TrackParticle mass based on hypo /fitter
The xAOD::TrackParticle mass has been more or less useless for "exclusive" /lepton hypotheses. I would have hoped no-one was using this method for leptons/exclusive things. It makes sense when we look inclusively e.g most things produced by InDet software are indeed assumed to be pions
But since what xAOD::Tracking does (which as said is pretty useless/confusing/wrong for exclusive/lepton cases)
seems to have managed to puzzle even the InDet Tracking convener in ATLMCPROD-10781
I also learned that GSFTrackParticles have the pion mass set, which I did not expect.
Lets try to change it as much as we can without breaking FT0
To be formally correct on the issue, for starters there is no mass setting going on for xAOD::TrackParticles. What happens is that :
-
Something like the GSF passes xAOD::electron here [https://acode-browser1.usatlas.bnl.gov/lxr/source/athena/Reconstruction/egamma/egammaAlgs/src/EMBremCollectionBuilder.cxx#0249]
-
Then this correct hypothesis is set. [https://acode-browser1.usatlas.bnl.gov/lxr/source/athena/Tracking/TrkTools/TrkParticleCreator/TrkParticleCreator/TrackParticleCreatorTool.icc#0013]
-
A mass is never set for xAOD:TrackParticle. There is not setter
-
Then the xAOD::TrackParticle mass just always returns charged pion mass.... no matter the hypothesis
Will this break something , who knows?
Lets assume that specific "exclusive" fitters (GSF / iPat ) are used only in the specific lepton context (this prb is reasonable). Then we can also assume and return the specific lepton mass for these .
PS : this seems to be there for quite some time (19.2.X?) [https://gitlab.cern.ch/atlas/athena/-/blob/archive/19.2.5/Event/xAOD/xAODTracking/Root/TrackParticle_v1.cxx?ref_type=tags#L92]
PS1 : https://gitlab.cern.ch/atlas/athena/-/blob/archive/19.2.5/Event/xAOD/xAODTracking/Root/NeutralParticle_v1.cxx#L85. [https://gitlab.cern.ch/atlas/athena/-/blob/main/Event/xAOD/xAODTracking/Root/NeutralParticle_v1.cxx?ref_type=heads#L54]
Is a bit more amusing in some sense . And prb will be for another MR. Why pion +/- and not pion 0 134.976, aka actually a mass of a neutral...? not sure ...