Problem: completion: not possible to limit number of matches
Solution: allow to limit the matches for 'complete' sources by using the
"{flag}^{limit}" notation (Girish Palya)
This change extends the 'complete' option to support limiting the
number of matches returned from individual completion sources.
**Rationale:** In large files, certain sources (such as the current
buffer) can generate an overwhelming number of matches, which may cause
more relevant results from other sources (e.g., LSP or tags) to be
pushed out of view. By specifying per-source match limits, the
completion menu remains balanced and diverse, improving visibility and
relevance of suggestions.
A caret (`^`) followed by a number can be appended to a source flag to
specify the maximum number of matches for that source. For example:
```
:set complete=.^9,w,u,t^5
```
In this configuration:
- The current buffer (`.`) will return up to 9 matches.
- The tag completion (`t`) will return up to 5 matches.
- Other sources (`w`, `u`) are not limited.
This feature is fully backward-compatible and does not affect behavior
when the `^count` suffix is not used.
The caret (`^`) was chosen as the delimiter because it is least likely
to appear in file names.
closes: vim/vim#170870ac1eb3555
Cherry-pick test_options.vim change from patch 9.1.1325.
Co-authored-by: Girish Palya <girishji@gmail.com>
Problem: Coverity warns about using uninitialized value
(Coverity, Tony Mechelynck, after v9.1.1301)
Solution: initialize callback pointer to NULL
d2079cff48
Co-authored-by: Christian Brabandt <cb@256bit.org>
Problem: completion: cannot configure completion functions with
'complete'
Solution: add support for setting completion functions using the f and o
flag for 'complete' (Girish Palya)
This change adds two new values to the `'complete'` (`'cpt'`) option:
- `f` – invokes the function specified by the `'completefunc'` option
- `f{func}` – invokes a specific function `{func}` (can be a string or `Funcref`)
These new flags extend keyword completion behavior (e.g., via `<C-N>` /
`<C-P>`) by allowing function-based sources to participate in standard keyword
completion.
**Key behaviors:**
- Multiple `f{func}` values can be specified, and all will be called in order.
- Functions should follow the interface defined in `:help complete-functions`.
- When using `f{func}`, escaping is required for spaces (with `\`) and commas
(with `\\`) in `Funcref` names.
- If a function sets `'refresh'` to `'always'`, it will be re-invoked on every
change to the input text. Otherwise, Vim will attempt to reuse and filter
existing matches as the input changes, which matches the default behavior of
other completion sources.
- Matches are inserted at the keyword boundary for consistency with other completion methods.
- If finding matches is time-consuming, `complete_check()` can be used to
maintain responsiveness.
- Completion matches are gathered in the sequence defined by the `'cpt'`
option, preserving source priority.
This feature increases flexibility of standard completion mechanism and may
reduce the need for external completion plugins for many users.
**Examples:**
Complete matches from [LSP](https://github.com/yegappan/lsp) client. Notice the use of `refresh: always` and `function()`.
```vim
set cpt+=ffunction("g:LspCompletor"\\,\ [5]). # maxitems = 5
def! g:LspCompletor(maxitems: number, findstart: number, base: string): any
if findstart == 1
return g:LspOmniFunc(findstart, base)
endif
return {words: g:LspOmniFunc(findstart, base)->slice(0, maxitems), refresh: 'always'}
enddef
autocmd VimEnter * g:LspOptionsSet({ autoComplete: false, omniComplete: true })
```
Complete matches from `:iabbrev`.
```vim
set cpt+=fAbbrevCompletor
def! g:AbbrevCompletor(findstart: number, base: string): any
if findstart > 0
var prefix = getline('.')->strpart(0, col('.') - 1)->matchstr('\S\+$')
if prefix->empty()
return -2
endif
return col('.') - prefix->len() - 1
endif
var lines = execute('ia', 'silent!')
if lines =~? gettext('No abbreviation found')
return v:none # Suppresses warning message
endif
var items = []
for line in lines->split("\n")
var m = line->matchlist('\v^i\s+\zs(\S+)\s+(.*)$')
if m->len() > 2 && m[1]->stridx(base) == 0
items->add({ word: m[1], info: m[2], dup: 1 })
endif
endfor
return items->empty() ? v:none :
items->sort((v1, v2) => v1.word < v2.word ? -1 : v1.word ==# v2.word ? 0 : 1)
enddef
```
**Auto-completion:**
Vim's standard completion frequently checks for user input while searching for
new matches. It is responsive irrespective of file size. This makes it
well-suited for smooth auto-completion. You can try with above examples:
```vim
set cot=menuone,popup,noselect inf
autocmd TextChangedI * InsComplete()
def InsComplete()
if getcharstr(1) == '' && getline('.')->strpart(0, col('.') - 1) =~ '\k$'
SkipTextChangedIEvent()
feedkeys("\<c-n>", "n")
endif
enddef
inoremap <silent> <c-e> <c-r>=<SID>SkipTextChangedIEvent()<cr><c-e>
def SkipTextChangedIEvent(): string
# Suppress next event caused by <c-e> (or <c-n> when no matches found)
set eventignore+=TextChangedI
timer_start(1, (_) => {
set eventignore-=TextChangedI
})
return ''
enddef
```
closes: vim/vim#17065cbe53191d0
Temporarily remove bufname completion with #if 0 to make merging easier.
Co-authored-by: Girish Palya <girishji@gmail.com>
Co-authored-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: glepnir <glephunter@gmail.com>
Problem: Unable to change the "relative" of a flag after its target
"win" no longer exists.
Solution: Unset target window if it is not present in config and
reconfigured "relative" no longer expects a target window.
Problem:
Developing/troubleshooting plugins has friction because "restarting"
Nvim requires quitting and manually starting again. #32484
Solution:
- Implement a `:restart` command which emits `restart` UI event.
- Handle the `restart` UI event in the builtin TUI client: stop the
`nvim --embed` server, start a new one, and attach to it.
Problem: When using conhost and pressing the 'x' button
to close it while nvim is open, nvim hangs up
while trying to reset the window icon, causing a big
delay before the terminal actually closes. #34171
Solution: Set the window handle to NULL after receiving SIGHUP
so that nvim will not try resetting the icon.
Problem: Non-visible/focusable windows are assigned a window number,
whereas commands that use this window number skip over them.
Solution: Skip over non-visible/focusable windows when computing
the window number, unless it is made the current window
through the API in which case an identifiable window number
is still useful. This also ensures it matches the window
number of the window entered by `<winnr>wincmd w` since
403fcacf.
Problem: nvim_create_user_command() Lua callbacks were missing the documented nargs field in the options table passed to the callback function.
Solution: Add nargs field derivation from command argument type flags in nlua_do_ucmd(), using the same logic as nvim_parse_cmd().
Problem: It is difficult to ignore all but some events.
Solution: Add support for a "-" prefix syntax in '(win)eventignore' that
subtracts an event from the ignored set if present
(Luuk van Baal).
8cc6d8b187
Problem: Blockwise Visual selection not redrawn correctly when moving
cursor for more than 1 cells with 'virtualedit'.
Solution: Restore the curswant update removed in 6679687bb3.
Problem:
Despite the name, `builtin.txt` only includes Vimscript functions, which
is confusing, especially to people who only use Lua to configure Nvim
Solution: From justinmk's suggestion
- Rename `builtin.txt` to `vimfn.txt`
- Rename `eval.txt` to `vimeval.txt`
- The tags `*builtin.txt*` and `*eval.txt*` should be kept for Vim-patches
Closes#33743
Problem:
Currently undefined behavior can occur when `string_fragment()` is
called with `OSC_COMMAND`. This is because when the state changes to
`OSC_COMMAND`, `string_initial` is set to true. Then in some cases,
directly after this `string_initial` will be set back to false before
the on_osc callback is called, this leads to `term_settermprop()` never
initializing the title.
Solution:
In all of the no-op cases in `string_fragment()` currently, we continue
to the end of the function where `vt->parser.string_initial` is set to
false. This change returns in the no-op cases instead since in these
cases the string has not yet been terminated and sent to the callback.
Note:
This change also adds a test with a byte sequence from the file
in #34028 that caused nvim to crash. This byte sequences is the shortest
sequence I could trim down from that file that still would trigger the
crash. There are also two other tests I added which validate that
setting the title with OSC-0 and OSC-2 still works.
Fixes: #34028
Problem: missing info about register completion in complete_info()
(after v9.1.1408)
Solution: update documentation and mention that register is used as
source, add a test (glepnir)
closes: vim/vim#1738949864aecd0
Co-authored-by: glepnir <glephunter@gmail.com>
Problem: not easily possible to complete from register content
Solution: add register-completion submode using i_CTRL-X_CTRL-R
(glepnir)
closes: vim/vim#173540546068aae
Problem: Unable to emit a message with arbitrary kind.
Solution: Add a "kind" opts field to nvim_echo().
Use it to set the "list_cmd" kind for vim.show_pos().
Problem: Search highlighting is applied strangely to the filler text of
transparent folds, and EOL list characters are not shown.
Solution: Don't apply search highlighting to the last column of the window row
if the last text char on the line is highlighted. Display the EOL list char if
needed. Don't highlight the entire filler text when matching EOL, just highlight
the EOL list char or the first filler char.
Problem: spurious CursorHold triggered in GUI on startup
Solution: init global did_cursorhold flag to true
(Gary Johnson)
When Vim is started in GUI mode, the CursorHold autocommand event is
triggered 'updatetime' milliseconds later, even when the user has not
pressed a key. This is different from the behavior of Vim in terminal
mode, which does not trigger a CursorHold autocommand event at startup,
and contradicts the description of the CursorHold event in ":help
CursorHold", which states that the event is "[n]ot triggered until the
user has pressed a key".
The fix is to change the initial value of did_cursorhold from FALSE to
TRUE. While it is true that the CursorDone event has not been done yet
at startup, it should appear to have been done until the user presses
a key.
fixesvim/vim#17350closes: vim/vim#17382318ff9c362
Co-authored-by: Gary Johnson <garyjohn@spocom.com>
Problem: Can't use getpos('v') in OptionSet when using setbufvar().
Solution: Don't reset Visual selection when switching to the same
buffer (zeertzjq).
closes: vim/vim#173735717ee33db
Problem: "99 searchcount ought to be enough for anyone."
Solution: Increase `SEARCH_STAT_DEF_MAX_COUNT` to 999, which I'm sure
will suffice for the next twenty years.
Co-authored-by: Sean Dewar <6256228+seandewar@users.noreply.github.com>
Problem: Consecutive "msg_show" events stemming from an `:echon`
command are supposed to be appended without a newline, this
information is not encoded in the "msg_show" event.
Solution: Add an "append" parameter to the "msg_show" event that is set
to true to indicate the message should not start on a new line.
Considered alternative: Emit a newline for the common case instead at the
start of a new message. That way UIs can more closely follow the logic
as it is implemented for the message grid currently. This would be a
breaking change. The "append" parameter seems OK.
Problem: When scrolling (the text) down with 'smoothscroll', filler
lines are revealed before the text skipped with `w_skipcol`.
Solution: Check `w_skipcol` before filler lines.
Problem: multi-byte mappings not properly stored in session file
Solution: unescape the mapping before writing out the mapping, prefer
single-byte mapping name if possible (Miguel Barro)
closes: vim/vim#173555b07aff2f6
Co-authored-by: GuyBrush <miguel.barro@live.com>
Copy whatever was made to work for generated headers:
(1) we need to consider all cmake targets not just main_lib
(2) we need to add the sysroot for macOS
Problem:
No way for a user to limit 'exrc' search in parent directories (compare
editorconfig.root).
Solution:
A configuration file can unset 'exrc', disabling the search for its
parent directories.
Problem: When items are combined with user-defined highlight attributes
(e.g., strikethrough), trunc inherits these attributes, making
the text difficult to read.
Solution: trunc now uses the original Pmenu and PmenuSel highlight
attributes (glepnir)
closes: vim/vim#173400816f17e9a
Co-authored-by: glepnir <glephunter@gmail.com>
This check was always broken. it will "detect" a file as
other-than-UTF-8 if the first line of a help file only is ASCII.
This only works by accident, as all our help files are UTF-8 (or
ASCII-only, which is fully compatible), but are all ASCII-only
on the first line of every help file which means that all helpfiles
gets detected as not-UTF8 which makes the "consistency" test pass
by accident even though the actual consistency is that every single
file is UTF-8 compatible. This means that the
"!_TAG_FILE_ENCODING\tutf-8\t" meta-tag already did not get emitted
but YAGNI in either case as no encoding tag just means that 'encoding'
is used which in neovim always is UTF-8 anyway.
An alternative approach would be to integrate the real encoding
detection already present in the codebase (an editor which edits text of
various encodings) which checks the entire file instead of a weird
first-line-only-hack, but as it happens to be 2025 the resolution of
encoding trouble is to just use UTF-8 everywhere. And if you use something
else you have to keep track yourself anyway it is not like we can detect
if one helpfile of your plugin is latin-1 and another is latin-2 or
whatever. Also, Nvim will detect the encoding of the file when you open
the file as a :help buffer anyway.
Problem: The 'grepformat' option is global option, but it would be
useful to have it buffer-local, similar to 'errorformat' and
other quickfix related options (Dani Dickstein)
Solution: Add the necessary code to support global-local 'grepformat',
allowing different buffers to parse different grep output
formats (glepnir)
fixes: vim/vim#17316closes: vim/vim#173157b9eb6389d
Co-authored-by: glepnir <glephunter@gmail.com>
Problem: search_stat not reset when pattern differs in case
(tahzibijafar)
Solution: use STRNCMP instead of MB_STRNICMP macro
There was a long standing todo comment, that using MB_STRNICMP is wrong.
So let's change it to STRNCMP() instead. Even if it not handle
multi-byte characters correctly, then Vim will rather recompute the
search stat, instead of re-using the old (and possibly wrong) value.
fixes: vim/vim#17312closes: vim/vim#17314670d0c1468
Co-authored-by: Christian Brabandt <cb@256bit.org>
This just cleans up some old definitions from before tui_set_term_mode
existed. There is no need to represent custom modes in a different
format just because they are safe without feature detection.
According to terminfo(5), these are obsolete:
Except for very old terminal descriptions, e.g., those developed
for SVr4, the scheme just described should be considered obsolete.
An improved set of capabilities was added late in the SVr4
releases (smglr and smgtb), which explicitly use two parameters
for setting the left/right or top/bottom margins.
Current "patch" asserts that EVERY terminal claiming to be an xterm has
broken vertical split scrolling which is far too an aggressive
workaround.
Instead, as left-right margins for scrolling regions use a
private DEC mode, we can use feature detection to use them in supported
terminals only. This way, users who use properly implemented terminals
can enjoy hardware accelerated scrolling even with vertical split
windows.
A primary device callback may set a new callback (e.g. when suspending),
so clearing it after invoking it is too late.
While at it, add missing reset of did_set_grapheme_cluster_mode in
terminfo_start().
Problem: One-off error in "count" to make "w_skipcol" zero with
'nosmoothscroll' page scrolling when last virtual line
in a buffer line is exactly the entire window width.
(Hirohito Higashi)
Solution: Properly compute the smallest integer value necessary
to make "w_skipcol" zero (Luuk van Baal)
c6c72d165c
Problem: buflist_new() leaks ffname and fails to reuse curbuf when
autocommands from buf_freeall change curbuf. Plus, a new
buffer is not allocated in this case, despite what the comment
above claims.
Solution: Remove the condition so ffname is not leaked and so a new
buffer is allocated like before v8.2.4791. It should not be
possible for undo_ftplugin or buf_freeall autocommands to
delete the buffer as they set b_locked, but to stay consistent
with other uses of buf_freeall, guard against that anyway
(Sean Dewar).
Note that buf is set to NULL if it was deleted to guard against the (rare)
possibility of messing up the "buf != curbuf" condition below if a new buffer
happens to be allocated at the same address.
closes: vim/vim#173190077282c82
Co-authored-by: Sean Dewar <6256228+seandewar@users.noreply.github.com>
Problem: Cannot get completion startcol when space is not the first
trigger character (after v9.1.1383)
Solution: Detect the next comma followed by a space in the option string
and use in next compare loop (glepnir)
closes: vim/vim#1731108db2f4f28
This is a left-over for when we ported to state-less scrolling events
(e.g. no scrolling regions as part of the ext_linegrid state).
We always reset the scrolling region after performing a non-fullscreen
scroll. Thus the extra check in tui_grid_resize() is not needed.
Problem: Loop that ensures "w_skipcol" is zero with 'nosmoothscroll'
for (half)-page scrolling is inefficient.
Solution: Calculate the required "count" instead of looping until
"w_skipcol" is zero (Luuk van Baal).
acf0ebe8a8
Problem: When a space character is used as a trigger in 'isexpand' option
it doesn't get recognized because skip_to_option_part() skips
spaces after a comma, treating them as option separators
rather than option value (after v9.1.1341)
Solution: manually set the part to a space character (glepnir).
closes: vim/vim#173058d0e42b710
