Skip to content

TMMT EventBuildingSeq: Add suffix to prevent InputMaker list merging in CA for separate build cfgs (ATR-27050)

Claire Antel requested to merge cantel/athena:23.0_EBSeq_fixcfgclash into 23.0

Purpose

Follow up of !61930 (merged) : Had meant to fix a oldJO config clash along with CA migration, but that was not done successfully. Thanks @fwinkl for pointing that out.

This is a bug fix. Current configs run two input makers for LArPEBHLT chains like HLT_j25_LArPEBHLT_L1J15, which exist in P1 menu.

Validation

I did config comparisons to oldJO config (before merge of !61930 (merged)) for cosmics and physics menu (until my eyes bled).

Expected difference: We have two separate configurations for chains like HLT_j25_LArPEBHLT_L1J15 (RoI based but full scan) and HLT_g80_loose_LArPEBHLT_L1EM22VHI (RoI based, not full scan) whereas before we just had one that was being overwritten.

Previous to !61930 (merged) (CA migration) :

pebSequence_LArPEBHLT
   DetStore = StoreGateSvc/DetectorStore
   EvtStore = StoreGateSvc
   ExtraDataForDynamicConsumers = []
   ExtraInputs = []
   ExtraOutputs = []
   Members = ['InputMakerForRoI/IMpeb_LArPEBHLT']
   ModeOR = False
   NeededResources = []
   Sequential = True
   StopOverride = False

where IMpeb_LArPEBHLT ended up using ViewCreatorCentredOnIParticleROITool.

Current bug:

pebMainSeq_LArPEBHLT
   DetStore = StoreGateSvc/DetectorStore
   EvtStore = StoreGateSvc
   ExtraDataForDynamicConsumers = []
   ExtraInputs = []
   ExtraOutputs = []
   Members = ['InputMakerForRoI/IMpeb_LArPEBHLT', 'InputMakerForRoI/IMpeb_LArPEBHLT_RoIBasedFS']
   NeededResources = []
   Sequential = True
   StopOverride = False

(2 input makers)

This MR:

pebMainSeq_LArPEBHLT
   DetStore = StoreGateSvc/DetectorStore
   EvtStore = StoreGateSvc
   ExtraDataForDynamicConsumers = []
   ExtraInputs = []
   ExtraOutputs = []
   Members = ['InputMakerForRoI/IMpeb_LArPEBHLT']
   NeededResources = []
   Sequential = True
   StopOverride = False
pebMainSeq_LArPEBHLT_RoIBasedFS
   DetStore = StoreGateSvc/DetectorStore
   EvtStore = StoreGateSvc
   ExtraDataForDynamicConsumers = []
   ExtraInputs = []
   ExtraOutputs = []
   Members = ['InputMakerForRoI/IMpeb_LArPEBHLT_RoIBasedFS']
   NeededResources = []
   Sequential = True
   StopOverride = False

Where IMpeb_LArPEBHLT uses ViewCreatorPreviousROITool and IMpeb_LArPEBHLT_RoIBasedFS uses ViewCreatorCentredOnIParticleROITool as was intended in the original configuration, see Athen 23.0.31 EventBuildingSequences.py#L259-268

The only other cfg differences I could spot was in feature collector tool configs, e.g. TrigSignatureMoni.FeatureDecisionCollectorStep5_reco:

Previous to !61930 (merged) (CA migration) :

TrigSignatureMoni.FeatureDecisionCollectorStep5_reco
   Decisions = ['_HLTNav_PEBInfoWriter_LArPEBHLT', '_HLTNav_PEBInfoWriter_IDCalibPEB']
   DetStore = StoreGateSvc/DetectorStore
   EvtStore = StoreGateSvc
   ExtraInputs = []
   ExtraOutputs = []

This MR:

TrigSignatureMoni.FeatureDecisionCollectorStep5_reco
   Decisions = ['_HLTNav_PEBInfoWriter_LArPEBHLT', '_HLTNav_PEBInfoWriter_LArPEBHLT_RoIBasedFS', '_HLTNav_PEBInfoWriter_IDCalibPEB_RoIBasedFS']
   DetStore = StoreGateSvc/DetectorStore
   EvtStore = StoreGateSvc
   ExtraInputs = []
   ExtraOutputs = []

I think this is expected, since it seems like this step is supposed to be configured with a list of all hypos:

Merge request reports