nrwt: use the default logic for skipping platform-specific tests for chromium
Created attachment 178279 [details] Patch
Created attachment 178280 [details] fix ChangeLogs
Comment on attachment 178280 [details] fix ChangeLogs View in context: https://bugs.webkit.org/attachment.cgi?id=178280&action=review > LayoutTests/platform/chromium-android/TestExpectations:-11 > -# Skip tests of other platforms to save time. > -platform/gtk [ WontFix ] > -platform/mac [ WontFix ] > -platform/mac-wk2 [ WontFix ] > -platform/qt [ WontFix ] > -platform/win [ WontFix ] If we don't change the code above (see below), we should probably change these to Skip instead. > LayoutTests/platform/chromium/TestExpectations:-241 > -# Run the Mac-specific platform tests, but only to check for crashes. > -platform/gtk [ WontFix ] > -platform/mac [ WontFix ] This looks like it was intentional, actually -- this comment suggests we really do want to run the off-platform tests if there's an expectation, just to look for crashes. I have no idea if it ever caught anything, though.
(In reply to comment #3) > (From update of attachment 178280 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=178280&action=review > > > LayoutTests/platform/chromium-android/TestExpectations:-11 > > -# Skip tests of other platforms to save time. > > -platform/gtk [ WontFix ] > > -platform/mac [ WontFix ] > > -platform/mac-wk2 [ WontFix ] > > -platform/qt [ WontFix ] > > -platform/win [ WontFix ] > > If we don't change the code above (see below), we should probably change these to Skip instead. > > > LayoutTests/platform/chromium/TestExpectations:-241 > > -# Run the Mac-specific platform tests, but only to check for crashes. > > -platform/gtk [ WontFix ] > > -platform/mac [ WontFix ] > > This looks like it was intentional, actually -- this comment suggests we really do want to run the off-platform tests if there's an expectation, just to look for crashes. I have no idea if it ever caught anything, though. Yes, we did intentionally use to run the platform-specific tests for other ports. At some point a while ago we decided to change this behavior to be more consistent with the other ports, but I never actually implemented it until now.
Comment on attachment 178280 [details] fix ChangeLogs OK, cool! r=me
Committed r137208: <http://trac.webkit.org/changeset/137208>