PaddleStroke

I don't know. I only moved this code when refactoring the `dropEvent` function. See the file before my PR : https://github.com/FreeCAD/FreeCAD/blob/560898907b317893d90e3a031c41ea1d63109dd9/src/Gui/Tree.cpp#L2321 https://github.com/FreeCAD/FreeCAD/blob/560898907b317893d90e3a031c41ea1d63109dd9/src/Gui/Tree.cpp#L2411

Interesting. I seem to recall that Qt 6.5 was implementing a dark mode that would make things simpler. Is that taking advantage of that? Are we even using 6.5 btw?

> But this makes it worse for everybody else. For them it makes the recompute always slower independent on whether the value is later used or not. Hmm I suspected...

@AjinkyaDahale regarding something you said on the forum. It appears that you had to make alignment geometry draw on top of others is that so? In this PR I was...

I would recommend that the features are left in experimental stage for about 2-3 months so that they can be tested by power users to make sure we iron out...

I think MeasureGui can know about SketcherGui. However that would create a dependency on SketcherGui which is IMO not necessary nor ideal. Besides there are other cases where this could...

So the bug is not on Measure, it is doing things correctly: - You select something it is added in the current references. - Then findType is determining the type...

I had a look but this is quite complex so @bgbsww could you take a look at this bug? It basically happen when you call getShape with a sketch object...

I see no crash on my side either (windows) but the this PR can be merged as it just adds a protection it's totally acceptable.

To achieve what you want, it needs a distance joint of 0 between the 2 conic surfaces. And currently distance joint doesn't work between 2 conic surfaces as the solver...