CPAlgorithms: small improvements to Event Selection

Use ROOT::Math::PtEtaPhiEVector instead of TLorentzVector (legacy) in a couple of algorithms, and explicitly cast selections to char so that they can be properly printed to file if requested (in the debugging mode)

This merge request affects 2 packages:

• PhysicsAnalysis/Algorithms/AsgAnalysisAlgorithms
• PhysicsAnalysis/Algorithms/EventSelectionAlgorithms

Affected files list will not be printed in this case

Adding @krumnack ,@tadej as watchers

added Analysis CPAlgorithms analysis-review-required main review-pending-level-1 labels

CI Result FAILURE (hash 41923532)

	Athena	AnalysisBase	AthAnalysis
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
AnalysisBase: number of compilation errors 0, warnings 0
AthAnalysis: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 9348] (remote access info)

Hi @krumnack maybe I'm missing something obvious here... What this EventSelection business is doing, is chaining together a number of very simple algorithms (e.g. count the number of jets and compare to some threshold). Each algorithm writes its decision onto the EventInfo via a CP::SysWriteSelectionHandle. The next algorithm in the chain then looks it up via CP::SysReadSelectionHandle. This is all completely fine.

Now, I have this debugMode option, which if True will save every single selection decision as a branch in the output ntuple. This is essentially a per-event cutflow. What I find is that these branches are currently broken (they all contain the same value, some very large int), and the fix is to append ,as_char to the name of the CP::SysWriteSelectionHandle. This works, as long as there is no further algorithm in the chain. That's what the CI test (only in Athena...) complains about:

Failure to parse expression: 'SUBcommon_NBJET_1_NOSYS': SG::ExcAuxTypeMismatch: Type mismatch for aux variable `::SUBcommon_NBJET_1_NOSYS' (305); old type is char new type is unsigned int

This comes from the second algorithm in the chain trying to read the decision from the previous one, but now that decision has ,as_char in it. I can't seem to find a way to get around it.

The only alternative I see at the moment, is to rewrite all these algorithms to have two outputs, one CP::SysWriteSelectionHandle and one CP::SysWriteDecorHandle, and only use the latter to create output branches. Do you have a better suggestion?

Switching tags while the discussions are ongoing.

added review-user-action-required label and removed review-pending-level-1 label

resolved all threads

added 1 commit

• cb57005a - as_char everywhere

Compare with previous version

build area is cleaned as the previous build was for the MR labeled as full-build

This merge request affects 2 packages:

• PhysicsAnalysis/Algorithms/AsgAnalysisAlgorithms
• PhysicsAnalysis/Algorithms/EventSelectionAlgorithms

Affected files list will not be printed in this case

Adding @krumnack ,@tadej as watchers

added review-pending-level-1 label and removed review-user-action-required label

CI Result SUCCESS (hash cb57005a)

	Athena	AnalysisBase	AthAnalysis
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
AnalysisBase: number of compilation errors 0, warnings 0
AthAnalysis: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 9549] (remote access info)

This seems fine to me. Approving for AR.

Nils [Analysis RC]

added analysis-review-approved label and removed analysis-review-required label

removed review-pending-level-1 label

added review-approved label

mentioned in commit 9edc4d43

merged

mentioned in merge request !71447 (merged)