fast/text/ruby-justification-flush.html fails on safari-600.6-branch

[Andy Estes]

fast/text/ruby-justification-flush.html has failed on safari-600.6-branch after it was introduced by merging r180278 in http://trac.webkit.org/changeset/180428.

The test says that it fails if the two black squares are not vertically aligned, which they are not.

[Andy Estes]

Created attachment 251789 [details]
ruby-justification-flush-diff

[Andy Estes]

Marked as ImageOnlyFailure in http://trac.webkit.org/changeset/183437

[Myles C. Maxfield]

Created attachment 251889 [details]
Patch

[Andy Estes]

Comment on attachment 251889 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=251889&action=review

r=me if you remove the code change. Thanks!

> Source/WebCore/rendering/RenderBlockLineLayout.cpp:767
> -                            opportunitiesInRun = Font::expansionOpportunityCount(renderText.characters8() + r->m_start, r->m_stop - r->m_start, r->box()->direction(), isAfterExpansion);
> +                            opportunitiesInRun = Font::expansionOpportunityCount(renderText.characters8(), renderText.textLength(), r->box()->direction(), isAfterExpansion);
>                          else
> -                            opportunitiesInRun = Font::expansionOpportunityCount(renderText.characters16() + r->m_start, r->m_stop - r->m_start, r->box()->direction(), isAfterExpansion);
> +                            opportunitiesInRun = Font::expansionOpportunityCount(renderText.characters16(), renderText.textLength(), r->box()->direction(), isAfterExpansion);

I'd create a separate bug for this, since it isn't necessary to fix the layout test, and we're not yet sure whether we want it in safari-600.6-branch.

[Andy Estes]

Does this test need updating on trunk, too?

[Myles C. Maxfield]

(In reply to comment #5)
> Does this test need updating on trunk, too?

Nope! This ruby stuff is pretty different on trunk.

[Myles C. Maxfield]

http://trac.webkit.org/changeset/183549