def createDatasetOutTag(dataset,tag='',datatier=None,year=None,version=""):
"""Create dataset output tag from DAS path."""
outtags = dataset.strip('/').split('/')
assert len(outtags)>=2, "Invalid DAS path '%s'!"%(dataset)
outtag = outtags[1]
outtag = outtag.replace(getUsernameFromSiteDB()+'-',"")
outtag = hashpattern.sub("",outtag)
dtype = 'data' if '/Run201' in dataset else 'mc'
if datatier=='nanoAOD':
if 'miniaod' in outtag.lower():
newtier = 'NanoAOD'+version
outtag = minipattern.sub(newtier,outtag)
globaltag = globaltags[dtype]['nanoAOD'].get(year,False)
if globaltag:
outtag = outtag[:outtag.index(newtier)+len(newtier)]+'_'+globaltag
if tag and not outtag.endswith(tag):
outtag += formatTag(tag)
return outtag
def submitSampleToCRAB(pset, year, samples, **kwargs):
"""Create a CRAB configuration and submit a given list of samples."""
assert isinstance(
samples,
list), "Samples list should be a list or tuple! Given %s" % samples
# USER OPTIONS
year = year
test = kwargs.get('test', 0)
force = kwargs.get('force', False)
datatier = 'nanoAOD' if 'nanoaod' in pset.lower() else 'miniAOD'
version = re.findall("(?<=AOD)v\d+", pset)
version = version[0] if version else ""
pluginName = 'Analysis' #'PrivateMC'
splitting = kwargs.get(
'split',
'FileBased') #if year==2018 or datatier=='nanoAOD' else 'Automatic')
tag = kwargs.get('tag', "")
instance = kwargs.get('instance', 'global')
nevents = -1
unitsPerJob = 1 # files per job for 'FileBased'
eventsPerJob = kwargs.get(
'eventsPerJob',
10000) # unitsPerJob for 'EventAwareLumiBased' splitting
njobs = -1
ncores = kwargs.get('ncores', 1) # make sure nCores > nThreads in pset.py
maxRunTime = kwargs.get('maxRunTime', 6 * 60) #1250 # minutes
maxMemory = kwargs.get('maxMemory', 3000) # MB
priority = kwargs.get('priority', 10)
workArea = "crab_tasks" #"crab_projects"
outdir = '/store/user/%s/%s_%s%s' % (getUsernameFromSiteDB(), datatier,
year, formatTag(tag))
publish = True #and False
site = 'T2_CH_CSCS'
# OVERRIDE
if test > 0:
splitting = 'FileBased'
unitsPerJob = 1 # files per job
njobs = int(test)
outdir += '_test'
publish = False
samples = samples[:1]
if nevents < 0:
nevents = 2500
if splitting == 'Automatic':
unitsPerJob = -1
njobs = -1
maxRunTime = -1
if splitting == 'EventAwareLumiBased':
unitsPerJob = eventsPerJob
njobs = -1
# PRINT
print ">>> " + '=' * 70
print ">>> year = %s" % year
print ">>> pset = '%s'" % bold(pset)
print ">>> pluginName = '%s'" % pluginName
print ">>> splitting = '%s'" % splitting
print ">>> unitsPerJob = %s" % unitsPerJob
print ">>> nevents = %s" % nevents
print ">>> tag = '%s'" % bold(tag)
print ">>> njobs = %s" % njobs
print ">>> nCores = %s" % ncores
print ">>> maxRunTime = %s" % maxRunTime
print ">>> maxMemory = %s" % maxMemory
print ">>> priority = %s" % priority
print ">>> workArea = '%s'" % workArea
print ">>> site = '%s'" % site
print ">>> outdir = '%s'" % outdir
print ">>> publish = %r" % publish
print ">>> test = %r" % test
print ">>> " + '=' * 70
if len(samples) == 0:
print ">>> No samples given..."
print ">>> "
return
# CRAB CONFIGURATION
config = crabconfig()
config.General.workArea = workArea
config.General.transferOutputs = True
config.General.transferLogs = False
config.JobType.pluginName = pluginName
config.JobType.psetName = pset
config.JobType.pyCfgParams = ["year=%s" % year, "nThreads=%s" % ncores]
config.JobType.numCores = ncores
if maxRunTime > 0:
config.JobType.maxJobRuntimeMin = maxRunTime # minutes
if maxMemory > 0:
config.JobType.maxMemoryMB = maxMemory # MB
config.JobType.priority = priority
config.Data.splitting = splitting
if unitsPerJob > 0:
config.Data.unitsPerJob = unitsPerJob
if njobs > 0:
config.Data.totalUnits = unitsPerJob * njobs
config.Site.storageSite = site
config.Data.outLFNDirBase = outdir
config.Data.publication = publish
for dataset in samples:
# INDIVIDUAL CONFIG
request = (datatier.lower().replace('aod', '') + '_' +
shortenDASPath(dataset))[:100]
private = dataset.endswith('/USER')
sites = getSampleSites(dataset, instance=None)
if private:
ignoreLocal = True
inputDBS = "https://cmsweb.cern.ch/dbs/prod/phys03/DBSReader/"
whitelist = getOptimalWhitelist(sites, instance=instance)
#whitelist = ['T2_CH_*','T2_DE_*','T2_IT_*']
else:
ignoreLocal = False
inputDBS = "https://cmsweb.cern.ch/dbs/prod/%s/DBSReader/" % instance
whitelist = []
outtag = createDatasetOutTag(dataset,
tag=tag,
datatier=datatier,
version=version,
year=year)
# PRINT
print ">>> " + '-' * 5 + " Submitting... " + '-' * 50
print ">>> request = '%s'" % bold(request)
print ">>> dataset = '%s'" % bold(dataset)
print ">>> inputDBS = '%s'" % inputDBS
print ">>> sites = %s" % sites
print ">>> whitelist = %s" % whitelist
print ">>> ignoreLocal = %s" % ignoreLocal
print ">>> outtag = '%s'" % outtag
print ">>> " + '-' * 70
# INDIVIDUAL CONFIG
config.General.requestName = request # max. 100 characters
config.Data.inputDataset = dataset
config.Data.inputDBS = inputDBS
#config.Data.outputPrimaryDataset = 'LQ_test' # only for 'PrivateMC'
config.Data.outputDatasetTag = outtag
config.Data.ignoreLocality = ignoreLocal # do not run on same site the dataset is stored on
if whitelist:
config.Site.whitelist = whitelist
print str(config).rstrip('\n')
print ">>> " + '-' * 70
# SUBMIT
if force:
print ">>> Do you want to submit this job to CRAB? [y/n]? force"
print ">>> Submitting..."
submitCRABConfig(config)
else:
while True:
submit = raw_input(
">>> Do you want to submit this job to CRAB? [y/n]? "
).strip().lower()
if any(s in submit for s in ['quit', 'exit']):
print ">>> Exiting..."
exit(0)
elif 'force' in submit:
submit = 'y'
force = True
if 'y' in submit:
print ">>> Submitting..."
submitCRABConfig(config)
break
elif 'n' in submit:
print ">>> Not submitting."
break
else:
print ">>> '%s' is not a valid answer, please choose 'y' or 'n'." % submit
print ">>> "
options.register('year', year, mytype=VarParsing.varType.int)
options.register('nevents', nevents, mytype=VarParsing.varType.int)
options.register('seed', seed, mytype=VarParsing.varType.int)
options.register('nThreads', nThreads, mytype=VarParsing.varType.int)
options.parseArguments()
sample = options.sample
index = options.index
gridpack = os.path.abspath(options.gridpack)
year = options.year
nevents = options.nevents
seed = options.seed
nThreads = options.nThreads
###globaltag = globaltags['miniAOD'].get(year,'auto:phase1_2017_realistic')
tag = sample
if index>0:
tag += formatTag('_'+str(index))
outfile_RAW = "file:GENSIM%s.root"%(tag)
outfile_LHE = "file:GENSIM_LHE%s.root"%(tag)
print ">>> sample = '%s'"%sample
print ">>> index = %s"%index
print ">>> gridpack = '%s'"%gridpack
print ">>> year = %s"%year
print ">>> nevents = %s"%nevents
print ">>> nThreads = %s"%nThreads
print ">>> globaltag = '%s'"%globaltag
print ">>> outfile_RAW = '%s'"%outfile_RAW
print ">>> outfile_LHE = '%s'"%outfile_LHE
print ">>> "+'-'*52
# PROCESS
options = VarParsing('analysis')
options.register('year', year, mytype=VarParsing.varType.int)
options.register('nThreads', nThreads, mytype=VarParsing.varType.int)
options.register('events', maxEvents, mytype=VarParsing.varType.int)
options.register('sample', sample, mytype=VarParsing.varType.string)
options.register('dtype', dtype, mytype=VarParsing.varType.string)
options.parseArguments()
year = options.year
dtype = options.dtype
nThreads = options.nThreads
maxEvents = options.events
sample = options.sample
globaltag = globaltags[dtype]['NanoAODv5'].get(year,'auto:phase1_2017_realistic')
era = eras['NanoAODv5'].get(year,None)
if index>0:
outfile = "file:nanoAOD_%s%s_%s.root"%(year,formatTag(sample),index)
else:
outfile = "file:nanoAOD_%s%s.root"%(year,formatTag(sample))
if year==2016: infiles = filter(lambda f: 'RunIISummer16' in f or '/Run2016' in f or '_2016_' in f,infiles)
elif year==2017: infiles = filter(lambda f: 'RunIIFall17' in f or '/Run2017' in f or '_2017_' in f,infiles)
elif year==2018: infiles = filter(lambda f: 'RunIIAutumn' in f or '/Run2018' in f or '_2018_' in f,infiles)
if dtype=='data': infiles = filter(lambda f: '/store/mc/' not in f,infiles)
elif dtype=='mc': infiles = filter(lambda f: '/store/data/' not in f,infiles)
print ">>> sample = '%s'"%sample
print ">>> index = %s"%index
print ">>> year = %s"%year
print ">>> dtype = '%s'"%dtype
print ">>> maxEvents = %s"%maxEvents
print ">>> globaltag = '%s'"%globaltag
print ">>> infiles = %s"%infiles
#director+"/store/mc/RunIIAutumn18MiniAOD/PairVectorLQ_InclusiveDecay_M-1100_TuneCP2_13TeV-madgraph-pythia8/MINIAODSIM/102X_upgrade2018_realistic_v15-v1/30000/F3649753-9038-CF4B-A178-BD03655BE7BD.root",
"file:input/VLQ-p_M1100_%s.root" % year,
#"file:miniAOD_%s_%s_%s.root"%(sample,year,index),
]
# USER OPTIONS
from FWCore.ParameterSet.VarParsing import VarParsing
options = VarParsing('analysis')
options.register('year', year, mytype=VarParsing.varType.int)
options.register('nThreads', nThreads, mytype=VarParsing.varType.int)
options.parseArguments()
year = options.year
nThreads = options.nThreads
globaltag = globaltags['mc']['miniAOD'].get(year, 'auto:phase1_2017_realistic')
if index > 0:
outfile = "file:miniAOD_rerun_%s%s_%s.root" % (year, formatTag(sample),
index)
else:
outfile = "file:miniAOD_rerun_%s%s.root" % (year, formatTag(sample))
print ">>> sample = '%s'" % sample
print ">>> index = '%s'" % index
print ">>> year = '%s" % year
print ">>> maxEvents = %s" % maxEvents
print ">>> globaltag = '%s'" % globaltag
print ">>> infiles = %s" % infiles
print ">>> outfile = %s" % outfile
print ">>> " + '-' * 69
# PROCESS
process = cms.Process('TauID')