Status of heppy?

[cbernet]

Dear all,

We would like to know what is the status of heppy at this stage, because:

• rerunning MVA tau isolation in heppy #737 : this pull request has been on hold for a while
• Fall17 MVA Electron ID v2 #742 : this issue did not trigger much interest while we believe people using heppy need it for 2017 data analysis.

We did the following things recently:

• implemented and validated electron ID v2 with the help of egamma. It's fully validated for the currently supported IDs, and the fwlite interface is now on the egamma side, which is very good.
• reviewed our analysis sequence and designed atomic analyzers and other generic tools that could be useful for others.

We would like to push these changes to cmssw and to cmgtools-lite.

Later on, we intend to:

• port heppy and CMGTools to CMSSW_10_X
• work towards moving the FWLite interface to the POG tools towards the POG realms, as we did for egamma
• keep contributing to the update of POG recipes, which should be easier when the previous item is done
• adapt heppy so that it can be used to analyse nanoAOD as well

If you're still using heppy, could you please give a thumbs up?

CMG, if you're still interested in managing this system, please tell us.
If not, we could do it in Lyon, and take again responsibility for CMS.

Cheers,

Colin

[gpetruc]

For #737 I was waiting for a validation, but it's not going to come so I just merged it.
If you have a PR for #742 please send it.

Regarding longer term maintenance, I have no idea if there's anyone in the group interested in contributing to it or not, I'll make some inquiries and let you know.

For 10_X, @peruzzim had some opinions about which base release to go to given where recipes are being integrated also on nanoAOD, so I let him comment.

cmgtools-lite is a bit more disorganized as the amount of code that is really shared across different projects is very small (just the sample files and a few DAS-related tools), so I don't know the upgrade path for the various subsystems (they're largely self-managed)
I can only say that we'll probably do some update of the friend-tree producing & plotting tools under TTHAnalysis, possibly integrating it with nanoAOD-tools & allowing them to run on top of plain nanoAOD.

[cbernet]

Thanks for the information Giovanni.

Indeed the amount of code in cmgtools-lite common to several projects is small,
and probably, common code can often simply be put in the PhysicsTools/HeppyCore and PhysicsTools/Heppy packages.

I guess the main issue is the maintenance of these two CMSSW packages.
packages. Do you know if there is an official maintainer for these packages at the moment?

We're going to send a new PR for electron ID v2 and some cleanup of the Electrons in PhysicsTools/Heppy soon.

Cheers,