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.
(cherry picked from commit f5930d0bb3)
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
(cherry picked from commit 4497d89267)
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.
(cherry picked from commit 4b374eb0a6)
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).
(cherry picked from commit c8e805a2fa)
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.
(cherry picked from commit 73986c03a1)
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.
(cherry picked from commit 88a18de44f)
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>
(cherry picked from commit 645e131739)
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.
(cherry picked from commit 568eccd7f8)
## 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>
(cherry picked from commit 7d2f28b046)
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.
(cherry picked from commit 3e2cea21ed)
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.
(cherry picked from commit cfa769fefc)
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`.
(cherry picked from commit f9647276d8)
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]`
(cherry picked from commit 4c8052a45b)
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.
(cherry picked from commit f2e4ae0016)
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.
(cherry picked from commit b73908a361)
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.
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.
- 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.
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`
Tagged `[backport]` in the commit message for inclusion in
`version-2-2`.
---------
Co-authored-by: puffball1567 <17452514+puffball1567@users.noreply.github.com>
(cherry picked from commit cbe02aa9de)
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.
(cherry picked from commit 98131a9fa1)
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.
(cherry picked from commit 2b2872928b)
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>
(cherry picked from commit 3eb4a60b6b)
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.
(cherry picked from commit 6353c4e5b0)
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.
(cherry picked from commit 9a2b0dd045)
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.
(cherry picked from commit 0dc577a4dc)
```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)
b(addr x.kind)
```
(cherry picked from commit 184d423779)
Fixes#25704
This makes sure that `iter` still has `tyGenericInst` skipped like
before, without skipping it for `iterType` which requires it
(cherry picked from commit f9524861f3)
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.
(cherry picked from commit 0028ea563c)
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:
* 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)
* 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.
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>
(cherry picked from commit 854c1f15ba)
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.
(cherry picked from commit be29bcd402)
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.
(cherry picked from commit 9c07bb94c1)
fixes#25632fixes#25631fixes#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.
(cherry picked from commit fb31e86537)
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.
(cherry picked from commit 7a82c5920c)
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.
(cherry picked from commit 5c86c1eda9)
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`.
(cherry picked from commit 7f6b76b34c)
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.
(cherry picked from commit 4bf44ca47f)
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.
(cherry picked from commit 1a1586a5fb)
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.
(cherry picked from commit 2db13e05ac)
`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
(cherry picked from commit edbb32e4c4)
