Luuk van Baal [Thu, 16 May 2024 18:44:09 +0000 (20:44 +0200)]
patch 9.1.0414: Unable to leave long line with 'smoothscroll' and 'scrolloff'
Problem: Unable to leave long line with 'smoothscroll' and 'scrolloff'.
Corrupted screen near the end of a long line with 'scrolloff'.
(Ernie Rael, after 9.1.0280)
Solution: Only correct cursor in case scroll_cursor_bot() was not itself
called to make the cursor visible. Avoid adjusting for
'scrolloff' beyond the text line height (Luuk van Baal)
fixes: #14726
closes: #14783
Signed-off-by: Luuk van Baal <luukvbaal@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
jbm950 [Thu, 16 May 2024 17:53:39 +0000 (13:53 -0400)]
runtime(doc): fix typo in vim9script help file (#14782)
This commit fixes a typo in the first example of the vim9script help
file. Trying to execute the given example before resulted in a "trailing
characters" error.
Signed-off-by: jbm950 <jmilam343@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
CI: Upload failed screendump tests when run in github actions
It's a bit of a pain to debug failing screendump tests without knowing
exactly what went wrong. Therefore include actions/upload-artifact for
the Github CI runners and have them uploaded those failing screen dump
tests automatically.
Let's add this step to each of the Linux/MacOS/Windows workflows but do
not duplicate the code, factor it out to a single file
.github/actions/screendump/action.yml and reference this one from the
main ci.yml file
patch 9.1.0413: smoothscroll may cause infinite loop
Problem: smoothscroll may cause infinite loop, with
very narrow windows
(Jaehwang Jung, after v9.1.0280)
Solution: Check for width1 being negative, verify
that win_linetabsize does not overflow
fixes: #14750
closes: #14772
Co-authored-by: zeertzjq <zeertzjq@outlook.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Cthulhux [Wed, 15 May 2024 18:53:52 +0000 (20:53 +0200)]
runtime(doc): update vi_diff.txt: add default value for 'flash'
The "flash" option was added relatively late and seems to be exclusive
to System V. (It's not in 4.4BSD and it's not in V8 UNIX, checked [on
TUHS](https://www.tuhs.org/cgi-bin/utree.pl).) The oldest occurrence of
`flash` in "a vi" I could find is in SysV R2 for the VAX, where it
[defaults to
1](https://github.com/ryanwoodsmall/oldsysv/blob/master/sysvr2-vax/src/cmd/vi/vax/ex_data.c)
= on.
closes: #14778
Signed-off-by: Cthulhux <github@tuxproject.de> Signed-off-by: Christian Brabandt <cb@256bit.org>
Aliaksei Budavei [Mon, 13 May 2024 19:06:05 +0000 (22:06 +0300)]
runtime(syntax-tests): Filter out non-Latin-1 characters for syntax tests (#14767)
Syntax tests are run with the LC_ALL=C environment variable
passed to "make". Occasionally, there are CI failures for
such test files containing non-Latin-1 characters with error
messages pointing to multi-byte characters:
these failures are unrelated to the nature of syntax tests
and should be considered false positives.
As a temporary workaround, all bytes of known non-Latin-1
characters can be replaced in memory with an arbitrary ASCII
byte (?) by applying a filter
> " To ignore part of the dump, provide a "dumps/{filename}.vim" file with
> " Vim commands to be applied to both the reference and the current dump, so
> " that parts that are irrelevant are not used for the comparison. The result
> " is NOT written, thus "term_dumpdiff()" shows the difference anyway.
before lines are compared between files.
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
The current logic is a bit funny, in that it checks for an executable of
scp, then pscp and if neither exists, it uses: scp :/
Anyway, let's fall back to the logic used before the above commit.
John Marriott [Sat, 11 May 2024 22:07:17 +0000 (00:07 +0200)]
patch 9.1.0409: too many strlen() calls in the regexp engine
Problem: too many strlen() calls in the regexp engine
Solution: refactor code to retrieve strlen differently, make use
of bsearch() for getting the character class
(John Marriott)
closes: #14648
Signed-off-by: John Marriott <basilisk@internode.on.net> Signed-off-by: Christian Brabandt <cb@256bit.org>
Tom Picton [Sat, 11 May 2024 18:26:06 +0000 (14:26 -0400)]
ftplugin(python): E16 fix, async keyword support for define (#14751)
This change includes the following changes:
- Fix "E16: Invalid range" when using a count with jump to start/end of class/method
- Update define with optional async keyword
- Update maintainer email
Signed-off-by: Tom Picton <tom@tompicton.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0408: configure fails on Fedora when including perl
Problem: configure fails on Fedora when including perl
(chesheer-smile)
Solution: Filter out -spec=<path> from $LIBS and $LDFLAGS to avoid
linking relocation errors for unrelated autoconf tests.
Luuk van Baal [Sat, 11 May 2024 09:27:52 +0000 (11:27 +0200)]
patch 9.1.0407: Stuck with long line and half-page scrolling
Problem: No scrolling happens with half-page scrolling with line
filling entire window when 'smoothscroll' is disabled.
(Mathias Rav, after v9.1.0285)
Solution: Adjust amount to move cursor by so that it is moved the same
number of lines as was scrolled, even when scrolling different
number of lines than requested with 'nosmoothscroll'.
fixes: #14743
closes: #14746
Signed-off-by: Luuk van Baal <luukvbaal@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Aliaksei Budavei [Fri, 10 May 2024 12:49:06 +0000 (15:49 +0300)]
runtime(java): Stop handpicking syntax groups for @javaTop (#14727)
* runtime(java): Stop handpicking syntax groups for @javaTop
Also:
- Remove the obsolete comment for g:java_allow_cpp_keywords.
- Remove the commented out groups java\%[Debug\]StringError.
- Infer and set the preferred formatting Vim options from
the modeline.
Since vim-6-0u, non-contained syntax groups can be referred
to by using the "contains=TOP..." argument.
* Set &encoding and &termencoding to "utf-8" for test files
* Limit non-ASCII charset to [§ƒɐɘʬʭΑ-Τα-μ] for test files
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
Wu, Zhenyu [Thu, 9 May 2024 18:35:13 +0000 (20:35 +0200)]
patch 9.1.0401: filetype: zsh module files are not recognized
Problem: filetype: zsh module files are not recognized
Solution: Detect '*.mdh' and '*.epro' as C filetype, '*.mdd' as zsh
filetype, determine zsh-modules '*.pro' from from it's content
(Wu, Zhenyu)
closes: #14737
Signed-off-by: Wu, Zhenyu <wuzhenyu@ustc.edu> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0400: Vim9: confusing error message for unknown type
Problem: Vim9: confusing error message for unknown type
(Doug Kearns)
Solution: For an unknown type, display only the type name in the error
message (Yegappan Lakshmanan)
fixes: #13153
closes: #14736
Signed-off-by: Yegappan Lakshmanan <yegappan@yahoo.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0399: block_editing errors out when using del
Problem: block_editing errors out when using del
(@Jamarley)
Solution: Change ins_len from size_t to int and
properly check that it doesn't become negative
There is a check in os.c that verifies that `ins_len` does not become
negative:
```
if (pre_textlen >= 0 && (ins_len = len - pre_textlen - offset) > 0)
```
However this only works, if ins_len can actually become negative and
unfortunately, ins_len has been declared as `size_t` so instead of
becoming negative it will wrap around and be very large.
So let's define it as integer, after which the condition above
properly catches this condition.
patch 9.1.0395: getregionpos() may leak memory on error
Problem: regionpos may leak memory on error, coverity
complains about dereferencing Null pointer
Solution: free all list pointers (after v9.1.394),
return early if buflist_findnr() returns NULL
Github is complaining about "This file has errors" and is just noisy
and for that reason does not ping maintainers if a new issue/PR is
created.
Let's just rename it to MAINTAINERS file instead and update the
documentation.
patch 9.1.0394: Cannot get a list of positions describing a region
Problem: Cannot get a list of positions describing a region
(Justin M. Keyes, after v9.1.0120)
Solution: Add the getregionpos() function
(Shougo Matsushita)
fixes: #14609
closes: #14617
Co-authored-by: Justin M. Keyes <justinkz@gmail.com> Signed-off-by: Shougo Matsushita <Shougo.Matsu@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
D. Ben Knoble [Mon, 6 May 2024 17:52:53 +0000 (19:52 +0200)]
runtime(sh,zsh): clear $MANPAGER in ftplugin before shelling out
Say you use Vim and set MANPAGER='vim -M +MANPAGER --not-a-term -'; then
:{Zs,S}hKeywordPrg (or K) will crap out and spew terminal garbage into
less when bash's "help" fails. This was introduced by 2f25e40b1
(runtime: configure keywordpg for some file types (#5566), 2023-08-23)
and may be present in other files touched by that commit.
Make the "man" invocation sensible by unsetting MANPAGER in the
environment.
Note that changing MANPAGER for `:terminal` is not needed; Vim within
Vim is perfectly fine.
closes: #14679
Signed-off-by: D. Ben Knoble <ben.knoble+github@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0392: tests: Vim9 debug tests may be flaky
Problem: tests: Vim9 debug tests may be flaky
(Shane-XB-Qian)
Solution: Give a few more lines so that line-wrapping won't
cause a hit-enter prompt
The two tests Run_Test_debug_running_out_of_lines() and
Run_Test_debug_with_lambda() test debugging of Vim script functions.
Depending from what file-path the tests are run, it may cause
line wrapping to occur on the following output:
Entering Debug mode. Type "cont" to continue.
command line..script
/home/chrisbra/code/vim-upstream/src/testdir/XdebugFunc[15]..function
<SNR>9_Crash
and if the window is too small, this will cause a hit-enter prompt and so
the WaitForAssert() fails, causing failure of the following tests.
So increase the (internal) Vim window by a few more lines, so that even
if line-wrapping occurs, no hit-enter prompts happens and so the tests
can finish.
Philip H [Wed, 1 May 2024 16:42:11 +0000 (18:42 +0200)]
patch 9.1.0386: filetype: stylus files not recognized
Problem: filetype: stylus files not recognized
Solution: Detect '*.styl' and '*.stylus' as stylus filetype,
include indent, filetype and syntax plugin
(Philip H)
closes: #14656
Signed-off-by: Philip H <47042125+pheiduck@users.noreply.github.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0381: cbuffer and similar commands don't accept a range
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
runtime(java): Improve the recognition of the "indent" method declarations (#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);
}
}
}
------------------------------------------------------------
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.
Signed-off-by: Aliaksei Budavei <0x000c70@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0380: Calculating line height for unnecessary amount of lines
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)
fixes: #14650
closes: #14652
Signed-off-by: Luuk van Baal <luukvbaal@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0378: Vim9: no comments allowed after class vars
Problem: Vim9: no comments allowed after class vars
(Christian Robinson, after 9.1.376)
Solution: Allow trailing comments after class vars
(Yegappan Lakshmanan)
closes: #14651
Signed-off-by: Yegappan Lakshmanan <yegappan@yahoo.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
patch 9.1.0377: Formatting text wrong when 'breakindent' is set
Problem: formatting text wrong when 'breakindent' is set
(Gary Johnson)
Solution: temporarily disable 'breakindent' option when formatting text,
so that the breakindent is not wrongly taken into account for
the line length
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: #14642
Signed-off-by: Jaehwang Jung <tomtomjhj@gmail.com> Signed-off-by: Christian Brabandt <cb@256bit.org>