hristov [Mon, 8 Dec 2014 14:18:37 +0000 (15:18 +0100)]
In the member function
GetLHCPeriodAgainstAlienFile a TUUID object is used to ensure
uniqueness of the target filename of the XML file to download.
However, TUUID is really only unique across hosts - not on the
same host. IF the above member function is executed by parallel
PROOF-Lite (same host) slaves at roughly the same time, they
will all get the same TUUID. The solution is to use
TSystem::TempDirectory to get a unique (to the system)
directory. In fact, one can do away with the TUUID altogether,
because all we need is a unique file name _on_ _the_ _system_ -
not world unique.
hristov [Mon, 8 Dec 2014 14:06:36 +0000 (15:06 +0100)]
In addition to producing 3 2D histograms of the _summed_ radiation and
absorption lengths and density, _also_ produce 3 2D (r,z) histograms
of the _local_ radiation and absorption lengths and density.
This is very useful to identify "hotspots" in the material.
hristov [Mon, 8 Dec 2014 14:01:40 +0000 (15:01 +0100)]
Use "AddFast" rather than "Add" when adding files to a collection.
In principle one _can_ get duplicates this way, but in practise it never
happens. The advantage is that it scales and one doesn't have to wait
minutesvor longer for the dataset to build.
zconesa [Mon, 8 Dec 2014 13:21:24 +0000 (14:21 +0100)]
Remove the possibility (flag) to recompute the PV as it rewrites the vertex information on the fly and affects other tasks. The method to recompute it is kept, but the flag to enable it is removed to avoid possible user mistakes.
mfasel [Sat, 6 Dec 2014 20:10:36 +0000 (12:10 -0800)]
Fix for broken compilation
=====================================
Add header files for types where functionality is used in the header file of AliJetContainer.
Fixes broken compilation on MacOS.
rbertens [Fri, 5 Dec 2014 14:27:04 +0000 (15:27 +0100)]
increase ease-of-use of jet v2 unfolding framework
- user can now more easily tune correlation between in-plane, out-of-plane yield
- make predictions of uncertainties for higher statistics
- minimize chi2 and caluclate p-values (is now done on-the-fly using static arrays)
- no more hardcoded parameters in .cxx or .h
- output to screen is less cluttered, some info printed in C style for further processing if desired
miweber [Fri, 5 Dec 2014 10:54:56 +0000 (11:54 +0100)]
Removing centrality check from default:
- return 1 for accepted events in case centrality is switched OFF
- switching on centrality with SetCentralityPercentileRange(min,max)
miweber [Thu, 4 Dec 2014 21:06:17 +0000 (22:06 +0100)]
changes in BF drawing macros:
- proper ROOT averaging in case of merging balance function histograms
- adding one more centrality class (0-10000) for MC V0M multiplicity binning
miweber [Thu, 4 Dec 2014 20:55:10 +0000 (21:55 +0100)]
Adding HF daughter tracks to filter bit QA task:
- avoid double counting (to be improved)
- possibility to exclude filter bits from QA (e.g. hybrids)
- so far working only for 2 prong decays (need to understand how subsequent decays are handled)
saiola [Thu, 4 Dec 2014 20:52:41 +0000 (15:52 -0500)]
Implement option to copy MC label from clusters to cells (for MC productions where MC labels are not available in cells). Copied and simplified from AliEMCALTenderSupply
mkrzewic [Thu, 4 Dec 2014 12:56:01 +0000 (13:56 +0100)]
provide better debug information and use the lib
- put log statistics in the email and attach: full log summary,
stack traces and stack trace plots
- use the PWGPP/scripts/utilities.sh lib instead of locally defined
functions
rbertens [Thu, 4 Dec 2014 13:05:17 +0000 (14:05 +0100)]
calibrate the vzero signal that goes into the total Q-vector which can be retrieved via AliFlowEventSimple::GetQ(...)
- the calibrated Q-vector is now called via AliFlowEvent::GetQ()
- calibration is performed performed automatically when the flow event is filled using the flow track cuts object that is set to provide VZERO data (i.e. via calling GetStandardVZEROOnlyTrackCuts() )
- correct calibration routine will be determined from the run-number
- 10h: Q2 and Q3 are calibrated
- 11h: only Q2 is calibrated
- change is backward compatible