[cairo] Slipping 1.2.0 out to 2005-12-15, and help needed
cworth at cworth.org
Thu Nov 10 16:47:09 PST 2005
Shortly after cairo 1.0.0 was released, I picked an out-of-the-hat
date of 2005-11-01 for cairo 1.2.0. That date has now come and gone,
so so much for my powers of prognostication.
I recently looked over the 1.2.0 roadmap [*] and saw that there's
quite a bit still left to do. Personally, I've been focusing on the PS
and PDF output quality issues. The hard parts of that are done now,
but there's probably at least a week or two of work still to be done
I had considered finishing those things up to get a 1.2.0 release out
as soon as possible (in the next week or two) and letting everything
else on the roadmap slip.
But it looks like that's not going to be possible. I probably won't
have much time to dedicate to this work until the beginning of
December. So 1.2.0 won't be happening before 2005-12-15.
This does provide more time for other interested people to jump in and
fill the gaps for the remaining pieces of 1.2.0. If any of the
following sound important to you, please consider getting involved:
* win32 backend
We basically just need to get it plugged into the test
suite. For a long time, the issue here was missing support in
the test suite for RGB24 output and for backend-specific
reference images. But thanks to Keith's efforts with getting
the PS backend into the test suite these two problems are
So, is there a cairo-win32 developer that could fill in the
two stubs in cairo/test/cairo-test.c:
create_win32_surface (int width, int height, void **closure);
cleanup_win32 (void *closure);
If so, that would be a great help and would definitely
contribute a lot toward making sure that cairo-win32 remains
stable in the future.
I sent a half-started patch for this to the list sometime
ago. I feel good about the API, (though maybe the patch was
before the addition of the current CONTENT support, and if so
it should be updated to that). But the real work here is to
make the intermediate surface auto-growable, which might
require a tweak in the pattern code to get the offset right
when the intermediate surface is finally painted.
I really like this API, and I'm looking forward to having it
available. I'd be glad to help anyone interested in working on
* Performance fixes
There's always more to be done here. I'm going to make that a
big focus of my efforts post-1.2.0 by working on a performance
regression suite and tackling some long-standing issues.
We probably don't need much here before 1.2.0 since people are
claiming that cairo 1.1.1 is already performing much better
than 1.0.2. (Of course, there's still a lot of room for
improvement, but this isn't something that should slow down
1.2.0 I think).
* SVG backend
The roadmap has this as experimental, so I think it can be
thrown in at pretty much any point. It just needs someone to
look at it first. I will do this after the PS and PDF stuff is
done if nobody else has by then.
* Bug fixes
There are a couple of fairly serious bugs listed in ROADMAP
already that I think really need to be fixed. A couple of the
XFAILs obviously need to be fixed as well, (such as the *ahem*
segfault from extend-reflect).
Meanwhile, there have been a bunch of little bug reports,
(slight build problems on various platforms for example), that
should be pretty easy picking for almost any willing
As usual, there's almost no end of fun work to be done. I appreciate
everything that so many of you are already doing. And this is also a
great chance for some new faces to jump in and make some contribution.
I'm glad to provide more details and guidance on any of the above as
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/cairo/attachments/20051110/89dd708a/attachment.pgp
More information about the cairo