bug#72296: 29.4; cperl-mode parsing error

[Harald Jörg]

Andrea Corallo <acorallo@gnu.org> writes:

> brian greenfield <dev@briang.org> writes:
>
>> On Thu, Jul 25, 2024 at 01:01:30PM -0400, Andrea Corallo wrote:
>>> I might be doing something wrong but I cannot reproduce on emacs-30,
>>> could you confirm?  29 is not under development anymore.
>>
>> No, it was me in the wrong. My emacs-30 was too old, retrying on a newer
>> emacs-30 has the bug fixed.
>>
>> Sorry for the noise.
>
> [re-adding the list, please keep it Cced in the replies]
>
> Actually the others can still reproduce on more recent versions (I
> can't), so I'm all but sure this is solved (and I'm a little puzzled on
> the reproducibility of this).

If either of you built your emacs-30 with the commit c27055a9 in place,
then it is supposed to be not reproducible: I fixed it yesterday.

Until the regular rebasing happens, it can still be reproduced on
master.  I may be stating the obvious, but the code is

for (2..$n/2) {}

without any enclosing quotes.  If you apply cperl-mode for this snippet,
you should see the message "End of ‘/ ... /’ string/RE not found:
(scan-error Unbalanced parentheses 11 17)" in the echo area, the slash
should be formatted with font-lock-constant-face, the 2 following it
with font-lock-string-face and the closing paren with
font-lock-warning-face.

If you can not reproduce the error on master, then I'd like to examine
this further before closing the bug.

-- 
Cheers,
haj