WebKit Bugzilla
9.1 of the layout algorithm says: "If a flexbox item has ‘flex-item-align:stretch' and its preferred cross size is 'auto' and its margin-box cross size is smaller than cross-size of its flexbox line, its final cross size is set so that its margin-box cross size is equal to cross-size of its flexbox line." http://dev.w3.org/csswg/css3-flexbox/#layout-algorithm
Created attachment 132390 [details] Patch
Comment on attachment 132390 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=132390&action=review > Source/WebCore/rendering/RenderFlexibleBox.cpp:1014 > +void RenderFlexibleBox::alignChildStretch(RenderBox* child, LayoutUnit lineCrossAxisExtent) Nit: I find this method name awkward. It reads to me as if it's aligning the stretch itself (which is nonsensical). applyChildStretch perhaps?
Created attachment 132403 [details] Patch for landing
(In reply to comment #2) > (From update of attachment 132390 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=132390&action=review > > > Source/WebCore/rendering/RenderFlexibleBox.cpp:1014 > > +void RenderFlexibleBox::alignChildStretch(RenderBox* child, LayoutUnit lineCrossAxisExtent) > > Nit: I find this method name awkward. It reads to me as if it's aligning the stretch itself (which is nonsensical). applyChildStretch perhaps? Now named applyStretchAlignmentToChild.
Comment on attachment 132403 [details] Patch for landing Clearing flags on attachment: 132403 Committed r111094: <http://trac.webkit.org/changeset/111094>
All reviewed patches have been landed. Closing bug.
Reverted r111094 for reason: broke Committed r111103: <http://trac.webkit.org/changeset/111103>
Committed r111214: <http://trac.webkit.org/changeset/111214>
css3/flexbox/child-overflow.html was failing due to https://bugs.webkit.org/show_bug.cgi?id=77368, so I just landed with it marked as failing in test_expectations.txt.