Sharing code between SBND and ICARUS

PAGE IN DEVELOPMENT. This page is dedicated to advice about using code from one experiment in the other. The hope is by noting things that change or may change going from one to the other will help avoid pit falls and headaches in the future. If you're stuck, don't hesitate to reach out! You can ask questions on various general purpose (e.g. #general) or analysis-related Slack channels. Likely some people have run into this issue before and found a fix, so it can be better to ask and save time than reworking everything! In particular, Yun-Tse Tsai (\@Yun-Tse) has done a lot of work on exchange of code between SBND and ICARUS and may have some insight.

Differences in icaruscode vs. sbndcode

Includes/Services between experiments (especially .fcl includes)

The fcl includes, services, etc. are different between the two code bases. For example, sbndcode has four services tables, two of which are most important to this discussion:

• sbnd_basic services: a very basic set of services that has very few functionalities
• sbnd_services: a more complete set of services (though there exist others too) that builds on the basic_services and adds many more features.

whereas icaruscode has

• icarus_basic_services: a set of services which has significantly more functionality than sbnd_basic_services, more like sbnd_services though not the same.

though depending on your specific needs you may need other services (simulation, wire calibration, etc.). While sbnd_services cannot be directly replaced by icarus_basic_services in all instances, this is probably a place to start. For more information on the services provided by the tables, see job/services_icarus.fcl in an icaruscode release (fcl/services_sbnd.fcl in an sbndcode release).

On the include side of things, there are three general categories of these. In some cases, an SBND-specific include maps directly to an ICARUS-specific include. In other cases, there is no ICARUS-specific include, but a base include can be used. In still other cases, an include used in SBND code appears to have no equivalent in ICARUS code. Below is a non-exhaustive list of such includes. The easiest way to see if your module is listed to do a ctrl-F (apple-F on Mac) for the include you are looking for:

Includes with direct ICARUS counter-part SBND includes with no ICARUS-specific analog, but a basic include Includes from SBND with no equivalent found ——————————————————————- ——————————————————————- ——————————————— hitfindermodules_sbnd.fcl -> hitfindermodules_icarus.fcl caldata_sbnd.fcl -> caldata.fcl rootoutput_sbnd.fcl cluster_sbnd.fcl -> cluster_icarus.fcl databaseutil_sbnd.fcl -> databaseutil.fcl sam_sbnd.fcl trackfindermodules_sbnd.fcl -> trackfindermodules_icarus.fcl particleid_sbnd.fcl -> particleid.fcl
calorimetry_sbnd.fcl -> calorimetry_icarus.fcl
showerfindermodules_sbnd.fcl -> showerfindermodules_icarus.fcl
pandoramodules_sbnd.fcl -> pandoramodules_icarus.fcl
simulationservices_sbnd.fcl -> simulationservices_icarus.fcl
services_sbnd.fcl -> services_icarus.fcl
simulationservices_sbnd.fcl -> simulationservices_icarus.fcl

TODO: cases where ICARUS includes have no direct counterpart in SBND?

Difference in module/product names between experiments

Often ICARUS modules and the products they create may differ from those used by SBND. The general procedure is quite similar, but in ICARUS we run 2 reconstruction steps, for example. One stage (reco1) is the "Gaus" or "Gauss" stage that uses Gauss deconvolved TPC hits and is more akin to what is used by SBND. The other stage (reco2) is the "Raw" stage, which uses the raw digits instead of performing a deconvolution. This has effects both on what you get in terms of files, but also in terms of the products available. Instead of just Pandora products, now there are PandoraGaus products (from the deconvolved reconstruction) and PandoraICARUS products (from the raw digits). In addition, the code also currently has versions split between the two active volumes, Cryo0 and Cryo1, so the following mappings exist for example:

sbndcode icaruscode (Gaus, example) icaruscode (Raw, example) ————– —————————– ——————————- pandora pandoraGausCryo0, etc. pandoraICARUSCryo0, etc. pandoraTrack pandoraTrackGausCryo0, etc. pandoraTrackICARUSCryo0, etc.

Detector agnostic code

In general, code pulled from the master branch of the LArSoft git repositories should work regardless of the experiment on which you are working. However, in a feature branch, this may not be true. Or, if you are trying to use code housed in SBND specific areas in ICARUS, for example from sbndcode, this will also not necessarily be the case.

One particular cause of frustration can be if code uses detector-specific plane information, since the wire planes are oriented differently between SBND and ICARUS. A signature of a mistake of this type is an error message that looks like the following:\

%MSG-s ArtException:  PostEndJob 12-Nov-2019 09:23:26 CST ModuleEndJob
cet::exception caught in art
---- OtherArt BEGIN
  ---- EventProcessorFailure BEGIN
    EventProcessor: an exception occurred during current event processing
    ---- ScheduleExecutionFailure BEGIN
      Path: ProcessingStopped.
      ---- OtherArt BEGIN
        ---- TPCGeo BEGIN
          TPCGeo[0x5ed6970]::Plane(): no plane for view #2
          cet::exception going through module 
        ---- TPCGeo END
      ---- OtherArt END
      Exception going through path reco
    ---- ScheduleExecutionFailure END
  ---- EventProcessorFailure END
---- OtherArt END
%MSG
Art has completed and will exit with status 1.

This particular failure mode was one specific call to fGeom->WirePitch(geo::kW) in code originally used with SBND.

General advice

Three pieces of general advice:

1. If you try to run an art (LArSoft) job but the includes aren't right or something isn't being located, it will generally complain and tell you what it can't find. You can use these print statements to guide your debugging!
2. Remember that Slack is just a click away!
3. If you find something that isn't on this page, add it. Then it's there for the next time!