Problem: wrong end_row when copying extmarks to the pager buffer when expanding
a message. Results in incorrect highlights or errors such as "invalid `end_col`
out of range".
Solution: don't forget to offset end_row by srow, as is already done for the
copied lines and extmark rows.
Problem:
Empty string is falsy, so the spec's label fallback applies, but there only checks
for nil. An empty filterText drops the item, an empty sortText sorts it
first, and an empty insertText leaves the word empty.
Solution:
Treat empty string as unset.
Instead of setting it in the test runner (which runs in an Nvim instance
that already has a log file), make run_tests.zig set a $NVIM_LOG_FILE
fallback path like what RunTests.cmake does.
Problem:
- `lhs` is not fully realized. E.g. for a "payload" mapping
`lhs` omits the `getchar()` payload during a mapping (vim-surround
`ds'` reports `lhs="ds"`). This means plugins like vim-repeat are
still needed...
- The "delta" fields of a CmdAtom are calculated too late.
- `<abuf>` and `changed` check whatever (wrong) buffer a command
(":bnext") might land in.
- CTRL-W_w between two windows on the same buffer reports type="motion".
Solution:
- `CmdOrigin` samples (buf/win/cursor/changedtick) at each "scope" entry
(CmdFrame, composite, Visual session, insert session).
- `dd<C-w>l` reports `changed=true` for the buffer it edited, regardless
of where the cursor ends up.
- New fields:
- `pos`: cursor position at command start.
- `moved`: indicates whether the cursor moved (in same buffer).
- `undoseq`: undo state at settlement.
- lhs now includes the payload: "ds)" reports lhs="ds)" instead of "ds".
- Easy for users to "replay" any atom.
- Rename: type "command" => "normal", "ex" => "excmd"; `arg` => `cmdarg`
- Drop `cascade` field (no reason to expose it)
Problem: Heap-buffer-overflow in spell_suggest() when the cursor is
beyond the end of the line, because a SpellFileMissing
autocommand changed the buffer (dvaave2025).
Solution: parse_spelllang() may run autocommands, so validate the cursor
position and re-take the saved position afterwards.
fixes: vim/vim#21097closes: vim/vim#21100
Supported by AI.
6073903cda
Co-authored-by: Christian Brabandt <cb@256bit.org>
Problem: Multiline messages exceeding 'cmdheight' not visible when a
mapping starts cmdline immediately after it (after 9.2.0967).
Solution: Revert patch 9.2.0967 and use a different solution (zeertzjq).
fixes: vim/vim#21098closes: vim/vim#21101fb4866a2dd
Problem:
An unreplayable Visual operation does not emit a CmdAtom event. That's
maybe not super important, but it hints at a flaw in how `vatom`
"voiding" is plumbed: `vatom.state=kVatomVoid` replaces the "kind", so
that info is lost to later parts in the lifecycle.
Solution:
- Define `VatomState` as "flags", so `kVatomVoid` can "poison"
`vatom.state` without losing its kind flag.
- Unify `lhs`: always report the original user input in `CmdAtom.lhs`,
for all kinds of user actions: visual, "translated"/"stuffed" cmds,
and dot-repeat (".") itself.
- Unreplayable Visual atom emits CmdAtom with non-empty `lhs` and empty
`keys`, like a mapping/macro composite.
Problem: tests: test_substitute leaves swapfiles behind
Solution: Close open buffer using :bw!
be66c98572
Co-authored-by: Christian Brabandt <cb@256bit.org>
Problem:
`signal_ignore_deadly` doesn't work for Windows, where the console still
may terminate Nvim during teardown (after `signal_teardown`), while it
is already trying to exit. Besides interrupting any housekeeping we are
doing, it results in an unpredictable exit code (flaky tests).
[Process exited -1073741510] // 0xC000013A STATUS_CONTROL_C_EXIT
Solution:
Register our own CTRL_CLOSE_EVENT handler which "blocks" the signal.
Note: if exit takes longer than 5s, Windows will consider the process
"hung" and kill it anyway.
Problem:
A deadly signal arriving during teardown can kill Nvim while it is
preserving swapfiles. `os_exit()` ignores deadly signals via
`signal_reject_deadly()`, but `signal_teardown()` => `uv_signal_stop()`
resets them to the default behavior, so SIGHUP arriving after that
kills the process:
[Process exited 129] // 128 + SIGHUP
This is a race when closing a pty: kernel sends SIGHUP to foreground
process group *and* the reads return EOF, so `chanclose()` on a TUI job
prepares to exit twice. This means it is unpredictable whether Nvim
exits 1 or is terminated.
Solution:
Ignore deadly signals once the watchers are closed. Only SIGKILL
interrupts it now.
Problem: Crash when closing the new current window during CTRL-W_x.
Solution: Remove a duplicate redraw of a window that should normally be
already mark for redraw in the previous call.
Problem: b:match_words groups "{" with the if/for/while/switch keywords
and "}" with "break" which breaks % matching on braces
Solution: Drop the brace and bracket groups, matchit appends
'matchpairs' by itself (Matthias Bruns).
matchit counts every alternative in a group instead of pairing the
alternatives with each other. Listing `{` alongside the if, for,
while, switch, struct and class keywords therefore makes a line such
as `for (...) {` count as two openers, and listing `break` alongside
`}` lets a brace pair with a break statement. As a result % on the
opening brace of a function does not move at all, and % on
`switch (x) {` jumps to `break;` instead of the closing brace.
Braces and brackets do not need to be listed: matchit appends
'matchpairs' to b:match_words by itself. Drop them and leave the
preprocessor group unchanged.
closes: vim/vim#2106408c74ce09a
Co-authored-by: Matthias Bruns <matthiasbruns35@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Problem:
Unreliable test on slow CI (ASAN/TSAN):
FAILED .../tui_spec.lua @ 3054: TUI exits immediately when stdin is closed
retry() attempts: 1
Expected: vim.NIL
Actual: { name = "nvim", pid = 33201, ppid = -1 }
The test asserts "immediate" exit of the Nvim process, but this may be
subject to OS delays outside of our control.
Solution:
Make the "timed out waiting for DA1" log conditional on the actual
timeout, and assert the logs in the test.
Problem:
- Dot-repeat of a Visual selection prepared by a `<Cmd>` mapping,
results in E1255 and leaves Visual mode active.
- Visual-mode capture is implemented as a second, parallel "capture
engine": it re-composes a CmdSpec from cmdarg_T per key and decides
replayability from its own table of "void" key classes.
Solution:
- Model nested `normal_execute()` as a `CmdFrame` stack,
instead of a single module-scoped `stage`.
- Capture `<Cmd>` ":norm …" commands as subatoms from its nested frames.
- Produce every command exactly once; atom_push_raw() routes the atom to
the Visual composite while a selection is open, like it already does
for mapping composites.
- `v/pat<CR>d` is now repeatable and cascades.
Problem:
Flaky test:
RUN T1158 nvim.zip reports an incorrect archive password: 11092.84 ms FAIL
The prompt is detected only if the pty output *ends* with "password: ",
but a read may return the prompt plus following bytes.
Solution:
- Match anywhere in the output since the last password was sent. The
buffer is cleared before each send, so won't match stale text.
- Assert on the reported message, so a failure shows what was reported.
Problem:
With the addition of the `:help al` text object, you can now easily
format the whole buffer with `gqal`. However, `vim.lsp.formatexpr` only
uses `textDocument/rangeFormatting` which some language servers (like
gopls) don't support.
Solution:
- Fall back to `textDocument/formatting` if the whole buffer is being formatted
and the server doesn't support `textDocument/rangeFormatting`.
- In theory, these two methods should return the same response if the whole
buffer is being formatted, but I preserved the existing behaviour of
prioritising `textDocument/rangeFormatting` in case that does not hold (i.e.
LS bug).
- Also: `vim.lsp.formatexpr` had no tests at all, so actually add tests for it.
Problem: The hit-enter prompt fires whenever a message scrolls the screen.
When this happens while a mapping is being processed, it consumes
the mapping's next key, causing unexpected behavior for users.
Solution: Similar to what 9.1.1969 did for stuffed characters, skip the
hit-enter prompt when there are still keys pending from a mapping
in the typeahead buffer.
related: neovim/neovim#38298
related: neovim/neovim#20635
related: neovim/neovim#30890
closes: vim/vim#20753
AI assisted.
6025ea9e02
Co-authored-by: XiaowenHu96 <me@xiaowenhu.com>
Problem: A SAL rule longer than MAXWLEN is silently truncated to an
empty lead. set_sal_first() then reorders the sl_sal entries
by their index byte and can move the terminating sentinel out
of the last slot, so spell_soundfold_wsal() reads past the end
of the array, e.g. when soundfold() or spellsuggest() is used
(Erick Alex).
Solution: Bound the sound-folding loops against sl_sal.ga_len.
closes: vim/vim#210766ac008db96
Co-authored-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Problem: string_reduce() copies *rettv into argv[0] before calling
eval_expr_typval(). When the evaluator fails early, rettv is
never reset and still aliases argv[0] v_string.
clear_tv(&argv[0]) frees it, leaving rettv dangling and when
in vim9script get_func_tv() frees it again (Ave Dva).
Solution: Set rettv->v_type = VAR_UNKNOWN like what is done in
list_reduce() and tuple_reduce(), use tv_get_string_strict()
in f_reduce()
closes: vim/vim#21048
Supported by AI.
cd59994c45
Co-authored-by: Christian Brabandt <cb@256bit.org>
Problem:
When `z=` delegates to `vim.ui.select()`, the picker may change the
current window before returning. `spell_suggest()` then continues to the
cursor restoration branch with the new window and assigns `prev_cursor`,
which belongs to the original window. This can leave Normal mode with an
invalid cursor position and produce E315.
Solution:
Clean up the spell suggestion state and return immediately after handing
control to `vim.ui.select()`.
Problem:
`<cmd>` mappings do not emit `CmdAtom.text`.
`<cmd>` and Lua-callback mappings that edit the buffer apply only at the
primary cursor, not cascaded (multicursor).
Solution:
Capture the `<cmd>` command in getcmdkeycmd().
Add kKeyOpaque ("no capturable keys"); narrow kKeySynthetic ("not
a keystroke") to K_EVENT/K_IGNORE, so an opaque mapping's edit still
sets `map_edit` and cascades via LHS-replay.
Problem:
`vim.fs.slug()` does not handle URIs like `term://foo//123:bash`,
so callers (e.g. terminal persistence) must strip the scheme before
calling `slug()`.
Solution:
Detect `scheme://` from the raw input before `normalize()` and
replace it with a `=uri-<scheme>-` prefix.
Problem:
`:rundo` on a corrupted undo file crashes or hangs, instead of failing
with E825. Patching one 4-byte field is enough:
ue_size = 0xFFFFFFFF " walks a NULL ue_array
ue_size = 0x7FFFFFF0 " 17 GB xmalloc + memset, then preserve_exit()
ue_top = 0xFFFFFFFB " negative lnum reaches ml_delete()
Analysis:
Every count in the file is read with `undo_read_4c()` and then checked,
differently at each site. None bounds the value by what the file can
hold, so a 2 GB count reaches `xmalloc()`.
Note:
- Vim doesn't have `bi_fsize` because it checks `U_ALLOC_LINE` result
everywhere (thus doesn't crash, but may thrash...); those checks were
dropped when Nvim moved to `xmalloc()`, and the `ue_size` loop counter
became unsigned.
- Vim *does* have the negative line numbers bug: `u_undoredo()` checks
`top > ml_line_count || top >= bot || bot > ml_line_count + 1`, which
rejects none of them.
Solution:
- Introduce `undo_read_len()` and use it to fail early instead of
continuing with nonsense.
- Validate `ue_top`/`ue_bot`/ `ue_lcount`.
- Use `xcalloc()`, so no site can proceed with a NULL array.
- Report a truncated "U" line, distinguish EOF from a 0xFFFFFFFF field,
and free the header on the extmark error path.
Problem:
A named mark updated after a change is moved back (treated as the
original mark) by undo:
:1mark d
:$
dw
:2mark d " 'd is on line 2
:undo " 'd is back on line 1
The undo header snapshots `b_namedm` when the change is recorded, and
`u_undoredo()` restores that snapshot indiscriminately.
Solution:
Update the pending header's snapshot when a mark is set explicitly.
Marks that the change itself moved go through mark_adjust(), not
setmark_pos(), so those are still reverted.
Similar to 2546741d1b (for extmarks): an explicit set inside an undo
block is confused with an edit-driven adjustment. But the extmarks case
is dealing with mid-edit moves, whereas named/regular marks only need
the stale snapshot dropped.
Problem:
Setting a :bcd buffer into another window, modifies the caller's CWD.
local b = vim.api.nvim_create_buf(true, true)
vim.api.nvim_buf_call(b, function() vim.cmd.bcd('..') end)
vim.cmd('vsplit')
vim.api.nvim_win_set_buf(vim.fn.win_getid(2), b)
:echo haslocaldir(0) haslocaldir(-1,0) haslocaldir(-1,-1,0)
0 0 0
:echo getcwd() ==# getcwd(-1,-1)
0
Analysis:
`ctx_dirs_save` only saves CWD if it predicts the switch can change it.
But win_set_buf() replaces the target window's buffer *after* the
switch, which cannot be "predicted" from `ctx_dirs_save`.
Solution:
Always snapshot whenever the switch enters another window.
Skipping `os_dirname` was a micro-optimization.
Problem:
nvim_win_set_buf() on a non-current win, while the current win has
a win-local dir, changes the global CWD:
:vsplit | lcd ..
:call nvim_win_set_buf(other_win, buf)
:wincmd l
:verbose pwd
[global] /parent " expected: the initial cwd
Analysis:
`globaldir` is where to return when no local dir applies; NULL means the
process CWD is already there. Switching to a window with no local dir
makes update_cwd() chdir back to `globaldir` and clear it. kCtxKeepCwd
restores the process CWD but not that bookkeeping, so the restored
window-local dir is mistaken for the global one.
Solution:
Save/restore `globaldir` with the CWD.
Problem:
A mark created by `nvim_buf_set_extmark()` while an undo block is open
never comes back on redo.
Analysis:
Undo deletes the text it covers and collapses the range; redo replays
the splices, which re-insert the text but cannot re-expand the mark.
`extmark_set()` records a position only for a mark it moves, not for one
it creates.
Solution:
Record the created position for redo; undo leaves the mark to the splice
replay, since it did not exist before the edit. Each side of a paired
mark gets its own entry. Redo also revives a mark that undo invalidated,
else its position returns but its highlight does not.
Problem:
Directory listing entries cannot be customized (filtered, reordered).
Listings are read by a BufReadCmd, which suppresses BufReadPost, so they
are the only buffers with no post-read event to hook.
Solution:
Introduce a post-render User autocmd `DirReadPost`, marking the dir
buffer writable for the duration and before the cursor is placed, so
handlers can sort or filter it with ordinary commands. Document common
recipes
Problem:
There is no unified notion of a "user action".
Vim processes input by one-char-at-a-time, and mostly throws away any
hints it might gather about the user's action, with one exception: it
stores the last _edit_ action (the "redo buffer", encoded as
unstructured `["x][v][count]body` bytes).
Plugins can only observe individual keys (vim.on_key) and high-level
effects (TextChanged, CursorMoved).
Solution:
- Users can subscribe to `CmdAtom` events to handle any user action.
- Event is deferred; handlers cannot cancel or interfere with user
actions.
- Capture `CmdSpec` from the normal/insert/visual subsystems.
- typeahead/readahead stay unstructured (`buffheader_T`): they are key
streams, not commands.
- the redo/record buffers become `StringBuilder`: fewer
allocations/copies.
- Repurpose the input/redo engine to accept `CmdSpec` objects.
"atom": one repeatable unit of user input, as a resolved (post-mapping)
keysequence plus structured fields. Only user actions, not `:normal`,
API calls, or non-"t" `feedkeys`.
BREAKING: dot-repeat of an Insert session, replays the entire session
including cursor-moves (:help ins-repeat).
BREAKING: dot-repeat of a Visual operation, replays the selection
instead of operating on a fixed-size region.
Problem:
`inputlist()` advertises "click with the mouse" purely because it
implements click selection, so under the default `'mouse'` of "nvi" it
offers a click that command-line mode never receives.
Solution:
Only offer the mouse when `'mouse'` covers command-line mode, the same
condition `:help inputlist()` already documents.
Problem:
vim.diagnostic.set() defers extmark position computation for an
unloaded buffer via a once=true BufRead autocmd, registering a new one
on every call without replacing the previous one. Each pending autocmd
also retains that call's diagnostics.
Solution:
Instead of registering an autocmd per set() call, register a single
static BufRead autocmd that computes positions from the diagnostic
cache for any buffer with cached diagnostics when it is read. This
removes the per-call registration entirely (nothing left to
accumulate) and means diagnostics cleared while the buffer was
unloaded no longer produce stale extmarks.
Problem: transstr() has comments that do not add anything to what the
code says, and it casts a length to int only to cast it back to
size_t.
Solution: Drop the comments and keep the length in a size_t
(Hirohito Higashi).
related: vim/vim#20925
closes: vim/vim#21026fe65307d49
Co-authored-by: Hirohito Higashi <h.east.727@gmail.com>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Problem: sort() with "n", "N" or "f" converts an item to its number on
every comparison. For "n" that is a tv2string() plus strtod()
per comparison, so sorting a list of numbers turns each number
into a string and back O(n log n) times, dwarfing the sort.
Solution: Compute the numeric key of each item once, before the sort,
and compare the stored key (Samuel Schlesinger). Only the
builtin numeric compare modes are affected; uniq(), which
passes a bare list item to the compare function, and the
string and user-function paths are unchanged.
Sorting a list of 100000 numbers (min of 3, macOS arm64):
- sort(l, 'n'): 0.205s -> 0.017s
- sort(l, 'N'): 0.017s -> 0.010s
- sort(l, 'f'): 0.014s -> 0.010s
The result is identical, including that a string is still treated as 0
in "n" mode and that "N" keeps full 64-bit precision.
Add Test_sort_numeric_precomputed(): a large shuffled list sorted with
"n", mixed integers and floats, int64 values beyond the exact range of
a double for "N", and uniq() over the non-precomputed path.
closes: vim/vim#21003c8c59db9df
Co-authored-by: Samuel Schlesinger <sgschlesinger@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Problem: Reading an undo file resolves every stored sequence number
with a linear scan over all headers, making loading
quadratic in the number of undo states.
Solution: Sort uhp_table on uh_seq once and resolve each reference
with a binary search; the duplicate uh_seq check becomes a
single pass over the sorted table (Samuel Schlesinger).
At the default 'undolevels' of 1000 the quadratic cost is not
measurable; it takes 'undolevels' in the tens of thousands to matter.
Loading an undo file with 20000 states and 50 alternate branches with
:rundo goes from 1.49s to 0.11s (min of 3, macOS arm64), with the
same undotree().
Also make old_idx/new_idx/cur_idx and the loop index "i" long instead
of short/int: they index uhp_table, whose length num_head is a long
read from the file. A short index truncated above 32767 headers,
making the restored b_u_oldhead/b_u_newhead/b_u_curhead pointers
wrong in exactly the many-headers case this change is about.
Add tests: a round-trip test with alternate branches that compares
the entries of the tree and the text at every sequence number, a
corruption test with a duplicated uh_seq, and a test for reading an
undo file with zero headers, which is written when only the line for
the "U" command is saved.
closes: vim/vim#20942fccf613c8f
Co-authored-by: Samuel Schlesinger <sgschlesinger@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Problem: In diff mode with 'cursorbind' the cursor in the other window is
not updated after an undo that changes which lines correspond.
Solution: Also check whether the text changed before skipping the update
(Hirohito Higashi).
fixes: vim/vim#20982
related: vim/vim#13219
related: vim/vim#13210
closes: vim/vim#210042045a20d4b
Co-authored-by: Hirohito Higashi <h.east.727@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Problem:
`1-` does nothing from a directory buffer, because we are already in the
buffer-local CWD. It's also unintuitive that this mapping behaves
differently based on the resolved CWD.
Solution:
Have `1-` open the global CWD.
Now this was a cargo cult anti-pattern to write home about.
Doing painful save-and-restore bookkeeping around a separate
`magic_overruled` decoy global is just as messy as doing painful
save-and-restore logic around `p_magic` itself. only that now you need
to wrap every access to the effective value in a function call.
This replaces this with a marvellous new Clean Code technique™:
passing in the intended behavior as a function parameter to functions
where either the option or an explicit value might be used.
Problem:
A mark moved by nvim_buf_set_extmark() during an edit is misplaced by
undo and redo. Only splices ("edits") are recorded, and replaying them
reproduces the shifts they caused, never the explicit set: the mark ends
up wherever the text pushed it.
Solution:
When an open undo block moves an existing mark, record both positions.
Undo restores the pre-set position, redo re-applies the set.
Partially reverts 18334a4a0c ; ExtmarkSavePos.row/col were unused
because nothing recorded an explicit move, but now `extmark_set()` does.
Problem:
After #40270, events are no longer emitted from the automatic background
detection. This applies not just during startup, but also if the user
manually changes the background of their terminal.
Solution:
Set the background as normal, assuming that a normal terminal will
respond within 100 ms. Change test to match expected behavior:
- BG set during startup won't trigger user autocmds since it runs before
any user config
- If the terminal takes longer than 100 ms to respond to initial OSC 11,
it does trigger the OptionSet, but it is triggered through the normal
path to ensure values like v:option_new are set #38551
- BG change after startup still triggers autocmds #41146
