Why does yt clump_find crash on native Parthenon data?

0
0
Asked By VelvetKite42 On

I'm trying to run yt's clump_find function directly on a Parthenon .rhdf file, but it crashes as soon as it starts generating the first contours. Has anyone run into this with native Parthenon data, and is there a way to fix it without regridding? Regridding avoids the crash, but it also reduces the resolution and loses detail.

2 Answers

Answered By CopperMango7 On

This sounds like a possible edge case in yt’s contour-linking code rather than a problem with clump finding itself. The crash appears to occur in link_node_contours, possibly inside construct_boundary_relationships, where contour bounds, joins between objects, or the calculated contour count could be going wrong. Try rebuilding the relevant Cython code with optimization disabled, such as -O0, so bounds and indexing checks can expose the exact failure. A traceback showing the precise Cython line would help distinguish a bounds issue from a bad contour relationship.

Answered By QuietLantern18 On

Regridding can work around it, but it’s definitely not ideal when preserving the native resolution matters. I’d investigate the contour-linking routine first instead of treating the regridded result as the real fix.

Related Questions

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.