the unix.vim file was probably accidentally ignored at some point.
An actual invokation of nvim-under-test would in practice look like
["/path/to/neovim/build/bin/nvim", "-u", "unix.vim", "-U", "NONE", "-i", "NONE", "--noplugin", "--headless", "-u", "NONE", "--cmd", "set shortmess-=F", "-S", "runtest.vim", "test_arabic.vim"]
but -u NONE cancels out the earlier -u unix.vim
By now, too many tests rely on specific behavior from "NONE", so copy in
the useful parts of unix.vim to the cmdline again. also, some tests
conflict with `directory=.` (or even `directory=Xtempswapdir`) so don't use that.
`-U NONE` is dead code in Nvim, remove it.
Problem: When Neovim is built with Zig, `:checkhealth` falsely reports
"Non-optimized debug build" for release builds. The extraction regex
stops at the first space, and the validation regex only lists CMake
build type names.
Solution: Fix the extraction regex to capture the full build type string
and add Zig optimization levels (ReleaseFast, ReleaseSafe, ReleaseSmall)
to the validation regex.
AI-assisted: Claude Code
Problem: pum_col goes negative when item width + border exceeds screen.
Solution: account for border_width in pum_compute_horizontal_placement()
instead of adjusting pum_col after the fact
Problem:
On exit, rpc_free() is called when processing main_loop.events after
libuv calls close callbacks of the channel's stream. However, when there
are no child processes, these libuv callbacks are called in loop_close()
instead of proc_teardown(), and main_loop.events isn't processed after
loop_close(). As a result, calling remote_ui_disconnect() in rpc_free()
causes UILeave to depend on the presence of child processes.
Solution:
Always call remote_ui_disconnect() in rpc_close_event(), and remove the
call in rpc_free().
Problem: When emitting a msg_show event with the "empty" kind,
there may still be messages waiting to be emitted, which
are then dropped as a result of recursion protection.
Solution: Flush messages before emitting "empty" message show.
Problem: Undefined behavior when 'undodir' contains empty entry.
Solution: Don't try to remove trailing slashes from empty path. Also
don't remove a colon on Windows while at it.
Fixes the following Coverity warning:
*** CID 549779: Integer handling issues (INTEGER_OVERFLOW)
/src/nvim/undo.c: 717 in u_get_undo_file_name()
711 dir_name[dir_len] = NUL;
712
713 // Remove trailing pathseps from directory name
714 char *p = &dir_name[dir_len - 1];
715 while (vim_ispathsep(*p)) {
716 *p-- = NUL;
>>> CID 549779: Integer handling issues (INTEGER_OVERFLOW)
>>> Expression "dir_len--", where "dir_len" is known to be equal to 0, underflows the type of "dir_len--", which is type "size_t".
717 dir_len--;
718 }
719
720 bool has_directory = os_isdir(dir_name);
721 if (!has_directory && *dirp == NUL && !reading) {
722 // Last directory in the list does not exist, create it.
Problem: 'inccommand' preview is not executed after setcmdline(),
and as a result cmdline_show event is emitted when redrawing
is not allowed (5b6477be).
Solution: Call command_line_changed() when ccline.cmdbuff_replaced is
set (by setcmdline()).
Problem: No way of inspecting the (user-added) filetype detection rules.
Solution: Add `vim.filetype.inspect()` returning copies of the internal
`extension`, `filename`, `pattern` tables. Due to the dynamic nature of
filetype detection, this will in general not allow getting the list of
known filetypes, but at least one can see if a given extension is known.
Problem:
We normally get the background color via continuous reporting. However,
if we were backgrounded while the light/dark mode changed, we won't have
received the report, and we'll have the wrong background color.
Without this change, if you background nvim, toggle the light/dark mode,
resume, and check `:set bg`, it will not match the current state.
Solution:
Query it on resume as well. (This requires separating the query from the
flush, to just do the query along with all the others, while waiting to
flush until we've set up uv.)
With this change, if you background nvim, toggle the light/dark mode,
resume, and check `:set bg`, it will have updated.
Problem: `buf` is optional even though its needed to perform conversions
and the ordering of `(buf, row, col)` is not consistent.
Solution: make `buf` mandatory on `vim.range` and `vim.pos` and enforce
the `buf, row, col` ordering
Problem: runtime(tar): some issues with lz4 support
Solution: Fix bugs (see below) (Aaron Burrow)
The tar plugin allows users to extract files from tar archives that are
compressed with lz4. But, tar#Extract() builds malformed extraction commands
for lz4-compressed tar archives. This commit fixes three issues in that code.
The first affects archives with a .tlz4 extension and the other two affect
archives with .tar.lz4 extension (but one of these is symmetric to the issue
that .tlz4 archives had).
(1) When trying to extract .tlz4 archives the command created by
tar#Extract looked like this:
tar -I lz4pxf foo.tlz4 foo
This isn't right. It should be something like this:
tar -I lz4 -pxf foo.tlz4 foo
This was happening because tar.plugin is just substituting on the
first - in "tar -pxf". This works fine if we just add a simple flag for
extraction (eg, z for .tgz), but for lz4 we need to add "-I lz4".
I don't believe that there is an obvious good way to fix this without
reworking the way the command is generated. Probably we should collect
the command and flags separately and the flags should be stored in a
set. Then put everything together into a string just before issuing it
as an extraction command. Unfortunately, this might break things for users
because they have access to tar_extractcmd.
This patch just makes the substitution a little bit more clever so that it
does the right thing when substituting on a string like "tar -pxf".
(2) .tar.lz4 extractions had the same issue, which my patch fixes in
the same way.
(3) .tar.lz4 extractions had another issue. There was a space missing
in the command generated by tar#Extract. This meant that commands
looked like this (notice the lack of space between the archive and output
file names):
tar -I lz4pxf foo.tar.lz4foo
This patch just puts a space where it should be.
Finally, I should note that ChatGPT 5.4 initially identified this issue
in the code and generated the test cases. I reviewed the test cases,
wrote the patch, and actually ran vim against the tests (both with and
without the patch).
closes: vim/vim#1992578954f86c2
Co-authored-by: Aaron Burrow <burrows@fastmail.com>
Problem: When the terminal is very large, test for 9.2.0285 doesn't
trigger an ASAN error without the fix.
Solution: Use a window with fixed height (zeertzjq)
closes: vim/vim#19924b03970f41f
Problem: zip plugin tests may match messages from previous test cases
when checking for warning message.
Solution: Clear messages at the start of these tests (zeertzjq).
closes: vim/vim#19926a1f4259e68
Problem: vim.ui_attach() msg_show callback runs the risk of a recursive
loop_uv_run() when trying to display a message from a shell
command stream.
Solution: Schedule the message callback on the fast_events queue.
Problem: Normal Windows builtin-TUI startup spawns the embedded server as DETACHED_PROCESS, which breaks Ctrl-C delivery to :terminal jobs.
Solution: Restores the default behavior once the embedded server has a
console so terminal jobs inherit it.
Problem: :packadd may lead to heap-buffer-overflow when all entries in
'runtimepath' have the same length (after 9.2.0291).
Solution: Check for comma after current entry properly (zeertzjq).
related: vim/vim#19854
closes: vim/vim#19911bc182ae56e
Problem: too many strlen() calls
Solution: refactor concat_fname() and remove calls to strlen()
(John Marriott)
Function `concat_fnames()` can make up to 5 calls to `STRLEN()` (either
directly or indirectly via `STRCAT()`). In many cases the lengths of
arguments `fname1` and/or `fname2` are either known or can simply be
calculated.
This Commit refactors this function to accept the lengths of arguments
`fname1` and `fname2` as arguments. It also adds new argument `ret` to
return the resulting string as a `string_T`.
Additionally:
- function `add_pack_dir_to_rtp()` in `scriptfile.c`:
Use a `string_T` to store local variables `new_rtp` and `afterdir`.
Replace calls to `STRCAT()` with calls to `STRCPY()`.
Change type of variable `keep` to `size_t` for consistency with
other lengths.
- function `qf_get_fnum()` in `quickfix.c`:
Use a `string_T` to store local variables `ptr` and `bufname`
- function `qf_push_dir()` in `quickfix.c`:
Use a `string_T` to store local variable `dirname`.
Replace call to `vim_strsave()` with `vim_strnsave()`.
- function `qf_guess_filepath()` in `quickfix.c`:
Use a `string_T` to store local variable `fullname`.
- function `make_percent_swname()` in `memline.c`:
Rename some variables to better reflect their use.
Use a `string_T` to store local variables `d` and `fixed_name`.
Slightly refactor to remove need to create an extra string.
- function `get_file_in_dir()` in `memline.c`:
Use a `string_T` to store local variables `tail` and `retval`.
Move some variables closer to where they are used.
- function `cs_resolve_file()` in `if_cscope.c`:
Use a `string_T` to store local variable `csdir`.
Remove one call to `STRLEN()`.
- function `add_pathsep()` in `filepath.c`:
Refactor and remove 1 call to `STRLEN()`
- function `set_init_xdg_rtp()` in `option.c`:
Use a `string_T` to store local variable `vimrc_xdg`.
closes: vim/vim#19854cb51add7ae
Co-authored-by: John Marriott <basilisk@internode.on.net>
Co-authored-by: Christian Brabandt <cb@256bit.org>
Problem: runtime(zip): may write using absolute paths
(syndicate)
Solution: Detect this case and abort on Unix, warn in the documentation
about possible issues
46f530e517
Co-authored-by: Christian Brabandt <cb@256bit.org>
Problem: Commands that rely on Git may need its version to perform more
targeted actions (like decide which arguments are safe to use).
For performance, computing this version is delayed up until it is
needed (like to not compute on regular startup), but not done before
every Git operation (as it is too much and can be done better).
This requires storing the Git version in a variable which is currently
initiated via `vim.version.parse()` call (most probably because it was
easier to handle Lua types this way).
However, the problem is that this results in sourcing `vim.version`
and computing `vim.version.parse` on every startup even if no Git
operation would be done.
Solution: Don't call `vim.version.parse()` during `require('vim.pack')`
and ensure its more precise lazy computation.
Problem:
Running `:Open` on an open buffer does not run `vim.ui.open()` on that file, requiring the user to type `:Open %`. This is inconsistent with most other vim commands which accept files, which default to the current buffer's file.
Solution:
Default to the current file when `:Open` is used without arguments.
Problem:
On Windows, `:!echo тест` shows `????` because the console code page defaults to a legacy ANSI encoding (e.g. CP1252) instead of `UTF-8`
Solution:
Call `SetConsoleOutputCP(CP_UTF8)` and `SetConsoleCP(CP_UTF8)` in `do_os_system()` before spawning child processes, and restore the original values after. It covers both `:!` and `system()` since they both go through `do_os_system()`
Problem:
Opening .tar.gz or .zip URLs shows raw binary instead of using the archive plugins.
Solution:
Similar to the original netrw implementation, the autocmd should detect
archive URLs, download them to a temp file and the open them with
tar/zip handlers already bundled as vim plugins.
Problem:
Apparently vim.SystemCompleted.stdout can also be nil, even without a
custom stdout handler. (Although the docs can be interpreted otherwise).
Solution:
Explicitly check for nil and set the result body to an empty string if
stdout was nil.
Problem: Computing number of threads for parallel asynchronous
computation using `uv.cpu_info()` can be slow. This is especially
noticeable since it is pre-computed on every `require('vim.pack')` and
not only when parallelism is needed.
Solution: Use `uv.available_parallelism()` to compute number of threads
in a helper function.
vim-patch:9.2.0286: still some unnecessary (int) casts in alloc()
vim-patch:9.2.0288: libvterm: signed integer overflow parsing long CSI args
vim-patch:bd8b6c6b0 CI: Bump codecov/codecov-action
vim-patch:9.2.0294: if_lua: lua interface does not work with lua 5.5
vim-patch:9.2.0297: libvterm: can improve CSI overflow code
vim-patch:8.2.0356: MS-Windows: feedkeys() with VIMDLL cannot handle CSI
vim-patch:8.2.0647: MS-Windows: repeat count for events was not used
Problem: When messages are appended to an already expanded cmdline,
the spilled lines indicator is not updated.
Solution: Remove early return for updating virtual text while cmdline is
expanded, guard updating "msg" virt_text at callsite instead.
Problem: Specifying an URL (e.g., Github URL for a PR commit) on the
command line does not work since it will check it against the pinned
checksum.
Solution: Allow overriding `DEPS_IGNORE_SHA` from the command line as
well.
Problem: Crash on exit after closing v:stderr channel when piping
to stdin.
Solution: Reopen stderr as /dev/null or NUL instead of closing it.
This also avoids writing to an related file if one is opened
after closing v:stderr.
