From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.org/gmane.linux.lib.musl.general/2824 Path: news.gmane.org!not-for-mail From: Rich Felker Newsgroups: gmane.linux.lib.musl.general Subject: Re: strcasestr.c Date: Wed, 20 Feb 2013 20:34:17 -0500 Message-ID: <20130221013417.GO20323@brightrain.aerifal.cx> References: <201302141459.r1EExuU3024259@core.courtesan.com> <20130214152349.GA20323@brightrain.aerifal.cx> <20130217190452.GH20323@brightrain.aerifal.cx> <51254E0F.7010206@barfooze.de> <20130221010328.GN20323@brightrain.aerifal.cx> <20130221013054.GA90956@intma.in> Reply-To: musl@lists.openwall.com NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: ger.gmane.org 1361410468 16735 80.91.229.3 (21 Feb 2013 01:34:28 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Thu, 21 Feb 2013 01:34:28 +0000 (UTC) To: musl@lists.openwall.com Original-X-From: musl-return-2825-gllmg-musl=m.gmane.org@lists.openwall.com Thu Feb 21 02:34:51 2013 Return-path: Envelope-to: gllmg-musl@plane.gmane.org Original-Received: from mother.openwall.net ([195.42.179.200]) by plane.gmane.org with smtp (Exim 4.69) (envelope-from ) id 1U8L3i-0007dR-8M for gllmg-musl@plane.gmane.org; Thu, 21 Feb 2013 02:34:50 +0100 Original-Received: (qmail 13339 invoked by uid 550); 21 Feb 2013 01:34:29 -0000 Mailing-List: contact musl-help@lists.openwall.com; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: Original-Received: (qmail 13331 invoked from network); 21 Feb 2013 01:34:29 -0000 Content-Disposition: inline In-Reply-To: <20130221013054.GA90956@intma.in> User-Agent: Mutt/1.5.21 (2010-09-15) Xref: news.gmane.org gmane.linux.lib.musl.general:2824 Archived-At: On Wed, Feb 20, 2013 at 08:30:54PM -0500, Kurt H Maier wrote: > On Wed, Feb 20, 2013 at 08:03:28PM -0500, Rich Felker wrote: > > > > The other somewhat reasonable option would be removing the function, > > which would expose breakage in programs that were already using the > > broken version in musl. I'm mildly against this, but I'd be interested > > in hearing arguments either way. > > Please remove it. It's not world-ending to ask people to relink; > programs known to require the broken functionality can run against an > older version of musl until they're repaired. If you start enshrining > mistakes now you're just driving down glibc avenue. I'd rather have > musl be a correct libc than a quirky libc. Well the other issue is using glibc-linked binaries/libs. If any of them reference strcasestr (which they might in all those gnulib packages mentioned earlier in this thread once gnulib detects that glibc has its own strcasestr...), then we'd potentially want to provide strcasestr to be able to use them.. I don't think "enshrining mistakes" is a good policy in itself, but if they can be fixed (unbreaking the function) in ways that have other practical benefits, that might be a better alternative. With that said, opinion #1 noted. :) Rich