Date: Fri, 28 Sep 2007 13:53:48 -0700 (PDT) From: Don Lewis <truckman@FreeBSD.org> To: chat95@mac.com Cc: maho@FreeBSD.org, freebsd-openoffice@FreeBSD.org Subject: Re: Build of 2.3.0 on 6-stable fails Message-ID: <200709282053.l8SKrmra025850@gw.catspoiler.org> In-Reply-To: <20070928.131141.124084225.chat95@mac.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 28 Sep, Maho NAKATA wrote: > From: Don Lewis <truckman@FreeBSD.org> > Subject: Re: Build of 2.3.0 on 6-stable fails > Date: Tue, 25 Sep 2007 21:11:13 -0700 (PDT) > >> The trigger is WITH_TTF_BYTECODE_ENABLED=yes, which causes the patch >> file files/optpatch-freetype to be applied to the freetype-2.2.1.patch >> file in the OO source. The freetype-2.2.1.patch patch file was >> apparently changed for the OO 2.3.0 release, and the optpatch-freetype >> patch file in the port was not updated to match. > > Right. > >> Where things get interesting is that the patch program applies the first >> chunk of optpatch-freetype and doesn't reject it, even though the lines >> being changed don't match those in the patch. This causes the first >> four chunks of freetype-2.2.1.patch to be applied to the wrong file in >> the freetype source and to be rejected. This apparently confuses >> gpatch, causing it to think that the rest of the chunks of >> freetype-2.2.1.patch have already been applied. > :) > > thanks for clarifying about that. > >> Here's a patch to a patch to a patch that should fix this problem: > > Huge thanks, and as this will never be covered by JCA, > I can directly commit it to our FBSD port cvs repo. Please verify. It looked to me like you had some other way of generating optpatch-freetype using "cvs diff" on a checked-in copy of the OO source tree. If so, you should be able to regenerate the file, otherwise you are welcome to commit my patch.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200709282053.l8SKrmra025850>