Burst summary mis-counts patterns in mulc_dit

[leafybillow]

Environment: (where does this bug occur, have you tried other environments)

• branch (master for latest released): prex_respin1
• revision (HEAD for most recent): HEAD
• OS or system: linux
• Special ROOT or Geant4 versions?

Steps to reproduce: (give a step by step account of how to trigger the bug)

1. ./build/qwparity -r 3305 -c prex_prompt.conf where a second combiner map for dithering correction is added in datahandler array.
2. burst_mulc_dit->Scan("dit_asym_usl.hw_sum : dit_asym_usl.num_samples")
3. mulc_dit->Draw("dit_asym_usl","BurstCounter==1 && ErrorFlag==0")

Expected Result: (what do you expect when you execute the steps above)

We expect number of entries in histogram matches number of samples in burst summary tree.

Actual Result: (what do you get when you execute the steps above)

It seems burst summary lost one pattern from the second burst and ones afterwards, while the histogram drawn from mulc_dit tree looks normal. The burst summary for 'mul' and 'mulc' has no such problem.

Burst summary missing one pattern in the second burst and afterwards. Also central value mismathced with that from histogram draw.

Histogram drawn from tree looks normal.

[paulmking]

@leafybillow Can you check the result of adding a device_error_code test on the dit_asym_usl plot from the tree? It may be that one of the beamline channels which contributes to dit_asym_usl is not in the global error definition.

[leafybillow]

@leafybillow Can you check the result of adding a device_error_code test on the dit_asym_usl plot from the tree? It may be that one of the beamline channels which contributes to dit_asym_usl is not in the global error definition.

Ok, with device_error_code flag, now the numbers agree with each other. @paulmking And the event cut of this run is defined by prexCH_beamline_eventcuts.2948-.map where regression/dithering IV channels are not fully included. I think we need to review the event cuts in this run range. @Gongzi-Zhang