From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7793 invoked by alias); 9 Apr 2003 17:44:10 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 7784 invoked from network); 9 Apr 2003 17:44:09 -0000 Received: from unknown (HELO jackfruit.Stanford.EDU) (171.64.38.136) by sources.redhat.com with SMTP; 9 Apr 2003 17:44:09 -0000 Received: (from carlton@localhost) by jackfruit.Stanford.EDU (8.11.6/8.11.6) id h39Hhrp01762; Wed, 9 Apr 2003 10:43:53 -0700 X-Authentication-Warning: jackfruit.Stanford.EDU: carlton set sender to carlton@math.stanford.edu using -f To: Elena Zannoni Cc: gdb-patches@sources.redhat.com Subject: Re: [RFA/TESTSUITE] annota1.exp fixes References: <16019.21217.67067.495012@localhost.redhat.com> From: David Carlton Date: Wed, 09 Apr 2003 17:44:00 -0000 In-Reply-To: <16019.21217.67067.495012@localhost.redhat.com> Message-ID: User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Common Lisp) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-SW-Source: 2003-04/txt/msg00167.txt.bz2 On Tue, 8 Apr 2003 18:53:21 -0400, Elena Zannoni said: > The test was too eager in matching the whole path to the source file. > What's really happening is that it wouldn't match a '+' in the > directory name. (In RedHat gdb the source directory is named the same > as the sourceware snapshots i.e. gdb+dejagnu, which now I know not to > do again). For what it's worth, I had another problem with those regexps a while ago: I'm getting output that included subdir but not srcdir. (Presumably I'm the only one seeing this because I actually build in the source directory.) I submitted a patch to allow that; Fernando's response was that he didn't know if that was allowable output, so we should check with annotation users to make sure. And nobody ever did that. Your regexps accept the output that I'm seeing. So if we're really nervous about not wanting to accept my output, then probably your patches aren't a great idea, and we should instead modify them to create new variables srcdir-regexp and subdir-regexp (or whatever) that are regexp-protected, and use them. On the other hand, I'm inclined to go with your patches: the regexps already accept it if both srcdir and subdir are missing, so could it really be any worse if just srcdir is missing? By the way, my original message was in a thread titled "[rfa/testsuite] make annota1 regexps more generous"; see also a threat titled "[RFA] annota1.exp: Don't require complete pathname in info break test". David Carlton carlton@math.stanford.edu