Qt closeevent fixes for v1.2.x #1705
Conversation
added QtGui.QMainWindow.closeEvent() to make sure the close event
Gcf when it is closed. In PR matplotlib#1498 the attribute WA_DeleteOnClose was no longer set on the QtMainWindow object in QtFigureManager. The signal connection that was being used to remove the figure from Gcf when the window was closed was tied to the `destroyed()` signal of QtMainWindow, which is no longer being destroyed. Thus, gca and gcf would return references to no-longer visible figures/axes. _widgetclosed is now called when MainWindow emits 'closing()'.
as this should hit either PySide or PyQt4, depending on which is installed. (I have faith that the setup code that will make sure that if only one of them is installed, it defaults to using that one)
|
You can cherry-pick the merge commit (the one that's displayed in the PR). Generally, we just do this manually rather than through a PR, but it doesn't really matter. |
|
@tacaswell I think it's possible to cherry-pick a range of commits, but I haven't had success with it. Also, when I said 'rebase' I didn't actually mean To be honest, I think I'm just used to thinking to myself, "Are my changes going to fix a bug or enhance the codebase?" before I branch. I realise it may take a while to force yourself to do that, though. I also realise you may enhance code while fixing a bug. If in doubt, branch off of the release branch because then you can rebase it onto master (or merge it in) if you end up making a bigger change than you had initially envisaged. As usual, thanks for the hard work an perseverance. |
|
OK, I will be more careful in the future. Sorry for learning via destruction. If you do these moves by hand, PR #1678 should also be marked as v1.2.x milestone. |
But it's the best way! Joking aside, it's not a problem. I'll leave the merging decision to @mdboom, but this gets my +1. |
|
Ok. I'll merge this and then merge it onto master (and just close #1683). That follows our usual procedure. Thanks for getting this all into shape! |
This rolls up PR #1678 and PR #1683 cherry-picked on to v1.2.x.
For future reference, is there a better way to do something like this than cherry-picking each commit by hand? (my naive attempt to re-base blew up as it tried to move all of master over).