Since this new version (or attached frameworks), folding code is not working correctly with python files. It becomes simply unusable. It detects nodes in many many places where there are not, and it becomes impossible to fold/unfold the true nodes. Clicking on a wrong node doesn't make it disappears anymore.
I suggest specifying your version of KDE Frameworks, since there's no such thing as "attached frameworks".
Already fixed: https://cgit.kde.org/ktexteditor.git/commit/?id=6a35a902c9697b16376fc68848fee5b978881b71
Perhaps that was fixed, but actually one would need some example what precisely was wrong.
Hm, the change is unrelated. Can you provide a screenshot?
Darn, let's re-open then.
Created attachment 115022 [details] Screen capture of python code and nodes
Yep, pretty sure that was fixed by https://cgit.kde.org/ktexteditor.git/commit/?id=6a35a902c9697b16376fc68848fee5b978881b71, but I'll let Dominik confirm. :)
So I use a neon based distribution, currently with KDE Frameworks 5.50.0, kate version is 18.08.1. I didn't have time for coding this summer, so I can't know sinche when this problem is here. On the screenshot you can see that there are many nodes for folding codes that shouldn't exists. Previously, a wrong node was something that could appears, but clicking on it made it erase. Now It doesn't make anything.
Yeah, that is fixed. But just to confirm: you can still fold all things that should be foldable, the only issue that is there is that some lines have markers that have nothing to fold (as e.g. for (......) the region ends in the same line as it starts).
I can fold/Unfold true nodes only in the rare case that there is no wrong node inside. If there is one or more wrong node inside, click on a true node makes nothing.
OK, so then this is indeed already fixed by https://cgit.kde.org/ktexteditor.git/commit/?id=6a35a902c9697b16376fc68848fee5b978881b71. :)
Sorry for the noise :-)
Not a problem!
*** Bug 398840 has been marked as a duplicate of this bug. ***
*** Bug 398845 has been marked as a duplicate of this bug. ***
*** Bug 398884 has been marked as a duplicate of this bug. ***