----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://git.reviewboard.kde.org/r/103478/#review57846 -----------------------------------------------------------
Ship it! Even if it does not crash, the --waitforwm key is completely broken w/o this patch since the application will wait for some random property forever. -> No idea what --waitforwm is good for, but either this patch should go in or the option removed. For testing the application has likely to be run from ~/.kde/env since the standard does not mandate to withdraw it (and kwin doesn't), so it's not a secure check whether a NETWM compliant WM is *still* present at all. openbox withdraws it, so you could openbox --replace & sleep 5; pkill openbox; sleep 5; klipper --waitforwm & sleep 30; kwin & or alternatively just kquitapp kwin; sleep 1; xprop -remove -root _NET_SUPPORTED; sleep 1; klipper --waitforwm & sleep 30; kwin & Assuming it would be relevant, you got a shipIt! from at least here - sorry for not reviewing before. - Thomas Lübking On May 12, 2014, 10:25 p.m., Christoph Feck wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://git.reviewboard.kde.org/r/103478/ > ----------------------------------------------------------- > > (Updated May 12, 2014, 10:25 p.m.) > > > Review request for kdelibs. > > > Bugs: 287364 > http://bugs.kde.org/show_bug.cgi?id=287364 > > > Repository: kdelibs > > > Description > ------- > > KApplication references atom_NetSupported in parseCommandLine() before it is > initialized. This patch fixes it by reordering the code. > > I verified it no longer crashes when invoking with "--waitforwm", but I have > no idea how to test if it actually works. > > > Diffs > ----- > > kdeui/kernel/kapplication.cpp cc20f05 > > Diff: https://git.reviewboard.kde.org/r/103478/diff/ > > > Testing > ------- > > > Thanks, > > Christoph Feck > >