbuilder.open(outfile, writeMode = OnlyIfChanged)
b.withTree "stmts":
b.withTree "meta":
b.addStrLit IcConfigVersion
b.withTree "sources":
# Every config file read while loading (nim.cfg chain + config.nims), so a
# later run can decide via mtimes whether this artifact is still current
# (see `sourcesChanged`).
for f in conf.configFiles:
b.addStrLit f.string
b.withTree "nimcache":
# Resolved build nimcache. Recorded (unlike the path-search switches) so the
# driver, which replays this artifact instead of parsing `nim.cfg`, still
# learns a `--nimcache:` set inside `nim.cfg` and builds in the right place.
b.addStrLit conf.nimcacheDir.string
b.withTree "cppdefines":
# HashSet iteration order is unspecified; sort so the artifact is
# byte-stable across runs (nifmake keys rebuilds off content changes).
var defs: seq[string] = @[]
for d in conf.cppDefines: defs.add d
sort defs
for d in defs: b.addStrLit d
b.withTree "searchpaths":
# The resolved (absolute) search paths. Path-search *switches* are skipped
# below because their raw arguments are config-dir-relative; the net effect
# lives here instead, so a replayer with no `--path` command-line arguments
# (the `nim ic` driver itself) still resolves imports. `nim m`/`nim nifc`
# children also receive these as forwarded `--path` args; the dedup on
# replay makes the overlap harmless.
for p in conf.searchPaths:
b.addStrLit p.string
b.withTree "switches":
for sw in conf.icConfigSwitches:
b.addTree "sw"
b.addStrLit sw.switch
b.addStrLit sw.arg
b.endTree()
b.close()
proc applyIcConfig*(conf: ConfigRef; infile: string): bool =
## Replay the precompiled config into `conf`. Returns false (and applies
## nothing meaningful) when the artifact is missing or written by a compiler
## with an incompatible format version, so the caller can fall back to reading
## the config files normally.
if not fileExists(infile): return false
var pool = newPool()
var tags = newTagPool()
let
stmtsTag = tags.registerTag("stmts")
metaTag = tags.registerTag("meta")
sourcesTag = tags.registerTag("sources")
nimcacheTag = tags.registerTag("nimcache")
cppTag = tags.registerTag("cppdefines")
pathsTag = tags.registerTag("searchpaths")
switchesTag = tags.registerTag("switches")
swTag = tags.registerTag("sw")
var buf = parseFromFile(infile, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != stmtsTag:
endRead(c)
return false
var version = ""
var sawMeta = false
let info = unknownLineInfo
c.loopInto:
if c.kind == TagLit:
if c.cursorTagId == metaTag:
sawMeta = true
c.loopInto:
if c.kind == StrLit:
version = strVal(c)
inc c
else:
skip c
elif c.cursorTagId == nimcacheTag:
c.loopInto:
if c.kind == StrLit:
let nc = strVal(c)
# Only when nimcache was not already pinned on the command line: a
# `--nimcache:` argument the driver/child was launched with must win
# over whatever `nim.cfg` recorded into the artifact.
if nc.len > 0 and conf.nimcacheDir.isEmpty:
conf.nimcacheDir = AbsoluteDir(nc)
inc c
else:
skip c
elif c.cursorTagId == sourcesTag:
# Replay does not need the source list; it exists only for
# `sourcesChanged`. Skip the whole section.
skip c
elif c.cursorTagId == cppTag:
c.loopInto:
if c.kind == StrLit:
cppDefine(conf, strVal(c))
inc c
else:
skip c
elif c.cursorTagId == pathsTag:
c.loopInto:
if c.kind == StrLit:
# Append preserving the serialised order (which already reflects the
# driver's addPath insert-at-front sequence), deduping against any
# path a child already received via a forwarded `--path` argument.
let d = AbsoluteDir(strVal(c))
if not conf.searchPaths.contains(d): conf.searchPaths.add d
inc c
else:
skip c
elif c.cursorTagId == switchesTag:
c.loopInto:
if c.kind == TagLit and c.cursorTagId == swTag:
var sw = ""
var arg = ""
var idx = 0
c.loopInto:
if c.kind == StrLit:
if idx == 0: sw = strVal(c)
else: arg = strVal(c)
inc idx
inc c
else:
skip c
processSwitch(sw, arg, passPP, info, conf)
else:
skip c
else:
skip c
else:
skip c
endRead(c)
result = sawMeta and version == IcConfigVersion
proc sourcesChanged*(configFile: string): bool =
## True when the precompiled config at `configFile` is missing, malformed,
## written by an incompatible version, or any recorded config *source* file is
## newer than it (or has vanished) — i.e. the artifact must be regenerated.
## Mirrors nifler's `sourcesChanged`: the source list lives inside the artifact
## so this needs no out-of-band knowledge of which `nim.cfg`s were read.
if not fileExists(configFile): return true
let modtime = getLastModificationTime(configFile)
var pool = newPool()
var tags = newTagPool()
let
stmtsTag = tags.registerTag("stmts")
metaTag = tags.registerTag("meta")
sourcesTag = tags.registerTag("sources")
var buf = parseFromFile(configFile, 1000, pool, tags)
var c = beginRead(buf)
if c.kind != TagLit or c.cursorTagId != stmtsTag:
endRead(c)
return true
var version = ""
var depsChanged = false
c.loopInto:
if c.kind == TagLit and c.cursorTagId == metaTag:
c.loopInto:
if c.kind == StrLit:
version = strVal(c)
inc c
else:
skip c
elif c.kind == TagLit and c.cursorTagId == sourcesTag:
c.loopInto:
if c.kind == StrLit:
let dep = strVal(c)
if not fileExists(dep) or getLastModificationTime(dep) >= modtime:
depsChanged = true
inc c
else:
skip c
else:
skip c
endRead(c)
result = depsChanged or version != IcConfigVersion
proc produceIcConfig*(conf: ConfigRef) =
## The `cmdIcConfig` command. By the time it runs, the normal pipeline has
## already fully parsed the `nim.cfg` chain and run `config.nims`, so the
## resolved config is sitting in `conf`; just serialise it to `--o`.
let outPath = conf.icConfigOut
if outPath.len == 0:
rawMessage(conf, errGenerated, "icconfig: missing output path (--icConfigOut)")
return
createDir(parentDir(outPath))
writeIcConfig(conf, outPath)
proc ensureIcConfig*(conf: ConfigRef) =
## Driver-side (`cmdIc`). Make sure an up-to-date precompiled config exists,
## (re)producing it in a *separate* process when missing or stale, then point
## `conf.icPreparsedConfig` at it so the driver replays the very same config its
## `nim m`/`nim nifc` children will — perfect speed (config parsed at most once,
## skipped entirely when nothing changed) and consistency (one producer, every
## process replays its output). The artifact lives in the nimcache derived from
## the command line (pre-config-parse), which is the one the children are told;
## a `--nimcache:` set inside `nim.cfg` is recovered from the artifact itself.
let cacheDir = getNimcacheDir(conf).string
# Start from a clean cache when the on-disk NIF format stamp is absent or stale
# (see `icFormatVersion`). This must happen HERE, before the config artifact is
# produced — `commandIc` performs the same check later, but by then the artifact
# would already live in the cache and the wipe would delete it.
createDir(cacheDir)
let versionFile = cacheDir / "ic.version"
let stamp = if fileExists(versionFile): readFile(versionFile) else: ""
if stamp != icFormatVersion:
removeDir(cacheDir)
createDir(cacheDir)
writeFile(versionFile, icFormatVersion)
let outPath = cacheDir / "ic_config.cfg.nif"
if not fileExists(outPath) or sourcesChanged(outPath):
createDir(cacheDir)
# Re-invoke ourselves as the config producer: reuse this process's command
# line, dropping the command argument (`ic`) in favour of `icconfig` and the
# explicit output path, both BEFORE the project file (anything after the
# project is swallowed into `config.arguments` by `cmdLineRest`). The
# producer re-reads `nim.cfg` itself.
var pargs = @["icconfig", "--icConfigOut:" & outPath]
var droppedCmd = false
for a in commandLineParams():
if not droppedCmd and a.len > 0 and a[0] != '-':
droppedCmd = true # drop the original command token (`ic`)
else:
pargs.add a
let p = startProcess(getAppFilename(), args = pargs,
options = {poStdErrToStdOut})
let outp = p.outputStream.readAll()
let code = p.waitForExit()
p.close()
if code != 0 or not fileExists(outPath):
rawMessage(conf, errGenerated,
"failed to produce precompiled config (exit code " & $code & "):\n" & outp)
return
conf.icPreparsedConfig = outPath

View File

@@ -10,10 +10,10 @@
## This module implements the symbol importing mechanism.
import
ast, astalgo, msgs, options, idents, lookups,
ast, msgs, options, idents, lookups,
semdata, modulepaths, sigmatch, lineinfos,
modulegraphs, wordrecg
from std/strutils import `%`, startsWith
from std/strutils import `%`, startsWith, replace
from std/sequtils import addUnique
import std/[sets, tables, intsets]
@@ -108,8 +108,8 @@ proc rawImportSymbol(c: PContext, s, origin: PSym; importSet: var IntSet) =
else:
importPureEnumField(c, e)
else:
if s.kind == skConverter: addConverter(c, LazySym(sym: s))
if hasPattern(s): addPattern(c, LazySym(sym: s))
if s.kind == skConverter: addConverter(c, s)
if hasPattern(s): addPattern(c, s)
if s.owner != origin:
c.exportIndirections.incl((origin.id, s.id))
@@ -190,22 +190,19 @@ proc addImport(c: PContext; im: sink ImportedModule) =
template addUnnamedIt(c: PContext, fromMod: PSym; filter: untyped) {.dirty.} =
for it in mitems c.graph.ifaces[fromMod.position].converters:
if filter:
loadPackedSym(c.graph, it)
if sfExported in it.sym.flags:
if sfExported in it.flags:
addConverter(c, it)
for it in mitems c.graph.ifaces[fromMod.position].patterns:
if filter:
loadPackedSym(c.graph, it)
if sfExported in it.sym.flags:
if sfExported in it.flags:
addPattern(c, it)
for it in mitems c.graph.ifaces[fromMod.position].pureEnums:
if filter:
loadPackedSym(c.graph, it)
importPureEnumFields(c, it.sym, it.sym.typ)
importPureEnumFields(c, it, it.typ)
proc importAllSymbolsExcept(c: PContext, fromMod: PSym, exceptSet: IntSet) =
c.addImport ImportedModule(m: fromMod, mode: importExcept, exceptSet: exceptSet)
addUnnamedIt(c, fromMod, it.sym.name.id notin exceptSet)
addUnnamedIt(c, fromMod, it.name.id notin exceptSet)
proc importAllSymbols*(c: PContext, fromMod: PSym) =
c.addImport ImportedModule(m: fromMod, mode: importAll)
@@ -231,7 +228,7 @@ proc importForwarded(c: PContext, n: PNode, exceptSet: IntSet; fromMod: PSym; im
for i in 0..n.safeLen-1:
importForwarded(c, n[i], exceptSet, fromMod, importSet)
proc importModuleAs(c: PContext; n: PNode, realModule: PSym, importHidden: bool): PSym =
proc importModuleAs(c: PContext; n: PNode, realModule: PSym, importHidden, trackUnusedImport: bool): PSym =
result = realModule
template createModuleAliasImpl(ident): untyped =
createModuleAlias(realModule, c.idgen, ident, n.info, c.config.options)
@@ -245,9 +242,12 @@ proc importModuleAs(c: PContext; n: PNode, realModule: PSym, importHidden: bool)
# avoids modifying `realModule`, see D20201209T194412 for `import {.all.}`
result = createModuleAliasImpl(realModule.name)
if importHidden:
result.options.incl optImportHidden
let moduleIdent = if n.kind == nkInfix: n[^1] else: n
c.unusedImports.add((result, moduleIdent.info))
ensureMutable result
result.optionsImpl.incl optImportHidden
let moduleIdent = if n.kind in {nkInfix, nkImportAs}: n[^1] else: n
result.info = moduleIdent.info
if trackUnusedImport:
c.unusedImports.add((result, result.info))
c.importModuleMap[result.id] = realModule.id
c.importModuleLookup.mgetOrPut(result.name.id, @[]).addUnique realModule.id
@@ -288,10 +288,10 @@ proc myImportModule(c: PContext, n: var PNode, importStmtResult: PNode): PSym =
toFullPath(c.config, c.graph.importStack[i+1])
c.recursiveDep = err
var realModule: PSym
let trackUnusedImport = warnUnusedImportX in c.config.notes
discard pushOptionEntry(c)
realModule = c.graph.importModuleCallback(c.graph, c.module, f)
result = importModuleAs(c, n, realModule, transf.importHidden)
let realModule = c.graph.importModuleCallback(c.graph, c.module, f)
result = importModuleAs(c, n, realModule, transf.importHidden, trackUnusedImport)
popOptionEntry(c)
#echo "set back to ", L
@@ -304,9 +304,9 @@ proc myImportModule(c: PContext, n: var PNode, importStmtResult: PNode): PSym =
var prefix = ""
if realModule.constraint != nil: prefix = realModule.constraint.strVal & "; "
message(c.config, n.info, warnDeprecated, prefix & realModule.name.s & " is deprecated")
let moduleName = getModuleName(c.config, n)
if belongsToStdlib(c.graph, result) and not startsWith(moduleName, stdPrefix) and
not startsWith(moduleName, "system/") and not startsWith(moduleName, "packages/"):
let moduleNameNorm = getModuleName(c.config, n).replace("\\", "/")
if belongsToStdlib(c.graph, result) and not startsWith(moduleNameNorm, stdPrefix) and
not startsWith(moduleNameNorm, "system/") and not startsWith(moduleNameNorm, "packages/"):
message(c.config, n.info, warnStdPrefix, realModule.name.s)
proc suggestMod(n: PNode; s: PSym) =
@@ -335,7 +335,7 @@ proc impMod(c: PContext; it: PNode; importStmtResult: PNode) =
let m = myImportModule(c, it, importStmtResult)
if m != nil:
# ``addDecl`` needs to be done before ``importAllSymbols``!
addDecl(c, m, it.info) # add symbol to symbol table of module
addDecl(c, m) # add symbol to symbol table of module
importAllSymbols(c, m)
#importForwarded(c, m.ast, emptySet, m)
afterImport(c, m)
@@ -372,7 +372,7 @@ proc evalFrom*(c: PContext, n: PNode): PNode =
var m = myImportModule(c, n[0], result)
if m != nil:
n[0] = newSymNode(m)
addDecl(c, m, n.info) # add symbol to symbol table of module
addDecl(c, m) # add symbol to symbol table of module
var im = ImportedModule(m: m, mode: importSet, imported: initIntSet())
for i in 1..<n.len:
@@ -387,7 +387,7 @@ proc evalImportExcept*(c: PContext, n: PNode): PNode =
var m = myImportModule(c, n[0], result)
if m != nil:
n[0] = newSymNode(m)
addDecl(c, m, n.info) # add symbol to symbol table of module
addDecl(c, m) # add symbol to symbol table of module
importAllSymbolsExcept(c, m, readExceptSet(c, n))
#importForwarded(c, m.ast, exceptSet, m)
afterImport(c, m)

View File

@@ -24,7 +24,7 @@ import std/[strtabs, tables, strutils, intsets]
when defined(nimPreviewSlimSystem):
import std/assertions
from trees import exprStructuralEquivalent, getRoot, whichPragma
from trees import exprStructuralEquivalent, getRoot, whichPragma, getPotentialWrites
type
Con = object
@@ -69,12 +69,14 @@ proc hasDestructor(c: Con; t: PType): bool {.inline.} =
result = ast.hasDestructor(t)
when toDebug.len > 0:
# for more effective debugging
if not result and c.graph.config.selectedGC in {gcArc, gcOrc, gcAtomicArc}:
if not result and c.graph.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc}:
assert(not containsGarbageCollectedRef(t))
proc getTemp(c: var Con; s: var Scope; typ: PType; info: TLineInfo): PNode =
proc getTemp(c: var Con; s: var Scope; typ: PType; info: TLineInfo; needsInit: bool): PNode =
let sym = newSym(skTemp, getIdent(c.graph.cache, ":tmpD"), c.idgen, c.owner, info)
sym.typ = typ
if not needsInit:
sym.incl sfNoInit
s.vars.add(sym)
result = newSymNode(sym)
@@ -100,6 +102,7 @@ when false:
proc isLastReadImpl(n: PNode; c: var Con; scope: var Scope): bool =
let root = parampatterns.exprRoot(n, allowCalls=false)
if root == nil: return false
elif sfSingleUsedTemp in root.flags: return true
var s = addr(scope)
while s != nil:
@@ -162,13 +165,16 @@ proc isLastReadImpl(n: PNode; c: var Con; scope: var Scope): bool =
else:
result = false
template hasDestructorOrAsgn(c: var Con, typ: PType): bool =
# bug #23354; an object type could have a non-trivial assignements when it is passed to a sink parameter
hasDestructor(c, typ) or (c.graph.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc} and
typ.kind == tyObject and not isTrivial(getAttachedOp(c.graph, typ, attachedAsgn)))
proc isLastRead(n: PNode; c: var Con; s: var Scope): bool =
# bug #23354; an object type could have a non-trival assignements when it is passed to a sink parameter
if not hasDestructor(c, n.typ) and (n.typ.kind != tyObject or isTrival(getAttachedOp(c.graph, n.typ, attachedAsgn))): return true
if not hasDestructorOrAsgn(c, n.typ): return true
let m = skipConvDfa(n)
result = (m.kind == nkSym and sfSingleUsedTemp in m.sym.flags) or
isLastReadImpl(n, c, s)
result = isLastReadImpl(n, c, s)
proc isFirstWrite(n: PNode; c: var Con): bool =
let m = skipConvDfa(n)
@@ -185,10 +191,11 @@ proc isCursor(n: PNode): bool =
else:
false
template isUnpackedTuple(n: PNode): bool =
template isFullyUnpackedTuple(n: PNode): bool =
## we move out all elements of unpacked tuples,
## hence unpacked tuples themselves don't need to be destroyed
## except it's already a cursor
## restricted to `skTemp`, tuple temps where not every field is unpacked should not use `skTemp`
(n.kind == nkSym and n.sym.kind == skTemp and
n.sym.typ.kind == tyTuple and sfCursor notin n.sym.flags)
@@ -197,7 +204,8 @@ proc checkForErrorPragma(c: Con; t: PType; ri: PNode; opname: string; inferredFr
if inferredFromCopy:
m.add ", which is inferred from unavailable '=copy'"
if (opname == "=" or opname == "=copy" or opname == "=dup") and ri != nil:
if (opname == "=" or opname == "=copy" or opname == "=dup") and
ri != nil:
m.add "; requires a copy because it's not the last read of '"
m.add renderTree(ri)
m.add '\''
@@ -247,7 +255,12 @@ proc genOp(c: var Con; t: PType; kind: TTypeAttachedOp; dest, ri: PNode): PNode
dbg:
if kind == attachedDestructor:
echo "destructor is ", op.id, " ", op.ast
if sfError in op.flags: checkForErrorPragma(c, t, ri, AttachedOpToStr[kind])
if sfError in op.flags:
if ri != nil:
checkForErrorPragma(c, t, ri, AttachedOpToStr[kind])
else:
# uses the lineinfos of `dest` is `ri` is not available
checkForErrorPragma(c, t, dest, AttachedOpToStr[kind])
c.genOp(op, dest)
proc genDestroy(c: var Con; dest: PNode): PNode =
@@ -275,7 +288,7 @@ proc deepAliases(dest, ri: PNode): bool =
return aliases(dest, ri) != no
proc genSink(c: var Con; s: var Scope; dest, ri: PNode; flags: set[MoveOrCopyFlag] = {}): PNode =
if (c.inLoopCond == 0 and (isUnpackedTuple(dest) or IsDecl in flags or
if (c.inLoopCond == 0 and (isFullyUnpackedTuple(dest) or IsDecl in flags or
(isAnalysableFieldAccess(dest, c.owner) and isFirstWrite(dest, c)))) or
isNoInit(dest) or IsReturn in flags:
# optimize sink call into a bitwise memcopy
@@ -291,7 +304,7 @@ proc genSink(c: var Con; s: var Scope; dest, ri: PNode; flags: set[MoveOrCopyFla
if deepAliases(dest, ri):
# consider: x = x + y, it is wrong to destroy the destination first!
# tmp to support self assignments
let tmp = c.getTemp(s, dest.typ, dest.info)
let tmp = c.getTemp(s, dest.typ, dest.info, needsInit = false)
result = newTree(nkStmtList, newTree(nkFastAsgn, tmp, dest), newTree(nkFastAsgn, dest, ri),
c.genDestroy(tmp))
else:
@@ -318,21 +331,21 @@ proc isCriticalLink(dest: PNode): bool {.inline.} =
result = dest.kind != nkSym
proc finishCopy(c: var Con; result, dest: PNode; flags: set[MoveOrCopyFlag]; isFromSink: bool) =
if c.graph.config.selectedGC == gcOrc and IsExplicitSink notin flags:
if c.graph.config.selectedGC in {gcOrc, gcYrc} and IsExplicitSink notin flags:
# add cyclic flag, but not to sink calls, which IsExplicitSink generates
let t = dest.typ.skipTypes(tyUserTypeClasses + {tyGenericInst, tyAlias, tySink, tyDistinct})
if cyclicType(c.graph, t):
result.add boolLit(c.graph, result.info, isFromSink or isCriticalLink(dest))
proc genMarkCyclic(c: var Con; result, dest: PNode) =
if c.graph.config.selectedGC == gcOrc:
if c.graph.config.selectedGC in {gcOrc, gcYrc}:
let t = dest.typ.skipTypes({tyGenericInst, tyAlias, tySink, tyDistinct})
if cyclicType(c.graph, t):
if t.kind == tyRef:
result.add callCodegenProc(c.graph, "nimMarkCyclic", dest.info, dest)
else:
let xenv = genBuiltin(c.graph, c.idgen, mAccessEnv, "accessEnv", dest)
xenv.typ() = getSysType(c.graph, dest.info, tyPointer)
xenv.typ = getSysType(c.graph, dest.info, tyPointer)
result.add callCodegenProc(c.graph, "nimMarkCyclic", dest.info, xenv)
proc genCopyNoCheck(c: var Con; dest, ri: PNode; a: TTypeAttachedOp): PNode =
@@ -360,7 +373,7 @@ proc genDiscriminantAsgn(c: var Con; s: var Scope; n: PNode): PNode =
# but fields within active case branch might need destruction
# tmp to support self assignments
let tmp = c.getTemp(s, n[1].typ, n.info)
let tmp = c.getTemp(s, n[1].typ, n.info, needsInit = false)
result = newTree(nkStmtList)
result.add newTree(nkFastAsgn, tmp, p(n[1], c, s, consumed))
@@ -400,7 +413,7 @@ proc genWasMoved(c: var Con, n: PNode): PNode =
result = genOp(c, op, n)
else:
result = newNodeI(nkCall, n.info)
result.add(newSymNode(createMagic(c.graph, c.idgen, "`=wasMoved`", mWasMoved)))
result.add(newSymNode(createMagic(c.graph, c.idgen, "wasMoved", mWasMoved)))
result.add copyTree(n) #mWasMoved does not take the address
#if n.kind != nkSym:
# message(c.graph.config, n.info, warnUser, "wasMoved(" & $n & ")")
@@ -408,7 +421,7 @@ proc genWasMoved(c: var Con, n: PNode): PNode =
proc genDefaultCall(t: PType; c: Con; info: TLineInfo): PNode =
result = newNodeI(nkCall, info)
result.add(newSymNode(createMagic(c.graph, c.idgen, "default", mDefault)))
result.typ() = t
result.typ = t
proc destructiveMoveVar(n: PNode; c: var Con; s: var Scope): PNode =
# generate: (let tmp = v; reset(v); tmp)
@@ -446,49 +459,50 @@ proc isCapturedVar(n: PNode): bool =
else: result = false
proc passCopyToSink(n: PNode; c: var Con; s: var Scope): PNode =
result = newNodeIT(nkStmtListExpr, n.info, n.typ)
let nTyp = n.typ.skipTypes(tyUserTypeClasses)
let tmp = c.getTemp(s, nTyp, n.info)
if hasDestructor(c, nTyp):
let typ = nTyp.skipTypes({tyGenericInst, tyAlias, tySink})
let op = getAttachedOp(c.graph, typ, attachedDup)
if op != nil and tfHasOwned notin typ.flags:
if sfError in op.flags:
c.checkForErrorPragma(nTyp, n, "=dup")
else:
let copyOp = getAttachedOp(c.graph, typ, attachedAsgn)
if copyOp != nil and sfError in copyOp.flags and
sfOverridden notin op.flags:
c.checkForErrorPragma(nTyp, n, "=dup", inferredFromCopy = true)
let src = p(n, c, s, normal)
var newCall = newTreeIT(nkCall, src.info, src.typ,
newSymNode(op),
src)
c.finishCopy(newCall, n, {}, isFromSink = true)
result.add newTreeI(nkFastAsgn,
src.info, tmp,
newCall
)
else:
result.add c.genWasMoved(tmp)
var m = c.genCopy(tmp, n, {})
m.add p(n, c, s, normal)
c.finishCopy(m, n, {}, isFromSink = true)
result.add m
if isLValue(n) and not isCapturedVar(n) and nTyp.skipTypes(abstractInst).kind != tyRef and c.inSpawn == 0:
message(c.graph.config, n.info, hintPerformance,
("passing '$1' to a sink parameter introduces an implicit copy; " &
"if possible, rearrange your program's control flow to prevent it") % $n)
if c.inEnsureMove > 0:
localError(c.graph.config, n.info, errFailedMove,
("cannot move '$1', passing '$1' to a sink parameter introduces an implicit copy") % $n)
else:
if c.graph.config.selectedGC in {gcArc, gcOrc, gcAtomicArc}:
if not hasDestructorOrAsgn(c, nTyp):
# Non-managed (plain-old-data) type: no ownership transfer is needed.
# Return the expression directly — no temp required.
if c.graph.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc}:
assert(not containsManagedMemory(nTyp))
if nTyp.skipTypes(abstractInst).kind in {tyOpenArray, tyVarargs}:
localError(c.graph.config, n.info, "cannot create an implicit openArray copy to be passed to a sink parameter")
result.add newTree(nkAsgn, tmp, p(n, c, s, normal))
return p(n, c, s, normal)
result = newNodeIT(nkStmtListExpr, n.info, n.typ)
let tmp = c.getTemp(s, nTyp, n.info, needsInit = false)
let typ = nTyp.skipTypes({tyGenericInst, tyAlias, tySink})
let op = getAttachedOp(c.graph, typ, attachedDup)
if op != nil and tfHasOwned notin typ.flags:
if sfError in op.flags:
c.checkForErrorPragma(nTyp, n, "=dup")
else:
let copyOp = getAttachedOp(c.graph, typ, attachedAsgn)
if copyOp != nil and sfError in copyOp.flags and
sfOverridden notin op.flags:
c.checkForErrorPragma(nTyp, n, "=dup", inferredFromCopy = true)
let src = p(n, c, s, normal)
var newCall = newTreeIT(nkCall, src.info, src.typ,
newSymNode(op),
src)
c.finishCopy(newCall, n, {}, isFromSink = true)
result.add newTreeI(nkFastAsgn,
src.info, tmp,
newCall
)
else:
result.add c.genWasMoved(tmp)
var m = c.genCopy(tmp, n, {})
m.add p(n, c, s, normal)
c.finishCopy(m, n, {}, isFromSink = true)
result.add m
if isLValue(n) and not isCapturedVar(n) and nTyp.skipTypes(abstractInst).kind != tyRef and c.inSpawn == 0:
message(c.graph.config, n.info, hintPerformance,
("passing '$1' to a sink parameter introduces an implicit copy; " &
"if possible, rearrange your program's control flow to prevent it") % $n)
if c.inEnsureMove > 0:
localError(c.graph.config, n.info, errFailedMove,
("cannot move '$1', passing '$1' to a sink parameter introduces an implicit copy") % $n)
# Since we know somebody will take over the produced copy, there is
# no need to destroy it.
result.add tmp
@@ -519,7 +533,7 @@ proc ensureDestruction(arg, orig: PNode; c: var Con; s: var Scope): PNode =
# produce temp creation for (fn, env). But we need to move 'env'?
# This was already done in the sink parameter handling logic.
result = newNodeIT(nkStmtListExpr, arg.info, arg.typ)
let tmp = c.getTemp(s, arg.typ, arg.info)
let tmp = c.getTemp(s, arg.typ, arg.info, true)
result.add c.genSink(s, tmp, arg, {IsDecl})
result.add tmp
s.final.add c.genDestroy(tmp)
@@ -559,7 +573,7 @@ proc cycleCheck(n: PNode; c: var Con) =
proc pVarTopLevel(v: PNode; c: var Con; s: var Scope; res: PNode) =
# move the variable declaration to the top of the frame:
s.vars.add v.sym
if isUnpackedTuple(v):
if isFullyUnpackedTuple(v):
if c.inLoop > 0:
# unpacked tuple needs reset at every loop iteration
res.add newTree(nkFastAsgn, v, genDefaultCall(v.typ, c, v.info))
@@ -598,7 +612,7 @@ template processScopeExpr(c: var Con; s: var Scope; ret: PNode, processCall: unt
# There is a possibility to do this check: s.wasMoved.len > 0 or s.final.len > 0
# later and use it to eliminate the temporary when theres no need for it, but its
# tricky because you would have to intercept moveOrCopy at a certain point
let tmp = c.getTemp(s.parent[], ret.typ, ret.info)
let tmp = c.getTemp(s.parent[], ret.typ, ret.info, needsInit = true)
tmp.sym.flags = tmpFlags
let cpy = if hasDestructor(c, ret.typ) and
ret.typ.kind notin {tyOpenArray, tyVarargs}:
@@ -759,7 +773,7 @@ proc pRaiseStmt(n: PNode, c: var Con; s: var Scope): PNode =
result = copyNode(n)
result.add call
else:
let tmp = c.getTemp(s, n[0].typ, n.info)
let tmp = c.getTemp(s, n[0].typ, n.info, needsInit = true)
var m = c.genCopyNoCheck(tmp, n[0], attachedAsgn)
m.add p(n[0], c, s, normal)
c.finishCopy(m, n[0], {}, isFromSink = false)
@@ -789,6 +803,23 @@ proc hasCustomDestructor(c: Con, t: PType): bool =
obj = skipTypes(obj.baseClass, abstractPtrs)
result = result or isCustomDestructor(c, obj)
const
exprBranchKinds = {nkStmtListExpr, nkBlockExpr, nkIfExpr, nkCaseStmt,
nkTryStmt, nkPragmaBlock}
proc distributeAsgn(asgnKind: TNodeKind; dest, ri: PNode; c: var Con; s: var Scope): PNode =
## Distributes an assignment ``dest = ri`` into the leaf expressions of
## ``ri`` when ``ri`` is an expression-based control flow construct. This
## avoids creating pointless intermediate temporaries (bug #25850). The
## descent is recursive so that nestings like ``block: ...; if c: a else: b``
## assign directly to ``dest`` instead of going through a temp per branch.
if ri.kind in exprBranchKinds:
template process(child, s): untyped =
distributeAsgn(asgnKind, dest, child, c, s)
handleNestedTempl(ri, process, willProduceStmt = true)
else:
result = newTree(asgnKind, dest, p(ri, c, s, consumed))
proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSingleUsedTemp}; inReturn = false): PNode =
if n.kind in {nkStmtList, nkStmtListExpr, nkBlockStmt, nkBlockExpr, nkIfStmt,
nkIfExpr, nkCaseStmt, nkWhen, nkWhileStmt, nkParForStmt, nkTryStmt, nkPragmaBlock}:
@@ -802,15 +833,7 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
result = passCopyToSink(n, c, s)
elif n.kind in {nkBracket, nkObjConstr, nkTupleConstr, nkClosure, nkNilLit} +
nkCallKinds + nkLiterals:
if n.kind in nkCallKinds and n[0].kind == nkSym:
if n[0].sym.magic == mEnsureMove:
inc c.inEnsureMove
result = p(n[1], c, s, sinkArg)
dec c.inEnsureMove
else:
result = p(n, c, s, consumed)
else:
result = p(n, c, s, consumed)
result = p(n, c, s, consumed)
elif ((n.kind == nkSym and isSinkParam(n.sym)) or isAnalysableFieldAccess(n, c.owner)) and
isLastRead(n, c, s) and not (n.kind == nkSym and isCursor(n)):
# Sinked params can be consumed only once. We need to reset the memory
@@ -822,9 +845,9 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
n[1].typ.skipTypes(abstractInst-{tyOwned}).kind == tyOwned:
# allow conversions from owned to unowned via this little hack:
let nTyp = n[1].typ
n[1].typ() = n.typ
n[1].typ = n.typ
result[1] = p(n[1], c, s, sinkArg)
result[1].typ() = nTyp
result[1].typ = nTyp
else:
result[1] = p(n[1], c, s, sinkArg)
elif n.kind in {nkObjDownConv, nkObjUpConv}:
@@ -886,12 +909,6 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
if mode == normal and (isRefConstr or hasCustomDestructor(c, t)):
result = ensureDestruction(result, n, c, s)
of nkCallKinds:
if n[0].kind == nkSym and n[0].sym.magic == mEnsureMove:
inc c.inEnsureMove
result = p(n[1], c, s, sinkArg)
dec c.inEnsureMove
return
let inSpawn = c.inSpawn
if n[0].kind == nkSym and n[0].sym.magic == mSpawn:
c.inSpawn.inc
@@ -909,13 +926,19 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
isDangerous = true
result = shallowCopy(n)
for i in 1..<n.len:
if i < L and isCompileTimeOnly(parameters[i]):
result[i] = n[i]
elif i < L and (isSinkTypeForParam(parameters[i]) or inSpawn > 0):
result[i] = p(n[i], c, s, sinkArg)
else:
result[i] = p(n[i], c, s, normal)
if n[0].kind == nkSym and n[0].sym.magic == mEnsureMove:
inc c.inEnsureMove
result[1] = p(n[1], c, s, sinkArg)
dec c.inEnsureMove
else:
for i in 1..<n.len:
if i < L and isCompileTimeOnly(parameters[i]):
result[i] = n[i]
elif i < L and (isSinkTypeForParam(parameters[i]) or inSpawn > 0):
result[i] = p(n[i], c, s, sinkArg)
else:
result[i] = p(n[i], c, s, normal)
when false:
if isDangerous:
@@ -923,7 +946,7 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
if n[0].kind == nkSym and n[0].sym.magic in {mNew, mNewFinalize}:
result[0] = copyTree(n[0])
if c.graph.config.selectedGC in {gcHooks, gcArc, gcAtomicArc, gcOrc}:
if c.graph.config.selectedGC in {gcHooks, gcArc, gcAtomicArc, gcOrc, gcYrc}:
let destroyOld = c.genDestroy(result[1])
result = newTree(nkStmtList, destroyOld, result)
else:
@@ -943,6 +966,9 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
of nkVarSection, nkLetSection:
# transform; var x = y to var x; x op y where op is a move or copy
result = newNodeI(nkStmtList, n.info)
let isInProc = c.owner.kind in {skProc, skFunc, skMethod, skIterator, skConverter}
for it in n:
var ri = it[^1]
if it.kind == nkVarTuple and hasDestructor(c, ri.typ):
@@ -958,7 +984,15 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
s.locals.add v.sym
pVarTopLevel(v, c, s, result)
if ri.kind != nkEmpty:
result.add moveOrCopy(v, ri, c, s, if v.kind == nkSym: {IsDecl} else: {})
let isGlobalPragma = v.kind == nkSym and
{sfPure, sfGlobal} <= v.sym.flags and
isInProc
if isGlobalPragma:
c.graph.procGlobals.add newTree(nkFastAsgn, v, ri)
else:
let value = moveOrCopy(v, ri, c, s, if v.kind == nkSym: {IsDecl} else: {})
result.add value
elif ri.kind == nkEmpty and c.inLoop > 0:
let skipInit = v.kind == nkDotExpr and # Closure var
sfNoInit in v[1].sym.flags
@@ -987,6 +1021,11 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
result = moveOrCopy(p(n[0], c, s, mode), n[1], c, s, flags)
elif isDiscriminantField(n[0]):
result = c.genDiscriminantAsgn(s, n)
elif n[1].kind in exprBranchKinds:
# Distribute the assignment into each branch to avoid
# creating pointless temporaries for expression-based control flow.
let dest = p(n[0], c, s, mode)
result = distributeAsgn(n.kind, dest, n[1], c, s)
else:
result = copyNode(n)
result.add p(n[0], c, s, mode)
@@ -1022,9 +1061,9 @@ proc p(n: PNode; c: var Con; s: var Scope; mode: ProcessMode; tmpFlags = {sfSing
n[1].typ.skipTypes(abstractInst-{tyOwned}).kind == tyOwned:
# allow conversions from owned to unowned via this little hack:
let nTyp = n[1].typ
n[1].typ() = n.typ
n[1].typ = n.typ
result[1] = p(n[1], c, s, mode)
result[1].typ() = nTyp
result[1].typ = nTyp
else:
result[1] = p(n[1], c, s, mode)
@@ -1130,6 +1169,25 @@ proc genFieldAccessSideEffects(c: var Con; s: var Scope; dest, ri: PNode; flags:
var snk = c.genSink(s, dest, newAccess, flags)
result = newTree(nkStmtList, v, snk, c.genWasMoved(newAccess))
proc ownsData(c: var Con; s: var Scope; orig: PNode; flags: set[MoveOrCopyFlag]): PNode =
var n = orig
while true:
case n.kind
of nkDotExpr, nkCheckedFieldExpr, nkBracketExpr:
n = n[0]
else:
break
if n.kind in nkCallKinds and n.typ != nil and hasDestructor(c, n.typ):
result = newNodeIT(nkStmtListExpr, orig.info, orig.typ)
let tmp = c.getTemp(s, n.typ, n.info, needsInit = true)
tmp.sym.flagsImpl.incl sfSingleUsedTemp
result.add newTree(nkFastAsgn, tmp, copyTree(n))
s.final.add c.genDestroy(tmp)
n[] = tmp[]
result.add copyTree(orig)
else:
result = nil
proc moveOrCopy(dest, ri: PNode; c: var Con; s: var Scope, flags: set[MoveOrCopyFlag] = {}): PNode =
var ri = ri
var isEnsureMove = 0
@@ -1156,7 +1214,7 @@ proc moveOrCopy(dest, ri: PNode; c: var Con; s: var Scope, flags: set[MoveOrCopy
of nkCallKinds:
result = c.genSink(s, dest, p(ri, c, s, consumed), flags)
of nkBracketExpr:
if isUnpackedTuple(ri[0]):
if isFullyUnpackedTuple(ri[0]):
# unpacking of tuple: take over the elements
result = c.genSink(s, dest, p(ri, c, s, consumed), flags)
elif isAnalysableFieldAccess(ri, c.owner) and isLastRead(ri, c, s):
@@ -1205,15 +1263,22 @@ proc moveOrCopy(dest, ri: PNode; c: var Con; s: var Scope, flags: set[MoveOrCopy
result.add p(ri, c, s, consumed)
c.finishCopy(result, dest, flags, isFromSink = false)
of nkHiddenSubConv, nkHiddenStdConv, nkConv, nkObjDownConv, nkObjUpConv, nkCast:
result = c.genSink(s, dest, p(ri, c, s, sinkArg), flags)
of nkStmtListExpr, nkBlockExpr, nkIfExpr, nkCaseStmt, nkTryStmt:
if IsExplicitSink in flags:
result = c.genSink(s, dest, p(ri, c, s, consumed), flags)
else:
result = c.genSink(s, dest, p(ri, c, s, sinkArg), flags)
of nkStmtListExpr, nkBlockExpr, nkIfExpr, nkCaseStmt, nkTryStmt, nkPragmaBlock:
template process(child, s): untyped = moveOrCopy(dest, child, c, s, flags)
# We know the result will be a stmt so we use that fact to optimize
handleNestedTempl(ri, process, willProduceStmt = true)
of nkRaiseStmt:
result = pRaiseStmt(ri, c, s)
else:
if isAnalysableFieldAccess(ri, c.owner) and isLastRead(ri, c, s) and
let isOwnsData = ownsData(c, s, ri2, flags)
if isOwnsData != nil:
result = moveOrCopy(dest, isOwnsData, c, s, flags)
elif isAnalysableFieldAccess(ri, c.owner) and isLastRead(ri, c, s) and
canBeMoved(c, dest.typ):
# Rule 3: `=sink`(x, z); wasMoved(z)
let snk = c.genSink(s, dest, ri, flags)
@@ -1252,6 +1317,55 @@ when false:
for i in 0..<n.safeLen:
injectDefaultCalls(n[i], c)
proc replaceSinkParam(n: PNode, mapping: Table[int, PSym]): PNode =
case n.kind
of nkSym:
if n.sym.id in mapping:
result = newSymNode(mapping[n.sym.id])
else:
result = n
of nkVarSection, nkLetSection:
result = copyNode(n)
newSons(result, n.len)
for i in 0..<n.len:
result[i] = copyNode(n[i])
for j in 0..<n[i].len-1:
result[i].add n[i][j]
result[i].add replaceSinkParam(n[i][^1], mapping)
of {nkNone..nkNilLit}-{nkSym}, nkTypeSection, nkProcDef, nkConverterDef,
nkMethodDef, nkIteratorDef, nkMacroDef, nkTemplateDef, nkLambda, nkDo,
nkFuncDef, nkConstSection, nkConstDef, nkIncludeStmt, nkImportStmt,
nkExportStmt, nkPragma, nkCommentStmt, nkBreakState,
nkTypeOfExpr, nkMixinStmt, nkBindStmt:
result = n
else:
result = copyNode(n)
for i in 0..<n.len:
result.add replaceSinkParam(n[i], mapping)
proc addSinkCopy(c: var Con; s: var Scope; sinkParams: seq[PSym]; n: PNode): PNode =
result = newNodeI(nkStmtList, n.info)
var mapping = initTable[int, PSym]()
var mutated = newSeq[PNode]()
getPotentialWrites(n, false, mutated)
var mutatedSet = initIntSet()
for m in mutated:
mutatedSet.incl m.sym.id
for param in sinkParams:
if param.id in mutatedSet:
let newSym = newSym(skTemp, getIdent(c.graph.cache, "sinkCopy"), c.idgen, param.owner, n.info)
newSym.flagsImpl.incl sfFromGeneric
newSym.typ = param.typ.elementType
mapping[param.id] = newSym
let v = newNodeI(nkVarSection, n.info)
v.addVar(newSymNode(newSym), newSymNode(param))
result.add v
if mapping.len > 0:
result.add replaceSinkParam(n, mapping)
else:
result = n
proc injectDestructorCalls*(g: ModuleGraph; idgen: IdGenerator; owner: PSym; n: PNode): PNode =
when toDebug.len > 0:
shouldDebug = toDebug == owner.name.s or toDebug == "always"
@@ -1265,15 +1379,24 @@ proc injectDestructorCalls*(g: ModuleGraph; idgen: IdGenerator; owner: PSym; n:
var scope = Scope(body: n)
let body = p(n, c, scope, normal)
var sinkParams = newSeq[PSym]()
if owner.kind in {skProc, skFunc, skMethod, skIterator, skConverter}:
let params = owner.typ.n
for i in 1..<params.len:
let t = params[i].sym.typ
if isSinkTypeForParam(t) and hasDestructor(c, t.skipTypes({tySink})):
scope.final.add c.genDestroy(params[i])
if isSinkTypeForParam(t):
let baseType = t.skipTypes({tySink})
if baseType.kind in {tyString, tySequence, tyArray, tyTuple, tyObject}:
sinkParams.add params[i].sym
if hasDestructor(c, baseType):
scope.final.add c.genDestroy(params[i])
#if optNimV2 in c.graph.config.globalOptions:
# injectDefaultCalls(n, c)
result = optimize processScope(c, scope, body)
if sinkParams.len > 0:
result = addSinkCopy(c, scope, sinkParams, result)
dbg:
echo ">---------transformed-to--------->"
echo renderTree(result, {renderIds})

122
compiler/inliner.nim Normal file
View File

@@ -0,0 +1,122 @@
proc copySymdef(n: PNode; locals: var Table[int, PSym]; idgen: IdGenerator; owner: PSym): PNode =
case n.kind
of nkEmpty..pred(nkSym), succ(nkSym)..nkNilLit:
result = n
of nkSym:
let oldSym = n.sym
let newSym = copySym(oldSym, idgen)
setOwner(newSym, owner)
locals[oldSym.id] = newSym
result = newSymNode(newSym, oldSym.info)
else:
result = shallowCopy(n)
for i in 0..<n.len:
result[i] = copySymdef(n[i], locals, idgen, owner)
proc copyInlineProcBody(n: PNode; locals: var Table[int, PSym]; idgen: IdGenerator; owner: PSym): PNode =
case n.kind
of nkEmpty..pred(nkSym), succ(nkSym)..nkNilLit:
result = n
of nkSym:
let sym = locals.getOrDefault(n.sym.id)
if sym != nil:
result = newSymNode(sym, n.info)
else:
result = n
of nkLetSection, nkVarSection:
result = shallowCopy(n)
for i in 0..<n.len:
let it = n[i]
if it.kind == nkCommentStmt:
result[i] = it
elif it.kind in {nkIdentDefs, nkConstDef}:
result[i] = shallowCopy(it)
for j in 0..<it.len-2:
result[i][j] = copySymdef(it[j], locals, idgen, owner)
for j in it.len-2..<it.len:
result[i][j] = copyInlineProcBody(it[j], locals, idgen, owner)
else:
assert it.kind == nkVarTuple
result[i] = shallowCopy(it)
for j in 0..<it.len-2:
assert it[j].kind == nkSym
let oldSym = it[j].sym
let newSym = copySym(oldSym, idgen)
setOwner(newSym, owner)
locals[oldSym.id] = newSym
result[i][j] = newSymNode(newSym, oldSym.info)
for j in it.len-2..<it.len:
result[i][j] = copyInlineProcBody(it[j], locals, idgen, owner)
of nkForStmt, nkParForStmt:
result = shallowCopy(n)
for i in 0..<n.len-2:
assert n[i].kind == nkSym
let oldSym = n[i].sym
let newSym = copySym(oldSym, idgen)
setOwner(newSym, owner)
locals[oldSym.id] = newSym
result[i] = newSymNode(newSym, oldSym.info)
result[n.len-2] = copyInlineProcBody(n[n.len-2], locals, idgen, owner)
result[n.len-1] = copyInlineProcBody(n[n.len-1], locals, idgen, owner)
of routineDefs, nkTypeSection, nkTypeOfExpr, nkMixinStmt, nkBindStmt, nkConstSection:
result = n
else:
result = shallowCopy(n)
for i in 0..<n.len:
result[i] = copyInlineProcBody(n[i], locals, idgen, owner)
proc copyParams(n: PNode; locals: var Table[int, PSym]; idgen: IdGenerator; owner: PSym): PNode =
result = shallowCopy(n)
result[0] = n[0] # return type
for i in 1..<n.len:
let it = n[i]
assert it.kind == nkIdentDefs
result[i] = shallowCopy(it)
for j in 0..<it.len-2:
assert it[j].kind == nkSym
let oldSym = it[j].sym
let newSym = copySym(oldSym, idgen)
setOwner(newSym, owner)
locals[oldSym.id] = newSym
result[i][j] = newSymNode(newSym, oldSym.info)
owner.typ.addParam newSym
for j in it.len-2..<it.len:
result[i][j] = copyInlineProcBody(it[j], locals, idgen, owner)
proc copyInlineProc(prc: PSym; idgen: IdGenerator): PSym =
result = copySym(prc, idgen)
var locals = initTable[int, PSym]()
var a = shallowCopy(prc.ast)
if resultPos < prc.ast.len and prc.ast[resultPos].kind == nkSym:
let oldRes = prc.ast[resultPos].sym
let newRes = copySym(oldRes, idgen)
setOwner(newRes, result)
locals[oldRes.id] = newRes
a[resultPos] = newSymNode(newRes, oldRes.info)
result.typ = copyType(prc.typ, idgen, result)
result.typ.n = newNodeI(prc.typ.n.kind, prc.typ.n.info)
if prc.typ.n.len > 0:
result.typ.n.add copyNode(prc.typ.n[0])
for i in 1..<prc.typ.n.len:
let it = prc.typ.n[i]
assert it.kind == nkSym
let oldSym = it.sym
let newSym = copySym(oldSym, idgen)
setOwner(newSym, result)
locals[oldSym.id] = newSym
result.typ.addParam newSym
for i in 0..<prc.ast.len:
if i == paramsPos:
a[i] = copyTree(prc.ast[i])
elif i == resultPos and prc.ast[i].kind == nkSym:
discard "handled above"
else:
a[i] = copyInlineProcBody(prc.ast[i], locals, idgen, result)
result.ast = a
#echo "Produced: ", renderTree(result.ast, {renderIds})

View File

@@ -6,7 +6,7 @@ Name: "Nim"
Version: "$version"
Platforms: """
windows: i386;amd64
linux: i386;hppa;ia64;alpha;amd64;powerpc64;arm;sparc;sparc64;m68k;mips;mipsel;mips64;mips64el;powerpc;powerpc64el;arm64;riscv32;riscv64;loongarch64
linux: i386;hppa;ia64;alpha;amd64;powerpc64;arm;sparc;sparc64;s390x;m68k;mips;mipsel;mips64;mips64el;powerpc;powerpc64el;arm64;riscv32;riscv64;loongarch64
macosx: i386;amd64;powerpc64;arm64
solaris: i386;amd64;sparc;sparc64
freebsd: i386;amd64;powerpc64;arm;arm64;riscv64;sparc64;mips;mipsel;mips64;mips64el;powerpc;powerpc64el
@@ -80,6 +80,7 @@ Files: "lib"
Files: "examples"
Files: "dist/nimble"
Files: "dist/checksums"
Files: "dist/nimony"
Files: "tests"

View File

@@ -460,7 +460,9 @@ proc addInt128*(result: var string; value: Int128) =
var i = initialSize
var j = high(result)
while i < j:
swap(result[i], result[j])
let tmp = result[i]
result[i] = result[j]
result[j] = tmp
i += 1
j -= 1

View File

@@ -166,7 +166,7 @@ proc containsVariable(n: PNode): bool =
proc checkIsolate*(n: PNode): bool =
if types.containsTyRef(n.typ):
# XXX Maybe require that 'n.typ' is acyclic. This is not much
# worse than the already exisiting inheritance and closure restrictions.
# worse than the already existing inheritance and closure restrictions.
case n.kind
of nkCharLit..nkNilLit:
result = true

98
compiler/itemids.nim Normal file
View File

@@ -0,0 +1,98 @@
#
#
# The Nim Compiler
# (c) Copyright 2026 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## `ItemId` is the identity of a symbol or type: a `(module, item)` pair.
##
## The fields are private on purpose: the module half reserves bit 30 as the
## "backend minted" marker, so all construction and inspection has to go
## through this module's API and the marker bit can never leak into module
## indexing or arithmetic.
##
## Three id spaces coexist per module:
## - Semantic-phase and NIF-loader ids: `itemId(module, item)` with `item > 0`.
## - Backend-minted ids (IC codegen, `nim nifc`: transf labels and temps,
## lifted hooks): `backendItemId` sets `BackendModuleBit`, so these can
## never compare equal to a loader id even though both counters mint the
## same small `item` range in one process. They never cross a process
## boundary and must never be written to a NIF file.
## - Derived env/tuple-field ids (`lowerings.addField`): the source local's
## id with `item` negated. `derivedFieldId` preserves the backend marker,
## keeping the derivation collision-free for both id spaces above.
import std/hashes
when defined(nimPreviewSlimSystem):
import std/assertions
const
BackendModuleBit = 0x4000_0000'i32
# Bit 30 of the module field. Bit 31 stays clear so marked module values
# remain non-negative and cannot be mistaken for the special negative
# module ids like `PackageModuleId`.
PackageModuleId* = -3'i32
type
ItemId* = object
moduleBits: int32
itemBits: int32
proc itemId*(module, item: int32): ItemId {.inline.} =
assert module < 0 or (module and BackendModuleBit) == 0
ItemId(moduleBits: module, itemBits: item)
proc backendItemId*(module, item: int32): ItemId {.inline.} =
## An id minted during IC codegen; distinct from every `itemId` of the
## same module so that the loader's stub counter and the backend's counter
## cannot collide in id-keyed tables.
assert module >= 0 and (module and BackendModuleBit) == 0
ItemId(moduleBits: module or BackendModuleBit, itemBits: item)
proc module*(x: ItemId): int32 {.inline.} =
if x.moduleBits >= 0: x.moduleBits and not BackendModuleBit
else: x.moduleBits
proc item*(x: ItemId): int32 {.inline.} = x.itemBits
proc isBackendMinted*(x: ItemId): bool {.inline.} =
x.moduleBits >= 0 and (x.moduleBits and BackendModuleBit) != 0
proc derivedFieldId*(source: ItemId): ItemId {.inline.} =
## The id of the env/tuple field that `lowerings.addField` derives for a
## captured local: `item` negated, module bits (including the backend
## marker) preserved.
ItemId(moduleBits: source.moduleBits, itemBits: -abs(source.itemBits))
proc matchesDerivedFieldId*(field, source: ItemId): bool {.inline.} =
## Does `field` carry the id `derivedFieldId` would derive for `source`?
## `source` may itself already be the derived field id.
field.moduleBits == source.moduleBits and
field.itemBits == -abs(source.itemBits)
proc `==`*(a, b: ItemId): bool {.inline.} =
# raw bit comparison: a backend-minted id never equals a loader id
a.itemBits == b.itemBits and a.moduleBits == b.moduleBits
proc hash*(x: ItemId): Hash =
var h: Hash = hash(x.moduleBits)
h = h !& hash(x.itemBits)
result = !$h
proc `$`*(x: ItemId): string =
result = "(module: " & $x.module & ", item: " & $x.itemBits
if x.isBackendMinted: result.add ", backend"
result.add ")"
const
moduleShift = when defined(cpu32): 20 else: 24
proc toId*(a: ItemId): int {.inline.} =
## Packs an ItemId into a single int. Uses the raw module bits so the
## backend marker keeps the two id spaces disjoint (bit 30 shifts to
## bit 54; like the module/item split itself this needs a 64-bit int).
(a.moduleBits.int shl moduleShift) + a.itemBits.int

View File

@@ -277,7 +277,8 @@ proc mangleName(m: BModule, s: PSym): Rope =
else:
result.add("_")
result.add(rope(s.id))
s.loc.snippet = result
ensureMutable s
s.locImpl.snippet = result
proc escapeJSString(s: string): string =
result = newStringOfCap(s.len + s.len shr 2)
@@ -728,44 +729,47 @@ proc arithAux(p: PProc, n: PNode, r: var TCompRes, op: TMagic) =
of mShrI:
let typ = n[1].typ.skipTypes(abstractVarRange)
if typ.kind == tyInt64 and optJsBigInt64 in p.config.globalOptions:
applyFormat("BigInt.asIntN(64, BigInt.asUintN(64, $1) >> BigInt($2))")
applyFormat("BigInt.asIntN(64, BigInt.asUintN(64, $1) >> (BigInt($2) & 63n))")
elif typ.kind == tyUInt64 and optJsBigInt64 in p.config.globalOptions:
applyFormat("($1 >> BigInt($2))")
applyFormat("($1 >> (BigInt($2) & 63n))")
else:
let bitmask = typ.size * 8 - 1
if typ.kind in {tyInt..tyInt32}:
let trimmerU = unsignedTrimmer(typ.size)
let trimmerS = signedTrimmer(typ.size)
r.res = "((($1 $2) >>> $3) $4)" % [xLoc, trimmerU, yLoc, trimmerS]
r.res = "((($1 $2) >>> ($3 & $5)) $4)" % [xLoc, trimmerU, yLoc, trimmerS, $bitmask]
else:
applyFormat("($1 >>> $2)")
r.res = "($1 >>> ($2 & $3))" % [xLoc, yLoc, $bitmask]
of mShlI:
let typ = n[1].typ.skipTypes(abstractVarRange)
if typ.size == 8:
if typ.kind == tyInt64 and optJsBigInt64 in p.config.globalOptions:
applyFormat("BigInt.asIntN(64, $1 << BigInt($2))")
applyFormat("BigInt.asIntN(64, $1 << (BigInt($2) & 63n))")
elif typ.kind == tyUInt64 and optJsBigInt64 in p.config.globalOptions:
applyFormat("BigInt.asUintN(64, $1 << BigInt($2))")
applyFormat("BigInt.asUintN(64, $1 << (BigInt($2) & 63n))")
else:
applyFormat("($1 * Math.pow(2, $2))")
applyFormat("($1 * Math.pow(2, ($2 & 63)))")
else:
let bitmask = typ.size * 8 - 1
if typ.kind in {tyUInt..tyUInt32}:
let trimmer = unsignedTrimmer(typ.size)
r.res = "(($1 << $2) $3)" % [xLoc, yLoc, trimmer]
r.res = "(($1 << ($2 & $4)) $3)" % [xLoc, yLoc, trimmer, $bitmask]
else:
let trimmer = signedTrimmer(typ.size)
r.res = "(($1 << $2) $3)" % [xLoc, yLoc, trimmer]
r.res = "(($1 << ($2 & $4)) $3)" % [xLoc, yLoc, trimmer, $bitmask]
of mAshrI:
let typ = n[1].typ.skipTypes(abstractVarRange)
if typ.size == 8:
if optJsBigInt64 in p.config.globalOptions:
applyFormat("($1 >> BigInt($2))")
applyFormat("($1 >> (BigInt($2) & 63n))")
else:
applyFormat("Math.floor($1 / Math.pow(2, $2))")
applyFormat("Math.floor($1 / Math.pow(2, ($2 & 63)))")
else:
let bitmask = typ.size * 8 - 1
if typ.kind in {tyUInt..tyUInt32}:
applyFormat("($1 >>> $2)")
r.res = "($1 >>> ($2 & $3)))" % [xLoc, yLoc, $bitmask]
else:
applyFormat("($1 >> $2)")
r.res = "($1 >> ($2 & $3))" % [xLoc, yLoc, $bitmask]
of mBitandI: bitwiseExpr("&")
of mBitorI: bitwiseExpr("|")
of mBitxorI: bitwiseExpr("^")
@@ -1002,7 +1006,8 @@ proc genTry(p: PProc, n: PNode, r: var TCompRes) =
# If some branch requires a local alias introduce it here. This is needed
# since JS cannot do ``catch x as y``.
if excAlias != nil:
excAlias.sym.loc.snippet = mangleName(p.module, excAlias.sym)
ensureMutable excAlias.sym
excAlias.sym.locImpl.snippet = mangleName(p.module, excAlias.sym)
lineF(p, "var $1 = lastJSError;$n", excAlias.sym.loc.snippet)
gen(p, n[i][^1], a)
moveInto(p, a, r)
@@ -1135,7 +1140,8 @@ proc genBlock(p: PProc, n: PNode, r: var TCompRes) =
# named block?
if (n[0].kind != nkSym): internalError(p.config, n.info, "genBlock")
var sym = n[0].sym
sym.loc.k = locOther
ensureMutable sym
sym.locImpl.k = locOther
sym.position = idx+1
let labl = p.unique
lineF(p, "Label$1: {$n", [labl.rope])
@@ -1234,7 +1240,8 @@ proc generateHeader(p: PProc, prc: PSym): Rope =
# to keep it simple
let env = prc.ast[paramsPos].lastSon
assert env.kind == nkSym, "env is missing"
env.sym.loc.snippet = "this"
ensureMutable env.sym
env.sym.locImpl.snippet = "this"
for i in 1..<typ.n.len:
assert(typ.n[i].kind == nkSym)
@@ -1371,12 +1378,15 @@ proc genFieldAddr(p: PProc, n: PNode, r: var TCompRes) =
r.typ = etyBaseIndex
let b = if n.kind == nkHiddenAddr: n[0] else: n
gen(p, b[0], a)
if skipTypes(b[0].typ, abstractVarRange).kind == tyTuple:
if skipTypes(b[0].typ, abstractVarRange + tyTypeClasses).kind == tyTuple:
# ref #25227 about `+ tyTypeClasses`
r.res = makeJSString("Field" & $getFieldPosition(p, b[1]))
else:
if b[1].kind != nkSym: internalError(p.config, b[1].info, "genFieldAddr")
var f = b[1].sym
if f.loc.snippet == "": f.loc.snippet = mangleName(p.module, f)
if f.loc.snippet == "":
ensureMutable f
f.locImpl.snippet = mangleName(p.module, f)
r.res = makeJSString($f.loc.snippet)
internalAssert p.config, a.typ != etyBaseIndex
r.address = a.res
@@ -1404,7 +1414,9 @@ proc genFieldAccess(p: PProc, n: PNode, r: var TCompRes) =
else:
if n[1].kind != nkSym: internalError(p.config, n[1].info, "genFieldAccess")
var f = n[1].sym
if f.loc.snippet == "": f.loc.snippet = mangleName(p.module, f)
if f.loc.snippet == "":
ensureMutable f
f.locImpl.snippet = mangleName(p.module, f)
r.res = "$1.$2" % [r.res, f.loc.snippet]
mkTemp(1)
r.kind = resExpr
@@ -1425,11 +1437,15 @@ proc genCheckedFieldOp(p: PProc, n: PNode, addrTyp: PType, r: var TCompRes) =
# Field symbol
var field = accessExpr[1].sym
internalAssert p.config, field.kind == skField
if field.loc.snippet == "": field.loc.snippet = mangleName(p.module, field)
if field.loc.snippet == "":
ensureMutable field
field.locImpl.snippet = mangleName(p.module, field)
# Discriminant symbol
let disc = checkExpr[2].sym
internalAssert p.config, disc.kind == skField
if disc.loc.snippet == "": disc.loc.snippet = mangleName(p.module, disc)
if disc.loc.snippet == "":
ensureMutable disc
disc.locImpl.snippet = mangleName(p.module, disc)
var setx: TCompRes = default(TCompRes)
gen(p, checkExpr[1], setx)
@@ -1528,7 +1544,7 @@ proc genSymAddr(p: PProc, n: PNode, typ: PType, r: var TCompRes) =
r.res = s.loc.snippet
r.address = ""
r.typ = etyNone
of skVar, skLet, skResult:
of skVar, skLet, skResult, skTemp, skForVar:
r.kind = resExpr
let jsType = mapType(p):
if typ.isNil:
@@ -1586,7 +1602,11 @@ proc genAddr(p: PProc, n: PNode, r: var TCompRes) =
of nkObjDownConv:
gen(p, n[0], r)
of nkHiddenDeref, nkDerefExpr:
gen(p, n[0], r)
if n.kind in {nkAddr, nkHiddenAddr}:
# addr ( deref ( x )) --> x
gen(p, n[0][0], r)
else:
gen(p, n[0], r)
of nkHiddenAddr:
gen(p, n[0], r)
of nkConv:
@@ -1837,7 +1857,9 @@ proc genPatternCall(p: PProc; n: PNode; pat: string; typ: PType;
proc genInfixCall(p: PProc, n: PNode, r: var TCompRes) =
# don't call '$' here for efficiency:
let f = n[0].sym
if f.loc.snippet == "": f.loc.snippet = mangleName(p.module, f)
if f.loc.snippet == "":
ensureMutable f
f.locImpl.snippet = mangleName(p.module, f)
if sfInfixCall in f.flags:
let pat = $n[0].sym.loc.snippet
internalAssert p.config, pat.len > 0
@@ -1996,8 +2018,12 @@ proc createVar(p: PProc, typ: PType, indirect: bool): Rope =
if indirect: result = "[$1]" % [result]
of tyTuple:
result = rope("{")
var first = true
for i in 0..<t.len:
if i > 0: result.add(", ")
# Do not produce code for void types
if isEmptyType(t[i]): continue
if not first: result.add(", ")
first = false
result.addf("Field$1: $2", [i.rope,
createVar(p, t[i], false)])
result.add("}")
@@ -2329,8 +2355,8 @@ proc genMagic(p: PProc, n: PNode, r: var TCompRes) =
r.res = "if (null != $1) { if (null == $2) $2 = $3; else $2 += $3; }" %
[b, lhs.rdLoc, tmp]
else:
let (a, tmp) = maybeMakeTemp(p, n[1], lhs)
r.res = "$1.push.apply($3, $2);" % [a, rhs.rdLoc, tmp]
useMagic(p, "nimAddStrStr")
r.res = "nimAddStrStr($1, $2);" % [lhs.rdLoc, rhs.rdLoc]
r.kind = resExpr
of mAppendSeqElem:
var x, y: TCompRes = default(TCompRes)
@@ -2434,7 +2460,7 @@ proc genMagic(p: PProc, n: PNode, r: var TCompRes) =
binaryExpr(p, n, r, "mnewString",
"""if ($1.length < $2) { for (var i = $3.length; i < $4; ++i) $3.push(0); }
else {$3.length = $4; }""")
of mSetLengthSeq:
of mSetLengthSeq, mSetLengthSeqUninit:
var x, y: TCompRes = default(TCompRes)
gen(p, n[1], x)
gen(p, n[2], y)
@@ -2573,7 +2599,9 @@ proc genObjConstr(p: PProc, n: PNode, r: var TCompRes) =
let val = it[1]
gen(p, val, a)
var f = it[0].sym
if f.loc.snippet == "": f.loc.snippet = mangleName(p.module, f)
if f.loc.snippet == "":
ensureMutable f
f.locImpl.snippet = mangleName(p.module, f)
fieldIDs.incl(lookupFieldAgain(n.typ.skipTypes({tyDistinct}), f).id)
let typ = val.typ.skipTypes(abstractInst)
@@ -2875,6 +2903,8 @@ proc genCast(p: PProc, n: PNode, r: var TCompRes) =
elif dest.kind in tyFloat..tyFloat64:
if src.kind in {tyInt64, tyUInt64} and optJsBigInt64 in p.config.globalOptions:
r.res = "Number($1)" % [r.res]
elif dest.kind == tyChar and (fromInt or fromUint):
r.res = "($1 & 255)" % [r.res]
elif (src.kind == tyPtr and mapType(p, src) == etyObject) and dest.kind == tyPointer:
r.address = r.res
r.res = "null"

View File

@@ -122,7 +122,7 @@ proc genEnumInfo(p: PProc, typ: PType, name: Rope) =
[name, genTypeInfo(p, typ.baseClass)])
proc genTypeInfo(p: PProc, typ: PType): Rope =
let t = typ.skipTypes({tyGenericInst, tyDistinct, tyAlias, tySink, tyOwned})
let t = typ.skipTypes({tyGenericInst, tyDistinct, tyAlias, tySink, tyOwned} + tyUserTypeClasses)
result = "NTI$1" % [rope(t.id)]
if containsOrIncl(p.g.typeInfoGenerated, t.id): return
case t.kind

View File

@@ -150,6 +150,7 @@ template isIterator*(owner: PSym): bool =
proc createEnvObj(g: ModuleGraph; idgen: IdGenerator; owner: PSym; info: TLineInfo): PType =
result = createObj(g, idgen, owner, info, final=false)
result.incl tfFinal
if owner.isIterator:
rawAddField(result, createStateField(g, owner, idgen))
@@ -160,12 +161,24 @@ proc getClosureIterResult*(g: ModuleGraph; iter: PSym; idgen: IdGenerator): PSym
# XXX a bit hacky:
result = newSym(skResult, getIdent(g.cache, ":result"), idgen, iter, iter.info, {})
result.typ = iter.typ.returnType
incl(result.flags, sfUsed)
incl(result.flagsImpl, sfUsed)
iter.ast.add newSymNode(result)
proc closureParams(routine: PSym): PNode =
## The formal parameters node lambda lifting reads and extends. In a
## from-source compilation `routine.ast[paramsPos]` and `routine.typ.n` are the
## very same node (see the `typ.n.len` based position math below). Under IC the
## loaded proc AST omits the parameters (they are kept only in `typ.n`), so
## restore the shared node here.
result = routine.ast[paramsPos]
if (result == nil or result.kind == nkEmpty) and routine.typ != nil and
routine.typ.n != nil and routine.ast.len > paramsPos:
result = routine.typ.n
routine.ast[paramsPos] = result
proc addHiddenParam(routine: PSym, param: PSym) =
assert param.kind == skParam
var params = routine.ast[paramsPos]
var params = closureParams(routine)
# -1 is correct here as param.position is 0 based but we have at position 0
# some nkEffect node:
param.position = routine.typ.n.len-1
@@ -176,7 +189,8 @@ proc addHiddenParam(routine: PSym, param: PSym) =
proc getEnvParam*(routine: PSym): PSym =
if routine.ast.isNil: return nil
let params = routine.ast[paramsPos]
let params = closureParams(routine)
if params == nil or params.len == 0: return nil
let hidden = lastSon(params)
if hidden.kind == nkSym and hidden.sym.kind == skParam and hidden.sym.name.s == paramName:
result = hidden.sym
@@ -199,7 +213,7 @@ proc interestingVar(s: PSym): bool {.inline.} =
proc illegalCapture(s: PSym): bool {.inline.} =
result = classifyViewType(s.typ) != noView or s.kind == skResult
proc isInnerProc(s: PSym): bool =
proc isInnerProc*(s: PSym): bool =
if s.kind in {skProc, skFunc, skMethod, skConverter, skIterator} and s.magic == mNone:
result = s.skipGenericOwner.kind in routineKinds
else:
@@ -215,6 +229,10 @@ proc newAsgnStmt(le, ri: PNode, info: TLineInfo): PNode =
result[0] = le
result[1] = ri
proc markInjectDestructors(s: PSym) {.inline.} =
backendEnsureMutable s
s.flagsImpl.incl sfInjectDestructors
proc makeClosure*(g: ModuleGraph; idgen: IdGenerator; prc: PSym; env: PNode; info: TLineInfo): PNode =
result = newNodeIT(nkClosure, info, prc.typ)
result.add(newSymNode(prc))
@@ -227,7 +245,7 @@ proc makeClosure*(g: ModuleGraph; idgen: IdGenerator; prc: PSym; env: PNode; inf
#if isClosureIterator(result.typ):
createTypeBoundOps(g, nil, result.typ, info, idgen)
if tfHasAsgn in result.typ.flags or optSeqDestructors in g.config.globalOptions:
prc.flags.incl sfInjectDestructors
markInjectDestructors(prc)
template liftingHarmful(conf: ConfigRef; owner: PSym): bool =
## lambda lifting can be harmful for JS-like code generators.
@@ -239,7 +257,7 @@ proc createTypeBoundOpsLL(g: ModuleGraph; refType: PType; info: TLineInfo; idgen
createTypeBoundOps(g, nil, refType.elementType, info, idgen)
createTypeBoundOps(g, nil, refType, info, idgen)
if tfHasAsgn in refType.flags or optSeqDestructors in g.config.globalOptions:
owner.flags.incl sfInjectDestructors
markInjectDestructors(owner)
proc genCreateEnv(env: PNode): PNode =
var c = newNodeIT(nkObjConstr, env.info, env.typ)
@@ -289,7 +307,8 @@ proc markAsClosure(g: ModuleGraph; owner: PSym; n: PNode) =
elif not (owner.typ.isClosure or owner.isNimcall and not owner.isExplicitCallConv or isEnv):
localError(g.config, n.info, "illegal capture '$1' because '$2' has the calling convention: <$3>" %
[s.name.s, owner.name.s, $owner.typ.callConv])
incl(owner.typ.flags, tfCapturesEnv)
unsealForTransform(owner.typ)
incl(owner.typ, tfCapturesEnv)
if not isEnv:
owner.typ.callConv = ccClosure
@@ -335,7 +354,7 @@ proc asOwnedRef(c: var DetectionPass; t: PType): PType =
if optOwnedRefs in c.graph.config.globalOptions:
assert t.kind == tyRef
result = newType(tyOwned, c.idgen, t.owner)
result.flags.incl tfHasOwned
result.incl tfHasOwned
result.rawAddSon t
else:
result = t
@@ -407,17 +426,29 @@ Consider:
proc isTypeOf(n: PNode): bool =
n.kind == nkSym and n.sym.magic in {mTypeOf, mType}
proc isEnvTypeForRoutine(envTyp: PType; routine: PSym): bool =
## True if `envTyp` is (maybe wrapped) env object type owned by `routine`, as
## created by `getEnvTypeForOwner` / `createEnvObj`.
let obj = envTyp.skipTypes({tyOwned, tyRef, tyPtr})
result = obj.kind == tyObject and obj.owner.id == routine.id
proc addClosureParam(c: var DetectionPass; fn: PSym; info: TLineInfo) =
var cp = getEnvParam(fn)
let owner = if fn.kind == skIterator: fn else: fn.skipGenericOwner
let t = c.getEnvTypeForOwner(owner, info)
if cp == nil:
cp = newSym(skParam, getIdent(c.graph.cache, paramName), c.idgen, fn, fn.info)
incl(cp.flags, sfFromGeneric)
incl(cp.flagsImpl, sfFromGeneric)
cp.typ = t
addHiddenParam(fn, cp)
elif cp.typ != t and fn.kind != skIterator:
localError(c.graph.config, fn.info, "internal error: inconsistent environment type")
# Nested `liftLambdas` uses a fresh `DetectionPass`, so `getEnvTypeForOwner`
# can allocate another PType for the same logical env; the hidden param from
# the inner pass is authoritative (bug #21242).
if isEnvTypeForRoutine(cp.typ, owner) and isEnvTypeForRoutine(t, owner):
c.ownerToType[owner.id] = cp.typ
else:
localError(c.graph.config, fn.info, "internal error: inconsistent environment type")
#echo "adding closure to ", fn.name.s
proc iterEnvHasUpField(g: ModuleGraph, iter: PSym): bool =
@@ -609,7 +640,7 @@ proc rawClosureCreation(owner: PSym;
let unowned = c.unownedEnvVars[owner.id]
assert unowned != nil
let env2 = copyTree(env)
env2.typ() = unowned.typ
env2.typ = unowned.typ
result.add newAsgnStmt(unowned, env2, env.info)
createTypeBoundOpsLL(d.graph, unowned.typ, env.info, d.idgen, owner)
@@ -623,7 +654,7 @@ proc rawClosureCreation(owner: PSym;
if owner.kind != skMacro:
createTypeBoundOps(d.graph, nil, fieldAccess.typ, env.info, d.idgen)
if tfHasAsgn in fieldAccess.typ.flags or optSeqDestructors in d.graph.config.globalOptions:
owner.flags.incl sfInjectDestructors
markInjectDestructors(owner)
let upField = lookupInRecord(env.typ.skipTypes({tyOwned, tyRef, tyPtr}).n, getIdent(d.graph.cache, upName))
if upField != nil:
@@ -665,7 +696,7 @@ proc closureCreationForIter(owner: PSym, iter: PNode;
result = newNodeIT(nkStmtListExpr, iter.info, iter.sym.typ)
let iterOwner = iter.sym.skipGenericOwner
var v = newSym(skVar, getIdent(d.graph.cache, envName), d.idgen, iterOwner, iter.info)
incl(v.flags, sfShadowed)
incl(v.flagsImpl, sfShadowed)
v.typ = asOwnedRef(d, getHiddenParam(d.graph, iter.sym).typ)
var vnode: PNode
if iterOwner.isIterator:
@@ -786,7 +817,7 @@ proc liftCapturedVars(n: PNode; owner: PSym; d: var DetectionPass;
let oldInContainer = c.inContainer
c.inContainer = 0
let m = newSymNode(n[namePos].sym)
m.typ() = n.typ
m.typ = n.typ
result = liftCapturedVars(m, owner, d, c)
c.inContainer = oldInContainer
of nkHiddenStdConv:
@@ -974,6 +1005,20 @@ proc liftForLoop*(g: ModuleGraph; body: PNode; idgen: IdGenerator; owner: PSym):
for i in 0..<op.len-1:
result.add op[i]
elif op.kind != nkSym: # might have side effects
# bug #25046
# create a temp for the closure
# var :closureTemp
# :closureTemp = ...
let tempSym = newSym(skLet, getIdent(g.cache, ":closureTemp"), idgen, owner, body.info)
tempSym.typ = call[0].typ
let temp = newSymNode(tempSym)
var v = newNodeI(nkVarSection, body.info)
addVar(v, temp)
result.add(v)
result.add newAsgnStmt(temp, call[0], body.info)
call[0] = temp
var loopBody = newNodeI(nkStmtList, body.info, 3)
var whileLoop = newNodeI(nkWhileStmt, body.info, 2)
whileLoop[0] = newIntTypeNode(1, getSysType(g, body.info, tyBool))

View File

@@ -1,5 +1,4 @@
import std/tables
import ast
import ast, astalgo
type
LayeredIdTableObj* {.acyclic.} = object
@@ -28,14 +27,15 @@ proc shallowCopy*(pt: LayeredIdTable): LayeredIdTable {.inline.} =
## copies only the type bindings of the current layer, but not any parent layers,
## useful for write-only bindings
result = LayeredIdTable(topLayer: pt.topLayer, nextLayer: pt.nextLayer, previousLen: pt.previousLen)
#copyIdTable(result.topLayer, pt.topLayer)
proc currentLen*(pt: LayeredIdTable): int =
## the sum of the cached total binding count of the parents and
## the current binding count, just used to track if bindings were added
pt.previousLen + pt.topLayer.len
pt.previousLen + pt.topLayer.counter
proc newTypeMapLayer*(pt: LayeredIdTable): LayeredIdTable =
result = LayeredIdTable(topLayer: initTable[ItemId, PType](), previousLen: pt.currentLen)
result = LayeredIdTable(topLayer: initTypeMapping(), previousLen: pt.currentLen)
when useRef:
result.nextLayer = pt
else:
@@ -46,12 +46,20 @@ proc setToPreviousLayer*(pt: var LayeredIdTable) {.inline.} =
when useRef:
pt = pt.nextLayer
else:
when defined(gcDestructors):
pt = pt.nextLayer[]
else:
# workaround refc
let tmp = pt.nextLayer[]
pt = tmp
# Must read nextLayer into a temp before destroying pt:
# `pt = pt.nextLayer[]` would call eqcopy(&pt, &(*pt.nextLayer)) which
# decrements pt.nextLayer's rc (freeing it) before reading pt.nextLayer.nextLayer.
let tmp = pt.nextLayer[]
pt = tmp
iterator pairs*(pt: LayeredIdTable): (ItemId, PType) =
var tm = pt
while true:
for (k, v) in idTablePairs(tm.topLayer):
yield (k, v)
if tm.nextLayer == nil:
break
tm.setToPreviousLayer
proc lookup(typeMap: ref LayeredIdTableObj, key: ItemId): PType =
result = nil

View File

@@ -316,6 +316,28 @@ proc getNumber(L: var Lexer, result: var Token) =
L.bufpos = msgPos
lexMessage(L, msgKind, msg % t.literal)
proc checkBitWidth(L: var Lexer, base: NumericalBase, tokType: TokType,
numDigits: int, startpos: int) =
# Check bit width for non-base-10 literals
# Warn if the digit count exceeds what can fit in the target type
let bitsPerDigit = case base
of base2: 1
of base8: 3
of base16: 4
else: raiseAssert "unreachable"
let bitWidth = case tokType
of tkInt8Lit, tkUInt8Lit: 8
of tkInt16Lit, tkUInt16Lit: 16
of tkInt32Lit, tkUInt32Lit: 32
of tkInt64Lit, tkUIntLit, tkIntLit, tkUInt64Lit: 64
else: raiseAssert "unreachable"
# Maximum digits = ceil(bitWidth / bitsPerDigit) = (bitWidth + bitsPerDigit - 1) div bitsPerDigit
let maxDigits = (bitWidth + bitsPerDigit - 1) div bitsPerDigit
if numDigits > maxDigits:
lexMessageLitNum(L,
"number has " & $numDigits & " digits but type only supports " &
$maxDigits & " digits: '$1'", startpos, warnLongLiterals)
var
xi: BiggestInt
isBase10 = true
@@ -491,6 +513,11 @@ proc getNumber(L: var Lexer, result: var Token) =
setNumber result.fNumber, (cast[ptr float64](addr(xi)))[]
else: internalError(L.config, getLineInfo(L), "getNumber")
# Check bit width for non-base-10 literals
# Warn if the digit count exceeds what can fit in the target type
if result.base != base10 and result.tokType in {tkIntLit..tkUInt64Lit} and numDigits > 0:
checkBitWidth(L, result.base, result.tokType, numDigits, startpos)
# Bounds checks. Non decimal literals are allowed to overflow the range of
# the datatype as long as their pattern don't overflow _bitwise_, hence
# below checks of signed sizes against uint*.high is deliberate:
@@ -896,7 +923,7 @@ proc getSymbol(L: var Lexer, tok: var Token) =
tok.tokType = tkSymbol
else:
tok.tokType = TokType(tok.ident.id + ord(tkSymbol))
if suspicious and {optStyleHint, optStyleError} * L.config.globalOptions != {}:
if suspicious and {optStyleHint, optStyleError, optStyleWarning} * L.config.globalOptions != {}:
lintReport(L.config, getLineInfo(L), tok.ident.s.normalize, tok.ident.s)
L.bufpos = pos
@@ -1322,7 +1349,7 @@ proc rawGetTok*(L: var Lexer, tok: var Token) =
lexMessage(L, errGenerated, "invalid token: no whitespace between number and identifier")
of '-':
if L.buf[L.bufpos+1] in {'0'..'9'} and
(L.bufpos-1 == 0 or L.buf[L.bufpos-1] in UnaryMinusWhitelist):
(L.bufpos == 0 or L.buf[L.bufpos-1] in UnaryMinusWhitelist):
# x)-23 # binary minus
# ,-23 # unary minus
# \n-78 # unary minus? Yes.

View File

@@ -40,7 +40,7 @@ template asink*(t: PType): PSym = getAttachedOp(c.g, t, attachedSink)
proc fillBody(c: var TLiftCtx; t: PType; body, x, y: PNode)
proc produceSym(g: ModuleGraph; c: PContext; typ: PType; kind: TTypeAttachedOp;
info: TLineInfo; idgen: IdGenerator; isDistinct = false): PSym
info: TLineInfo; idgen: IdGenerator): PSym
proc createTypeBoundOps*(g: ModuleGraph; c: PContext; orig: PType; info: TLineInfo;
idgen: IdGenerator)
@@ -49,7 +49,7 @@ proc at(a, i: PNode, elemType: PType): PNode =
result = newNodeI(nkBracketExpr, a.info, 2)
result[0] = a
result[1] = i
result.typ() = elemType
result.typ = elemType
proc destructorOverridden(g: ModuleGraph; t: PType): bool =
let op = getAttachedOp(g, t, attachedDestructor)
@@ -68,7 +68,7 @@ proc dotField(x: PNode, f: PSym): PNode =
else:
result[0] = x
result[1] = newSymNode(f, x.info)
result.typ() = f.typ
result.typ = f.typ
proc newAsgnStmt(le, ri: PNode): PNode =
result = newNodeI(nkAsgn, le.info, 2)
@@ -88,10 +88,10 @@ proc defaultOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.add newAsgnStmt(x, y)
elif c.kind == attachedDestructor and c.addMemReset:
let call = genBuiltin(c, mDefault, "default", x)
call.typ() = t
call.typ = t
body.add newAsgnStmt(x, call)
elif c.kind == attachedWasMoved:
body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc genAddr(c: var TLiftCtx; x: PNode): PNode =
if x.kind == nkHiddenDeref:
@@ -105,7 +105,7 @@ proc genWhileLoop(c: var TLiftCtx; i, dest: PNode): PNode =
result = newNodeI(nkWhileStmt, c.info, 2)
let cmp = genBuiltin(c, mLtI, "<", i)
cmp.add genLen(c.g, dest)
cmp.typ() = getSysType(c.g, c.info, tyBool)
cmp.typ = getSysType(c.g, c.info, tyBool)
result[0] = cmp
result[1] = newNodeI(nkStmtList, c.info)
@@ -127,10 +127,10 @@ proc genContainerOf(c: var TLiftCtx; objType: PType, field, x: PSym): PNode =
dotExpr.add newSymNode(field)
let offsetOf = genBuiltin(c, mOffsetOf, "offsetof", dotExpr)
offsetOf.typ() = intType
offsetOf.typ = intType
let minusExpr = genBuiltin(c, mSubI, "-", castExpr1)
minusExpr.typ() = intType
minusExpr.typ = intType
minusExpr.add offsetOf
let objPtr = makePtrType(objType.owner, objType, c.idgen)
@@ -148,7 +148,7 @@ proc destructorCall(c: var TLiftCtx; op: PSym; x: PNode): PNode =
if sfNeverRaises notin op.flags:
c.canRaise = true
if c.addMemReset:
result = newTree(nkStmtList, destroy, genBuiltin(c, mWasMoved, "`=wasMoved`", x))
result = newTree(nkStmtList, destroy, genBuiltin(c, mWasMoved, "wasMoved", x))
else:
result = destroy
@@ -163,12 +163,12 @@ proc fillBodyObj(c: var TLiftCtx; n, body, x, y: PNode; enforceDefaultOp: bool,
if c.filterDiscriminator != nil: return
let f = n.sym
let b = if c.kind == attachedTrace: y else: y.dotField(f)
if (sfCursor in f.flags and c.g.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcHooks}) or
if (sfCursor in f.flags and c.g.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc, gcHooks}) or
enforceDefaultOp:
defaultOp(c, f.typ, body, x.dotField(f), b)
else:
if enforceWasMoved:
body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x.dotField(f))
body.add genBuiltin(c, mWasMoved, "wasMoved", x.dotField(f))
fillBody(c, f.typ, body, x.dotField(f), b)
of nkNilLit: discard
of nkRecCase:
@@ -218,23 +218,38 @@ proc fillBodyObj(c: var TLiftCtx; n, body, x, y: PNode; enforceDefaultOp: bool,
fillBodyObj(c, n[0], body, x, y, enforceDefaultOp = false)
c.filterDiscriminator = oldfilterDiscriminator
of nkRecList:
for t in items(n): fillBodyObj(c, t, body, x, y, enforceDefaultOp, enforceWasMoved)
# destroys in reverse order #24719
if c.kind == attachedDestructor:
for i in countdown(n.len-1, 0):
fillBodyObj(c, n[i], body, x, y, enforceDefaultOp, enforceWasMoved)
else:
for t in items(n): fillBodyObj(c, t, body, x, y, enforceDefaultOp, enforceWasMoved)
else:
illFormedAstLocal(n, c.g.config)
proc fillBodyObjTImpl(c: var TLiftCtx; t: PType, body, x, y: PNode) =
if t.baseClass != nil:
let dest = newNodeIT(nkHiddenSubConv, c.info, t.baseClass)
dest.add newNodeI(nkEmpty, c.info)
dest.add x
var src = y
if c.kind in {attachedAsgn, attachedDeepCopy, attachedSink}:
src = newNodeIT(nkHiddenSubConv, c.info, t.baseClass)
src.add newNodeI(nkEmpty, c.info)
src.add y
template fillBase =
if t.baseClass != nil:
let dest = newNodeIT(nkHiddenSubConv, c.info, t.baseClass)
dest.add newNodeI(nkEmpty, c.info)
dest.add x
var src = y
if c.kind in {attachedAsgn, attachedDeepCopy, attachedSink}:
src = newNodeIT(nkHiddenSubConv, c.info, t.baseClass)
src.add newNodeI(nkEmpty, c.info)
src.add y
fillBody(c, skipTypes(t.baseClass, abstractPtrs), body, dest, src)
fillBodyObj(c, t.n, body, x, y, enforceDefaultOp = false)
fillBody(c, skipTypes(t.baseClass, abstractPtrs), body, dest, src)
template fillFields =
fillBodyObj(c, t.n, body, x, y, enforceDefaultOp = false)
if c.kind == attachedDestructor:
# destroys in reverse order #24719
fillFields()
fillBase()
else:
fillBase()
fillFields()
proc fillBodyObjT(c: var TLiftCtx; t: PType, body, x, y: PNode) =
var hasCase = isCaseObj(t.n)
@@ -265,11 +280,11 @@ proc fillBodyObjT(c: var TLiftCtx; t: PType, body, x, y: PNode) =
# because the wasMoved(dest) call would zero out src, if dest aliases src.
var cond = newTree(nkCall, newSymNode(c.g.getSysMagic(c.info, "==", mEqRef)),
newTreeIT(nkAddr, c.info, makePtrType(c.fn, x.typ, c.idgen), x), newTreeIT(nkAddr, c.info, makePtrType(c.fn, y.typ, c.idgen), y))
cond.typ() = getSysType(c.g, x.info, tyBool)
cond.typ = getSysType(c.g, x.info, tyBool)
body.add genIf(c, cond, newTreeI(nkReturnStmt, c.info, newNodeI(nkEmpty, c.info)))
var temp = newSym(skTemp, getIdent(c.g.cache, lowerings.genPrefix), c.idgen, c.fn, c.info)
temp.typ = x.typ
incl(temp.flags, sfFromGeneric)
incl(temp, sfFromGeneric)
var v = newNodeI(nkVarSection, c.info)
let blob = newSymNode(temp)
v.addVar(blob, x)
@@ -277,7 +292,8 @@ proc fillBodyObjT(c: var TLiftCtx; t: PType, body, x, y: PNode) =
#body.add newAsgnStmt(blob, x)
var wasMovedCall = newNodeI(nkCall, c.info)
wasMovedCall.add(newSymNode(createMagic(c.g, c.idgen, "`=wasMoved`", mWasMoved)))
wasMovedCall.add(newSymNode(createMagic(c.g, c.idgen, "wasMoved", mWasMoved)))
wasMovedCall.add x # mWasMoved does not take the address
body.add wasMovedCall
@@ -296,7 +312,7 @@ proc fillBodyObjT(c: var TLiftCtx; t: PType, body, x, y: PNode) =
proc boolLit*(g: ModuleGraph; info: TLineInfo; value: bool): PNode =
result = newIntLit(g, info, ord value)
result.typ() = getSysType(g, info, tyBool)
result.typ = getSysType(g, info, tyBool)
proc getCycleParam(c: TLiftCtx): PNode =
assert c.kind in {attachedAsgn, attachedDup}
@@ -364,6 +380,10 @@ proc requiresDestructor(c: TLiftCtx; t: PType): bool {.inline.} =
proc instantiateGeneric(c: var TLiftCtx; op: PSym; t, typeInst: PType): PSym =
if c.c != nil and typeInst != nil:
result = c.c.instTypeBoundOp(c.c, op, typeInst, c.info, attachedAsgn, 1)
elif typeInst != nil and getAttachedOp(c.g, typeInst, c.kind) != nil:
# c.c == nil in lambdalifting
# hooks are already insted
result = getAttachedOp(c.g, typeInst, c.kind)
else:
localError(c.g.config, c.info,
"cannot generate destructor for generic type: " & typeToString(t))
@@ -377,7 +397,8 @@ proc considerAsgnOrSink(c: var TLiftCtx; t: PType; body, x, y: PNode;
if op != nil and op != c.fn and
(sfOverridden in op.flags or destructorOverridden):
if sfError in op.flags:
incl c.fn.flags, sfError
ensureMutable c.fn
incl c.fn.flagsImpl, sfError
#else:
# markUsed(c.g.config, c.info, op, c.g.usageSym)
onUse(c.info, op)
@@ -403,7 +424,8 @@ proc considerAsgnOrSink(c: var TLiftCtx; t: PType; body, x, y: PNode;
if op == nil:
op = produceSym(c.g, c.c, t, c.kind, c.info, c.idgen)
if sfError in op.flags:
incl c.fn.flags, sfError
ensureMutable c.fn
incl c.fn.flagsImpl, sfError
#else:
# markUsed(c.g.config, c.info, op, c.g.usageSym)
onUse(c.info, op)
@@ -519,7 +541,7 @@ proc considerUserDefinedOp(c: var TLiftCtx; t: PType; body, x, y: PNode): bool =
proc declareCounter(c: var TLiftCtx; body: PNode; first: BiggestInt): PNode =
var temp = newSym(skTemp, getIdent(c.g.cache, lowerings.genPrefix), c.idgen, c.fn, c.info)
temp.typ = getSysType(c.g, body.info, tyInt)
incl(temp.flags, sfFromGeneric)
incl(temp.flagsImpl, sfFromGeneric)
var v = newNodeI(nkVarSection, c.info)
result = newSymNode(temp)
@@ -529,13 +551,29 @@ proc declareCounter(c: var TLiftCtx; body: PNode; first: BiggestInt): PNode =
proc declareTempOf(c: var TLiftCtx; body: PNode; value: PNode): PNode =
var temp = newSym(skTemp, getIdent(c.g.cache, lowerings.genPrefix), c.idgen, c.fn, c.info)
temp.typ = value.typ
incl(temp.flags, sfFromGeneric)
incl(temp.flagsImpl, sfFromGeneric)
var v = newNodeI(nkVarSection, c.info)
result = newSymNode(temp)
v.addVar(result, value)
body.add v
proc considerInferDupFromCopy(c: var TLiftCtx; t: PType; body, x, y: PNode): bool =
## For `=dup`, if no explicit hook exists, try to infer from `=copy` hook
## to maintain backward compatibility. Returns true if inference was applied.
if c.kind == attachedDup:
var op2 = getAttachedOp(c.g, t, attachedAsgn)
if op2 != nil and sfOverridden in op2.flags:
#markUsed(c.g.config, c.info, op, c.g.usageSym)
onUse(c.info, op2)
body.add genBuiltin(c, mWasMoved, "wasMoved", x)
body.add newHookCall(c, op2, x, y)
result = true
else:
result = false
else:
result = false
proc addIncStmt(c: var TLiftCtx; body, i: PNode) =
let incCall = genBuiltin(c, mInc, "inc", i)
incCall.add lowerings.newIntLit(c.g, c.info, 1)
@@ -545,19 +583,21 @@ proc newSeqCall(c: var TLiftCtx; x, y: PNode): PNode =
# don't call genAddr(c, x) here:
result = genBuiltin(c, mNewSeq, "newSeq", x)
let lenCall = genBuiltin(c, mLengthSeq, "len", y)
lenCall.typ() = getSysType(c.g, x.info, tyInt)
lenCall.typ = getSysType(c.g, x.info, tyInt)
result.add lenCall
proc setLenStrCall(c: var TLiftCtx; x, y: PNode): PNode =
let lenCall = genBuiltin(c, mLengthStr, "len", y)
lenCall.typ() = getSysType(c.g, x.info, tyInt)
lenCall.typ = getSysType(c.g, x.info, tyInt)
result = genBuiltin(c, mSetLengthStr, "setLen", x) # genAddr(g, x))
result.add lenCall
proc setLenSeqCall(c: var TLiftCtx; t: PType; x, y: PNode): PNode =
proc setLenSeqCall(c: var TLiftCtx; t: PType; x, y: PNode; noinit = false): PNode =
let lenCall = genBuiltin(c, mLengthSeq, "len", y)
lenCall.typ() = getSysType(c.g, x.info, tyInt)
var op = getSysMagic(c.g, x.info, "setLen", mSetLengthSeq)
lenCall.typ = getSysType(c.g, x.info, tyInt)
let name = if noinit: "setLenUninit" else: "setLen"
let magic = if noinit: mSetLengthSeqUninit else: mSetLengthSeq
var op = getSysMagic(c.g, x.info, name, magic)
op = instantiateGeneric(c, op, t, t)
result = newTree(nkCall, newSymNode(op, x.info), x, lenCall)
@@ -579,14 +619,38 @@ proc checkSelfAssignment(c: var TLiftCtx; t: PType; body, x, y: PNode) =
newTreeIT(nkAddr, c.info, makePtrType(c.fn, x.typ, c.idgen), x),
newTreeIT(nkAddr, c.info, makePtrType(c.fn, y.typ, c.idgen), y)
)
cond.typ() = getSysType(c.g, c.info, tyBool)
cond.typ = getSysType(c.g, c.info, tyBool)
body.add genIf(c, cond, newTreeI(nkReturnStmt, c.info, newNodeI(nkEmpty, c.info)))
proc genBulkCopySeq(c: var TLiftCtx; t: PType; body, x, y: PNode) =
## Generates a call to nimCopySeqPayload for bulk memcpy of seq data.
let elemType = t.elementType
let sym = magicsys.getCompilerProc(c.g, "nimCopySeqPayload")
if sym == nil:
localError(c.g.config, c.info, "system module needs: nimCopySeqPayload")
return
var sizeOf = genBuiltin(c, mSizeOf, "sizeof", newNodeIT(nkType, c.info, elemType))
sizeOf.typ = getSysType(c.g, c.info, tyInt)
var alignOf = genBuiltin(c, mAlignOf, "alignof", newNodeIT(nkType, c.info, elemType))
alignOf.typ = getSysType(c.g, c.info, tyInt)
let call = newNodeI(nkCall, c.info)
call.add newSymNode(sym)
call.add newTreeIT(nkAddr, c.info, makePtrType(c.fn, x.typ, c.idgen), x)
call.add newTreeIT(nkAddr, c.info, makePtrType(c.fn, y.typ, c.idgen), y)
call.add sizeOf
call.add alignOf
call.typ = sym.typ.returnType
body.add call
proc fillSeqOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
case c.kind
of attachedDup:
body.add setLenSeqCall(c, t, x, y)
forallElements(c, t, body, x, y)
let bulkCopy = supportsCopyMem(t.elementType)
body.add setLenSeqCall(c, t, x, y, noinit = bulkCopy)
if bulkCopy:
genBulkCopySeq(c, t, body, x, y)
else:
forallElements(c, t, body, x, y)
of attachedAsgn, attachedDeepCopy:
# we generate:
# if x.p == y.p:
@@ -595,9 +659,14 @@ proc fillSeqOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
# var i = 0
# while i < y.len: dest[i] = y[i]; inc(i)
# This is usually more efficient than a destroy/create pair.
# For trivially copyable types, use bulk copyMem instead of element loop.
checkSelfAssignment(c, t, body, x, y)
body.add setLenSeqCall(c, t, x, y)
forallElements(c, t, body, x, y)
let bulkCopy = supportsCopyMem(t.elementType)
body.add setLenSeqCall(c, t, x, y, noinit = bulkCopy)
if bulkCopy:
genBulkCopySeq(c, t, body, x, y)
else:
forallElements(c, t, body, x, y)
of attachedSink:
let moveCall = genBuiltin(c, mMove, "move", x)
moveCall.add y
@@ -612,7 +681,7 @@ proc fillSeqOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if canFormAcycle(c.g, t.elemType):
# follow all elements:
forallElements(c, t, body, x, y)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc useSeqOrStrOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
createTypeBoundOps(c.g, c.c, t, body.info, c.idgen)
@@ -642,6 +711,11 @@ proc useSeqOrStrOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
doAssert t.asink != nil
body.add newHookCall(c, t.asink, x, y)
of attachedDestructor:
when defined(icDbg):
if t.destructor == nil:
echo "MISSING destructor: ", typeToString(t), " kind=", t.kind,
" itemId=", t.itemId, " uniqueId=", t.uniqueId, " state=", t.state,
" owner=", (if t.owner != nil: t.owner.name.s else: "nil")
doAssert t.destructor != nil
body.add destructorCall(c, t.destructor, x)
of attachedTrace:
@@ -650,7 +724,7 @@ proc useSeqOrStrOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if op == nil:
return # protect from recursion
body.add newHookCall(c, op, x, y)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
of attachedDup:
# XXX: replace these with assertions.
let op = getAttachedOp(c.g, t, c.kind)
@@ -663,16 +737,23 @@ proc fillStrOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
of attachedAsgn, attachedDeepCopy, attachedDup:
body.add callCodegenProc(c.g, "nimAsgnStrV2", c.info, genAddr(c, x), y)
of attachedSink:
let moveCall = genBuiltin(c, mMove, "move", x)
moveCall.add y
doAssert t.destructor != nil
moveCall.add destructorCall(c, t.destructor, x)
body.add moveCall
if c.g.config.usesSso():
# SmallString: destroy old dst, then bit-copy src (no rc increment — this is a move).
# No .p aliasing check needed; rc-based destroy handles COW sharing correctly.
doAssert t.destructor != nil
body.add destructorCall(c, t.destructor, x)
body.add newAsgnStmt(x, y)
else:
let moveCall = genBuiltin(c, mMove, "move", x)
moveCall.add y
doAssert t.destructor != nil
moveCall.add destructorCall(c, t.destructor, x)
body.add moveCall
of attachedDestructor:
body.add genBuiltin(c, mDestroy, "destroy", x)
of attachedTrace:
discard "strings are atomic and have no inner elements that are to trace"
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc cyclicType*(g: ModuleGraph, t: PType): bool =
case t.kind
@@ -699,14 +780,43 @@ proc atomicRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
dest[] = source
decRef tmp
For YRC the write barrier is more complicated still and must be:
let tmp = dest
# assignment must come first so that the collector sees the most-recent graph:
atomic: dest[] = source
# Then teach the cycle collector about the changes edge (these use locks, see yrc.nim):
incRef source
decRef tmp
This is implemented as a single runtime call (nimAsgnYrc / nimSinkYrc).
]#
var actions = newNodeI(nkStmtList, c.info)
let elemType = t.elementType
createTypeBoundOps(c.g, c.c, elemType, c.info, c.idgen)
let isCyclic = c.g.config.selectedGC == gcOrc and types.canFormAcycle(c.g, elemType)
let isInheritableAcyclicRef = c.g.config.selectedGC == gcOrc and
# YRC uses dedicated runtime procs for the entire write barrier:
if c.g.config.selectedGC == gcYrc:
let desc =
if isFinal(elemType):
let ti = genBuiltin(c, mGetTypeInfoV2, "getTypeInfoV2", newNodeIT(nkType, x.info, elemType))
ti.typ = getSysType(c.g, c.info, tyPointer)
ti
else:
newNodeIT(nkNilLit, c.info, getSysType(c.g, c.info, tyPointer))
case c.kind
of attachedAsgn, attachedDup:
body.add callCodegenProc(c.g, "nimAsgnYrc", c.info, genAddr(c, x), y, desc)
return
of attachedSink:
body.add callCodegenProc(c.g, "nimSinkYrc", c.info, genAddr(c, x), y, desc)
return
else: discard # fall through for destructor, trace, wasMoved
let isCyclic = c.g.config.selectedGC in {gcOrc, gcYrc} and types.canFormAcycle(c.g, elemType)
let isInheritableAcyclicRef = c.g.config.selectedGC in {gcOrc, gcYrc} and
(not isPureObject(elemType)) and
tfAcyclic in skipTypes(elemType, abstractInst+{tyOwned}-{tyTypeDesc}).flags
# dynamic Acyclic refs need to use dyn decRef
@@ -720,7 +830,7 @@ proc atomicRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if isFinal(elemType):
addDestructorCall(c, elemType, actions, genDeref(tmp, nkDerefExpr))
var alignOf = genBuiltin(c, mAlignOf, "alignof", newNodeIT(nkType, c.info, elemType))
alignOf.typ() = getSysType(c.g, c.info, tyInt)
alignOf.typ = getSysType(c.g, c.info, tyInt)
actions.add callCodegenProc(c.g, "nimRawDispose", c.info, tmp, alignOf)
else:
addDestructorCall(c, elemType, newNodeI(nkStmtList, c.info), genDeref(tmp, nkDerefExpr))
@@ -730,7 +840,7 @@ proc atomicRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if isCyclic:
if isFinal(elemType):
let typInfo = genBuiltin(c, mGetTypeInfoV2, "getTypeInfoV2", newNodeIT(nkType, x.info, elemType))
typInfo.typ() = getSysType(c.g, c.info, tyPointer)
typInfo.typ = getSysType(c.g, c.info, tyPointer)
cond = callCodegenProc(c.g, "nimDecRefIsLastCyclicStatic", c.info, tmp, typInfo)
else:
cond = callCodegenProc(c.g, "nimDecRefIsLastCyclicDyn", c.info, tmp)
@@ -738,7 +848,7 @@ proc atomicRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
cond = callCodegenProc(c.g, "nimDecRefIsLastDyn", c.info, x)
else:
cond = callCodegenProc(c.g, "nimDecRefIsLast", c.info, x)
cond.typ() = getSysType(c.g, x.info, tyBool)
cond.typ = getSysType(c.g, x.info, tyBool)
case c.kind
of attachedSink:
@@ -765,13 +875,13 @@ proc atomicRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if isCyclic:
if isFinal(elemType):
let typInfo = genBuiltin(c, mGetTypeInfoV2, "getTypeInfoV2", newNodeIT(nkType, x.info, elemType))
typInfo.typ() = getSysType(c.g, c.info, tyPointer)
typInfo.typ = getSysType(c.g, c.info, tyPointer)
body.add callCodegenProc(c.g, "nimTraceRef", c.info, genAddrOf(x, c.idgen), typInfo, y)
else:
# If the ref is polymorphic we have to account for this
body.add callCodegenProc(c.g, "nimTraceRefDyn", c.info, genAddrOf(x, c.idgen), y)
#echo "can follow ", elemType, " static ", isFinal(elemType)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
of attachedDup:
if isCyclic:
body.add newAsgnStmt(x, y)
@@ -786,9 +896,28 @@ proc atomicClosureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
## Closures are really like refs except they always use a virtual destructor
## and we need to do the refcounting only on the ref field which we call 'xenv':
let xenv = genBuiltin(c, mAccessEnv, "accessEnv", x)
xenv.typ() = getSysType(c.g, c.info, tyPointer)
xenv.typ = getSysType(c.g, c.info, tyPointer)
let isCyclic = c.g.config.selectedGC == gcOrc
# Closures are (fnPtr, env) pairs. nimAsgnYrc/nimSinkYrc handle the env pointer
# (atomic store + buffered inc/dec). We also need newAsgnStmt to copy the fnPtr.
if c.g.config.selectedGC == gcYrc:
let nilDesc = newNodeIT(nkNilLit, c.info, getSysType(c.g, c.info, tyPointer))
let yenv = genBuiltin(c, mAccessEnv, "accessEnv", y)
yenv.typ = getSysType(c.g, c.info, tyPointer)
case c.kind
of attachedAsgn, attachedDup:
# nimAsgnYrc: save old env, atomic store new env, inc new env, dec old env
body.add callCodegenProc(c.g, "nimAsgnYrc", c.info, genAddr(c, xenv), yenv, nilDesc)
# Raw struct copy to also update the function pointer (env write is redundant but benign)
body.add newAsgnStmt(x, y)
return
of attachedSink:
body.add callCodegenProc(c.g, "nimSinkYrc", c.info, genAddr(c, xenv), yenv, nilDesc)
body.add newAsgnStmt(x, y)
return
else: discard # fall through for destructor, trace, wasMoved
let isCyclic = c.g.config.selectedGC in {gcOrc, gcYrc}
let tmp =
if isCyclic and c.kind in {attachedAsgn, attachedSink, attachedDup}:
declareTempOf(c, body, xenv)
@@ -802,7 +931,7 @@ proc atomicClosureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if isCyclic: "nimDecRefIsLastCyclicDyn"
else: "nimDecRefIsLast"
let cond = callCodegenProc(c.g, decRefProc, c.info, tmp)
cond.typ() = getSysType(c.g, x.info, tyBool)
cond.typ = getSysType(c.g, x.info, tyBool)
case c.kind
of attachedSink:
@@ -814,19 +943,18 @@ proc atomicClosureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.add newAsgnStmt(x, y)
of attachedAsgn:
let yenv = genBuiltin(c, mAccessEnv, "accessEnv", y)
yenv.typ() = getSysType(c.g, c.info, tyPointer)
yenv.typ = getSysType(c.g, c.info, tyPointer)
if isCyclic:
body.add genIf(c, yenv, callCodegenProc(c.g, "nimIncRefCyclic", c.info, yenv, getCycleParam(c)))
body.add newAsgnStmt(x, y)
body.add genIf(c, cond, actions)
else:
body.add genIf(c, yenv, callCodegenProc(c.g, "nimIncRef", c.info, yenv))
body.add genIf(c, cond, actions)
body.add newAsgnStmt(x, y)
of attachedDup:
let yenv = genBuiltin(c, mAccessEnv, "accessEnv", y)
yenv.typ() = getSysType(c.g, c.info, tyPointer)
yenv.typ = getSysType(c.g, c.info, tyPointer)
if isCyclic:
body.add newAsgnStmt(x, y)
body.add genIf(c, yenv, callCodegenProc(c.g, "nimIncRefCyclic", c.info, yenv, getCycleParam(c)))
@@ -838,7 +966,7 @@ proc atomicClosureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
of attachedDeepCopy: assert(false, "cannot happen")
of attachedTrace:
body.add callCodegenProc(c.g, "nimTraceRefDyn", c.info, genAddrOf(xenv, c.idgen), y)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc weakrefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
case c.kind
@@ -866,7 +994,7 @@ proc weakrefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.sons.insert(des, 0)
of attachedDeepCopy: assert(false, "cannot happen")
of attachedTrace: discard
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc ownedRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
var actions = newNodeI(nkStmtList, c.info)
@@ -878,7 +1006,7 @@ proc ownedRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if isFinal(elemType):
addDestructorCall(c, elemType, actions, genDeref(x, nkDerefExpr))
var alignOf = genBuiltin(c, mAlignOf, "alignof", newNodeIT(nkType, c.info, elemType))
alignOf.typ() = getSysType(c.g, c.info, tyInt)
alignOf.typ = getSysType(c.g, c.info, tyInt)
actions.add callCodegenProc(c.g, "nimRawDispose", c.info, x, alignOf)
else:
addDestructorCall(c, elemType, newNodeI(nkStmtList, c.info), genDeref(x, nkDerefExpr))
@@ -894,21 +1022,21 @@ proc ownedRefOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.add genIf(c, x, actions)
of attachedDeepCopy: assert(false, "cannot happen")
of attachedTrace: discard
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc closureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
if c.kind == attachedDeepCopy:
# a big problem is that we don't know the environment's type here, so we
# have to go through some indirection; we delegate this to the codegen:
let call = newNodeI(nkCall, c.info, 2)
call.typ() = t
call.typ = t
call[0] = newSymNode(createMagic(c.g, c.idgen, "deepCopy", mDeepCopy))
call[1] = y
body.add newAsgnStmt(x, call)
elif (optOwnedRefs in c.g.config.globalOptions and
optRefCheck in c.g.config.options) or c.g.config.selectedGC in {gcArc, gcAtomicArc, gcOrc}:
optRefCheck in c.g.config.options) or c.g.config.selectedGC in {gcArc, gcAtomicArc, gcOrc, gcYrc}:
let xx = genBuiltin(c, mAccessEnv, "accessEnv", x)
xx.typ() = getSysType(c.g, c.info, tyPointer)
xx.typ = getSysType(c.g, c.info, tyPointer)
case c.kind
of attachedSink:
# we 'nil' y out afterwards so we *need* to take over its reference
@@ -917,13 +1045,13 @@ proc closureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.add newAsgnStmt(x, y)
of attachedAsgn:
let yy = genBuiltin(c, mAccessEnv, "accessEnv", y)
yy.typ() = getSysType(c.g, c.info, tyPointer)
yy.typ = getSysType(c.g, c.info, tyPointer)
body.add genIf(c, yy, callCodegenProc(c.g, "nimIncRef", c.info, yy))
body.add genIf(c, xx, callCodegenProc(c.g, "nimDecWeakRef", c.info, xx))
body.add newAsgnStmt(x, y)
of attachedDup:
let yy = genBuiltin(c, mAccessEnv, "accessEnv", y)
yy.typ() = getSysType(c.g, c.info, tyPointer)
yy.typ = getSysType(c.g, c.info, tyPointer)
body.add newAsgnStmt(x, y)
body.add genIf(c, yy, callCodegenProc(c.g, "nimIncRef", c.info, yy))
of attachedDestructor:
@@ -934,11 +1062,11 @@ proc closureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.sons.insert(des, 0)
of attachedDeepCopy: assert(false, "cannot happen")
of attachedTrace: discard
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc ownedClosureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
let xx = genBuiltin(c, mAccessEnv, "accessEnv", x)
xx.typ() = getSysType(c.g, c.info, tyPointer)
xx.typ = getSysType(c.g, c.info, tyPointer)
var actions = newNodeI(nkStmtList, c.info)
#discard addDestructorCall(c, elemType, newNodeI(nkStmtList, c.info), genDeref(xx))
actions.add callCodegenProc(c.g, "nimDestroyAndDispose", c.info, xx)
@@ -952,16 +1080,25 @@ proc ownedClosureOp(c: var TLiftCtx; t: PType; body, x, y: PNode) =
body.add genIf(c, xx, actions)
of attachedDeepCopy: assert(false, "cannot happen")
of attachedTrace: discard
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
of attachedWasMoved: body.add genBuiltin(c, mWasMoved, "wasMoved", x)
proc fillBody(c: var TLiftCtx; t: PType; body, x, y: PNode) =
case t.kind
of tyNone, tyEmpty, tyVoid: discard
of tyUncheckedArray:
# An UncheckedArray has no known length, so it cannot be copied, moved or
# destroyed as a value: it only ever lives behind a pointer and its bytes
# are managed manually (element ops for seqs/strings go through the
# seq/string hooks, which know the length). Emitting `x = y` for it (as the
# pointer-like group below does) produces an assignment of an unsized array,
# which the C backend cannot lower (genAssignment: tyUncheckedArray). So all
# value hooks for it are no-ops.
discard
of tyPointer, tySet, tyBool, tyChar, tyEnum, tyInt..tyUInt64, tyCstring,
tyPtr, tyUncheckedArray, tyVar, tyLent:
tyPtr, tyVar, tyLent:
defaultOp(c, t, body, x, y)
of tyRef:
if c.g.config.selectedGC in {gcArc, gcOrc, gcAtomicArc}:
if c.g.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc}:
atomicRefOp(c, t, body, x, y)
elif (optOwnedRefs in c.g.config.globalOptions and
optRefCheck in c.g.config.options):
@@ -970,7 +1107,7 @@ proc fillBody(c: var TLiftCtx; t: PType; body, x, y: PNode) =
defaultOp(c, t, body, x, y)
of tyProc:
if t.callConv == ccClosure:
if c.g.config.selectedGC in {gcArc, gcOrc, gcAtomicArc}:
if c.g.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc}:
atomicClosureOp(c, t, body, x, y)
else:
closureOp(c, t, body, x, y)
@@ -1002,9 +1139,13 @@ proc fillBody(c: var TLiftCtx; t: PType; body, x, y: PNode) =
# 'selectedGC' here to determine if we have the new runtime.
discard considerUserDefinedOp(c, t, body, x, y)
elif tfHasAsgn in t.flags:
# seqs with elements using custom hooks in refc
if c.kind in {attachedAsgn, attachedSink, attachedDeepCopy}:
body.add newSeqCall(c, x, y)
forallElements(c, t, body, x, y)
if c.kind == attachedWasMoved:
body.add genBuiltin(c, mWasMoved, "wasMoved", x)
else:
forallElements(c, t, body, x, y)
else:
defaultOp(c, t, body, x, y)
of tyString:
@@ -1021,23 +1162,18 @@ proc fillBody(c: var TLiftCtx; t: PType; body, x, y: PNode) =
of {attachedAsgn, attachedSink, attachedDup}:
body.add newAsgnStmt(x, y)
of attachedWasMoved:
body.add genBuiltin(c, mWasMoved, "`=wasMoved`", x)
body.add genBuiltin(c, mWasMoved, "wasMoved", x)
else:
fillBodyObjT(c, t, body, x, y)
elif tfUnion in t.flags: # bug #25236
defaultOp(c, t, body, x, y)
else:
if c.kind == attachedDup:
var op2 = getAttachedOp(c.g, t, attachedAsgn)
if op2 != nil and sfOverridden in op2.flags:
#markUsed(c.g.config, c.info, op, c.g.usageSym)
onUse(c.info, op2)
body.add newHookCall(c, t.assignment, x, y)
else:
fillBodyObjT(c, t, body, x, y)
else:
if not considerInferDupFromCopy(c, t, body, x, y):
fillBodyObjT(c, t, body, x, y)
of tyDistinct:
if not considerUserDefinedOp(c, t, body, x, y):
fillBody(c, t.elementType, body, x, y)
if not considerInferDupFromCopy(c, t, body, x, y):
fillBody(c, t.elementType, body, x, y)
of tyTuple:
fillBodyTup(c, t, body, x, y)
of tyVarargs, tyOpenArray:
@@ -1063,9 +1199,7 @@ proc produceSymDistinctType(g: ModuleGraph; c: PContext; typ: PType;
assert typ.kind == tyDistinct
let baseType = typ.elementType
if getAttachedOp(g, baseType, kind) == nil:
# TODO: fixme `isDistinct` is a fix for #23552; remove it after
# `-d:nimPreviewNonVarDestructor` becomes the default
discard produceSym(g, c, baseType, kind, info, idgen, isDistinct = true)
discard produceSym(g, c, baseType, kind, info, idgen)
result = getAttachedOp(g, baseType, kind)
setAttachedOp(g, idgen.module, typ, kind, result)
@@ -1086,7 +1220,7 @@ proc symDupPrototype(g: ModuleGraph; typ: PType; owner: PSym; kind: TTypeAttache
result.typ.addParam src
if g.config.selectedGC == gcOrc and
if g.config.selectedGC in {gcOrc, gcYrc} and
cyclicType(g, typ.skipTypes(abstractInst)):
let cycleParam = newSym(skParam, getIdent(g.cache, "cyclic"),
idgen, result, info)
@@ -1100,11 +1234,11 @@ proc symDupPrototype(g: ModuleGraph; typ: PType; owner: PSym; kind: TTypeAttache
n[bodyPos] = newNodeI(nkStmtList, info)
n[resultPos] = newSymNode(res)
result.ast = n
incl result.flags, sfFromGeneric
incl result.flags, sfGeneratedOp
incl result.flagsImpl, {sfFromGeneric, sfGeneratedOp}
setHookDisamb(g, result, AttachedOpToStr[kind], typ)
proc symPrototype(g: ModuleGraph; typ: PType; owner: PSym; kind: TTypeAttachedOp;
info: TLineInfo; idgen: IdGenerator; isDiscriminant = false; isDistinct = false): PSym =
info: TLineInfo; idgen: IdGenerator; isDiscriminant = false): PSym =
if kind == attachedDup:
return symDupPrototype(g, typ, owner, kind, info, idgen)
@@ -1114,8 +1248,8 @@ proc symPrototype(g: ModuleGraph; typ: PType; owner: PSym; kind: TTypeAttachedOp
let src = newSym(skParam, getIdent(g.cache, if kind == attachedTrace: "env" else: "src"),
idgen, result, info)
if kind == attachedDestructor and g.config.selectedGC in {gcArc, gcOrc, gcAtomicArc} and
((g.config.isDefined("nimPreviewNonVarDestructor") and not isDiscriminant) or (typ.kind in {tyRef, tyString, tySequence} and not isDistinct)):
if kind == attachedDestructor and g.config.selectedGC in {gcArc, gcOrc, gcYrc, gcAtomicArc} and
((g.config.isDefined("nimPreviewNonVarDestructor") and not isDiscriminant) or (typ.kind in {tyRef, tyString, tySequence})):
dest.typ = typ
else:
dest.typ = makeVarType(typ.owner, typ, idgen)
@@ -1130,7 +1264,7 @@ proc symPrototype(g: ModuleGraph; typ: PType; owner: PSym; kind: TTypeAttachedOp
if kind notin {attachedDestructor, attachedWasMoved}:
result.typ.addParam src
if kind == attachedAsgn and g.config.selectedGC == gcOrc and
if kind == attachedAsgn and g.config.selectedGC in {gcOrc, gcYrc} and
cyclicType(g, typ.skipTypes(abstractInst)):
let cycleParam = newSym(skParam, getIdent(g.cache, "cyclic"),
idgen, result, info)
@@ -1143,27 +1277,41 @@ proc symPrototype(g: ModuleGraph; typ: PType; owner: PSym; kind: TTypeAttachedOp
n[paramsPos] = result.typ.n
n[bodyPos] = newNodeI(nkStmtList, info)
result.ast = n
incl result.flags, sfFromGeneric
incl result.flags, sfGeneratedOp
incl result.flagsImpl, sfFromGeneric
incl result.flagsImpl, sfGeneratedOp
if kind == attachedWasMoved:
incl result.flags, sfNoSideEffect
incl result.typ.flags, tfNoSideEffect
incl result.flagsImpl, sfNoSideEffect
incl result.typ, tfNoSideEffect
if not isDiscriminant:
# discriminant destructors derive their body from the enclosing object
# AND the selected field; their key is set at the call site
setHookDisamb(g, result, AttachedOpToStr[kind], typ)
proc genTypeFieldCopy(c: var TLiftCtx; t: PType; body, x, y: PNode) =
let xx = genBuiltin(c, mAccessTypeField, "accessTypeField", x)
let yy = genBuiltin(c, mAccessTypeField, "accessTypeField", y)
xx.typ() = getSysType(c.g, c.info, tyPointer)
yy.typ() = xx.typ
xx.typ = getSysType(c.g, c.info, tyPointer)
yy.typ = xx.typ
body.add newAsgnStmt(xx, yy)
proc produceSym(g: ModuleGraph; c: PContext; typ: PType; kind: TTypeAttachedOp;
info: TLineInfo; idgen: IdGenerator; isDistinct = false): PSym =
info: TLineInfo; idgen: IdGenerator): PSym =
if typ.kind == tyDistinct:
return produceSymDistinctType(g, c, typ, kind, info, idgen)
# For =dup, if the distinct type has a user-defined =copy, don't delegate
# to the base type. Instead fall through to the normal produceSym logic
# so that fillBody -> considerInferDupFromCopy can synthesize =dup from =copy.
if kind == attachedDup:
let copyOp = getAttachedOp(g, typ, attachedAsgn)
if copyOp != nil and sfOverridden in copyOp.flags:
discard "fall through to normal produceSym logic"
else:
return produceSymDistinctType(g, c, typ, kind, info, idgen)
else:
return produceSymDistinctType(g, c, typ, kind, info, idgen)
result = getAttachedOp(g, typ, kind)
if result == nil:
result = symPrototype(g, typ, typ.owner, kind, info, idgen, isDistinct = isDistinct)
result = symPrototype(g, typ, typ.owner, kind, info, idgen)
var a = TLiftCtx(info: info, g: g, kind: kind, c: c, asgnForType: typ, idgen: idgen,
fn: result)
@@ -1180,33 +1328,48 @@ proc produceSym(g: ModuleGraph; c: PContext; typ: PType; kind: TTypeAttachedOp;
if kind == attachedSink and destructorOverridden(g, typ):
## compiler can use a combination of `=destroy` and memCopy for sink op
dest.flags.incl sfCursor
ensureMutable dest
dest.flagsImpl.incl sfCursor
let op = getAttachedOp(g, typ, attachedDestructor)
result.ast[bodyPos].add newOpCall(a, op, if op.typ.firstParamType.kind == tyVar: d[0] else: d)
result.ast[bodyPos].add newAsgnStmt(d, src)
else:
var tk: TTypeKind
if g.config.selectedGC in {gcArc, gcOrc, gcHooks, gcAtomicArc}:
tk = skipTypes(typ, {tyOrdinal, tyRange, tyInferred, tyGenericInst, tyStatic, tyAlias, tySink}).kind
var skipped: PType = nil
if g.config.selectedGC in {gcArc, gcOrc, gcYrc, gcHooks, gcAtomicArc}:
skipped = skipTypes(typ, {tyOrdinal, tyRange, tyInferred, tyGenericInst, tyStatic, tyAlias, tySink})
tk = skipped.kind
else:
tk = tyNone # no special casing for strings and seqs
case tk
of tySequence:
let needsYrcLock = g.config.selectedGC == gcYrc and
kind in {attachedDestructor, attachedSink, attachedAsgn, attachedDeepCopy, attachedDup} and
types.canFormAcycle(g, skipped.elementType)
# YRC: topology-changing seq ops must hold the mutator (read) lock
if needsYrcLock:
result.ast[bodyPos].add callCodegenProc(g, "acquireMutatorLock", info)
fillSeqOp(a, typ, result.ast[bodyPos], d, src)
if needsYrcLock:
result.ast[bodyPos].add callCodegenProc(g, "releaseMutatorLock", info)
of tyString:
fillStrOp(a, typ, result.ast[bodyPos], d, src)
else:
fillBody(a, typ, result.ast[bodyPos], d, src)
if tk == tyObject and a.kind in {attachedAsgn, attachedSink, attachedDeepCopy, attachedDup} and not lacksMTypeField(typ):
if tk == tyObject and a.kind in {attachedAsgn, attachedSink, attachedDeepCopy, attachedDup} and not isObjLackingTypeField(skipped):
# bug #19205: Do not forget to also copy the hidden type field:
genTypeFieldCopy(a, typ, result.ast[bodyPos], d, src)
if not a.canRaise:
incl result.flags, sfNeverRaises
ensureMutable result
incl result.flagsImpl, sfNeverRaises
result.ast[pragmasPos] = newNodeI(nkPragma, info)
result.ast[pragmasPos].add newTree(nkExprColonExpr,
newIdentNode(g.cache.getIdent("raises"), info), newNodeI(nkBracket, info))
if kind == attachedDestructor:
ensureMutable result
incl result.optionsImpl, optQuirky
completePartialOp(g, idgen.module, typ, kind, result)
@@ -1215,6 +1378,7 @@ proc produceDestructorForDiscriminator*(g: ModuleGraph; typ: PType; field: PSym,
assert(typ.skipTypes({tyAlias, tyGenericInst}).kind == tyObject)
# discrimantor assignments needs pointers to destroy fields; alas, we cannot use non-var destructor here
result = symPrototype(g, field.typ, typ.owner, attachedDestructor, info, idgen, isDiscriminant = true)
setHookDisamb(g, result, "=destroy¦" & field.name.s & "¦" & $field.position, typ)
var a = TLiftCtx(info: info, g: g, kind: attachedDestructor, asgnForType: typ, idgen: idgen,
fn: result)
a.asgnForType = typ
@@ -1231,7 +1395,9 @@ proc produceDestructorForDiscriminator*(g: ModuleGraph; typ: PType; field: PSym,
result.ast[bodyPos].add v
let placeHolder = newNodeIT(nkSym, info, getSysType(g, info, tyPointer))
fillBody(a, typ, result.ast[bodyPos], d, placeHolder)
if not a.canRaise: incl result.flags, sfNeverRaises
if not a.canRaise:
ensureMutable result
incl result.flagsImpl, sfNeverRaises
template liftTypeBoundOps*(c: PContext; typ: PType; info: TLineInfo) =
@@ -1266,7 +1432,7 @@ proc inst(g: ModuleGraph; c: PContext; t: PType; kind: TTypeAttachedOp; idgen: I
else:
localError(g.config, info, "unresolved generic parameter")
proc isTrival*(s: PSym): bool {.inline.} =
proc isTrivial*(s: PSym): bool {.inline.} =
s == nil or (s.ast != nil and s.ast[bodyPos].len == 0)
proc createTypeBoundOps(g: ModuleGraph; c: PContext; orig: PType; info: TLineInfo;
@@ -1275,11 +1441,13 @@ proc createTypeBoundOps(g: ModuleGraph; c: PContext; orig: PType; info: TLineInf
## to ensure we lift assignment, destructors and moves properly.
## The later 'injectdestructors' pass depends on it.
if orig == nil or {tfCheckedForDestructor, tfHasMeta} * orig.flags != {}: return
incl orig.flags, tfCheckedForDestructor
# IC: review this solution again later
incl orig.flagsImpl, tfCheckedForDestructor
# for user defined generic destructors:
let origRoot = genericRoot(orig)
if origRoot != nil:
incl origRoot.flags, tfGenericHasDestructor
# IC: review this solution again later
incl origRoot.flagsImpl, tfGenericHasDestructor
let skipped = orig.skipTypes({tyGenericInst, tyAlias, tySink})
if isEmptyContainer(skipped) or skipped.kind == tyStatic: return
@@ -1299,13 +1467,13 @@ proc createTypeBoundOps(g: ModuleGraph; c: PContext; orig: PType; info: TLineInf
# we do not generate '=trace' procs if we
# have the cycle detection disabled, saves code size.
let lastAttached = if g.config.selectedGC == gcOrc: attachedTrace
let lastAttached = if g.config.selectedGC in {gcOrc, gcYrc}: attachedTrace
else: attachedSink
# bug #15122: We need to produce all prototypes before entering the
# mind boggling recursion. Hacks like these imply we should rewrite
# this module.
var generics: array[attachedWasMoved..attachedTrace, bool] = default(array[attachedWasMoved..attachedTrace, bool])
var generics = default(array[attachedWasMoved..attachedTrace, bool])
for k in attachedWasMoved..lastAttached:
generics[k] = getAttachedOp(g, canon, k) != nil
if not generics[k]:
@@ -1321,8 +1489,9 @@ proc createTypeBoundOps(g: ModuleGraph; c: PContext; orig: PType; info: TLineInf
if canon != orig:
setAttachedOp(g, idgen.module, orig, k, getAttachedOp(g, canon, k))
if not isTrival(getAttachedOp(g, orig, attachedDestructor)):
#or not isTrival(orig.assignment) or
# not isTrival(orig.sink):
orig.flags.incl tfHasAsgn
if not isTrivial(getAttachedOp(g, orig, attachedDestructor)):
#or not isTrivial(orig.assignment) or
# not isTrivial(orig.sink):
# IC: review this solution again later
orig.flagsImpl.incl tfHasAsgn
# ^ XXX Breaks IC!

View File

@@ -32,12 +32,12 @@ proc interestingVar(s: PSym): bool {.inline.} =
proc lookupOrAdd(c: var Ctx; s: PSym; info: TLineInfo): PNode =
let field = addUniqueField(c.objType, s, c.cache, c.idgen)
var deref = newNodeI(nkHiddenDeref, info)
deref.typ() = c.objType
deref.typ = c.objType
deref.add(newSymNode(c.partialParam, info))
result = newNodeI(nkDotExpr, info)
result.add(deref)
result.add(newSymNode(field))
result.typ() = field.typ
result.typ = field.typ
proc liftLocals(n: PNode; i: int; c: var Ctx) =
let it = n[i]

View File

@@ -93,9 +93,14 @@ type
warnBareExcept = "BareExcept",
warnImplicitDefaultValue = "ImplicitDefaultValue",
warnIgnoredSymbolInjection = "IgnoredSymbolInjection",
warnStdPrefix = "StdPrefix"
warnStdPrefix = "StdPrefix",
warnUnknownNotes = "UnknownNotes",
warnLongLiterals = "LongLiterals",
warnUser = "User",
warnGlobalVarConstructorTemporary = "GlobalVarConstructorTemporary",
warnImplicitRangeConversion = "ImplicitRangeConversion",
warnSystemRangeConversion = "SystemRangeConversion",
warnInvalidCmpOp = "InvalidCmpOp",
# hints
hintSuccess = "Success", hintSuccessX = "SuccessX",
hintCC = "CC",
@@ -109,9 +114,9 @@ type
hintSource = "Source", hintPerformance = "Performance", hintStackTrace = "StackTrace",
hintGCStats = "GCStats", hintGlobalVar = "GlobalVar", hintExpandMacro = "ExpandMacro",
hintUser = "User", hintUserRaw = "UserRaw", hintExtendedContext = "ExtendedContext",
hintUnknownRaises = "UnknownRaises",
hintMsgOrigin = "MsgOrigin", # since 1.3.5
hintDeclaredLoc = "DeclaredLoc", # since 1.5.1
hintUnknownHint = "UnknownHint"
const
MsgKindToStr*: array[TMsgKind, string] = [
@@ -200,8 +205,13 @@ const
warnImplicitDefaultValue: "$1",
warnIgnoredSymbolInjection: "$1",
warnStdPrefix: "$1 needs the 'std' prefix",
warnUnknownNotes: "$1",
warnLongLiterals: "$1",
warnUser: "$1",
warnGlobalVarConstructorTemporary: "global variable '$1' initialization requires a temporary variable",
warnImplicitRangeConversion: "implicit range conversion $1",
warnSystemRangeConversion: "implicit range conversion $1",
warnInvalidCmpOp: "$1",
hintSuccess: "operation successful: $#",
# keep in sync with `testament.isSuccess`
hintSuccessX: "$build\n$loc lines; ${sec}s; $mem; proj: $project; out: $output",
@@ -235,9 +245,9 @@ const
hintUser: "$1",
hintUserRaw: "$1",
hintExtendedContext: "$1",
hintUnknownRaises: "$1 is a forward declaration without explicit .raises, assuming it can raise anything",
hintMsgOrigin: "$1",
hintDeclaredLoc: "$1",
hintUnknownHint: "unknown hint: $1"
hintDeclaredLoc: "$1"
]
const
@@ -256,9 +266,9 @@ type
proc computeNotesVerbosity(): array[0..3, TNoteKinds] =
result = default(array[0..3, TNoteKinds])
result[3] = {low(TNoteKind)..high(TNoteKind)} - {warnObservableStores, warnResultUsed, warnAnyEnumConv, warnBareExcept, warnStdPrefix}
result[3] = {low(TNoteKind)..high(TNoteKind)} - {warnObservableStores, warnResultUsed, warnAnyEnumConv, warnBareExcept, warnStdPrefix, warnSystemRangeConversion}
result[2] = result[3] - {hintStackTrace, hintExtendedContext, hintDeclaredLoc, hintProcessingStmt}
result[1] = result[2] - {warnProveField, warnProveIndex,
result[1] = result[2] - {warnImplicitRangeConversion, warnProveField, warnProveIndex,
warnGcUnsafe, hintPath, hintDependency, hintCodeBegin, hintCodeEnd,
hintSource, hintGlobalVar, hintGCStats, hintMsgOrigin, hintPerformance}
result[0] = result[1] - {hintSuccessX, hintSuccess, hintConf,
@@ -271,6 +281,10 @@ const
errFloatToString* = "cannot convert '$1' to '$2'"
type
FileInfoKind* = enum
fikSource, ## A real source file path
fikNifModule ## A NIF module suffix (not a real path)
TFileInfo* = object
fullPath*: AbsoluteFile # This is a canonical full filesystem path
projPath*: RelativeFile # This is relative to the project's root
@@ -289,6 +303,7 @@ type
# for 'nimsuggest'
hash*: string # the checksum of the file
dirty*: bool # for 'nimpretty' like tooling
kind*: FileInfoKind # distinguishes real files from NIF suffixes
when defined(nimpretty):
fullContent*: string
FileIndex* = distinct int32

View File

@@ -95,7 +95,7 @@ proc nep1CheckDefImpl(conf: ConfigRef; info: TLineInfo; s: PSym; k: TSymKind) =
template styleCheckDef*(ctx: PContext; info: TLineInfo; sym: PSym; k: TSymKind) =
## Check symbol definitions adhere to NEP1 style rules.
if optStyleCheck in ctx.config.options and # ignore if styleChecks are off
{optStyleHint, optStyleError} * ctx.config.globalOptions != {} and # check only if hint/error is enabled
{optStyleHint, optStyleError, optStyleWarning} * ctx.config.globalOptions != {} and # check only if hint/error/warning is enabled
hintName in ctx.config.notes and # ignore if name checks are not requested
ctx.config.belongsToProjectPackageMaybeNil(getModule(ctx.graph, info.fileIndex)) and # ignore foreign packages
optStyleUsages notin ctx.config.globalOptions and # ignore if requested to only check name usage
@@ -136,7 +136,7 @@ proc styleCheckUseImpl(conf: ConfigRef; info: TLineInfo; s: PSym) =
template styleCheckUse*(ctx: PContext; info: TLineInfo; sym: PSym) =
## Check symbol uses match their definition's style.
if {optStyleHint, optStyleError} * ctx.config.globalOptions != {} and # ignore if styleChecks are off
if {optStyleHint, optStyleError, optStyleWarning} * ctx.config.globalOptions != {} and # ignore if styleChecks are off
hintName in ctx.config.notes and # ignore if name checks are not requested
ctx.config.belongsToProjectPackageMaybeNil(getModule(ctx.graph, info.fileIndex)) and # ignore foreign packages
sym.kind != skTemp and # ignore temporary variables created by the compiler
@@ -152,7 +152,7 @@ proc checkPragmaUseImpl(conf: ConfigRef; info: TLineInfo; w: TSpecialWord; pragm
template checkPragmaUse*(ctx: PContext; info: TLineInfo; w: TSpecialWord; pragmaName: string, sym: PSym) =
## Check builtin pragma uses match their definition's style.
## Note: This only applies to builtin pragmas, not user pragmas.
if {optStyleHint, optStyleError} * ctx.config.globalOptions != {} and # ignore if styleChecks are off
if {optStyleHint, optStyleError, optStyleWarning} * ctx.config.globalOptions != {} and # ignore if styleChecks are off
hintName in ctx.config.notes and # ignore if name checks are not requested
ctx.config.belongsToProjectPackageMaybeNil(getModule(ctx.graph, info.fileIndex)): # ignore foreign packages
checkPragmaUseImpl(ctx.config, info, w, pragmaName)

View File

@@ -11,7 +11,7 @@
import
pathutils
import std/strutils
when defined(nimPreviewSlimSystem):
import std/syncio
@@ -86,6 +86,47 @@ const
LineContinuationOprs = {'+', '-', '*', '/', '\\', '<', '>', '!', '?', '^',
'|', '%', '&', '$', '@', '~', ','}
AdditionalLineContinuationOprs = {'#', ':', '='}
LineContinuationTokens = [
"let", "var", "const", "type", # section
"object", "tuple",
# from ./layouter.oprSet
"div", "mod", "shl", "shr", "in", "notin", "is",
"isnot", "not", "of", "as", "from", "..", "and", "or", "xor",
] # must be all `nimIdentNormalized`-ed
proc eqIdent(a, bNormalized: string): bool =
a.nimIdentNormalize == bNormalized
proc endsWithIdent(s, subs: string): bool =
let le = subs.len
if le > s.len: return false
s[^le .. ^1].eqIdent subs
proc continuesWithIdent(s, subs: string, start: int): bool =
s.substr(start, start+subs.high).eqIdent subs
proc endsWithIdent(s, subs: string, endIdx: var int): bool =
endIdx.dec subs.len
result = s.continuesWithIdent(subs, endIdx+1)
proc containsObjectOf(x: string): bool =
const sep = ' '
var idx = x.rfind(sep)
if idx == -1: return
template eatWord(word) =
while x[idx] == sep: idx.dec
result = x.endsWithIdent(word, idx)
if not result: return
eatWord "of"
eatWord "object"
result = true
proc endsWithLineContinuationToken(x: string): bool =
result = false
for tok in LineContinuationTokens:
if x.endsWithIdent(tok):
return true
result = x.containsObjectOf
proc endsWithOpr*(x: string): bool =
result = x.endsWith(LineContinuationOprs)
@@ -93,7 +134,9 @@ proc endsWithOpr*(x: string): bool =
proc continueLine(line: string, inTripleString: bool): bool {.inline.} =
result = inTripleString or line.len > 0 and (
line[0] == ' ' or
line.endsWith(LineContinuationOprs+AdditionalLineContinuationOprs))
line.endsWith(LineContinuationOprs+AdditionalLineContinuationOprs) or
line.endsWithLineContinuationToken()
)
proc countTriples(s: string): int =
result = 0
@@ -109,7 +152,10 @@ proc llReadFromStdin(s: PLLStream, buf: pointer, bufLen: int): int =
s.rd = 0
var line = newStringOfCap(120)
var triples = 0
while readLineFromStdin(if s.s.len == 0: ">>> " else: "... ", line):
while true:
if not readLineFromStdin(if s.s.len == 0: ">>> " else: "... ", line):
# now readLineFromStdin meets EOF (ctrl-D/Z) or ctrl-C
quit()
s.s.add(line)
s.s.add("\n")
inc triples, countTriples(line)
@@ -117,7 +163,7 @@ proc llReadFromStdin(s: PLLStream, buf: pointer, bufLen: int): int =
inc(s.lineOffset)
result = min(bufLen, s.s.len - s.rd)
if result > 0:
copyMem(buf, addr(s.s[s.rd]), result)
copyMem(buf, readRawData(s.s, s.rd), result)
inc(s.rd, result)
proc llStreamRead*(s: PLLStream, buf: pointer, bufLen: int): int =
@@ -127,7 +173,7 @@ proc llStreamRead*(s: PLLStream, buf: pointer, bufLen: int): int =
of llsString:
result = min(bufLen, s.s.len - s.rd)
if result > 0:
copyMem(buf, addr(s.s[0 + s.rd]), result)
copyMem(buf, readRawData(s.s, s.rd), result)
inc(s.rd, result)
of llsFile:
result = readBuffer(s.f, buf, bufLen)

View File

@@ -58,13 +58,11 @@ proc considerQuotedIdent*(c: PContext; n: PNode, origin: PNode = nil): PIdent =
of nkLiterals - nkFloatLiterals: id.add(x.renderTree)
else: handleError(n, origin)
result = getIdent(c.cache, id)
of nkOpenSymChoice, nkClosedSymChoice:
of nkOpenSymChoice, nkClosedSymChoice, nkOpenSym:
if n[0].kind == nkSym:
result = n[0].sym.name
else:
handleError(n, origin)
of nkOpenSym:
result = considerQuotedIdent(c, n[0], origin)
else:
handleError(n, origin)
@@ -77,10 +75,13 @@ proc addUniqueSym*(scope: PScope, s: PSym): PSym =
proc openScope*(c: PContext): PScope {.discardable.} =
result = PScope(parent: c.currentScope,
symbols: initStrTable(),
depthLevel: c.scopeDepth + 1)
depthLevel: c.scopeDepth + 1,
optionStackLen: c.optionStack.len)
c.currentScope = result
proc rawCloseScope*(c: PContext) =
if c.currentScope.optionStackLen >= 1:
c.optionStack.setLen(c.currentScope.optionStackLen)
c.currentScope = c.currentScope.parent
proc closeScope*(c: PContext) =
@@ -221,7 +222,14 @@ proc debugScopes*(c: PContext; limit=0, max = int.high) {.deprecated.} =
if i == limit: return
inc i
proc searchInScopesAllCandidatesFilterBy*(c: PContext, s: PIdent, filter: TSymKinds): seq[PSym] =
proc searchImportsAll*(c: PContext, s: PIdent, filter: TSymKinds, holding: var seq[PSym]) =
var marked = initIntSet()
for im in c.imports.mitems:
for s in symbols(im, marked, s, c.graph):
if s.kind in filter:
holding.add s
proc searchScopes*(c: PContext, s: PIdent, filter: TSymKinds): seq[PSym] =
result = @[]
for scope in allScopes(c.currentScope):
var ti: TIdentIter = default(TIdentIter)
@@ -231,14 +239,12 @@ proc searchInScopesAllCandidatesFilterBy*(c: PContext, s: PIdent, filter: TSymKi
result.add candidate
candidate = nextIdentIter(ti, scope.symbols)
proc searchScopesAll*(c: PContext, s: PIdent, filter: TSymKinds): seq[PSym] =
result = searchScopes(c,s,filter)
if result.len == 0:
var marked = initIntSet()
for im in c.imports.mitems:
for s in symbols(im, marked, s, c.graph):
if s.kind in filter:
result.add s
searchImportsAll(c, s, filter, result)
proc searchInScopesFilterBy*(c: PContext, s: PIdent, filter: TSymKinds): seq[PSym] =
proc selectFromScopesElseAll*(c: PContext, s: PIdent, filter: TSymKinds): seq[PSym] =
result = @[]
block outer:
for scope in allScopes(c.currentScope):
@@ -252,11 +258,7 @@ proc searchInScopesFilterBy*(c: PContext, s: PIdent, filter: TSymKinds): seq[PSy
candidate = nextIdentIter(ti, scope.symbols)
if result.len == 0:
var marked = initIntSet()
for im in c.imports.mitems:
for s in symbols(im, marked, s, c.graph):
if s.kind in filter:
result.add s
searchImportsAll(c, s, filter, result)
proc cmpScopes*(ctx: PContext, s: PSym): int =
# Do not return a negative number
@@ -309,7 +311,7 @@ proc errorSym*(c: PContext, ident: PIdent, info: TLineInfo): PSym =
## creates an error symbol to avoid cascading errors (for IDE support)
result = newSym(skError, ident, c.idgen, getCurrOwner(c), info, {})
result.typ = errorType(c)
incl(result.flags, sfDiscardable)
incl(result.flagsImpl, sfDiscardable)
# pretend it's from the top level scope to prevent cascading errors:
if c.config.cmd != cmdInteractive and c.compilesContextId == 0:
c.moduleScope.addSym(result)
@@ -376,6 +378,9 @@ proc wrongRedefinition*(c: PContext; info: TLineInfo, s: string;
conflictsWith: TLineInfo, note = errGenerated) =
## Emit a redefinition error if in non-interactive mode
if c.config.cmd != cmdInteractive:
when defined(icDbgRefc):
echo "[icRedef] ", s
echo getStackTrace()
localError(c.config, info, note,
"redefinition of '$1'; previous declaration here: $2" %
[s, c.config $ conflictsWith])
@@ -386,7 +391,7 @@ proc addDeclAt*(c: PContext; scope: PScope, sym: PSym, info: TLineInfo) =
if sym.name.id == ord(wUnderscore): return
let conflict = scope.addUniqueSym(sym)
if conflict != nil:
if sym.kind == skModule and conflict.kind == skModule:
if sym.kind == skModule and conflict.kind == skModule and not c.config.isDefined("nimPreviewDuplicateModuleError"):
# e.g.: import foo; import foo
# xxx we could refine this by issuing a different hint for the case
# where a duplicate import happens inside an include.
@@ -410,8 +415,6 @@ proc addDecl*(c: PContext, sym: PSym) {.inline.} =
proc addPrelimDecl*(c: PContext, sym: PSym) =
discard c.currentScope.addUniqueSym(sym)
from ic / ic import addHidden
proc addInterfaceDeclAux(c: PContext, sym: PSym) =
## adds symbol to the module for either private or public access.
if sfExported in sym.flags:
@@ -420,8 +423,6 @@ proc addInterfaceDeclAux(c: PContext, sym: PSym) =
else: internalError(c.config, sym.info, "addInterfaceDeclAux")
elif sym.kind in ExportableSymKinds and c.module != nil and isTopLevelInsideDeclaration(c, sym):
strTableAdd(semtabAll(c.graph, c.module), sym)
if c.config.symbolFiles != disabledSf:
addHidden(c.encoder, c.packedRepr, sym)
proc addInterfaceDeclAt*(c: PContext, scope: PScope, sym: PSym) =
## adds a symbol on the scope and the interface if appropriate
@@ -461,6 +462,15 @@ proc openShadowScope*(c: PContext) =
symbols: initStrTable(),
depthLevel: c.scopeDepth)
proc rememberShadowDefs*(c: PContext) =
## bug #25693: a template/macro operand's local definitions are sem-checked in
## a shadow scope that is then discarded. Record those definitions so that a
## later re-emission (e.g. a captured `typed` fragment expanded more than once)
## can be detected as a redefinition rather than silently miscompiled.
for s in c.currentScope.symbols:
if s.kind in {skVar, skLet, skForVar} and {sfGenSym, sfWasGenSym} * s.flags == {}:
c.shadowDiscardedDefs.incl s.id
proc closeShadowScope*(c: PContext) =
## closes the shadow scope, but doesn't merge any of the symbols
## Does not check for unused symbols or missing forward decls since a macro
@@ -644,7 +654,7 @@ const allExceptModule = {low(TSymKind)..high(TSymKind)} - {skModule, skPackage}
proc lookUpCandidates*(c: PContext, ident: PIdent, filter: set[TSymKind],
includePureEnum = false): seq[PSym] =
result = searchInScopesFilterBy(c, ident, filter)
result = selectFromScopesElseAll(c, ident, filter)
if skEnumField in filter and (result.len == 0 or includePureEnum):
result.add allPureEnumFields(c, ident)
@@ -725,7 +735,7 @@ proc qualifiedLookUp*(c: PContext, n: PNode, flags: set[TLookupFlag]): PSym =
proc initOverloadIter*(o: var TOverloadIter, c: PContext, n: PNode): PSym =
if n.kind == nkOpenSym:
# maybe the logic in semexprs should be mirrored here instead
# for now it only seems this is called for `pickSym` in `getTypeIdent`
# for now it only seems this is called for `pickSym` in `getTypeIdent`
return initOverloadIter(o, c, n[0])
o.importIdx = -1
o.marked = initIntSet()

View File

@@ -82,7 +82,7 @@ proc lowerTupleUnpacking*(g: ModuleGraph; n: PNode; idgen: IdGenerator; owner: P
var temp = newSym(skTemp, getIdent(g.cache, genPrefix), idgen,
owner, value.info, g.config.options)
temp.typ = skipTypes(value.typ, abstractInst)
incl(temp.flags, sfFromGeneric)
incl(temp.flagsImpl, sfFromGeneric)
tempAsNode = newSymNode(temp)
var v = newNodeI(nkVarSection, value.info)
@@ -103,7 +103,7 @@ proc evalOnce*(g: ModuleGraph; value: PNode; idgen: IdGenerator; owner: PSym): P
var temp = newSym(skTemp, getIdent(g.cache, genPrefix), idgen,
owner, value.info, g.config.options)
temp.typ = skipTypes(value.typ, abstractInst)
incl(temp.flags, sfFromGeneric)
incl(temp.flagsImpl, sfFromGeneric)
var v = newNodeI(nkLetSection, value.info)
let tempAsNode = newSymNode(temp)
@@ -127,8 +127,8 @@ proc lowerSwap*(g: ModuleGraph; n: PNode; idgen: IdGenerator; owner: PSym): PNod
# note: cannot use 'skTemp' here cause we really need the copy for the VM :-(
var temp = newSym(skVar, getIdent(g.cache, genPrefix), idgen, owner, n.info, owner.options)
temp.typ = n[1].typ
incl(temp.flags, sfFromGeneric)
incl(temp.flags, sfGenSym)
incl(temp.flagsImpl, sfFromGeneric)
incl(temp.flagsImpl, sfGenSym)
var v = newNodeI(nkVarSection, n.info)
let tempAsNode = newSymNode(temp)
@@ -147,13 +147,13 @@ proc createObj*(g: ModuleGraph; idgen: IdGenerator; owner: PSym, info: TLineInfo
result = newType(tyObject, idgen, owner)
if final:
rawAddSon(result, nil)
incl result.flags, tfFinal
incl result, tfFinal
else:
rawAddSon(result, getCompilerProc(g, "RootObj").typ)
result.n = newNodeI(nkRecList, info)
let s = newSym(skType, getIdent(g.cache, "Env_" & toFilename(g.config, info) & "_" & $owner.name.s),
idgen, owner, info, owner.options)
incl s.flags, sfAnon
incl s.flagsImpl, sfAnon
s.typ = result
result.sym = s
@@ -174,12 +174,12 @@ proc rawIndirectAccess*(a: PNode; field: PSym; info: TLineInfo): PNode =
# returns a[].field as a node
assert field.kind == skField
var deref = newNodeI(nkHiddenDeref, info)
deref.typ() = a.typ.skipTypes(abstractInst)[0]
deref.typ = a.typ.skipTypes(abstractInst)[0]
deref.add a
result = newNodeI(nkDotExpr, info)
result.add deref
result.add newSymNode(field)
result.typ() = field.typ
result.typ = field.typ
proc rawDirectAccess*(obj, field: PSym): PNode =
# returns a.field as a node
@@ -187,7 +187,7 @@ proc rawDirectAccess*(obj, field: PSym): PNode =
result = newNodeI(nkDotExpr, field.info)
result.add newSymNode(obj)
result.add newSymNode(field)
result.typ() = field.typ
result.typ = field.typ
proc lookupInRecord(n: PNode, id: ItemId): PSym =
result = nil
@@ -207,7 +207,7 @@ proc lookupInRecord(n: PNode, id: ItemId): PSym =
if result != nil: return
else: discard
of nkSym:
if n.sym.itemId.module == id.module and n.sym.itemId.item == -abs(id.item): result = n.sym
if matchesDerivedFieldId(n.sym.itemId, id): result = n.sym
else: discard
proc addField*(obj: PType; s: PSym; cache: IdentCache; idgen: IdGenerator): PSym =
@@ -215,7 +215,7 @@ proc addField*(obj: PType; s: PSym; cache: IdentCache; idgen: IdGenerator): PSym
# This is hacky but the clean solution is much more complex than it looks.
var field = newSym(skField, getIdent(cache, s.name.s & $obj.n.len),
idgen, s.owner, s.info, s.options)
field.itemId = ItemId(module: s.itemId.module, item: -s.itemId.item)
field.itemId = derivedFieldId(s.itemId)
let t = skipIntLit(s.typ, idgen)
field.typ = t
if s.kind in {skLet, skVar, skField, skForVar}:
@@ -235,7 +235,7 @@ proc addUniqueField*(obj: PType; s: PSym; cache: IdentCache; idgen: IdGenerator)
if result == nil:
var field = newSym(skField, getIdent(cache, s.name.s & $obj.n.len), idgen,
s.owner, s.info, s.options)
field.itemId = ItemId(module: s.itemId.module, item: -s.itemId.item)
field.itemId = derivedFieldId(s.itemId)
let t = skipIntLit(s.typ, idgen)
field.typ = t
assert t.kind != tyTyped
@@ -250,12 +250,12 @@ proc newDotExpr*(obj, b: PSym): PNode =
assert field != nil, b.name.s
result.add newSymNode(obj)
result.add newSymNode(field)
result.typ() = field.typ
result.typ = field.typ
proc indirectAccess*(a: PNode, b: ItemId, info: TLineInfo): PNode =
# returns a[].b as a node
var deref = newNodeI(nkHiddenDeref, info)
deref.typ() = a.typ.skipTypes(abstractInst).elementType
deref.typ = a.typ.skipTypes(abstractInst).elementType
var t = deref.typ.skipTypes(abstractInst)
var field: PSym
while true:
@@ -273,12 +273,12 @@ proc indirectAccess*(a: PNode, b: ItemId, info: TLineInfo): PNode =
result = newNodeI(nkDotExpr, info)
result.add deref
result.add newSymNode(field)
result.typ() = field.typ
result.typ = field.typ
proc indirectAccess*(a: PNode, b: string, info: TLineInfo; cache: IdentCache): PNode =
# returns a[].b as a node
var deref = newNodeI(nkHiddenDeref, info)
deref.typ() = a.typ.skipTypes(abstractInst).elementType
deref.typ = a.typ.skipTypes(abstractInst).elementType
var t = deref.typ.skipTypes(abstractInst)
var field: PSym
let bb = getIdent(cache, b)
@@ -297,7 +297,7 @@ proc indirectAccess*(a: PNode, b: string, info: TLineInfo; cache: IdentCache): P
result = newNodeI(nkDotExpr, info)
result.add deref
result.add newSymNode(field)
result.typ() = field.typ
result.typ = field.typ
proc getFieldFromObj*(t: PType; v: PSym): PSym =
assert v.kind != skField
@@ -320,7 +320,7 @@ proc indirectAccess*(a, b: PSym, info: TLineInfo): PNode =
proc genAddrOf*(n: PNode; idgen: IdGenerator; typeKind = tyPtr): PNode =
result = newNodeI(nkAddr, n.info, 1)
result[0] = n
result.typ() = newType(typeKind, idgen, n.typ.owner)
result.typ = newType(typeKind, idgen, n.typ.owner)
result.typ.rawAddSon(n.typ)
proc genDeref*(n: PNode; k = nkHiddenDeref): PNode =
@@ -344,18 +344,18 @@ proc callCodegenProc*(g: ModuleGraph; name: string;
if optionalArgs != nil:
for i in 1..<optionalArgs.len-2:
result.add optionalArgs[i]
result.typ() = sym.typ.returnType
result.typ = sym.typ.returnType
proc newIntLit*(g: ModuleGraph; info: TLineInfo; value: BiggestInt): PNode =
result = nkIntLit.newIntNode(value)
result.typ() = getSysType(g, info, tyInt)
result.typ = getSysType(g, info, tyInt)
proc genHigh*(g: ModuleGraph; n: PNode): PNode =
if skipTypes(n.typ, abstractVar).kind == tyArray:
result = newIntLit(g, n.info, toInt64(lastOrd(g.config, skipTypes(n.typ, abstractVar))))
else:
result = newNodeI(nkCall, n.info, 2)
result.typ() = getSysType(g, n.info, tyInt)
result.typ = getSysType(g, n.info, tyInt)
result[0] = newSymNode(getSysMagic(g, n.info, "high", mHigh))
result[1] = n
@@ -364,7 +364,7 @@ proc genLen*(g: ModuleGraph; n: PNode): PNode =
result = newIntLit(g, n.info, toInt64(lastOrd(g.config, skipTypes(n.typ, abstractVar)) + 1))
else:
result = newNodeI(nkCall, n.info, 2)
result.typ() = getSysType(g, n.info, tyInt)
result.typ = getSysType(g, n.info, tyInt)
result[0] = newSymNode(getSysMagic(g, n.info, "len", mLengthSeq))
result[1] = n

View File

@@ -10,8 +10,8 @@
# Built-in types and compilerprocs are registered here.
import
ast, astalgo, msgs, platform, idents,
modulegraphs, lineinfos
ast, msgs, platform, idents,
modulegraphs, lineinfos, types
export createMagic
@@ -134,7 +134,7 @@ proc getNimScriptSymbol*(g: ModuleGraph; name: string): PSym =
proc resetNimScriptSymbols*(g: ModuleGraph) = g.exposed = initStrTable()
proc getMagicEqSymForType*(g: ModuleGraph; t: PType; info: TLineInfo): PSym =
case t.kind
case t.skipTypes(abstractRange).kind
of tyInt, tyInt8, tyInt16, tyInt32, tyInt64,
tyUInt, tyUInt8, tyUInt16, tyUInt32, tyUInt64:
result = getSysMagic(g, info, "==", mEqI)
@@ -166,4 +166,4 @@ proc makeAddr*(n: PNode; idgen: IdGenerator): PNode =
result = n
else:
result = newTree(nkHiddenAddr, n)
result.typ() = makePtrType(n.typ, idgen)
result.typ = makePtrType(n.typ.skipTypes({tySink}), idgen)

View File

@@ -26,11 +26,14 @@ import
when defined(nimPreviewSlimSystem):
import std/[syncio, assertions]
import ic / [cbackend, integrity, navigator, ic]
import ../dist/checksums/src/checksums/sha1
import pipelines
from icconfig import produceIcConfig
when not defined(nimKochBootstrap):
import nifbackend
import deps
when not defined(leanCompiler):
import docgen
@@ -95,14 +98,6 @@ proc commandCheck(graph: ModuleGraph) =
setPipeLinePass(graph, SemPass)
compilePipelineProject(graph)
if conf.symbolFiles != disabledSf:
case conf.ideCmd
of ideDef: navDefinition(graph)
of ideUse: navUsages(graph)
of ideDus: navDefusages(graph)
else: discard
writeRodFiles(graph)
when not defined(leanCompiler):
proc commandDoc2(graph: ModuleGraph; ext: string) =
handleDocOutputOptions graph.config
@@ -117,6 +112,38 @@ when not defined(leanCompiler):
else: raiseAssert $ext
compilePipelineProject(graph)
proc commandCompileToNif(graph: ModuleGraph) =
let conf = graph.config
extccomp.initVars(conf)
if conf.symbolFiles == disabledSf:
if {optRun, optForceFullMake} * conf.globalOptions == {optRun} or isDefined(conf, "nimBetterRun"):
if not changeDetectedViaJsonBuildInstructions(conf, conf.jsonBuildInstructionsFile):
# nothing changed
graph.config.notes = graph.config.mainPackageNotes
return
if not extccomp.ccHasSaneOverflow(conf):
conf.symbols.defineSymbol("nimEmulateOverflowChecks")
setPipeLinePass(graph, NifgenPass)
compilePipelineProject(graph)
proc commandNifC(graph: ModuleGraph) =
## Generate C code from precompiled NIF files.
## This is the new IC approach: compile modules to NIF first with `nim m`,
## then generate C code from the entry.nif file with whole-program DCE.
when not defined(nimKochBootstrap):
let conf = graph.config
extccomp.initVars(conf)
if not extccomp.ccHasSaneOverflow(conf):
conf.symbols.defineSymbol("nimEmulateOverflowChecks")
# Use the NIF backend to generate C code
nifbackend.generateCode(graph, conf.projectMainIdx)
else:
rawMessage(graph.config, errGenerated, "NIF backend not available during bootstrap build")
proc commandCompileToC(graph: ModuleGraph) =
let conf = graph.config
extccomp.initVars(conf)
@@ -137,15 +164,7 @@ proc commandCompileToC(graph: ModuleGraph) =
compilePipelineProject(graph)
if graph.config.errorCounter > 0:
return # issue #9933
if conf.symbolFiles == disabledSf:
cgenWriteModules(graph.backend, conf)
else:
if isDefined(conf, "nimIcIntegrityChecks"):
checkIntegrity(graph)
generateCode(graph)
# graph.backend can be nil under IC when nothing changed at all:
if graph.backend != nil:
cgenWriteModules(graph.backend, conf)
cgenWriteModules(graph.backend, conf)
if conf.cmd != cmdTcc and graph.backend != nil:
extccomp.callCCompiler(conf)
# for now we do not support writing out a .json file with the build instructions when HCR is on
@@ -184,7 +203,7 @@ proc commandInteractive(graph: ModuleGraph) =
discard graph.compilePipelineModule(fileInfoIdx(graph.config, graph.config.projectFull), {})
else:
var m = graph.makeStdinModule()
incl(m.flags, sfMainModule)
incl(m, sfMainModule)
var idgen = IdGenerator(module: m.itemId.module, symId: m.itemId.item, typeId: 0)
let s = llStreamOpenStdIn(onPrompt = proc() = flushDot(graph.config))
discard processPipelineModule(graph, m, idgen, s)
@@ -205,10 +224,6 @@ proc commandScan(cache: IdentCache, config: ConfigRef) =
else:
rawMessage(config, errGenerated, "cannot open file: " & f.string)
proc commandView(graph: ModuleGraph) =
let f = toAbsolute(mainCommandArg(graph.config), AbsoluteDir getCurrentDir()).addFileExt(RodExt)
rodViewer(f, graph.config, graph.cache)
const
PrintRopeCacheStats = false
@@ -257,7 +272,7 @@ proc mainCommand*(graph: ModuleGraph) =
if conf.exc == excNone: conf.exc = excSetjmp
of backendCpp:
if conf.exc == excNone: conf.exc = excCpp
of backendObjc: discard
of backendObjc, backendNif: discard
of backendJs:
if conf.hcrOn:
# XXX: At the moment, system.nim cannot be compiled in JS mode
@@ -275,6 +290,7 @@ proc mainCommand*(graph: ModuleGraph) =
of backendCpp: commandCompileToC(graph)
of backendObjc: commandCompileToC(graph)
of backendJs: commandCompileToJS(graph)
of backendNif: commandCompileToNif(graph)
of backendInvalid: raiseAssert "unreachable"
template docLikeCmd(body) =
@@ -305,8 +321,6 @@ proc mainCommand*(graph: ModuleGraph) =
case conf.cmd
of cmdBackends:
compileToBackend()
when BenchIC:
echoTimes graph.packed
of cmdTcc:
when hasTinyCBackend:
extccomp.setCC(conf, "tcc", unknownLineInfo)
@@ -403,16 +417,37 @@ proc mainCommand*(graph: ModuleGraph) =
of cmdCheck:
commandCheck(graph)
of cmdM:
graph.config.symbolFiles = v2Sf
setUseIc(graph.config.symbolFiles != disabledSf)
# cmdM uses NIF files, not ROD files
graph.config.symbolFiles = disabledSf
setUseIc(true)
# vtable dispatch needs a whole-program vtable layout, which the
# per-module compilation model cannot provide (yet); methods dispatch
# through the classic if-chain dispatchers instead
excl conf.features, Feature.vtables
commandCheck(graph)
of cmdNifC:
setUseIc(true)
excl conf.features, Feature.vtables
# Generate C code from NIF files
wantMainModule(conf)
setOutFile(conf)
commandNifC(graph)
of cmdIc:
# Generate .build.nif for nifmake
setUseIc(true)
wantMainModule(conf)
when not defined(nimKochBootstrap):
commandIc(conf)
else:
rawMessage(conf, errGenerated, "nim deps not available in bootstrap build")
of cmdIcConfig:
# Produce the precompiled config artifact for `nim ic` (config already
# parsed by the normal pipeline); a separate process spawned by the driver.
wantMainModule(conf)
produceIcConfig(conf)
of cmdParse:
wantMainModule(conf)
discard parseFile(conf.projectMainIdx, cache, conf)
of cmdRod:
wantMainModule(conf)
commandView(graph)
#msgWriteln(conf, "Beware: Indentation tokens depend on the parser's state!")
of cmdInteractive: commandInteractive(graph)
of cmdNimscript:
if conf.projectIsCmd or conf.projectIsStdin: discard
@@ -426,7 +461,14 @@ proc mainCommand*(graph: ModuleGraph) =
of cmdUnknown, cmdNone, cmdIdeTools:
rawMessage(conf, errGenerated, "invalid command: " & conf.command)
if conf.errorCounter == 0 and conf.cmd notin {cmdTcc, cmdDump, cmdNop}:
if conf.errorCounter == 0 and conf.cmd notin {cmdTcc, cmdDump, cmdNop, cmdM} and
not (conf.cmd == cmdNifC and conf.icBackendStage.len > 0):
# The IC build runs hundreds of internal per-module child processes — the
# frontend `nim m` (cmdM) and the per-module backend stages (cg/emit/merge/
# link). Each would print a `[SuccessX]` summary that is pure noise (and
# misleading: `out: unknownOutput`, or `out: <the whole compiler>` for a
# step that only wrote one `.c.nif`/`.c`). The driving `nim ic` (and koch)
# reports the real result.
if optProfileVM in conf.globalOptions:
echo conf.dump(conf.vmProfileData)
genSuccessX(conf)

View File

@@ -53,7 +53,29 @@ proc mangleParamExt*(s: PSym): string =
result.addInt s.position
proc mangleProcNameExt*(graph: ModuleGraph, s: PSym): string =
result = "__"
# The disambiguator comes first and the module suffix LAST, so the suffix is
# a strippable trailing token: content-addressed cross-module merging chops
# everything from the final `__` to recover a mint-site-independent name.
if s.itemId.isBackendMinted:
# A symbol minted during IC codegen (`idGeneratorForBackend`): its idgen
# starts with an EMPTY per-name disamb table, so its `disamb` restarts at 0
# and collides with same-named sem-time symbols loaded from NIFs (two
# `=destroy` hooks both mangling to `_u2` → "conflicting types for ..." in
# the generated C). These symbols never cross a process boundary (nifc
# lifts, emits and compiles them in one run), so the per-module-unique
# item id is a safe and deterministic discriminator; the `_c` marker keeps
# the namespace disjoint from `_u<disamb>`.
result = "_c"
result.addInt s.itemId.item
else:
result = "_u"
# Use `disamb` rather than `itemId.item`: under incremental compilation a
# symbol loaded from a NIF file gets a fresh, load-order-dependent `itemId.item`
# (from the per-module symbol counter), which is neither stable across the
# processes that compile vs. use a module nor guaranteed distinct from another
# loaded symbol's. `disamb` is assigned deterministically per (module, name)
# and, together with the already-prepended mangled name, yields a unique and
# stable C identifier.
result.addInt s.disamb
result.add "__"
result.add graph.ifaces[s.itemId.module].uniqueName
result.add "_u"
result.addInt s.itemId.item # s.disamb #

File diff suppressed because it is too large Load Diff

View File

@@ -79,6 +79,10 @@ proc checkModuleName*(conf: ConfigRef; n: PNode; doLocalError=true): FileIndex =
else:
result = fileInfoIdx(conf, fullPath)
type
SelectedBase = enum
FromProject, FromSearchPath, FromNimblePath
proc mangleModuleName*(conf: ConfigRef; path: AbsoluteFile): string =
## Mangle a relative module path to avoid path and symbol collisions.
##
@@ -87,9 +91,29 @@ proc mangleModuleName*(conf: ConfigRef; path: AbsoluteFile): string =
##
## Example:
## `foo-#head/../bar` becomes `@foo-@hhead@s..@sbar`
"@m" & relativeTo(path, conf.projectPath).string.multiReplace(
{$os.DirSep: "@s", $os.AltSep: "@s", "#": "@h", "@": "@@", ":": "@c"})
var best = relativeTo(path, conf.projectPath).string
var selectedBase = FromProject
for x in conf.searchPaths:
let other = relativeTo(path, x).string
if other.len < best.len:
best = other
selectedBase = FromSearchPath
for x in conf.nimblePaths:
let other = relativeTo(path, x).string
if other.len < best.len:
best = other
selectedBase = FromNimblePath
let prefix =
case selectedBase
of FromProject: "@m"
of FromSearchPath: "@p"
of FromNimblePath: "@n"
# Note: We encode ".." specially as "@d" to avoid issues with changeFileExt
# which would misinterpret ".." as "name.ext" and strip the second part.
prefix & best.multiReplace(
{"..": "@d", $os.DirSep: "@s", $os.AltSep: "@s", "#": "@h", "@": "@@", ":": "@c"})
proc demangleModuleName*(path: string): string =
## Demangle a relative module path.
result = path.multiReplace({"@@": "@", "@h": "#", "@s": "/", "@m": "", "@c": ":"})
result = path.multiReplace({"@@": "@", "@d": "..", "@h": "#", "@s": "/", "@m": "", "@p": "", "@n": "", "@c": ":"})

View File

@@ -32,9 +32,9 @@ proc newModule*(graph: ModuleGraph; fileIdx: FileIndex): PSym =
let filename = AbsoluteFile toFullPath(graph.config, fileIdx)
# We cannot call ``newSym`` here, because we have to circumvent the ID
# mechanism, which we do in order to assign each module a persistent ID.
result = PSym(kind: skModule, itemId: ItemId(module: int32(fileIdx), item: 0'i32),
result = PSym(kindImpl: skModule, itemId: itemId(int32(fileIdx), 0'i32),
name: getModuleIdent(graph, filename),
info: newLineInfo(fileIdx, 1, 1))
infoImpl: newLineInfo(fileIdx, 1, 1))
if not isNimIdentifier(result.name.s):
rawMessage(graph.config, errGenerated, "invalid module name: '" & result.name.s &
"'; a module name must be a valid Nim identifier.")

View File

@@ -30,7 +30,6 @@ proc toLowerAscii(a: var string) {.inline.} =
proc flushDot*(conf: ConfigRef) =
## safe to call multiple times
# xxx one edge case not yet handled is when `printf` is called at CT with `compiletimeFFI`.
let stdOrr = if optStdout in conf.globalOptions: stdout else: stderr
let stdOrrKind = toStdOrrKind(stdOrr)
if stdOrrKind in conf.lastMsgWasDot:
@@ -52,7 +51,7 @@ proc makeCString*(s: string): Rope =
result = newStringOfCap(int(s.len.toFloat * 1.1) + 1)
result.add("\"")
for i in 0..<s.len:
# line wrapping of string litterals in cgen'd code was a bad idea, e.g. causes: bug #16265
# line wrapping of string literals in cgen'd code was a bad idea, e.g. causes: bug #16265
# It also makes reading c sources or grepping harder, for zero benefit.
# const MaxLineLength = 64
# if (i + 1) mod MaxLineLength == 0:
@@ -60,12 +59,12 @@ proc makeCString*(s: string): Rope =
toCChar(s[i], result)
result.add('\"')
proc newFileInfo(fullPath: AbsoluteFile, projPath: RelativeFile): TFileInfo =
proc newFileInfo(fullPath: AbsoluteFile, projPath: RelativeFile; kind = fikSource): TFileInfo =
result = TFileInfo(fullPath: fullPath, projPath: projPath,
shortName: fullPath.extractFilename,
quotedFullName: fullPath.string.makeCString,
lines: @[]
)
lines: @[],
kind: kind)
result.quotedName = result.shortName.makeCString
when defined(nimpretty):
if not result.fullPath.isEmpty:
@@ -126,12 +125,42 @@ proc fileInfoIdx*(conf: ConfigRef; filename: AbsoluteFile): FileIndex =
var dummy: bool = false
result = fileInfoIdx(conf, filename, dummy)
proc expandOrPseudo(filename: string): AbsoluteFile =
# `expandFilename` raises OSError when the path does not exist on disk. That is
# fine for a real source path, but a macro can legitimately set a node's
# line-info file to a name that has no file behind it — e.g. the `???` sentinel
# produced by `toFilename` for a NIF-loaded node whose `fileIndex` is unknown
# (FileIndex(-1)). Falling back to the raw name lets the `AbsoluteFile` overload
# register it as a pseudo-path (like `command line`/`stdin`) instead of crashing
# the whole `nim m` child with an unhandled OSError.
try:
result = AbsoluteFile expandFilename(filename)
except OSError:
result = AbsoluteFile filename
proc fileInfoIdx*(conf: ConfigRef; filename: RelativeFile; isKnownFile: var bool): FileIndex =
fileInfoIdx(conf, AbsoluteFile expandFilename(filename.string), isKnownFile)
fileInfoIdx(conf, expandOrPseudo(filename.string), isKnownFile)
proc fileInfoIdx*(conf: ConfigRef; filename: RelativeFile): FileIndex =
var dummy: bool = false
fileInfoIdx(conf, AbsoluteFile expandFilename(filename.string), dummy)
fileInfoIdx(conf, expandOrPseudo(filename.string), dummy)
proc registerNifSuffix*(conf: ConfigRef; suffix: string; isKnownFile: var bool): FileIndex =
result = conf.m.filenameToIndexTbl.getOrDefault(suffix, InvalidFileIdx)
if result == InvalidFileIdx:
isKnownFile = false
result = conf.m.fileInfos.len.FileIndex
conf.m.fileInfos.add(newFileInfo(AbsoluteFile suffix, RelativeFile suffix, fikNifModule))
conf.m.filenameToIndexTbl[suffix] = result
else:
isKnownFile = true
proc fileInfoKind*(conf: ConfigRef; fileIdx: FileIndex): FileInfoKind =
## Returns the kind of a FileIndex (source file or NIF module suffix).
if fileIdx.int >= 0 and fileIdx.int < conf.m.fileInfos.len:
result = conf.m.fileInfos[fileIdx.int].kind
else:
result = fikSource # Default to source for unknown indices
proc newLineInfo*(fileInfoIdx: FileIndex, line, col: int): TLineInfo =
result = TLineInfo(fileIndex: fileInfoIdx)
@@ -224,7 +253,7 @@ proc setDirtyFile*(conf: ConfigRef; fileIdx: FileIndex; filename: AbsoluteFile)
proc setHash*(conf: ConfigRef; fileIdx: FileIndex; hash: string) =
assert fileIdx.int32 >= 0
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc):
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc) or defined(gcYrc):
conf.m.fileInfos[fileIdx.int32].hash = hash
else:
shallowCopy(conf.m.fileInfos[fileIdx.int32].hash, hash)
@@ -232,7 +261,7 @@ proc setHash*(conf: ConfigRef; fileIdx: FileIndex; hash: string) =
proc getHash*(conf: ConfigRef; fileIdx: FileIndex): string =
assert fileIdx.int32 >= 0
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc):
when defined(gcArc) or defined(gcOrc) or defined(gcAtomicArc) or defined(gcYrc):
result = conf.m.fileInfos[fileIdx.int32].hash
else:
shallowCopy(result, conf.m.fileInfos[fileIdx.int32].hash)
@@ -495,6 +524,9 @@ proc sourceLine*(conf: ConfigRef; i: TLineInfo): string =
## 1-based index (matches editor line numbers); 1st line is for i.line = 1
## last valid line is `numLines` inclusive
if i.fileIndex.int32 < 0: return ""
# line 0 means "unknown": nodes synthesized from an IC-loaded template or
# macro body carry no source position.
if i.line.int < 1: return ""
let num = numLines(conf, i.fileIndex)
# can happen if the error points to EOF:
if i.line.int > num: return ""
@@ -648,7 +680,9 @@ template internalAssert*(conf: ConfigRef, e: bool) =
template lintReport*(conf: ConfigRef; info: TLineInfo, beau, got: string, extraMsg = "") =
let m = "'$1' should be: '$2'$3" % [got, beau, extraMsg]
let msg = if optStyleError in conf.globalOptions: errGenerated else: hintName
let msg = if optStyleError in conf.globalOptions: errGenerated
elif optStyleWarning in conf.globalOptions: warnUser
else: hintName
liMessage(conf, info, msg, m, doNothing, instLoc())
proc quotedFilename*(conf: ConfigRef; fi: FileIndex): Rope =

517
compiler/nifbackend.nim Normal file
View File

@@ -0,0 +1,517 @@
#
#
# The Nim Compiler
# (c) Copyright 2025 Andreas Rumpf
#
# See the file "copying.txt", included in this
# distribution, for details about the copyright.
#
## NIF-based C/C++ code generator backend.
##
## This module implements C code generation from precompiled NIF files.
## It traverses the module dependency graph starting from the main module
## and generates C code for all reachable modules.
##
## Usage:
## 1. Compile modules to NIF: nim m mymodule.nim
## 2. Generate C from NIF: nim nifc myproject.nim
import std/[intsets, tables, sets, os, algorithm, syncio, times, strutils]
when defined(nimPreviewSlimSystem):
import std/assertions
import ast, options, lineinfos, modulegraphs, cgendata, cgen,
pathutils, extccomp, msgs, modulepaths, idents, types, ast2nif, typekeys,
cnif
from cgmeth import generateIfMethodDispatchers
import ic / replayer
proc loadModuleDependencies(g: ModuleGraph; mainFileIdx: FileIndex;
nifFiles: var seq[string];
depFlags: set[LoadFlag] = {LoadFullAst}): seq[PrecompiledModule] =
## Traverse the module dependency graph using a stack.
## Returns all modules that need code generation, in dependency order.
##
## The main module is always loaded with its full AST (it is the codegen
## target). `depFlags` governs the rest: the whole-program backend needs every
## module's full AST (it generates code for all of them), but a per-module
## stage codegens only one target, so it loads the others interface-only
## (`depFlags = {}`) — the interface, hooks, methods and the `(replay ...)`
## directives are loaded regardless of `LoadFullAst`, and demanded bodies are
## fetched lazily from the kept-open stream, so the per-module proc-body ASTs
## (the bulk of the memory) are never materialized for non-targets.
# The main module is loaded by its SOURCE FileIndex, but its serialized
# symbols carry the module's NIF suffix. Pre-alias the suffix to the source
# index so that `registerNifSuffix` does not allocate a second FileIndex for
# the same module, which would split its codegen across two C translation
# units (top-level globals in one, procs in the other → undeclared symbols).
g.config.m.filenameToIndexTbl[cachedModuleSuffix(g.config, mainFileIdx)] = mainFileIdx
let mainModule = moduleFromNifFile(g, mainFileIdx, {LoadFullAst})
nifFiles.add toNifFilename(g.config, mainFileIdx)
var stack: seq[ModuleSuffix] = @[]
result = @[]
if mainModule.module != nil:
incl mainModule.module.flagsImpl, sfMainModule
for dep in mainModule.deps:
stack.add dep
var visited = initHashSet[string]()
while stack.len > 0:
let suffix = stack.pop()
if not visited.containsOrIncl(suffix.string):
var isKnownFile = false
let fileIdx = g.config.registerNifSuffix(suffix.string, isKnownFile)
let precomp = moduleFromNifFile(g, fileIdx, depFlags)
if precomp.module != nil:
result.add precomp
nifFiles.add toNifFilename(g.config, fileIdx)
for dep in precomp.deps:
if not visited.contains(dep.string):
stack.add dep
else:
assert false, "Recompiling module is not implemented."
if mainModule.module != nil:
result.add mainModule
proc setupNifBackendModule(g: ModuleGraph; module: PSym): BModule =
## Set up a BModule for code generation from a NIF module.
if g.backend == nil:
g.backend = cgendata.newModuleList(g)
result = cgen.newModule(BModuleList(g.backend), module, g.config, idGeneratorForBackend(module))
proc isMetaIter(t: PType, closure: RootRef): bool =
# openArray/varargs hooks are sem bookkeeping: no real flow ever demands
# them, and generating one pollutes the TU's type cache with a struct
# descriptor for what must remain a (ptr, len) parameter expansion
t.kind in tyMetaTypes + {tyTyped, tyUntyped, tyNone, tyVarargs, tyOpenArray}
proc finishModule(g: ModuleGraph; bmod: BModule) =
# Finalize the module (this adds it to modulesClosed)
# Create an empty stmt list as the init body - genInitCode in writeModule will set it up properly
let initStmt = newNode(nkStmtList)
finalCodegenActions(g, bmod, initStmt)
# NB: the method dispatchers are emitted in `emitMethodDispatchers`,
# between the module loop and this finish loop: their bodies demand the
# method definitions, which can in turn demand definitions from modules
# the backend never loaded — and a TU demand-created during the LAST
# finishModule call would miss `modulesClosed` and never be written.
proc emitMethodDispatchers(g: ModuleGraph) =
## Synthesizes the method dispatcher bodies from the replayed dispatch
## buckets (`registerLoadedMethod`) and emits their definitions into the
## main TU. Main is regenerated on every run, so a dispatcher — whose
## body enumerates the whole program's method set — can never go stale
## inside a cached TU; cross-TU callers prototype it (see genProcLvl3).
let bl = BModuleList(g.backend)
var mainMod: BModule = nil
for m in bl.mods:
if m != nil and m.module != nil and sfMainModule in m.module.flags:
mainMod = m
break
if mainMod == nil: return
generateIfMethodDispatchers(g, mainMod.idgen)
for disp in getDispatchers(g):
if not containsOrIncl(mainMod.declaredThings, disp.id):
genProcLvl3(mainMod, disp)
proc signatureHasMetaType(t: PType; depth: int = 0): bool =
## Whether a routine signature mentions a compile-time/meta element type
## (`typed`/`untyped` — e.g. `echo`'s `varargs[typed]` — typedesc, static,
## generic param). Such routines are expanded at their call sites and never
## emitted standalone, so the per-module owned-routine seeding must skip them
## (`getTypeDescAux(tyTyped)` otherwise). `tfHasMeta` alone misses the varargs
## element case, hence the explicit scan.
result = false
if t == nil or depth > 8: return false
if t.kind == tyGenericBody:
# The uninstantiated template carried as a `tyGenericInst`'s first child
# always mentions its `tyGenericParam` placeholders, but the instance
# itself is fully concrete (e.g. `var CountTable[SigHash]`). Descending
# here would wrongly flag every routine with a generic-instance parameter
# as meta and drop it from the owned-routine seeding -> undefined symbols
# at link (its only definer never emits it).
return false
if t.kind in {tyTyped, tyUntyped, tyTypeDesc, tyStatic, tyGenericParam,
tyAnything, tyFromExpr, tyError}:
return true
for k in t.kids:
if signatureHasMetaType(k, depth + 1): return true
proc generateCodeForModule(g: ModuleGraph; precomp: PrecompiledModule) =
## Generate C code for a single module.
let moduleId = precomp.module.position
var bmod = BModuleList(g.backend).mods[moduleId]
if bmod == nil:
bmod = setupNifBackendModule(g, precomp.module)
# Apply the module's recorded C compile/link directives (passl/passc/...)
# before generating code: the link step needs them (e.g. math's -lm).
replayBackendActions(g, precomp.module, precomp.topLevel)
# Generate code for the module's top-level statements
if precomp.topLevel != nil:
cgen.genTopLevelStmt(bmod, precomp.topLevel)
# Per-module backend: emit the bodies of the routines this module OWNS, not
# only the ones its top-level happens to demand. Procs are serialized as lazy
# `(sd ...)` defs (never as `nkProcDef` statements), so `genTopLevelStmt` never
# reaches them; a routine called only from *other* modules would otherwise be
# emitted by nobody, because every module now merely prototypes its foreign
# callees instead of funnelling their bodies (see `cgen.emitsBodyInThisModule`).
# The merge stage's DCE drops whatever turns out globally dead.
if g.config.cmd == cmdNifC and g.config.icBackendStage == "cg":
let modPos = precomp.module.position
for s in moduleSymbolStubs(ast.program, FileIndex modPos):
if s.itemId.module == modPos and
s.kind in {skProc, skFunc, skConverter, skMethod} and
# Only MODULE-level routines: a nested/closure proc (its owner is a
# proc) captures its enclosing scope and cannot be emitted standalone —
# the captured params have no loc → `expr: param not init`. Nested procs
# are emitted via their enclosing routine's lambda-lifting, so seeding
# the enclosing (module-level) routine already covers them.
s.skipGenericOwner != nil and s.skipGenericOwner.kind == skModule and
s.magic == mNone and
# Skip generic instances: they have no single owning-module top-level
# and are emitted by demand (emit-everywhere, deduped by the merge
# stage). An instance has an empty `genericParamsPos` just like a plain
# concrete proc, so only `sfFromGeneric` tells them apart; seeding one
# would force standalone codegen of an instance body whose `when T is X`
# branches were never folded for this path → `genMagicExpr: mIs`.
sfFromGeneric notin s.flags and
# Every other routine the module owns must be emitted here, exported or
# not: a non-exported helper is still reached from another module when a
# `template`/inline routine expands at a call site there (e.g. msgs'
# `internalErrorImpl` behind the `internalError` template), and that
# caller now only prototypes it. `{.error.}`/`compileTime` sentinels and
# bodyless forward decls are not real codegen targets.
{sfForward, sfImportc, sfCompileTime, sfError} * s.flags == {} and
s.typ != nil and not signatureHasMetaType(s.typ) and
s.ast != nil and s.ast.safeLen > bodyPos and
s.ast[genericParamsPos].kind == nkEmpty and
s.ast[bodyPos].kind != nkEmpty:
# a concrete, non-generic, runtime routine with a real body, owned here
requestProcDef(bmod, s)
proc loadBackendModules(g: ModuleGraph; mainFileIdx: FileIndex):
tuple[modules: seq[PrecompiledModule], precompSys: PrecompiledModule,
nifFiles: seq[string]] =
## Shared by the per-module `cg` and `emit` stages: load system + the main
## module's whole import closure and set up a `BModule` for each, so every
## type/symbol resolves and `getCFile` yields the same path both stages use.
## The main module is loaded by its source index (its NIF suffix is aliased to
## it in `loadModuleDependencies`), so it gets exactly one `BModule`.
##
## Only the main module — the codegen target of the stages that use this — is
## loaded with its full AST; every other module is loaded interface-only so
## the whole program's proc bodies are not materialized into this process (that
## was ~1.8 GB for the compiler's main `cg`). The `link` stage codegens nothing
## and only needs each module's `(replay ...)` directives, which load anyway.
resetForBackend(g)
var isKnownFile = false
let systemFileIdx = registerNifSuffix(g.config, "sysma2dyk", isKnownFile)
g.config.m.systemFileIdx = systemFileIdx
var precompSys = moduleFromNifFile(g, systemFileIdx, {AlwaysLoadInterface})
g.systemModule = precompSys.module
var nifFiles: seq[string] = @[toNifFilename(g.config, systemFileIdx)]
var modules = loadModuleDependencies(g, mainFileIdx, nifFiles, depFlags = {})
# loadModuleDependencies traverses the project's import closure and stops at
# system. The whole-program backend then demand-loads system's own closure
# (locks, allocators, threads, …) during codegen; the per-module backend
# instead makes every one of those a first-class cg/emit target, so load that
# closure here too — otherwise `findTargetModule` cannot resolve their suffix.
block:
var visited = initHashSet[string]()
visited.incl "sysma2dyk"
for m in modules:
visited.incl cachedModuleSuffix(g.config, FileIndex m.module.position)
var stack: seq[ModuleSuffix] = @[]
if precompSys.module != nil:
for dep in precompSys.deps: stack.add dep
while stack.len > 0:
let suffix = stack.pop()
if not visited.containsOrIncl(suffix.string):
var isKnown = false
let fileIdx = registerNifSuffix(g.config, suffix.string, isKnown)
let precomp = moduleFromNifFile(g, fileIdx, {})
if precomp.module != nil:
modules.add precomp
nifFiles.add toNifFilename(g.config, fileIdx)
for dep in precomp.deps: stack.add dep
flushMethodReplays(g)
for m in modules:
discard setupNifBackendModule(g, m.module)
if precompSys.module != nil:
discard setupNifBackendModule(g, precompSys.module)
result = (modules, precompSys, nifFiles)
proc loadDepClosure(g: ModuleGraph; targetSuffix: string):
tuple[modules: seq[PrecompiledModule], precompSys: PrecompiledModule,
target: PrecompiledModule] =
## Per-module `cg`/`emit` for a NON-main target: load system + the target
## module + the target's transitive import closure ONLY — not the whole
## program. This is the "process the one file it is passed" model (à la
## Nimony's `hexer c file.nif`): the foreign symbols the target's codegen
## demands are loaded lazily by `ast2nif.moduleId`, which opens any referenced
## module's NIF index on first touch, so a body in a not-loaded module still
## resolves. The closure is loaded as full `BModule`s only so that the
## incidental `g.mods[pos]` accesses during codegen resolve; system's own
## internal closure (allocators, locks, …) is included because a target's
## emit-everywhere codegen can demand those without importing them directly.
##
## The whole program is no longer loaded in this process, which is what bounds
## per-process memory under nifmake's parallel fan-out (the main module's `cg`,
## which still loads everything for NimMain's init list and the method
## dispatchers, runs essentially alone since every other `.c.nif` precedes it).
resetForBackend(g)
var isKnownFile = false
let systemFileIdx = registerNifSuffix(g.config, "sysma2dyk", isKnownFile)
g.config.m.systemFileIdx = systemFileIdx
let precompSys = moduleFromNifFile(g, systemFileIdx, {AlwaysLoadInterface})
g.systemModule = precompSys.module
var modules: seq[PrecompiledModule] = @[]
var visited = initHashSet[string]()
visited.incl "sysma2dyk"
# Only the target is codegen'd, so only it needs its full AST; the closure is
# loaded interface-only (demanded bodies come lazily from the kept-open
# streams), which is what keeps a per-module process light under parallel fan-out.
var isKnown = false
let targetIdx = registerNifSuffix(g.config, targetSuffix, isKnown)
let target = moduleFromNifFile(g, targetIdx, {LoadFullAst})
visited.incl targetSuffix
var stack: seq[ModuleSuffix] = @[]
if target.module != nil:
modules.add target
for dep in target.deps: stack.add dep
if precompSys.module != nil:
for dep in precompSys.deps: stack.add dep
while stack.len > 0:
let suffix = stack.pop()
if not visited.containsOrIncl(suffix.string):
var isKnown2 = false
let fileIdx = registerNifSuffix(g.config, suffix.string, isKnown2)
let precomp = moduleFromNifFile(g, fileIdx, {})
if precomp.module != nil:
modules.add precomp
for dep in precomp.deps: stack.add dep
flushMethodReplays(g)
for m in modules:
discard setupNifBackendModule(g, m.module)
if precompSys.module != nil:
discard setupNifBackendModule(g, precompSys.module)
result = (modules, precompSys, target)
proc findTargetModule(g: ModuleGraph; modules: seq[PrecompiledModule];
precompSys: PrecompiledModule; suffix: string): PrecompiledModule =
## The loaded module whose NIF suffix is `suffix` (the `--icBackendModule`
## value), or a nil module if none matches.
result = PrecompiledModule(module: nil)
for m in modules:
if cachedModuleSuffix(g.config, FileIndex m.module.position) == suffix:
return m
if precompSys.module != nil and
cachedModuleSuffix(g.config, FileIndex precompSys.module.position) == suffix:
return precompSys
proc generateCgStage(g: ModuleGraph; mainFileIdx: FileIndex) =
## Per-module backend codegen (`--icBackendStage:cg --icBackendModule:<suffix>`):
## generate C for the single module named by `icBackendModule` and write only
## its `.c.nif` artifact (no merge, no `.c` render, no cc/link — those are
## separate nifmake rules).
##
## `findPendingModule` routes every demand into the target (emit-everywhere).
##
## A NON-main target loads only its own import closure (`loadDepClosure`); the
## whole program is no longer pulled into every parallel `cg` process. The main
## module still loads everything (`loadBackendModules`) because NimMain's init
## list and the method dispatchers are whole-program; its `cg` runs essentially
## alone (every other `.c.nif` precedes it), so it does not contend for memory.
let mainSuffix = cachedModuleSuffix(g.config, mainFileIdx)
let targetIsMain = g.config.icBackendModule.len == 0 or
g.config.icBackendModule == mainSuffix
var modules: seq[PrecompiledModule]
var precompSys: PrecompiledModule
var target: PrecompiledModule
if targetIsMain:
var nifFiles: seq[string]
(modules, precompSys, nifFiles) = loadBackendModules(g, mainFileIdx)
if modules.len == 0:
rawMessage(g.config, errGenerated,
"Cannot load NIF file for main module: " & toFullPath(g.config, mainFileIdx))
return
# No whole-program DCE here: each module emits the routines it owns and the
# MERGE stage recomputes the one program-wide live set across all `.c.nif`s.
# Running a whole-program liveness pass over all ~260 NIFs in the main `cg`
# would cost ~900 MB for a result the merge stage throws away.
target = findTargetModule(g, modules, precompSys, g.config.icBackendModule)
else:
# No whole-program load, hence no whole-program DCE: the target emits its
# full demanded closure and the merge stage drops what is globally dead.
(modules, precompSys, target) = loadDepClosure(g, g.config.icBackendModule)
if target.module == nil:
rawMessage(g.config, errGenerated,
"per-module codegen: module not found for suffix: " & g.config.icBackendModule)
return
generateCodeForModule(g, target)
let bl = BModuleList(g.backend)
# The main module also owns the whole-program method dispatchers + NimMain.
if sfMainModule in target.module.flags:
emitMethodDispatchers(g)
# NimMain (generated when the main module is finished) must call every other
# module's init/datInit. Those translation units are produced by their own
# `cg` processes, so the calls are registered here from each `.c.nif` meta
# head — which is why the main module's `cg` runs last, after every other
# `.c.nif` exists. Modules without init code (no `.c.nif`) register nothing.
for m in bl.mods:
if m != nil and sfMainModule notin m.module.flags:
let heads = readCnifHeads(getCFile(m).string & ".nif")
registerReusedModuleToMain(bl, m, heads.initRequired, heads.datInitRequired)
let tb = bl.mods[target.module.position]
if tb != nil:
finishModule(g, tb)
# Writes only the target's `.c.nif` (every other loaded module's TU is empty,
# so `cgenWriteModules` emits no artifact for it). cc/link are NOT run here.
cgenWriteModules(g.backend, g.config)
# Always leave a `.c.nif` for the target, even when the module has no code
# (a leaf library whose procs all emit into their users): the per-module
# nifmake graph declares one `.c.nif` output per `cg` rule, so a missing one
# would re-fire the rule forever. An empty artifact renders to an empty `.c`.
if tb != nil:
let artifact = getCFile(tb).string & ".nif"
if not fileExists(artifact):
writeCnifArtifact("", artifact,
semmedNif = toNifFilename(g.config, FileIndex target.module.position),
moduleBase = $getSomeNameForModule(tb))
proc generateMergeStage(g: ModuleGraph) =
## Per-module backend merge (`--icBackendStage:merge`): a pure artifact
## operation, no module graph loaded. Reads every `.c.nif` the `cg` stages
## wrote, computes the global live set and — for each `'u'`-flagged unique
## definition that several `cg` processes emitted (emit-everywhere) — the one
## artifact allowed to embed its body, and writes the decision the `emit`
## stages consume — the cross-process replacement for what used to be
## in-process first-claimant/DCE coordination.
let nimcache = getNimcacheDir(g.config).string
var files: seq[string] = @[]
for artifact in walkFiles(nimcache / "*.c.nif"):
files.add artifact
sort files
let decision = computeMergeDecision(files)
if decision.broken:
rawMessage(g.config, errGenerated,
"per-module backend merge: a .c.nif artifact is missing or unparsable")
return
writeMergeDecision(nimcache / MergeDecisionFile, decision)
if isDefined(g.config, "icDceCheck"):
stderr.writeLine "[icMerge] artifacts: " & $files.len &
" live: " & $decision.live.len & " defs: " & $decision.defs &
" liveDefs: " & $decision.liveDefs & " owned: " & $decision.owners.len
proc generateEmitStage(g: ModuleGraph; mainFileIdx: FileIndex) =
## Per-module backend emit (`--icBackendStage:emit --icBackendModule:<suffix>`):
## render the target module's final `.c` from its `.c.nif` and the merge
## decision. Loads the target the same way `cg` does so `getCFile` returns the
## identical path `cg` wrote to (the main module's source-vs-suffix aliasing in
## particular); no codegen runs. A non-main target loads only its own closure
## (`loadDepClosure`) so emit, like `cg`, stays bounded under parallel fan-out.
let mainSuffix = cachedModuleSuffix(g.config, mainFileIdx)
let targetIsMain = g.config.icBackendModule.len == 0 or
g.config.icBackendModule == mainSuffix
var modules: seq[PrecompiledModule]
var precompSys: PrecompiledModule
var target: PrecompiledModule
if targetIsMain:
var nifFiles: seq[string]
(modules, precompSys, nifFiles) = loadBackendModules(g, mainFileIdx)
if modules.len == 0:
rawMessage(g.config, errGenerated,
"Cannot load NIF file for main module: " & toFullPath(g.config, mainFileIdx))
return
target = findTargetModule(g, modules, precompSys, g.config.icBackendModule)
else:
(modules, precompSys, target) = loadDepClosure(g, g.config.icBackendModule)
if target.module == nil:
rawMessage(g.config, errGenerated,
"per-module emit: module not found for suffix: " & g.config.icBackendModule)
return
let decision = readMergeDecision(getNimcacheDir(g.config).string / MergeDecisionFile)
if decision.broken:
rawMessage(g.config, errGenerated,
"per-module emit: missing or unparsable merge decision " & MergeDecisionFile)
return
let bmod = BModuleList(g.backend).mods[target.module.position]
let cfile = getCFile(bmod).string
let artifact = cfile & ".nif"
var dropped = 0
let code = renderCFromArtifact(artifact, decision, extractFilename(artifact), dropped)
writeFile(cfile, code)
if isDefined(g.config, "icDceCheck"):
stderr.writeLine "[icEmit] " & extractFilename(cfile) & " dropped " &
$dropped & " bodies (" & $code.len & " bytes)"
proc generateLinkStage(g: ModuleGraph; mainFileIdx: FileIndex) =
## Per-module backend link (`--icBackendStage:link`): the `emit` stages have
## written every module's `.c`; register them and run the C compiler + linker
## once via `extccomp.callCCompiler` (which parallelizes the per-file cc and
## skips up-to-date objects itself). No codegen runs — the graph is loaded only
## so `getCFile` yields each module's emitted `.c` path.
let (modules, precompSys, _) = loadBackendModules(g, mainFileIdx)
if modules.len == 0:
rawMessage(g.config, errGenerated,
"Cannot load NIF file for main module: " & toFullPath(g.config, mainFileIdx))
return
# The per-module `cg` processes each collect their module's C compile/link
# directives (`{.passL: "-lm".}` etc.) via `replayBackendActions`, but those
# live in the cg process and never reach this separate link process. Re-collect
# every loaded module's directives here so the final `callCCompiler` sees them
# (without this, math's `-lm` is lost → undefined `floor`/`pow`/… at link).
for m in modules:
replayBackendActions(g, m.module, m.topLevel)
if precompSys.module != nil:
replayBackendActions(g, precompSys.module, precompSys.topLevel)
let bl = BModuleList(g.backend)
for m in bl.mods:
if m != nil:
let cfile = getCFile(m)
# Only modules that are their own cg/emit target produced a `.c`; the rest
# (extra members of system's closure that no build rule targets) had their
# code emit-everywhere'd into the targets, so they have no file to compile.
if not fileExists(cfile.string): continue
var cf = Cfile(nimname: m.module.name.s, cname: cfile,
obj: completeCfilePath(g.config, toObjFile(g.config, cfile)),
flags: {})
addFileToCompile(g.config, cf)
if g.config.cmd != cmdTcc:
extccomp.callCCompiler(g.config)
proc generateCode*(g: ModuleGraph; mainFileIdx: FileIndex) =
## Main entry point for NIF-based C code generation.
## Traverses the module dependency graph and generates C code.
if g.config.icBackendStage == "cg":
generateCgStage(g, mainFileIdx)
return
elif g.config.icBackendStage == "merge":
generateMergeStage(g)
return
elif g.config.icBackendStage == "emit":
generateEmitStage(g, mainFileIdx)
return
elif g.config.icBackendStage == "link":
generateLinkStage(g, mainFileIdx)
return
else:
rawMessage(g.config, errGenerated,
"the per-module NIF backend requires --icBackendStage:cg|merge|emit|link")

1690
compiler/nifgen.nim Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -183,12 +183,6 @@ func `<`*(a: ExprIndex, b: ExprIndex): bool =
func `<=`*(a: ExprIndex, b: ExprIndex): bool =
a.int16 <= b.int16
func `>`*(a: ExprIndex, b: ExprIndex): bool =
a.int16 > b.int16
func `>=`*(a: ExprIndex, b: ExprIndex): bool =
a.int16 >= b.int16
func `==`*(a: ExprIndex, b: ExprIndex): bool =
a.int16 == b.int16
@@ -919,7 +913,7 @@ proc infix(ctx: NilCheckerContext, l: PNode, r: PNode, magic: TMagic): PNode =
newSymNode(op, r.info),
l,
r)
result.typ() = newType(tyBool, ctx.idgen, nil)
result.typ = newType(tyBool, ctx.idgen, nil)
proc prefixNot(ctx: NilCheckerContext, node: PNode): PNode =
var cache = newIdentCache()
@@ -929,7 +923,7 @@ proc prefixNot(ctx: NilCheckerContext, node: PNode): PNode =
result = nkPrefix.newTree(
newSymNode(op, node.info),
node)
result.typ() = newType(tyBool, ctx.idgen, nil)
result.typ = newType(tyBool, ctx.idgen, nil)
proc infixEq(ctx: NilCheckerContext, l: PNode, r: PNode): PNode =
infix(ctx, l, r, mEqRef)

View File

@@ -10,6 +10,11 @@ define:nimPreviewProcConversion
define:nimPreviewRangeDefault
define:nimPreviewNonVarDestructor
define:nimPreviewCheckedClose
define:nimPreviewAsmSemSymbol
define:nimPreviewCStringComparisons
#define:nimPreviewDuplicateModuleError
# Incompatible with Nimony's compat2.nim for now
threads:off
#import:"$projectpath/testability"
@@ -48,7 +53,7 @@ define:useStdoutAsStdmsg
@if nimUseStrictDefs:
experimental:strictDefs # deadcode
experimental:strictDefs
warningAsError[Uninit]:on
warningAsError[ProveInit]:on
@end
@@ -61,3 +66,7 @@ define:useStdoutAsStdmsg
@if nimHasVtables:
experimental:vtables
@end
@if nimHasImplicitRangeConversion:
warning[ImplicitRangeConversion]:off
@end

View File

@@ -28,10 +28,13 @@ import
commands, options, msgs, extccomp, main, idents, lineinfos, cmdlinehelper,
pathutils, modulegraphs
from ast2nif import registerNifAstTags
from icconfig import ensureIcConfig
from std/browsers import openDefaultBrowser
from nodejs import findNodeJs
when hasTinyCBackend:
when defined(tinyc): # == hasTinyCBackend; spelled out for the IC dep scanner
import tccgen
when defined(profiler) or defined(memProfiler):
@@ -96,6 +99,11 @@ proc getNimRunExe(conf: ConfigRef): string =
result = ""
proc handleCmdLine(cache: IdentCache; conf: ConfigRef) =
# NIF tag registration must not depend on module init order — the IC-built
# compiler orders module init calls differently and the top-level
# `registerTag` initializers then ran against a not-yet-initialized pool,
# corrupting every written NIF (see registerNifAstTags).
registerNifAstTags()
let self = NimProg(
supportsStdinFile: true,
processCmdLine: processCmdLine
@@ -107,6 +115,14 @@ proc handleCmdLine(cache: IdentCache; conf: ConfigRef) =
self.processCmdLineAndProjectPath(conf)
# `nim ic` driver: ensure the precompiled config exists (produced by a separate
# `nim icconfig` process, skipped when nothing changed) BEFORE config loading,
# so `loadConfigs` replays it instead of re-parsing the `nim.cfg` chain — the
# driver runs on the exact same config its children will. See icconfig.nim.
when not defined(nimKochBootstrap):
if conf.cmd == cmdIc:
ensureIcConfig(conf)
var graph = newModuleGraph(cache, conf)
if not self.loadConfigsAndProcessCmdLine(cache, conf, graph):
return
@@ -118,9 +134,14 @@ proc handleCmdLine(cache: IdentCache; conf: ConfigRef) =
if conf.selectedGC == gcUnselected:
if conf.backend in {backendC, backendCpp, backendObjc} or
(conf.cmd in cmdDocLike and conf.backend != backendJs) or
conf.cmd == cmdGendepend:
conf.cmd in {cmdGendepend, cmdNifC, cmdIc, cmdM}:
initOrcDefines(conf)
if conf.selectedStrings == stringSso and
conf.selectedGC notin {gcArc, gcOrc, gcYrc, gcAtomicArc}:
rawMessage(conf, errGenerated,
"--strings:sso requires --mm:arc, --mm:orc, --mm:yrc, or --mm:atomicArc")
mainCommand(graph)
if conf.hasHint(hintGCStats): echo(GC_getStatistics())
#echo(GC_getStatistics())

View File

@@ -11,7 +11,7 @@
import
llstream, commands, msgs, lexer, ast,
options, idents, wordrecg, lineinfos, pathutils, scriptconfig
options, idents, wordrecg, lineinfos, pathutils, scriptconfig, icconfig
import std/[os, strutils, strtabs]
@@ -207,7 +207,6 @@ proc parseAssignment(L: var Lexer, tok: var Token;
checkSymbol(L, tok)
val.add($tok)
confTok(L, tok, config, condStack)
config.currentConfigDir = parentDir(filename.string)
if percent:
processSwitch(s, strtabs.`%`(val, config.configVars,
{useEnvironment, useEmpty}), passPP, info, config)
@@ -247,8 +246,20 @@ proc getSystemConfigPath*(conf: ConfigRef; filename: RelativeFile): AbsoluteFile
proc loadConfigs*(cfg: RelativeFile; cache: IdentCache; conf: ConfigRef; idgen: IdGenerator) =
setDefaultLibpath(conf)
# The `nim ic` driver and its `nim m`/`nim nifc` children replay the precompiled
# config (produced once by a separate `nim icconfig` process — see
# `icconfig.ensureIcConfig`, which sets `icPreparsedConfig` for the driver
# before this runs; the children get it as a forwarded `--icPreparsedConfig`
# argument) instead of re-reading the `nim.cfg` chain and re-running
# `config.nims` in the VM. A missing/format-incompatible artifact returns false:
# fall through to a normal parse (this is also the path the `nim icconfig`
# producer itself takes, since it runs with no `icPreparsedConfig`).
if conf.icPreparsedConfig.len > 0 and applyIcConfig(conf, conf.icPreparsedConfig):
return
template readConfigFile(path) =
let configPath = path
conf.currentConfigDir = configPath.splitFile.dir.string
setConfigVar(conf, "selfDir", conf.currentConfigDir)
if readConfigFile(configPath, cache, conf):
conf.configFiles.add(configPath)

View File

@@ -128,7 +128,7 @@ proc createInterpreter*(scriptName: string;
if conf.libpath.isEmpty: conf.libpath = AbsoluteDir p
var m = graph.makeModule(scriptName)
incl(m.flags, sfMainModule)
incl(m, sfMainModule)
var idgen = idGeneratorFromModule(m)
var vm = newCtx(m, cache, graph, idgen)
vm.mode = emRepl
@@ -168,7 +168,7 @@ proc runRepl*(r: TLLRepl;
if supportNimscript: defineSymbol(conf.symbols, "nimconfig")
when hasFFI: defineSymbol(graph.config.symbols, "nimffi")
var m = graph.makeStdinModule()
incl(m.flags, sfMainModule)
incl(m, sfMainModule)
var idgen = idGeneratorFromModule(m)
if supportNimscript: graph.vm = setupVM(m, cache, "stdin", graph, idgen)

View File

@@ -84,7 +84,7 @@ proc toTreeSet*(conf: ConfigRef; s: TBitSet, settype: PType, info: TLineInfo): P
elemType = settype[0]
first = firstOrd(conf, elemType).toInt64
result = newNodeI(nkCurly, info)
result.typ() = settype
result.typ = settype
result.info = info
e = 0
while e < s.len * ElemSize:
@@ -101,7 +101,7 @@ proc toTreeSet*(conf: ConfigRef; s: TBitSet, settype: PType, info: TLineInfo): P
result.add aa
else:
n = newNodeI(nkRange, info)
n.typ() = elemType
n.typ = elemType
n.add aa
let bb = newIntTypeNode(b + first, elemType)
bb.info = info

View File

@@ -28,11 +28,14 @@ type
hasReturn, hasBreak: bool
label: PSym # can be nil
parent: ptr BasicBlock
symToDel: seq[PNode]
Con = object
somethingTodo: bool
inFinally: int
proc invalidateWasMoved(c: var BasicBlock; x: PNode)
proc nestedBlock(parent: var BasicBlock; kind: TNodeKind): BasicBlock =
BasicBlock(wasMovedLocs: @[], kind: kind, hasReturn: false, hasBreak: false,
label: nil, parent: addr(parent))
@@ -62,6 +65,10 @@ proc mergeBasicBlockInfo(parent: var BasicBlock; this: BasicBlock) {.inline.} =
if this.hasReturn:
parent.wasMovedLocs.setLen 0
parent.hasReturn = true
elif this.symToDel.len > 0:
parent.symToDel = this.symToDel
for i in this.symToDel:
invalidateWasMoved(parent, i)
proc wasMovedTarget(matches: var IntSet; branch: seq[PNode]; moveTarget: PNode): bool =
result = false
@@ -149,6 +156,7 @@ proc analyse(c: var Con; b: var BasicBlock; n: PNode) =
# any usage of the location before destruction implies we
# cannot elide the 'wasMoved(x)':
b.invalidateWasMoved n
b.symToDel.add n
of nkNone..pred(nkSym), succ(nkSym)..nkNilLit, nkTypeSection, nkProcDef, nkConverterDef,
nkMethodDef, nkIteratorDef, nkMacroDef, nkTemplateDef, nkLambda, nkDo,

View File

@@ -25,10 +25,36 @@ const
useEffectSystem* = true
useWriteTracking* = false
hasFFI* = defined(nimHasLibFFI)
copyrightYear* = "2024"
copyrightYear* = "2026"
nimEnableCovariance* = defined(nimEnableCovariance)
icFormatVersion* = "6"
## Version of the IC cache format (the sem-NIF module layout written by
## ast2nif.nim plus the iface/impl/edges side files). Bump it whenever
## that layout changes: `commandIc` wipes a nimcache whose `ic.version`
## stamp differs, instead of letting a newer reader mis-parse records
## written by an older compiler (nifmake's rebuild check is mtime-only
## and knows nothing about format changes).
## v2: iface cookie hashes routine SIGNATURES only (no inline-semantics
## body folding); body access now records a NeedsImpl edge instead. A v1
## cache mixes body-sensitive and body-insensitive cookies, so it must be
## wiped rather than warm-rebuilt.
## v3: added the `.s.deps` sidecar (real post-sem imports) and switched the
## macro-generated-import discovery from `icmissing.txt` to it.
## v4: backend C-name scheme change — the module suffix is now the trailing
## token (`name_u<disamb>__<suffix>`, was `name__<suffix>_u<disamb>`), so
## cached `.c.nif` artifacts hold incompatible names and must be wiped.
## v5: data definitions (consts, RTTI) are now wrapped in droppable `'d'`
## cdef directives with an always-present extern declaration, so the
## per-module merge stage can assign them a single owner; old `.c.nif`
## artifacts lack the wrappers.
## v6: `signatureHash`/`hashType` of a builtin type class (`object`, `tuple`,
## `proc`, ...) no longer mixes in the placeholder son's process-local type
## id, so its hash is stable across the NIF boundary (was breaking
## nim-serialization's auto-serialization lookup under IC). The sem-NIF
## macrocache entries and baked generic-instance bodies hold the old hashes.
type # please make sure we have under 32 options
# (improves code efficiency a lot!)
TOption* = enum # **keep binary compatible**
@@ -68,6 +94,7 @@ type # please make sure we have under 32 options
optUseNimcache, # save artifacts (including binary) in $nimcache
optStyleHint, # check that the names adhere to NEP-1
optStyleError, # enforce that the names adhere to NEP-1
optStyleWarning, # emit style checks as warnings
optStyleUsages, # only enforce consistent **usages** of the symbol
optSkipSystemConfigFile, # skip the system's cfg/nims config file
optSkipProjConfigFile, # skip the project's cfg/nims config file
@@ -110,6 +137,10 @@ type # please make sure we have under 32 options
optEnableDeepCopy # ORC specific: enable 'deepcopy' for all types.
optShowNonExportedFields # for documentation: show fields that are not exported
optJsBigInt64 # use bigints for 64-bit integers in JS
optDocRaw # for documentation: Don't render markdown for JSON output
optItaniumMangle # mangling follows the Itanium spec
optCompress # turn on AST compression by converting it to NIF
optWithinConfigSystem # we still compile within the configuration system
TGlobalOptions* = set[TGlobalOption]
@@ -139,6 +170,7 @@ type
backendCpp = "cpp"
backendJs = "js"
backendObjc = "objc"
backendNif = "nif"
# backendNimscript = "nimscript" # this could actually work
# backendLlvm = "llvm" # probably not well supported; was cmdCompileToLLVM
@@ -151,7 +183,6 @@ type
cmdCheck # semantic checking for whole project
cmdM # only compile a single
cmdParse # parse a single file (for debugging)
cmdRod # .rod to some text representation (for debugging)
cmdIdeTools # ide tools (e.g. nimsuggest)
cmdNimscript # evaluate nimscript
cmdDoc0
@@ -171,10 +202,14 @@ type
cmdNop
cmdJsonscript # compile a .json build file
# old unused: cmdInterpret, cmdDef: def feature (find definition for IDEs)
cmdCompileToNif
cmdNifC # generate C code from NIF files
cmdIc # generate .build.nif for nifmake
cmdIcConfig # `nim ic`'s precompiled-config producer (writes ic_config.cfg.nif)
const
cmdBackends* = {cmdCompileToC, cmdCompileToCpp, cmdCompileToOC,
cmdCompileToJS, cmdCrun}
cmdCompileToJS, cmdCrun, cmdCompileToNif}
cmdDocLike* = {cmdDoc0, cmdDoc, cmdDoc2tex, cmdJsondoc0, cmdJsondoc,
cmdCtags, cmdBuildindex}
@@ -187,6 +222,7 @@ type
gcRegions = "regions"
gcArc = "arc"
gcOrc = "orc"
gcYrc = "yrc" # thread-safe ORC (concurrent cycle collector)
gcAtomicArc = "atomicArc"
gcMarkAndSweep = "markAndSweep"
gcHooks = "hooks"
@@ -248,8 +284,16 @@ type
## Useful for libraries that rely on local passC
jsNoLambdaLifting
## Old transformation for closures in JS backend
noStrictDefs
## disable "strictdefs"
noPanicOnExcept
## don't panic on bare except
procParamTypeBackendAliases
## Keep the old proc type compatibility rules that ignore backend
## c type aliases.
injectedSymbolRedefinition
## Allow a template to inject a symbol *definition* that is then emitted
## more than once (e.g. a `typed` argument captured by a `{.dirty.}`
## template and re-emitted). This is a redefinition and rejected by
## default; enabling this restores the old, unsound behavior. See #25693.
SymbolFilesOption* = enum
disabledSf, writeOnlySf, readOnlySf, v2Sf, stressTest
@@ -258,6 +302,10 @@ type
ccNone, ccGcc, ccNintendoSwitch, ccLLVM_Gcc, ccCLang, ccBcc, ccVcc,
ccTcc, ccEnv, ccIcl, ccIcc, ccClangCl, ccHipcc, ccNvcc
StringsMode* = enum
stringDefault = "default"
stringSso = "sso"
ExceptionSystem* = enum
excNone, # no exception system selected yet
excSetjmp, # setjmp based exception handling
@@ -357,11 +405,55 @@ type
implicitCmd*: bool # whether some flag triggered an implicit `command`
selectedGC*: TGCMode # the selected GC (+)
exc*: ExceptionSystem
selectedStrings*: StringsMode
hintProcessingDots*: bool # true for dots, false for filenames
verbosity*: int # how verbose the compiler is
numberOfProcessors*: int # number of processors
lastCmdTime*: float # when caas is enabled, we measure each command
symbolFiles*: SymbolFilesOption
ic*: bool # whether ic is enabled
icGroup*: HashSet[string] # under `nim m`: absolute paths of the modules in
# this strongly-connected import group. They are all
# compiled from source in one process (so mutual
# recursion resolves in-memory) and each gets its NIF
# written, instead of being loaded from a precompiled
# NIF. See `compiler/deps.nim` (SCC grouping).
icProject*: string # under `nim m`/`nim nifc`: absolute path of the
# ORIGINAL project file. The child's own project file
# is the module being compiled, which would make that
# module's package the "main package" and unfilter
# foreign-package diagnostics; the real project
# restores whole-program filtering semantics.
icPreparsedConfig*: string # under the `nim ic` driver and its `nim m`/`nim nifc`
# children: path of the precompiled config artifact.
# When set, `loadConfigs` replays the recorded
# config-file switches from it instead of re-reading
# the `nim.cfg` chain and re-running `config.nims`
# (which the VM makes expensive) per process. The
# artifact itself is produced by a separate
# `nim icconfig` process (see `cmdIcConfig`).
icConfigOut*: string # under `nim icconfig`: the path to write the
# precompiled config artifact to (set via `--o`).
icConfigSwitches*: seq[tuple[switch, arg: string]]
# the config-file (`passPP`) switches applied while
# loading config, in order. Recorded by every nim
# process; only the `ic` driver serialises them.
# Path-search switches are excluded — the driver
# forwards the resolved `searchPaths` as `--path`.
icBackendStage*: string # under `nim nifc`: which stage of the per-module
# backend this invocation runs — "cg" (codegen one
# module to its `.c.nif`), "merge" (global liveness
# + owner assignment across all `.c.nif`), "emit"
# (render one module's `.c` from its `.c.nif` + the
# merge decision), "link" (cc + link every emitted
# `.c`). Empty = whole-program backend (load all,
# codegen+DCE+cc+link in one process). The stages
# are wired as nifmake rules by `deps.nim`'s backend
# build file. See `compiler/nifbackend.nim`.
icBackendModule*: string # under `nim nifc` with icBackendStage in {cg,emit}:
# the NIF module suffix this invocation codegens or
# emits. The other modules are loaded only so types
# resolve; their definitions are referenced extern.
spellSuggestMax*: int # max number of spelling suggestions for typos
cppDefines*: HashSet[string] # (*)
@@ -385,6 +477,7 @@ type
warnCounter*: int
errorMax*: int
maxLoopIterationsVM*: int ## VM: max iterations of all loops
maxCallDepthVM*: int ## VM: max call depth
isVmTrace*: bool
configVars*: StringTableRef
symbols*: StringTableRef ## We need to use a StringTableRef here as defined
@@ -403,9 +496,16 @@ type
projectPath*: AbsoluteDir # holds a path like /home/alice/projects/nim/compiler/
projectFull*: AbsoluteFile # projectPath/projectName
projectIsStdin*: bool # whether we're compiling from stdin
stdinFile*: AbsoluteFile # Filename to use in messages for stdin
lastMsgWasDot*: set[StdOrrKind] # the last compiler message was a single '.'
projectMainIdx*: FileIndex # the canonical path id of the main module
projectMainIdx2*: FileIndex # consider merging with projectMainIdx
isMainModule*: bool # `nim m`/IC only: whether the single module being
# semantically checked is the program's real entry point.
# Under IC every module is compiled via `nim m` (which sets
# `sfMainModule` so the module writes its own NIF), so
# `sfMainModule` can no longer answer `isMainModule`. The IC
# build file passes `--isMainModule:on` for the root module.
command*: string # the main command (e.g. cc, check, scan, etc)
commandArgs*: seq[string] # any arguments after the main command
commandLine*: string
@@ -504,7 +604,7 @@ const
optHints, optStackTrace, optLineTrace, # consider adding `optStackTraceMsgs`
optTrMacros, optStyleCheck, optCursorInference}
DefaultGlobalOptions* = {optThreadAnalysis, optExcessiveStackTrace,
optJsBigInt64}
optJsBigInt64, optItaniumMangle}
proc getSrcTimestamp(): DateTime =
try:
@@ -562,6 +662,7 @@ proc newConfigRef*(): ConfigRef =
arcToExpand: newStringTable(modeStyleInsensitive),
m: initMsgConfig(),
cppDefines: initHashSet[string](),
icGroup: initHashSet[string](),
headerFile: "", features: {}, legacyFeatures: {},
configVars: newStringTable(modeStyleInsensitive),
symbols: newStringTable(modeStyleInsensitive),
@@ -579,6 +680,7 @@ proc newConfigRef*(): ConfigRef =
projectPath: AbsoluteDir"", # holds a path like /home/alice/projects/nim/compiler/
projectFull: AbsoluteFile"", # projectPath/projectName
projectIsStdin: false, # whether we're compiling from stdin
stdinFile: AbsoluteFile"stdinfile",
projectMainIdx: FileIndex(0'i32), # the canonical path id of the main module
command: "", # the main command (e.g. cc, check, scan, etc)
commandArgs: @[], # any arguments after the main command
@@ -600,6 +702,7 @@ proc newConfigRef*(): ConfigRef =
arguments: "",
suggestMaxResults: 10_000,
maxLoopIterationsVM: 10_000_000,
maxCallDepthVM: 2_000,
vmProfileData: newProfileData(),
spellSuggestMax: spellSuggestSecretSauce,
currentConfigDir: ""
@@ -635,6 +738,7 @@ proc isDefined*(conf: ConfigRef; symbol: string): bool =
of "x86": result = conf.target.targetCPU == cpuI386
of "itanium": result = conf.target.targetCPU == cpuIa64
of "x8664": result = conf.target.targetCPU == cpuAmd64
of "wasm": result = conf.target.targetCPU in {cpuWasm32, cpuWasm64}
of "posix", "unix":
result = conf.target.targetOS in {osLinux, osMorphos, osSkyos, osIrix, osPalmos,
osQnx, osAtari, osAix,
@@ -684,6 +788,7 @@ template quitOrRaise*(conf: ConfigRef, msg = "") =
proc importantComments*(conf: ConfigRef): bool {.inline.} = conf.cmd in cmdDocLike + {cmdIdeTools}
proc usesWriteBarrier*(conf: ConfigRef): bool {.inline.} = conf.selectedGC >= gcRefc
proc usesSso*(conf: ConfigRef): bool {.inline.} = conf.selectedStrings == stringSso
template compilationCachePresent*(conf: ConfigRef): untyped =
false
@@ -1034,6 +1139,9 @@ proc isDynlibOverride*(conf: ConfigRef; lib: string): bool =
proc showNonExportedFields*(conf: ConfigRef) =
incl(conf.globalOptions, optShowNonExportedFields)
proc docRawOutput*(conf: ConfigRef) =
incl(conf.globalOptions, optDocRaw)
proc expandDone*(conf: ConfigRef): bool =
result = conf.ideCmd == ideExpand and conf.expandLevels == 0 and conf.expandProgress

View File

@@ -33,7 +33,7 @@ proc getPackage*(conf: ConfigRef; cache: IdentCache; fileIdx: FileIndex): PSym =
pkgIdent = getIdent(cache, pkgName)
newSym(skPackage, pkgIdent, idGeneratorForPackage(int32(fileIdx)), nil, info)
func getPackageSymbol*(sym: PSym): PSym =
proc getPackageSymbol*(sym: PSym): PSym =
## Return the owning package symbol.
assert sym != nil
result = sym
@@ -41,18 +41,18 @@ func getPackageSymbol*(sym: PSym): PSym =
result = result.owner
assert result != nil, repr(sym.info)
func getPackageId*(sym: PSym): int =
proc getPackageId*(sym: PSym): int =
## Return the owning package ID.
sym.getPackageSymbol.id
func belongsToProjectPackage*(conf: ConfigRef, sym: PSym): bool =
proc belongsToProjectPackage*(conf: ConfigRef, sym: PSym): bool =
## Return whether the symbol belongs to the project's package.
##
## See Also:
## * `modulegraphs.belongsToStdlib`
conf.mainPackageId == sym.getPackageId
func belongsToProjectPackageMaybeNil*(conf: ConfigRef, sym: PSym): bool =
proc belongsToProjectPackageMaybeNil*(conf: ConfigRef, sym: PSym): bool =
## Return whether the symbol belongs to the project's package.
## Returns `false` if `sym` is nil.
##

View File

@@ -13,7 +13,7 @@
import ast, types, msgs, idents, renderer, wordrecg, trees,
options
import std/strutils
import std/[strutils, assertions]
# we precompile the pattern here for efficiency into some internal
# stack based VM :-) Why? Because it's fun; I did no benchmarks to see if that
@@ -216,6 +216,11 @@ proc exprRoot*(n: PNode; allowCalls = true): PSym =
else:
break
proc isAssignable*(owner: PSym, n: PNode): TAssignableResult
proc isLentableBranch(owner: PSym, n: PNode): bool =
result = isAssignable(owner, n) in {arLentValue, arAddressableConst, arLentValue}
proc isAssignable*(owner: PSym, n: PNode): TAssignableResult =
## 'owner' can be nil!
result = arNone
@@ -308,6 +313,35 @@ proc isAssignable*(owner: PSym, n: PNode): TAssignableResult =
# nkVarTy denotes an lvalue, but the example above is the only
# possible code which will get us here
result = arLValue
of nkIfExpr, nkIfStmt:
# allow 'if' expressions to be lent if all branches are lentable
for branch in n:
if branch.len == 2:
if not isLentableBranch(owner, branch[1]):
return
elif branch.len == 1:
if not isLentableBranch(owner, branch[0]):
return
else:
raiseAssert "Malformed `if` statement in isAssignable"
result = arLentValue
of nkCaseStmt:
# allow 'case' expressions to be lent if all branches are lentable
for i in 1 ..< n.len:
let branch = n[i]
case branch.kind
of nkOfBranch:
if not isLentableBranch(owner, branch[^1]):
return
of nkElifBranch:
if not isLentableBranch(owner, branch[1]):
return
of nkElse:
if not isLentableBranch(owner, branch[0]):
return
else:
raiseAssert "Malformed `case` statement in isAssignable"
result = arLentValue
else:
discard

View File

@@ -54,7 +54,10 @@ import
when not defined(nimCustomAst):
import ast
else:
when defined(nimCustomAst):
# NOTE: explicit negated `when` rather than `else:` — nifler's dep scanner
# guards `when`/`elif` imports with their condition but emits `else:` imports
# unconditionally, which would wrongly schedule this module under `nim ic`.
import plugins / customast
import std/strutils
@@ -638,8 +641,9 @@ proc semiStmtList(p: var Parser, result: PNode) =
getTok(p)
if p.tok.tokType == tkParRi:
break
elif not (sameInd(p) or realInd(p)):
parMessage(p, errInvalidIndentation)
# ignore indent:
#elif not (sameOrNoInd(p) or realInd(p)):
# parMessage(p, errInvalidIndentation)
let a = complexOrSimpleStmt(p)
if a.kind == nkEmpty:
parMessage(p, errExprExpected, p.tok)
@@ -699,10 +703,12 @@ proc parsePar(p: var Parser): PNode =
asgn.add b
result.add(asgn)
if p.tok.tokType == tkSemiColon:
getTok(p)
semiStmtList(p, result)
elif p.tok.tokType == tkSemiColon:
# stmt context:
result.add(a)
getTok(p)
semiStmtList(p, result)
else:
a = colonOrEquals(p, a)
@@ -1153,7 +1159,10 @@ proc parseParamList(p: var Parser, retColon = true): PNode =
parMessage(p, errGenerated, "the syntax is 'parameter: var T', not 'var parameter: T'")
break
else:
parMessage(p, "expected closing ')'")
if p.tok.tokType in tokKeywordLow..tokKeywordHigh:
parMessage(p, errGenerated, "'" & $p.tok.ident.s & "' is a keyword and cannot be used as a parameter name")
else:
parMessage(p, "expected closing ')'")
break
result.add(a)
if p.tok.tokType notin {tkComma, tkSemiColon}: break
@@ -2107,12 +2116,28 @@ proc parseObjectCase(p: var Parser): PNode =
#| objectBranches = objectBranch (IND{=} objectBranch)*
#| (IND{=} 'elif' expr colcom objectPart)*
#| (IND{=} 'else' colcom objectPart)?
#| objectCase = 'case' declColonEquals ':'? COMMENT?
#| objectCase = 'case' (declColonEquals / pragma)? ':'? COMMENT?
#| (IND{>} objectBranches DED
#| | IND{=} objectBranches)
result = newNodeP(nkRecCase, p)
getTokNoInd(p)
var a = parseIdentColonEquals(p, {withPragma})
getTok(p)
if p.tok.tokType != tkOf:
# of case will be handled later
if p.tok.indent >= 0: parMessage(p, errInvalidIndentation)
var a: PNode
if p.tok.tokType in {tkSymbol, tkAccent}:
a = parseIdentColonEquals(p, {withPragma})
else:
a = newNodeP(nkIdentDefs, p)
if p.tok.tokType == tkCurlyDotLe:
var prag = newNodeP(nkPragmaExpr, p)
prag.add(p.emptyNode)
prag.add(parsePragma(p))
a.add(prag)
else:
a.add(p.emptyNode)
a.add(p.emptyNode)
a.add(p.emptyNode)
result.add(a)
if p.tok.tokType == tkColon: getTok(p)
flexComment(p, result)
@@ -2219,14 +2244,17 @@ proc parseTypeClassParam(p: var Parser): PNode =
proc parseTypeClass(p: var Parser): PNode =
#| conceptParam = ('var' | 'out' | 'ptr' | 'ref' | 'static' | 'type')? symbol
#| conceptDecl = 'concept' conceptParam ^* ',' (pragma)? ('of' typeDesc ^* ',')?
#| conceptDecl = 'concept' (conceptParam ^* ',' (pragma)?)? ('of' typeDesc ^* ',')?
#| &IND{>} stmt
result = newNodeP(nkTypeClassTy, p)
getTok(p)
if p.tok.tokType == tkComment:
skipComment(p, result)
if p.tok.indent < 0:
if p.tok.tokType == tkOf and p.tok.indent < 0:
# new-styled `concept of A, B` on the same line as `concept`
result.add(p.emptyNode)
elif p.tok.indent < 0:
var args = newNodeP(nkArgList, p)
result.add(args)
args.add(p.parseTypeClassParam)
@@ -2252,9 +2280,10 @@ proc parseTypeClass(p: var Parser): PNode =
result.add(p.emptyNode)
if p.tok.tokType == tkComment:
skipComment(p, result)
# an initial IND{>} HAS to follow:
# an initial IND{>} HAS to follow, unless this concept inherits requirements:
if not realInd(p):
if result.isNewStyleConcept:
let hasParents = result[2].kind != nkEmpty
if result.isNewStyleConcept and not hasParents:
parMessage(p, "routine expected, but found '$1' (empty new-styled concepts are not allowed)", p.tok)
result.add(p.emptyNode)
else:

View File

@@ -148,11 +148,6 @@ proc processModule*(graph: ModuleGraph; module: PSym; idgen: IdGenerator;
closeParser(p)
if s.kind != llsStdIn: break
closePasses(graph, a)
if graph.config.backend notin {backendC, backendCpp, backendObjc}:
# We only write rod files here if no C-like backend is active.
# The C-like backends have been patched to support the IC mechanism.
# They are responsible for closing the rod files. See `cbackend.nim`.
closeRodFile(graph, module)
result = true
proc compileModule*(graph: ModuleGraph; fileIdx: FileIndex; flags: TSymFlags, fromModule: PSym = nil): PSym =
@@ -168,24 +163,12 @@ proc compileModule*(graph: ModuleGraph; fileIdx: FileIndex; flags: TSymFlags, fr
elif graph.config.projectIsCmd: s = llStreamOpen(graph.config.cmdInput)
discard processModule(graph, result, idGeneratorFromModule(result), s)
if result == nil:
var cachedModules: seq[FileIndex] = @[]
result = moduleFromRodFile(graph, fileIdx, cachedModules)
let filename = AbsoluteFile toFullPath(graph.config, fileIdx)
if result == nil:
result = newModule(graph, fileIdx)
result.flags.incl flags
registerModule(graph, result)
processModuleAux("import")
else:
if sfSystemModule in flags:
graph.systemModule = result
partialInitModule(result, graph, fileIdx, filename)
for m in cachedModules:
registerModuleById(graph, m)
replayStateChanges(graph.packed.pm[m.int].module, graph)
replayGenericCacheInformation(graph, m.int)
result = newModule(graph, fileIdx)
result.incl flags
registerModule(graph, result)
processModuleAux("import")
elif graph.isDirty(result):
result.flags.excl sfDirty
result.excl sfDirty
# reset module fields:
initStrTables(graph, result)
result.ast = nil

View File

@@ -77,7 +77,7 @@ proc inSymChoice(sc, x: PNode): bool =
result = false
for i in 0..<sc.len:
if sc[i].sym == x.sym: return true
elif sc.kind == nkOpenSymChoice:
elif sc.kind in {nkOpenSymChoice, nkOpenSym}:
# same name suffices for open sym choices!
result = sc[0].sym.name.id == x.sym.name.id
else:

Some files were not shown because too many files have changed in this diff Show More