Problem: Temporarily changing current window in a script causes
CursorMoved to be triggerd.
Solution: Don't trigger CursorMoved if neither curwin nor cursor
changed between two checks.
* feat(treesitter): add injection language fallback
Problem: injection languages are often specified via aliases (e.g.,
filetype or in upper case), requiring custom directives.
Solution: include lookup logic (try as parser name, then filetype, then
lowercase) in LanguageTree itself and remove `#inject-language`
directive.
Co-authored-by: Lewis Russell <me@lewisr.dev>
If an iterator pipeline stage returns nil as its first return value, the
other return values are ignored and it is treated as if that stage
returned only nil (the semantics of returning nil are different between
different stages). This is consistent with how for loops work in Lua
more generally, where the for loop breaks when the first return value
from the function iterator is nil (see :h for-in for details).
Currently (as of nvim 0.9), the behavior of boolean params in
vim.api lua wrappers is inconsistent for optional parameters
(part of an `opts` dict) compared to positional parameters.
This was inadvertently changed in #24524 . While cleaning up this
inconsistency is something we might want eventually, it needs
to be discussed separately and the impact of existing code considered.
When an injection has not set include children, make sure not to add
the injection if no ranges are determined.
This could happen when there is an injection with a child that has the
same range as itself. e.g. consider this Makefile snippet
```make
foo:
$(VAR)
```
Line 2 has an injection for bash and a make variable reference. If
include-children isn't set (default), then there is no range on line 2
to inject since the variable reference needs to be excluded.
This caused the language tree to return an empty range, which the parser
now interprets to mean the full buffer. This caused makefiles to have
completely broken highlighting.
Problem: Cannot break undo by setting 'undolevels' to itself in
'inccommand' preview callback.
Solution: Don't set an invalid 'undolevels' value.
Co-authored-by: Michael Henry <drmikehenry@drmikehenry.com>
PR #23689 assumes `client.config.capabilities.workspace.didChangeWatchedFiles`
exists when checking `dynamicRegistration`, but thats's true only if it was
passed to `vim.lsp.start{_client}`.
This caused #23806 (still an issue in v0.9.1; needs manual backport), but #23681
fixed it by defaulting `config.capabilities` to `make_client_capabilities` if
not passed to `vim.lsp.start{_client}`.
However, the bug resurfaces on HEAD if you provide a non-nil `capabilities` to
`vim.lsp.start{_client}` with missing fields (e.g: not made via
`make_client_capabilities`).
From what I see, the spec says such missing fields should be interpreted as an
absence of the capability (including those indicated by missing sub-fields):
https://microsoft.github.io/language-server-protocol/specifications/lsp/3.17/specification/#clientCapabilities
Also, suggest `vim.empty_dict()` for an empty dict in
`:h vim.lsp.start_client()` (`{[vim.type_idx]=vim.types.dictionary}`
no longer works anyway, probably since the cjson switch).
Problem: cache paths are derived by replacing each reserved/filesystem-
path-sensitive char with a `%` char in the original path. With this
method, two different files at two different paths (each containing `%`
chars) can erroneously resolve to the very same cache path in certain
edge-cases.
Solution: derive cache paths by url-encoding the original (path) instead
using `vim.uri_encode()` with `"rfc2396"`. Increment `Loader.VERSION` to
denote this change.
Problem: 'inccommand' doesn't restore 'undolevels' properly for a
buffer shown in multiple windows.
Solution: Don't save information of a buffer twice.
Currently it only skips if `Rows` changed, but it's possible for the height of
the usable area for windows to change (e.g: via `&ch`, `&stal` or `&ls`), which
can cause the value of `&cmdheight` to change when the sizes are restored.
This is a Vim bug, so I've submitted a PR there too. No telling when it'll be
merged though, given the current lack of activity there.
`ROWS_AVAIL` is convenient here, but also subtracts the `global_stl_height()`.
Not ideal, as we also care about the height of the last statusline for other
values of `&ls`. Meh.
Introduce `last_stl_height` for getting the height of the last statusline and
use it in `win_size_save/restore` and `last_status` (means
`last_status_rec`'s `statusline` argument will now be true if `&ls` is 3,
but that does not change the behaviour).
Also corrects the logic in `comp_col` to not assume there's a last statusline
if `&ls` is 1 and the last window is floating.
This aligns its behaviour better with `nvim_win_close`.
Note that `:hide` is actually incapable of closing the cmdwin, unlike `:close`
and `:quit`, so this is a bit of a difference in behaviour.
Disallow closing the previous window from `nvim_win_close`, as this will cause
issues.
Again, no telling how safe this is. It also requires exposing old_curwin. :/
Also note that it's possible for the `&cmdheight` to change if, for example,
there are 2 tabpages and `nvim_win_close` is used to close the last window in
the other tabpage while `&stal` is 1. This is addressed in a later commit.
Problem: As discussed on Matrix, there was some interest in having
`nvim_open_win` again be able to open floats in the cmdwin (e.g: displaying a
hover doc related to what's in the cmdwin). After #23228, this was disallowed.
Solution: Allow `nvim_open_win` in the cmdwin as long as `!enter` and
`buffer != curbuf` (the former can cause all sorts of issues, and the latter
can crash Nvim after closing cmdwin). Also allow `nvim_win_set_buf` in a similar
fashion.
Note that we're not *entirely* sure if this is 100% safe (cmdwin is a
global-state-using-main-loop-calling beast), but this seems to work OK..?
Also:
- Check the buffer argument of `nvim_open_win` earlier, and abort if it's
invalid (it used to still open a window in this case).
- Untranslate `e_cmdwin` errors in the API (other errors in the API are not
translated: although not detailed in the API contract yet, errors are
supposed to be stable).
Problem:
The cmdpreview saved undo nodes on cmdpreview_prepare() from ex_getln.c may
become invalid (free) if the preview function makes undo operations, causing
heap-use-after-free errors.
Solution:
Save the buffer undo list on cmdpreview_prepare)_ and start a new empty one. On
cmdpreview_restore_state(), undo all the entries in the new undo list and
restore the original one. With this approach, the preview function will be
allowed to undo only its own changes.
Fix#20036Fix#20248
Problem: Cannot drag a vertical separator to the right of a window
whose 'statuscolumn' is wider than itself.
Solution: Never treat a click on a vertical separator as a click on
'statuscolumn'.
Avoid consecutive RPC requests involving :startinsert or :stopinsert,
because consecutive RPC requests may be processed together, before the
:startinsert or :stopinsert takes effect.
Also change some feed_command() to command() to make tests faster.
Problem:
On Windows, `rundll32` exits zero (success) even when given
a non-existent file.
Solution:
Mock vim.system() on Windows to force a "failure" case.
Previously, when using foldmethod=indent, inserting an unindented line
would inadvertently open closed folds below it.
As a performance improvement, folds were only updated once, across all
lines, after Insert mode was exited.
Now, the performance improvement is no longer being used when
foldmethod=indent, so folds are updated multiple times during Insert
mode, but only across the lines that are changing, which preserves the
folds (and their open/close states) instead of recreating them.
Problem:
helpers.tmpname() may create a local file, depending on circumstances.
Solution:
Only use helpers.tmpname() for its parent directory (the "temp root").
Use fs_mkdtemp() to actually get a unique name.
Problem:
Bash language server returns "hover" markdown content that starts with
a code fence and info string of `man` preceded by whitespace, which Nvim
does not render properly.
See 0ee73c53ce/server/src/server.ts (L821C15-L821C15)
```typescript
function getMarkdownContent(documentation: string, language?: string): LSP.MarkupContent {
return {
value: language
? // eslint-disable-next-line prefer-template
['``` ' + language, documentation, '```'].join('\n')
: documentation,
kind: LSP.MarkupKind.Markdown,
}
}
```
For example,
```
``` man
NAME
git - the stupid content tracker
```
```
If I remove the white space, then it is properly formatted.
```
```man instead of ``` man
```
Per CommonMark Spec https://spec.commonmark.org/0.30/#info-string
whitespace is allowed before and after the `info string` which
identifies the language in a codeblock.
> The line with the opening code fence may optionally contain some text
> following the code fence; this is trimmed of leading and trailing
> spaces or tabs and called the [info
> string](https://spec.commonmark.org/0.30/#info-string). If the [info
> string](https://spec.commonmark.org/0.30/#info-string) comes after
> a backtick fence, it may not contain any backtick characters. (The
> reason for this restriction is that otherwise some inline code would
> be incorrectly interpreted as the beginning of a fenced code block.)
Solution:
Adjust stylize_markdown() to allow whitespace before codeblock info.
