Re: event IO

From: Jens Ivar Jordre (jensivar.jordre@fi.uib.no)
Date: Mon Nov 05 2001 - 10:57:45 EST

  • Next message: Christian Holm Christensen: "Re: event IO"

    Howdy Ian, Djam ++
    
    My changes last week were only in BrIOModule's End and Finish methods to handle 
    successful file closure. Nothing was done to BrMainModule (See
    http://www.sdcc.bnl.gov/brahms/private/list_hyper/brahms-dev-l/1524.html)
    
    And regarding Djamel's fix I'm happy. A missing file should of course not stop the 
    whole analysis job.
    
    Ian Bearden wrote:
    > Hi Djam, others,
    > Is this problem related to the changes made recently by Jens Ivar?
    > I don't think you should wait Djam, since we *absolutely* **MUST** be able
    > to
    > skip missing sequences.
    > If someone knows a more elegant way of doing it, let him or her step forward
    > now...but I guess at present working, inelegant code is prefered to elegant
    > non-working code.
    > Ian
    > 
    > 
    >>-----Original Message-----
    >>From: owner-brahms-dev-l@bnl.gov [mailto:owner-brahms-dev-l@bnl.gov]On
    >>Behalf Of Djamel Ouerdane
    >>Sent: Saturday, November 03, 2001 9:26 AM
    >>To: brahms-dev-l@bnl.gov
    >>Subject: Re: event IO
    >>
    >>
    >>Ok, I've found the stuff that stopped the jobs:
    >>
    >>in BrMainModule.cxx, someone changed this:
    >>
    >>    if (fStatus >= kStop)
    >>      continue;
    >>
    >>to that
    >>
    >>    if (fStatus >= kStop) {
    >>      cout << "Begin Stopped " << endl;
    >>      return fStatus;
    >>    }
    >>
    >>The reason is that BrIOModule sends a stop signal if Open returns kFALSE
    >>in the method OpenNext(). I'm not sure it's the right thing to do in the
    >>case i mentioned previously (scan of many sequences with one missing):
    >>if you're in the middle of a calibration that requires a lot of
    >>statistics, I can tell you you're not happy to see your job stopped
    >>because of this kind of things... if you're patient enough to add by hand
    >>all the sequences physically present to the IO module (AddFile(...) 25
    >>times or more) or check systematically the missing sequences and make
    >>some special case in your loop over the files, well, I envy you, I'm not
    >>that patient.
    >>
    >>I don't know the reason for this change in BrMainModule and I'm waiting
    >>for comments on this before I check in back the previous stuff.
    >>
    >>Djam
    >>
    >>Djamel Ouerdane ------------------------------------------o
    >>|  Niels Bohr Intstitute     |  Home:                     |
    >>|  Blegdamsvej 17, DK-2100 Ř |  Jagtvej 141 2D,           |
    >>|  Fax: +45 35 32 50 16      |  DK-2200 Copenhagen N      |
    >>|  Tel: +45 35 32 52 69      |  +45 35 86 19 74           |
    >>|                  http://www.nbi.dk/~ouerdane            |
    >>|                  ouerdane@nbi.dk                        |
    >>o---------------------------------------------------------o
    >>
    >>
    >>
    >>
    >>
    > 
    > 
    > 
    
    
    
    -- 
    Jens Ivar Jřrdre
    e-mail:    JensIvar.Jordre@fi.uib.no
    usually:   Dep. of Phys., Allégt. 55, N-5007 BERGEN, NORWAY
    currently: Bldg 510D, P.O.Box 5000, Upton, NY 11973-5000, USA
    phone:     +1-631-344-4223
    



    This archive was generated by hypermail 2b30 : Mon Nov 05 2001 - 10:56:10 EST