Bugfixes and much more extensive testing #16
Conversation
| w=WeakRef(w) | ||
| gc(); gc(); sleep(.1); gc() | ||
| @assert w.value.handle == C_NULL | ||
| # FIXME: I get a segfault now if I uncomment the next lines |
There was a problem hiding this comment.
I don't get a segfault here
There was a problem hiding this comment.
Now I'm not getting it either. Weird. It was quite reliable (3x in a row), but it had never happened before, either.
There was a problem hiding this comment.
strange. are you sure you weren't getting the segfault further on (in signal_emit), when you were feeding Gtk uninitialized memory?
There was a problem hiding this comment.
I can't remember whether I was running the whole script or just the top part. But that explanation seems so plausible that I certainly can't deny it; perhaps just the randomness of what memory got fed.
Didn't realize I had to construct the event itself, that's good to know (and makes sense).
|
looks good to me. i'll leave it open in case you wanted to add anything else before you merge |
|
I'm done for the day, so feel free to merge or just let sit until tomorrow or later. There are a few more exported items that should be tested. |
No description provided.