[Halld-offline] DDetectorMatches, REST

[Matthew Shepherd]

Paul and others,

For reference, in CLEO we used a template to generate these link classes.  You may read the proceedings here by Jon Thaler:

http://scitation.aip.org/content/aip/proceeding/aipcp/10.1063/1.1405289

You may also find his web page:

http://web.hep.uiuc.edu/home/jjt/Lattice/

This template provides a mechanism to do many to many matching with some additional class that stores "link data" (user defined data about the match).  We used this template for many things (examples in Jon's document) in addition to track/shower match.

I'm not suggesting that you adopt the CLEO code.  Since you are volunteering to develop this, you should do it as you see fit -- I think you've thought most about the problem.  I would consider the use of a template though.  It is ideal for this sort of thing, and, when done well, provides an easy common interface that can be reused by others for many to many matching.  (Unfortunately, there is no standard template for such a thing.)

Matt

---------------------------------------------------------------------
Matthew Shepherd, Associate Professor
Department of Physics, Indiana University, Swain West 265
727 East Third Street, Bloomington, IN 47405

Office Phone:  +1 812 856 5808

On Dec 16, 2013, at 9:28 PM, Paul Mattione <pmatt at jlab.org> wrote:

> As discussed at the last offline meeting, I'm creating a new JANA object (DDetectorMatches) that will store the information regarding the matching between a time-based track and the different detector elements (BCAL, FCAL, SC, TOF).  These objects will be saved to the REST file so that the tracks don't need to be re-swimmed through the detector during the analysis stage.  
> 
> Before finalizing the format of the DDetectorMatches class and corresponding REST structure, I want to see what you guys think about them.  The idea isn't just to store the matches themselves (where matching is many <--> many), but also additional information that is needed for calculating PID confidence levels, constraining timing in a kinematic fit, and other information that may be for signal/background separation with a BDT.  
> 
> I've attached the DDetectorMatches.h header file and a modified rest.xml file with the new "detectorMatch" structure added.  Comments are included in the files.  I think the JANA object is in pretty good shape, but I think the method of pointing between matching objects in the REST file warrants some discussion.  Please let me know what you think.  
> 
> - Paul
> 
> <DDetectorMatches.h><rest.xml>
> 
> _______________________________________________
> Halld-offline mailing list
> Halld-offline at jlab.org
> https://mailman.jlab.org/mailman/listinfo/halld-offline