Various text-related reftest failures on geckoview (x86-64 Android 7.0 emulator)
Categories
(Core :: Layout: Text and Fonts, defect, P3)
Tracking
()
Tracking | Status | |
---|---|---|
firefox69 | --- | affected |
People
(Reporter: kats, Unassigned)
References
Details
I'm turning on gecko reftests on Geckoview in bug 1501582 and as part of that annotating any failures I'm running into. There's a few failures where the test/reference images are just vertically offset by a few pixels:
== 617869-1.html 617869-1-ref.html
== emoji-fallback-1.html emoji-fallback-1-ref.html
== emoji-fallback-3.html emoji-fallback-3-ref.html
== stretchy-mfenced-4a.html stretchy-mfenced-4-ref.html
== stretchy-mfenced-4b.html stretchy-mfenced-4-ref.html
1135361-ruby-justify-1.html 1135361-ruby-justify-1-ref.html
and a couple of synthetic font !=
tests that end up rendering the same for test/reference:
!= synthetic-style-1.html synthetic-style-1-notref.html
!= synthetic-style-2.html synthetic-style-2-notref.html
and some tests where a single glyph appears mis-positioned by a pixel, maybe rounding error but it looks visually different so I figured I'd mark it as a failure to be safe:
== multiple-chunks-bidi.svg multiple-chunks-bidi-ref.svg
== multiple-chunks-different-anchor-bidi.svg multiple-chunks-different-anchor-bidi-ref.svg
== multiple-chunks-multiple-dx-bidi.svg multiple-chunks-multiple-dx-bidi-ref.svg
== multiple-chunks-multiple-dx-different-anchor-bidi.svg multiple-chunks-multiple-dx-different-anchor-bidi-ref.svg
Updated•5 years ago
|
Updated•5 years ago
|
Comment 1•5 years ago
|
||
The priority flag is not set for this bug.
:jfkthame, could you have a look please?
For more information, please visit auto_nag documentation.
Comment 2•5 years ago
|
||
These look like mostly fragile tests or minor rounding issues, not serious bugs we need to fix -- but we should try to resolve them and remove failure annotations wherever possible.
In particular, the synthetic-style-{1,2}.html notref-tests may be failing because the emulator environment doesn't provide any of the font names that the testcase calls for. If that's the case, we should be able to fix it by adding something appropriate to the @font-face rules.
Updated•1 year ago
|
Description
•