Only first mesh detected in multiple boolean intersections

[HideoShiraishi]

Hi all,

I’m working with a Grasshopper Python component where I loop over a set of street meshes (provided as a DataTree) and compare them against a single view mesh (provided as a list) using
rg.Mesh.CreateBooleanIntersection() to check overlaps.

However, I’ve run into a puzzling issue:

• Only the first branch (idx 0) correctly reports a successful intersection.
• All subsequent branches return no intersection, even though visually they overlap the view mesh.

In the screenshot:

• The orange-colored rectangles are the street meshes (organized as separate branches).
• The green circular mesh is the view mesh used as the intersection check target.
• The blue-colored patch is the intersection result (what my script currently detects).
  
  

What I’ve already checked:

• Each street mesh is correctly retrieved (I print out bounding boxes, vertex counts, etc., and they match what I see visually).
• The view mesh is a single, fixed mesh used for all comparisons.
• I apply Z-axis flattening (Min.Z or absolute 0) to ensure everything is aligned on the XY plane.
• The intersection call rg.Mesh.CreateBooleanIntersection([street], [view]) always works only once (on the first hit) and then returns None or an empty result for the others.

For testing, I even set up a simple case:

• Four rectangular street meshes laid out in a row.
• One circular view mesh overlapping two of them.

Yet only the first street mesh is ever reported as intersecting.

My questions:

• Does rg.Mesh.CreateBooleanIntersection() have any internal state or caching that makes repeated calls inside a loop unreliable?
• Is there a recommended or more robust way to perform per-branch mesh intersection tests inside a Grasshopper Python component?

Here is the .gh file which I am working with:
test_stClip.gh (31.9 KB)

Any advice or insights would be greatly appreciated — thanks!