class XbbRun:
def __init__(self, opts):
# get file list
self.filelist = FileList.decompress(opts.fileList) if len(opts.fileList) > 0 else None
print "len(filelist)",len(self.filelist),
if len(self.filelist) > 0:
print "filelist[0]:", self.filelist[0]
else:
print ''
# config
self.debug = 'XBBDEBUG' in os.environ
self.verifyCopy = True
self.opts = opts
self.config = BetterConfigParser()
self.config.read(opts.config)
self.channel = self.config.get('Configuration', 'channel')
# load namespace, TODO
VHbbNameSpace = self.config.get('VHbbNameSpace', 'library')
ROOT.gSystem.Load(VHbbNameSpace)
# directories
self.pathIN = self.config.get('Directories', opts.inputDir)
self.pathOUT = self.config.get('Directories', opts.outputDir)
self.tmpDir = self.config.get('Directories', 'scratch')
print 'INput samples:\t%s'%self.pathIN
print 'OUTput samples:\t%s'%self.pathOUT
self.fileLocator = FileLocator(config=self.config)
# check if given sample identifier uniquely matches a samples from config
matchingSamples = ParseInfo(samples_path=self.pathIN, config=self.config).find(identifier=opts.sampleIdentifier)
if len(matchingSamples) != 1:
print "ERROR: need exactly 1 sample identifier as input with -S !!"
print matchingSamples
exit(1)
self.sample = matchingSamples[0]
# collections
self.collections = [x.strip() for x in opts.addCollections.split(',') if len(x.strip()) > 0] if len(opts.addCollections.strip())>0 else []
if len(self.collections) < 1:
print "\x1b[31mWARNING: no collections added! Specify the collections to add with the --addCollections option!\x1b[0m"
print 'collections to add:', self.collections
self.collections = self.parseCollectionList(self.collections)
print 'after parsing:', self.collections
# temorary folder to save the files of this job on the scratch
temporaryName = self.sample.identifier + '/' + uuid.uuid4().hex
# input files
self.subJobs = []
if opts.join:
print("INFO: join input files! This is an experimental feature!")
# translate naming convention of .txt file to imported files after the prep step
inputFileNamesAfterPrep = [self.fileLocator.getFilenameAfterPrep(x) for x in self.filelist]
self.subJobs.append({
'inputFileNames': self.filelist,
'localInputFileNames': ["{path}/{subfolder}/{filename}".format(path=self.pathIN, subfolder=self.sample.identifier, filename=localFileName) for localFileName in inputFileNamesAfterPrep],
'outputFileName': "{path}/{subfolder}/{filename}".format(path=self.pathOUT, subfolder=self.sample.identifier, filename=inputFileNamesAfterPrep[0]),
'tmpFileName': "{path}/{subfolder}/{filename}".format(path=self.tmpDir, subfolder=temporaryName, filename=inputFileNamesAfterPrep[0]),
})
else:
# create separate subjob for all files (default!)
for inputFileName in self.filelist:
inputFileNamesAfterPrep = [self.fileLocator.getFilenameAfterPrep(inputFileName)]
self.subJobs.append({
'inputFileNames': [inputFileName],
'localInputFileNames': ["{path}/{subfolder}/{filename}".format(path=self.pathIN, subfolder=self.sample.identifier, filename=localFileName) for localFileName in inputFileNamesAfterPrep],
'outputFileName': "{path}/{subfolder}/{filename}".format(path=self.pathOUT, subfolder=self.sample.identifier, filename=inputFileNamesAfterPrep[0]),
'tmpFileName': "{path}/{subfolder}/{filename}".format(path=self.tmpDir, subfolder=temporaryName, filename=inputFileNamesAfterPrep[0]),
})
# lists of single modules can be given instead of a module, "--addCollections Sys.all"
# [Sys]
# all = ['Sys.Vtype', 'Sys.Leptons', ...]
# TODO: make it fully recursive
def parseCollectionList(self, collections):
collectionsListsReplaced = []
for collection in collections:
if '.' in collection:
section = collection.split('.')[0]
key = collection.split('.')[1]
listExpression = self.config.get(section, key).strip()
if listExpression.startswith('[') and listExpression.endswith(']'):
listParsed = eval(listExpression)
for i in listParsed:
collectionsListsReplaced.append(i)
else:
collectionsListsReplaced.append(collection)
else:
collectionsListsReplaced.append(collection)
return collectionsListsReplaced
# run all subjobs
def run(self):
nFilesProcessed = 0
nFilesFailed = 0
for subJob in self.subJobs:
# only process if output is non-existing/broken or --force was used
if self.opts.force or not self.fileLocator.isValidRootFile(subJob['outputFileName']):
# create directories
outputFolder = '/'.join(subJob['outputFileName'].split('/')[:-1])
tmpFolder = '/'.join(subJob['tmpFileName'].split('/')[:-1])
self.fileLocator.makedirs(outputFolder)
self.fileLocator.makedirs(tmpFolder)
# load sample tree
sampleTree = SampleTree(subJob['localInputFileNames'], config=self.config)
if not sampleTree.tree:
print "trying fallback...", len(subJob['inputFileNames'])
if len(subJob['inputFileNames']) == 1:
# try original naming scheme if reading directly from Heppy/Nano ntuples (without prep)
fileNameOriginal = self.pathIN + '/' + subJob['inputFileNames'][0]
print "FO:", fileNameOriginal
xrootdRedirector = self.fileLocator.getRedirector(fileNameOriginal)
sampleTree = SampleTree([fileNameOriginal], config=self.config, xrootdRedirector=xrootdRedirector)
if not sampleTree.tree:
print "\x1b[31mERROR: file does not exist or is broken, will be SKIPPED!\x1b[0m"
nFilesFailed += 1
continue
else:
print "\x1b[31mERROR: file does not exist or is broken, will be SKIPPED! (old naming scheme not supported for joining multipel files)\x1b[0m"
nFilesFailed += 1
continue
# to use this syntax, use "--addCollections Sys.Vtype" for a config file entry like this:
# [Sys]
# Vtype = VtypeCorrector.VtypeCorrector(channel='Zll')
# (instead of passing the tree in the constructor, the setTree method can be used)
pyModules = []
versionTable = []
for collection in self.collections:
if '.' in collection:
section = collection.split('.')[0]
key = collection.split('.')[1]
if self.config.has_section(section) and self.config.has_option(section, key):
pyCode = self.config.get(section, key)
elif '(' in collection and collection.endswith(')'):
print "WARNING: config option", collection, " not found, interpreting it as Python code!"
pyCode = collection
else:
print "\x1b[31mERROR: config option not found:", collection, ". To specify Python code directly, pass a complete constructor, e.g. --addCollections 'Module.Class()'. Module has to be placed in python/myutils/ folder.\x1b[0m"
raise Exception("ConfigError")
# import module from myutils
moduleName = pyCode.split('(')[0].split('.')[0].strip()
if self.debug:
print "DEBUG: import module:", moduleName
print("\x1b[33mDEBUG: " + collection + ": run PYTHON code:\n"+pyCode+"\x1b[0m")
globals()[moduleName] = importlib.import_module(".{module}".format(module=moduleName), package="myutils")
# get object
wObject = eval(pyCode)
# pass the tree and other variables if needed to finalize initialization
if hasattr(wObject, "customInit") and callable(getattr(wObject, "customInit")):
wObject.customInit({'config': self.config,
'sampleTree': sampleTree,
'tree': sampleTree.tree,
'sample': self.sample,
'channel': self.channel,
'pathIN': self.pathIN,
'pathOUT': self.pathOUT,
})
# add callbacks if the objects provides any
if hasattr(wObject, "processEvent") and callable(getattr(wObject, "processEvent")):
sampleTree.addCallback('event', wObject.processEvent)
for cb in ["finish", "prepareOutput"]:
if hasattr(wObject, cb) and callable(getattr(wObject, cb)):
sampleTree.addCallback(cb, getattr(wObject, cb))
# add branches
if hasattr(wObject, "getBranches") and callable(getattr(wObject, "getBranches")):
sampleTree.addOutputBranches(wObject.getBranches())
pyModules.append(wObject)
versionTable.append([moduleName, wObject.getVersion() if hasattr(wObject, "getVersion") else 0])
else:
print "\x1b[31mERROR: config option not found:", collection, " the format should be: [Section].[Option]\x1b[0m"
raise Exception("ConfigError")
for moduleName, moduleVersion in versionTable:
print " > {m}:{v}".format(m=moduleName, v=moduleVersion)
# DEPRECATED, do not use anymore ---> use BranchTools.TreeFormulas()
if 'addbranches' in self.collections:
writeNewVariables = eval(self.config.get("Regression", "writeNewVariablesDict"))
sampleTree.addOutputBranches(writeNewVariables)
# DEPRECATED, do not use anymore ---> use BranchTools.Drop()
if 'removebranches' in self.collections:
bl_branch = eval(config.get('Branches', 'useless_branch'))
for br in bl_branch:
sampleTree.addBranchToBlacklist(br)
bl_branch = eval(config.get('Branches', 'useless_after_sys'))
for br in bl_branch:
sampleTree.addBranchToBlacklist(br)
# define output file
sampleTree.addOutputTree(subJob['tmpFileName'], cut='1', branches='*', friend=self.opts.friend)
# run processing
for pyModule in pyModules:
if hasattr(pyModule, "beforeProcessing"):
getattr(pyModule, "beforeProcessing")()
sampleTree.process()
for pyModule in pyModules:
if hasattr(pyModule, "afterProcessing"):
getattr(pyModule, "afterProcessing")()
# if output trees have been produced: copy temporary file to output folder
if sampleTree.getNumberOfOutputTrees() > 0:
try:
self.fileLocator.cp(subJob['tmpFileName'], subJob['outputFileName'], force=True)
print 'copy ', subJob['tmpFileName'], subJob['outputFileName']
if self.verifyCopy:
if not self.fileLocator.isValidRootFile(subJob['outputFileName']):
print 'INFO: output at final destination broken, try to copy again from scratch disk to final destination...'
self.fileLocator.cp(subJob['tmpFileName'], subJob['outputFileName'], force=True)
print 'INFO: second attempt copy done!'
if not self.fileLocator.isValidRootFile(subJob['outputFileName']):
print '\x1b[31mERROR: output still broken!\x1b[0m'
nFilesFailed += 1
raise Exception("FileCopyError")
else:
print 'INFO: file is good after second attempt!'
except Exception as e:
print e
print "\x1b[31mERROR: copy from scratch to final destination failed!!\x1b[0m"
# delete temporary file
try:
self.fileLocator.rm(subJob['tmpFileName'])
except Exception as e:
print e
print "WARNING: could not delete file on scratch!"
# clean up
if hasattr(wObject, "cleanUp") and callable(getattr(wObject, "cleanUp")):
getattr(wObject, "cleanUp")()
else:
print 'SKIP:', subJob['inputFileNames']
if nFilesFailed > 0:
raise Exception("ProcessingIncomplete")
class PartialFileMerger(object):
def __init__(self,
fileNames,
chunkNumber,
submitTime='000000_000000',
force=False,
config=None,
sampleIdentifier=None):
self.fileNames = fileNames
self.debug = 'XBBDEBUG' in os.environ
self.submitTime = submitTime
self.chunkNumber = chunkNumber
self.config = config
self.fileLocator = FileLocator(config=self.config)
# -O option (reoptimizing baskets) leads to crashes...
self.commandTemplate = "hadd -k -ff {output} {inputs}"
self.sampleIdentifier = sampleIdentifier
self.force = force
# use sampleTree class as replacement for hadd
self.useChain = True
treeHashes = []
for fileName in self.fileNames:
treeHashes.append(hashlib.sha224(fileName).hexdigest())
totalHash = hashlib.sha224('-'.join(sorted(treeHashes))).hexdigest()
self.mergedFileName = '/'.join(
self.fileNames[0].split('/')[:-4]
) + '/' + totalHash + '/' + self.submitTime + '/0000/tree_%d.root' % chunkNumber
# return a fake name which is written to sample list .txt files in order to keep compatibility to the method of converting file names in .txt
# files to file names after prep step. This conversion applied to the fake name will give the real file name.
def getMergedFakeFileName(self):
return self.mergedFileName
# real output file name where the file is stored
def getOutputFileName(self):
fakeFileName = self.getMergedFakeFileName()
outputFileName = self.fileLocator.getFilenameAfterPrep(fakeFileName)
return "{path}/{sample}/{fileName}".format(
path=self.config.get('Directories', 'HADDout'),
sample=self.sampleIdentifier,
fileName=outputFileName)
def getTemporaryFileName(self):
fakeFileName = self.getMergedFakeFileName()
outputFileName = self.fileLocator.getFilenameAfterPrep(fakeFileName)
return "{path}/hadd/{sample}/{fileName}".format(
path=self.config.get('Directories', 'scratch'),
sample=self.sampleIdentifier,
fileName=outputFileName)
def run(self):
inputFileNames = [
"{path}/{sample}/{fileName}".format(
path=self.config.get('Directories', 'HADDin'),
sample=self.sampleIdentifier,
fileName=self.fileLocator.getFilenameAfterPrep(fileName))
for fileName in self.fileNames
]
outputFileName = self.getTemporaryFileName()
self.fileLocator.makedirs('/'.join(outputFileName.split('/')[:-1]))
command = self.commandTemplate.format(output=outputFileName,
inputs=' '.join(inputFileNames),
f="-f" if self.force else "")
if self.debug:
print("DEBUG: run \x1b[34m", command, "\x1b[0m")
if self.useChain:
# use sampleTree class (can e.g. drop branches at the same time)
sampleTree = SampleTree(inputFileNames, config=self.config)
try:
removeBranches = eval(
self.config.get('General', 'remove_branches'))
for removeBranch in removeBranches:
sampleTree.addBranchToBlacklist(removeBranch)
print("DEBUG: disable branch ", removeBranch)
except Exception as e:
print("DEBUG: could not disable branch:", e)
sampleTree.addOutputTree(outputFileName, cut='1', branches='*')
sampleTree.process()
result = 0
else:
# standard hadd
result = self.fileLocator.runCommand(command)
print("INFO: hadd returned ", result)
if result == 0:
finalOutputFileName = self.getOutputFileName()
print("move file to final destination: \x1b[34m",
finalOutputFileName, "\x1b[0m")
self.fileLocator.makedirs('/'.join(
finalOutputFileName.split('/')[:-1]))
resultCopy = self.fileLocator.cp(outputFileName,
finalOutputFileName, self.force)
if not resultCopy:
print("\x1b[31mERROR: copy failed\n from:", outputFileName,
"\n to:", finalOutputFileName, "\n force:", self.force,
"\x1b[0m")
raise Exception("FileCopyError")
# try to delete temporary file
try:
self.fileLocator.rm(outputFileName)
except Exception as e:
print("ERROR: could not delete temporary file:",
outputFileName, " => ", e)
print("INFO: done.")
else:
raise Exception("HaddError")
# TODO:
collections = [x.strip() for x in opts.addCollections.split(',') if len(x.strip()) > 0] if len(opts.addCollections.strip())>0 else []
if len(collections) < 1:
print "\x1b[31mWARNING: no collections added! Specify the collections to add with the --addCollections option!\x1b[0m"
print 'collections to add:', collections
for fileName in filelist:
localFileName = fileLocator.getFilenameAfterPrep(fileName)
inputFileName = "{path}/{subfolder}/{filename}".format(path=pathIN, subfolder=sample.identifier, filename=localFileName)
outputFileName = "{path}/{subfolder}/{filename}".format(path=pathOUT, subfolder=sample.identifier, filename=localFileName)
tmpFileName = "{path}/{subfolder}/{filename}".format(path=tmpDir, subfolder=sample.identifier, filename=localFileName)
outputFolder = '/'.join(outputFileName.split('/')[:-1])
tmpFolder = '/'.join(tmpFileName.split('/')[:-1])
fileLocator.makedirs(tmpFolder)
fileLocator.makedirs(outputFolder)
if opts.force or not fileLocator.isValidRootFile(outputFileName):
# load sample tree and initialize vtype corrector
sampleTree = SampleTree([inputFileName], config=config)
if not sampleTree.tree:
# try original naming scheme if reading directly from Heppy/Nano ntuples (without prep)
fileNameOriginal = pathIN + '/' + fileName
print "FO:", fileNameOriginal
xrootdRedirector = fileLocator.getRedirector(fileNameOriginal)
sampleTree = SampleTree([fileNameOriginal], config=config, xrootdRedirector=xrootdRedirector)
if not sampleTree.tree:
print "\x1b[31mERROR: file does not exist or is broken, will be SKIPPED!\x1b[0m"
continue
# TODO:
collections = [x.strip() for x in opts.addCollections.split(',') if len(x.strip()) > 0] if len(opts.addCollections.strip())>0 else []
if len(collections) < 1:
print "\x1b[31mWARNING: no collections added! Specify the collections to add with the --addCollections option!\x1b[0m"
print 'collections to add:', collections
for fileName in filelist:
localFileName = fileLocator.getFilenameAfterPrep(fileName)
inputFileName = "{path}/{subfolder}/{filename}".format(path=pathIN, subfolder=sample.identifier, filename=localFileName)
outputFileName = "{path}/{subfolder}/{filename}".format(path=pathOUT, subfolder=sample.identifier, filename=localFileName)
tmpFileName = "{path}/{subfolder}/{filename}".format(path=tmpDir, subfolder=sample.identifier, filename=localFileName)
outputFolder = '/'.join(outputFileName.split('/')[:-1])
tmpFolder = '/'.join(tmpFileName.split('/')[:-1])
fileLocator.makedirs(tmpFolder)
fileLocator.makedirs(outputFolder)
if not fileLocator.exists(outputFileName) or opts.force:
# load sample tree and initialize vtype corrector
sampleTree = SampleTree([inputFileName], config=config)
if not sampleTree.tree:
print "\x1b[31mERROR: file does not exist or is broken, will be SKIPPED!\x1b[0m"
continue
# lists of single modules can be given instead of a module, "--addCollections Sys.all"
# [Sys]
# all = ['Sys.Vtype', 'Sys.Leptons', ...]
collectionsListsReplaced = []
for collection in collections:
if '.' in collection:
class PartialFileMerger(object):
def __init__(self, fileNames, chunkNumber, submitTime='000000_000000', force=False, config=None, sampleIdentifier=None):
self.fileNames = fileNames
self.debug = 'XBBDEBUG' in os.environ
self.submitTime = submitTime
self.chunkNumber = chunkNumber
self.config = config
self.fileLocator = FileLocator(config=self.config)
# -O option (reoptimizing baskets) leads to crashes...
self.commandTemplate = "hadd -k -ff {output} {inputs}"
self.sampleIdentifier = sampleIdentifier
self.force = force
# use sampleTree class as replacement for hadd
self.useChain = True
treeHashes = []
for fileName in self.fileNames:
treeHashes.append(hashlib.sha224(fileName).hexdigest())
totalHash = hashlib.sha224('-'.join(sorted(treeHashes))).hexdigest()
self.mergedFileName = '/'.join(self.fileNames[0].split('/')[:-4]) + '/' + totalHash + '/' + self.submitTime + '/0000/tree_%d.root'%chunkNumber
# return a fake name which is written to sample list .txt files in order to keep compatibility to the method of converting file names in .txt
# files to file names after prep step. This conversion applied to the fake name will give the real file name.
def getMergedFakeFileName(self):
return self.mergedFileName
# real output file name where the file is stored
def getOutputFileName(self):
fakeFileName = self.getMergedFakeFileName()
outputFileName = self.fileLocator.getFilenameAfterPrep(fakeFileName)
return "{path}/{sample}/{fileName}".format(path=self.config.get('Directories','HADDout'), sample=self.sampleIdentifier, fileName=outputFileName)
def getTemporaryFileName(self):
fakeFileName = self.getMergedFakeFileName()
outputFileName = self.fileLocator.getFilenameAfterPrep(fakeFileName)
return "{path}/hadd/{sample}/{fileName}".format(path=self.config.get('Directories','scratch'), sample=self.sampleIdentifier, fileName=outputFileName)
def run(self):
inputFileNames = ["{path}/{sample}/{fileName}".format(path=self.config.get('Directories','HADDin'), sample=self.sampleIdentifier, fileName=self.fileLocator.getFilenameAfterPrep(fileName)) for fileName in self.fileNames]
outputFileName = self.getTemporaryFileName()
self.fileLocator.makedirs('/'.join(outputFileName.split('/')[:-1]))
command = self.commandTemplate.format(output=outputFileName, inputs=' '.join(inputFileNames), f="-f" if self.force else "")
if self.debug:
print ("DEBUG: run \x1b[34m", command, "\x1b[0m")
if self.useChain:
# use sampleTree class (can e.g. drop branches at the same time)
sampleTree = SampleTree(inputFileNames, config=self.config)
try:
removeBranches = eval(self.config.get('General', 'remove_branches'))
for removeBranch in removeBranches:
sampleTree.addBranchToBlacklist(removeBranch)
print("DEBUG: disable branch ", removeBranch)
except Exception as e:
print("DEBUG: could not disable branch:", e)
sampleTree.addOutputTree(outputFileName, cut='1', branches='*')
sampleTree.process()
result = 0
else:
# standard hadd
result = self.fileLocator.runCommand(command)
print ("INFO: hadd returned ", result)
if result == 0:
finalOutputFileName = self.getOutputFileName()
print("move file to final destination: \x1b[34m", finalOutputFileName, "\x1b[0m")
self.fileLocator.makedirs('/'.join(finalOutputFileName.split('/')[:-1]))
resultCopy = self.fileLocator.cp(outputFileName, finalOutputFileName, self.force)
if not resultCopy:
print("\x1b[31mERROR: copy failed\n from:", outputFileName, "\n to:", finalOutputFileName, "\n force:", self.force, "\x1b[0m")
raise Exception("FileCopyError")
# try to delete temporary file
try:
self.fileLocator.rm(outputFileName)
except Exception as e:
print("ERROR: could not delete temporary file:", outputFileName, " => ", e)
print("INFO: done.")
else:
raise Exception("HaddError")
class XbbRun:
def __init__(self, opts):
# get file list
self.filelist = FileList.decompress(opts.fileList) if len(opts.fileList) > 0 else None
print "len(filelist)",len(self.filelist),
if len(self.filelist) > 0:
print "filelist[0]:", self.filelist[0]
else:
print ''
# config
self.debug = 'XBBDEBUG' in os.environ
self.verifyCopy = True
self.opts = opts
self.config = BetterConfigParser()
self.config.read(opts.config)
samplesinfo = self.config.get('Directories', 'samplesinfo')
self.channel = self.config.get('Configuration', 'channel')
# load namespace, TODO
VHbbNameSpace = self.config.get('VHbbNameSpace', 'library')
ROOT.gSystem.Load(VHbbNameSpace)
# directories
self.pathIN = self.config.get('Directories', opts.inputDir)
self.pathOUT = self.config.get('Directories', opts.outputDir)
self.tmpDir = self.config.get('Directories', 'scratch')
print 'INput samples:\t%s'%self.pathIN
print 'OUTput samples:\t%s'%self.pathOUT
self.fileLocator = FileLocator(config=self.config)
# check if given sample identifier uniquely matches a samples from config
matchingSamples = ParseInfo(samplesinfo, self.pathIN).find(identifier=opts.sampleIdentifier)
if len(matchingSamples) != 1:
print "ERROR: need exactly 1 sample identifier as input with -S !!"
print matchingSamples
exit(1)
self.sample = matchingSamples[0]
# collections
self.collections = [x.strip() for x in opts.addCollections.split(',') if len(x.strip()) > 0] if len(opts.addCollections.strip())>0 else []
if len(self.collections) < 1:
print "\x1b[31mWARNING: no collections added! Specify the collections to add with the --addCollections option!\x1b[0m"
print 'collections to add:', self.collections
self.collections = self.parseCollectionList(self.collections)
print 'after parsing:', self.collections
# temorary folder to save the files of this job on the scratch
temporaryName = self.sample.identifier + '/' + uuid.uuid4().hex
# input files
self.subJobs = []
if opts.join:
print("INFO: join input files! This is an experimental feature!")
# translate naming convention of .txt file to imported files after the prep step
inputFileNamesAfterPrep = [self.fileLocator.getFilenameAfterPrep(x) for x in self.filelist]
self.subJobs.append({
'inputFileNames': self.filelist,
'localInputFileNames': ["{path}/{subfolder}/{filename}".format(path=self.pathIN, subfolder=self.sample.identifier, filename=localFileName) for localFileName in inputFileNamesAfterPrep],
'outputFileName': "{path}/{subfolder}/{filename}".format(path=self.pathOUT, subfolder=self.sample.identifier, filename=inputFileNamesAfterPrep[0]),
'tmpFileName': "{path}/{subfolder}/{filename}".format(path=self.tmpDir, subfolder=temporaryName, filename=inputFileNamesAfterPrep[0]),
})
else:
# create separate subjob for all files (default!)
for inputFileName in self.filelist:
inputFileNamesAfterPrep = [self.fileLocator.getFilenameAfterPrep(inputFileName)]
self.subJobs.append({
'inputFileNames': [inputFileName],
'localInputFileNames': ["{path}/{subfolder}/{filename}".format(path=self.pathIN, subfolder=self.sample.identifier, filename=localFileName) for localFileName in inputFileNamesAfterPrep],
'outputFileName': "{path}/{subfolder}/{filename}".format(path=self.pathOUT, subfolder=self.sample.identifier, filename=inputFileNamesAfterPrep[0]),
'tmpFileName': "{path}/{subfolder}/{filename}".format(path=self.tmpDir, subfolder=temporaryName, filename=inputFileNamesAfterPrep[0]),
})
# lists of single modules can be given instead of a module, "--addCollections Sys.all"
# [Sys]
# all = ['Sys.Vtype', 'Sys.Leptons', ...]
# TODO: make it fully recursive
def parseCollectionList(self, collections):
collectionsListsReplaced = []
for collection in collections:
if '.' in collection:
section = collection.split('.')[0]
key = collection.split('.')[1]
listExpression = self.config.get(section, key).strip()
if listExpression.startswith('[') and listExpression.endswith(']'):
listParsed = eval(listExpression)
for i in listParsed:
collectionsListsReplaced.append(i)
else:
collectionsListsReplaced.append(collection)
else:
collectionsListsReplaced.append(collection)
return collectionsListsReplaced
# run all subjobs
def run(self):
nFilesProcessed = 0
nFilesFailed = 0
for subJob in self.subJobs:
# only process if output is non-existing/broken or --force was used
if self.opts.force or not self.fileLocator.isValidRootFile(subJob['outputFileName']):
# create directories
outputFolder = '/'.join(subJob['outputFileName'].split('/')[:-1])
tmpFolder = '/'.join(subJob['tmpFileName'].split('/')[:-1])
self.fileLocator.makedirs(outputFolder)
self.fileLocator.makedirs(tmpFolder)
# load sample tree
sampleTree = SampleTree(subJob['localInputFileNames'], config=self.config)
if not sampleTree.tree:
print "trying fallback...", len(subJob['inputFileNames'])
if len(subJob['inputFileNames']) == 1:
# try original naming scheme if reading directly from Heppy/Nano ntuples (without prep)
fileNameOriginal = self.pathIN + '/' + subJob['inputFileNames'][0]
print "FO:", fileNameOriginal
xrootdRedirector = self.fileLocator.getRedirector(fileNameOriginal)
sampleTree = SampleTree([fileNameOriginal], config=self.config, xrootdRedirector=xrootdRedirector)
if not sampleTree.tree:
print "\x1b[31mERROR: file does not exist or is broken, will be SKIPPED!\x1b[0m"
nFilesFailed += 1
continue
else:
print "\x1b[31mERROR: file does not exist or is broken, will be SKIPPED! (old naming scheme not supported for joining multipel files)\x1b[0m"
nFilesFailed += 1
continue
# to use this syntax, use "--addCollections Sys.Vtype" for a config file entry like this:
# [Sys]
# Vtype = VtypeCorrector.VtypeCorrector(channel='Zll')
# (instead of passing the tree in the constructor, the setTree method can be used)
pyModules = []
for collection in self.collections:
if '.' in collection:
section = collection.split('.')[0]
key = collection.split('.')[1]
pyCode = self.config.get(section, key)
# import module from myutils
moduleName = pyCode.split('(')[0].split('.')[0].strip()
if self.debug:
print "DEBUG: import module:", moduleName
print("\x1b[33mDEBUG: " + collection + ": run PYTHON code:\n"+pyCode+"\x1b[0m")
globals()[moduleName] = importlib.import_module(".{module}".format(module=moduleName), package="myutils")
# get object
wObject = eval(pyCode)
# pass the tree and other variables if needed to finalize initialization
if hasattr(wObject, "customInit") and callable(getattr(wObject, "customInit")):
wObject.customInit({'config': self.config,
'sampleTree': sampleTree,
'tree': sampleTree.tree,
'sample': self.sample,
'channel': self.channel,
'pathIN': self.pathIN,
'pathOUT': self.pathOUT,
})
# add callbacks if the objects provides any
if hasattr(wObject, "processEvent") and callable(getattr(wObject, "processEvent")):
sampleTree.addCallback('event', wObject.processEvent)
# add branches
if hasattr(wObject, "getBranches") and callable(getattr(wObject, "getBranches")):
sampleTree.addOutputBranches(wObject.getBranches())
pyModules.append(wObject)
# DEPRECATED, do not use anymore ---> use BranchTools.TreeFormulas()
if 'addbranches' in self.collections:
writeNewVariables = eval(self.config.get("Regression", "writeNewVariablesDict"))
sampleTree.addOutputBranches(writeNewVariables)
# DEPRECATED, do not use anymore ---> use BranchTools.Drop()
if 'removebranches' in self.collections:
bl_branch = eval(config.get('Branches', 'useless_branch'))
for br in bl_branch:
sampleTree.addBranchToBlacklist(br)
bl_branch = eval(config.get('Branches', 'useless_after_sys'))
for br in bl_branch:
sampleTree.addBranchToBlacklist(br)
# define output file
sampleTree.addOutputTree(subJob['tmpFileName'], cut='1', branches='*', friend=self.opts.friend)
# run processing
for pyModule in pyModules:
if hasattr(pyModule, "beforeProcessing"):
getattr(pyModule, "beforeProcessing")()
sampleTree.process()
for pyModule in pyModules:
if hasattr(pyModule, "afterProcessing"):
getattr(pyModule, "afterProcessing")()
# if output trees have been produced: copy temporary file to output folder
if sampleTree.getNumberOfOutputTrees() > 0:
try:
self.fileLocator.cp(subJob['tmpFileName'], subJob['outputFileName'], force=True)
print 'copy ', subJob['tmpFileName'], subJob['outputFileName']
if self.verifyCopy:
if not self.fileLocator.isValidRootFile(subJob['outputFileName']):
print 'INFO: output at final destination broken, try to copy again from scratch disk to final destination...'
self.fileLocator.cp(subJob['tmpFileName'], subJob['outputFileName'], force=True)
print 'INFO: second attempt copy done!'
if not self.fileLocator.isValidRootFile(subJob['outputFileName']):
print '\x1b[31mERROR: output still broken!\x1b[0m'
nFilesFailed += 1
raise Exception("FileCopyError")
else:
print 'INFO: file is good after second attempt!'
except Exception as e:
print e
print "\x1b[31mERROR: copy from scratch to final destination failed!!\x1b[0m"
# delete temporary file
try:
self.fileLocator.rm(subJob['tmpFileName'])
except Exception as e:
print e
print "WARNING: could not delete file on scratch!"
# clean up
if hasattr(wObject, "cleanUp") and callable(getattr(wObject, "cleanUp")):
getattr(wObject, "cleanUp")()
else:
print 'SKIP:', subJob['inputFileNames']
if nFilesFailed > 0:
raise Exception("ProcessingIncomplete")
