[PWGHF] Add a mass axis to final Thnsparse Correlation Data

[zz951]

In correlatorLcHadrons.cxx, add a histogram to be filled with mass, pt, which will be filled only once in same delta phi bin even.
In taskCorrelationLcHadrons.cxx, add a histogram to be filled in a global region also add anther mass axis.

[github-actions]

O2 linter results: ❌ 0 errors, ⚠️ 0 warnings, 🔕 0 disabled

[gluparel]

Hi @zz951,
it is not really clear to me why you want to add the TH3 histogram and in particular the additional ThNSparse in the task + the additional mass axis in the other 3 ThNSparse because it seems to me that you are only duplicating the information.

In fact, if you add the ThNSparse hCorrel2DVsPtGlobalRegion with the mass axis you can then select on the invariant mass at analysis level and you do not need any more the other 3 ThNsparse (signal, sideband left and sideband right).

Do I understand correctly?

Grazia

[zz951]

Hi @zz951, it is not really clear to me why you want to add the TH3 histogram and in particular the additional ThNSparse in the task + the additional mass axis in the other 3 ThNSparse because it seems to me that you are only duplicating the information.

In fact, if you add the ThNSparse hCorrel2DVsPtGlobalRegion with the mass axis you can then select on the invariant mass at analysis level and you do not need any more the other 3 ThNsparse (signal, sideband left and sideband right).

Do I understand correctly?

Grazia

Ciao @gluparel , because of the traditional method to subtract background using left or right sideband only and both of them is different from each other, there are a mass dependence. Therefore, I am thinking to using the fitting method to extract signal correlation, which we need to project correlations on mass axis with 32 delta_phi bins to get 32 per-trigger yield of signal, and for this reason we need a raw yield within 32 different delta_phi bins. In ThNsparse, what we stored is the information for each pair(Lambda_c-Hadron), some candidates may be associated different charge particles in the same delta_phi bins, so I need a TH3 histogram with mass, pt_lambda_c and delta_phi (the candidates in the same delta_phi bin will be filled only once). As for the other 3 ThNsparse, we keep them now for final comparison to check the performance of the fitting method.
Zhen

[apalasciano]

Hi @zz951,
Thanks for the implementation.
I share the same doubt as @gluparel .
It would be enough of an implementation like:
if (storeMass) { registry.fill(HIST("hCorrel2DVsPtGlobalRegion"), deltaPhi, deltaEta, ptLc, ptHadron, poolBin, cent, massLc, efficiencyWeight); } else{ ... (same lines as before) ... }
where storeMass would be a configurable, which, in case you can switch on and off.
You don't need to add the mass axis on the signal and sidebands ThNSparse, since they are specific regions of what you store in the 'Global' ThNSparse (duplicating information).
No need to introduce the modifications in the correlator if you are interested in the fitting method.

[zz951]

Hi @zz951, Thanks for the implementation. I share the same doubt as @gluparel . It would be enough of an implementation like: if (storeMass) { registry.fill(HIST("hCorrel2DVsPtGlobalRegion"), deltaPhi, deltaEta, ptLc, ptHadron, poolBin, cent, massLc, efficiencyWeight); } else{ ... (same lines as before) ... } where storeMass would be a configurable, which, in case you can switch on and off. You don't need to add the mass axis on the signal and sidebands ThNSparse, since they are specific regions of what you store in the 'Global' ThNSparse (duplicating information). No need to introduce the modifications in the correlator if you are interested in the fitting method.

Ciao @apalasciano thank you for the comments, you are right, I misunderstood before. Here I give a new commit as you mentioned.
Zhen

[apalasciano]

Hi Zhen,
Thanks for implementing the suggestion.
I'm approving it

[alibuild]

Error while checking build/O2Physics/o2 for 8472938 at 2025-11-07 09:04:

## sw/BUILD/O2-latest/log
/sw/SOURCES/O2/daily-20251107-0000/daily-20251107-0000/Framework/Core/include/Framework/InputRecord.h:22:10: fatal error: Framework/ObjectCache.h: No such file or directory
/sw/SOURCES/O2/daily-20251107-0000/daily-20251107-0000/Framework/Core/src/ControlServiceHelpers.h:15:10: fatal error: Framework/DataProcessingStates.h: No such file or directory
ninja: build stopped: subcommand failed.

Full log here.