Closed lassoan closed 6 years ago
Thanks for the patch. I was off yesterday. Just starting to spin up on this issue...
@jcfr @lassoan Was this ever updated in Slicer master?
Not yet. Would be great if someone could take care of it.
I was out at a hackathon last week and now in vacation.
On Tue, Apr 10, 2018, 2:46 PM Bradley Lowekamp notifications@github.com wrote:
@jcfr https://github.com/jcfr @lassoan https://github.com/lassoan Was this ever updated in Slicer master?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/SimpleITK/SlicerSimpleFilters/pull/10#issuecomment-380224190, or mute the thread https://github.com/notifications/unsubscribe-auth/AANXo9dYeoHBYUsiji353UiVRf6_pQ3eks5tnQwHgaJpZM4TAxAT .
I'll take care of it.
Fixed in rev 27138.
The update has fixed py_SimpleFiltersModuleTest, which has been failing for a long time.
Thanks again to fixing this. I still don't understand the origins of this problem and where the behavior changed.
I thought that with Qt slot and signals, when were automatically deleted when the QtObject emitting the signal was deconstructed. So the need for this fix just seems little odd to me and makes me thing there may be something improper with the destructor of the ctkCoordinateWidget.
I don't understand the root cause either and why the problem is only with ctkCoordinateWidget. The issue came up when we switched to Qt5.
In Slicer-4.9 (with Qt5), SimpleFilters module made Slicer crash when switching to a widget that contains ctkCoordinatesWidget. For example, switching to AdaptiveHistogramEqualizationImageFilter and then to another filter caused crash.
The problem was caused by destroying ctkCoordinateWidget while still having connections.
Fixed by removing all connections of filter parameter widgets before the widgets are destroyed.