Particle flow output from PMTrackMaker with unexpected tree structure
I am having problems with PMA output.
I am running
PMTrackMaker producer on a MicroBooNE simulated
cosmicPandora output (
reco_uboone_stage_1.fcl on file generated by
I observe the following flaws in the hierarchy of particles in
- no primary particles; primary particles have parent ID
recob::PFParticle::kPFParticlePrimary; it seems it is set to
- the endpoint particles in the tree refer to their ancestor as daughter; this creates a fatal circular dependency. More in general, there are multiple connections I don't grasp the logic of.
The first point was undocumented (until LArSoft issue #11627).
The second is particularly bad for me, since when I try to follow the tree I get lost in infinite recursion.