PWGHF: Adding flag selection in the hadron tasks

[DelloStritto]

No description provided.

[vkucera]

If this works, it works by sheer luck.
There is currently no correspondence between the hfflag() values set in the index skimming and the DecayType values used in MC matching.
@nzardosh I suggest to use the DecayType values for the skimming bits as well.
Same applies to #5204.

[nzardosh]

If this works, it works by sheer luck.
There is currently no correspondence between the hfflag() values set in the index skimming and the DecayType values used in MC matching.
@nzardosh I suggest to use the DecayType values for the skimming bits as well.
Same applies to #5204.

Hi @vkucera , yes I agree. This was the idea behind Luigi's change too.

[vkucera]

-> JpsiToEE

[vkucera]

I suppose the same bit check should be added in the rec. part of the MC tasks as well, right?

[DelloStritto]

I suppose the same bit check should be added in the rec. part of the MC tasks as well, right?

I think it is not strictly needed because we check the bit here:
https://github.com/AliceO2Group/AliceO2/blob/dev/Analysis/Tasks/PWGHF/taskLc.cxx#L127

[vkucera]

I suppose the same bit check should be added in the rec. part of the MC tasks as well, right?

I think it is not strictly needed because we check the bit here:
https://github.com/AliceO2Group/AliceO2/blob/dev/Analysis/Tasks/PWGHF/taskLc.cxx#L127

But maybe we can add it so we do not loop over all the candidates.

I think it is needed if we want to fill the MC background histograms with the same candidates that populate the combinatorial background in the non-MC task. Isn't it?

[DelloStritto]

I suppose the same bit check should be added in the rec. part of the MC tasks as well, right?

I think it is not strictly needed because we check the bit here:
https://github.com/AliceO2Group/AliceO2/blob/dev/Analysis/Tasks/PWGHF/taskLc.cxx#L127
But maybe we can add it so we do not loop over all the candidates.

I think it is needed if we want to fill the MC background histograms with the same candidates that populate the combinatorial background in the non-MC task. Isn't it?

I'm not sure about this. In this way we would not compute the loss in efficiency due to the skimming cuts and I'm not sure this is what we want.

I missed that you specified "rec part of the MC" in your comment. I think that you are right

[iarsene]

Hi, is this PR ready now ?

[ginnocen]

@iarsene, please wait. We have to make a few extra tests.

[iarsene]

Maybe switch to draft so we avoid mistakenly merging these PRs