Skip to content

keep dd4hep id as is from gean4 in FT

Menglin Xu requested to merge mexu_g4detit_FT into master

  • Issue: Detector!556 (merged) changes the FT main volume to be an Assembly instead of a Volume with a defined solid, in order to avoid overlaps between the FT and MP. However there was a crash in the Gauss test gauss-2023-dd4hep (details). @tlatham found it is about GetTrackerHitsAlgBase::fillHit

  • Reason: !979 (merged) allows us to use the same method for DetDesc and DD4Hep as well as the generic sensitiveDetector implementation for DD4Hep. Before calling it it checks the hit isInside the FT sensitive detector, but the DD4Hep implementation does not seem to work with the assembly...

  • Solution: The implementation of sensitiveDetector needs to be overwritten by each subdetector and that is not the case for FT at the moment. We have 3 templates of GetTrackerHitsAlg after !979 (merged), one for DetDesc, one for DD4Hep and a third for the native Gaussino geometry defined via python where no DeElement is implemented. For DD4hep we can configure GetTrackerHitsAlg to:

    • retrieve it from DD4Hep that now has the same method as DetDesc
    • keep what Geant4 has internally (and that is the coded value from DD4hep) as done in !1063 (merged) for the Muon system

  • Questions: are these changes ok in Boole? @zexu @lohenry (close #130 (closed)).

  • Following: If it is fine for FT in Boole, to not block MP merge requests, we can merge this request in, but then need a wider discussion with all detectors including U2 detectors

Edited by Gloria Corti

Merge request reports