mutt/1536: Segment fault with long lines when LANG=*.UTF-8
On Wednesday, 03 August 2005 at 08:25, Brendan Cully wrote:
> The following reply was made to PR mutt/1536; it has been noted by GNATS.
>
> From: Brendan Cully <brendan@xxxxxxxxxx>
> To: bug-any@xxxxxxxxxxxxx
> Cc:
> Subject: mutt/1536: Segment fault with long lines when LANG=*.UTF-8
> Date: Tue, 02 Aug 2005 23:20:12 -0700
>
> On Tuesday, 02 August 2005 at 07:05, Anders Helmersson wrote:
> > The following reply was made to PR mutt/1536; it has been noted by GNATS.
> >
> > From: Anders Helmersson <anders.helmersson.utsikt@xxxxxxxxxxxx>
> > To: bug-any@xxxxxxxxxxxxx
> > Cc:
> > Subject: Re: mutt/1536: Segment fault with long lines when LANG=*.UTF-8
> > Date: Tue, 2 Aug 2005 06:24:52 +0200
> >
> > On Tue, 2005-08-02 at 04:31:44 +0200, Brendan Cully wrote:
> > > This seems to have been a libc bug. Can you reproduce with a current
> version?
> >
> > Yes, the segfault still appears with 1.5.9i (2005-03-13).
> >
> > The debian version (1.5.9-2) with the patch
> > patch.helmersson.incomplete-mbyte.2 does not segfault.
>
> The patch seems reasonable to me, but I'm not too conversant in this
> multibyte stuff. Can anyone with more authority comment on why it
> shouldn't be committed?
I've had no luck reproducing the error according to your instructions
on debian with CVS mutt, LANG=en_US.UTF-8, color body magenta default
[a-z] and the mail you uploaded. I'm on sid and my libc is
2.3.2.ds1-22.
I also can't reproduce this on ubuntu hoary or OS X. Any clues?