Remove

- `Trigger/TrigEvent/TrigDecision`
- `PhysicsAnalysis/AnalysisTrigger/AnalysisTriggerTools`

packages as they only contained deprecated code.

The '_load.cxx' files are no longer needed to instantiate Gaudi
components.

  * Correctly implement the use of PseudoJetGetters and algs

Remove remaining remnants of Run 1 EDM.

Remove remaining remnants of Run 1 EDM.

James Beacham authored 6 years ago and Atlas Nightlybuild committed 6 years ago

Sweeping !14558 from 21.2 to 21.0.
Fixing the PATJobTransforms to allow the MCP code to be processed on tier0

See merge request atlas/athena!14570

(cherry picked from commit 3166a6a4567ff132f9a033028e4d0b84adb8e016)

41b54058 Merge branch 'reco_tf_fixes' into '21.2'

James Beacham authored 6 years ago and Atlas Nightlybuild committed 6 years ago

21.0 desdm eoverp addition

See merge request atlas/athena!14267

(cherry picked from commit 43db1c574e717488a1d82efd5d61e57e7beb87d6)

d132ff55 Adding a new output type which saves ALLCELLS information and skims on triggers…
55a1d36b Update PrimaryDPDFlags.py
89fc36a1 Update UploadTfAMITag.py
59561d2e Update UploadTfAMITag.py

    
Create MCTruthClassifier cache where needed.
Fixes crashes observed in master.

    
Create MCTruthClassifier cache where needed.
Fixes crashes observed in master.

Create MCTruthClassifier cache where needed.
Fixes crashes observed in master.

clang warnings: Inconsistent use of struct/class keywords.

With clang, having `using base::procType' in a derived class is not
sufficient to resolve ambiguities.  Use an explicit forwarding
function instead.

Remove reference to nonexistent path.

Ignore some Athena/ARA differences seen when reading some old data versions
that are not worth fixing.

Started using std::string::empty() instead of a comparison with an empty
string, and removed the meaningless greeting INFO message from the
initialize() function.

This merge included, for b-tagging purposes, the aliasing of
AntiKt4HI jet collection to AntiKt4EMTopo, which (at least
in a 21.0 test) was not enough for the EM Topo jet reference
data to be recognized. Aliasing is now extended to
AntiKt4TopoEM and AntiKt4LCTopo, to solve this issue and be
fully consistent with the scheme used for AntiKt4EMTopo.

But didn't do anything significant to it in the end, as the dependencies
of the package didn't change.

It turns out that the L1Calo tools very much have an internal state, so
the algorithm can't be reentrant.

Kept the structure of the code such that it could be made reentrant easily
in the future once the tools will allow to do that.

Turns out that the python config was wrong for this class since a long-long time...