> On Nov. 17, 2011, 9:28 p.m., Ruurd Pels wrote: > > konqueror/src/konqview.h, line 280 > > <http://git.reviewboard.kde.org/r/103176/diff/1/?file=41389#file41389line280> > > > > OK I can agree with returning an empty list if m_service == null > > however I also think we need to come up with reasons WHY m_service == null > > before serviceTypes() is called. > > > > From the original code it looks like m_service should always be > > initialized beforehand, so why is it that that did not happen for that > > particular website. Maybe an errormessage is in order?
Honestly that part of the patch is unecessary. The only part that is applicable to this bug is the patch for KonqRun which uses a QPointer object without first checking to see if it is still valid. I will simply discard this portion. - Dawit ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://git.reviewboard.kde.org/r/103176/#review8281 ----------------------------------------------------------- On Nov. 17, 2011, 8:33 p.m., Dawit Alemayehu wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://git.reviewboard.kde.org/r/103176/ > ----------------------------------------------------------- > > (Updated Nov. 17, 2011, 8:33 p.m.) > > > Review request for KDE Base Apps and David Faure. > > > Description > ------- > > Though I cannot reproduce the bug given above, the backtrace attached to the > report seems to show a NULL object is being accessed. This patch is intended > to prevent such scenarios and in the process fix bugs like the one reported > in bug# 285844. > > > This addresses bug 285844. > http://bugs.kde.org/show_bug.cgi?id=285844 > > > Diffs > ----- > > konqueror/src/konqrun.cpp 9e9310c > konqueror/src/konqview.h 45c5bde > > Diff: http://git.reviewboard.kde.org/r/103176/diff/diff > > > Testing > ------- > > > Thanks, > > Dawit Alemayehu > >
