ATLPHYSVAL-797 add Detailed content for GSFTrackParticles. Includes Parametes and TRT hits variables

This merge request corrects an oversight where EventInfo was not added to the AllVariables list in DAOD_PHYSVAL, as it now needs to be given that EventInfo is handled by the usual slimming machinery since !44456
Without this MR, EventInfo will be completely absent from PHYSVAL.

ATLASG-1600

The DAOD_PHYSVAL format should have been updated in !44343 to accord with the re-organisation of the imported modules, but this was forgotten so the format broke in master. This MR fixes that by adding the appropriate method calls.

Arnaud Duperrin authored 3 years ago and Edward Moyse committed 3 years ago

Add btagging of AntiKtVR30Rmax4Rmin02PV0TrackJets VR jets in DAOD_PHYSVAL derivation (using update/fix in ATLJETMET-1326 from Jet/Met group) + allow mutiple jet collections btagging (PFlow and VR)

Dan Guest authored 3 years ago and Walter Lampl committed 3 years ago

The idea is to return a component accumulator for each jet collection, so that
we can merge all of them together. There are still some points that need to be
cleaned up a bit, e.g. the calib service is just scheduled once, not in the
accumulator. Also it crashes right now...

This MR imports the AtlasTrackingGeometrySvc into DAOD_PHYSVAL to allow it to process data. The service is needed by the flavour tagging component of the job. In DAOD_PHYS other components pull it in, but these components aren't used in PHYSVAL, hence the need for this addition.

ATLPHYSVAL-696

The adds the DAOD_PHYSVAL format to master, enabling derivation-level containers to be validated using the same machinery as is used for AOD. The format has been completely re-written versus the r21 version, starting from DAOD_PHYS and adding any/all low-level variables needed to make the physics validation n-tuple maker run.

This MR makes a small fix to the TriggerMatchingHelper such that it is compatible with Python 3. It restores the remaining lines in DAOD_PHYS needed to restore the trigger matching decorations to DAOD_PHYS.

With this change the trigger matching information as produced by DAOD_PHYS in release 22 should now be equivalent to release 21.

This is the first step in getting trigger matching to work in DAOD_PHYS in master. It restores the machinery for getting the required list of trigger that was broken by changes to the HLT menu metadata in master. It does not yet restore the trigger matching itself, which will need further work. Consequently this MR has no impact on the output of DAOD_PHYS, because the active lines are still commented.

https://its.cern.ch/jira/browse/ATR-21817

The way the trigger matching is set up in DAOD_PHYS no longer works because of a change in the trigger menu metadata. In order to allow the validation with DAOD_PHYS to proceed the relevant lines are commented out.

The version of SGxAODProxyLoader in master, which defines the type for each container in the ExpressionEvaluation, has not been touched for four years. In consequence it misses several containers that were added later (particularly the DiTauJets) This MR sweeps the version from 21.2 into master. DAOD_PHYS is adjusted to uncomment some lines that were broken by this issue and are now able to work.

Hello,

This MR enables the on-the-fly reconstruction of low-pt di-taus in DAOD_PHYS.
A small change was made in the DiTauRec tool scheduling, to avoid a "spurious JetAlgorithm addition" to AlgSequence, which could cause some conflicts when scheduling ditau reconstruction on-the-fly, depending when the DAOD_PHYS AthSequencer is added to DerivationFrameworkJob.
Other small changes: the import of `DerivationFramework__CommonAugmentation` and `DerivationFramework__DerivationKernel` was commented out, it did not seem mandatory, please let me know if that is wrong.
Also, the `DerivationFramework__DerivationKernel` was added to `DerivationFrameworkJob` instead of `SeqPHYS`, in case that was intended, I am happy to put this back.

Besides, TauCommon has been better structured, so that functions can be imported and more specific actions can be made, such as decorate only the desired tau ID WP, instead of all 4 WPs.

What is not functional yet is the thinning of low-pt ditaus. No matter what I tried, I end up with `ERROR MultipleProxyLoader: unable to find valid proxy loader for DiTauJetsLowPt.nSubjets`.
Not sure why, the ditau EDM is defined in xAODTau, and that was working in 21.2. To be followed up.

Cheers,
Bertrand

Hello,

This MR imports from 21.2 the functionality to build truth taus when running on xAODs.
At the moment, the production of TruthTaus from EVNT does not work, because for that we'll need to migrate DerivationFrameworkMCTruth/src/TruthCollectionMakerTau.cxx.
The TruthTaus collection is now enabled in MCTruthCommon.py, and in DAOD_PHYS as well.
I took this opportunity to import the smart slimming lists for DiTauJetsCP and DiTauJetsLowPt.
Some functionality in TauCommon has also been imported from 21.2.
The on-the-fly scheduling of DiTauJetsLowPt does not work yet, due to a missing dependency in DerivationFrameworkJetEtMiss/python/ExtendedJetCommon.py (addCHSPFlowObjects).
Not sure if this workflow is supposed to be different in R22.

Naive remark: the truth tau building has not been extensively validated. To come: the comparison of TruthTaus built from EVNT, in R21.2 and R22, when we'll have TruthCollectionMakerTau (the comparison of TruthTaus on xAOD is more cumbersome).

Cheers,
Bertrand

This MR makes use of !33167 and !33139 and introduces MC truth into DAOD_PHYS in master. Necessarily a number of components have been commented out as they rely on other software that has not been migrated into release 22 and will need to be added in further MRs. The missing components are as follows. If no file name is provided the commented lines are in MCTruthCommon.py.
- Q/G decoration jets (TruthDerivationTools.py)
- TruthMetaDataWriter
- TruthCategories (PHYS.py)
- CopyTruthJetParticles
- addAntiKt4TruthDressedWZJets
- truth_groomed
- scheduleTauTruthTools
- GenFilterTools in GenFilterToolSetup.py
- addTruthEnergyDensity
- TauTruth (PHYS.py)
- LHE3 weights (PHYS.py)
- addTruthJets (PHYS.py) - seg fault if uncommented
- TruthQGDecorationTool
- TauTruth omitted from TruthNavigationDecorator
- addLargeRJetD2 (complains about missing AntiKt10...R20Jets

This MR cherry picks the 21.2 MR !26222 into master, ensuring that the TauTrackParticleThinningTools in master, recently updated to include DataHandles and the thread-safe thinning machinery, also have the TauTrack thinning features of the 21.2. version which were implemented in !26222
This MR allows the tau track thinning to be restored in DAOD_PHYS.

The tool InDetTrackSelectionToolWrapper is required to allow the derivation framework to decorate InDetTrackParticles with track selection flags and thereby allow their selection. This MR sweeps the tool from 21.2 to master. In consequence of this MR the full set of InDetTrackParticle thinning criteria can be applied in DAOD_PHYS; this is also set in the MR.

This MR restores the thinning to DAOD_PHYS, removing the lines relevant to the old thinning service, which no longer exists, dropping the thinning helper which is no longer needed, and passing to the thinning tools the stream name rather than a service. Some aspects of the thinning tools still need to be synchronised with 21.2 and so some lines are commented out. An as-yet unmigrated tool called in MuonsCommon is also commented out until the package is swept into master.