def __init__(self, dataset, user, options):
Task.__init__(self, 'ParseOptions', dataset, user, options)
usage = """%prog [options] <dataset>
The %prog script aims to take a list of samples and process them on the batch system. Submission
may be done serially (by setting --max_threads to 1), or in parallel (the default).
The basic flow is:
1) Check that the sample to run on exists
2) Generate a source CFG
3) Run locally and check everything works with a small number of events
4) Submit to the batch system
5) Wait until the jobs are finished
6) Check the jobs ran OK and that the files are good
Example:
ProductionTasks.py -u cbern -w 'PFAOD*.root' -c -N 1 -q 8nh -t PAT_CMG_V5_10_0 --output_wildcard '*.root' --cfg PATCMG_cfg.py /QCD_Pt-1800_TuneZ2_7TeV_pythia6/Summer11-PU_S3_START42_V11-v2/AODSIM/V2
It is often useful to store the sample names in a file, in which case you could instead do:
ProductionTasks.py -w '*.root' -c -N 1 -q 8nh -t PAT_CMG_V5_10_0 --output_wildcard '*.root' --cfg PATCMG_cfg.py `cat samples_mc.txt`
An example file might contain:
palencia%/Tbar_TuneZ2_tW-channel-DR_7TeV-powheg-tauola/Summer11-PU_S4_START42_V11-v1/AODSIM/V2
benitezj%/ZZ_TuneZ2_7TeV_pythia6_tauola/Summer11-PU_S4_START42_V11-v1/AODSIM/V2
wreece%/ZJetsToNuNu_100_HT_200_7TeV-madgraph/Summer11-PU_S4_START42_V11-v1/AODSIM/V2
The CASTOR username for each sample is given before the '%'.
Each step in the flow has a task associated with it, which may set options. The options for each task are
documented below.
"""
self.das = Das.DASOptionParser(usage=usage)
from __future__ import print_function
from PhysicsTools.HeppyCore.utils.edmIntegrityCheck import PublishToFileSystem, IntegrityCheck
import das
import copy, os
if __name__ == '__main__':
from optparse import OptionParser, OptionGroup
usage = """usage: %prog [options] /Sample/Name/On/Castor
e.g.: %prog -u wreece -p -w 'PFAOD_*.root' /MultiJet/Run2011A-05Aug2011-v1/AOD/V2
"""
das = das.DASOptionParser(usage=usage)
group = OptionGroup(das.parser, 'edmIntegrityCheck Options',
'Options related to checking files on CASTOR')
group.add_option("-d",
"--device",
dest="device",
default='cmst3',
help="The storage device to write to, e.g. 'cmst3'")
group.add_option(
"-n",
"--name",
dest="name",
default=None,
help='The name of the dataset in DAS. Will be guessed if not specified'
)