Testing output from each backend (was: [cairo] State of Win32 backend?)

[Owen Taylor]

On Wed, 2005-05-04 at 09:18 -0400, Carl Worth wrote:
> On Wed, 4 May 2005 11:33:39 +0300, Tor Lillqvist wrote:
> > Cairo HEAD has for some time not built on Win32. The below makes it
> > build, but I can't really say whether this is correct or not.
> 
> Tor,
> 
> Thanks for looking into this.
> 
> I can't really comment too much on the correctness of this patch
> myself.
> 
> One thing that would help those people who want to work on the win32
> (or almost any other backend) is to tweak the test suite so that it
> will test these backends. I've done most of the work to make the test
> suite test multiple backends. All that is still needed is to fill out
> the following stubs in cairo/test/cairo-test.c:

[...]

> We could use this for the glitz, quartz, win32, and xcb backends right
> away. It will be nice to get to the point where we trust pixel-
> accurate consistency among the various backends.

Except that we won't *get* pixel accurate consistency between backends.
If we always do path rasterization in software, we might reduce the
problem to a few problems with the last bits from compositing
differences, except for the text test cases, which are never going
to pass.

Regards,
						Owen

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://lists.freedesktop.org/archives/cairo/attachments/20050505/831163aa/attachment.pgp