Python FileSpec.fsize方法代码示例

# 需要导入模块: from taskbuffer.FileSpec import FileSpec [as 别名]
# 或者: from taskbuffer.FileSpec.FileSpec import fsize [as 别名]
 def convertToJobFileSpec(self,datasetSpec,setType=None,useEventService=False):
     jobFileSpec = JobFileSpec()
     jobFileSpec.fileID     = self.fileID
     jobFileSpec.datasetID  = datasetSpec.datasetID
     jobFileSpec.jediTaskID = datasetSpec.jediTaskID
     jobFileSpec.lfn        = self.lfn
     jobFileSpec.GUID       = self.GUID
     if setType == None:
         jobFileSpec.type   = self.type
     else:
         jobFileSpec.type   = setType
     jobFileSpec.scope      = self.scope
     jobFileSpec.fsize      = self.fsize
     jobFileSpec.checksum   = self.checksum
     jobFileSpec.attemptNr  = self.attemptNr
     # dataset attribute
     if datasetSpec != None:
         # dataset
         if not datasetSpec.containerName in [None,'']:
             jobFileSpec.dataset = datasetSpec.containerName
         else:
             jobFileSpec.dataset = datasetSpec.datasetName
         if self.type in datasetSpec.getInputTypes() or setType in datasetSpec.getInputTypes():
             # prodDBlock
             jobFileSpec.prodDBlock = datasetSpec.datasetName
             # storage token    
             if not datasetSpec.storageToken in ['',None]:
                 jobFileSpec.dispatchDBlockToken = datasetSpec.storageToken 
         else:
             # destinationDBlock
             jobFileSpec.destinationDBlock = datasetSpec.datasetName
             # storage token    
             if not datasetSpec.storageToken in ['',None]:
                 jobFileSpec.destinationDBlockToken = datasetSpec.storageToken.split('/')[0] 
             # destination
             if not datasetSpec.destination in ['',None]:
                 jobFileSpec.destinationSE = datasetSpec.destination
             # set prodDBlockToken for Event Service
             if useEventService and datasetSpec.getObjectStore() != None:
                 jobFileSpec.prodDBlockToken = 'objectstore^{0}'.format(datasetSpec.getObjectStore())
             # allow no output
             if datasetSpec.isAllowedNoOutput():
                 jobFileSpec.allowNoOutput()
     # return
     return jobFileSpec

# 需要导入模块: from taskbuffer.FileSpec import FileSpec [as 别名]
# 或者: from taskbuffer.FileSpec.FileSpec import fsize [as 别名]

#.........这里部分代码省略.........
        if app.dbrelease == 'LATEST':
            try:
                latest_dbrelease = getLatestDBReleaseCaching()
            except:
                from pandatools import Client
                latest_dbrelease = Client.getLatestDBRelease()
            m = re.search('(.*):DBRelease-(.*)\.tar\.gz', latest_dbrelease)
            if m:
                self.dbrelease_dataset = m.group(1)
                self.dbrelease = m.group(2)
            else:
                raise ApplicationConfigurationError(None, "Error retrieving LATEST DBRelease. Try setting application.dbrelease manually.")
        else:
            self.dbrelease_dataset = app.dbrelease_dataset
            self.dbrelease = app.dbrelease
        jspec.jobParameters = app.job_parameters

        if self.dbrelease:
            if self.dbrelease == 'current':
                jspec.jobParameters += ' --DBRelease=current' 
            else:
                if jspec.transformation.endswith("_tf.py") or jspec.transformation.endswith("_tf"):
                    jspec.jobParameters += ' --DBRelease=DBRelease-%s.tar.gz' % (self.dbrelease,)
                else:
                    jspec.jobParameters += ' DBRelease=DBRelease-%s.tar.gz' % (self.dbrelease,)
                dbspec = FileSpec()
                dbspec.lfn = 'DBRelease-%s.tar.gz' % self.dbrelease
                dbspec.dataset = self.dbrelease_dataset
                dbspec.prodDBlock = jspec.prodDBlock
                dbspec.type = 'input'
                jspec.addFile(dbspec)

        if job.inputdata:
            m = re.search('(.*)\.(.*)\.(.*)\.(.*)\.(.*)\.(.*)',
                          job.inputdata.dataset[0])
            if not m:
                logger.error("Error retrieving run number from dataset name")
                #raise ApplicationConfigurationError(None, "Error retrieving run number from dataset name")
                runnumber = 105200
            else:
                runnumber = int(m.group(2))
            if jspec.transformation.endswith("_tf.py") or jspec.transformation.endswith("_tf"):
                jspec.jobParameters += ' --runNumber %d' % runnumber
            else:
                jspec.jobParameters += ' RunNumber=%d' % runnumber
        
        # Output files.
        randomized_lfns = []
        ilfn = 0
        for lfn, lfntype in zip(app.output_files,app.output_type):
            ofspec = FileSpec()
            if app.randomize_lfns:
                randomized_lfn = lfn + ('.%s.%d.%s' % (job.backend.site, int(time.time()), commands.getoutput('uuidgen 2> /dev/null')[:4] ) )
            else:
                randomized_lfn = lfn
            ofspec.lfn = randomized_lfn
            randomized_lfns.append(randomized_lfn)
            ofspec.destinationDBlock = jspec.destinationDBlock
            ofspec.destinationSE = jspec.destinationSE
            ofspec.dataset = jspec.destinationDBlock
            ofspec.type = 'output'
            jspec.addFile(ofspec)
            if jspec.transformation.endswith("_tf.py") or jspec.transformation.endswith("_tf"):
                jspec.jobParameters += ' --output%sFile %s' % (lfntype, randomized_lfns[ilfn])
            else:
                jspec.jobParameters += ' output%sFile=%s' % (lfntype, randomized_lfns[ilfn])
            ilfn=ilfn+1

        # Input files.
        if job.inputdata:
            for guid, lfn, size, checksum, scope in zip(job.inputdata.guids, job.inputdata.names, job.inputdata.sizes, job.inputdata.checksums, job.inputdata.scopes):
                ifspec = FileSpec()
                ifspec.lfn = lfn
                ifspec.GUID = guid
                ifspec.fsize = size
                ifspec.md5sum = checksum
                ifspec.scope = scope
                ifspec.dataset = jspec.prodDBlock
                ifspec.prodDBlock = jspec.prodDBlock
                ifspec.type = 'input'
                jspec.addFile(ifspec)
            if app.input_type:
                itype = app.input_type
            else:
                itype = m.group(5)
            if jspec.transformation.endswith("_tf.py") or jspec.transformation.endswith("_tf"):
                jspec.jobParameters += ' --input%sFile %s' % (itype, ','.join(job.inputdata.names))
            else:
                jspec.jobParameters += ' input%sFile=%s' % (itype, ','.join(job.inputdata.names))

        # Log files.
        lfspec = FileSpec()
        lfspec.lfn = '%s.job.log.tgz' % jspec.jobName
        lfspec.destinationDBlock = jspec.destinationDBlock
        lfspec.destinationSE  = jspec.destinationSE
        lfspec.dataset = jspec.destinationDBlock
        lfspec.type = 'log'
        jspec.addFile(lfspec)
        
        return jspec

# 需要导入模块: from taskbuffer.FileSpec import FileSpec [as 别名]
# 或者: from taskbuffer.FileSpec.FileSpec import fsize [as 别名]
job.prodDBlock        = 'user.mlassnig:user.mlassnig.pilot.test.single.hits'
job.currentPriority   = 1000
#job.prodSourceLabel   = 'ptest'
job.prodSourceLabel   = 'user'
job.computingSite     = site
job.cloud             = cloud
job.cmtConfig         = 'x86_64-slc6-gcc48-opt'
job.specialHandling   = 'ddm:rucio'
#job.transferType      = 'direct'

ifile = 'HITS.06828093._000096.pool.root.1'
fileI = FileSpec()
fileI.GUID = 'AC5B3759-B606-BA42-8681-4BD86455AE02'
fileI.checksum = 'ad:5d000974'
fileI.dataset = 'user.mlassnig:user.mlassnig.pilot.test.single.hits'
fileI.fsize = 94834717
fileI.lfn = ifile
fileI.prodDBlock = job.prodDBlock
fileI.scope = 'mc15_13TeV'
fileI.type = 'input'
job.addFile(fileI)

ofile = 'RDO_%s.root' % commands.getoutput('uuidgen')
fileO = FileSpec()
fileO.dataset = job.destinationDBlock
fileO.destinationDBlock = job.destinationDBlock
fileO.destinationSE = job.destinationSE
fileO.lfn = ofile
fileO.type = 'output'
job.addFile(fileO)