Need to determine which ones need platform results and which ones are real failures.
Added expected failures in http://trac.webkit.org/changeset/91645. I think the one that has pixel result (fast/multicol/client-rects.html) looks totally wrong, the rest likely just need platform results. Ossy, could you take a look?
Turn the skipping into orwt style to make all of our platforms happy: http://trac.webkit.org/changeset/91650 We should really switch to NRWT!
Fix landed in https://trac.webkit.org/changeset/91662. Balázs, feel free to fix NRWT for ARM, Qt 4.8 and wk2 if you would like to do the switch as soon as possible: https://bugs.webkit.org/show_bug.cgi?id=64071
(In reply to comment #3) > Fix landed in https://trac.webkit.org/changeset/91662. Have you checked the client-rects test manually? In the expected png the boxes are in the center of the rectangular but with Qt they are outside of it. I don't think it is correct result :-)
(In reply to comment #4) > (In reply to comment #3) > > Fix landed in https://trac.webkit.org/changeset/91662. > > Have you checked the client-rects test manually? In the expected png the boxes are in the center of the rectangular but with Qt they are outside of it. I don't think it is correct result :-) Nevermind, I had to use the test fonts.
(In reply to comment #4) > (In reply to comment #3) > > Fix landed in https://trac.webkit.org/changeset/91662. > > Have you checked the client-rects test manually? In the expected png the boxes are in the center of the rectangular but with Qt they are outside of it. I don't think it is correct result :-) Of course. :) DRT generated correct results, but QtTestBrowser doesn't, because it doesn't use testfont by default. You can enable it with -use-test-fonts command line option.