particle set gets reordered

[Bijan Chokoufe Nejad]

I thought that we have removed all reordering of the particle set before event output occurs, but maybe this got reintroduced with the rescanning stuff? For the context, stdhep requires children to be adjacent and has two numbers to identify the start and the end of the children.

With the given sindarin, we see in the .debug output that after hadronization the particle set has the correct ordering (the H is at pos 5 and 14 with [5] -> [10-14], [14]->[] this is a bogus copy but with status virtual so its fine). In the .lhe as well as in the .stdhep (I attach output from stdhep_rd), the Higgs is at position 3 and 27 with [3]->[6-27] and the pions in between.

[Juergen Reuter]

So which format is failing now? Mo Xin said LHE is ok.

[Bijan Chokoufe Nejad]

Please read the ticket before posting. Also, Mo Xin did not say LHE is ok in his mail. He just said it's broken in stdhep. As far as I can tell, the particle set is identical between lhe and stdhep and different from the ordering in the debug output that is generated by the hadronization evt.

[Juergen Reuter]

He said this a few hours ago to me personally. I guess we just close this ticket, and admit that the output is as it is.

[Bijan Chokoufe Nejad]

Maybe he didn't recognize it because in LHE you see the mothers and not the children. As you can see from the files, the particle set is obviously mangled. I was suspecting, it happens here

    call pset_in%apply_keep_beams (pset, keep_beams = keep_beams)

which is in hepeup_from_particle_set and

    call particle_set%apply_keep_beams (pset_tmp, keep_beams = keep_beams)

which is in hepevt_from_particle_set. But apply_keep_beams should only use the infamous get_real_parents/children when the optional real_parents is given and true, which doesn't seem to be the case ever.

I don't know where else to look but if you prefer not to fix, I don't care.

[Juergen Reuter]

Of course I want this fixed, but I also have no clue. Particularly not where it broke down (was it ever right?) WK??? What shall we do?

[Juergen Reuter]

What about the Higgs boson in this event sample? It has not been decayed, neither by WHIZARD nor by PYTHIA, how can that be?

[Bijan Chokoufe Nejad]

? Yes it has. By Whizard. The only undecayed Higgs is the virtual copy that pythia has made as described above. Or at what particle number do you see any more?

[Juergen Reuter]

Oh, I see now, I overlooked one of them. There is one with status [r] as entry 5 and one with status [v] as entry 14. The latter one doesn't have children, and is probably the one from PYTHIA. Wouldn't the solution be to filter out that one before writing to the event common blocks?

[Juergen Reuter]

Actually, this would probably interfere now with Christian's changes for Mo Xin which have to be revised in any case.

[kilian]

Who's now actually in charge of this? Christian?

I don't interfere right now (won't be able due to meetings anyway). From the comments above, it looks like we're now debugging Pythia 8 ...? That program appears to act smarter that it is supposed to be.

[Juergen Reuter]

So my suggestion is that CW, BACN and JRR are discussing these things when I'm back in the office which will we be Wednesday morning. This is not sooooooo super-urgent.

[Juergen Reuter]

This has been (re)solved in r7327. Thanks to CW, and to Mo Xin for reporting. Closing for now. Please reopen in case of new issues.