Problem:
Cannot remove a `@conceal` highlight when defined in highlights.scm.
Solution:
Support a `@noconceal` highlight that works similarly to `@nospell` where it
overrides the conceal set on the range to remove it. Additionally, can
set the conceal metadata field to false for the same behavior.
Problem:
Invalid `nvim_create_user_command` calls can leak the
`preview` callback reference after Neovim has taken ownership of it.
1. build with {a,l}san
2. run:
```sh
<path/to/nvim> --headless -u NONE --clean +'lua
for i = 1, 100 do
pcall(vim.api.nvim_create_user_command,
"some very epic stuff" .. i,
{}, -- NOTE: this is INVALID (not a function or string)
{ preview = function() end })
end
vim.cmd("qa!")
' +qa
```
3. see:
```
100 lua references were leaked!
```
Solution:
Clear `preview_luaref` in `err:`.
Problem: fg_indexed/bg_indexed were dropped from nvim_get_hl output due
to a wrong short_keys guard. HL_FG_INDEXED also wasn't cleared in
hl_blend_attrs, and HLATTRS_DICT_SIZE was too small.
Solution: Remove the short_keys guard, clear HL_FG_INDEXED in
hl_blend_attrs, bump HLATTRS_DICT_SIZE to 24, and clarify docs that
these flags mean rgb is an approximation of the cterm palette index.
Problem: too many strlen() calls when adding strings to dicts
Solution: Refactor code to use string_T, use dict_add_string_len()
instead of dict_add_string() (John Marriott)
Additionally:
- In textprop.c, in function prop_fill_dict() use a string_T to store
local variable text_align.
- In popupwin.c, use a string_T to store struct member pp_name in struct
poppos_entry_T.
- In mark.c, refactor function add_mark() to pass in the length of
argument mname.
- In insexpand.c:
->Use a string_T to store the elements of static array
ctrl_x_mode_names.
->Refactor function trigger_complete_done_event():
->->change type of argument char_u *word to string_T *word.
->->make one access of array ctrl_x_mode_names instead of two.
->Refactor function ins_compl_mode() to accept a string_T to return the
resulting string.
- In fileio.c:
->Refactor function getftypewfd() to accept a string_T to return the
resulting string.
->In function create_readdirex_item() use a string_T to store local
variable q.
- In cmdexpand.c, store global cmdline_orig as a string_T.
- In autocmd.c, in function f_autocmd_get() use a string_T to store local
variables event_name and group_name. Measure their lengths once when
they are assigned so they are not remeasured on each call to
dict_add_string() in the subsequent for loop.
- In channel.c, in function channel_part_info() drop local variable status
and use s instead. Make s a string_T.
closes: vim/vim#19999c13232699d
Co-authored-by: John Marriott <basilisk@internode.on.net>
* fix(api): allow silencing "Too many highlight groups" error
Problem: Using Lua's `vim.api.nvim_set_hl(0, 'New', {...})` can fail if
there are too many existing highlight groups. However, this error can
not be silenced with `pcall`.
Solution: Make it possible to silence in `nvim_set_hl` and
`nvim_get_hl_id_by_name`.
* fix(lsp): limit number of groups created by `document_color()`
Problem: A file can contain many string colors that would be highlighted
by an LSP server. If this number crosses 19999 (maximum number of
allowed highlight groups), there are general issues with creating
other highlight groups, which can break functionality outside of
`vim.lsp.document_color`.
Solution: Limit number of highlight groups that are created by
`vim.lsp.document_color` to 10000 (half of allowed maximum).
This is not a 100% solution (since there can exist more than 10000
other highlight groups), but explicitly checking number of groups is
slow and 10000 should (hopefully) be enough for most use cases.
Problem:
Naming conventions are not automatically checked.
Solution:
Add a check to the doc generator. Eventually we should extract this
somehow, but that will require refactoring the doc generator...
Note: this also checks non-public functions, basically anything that
passes through `gen_eval_files.lua` and `gen_vimdoc.lua`. And that's
a good thing.
continues d0af4cd909.
This commit renames positional parameters. This is only "cosmetic", but
is intended to make it extra clear which name is preferred, since people
often copy existing code despite the guidelines in `:help dev-naming`.
In 3a4a66017b, 4d3a67cd62, df8d98173c we renamed "buffer" to "buf"
in dict parameters and return-values.
This commit renames positional parameters. This is only "cosmetic", but
is intended to make it extra clear which name is preferred, since people
often copy existing code despite the guidelines in `:help dev-naming`.
Problem: The cursor shape is changed to indicate when it is behind an
unfocused floating window (since a2b92a5e). This behavior
cannot be controlled by a floating window that doesn't want
to dim the cursor.
Solution: Assign a zindex-offset of 50 to the zindex of the current
window. To not dim the cursor when creating a floating window
on top of the current window one can assign the zindex
accordingly.
Problem:
clang 21 promoted alpha.security.ArrayBoundV2 to security.ArrayBound
(stable). This new check reports false-positive "out of bound access"
errors in drawline.c and vimscript.c, where the analyzer constructs
impossible paths (e.g. concealed line with draw_text=false yet ptr
advanced past the NUL terminator, or root AST node with a "next"
sibling).
Per-line NOLINT suppression doesn't work because the analyzer finds
multiple paths to the same false positive.
Solution:
Disable clang-analyzer-security.ArrayBound globally in the
clang-analyzer cmake target until the check matures.
Co-Authored-By: Claude
Problem: hlgroup2dict passes &ns_id to ns_get_hl twice. The first call
(link=true) sets *ns_hl = 0 when link_global is set, so the second call
and the sg_cleared guard both see ns_id == 0 and bail out. The group is
silently dropped from the result.
Solution: use a temporary copy of ns_id for each ns_get_hl call so the
original value is preserved.
Problem:
`:help dev-name-common` states that "buf" should be used instead of
"buffer" but there are cases where buffer is mentioned in the lua API.
Solution:
- Rename occurrences of "buffer" to "buf" for consistency with the
documentation.
- Support (but deprecate) "buffer" for backwards compatibility.
Co-authored-by: Justin M. Keyes <justinkz@gmail.com>
Problem: nvim_clear_autocmds() does not type check "event" correctly, and also
treats an empty array "event" like nil.
Solution: fix type checking. Treat empty array "event" as a no-op, like
nvim_exec_autocmds(). Add some extra tests.
Likewise the nil handling change may be considered breaking if anyone
(unintentionally) relied on that. It was also true that integer, function, etc.
"event"s would also be treated like nil!
Note that an empty string "event" is still an error, as that's must be an exact
match on an event name.
Problem: nvim_exec_autocmds() documentation incorrectly describes the default
for "pattern" as *, when it's actually the current file name (like :doautocmd).
Solution: correct it. Add a test.
Problem: in autocmd APIs, a non-nil "pattern" containing only empty
'sub'-patterns is silently treated as nil, causing the fallback value to be
unexpectedly used instead.
Solution: for nvim_create_autocmd(), raise a validation error (as no autocmds
would be created). For nvim_{exec,clear}_autocmds(), make it a no-op (as
matching no autocmds is not an error).
Problem:
The "restart" event has some problems:
- all UI clients must implement a somewhat complex set of setups
- UI must be on the same machine as the server
- only works for the "current" UI
- race/edge case: If the user config has errors / waiting for input, are
all UIs able to attach while Nvim is waiting for input?
Solution:
- Perform the restart on the server, not the client.
- Pass listen address (instead of CLI args) in the UI event.
- Simplifies UI logic: they only need to attach to new address.
- Opens the door for more enhancements in the future, such as allowing
all UIs to reattach instead of only the "current" UI.
Co-authored-by: zeertzjq <zeertzjq@outlook.com>
Co-authored-by: Justin M. Keyes <justinkz@gmail.com>
Problem:
- Progress-events are filtered by "source". But "source" is not required by nvim_echo.
- Without "++nested" (force=false), nvim_echo in an event-handler does not trigger Progress events.
- vim.health does not declare a "source".
Solution:
- Make source mandatory for progress-messages
- Enable ++nested (force=true) by default when firing Progress event.
- Set "source" in vim.health module.
Problem:
Currently, there's no way to distinguish progress messages coming from
different sources. Nor can Progress event be easily filtered based on
source.
Solution:
- Add "source" field to nvim_echo-opts.
- The Progress event pattern is now defined by the "source" field.
- Include the "title" as ev.data.
- Unrelated change: set force=false to disable nesting.
Problem: nvim_set_hl always replaces all attributes.
Solution: Add update field. When true, merge with existing
attributes instead of replacing. Unspecified attributes are preserved.
If highlight group doesn't exist, falls back to reset mode.
Problem:
`vim.keymap.*.Opts.buf` allows `boolean` aliases for more widely
used `integer?` values, `true` -> `0` and `false` -> `nil`. This
conversion is unnecessary and can be handled at call sites.
Solution:
As a follow-up to deprecating the `buffer` option, drop support for
boolean values for the new `buf` option. The deprecated `buffer`
continues to support booleans for backward compatibility.
The `buffer` option remains functional but is now undocumented.
Providing both will raise an error. Since providing `buf` was disallowed
before, there is no code that will break due to using `buffer` alongside
`buf`.
Problem:
If buffer update callbacks poll for uv events during terminal scrollback
refresh, new output from PTY process may lead to incorrect scrollback.
Solution:
Don't poll for output to the same terminal as the one being refreshed.
Problem:
`nvim_echo(…, {id=…})` accepts user-defined id as a string or integer.
Generated ids are always higher than last highest msg-id used. Thus
plugins may accidentally advance the integer id "address space", which,
at minimum, could lead to confusion when troubleshooting, or in the
worst case, could overflow or "exhaust" the id address space.
There's no use-case for it, and it could be the mildly confusing, so we
should just disallow it.
Solution:
Disallow *integer* user-defined message-id.
Only allow *string* user-defined message-id.
Problem:
nvim_get_option_value with "filetype" set silently returns incorrect
defaults if autocommands are blocked, like when they're already running.
Solution:
Allow its FileType autocommands to nest: `do_filetype_autocmd(force=true)`.
Also error if executing them fails, rather than silently return wrong defaults.
Endless nesting from misbehaving scripts should be prevented by the recursion
limit in apply_autocmds_group, which is 10.
Problem: nvim_open_tabpage's "enter" argument is optional, which is inconsistent
with nvim_open_win.
Solution: make it a (non-optional) positional argument, like nvim_open_win.
Also change "enter"'s description to be more like nvim_open_win's doc.
Problem: "after" in nvim_open_tabpage is inconsistent with how a count works
with :tab, :tabnew, etc. Plus, the name "after" implies it's inserted after that
number.
Solution: internally offset by 1. Allow negative numbers to mean after current.
Hmm, should we even reserve sentinels for after current? Callers can probably
just use nil...
- Cleanup, remove redundant comments, add more tests.
- Enhance win_new_tabpage rather than create a new function for !enter, and use
a different approach that minimizes side-effects. Return the tabpage_T * and
first win_T * it allocated.
- Disallow during textlock, like other APIs that open windows.
- Remove existing win_alloc_firstwin error handling from win_new_tabpage; it's
not needed, and looks incorrect. (enter_tabpage is called for curtab, which is
not the old tabpage! Plus newtp is not freed)
- Fix checks after creating the tabpage:
- Don't fail if buf wasn't set successfully; the tab page may still be valid
regardless. Set buffer like nvim_open_win, possibly blocking Enter/Leave
events. (except BufWinEnter)
- tp_curwin may not be the initial window opened by win_new_tabpage. Use the
win_T * it returns instead, which is the real first window it allocated,
regardless of autocmd shenanigans.
- Properly check whether tab page was freed; it may have also been freed
before win_set_buf. Plus, it may not be safe to read its handle!
Problem: "win" is allowed in external window configs in some cases. External
window converted to normal float can't move tabpages in one nvim_win_set_config
call. External window can't be turned into a normal split.
Solution: disallow setting "win" for external windows. Allow external window to
move tabpages, which turns it non-external. Allow external window to be turned
into a (non-external) split.
parse_win_config has more validation issues from not considering the window's
existing config enough (not from this PR). For example, zindex can be set for an
existing split if "split"/"vertical" isn't given, despite intending for that to
be an error. Plus the logic is confusing.
It could do with a refactor at some point...
