After https://trac.webkit.org/changeset/129788 many tests starting to fail on Qt WK2, because now NRWT uses only wk2/Skipped file, but not wk2/TestExpectations. Fix is coming soon.
Created attachment 166224 [details] Patch
Additionally I removed the unnecessary skip list functions.
View in context: https://bugs.webkit.org/attachment.cgi?id=166224&action=review > Tools/Scripts/webkitpy/layout_tests/port/qt.py:148 > + if self.get_option('webkit_test_runner'): > + paths.append('wk2') Here is a loss of check for Qt 5.0, but if the webkit_test_runner is true it is wk2 by default. Am I right? > Tools/Scripts/webkitpy/layout_tests/port/qt_unittest.py:98 > + if case['use_webkit2']: > + case['search_paths'].append("wk2") This could lead to some strange things if I'm not mistaken. Like: if the test_baseline_search_path method executes after this.
(In reply to comment #3) > View in context: https://bugs.webkit.org/attachment.cgi?id=166224&action=review > > > Tools/Scripts/webkitpy/layout_tests/port/qt.py:148 > > + if self.get_option('webkit_test_runner'): > > + paths.append('wk2') > > Here is a loss of check for Qt 5.0, but if the webkit_test_runner is true it is wk2 by default. Am I right? WK2 build is impossible with Qt 4.8, so running WK2 test with Qt 4.8 will fail because of missing WTR, search path won't be real problem. :) > > Tools/Scripts/webkitpy/layout_tests/port/qt_unittest.py:98 > > + if case['use_webkit2']: > > + case['search_paths'].append("wk2") > > This could lead to some strange things if I'm not mistaken. Like: if the test_baseline_search_path method executes after this. Good point. Will fix.
Created attachment 166237 [details] updated patch
(In reply to comment #5) > Created an attachment (id=166237) [details] > updated patch LBTM :)
Comment on attachment 166237 [details] updated patch rs=me
Comment on attachment 166237 [details] updated patch Clearing flags on attachment: 166237 Committed r129887: <http://trac.webkit.org/changeset/129887>
All reviewed patches have been landed. Closing bug.
Sorry about this ... I forgot to check the Qt waterfall for failures :(.