From owner-cvs-gnu  Mon May 22 16:31:04 1995
Return-Path: cvs-gnu-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.10/8.6.6) id QAA01063
          for cvs-gnu-outgoing; Mon, 22 May 1995 16:31:04 -0700
Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241])
          by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id MAA03155
          ; Mon, 22 May 1995 12:34:26 -0700
Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id MAA05512; Mon, 22 May 1995 12:33:08 -0700
From: "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com>
Message-Id: <199505221933.MAA05512@gndrsh.aac.dev.com>
Subject: Re: cvs commit: src/gnu/lib/libreadline display.c
To: adam@veda.is (Adam David)
Date: Mon, 22 May 1995 12:33:08 -0700 (PDT)
Cc: adam@freefall.cdrom.com, CVS-commiters@freefall.cdrom.com,
        cvs-gnu@freefall.cdrom.com
In-Reply-To: <199505221924.TAA09678@veda.is> from "Adam David" at May 22, 95 07:24:30 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1488      
Sender: cvs-gnu-owner@freebsd.org
Precedence: bulk

> 
> > >   Modified:    gnu/lib/libreadline  display.c
> > >   Log:
> > >   eliminate coredump for readline(NULL) case
> > 
> > Did I miss a message about code freeze being lifted??? 
> 
> Bugfix only, as explicitly permitted by freefall's motd.

There was a full page message sent to -committers describing
in detail on how thing are to be done.  Changes in any form
to libraries is a reviewable change.

> > A library is defanitly a critical place to be patching.
> 
> There is no change of interface, only an advertised functionality
> that dumped core instead of the expected behaviour.

Change is change, any and every change *can* cause a new bug,
the posted process did not say anything about interface changes
or not, it said critical system components, and a library is
defanitly a critical system component.

> > Reviewd by: ??????
> 
> Is this necessary for a trivial bugfix?

YES!  No mater how trivial it may look, it can cause a problem if not
carefully looked at.  I have seen some of the most trivial bug fixes
break tons of things due to an oversight.

> old: strlen(local_prompt)
> new: local_prompt ? strlen(local_prompt) : 0
> 
> local_prompt == NULL  is a valid state, as empty prompts may be represented
> as either "" or NULL in the rest of the code.

The above is pretty useless, how about a context diff???


-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                   Custom computers for FreeBSD