Master egamma eta restriction

This is an attempt to address ATLASRECTS-4233, in order to address pile-up dependencies of superclusters. @christos and @rnaranjo should comment on the direction this is going and suggest any code improvements. (One potential improvement is to make the L1 restriction more controllable by parameters--I might try to do that.)

This code has been tested to run, but there hasn't been any performance testing.

This merge request affects 1 package:

• Reconstruction/egamma/egammaAlgs

added Egamma Reconstruction master review-pending-level-1 labels

added 1 commit

• f490d8e9 - improve unique_ptr usage

Compare with previous version

This merge request affects 1 package:

• Reconstruction/egamma/egammaAlgs

CI Result FAILURE

	Athena	AthSimulation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-21908-2019-03-14-15-45
Athena: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST 35410]

CI Result SUCCESS

	Athena	AthSimulation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-21908-2019-03-14-16-16
Athena: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST 35413]

Looks fine from MR shifter point of view.... So approving, pinging @christos and @rnaranjo if they want comment.

Pavol [as L2 MR shifter]

added review-approved label and removed review-pending-level-1 label

Hey @jmitrevs

3 questions.

• One is obviously silly, but the addL0L1, actually removed/prunes L0L1?

• What happens in the crack when we have both EME2 and EMB2? In the past they were different clusters and everything was added with the corresponding window. Or this https://acode-browser1.usatlas.bnl.gov/lxr/source/athena/Calorimeter/CaloClusterCorrection/src/CaloFillRectangularCluster.cxx

• auto diff = cell_phi - phiRef; are we sure this can not wrap in -pi,pi or so ?

Here are answers:

• addL0L1 still adds the unpruned L0 and L1 cells. They are explicitly not included in addEMCells (unless I messed up).
• I don't do anything in particular for the crack. The eta range (whether barrel or end-cap) is set by where the hottest cell is
• I was thinking that the proxim would keep it from wrapping around. But maybe it's worth confirming.

• yea I see it is addEMCellsToCluster that actually is doing the different thing here (add only L2L3 or so)

• For two I think what you do is that you take the one with higher etamax, so for crack it is the higher either for EMB2 or EME2 not both. But then you sum both Barrel/E ndcap which should be again what we want here.

• For three I guess proxim returns b in the window = a +- M_PI; so I guess it can no not go out again ...

merged

mentioned in commit c1599f09

This is merged but for the auto I am more on the side of the ATLAS guide. Be conservative in using auto. [using-auto]

" Some authorities have recommended using auto pretty much everywhere you can (calling it ``auto almost always''). However, our experience has been that this adversely affects the readability and robustness of the code. It generally helps a reader to understand what the code is doing if the type is apparent, but with auto, it often isn't. Using auto also makes it more difficult to find places where a particular type is used when searching the code with tools like lxr. It can also make it more difficult to track errors back to their source: "

I think in only one case was used for iterator. But OK it does not matter perhaps.

We should look at the ART tests to make sure this looks reasonable.

added sweep:ignore label