def submit_job(brunel_app, reco_type, input_files=None, local=RUN_LOCAL):
# Set EvtMax depending on if this is a local job
brunel_app.extraOpts += 'from Configurables import Brunel\n'
brunel_app.extraOpts += 'Brunel().EvtMax = {}'.format(2 * int(local) - 1)
# Configure the corresponding Job
job = Job(name='VP hybrid distortions',
comment='{reco_type} reconstruction {suffix}'.format(
reco_type=reco_type, suffix=['', '(local)'][local]),
application=brunel_app,
splitter=SplitByFiles(filesPerJob=1, ignoremissing=True),
parallel_submit=True)
if local:
job.backend = Local()
job.outputfiles = [LocalFile('*.xdst'), LocalFile('*.root')]
job.inputdata = dataset[:1]
else:
job.backend = Dirac()
job.outputfiles = [DiracFile('*.xdst'), DiracFile('*.root')]
job.inputdata = dataset
job.inputfiles = input_files or []
queues.add(job.submit)
return True
def test_g_Splitters(self):
from Ganga.GPI import Job, GenericSplitter, GangaDataset, GangaDatasetSplitter, LocalFile
# -- SPLITTERS BASICUSE START
j = Job()
j.splitter = GenericSplitter()
j.splitter.attribute = 'application.args'
j.splitter.values = [['hello', 1], ['world', 2], ['again', 3]]
j.submit()
# -- SPLITTERS BASICUSE STOP
# -- SPLITTERS SUBJOBCHECK START
j.subjobs
j.subjobs(0).peek("stdout")
# -- SPLITTERS SUBJOBCHECK STOP
# -- SPLITTERS MULTIATTRS START
j = Job()
j.splitter = GenericSplitter()
j.splitter.multi_attrs = {'application.args': ['hello1', 'hello2'],
'application.env': [{'MYENV':'test1'}, {'MYENV':'test2'}]}
j.submit()
# -- SPLITTERS MULTIATTRS STOP
# -- SPLITTERS DATASETSPLITTER START
j = Job()
j.application.exe = 'more'
j.application.args = ['__GangaInputData.txt__']
j.inputdata = GangaDataset( files=[ LocalFile('*.txt') ] )
j.splitter = GangaDatasetSplitter()
j.splitter.files_per_subjob = 2
j.submit()
def testDatasets(self):
from Ganga.GPI import DiracFile, PhysicalFile, LHCbDataset, Job, LocalFile
# test constructors/setters
ds = LHCbDataset(['lfn:a', 'pfn:b'])
assert len(ds) == 2
print(ds[0])
assert isinstance(ds[0], DiracFile)
assert isinstance(ds[1], PhysicalFile)
ds = LHCbDataset()
ds.files = ['lfn:a', 'pfn:b']
assert isinstance(ds[0], DiracFile)
assert isinstance(ds[1], PhysicalFile)
ds.files.append('lfn:c')
assert isinstance(ds[-1], DiracFile)
d = OutputData(['a', 'b'])
assert isinstance(d.files[0],str)
assert isinstance(d.files[1],str)
# check job assignments
j = Job()
j.inputdata = ['lfn:a', 'pfn:b']
assert isinstance(j.inputdata, LHCbDataset)
j.outputfiles = ['a', DiracFile('b')]
assert isinstance(j.outputfiles[0], LocalFile)
print(type(j.outputfiles[1]))
assert isinstance(j.outputfiles[1], DiracFile)
LFN_DATA = ['LFN:/lhcb/LHCb/Collision11/DIMUON.DST/00016768/0000/00016768_00000005_1.dimuon.dst',
'LFN:/lhcb/LHCb/Collision11/DIMUON.DST/00016768/0000/00016768_00000006_1.dimuon.dst']
ds = LHCbDataset(LFN_DATA)
assert len(ds.getReplicas().keys()) == 2
assert ds.getCatalog()
def testJobCopy(self):
"""Test that a job copy copies everything properly"""
from Ganga.GPI import Job, ARC, GenericSplitter, GangaDataset, LocalFile, FileChecker
from Ganga.GPIDev.Base.Proxy import isType
j = Job()
j.application.exe = "sleep"
j.application.args = ['myarg']
j.backend = ARC()
j.backend.CE = "my.ce"
j.inputdata = GangaDataset()
j.inputdata.files = [ LocalFile("*.txt") ]
j.inputfiles = [ LocalFile("*.txt") ]
j.name = "testname"
j.outputfiles = [ LocalFile("*.txt") ]
j.postprocessors = FileChecker(files=['stdout'], searchStrings = ['my search'])
j.splitter = GenericSplitter()
j.splitter.attribute = "application.args"
j.splitter.values = ['arg 1', 'arg 2', 'arg 3']
j2 = j.copy()
# test the copy has worked
self.assertTrue( isType(j2, Job) )
self.assertEqual( j2.application.exe, "sleep" )
self.assertEqual( j2.application.args, ["myarg"] )
self.assertTrue( isType(j2.backend, ARC) )
self.assertEqual( j2.backend.CE, "my.ce" )
self.assertTrue( isType(j2.inputdata, GangaDataset) )
self.assertEqual( len(j2.inputdata.files), 1 )
self.assertTrue( isType(j2.inputdata.files[0], LocalFile) )
self.assertEqual( j2.inputdata.files[0].namePattern, "*.txt" )
self.assertEqual( len(j2.inputfiles), 1 )
self.assertTrue( isType(j2.inputfiles[0], LocalFile) )
self.assertEqual( j2.inputfiles[0].namePattern, "*.txt" )
self.assertEqual( j2.name, "testname" )
self.assertEqual( len(j2.outputfiles), 1 )
self.assertTrue( isType(j2.outputfiles[0], LocalFile) )
self.assertEqual( j2.outputfiles[0].namePattern, "*.txt" )
self.assertEqual( len(j2.postprocessors), 1 )
self.assertTrue( isType(j2.postprocessors[0], FileChecker) )
self.assertEqual( j2.postprocessors[0].files, ["stdout"] )
self.assertEqual( j2.postprocessors[0].searchStrings, ["my search"] )
self.assertTrue( isType(j2.splitter, GenericSplitter) )
self.assertEqual( j2.splitter.attribute, "application.args" )
self.assertEqual( j2.splitter.values, ['arg 1', 'arg 2', 'arg 3'])
def testInterfaceLookFeel(self):
# Just testing that the job construction works
from Ganga.GPI import Job, Im3ShapeApp, Im3ShapeSplitter, DiracFile, LocalFile, GangaDataset, Dirac
j=Job()
app = Im3ShapeApp(
im3_location=DiracFile(lfn='/lsst/y1a1-v2-z/software/2016-02-24/im3shape-grid.tar.gz'),
ini_location=LocalFile('/afs/cern.ch/user/r/rcurrie/cmtuser/GANGA/GANGA_LSST/install/ganga/python/params_disc.ini'),
blacklist=LocalFile('/afs/cern.ch/user/r/rcurrie/cmtuser/GANGA/GANGA_LSST/install/ganga/python/blacklist-y1.txt')
)
j.application = app
j.backend=Dirac()
mydata=GangaDataset()
mydata.append(DiracFile(lfn='/lsst/DES0005+0043-z-meds-y1a1-gamma.fits.fz'))
j.inputdata = mydata
j.splitter=Im3ShapeSplitter(size=20)
j.outputfiles = [DiracFile('*.main.txt'), DiracFile('*.epoch.txt')]
def testDatasetsFunctions(self):
from Ganga.GPI import DiracFile, PhysicalFile, LHCbDataset, Job, LocalFile
# test constructors/setters
ds = LHCbDataset(['lfn:a', 'pfn:b'])
assert len(ds) == 2
print(ds[0])
assert isinstance(ds[0], DiracFile)
assert isinstance(ds[1], LocalFile)
ds = LHCbDataset()
ds.files = ['lfn:a', 'pfn:b']
assert isinstance(ds[0], DiracFile)
assert isinstance(ds[1], LocalFile)
assert ds.getFullFileNames() == ['LFN:a', 'PFN:b']
ds.files.append('lfn:c')
assert isinstance(ds[-1], DiracFile)
# check job assignments
j = Job()
j.inputdata = ['lfn:a', 'pfn:b']
assert isinstance(j.inputdata, LHCbDataset)
j.outputfiles = ['a', DiracFile('b')]
assert isinstance(j.outputfiles[0], LocalFile)
print(type(j.outputfiles[1]))
assert isinstance(j.outputfiles[1], DiracFile)
# check the LHCbDataset functions:
assert ds.getLFNs() == ['a', 'c']
assert ds.getPFNs() == ['b']
ds2 = LHCbDataset(['lfn:a', 'lfn:d'])
ds.extend(ds2, True)
assert len(ds) == 4
# check the useful difference functions etc
assert sorted(ds.difference(ds2).getFileNames()) == ['b', 'c']
assert sorted(ds.symmetricDifference(ds2).getFileNames()) == ['b', 'c']
assert sorted(ds.intersection(ds2).getFileNames()) == ['a', 'd']
assert sorted(ds.union(ds2).getFileNames()) == ['a', 'b', 'c', 'd']
def test_g_Splitters(self):
from Ganga.GPI import Job, GenericSplitter, GangaDataset, GangaDatasetSplitter, LocalFile
# -- SPLITTERS BASICUSE START
j = Job()
j.splitter = GenericSplitter()
j.splitter.attribute = 'application.args'
j.splitter.values = [['hello', 1], ['world', 2], ['again', 3]]
j.submit()
# -- SPLITTERS BASICUSE STOP
# -- SPLITTERS SUBJOBCHECK START
j.subjobs
j.subjobs(0).peek("stdout")
# -- SPLITTERS SUBJOBCHECK STOP
# -- SPLITTERS MULTIATTRS START
j = Job()
j.splitter = GenericSplitter()
j.splitter.multi_attrs = {
'application.args': ['hello1', 'hello2'],
'application.env': [{
'MYENV': 'test1'
}, {
'MYENV': 'test2'
}]
}
j.submit()
# -- SPLITTERS MULTIATTRS STOP
# -- SPLITTERS DATASETSPLITTER START
j = Job()
j.application.exe = 'more'
j.application.args = ['__GangaInputData.txt__']
j.inputdata = GangaDataset(files=[LocalFile('*.txt')])
j.splitter = GangaDatasetSplitter()
j.splitter.files_per_subjob = 2
j.submit()
def testDatasets(self):
from Ganga.GPI import DiracFile, PhysicalFile, LHCbDataset, Job, LocalFile
# test constructors/setters
ds = LHCbDataset(['lfn:a', 'pfn:b'])
assert len(ds) == 2
print(ds[0])
assert isinstance(ds[0], DiracFile)
assert isinstance(ds[1], PhysicalFile)
ds = LHCbDataset()
ds.files = ['lfn:a', 'pfn:b']
assert isinstance(ds[0], DiracFile)
assert isinstance(ds[1], PhysicalFile)
ds.files.append('lfn:c')
assert isinstance(ds[-1], DiracFile)
d = OutputData(['a', 'b'])
assert isinstance(d.files[0], str)
assert isinstance(d.files[1], str)
# check job assignments
j = Job()
j.inputdata = ['lfn:a', 'pfn:b']
assert isinstance(j.inputdata, LHCbDataset)
j.outputfiles = ['a', DiracFile('b')]
assert isinstance(j.outputfiles[0], LocalFile)
print(type(j.outputfiles[1]))
assert isinstance(j.outputfiles[1], DiracFile)
LFN_DATA = [
'LFN:/lhcb/LHCb/Collision11/DIMUON.DST/00016768/0000/00016768_00000005_1.dimuon.dst',
'LFN:/lhcb/LHCb/Collision11/DIMUON.DST/00016768/0000/00016768_00000006_1.dimuon.dst'
]
ds = LHCbDataset(LFN_DATA)
assert len(ds.getReplicas().keys()) == 2
assert ds.getCatalog()
def makeIPResolutionsJob( jobName, dataFile, brunelVersion="", dataType = '2012', extraFiles = [],
ddDBTag = None, condDBTag = None ) :
"""Call this method to make a job that will run Brunel with the IP resolutions
ntupling algorithm, using the given config file (data type, etc) and data file.
Add this method to your ~/.ganga.py file to have it automatically loaded whenever
you run ganga."""
print "Creating an IP resolutions monitoring job named \'%s\'" % jobName
dataFile = os.path.expandvars(dataFile)
if not os.path.exists(dataFile) :
print "Could not find the data file \"%s\"!" % dataFile
return None
dataFile = os.path.abspath(dataFile)
print "Using data file \'%s\'" % dataFile
print "Parsing data file for options."
stepInfos = parseDataFileForSteps(dataFile)
productionOptsFiles = None
if len(stepInfos) > 0 :
for step in stepInfos :
# Get any additional options files used for the Brunel step.
if step['ApplicationName'] == "Brunel" :
productionOptsFiles = []
for optsFile in step['OptionFiles'].split(";") :
productionOptsFiles.append(optsFile.replace(" \n",""))
# Get the DB tags used for the Gauss step. This should be the
# same as for the Brunel step but it seems there's a bug in bkk.
if step['ApplicationName'] == "Gauss" or step['ApplicationName'] == 'DaVinci' or step['ApplicationName'] == 'Brunel' :
if ddDBTag == None :
ddDBTag = step['DDDB']
if condDBTag == None :
condDBTag = step['CONDDB']
if None == ddDBTag and len(extraFiles) == 0 :
print "The DB tags could not be retrieved from the data file and no extra options\
files have been specified! The job cannot be made."
return None
j = Job( name = jobName )
if brunelVersion == "" :
j.application = Brunel()
else :
j.application = Brunel(version=brunelVersion)
j.application.extraopts = 'Brunel().OutputLevel = 5\nBrunel().PrintFreq = 10000\nBrunel().DataType = "{0}"\n'.format(dataType)
if None != ddDBTag :
print "Using DDDBTag \"%s\" and CondDBTag \"%s\"" % (ddDBTag, condDBTag)
j.application.extraopts += "\nBrunel().DDDBtag = \"%s\"\nBrunel().CondDBtag = \"%s\"\n" % (ddDBTag, condDBTag)
if 'sim' in ddDBTag.lower() :
j.application.extraopts += '\nBrunel().Simulation = True\n'
#j.application.extraopts += '\nBrunel().WithMC = True\n'
else :
print "The DB tags could not be retrieved from the data file."
print "If they are not defined in one of the extra options files default values will be used."
j.application.optsfile = [ mainIPConfigFile ] + extraFiles
print "Using options files:"
for optsFile in j.application.optsfile :
print optsFile.name
if None != productionOptsFiles and len(productionOptsFiles) > 0 :
for optsFile in productionOptsFiles :
print optsFile
j.application.extraopts += "\nimportOptions(\"%s\")\n" % optsFile
print "Reading in data ..."
j.inputdata = j.application.readInputData( dataFile )
print "Data read. %s files found." % len(j.inputdata.files)
if len(j.inputdata.files) > 0 :
j.application.extraopts += '\nBrunel().InputType = "{0}"\n'.format(j.inputdata.files[0].name.split('.')[-1].upper())
j.splitter = SplitByFiles( filesPerJob = 10 )
j.backend = Dirac()
j.outputfiles = [DiracFile('*.root')]
return j
def test_f_InputAndOutputData(self):
from Ganga.GPI import Job, File, LocalFile, GangaDataset, Local, plugins
# -- INPUTANDOUTPUTDATA BASIC START
# create a script to send
open('my_script2.sh', 'w').write("""#!/bin/bash
ls -ltr
more "my_input.txt"
echo "TESTING" > my_output.txt
""")
import os
os.system('chmod +x my_script2.sh')
# create a script to send
open('my_input.txt', 'w').write('Input Testing works!')
j = Job()
j.application.exe = File('my_script2.sh')
j.inputfiles = [ LocalFile('my_input.txt') ]
j.outputfiles = [ LocalFile('my_output.txt') ]
j.submit()
# -- INPUTANDOUTPUTDATA BASIC STOP
# -- INPUTANDOUTPUTDATA PEEKOUTPUT START
j.peek() # list output dir contents
j.peek('my_output.txt')
# -- INPUTANDOUTPUTDATA PEEKOUTPUT STOP
# -- INPUTANDOUTPUTDATA FAILJOB START
# This job will fail
j = Job()
j.application.exe = File('my_script2.sh')
j.inputfiles = [ LocalFile('my_input.txt') ]
j.outputfiles = [ LocalFile('my_output_FAIL.txt') ]
j.submit()
# -- INPUTANDOUTPUTDATA FAILJOB STOP
# -- INPUTANDOUTPUTDATA WILDCARD START
# This job will pick up both 'my_input.txt' and 'my_output.txt'
j = Job()
j.application.exe = File('my_script2.sh')
j.inputfiles = [LocalFile('my_input.txt')]
j.outputfiles = [LocalFile('*.txt')]
j.submit()
# -- INPUTANDOUTPUTDATA WILDCARD STOP
# -- INPUTANDOUTPUTDATA OUTPUTFILES START
j.outputfiles
# -- INPUTANDOUTPUTDATA OUTPUTFILES STOP
# -- INPUTANDOUTPUTDATA INPUTDATA START
# Create a test script
open('my_script3.sh', 'w').write("""#!/bin/bash
echo $PATH
ls -ltr
more __GangaInputData.txt__
echo "MY TEST FILE" > output_file.txt
""")
import os
os.system('chmod +x my_script3.sh')
# Submit a job
j = Job()
j.application.exe = File('my_script3.sh')
j.inputdata = GangaDataset(files=[LocalFile('*.sh')])
j.backend = Local()
j.submit()
# -- INPUTANDOUTPUTDATA INPUTDATA STOP
# -- INPUTANDOUTPUTDATA GANGAFILES START
plugins('gangafiles')
def test_f_InputAndOutputData(self):
from Ganga.GPI import Job, File, LocalFile, GangaDataset, Local, plugins
# -- INPUTANDOUTPUTDATA BASIC START
# create a script to send
open('my_script2.sh', 'w').write("""#!/bin/bash
ls -ltr
more "my_input.txt"
echo "TESTING" > my_output.txt
""")
import os
os.system('chmod +x my_script2.sh')
# create a script to send
open('my_input.txt', 'w').write('Input Testing works!')
j = Job()
j.application.exe = File('my_script2.sh')
j.inputfiles = [LocalFile('my_input.txt')]
j.outputfiles = [LocalFile('my_output.txt')]
j.submit()
# -- INPUTANDOUTPUTDATA BASIC STOP
# -- INPUTANDOUTPUTDATA PEEKOUTPUT START
j.peek() # list output dir contents
j.peek('my_output.txt')
# -- INPUTANDOUTPUTDATA PEEKOUTPUT STOP
# -- INPUTANDOUTPUTDATA FAILJOB START
# This job will fail
j = Job()
j.application.exe = File('my_script2.sh')
j.inputfiles = [LocalFile('my_input.txt')]
j.outputfiles = [LocalFile('my_output_FAIL.txt')]
j.submit()
# -- INPUTANDOUTPUTDATA FAILJOB STOP
# -- INPUTANDOUTPUTDATA WILDCARD START
# This job will pick up both 'my_input.txt' and 'my_output.txt'
j = Job()
j.application.exe = File('my_script2.sh')
j.inputfiles = [LocalFile('my_input.txt')]
j.outputfiles = [LocalFile('*.txt')]
j.submit()
# -- INPUTANDOUTPUTDATA WILDCARD STOP
# -- INPUTANDOUTPUTDATA OUTPUTFILES START
j.outputfiles
# -- INPUTANDOUTPUTDATA OUTPUTFILES STOP
# -- INPUTANDOUTPUTDATA INPUTDATA START
# Create a test script
open('my_script3.sh', 'w').write("""#!/bin/bash
echo $PATH
ls -ltr
more __GangaInputData.txt__
echo "MY TEST FILE" > output_file.txt
""")
import os
os.system('chmod +x my_script3.sh')
# Submit a job
j = Job()
j.application.exe = File('my_script3.sh')
j.inputdata = GangaDataset(files=[LocalFile('*.sh')])
j.backend = Local()
j.submit()
# -- INPUTANDOUTPUTDATA INPUTDATA STOP
# -- INPUTANDOUTPUTDATA GANGAFILES START
plugins('gangafiles')
print('Options files:' + ' '.join([s.format(path=base, year=year) for s in OPTIONS])) # NOQA
j = Job(name=JNAME.format(
polarity, year, mode
))
j.comment = (
'{1} {2} MC {0} ntuple creation for k3pi mixing measurement.'
.format(event_type, year, polarity)
)
j.application = DaVinci(version='v41r3')
j.application.optsfile = [s.format(path=base, year=year) for s in OPTIONS]
if args.test:
# If testing, run over a couple of files locally,
# saving the results to the sandbox
j.inputdata = dataset[0:1]
j.backend = Local()
# Prepend test string to job name
j.name = 'TEST_{0}'.format(j.name)
j.outputfiles = [LocalFile(tfn)]
else:
# If not testing, run over everything on the grid, splitting jobs
# into groups of 10 files, notifying me on job completion/subjob failure,
# and save the results on the grid storage
j.inputdata = dataset
j.backend = Dirac()
j.backend.settings['CPUTime'] = 60*60*24*7
j.do_auto_resubmit = True
j.splitter = SplitByFiles(filesPerJob=5, ignoremissing=True)
j.postprocessors = [Notifier(address=email)]
j.outputfiles = [DiracFile(tfn)]
