fix GenKt not recognized as pp algorithm

[m-fila]

GenKt wasn't recognized as a pp algorithm and jet_reconstruction was failing to resolve which algorithm function to use in that case.
The tests for GenKt seemed to be missing

Reference file jet-collections-fastjet-inclusive-genkt-p1.5.json.gz was using different capitalization than files for the algorithms. I couldn't find anything relaying on the filename being spelled this way so renamed for consistency

There is also do_test_compare_to_fastjet function that seems to be unused in the current tests and is probably a relic from #73

JetReconstruction.jl/test/runtests.jl

Line 66 in 96bde1b

function do_test_compare_to_fastjet(strategy::RecoStrategy.Strategy, fastjet_jets;

Let me know if we could remove it

[codecov]

Codecov Report

Attention: Patch coverage is 85.71429% with 1 line in your changes missing coverage. Please review.

Project coverage is 73.86%. Comparing base (164c6d2) to head (6196918).
Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
src/ClusterSequence.jl	83.33%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #119      +/-   ##
==========================================
+ Coverage   73.49%   73.86%   +0.36%     
==========================================
  Files          18       18              
  Lines        1249     1251       +2     
==========================================
+ Hits          918      924       +6     
+ Misses        331      327       -4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[m-fila]

Currently GenKt can't be also used with exclusive_jets and n_exclusive_jets. I think there should be a check for power > 0 if GenKt is used, right?

JetReconstruction.jl/src/ClusterSequence.jl

Lines 323 to 327 in 96bde1b

	# Check that an algorithm was used that makes sense for exclusive jets
	if !(clusterseq.algorithm ∈
	(JetAlgorithm.CA, JetAlgorithm.Kt, JetAlgorithm.EEKt, JetAlgorithm.Durham))
	throw(ArgumentError("Algorithm used is not suitable for exclusive jets ($(clusterseq.algorithm))"))
	end

JetReconstruction.jl/src/ClusterSequence.jl

Lines 380 to 384 in 96bde1b

	# Check that an algorithm was used that makes sense for exclusive jets
	if !(clusterseq.algorithm ∈
	(JetAlgorithm.CA, JetAlgorithm.Kt, JetAlgorithm.EEKt, JetAlgorithm.Durham))
	throw(ArgumentError("Algorithm used is not suitable for exclusive jets ($(clusterseq.algorithm))"))
	end

[graeme-a-stewart]

Hello @m-fila

Thanks for this - when I generalised the test setup for e+e- in #73 the GenKt test got lost!

Also, the do_test_compare_to_fastjet is indeed and anachronism now, so it can be deleted.

[graeme-a-stewart]

Currently GenKt can't be also used with exclusive_jets and n_exclusive_jets. I think there should be a check for power > 0 if GenKt is used, right?

Yes, the merging has to favour small energy depositions first for the exclusive jets selection to make sense. So it should be valid for GenKt with a power >= 0.

[graeme-a-stewart]

Just a small additional improvement suggested

[graeme-a-stewart]

Hi @m-fila - there's a logic flaw now that prevents GenKt with p >= 0 from running exclusive selections.

I also realised that the same argument applied to EEKt - it should be tested for p >= 0.

So the logic should be more like:

    # Check that an algorithm was used that makes sense for exclusive jets
    if (clusterseq.algorithm ∈ (JetAlgorithm.GenKt, JetAlgorithm.EEKt)) && clusterseq.power < 0
        throw(ArgumentError("Algorithm $(clusterseq.algorithm) requires power >= 0 (power=$(clusterseq.power))"))
    elseif clusterseq.algorithm ∉
           (JetAlgorithm.CA, JetAlgorithm.Kt, JetAlgorithm.Durham, JetAlgorithm.GenKt, JetAlgorithm.EEKt)
        throw(ArgumentError("Algorithm used is not suitable for exclusive jets ($(clusterseq.algorithm))"))
    end

[graeme-a-stewart]

It might be useful to write a few tests against the wrong setup, checking exceptions are thrown correctly (would help codecov!).

[m-fila]

Thanks, fixed message and logic. Added standalone file with tests - I didn't like splitting them and putting in existing pp and ee tests

[graeme-a-stewart]

Just one tiny little snag - the comment at the top of the test-selection file. Otherwise LGTM!

[graeme-a-stewart]

Filename is wrong.

The comment should describe better the aim of these tests of exclusive jet selection