cc-mode-help — Bug reports, feature requests, and general talk about CC Mode.

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Mattias E. <mat...@gm...>]

30 apr. 2025 kl. 20.58 skrev Alan Mackenzie <ac...@mu...>:

> DONE.

Thank you!

Re: C23 support in CC Mode.

[Alan M. <ac...@mu...>]

Hello, Arsen.

As half-promised, I've committed a large CC Mode patch which is intended
to cover C23.  It is now in the Emacs master branch at savannah.

On Mon, Dec 23, 2024 at 11:19:37 +0100, Arsen Arsenović wrote:
> Hi Alan,

> Alan Mackenzie <ac...@mu...> writes:

[ .... ]

> >> > BTW, C has recently gained auto type deduction similar to C++

> > Yes.  What is the world coming to?  ;-)

> >> Seems that we forgot about this one!  Whoops.

> >> Could I ask for this one being applied too?

> > Certainly!  There are actually a few new things in C23 I need to attend
> > to.

[ .... ]

> > There's numbers like 1'000'000, for example.  That's already been
> > implemented for C++ Mode, so shouldn't be much trouble, but I need to
> > get down to this and do it systematically.  So give me just a little
> > time here, please.

> No worries!  I didn't mean to hurry you along, sorry!

If you could find time to download the latest master and give the C23
features a run through, that would be great.

Thanks!

> -- 
> Arsen Arsenović

-- 
Alan Mackenzie (Nuremberg, Germany).

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Alan M. <ac...@mu...>]

Hello again, Mattias.

On Mon, Apr 21, 2025 at 20:35:22 +0000, Alan Mackenzie wrote:
> On Thu, Apr 17, 2025 at 15:42:59 +0200, Mattias Engdegård wrote:
> > Thank you for fixing this. A minor oversight in the committed patch, in cc-align.el:

> >  351        (skip-syntax-backward " \t([{" (c-point 'boi))

> > This call should probably use `skip-char-backward`, because `\t`, `[`, and `{` aren't syntax chars.
> > It currently works by accident because of the ` ` and `(` syntax classes.

> Thanks for spotting this and telling me about it.  I'll get around to
> fixing it in the next few days.

DONE.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: bug#77969: [PATCH] Add buttons to manuals in some commentary sections

[Eli Z. <el...@gn...>]

> From: Morgan Smith <Mor...@ou...>
> Date: Mon, 21 Apr 2025 13:54:47 -0400
> 
> Tags: patch
> 
> This patch will affect how a lot of packages get presented to the user.
> 
> Run `M-x package-list-packages'.
> 
> Now go down to 'use-package' and press the button (RET) to pull up the
> package summary.
> 
> At the bottom of that summary it says "See the `use-package' info manual
> for more information.".  The "`use-package'" bit is a button will take
> you to a Help buffer describing the macro.  This is not ideal as the
> sentence is talking about the Info manual.
> 
> After my change the sentence will change too "See Info node
> `(use-package)' for more information." where the "`(use-package)'" part
> is a button that takes you to the info manual.

If we want something like this at the beginning of each Lisp package,
we need to augment our conventions and tips to say that.

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Alan M. <ac...@mu...>]

Hello, Mattias.

On Thu, Apr 17, 2025 at 15:42:59 +0200, Mattias Engdegård wrote:
> Thank you for fixing this. A minor oversight in the committed patch, in cc-align.el:

>  351        (skip-syntax-backward " \t([{" (c-point 'boi))

> This call should probably use `skip-char-backward`, because `\t`, `[`, and `{` aren't syntax chars.
> It currently works by accident because of the ` ` and `(` syntax classes.

Thanks for spotting this and telling me about it.  I'll get around to
fixing it in the next few days.

-- 
Alan Mackenzie (Nuremberg, Germany).

bug#77969: [PATCH] Add buttons to manuals in some commentary sections

[Morgan S. <Mor...@ou...>]

Tags: patch

This patch will affect how a lot of packages get presented to the user.

Run `M-x package-list-packages'.

Now go down to 'use-package' and press the button (RET) to pull up the
package summary.

At the bottom of that summary it says "See the `use-package' info manual
for more information.".  The "`use-package'" bit is a button will take
you to a Help buffer describing the macro.  This is not ideal as the
sentence is talking about the Info manual.

After my change the sentence will change too "See Info node
`(use-package)' for more information." where the "`(use-package)'" part
is a button that takes you to the info manual.

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Mattias E. <mat...@gm...>]

Thank you for fixing this. A minor oversight in the committed patch, in cc-align.el:

 351        (skip-syntax-backward " \t([{" (c-point 'boi))

This call should probably use `skip-char-backward`, because `\t`, `[`, and `{` aren't syntax chars.
It currently works by accident because of the ` ` and `(` syntax classes.

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Stefan K. <ste...@gm...>]

Alan Mackenzie <ac...@mu...> writes:

> I've just committed a fix (a ~700 line patch ;-), so I think this bug is
> finally solved.

Thank you!

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Alan M. <ac...@mu...>]

Hello, Stefan.

On Tue, Mar 11, 2025 at 16:23:13 -0400, Stefan Kangas wrote:
> Alan Mackenzie <ac...@mu...> writes:

> > Yes, I'll do this.  Sorry I never replied to your post in 2020.  I
> > started looking at the bug then, but somehow didn't get very far with it.
> > Maybe it was because there were so many individual glitches in the
> > report.  Thanks for distilling the cases which still remain unfixed.

> Thanks!

I've just committed a fix (a ~700 line patch ;-), so I think this bug is
finally solved.

I'm closing the bug with this post.

-- 
Alan Mackenzie (Nuremberg, Germany).

bug#19867: c++-mode indentation issues with C++1x initializer lists

[Alan M. <ac...@mu...>]

Hello, Stefan.

On Tue, Mar 11, 2025 at 09:54:20 -0700, Stefan Kangas wrote:
> Stefan Kangas <ste...@gm...> writes:

> > Simon <tur...@gm...> writes:

> >> Initializer lists use curly braces, but their contents do not indent properly with emacs' c++-mode.
> >> In short, one may use an initializer list to declare and initialize a vector of integers as such:
> >>   std::vector<int> Foo( { 1, 2, 3, 4, 5 } );

> >> Problems arise when the elements of the list span on multiple line and it gets even worse when the elements are lambda-expressions
> >> and nested initializer lists.
> >> The following code illustrate most cases and related situations.  The code below compiles without error or warning with gcc 4.8.3.

> > I had a look at the fairly long example provided here, and AFAICT, the
> > indentation is incorrect in the below cases (trimmed down from the
> > original).  Some of the examples of incorrect indentation were already
> > fixed.

> > Alan, could you perhaps take a look at this and see if this is something
> > that is fixable?  Thanks in advance.

> Alan, any chance you could look into the below cases?  Thanks in advance.

Yes, I'll do this.  Sorry I never replied to your post in 2020.  I
started looking at the bug then, but somehow didn't get very far with it.
Maybe it was because there were so many individual glitches in the
report.  Thanks for distilling the cases which still remain unfixed.

[ .... ]

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Java var and final keywords [was: Re: [PATCH] Java assert keyword]

[Arsen A. <ar...@aa...>]

Hi Alan,

Alan Mackenzie <ac...@mu...> writes:

> Hello, Arsen.
>
> On Thu, Jan 02, 2025 at 21:09:26 +0100, Arsen Arsenović wrote:
>> Hi Alan,
>
>> Happy new year!
>
> And a HNY to yourself, too!
>
>> ... and sorry to bother you so early in 2025!
>
> No bother at all!
>
>> Arsen Arsenović <cc-...@li...> writes:
>
>> > I was referring to the patch, sorry for not being clear:
>
>> >   https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
>
>> > ... it adds the relatively new 'var' keyword to the Java support.
>
>> Gentle ping on this:
>> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
>
>> It should be another easy patch to merge.
>
> Indeed it was, thanks!  I've committed it to all the usual places, and I
> think I've spelt your name correctly this time.

Thanks!  I just rebuilt Emacs and got it.

And yes, the name is fine this time :-)

Have a lovely day!
-- 
Arsen Arsenović

Java var and final keywords [was: Re: [PATCH] Java assert keyword]

[Alan M. <ac...@mu...>]

Hello, Arsen.

On Thu, Jan 02, 2025 at 21:09:26 +0100, Arsen Arsenović wrote:
> Hi Alan,

> Happy new year!

And a HNY to yourself, too!

> ... and sorry to bother you so early in 2025!

No bother at all!

> Arsen Arsenović <cc-...@li...> writes:

> > I was referring to the patch, sorry for not being clear:

> >   https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

> > ... it adds the relatively new 'var' keyword to the Java support.

> Gentle ping on this:
> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

> It should be another easy patch to merge.

Indeed it was, thanks!  I've committed it to all the usual places, and I
think I've spelt your name correctly this time.

> Thanks in advance.

Thank you for the patch.

> -- 
> Arsen Arsenović

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [PATCH] Java assert keyword

[Alan M. <ac...@mu...>]

Hello, Arsen.

Happy new year to you, too!

On Thu, Jan 02, 2025 at 21:09:26 +0100, Arsen Arsenović wrote:
> Hi Alan,

> Happy new year!

> ... and sorry to bother you so early in 2025!

> Arsen Arsenović <cc-...@li...> writes:

> > I was referring to the patch, sorry for not being clear:

> >   https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

> > ... it adds the relatively new 'var' keyword to the Java support.

> Gentle ping on this:
> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

> It should be another easy patch to merge.

I'll get it seen to, soon.  I'll try to get the commit message correct
this time, too.  ;-)

> Thanks in advance.
> -- 
> Arsen Arsenović

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Hi Alan,

Happy new year!

... and sorry to bother you so early in 2025!

Arsen Arsenović <cc-...@li...> writes:

> I was referring to the patch, sorry for not being clear:
>
>   https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
>
> ... it adds the relatively new 'var' keyword to the Java support.

Gentle ping on this:
https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

It should be another easy patch to merge.

Thanks in advance.
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Hi Alan,

Alan Mackenzie <ac...@mu...> writes:

>> > I just noticed the same about var and, similarly:
>> > https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
>
>> > BTW, C has recently gained auto type deduction similar to C++
>
> Yes.  What is the world coming to?  ;-)
>
>> Seems that we forgot about this one!  Whoops.
>
>> Could I ask for this one being applied too?
>
> Certainly!  There are actually a few new things in C23 I need to attend
> to.

I was referring to the patch, sorry for not being clear:

  https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

... it adds the relatively new 'var' keyword to the Java support.

> There's numbers like 1'000'000, for example.  That's already been
> implemented for C++ Mode, so shouldn't be much trouble, but I need to
> get down to this and do it systematically.  So give me just a little
> time here, please.

No worries!  I didn't mean to hurry you along, sorry!
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Alan M. <ac...@mu...>]

Hello, Arsen.

On Sun, Dec 22, 2024 at 23:55:54 +0100, Arsen Arsenović wrote:
> Hi Alan,

> Arsen Arsenović via CC-Mode-help <cc-...@li...> writes:

> > Arsen Arsenović <ar...@aa...> writes:

> >> Hi Alan!

> >> I've, regrettably, been writing Java lately and I noticed CC-mode
> >> doesn't recognize asserts.  So I added it, hopefully correctly.

> > I just noticed the same about var and, similarly:
> > https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

> > BTW, C has recently gained auto type deduction similar to C++

Yes.  What is the world coming to?  ;-)

> Seems that we forgot about this one!  Whoops.

> Could I ask for this one being applied too?

Certainly!  There are actually a few new things in C23 I need to attend
to.  There's numbers like 1'000'000, for example.  That's already been
implemented for C++ Mode, so shouldn't be much trouble, but I need to
get down to this and do it systematically.  So give me just a little
time here, please.

> Thanks in advance, and sorry for bothering.

No bother at all!

> Have a lovely day!

And yourself, too!

> -- 
> Arsen Arsenović

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Hi Alan,

Arsen Arsenović via CC-Mode-help <cc-...@li...> writes:

> Arsen Arsenović <ar...@aa...> writes:
>
>> Hi Alan!
>>
>> I've, regrettably, been writing Java lately and I noticed CC-mode
>> doesn't recognize asserts.  So I added it, hopefully correctly.
>
> I just noticed the same about var and, similarly:
> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
>
> BTW, C has recently gained auto type deduction similar to C++

Seems that we forgot about this one!  Whoops.

Could I ask for this one being applied too?

Thanks in advance, and sorry for bothering.

Have a lovely day!
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Alan Mackenzie <ac...@mu...> writes:

> Great news, thanks!
>
> I've committed your patch (trivially altered to keep things in
> alphabetical order), and committed it under your name in stand alone CC
> Mode, XEmacs, and the Emacs master branch.

Thanks!  Just rebuilt Emacs, so I got it now.  The keyword is nice and
fontified in my files :)

> Have a great Christmas!

Likewise!
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Alan M. <ac...@mu...>]

Hello, Arsen.

On Sun, Dec 22, 2024 at 15:29:47 +0100, Arsen Arsenović wrote:
> Hi Alan,

> Arsen Arsenović <ar...@aa...> writes:

> > Arsen Arsenović via CC-Mode-help <cc-...@li...> writes:

> >> Arsen Arsenović <ar...@aa...> writes:

> >>> Hi Alan!

> >>> I've, regrettably, been writing Java lately and I noticed CC-mode
> >>> doesn't recognize asserts.  So I added it, hopefully correctly.

> >> I just noticed the same about var and, similarly:
> >> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

> A ping again - the copyright assignment went through.

Great news, thanks!

I've committed your patch (trivially altered to keep things in
alphabetical order), and committed it under your name in stand alone CC
Mode, XEmacs, and the Emacs master branch.

Have a great Christmas!

> -- 
> Arsen Arsenović

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Hi Alan,

Arsen Arsenović <ar...@aa...> writes:

> Arsen Arsenović via CC-Mode-help <cc-...@li...> writes:
>
>> Arsen Arsenović <ar...@aa...> writes:
>>
>>> Hi Alan!
>>>
>>> I've, regrettably, been writing Java lately and I noticed CC-mode
>>> doesn't recognize asserts.  So I added it, hopefully correctly.
>>
>> I just noticed the same about var and, similarly:
>> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e

A ping again - the copyright assignment went through.
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Eli Zaretskii <el...@gn...> writes:

> Thanks, form sent off-list.

Seems to be finally done, at long last.  \o/
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Eli Z. <el...@gn...>]

> From: Arsen Arsenović <ar...@aa...>
> Cc: ac...@mu...,  cc-...@li...
> Date: Sat, 14 Dec 2024 13:23:41 +0100
> 
> Eli Zaretskii <el...@gn...> writes:
> 
> > Thanks, form sent off-list.
> 
> Seems to be finally done, at long last.  \o/

Yes, thanks.

Re: [PATCH] Java assert keyword

[Eli Z. <el...@gn...>]

> From: Arsen Arsenović <ar...@aa...>
> Cc: ac...@mu...,  cc-...@li...
> Date: Sun, 27 Oct 2024 20:43:18 +0100
> 
> >> Eli, to save you some time digging, here are the patches in question:
> >> https://hg.sr.ht/~arsen/cc-mode/rev/15bcae1c6ae0f68c75791f9239879af71c4f237c
> >> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
> >> 
> >> Do these need copyright assignment?  How should I go about it?  TIA
> >
> > They are borderline.  I'd prefer that you sign the copyright
> > assignment, definitely if you intend to contribute to Emacs in the
> > future.
> 
> Works for me.  Do send the form.

Thanks, form sent off-list.

Re: [PATCH] Java assert keyword

[Arsen A. <ar...@aa...>]

Eli Zaretskii <el...@gn...> writes:

>> From: Arsen Arsenović <ar...@aa...>
>> Cc: CC-Mode-help <cc-...@li...>
>> Date: Sun, 27 Oct 2024 18:41:55 +0100
>> 
>> > If not, could I ask you to do that, please?  I've forgotten the exact
>> > procedure, but I can ask Eli Zaretskii (chief Emacs maintainer), who
>> > knows it well.
>> 
>> I have not.
>> 
>> Eli, to save you some time digging, here are the patches in question:
>> https://hg.sr.ht/~arsen/cc-mode/rev/15bcae1c6ae0f68c75791f9239879af71c4f237c
>> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
>> 
>> Do these need copyright assignment?  How should I go about it?  TIA
>
> They are borderline.  I'd prefer that you sign the copyright
> assignment, definitely if you intend to contribute to Emacs in the
> future.

Works for me.  Do send the form.

Thanks.
-- 
Arsen Arsenović

Re: [PATCH] Java assert keyword

[Eli Z. <el...@gn...>]

> From: Arsen Arsenović <ar...@aa...>
> Cc: CC-Mode-help <cc-...@li...>
> Date: Sun, 27 Oct 2024 18:41:55 +0100
> 
> > If not, could I ask you to do that, please?  I've forgotten the exact
> > procedure, but I can ask Eli Zaretskii (chief Emacs maintainer), who
> > knows it well.
> 
> I have not.
> 
> Eli, to save you some time digging, here are the patches in question:
> https://hg.sr.ht/~arsen/cc-mode/rev/15bcae1c6ae0f68c75791f9239879af71c4f237c
> https://hg.sr.ht/~arsen/cc-mode/rev/ce95a9f9caa050602c107f61dbe933e31338674e
> 
> Do these need copyright assignment?  How should I go about it?  TIA

They are borderline.  I'd prefer that you sign the copyright
assignment, definitely if you intend to contribute to Emacs in the
future.

> > Before saying yes, please bear in mind that CC Mode is being
> > superseded by the Tree Sitter modes (like c-ts-mode and c++-ts-mode)
> > and its future as a prime set of Emacs mode is somewhat in doubt.
> 
> This surprises me.
> 
> To be clear, I'm not surprised by tree-sitter superseding a _parser_,
> but a whole mode is not a just a parser, nor just a highlighting engine.
> 
> I primarily don't use the -ts- mode (as someone who is a fan of
> Tree-sitter) because their implementations of other parts of what
> constitutes a mode are not as good (the indentation engines are worse,
> handling of comments is worse, probably some other bits).
> 
> I expected that, eventually, the "old-style" modes would start using
> tree-sitter information to do indentation.

>From my POV, the TS mode do not supersede the "traditional" modes for
the same languages, they are alternative implementations.  I envision
both to coexist in Emacs for many years to come, and in that time each
mode learning from its sibling and improving.  The main advantage of a
TS based mode is that we (the Emacs developers) don't need to have
experts for the language parsing and don't need to play catch-up will
all the latest developments of the language.  But it also has
disadvantages, and generally suffers from being still very new and
lacking some important features.