Change tau trigger bits for nanoAOD

[lwezenbe]

Description
The tau trigger bits for NanoAOD never provided a match in 2018 trigger because the 'Hps' part in the name was forgotten. This PR is meant as both a fix for that and as a preparation for Run III. We have decided to go with a unified strategy that will work for both Run II and Run III triggers, hence the addition of extra bits containing DeepTau. It now also contains trigger bits for taus in the PAG VBF and displaced tau triggers. A full overview of all trigger filters in Run II and Run III was made and is contained in the following google sheets:
https://docs.google.com/spreadsheets/d/1L8P4HeiUSaisewuSFBEJOM5t5xXcDjNbfBtD9hEE5F8/edit#gid=0

This might need a review from a few more people to make sure that we have everything we need. Also bits 65536 and beyond were added for completeness but I am not fully sure if we need them

Changes

• Addition of wildcards to original trigger bits to work for both Hps and non-Hps paths
• Addition of deeptau trigger bits
• Addition of bits 4096 to 32768 to include PAG paths
• Addition of bits 65536 for completeness, they allow to go a bit more specific in the triggers to select
• Addition of boosted tau block. Right now the bits are redundant but we expect more to be added in the future.
• Added bits to non-tau blocks from cross triggers

Tests

• Reran NanoAOD for all three Run II years and in a Run III sample, everything seemed to act as expected

[mbluj]

@lwezenbe, @kandrosov I have just went through the changes and they seem sensible for me. However, my knowledge on current tau triggers does not allow to check if there are not missing patters of trigger filter names or so.
Have you run this setup on any new samples with current trigger menu? Do you confirm that it is successful? What about historic run-2 menus? It seems that they should be "phased" correctly, but has it been already checked?

[lwezenbe]

@mbluj I have checked this within my capabilities. For 2016, 2017 and 2018 I have run it on miniAOD before with the version of the trigger menu used for that specific sample and it worked if I used the appropriate CMSSW version and ported the changes into it. For Run III samples I have also tried to run it on the already existing MiniAOD that has already been produced but I wasn't sure what menu they had used exactly, I was trying to figure that out earlier. I wanted to recheck if everything is fine after the changes and just do another sanity check today but I had a lot of interference with from other projects/teaching this week.

[mbluj]

@mbluj I have checked this within my capabilities. For 2016, 2017 and 2018 I have run it on miniAOD before with the version of the trigger menu used for that specific sample and it worked if I used the appropriate CMSSW version and ported the changes into it. For Run III samples I have also tried to run it on the already existing MiniAOD that has already been produced but I wasn't sure what menu they had used exactly, I was trying to figure that out earlier. I wanted to recheck if everything is fine after the changes and just do another sanity check today but I had a lot of interference with from other projects/teaching this week.

@lwezenbe thank you. I think these tests are good enough. I will merge this development and prepare a PR to official CMSSW. I suppose that this PR will be processed some time (~1 week or so) so additional last minute fixes will be still possible.

@kandrosov, is backport of this development to 10_4_X foreseen?

[mbluj]

+1

[mbluj]

I went through this development, I found proposed changes sensible and I approve them.

[kandrosov]

@kandrosov, is backport of this development to 10_4_X foreseen?

For now, no backports are planned. We going to follow XPOG guidelines once they are defined. Currently, PPD does not plan to reprocess Run2 in nano v10 format (they want to accumulate more changes before), so this change will only go for Run 3 + private nano productions for Run 2. @lwezenbe we can point to this code in the tau trigger TWiki for the analysers who plans to have private UL nanoAODs with correct trigger bits.

[mbluj]

@kandrosov, is backport of this development to 10_4_X foreseen?

For now, no backports are planned. We going to follow XPOG guidelines once they are defined. Currently, PPD does not plan to reprocess Run2 in nano v10 format (they want to accumulate more changes before), so this change will only go for Run 3 + private nano productions for Run 2. @lwezenbe we can point to this code in the tau trigger TWiki for the analysers who plans to have private UL nanoAODs with correct trigger bits.

OK, fine. But as far as I know with current production setup nanoAOD is produced together with RECO/AOD/MiniAOD, so the first, prompt nanoAOD will still contain tau trigger bits with old (partially buggy) setup. On the other hand, backport to 12_4_X will require special switch to ensure backward compatibility of the nanoAOD content so most likely prompt nanoAOD will be also produced with the old setup...

[kandrosov]

@mbluj good point. Perhaps we will need to clarify with XPOG what would be the best way to proceed. But I think it is better to accumulate all modifications that we want to have for Run 3 nanos (e.g. DeepTau v2p5) before. I expect that we'll open a PR for v2p5 for nano (into the cms-tau-pog fork) the next week at the latest. Could you, please, prepare a branch for it?

[mbluj]

@mbluj good point. Perhaps we will need to clarify with XPOG what would be the best way to proceed. But I think it is better to accumulate all modifications that we want to have for Run 3 nanos (e.g. DeepTau v2p5) before. I expect that we'll open a PR for v2p5 for nano (into the cms-tau-pog fork) the next week at the latest. Could you, please, prepare a branch for it?

It sounds as a good plan.

I will prepare a new branch for the WPs; whom I should inform that it is ready?

[mbluj]

@lwezenbe, PR to the official CMSSW is being reviewed. In particluar they ask if observed changes here: https://gitlab.cern.ch/cms-nanoAOD/nanoAOD-integration/-/issues/155 are as expected. I am going to check this, but a cross check from your side is welcome. Thanks!

[kandrosov]

whom I should inform that it is ready?

me and Oleg. Thank you!