ghep tree takes the entry corresponding to the EventId

[giacomo-santoni]

The EDEPSIM step takes just the events with hits, as in the .mac files there is the condition: /edep/db/set/requireEventsWithHits true. This could lead to have less events in the edep files with respect to the ghep ones (gtrac files instead are copied into edep files maintaining a 1-1 correspondence). This possible mismatch could be a problem in the spill-build step: when the interaction time is computed, flux information from the ghep files are taken, so we need to have the right correspondence between edep and ghep files.
With this PR I have just changed a line in the spill-build code: instead of taking the entry I take the corresponding EventId

[alexbooth92]

Thanks @giacomo-santoni! Before merging, I like to discuss this a little more.

Firstly, if we decide to go with this change, the spill building macros that we have used historically (the various .C) should also be updated.

Separately, @jdkio made a comment in a CAFMaker scrum meeting a few weeks ago relating IRRC to the possibility of a GENIE index / EventID skew coming from the use of rock box? Would this not have shown up in here?

[giacomo-santoni]

Hi @alexbooth92, sorry for the late response.

Concerning the first point, actually the ghep - edepsim matching is present only in the "...WithNuIntTime" macro, where ghep files are needed to compute the nu interaction time. In the other macros, there is a gtrac - edepsim matching which it's ok.

About the second point, is there any additional information I can have a look at?