I am forwarding the following update message from Djamel in rgerad to Brat. > Hi Flemming, > > For some reasons, my emails could not be sent > to the soft and dev lists. Could you please forward > them ? Here they are: I hope you do not use warden as e-mail address. > > Hi all, > > I released a new version of brat, namely 2.0.11, tag > BRAT-2-0-11 > The changes are: > > * updates of BrTofTdcGainModule.cxx/h > > added methods SetMaxGain and SetMinGain and > corresponding private > members to set a range of reasonable > values. If the calculated gain is outside this > range, the > BrTofCalibration::kCalException is used. > > added method SetTimeStep and fTimeStep to give the > user the > possibility to set the pulser time step (the > default is 10 ns) > > in the Event methods, histograms are filled only > if the tdc values are > greater than 1 and lower than 4095. > > in the FindMax method, cleaned up the algorithm > since I got > sometimes some nan values. Removed the bug. > > * added 2 classes in the dir. modules/util: > > BrDbUpdateModule to update the calibration > parameter manager and the > run info manager. > > BrDbCommitModule to commit a new calibration > revision to the database. > > So far I used these modules only in a config. script > working with > the bratmain application. > The BrDbUpdateModule has to be placed carefully in the > module > pipeline for the following reason: > > the run info manager should be the first module to be > initialized > and updated after the main module and the input > module (if you read input data) > whereas the calibration manager should be the last > because the > calibration modules (eg for TOF) have first to > register the parameters > they need (done usually in the Init method). I used a > trick to do that. I > put the db update module after the input module in > the main module and > update the run info manager once in its init method. > Then, come the > calibration modules registering their parameters and > finally the > calibration parameter manager is initialized and > updated once in the > Begin method of the db update module. > > if you scan data files with events coming from > different runs (let's say > 2530, 2531, 2535), provided that you have registered > these runs in the > run info manager in the right order, the db update > module will know when > to update the calibration parameter manager and the > run info manager by > comparing the run number given by the BrEvent and the > one returned by the > current run of the run info manager. If it's > different, then the db > update module calls a private method to update these > managers. > > if you scan events coming from a single run, this > method will never be > called, the appropriate updates have been alreay made > in the Init > and Begin methods. > > If you know a better way to handle more than one run, > you're welcome to > suggest it (I know that Christian is thinking about it > but it's not > trivial at all) but so far, it worked fine in the way > I used it. > > The db commit module is much simpler, the user can set > some start run and > end run. This module will then take the start time of > the first one and > the end time of the last one as calibration revision > bounds. > > > You can of source use the same run as start and end. > I insist on this point: if you use this module in the > main module, make > sure that it is the LAST one in the module list. > > > * updated modules/util/Makefile.am, LinkDef.h, > Include.h > * updated configure.in > > I'll will also commit the bratmain scripts I used for > the tof tdc gain > calibration in my brahms_app area > (do_app/tof/mm_scripts) > I'll let you know when they are committed. > > If you have any problems compiling brat, tell me. > > A last thing: I'm presently tagging brat but got this > message when it came > to bratroot: > > cvs tag: [18:33:16] waiting for uid2485's lock in > /afs/rhic/brahms/BRAHMS_CVS/brat/applications/bratroot > > This user is welcome to unlock cvs. > > Ciao > Djam > > >
This archive was generated by hypermail 2b30 : Wed Jul 18 2001 - 15:44:21 EDT