Bug for EEJet(EDM4hep.ReconstructionParticle)

[HarryWang0619]

Currently, in latest two versions of EEJet, when the input is EDM4hep.ReconstructedParticle, as the method in EDM4hepJet Extension doesn't have the cluster_hist_index, the method goes automatically to the function EEJet(jet::Any; cluster_hist_index::Int = 0) in src/EEJet.jl, line 113, and falls in error of line 120.

This occurs when perform this line: cs = jet_reconstruct(recps; p = 1.0, R = 2.0, algorithm = JetAlgorithm.EEKt)
where recps is the reconstructed particle from edm4hep event.

[graeme-a-stewart]

Hello @HarryWang0619 - oh, yeah, this isn't great!

This should be taken care of by the preprocess_escheme filter when the input particles are not themselves EEJets. But something may be going wrong, I will check.

We need to revisit running a check for this in CI (we don't run checks where heavy external dependencies are needed, but this shows the issues with that approach).

[graeme-a-stewart]

I understand the problem - it's the preprocessor that is calling the constructor, with the cluster_hist_index argument. There is no constructor with this set of arguments, so we get the error

EEJet cannot be constructed from object of type 'EDM4hep.edm4hep!ReconstructedParticle'