[Webkit-unassigned] [Bug 181360] REGRESSION(r225736)[Win] mathml/opentype/opentype-stretchy-horizontal.html fails to pass

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Sat Jan 6 02:41:19 PST 2018


https://bugs.webkit.org/show_bug.cgi?id=181360

--- Comment #2 from Frédéric Wang (:fredw) <fred.wang at free.fr> ---
(In reply to Minsheng Liu from comment #0)
> The logical widths are way off, and the glyph assembly also fails.

The width seems too small (at least the middle glyphs of the assembly are missing and the start & end glyphs are too close). The logical height is too large.

git log -- LayoutTests/platform/win/mathml/opentype/opentype-stretchy-horizontal-expected.txt

gives me the following history:

https://trac.webkit.org/changeset/169644
https://trac.webkit.org/changeset/202776
https://trac.webkit.org/changeset/202794
https://trac.webkit.org/changeset/203212
https://trac.webkit.org/changeset/225736

I am not sure the painting of stretchy operator has ever worked well on AppleWin (there are no PNG references in the history) and I think the test even randomly crashed in the past. The crazy logical height was already here when I first removed the failure expectation and added back the txt reference ( https://trac.webkit.org/changeset/202794 ). So I don't think new issues were caused by r225736.

So maybe just re-baseline the test with the current txt output for now.

Note: Personally, I won't spend too much time on the AppleWin port... I would even actually recommend you to get more freedom and try a nice operating system called Linux at some point :-) That will also allow you to update tests for the GTK port ;-)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20180106/6a1958f5/attachment-0001.html>


More information about the webkit-unassigned mailing list