Filtering content by layer is not reliable

My definition filters referenced objects by layer.
During a presentation, the definiton started misbehaving, and the problem was narrowed down to objects from a certain layer no longer being output by the content filter, just like if that particular layer didn’t exist anymore.

Was workink like a charm minutes before.
A recompute failed to do fix the problem, so I referenced the objects again, but to no avail.
What fixed the problem was to close and re-open the Rhinoand GH files.

Can’t share the curves publicly, and I gess that it would not help much anyways, because it happens randomly.

Has anyone else encountered glitches with this component ?

Here’s an example, but I’m sure that the problem will not manifest itself on your side…


240626_PROG_Filter content bug.gh (4.8 KB)

EDIT : I just tried by setting the “Case sensitive” input to "False, and now it seems to work reliably.
This is buggy because I scrupulously checked the cases and they were identical.

Can you send us the file but empty of the curves. You can PM it to me if you want.

I want to test the Layer name capitalization against the character set being used.

I can fill in the layers with the curves to test.

OK, sending the Rhino file in PM

So, normally this is some cache or list filling up over time. I expect it is a bug. Is there something that you were doing a lot during the presentation? I would need to search for ways to create the overflow.

Was the model changing a lot? Or were layers being turned off and on? Lots of models being opened? Curves being edited a lot.

No, just me yammering about how good my definition was, until I made a complete fool of myself as the data wasn’t going though anymore and everything brok e a p a r t.

Yuk. We will see if there is a hint in the capitalization code.