[Halld-physics] Updates to Analysis Library & Scripts

[Paul Mattione]

Sorry, I've just checked in a few more important updates that improve things like thrown <--> reconstructed matching (max error, cutoff value).  Please re-checkout the analysis library and perl scripts.  Thanks.  

 - Paul

On May 19, 2014, at 9:56 PM, Paul Mattione wrote:

> I've updated the physics analysis library with some bug fixes (thanks Beni & Justin) and some improvements (see below).  Please checkout the latest version if you are analyzing the data challenge 2 data, along with the latest version of the perl scripts for generating your plugins/actions.  
> 
> Analysis plugin perl scripts: https://halldsvn.jlab.org/repos/trunk/scripts/analysis/
> Wiki documentation: https://halldweb1.jlab.org/wiki/index.php/GlueX_Analysis_Software
> 
> ****** IMPORTANT NOTES ******
> 
> 1) Potential plugin-breaker: If you created custom DAnalysisActions in your plugin, delete all calls to Get_PreviousParticleCombos() in them (this function no longer exists).  Instead call the new Get_NumPreviousParticleCombos() to see if the given action has been called for your event yet (if you don't know why then don't worry about it).  This change was made to get a very large speed increase in events with thousands of combos.  
> 
> 2) A new perl script has been created: MakeSkimPlugin.pl.  It will create an ".idxa" (text) file containing the events (Run, Event, File #'s) that passed your DReaction cuts on the data challenge 2 data.  The EventStore database will use these files to keep track of events that pass skim cuts.  All you have to do is add the cuts that you want to your DReaction.  
> 
> 3) The cuts for pre-selecting tracks & showers have been updated.  If you run the latest MakeSkimPlugin.pl and MakeReactionPlugin.pl scripts, they automatically include the below "highly recommended" selection cuts.  It is extremely important to include these cuts to reduce the extremely large combinatorics, and for consistency between analyses: 
>  a) Tracking FOM >= 3-sigma
>  b) Neutral PID FOM >= 5-sigma
>  c) Tracks have at least one detector match (ST/TOF/BCAL/FCAL)  
> 
> 4) New Analysis Actions: 
>  a) DHistogramAction_NumParticleCombos
>  b) DCutAction_TransverseMomentum
>  c) DCutAction_AllTracksHaveDetectorMatch
> 
> Also, in case anyone is interested, I checked all of the plugins I used to make the plots for my talks.  In addition to monitoring_hists, they are the SkimStudies, n3pi_skim, p2pi, p3pi, & p4pi plugins at:
> 
> https://halldsvn.jlab.org/repos/trunk/home/pmatt/plugins/
> 
> - Paul
>