----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://git.reviewboard.kde.org/r/111951/#review38306 -----------------------------------------------------------
Hmm, what's the bug number? 321706 doesn't exist, must be a typo. I assume the unittest from commit 871cccc8a still works with this fix? It would be good to unittest this fix too, to ensure no regressions. I see that kde-runtime/keditfiletype/tests/ still passes, so if I broke something, we need to extend that test... - David Faure On Aug. 9, 2013, 12:22 p.m., Mathias Tillman wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://git.reviewboard.kde.org/r/111951/ > ----------------------------------------------------------- > > (Updated Aug. 9, 2013, 12:22 p.m.) > > > Review request for kdelibs. > > > Description > ------- > > This fixes a bug introduced by > https://projects.kde.org/projects/kde/kdelibs/repository/revisions/871cccc8a88a600c8f850a020d44bfc5f5858caa > that made it impossible to re-order file type associations both in System > settings and in the open with list. Hence it contains a new way of detecting > duplicate (inherited) mimetype entries, that the original was supposed to fix. > > > This addresses bug https://bugs.kde.org/show_bug.cgi?id=321706. > > http://bugs.kde.org/show_bug.cgi?id=https://bugs.kde.org/show_bug.cgi?id=321706 > > > Diffs > ----- > > kdecore/services/kservice.cpp 8e81929b91803a3eed586d9fc15cdd78165b6bce > kded/kbuildservicefactory.cpp 7f89a991d088476d8ed94763e6fa65dcc3d0603c > > Diff: http://git.reviewboard.kde.org/r/111951/diff/ > > > Testing > ------- > > I have tested so that file type associations can be ordered correctly, in > addition to making sure that parent mimetypes have precedence when an app > lists two or more mimetypes where one is the parent of the other. > > > Thanks, > > Mathias Tillman > >