[Webkit-unassigned] [Bug 139905] REGRESSION(r177637) [HarfBuzz][GTK][EFL] It made 3 performance tests crash and +24 layout tests crashes/failures

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=139905

--- Comment #8 from Carlos Garcia Campos <cgarcia at igalia.com> ---
(In reply to comment #7)
> (In reply to comment #6)
> > Created attachment 244254 [details]
> > Patch
> > 
> > I think this is equivalent to what we did before r177637, falling back to
> > the primary font data for missing glhyps. I'm still getting a few failures,
> > though, but I'm not sure if it's something related to my system, because I
> > see the missing glyhps but rendered slightly different. I think we can land
> > this anyway for now, since it fixes the crashes, most of the layout tests
> > for sure, and the perf tests.
> 
> That fixes the crashes, but there are still differences on the expected
> results of some tests. I'm not sure if that differences still indicate an
> incomplete fix, or they just need to be rebaselined.
> 
> With your patch:
> 
> Regressions: Unexpected text-only failures (15)
>   fast/css/line-height-determined-by-primary-font.html [ Failure ]
>   fast/ruby/bopomofo-letter-spacing.html [ Failure ]
>   fast/ruby/bopomofo-rl.html [ Failure ]
>   fast/ruby/bopomofo.html [ Failure ]
>   fast/text/decorations-with-text-combine.html [ Failure ]
>   fast/text/emphasis-combined-text.html [ Failure ]
>   fast/text/emphasis-vertical.html [ Failure ]
>   fast/text/international/khmer-selection.html [ Failure ]
>   fast/text/international/plane2.html [ Failure ]
>   fast/text/justify-ideograph-complex.html [ Failure ]
>   fast/text/justify-ideograph-simple.html [ Failure ]
>   fast/text/justify-ideograph-vertical.html [ Failure ]
>   fast/text/khmer-lao-font.html [ Failure ]
>   fast/text/midword-break-before-surrogate-pair-2.html [ Failure ]
>   svg/custom/svg-fonts-fallback.xhtml [ Failure ]
> 
> Check here the diffs:
> http://people.igalia.com/clopez/bug-139905-results/results.html

I see, I'm not a fonts expert so I don't know how to properly fix this, so I suggest to land this patch for now to fix the crashes, and open a new bug for the remaining layout test failures.

-- 
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/20150108/c6a490ad/attachment-0002.html>