bugfix https://savannah.cern.ch/bugs/?66219

update the track parameters at kTPCin after kTPCout
the exact sequence needs to be investigated

diff --git a/HLT/global/AliHLTGlobalEsdConverterComponent.cxx b/HLT/global/AliHLTGlobalEsdConverterComponent.cxx
index 7b2d391afa4dd3b7e9732832511a46c25d2efefa..d37123c3c8c49494633a692cad216538d08e747b 100644 (file)
--- a/HLT/global/AliHLTGlobalEsdConverterComponent.cxx
+++ b/HLT/global/AliHLTGlobalEsdConverterComponent.cxx
@@ -441,14 +441,21 @@ int AliHLTGlobalEsdConverterComponent::ProcessBlocks(TTree* pTree, AliESDEvent*
        //   pid signal
        // The first one can be updated already at that stage here, while the two others
        // eventually require to update from the ITS tracks before. The exact scheme
-       // needs to be checked 
+       // needs to be checked
+       // 2010-05-12 TODO: the outer parameter is set when updating with kTPCout but
+       // also when updated with kITSout. So the value from here is overwritten further
+       // down. Comes along with the necessity to check the full sequence.
        iotrack.SetID( element->TrackID() );
-       iotrack.UpdateTrackParams(&(*element),AliESDtrack::kTPCin);
        {
          AliHLTGlobalBarrelTrack outPar(*element);       
          outPar.AliExternalTrackParam::PropagateTo( element->GetLastPointX(), fSolenoidBz );
          iotrack.UpdateTrackParams(&outPar,AliESDtrack::kTPCout);
        }
+       iotrack.UpdateTrackParams(&(*element),AliESDtrack::kTPCin);
+       // 2010-05-12: investigations ongoing to find out what needs to be set in
+       // the ESD track in order to get the tracks correctly displayed by the offline
+       // macro.
+       //iotrack.SetStatus(AliESDtrack::kTPCrefit);
        iotrack.SetTPCPoints(points);
        if( element->TrackID()<ndEdxTPC ){
          iotrack.SetTPCsignal( dEdxTPC[element->TrackID()], 0, 0 );