Unfortunately, given the current state of the Windows testers, I am not sure when this started. fast/text/complex-preferred-logical-widths.html fast/text/complex-synthetic-bold-space-width.html fast/text/line-initial-and-final-swashes.html fast/text/midword-break-before-surrogate-pair-2.html fast/text/unicode-variation-selector.html fast/text/international/arabic-justify.html fast/text/international/bidi-mirror-he-ar.html fast/text/international/float-as-only-child-of-isolate-crash.html fast/text/international/text-combine-image-test.html fast/text/international/text-spliced-font.html I am going to add expected (failing?) results for Windows, but someone with more knowledge about text needs to look over them. Here is one example: --- /home/buildbot/slave/win-release-tests/build/layout-test-results/fast/text/complex-preferred-logical-widths-expected.txt 2012-05-31 11:10:12.893989000 -0700 +++ /home/buildbot/slave/win-release-tests/build/layout-test-results/fast/text/complex-preferred-logical-widths-actual.txt 2012-05-31 11:10:12.892988900 -0700 @@ -3,6 +3,6 @@ layer at (0,0) size 800x600 RenderBlock {HTML} at (0,0) size 800x600 RenderBody {BODY} at (8,8) size 784x584 - RenderBlock (floating) {DIV} at (0,0) size 308x24 [border: (3px solid #0000FF)] - RenderText {#text} at (3,3) size 302x18 - text run at (3,3) width 302: "The number of lines in this box should be one." + RenderBlock (floating) {DIV} at (0,0) size 299x24 [border: (3px solid #0000FF)] + RenderText {#text} at (3,3) size 293x18 + text run at (3,3) width 293: "The number of lines in this box should be one."
<rdar://problem/11567490>
Added Windows specific expected (failing?) results in http://trac.webkit.org/changeset/119130
Added pixel results so that it is easier to determine if these are major regressions: http://trac.webkit.org/changeset/119138
fast/text/international/float-as-only-child-of-isolate-crash.html results now matches original results. Deleting Windows specific results for this test: http://trac.webkit.org/changeset/129398