hristov [Tue, 23 Mar 2010 13:14:08 +0000 (13:14 +0000)]
Add a new variable, that is the radial position of the track at the end
of the muon absorber, both in ESD and AOD. This variable has proven to
be extremely useful to better understand the effects of the absorber in
the data taken by the muon spectrometer.
(Philippe P.)
as their only function was to store temporarily the beam type in AliRun to be transferred to AliSimulation. This information is used for the trigger configuration. Instead, we have introduced the two methods
To store and get directly this info from the AliSimulation class. This change is completely transparent, however what is NOT transparent is the setting of the beam type which is done in the Config.C. Typically the change is the following
All the macros in AliRoot have been changed, but also all private macros will have to be changed. The symptom will be very easy to detect, as CINT will complain that there is no method SetTriggerConfig in AliRun.
One subtle point here is that with this change Config.C will require the presence of AliSimulation. If it is not there, it will be created by the Instance() call. This is not a problem in normal usage, but if Config.C is not used for simulation, it will create an instance of AliSimulation, that will not be used. Again, this should not be a problem, but users might as well be aware o fit.
richterm [Tue, 23 Mar 2010 10:55:17 +0000 (10:55 +0000)]
bugfix for an issue with recent autoconf versions on Ubuntu 10
with autoconf 2.64 AC_CHECK_HEADERS behaves undefined if the first parameter is empty
it looks like some arbitrary content is taken from somewhere resulting in wrong checks
for include files 'dummy' and 'rootcint'. Added a check and calling the macro only if
there is something to check.
richterm [Mon, 22 Mar 2010 17:39:18 +0000 (17:39 +0000)]
updating macros (Theo)
- no necessity to circumvent the reconstruction problem anymore, as it was solved
- add possibility to use disableHLTflag without adding custom arguments
richterm [Mon, 22 Mar 2010 17:33:49 +0000 (17:33 +0000)]
updating build checks for the HLT TRD (Theo)
- adding check for the availibility of improved AliTRDrawStream (r39608)
- removing check for AliTRDrecoParam r34995
- requirering v4-18-Release
ivana [Mon, 22 Mar 2010 11:48:16 +0000 (11:48 +0000)]
In AliMUONSDigitizerV2:
For the event interaction time, SDigitizer now gets the interaction time stored in the event header instead of loading the stack
(Mercedes L.N.)
kleinb [Mon, 22 Mar 2010 10:06:31 +0000 (10:06 +0000)]
Do not stream event by event pointers and variables, avoid errors when writing the object with the AnalysisManager
E-TStreamerInfo::Build:: AliCdfJetFinder: varContainer** has no streamer or dictionary, data member fVectParticle will not be saved (etc.)
bhippoly [Fri, 19 Mar 2010 17:16:57 +0000 (17:16 +0000)]
SUMMARY = CFContainer improvement + step towards bookkeeping
-------
1. CheckCascade :
* Andrei's advice : 1. implementation of a destructor /2. TList::SetOwner /3. PostData in UserCreateOutputObjects /4. PostData in accordance with return wihtin UserExec;
* Settings related to (quality) analysis cuts now as data members (needed for bookkeeping);
* Step towards bookkeeping with a TPaveText (commented ! not supported by the framework);
* Introduction of new TH1F or change of name :
- fHistCascadeMultiplicityBeforeTrigSel,
- fHistCascadeMultiplicity becomes fHistCascadeMultiplicityForTrigEvt,
- fHistTrackMultiplicity becomes fHistTrackMultiplicityForTrigEvt,
- fHistTPCrefitTrackMultiplicity becomes fHistTPCrefitTrackMultiplicityForTrigEvt,
- fHistCascadeMultiplicityForSelEvt, - fHistPosBestPrimaryVtx(X,Y,Z)ForSelEvt,
- fHistPosTrkgPrimaryVtx(X,Y,Z) become fHistPosTrkgPrimaryVtx(X,Y,Z)ForCascadeEvt,
- fHistPosBestPrimaryVtx(X,Y,Z) become fHistPosBestPrimaryVtx(X,Y,Z)ForCascadeEvt;
* fCFContCascadeCuts reshuffled : 1. 20 dimensions [+Y(Xi) +Y(Omega)] / 2. 1 step = 1 species.
2. CheckPerformanceCascade :
* Andrei's advice : 1. implementation of a destructor /2. TList::SetOwner /3. PostData in UserCreateOutputObjects /4. PostData in accordance with return wihtin UserExec;
* Settings related to (quality) analysis cuts now as data members (needed for bookkeeping);
* Step towards bookkeeping with a TPaveText (commented ! not supported by the framework);
* AliCFContainer: Introduction of fCFCont*As*CascadeCuts = like fCFContCascadeCuts but for associated-to-MC particles
3. AddTaskCheckCascade + AddTaskCheckPerformanceCascade
* Settings corresponding to the new features above (choices hard-coded into each macro, != arguments)
(A.Maire)
morsch [Fri, 19 Mar 2010 14:34:36 +0000 (14:34 +0000)]
Updates for SPD QA
I have modified the SPD task for the QA train for two reasons :
- revisiting included histograms
- getting rid of all the online-offline numbering conventions in the task.
For the second item I have created another class, AliSPDUtils, which makes the all the conversions (it is made up of static methods only).
I attached my modified runPilot.C macro which I used to test everything locally (basically I removed the loading of libraries and I commented
out the setcollisioncandidates for my task).
richterm [Fri, 19 Mar 2010 10:10:50 +0000 (10:10 +0000)]
adding option to skip some of the std content of the ESD, by default
AliESDZDC,AliESDFMD,Cascades,Kinks,AliRawDataErrorLogs,AliESDACORDE are skipped
component argument '-skipobject=name1,name2' can be used the change the list of
objects, names refer to the ESD std object identifiers.
Speedup for pp min bias 2.5, for Hijing 1000 PbPb 1.4
dsilverm [Fri, 19 Mar 2010 00:34:32 +0000 (00:34 +0000)]
more standardizing of result return codes + always return tmax info + loose parameter limits and consistency checks + ADC smearing for no fits + try-catch for possible minuit exceptions