Additions to brat

From: Kris Hagel (hagel@comp.tamu.edu)
Date: Tue Aug 06 2002 - 18:32:23 EDT

  • Next message: Pawel Staszel: "changes to brat"

    Hello,
    It is almost as if this list was taken off the air on 12-July.  Not a 
    single message since then. Consider this the start of its use again.
    
    Anyway, this message is to inform that I checked in several minor 
    changes into BRAT.  Most have to do with increasing the verbosity of 
    different modules I used in debugging the pp data chain. BrVertexFilter 
    was modified appropriately to handle vertices from the Inel counters 
    used in the pp measurement.
    
    A more long term important addition was something I named BrRecoPackage. 
     This is a reaction to the reco scripts used to reconstruct the data.  I 
    essentially put into the the package the analysis chain used for 
    reconstructing local tracks.  I think the reco configuration scripts are 
    becoming more or less standard and I think we ought to move in that 
    direction because those configuration scripts are becoming too complex 
    for my taste.  I should note, however, that the version in the 
    repository is a compilable placeholder.  It is more or less unusable 
    right now, so don't try unless you want to do some work to make it 
    functional.  There is a warning on the top indicating exactly that.  I 
    should also note that I came across BrFfsPackage which is kind of what I 
    had envisioned, but does not go as far.  I did not want to modify it 
    because I thought the author of that package might have more plans for 
    it even though the last time it was modified was 5-Nov-2001.  That might 
    have to do with finishing a thesis or something like that.
    
    The chain envisioned to then reconstruct local tracks would be
    input = new BrEventIO(....)
    mainModule->AddModule(input);
    
    reco = new BrRecoPackage(...)
    mainModule->AddModule(reco)
    
    output = new BrEventIO(....)
    mainModule->AddModule(output)
    
    Of course, there are the usual connects to the DB and there will be 
    options in the reco package to get it to do the kind of reco you want.
    
    I would then envision making a BrMatchPackage and a BrGenerateDst 
    package.  The last will depend on the Dst software now residing in 
    brahms_app/pc_app/dst to be incorporated into brat.  I had discussed 
    with someone at the collaboration meeting the possibility of 
    incorporating it (dst) into brat and we had thought to let it mature. 
     What is the feeling of the movers and shakers of the dst software?  Is 
    it ready for full bonafide incorporation into brat?  My personal opinion 
    from experience with pp analysis is that it is ready.
    
    Comments are welcome as I am not yet married to any particular way of 
    doing this module except that I want to streamline things to 
    reconstruction easier and more uniform across different people doing it. 
     And if you want to make comments, do it now as a number of you know 
    that when I marry myself to a particular way of doing something, I hold 
    on to it until I nearly get dragged to death.
    
    Enjoy
    
    Kris
    



    This archive was generated by hypermail 2b30 : Tue Aug 06 2002 - 18:33:05 EDT