Clean-up CSC Examiner related code and configs

[perrotta]

While reviewing #24295 it was noticed that (see #24295 (comment) and #24295 (comment)):

• There are several (test and DQM) configs which define different ExaminerMask's:
  http://cmslxr.fnal.gov/search?_filestring=&_string=ExaminerMask&_casesensitive=1
  Those masks are ideally supposed to be the same. The problem is that that most of those configs are obsolete, and need to be removed from the repository (or alternatively brought up to date)

• There is a fTMB_Scope flag in the CSCDCCExaminer which was originally supposed to be for a handling of the debugging mode feature of CSC hardware, which is not used in the production. It being defined and never used generates a complain in the static analyzer. The code in /EventFilter/CSCRawToDigi/src/CSCDCCExaminer.cc must be cleaned up of it, and of whatever else obsolete is in it

@barvic @Martin-Grunewald @slava77

[perrotta]

assign reconstruction

[cmsbuild]

New categories assigned: reconstruction

@slava77,@perrotta you have been requested to review this Pull request/Issue and eventually sign? Thanks

[cmsbuild]

A new Issue was created by @perrotta .

@davidlange6, @Dr15Jones, @smuzaffar, @fabiocos, @kpedro88 can you please review it and eventually sign/assign? Thanks.

cms-bot commands are listed here

[perrotta]

The dead assignment of fTMB_Scope and related flags was fixed by #26574

The cleaning of the configs with different ExaminerMask's implemented in them is still pending. According to what was written by @barvic in #24295 (comment):

• Those test config files with .cfg extensions are obsolete and unusable (from pre-python era). Most likely will be just removed for general clean up purposes at some point
• Those in DQM looks like old L1T monitoring-related and in principle it would be better to be updated as well for consistency, but I think that particular DQM module is obsolete in Run2

@ptcox

[ptcox]

I have to say I do not think it is a smart idea to blithely delete complex code like this. You have a lot more confidence than me that such code will never need to be reimplemented. And likewise with old config files. To me it seems better to leave them around because they show what once WORKED even if they no longer do. Better to comment out code that is unused but now troublesome rather than completely erase it. Tim perrotta wrote on 5/13/19 11:35:

…

The dead assignment of |fTMB_Scope| and related flags was fixed by #26574 <#26574> The cleaning of the configs with different ExaminerMask's implemented in them is still pending. According to what was written by @barvic <https://github.com/barvic> in #24295 (comment) <#24295 (comment)>: * Those test config files with .cfg extensions are obsolete and unusable (from pre-python era). Most likely will be just removed for general clean up purposes at some point * Those in DQM looks like old L1T monitoring-related and in principle it would be better to be updated as well for consistency, but I think that particular DQM module is obsolete in Run2 @ptcox <https://github.com/ptcox> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#24318 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABGYLHWPDAD3AI7UJT4UPB3PVEY4NANCNFSM4FQACMUA>.

[kpedro88]

@ptcox old code is always available in the git history in case it needs to be revived. Leaving large chunks of commented-out code (or completely broken configs) is discouraged because it is messy and confusing for any future observer or developer.

[ptcox]

I find untangling old code from git history next to impossible. And that is only feasible anyway if somebody KNOWS that some particular feature was once implemented and then deleted. The same is the case for deleted config files. Their existence, even if they no longer operate, can illuminate the fact that once some particular feature or set of features WAS operational, information that can otherwise get lost to institutional memory. I do not consider this as trivial as apparently you do. Kevin Pedro wrote on 5/13/19 15:48:

…

@ptcox <https://github.com/ptcox> old code is always available in the git history in case it needs to be revived. Leaving large chunks of commented-out code (or completely broken configs) is discouraged because it is messy and confusing for any future observer or developer. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#24318?email_source=notifications&email_token=ABGYLHUAXWRI5B3JNUAFLXTPVFWTZA5CNFSM4FQACMUKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVILTPI#issuecomment-491829693>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABGYLHT3ZYSQ2NTXKJWK4YLPVFWTZANCNFSM4FQACMUA>.

[kpedro88]

@ptcox this is one of the primary purposes of version control systems. There are numerous resources available to become more familiar with Git, including a tutorial I wrote specifically for CMS: https://twiki.cern.ch/twiki/bin/view/CMS/CMSGitTutorial.

Additionally, we usually clean up old code only in the development branch. All the previous branches, corresponding to older release cycles of CMSSW, will still have the old code present, so it is quite easy to locate it (without e.g. memorizing some commit hash).