Date: Tue, 28 Jan 1997 10:50:38 -0700 (MST)
From: Marc Slemko <marcs@znep.com>
To: Apache Developer ML <new-httpd@hyperreal.com>
Subject: Re: Agenda for 1.2b7
In-Reply-To: <199701281201.NAA11486@en1.engelschall.com>
Message-ID: <Pine.BSF.3.95.970128104342.18522B-100000@alive.ampr.ab.ca>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

On Tue, 28 Jan 1997, Ralf S. Engelschall wrote:

> 
> In article <9701271900.aa15986@paris.ics.uci.edu> you wrote:
> 
> [...]
> > Should be done before 1.2b7:
> [...]
> >   * mod_rewrite is incorrectly rewriting things in 1.2b6 from 
> >     /cgi-bin/whatever?parms      to...     /http?parms
> >         Status: Jake Buchholz reported, may have to do with snprintf call.
> 
> Have I missed a CVS commit posting or why was mod_rewrite again changed?  You

It was changed along with most other things with my large buffer overflow
patch.

> know that I incorporate all changes into the official mod_rewrite distribution
> to be able to bring mod_rewrite 2.4 into Apache the next time.  Hmmm..
> 
> This time it is not nice because the changes broke mod_rewrite in the 1.2b6
> release and I received three bugreports today. And now I discovered that the
> used mod_rewrite is not my version. Instead the users used the broken one.
> Hmmmm... I would really prefer if the Apache Group will ask me to patch
> mod_rewrite.c myself so I could avoid to monitor the CVS commit messages and
> being not longer one step behind.

The problem is that if there are a dozen modules distributed with apache
and each one is maintained by someone, getting anything done is quite
difficult.  There is also the problem of conflicting release schedules.
We are _supposed_ to be getting ready for a release, which would seem to
indicate that we can't just be incorporating new versions of modules all
the time.

> 
> What is the status of the current snprintf/strncpy patches for mod_rewrite? 
> 

They were done along with the other changes between b4 and b5.  The
patches were posted at least three times to the list.  The changes made do
_not_ directly fit back into 1.1 because they need ap_snprintf.  You can
ifdef around that though I guess.  There is currently a bug as posted, and
I'm waiting for a couple of +1s to my fix before I commit it. 

I'm not trying to discourage you from maintaining the module or discourage
it from being distributed with Apache, but there are problems with this
approach.