Tidy up after validation

[tdealtry]

Validation almost done! All comparisons were done using a NoTrigger WCSim file and run through complete_comparison.C

• pass_all exactly reproduces WCSim NoTrigger
• nhits CPU exactly reproduces WCSim NDigits trigger (run with save_multiple_digits_per_trigger and /DAQ/MultiDigitsPerTrigger set to both true and false) when running with electron particle gun
  • This was one of the problems yesterday (I forgot what the WCSim default was)
  • Not that there is a ~1E-4 shift in digit times. This is due to the 950 ns offset WCSimWCTriggerBase::offset that needs to be subtracted from digit times in DataOut. If this offset is set to 0, everything is exact to better than 1E-6 (the default precision of complete_comparison.C)
  • I added a parameter DataOut called trigger_offset to change this offset
• Note for now that nhits CPU has some discrepancies relative to WCSim NDigits trigger (run with save_multiple_digits_per_trigger and /DAQ/MultiDigitsPerTrigger set to both true and false) when running with muon particle gun
  • Some muon events have subevents due to decay electron
  • This requires some more changes, related to having to move tracks from trigger 0 to trigger 1
  • I'll get to this...

Anyway, what this PR does:

• Adds save_multiple_digits_per_trigger and trigger_offset options to DataOut
• Sanitise window_end_time in nhits::AlgNDigits()
• Adds a .gitignore
• Adds a pass_all trigger (useful for debugging against WCSim's NoTrigger)

Note that there are changes in my WCSim trigger branch. The major one being:

• Making sure NoTrigger saves everything
  • Negative digits being dropped was the other problem from yesterday

[brichards64]

Hi Tom,

what does pass all Tool do? if its just no trigger then surely thats covered by not having any trigger tools?

With respect to your ReadMe issue, If i institute the policy that i wont merge new tools unless the readme for that tool is filled out, that will help. So could you fill out the readme for the pass all tool :).

does this need another update of your branch on the docker container or not? It seemed to compile without

[tdealtry]

DataOut removes all digits that don't live inside a trigger window pass_all adds a trigger window that runs from ±10 ms

There is an important change in WCSim, but it was just a change of parameter value (hence why it still compiles without the change). Without it, WCSim NDigits & nhits give different answers using the default WCSim.mac (NoTrigger throws some digits away with negative times (so nhits can't see them). NDigits doesn't throw them away)

That policy sounds good

P.S. Enjoy the READMEs!

[brichards64]

Hi Tom,

looks good to me, Iv added an auto update to the WCSim branch so on each PR and docker build it pulls your latest changes and rebuilds your branch. So we don't ahve to think about it (as long as your your branch builds ;) )

PR approved