[PWGHF] Use the option of occupancy estimator from HfEventSelection in candidate creators

[wuctlby]

Hi @zhangbiao-phy, this PR is for adding a configurable occupancy estimator in the candidate creator. Since the estimator was hard-coded as ITS from the beginning, this could cause some potential problems.

Here, I would also like to @maciacco , even though he is on vacation, in case he wants to further check something when he returns.

I already tested it locally, and everything went smoothly.

PS: I couldn't find a method to read the estimator option directly from the 'hfEvSel', so I added a new variable instead.
NOTE: If the FT0c is used, the occupancy range must also be changed accordingly.

Please let me know if you have any suggestions. Thanks a lot!

[github-actions]

O2 linter results: ❌ 3 errors, ⚠️ 0 warnings, 🔕 13 disabled

[wuctlby]

O2 linter results: ❌ 3 errors, ⚠️ 0 warnings, 🔕 12 disabled

Regarding the errors, as I am not the developer of either code, I am not sure if it's okay that I directly disable the check for the corresponding parts.

[vkucera]

O2 linter results: ❌ 3 errors, ⚠️ 0 warnings, 🔕 12 disabled

Regarding the errors, as I am not the developer of either code, I am not sure if it's okay that I directly disable the check for the corresponding parts.

They are already disabled.

[vkucera]

occEstimator is already included in HfEventSelection and you literally copy-pasted its definition from there.
What's the point of this PR?

[wuctlby]

occEstimator is already included in HfEventSelection and you literally copy-pasted its definition from there. What's the point of this PR?

Yes, it has in HfEventSelection. But we should also use the occEstimator, the one we want to use, in the candidate creator, instead of hard-coding it as ITS. Otherwise, we won't be able to obtain the occupancy distribution at the event level based on FT0c. In this case, if we used the FT0c in the HfEventSelection and applied the occupancy cut or a narrow occupancy range, we would only obtain the occupancy distribution at the event level based on the ITS. And here the events are those that pass the FT0c-based occupancy cut. It is not correct.

[wuctlby]

occEstimator is already included in HfEventSelection and you literally copy-pasted its definition from there. What's the point of this PR?

Or, could you kindly tell me how I can read the occEstimator from the HfEventSelection instead of adding a new one in creator? I think it's better, in case someone changed the occupancy estimator in HfEventSelection but did not change it in creator. But I didn't find that there is this kind of usage.

[vkucera]

I see your point now.
Please get the estimator using hfEvSel.occEstimator.

[wuctlby]

I see your point now. Please get the estimator using hfEvSel.occEstimator.

Yes, I also tried it. You mean something like occEstimator = hfEvSel.occEstimator?

[vkucera]

I see your point now. Please get the estimator using hfEvSel.occEstimator.

Yes, I also tried it. You mean something like occEstimator = hfEvSel.occEstimator?

That should work too but I don't see why you need the new variable.

[wuctlby]

I see your point now. Please get the estimator using hfEvSel.occEstimator.

Yes, I also tried it. You mean something like occEstimator = hfEvSel.occEstimator?

That should work too but I don't see why you need the new variable.

It's due to that I tried the occEstimator = hfEvSel.occEstimator, but when compiling it, it fails. So ..., I will add the variable type and try it again in the docker of ubuntu 22.04

[vkucera]

occEstimator

I cannot know what type your variable occEstimator is, but you can try hfEvSel.occEstimator.value which should return int.

[wuctlby]

Hi @vkucera , the new commit is about directly using the hfEvSel.occEstimator as the occupancy estimator in the candidate creator. Sorry, I misunderstood what you meant. I thought you were still thinking why we need to specify the estimator. After I confirmed that it can be directly used in the `getOccupancyColl, then I just got why you just asked why I need a new variable.
Please let me know if there are still any problems.

[zhangbiao-phy]

hi @wuctlby, many thanks for your commits! Maybe also modified candidateCreatorXicToXiPiPi.cxx accordingly Hi @vkucera, if you don't have further comments, I would like to approve it now. we are doing the study of D0 production vs occupancy.