Ah. Old fashioned printf/gdb debugging. I did many years of that.
Looking at my previous gdb backtrace, I see that the last breaths of the tool were taken on QWidget, QObject, QStandardItem etc. So i agree that the final crash is buried in Qt. However the situation that sets up the crash is probably created far removed from the crash point, which makes it difficult to find. I suspect a damaged tree. I would start by looking at the manipulation of the tree in the Widget. But, given that the tree is manipulated with Qt, the error is buried inside Qt anyway. Sigh.
I will fool around with this for a while to see what I can puzzle out. However, I am:
1. unfamiliar with Qt
2. unfamiliar with Sigil
3. unfamiliar with Cmake
4. stale in gdb skills
5. weak in c++ skills (java guy)
so don't hold your breath while I do this. You might turn blue.
Banjo
(_)=='=~