initial commit of 2016 L1T muon emulation #11504
Conversation
|
A new Pull Request was created by @mulhearn for CMSSW_7_6_X. initial commit of 2016 L1T muon emulation It involves the following packages: DataFormats/L1TMuon The following packages do not have a category, yet: DataFormats/L1TMuon @cmsbuild, @mulhearn can you please review it and eventually sign? Thanks. |
|
please test |
|
The tests are being triggered in jenkins. |
|
+1 |
|
@davidlange6 late as usual but this has no effect on standard sequences and is reasonably sized for a change... please put into 76X if possible. |
| void GeometryTranslator::checkAndUpdateGeometry(const edm::EventSetup& es) { | ||
| const MuonGeometryRecord& geom = es.get<MuonGeometryRecord>(); | ||
| unsigned long long geomid = geom.cacheIdentifier(); | ||
| if( _geom_cache_id != geomid ) { |
There was a problem hiding this comment.
@Dr15Jones - is this the construct that we want to advise against ? I'm not sure I got the valid usecases of cacheIdentifier()
There was a problem hiding this comment.
@davidlange6 It is a proper use of cacheIdentifier. However, I think the interface design makes it very easy to make a mistake. Essentially one must always be sure to call checkAndUpdateGeometry before ever calling any of the other methods. I would either pass the EventSetup in to all the calls or put the object into the EventSetup itself.
There was a problem hiding this comment.
Another option is to construct a GeometryTranslator by passing it the EventSetup and then just make a new one each event.
initial commit of 2016 L1T muon emulation
|
@mulhearn - could you have a look at Chris's suggestions in a new PR? |
|
@davidlange6 OK, will look into it. |
This is the initial batch of L1T Muon emulation for 2016 run. It includes the GMT emulator and data formats needed by GMT.
This code is unknown to standard workflows, so should have no effect.