| Commit message (Collapse) | Author | Age |
|---|
| | |
|
| |
|
|
| |
Now it shows something like `5 lines yanked into "~my.extreg`
|
| | |
|
| | |
|
| |
|
|
| |
Alias semantis are identical to using the full extreg name.
|
| | |
|
| | |
|
| |
|
|
| |
Added a big test suite for extregs
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Implements "extended registers"
Extended registers provide access to a separate namespace of registers
backed by custom lua callback functions.
A new special register ("~) is introduced, which prompts the user for
the name of an extregistr to yank or paste from.
This enables file-backed or cloud-synced or other arbitrary register
behavior.
One can now use:
`"~file_extreg<cr>yy`
to yank into the "file_extreg" register.
AI-assisted: Qwen3-Coder
|
| |
|
|
|
|
|
|
| |
Problem:
By default, 'termpastefilter' removes TAB chars when pasting to a :terminal.
Very annoying, and does not improve "security".
Solution:
Drop "HT" from the 'termpastefilter' default.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem:
A number in 'iskeyword', 'isfname' or 'isident' that overflows (e.g. via
a modeline) hits abort() or an assert in getdigits_int().
Solution:
Parse the numbers non-strictly so an out-of-range value gives E474.
Fixes #42409
AI-assisted
|
| |
|
|
|
|
|
|
|
|
| |
Problem:
TM_DIRECTORY expands to the directory's basename, losing its parent path.
Solution:
Remove the tail modifier when expanding the current file's directory.
Extend the existing known-variable test to check the full directory path.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem: watchdirs() keeps its state in large nested callbacks in
_watch.lua, which makes its directory tracking logic harder to change.
Solution: Move the backend into its own module and store its state in a
WatchDirs instance. Split setup, event handling and cleanup into
methods. Extend the existing test to cover cancellation while a
directory stat is pending.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
After edits move diagnostic ranges, counted jumps resume searching from
their original positions. This can revisit the same diagnostic and hang
the default first/last diagnostic mappings.
Solution:
Advance the search origin using the diagnostic's current logical position.
Test forward and backward counted jumps after inserting or deleting a
line, including jumps to the last or first diagnostic without wrapping.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Accepting an item from another completion source raises an error when its
user_data.nvim table does not contain LSP metadata.
Solution:
Retrieve and guard the nested LSP metadata before accessing its fields.
Cover acceptance of a foreign completion item without LSP metadata.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem:
An empty snippet insertText falls back to the label during selection, but
acceptance clears that label and expands an empty string.
Solution:
Require nonempty insertText before expanding a snippet without textEdit.
Extend the omnifunc acceptance fixture with an empty snippet insertion.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem:
A completion list default edit range can replace the explicit range of an
InsertReplaceEdit, filtering out valid items or changing their replacement.
Solution:
Only synthesize a default text edit when the completion item has none.
Extend the existing itemDefaults test for explicit edits with both range variants.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Snippet completion words and previews prefer insertText even when textEdit
is present, although the actual expansion uses textEdit.newText.
Solution:
Use the text edit before the fallback insertion text in both selectors.
Extend the existing precedence test to check the snippet word and preview.
AI-assisted
|
| |
|
|
|
|
|
|
| |
Problem: Custom ui2 window tracks 'fillchars' option to adjust to the
change of `msgsep` flag. However, 'fillchars' is `:h global-local`
window option. If it is set locally to the window, ui2 reacts to its
`msgsep` value and keeps using it afterwards even in other windows.
Solution: Use and react to only global value of 'fillchars'.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: Highlight inspection is limited to a single position, and its
result types do not document the fields shared by different highlight
sources.
Solution: Accept end_row/end_col in vim.inspect_pos to inspect a range.
Keep range ends exclusive and retain the existing single-position
behavior.
Support Ex line ranges in :Inspect, including Visual selections.
Document named result types, expose shared highlight fields, and keep
the existing extmark details and buffer field for compatibility.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
When starting Nvim with both `--clean` and `--noplugin` (in either
order), `--noplugin` is silently ignored and built-in runtime plugins
are loaded anyway. `command_line_scan()` sets `p_lpl = false` for
`--noplugin`, and `--clean` sets `params.use_vimrc = "NONE"` and
`params.clean = true`, which later overwrites `p_lpl` back to `true` via
`p_lpl = vimrc_none ? params.clean : p_lpl`.
Solution:
Only reset `p_lpl = false` when `vimrc_none && !params.clean`,
preserving `p_lpl = false` if `--noplugin` was passed.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Provider checks start and wait for a job, then execute the same command
again to obtain its output. The returned output can disagree with the
first process's exit status, and the second execution has no timeout.
Solution:
Return the first job's captured output. Bind the output callbacks to
their state and preserve line breaks between callback data entries.
Cover invocation count, multiline stdout, and merged stderr.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: nvim_set_option_value() dry runs skip checks in did_set_*
callbacks. This logic also prepares values needed to apply the option,
such as tab widths or a compiled regexp. Moving the checks into
validation callbacks needs a way to save that work for did_set_*.
Solution: Store prepared values in optset_T so did_set_* can consume
them after the option is stored. Dry runs perform the same validation,
then free the prepared values without changing live state.
This extends dry-run validation for:
- 'cedit': reject invalid command-line window keys.
- 'completeitemalign': require each completion column exactly once.
- 'cursorlineopt': reject empty values and conflicting line modes.
- 'encoding': accept only UTF-8 and its aliases.
- 'fileencoding': reject lists and local changes with 'nomodifiable'.
- 'messagesopt': require mandatory fields and enforce numeric limits.
- 'spellcapcheck': return regexp errors, including under :silent!.
- 'spellsuggest': reject conflicting suggestion modes.
- 'varsofttabstop' and 'vartabstop': reject invalid tab widths.
- 'wildmode': reject more than four completion stages.
'makeencoding' also uses the shared encoding preparation.
Separating value preparation exposed a bug in 'vartabstop' and
'varsofttabstop': failed assignments could discard local tab widths, and
global assignments could overwrite them. Fix this by preparing new
widths separately and applying them only after a valid local assignment.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
| |
vim.diagnostic.Opts.VirtualText has a `source` option to prefix
diagnostic messages with their source (`'if_many'` or always), but the
newer virtual_lines handler never got the same option, even though the
shared helpers it needs (diagnostic_shared.prefix_source/count_sources)
are already render-target-agnostic and used by virtual_text as-is.
Add the same `source` field to vim.diagnostic.Opts.VirtualLines and
apply it in the virtual_lines handler right after message formatting,
mirroring virtual_text's own ordering (format first, then prefix the
source onto the already-formatted message).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
vim-patch:9.2.1191: crash with "//e" when only a substitute pattern exists
Problem: Crash when using "//e" or "//" while there is a substitute
pattern but no previous search pattern (after v9.2.0374).
Solution: Use the previous pattern selected for an empty pattern
(Jonas Ornfelt).
related: vim/vim#19998
closes: vim/vim#21533
AI-assisted
https://github.com/vim/vim/commit/c58607c9a0579de0da909fc46f68144a01378af1
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
#42352
Problem:
- nvim_buf_call() aborts (assert) when called from an expression evaluated
for a window in another tabpage ('foldexpr', 'statusline'), because
`curwin` is not in `curtab`.
- After it returns, `curwin` is wrong if it switched windows.
- `vim._with{win=…, keepcwd=true}` does not undo `:lcd` if `win` is in
another tabpage.
Solution:
- Allow `curwin` outside `curtab`, and restore if it was outside
`curtab` at `ctx_switch()`.
- Find the target window in any tabpage when restoring its dirs.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
`nvim_set_option_value('fo', {'c','r','o'}, {operation='remove'})` does
nothing on "jcqrol": the items are joined into one `:set-=` value, which
only matches a contiguous substring. Removing map keys, e.g. {'eol','space'}
from 'listchars', never matches. `vim.opt:remove{…}` inherits this.
(Regression from f34ee3da80e8; 429338e74bf4 fixed the case where the
flags are adjacent and in order.)
Solution:
Remove each item separately.
|
| |
|
|
|
|
|
|
|
| |
Problem:
If a job exits with cursor at last row, the `[Process exited]` message
covers the text of that row.
Solution:
Start a new line before showing the message, like the old
`\r\n[Process exited]` did.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Problem:
After clearing cursors, "]C" "[C" do nothing, although the previous
cursor positions ("gQ") may still be useful as marks.
Solution:
If multicursor is not active, "]C" "[C" iterate the
"nvim.multicursor.last" extmarks. They also work as an operator motion
("d]C"), since there is no cascade.
AI-assisted
|
| |
|
|
|
|
|
|
|
| |
Problem: :detach! changes the channel's detach property without emitting
ChanInfo, so plugins watching channel updates cannot react to changes.
Solution: Schedule ChanInfo after setting detach=true. Add a test
covering emit on :detach!.
AI-assisted
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
`g@<C-v>j` and `y<C-v>j` set '] one char short of the block corner, so
'operatorfunc' cannot get the block. With 'virtualedit' "block", "y"
also yanks a short last line one column narrow.
Solution:
- Decide `inclusive` from the motion's own type, before get_op_vcol()
makes it blockwise (like Vim).
- Never exclude the end of a block from '] (`g@<C-v>l`): a block
includes its corner.
Co-authored-by: IcyJerry <263879414+Icyjerry@users.noreply.github.com>
Co-authored-by: Nagato Yuzuru <nagato@yuzuru.me>
|
| | |
|
| |
|
|
|
|
|
|
| |
Problem:
nvim_parse_cmd("'btab", {}) and typing ":'btab" crash, parse_cmdline()
continues with a NULL eap->cmd after an invalid :tab address.
Solution:
Fail if parse_command_modifiers() set an error.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
With 'ruler' in the cmdline ('laststatus=0') and ext_messages (ui2), a
'rulerformat' "%m" does not update the "[+]" indicator after ":write",
until a cursor-move.
Analysis:
`redraw_buf_status_later()` only marks windows with a statusline or
winbar. The symptom is hidden by the legacy message grid because the
write message overwrites the cmdline, which redraws the ruler.
Solution:
Also mark the cmdline 'ruler', like `status_redraw_buf()`.
|
| |
|
|
|
|
|
|
|
|
| |
Problem:
A Visual-mode mapping that reads input (`getchar()`) does not include
that input when cascading to other cursors (they only see ESC). OTOH,
Normal-mode mappings do replay the input.
Solution:
Append the payload to the Visual session's Lua-mapping subatom, and to the
session-ending command (`:<C-u>…` mapping).
|
| |
|
|
|
|
|
|
|
|
|
| |
Problem:
- `v:cascading` is false during the Insert-mode live-preview.
- `vim.on_key()` sees the internal keys of the Visual-selection preview
and "zq" steps.
Solution:
- Set `v:cascading` during the Insert-mode preview.
- Don't send internal keys to `on_key` callbacks.
- Use `v:cascading` as the source of truth for `mc_replaying()`.
|
| |
|
|
|
|
|
|
| |
Problem:
After `/\%Vfoo<CR>`, `zqn` places no cursors. `mc_zq_target()` clears
`'< '>` before `gn` (to detect a failed match), so `\%V` never matches.
Solution:
Keep the user's `'< '>` during `gn`; detect failure by its beep.
|
| |
|
|
|
| |
Problem: No way to check if cascade is in action.
Solution: Add `v:cascading` variable.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
With multicursors, an Insert-mode edit by a Lua/`<Cmd>` mapping or API
(nvim_put(), …) is live-previewed at the other cursors, but dropped at
ESC: the cascade replays _redo_, which does not contain such edits.
Solution:
When `<Cmd>`/Lua/RPC changes the buffer during Insert-mode, sync the
previews and restart the capture (like a non-captured cursor-move). The
edit stays as text.
- The mapping runs only at the primary. Its text is mirrored as-is
(similar to insert-completion). Re-running it per cursor would require
a "partial-remap" span replay, and breaks single-session plugins
(vim.snippet).
- Not during completion; redo already captures the pre-inserted
candidate on accept.
- Text typed before the edit is kept as mirrored, not replayed, so
per-cursor effects ('textwidth' wrapping, …) don't apply to it.
Known issues:
- "o"/"O"/"cc" with 'autoindent' followed by the mapping: the per-cursor
insert-entry replay is still pending, so the edit is lost.
|
| |
|
|
|
|
|
|
|
|
|
| |
Problem:
Handlers defer `show()` on unloaded buffers with a buffer-local BufRead
autocmd whose id is stored per-namespace:
- `noautocmd bwipeout` deletes the autocmd, so the next show/hide fails.
- Showing another buffer in the same namespace cancels the pending show.
Solution:
Handlers skip unloaded buffers. The existing global BufRead autocmd
shows cached diagnostics when the buffer loads.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
vim-patch:9.2.1175: setreg() block width wrong for unprintable chars
When yanking, the width of a <c-v>{motion}y register
is determined by the virtual colums of the text. However
in setreg() in 'b' mode use an idosyncratic width measurement
where C0 control chars (ASCII 0x00 - 0x19) and invalid UTF-8 bytes
have width 1 instead of their display width (2 or 4 cells).
Consider the following test case:
let strings = ['"reference', '"Übel', "\"aa\x07bb", "\"qq\x82", "\"x\nbabbel"]
for string in strings
call append('$', string)
exec "normal G0\<c-v>$\"ay"
let one = getregtype('a')
if string != strings[-1]
call setreg("b", string, 'b')
let two = getregtype('b')
else
let two = "na"
endif
call setreg("c", [string], 'b')
let three = getregtype('c')
call append('$', '"yank: '.one.", string: ".two.", list: ".three)
endfor
AFter this fix, all the three ways of measuring the width gives
equal values (or two, in the case of not being able to use setreg()
with a plain string in the first case)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
`nvim --embed --listen addr` always treats stdin as an RPC channel. If
stdin is /dev/null (e.g. backgrounded over ssh), the channel closes
before a UI can attach, so Nvim exits (detach=false behavior). If stdin
is a tty, keystrokes are read as RPC.
Solution:
With `--embed --listen`, open the stdio RPC channel only if stdin is
a pipe or socket. Else wait for a UI at the `--listen` address.
$ nvim --embed --listen ./nvim.sock
Nvim server (pid: 71984) waiting for UI, listening on: "./nvim.sock"
Note:
- No change when stdin is a pipe. So the test harness, TUI, and :restart
all behave exactly as before.
- The #29634 use-case of configuring a UI that attaches *after* startup
(a second UI, or reattach) still requires #25377 (reliable UIEnter).
Not affected by this change.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Only the stdio channel can "own" a server (self-exit on close). After
:restart the TUI reconnects over a socket, so the restarted server does
not have "self-exit" behavior when the channel closes (e.g. by UI
disconnect, host-terminal close, etc.), which leaves a dangling server.
Solution:
- Introduce `nvim_chan_set`. (`nvim_set_client_info` doesn't fit:
exposing it to Lua would require a `channel` param, which is not
backwards-compatible, plus it's awkward because it doesn't merge info.)
- Broadcast per-channel properties in the "restart" UI event as an
opaque dict (a potentially useful pattern for other UI-events in the
future).
- UIs are expected to restore the channel properties on restart.
- `nvim_get_chan_info()` reports "detach".
- Remove `nvim__chan_set_detach()`.
|
| |\
| |
| | |
fix(lua): regression with `print()` in headless mode when redirecting
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
This is a bunch of hacks so we can override msg_puts_printf specifically
while still allowing redirecting and "normal" message flow to happen,
as this still has observable side-effects (even with no screen attached,
screenstring() is still valid..)
fixes #42218
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Problem:
nvim_open_term() sizes the terminal from the current window, which may
have 0 rows (bordered float, 'cmdheight' taking all rows). libvterm
aborts when resizing from 0 rows.
Solution:
In `terminal_alloc()`, treat zero width/height as the pty default
(80x24), like jobstart(term=true).
Open question:
nvim_win_get_height() on such a float returns -1 (winheight() returns 0).
|
| |\ \
| | |
| | | |
perf(watch): skip directories outside include glob prefixes
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
|
