mg5 generator issue summary...

[wdconinc]

Environment: (where does this bug occur, have you tried other environments)

• branch (master for latest released): develop
• revision (HEAD for most recent): HEAD e40614d0f1865bb66dba012ea1005f5d1c638f62
• OS or system: gfortran-8
• Special ROOT or Geant4 versions? latest mg5 from bzr on launchpad

Steps to reproduce: (give a step by step account of how to trigger the bug)

1. cd generators/MG5/ee && mg5_aNL proc_card_mg5_ee_ee.dat
2. cd ee_ee && bin/generate_events run_01 -f
3. gunzip Events/run_01/unweighted_events.lhe.gz
4. lhef2hepmc Events/run_01/unweighted_events.lhe Events/run_01/unweighted_events.hepmc
5. build/remoll macros/hepmc/ee_ee.mac
6. build/reroot -l remollout_ee_ee.root
7. T->Draw("part.p:part.th/deg")

Expected Result: (what do you expect when you execute the steps above)

• No nan, but ee_ee produces nan for xsection.
• Cuts on CoM angles between 30 and 150 deg, but corresponding eta = -ln tan th_com/2 doesn't seem to result in anything
• Similar procedure on ep_ep returns actual xsection
• ep_ep only knocks out u valence quarks, no d valence quarks (but it does do u,d,~u,~d sea quarks) (at this point I'm not worried about hadronization which, technically, could be done with pythia)
• ep_ep needs energy minimum on the lepton side to get anything on detectors

[wdconinc]

Fixed target cuts ala https://github.com/JeffersonLab/hps-mc are hard to maintain with upgrades to mg5. Probably needs a patch file to update the cuts.inc in Source and cuts.f in SubProcess, which needs to be applied before bin/generate_events.

[wdconinc]

The nan in the MG5 output is not cross section but alpha_s which is indeed undefined below Lambda_QCD which is where we are when moller scattering (based on Mandelstam s).

Cuts on CM angle are better implemented through pT, not eta. That works as expected now.