`[Backport release-x.y]` will no longer be part of the pull request
title. This means the PR titles will go from looking like
```
[Backport release-0.9] fix(languagetree): remove double recursion in LanguageTree:parse
```
to
```
fix(languagetree): remove double recursion in LanguageTree:parse
```
The benefit of this is that pull requests merged with the "Squash and
Merge" strategy (which uses the PR title as the commit message), will
still follow the conventional commits specification. This will help
tools that rely on conventional commits such as git-cliff.
The `backport` label is added to backported PRs to help distinguish
between backport PRs with regular PRs in the "Pull Requests" tab on
github.
To reduce confusion with the `backport` label, the label to trigger the
backporting has been changed from `backport release-x.y` to
`ci:backport release-x.y`. This is also more consistent with other
labels that trigger a CI job which all use the `ci:` prefix.
This reverts commit 15e77a56b7.
Subpriorities were added in https://github.com/neovim/neovim/pull/27131
as a mechanism for enforcing query order when using iter_matches in the
Tree-sitter highlighter. However, iter_matches proved to have too many
complications to use in the highlighter so we eventually reverted back
to using iter_captures (https://github.com/neovim/neovim/pull/27901).
Thus, subpriorities are no longer needed and can be removed.
Problem: filetype: .out files recognized as tex files
Solution: Do not set an explicit filetype until it is clear what this
should be (shane.xb.qian)
closes: vim/vim#14670e35478bc9d
Co-authored-by: shane.xb.qian <shane.qian@foxmail.com>
Problem: Kbuild files are not recognized.
Solution: Detect Kbuild files as make files.
(Bruno Belanyi)
closes: vim/vim#146765cbc9a69e5
Co-authored-by: Bruno BELANYI <bruno@belanyi.fr>
Follow-up to #28490
Problem:
The new behaviour of goto_next/prev() of navigating to the next highest
severity doesn't work well when diagnostic providers have different
interpretations of severities. E.g. the user may be blocked from
navigating to a useful LSP warning, due to some linter error.
Solution:
The behaviour of next highest severity is now a hidden option
`_highest = true`. We can revisit how to integrate this behaviour
during the 0.11 cycle.
runtime(java): Improve the recognition of the "indent" method declarations (vim/vim#14659)
There is a flaw in the current implementation that has been
exacerbated around v5.2. It lies in the recognition of all
three indentation styles simultaneously: a tab, two space,
and eight space character(s). With it, it is not uncommon
to misidentify various constructs as method declarations
when they belong to two-space indented members and other
blocks of a type and are offset at eight space characters or
a tab from the start of the line.
For example,
------------------------------------------------------------
class Test
{
static String hello() { return "hello"; }
public static void main(String[] args)
{
try {
if (args.length > 0) {
// FIXME: eight spaces.
System.out.println(args[0]);
} else {
// FIXME: a tab.
System.out.println(hello());
}
} catch (Exception e) {
throw new Error(e);
}
}
}
------------------------------------------------------------
------------------------------------------------------------
:let g:java_highlight_functions = 'indent'
:doautocmd Syntax
------------------------------------------------------------
A better approach is to pick an only indentation style out
of all supported styles (so either two spaces _or_ eight
spaces _or_ a tab). Note that tabs and spaces can still be
mixed, only the leading tab or the leading run of spaces
matters for the recognition. And there is no reason to not
complement the set of valid styles with any number of spaces
from 1 to 8, inclusively.
Please proceed with the necessary change as follows:
- rename from "indent" to "indent2" for a 2-space run;
- rename from "indent" to "indent8" for an 8-space run;
- continue to have "indent" for a tab run;
- define an "indent" variable with a suffix number denoting
the preferred amount of indentation for any other run of
spaces [1-8].
As before, this alternative style of recognition of method
declarations still does not prescribe naming conventions and
still cannot recognise method declarations in nested types
that are conventionally indented.
The proposed changes also follow suit of "style" in stopping
the claiming of constructor and enum constant declarations.
c4d0c8c812
Co-authored-by: Aliaksei Budavei <32549825+zzzyxwvut@users.noreply.github.com>
Problem: cbuffer and similar quickfix and locationlist commands don't
accept a range, even so it is documented they should
(ilan-schemoul, after 8.1.1241)
Solution: Define ex commands with ADDR_LINES instead of ADDR_OTHER
fixes: vim/vim#14638closes: vim/vim#14657652c821366
Co-authored-by: Christian Brabandt <cb@256bit.org>
Instead of painfully messing with timing to determine if queries were
reparsed, we can simply keep a counter next to the call to ts_query_new
Also memoization had a hidden dependency on the garbage collection of
the the key, a hash value which never is kept around in memory. this was
done intentionally as the hash does not capture all relevant state for the
query (external included files) even if actual query objects still
would be reachable in memory. To make the test fully deterministic in
CI, we explicitly control GC.
In particular, TUI manages its own screen buffers and highlight table, so we don't need
to run init_highlight() and default_grid_alloc() in the ui client process.
runtime(asm): missing setlocal in indent plugin (vim/vim#14658)
2e9b9e9a9e
Co-authored-by: Marc Sven Schulte <167623652+msschulte@users.noreply.github.com>
Problem: Calculating line height for unnecessary amount of lines with
half-page scrolling (zhscn, after 9.1.0280)
Solution: Replace "limit_winheight" argument with higher resolution
"max" argument to which to limit the calculated line height
in plines_m_win() to (Luuk van Baal)
32d701f51b
Problem:
The new LSP "refactor menu" keybinding "crr" is also defined in visual
mode, which overlaps with the builtin "c".
Solution:
Use CTRL-R instead of "crr" for visual mode.
fix#28528
Based on feedback from #28324, pass -H and -I to regular grep
(available on all platforms officially supported by Neovim), and
only pass -uu to ripgrep. This makes :grep ignore binary files by
default in both cases.
* fix(treesitter): enforce lowercase language names
Problem: On case-insensitive file systems (e.g., macOS), `has_parser`
will return `true` for uppercase aliases, which will then try to inject
the uppercase language unsuccessfully.
Solution: Enforce and assume parser names to be lowercase when
resolving language names.
Reverts parts of https://github.com/neovim/neovim/pull/27674
LSP snippets typically do include tabs or spaces to add extra
indentation and don't rely on the client using `autoindent`
functionality.
For example:
public static void main(String[] args) {\n\t${0}\n}
Notice the `\t` after `{\n`
Adding spaces or tabs independent of that breaks snippets for languages
like Haskell where you can have snippets like:
${1:name} :: ${2}\n${1:name} ${3}= ${0:undefined}
To generate:
name ::
name = undefined
Problem: when line is blank link then there will got an invalid column number in math.min compare.
Solution: make sure the min column number is 0 not an illegal number.
There's no "rule" or bad practice or whatever that says we cannot
generate c files. it is is just that we have ~20 generated headers
and ~2 generated sources and there is nothing in these two generated
source files which sets them aparts. Lua bindings are not different from
rpc bindings, and pathdef is not different from versiondef.
So to simplify build logic and ease the future port to build.zig,
streamline the build to only have generated headers, no direct generated
.c files.
Also "nlua_add_api_functions" had its prototype duplicated twice which
defeated the point of having mandatory prototypes (one source of truth).
Problem: Enabling ext_messages claims to set 'cmdheight' to zero, but
only does so for the current tabpage.
Solution: Set stored 'cmdheight' value to zero for all tabpages.
Problem: When :edit an existing buffer, line('w$') may return a
wrong result.
Solution: Reset w_valid in curwin_init() (Jaehwang Jung)
`do_ecmd()` reinitializes the current window (`curwin_init()`) whose
`w_valid` field may have `VALID_BOTLINE` set. Resetting `w_botline`
without marking it as invalid makes subsequent `validate_botline()`
calls a no-op, thus resulting in wrong `line('w$')` value.
closes: vim/vim#14642eb80b8304e
Co-authored-by: Jaehwang Jung <tomtomjhj@gmail.com>
Problem:
vim.iter has both `rfind()` and various `*back()` methods, which work
in "reverse" or "backwards" order. It's inconsistent to have both kinds
of names, and "back" is fairly uncommon (rust) compared to python
(rfind, rstrip, rsplit, …).
Solution:
- Remove `nthback()` and let `nth()` take a negative index.
- Because `rnth()` looks pretty obscure, and because it's intuitive
for a function named `nth()` to take negative indexes.
- Rename `xxback()` methods to `rxx()`.
- This informally groups the "list-iterator" functions under a common
`r` prefix, which helps discoverability.
- Rename `peekback()` to `pop()`, in duality with the existing `peek`.
Fixes regression introduced in #28030
If an LSP server is restarted, then the associated `nvim_buf_attach`
call will not detach if no buffer changes are sent between the client
stopping and a new one being created. This leads to `nvim_buf_attach`
being called multiple times for the same buffer, which then leads to
changetracking sending duplicate requests to the server (one per
attach).
To solve this, introduce separate tracking (client agnostic) on which
buffers have had calls to `nvim_buf_attach`.
