Problem: Some tests are slow.
Solution: Make a few test cases faster.
bf63011a52
Only include the RunDbgCmd() to s:RunDbgCmd() change.
Co-authored-by: Bram Moolenaar <Bram@vim.org>
Problem: Debugger does not display the whole command.
Solution: Set ea.cmd before checking for a breakpoint.
33b55b562b
The code change is already covered by #35019.
Co-authored-by: Bram Moolenaar <Bram@vim.org>
Problem:
MessagePack fixstr format supports string lengths 0-31, but mpack_str()
only used fixstr for lengths < 20. Strings of 20-31 bytes were
incorrectly encoded as str8 (2-byte header) instead of fixstr (1-byte
header).
Solution:
Change the condition from `len < 20` to `len < 32` to match the
MessagePack specification.
This fix affects message timing which exposed a pre-existing race
condition in the channels_spec PTY test. The test now uses a helper
function to accumulate partial PTY reads, making it more robust.
Fixes#32784
Problem: tests: excessive wait in Test_matchfuzzy_initialized
Solution: Use term_wait() instead of the TermWait() wrapper
(Corey Hickey)
Test_matchfuzzy_initialized is a terminal test, which are specified to be
"flaky" and automatically retried. The TermWait wrapper multiplies the
specified wait time by higher values for later retries, maxing out at
10x the specified value. This makes tries vim/vim#3 to vim/vim#6 sleep for 20 seconds
each, which makes the test very slow to work with. The specified intent
of the test (as noted in a comment eight lines above here) is to
sleep for 2s.
closes: vim/vim#18822d4f9de889b
Co-authored-by: Corey Hickey <bugfood-c@fatooh.org>
Problem: tests: Test_matchfuzzy_initialized() fails
Solution: Send a dummy key (Corey Hickey)
Test_matchfuzzy_initialized seems to expect that the 'lvimgrep' will be
interrupted by sending a SIGINT. If the search finishes beforehand,
though, then the SIGINT triggers vim to tell the user how to quit.
Vim does not show this message immediately, though; instead, vim shows
the message next time it is active. When StopVimInTerminal() sends a key
sequence intended to cause vim to quit, this activates vim to show the
message instead of quitting.
I do not understand every detail of the problem fully--if I type the
characters from StopVimInTerminal() into a post-SIGTERM terminal
directly, that seems to work ok; there seems to be a timing issue due to
sending all the characters at once.
This fix does make the test work reliably for me, and the test still
works even if I limit my CPU frequency so that the search is interrupted
by the SIGINT.
fixes: vim/vim#18821
related: vim/vim#18822a3925d783a
Co-authored-by: Corey Hickey <bugfood-c@fatooh.org>
Problem: filetype: Erlang lexical files are not recognized
Solution: Detect *.xrl files as leex filetype, include syntax and
filetype plugins (Jon Parise).
leex is the lexical analyzer generator for Erlang. Its input file format
follows a section-based structure and uses the `.xrl` file extension.
This initial work includes file detection, an ftplugin (which inherits
the Erlang configuration), and a syntax definition.
Reference:
- https://www.erlang.org/doc/apps/parsetools/leex.html
related: vim/vim#18819closes: vim/vim#18832b087c5452b
Co-authored-by: Jon Parise <jon@indelible.org>
Problem: filetype: not all Erlang files are recognized
Solution: Detect *.app.src and rebar.config files as erlang filetype
(John Parise).
*.app.src files contain Erlang application definitions. (There are also
*.app files, which are similar but more often build artifacts, and that
file extension is too ambiguous to be recognized by default.)
Reference:
- https://www.erlang.org/doc/system/applications.html
Rebar is the Erlang build tool. rebar.config uses Erlang syntax.
Reference:
- https://rebar3.org/docs/configuration/configuration/closes: vim/vim#188352b2580e61a
Co-authored-by: Jon Parise <jon@indelible.org>
Problem:
Nvim supports `textDocument/semanticTokens/full` and `…/full/delta`
already, but most servers don't support `…/full/delta` and Nvim will try
to request and process full semantic tokens response on every buffer
change. Even though the request is debounced, there is noticeable lag if
the token response is large (in a big file).
Solution:
Support `textDocument/semanticTokens/range`, which requests semantic
tokens for visible screen only.
Problem:
Currently we fetch a prebuilt xxd.exe from neovim/deps for Windows,
which is not ideal in terms of obviousness, misses updates, and is
clumsy to update.
Solution:
Similar to tee.c (#36363), vendor xxd.c from Vim and build it as part
of the Neovim build process. This makes the source obvious, enables
analyzer checks, and simplifies updates.
Fixes#36664
Problem: Returning 0 for has('patch-9.0.0') is inconsistent.
Solution: Make it return 1. (closesvim/vim#10640)
b0375d466e
Co-authored-by: Bram Moolenaar <Bram@vim.org>
Problem: has() is not strict about parsing the patch version.
Solution: Check the version more strictly. (Ken Takata, closesvim/vim#10752)
d90f91fe30
Co-authored-by: K.Takata <kentkt@csc.jp>
Problem: It is desirable for the default statusline to contain colored
diagnostics information. However, current `StatusLine` group is
purposefully defined as almost inverted `Normal` to "make current
window obvious". This makes diagnostic information from
`vim.diagnostic.status()` barely visible: it uses established
`DiagnosticSignXxx` groups which have colored foreground with
lightness close to `StatusLine` background.
Also the `StatusLineNC` group is fairly different from `Normal` in
order to both "makes window separators clear" and "be different from
`CursorLine`". But not as mush different as `StatusLine` because
"`StatusLine` and `StatusLineNC` should be clearly different".
Solution: Make both `StatusLine` and `StatusLineNC` be slightly closer
in lightness to `Normal`. This makes `StatusLine` and `StatusLineNC`
groups satisfy their conditions in the following way:
- `vim.diagnostic.count()` is readable on `StatusLine` - yes.
- `vim.diagnostic.count()` is readable on `StatusLineNC` - yes.
- `StatusLine` makes current window obvious - I'd say yes.
- `StatusLine` and `StatusLineNC` are clearly different - it depends
on the eyes and monitor. The current is clearly better, but the new
ones I'd say are still visibly different.
- `StatusLineNC` makes window separators clear - I'd say yes, but
depends on the eyes and monitor.
- `StatuslineNC` is different from `CursorLine` - NO, they are same.
Another approach to solve this would be to introduce dedicated
`DiagnosticStatuslineXxx` groups to use in `vim.diagnostics.status()`.
They can be defined using foreground colors from the same lightness as
`Normal`. This would make them readable in `StatusLine`. But not
`StatusLineNC`, though.
Problem: completion: complete_match() Vim script function and
'isexpand' option are not that useful and confusing
(after v9.1.1341)
Solution: Remove function and option and clean up code and documentation
(Girish Palya).
complete_match() and 'isexpand' add no real functionality to Vim. They
duplicate what `strridx()` already does, yet pretend to be part of the
completion system. They have nothing to do with the completion mechanism.
* `f_complete_match()` in `insexpand.c` does not call any completion code.
It’s just a `STRNCMP()` wrapper with fluff logic.
* `'isexpand'` exists only as a proxy argument to that function.
It does nothing on its own and amounts to misuse of a new option.
The following Vim script function can be used to implement the same
functionality:
```vim
func CompleteMatch(triggers, sep=',')
let line = getline('.')->strpart(0, col('.') - 1)
let result = []
for trig in split(a:triggers, a:sep)
let idx = strridx(line, trig)
if l:idx >= 0
call add(result, [idx + 1, trig])
endif
endfor
return result
endfunc
```
related: vim/vim#16716fixes: vim/vim#18563closes: vim/vim#18790cbcbff8712
Co-authored-by: Girish Palya <girishji@gmail.com>
Problem: completion: 'completefuzzycollect' option is too obscure
Solution: Deprecate the option, but don't error out for existing scripts,
behave like 'completefuzzycollect' is set when fuzzy
completion is enabled (Girish Palya).
fixes: vim/vim#18498closes: vim/vim#1878833fbfe003c
Remove this option completely, as it's introduced in Nvim v0.12 cycle.
Co-authored-by: Girish Palya <girishji@gmail.com>
Problem: When creating floating window from existing window, the
window-local 'statusline' option is inherited, causing unwanted
statusline display. Additionally, with laststatus=3 (global statusline),
the early return skipped clearing wp->w_redr_status flag.
Solution: Clear inherited window-local 'statusline' in win_new_float
for style="minimal". Restructure win_redr_status to always clear the
flag before any early returns, and only render floating window statusline
when window-local 'statusline' is explicitly set. And respect 'laststatus'
option.
Don't handle cindent in insert_check(). Instead, do that just before
returning from insert_execute() if required.
This also makes the in_cinkeys() change from #12894 unnecessary.
Problem:
nvim_get_commands does not return callbacks defined for
"preview", "complete", or the command itself.
Solution:
- Return Lua function as "callback" field in a Lua context.
- Return "preview" function in a Lua context.
- BREAKING: Return "complete" as a function instead of a boolean.
Problem:
`diagnostic.status()` is configured via `config.signs`, but users may
want diagnostics only in statusline, not in the gutter (signs).
Solution:
Add `config.status`.
Problem: apply_autocmds function can free both buf_T and win_T pointers
Solution: instead retain winids for WinResized and WinScrolled
autocmds and use curbuf pointer, which is consistent with other uses
of apply_autocmds function
Problem:
No way to detect at runtime if the build includes unibilium (or whatever
terminfo layer we swap it with later).
Solution:
Support `has('terminfo')`.
Problem: `extmark_splice()` was being called before `ml_replace()`,
which caused the on_bytes callback to be invoked with the old buffer
text instead of the new text.
Solution: store metadata for each match in a growing array, call
`ml_replace()` once to update the buffer, then call `extmark_splice()`
once per match.
Closes https://github.com/neovim/neovim/issues/36370.
Problem: When assigning to @. in a :let command an incorrect "E15"
error is emitted.
Solution: Emit the correct "E354" error. (Doug Kearns).
An incorrect "E488" error was also emitted in Vim9 script assignments.
It appears that the code deleted in this commit was added to work around
a limitation in the returned value from find_name_end() that no longer
exists.
See commit 76b92b2830841fd4e05006cc3cad1d8f0bc8101b (tag: v7.0b).
closes: vim/vim#187572447131e00
Co-authored-by: Doug Kearns <dougkearns@gmail.com>
Problem: Wrong virtcol('$') with virtual text at EOL (rickhowe).
Solution: Also add 1 to end virtcol when there is virtual text.
(zeertzjq)
fixes: vim/vim#18761closes: vim/vim#18762d434f6c2a5
Problem: tests: not enough testing for wildtrigger() pum redrawing.
Solution: Also test redrawing when leaving cmdline mode (zeertzjq).
closes: vim/vim#18773eb33c2eb28
Problem:
Trying to match the search highlight groups to the Normal highlight for
the window can fail when the message highlighting contains a fg/bg that
the Normal highlight doesn't (like an error message in cmd will have
ErrorMsg highlight instead of MsgArea - which is Normal in cmd.)
Solution:
Link the search highlight groups to an empty group in 'winhighlight'
thus disabling them instead of overriding them with Normal/MsgArea/etc.
Limit the default truncation item to the current recursion range so
nested `nvim_eval_statusline()` calls don't reuse stale `stl_items`
pointers. Add a functional regression that evaluates a Lua statusline
helper which forces truncation to ensure the nested scenario stays
stable.
AI-Assist: OpenAI ChatGPT
Fixes#36616
Problem:
When running ":edit <url>", filetype detection is not triggered.
Solution:
Run the autocmds in the filetypedetect group after loading the content.
Problem:
After fetching remote content from a URL and adding it to the buffer,
the buffer is marked as modified. This is inconsistent with the original
netrw behavior, and it causes problems with `:e` to refresh or `:q` as
it prompts for saving the file even if the user hasn't touched the
content at all.
Solution:
Mark the buffer as unmodified right after adding the remote content to
the buffer.
Technically the current behavior does match documentation. However, the
keys following <Cmd>/K_LUA aren't normally received by vim.on_key()
callbacks either, so it does makes sense to discard them along with the
preceding key.
One may also argue that vim.on_key() callbacks should instead receive
the following keys together with the <Cmd>/K_LUA, but doing that may
cause some performance problems, and even in that case the keys should
still be discarded together.
fix(ui2): hide search highlights in msg window.
Problem: Search highlighting is shown in the msg (and dialog) window.
Solution: Hide search highlighting in all but the pager window.
Problem:
With the typescript LSes typescript-language-server and vtsls,
omnicompletion on partial tokens for certain types, such as array
methods, and functions that are attached as attributes to other
functions, either results in no entries populated in the completion menu
(typescript-language-server), or an unfiltered completion menu with all
array methods included, even if they don't share the same prefix as the
partial token being completed (vtsls).
Solution:
Enable insertReplaceSupport and uses the insert portion of the lsp
completion response in adjust_start_col if it's included in the
response.
Completion results are still filtered client side.
Problem: No way to customize completion order across multiple servers.
Solution: Add `cmp` function to `vim.lsp.completion.enable()` options
for custom sorting logic.
Problem: Cannot reuse same config with noautocmd for both window
creation and updates, even when value is unchanged.
Solution: Only reject noautocmd changes for existing windows.
