bug#81517: 31.0.90; executable-find on remote buffer fails
Eli Zaretskii <[email protected]> Thu, 30 Jul 2026 10:30:59 +0300
| Newsgroups | gmane.emacs.bugs |
|---|---|
| Message-ID | <[email protected]> |
> From: Michael Albinus <[email protected]> > Cc: Shun-ichi TAHARA <[email protected]>, [email protected] > Date: Thu, 30 Jul 2026 09:01:39 +0200 > > Eli Zaretskii <[email protected]> writes: > > Hi Eli, > > >> If this commit is merged to Emacs31, many Windows users are happy. > > > > I imagine those users to be less happy if that commit, which is far > > from trivial, causes regressions elsewhere. > > > > Being very close to a release, this is not a simple obvious decision > > to make, it involves risks that need to be assessed. By contrast, the > > offending commit was with us for more than a year, with no complaints > > from the users of what is now the Emacs 31 pretest. > > As a compromise, we might backport only the fix in executable-find (a > one-liner), and backport the other changes to Emacs 31.2. It looks > simple enough (I know, famous last words ...). I don't think it's a good idea to have executable-find behave differently from call-process etc. No, if we backport, we should backport all of it, IMO. Let's wait for Sean to chime in.