Andrew Benton wrote: > Hello, > Mozilla have altered the copy of Cairo that they use (to fix a bug > https://bugzilla.mozilla.org/show_bug.cgi?id=715658 ) and the effect is > that for more than a month compiling with --enable-system-cairo has > been broken ( https://bugzilla.mozilla.org/show_bug.cgi?id=722975 ). > It's only a matter of time before this change works its way through to > a released version of Firefox and I thought I should mention it on the > list ahead of that. > I can see 3 options:
> 1) We give up on --enable-system-cairo and statically compile their > version of Cairo into Firefox. (We decided to do a similar thing with > Mplayer and FFmpeg) > 2) We patch Firefox to revert the change (it's a tiny 2 line patch > https://bugzilla.mozilla.org/attachment.cgi?id=595000 ) > 3) We patch Cairo to add the mozilla code into the system Cairo > https://bugzilla.mozilla.org/attachment.cgi?id=609898 > It's like with the animated png patch we apply. We use that patch so > Firefox will build against the system library, in that case the > upstream (libpng) devs have rejected the patch but we use it anyway. > I've been using this for a month and it seems to work fine, I'm not > aware of any problems, however, on > https://bugzilla.mozilla.org/show_bug.cgi?id=722975 Jeff Muizelaar was > asked why he hadn't pushed the changes upstream to the Cairo people and > he said "I've been hesitant to push this upstream because I'm not > really sure it's appropriate", whatever that means. > So I vote for option 3. Looking at the discussion, I agree with option 3. From the patch, it looks like it is only making a couple of private functions into public functions. I'd think that would be OK with the cairo developers if someone as prominent as Mozilla needed them. -- Bruce -- http://linuxfromscratch.org/mailman/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
