Hi all,
I've found a minor bug in one of the SVG 1.1 Second Edition tests,
filters-felem-02-f.svg [1]. The test doesn't take the boundingbox offsets
into account, same thing roc discovered in the filters-light-03 test at
the F2F in Auckland (that one has already been fixed).
Changing the test to be correct means we would need to update the
implementation report[2]. I don't believe it's possible to pass both
filters-light-03 and filters-felem-02 unless the filters-felem-02 test is
changed, but we do have some builds passing that filters-light-03 test,
and speaking for Opera the same build also passes the modified
filters-felem-02 test. I'd like to hear if the mozilla build that passed
the filters-light-03 test also passes the modified version of
filters-felem-02.
I've attached the modified version of the testcase to this email, along
with the new reference png.
Either we drop the test back to draft and remove it from the
implementation report, or we review the changes and test to see if the
mozilla build passes, and then update the report accordingly.
I have all the files ready to be checked in (including an svg imagepatch).
[1] http://dev.w3.org/SVG/profiles/1.1F2/test/svg/filters-felem-02-f.svg
[2]
http://dev.w3.org/SVG/profiles/1.1F2/test/status/implementation_matrix.html
--
Erik Dahlstrom, Core Technology Developer, Opera Software
Co-Chair, W3C SVG Working Group
Personal blog: http://my.opera.com/macdev_ed
