[cairo] flagging Quartz backend as supported for 1.6

Vladimir Vukicevic vladimir at pobox.com
Fri Mar 14 19:58:45 PDT 2008


Behdad and I had a brief conversation about this in the other thread  
(related to the quick iPhone hack I did), and I wanted to get some  
more feedback directly on this topic.  I've pushed out a few  
enhancements to the quartz backend that cause the unbounded operators  
to be handled more correctly than before, but with color fringing at  
edges (I basically clear a mask after each operation).  I've also  
implemented reflect and repeat for gradients.  I believe everything  
else works well, though there are some differences in how radial  
gradients are handled in some cases, and some of the pixman-operation- 
specific tests are not giving identical results.  I've also split out  
the experimental quartz-image stuff out into a separate backend header.

So, any objections to marking the quartz surface and atsui font  
backend as supported?  I'm not going to get a chance to finish the  
CGFont work before 1.6 (though it's very nearly there), but even then,  
the current atsui entry point can be supported trivially even with a  
wholesale switch to CGFont for cairo 1.8.  (However, the backend font  
identifier would change.. hmm.. maybe we should call it a QUARTZ font,  
instead of an ATSUI font even now?)

     - Vlad

More information about the cairo mailing list