bug#71883: [PATCH] Fix tab-bar-auto-width with customized tab-bar-tab-face-function

[Joseph Turner via "Bug reports for GNU Emacs, the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>]

Juri Linkov <juri@linkov.net> writes:

> Thanks for the patches.
>
>> -   `((,(intern (format "group-%i" i))
>> +   `((,(intern (if current-p "current-group" (format "group-%i" i)))
>
> I pushed this part to master now.
>
>> +(defun tab-bar-auto-width-predicate-default (item)
>> +  "Accepts tab ITEM and returns non-nil for tabs and tab groups."
>> +  (string-match-p
>> +   ;; (rx bos (or "current-tab" "current-group" "tab-" "group-"))
>> +   "\\`\\(?:current-\\(?:group\\|tab\\)\\|\\(?:group\\|tab\\)-\\)"
>> +   (symbol-name (nth 0 item))))
>
> In this part please remove the current group from the default implementation,
> because it looks too ugly, and it was not resized before this change, since
> tab-bar-auto-width-faces didn't contain the tab-bar-tab-group-current face.

If the simpler change offers feature parity, that sounds good to me.

>> -        (if (memq (get-text-property 0 'face (nth 2 item))
>> -                  tab-bar-auto-width-faces)
>> +        (if (run-hook-with-args-until-success 'tab-bar-auto-width-functions item)
>
> I wonder how users are supposed to handle tab-bar-tab-ungrouped now?
> Since it can't be distinguished from grouped tabs (both have tab- symbols),
> is this how users should customize this now:
>
>   (setq tab-bar-auto-width-functions
>         `(,(lambda (item)
>              (and (string-match-p
>                    "\\`\\(?:current-tab\\|\\(?:group\\|tab\\)-\\)" (symbol-name (nth 0 item)))
>                   (not (eq (get-text-property 0 'face (nth 2 item))
>                            'tab-bar-tab-ungrouped))))))

I don't use tab groups, so I don't know if anyone will want to do this.

>> +(make-obsolete-variable 'tab-bar-auto-width-faces 'tab-bar-auto-width-functions "30")
>> [...]
>> +*** The 'tab-bar-auto-width-faces' variable is now obsolete.
>> +Use 'tab-bar-auto-width-functions' instead.
>
> Actually I see no need to obsolete the variable instead of deleting it right away
> since it's not used anywhere anymore.

If it's okay to remove defvars without obsoletion, I'm in favor of it.

Thanks!

Joseph