patch 9.1.0737: tests: screendump tests may require a bit more time
Problem: Asynchronous terminal jobs occassionally require more time
to complete and redraw the window
Solution: increase the sleep value from 10 to 50 milliseconds
(Aliaksei Budavei)
closes: #15695
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
runtime(java): Optionally recognise all primitive constants in _switch-case_ labels
Define "g:java_syntax_previews" and include number 455 in
its list to enable this recognition:
------------------------------------------------------------
let g:java_syntax_previews = [455]
------------------------------------------------------------
Reference:
https://openjdk.org/jeps/455
closes: #15698
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
runtime(java): Highlight javaConceptKind modifiers with StorageClass
Stop assigning by default the NonText highlighting group for
javaConceptKind modifiers since its colour is hardly
distinguishable from a background colour for a range of
colour schemes.
patch 9.1.0731: inconsistent case sensitive extension matching
Problem: inconsistent case sensitive extension matching
Solution: unify case sensitive extension matching (Evgeni Chasnovski).
There are different approaches of how extensions are matched with
respect to case sensitivity. In particular, '\c' flag is used in pattern
whereas in most places case sensitive matching is guarded behind
`has("fname_case")` condition.
Replace all instances of '\c' with an explicit case sensitive pattern
variants guarded by `has("fname_case")`. Strictly speaking, this is a
breaking change because only two (most common and prevailingly tested)
variants are now matched: upper first letter and upper all letters.
closes: #15672
Signed-off-by: Evgeni Chasnovski <evgeni.chasnovski@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0728: [security]: heap-use-after-free in garbage collection with location list user data
Problem: heap-use-after-free in garbage collection with location list
user data.
Solution: Mark user data as in use when no other window is referencing
the location list (zeertzjq)
fixes: neovim/neovim#30371
closes: #15683
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
runtime(java): Recognise all available standard doclet tags
* Complement the tag set with @spec, {@systemProperty},
{@summary}, @provides, @uses, @hidden, and {@index}.
* Do not hoard all tags under a single highlighting group.
* Skip over nested balanced braces in inline tags.
* Observe that tag names are case sensitive: both {@docRoot}
and {@inheritDoc} are valid, whereas {@inheritdoc} and
{@docroot} are not.
* In the @see tag arguments, allow for:
- module name prefixes (e.g. java.base/java.lang.String);
- references to arbitrary URI fragments (e.g. ##foo);
- matching any tag variation arguments on the next line.
* Test directives and tags for Java module declarations.
* Enforce the word end for "module-info" candidates.
patch 9.1.0726: not using correct python3 API with dynamic linking
Problem: not using correct python3 API with dynamic linking
Solution: Use stable IncRef / DecRef in Python 3 dynamic bindings
(Yee Cheng Chin)
Previously, we were using the Py_DECREF macros even when using dynamic
linking of Python 3. This caused issues because Python's headers contain
references to internal APIs like `_Py_Dealloc` and in v8.1.2201 and
v8.2.1225 we simply hacked around the issue by manually copying the
Python header implementation to Vim and linking in the private APIs.
This is fragile and prone to break. In fact, the Py_DECREF
implementation is different in newer versions of Python meaning that
this could potentially cause memory issues.
Instead, simply use the API versions (`Py_DECREF` and `Py_INCREF`) which
are functions exposed by the Python library. They could be slightly
slower since they require a function call instead of a macro, but are
much more reliable and we should only be calling these when the Python
Vim plugins are crossing the language boundary anyway which are always
going to be slow.
Note that this only affects dynamically linked Python builds that are
not using stable ABI.
Also see #15648
closes: #15653
Signed-off-by: Yee Cheng Chin <ychin.git@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
John Tobin [Tue, 10 Sep 2024 18:52:15 +0000 (20:52 +0200)]
runtime(dosini): Update syntax script, spellcheck comments only
By default spell checking is enabled for all text, but adding
`contains=@Spell` to syntax rules restricts spell checking to those
syntax rules. See `:help spell-syntax` for full details.
Variable names and headers are far more likely than comments to contain
spelling errors, so only enable spell checking in comments.
Introduced in https://github.com/xuhdev/syntax-dosini.vim/pull/8
cc @tobinjt
closes: #15655
Signed-off-by: John Tobin <johntobin@johntobin.ie> Signed-off-by: Hong Xu <hong@topbug.net> Signed-off-by: Christian Brabandt <cb@256bit.org>
runtime(doc): Revert outdated comment in completeopt's fuzzy documentation
Originally, `:set completeopt+=fuzzy` did not affect how the candidate
list is collected in default keyword completion. A comment was added to
documentation as part of #14912 to clarify it. #15193 later changed the
fuzzy behavior to now change the candidate collection behavior as well
so the clarification in docs is now wrong. Remove them here.
closes: #15656
Signed-off-by: Yee Cheng Chin <ychin.git@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Groff MOM (Macros for Manuscripts) is a macro package for the GNU
troff (groff) typesetting system, a light-weight alternative
to LaTeX for professional-quality documents.
closes: #15646
Signed-off-by: Konfekt <Konfekt@users.noreply.github.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0724: if_python: link error with python 3.13 and stable ABI
Problem: if_python: link error with python 3.13 and stable ABI
(zdohnal)
Solution: Use the correct stable APIs Py_IncRef and Py_DecRef instead
(Yee Cheng Chin)
This fixes #15460 properly. There was an attempt to fix it in v9.1.0668,
but it did it by manually copy and pasting definitions from Python 3.13
headers, which is problematic as it makes Vim dependent on low-level
implementation details which are subject to change. That change also
pulls in dependencies to private APIs (`_Py_IncRef`) which is a very bad
idea as the next version of Python could very well remove that.
The core issue was simply that `Py_INCREF` and similar functions are not
part of the stable API. We are supposed to be using `Py_IncRef` instead
which performs null-check (similar to `Py_XINCREF`) and is available as
a linkable function. We simply need to call it instead of the macro.
We simply remap `Py_INCREF` (and friends) to the function version in
stable API similar to how we mapped other functions.
related #15460
closes: #15648
Signed-off-by: Yee Cheng Chin <ychin.git@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0723: if_python: dynamic linking fails with python3 >= 3.13
Problem: if_python: dynamic linking fails with python3 >= 3.13
when using non-stable ABI (zdohnal)
Solution: do not try to import the privat python symbol
_PyObject_NextNotImplemented
(Yee Cheng Chin)
Vim is importing a private Python symbol `_PyObject_NextNotImplemented`
because it used to be required as part of the `PyIter_Check()` macro in
an abstraction breaking way. Python eventually fixed the issue and in
3.13 it removed the private symbol export, which broke Vim. Simply
remove importing this private symbol in newer Python versions as it's no
longer needed for PyIter_Check to work.
fixes: #15457
closes: #15649
Signed-off-by: Yee Cheng Chin <ychin.git@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
John M Devin [Sun, 8 Sep 2024 18:00:38 +0000 (20:00 +0200)]
patch 9.1.0721: tests: test_mksession does not consider XDG_CONFIG_HOME
Problem: tests: test_mksession does not consider XDG_CONFIG_HOME
Solution: allow to match $HOME/.vim/ and $HOME/.config/vim for &viewdir
(John M Devin)
closes: #15639
Signed-off-by: John M Devin <john.m.devin@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
GetLatestVimScripts builtin plugin required several fixes:
* Support for the new vimbal `.vmb` extension introduced in
[patch 9.0.1797](https://github.com/vim/vim/commit/f97f6bbf56408c0c97b4ddbe81fba858d7455b0d):
Vimball/Visual Basic filetype detection conflict
* Update the urls from the old `sourceforge.net` to `vim.org`.
The download url was hardcoded and a new variable is introduced.
* Fix `curl` command line option to set a filename (`-O` uses the remote
filename and `-o` specifies a filename).
* Replace windows' command to move files. `REN` can only rename files and
the script actually moves them. My educated guess was that originally
only renaming was necessary. When the script was modified to move
files no change was required on linux because `mv` does both.
* Fix Autoinstall support to check `ftplugins` and `pack-plugins` too
(`pack-plugins` did not exist when the plugin was created).
closes: #15640
Signed-off-by: GuyBrush <miguel.barro@live.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0720: Wrong breakindentopt=list:-1 with multibyte or TABs
Problem: Wrong breakindentopt=list:-1 with multibyte chars or TABs in
text matched by 'formatlistpat' (John M Devin)
Solution: Use the width of the match text (zeertzjq)
fixes: #15634
closes: #15635
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0718: hard to know the users personal Vim Runtime Directory
Problem: hard to guess the Vim Runtime Directory
Solution: Set the $MYVIMDIR environment variable to the users
personal runtime directory (e.g. ~/.vim on Linux)
patch 9.1.0717: Unnecessary nextcmd NULL checks in parse_command_modifiers()
Problem: Unnecessary nextcmd NULL checks in parse_command_modifiers().
Solution: Remove them (zeertzjq)
Every place parse_command_modifiers() is called, nextcmd is NULL, and
after it's set to non-NULL the function returns very soon.
Even if one day nextcmd may be non-NULL, the NULL checks may still be
wrong as the correct behavior may be overriding nextcmd.
closes: #15620
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0715: Not correctly parsing color names (after v9.1.0709)
Problem: Not correctly parsing color names (chdiza, after v9.1.0709)
Solution: Revert part of the patch that compares the color names and
fall-back to the macro STRICMP
Ken Takata [Tue, 3 Sep 2024 21:01:55 +0000 (23:01 +0200)]
patch 9.1.0714: tests: GuiEnter_Turkish test may fail
Problem: tests: GuiEnter_Turkish test may fail
Solution: the message will be translated if the Turkish message file is
installed. Use gettext() to get the expected message
(Ken Takata)
closes: #15615
Signed-off-by: Ken Takata <kentkt@csc.jp> Signed-off-by: Christian Brabandt <cb@256bit.org>
James McCoy [Tue, 3 Sep 2024 20:46:31 +0000 (22:46 +0200)]
patch 9.1.0711: tests: test_xxd may file when using different xxd
Problem: tests: test_xxd may file when using different xxd
Solution: Make Test_xxd_color_0 agnostic to xxd's path, similar to
Test_xxd_color_1 by filtering out shell command prompt in
screen dump file (James McCoy)
If an alternative xxd path is provided, e.g., the system installed xxd,
then the screen dump fails due to the difference in path.
From test_xxd.vim:
Found errors in Test_xxd_color2():
Run 1, 15:17:03 - 15:17:04:
command line..script /tmp/autopkgtest-lxc.1auv5tlk/downtmp/autopkgtest_tmp/vim/src/testdir/runtest.vim[607]..function RunTheTest[57]..Test_xxd_color2[36]..VerifyScreenDump line 67: See dump file difference: call term_dumpdiff("testdir/failed/Test_xxd_color_0.dump", "testdir/dumps/Test_xxd_color_0.dump"); difference in line 1: "|$+0&#ffffff0| |/|u|s|r|/|b|i|n|/|x@1|d| |-|R| |n|e|v|e|r| @1|<| |X@1|D|f|i|l|e|_|c|o|l|o|r|s| @33"
related: #15612
Signed-off-by: James McCoy <jamessan@jamessan.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0710: popup window may hide part of Command line
Problem: when a popup window covers the command line,
the command line is not completely cleared on popup_hide()
(yu3s)
Solution: Check if the popup window covers the command line and if it
does, set the clear_cmdline flag.
- Match -addr and -keepscript attributes and generate -addr values.
- Match attribute errors where = is specified.
- Highlight attributes with Special like other Ex command options.
- Don't highlight user-specified completion function args.
- Match :delcommand -buffer attribute.
closes: #15586
Signed-off-by: Doug Kearns <dougkearns@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0709: GUIEnter event not found in Turkish locale
Problem: GUIEnter not found in Turkish locale
(James McCoy, after v9.1.0256, the issue was there before,
but v9.1.0256 made it more apparent)
Solution: explicitly compare autocommand events by ASCII value and
ignoring locale, because according to the documentation,
events are case insensitive (:h autocommand-events)
runtime(sudoers): improve recognized Runas_Spec and Tag_Spec items
Recognize colon-delimited second part of Runas_Spec that specifies
permitted groups, e.g.:
alan ALL = (root, bin : operator, system) ALL
This implementation is sloppy because it accepts any amount of colons
delimiting further Runas_Lists, but for now that's better than bailing
out completely as soon as a colon is encountered (esp. given that the
default sudoers uses these colons, breaking highlighting OOTB).
Also, while at it, make Vim recognize all Tag_Spec items, not just
{,NO}PASSWD
closes: #15607
Signed-off-by: Ivan Shapovalov <intelfx@intelfx.name> Signed-off-by: Eisuke Kawashima <e-kwsm@users.noreply.github.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0708: Recursive window update does not account for reset skipcol
Problem: Window is updated with potentially invalid skipcol in recursive
window update path. I.e. cursor outside of visible range in
large line that does not fit.
Solution: Make sure it is valid (Luuk van Baal).
closes: #15605
Signed-off-by: Luuk van Baal <luukvbaal@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0707: [security]: invalid cursor position may cause a crash
Problem: [security]: invalid cursor position may cause a crash
(after v9.1.0038)
Solution: Set cursor to the last character in a line, if it would
otherwise point to beyond the line; no tests added, as it
is unclear how to reproduce this.
James McCoy [Sat, 31 Aug 2024 15:10:26 +0000 (17:10 +0200)]
patch 9.1.0706: tests: test_gettext fails when using shadow dir
Problem: tests: test_gettext fails when using shadow dir
Solution: Link the ru_RU directory into the shadow testdir
(James McCoy)
Link the ru_RU directory into shadow testdir
When the ru_RU locale is present, but the build is using a shadowdir,
then test_gettext_cp1251.vim and test_gettext_utf8.vim fail:
From test_gettext_cp1251.vim:
Executed Test_gettext() in 0.000848 seconds
Executed 1 test in 0.007010 seconds
1 FAILED:
Found errors in Test_gettext():
command line..script /home/runner/work/vim/vim/src/shadow/testdir/runtest.vim[607]..function RunTheTest[57]..Test_gettext line 9: Expected '������: ' but got 'ERROR: '
From test_gettext_utf8.vim:
Executed Test_gettext() in 0.000908 seconds
Executed 1 test in 0.007339 seconds
1 FAILED:
Found errors in Test_gettext():
command line..script /home/runner/work/vim/vim/src/shadow/testdir/runtest.vim[607]..function RunTheTest[57]..Test_gettext line 9: Expected '������: ' but got 'ERROR: '
This is because it's unable to load the translations from the ru_RU test
directory, since it wasn't symlinked into the shadow directory.
closes: #15591
Signed-off-by: James McCoy <jamessan@jamessan.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0703: crash with 2byte encoding and glob2regpat()
Problem: possible crash with 2-byte encoding and glob2regpat()
(after v9.1.0700, v9.1.0702)
Solution: include both bytes for a multi-byte character for an
escaped character
patch 9.1.0701: crash with NFA regex engine when searching for composing chars
Problem: crash with NFA regex engine when searching for composing chars
(SuyueGuo)
Solution: When there is no composing character, break out of the loop
and check that out1 state is not null
Aliaksei Budavei [Tue, 27 Aug 2024 20:32:13 +0000 (22:32 +0200)]
runtime(java): Provide support for syntax preview features
Introduce a new API variable "g:java_syntax_previews" whose
value must be a list of syntax preview feature numbers.
Enumerate the currently supported numbers in a table at the
end of the documentation entry for "ft-java-syntax".
Also, disable the recognition of String Templates. Despite
the withdrawal of this preview feature in its proposed form
from the upcoming JDK 23 release and the fact that the JDK
22 release is coming to EOL this September, an earlier
iteration of this preview feature was included in JDK 21
(LTS) whose EOL is projected to fall due in late 2028 and,
therefore, retain the current implementation.
Define "g:java_syntax_previews" and include number 430 in
its list to enable the recognition of String Templates:
------------------------------------------------------------
let g:java_syntax_previews = [430]
------------------------------------------------------------
zeertzjq [Mon, 26 Aug 2024 16:45:37 +0000 (18:45 +0200)]
patch 9.1.0698: tests: "Untitled" file not removed when running Test_crash1_3 alone
Problem: tests: "Untitled" file not removed when running Test_crash1_3 alone
with TEST_FILTER (after v9.1.0695)
Solution: Use a TearDown function instead of another test.
(zeertzjq)
closes: #15578
closes: #15577
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
James McCoy [Sun, 25 Aug 2024 18:22:11 +0000 (20:22 +0200)]
patch 9.1.0696: installing runtime files fails when using SHADOWDIR
Problem: installing runtime files fails when using SHADOWDIR
Solution: revert part of v9.1.0609, since runtime/doc/Makefile's default
value for VIMPROG does not work if vim was built in a SHADOWDIR.
(James McCoy)
closes: #15575
Signed-off-by: James McCoy <jamessan@jamessan.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0691: python3: stable-abi may cause segfault on Python 3.11
Problem: python3: stable-abi may cause segfault on Python 3.11
(Audrius Kažukauskas, after v9.1.0668)
Solution: do not enable the stable Python ABI by default, only when used
with --with-python3-stable-abi argument is given
translation(pt): Revert and fix wrong Portuguese menu translation files
In #14674, the Brazilian and Portuguese localization files were
erroneously combined, even though those are separate locales. This
reverts that change, but maintains some of the original typo fixes in
it.
closes: #15557
related: #14674
Signed-off-by: Yee Cheng Chin <ychin.git@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0689: [security]: buffer-overflow in do_search() with 'rightleft'
Problem: buffer-overflow in do_search() with 'rightleft'
(SuyueGuo)
Solution: after reversing the text (which allocates a new buffer),
re-calculate the text length
Ken Takata [Thu, 22 Aug 2024 19:29:39 +0000 (21:29 +0200)]
runtime(vim): Improve heredoc handling for all embedded scripts
* Improve heredoc handling
- Support "trim" for all the embedded scripts.
- Check the indent of "trim" for "let" and all the embedded scripts.
* Update missing part of vim.vim.base in the commit d164f2a521f8e52e587727657fb1c19e9a25f32a.
* Update gen_syntax_vim.vim to catch up with 9.1.0685's source code.
projects / thirdparty / vim.git / log
