bug#66187: read-file-name unexpected behavior when MUSTMATCH is a function

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: joseph@breatheoutbreathe.in,  michael_heerdegen@web.de,
>   philipk@posteo.net,  66187@debbugs.gnu.org
> Date: Fri, 06 Oct 2023 09:01:23 -0400
> 
> >> >> For example, given:
> >> >> 
> >> >> (completing-read "Prompt: " '("a" "b") nil
> >> >>                  (lambda (input)
> >> >>                    (string= "a" input)))
> >> >> 
> >> >> I expected that the prompt would refuse to complete "b". I was wrong.
> >> >> 
> >> >> Since completing-read is such a fundamental part of Emacs, I doubt it
> >> >> will be possible to change this behavior.  What do you think about the
> >> >> attached patch to clarify the completing-read docstring?
> >> 
> >> The use of a function as `require-match` is brand new in Emacs-29, so
> >> I think it's not too late to fix it.  I think rather than fixing the doc
> >> we should fix the behavior, e.g. with the patch below.
> >
> > What change in behavior (wrt Emacs 29.1) does this patch cause?
> 
> In the example he gave above it makes it so completion can be used to
> insert "b" but when you try to exit it refuses to exit.
> 
> IOW it gives primacy to the `require-match` function over the
> completion-table w.r.t deciding when the minibuffer's content is
> acceptable or not.
> 
> Code which wants the 29.1 behavior can easily get it by modifying their
> `require-match` function accordingly (and such code then works
> correctly both with Emacs-29.1 and with the new behavior).

Do we have any users of this feature in the tree?