Difference: TextToNtupleParser (14 vs. 15)

Set up the environment - AFS

If you have a Glasgow AFS account you can set up the environment by sourcing the script v16g.sh in /afs/phas.gla.ac.uk/user/g/gkirby/public. This will then allow you to use the runFillFlatFromAscii command:

runFillFlatFromAscii teststeerFillFlatFromAscii.txt PythiaOutput.dat FlatFillATLAS.root

The first argument is the steering file - this points to the VariableTree and TreeSpec files which identify the variables that the parser will read. It is important to ensure that the correct files are pointed to and that their contents are up to date, or the flat tuple will not be correctly filled. It is convenient to use different steering files to point to different "event" TreeSpec files (e.g. for different processes). The "global" file contains metadata relevant to downstream analysis.

The second argument is the output of the PythiattH code, i.e. a text file containing the relevant event output variables.

The last argument is the name of the flat tuple that will be created.

Important notes on running the parser

• Remember to increment the number after EvInfoTree for each variable that is added to the VariableTree file. If conflicting numbers are assigned to two different variables the tool will fail to run.

• The value to the left of the colon in the VariableTree file has to match the name in the text file from the event generator, and must also match what is in the TreeSpec file. It does not have to be the same as the stuff to the right. In the downstream analysis this is used as the input variable to plotting and the neural net training. This however is the point where you are flexible in mapping the branch/leaf to that name of the varaible later.

• The stuff to the right of the colon is the branch/leaf as it appears in root.

Set up the environment - CDF (obsolete)