View Single Post
Old 12-22-2019, 07:12 PM   #65
KevinH
Wizard
KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.KevinH ought to be getting tired of karma fortunes by now.
 
Posts: 4,251
Karma: 2621106
Join Date: Nov 2009
Device: many
Yes, I agree, it is probably due to signals not being properly disconnected when the current tab is closed. Which I think is related to the fast delete of tabs "fix" issue we had earlier in 0.9.991 alpha.

So my fix probably broke tab signal disconnect in MainWindow.cpp.

Arrrrggggghhhhh!

Kevin


Quote:
Originally Posted by DiapDealer View Post
No. It doesn't seem to with similar steps anyway. Tested 0.9.18 on Windows with Qt5.12.3 and Linux with Qt5.13.2.

The bug hitting Sigil 1.0.0 on Linux with Qt5.13.2, but not with Sigil 0.9.18 on Linux with Qt5.13.2 would seem to clear Qt of any fault, I think.

EDIT: Debug output shows that FlowTab::SplitSection() is definitely being called multiple times. As is MainWIndow::CreateSectionBreakOldTab()

TabManager's OldTabRequest signal is hardwired to FlowTab's OldTabRequest signal. FlowTab emits its own OldTabRequest signal in SplitSection(), and MainWindow connects TabManager's OldTabRequest signal to it's own CreateSectionBreakOldTab method.

Not sure if that helps, but it made my head hurt to follow that all around!
.
It's almost like closed tabs aren't getting properly disconnected from signals or something (that's a WAG on my part, though).
KevinH is offline   Reply With Quote