Home Computing DAQ E-mail Notes Meetings Subsystems Search

 

Version

v12r6
Release Date 19 January 2005
Description

 

Backward-compatible HLT development version.

Based on Gaudi v15r5 and LHCb v16r4. Includes LoKi v3r7.

 

New version with some modifications and many additions.

First full release of HLT. For experts only.

 
Main modifications Major Modifications:
  • CombinedParticleMaker:  Set ExclusiveSelection to false by default.
     !!! WARNING !!! This is a major change that will change the result of all selections using the CombinedParticleMaker without resetting this option. Actually, most stripping preselections use ExclusiveSelection = true ;
     
  • All Interfaces are moved from $DAVINCIKERNELROOT/DaVinciTools/ to $DAVINCIKERNELROOT/Kernel/ for compatibility with LHCb. 
    Hence: Replace all ``DaVinciTools/'' by ``Kernel/'' in all code. The DaVinciTools/ directory is still there as long as there is some code not yet adapted. It contains dummy files including the proper file.
     
  • New doxygen style using dogygen 1.5. The main "feature" is that now classes taken from LHCb and Gaudi are linked to and not included. A disadvantage is that they do not appear in the alphabetical list anymore. If you want to look at the "GaudiTool" doc, either go to Gaudi under "See also:" or follow a link from any tool. Have a look here.

More new features:

  • On popular demand, the IParticleFilter now includes methods
      StatusCode filterByPID (const ParticleVector &, ParticleVector &, const std::string &, const bool alsoCC=true)=0
      StatusCode filterByPID (const ParticleVector &, ParticleVector &, const int &pid, const bool alsoCC=true)=0
     
  • The code
          ParticleVector myKaons, myPiPlus, myPiMinus;
        sc = particleFilter()->filterByPID(parts, myKaons,   "K+");
        sc = particleFilter()->filterByPID(parts, myPiPlus,  "pi+", false);
        sc = particleFilter()->filterByPID(parts, myPiMinus, "pi-", false);
       
        would return a vector of kaons (both signs) and pions.
       
  • New algorithm IsBEvent in DaVinciMCTools that easily allows to select events containing a b quark. Actually works for any quark or particle(s).
     
  • New tool family "IPlotTool" with implementations SimplePlotTool and RecursivePlotTool. Make plots for a given Particle or ParticleVector. The plots have to be configured by options. The algorithm SimplePlots does nothing more than calling such tools:   
          HLTPlots.Members += { "SimplePlots/HLTB2HHPlots" };
          HLTB2HHPlots.PhysDesktop.InputLocations = { "Phys/HLTFilterB2HH" } ;
          HLTB2HHPlots.Plots.Variables = { "M", "Chi2", "Pt" } ;   
        plots the mass, chi2 and Pt of all particles in "Phys/HLTFilterB2HH".
       
  • Small changes in FlavourTagging. No need to run L1 anymore.
     
  • The Tampering is released. Ask Mariusz Witek for help in you want to run it.
     
  • First full pre-release of the HLT
    • New version of HltGeneric
    • Use of TrgMuon that allows to skip TT for muons
    • First release of HltSelections
    • First release of HltSelChecker
    • All this is for experts. Contact Patrick.Koppenburg@cern.ch if you need to run the HLT prototype.
       
  • New preselections DsSelector and Bc2RhoPi.
     
  • Few modifications in LoKi.
Known problems: Troubles with RFIO: Crashes in PoolRootTreeEvtCnvSvc. Try to use the following trick, which seems to work fine:
  • Copy the file
    /afs/cern.ch/user/p/pkoppenb/public/.rootauthrc
    to your home directory
  • Make sure there is no .rootrc file there, or at least that you know what it is doing.
  • Replace all the "rfio:" by "castor:" in your input data. It will look like
    EventSelector.Input   = {
    "DATAFILE='PFN:castor:/castor/cern.ch/lhcb/DC04/00000541_00000665_9.dst'
    TYP='POOL_ROOTTREE' OPT='READ'"};
    This will pick up rootd instead of rfio.
If your execution hangs after the DaVinci welcome-banner on slc3, it means you have made a typo in the name of some option file you are including, or you are using an undefined environment variable. There's no straightforward way to debug this.
Documentation Doxygen
Release notes (for detailed release notes of each packages look in the release area)
Requirements
Installation follow the LHCb installation procedure
tar files of the sources
web access to CVS repository
web access to AFS release area

DaVinci packages modified with respect to v12r5

 
DaVinciSys v12r6 New doxygen style and location
Calo/ CaloReco v2r7 A few steps to fast algorithm
Event/ TrgEvent v5r3 Links from TrgTrack to Muon
Hlt/ HltGeneric v2r1 Reshuffle sequencing
HltSelections v2r0 New
HltSelChecker v2r0 New
L0/ L0MuonKernel v2r4 Some cleaning
Muon/

No changes
Phys/ DaVinciAssociators v5r2p2  
DaVinciFilter v1r4 New ParticleFilter methods
DaVinciKernel v3r0 Move DaVinciTools/ to Kernel/
DaVinciMCTools v5r8 New IsBEvent
DaVinciTest v5r5 Use Trg in PrimVertexAnalysis
DaVinciTools v11r4 New plot tools
DaVinciTransporter v2r0 Some rewriting
DaVinci v12r6 New doxygen location
FlavourTagging v5r3p1 Don't need L1 anymore
ParticleMaker v1r2 ExclusiveSelection = false
Tampering v1r0 New
VertexFit v1r3 Treat photons correctly in the unconstrained vertex fitter
PhysSel/ PhysSelSys v7r14
Bc2RhoPi v1r0
Bs2DsDs v2r0
Bs2JpsiPhi v2r3p2
DsSelector v1r1
Rec/

No changes
Tools/ LoKi v3r7
LoKiHlt v2r0p1
Stripping v1r6 Reflect ExclusiveSelection
Utilities v1r13 Minor changes
Tr/

No changes
Trg/ L1Decision v4r1 Checking
TrgCalo v1r0p1 Make sequencer
TrgMuon v1r2 New
TrgVelo v6r0p1 Bug fix for Primary Vertex
TrgSys v1r2 Reshuffle sequencing
LHCb  packages used LHCb v16r4 (refer to the requirements files for full package list and corresponding release notes for details)
Gaudi packages used based on Gaudi v15r5.