When I did the cpp version above I got an error trying to push something with ’ around it onto the string-vector, could there be some problem due to how python represents strings?
Everything looks fine here, and the C++ version works correctly.
I’ll need to look into this a bit more closely: could you provide a very small file (even 1 event is fine, if the problem presents itself with 1 event) that I can use to reproduce the issue?
Actually, I’ve restarted the notebook and now I no longer get the original problem: it just seems to hang forever. The cpp example took a few minutes at most, but the notebook is now stuck on the snapshot line at least half an hour, and I have probably let it run for two hours at least once.
I’m not sure what I’ve changed, maybe there was some old variable shadowing good_columns before the restart.
Ok, good I guess, unless the hanging you see now is a different manifestation of the same issue
You can try not to call ROOT::EnableImplicitMT to disable multi-threading and to prepend a Range(10) to the Snapshot to run on very few events, and to snapshot only one branch. If that hangs for half an hour, we have a problem – one that we can only debug with a reproducer in hand.