bug#61893: 29.0.60; Indentation in c-ts-mode broken again

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

Indentation in c-ts-mode on today's emacs-29 branch is in muc worse
shape than it was a few days ago.

To reproduce:

  emacs -Q
  C-x C-f src/dispnew.c RET
  M-x c-ts-mode RET

Go to line 170 and type "C-e RET": point goes to BOL instead of column
2.  Similar problems happen in many other places in the file.  It also
sometimes happens inside a block: for example, go to line 290 and type
"C-e RET".  But in other places within a block RET behaves correctly.

What happened?



In GNU Emacs 29.0.60 (build 363, i686-pc-mingw32) of 2023-03-01 built on
 HOME-C4E4A596F7
Repository revision: 40c9fc8e3b3f55c9122b95e21660b5709109bd1a
Repository branch: emacs-29
Windowing system distributor 'Microsoft Corp.', version 5.1.2600
System Description: Microsoft Windows XP Service Pack 3 (v5.1.0.2600)

Configured using:
 'configure -C --prefix=/d/usr --with-wide-int
 --enable-checking=yes,glyphs 'CFLAGS=-O0 -gdwarf-4 -g3''

Configured features:
ACL GIF GMP GNUTLS HARFBUZZ JPEG JSON LCMS2 LIBXML2 MODULES NOTIFY
W32NOTIFY PDUMPER PNG RSVG SOUND SQLITE3 THREADS TIFF
TOOLKIT_SCROLL_BARS TREE_SITTER WEBP XPM ZLIB

Important settings:
  value of $LANG: ENU
  locale-coding-system: cp1255

Major mode: C/* 

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  show-paren-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  line-number-mode: t
  indent-tabs-mode: t
  transient-mark-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message mailcap yank-media puny dired
dired-loaddefs rfc822 mml mml-sec password-cache epa derived epg rfc6068
epg-config gnus-util text-property-search time-date subr-x mm-decode
mm-bodies mm-encode mail-parse rfc2231 mailabbrev gmm-utils mailheader
sendmail rfc2047 rfc2045 ietf-drums mm-util mail-prsvr mail-utils
c-ts-mode c-ts-common treesit cl-seq vc-git diff-mode easy-mmode vc
vc-dispatcher bug-reference byte-opt gv bytecomp byte-compile cc-mode
cc-fonts cc-guess cc-menus cc-cmds cc-styles cc-align cc-engine cc-vars
cc-defs cl-loaddefs cl-lib rmc iso-transl tooltip cconv eldoc paren
electric uniquify ediff-hook vc-hooks lisp-float-type elisp-mode mwheel
dos-w32 ls-lisp disp-table term/w32-win w32-win w32-vars term/common-win
tool-bar dnd fontset image regexp-opt fringe tabulated-list replace
newcomment text-mode lisp-mode prog-mode register page tab-bar menu-bar
rfn-eshadow isearch easymenu timer select scroll-bar mouse jit-lock
font-lock syntax font-core term/tty-colors frame minibuffer nadvice seq
simple cl-generic indonesian philippine cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese eucjp-ms
cp51932 hebrew greek romanian slovak czech european ethiopic indian
cyrillic chinese composite emoji-zwj charscript charprop case-table
epa-hook jka-cmpr-hook help abbrev obarray oclosure cl-preloaded button
loaddefs theme-loaddefs faces cus-face macroexp files window
text-properties overlay sha1 md5 base64 format env code-pages mule
custom widget keymap hashtable-print-readable backquote threads
w32notify w32 lcms2 multi-tty make-network-process emacs)

Memory information:
((conses 16 81788 11421)
 (symbols 48 9613 0)
 (strings 16 28450 3239)
 (string-bytes 1 916865)
 (vectors 16 15810)
 (vector-slots 8 208511 14555)
 (floats 8 28 75)
 (intervals 40 2215 89)
 (buffers 888 11))

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Theodor Thornhill via Bug reports for GNU Emacs, the Swiss army knife of text editors]

On 1 March 2023 15:37:43 CET, Eli Zaretskii <eliz@gnu.org> wrote:
>Indentation in c-ts-mode on today's emacs-29 branch is in muc worse
>shape than it was a few days ago.
>
>To reproduce:
>
>  emacs -Q
>  C-x C-f src/dispnew.c RET
>  M-x c-ts-mode RET
>


Not sure, don't believe I've touched it last couple of days. I can take a look this evening!

Theo

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

> Date: Wed, 01 Mar 2023 17:35:55 +0100
> From: Theodor Thornhill <theo@thornhill.no>
> CC: Yuan Fu <casouri@gmail.com>
> 
> 
> 
> On 1 March 2023 15:37:43 CET, Eli Zaretskii <eliz@gnu.org> wrote:
> >Indentation in c-ts-mode on today's emacs-29 branch is in muc worse
> >shape than it was a few days ago.
> >
> >To reproduce:
> >
> >  emacs -Q
> >  C-x C-f src/dispnew.c RET
> >  M-x c-ts-mode RET
> >
> 
> 
> Not sure, don't believe I've touched it last couple of days. I can take a look this evening!

Thanks.  It's always possible that it behaved like that back then,
although I'm almost sure it didn't.

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Daniel Martín via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Eli Zaretskii <eliz@gnu.org> writes:

>> Date: Wed, 01 Mar 2023 17:35:55 +0100
>> From: Theodor Thornhill <theo@thornhill.no>
>> CC: Yuan Fu <casouri@gmail.com>
>> 
>> 
>> 
>> On 1 March 2023 15:37:43 CET, Eli Zaretskii <eliz@gnu.org> wrote:
>> >Indentation in c-ts-mode on today's emacs-29 branch is in muc worse
>> >shape than it was a few days ago.
>> >
>> >To reproduce:
>> >
>> >  emacs -Q
>> >  C-x C-f src/dispnew.c RET
>> >  M-x c-ts-mode RET
>> >
>> 
>> 
>> Not sure, don't believe I've touched it last couple of days. I can take a look this evening!
>
> Thanks.  It's always possible that it behaved like that back then,
> although I'm almost sure it didn't.

A git bisect pointed to this commit, in case that helps:

commit 4a25fa4586c9d419ca96441120f4b9189e13ae1c
Author: Yuan Fu <casouri@gmail.com>
Date:   Fri Feb 24 18:40:52 2023 -0800
    Align to prev sibling for c-ts-mode indentation
    
    As requested by a discussion on emacs-devel.
    
    * lisp/progmodes/c-ts-mode.el:
    (c-ts-mode--anchor-prev-sibling): New function.
    (c-ts-mode--indent-styles): Indent first child as before, indent the
    rest to their previous sibling.  Remove bracketless statement rules
    that I forgot to remove previously.
 lisp/progmodes/c-ts-mode.el | 46 ++++++++++++++++++++++++++++++++++-----------
 1 file changed, 35 insertions(+), 11 deletions(-)

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

> From: Daniel Martín <mardani29@yahoo.es>
> Cc: Theodor Thornhill <theo@thornhill.no>,  61893@debbugs.gnu.org,
>   casouri@gmail.com
> Date: Wed, 01 Mar 2023 18:13:37 +0100
> 
> >> Not sure, don't believe I've touched it last couple of days. I can take a look this evening!
> >
> > Thanks.  It's always possible that it behaved like that back then,
> > although I'm almost sure it didn't.
> 
> A git bisect pointed to this commit, in case that helps:

Thanks.  So I wasn't dreaming after all.

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Yuan Fu]

> On Mar 1, 2023, at 10:26 AM, Eli Zaretskii <eliz@gnu.org> wrote:
> 
>> From: Daniel Martín <mardani29@yahoo.es>
>> Cc: Theodor Thornhill <theo@thornhill.no>,  61893@debbugs.gnu.org,
>>  casouri@gmail.com
>> Date: Wed, 01 Mar 2023 18:13:37 +0100
>> 
>>>> Not sure, don't believe I've touched it last couple of days. I can take a look this evening!
>>> 
>>> Thanks.  It's always possible that it behaved like that back then,
>>> although I'm almost sure it didn't.
>> 
>> A git bisect pointed to this commit, in case that helps:
> 
> Thanks.  So I wasn't dreaming after all.

Sorry about that. It should be fixed now.

Yuan

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

> From: Yuan Fu <casouri@gmail.com>
> Date: Wed, 1 Mar 2023 14:08:23 -0800
> Cc: Daniel Martín <mardani29@yahoo.es>,
>  Theodor Thornhill <theo@thornhill.no>,
>  61893@debbugs.gnu.org
> 
> > Thanks.  So I wasn't dreaming after all.
> 
> Sorry about that. It should be fixed now.

Thanks, it's much better now.

I spotted the following problems, not sure if they are new or not: go
to line 297 of dispnew.c, which is "#endif".  Then type "C-e RET" --
point goes to column zero instead of the expected column 6.  This
affects the following 3 lines as well; only the next 'if' fixes the
indentation.  I see the same problem with every "#endif" line.

Line 6584 of dispnew.c, which is this:

#endif /* HAVE_WINDOW_SYSTEM */

exhibits a slightly different incorrect behavior: "C-e RET" on that
line moves point to column 7, not column 6, presumably to align with
the "/*" of the comment?

Line 6608 of dispnew.c is just "#else"; typing "C-e RET" there
unexpectedly goes to column 2 instead of 4.

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Yuan Fu]

> On Mar 1, 2023, at 10:31 PM, Eli Zaretskii <eliz@gnu.org> wrote:
> 
>> From: Yuan Fu <casouri@gmail.com>
>> Date: Wed, 1 Mar 2023 14:08:23 -0800
>> Cc: Daniel Martín <mardani29@yahoo.es>,
>> Theodor Thornhill <theo@thornhill.no>,
>> 61893@debbugs.gnu.org
>> 
>>> Thanks.  So I wasn't dreaming after all.
>> 
>> Sorry about that. It should be fixed now.
> 
> Thanks, it's much better now.
> 
> I spotted the following problems, not sure if they are new or not: go
> to line 297 of dispnew.c, which is "#endif".  Then type "C-e RET" --
> point goes to column zero instead of the expected column 6.  This
> affects the following 3 lines as well; only the next 'if' fixes the
> indentation.  I see the same problem with every "#endif" line.
> 
> Line 6584 of dispnew.c, which is this:
> 
> #endif /* HAVE_WINDOW_SYSTEM */
> 
> exhibits a slightly different incorrect behavior: "C-e RET" on that
> line moves point to column 7, not column 6, presumably to align with
> the "/*" of the comment?
> 
> Line 6608 of dispnew.c is just "#else"; typing "C-e RET" there
> unexpectedly goes to column 2 instead of 4.

That’s indeed the “align to previous statement”’s fault, caused by preproc directives Theo was working on. I’ve fixed it. Theo, while working on this, I found some problem with the current indent for preproc directives. In the following example:

static void
free_glyph_pool (struct glyph_pool *pool)
{
  if (pool)
    {
#if defined GLYPH_DEBUG
      int c = 1;
#endif
      int check_this = 3;
#ifdef stuff
      int c = 1;
#elif defined stuff
  int e = 5;
#else
  int d = 11;
#endif
  int check_this = 3;
    }
}

The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.

Yuan

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Theodor Thornhill via Bug reports for GNU Emacs, the Swiss army knife of text editors]

On 3 March 2023 05:07:44 CET, Yuan Fu <casouri@gmail.com> wrote:
>
>
>> On Mar 1, 2023, at 10:31 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>> 
>>> From: Yuan Fu <casouri@gmail.com>
>>> Date: Wed, 1 Mar 2023 14:08:23 -0800
>>> Cc: Daniel Martín <mardani29@yahoo.es>,
>>> Theodor Thornhill <theo@thornhill.no>,
>>> 61893@debbugs.gnu.org
>>> 
>>>> Thanks.  So I wasn't dreaming after all.
>>> 
>>> Sorry about that. It should be fixed now.
>> 
>> Thanks, it's much better now.
>> 
>> I spotted the following problems, not sure if they are new or not: go
>> to line 297 of dispnew.c, which is "#endif".  Then type "C-e RET" --
>> point goes to column zero instead of the expected column 6.  This
>> affects the following 3 lines as well; only the next 'if' fixes the
>> indentation.  I see the same problem with every "#endif" line.
>> 
>> Line 6584 of dispnew.c, which is this:
>> 
>> #endif /* HAVE_WINDOW_SYSTEM */
>> 
>> exhibits a slightly different incorrect behavior: "C-e RET" on that
>> line moves point to column 7, not column 6, presumably to align with
>> the "/*" of the comment?
>> 
>> Line 6608 of dispnew.c is just "#else"; typing "C-e RET" there
>> unexpectedly goes to column 2 instead of 4.
>
>That’s indeed the “align to previous statement”’s fault, caused by preproc directives Theo was working on. I’ve fixed it. Theo, while working on this, I found some problem with the current indent for preproc directives. In the following example:
>
>static void
>free_glyph_pool (struct glyph_pool *pool)
>{
>  if (pool)
>    {
>#if defined GLYPH_DEBUG
>      int c = 1;
>#endif
>      int check_this = 3;
>#ifdef stuff
>      int c = 1;
>#elif defined stuff
>  int e = 5;
>#else
>  int d = 11;
>#endif
>  int check_this = 3;
>    }
>}
>
>The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.
>
>Yuan

Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)

Theo

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Yuan Fu]

> On Mar 2, 2023, at 10:07 PM, Theodor Thornhill <theo@thornhill.no> wrote:
> 
> 
> 
> On 3 March 2023 05:07:44 CET, Yuan Fu <casouri@gmail.com> wrote:
>> 
>> 
>>> On Mar 1, 2023, at 10:31 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>>> 
>>>> From: Yuan Fu <casouri@gmail.com>
>>>> Date: Wed, 1 Mar 2023 14:08:23 -0800
>>>> Cc: Daniel Martín <mardani29@yahoo.es>,
>>>> Theodor Thornhill <theo@thornhill.no>,
>>>> 61893@debbugs.gnu.org
>>>> 
>>>>> Thanks.  So I wasn't dreaming after all.
>>>> 
>>>> Sorry about that. It should be fixed now.
>>> 
>>> Thanks, it's much better now.
>>> 
>>> I spotted the following problems, not sure if they are new or not: go
>>> to line 297 of dispnew.c, which is "#endif".  Then type "C-e RET" --
>>> point goes to column zero instead of the expected column 6.  This
>>> affects the following 3 lines as well; only the next 'if' fixes the
>>> indentation.  I see the same problem with every "#endif" line.
>>> 
>>> Line 6584 of dispnew.c, which is this:
>>> 
>>> #endif /* HAVE_WINDOW_SYSTEM */
>>> 
>>> exhibits a slightly different incorrect behavior: "C-e RET" on that
>>> line moves point to column 7, not column 6, presumably to align with
>>> the "/*" of the comment?
>>> 
>>> Line 6608 of dispnew.c is just "#else"; typing "C-e RET" there
>>> unexpectedly goes to column 2 instead of 4.
>> 
>> That’s indeed the “align to previous statement”’s fault, caused by preproc directives Theo was working on. I’ve fixed it. Theo, while working on this, I found some problem with the current indent for preproc directives. In the following example:
>> 
>> static void
>> free_glyph_pool (struct glyph_pool *pool)
>> {
>> if (pool)
>>   {
>> #if defined GLYPH_DEBUG
>>     int c = 1;
>> #endif
>>     int check_this = 3;
>> #ifdef stuff
>>     int c = 1;
>> #elif defined stuff
>> int e = 5;
>> #else
>> int d = 11;
>> #endif
>> int check_this = 3;
>>   }
>> }
>> 
>> The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.
>> 
>> Yuan
> 
> Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)

Absolutely, I’ll work on it. 

Yuan

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Theodor Thornhill via Bug reports for GNU Emacs, the Swiss army knife of text editors]

On 3 March 2023 23:01:39 CET, Yuan Fu <casouri@gmail.com> wrote:
>
>
>> On Mar 2, 2023, at 10:07 PM, Theodor Thornhill <theo@thornhill.no> wrote:
>> 
>> 
>> 
>> On 3 March 2023 05:07:44 CET, Yuan Fu <casouri@gmail.com> wrote:
>>> 
>>> 
>>>> On Mar 1, 2023, at 10:31 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>>>> 
>>>>> From: Yuan Fu <casouri@gmail.com>
>>>>> Date: Wed, 1 Mar 2023 14:08:23 -0800
>>>>> Cc: Daniel Martín <mardani29@yahoo.es>,
>>>>> Theodor Thornhill <theo@thornhill.no>,
>>>>> 61893@debbugs.gnu.org
>>>>> 
>>>>>> Thanks.  So I wasn't dreaming after all.
>>>>> 
>>>>> Sorry about that. It should be fixed now.
>>>> 
>>>> Thanks, it's much better now.
>>>> 
>>>> I spotted the following problems, not sure if they are new or not: go
>>>> to line 297 of dispnew.c, which is "#endif".  Then type "C-e RET" --
>>>> point goes to column zero instead of the expected column 6.  This
>>>> affects the following 3 lines as well; only the next 'if' fixes the
>>>> indentation.  I see the same problem with every "#endif" line.
>>>> 
>>>> Line 6584 of dispnew.c, which is this:
>>>> 
>>>> #endif /* HAVE_WINDOW_SYSTEM */
>>>> 
>>>> exhibits a slightly different incorrect behavior: "C-e RET" on that
>>>> line moves point to column 7, not column 6, presumably to align with
>>>> the "/*" of the comment?
>>>> 
>>>> Line 6608 of dispnew.c is just "#else"; typing "C-e RET" there
>>>> unexpectedly goes to column 2 instead of 4.
>>> 
>>> That’s indeed the “align to previous statement”’s fault, caused by preproc directives Theo was working on. I’ve fixed it. Theo, while working on this, I found some problem with the current indent for preproc directives. In the following example:
>>> 
>>> static void
>>> free_glyph_pool (struct glyph_pool *pool)
>>> {
>>> if (pool)
>>>   {
>>> #if defined GLYPH_DEBUG
>>>     int c = 1;
>>> #endif
>>>     int check_this = 3;
>>> #ifdef stuff
>>>     int c = 1;
>>> #elif defined stuff
>>> int e = 5;
>>> #else
>>> int d = 11;
>>> #endif
>>> int check_this = 3;
>>>   }
>>> }
>>> 
>>> The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.
>>> 
>>> Yuan
>> 
>> Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)
>
>Absolutely, I’ll work on it. 
>
>Yuan

Thanks :)

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Yuan Fu]

> On Mar 4, 2023, at 12:54 AM, Theodor Thornhill <theo@thornhill.no> wrote:
> 
> 
> 
> On 3 March 2023 23:01:39 CET, Yuan Fu <casouri@gmail.com> wrote:
>> 
>> 
>>> On Mar 2, 2023, at 10:07 PM, Theodor Thornhill <theo@thornhill.no> wrote:
>>> 
>>> 
>>> 
>>> On 3 March 2023 05:07:44 CET, Yuan Fu <casouri@gmail.com> wrote:
>>>> 
>>>> 
>>>>> On Mar 1, 2023, at 10:31 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>>>>> 
>>>>>> From: Yuan Fu <casouri@gmail.com>
>>>>>> Date: Wed, 1 Mar 2023 14:08:23 -0800
>>>>>> Cc: Daniel Martín <mardani29@yahoo.es>,
>>>>>> Theodor Thornhill <theo@thornhill.no>,
>>>>>> 61893@debbugs.gnu.org
>>>>>> 
>>>>>>> Thanks.  So I wasn't dreaming after all.
>>>>>> 
>>>>>> Sorry about that. It should be fixed now.
>>>>> 
>>>>> Thanks, it's much better now.
>>>>> 
>>>>> I spotted the following problems, not sure if they are new or not: go
>>>>> to line 297 of dispnew.c, which is "#endif".  Then type "C-e RET" --
>>>>> point goes to column zero instead of the expected column 6.  This
>>>>> affects the following 3 lines as well; only the next 'if' fixes the
>>>>> indentation.  I see the same problem with every "#endif" line.
>>>>> 
>>>>> Line 6584 of dispnew.c, which is this:
>>>>> 
>>>>> #endif /* HAVE_WINDOW_SYSTEM */
>>>>> 
>>>>> exhibits a slightly different incorrect behavior: "C-e RET" on that
>>>>> line moves point to column 7, not column 6, presumably to align with
>>>>> the "/*" of the comment?
>>>>> 
>>>>> Line 6608 of dispnew.c is just "#else"; typing "C-e RET" there
>>>>> unexpectedly goes to column 2 instead of 4.
>>>> 
>>>> That’s indeed the “align to previous statement”’s fault, caused by preproc directives Theo was working on. I’ve fixed it. Theo, while working on this, I found some problem with the current indent for preproc directives. In the following example:
>>>> 
>>>> static void
>>>> free_glyph_pool (struct glyph_pool *pool)
>>>> {
>>>> if (pool)
>>>>  {
>>>> #if defined GLYPH_DEBUG
>>>>    int c = 1;
>>>> #endif
>>>>    int check_this = 3;
>>>> #ifdef stuff
>>>>    int c = 1;
>>>> #elif defined stuff
>>>> int e = 5;
>>>> #else
>>>> int d = 11;
>>>> #endif
>>>> int check_this = 3;
>>>>  }
>>>> }
>>>> 
>>>> The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.
>>>> 
>>>> Yuan
>>> 
>>> Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)
>> 
>> Absolutely, I’ll work on it. 
>> 
>> Yuan
> 
> Thanks :)

Done!

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

> From: Yuan Fu <casouri@gmail.com>
> Date: Sat, 4 Mar 2023 15:03:46 -0800
> Cc: Eli Zaretskii <eliz@gnu.org>,
>  Daniel Martín <mardani29@yahoo.es>,
>  61893@debbugs.gnu.org
> 
> >>>> The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.
> >>>> 
> >>>> Yuan
> >>> 
> >>> Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)
> >> 
> >> Absolutely, I’ll work on it. 
> >> 
> >> Yuan
> > 
> > Thanks :)
> 
> Done!

Thanks.

Some problems still remain.

Line 1761 of dispnew.c: type "C-e RET" -- point goes to column 7
instead of the expected column 2.

Line 3381 of dispnew.c: type "C-e RET" -- point goes to the correct
column, but line 3381 is reindented incorrectly.

Line 6611 of dispnew.e: type "C-e RET" -- point goes to column 0
instead of the expected column 2.  If you type "C-e RET" on the next
line 6612, point goes to column 0 and the line is reindented
incorrectly.

Line 13925 of xdisp.c: type "C-e RET" -- point goes to column 33(!)
and line 13925 is reindented incorrectly.

Lines 34529 and 34530 of xdisp.c: "C-e RET" produces incorrect column
of point and reindents line 34530.

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Yuan Fu]

> On Mar 5, 2023, at 12:22 AM, Eli Zaretskii <eliz@gnu.org> wrote:
> 
>> From: Yuan Fu <casouri@gmail.com>
>> Date: Sat, 4 Mar 2023 15:03:46 -0800
>> Cc: Eli Zaretskii <eliz@gnu.org>,
>> Daniel Martín <mardani29@yahoo.es>,
>> 61893@debbugs.gnu.org
>> 
>>>>>> The elif directive’s and subsequent directives aren’t indented properly. Looking at the parsed tree, the elif directives are nested in the if directive, and the nesting can be arbitrarily deep. We probably need a custom function rather than great-grand-parent to find the anchor, like a function that keeps going up the tree until the node isn’t a preproc directive anymore.
>>>>>> 
>>>>>> Yuan
>>>>> 
>>>>> Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)
>>>> 
>>>> Absolutely, I’ll work on it. 
>>>> 
>>>> Yuan
>>> 
>>> Thanks :)
>> 
>> Done!
> 
> Thanks.
> 
> Some problems still remain.

Thanks, there are so many edge cases :-)

> 
> Line 1761 of dispnew.c: type "C-e RET" -- point goes to column 7
> instead of the expected column 2.

Fixed.
> 
> Line 3381 of dispnew.c: type "C-e RET" -- point goes to the correct
> column, but line 3381 is reindented incorrectly.

This is due to it aligning to the previous sibling, aka the comment. Do we want to skip comments for the “align to the prev sibling” rule?

> 
> Line 6611 of dispnew.e: type "C-e RET" -- point goes to column 0
> instead of the expected column 2.  If you type "C-e RET" on the next
> line 6612, point goes to column 0 and the line is reindented
> incorrectly.

Fixed.
> 
> Line 13925 of xdisp.c: type "C-e RET" -- point goes to column 33(!)
> and line 13925 is reindented incorrectly.

Fixed.
> 
> Lines 34529 and 34530 of xdisp.c: "C-e RET" produces incorrect column
> of point and reindents line 34530.

The preproc directive messed the parse tree, and tree-sitter can’t tell that it’s an else statement, so it can’t be easily fixed sadly.

Yuan

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Theodor Thornhill via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Yuan Fu <casouri@gmail.com> writes:

>> On Mar 5, 2023, at 12:22 AM, Eli Zaretskii <eliz@gnu.org> wrote:
>> 
>>> From: Yuan Fu <casouri@gmail.com>
>>> Date: Sat, 4 Mar 2023 15:03:46 -0800
>>> Cc: Eli Zaretskii <eliz@gnu.org>,
>>> Daniel Martín <mardani29@yahoo.es>,
>>> 61893@debbugs.gnu.org
>>> 
>>>>>>> The elif directive’s and subsequent directives aren’t indented
>>>>>>> properly. Looking at the parsed tree, the elif directives are nested in
>>>>>>> the if directive, and the nesting can be arbitrarily deep. We probably
>>>>>>> need a custom function rather than great-grand-parent to find the anchor,
>>>>>>> like a function that keeps going up the tree until the node isn’t a
>>>>>>> preproc directive anymore.
>>>>>>> 
>>>>>>> Yuan
>>>>>> 
>>>>>> Yeah, you're right. I'll see what I can do, but bandwidth is a little low at the moment, so if you have some extra time, don't hesitate to improve it :)
>>>>> 
>>>>> Absolutely, I’ll work on it. 
>>>>> 
>>>>> Yuan
>>>> 
>>>> Thanks :)
>>> 
>>> Done!
>> 
>> Thanks.
>> 
>> Some problems still remain.
>
> Thanks, there are so many edge cases :-)
>
>> 
>> Line 1761 of dispnew.c: type "C-e RET" -- point goes to column 7
>> instead of the expected column 2.
>
> Fixed.
>> 
>> Line 3381 of dispnew.c: type "C-e RET" -- point goes to the correct
>> column, but line 3381 is reindented incorrectly.
>
> This is due to it aligning to the previous sibling, aka the comment. Do we want to skip comments for the “align to the prev sibling” rule?
>
>> 
>> Line 6611 of dispnew.e: type "C-e RET" -- point goes to column 0
>> instead of the expected column 2.  If you type "C-e RET" on the next
>> line 6612, point goes to column 0 and the line is reindented
>> incorrectly.
>
> Fixed.
>> 
>> Line 13925 of xdisp.c: type "C-e RET" -- point goes to column 33(!)
>> and line 13925 is reindented incorrectly.
>
> Fixed.
>> 
>> Lines 34529 and 34530 of xdisp.c: "C-e RET" produces incorrect column
>> of point and reindents line 34530.
>
> The preproc directive messed the parse tree, and tree-sitter can’t tell that it’s an else statement, so it can’t be easily fixed sadly.
>
> Yuan

Now this isn't indenting how I expect:

int foo
{
|      // <-- Now point is indented to column 0
}

It seems like `c-ts-mode--anchor-prev-sibling` isn't handling this?

Theo

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

> From: Theodor Thornhill <theo@thornhill.no>
> Cc: Eli Zaretskii <eliz@gnu.org>,  61893@debbugs.gnu.org,  Daniel
>  Martín
>  <mardani29@yahoo.es>
> Date: Mon, 06 Mar 2023 07:03:33 +0100
> 
> Now this isn't indenting how I expect:
> 
> int foo
> {
> |      // <-- Now point is indented to column 0
> }

Seems to work for me?

bug#61893: 29.0.60; Indentation in c-ts-mode broken again

[Eli Zaretskii]

> From: Yuan Fu <casouri@gmail.com>
> Date: Sun, 5 Mar 2023 15:26:45 -0800
> Cc: Theodor Thornhill <theo@thornhill.no>,
>  Daniel Martín <mardani29@yahoo.es>,
>  61893@debbugs.gnu.org
> 
> > Line 3381 of dispnew.c: type "C-e RET" -- point goes to the correct
> > column, but line 3381 is reindented incorrectly.
> 
> This is due to it aligning to the previous sibling, aka the comment. Do we want to skip comments for the “align to the prev sibling” rule?

No, this behavior is okay.

> > Lines 34529 and 34530 of xdisp.c: "C-e RET" produces incorrect column
> > of point and reindents line 34530.
> 
> The preproc directive messed the parse tree, and tree-sitter can’t tell that it’s an else statement, so it can’t be easily fixed sadly.

I see.

Thanks for the other fixes.  I'm closing this bug.