Aliaksei Budavei [Mon, 19 Aug 2024 19:33:26 +0000 (21:33 +0200)]
patch 9.1.0681: tests: Analyzing failed screendumps is hard
Problem: tests: Analyzing failed screendumps is hard
Solution: Facilitate the viewing of rendered screendumps under src/
add some documentation on how to use the viewdumps.vim
script (Aliaksei Budavei)
With the submitted "viewdumps.vim" script, a few manual
steps in typical workflows (see below) can be automated.
The updated "README.txt" contains additional information.
Reviewing LOCAL failed screendump tests can be arranged as
follows:
1) Run tests and generate screendumps:
------------------------------------------------------------
cd /path/to/fork/src/testdir
make
------------------------------------------------------------
2) Examine the screendumps from the "failed" directory:
------------------------------------------------------------
../vim -u NONE -S viewdumps.vim
------------------------------------------------------------
3) Set up the "dumps" directory. Create a symlink to
"/path/to/fork/dirs/dumps" in the extracted directories so
that term_dumpdiff() can be used. (The lookup algorithm
resolves "dumps" for every loaded filename. So, with
"/tmp/src/testdir/failed/*.dump" files passed as script
arguments, the algorithm will make the files in
"/tmp/src/testdir/dumps" queried.)
------------------------------------------------------------
cd /path/to/fork
ln -s $(pwd)/src/testdir/dumps /tmp/src/testdir/dumps
------------------------------------------------------------
Reviewing SUBMITTED FOR PULL REQUEST screendump tests can be
arranged as follows (it can be further locally scripted):
1) List the fetched changeset and write the changed "dumps"
filenames to "/tmp/filelist":
------------------------------------------------------------
cd /path/to/fork
git switch prs/1234
git diff-index --relative=src/testdir/dumps/ \
--name-only prs/1234~1 > /tmp/filelist
------------------------------------------------------------
2) Reconcile relative filepaths, and copy next-to-be-updated
"dumps" files in the "failed" directory (note the missing
new screendumps, if any):
------------------------------------------------------------
git switch master
cd src/testdir/dumps
test -d ../failed || mkdir ../failed
cp -t ../failed $(cat /tmp/filelist)
------------------------------------------------------------
3) Remember about the introduced INVERTED relation between
"dumps" and "failed", i.e. the files to be committed are in
"dumps" already and their previous versions are in "failed";
therefore, copy the missing new screendumps from "dumps" to
"failed" (otherwise these won't be shown):
------------------------------------------------------------
git switch prs/1234
cp -t ../failed foo_10.dump foo_11.dump foo_12.dump
------------------------------------------------------------
4) Examine the screendumps from the "failed" directory (new
screendumps will be shown with no difference between their
versions):
------------------------------------------------------------
cd ..
../vim -u NONE -S viewdumps.vim
------------------------------------------------------------
closes: #15515
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Otherwise, if the executable to be verified does not exist,
this would cause a false-positive in the 'IsSafeExecutable()' check,
because 'exepath(executable)' returns an empty string and
'fnamemodify('', ':p:h')' returns the current directory and as a result
the 'IsSafeExecutable()' returns false (for the wrong reason).
zeertzjq [Fri, 16 Aug 2024 19:11:31 +0000 (21:11 +0200)]
patch 9.1.0679: Rename from w_closing to w_locked is incomplete
Problem: Rename from w_closing to w_locked is incomplete
(after 9.1.0678).
Solution: Rename remaining occurrences of w_closing to w_locked and
update comments (zeertzjq).
closes: #15504
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Maxim Kim [Fri, 16 Aug 2024 19:09:50 +0000 (21:09 +0200)]
runtime(colors): update colorschemes
- all: PMenuMatch and PMenuMatchSel for 8c/16c
- habamax:
- revert VertSplit to solid background color
- remove gitCommitSummary link to Title
- make TabLineFill same as StatuslineNC
closes: #15506
Signed-off-by: Maxim Kim <habamax@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
The end marker must appear on line of its own without any trailing
whitespace.
Whitespace is incorrectly allowed before all end markers. Limiting this
only to heredocs where "trim" was specified, and with the correct
indent, is currently an intractable problem given that contained syntax
groups (in this case :let) cannot be limited to start patterns.
Highlight interpolated expressions when "eval" is specified.
cloess: #15511
Signed-off-by: Doug Kearns <dougkearns@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
runtime(yaml): do not re-indent when commenting out lines
It's a personal annoyance for me. I have to edit yaml files on a lot of
customer environments and whenever you type '#' at the start of the
line, the commented line will be indented by whatever indent the
previous line had.
I hate this seriously, because it makes un-commenting painful. So let's
fix this. But instead of messing with the indent function, let's just
remove the '0#' from cinkeys, so that Vim won't perform re-indenting
when commenting out such a yaml file.
patch 9.1.0678: [security]: use-after-free in alist_add()
Problem: [security]: use-after-free in alist_add()
(SuyueGuo)
Solution: Lock the current window, so that the reference to
the argument list remains valid.
Gregory Anders [Thu, 15 Aug 2024 20:04:22 +0000 (22:04 +0200)]
patch 9.1.0677: :keepp does not retain the substitute pattern
Problem: :keeppatterns does not retain the substitute pattern
for a :s command
Solution: preserve the last substitute pattern when used with the
:keeppatterns command modifier (Gregory Anders)
closes: #15497
Signed-off-by: Gregory Anders <greg@gpanders.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Ken Takata [Wed, 14 Aug 2024 19:56:42 +0000 (21:56 +0200)]
runtime(dosbatch): Add syn-sync to syntax file
Closing parentheses were often highlighted as errors.
Add a syntax sync command to reduce the error.
Also fix that `defined` was not highlighted as an operator inside
parentheses. E.g.:
```
if defined foo (
if defined bar (
...
)
)
```
The first `defined` was highlighted but the second one was not.
related: #15453
Signed-off-by: Ken Takata <kentkt@csc.jp> Signed-off-by: Christian Brabandt <cb@256bit.org>
Yinzuo Jiang [Wed, 14 Aug 2024 19:49:00 +0000 (21:49 +0200)]
runtime(sql, mysql): fix E169: Command too recursive with sql_type_default = "mysql"
Problem: When setting "let g:sql_type_default = "mysql", editing .sql
file reports "E169: Command too recursive" error
Solution:
- Add 'let b:did_ftplugin = 1' at the top of ftplugin/sql.vim
- Add 'if exists("b:did_ftplugin") | finish | endif' in ftplugin/mysql.vim
- Add missing header information in ftplugin/mysql.vim
- Remove redundant code in ftplugin/sql.vim
Ernie Rael [Wed, 14 Aug 2024 12:53:55 +0000 (14:53 +0200)]
patch 9.1.0674: Vim9: compiling abstract method fails because of missing return
Problem: Vim9: compiling abstract method fails because of missing
return (Aliaksei Budavei)
Solution: don't require a return statement for an abstract method when
compiling (Ernie Rael)
fixes: #15432
closes: #15441
Signed-off-by: Ernie Rael <errael@raelity.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Ernie Rael [Tue, 13 Aug 2024 21:27:22 +0000 (23:27 +0200)]
patch 9.1.0673: Vim9: too recursive func calls when calling super-class method
Problem: Vim9: too recursive func calls when calling super-class method
with non-overriden super-call methods. (Aliaksei Budavei)
Solution: use interface method, when super is to be used (Ernie Rael)
When compiling "super.Func()" force class context to class that defines
function that is doing "super.Func()".
ISN_METHODCALL arg "cmf_is_super" for specific ufunc.
fixes: #15448
fixes: #15463 (2) super.method may not execute in context of defining
class
closes: #15477
Signed-off-by: Ernie Rael <errael@raelity.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Aliaksei Budavei [Mon, 12 Aug 2024 16:37:15 +0000 (18:37 +0200)]
runtime(syntax-tests): Facilitate the viewing of rendered screendumps
With the submitted "viewdumps.vim" script, a few manual
steps in typical workflows (see below) can be automated.
The updated "README.txt" contains additional information.
Reviewing LOCAL failed syntax tests can be arranged as
follows:
1) Run tests and generate screendumps:
------------------------------------------------------------
cd /path/to/fork/runtime/syntax
make clean test
------------------------------------------------------------
2) Examine the screendumps from the "failed" directory:
------------------------------------------------------------
../../src/vim --clean -S testdir/viewdumps.vim
------------------------------------------------------------
3) Set up the "dumps" directory. Create a symlink to
"/path/to/fork/dirs/dumps" in the extracted directories so
that term_dumpdiff() can be used. (The lookup algorithm
resolves "dumps" for every loaded filename. So, with
"/tmp/runtime/syntax/testdir/failed/*.dump" files passed
as script arguments, the algorithm will make the files in
"/tmp/runtime/syntax/testdir/dumps" queried.)
------------------------------------------------------------
cd /path/to/fork
ln -s $(pwd)/runtime/syntax/testdir/dumps \
/tmp/runtime/syntax/testdir/dumps
------------------------------------------------------------
Reviewing SUBMITTED FOR PULL REQUEST syntax tests can be
arranged as follows (it can be further locally scripted):
1) List the fetched changeset and write the changed "dumps"
filenames to "/tmp/filelist":
------------------------------------------------------------
cd /path/to/fork
git switch prs/1234
git diff-index --relative=runtime/syntax/testdir/dumps/ \
--name-only prs/1234~1 > /tmp/filelist
------------------------------------------------------------
2) Reconcile relative filepaths, and copy next-to-be-updated
"dumps" files in the "failed" directory (note the missing
new screendumps, if any):
------------------------------------------------------------
git switch master
cd runtime/syntax/testdir/dumps
cp -t ../failed $(cat /tmp/filelist)
------------------------------------------------------------
3) Remember about the introduced INVERTED relation between
"dumps" and "failed", i.e. the files to be committed are in
"dumps" already and their previous versions are in "failed";
therefore, copy the missing new screendumps from "dumps" to
"failed" (otherwise these won't be shown):
------------------------------------------------------------
git switch prs/1234
cp -t ../failed foo_10.dump foo_11.dump foo_12.dump
------------------------------------------------------------
4) Examine the screendumps from the "failed" directory (new
screendumps will be shown with no difference between their
versions):
------------------------------------------------------------
cd ..
../../../src/vim --clean -S viewdumps.vim
------------------------------------------------------------
closes: #15476
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0672: marker folds may get corrupted on undo
Problem: marker folds may get corrupted on undo (Yousef Mohammed)
Solution: when adjusting folds, make sure that line1 is the lower limit
and line2 is the upper line limit. In particular, line2 should
not be able to get smaller than line1.
RestorerZ [Sun, 11 Aug 2024 18:03:35 +0000 (20:03 +0200)]
patch 9.1.0670: po file encoding fails on *BSD during make
Problem: po file encoding fails on *BSD during make
Solution: instead of using `$<` make use of variable `$?` which should
be equivalent when the rule is dependent on only a single
file (RestorerZ).
closes: #15471
Signed-off-by: RestorerZ <restorer@mail2k.ru> Signed-off-by: Christian Brabandt <cb@256bit.org>
Ken Takata [Sun, 11 Aug 2024 16:41:41 +0000 (18:41 +0200)]
patch 9.1.0669: if_python: stable python ABI not used by default
Problem: stable python ABI not used by default
Solution: Enable stable python ABI v3.8 when building with python3/dyn
by default, update the default Python3 version to 3.8 for
MS-Windows (Ken Takata)
closes: #15470
related: #15457
Signed-off-by: Ken Takata <kentkt@csc.jp> Signed-off-by: Christian Brabandt <cb@256bit.org>
zeertzjq [Sat, 10 Aug 2024 07:35:20 +0000 (09:35 +0200)]
patch 9.1.0667: Some other options reset curswant unnecessarily when set
Problem: Some other options reset curswant unnecessarily when set.
(Andrew Haust)
Solution: Don't reset curswant when setting 'comments', 'commentstring'
or 'define' (zeertzjq)
fixes: #15462
closes: #15467
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
zeertzjq [Thu, 8 Aug 2024 19:12:15 +0000 (21:12 +0200)]
runtime(doc): clarify directory of Vim's executable vs CWD
According to :h win32-PATH, "the same directory as Vim" means the same
directory as the Vim executable, not Vim's current directory. In patch
8.2.4860 these two concepts were mixed up.
closes: #15451
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
zeertzjq [Thu, 8 Aug 2024 19:05:57 +0000 (21:05 +0200)]
patch 9.1.0665: Locked variable can be changed in a :for loop
Problem: Locked variable can be changed in a :for loop.
Solution: Always do a full permission check on the first loop iteration
where ASSIGN_DECL is not set (zeertzjq).
related: #12470
fixes: #15450
closes: #15454
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0663: tests: zip test still resets 'shellslash' option
Problem: tests: zip test still resets 'shellslash' option
Solution: Remove resetting the 'shellslash' option, the zip
plugin should now be able to handle this options
Problem: the zip plugin uses features of Vim 9.0, but
checks for a Vim version >= 7.2
(interpolated strings and readblob() function)
Solution: raise the minimum Vim version to v9.0
runtime(zip): refactor save and restore of options
Problem: zip plugin has no way to set/restore option values
Solution: Add the SetSaneOpts() and RestoreOpts() functions,
so options that cause issues are set to sane values
and restored back to their initial values later on.
(this affects the 'shellslash' option on windows, which also
changes how the shellescape() function works)
Problem: zip plugin tests for fnameescape() function
Solution: Remove the check, fnameescape() has been available since
7.1.299, it should nowadays always be available
Problem: zip plugin uses :echo which does not store messages
Solution: use :echomsg instead of :echo so that messages are stored in
the message history
patch 9.1.0660: MS-Windows: Shift-Insert does work on old conhost
Problem: MS-Windows: Shift-Insert does not work on old conhost
(Nick Jensen, after 9.0.1146)
Solution: handle Shift-Insert specifically
(Christian Plewright)
fixes: #15326
closes: #15430
Signed-off-by: Christopher Plewright <chris@createng.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
zeertzjq [Sun, 4 Aug 2024 16:35:50 +0000 (18:35 +0200)]
runtime(zip): escape '[' on Unix as well
Problem: After 6f1cbfc9ab483a09877e153ad130164875c40b1d fnameescape()
is no longer called on the name of the file to be extracted.
However, while spaces indeed don't need to be escaped, unzip
treats '[' as a wildcard character, so it need to be escaped.
Solution: Escape '[' on both MS-Windows and Unix.
From the docs it seems '*' and '?' also need escaping, but they seem to
actually work without escaping.
fixes: neovim/neovim#29977
closes: #15427
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Ken Takata [Sat, 3 Aug 2024 12:40:35 +0000 (14:40 +0200)]
patch 9.1.0657: MS-Windows: MSVC build time can be optimized
Problem: MS-Windows: MSVC build time can be optimized
Solution: Stop using Powershell, show detected version, read _MSC_VER
and _MSC_FULL_VER in a single cl execution. (Ken Takata)
closes: #15421
Signed-off-by: Ken Takata <kentkt@csc.jp> Signed-off-by: Christian Brabandt <cb@256bit.org>
Ken Takata [Sat, 3 Aug 2024 12:34:33 +0000 (14:34 +0200)]
patch 9.1.0656: MS-Windows: MSVC Makefile CPU handling can be improved
Problem: MS-Windows: MSVC Makefile CPU handling can be improved
Solution: Update Makefile and also remove some old code
(Ken Takata)
* The variable `ASSEMBLY_ARCHITECTURE` was originally used to generate
the manifest file. However, this was removed in v7.0.198. Remove
`ASSEMBLY_ARCHITECTURE` and use `CPU` instead.
* `/arch:AVX512` is available starting from Visual C++ 2017.
* ARM64 was not considered when validating CPUNR.
Skip the validation for ARM64 for now.
* Remove old code for VC6/7 and Win9x.
closes: #15422
Signed-off-by: Ken Takata <kentkt@csc.jp> Signed-off-by: Christian Brabandt <cb@256bit.org>
Adam Monsen [Fri, 2 Aug 2024 17:54:15 +0000 (19:54 +0200)]
patch 9.1.0655: filetype: goaccess config file not recognized
Problem: filetype: goaccess config file not recognized
Solution: detect 'goaccess.conf' as goaccess filetype, also
include a basic syntax and ftplugin (Adam Monsen)
Add syntax highlighting for GoAccess configuration file.
GoAccess is a real-time web log analyzer and interactive viewer that
runs in a terminal in *nix systems or through your browser.
GoAccess home page: https://goaccess.io
closes: #15414
Signed-off-by: Adam Monsen <haircut@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Ivan Shapovalov [Fri, 2 Aug 2024 17:43:12 +0000 (19:43 +0200)]
runtime(netrw): honor `g:netrw_alt{o,v}` for `:{S,H,V}explore`
Make `:Sexplore` / `:Hexplore` / `:Vexplore` commands honor the user
`&split{right,below}` settings (or netrw-specific `g:netrw_alt{o,v}`)
instead of hardcoding a split direction. Similarly, update banged
variants of the two latter commands to follow the inverted preference.
closes: #15417
Signed-off-by: Ivan Shapovalov <intelfx@intelfx.name> Signed-off-by: Christian Brabandt <cb@256bit.org>
glepnir [Fri, 2 Aug 2024 17:14:38 +0000 (19:14 +0200)]
patch 9.1.0654: completion does not respect completeslash with fuzzy
Problem: completion does not respect completeslash with fuzzy
(egesip)
Solution: Change path separator on Windows, depending on 'completeslash'
option value (glepnir)
fixes: #15392
closes: #15418
Signed-off-by: glepnir <glephunter@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
John Marriott [Thu, 1 Aug 2024 21:01:25 +0000 (23:01 +0200)]
patch 9.1.0652: too many strlen() calls in syntax.c
Problem: too many strlen() calls in syntax.c
Solution: refactor code to reduce the number or strlen() calls,
get rid of un-used SYN_NAMELEN macro
(John Marriott)
closes: #15368
Signed-off-by: John Marriott <basilisk@internode.on.net> Signed-off-by: Christian Brabandt <cb@256bit.org>
zeertzjq [Thu, 1 Aug 2024 20:48:53 +0000 (22:48 +0200)]
patch 9.1.0650: Coverity warning in cstrncmp()
Problem: Coverity warning in cstrncmp()
(after v9.1.0645)
Solution: Change the type of n2 to int.
(zeertzjq)
________________________________________________________________________________________________________
*** CID 1615684: Integer handling issues (INTEGER_OVERFLOW)
/src/regexp.c: 1757 in cstrncmp()
1751 n1 -= mb_ptr2len(s1);
1752 MB_PTR_ADV(p);
1753 n2++;
1754 }
1755 // count the number of bytes to advance the same number of chars for s2
1756 p = s2;
>>> CID 1615684: Integer handling issues (INTEGER_OVERFLOW)
>>> Expression "n2--", which is equal to 18446744073709551615, where "n2" is known to be equal to 0, underflows the type that receives it, an unsigned integer 64 bits wide.
1757 while (n2-- > 0 && *p != NUL)
1758 MB_PTR_ADV(p);
1759
1760 n2 = p - s2;
1761
1762 result = MB_STRNICMP2(s1, s2, *n, n2);
closes: #15409
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
zeertzjq [Thu, 1 Aug 2024 20:46:54 +0000 (22:46 +0200)]
patch 9.1.0649: Wrong comment for "len" argument of call_simple_func()
Problem: Wrong comment for "len" argument of call_simple_func().
Solution: Remove the "or -1 to use strlen()". Also change its type to
size_t to remove one cast. (zeertzjq)
closes: #15410
Signed-off-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0648: [security] double-free in dialog_changed()
Problem: [security] double-free in dialog_changed()
(SuyueGuo)
Solution: Only clear pointer b_sfname pointer, if it is different
than the b_ffname pointer. Don't try to free b_fname,
set it to NULL instead.
patch 9.1.0647: [security] use-after-free in tagstack_clear_entry
Problem: [security] use-after-free in tagstack_clear_entry
(Suyue Guo )
Solution: Instead of manually calling vim_free() on each of the tagstack
entries, let's use tagstack_clear_entry(), which will
also free the stack, but using the VIM_CLEAR macro,
which prevents a use-after-free by setting those pointers
to NULL
- Prefix all global variables with "g:".
- Add spaces around each variable assignment operator.
- Remove extraneous whitespace characters.
- Remove a spurious _serializable_ Java keyword (since v1.1,
java.io.Serializable and java.io.Externalizable interfaces
provide an API for object serialization; see vim-6-0j).
- Normalise the syntax definition argument order by making
_contained_ the first argument of each such definition.
- Normalise the argument tabulation for highlighting group
definitions.
runtime(java): Tidy up the documentation for "ft-java-syntax"
- Reword a few sentences and reformat a few paragraphs.
- Supply absent capitalisation and punctuation.
- Make listed highlighting groups and code stand out.
- Prefix all Java-related global variables with "g:".
- Add spaces around each variable assignment operator.
- Acknowledge that some Javadoc variables are maintained in
the HTML syntax file.
Also, move the overridable _default_ HTML group links before
the HTML syntax file inclusion in order to implement the
documented diverged settings.
related: #15399
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
- Make diff colors more accessible, Green for added, Red for deleted, Blue for Changed
- Change Search to blue to be visible with Diff colors
- Change Todo to bright magenta
closes: #15400
Signed-off-by: Maxim Kim <habamax@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
projects / thirdparty / vim.git / log
