TrigInDetValidation ART tests - change TrigInDetAna flag "depends_on_previous"

See merge request !37899

InDetValidation: cmake cleanup

See merge request atlas/athena!37902

Dump all variations in BookkeeperDumperTool

See merge request !37856

JetTagging: cmake cleanup

See merge request atlas/athena!37901

HltEventLoopMgr: Add option to switch from pop-all mode to pop-one

See merge request !37885

Preliminary work on optimising for identical Conditions in the jet hypo

See merge request !37767

Peter Sherwood authored 4 years ago and Edward Moyse committed 4 years ago

Job group propagation is correct but inefficient in the presence of identical  conditions.

Jet group propagation is currently handled in the following manner.

Let a parent condition P have n siblings: S_1... S_n
Let J_1... J_2 be the jet groups that pass each  of the n siblings.
JetGroupProduct finds all the distinct job groups formed by selecting one jet group from is taken fron each of
the J_i. Only distinct jet groups are retained. Distinct job groups are those obtained by flattening the JobGroup
calculated by the combination process, and discounting differences due to the order of the jets.

When the siblings are identical Conditions, the same jets pass each of the conditions. The combination process
can be optimized. The optimazation exploits the repetion that is _always_ present for identical Conditions.

As an example, let J_1 = J_2 = J_3 = {1,2,3}. Then the combinations are	- excluding repeated jets - {1,2,3}, {1,3,2},
{2, 1, 3}, {2, 3, 1}, {3, 1, 2}, {3, 2, 1}. After repetition removal, we are left with {1,2,3}. Further we only need to
look at jets passing C_1.

This optimatiztion replaces n repeated Conditions of capacity = 1 by a single Condition of capacity n.
After initially allocating jets to Conditions, a further test needs now to be made: were there enough distinct
jets to satisfy the n conditions - i.e. were there at least n jets fulfilling the condition.

To this end, the IConditionMT interface is extended by ICheckedCapacityCondition. This PABC provides a checking function:

bool capacitySatisfied(const HypoJetVector&,..)

 which will be passed the selected jets as an argument.

Currently, the sole implementation is CheckedCapacityCondition.

This commit uses this CheckedCapacityCondition. However the hypo trees remain as before, and no checks are made.
It is therefor expected that they will be no visible effects at this stage.

TCC and UFO reco for r22

See merge request !37473

Refactor ParameterBase. Momentum and position to be handled by the derived classes

See merge request !37812

FillDistributions-FillbasicVars01

See merge request !37650

CaloEvent: Move locking out of loop.

See merge request !37892

Move l2io chains to muon slice

See merge request !37897

Remove unused SkimDecisionMultiFilter and FillEBCFromFlat

See merge request !37857

Update of Pixel Error Monitoring to match new Error Format of PixelRodDecoder

See merge request !37868

Only run CutFlowSvc when explicitly configured and wanted

See merge request !37858

TrigInDetValidation - add new test for ttbar_pu80

See merge request !37933

Shifting LCG version to LCG98_ATLAS_2 (python3).

Closes ATR-20582

See merge request !37429

Add TRT RNN PID

Closes ATLTRTSW-48

See merge request !37753

Preselection step for tracking in jet chains (ATR-21839)

See merge request atlas/athena!37032

Teng Jian Khoo authored 4 years ago and Edward Moyse committed 4 years ago

Deprecate jetMenuHelper, as it's only used in tests and is an annoying alternative to the standard jet reco config

Externals Update, master branch (2020.11.04.)

See merge request !37917

Fix some corner cases for weight parsing in BookkeeperTool

Closes ATLASSIM-4950

See merge request !37903

Add new sample for ttbar_pu80 - this one should really be pu80

	new file:   Trigger/TrigValidation/TrigInDetValidation/test/test_trigID_all_ttbar_pu80_new.py
	modified:   Trigger/TrigValidation/TrigValTools/share/TrigValInputs.json

StoreGate: Remove StoreGate.h.

See merge request !37894

TrigT1CaloCalibConditions: Enable thread-safety checking.

See merge request !37896

[ATR-22317] Further use of read handle in TDT

See merge request !37870

Tim Martin authored 4 years ago and Edward Moyse committed 4 years ago

Continuing to work around the legacy TDT interface, and the SG inspection needs of the failed features alg.

tauRec: save cells and cluster-to-cell links for photon shots

See merge request !37846

Sweeping !37849 from 21.2 to master.
Add MC calibration config files for new UFO large-R jet collection

See merge request !37889

Thread safe handling of CORAL transactions by RDBAccessSvc

See merge request !37847

ATLASRECTS-5743 : egammaART: Clean up a bit the way we remove/add to the CaloCalTopoClusterAux.

See merge request !37888

ISF transform integration beginnings

See merge request !37700

Attempt to clean up MuonCalibConfig

See merge request !36844

I don't think that the approach of modifying the class, rather than an instance, is a good idea.
I also worry that private tools are not being configured at all (specifically, the MdtCalibrationTool and MdtCalibrationDbTool). Attempting to fix this. This causes q431 changes, but this is probably expected.

fix PixelGangedClusterAmbiguities "feature"

See merge request !37766

Gabriel Facini authored 4 years ago and Edward Moyse committed 4 years ago

Some pixel clusters have the same identity
PixelGangedClusterAmbiguities confuses them to be the
same since it uses the identity to sort
now use the width to sort when the identity is the same