DD4hep Meeting March 19th, 2026

Present:     Andre, Juan, Markus, Peter, Thomas

Apologies: 

Link to minutes of previous meeting

Merged PRs

• PR1580:    fix: allow improved G4ParameterisedNavigation for pure 1D translation case (Wouter)

• PR1592:     Geant4ParticleHandler: Keep particles back-scattered from calorimeter...  (Markus)

• PR1595:     User particle hadler: fix for backscatters (Andre)
  ISSUE471:   missing MCParticle for particles scattered back into the calorimeter
              ===> Finally !

• PR1593:     Tests: ignore pseudotrap on aarch64, relax condition for voxelisationtest  (Andre)

• PR1594:     Implement overlayed constant field type limited to a given shape (Markus)

• PR1584:     feat: use axis-aligned bounding box to pre-filter MultipoleField calls  (Wouter)

• PR1589:     Clean up checks since the minimum C++ standard is 17 (Juan)

Closed PRs, closed issues and fixed problems

• ISSUE1588:  Placement of volumes using paramVolume1D
              ===> User error
• ISSUE1582:  Axis-aligned bounding box for MultipoleField

              ===> Fixed in PR1584

Pending Problems

• ISSUE1297:  Possible to dump all available readout elements?
                        ==> Possibly a copy of unsolved ISSUE580 from Valentin about enumerating all sensitive cells.
• ISSUE1335:   Uniform random picking of points on detector surfaces for noise studies
                         This problem is linked to #1297 and #580.
• ISSUE1458: XML file vs TGeoStored geometry in GenFit
• ISSUE1501: Bug/Feature: Condense duplicated particle contributions in calorimeter question
• ISSUE1514: How to convert Geant4TrackID into MCParticle Parent ID
• ISSUE1535: How to add new output variables in Hitsquestion (matclim -> DD4SHip)
                        ===> Debugging ongoing

New issues:

• ISSUE1594:  Test failures for tesselated volumes (Andre)
• ISSUE1597:  Remove LCDD data conversion.cleanup  (Markus)
              Not used since SLIC is not supported. Need to extract vis-extraction.
  
  Pending PR:

• PR1582:   Set the fast simulation flag and adjust step length for contributions created by fast simulation (Thomas)
• PR1587:   Fields/FieldTypes: performance related cleanups (David Lange)
• PR1590:   Fix more memory leaks (Juan)  [NEW]
• PR1598:   fix: randomize non-zero polarization in optical photon gun  (Wouter)  [NEW]

Round Table

Andre:            New tag coming up. 
Juan:.             NTR
Markus:          Somehow need to define a common place with minimal compile dependencies where to take pi, twopi, e, phi, etc from.
                        CLHEP is sort of not the best solution (see also PR1598).
                        math.h/cmath  TMath.h (looks very complete) ....
                        Thomas proposed to wait for the C++ 20 and the "numbers" header file.
Peter:             NTR
Thomas:        NTR

AOB

Next meeting:   April 2nd, 2026