On Fri, 16 May 2025 20:56:50 +0930 Jake <jforst.mailman@gmail.com> wrote:

> Thanks Steve.
> It respects the fill-column when point is in a comment, though.  Is that
> expected?
>
> Jake

Yes, because emacs-lisp-docstring-fill-column is used only in the
function lisp-fill-paragraph, which handles filling in emacs-lisp-mode
and lisp-interaction-mode and treats doc strings specially but delegates
filling in comments to fill-comment-paragraph, which uses the
buffer-local value of fill-column.

If you agree this is not a bug, you can close this issue by sending
email (e.g. as a reply to this) to 78451-done@debbugs.gnu.org (and not
to 78451@debbugs.gnu.org) (or I can close it if you prefer).

Steve Berman

> On Fri, 16 May 2025 at 6:00 pm, Stephen Berman <stephen.berman@gmx.net>
> wrote:
>
>> On Fri, 16 May 2025 07:47:42 +0000 Jake <jforst.mailman@gmail.com> wrote:
>>
>> > Hello
>> >
>> > `prog-fill-reindent-defun' does not respect the value of fill-column when
>> > inside a string (e.g. a docstring) in emacs-lisp-mode and
>> > lisp-interaction-mode.  I checked it is respected in python-mode, c-mode,
>> > and c++-mode.
>> >
>> > From emacs -Q:
>> > Yank into the *scratch* buffer:
>> >
>> > (defun a ()
>> >   "a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a a
>> a
>> > a a a a a a a a a a a a a a a a"
>> >   nil)
>> >
>> > C-x f 30 RET
>> > Put the point inside the docstring and M-q
>> > Observe that it is filled to something like column 70 to 75.
>>
>> That's because lisp-data-mode (which emacs-lisp-mode and
>> lisp-interaction-mode derive from) binds fill-column to the value of
>> emacs-lisp-docstring-fill-column, which is 72 by default.
>>
>> Steve Berman
>>