[cairo] ghostcript-8.54 causes most cairo-1.2.6 ps surface tests to fail

[Behdad Esfahbod]

On Wed, 2007-01-24 at 10:00 -0600, Gary V. Vaughan wrote:
> 
> If you decide to pursue this idea, I'd be happy to generate the
> reference images for any of the versions of ghostscript I have access
> to (7.07, 8.16 and 8.54 at least).

Eventually we want to.  But we first need to clean up our test code.
We've been adding parameters to it one after another and the number of
files in tests/ has been growing fast.

One thing I want to do is to separate text tests from other ones.
Another is to add an image comparison that doesn't work pixel by pixel
at all, but rather by finding the bounding box of non-background-colored
pixels and compares those.  That way you can test that text is correctly
positioned and ignore single pixel movements.

> Are the PDF surface test failures I posted about a similar known
> problem with the current testsuite? 

Not sure.  But it's typically the case that poppler HEAD does not
necessarily pass all tests.  krh should know better. (same for librsvg.
In fact installing librsvg trunk typically breaks lots of svg tests for
me.)

-- 
behdad
http://behdad.org/

"Those who would give up Essential Liberty to purchase a little
 Temporary Safety, deserve neither Liberty nor Safety."
        -- Benjamin Franklin, 1759