Add module to read recob::Wire back into dataprep.

[dladams]

The dataprep modules read raw data or digits and write recob::Wire. It would be useful if that structure could be fed back into dataprep, in particular, so dataprep could be used to further process the output of the Wirecell reco module. Slavic (@vgalymov) has recently requested this for his studies.

The most likely resolution here will be to modify one or add to the modules in the dataprep module directory.

[dladams]

I asked Slavic for an example input file and received this response:

You could grab this one:
/dune/data/users/vgalymov/vd/cb3/preprod/1727_3_a_cb_reco_63114101_0_2022-11-24T065140Z.root
it is from a pre-production set of coldbox data, so it is also on pnfs:
root://fndca1.fnal.gov:1094/pnfs/fnal.gov/usr/dune/tape_backed/dunepro/vd-coldbox-top/full-reconstructed/2022/detector/test/VD_coldbox_CRP2_CRP3_2022/00/00/17/27/1727_3_a_cb_reco_63114101_0_2022-11-24T065140Z.root

The recob::Wire product of interest is "wclsdatanfsp:gauss", which is what comes out of the wirecell.

Since the root file has also the raw digits, one can in principle make a comparison for ROIs found through dataprep and wirecell.  It is sufficient to put
source:
{
  module_type: RootInput
  maxEvents:  -1        # Number of events to create
}
and then your top fcl should run as before. For example:
In /dune/data/users/vgalymov/vd/cb3/preprod/vdt2proc-cnr-rootinput-digit.fcl

Thanks Slavic. This raises the question of whether we want to read the wires (meaning recob::Wire data) alone or in combination with the raw data. Or both.

[vgalymov]

That is a good question. Originally I envisioned two separate workflows: one where one runs through digits with some set of dataprep tools and the other when one reruns with wires with some set of tools (e.g., ROIs display and storage). Then offline one would try to compare the two results in some way. The benefit of being able to run a combination of the two at the same time might perhaps be that it would allow for an eventual possibility to make a tool where the two information could accessed and used.

[dladams]

I was distracted this week preparing to present talks on the new channel status tool/service and did not make ny progress here. But this remains a high (but not highest?) priority issue.

[dladams]

Slavic, sorry I have been distracted with vacation and many meetings but am back looking at this and hope to get something in place next week. LMK if you are no longer interested.

[vgalymov]

It is not at all urgent, but I think would be a good feature to have available eventually.