Time in Hits events produced with TRestDetectorSignalToHitsProcess

[DavidDiezIb]

I've seen that all four methods to produce Hits from Signal in TRestDetectorSignalToHitsProcess (https://github.com/rest-for-physics/detectorlib/blob/master/src/TRestDetectorSignalToHitsProcess.cxx) have time component manually set to 0: fHitsEvent->AddHit(x, y, z, energy, 0, type);
I'm not sure if this has any purpose but would be useful to have again a time value related with the time bins of the original RawSignal.

[jgalan]

This time at the hits structure was introduced by @ddn2 to be used in Garfield simulations so that we could keep the value of the drift time, as I am recalling it right now.

Since it is unused right now, I believe initialising the time-hit member to the physical time value would not be a problem.

Still, it is kind of redundant information, because one could always retrieve the physical time using the drift velocity stored at TRestDetectorSignalToHitsProcess, and the TRestDetectorReadout definition. But since we are already storing the hit-time.

A method TRestDetectorHitsEvent::GetHitTimePosition( Int_t n, Double_t driftvel, TRestDetectorReadout &readout ); could also be considered.

What do you think @rest-for-physics/core_dev ?