GNU bug report logs - #65049
29.1; vc-do-command fails in windows emacs 29.1

Previous Next

Full log

View this message in rfc822 format

From: Eli Zaretskii <eliz <at> gnu.org>
To: Dmitry Gutov <dmitry <at> gutov.dev>
Cc: rms <at> gnu.org, 65049 <at> debbugs.gnu.org
Subject: bug#65049: Minor update to the repro steps
Date: Thu, 31 Aug 2023 09:00:26 +0300

> Date: Thu, 31 Aug 2023 05:14:47 +0300
> Cc: 65049 <at> debbugs.gnu.org
> From: Dmitry Gutov <dmitry <at> gutov.dev>
> 
> On 31/08/2023 05:07, Richard Stallman wrote:
> > What effect would the proposed change have that I as a user of VC mode
> > notice?  I use it with CVS and with git.
> 
> If you have files in that repository which use a coding system that 
> doesn't match your configured language environment, it's possible that 
> the output of vc-diff might look inappropriate (decoded incorrectly).

Yes, but I think the existing code has the same issue.  IOW, the
situation where a VC fileset includes files with different encodings,
and the diffs of the entire fileset are shown in Emacs, is not well
supported by the Emacs decoding infrastructure, because that assumes
there's a single coding-system suitable to decode the entire text
shown in a buffer, and thus makes decisions about the encoding based
on sampling a small initial portion of the text.

Btw, the CVS case is slightly less problematic here, since in CVS one
usually looks at diffs of separate files, and the whole idea of VC
fileset is quite artificial, since CVS itself doesn't really support
multiple-file changesets, it treats each file separately.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.