Open davidlange6 opened 4 years ago
A new Issue was created by @davidlange6 David Lange.
@Dr15Jones, @silviodonato, @dpiparo, @smuzaffar, @makortel can you please review it and eventually sign/assign? Thanks.
cms-bot commands are listed here
assign l1
New categories assigned: l1
@benkrikler,@rekovic you have been requested to review this Pull request/Issue and eventually sign? Thanks
assign core
New categories assigned: core
@Dr15Jones,@smuzaffar,@makortel you have been requested to review this Pull request/Issue and eventually sign? Thanks
sounds reasonable to me.
Sounds good to me too.
A related issue might be the L1Trigger dependency in DataFormats (it's just this single one): https://github.com/cms-sw/cmssw/blob/master/DataFormats/L1TMuon/src/EMTFHit.cc#L2
Wouldn't the DataFormats have no dependencies in an ideal world, besides FWCore and maybe CommonTools for tests?
@guitargeek Good catch, that dependence is not allowed (any volunteers to move things around?).
The allowed dependencies on packages defining persisted data formats are listed here https://twiki.cern.ch/twiki/bin/view/CMSPublic/SWGuideCreatingNewProducts#Restrictions_on_the_package_defi (packages defining transient-only products have less constraints)
Thanks for the information. I don't know if it's good to move this code around without coordinating with the original developers, because it seems to be actively developed in the last years.
However, if I were to do it I would just move all of these member functions implemented in EMTFHit.cc
into some new file EventFilter/L1TRawToDigi/plugins/implementations_stage2/EMTFHitFunctions.h
(or something), because it's the only place where they are used. All of the member variables of EMTFHit
are public, so it should be no problem to turn the member functions into free functions.
more fun are the CondFormats dependencies in DataFormats/RPCDigi. If one wants to make a class that is both storable in the event and storable in the conditions db, what is the design for that? (or is there one?)
I've proposed L1 changes to #31346
Ping about DataFormats/RPCDigi.
+l1
I'd like to remove the dependency of DataFormats/PatCandidates on CondFormats/L1TObjects (which I believe are not supposed to be allowed...).
This can be accomplished by moving two enums from CondFormats/L1TObjects/interface/L1GtDefinitions.h to some DataFormats package. (presumably DataFormats/L1TGlobal). Are there reasons not to do this or to do it differently?