Re: grep 2.4.2 incorrect handling of '[x^]'

[Alain Magloire]

> 
> On Mon, 30 Jul 2001 14:27:19 -0700 (PDT), 
> Paul Eggert <address@hidden> wrote:
> >In the current POSIX standard, [a-c] matches any collating element in
> >the collating seqence from 'a' through 'c' inclusive.  This is not the
> >same as strcoll -- so the current code is indeed nonconforming.  Nor
> >is it the same as the set of characters that [a-c] matches -- so the
> >proposed patch is also nonconforming, even for unibyte locales.  (This
> >is because a collating element can contain more than one character,
> >e.g.  '[a-z]' can match 'ss' in a German locale.)
> >
> >Hence the proposed patch does not fix the whole problem.
> >
> >Also, I suspect that the proposed patch mishandles some cases,
> >e.g. range expressions involving '[' itself.
> 
> That is exactly why I logged the problem in the first place.  '[x^]'
> was misbehaving.  I searched Redhat's Bugzilla and found that somebody
> had already reported that the patch broke commands like
>   ps xa | grep '[^]]'ssh
> which searches for ssh but excludes the grep command itself.  Because
> that bug had already been reported against the RH patch, I did not add
> another bug report.
> 
> One thing that is worrying, Redhat claim that this bug is fixed in grep
> 2.5e, where do they get version 2.5 from?  It is not an official
> release number.

2.5 is not release.

2.5[a-z](C locale 8) are the codename, of {alpha,beta} snapshots of GNU grep.
The purpose is to make snapshots available for testers or anyone 
and to have feedback. The usual development cycle.

The code is available via anonymous CVS.

$ cvs -d :pserver:address@hidden:/cvs login
Password: [just hit enter]
$ cvs -d :pserver:address@hidden:/cvs checkout grep

> Also the latest RH grep I can find is still 2.4.2
> which has their buggy patch.  I think this needs to be sorted out
> between the grep maintainers and RH,

GNU/Linux distributions have there own priorities/schedule, which do not
necessary correspond to the available freetime of the maintainer 8-).

> it is time for me to bow out of this thread.

Thanks for the feedback.

--
alain