Embedded Files
To run an interactive STNTUPLE analysis job you need to open a ROOT session, and load and run the stnana.C script:
root [0] .L stnana.C root [1] stnana("file","nts.user.egun_stnmaker.xxx.000001.stn","","","val_stn(11,28)/save=egun_stnmaker.hist",100) stnana provides an interface to STNTUPLE event loop utility. The call parameters specify
- dataset to run on - first 4 parameters, see explanation below
- configuration of the analysis job - which analysis code should be executed within the event loop. In the given example
- number of events to run on
The example above runs the analysis job configured by val_stn(11,28) function on first 100 events of the file nts.user.egun_stnmaker.xxx.000001.stn. '11' and '28' are the call parameters passed to val_stn().
The following example
root[1] stnana("detsim","cnvs0202","","","track_ana()",10000)
the track_ana() function on first 10000 events of the dataset defined by the book="detsim" and dsname="cnvs0202".
- WARNING: track_ana() function is not defined by default, don't try to call it.
stnana call format:
void stnana (TString Book, TString Dataset, TString Fileset = "", TString File = "",TString JobName="lumi()",int NEvents = 0) ;
First 4 parameters describe the input dataset: (read more on STNTUPLE datasets)
Book : the book name, used in cataloging. One can have several books and have datasets with the same defined in different books
book = "file" : run on a single file, in this case the 2nd parameter (dataset) is the file name
Dataset : the dataset name
Fileset : fileset name, fileset is just a part of a dataset, several files located in the same directory. Division of a dataset into filesets is purely logical.
A fileset combines several files, such that it is convenient to run a single analysis job on a fileset
File : file name
JobName : name of the configuration script, for example, "catalog()".
JobName may include options separated by '/', i.e. "catalog()/save=a.hist/grl=none,101,500"
allowed options:
- /save: defines name of the histogram file. When the job finishes, the histograms will be saved into a file with the predefined name. The file name can't include '/' characters, so the file has to be stored in the current directory
NEvents : number of events to be processed, NEvents<=0: process all events ​
Google Sites
Report abuse