[ViewVC] Diff of: cvsroot/COMP/CRAB/python/cms

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.128 by fanzago, Thu Oct 11 16:23:44 2007 UTC vs.
Revision 1.211 by slacapra, Tue Jun 10 14:02:12 2008 UTC

#	Line 5 \| Line 5 \| from crab_util import *
5		from BlackWhiteListParser import BlackWhiteListParser
6		import common
7		import Scram
8	+	from LFNBaseName import *
9
10		import os, string, glob
11
12		class Cmssw(JobType):
13	<	def __init__(self, cfg_params, ncjobs):
13	>	def __init__(self, cfg_params, ncjobs,skip_blocks, isNew):
14		JobType.__init__(self, 'CMSSW')
15		common.logger.debug(3,'CMSSW::__init__')
16	+	self.skip_blocks = skip_blocks
17	+
18	+	self.argsList = []
19
20		self._params = {}
21		self.cfg_params = cfg_params
18	–
22		# init BlackWhiteListParser
23		self.blackWhiteListParser = BlackWhiteListParser(cfg_params)
24
25	<	try:
23	<	self.MaxTarBallSize = float(self.cfg_params['EDG.maxtarballsize'])
24	<	except KeyError:
25	<	self.MaxTarBallSize = 9.5
25	>	self.MaxTarBallSize = float(self.cfg_params.get('EDG.maxtarballsize',9.5))
26
27		# number of jobs requested to be created, limit obj splitting
28		self.ncjobs = ncjobs
29
30		log = common.logger
31	<
31	>
32		self.scram = Scram.Scram(cfg_params)
33		self.additional_inbox_files = []
34		self.scriptExe = ''
35		self.executable = ''
36		self.executable_arch = self.scram.getArch()
37		self.tgz_name = 'default.tgz'
38	–	self.additional_tgz_name = 'additional.tgz'
38		self.scriptName = 'CMSSW.sh'
39	<	self.pset = '' #scrip use case Da
40	<	self.datasetPath = '' #scrip use case Da
39	>	self.pset = ''
40	>	self.datasetPath = ''
41
42		# set FJR file name
43		self.fjrFileName = 'crab_fjr.xml'
44
45		self.version = self.scram.getSWVersion()
46	<
47	<	#
48	<	# Try to block creation in case of arch/version mismatch
49	<	#
50	<
51	<	a = string.split(self.version, "_")
52	<
53	<	if int(a[1]) == 1 and (int(a[2]) < 5 and self.executable_arch.find('slc4') == 0):
54	<	msg = "Error: CMS does not support %s with %s architecture"%(self.version, self.executable_arch)
55	<	raise CrabException(msg)
57	<	if int(a[1]) == 1 and (int(a[2]) >= 5 and self.executable_arch.find('slc3') == 0):
58	<	msg = "Error: CMS does not support %s with %s architecture"%(self.version, self.executable_arch)
46	>	version_array = self.version.split('_')
47	>	self.CMSSW_major = 0
48	>	self.CMSSW_minor = 0
49	>	self.CMSSW_patch = 0
50	>	try:
51	>	self.CMSSW_major = int(version_array[1])
52	>	self.CMSSW_minor = int(version_array[2])
53	>	self.CMSSW_patch = int(version_array[3])
54	>	except:
55	>	msg = "Cannot parse CMSSW version string: " + self.version + " for major and minor release number!"
56		raise CrabException(msg)
60	–
61	–	common.taskDB.setDict('codeVersion',self.version)
62	–	self.setParam_('application', self.version)
57
58		### collect Data cards
59
60	<	## get DBS mode
61	<	try:
68	<	self.use_dbs_1 = int(self.cfg_params['CMSSW.use_dbs_1'])
69	<	except KeyError:
70	<	self.use_dbs_1 = 0
71	<
72	<	try:
73	<	tmp = cfg_params['CMSSW.datasetpath']
74	<	log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
75	<	if string.lower(tmp)=='none':
76	<	self.datasetPath = None
77	<	self.selectNoInput = 1
78	<	else:
79	<	self.datasetPath = tmp
80	<	self.selectNoInput = 0
81	<	except KeyError:
82	<	msg = "Error: datasetpath not defined "
60	>	if not cfg_params.has_key('CMSSW.datasetpath'):
61	>	msg = "Error: datasetpath not defined "
62		raise CrabException(msg)
63	<
64	<	# ML monitoring
65	<	# split dataset path style: /PreProdR3Minbias/SIM/GEN-SIM
66	<	if not self.datasetPath:
67	<	self.setParam_('dataset', 'None')
89	<	self.setParam_('owner', 'None')
63	>	tmp = cfg_params['CMSSW.datasetpath']
64	>	log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
65	>	if string.lower(tmp)=='none':
66	>	self.datasetPath = None
67	>	self.selectNoInput = 1
68		else:
69	<	try:
70	<	datasetpath_split = self.datasetPath.split("/")
93	<	# standard style
94	<	self.setParam_('datasetFull', self.datasetPath)
95	<	if self.use_dbs_1 == 1 :
96	<	self.setParam_('dataset', datasetpath_split[1])
97	<	self.setParam_('owner', datasetpath_split[-1])
98	<	else:
99	<	self.setParam_('dataset', datasetpath_split[1])
100	<	self.setParam_('owner', datasetpath_split[2])
101	<	except:
102	<	self.setParam_('dataset', self.datasetPath)
103	<	self.setParam_('owner', self.datasetPath)
104	<
105	<	self.setTaskid_()
106	<	self.setParam_('taskId', self.cfg_params['taskId'])
69	>	self.datasetPath = tmp
70	>	self.selectNoInput = 0
71
72		self.dataTiers = []
73	<
73	>	self.debugWrap = ''
74	>	self.debug_wrapper = cfg_params.get('USER.debug_wrapper',False)
75	>	if self.debug_wrapper: self.debugWrap='--debug'
76		## now the application
77	<	try:
78	<	self.executable = cfg_params['CMSSW.executable']
113	<	self.setParam_('exe', self.executable)
114	<	log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
115	<	msg = "Default executable cmsRun overridden. Switch to " + self.executable
116	<	log.debug(3,msg)
117	<	except KeyError:
118	<	self.executable = 'cmsRun'
119	<	self.setParam_('exe', self.executable)
120	<	msg = "User executable not defined. Use cmsRun"
121	<	log.debug(3,msg)
122	<	pass
77	>	self.executable = cfg_params.get('CMSSW.executable','cmsRun')
78	>	log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
79
80	<	try:
125	<	self.pset = cfg_params['CMSSW.pset']
126	<	log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
127	<	if self.pset.lower() != 'none' :
128	<	if (not os.path.exists(self.pset)):
129	<	raise CrabException("User defined PSet file "+self.pset+" does not exist")
130	<	else:
131	<	self.pset = None
132	<	except KeyError:
80	>	if not cfg_params.has_key('CMSSW.pset'):
81		raise CrabException("PSet file missing. Cannot run cmsRun ")
82	+	self.pset = cfg_params['CMSSW.pset']
83	+	log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
84	+	if self.pset.lower() != 'none' :
85	+	if (not os.path.exists(self.pset)):
86	+	raise CrabException("User defined PSet file "+self.pset+" does not exist")
87	+	else:
88	+	self.pset = None
89
90		# output files
91		## stuff which must be returned always via sandbox
#	Line 140 \| Line 95 \| class Cmssw(JobType):
95		self.output_file_sandbox.append(self.fjrFileName)
96
97		# other output files to be returned via sandbox or copied to SE
98	<	try:
99	<	self.output_file = []
100	<	tmp = cfg_params['CMSSW.output_file']
101	<	if tmp != '':
102	<	tmpOutFiles = string.split(cfg_params['CMSSW.output_file'],',')
148	<	log.debug(7, 'cmssw::cmssw(): output files '+str(tmpOutFiles))
149	<	for tmp in tmpOutFiles:
150	<	tmp=string.strip(tmp)
151	<	self.output_file.append(tmp)
152	<	pass
153	<	else:
154	<	log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
155	<	pass
156	<	pass
157	<	except KeyError:
98	>	self.output_file = []
99	>	tmp = cfg_params.get('CMSSW.output_file',None)
100	>	if tmp :
101	>	self.output_file = [x.strip() for x in tmp.split(',')]
102	>	else:
103		log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
104	<	pass
104	>	pass
105
106		# script_exe file as additional file in inputSandbox
107	<	try:
108	<	self.scriptExe = cfg_params['USER.script_exe']
109	<	if self.scriptExe != '':
110	<	if not os.path.isfile(self.scriptExe):
111	<	msg ="ERROR. file "+self.scriptExe+" not found"
112	<	raise CrabException(msg)
168	<	self.additional_inbox_files.append(string.strip(self.scriptExe))
169	<	except KeyError:
170	<	self.scriptExe = ''
107	>	self.scriptExe = cfg_params.get('USER.script_exe',None)
108	>	if self.scriptExe :
109	>	if not os.path.isfile(self.scriptExe):
110	>	msg ="ERROR. file "+self.scriptExe+" not found"
111	>	raise CrabException(msg)
112	>	self.additional_inbox_files.append(string.strip(self.scriptExe))
113
172	–	#CarlosDaniele
114		if self.datasetPath == None and self.pset == None and self.scriptExe == '' :
115	<	msg ="Error. script_exe not defined"
116	<	raise CrabException(msg)
115	>	msg ="Error. script_exe not defined"
116	>	raise CrabException(msg)
117	>
118	>	# use parent files...
119	>	self.useParent = self.cfg_params.get('CMSSW.use_parent',False)
120
121		## additional input files
122	<	try:
122	>	if cfg_params.has_key('USER.additional_input_files'):
123		tmpAddFiles = string.split(cfg_params['USER.additional_input_files'],',')
124		for tmp in tmpAddFiles:
125		tmp = string.strip(tmp)
#	Line 192 \| Line 136 \| class Cmssw(JobType):
136		if not os.path.exists(file):
137		raise CrabException("Additional input file not found: "+file)
138		pass
195	–	# fname = string.split(file, '/')[-1]
196	–	# storedFile = common.work_space.pathForTgz()+'share/'+fname
197	–	# shutil.copyfile(file, storedFile)
139		self.additional_inbox_files.append(string.strip(file))
140		pass
141		pass
142		common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
143	<	except KeyError:
203	<	pass
204	<
205	<	# files per job
206	<	try:
207	<	if (cfg_params['CMSSW.files_per_jobs']):
208	<	raise CrabException("files_per_jobs no longer supported. Quitting.")
209	<	except KeyError:
210	<	pass
143	>	pass
144
145		## Events per job
146	<	try:
146	>	if cfg_params.has_key('CMSSW.events_per_job'):
147		self.eventsPerJob =int( cfg_params['CMSSW.events_per_job'])
148		self.selectEventsPerJob = 1
149	<	except KeyError:
149	>	else:
150		self.eventsPerJob = -1
151		self.selectEventsPerJob = 0
152	<
152	>
153		## number of jobs
154	<	try:
154	>	if cfg_params.has_key('CMSSW.number_of_jobs'):
155		self.theNumberOfJobs =int( cfg_params['CMSSW.number_of_jobs'])
156		self.selectNumberOfJobs = 1
157	<	except KeyError:
157	>	else:
158		self.theNumberOfJobs = 0
159		self.selectNumberOfJobs = 0
160
161	<	try:
161	>	if cfg_params.has_key('CMSSW.total_number_of_events'):
162		self.total_number_of_events = int(cfg_params['CMSSW.total_number_of_events'])
163		self.selectTotalNumberEvents = 1
164	<	except KeyError:
164	>	if self.selectNumberOfJobs == 1:
165	>	if (self.total_number_of_events != -1) and int(self.total_number_of_events) < int(self.theNumberOfJobs):
166	>	msg = 'Must specify at least one event per job. total_number_of_events > number_of_jobs '
167	>	raise CrabException(msg)
168	>	else:
169		self.total_number_of_events = 0
170		self.selectTotalNumberEvents = 0
171
172	<	if self.pset != None: #CarlosDaniele
172	>	if self.pset != None:
173		if ( (self.selectTotalNumberEvents + self.selectEventsPerJob + self.selectNumberOfJobs) != 2 ):
174		msg = 'Must define exactly two of total_number_of_events, events_per_job, or number_of_jobs.'
175		raise CrabException(msg)
#	Line 241 \| Line 178 \| class Cmssw(JobType):
178		msg = 'Must specify number_of_jobs.'
179		raise CrabException(msg)
180
181	<	## source seed for pythia
182	<	try:
183	<	self.sourceSeed = int(cfg_params['CMSSW.pythia_seed'])
184	<	except KeyError:
185	<	self.sourceSeed = None
186	<	common.logger.debug(5,"No seed given")
187	<
188	<	try:
189	<	self.sourceSeedVtx = int(cfg_params['CMSSW.vtx_seed'])
190	<	except KeyError:
191	<	self.sourceSeedVtx = None
192	<	common.logger.debug(5,"No vertex seed given")
193	<
194	<	try:
195	<	self.sourceSeedG4 = int(cfg_params['CMSSW.g4_seed'])
196	<	except KeyError:
197	<	self.sourceSeedG4 = None
198	<	common.logger.debug(5,"No g4 sim hits seed given")
199	<
200	<	try:
201	<	self.sourceSeedMix = int(cfg_params['CMSSW.mix_seed'])
202	<	except KeyError:
203	<	self.sourceSeedMix = None
204	<	common.logger.debug(5,"No mix seed given")
205	<
206	<	try:
207	<	self.firstRun = int(cfg_params['CMSSW.first_run'])
208	<	except KeyError:
209	<	self.firstRun = None
210	<	common.logger.debug(5,"No first run given")
211	<	if self.pset != None: #CarlosDaniele
212	<	ver = string.split(self.version,"_")
213	<	if (int(ver[1])>=1 and int(ver[2])>=5):
214	<	import PsetManipulator150 as pp
215	<	else:
216	<	import PsetManipulator as pp
217	<	PsetEdit = pp.PsetManipulator(self.pset) #Daniele Pset
181	>	## New method of dealing with seeds
182	>	self.incrementSeeds = []
183	>	self.preserveSeeds = []
184	>	if cfg_params.has_key('CMSSW.preserve_seeds'):
185	>	tmpList = cfg_params['CMSSW.preserve_seeds'].split(',')
186	>	for tmp in tmpList:
187	>	tmp.strip()
188	>	self.preserveSeeds.append(tmp)
189	>	if cfg_params.has_key('CMSSW.increment_seeds'):
190	>	tmpList = cfg_params['CMSSW.increment_seeds'].split(',')
191	>	for tmp in tmpList:
192	>	tmp.strip()
193	>	self.incrementSeeds.append(tmp)
194	>
195	>	## Old method of dealing with seeds
196	>	## FUTURE: This is for old CMSSW and old CRAB. Can throw exceptions after a couple of CRAB releases and then
197	>	## remove
198	>	self.sourceSeed = cfg_params.get('CMSSW.pythia_seed',None)
199	>	if self.sourceSeed:
200	>	print "pythia_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
201	>	self.incrementSeeds.append('sourceSeed')
202	>	self.incrementSeeds.append('theSource')
203	>
204	>	self.sourceSeedVtx = cfg_params.get('CMSSW.vtx_seed',None)
205	>	if self.sourceSeedVtx:
206	>	print "vtx_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
207	>	self.incrementSeeds.append('VtxSmeared')
208	>
209	>	self.sourceSeedG4 = cfg_params.get('CMSSW.g4_seed',None)
210	>	if self.sourceSeedG4:
211	>	print "g4_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
212	>	self.incrementSeeds.append('g4SimHits')
213	>
214	>	self.sourceSeedMix = cfg_params.get('CMSSW.mix_seed',None)
215	>	if self.sourceSeedMix:
216	>	print "mix_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
217	>	self.incrementSeeds.append('mix')
218	>
219	>	self.firstRun = cfg_params.get('CMSSW.first_run',None)
220	>
221	>
222	>	# Copy/return
223	>	self.copy_data = int(cfg_params.get('USER.copy_data',0))
224	>	self.return_data = int(cfg_params.get('USER.return_data',0))
225
226		#DBSDLS-start
227	<	## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
227	>	## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
228		self.maxEvents=0 # max events available ( --> check the requested nb. of evts in Creator.py)
229		self.DBSPaths={} # all dbs paths requested ( --> input to the site local discovery script)
230		self.jobDestination=[] # Site destination(s) for each job (list of lists)
#	Line 289 \| Line 233 \| class Cmssw(JobType):
233		blockSites = {}
234		if self.datasetPath:
235		blockSites = self.DataDiscoveryAndLocation(cfg_params)
236	<	#DBSDLS-end
236	>	#DBSDLS-end
237
294	–	self.tgzNameWithPath = self.getTarBall(self.executable)
295	–
238		## Select Splitting
239	<	if self.selectNoInput:
240	<	if self.pset == None: #CarlosDaniele
239	>	if self.selectNoInput:
240	>	if self.pset == None:
241		self.jobSplittingForScript()
242		else:
243		self.jobSplittingNoInput()
244		else:
245		self.jobSplittingByBlocks(blockSites)
246
247	<	# modify Pset
248	<	if self.pset != None: #CarlosDaniele
249	<	try:
250	<	if (self.datasetPath): # standard job
251	<	# allow to processa a fraction of events in a file
252	<	PsetEdit.inputModule("INPUT")
253	<	PsetEdit.maxEvent("INPUTMAXEVENTS")
254	<	PsetEdit.skipEvent("INPUTSKIPEVENTS")
255	<	else: # pythia like job
247	>	# modify Pset only the first time
248	>	if isNew:
249	>	if self.pset != None:
250	>	import PsetManipulator as pp
251	>	PsetEdit = pp.PsetManipulator(self.pset)
252	>	try:
253	>	# Add FrameworkJobReport to parameter-set, set max events.
254	>	# Reset later for data jobs by writeCFG which does all modifications
255	>	PsetEdit.addCrabFJR(self.fjrFileName) # FUTURE: Job report addition not needed by CMSSW>1.5
256		PsetEdit.maxEvent(self.eventsPerJob)
257	<	if (self.firstRun):
258	<	PsetEdit.pythiaFirstRun("INPUTFIRSTRUN") #First Run
259	<	if (self.sourceSeed) :
260	<	PsetEdit.pythiaSeed("INPUT")
261	<	if (self.sourceSeedVtx) :
262	<	PsetEdit.vtxSeed("INPUTVTX")
321	<	if (self.sourceSeedG4) :
322	<	PsetEdit.g4Seed("INPUTG4")
323	<	if (self.sourceSeedMix) :
324	<	PsetEdit.mixSeed("INPUTMIX")
325	<	# add FrameworkJobReport to parameter-set
326	<	PsetEdit.addCrabFJR(self.fjrFileName)
327	<	PsetEdit.psetWriter(self.configFilename())
328	<	except:
329	<	msg='Error while manipuliating ParameterSet: exiting...'
330	<	raise CrabException(msg)
257	>	PsetEdit.psetWriter(self.configFilename())
258	>	except:
259	>	msg='Error while manipulating ParameterSet: exiting...'
260	>	raise CrabException(msg)
261	>	## Prepare inputSandbox TarBall (only the first time)
262	>	self.tgzNameWithPath = self.getTarBall(self.executable)
263
264		def DataDiscoveryAndLocation(self, cfg_params):
265
266		import DataDiscovery
335	–	import DataDiscovery_DBS2
267		import DataLocation
268		common.logger.debug(10,"CMSSW::DataDiscoveryAndLocation()")
269
#	Line 341 \| Line 272 \| class Cmssw(JobType):
272		## Contact the DBS
273		common.logger.message("Contacting Data Discovery Services ...")
274		try:
275	<
345	<	if self.use_dbs_1 == 1 :
346	<	self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params)
347	<	else :
348	<	self.pubdata=DataDiscovery_DBS2.DataDiscovery_DBS2(datasetPath, cfg_params)
275	>	self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params,self.skip_blocks)
276		self.pubdata.fetchDBSInfo()
277
278		except DataDiscovery.NotExistingDatasetError, ex :
#	Line 357 \| Line 284 \| class Cmssw(JobType):
284		except DataDiscovery.DataDiscoveryError, ex:
285		msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
286		raise CrabException(msg)
360	–	except DataDiscovery_DBS2.NotExistingDatasetError_DBS2, ex :
361	–	msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
362	–	raise CrabException(msg)
363	–	except DataDiscovery_DBS2.NoDataTierinProvenanceError_DBS2, ex :
364	–	msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
365	–	raise CrabException(msg)
366	–	except DataDiscovery_DBS2.DataDiscoveryError_DBS2, ex:
367	–	msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
368	–	raise CrabException(msg)
287
288		self.filesbyblock=self.pubdata.getFiles()
289		self.eventsbyblock=self.pubdata.getEventsPerBlock()
290		self.eventsbyfile=self.pubdata.getEventsPerFile()
291	+	self.parentFiles=self.pubdata.getParent()
292
293		## get max number of events
294	<	self.maxEvents=self.pubdata.getMaxEvents() ## self.maxEvents used in Creator.py
294	>	self.maxEvents=self.pubdata.getMaxEvents()
295
296		## Contact the DLS and build a list of sites hosting the fileblocks
297		try:
#	Line 381 \| Line 300 \| class Cmssw(JobType):
300		except DataLocation.DataLocationError , ex:
301		msg = 'ERROR ***: failed Data Location in DLS \n %s '%ex.getErrorMessage()
302		raise CrabException(msg)
303	<
303	>
304
305		sites = dataloc.getSites()
306		allSites = []
#	Line 395 \| Line 314 \| class Cmssw(JobType):
314		common.logger.message("Requested dataset: " + datasetPath + " has " + str(self.maxEvents) + " events in " + str(len(self.filesbyblock.keys())) + " blocks.\n")
315
316		return sites
317	<
317	>
318		def jobSplittingByBlocks(self, blockSites):
319		"""
320		Perform job splitting. Jobs run over an integer number of files
#	Line 445 \| Line 364 \| class Cmssw(JobType):
364		totalNumberOfJobs = 999999999
365		else :
366		totalNumberOfJobs = self.ncjobs
448	–
367
368		blocks = blockSites.keys()
369		blockCount = 0
#	Line 465 \| Line 383 \| class Cmssw(JobType):
383		blockCount += 1
384		if block not in jobsOfBlock.keys() :
385		jobsOfBlock[block] = []
386	<
386	>
387		if self.eventsbyblock.has_key(block) :
388		numEventsInBlock = self.eventsbyblock[block]
389		common.logger.debug(5,'Events in Block File '+str(numEventsInBlock))
390	<
390	>
391		files = self.filesbyblock[block]
392		numFilesInBlock = len(files)
393		if (numFilesInBlock <= 0):
#	Line 477 \| Line 395 \| class Cmssw(JobType):
395		fileCount = 0
396
397		# ---- New block => New job ---- #
398	<	parString = "\\{"
398	>	parString = ""
399		# counter for number of events in files currently worked on
400		filesEventCount = 0
401		# flag if next while loop should touch new file
402		newFile = 1
403		# job event counter
404		jobSkipEventCount = 0
405	<
405	>
406		# ---- Iterate over the files in the block until we've met the requested ---- #
407		# ---- total # of events or we've gone over all the files in this block ---- #
408	+	pString=''
409		while ( (eventsRemaining > 0) and (fileCount < numFilesInBlock) and (jobCount < totalNumberOfJobs) ):
410		file = files[fileCount]
411	+	if self.useParent:
412	+	parent = self.parentFiles[file]
413	+	for f in parent :
414	+	pString += '\\\"' + f + '\\\"\,'
415	+	common.logger.debug(6, "File "+str(file)+" has the following parents: "+str(parent))
416	+	common.logger.write("File "+str(file)+" has the following parents: "+str(parent))
417		if newFile :
418		try:
419		numEventsInFile = self.eventsbyfile[file]
#	Line 500 \| Line 425 \| class Cmssw(JobType):
425		newFile = 0
426		except KeyError:
427		common.logger.message("File "+str(file)+" has unknown number of events: skipping")
503	–
428
429	+	eventsPerJobRequested = min(eventsPerJobRequested, eventsRemaining)
430		# if less events in file remain than eventsPerJobRequested
431	<	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested ) :
431	>	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested):
432		# if last file in block
433		if ( fileCount == numFilesInBlock-1 ) :
434		# end job using last file, use remaining events in block
435		# close job and touch new file
436		fullString = parString[:-2]
437	<	fullString += '\\}'
438	<	list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
437	>	if self.useParent:
438	>	fullParentString = pString[:-2]
439	>	list_of_lists.append([fullString,fullParentString,str(-1),str(jobSkipEventCount)])
440	>	else:
441	>	list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
442		common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
443		self.jobDestination.append(blockSites[block])
444		common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
#	Line 522 \| Line 450 \| class Cmssw(JobType):
450		eventsRemaining = eventsRemaining - filesEventCount + jobSkipEventCount
451		jobSkipEventCount = 0
452		# reset file
453	<	parString = "\\{"
453	>	pString = ""
454	>	parString = ""
455		filesEventCount = 0
456		newFile = 1
457		fileCount += 1
#	Line 534 \| Line 463 \| class Cmssw(JobType):
463		elif ( filesEventCount - jobSkipEventCount == eventsPerJobRequested ) :
464		# close job and touch new file
465		fullString = parString[:-2]
466	<	fullString += '\\}'
467	<	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
466	>	if self.useParent:
467	>	fullParentString = pString[:-2]
468	>	list_of_lists.append([fullString,fullParentString,str(eventsPerJobRequested),str(jobSkipEventCount)])
469	>	else:
470	>	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
471		common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
472		self.jobDestination.append(blockSites[block])
473		common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
#	Line 546 \| Line 478 \| class Cmssw(JobType):
478		eventsRemaining = eventsRemaining - eventsPerJobRequested
479		jobSkipEventCount = 0
480		# reset file
481	<	parString = "\\{"
481	>	pString = ""
482	>	parString = ""
483		filesEventCount = 0
484		newFile = 1
485		fileCount += 1
486	<
486	>
487		# if more events in file remain than eventsPerJobRequested
488		else :
489		# close job but don't touch new file
490		fullString = parString[:-2]
491	<	fullString += '\\}'
492	<	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
491	>	if self.useParent:
492	>	fullParentString = pString[:-2]
493	>	list_of_lists.append([fullString,fullParentString,str(eventsPerJobRequested),str(jobSkipEventCount)])
494	>	else:
495	>	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
496		common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
497		self.jobDestination.append(blockSites[block])
498		common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
#	Line 570 \| Line 506 \| class Cmssw(JobType):
506		jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
507		# remove all but the last file
508		filesEventCount = self.eventsbyfile[file]
509	<	parString = "\\{"
510	<	parString += '\\\"' + file + '\\\"\,'
509	>	if self.useParent:
510	>	for f in parent : pString += '\\\"' + f + '\\\"\,'
511	>	parString = '\\\"' + file + '\\\"\,'
512		pass # END if
513		pass # END while (iterate over files in the block)
514		pass # END while (iterate over blocks in the dataset)
#	Line 579 \| Line 516 \| class Cmssw(JobType):
516		if (eventsRemaining > 0 and jobCount < totalNumberOfJobs ):
517		common.logger.message("Could not run on all requested events because some blocks not hosted at allowed sites.")
518		common.logger.message(str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
519	<
519	>
520		# screen output
521		screenOutput = "List of jobs and available destination sites:\n\n"
522
#	Line 591 \| Line 528 \| class Cmssw(JobType):
528		for block in blocks:
529		if block in jobsOfBlock.keys() :
530		blockCounter += 1
531	<	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
531	>	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),
532	>	','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
533		if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0:
534	<	noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
534	>	noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
535		bloskNoSite.append( blockCounter )
536	<
536	>
537		common.logger.message(screenOutput)
538		if len(noSiteBlock) > 0 and len(bloskNoSite) > 0:
539		msg = 'WARNING: No sites are hosting any part of data for block:\n '
#	Line 611 \| Line 549 \| class Cmssw(JobType):
549		for range_jobs in noSiteBlock:
550		msg += str(range_jobs) + virgola
551		msg += '\n will not be submitted and this block of data can not be analyzed!\n'
552	+	if self.cfg_params.has_key('EDG.se_white_list'):
553	+	msg += 'WARNING: SE White List: '+self.cfg_params['EDG.se_white_list']+'\n'
554	+	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
555	+	msg += 'Please check if the dataset is available at this site!)\n'
556	+	if self.cfg_params.has_key('EDG.ce_white_list'):
557	+	msg += 'WARNING: CE White List: '+self.cfg_params['EDG.ce_white_list']+'\n'
558	+	msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
559	+	msg += 'Please check if the dataset is available at this site!)\n'
560	+
561		common.logger.message(msg)
562
563		self.list_of_args = list_of_lists
#	Line 621 \| Line 568 \| class Cmssw(JobType):
568		Perform job splitting based on number of event per job
569		"""
570		common.logger.debug(5,'Splitting per events')
571	<	common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
572	<	common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
573	<	common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
571	>
572	>	if (self.selectEventsPerJob):
573	>	common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
574	>	if (self.selectNumberOfJobs):
575	>	common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
576	>	if (self.selectTotalNumberEvents):
577	>	common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
578
579		if (self.total_number_of_events < 0):
580		msg='Cannot split jobs per Events with "-1" as total number of events'
#	Line 632 \| Line 583 \| class Cmssw(JobType):
583		if (self.selectEventsPerJob):
584		if (self.selectTotalNumberEvents):
585		self.total_number_of_jobs = int(self.total_number_of_events/self.eventsPerJob)
586	<	elif(self.selectNumberOfJobs) :
586	>	elif(self.selectNumberOfJobs) :
587		self.total_number_of_jobs =self.theNumberOfJobs
588	<	self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
588	>	self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
589
590		elif (self.selectNumberOfJobs) :
591		self.total_number_of_jobs = self.theNumberOfJobs
592		self.eventsPerJob = int(self.total_number_of_events/self.total_number_of_jobs)
593	<
593	>
594		common.logger.debug(5,'N jobs '+str(self.total_number_of_jobs))
595
596		# is there any remainder?
#	Line 655 \| Line 606 \| class Cmssw(JobType):
606		self.list_of_args = []
607		for i in range(self.total_number_of_jobs):
608		## Since there is no input, any site is good
609	<	# self.jobDestination.append(["Any"])
659	<	self.jobDestination.append([""]) #must be empty to write correctly the xml
609	>	self.jobDestination.append([""]) #must be empty to write correctly the xml
610		args=[]
611		if (self.firstRun):
612	<	## pythia first run
663	<	#self.list_of_args.append([(str(self.firstRun)+str(i))])
612	>	## pythia first run
613		args.append(str(self.firstRun)+str(i))
665	–	else:
666	–	## no first run
667	–	#self.list_of_args.append([str(i)])
668	–	args.append(str(i))
669	–	if (self.sourceSeed):
670	–	args.append(str(self.sourceSeed)+str(i))
671	–	if (self.sourceSeedVtx):
672	–	## + vtx random seed
673	–	args.append(str(self.sourceSeedVtx)+str(i))
674	–	if (self.sourceSeedG4):
675	–	## + G4 random seed
676	–	args.append(str(self.sourceSeedG4)+str(i))
677	–	if (self.sourceSeedMix):
678	–	## + Mix random seed
679	–	args.append(str(self.sourceSeedMix)+str(i))
680	–	pass
681	–	pass
614		self.list_of_args.append(args)
683	–	pass
684	–
685	–	# print self.list_of_args
615
616		return
617
618
619	<	def jobSplittingForScript(self):#CarlosDaniele
619	>	def jobSplittingForScript(self):
620		"""
621		Perform job splitting based on number of job
622		"""
#	Line 703 \| Line 632 \| class Cmssw(JobType):
632		# argument is seed number.$i
633		self.list_of_args = []
634		for i in range(self.total_number_of_jobs):
706	–	## Since there is no input, any site is good
707	–	# self.jobDestination.append(["Any"])
635		self.jobDestination.append([""])
709	–	## no random seed
636		self.list_of_args.append([str(i)])
637		return
638
639	<	def split(self, jobParams):
640	<
715	<	common.jobDB.load()
716	<	#### Fabio
639	>	def split(self, jobParams,firstJobID):
640	>
641		njobs = self.total_number_of_jobs
642		arglist = self.list_of_args
643		# create the empty structure
644		for i in range(njobs):
645		jobParams.append("")
722	–
723	–	for job in range(njobs):
724	–	jobParams[job] = arglist[job]
725	–	# print str(arglist[job])
726	–	# print jobParams[job]
727	–	common.jobDB.setArguments(job, jobParams[job])
728	–	common.logger.debug(5,"Job "+str(job)+" Destination: "+str(self.jobDestination[job]))
729	–	common.jobDB.setDestination(job, self.jobDestination[job])
646
647	<	common.jobDB.save()
647	>	listID=[]
648	>	listField=[]
649	>	for id in range(njobs):
650	>	job = id + int(firstJobID)
651	>	jobParams[id] = arglist[id]
652	>	listID.append(job+1)
653	>	job_ToSave ={}
654	>	concString = ' '
655	>	argu=''
656	>	if len(jobParams[id]):
657	>	argu += concString.join(jobParams[id] )
658	>	job_ToSave['arguments']= str(job+1)+' '+argu
659	>	job_ToSave['dlsDestination']= self.jobDestination[id]
660	>	listField.append(job_ToSave)
661	>	msg="Job "+str(job)+" Arguments: "+str(job+1)+" "+argu+"\n" \
662	>	+" Destination: "+str(self.jobDestination[id])
663	>	common.logger.debug(5,msg)
664	>	common._db.updateJob_(listID,listField)
665	>	self.argsList = (len(jobParams[0])+1)
666	>
667		return
668	<
734	<	def getJobTypeArguments(self, nj, sched):
735	<	result = ''
736	<	for i in common.jobDB.arguments(nj):
737	<	result=result+str(i)+" "
738	<	return result
739	<
668	>
669		def numberOfJobs(self):
741	–	# Fabio
670		return self.total_number_of_jobs
671
672		def getTarBall(self, exe):
673		"""
674		Return the TarBall with lib and exe
675		"""
748	–
749	–	# if it exist, just return it
750	–	#
751	–	# Marco. Let's start to use relative path for Boss XML files
752	–	#
676		self.tgzNameWithPath = common.work_space.pathForTgz()+'share/'+self.tgz_name
677		if os.path.exists(self.tgzNameWithPath):
678		return self.tgzNameWithPath
#	Line 763 \| Line 686 \| class Cmssw(JobType):
686
687		# First of all declare the user Scram area
688		swArea = self.scram.getSWArea_()
766	–	#print "swArea = ", swArea
767	–	# swVersion = self.scram.getSWVersion()
768	–	# print "swVersion = ", swVersion
689		swReleaseTop = self.scram.getReleaseTop_()
690	<	#print "swReleaseTop = ", swReleaseTop
771	<
690	>
691		## check if working area is release top
692		if swReleaseTop == '' or swArea == swReleaseTop:
693	+	common.logger.debug(3,"swArea = "+swArea+" swReleaseTop ="+swReleaseTop)
694		return
695
696		import tarfile
#	Line 781 \| Line 701 \| class Cmssw(JobType):
701		exeWithPath = self.scram.findFile_(executable)
702		if ( not exeWithPath ):
703		raise CrabException('User executable '+executable+' not found')
704	<
704	>
705		## then check if it's private or not
706		if exeWithPath.find(swReleaseTop) == -1:
707		# the exe is private, so we must ship
#	Line 790 \| Line 710 \| class Cmssw(JobType):
710		# distinguish case when script is in user project area or given by full path somewhere else
711		if exeWithPath.find(path) >= 0 :
712		exe = string.replace(exeWithPath, path,'')
713	<	tar.add(path+exe,os.path.basename(executable))
713	>	tar.add(path+exe,exe)
714		else :
715		tar.add(exeWithPath,os.path.basename(executable))
716		pass
717		else:
718		# the exe is from release, we'll find it on WN
719		pass
720	<
720	>
721		## Now get the libraries: only those in local working area
722		libDir = 'lib'
723		lib = swArea+'/' +libDir
724		common.logger.debug(5,"lib "+lib+" to be tarred")
725		if os.path.exists(lib):
726		tar.add(lib,libDir)
727	<
727	>
728		## Now check if module dir is present
729		moduleDir = 'module'
730		module = swArea + '/' + moduleDir
#	Line 812 \| Line 732 \| class Cmssw(JobType):
732		tar.add(module,moduleDir)
733
734		## Now check if any data dir(s) is present
735	<	swAreaLen=len(swArea)
736	<	for root, dirs, files in os.walk(swArea):
737	<	if "data" in dirs:
738	<	common.logger.debug(5,"data "+root+"/data"+" to be tarred")
739	<	tar.add(root+"/data",root[swAreaLen:]+"/data")
740	<
741	<	## Add ProdAgent dir to tar
742	<	paDir = 'ProdAgentApi'
743	<	pa = os.environ['CRABDIR'] + '/' + 'ProdAgentApi'
744	<	if os.path.isdir(pa):
745	<	tar.add(pa,paDir)
746	<
747	<	### FEDE FOR DBS PUBLICATION
748	<	## Add PRODCOMMON dir to tar
749	<	prodcommonDir = 'ProdCommon'
750	<	prodcommonPath = os.environ['CRABDIR'] + '/' + 'ProdCommon'
751	<	if os.path.isdir(prodcommonPath):
752	<	tar.add(prodcommonPath,prodcommonDir)
753	<	#############################
754	<
735	>	self.dataExist = False
736	>	srcArea=swArea+"/src/"
737	>	todo_list = [(i, i) for i in os.listdir(srcArea)]
738	>	while len(todo_list):
739	>	entry, name = todo_list.pop()
740	>	if name.startswith('crab_0_') or name.startswith('.') or name == 'CVS':
741	>	continue
742	>	if os.path.isdir(srcArea+entry):
743	>	entryPath = entry + '/'
744	>	todo_list += [(entryPath + i, i) for i in os.listdir(srcArea+entry)]
745	>	if name == 'data':
746	>	self.dataExist=True
747	>	common.logger.debug(5,"data "+entry+" to be tarred")
748	>	tar.add(entry)
749	>	pass
750	>	pass
751	>
752	>	### CMSSW ParameterSet
753	>	if not self.pset is None:
754	>	cfg_file = common.work_space.jobDir()+self.configFilename()
755	>	tar.add(cfg_file,self.configFilename())
756	>	common.logger.debug(5,"File added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
757	>
758	>
759	>	## Add ProdCommon dir to tar
760	>	prodcommonDir = './'
761	>	prodcommonPath = os.environ['CRABDIR'] + '/' + 'external/'
762	>	neededStuff = ['ProdCommon/FwkJobRep', 'IMProv']
763	>	for dir in neededStuff:
764	>	if os.path.isdir(prodcommonPath+dir):
765	>	tar.add(prodcommonPath+dir,prodcommonDir+dir)
766	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
767	>
768	>	##### ML stuff
769	>	ML_file_list=['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py']
770	>	path=os.environ['CRABDIR'] + '/python/'
771	>	for file in ML_file_list:
772	>	tar.add(path+file,file)
773	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
774	>
775	>	##### Utils
776	>	Utils_file_list=['parseCrabFjr.py','writeCfg.py', 'fillCrabFjr.py']
777	>	for file in Utils_file_list:
778	>	tar.add(path+file,file)
779	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
780	>
781	>	##### AdditionalFiles
782	>	for file in self.additional_inbox_files:
783	>	tar.add(file,string.split(file,'/')[-1])
784		common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
785	+
786		tar.close()
787	<	except :
788	<	raise CrabException('Could not create tar-ball')
787	>	except TarError:
788	>	raise CrabException('Could not create tar-ball '+self.tgzNameWithPath)
789
790		## check for tarball size
791		tarballinfo = os.stat(self.tgzNameWithPath)
#	Line 843 \| Line 793 \| class Cmssw(JobType):
793		raise CrabException('Input sandbox size of ' + str(float(tarballinfo.st_size)/1024.0/1024.0) + ' MB is larger than the allowed ' + str(self.MaxTarBallSize) + ' MB input sandbox limit and not supported by the used GRID submission system. Please make sure that no unnecessary files are in all data directories in your local CMSSW project area as they are automatically packed into the input sandbox.')
794
795		## create tar-ball with ML stuff
846	–	self.MLtgzfile = common.work_space.pathForTgz()+'share/MLfiles.tgz'
847	–	try:
848	–	tar = tarfile.open(self.MLtgzfile, "w:gz")
849	–	path=os.environ['CRABDIR'] + '/python/'
850	–	for file in ['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py', 'parseCrabFjr.py']:
851	–	tar.add(path+file,file)
852	–	common.logger.debug(5,"Files added to "+self.MLtgzfile+" : "+str(tar.getnames()))
853	–	tar.close()
854	–	except :
855	–	raise CrabException('Could not create ML files tar-ball')
856	–
857	–	return
858	–
859	–	def additionalInputFileTgz(self):
860	–	"""
861	–	Put all additional files into a tar ball and return its name
862	–	"""
863	–	import tarfile
864	–	tarName= common.work_space.pathForTgz()+'share/'+self.additional_tgz_name
865	–	tar = tarfile.open(tarName, "w:gz")
866	–	for file in self.additional_inbox_files:
867	–	tar.add(file,string.split(file,'/')[-1])
868	–	common.logger.debug(5,"Files added to "+self.additional_tgz_name+" : "+str(tar.getnames()))
869	–	tar.close()
870	–	return tarName
796
797	<	def wsSetupEnvironment(self, nj):
797	>	def wsSetupEnvironment(self, nj=0):
798		"""
799		Returns part of a job script which prepares
800		the execution environment for the job 'nj'.
801		"""
802	+	if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
803	+	psetName = 'pset.py'
804	+	else:
805	+	psetName = 'pset.cfg'
806		# Prepare JobType-independent part
807	<	txt = ''
808	<
809	<	## OLI_Daniele at this level middleware already known
881	<
882	<	txt += 'if [ $middleware == LCG ]; then \n'
883	<	txt += ' echo "### First set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
884	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
885	<	txt += ' export BUILD_ARCH='+self.executable_arch+'\n'
807	>	txt = '\n#Written by cms_cmssw::wsSetupEnvironment\n'
808	>	txt += 'echo ">>> setup environment"\n'
809	>	txt += 'if [ $middleware == LCG ]; then \n'
810		txt += self.wsSetupCMSLCGEnvironment_()
811		txt += 'elif [ $middleware == OSG ]; then\n'
812		txt += ' WORKING_DIR=`/bin/mktemp -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
813	<	txt += ' echo "Created working directory: $WORKING_DIR"\n'
814	<	txt += ' if [ ! -d $WORKING_DIR ] ;then\n'
815	<	txt += ' echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
816	<	txt += ' echo "JOB_EXIT_STATUS = 10016"\n'
893	<	txt += ' echo "JobExitCode=10016" \| tee -a $RUNTIME_AREA/$repo\n'
894	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
895	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
896	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
897	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
898	<	txt += ' exit 1\n'
813	>	txt += ' if [ ! $? == 0 ] ;then\n'
814	>	txt += ' echo "ERROR ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
815	>	txt += ' job_exit_code=10016\n'
816	>	txt += ' func_exit\n'
817		txt += ' fi\n'
818	+	txt += ' echo ">>> Created working directory: $WORKING_DIR"\n'
819		txt += '\n'
820		txt += ' echo "Change to working directory: $WORKING_DIR"\n'
821		txt += ' cd $WORKING_DIR\n'
822	<	txt += self.wsSetupCMSOSGEnvironment_()
823	<	txt += ' echo "### Set SCRAM ARCH to ' + self.executable_arch + ' ###"\n'
905	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
822	>	txt += ' echo ">>> current directory (WORKING_DIR): $WORKING_DIR"\n'
823	>	txt += self.wsSetupCMSOSGEnvironment_()
824		txt += 'fi\n'
825
826		# Prepare JobType-specific part
827		scram = self.scram.commandName()
828		txt += '\n\n'
829	<	txt += 'echo "### SPECIFIC JOB SETUP ENVIRONMENT ###"\n'
829	>	txt += 'echo ">>> specific cmssw setup environment:"\n'
830	>	txt += 'echo "CMSSW_VERSION = '+self.version+'"\n'
831		txt += scram+' project CMSSW '+self.version+'\n'
832		txt += 'status=$?\n'
833		txt += 'if [ $status != 0 ] ; then\n'
834	<	txt += ' echo "SET_EXE_ENV 10034 ==>ERROR CMSSW '+self.version+' not found on `hostname`" \n'
835	<	txt += ' echo "JOB_EXIT_STATUS = 10034"\n'
836	<	txt += ' echo "JobExitCode=10034" \| tee -a $RUNTIME_AREA/$repo\n'
918	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
919	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
920	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
921	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
922	<	## OLI_Daniele
923	<	txt += ' if [ $middleware == OSG ]; then \n'
924	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
925	<	txt += ' cd $RUNTIME_AREA\n'
926	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
927	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
928	<	txt += ' echo "SET_CMS_ENV 10018 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after CMSSW CMSSW_0_6_1 not found on `hostname`"\n'
929	<	txt += ' echo "JOB_EXIT_STATUS = 10018"\n'
930	<	txt += ' echo "JobExitCode=10018" \| tee -a $RUNTIME_AREA/$repo\n'
931	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
932	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
933	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
934	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
935	<	txt += ' fi\n'
936	<	txt += ' fi \n'
937	<	txt += ' exit 1 \n'
834	>	txt += ' echo "ERROR ==> CMSSW '+self.version+' not found on `hostname`" \n'
835	>	txt += ' job_exit_code=10034\n'
836	>	txt += ' func_exit\n'
837		txt += 'fi \n'
939	–	txt += 'echo "CMSSW_VERSION = '+self.version+'"\n'
838		txt += 'cd '+self.version+'\n'
941	–	########## FEDE FOR DBS2 ######################
839		txt += 'SOFTWARE_DIR=`pwd`\n'
840	<	txt += 'echo SOFTWARE_DIR=$SOFTWARE_DIR \n'
944	<	###############################################
945	<	### needed grep for bug in scramv1 ###
946	<	txt += scram+' runtime -sh\n'
840	>	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
841		txt += 'eval `'+scram+' runtime -sh \| grep -v SCRAMRT_LSB_JOBNAME`\n'
842	<	txt += 'echo $PATH\n'
843	<
842	>	txt += 'if [ $? != 0 ] ; then\n'
843	>	txt += ' echo "ERROR ==> Problem with the command: "\n'
844	>	txt += ' echo "eval \`'+scram+' runtime -sh \| grep -v SCRAMRT_LSB_JOBNAME \` at `hostname`"\n'
845	>	txt += ' job_exit_code=10034\n'
846	>	txt += ' func_exit\n'
847	>	txt += 'fi \n'
848		# Handle the arguments:
849		txt += "\n"
850		txt += "## number of arguments (first argument always jobnumber)\n"
851		txt += "\n"
852	<	# txt += "narg=$#\n"
955	<	txt += "if [ $nargs -lt 2 ]\n"
852	>	txt += "if [ $nargs -lt "+str(self.argsList)+" ]\n"
853		txt += "then\n"
854	<	txt += " echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$nargs+ \n"
855	<	txt += ' echo "JOB_EXIT_STATUS = 50113"\n'
856	<	txt += ' echo "JobExitCode=50113" \| tee -a $RUNTIME_AREA/$repo\n'
960	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
961	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
962	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
963	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
964	<	## OLI_Daniele
965	<	txt += ' if [ $middleware == OSG ]; then \n'
966	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
967	<	txt += ' cd $RUNTIME_AREA\n'
968	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
969	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
970	<	txt += ' echo "SET_EXE_ENV 50114 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Too few arguments for CRAB job wrapper"\n'
971	<	txt += ' echo "JOB_EXIT_STATUS = 50114"\n'
972	<	txt += ' echo "JobExitCode=50114" \| tee -a $RUNTIME_AREA/$repo\n'
973	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
974	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
975	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
976	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
977	<	txt += ' fi\n'
978	<	txt += ' fi \n'
979	<	txt += " exit 1\n"
854	>	txt += " echo 'ERROR ==> Too few arguments' +$nargs+ \n"
855	>	txt += ' job_exit_code=50113\n'
856	>	txt += " func_exit\n"
857		txt += "fi\n"
858		txt += "\n"
859
860		# Prepare job-specific part
861		job = common.job_list[nj]
862	<	### FEDE FOR DBS OUTPUT PUBLICATION
986	<	if (self.datasetPath):
862	>	if (self.datasetPath):
863		txt += '\n'
864		txt += 'DatasetPath='+self.datasetPath+'\n'
865
866		datasetpath_split = self.datasetPath.split("/")
867	<
867	>
868		txt += 'PrimaryDataset='+datasetpath_split[1]+'\n'
869		txt += 'DataTier='+datasetpath_split[2]+'\n'
994	–	#txt += 'ProcessedDataset='+datasetpath_split[3]+'\n'
870		txt += 'ApplicationFamily=cmsRun\n'
871
872		else:
873		txt += 'DatasetPath=MCDataTier\n'
874		txt += 'PrimaryDataset=null\n'
875		txt += 'DataTier=null\n'
1001	–	#txt += 'ProcessedDataset=null\n'
876		txt += 'ApplicationFamily=MCDataTier\n'
877	<	if self.pset != None: #CarlosDaniele
877	>	if self.pset != None:
878		pset = os.path.basename(job.configFilename())
879		txt += '\n'
880		txt += 'cp $RUNTIME_AREA/'+pset+' .\n'
881		if (self.datasetPath): # standard job
882	<	#txt += 'InputFiles=$2\n'
883	<	txt += 'InputFiles=${args[1]}\n'
884	<	txt += 'MaxEvents=${args[2]}\n'
885	<	txt += 'SkipEvents=${args[3]}\n'
882	>	txt += 'InputFiles=${args[1]}; export InputFiles\n'
883	>	if (self.useParent):
884	>	txt += 'ParentFiles=${args[2]}; export ParentFiles\n'
885	>	txt += 'MaxEvents=${args[3]}; export MaxEvents\n'
886	>	txt += 'SkipEvents=${args[4]}; export SkipEvents\n'
887	>	else:
888	>	txt += 'MaxEvents=${args[2]}; export MaxEvents\n'
889	>	txt += 'SkipEvents=${args[3]}; export SkipEvents\n'
890		txt += 'echo "Inputfiles:<$InputFiles>"\n'
891	<	txt += 'sed "s#{\'INPUT\'}#$InputFiles#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
891	>	if (self.useParent): txt += 'echo "ParentFiles:<$ParentFiles>"\n'
892		txt += 'echo "MaxEvents:<$MaxEvents>"\n'
1015	–	txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
893		txt += 'echo "SkipEvents:<$SkipEvents>"\n'
1017	–	txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
894		else: # pythia like job
895	<	seedIndex=1
895	>	txt += 'PreserveSeeds=' + ','.join(self.preserveSeeds) + '; export PreserveSeeds\n'
896	>	txt += 'IncrementSeeds=' + ','.join(self.incrementSeeds) + '; export IncrementSeeds\n'
897	>	txt += 'echo "PreserveSeeds: <$PreserveSeeds>"\n'
898	>	txt += 'echo "IncrementSeeds:<$IncrementSeeds>"\n'
899		if (self.firstRun):
900	<	txt += 'FirstRun=${args['+str(seedIndex)+']}\n'
900	>	txt += 'FirstRun=${args[1]}; export FirstRun\n'
901		txt += 'echo "FirstRun: <$FirstRun>"\n'
1023	–	txt += 'sed "s#\<INPUTFIRSTRUN\>#$FirstRun#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1024	–	seedIndex=seedIndex+1
902
903	<	if (self.sourceSeed):
1027	<	txt += 'Seed=${args['+str(seedIndex)+']}\n'
1028	<	txt += 'sed "s#\<INPUT\>#$Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1029	<	seedIndex=seedIndex+1
1030	<	## the following seeds are not always present
1031	<	if (self.sourceSeedVtx):
1032	<	txt += 'VtxSeed=${args['+str(seedIndex)+']}\n'
1033	<	txt += 'echo "VtxSeed: <$VtxSeed>"\n'
1034	<	txt += 'sed "s#\<INPUTVTX\>#$VtxSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1035	<	seedIndex += 1
1036	<	if (self.sourceSeedG4):
1037	<	txt += 'G4Seed=${args['+str(seedIndex)+']}\n'
1038	<	txt += 'echo "G4Seed: <$G4Seed>"\n'
1039	<	txt += 'sed "s#\<INPUTG4\>#$G4Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1040	<	seedIndex += 1
1041	<	if (self.sourceSeedMix):
1042	<	txt += 'mixSeed=${args['+str(seedIndex)+']}\n'
1043	<	txt += 'echo "MixSeed: <$mixSeed>"\n'
1044	<	txt += 'sed "s#\<INPUTMIX\>#$mixSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1045	<	seedIndex += 1
1046	<	pass
1047	<	pass
1048	<	txt += 'mv -f '+pset+' pset.cfg\n'
903	>	txt += 'mv -f ' + pset + ' ' + psetName + '\n'
904
1050	–	if len(self.additional_inbox_files) > 0:
1051	–	txt += 'if [ -e $RUNTIME_AREA/'+self.additional_tgz_name+' ] ; then\n'
1052	–	txt += ' tar xzvf $RUNTIME_AREA/'+self.additional_tgz_name+'\n'
1053	–	txt += 'fi\n'
1054	–	pass
905
906	<	if self.pset != None: #CarlosDaniele
907	<	txt += 'echo "### END JOB SETUP ENVIRONMENT ###"\n\n'
1058	<
1059	<	txt += '\n'
1060	<	txt += 'echo "*** cat pset.cfg *******"\n'
1061	<	txt += 'cat pset.cfg\n'
1062	<	txt += 'echo "**** end pset.cfg ******"\n'
906	>	if self.pset != None:
907	>	# FUTURE: Can simply for 2_1_x and higher
908		txt += '\n'
909	<	### FEDE FOR DBS OUTPUT PUBLICATION
910	<	txt += 'PSETHASH=`EdmConfigHash < pset.cfg` \n'
909	>	if self.debug_wrapper==True:
910	>	txt += 'echo "*** cat ' + psetName + ' *******"\n'
911	>	txt += 'cat ' + psetName + '\n'
912	>	txt += 'echo "**** end ' + psetName + ' ******"\n'
913	>	txt += '\n'
914	>	txt += 'PSETHASH=`edmConfigHash < ' + psetName + '` \n'
915		txt += 'echo "PSETHASH = $PSETHASH" \n'
1067	–	##############
916		txt += '\n'
1069	–	# txt += 'echo "*** cat pset1.cfg *******"\n'
1070	–	# txt += 'cat pset1.cfg\n'
1071	–	# txt += 'echo "**** end pset1.cfg ******"\n'
917		return txt
918
919	<	def wsBuildExe(self, nj=0):
919	>	def wsUntarSoftware(self, nj=0):
920		"""
921		Put in the script the commands to build an executable
922		or a library.
923		"""
924
925	<	txt = ""
925	>	txt = '\n#Written by cms_cmssw::wsUntarSoftware\n'
926
927		if os.path.isfile(self.tgzNameWithPath):
928	<	txt += 'echo "tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'"\n'
928	>	txt += 'echo ">>> tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+' :" \n'
929		txt += 'tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'\n'
930	+	if self.debug_wrapper:
931	+	txt += 'ls -Al \n'
932		txt += 'untar_status=$? \n'
933		txt += 'if [ $untar_status -ne 0 ]; then \n'
934	<	txt += ' echo "SET_EXE 1 ==> ERROR Untarring .tgz file failed"\n'
935	<	txt += ' echo "JOB_EXIT_STATUS = $untar_status" \n'
936	<	txt += ' echo "JobExitCode=$untar_status" \| tee -a $RUNTIME_AREA/$repo\n'
1090	<	txt += ' if [ $middleware == OSG ]; then \n'
1091	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
1092	<	txt += ' cd $RUNTIME_AREA\n'
1093	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
1094	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
1095	<	txt += ' echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
1096	<	txt += ' echo "JOB_EXIT_STATUS = 50999"\n'
1097	<	txt += ' echo "JobExitCode=50999" \| tee -a $RUNTIME_AREA/$repo\n'
1098	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1099	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1100	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1101	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1102	<	txt += ' fi\n'
1103	<	txt += ' fi \n'
1104	<	txt += ' \n'
1105	<	txt += ' exit 1 \n'
934	>	txt += ' echo "ERROR ==> Untarring .tgz file failed"\n'
935	>	txt += ' job_exit_code=$untar_status\n'
936	>	txt += ' func_exit\n'
937		txt += 'else \n'
938		txt += ' echo "Successful untar" \n'
939		txt += 'fi \n'
940		txt += '\n'
941	<	txt += 'echo "Include ProdAgentApi and PRODCOMMON in PYTHONPATH"\n'
941	>	txt += 'echo ">>> Include $RUNTIME_AREA in PYTHONPATH:"\n'
942		txt += 'if [ -z "$PYTHONPATH" ]; then\n'
943	<	#### FEDE FOR DBS OUTPUT PUBLICATION
1113	<	txt += ' export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon\n'
1114	<	#txt += ' export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon\n'
1115	<	#txt += ' export PYTHONPATH=ProdAgentApi\n'
943	>	txt += ' export PYTHONPATH=$RUNTIME_AREA/\n'
944		txt += 'else\n'
945	<	txt += ' export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
1118	<	#txt += ' export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon:${PYTHONPATH}\n'
1119	<	#txt += ' export PYTHONPATH=ProdAgentApi:${PYTHONPATH}\n'
945	>	txt += ' export PYTHONPATH=$RUNTIME_AREA/:${PYTHONPATH}\n'
946		txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
1121	–	###################
947		txt += 'fi\n'
948		txt += '\n'
949
950		pass
951	<
951	>
952		return txt
953
954	<	def modifySteeringCards(self, nj):
954	>	def wsBuildExe(self, nj=0):
955		"""
956	<	modify the card provided by the user,
957	<	writing a new card into share dir
956	>	Put in the script the commands to build an executable
957	>	or a library.
958		"""
959	<
959	>
960	>	txt = '\n#Written by cms_cmssw::wsBuildExe\n'
961	>	txt += 'echo ">>> moving CMSSW software directories in `pwd`" \n'
962	>
963	>	txt += 'rm -r lib/ module/ \n'
964	>	txt += 'mv $RUNTIME_AREA/lib/ . \n'
965	>	txt += 'mv $RUNTIME_AREA/module/ . \n'
966	>	if self.dataExist == True:
967	>	txt += 'rm -r src/ \n'
968	>	txt += 'mv $RUNTIME_AREA/src/ . \n'
969	>	if len(self.additional_inbox_files)>0:
970	>	for file in self.additional_inbox_files:
971	>	txt += 'mv $RUNTIME_AREA/'+os.path.basename(file)+' . \n'
972	>	txt += 'mv $RUNTIME_AREA/ProdCommon/ . \n'
973	>	txt += 'mv $RUNTIME_AREA/IMProv/ . \n'
974	>
975	>	txt += 'echo ">>> Include $RUNTIME_AREA in PYTHONPATH:"\n'
976	>	txt += 'if [ -z "$PYTHONPATH" ]; then\n'
977	>	txt += ' export PYTHONPATH=$RUNTIME_AREA/\n'
978	>	txt += 'else\n'
979	>	txt += ' export PYTHONPATH=$RUNTIME_AREA/:${PYTHONPATH}\n'
980	>	txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
981	>	txt += 'fi\n'
982	>	txt += '\n'
983	>
984	>	return txt
985	>
986	>
987		def executableName(self):
988	<	if self.scriptExe: #CarlosDaniele
988	>	if self.scriptExe:
989		return "sh "
990		else:
991		return self.executable
992
993		def executableArgs(self):
994	+	# FUTURE: This function tests the CMSSW version. Can be simplified as we drop support for old versions
995		if self.scriptExe:#CarlosDaniele
996		return self.scriptExe + " $NJob"
997		else:
998	<	# if >= CMSSW_1_5_X, add -e
999	<	version_array = self.scram.getSWVersion().split('_')
1000	<	major = 0
1001	<	minor = 0
1002	<	try:
1003	<	major = int(version_array[1])
1004	<	minor = int(version_array[2])
1005	<	except:
1153	<	msg = "Cannot parse CMSSW version string: " + "_".join(version_array) + " for major and minor release number!"
1154	<	raise CrabException(msg)
1155	<	if major >= 1 and minor >= 5 :
1156	<	return " -e -p pset.cfg"
998	>	ex_args = ""
999	>	# FUTURE: This tests the CMSSW version. Can remove code as versions deprecated
1000	>	# Framework job report
1001	>	if (self.CMSSW_major >= 1 and self.CMSSW_minor >= 5) or (self.CMSSW_major >= 2):
1002	>	ex_args += " -j $RUNTIME_AREA/crab_fjr_$NJob.xml"
1003	>	# Type of config file
1004	>	if self.CMSSW_major >= 2 :
1005	>	ex_args += " -p pset.py"
1006		else:
1007	<	return " -p pset.cfg"
1007	>	ex_args += " -p pset.cfg"
1008	>	return ex_args
1009
1010		def inputSandbox(self, nj):
1011		"""
1012		Returns a list of filenames to be put in JDL input sandbox.
1013		"""
1014		inp_box = []
1165	–	# # dict added to delete duplicate from input sandbox file list
1166	–	# seen = {}
1167	–	## code
1015		if os.path.isfile(self.tgzNameWithPath):
1016		inp_box.append(self.tgzNameWithPath)
1017	<	if os.path.isfile(self.MLtgzfile):
1018	<	inp_box.append(self.MLtgzfile)
1172	<	## config
1173	<	if not self.pset is None:
1174	<	inp_box.append(common.work_space.pathForTgz() + 'job/' + self.configFilename())
1175	<	## additional input files
1176	<	tgz = self.additionalInputFileTgz()
1177	<	inp_box.append(tgz)
1017	>	wrapper = os.path.basename(str(common._db.queryTask('scriptName')))
1018	>	inp_box.append(common.work_space.pathForTgz() +'job/'+ wrapper)
1019		return inp_box
1020
1021		def outputSandbox(self, nj):
#	Line 1185 \| Line 1026 \| class Cmssw(JobType):
1026
1027		## User Declared output files
1028		for out in (self.output_file+self.output_file_sandbox):
1029	<	n_out = nj + 1
1030	<	out_box.append(self.numberFile_(out,str(n_out)))
1029	>	n_out = nj + 1
1030	>	out_box.append(numberFile(out,str(n_out)))
1031		return out_box
1032
1192	–	def prepareSteeringCards(self):
1193	–	"""
1194	–	Make initial modifications of the user's steering card file.
1195	–	"""
1196	–	return
1033
1034		def wsRenameOutput(self, nj):
1035		"""
1036		Returns part of a job script which renames the produced files.
1037		"""
1038
1039	<	txt = '\n'
1040	<	txt += '# directory content\n'
1041	<	txt += 'ls \n'
1042	<
1043	<	txt += 'output_exit_status=0\n'
1044	<
1045	<	for fileWithSuffix in (self.output_file_sandbox):
1210	<	output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1211	<	txt += '\n'
1212	<	txt += '# check output file\n'
1213	<	txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1214	<	txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1215	<	txt += ' cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1216	<	txt += 'else\n'
1217	<	txt += ' exit_status=60302\n'
1218	<	txt += ' echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1219	<	if common.scheduler.boss_scheduler_name == 'condor_g':
1220	<	txt += ' if [ $middleware == OSG ]; then \n'
1221	<	txt += ' echo "prepare dummy output file"\n'
1222	<	txt += ' echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1223	<	txt += ' fi \n'
1224	<	txt += 'fi\n'
1225	<
1039	>	txt = '\n#Written by cms_cmssw::wsRenameOutput\n'
1040	>	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1041	>	txt += 'echo ">>> current directory content:"\n'
1042	>	if self.debug_wrapper:
1043	>	txt += 'ls -Al\n'
1044	>	txt += '\n'
1045	>
1046		for fileWithSuffix in (self.output_file):
1047	<	output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1047	>	output_file_num = numberFile(fileWithSuffix, '$NJob')
1048		txt += '\n'
1049		txt += '# check output file\n'
1050		txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1051	<	txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1052	<	txt += ' cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1051	>	if (self.copy_data == 1): # For OSG nodes, file is in $WORKING_DIR, should not be moved to $RUNTIME_AREA
1052	>	txt += ' mv '+fileWithSuffix+' '+output_file_num+'\n'
1053	>	txt += ' ln -s `pwd`/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1054	>	else:
1055	>	txt += ' mv '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1056	>	txt += ' ln -s $RUNTIME_AREA/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1057		txt += 'else\n'
1058	<	txt += ' exit_status=60302\n'
1059	<	txt += ' echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1060	<	txt += ' echo "JOB_EXIT_STATUS = $exit_status"\n'
1237	<	txt += ' output_exit_status=$exit_status\n'
1238	<	if common.scheduler.boss_scheduler_name == 'condor_g':
1058	>	txt += ' job_exit_code=60302\n'
1059	>	txt += ' echo "WARNING: Output file '+fileWithSuffix+' not found"\n'
1060	>	if common.scheduler.name().upper() == 'CONDOR_G':
1061		txt += ' if [ $middleware == OSG ]; then \n'
1062		txt += ' echo "prepare dummy output file"\n'
1063		txt += ' echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
#	Line 1243 \| Line 1065 \| class Cmssw(JobType):
1065		txt += 'fi\n'
1066		file_list = []
1067		for fileWithSuffix in (self.output_file):
1068	<	file_list.append(self.numberFile_(fileWithSuffix, '$NJob'))
1069	<
1068	>	file_list.append(numberFile(fileWithSuffix, '$NJob'))
1069	>
1070		txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1071	+	txt += '\n'
1072	+	txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1073	+	txt += 'echo ">>> current directory content:"\n'
1074	+	if self.debug_wrapper:
1075	+	txt += 'ls -Al\n'
1076	+	txt += '\n'
1077		txt += 'cd $RUNTIME_AREA\n'
1078	+	txt += 'echo ">>> current directory (RUNTIME_AREA): $RUNTIME_AREA"\n'
1079		return txt
1080
1252	–	def numberFile_(self, file, txt):
1253	–	"""
1254	–	append _'txt' before last extension of a file
1255	–	"""
1256	–	p = string.split(file,".")
1257	–	# take away last extension
1258	–	name = p[0]
1259	–	for x in p[1:-1]:
1260	–	name=name+"."+x
1261	–	# add "_txt"
1262	–	if len(p)>1:
1263	–	ext = p[len(p)-1]
1264	–	result = name + '_' + txt + "." + ext
1265	–	else:
1266	–	result = name + '_' + txt
1267	–
1268	–	return result
1269	–
1081		def getRequirements(self, nj=[]):
1082		"""
1083	<	return job requirements to add to jdl files
1083	>	return job requirements to add to jdl files
1084		"""
1085		req = ''
1086		if self.version:
1087		req='Member("VO-cms-' + \
1088		self.version + \
1089		'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1090	<	## SL add requirement for OS version only if SL4
1280	<	#reSL4 = re.compile( r'slc4' )
1281	<	if self.executable_arch: # and reSL4.search(self.executable_arch):
1090	>	if self.executable_arch:
1091		req+=' && Member("VO-cms-' + \
1092		self.executable_arch + \
1093		'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1094
1095		req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1096	+	if common.scheduler.name() == "glitecoll":
1097	+	req += ' && other.GlueCEStateStatus == "Production" '
1098
1099		return req
1100
1101		def configFilename(self):
1102		""" return the config filename """
1103	<	return self.name()+'.cfg'
1103	>	# FUTURE: Can remove cfg mode for CMSSW >= 2_1_x
1104	>	if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
1105	>	return self.name()+'.py'
1106	>	else:
1107	>	return self.name()+'.cfg'
1108
1294	–	### OLI_DANIELE
1109		def wsSetupCMSOSGEnvironment_(self):
1110		"""
1111		Returns part of a job script which is prepares
1112		the execution environment and which is common for all CMS jobs.
1113		"""
1114	<	txt = '\n'
1115	<	txt += ' echo "### SETUP CMS OSG ENVIRONMENT ###"\n'
1116	<	txt += ' if [ -f $GRID3_APP_DIR/cmssoft/cmsset_default.sh ] ;then\n'
1117	<	txt += ' # Use $GRID3_APP_DIR/cmssoft/cmsset_default.sh to setup cms software\n'
1118	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1119	<	txt += ' source $GRID3_APP_DIR/cmssoft/cmsset_default.sh '+self.version+'\n'
1306	<	txt += ' elif [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1114	>	txt = '\n#Written by cms_cmssw::wsSetupCMSOSGEnvironment_\n'
1115	>	txt += ' echo ">>> setup CMS OSG environment:"\n'
1116	>	txt += ' echo "set SCRAM ARCH to ' + self.executable_arch + '"\n'
1117	>	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1118	>	txt += ' echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1119	>	txt += ' if [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1120		txt += ' # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1121	<	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1122	<	txt += ' source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1123	<	txt += ' else\n'
1124	<	txt += ' echo "SET_CMS_ENV 10020 ==> ERROR $GRID3_APP_DIR/cmssoft/cmsset_default.sh and $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1125	<	txt += ' echo "JOB_EXIT_STATUS = 10020"\n'
1126	<	txt += ' echo "JobExitCode=10020" \| tee -a $RUNTIME_AREA/$repo\n'
1314	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1315	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1316	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1317	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1318	<	txt += ' exit 1\n'
1319	<	txt += '\n'
1320	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
1321	<	txt += ' cd $RUNTIME_AREA\n'
1322	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
1323	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
1324	<	txt += ' echo "SET_CMS_ENV 10017 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after $GRID3_APP_DIR/cmssoft/cmsset_default.sh and $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1325	<	txt += ' echo "JOB_EXIT_STATUS = 10017"\n'
1326	<	txt += ' echo "JobExitCode=10017" \| tee -a $RUNTIME_AREA/$repo\n'
1327	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1328	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1329	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1330	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1331	<	txt += ' fi\n'
1332	<	txt += '\n'
1333	<	txt += ' exit 1\n'
1334	<	txt += ' fi\n'
1121	>	txt += ' source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1122	>	txt += ' else\n'
1123	>	txt += ' echo "ERROR ==> $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1124	>	txt += ' job_exit_code=10020\n'
1125	>	txt += ' func_exit\n'
1126	>	txt += ' fi\n'
1127		txt += '\n'
1128	<	txt += ' echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1129	<	txt += ' echo " END SETUP CMS OSG ENVIRONMENT "\n'
1128	>	txt += ' echo "==> setup cms environment ok"\n'
1129	>	txt += ' echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1130
1131		return txt
1132	<
1341	<	### OLI_DANIELE
1132	>
1133		def wsSetupCMSLCGEnvironment_(self):
1134		"""
1135		Returns part of a job script which is prepares
1136		the execution environment and which is common for all CMS jobs.
1137		"""
1138	<	txt = ' \n'
1139	<	txt += ' echo " ### SETUP CMS LCG ENVIRONMENT ### "\n'
1140	<	txt += ' if [ ! $VO_CMS_SW_DIR ] ;then\n'
1141	<	txt += ' echo "SET_CMS_ENV 10031 ==> ERROR CMS software dir not found on WN `hostname`"\n'
1142	<	txt += ' echo "JOB_EXIT_STATUS = 10031" \n'
1143	<	txt += ' echo "JobExitCode=10031" \| tee -a $RUNTIME_AREA/$repo\n'
1144	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1145	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1146	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1147	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1148	<	txt += ' exit 1\n'
1149	<	txt += ' else\n'
1150	<	txt += ' echo "Sourcing environment... "\n'
1151	<	txt += ' if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1152	<	txt += ' echo "SET_CMS_ENV 10020 ==> ERROR cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1153	<	txt += ' echo "JOB_EXIT_STATUS = 10020"\n'
1154	<	txt += ' echo "JobExitCode=10020" \| tee -a $RUNTIME_AREA/$repo\n'
1155	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1156	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1157	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1158	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1159	<	txt += ' exit 1\n'
1160	<	txt += ' fi\n'
1161	<	txt += ' echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1162	<	txt += ' source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1163	<	txt += ' result=$?\n'
1164	<	txt += ' if [ $result -ne 0 ]; then\n'
1374	<	txt += ' echo "SET_CMS_ENV 10032 ==> ERROR problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1375	<	txt += ' echo "JOB_EXIT_STATUS = 10032"\n'
1376	<	txt += ' echo "JobExitCode=10032" \| tee -a $RUNTIME_AREA/$repo\n'
1377	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1378	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1379	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1380	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1381	<	txt += ' exit 1\n'
1382	<	txt += ' fi\n'
1383	<	txt += ' fi\n'
1384	<	txt += ' \n'
1385	<	txt += ' echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1386	<	txt += ' echo "### END SETUP CMS LCG ENVIRONMENT ###"\n'
1138	>	txt = '\n#Written by cms_cmssw::wsSetupCMSLCGEnvironment_\n'
1139	>	txt += ' echo ">>> setup CMS LCG environment:"\n'
1140	>	txt += ' echo "set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
1141	>	txt += ' export SCRAM_ARCH='+self.executable_arch+'\n'
1142	>	txt += ' export BUILD_ARCH='+self.executable_arch+'\n'
1143	>	txt += ' if [ ! $VO_CMS_SW_DIR ] ;then\n'
1144	>	txt += ' echo "ERROR ==> CMS software dir not found on WN `hostname`"\n'
1145	>	txt += ' job_exit_code=10031\n'
1146	>	txt += ' func_exit\n'
1147	>	txt += ' else\n'
1148	>	txt += ' echo "Sourcing environment... "\n'
1149	>	txt += ' if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1150	>	txt += ' echo "ERROR ==> cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1151	>	txt += ' job_exit_code=10020\n'
1152	>	txt += ' func_exit\n'
1153	>	txt += ' fi\n'
1154	>	txt += ' echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1155	>	txt += ' source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1156	>	txt += ' result=$?\n'
1157	>	txt += ' if [ $result -ne 0 ]; then\n'
1158	>	txt += ' echo "ERROR ==> problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1159	>	txt += ' job_exit_code=10032\n'
1160	>	txt += ' func_exit\n'
1161	>	txt += ' fi\n'
1162	>	txt += ' fi\n'
1163	>	txt += ' \n'
1164	>	txt += ' echo "==> setup cms environment ok"\n'
1165		return txt
1166
1389	–	### FEDE FOR DBS OUTPUT PUBLICATION
1167		def modifyReport(self, nj):
1168		"""
1169	<	insert the part of the script that modifies the FrameworkJob Report
1169	>	insert the part of the script that modifies the FrameworkJob Report
1170		"""
1171	+	txt = '\n#Written by cms_cmssw::modifyReport\n'
1172	+	publish_data = int(self.cfg_params.get('USER.publish_data',0))
1173	+	if (publish_data == 1):
1174	+	processedDataset = self.cfg_params['USER.publish_data_name']
1175	+	LFNBaseName = LFNBase(processedDataset)
1176
1177	<	txt = ''
1178	<	try:
1179	<	publish_data = int(self.cfg_params['USER.publish_data'])
1180	<	except KeyError:
1181	<	publish_data = 0
1182	<	if (publish_data == 1):
1183	<	txt += 'echo "Modify Job Report" \n'
1402	<	#txt += 'chmod a+x $RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1403	<	################ FEDE FOR DBS2 #############################################
1404	<	txt += 'chmod a+x $SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1405	<	#############################################################################
1406	<	#try:
1407	<	# publish_data = int(self.cfg_params['USER.publish_data'])
1408	<	#except KeyError:
1409	<	# publish_data = 0
1410	<
1411	<	txt += 'if [ -z "$SE" ]; then\n'
1412	<	txt += ' SE="" \n'
1413	<	txt += 'fi \n'
1414	<	txt += 'if [ -z "$SE_PATH" ]; then\n'
1415	<	txt += ' SE_PATH="" \n'
1416	<	txt += 'fi \n'
1417	<	txt += 'echo "SE = $SE"\n'
1418	<	txt += 'echo "SE_PATH = $SE_PATH"\n'
1177	>	txt += 'if [ $copy_exit_status -eq 0 ]; then\n'
1178	>	txt += ' FOR_LFN=%s_${PSETHASH}/\n'%(LFNBaseName)
1179	>	txt += 'else\n'
1180	>	txt += ' FOR_LFN=/copy_problems/ \n'
1181	>	txt += ' SE=""\n'
1182	>	txt += ' SE_PATH=""\n'
1183	>	txt += 'fi\n'
1184
1185	<	#if (publish_data == 1):
1186	<	#processedDataset = self.cfg_params['USER.processed_datasetname']
1422	<	processedDataset = self.cfg_params['USER.publish_data_name']
1185	>	txt += 'echo ">>> Modify Job Report:" \n'
1186	>	txt += 'chmod a+x $SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py\n'
1187		txt += 'ProcessedDataset='+processedDataset+'\n'
1424	–	#### LFN=/store/user/<user>/processedDataset_PSETHASH
1425	–	txt += 'if [ "$SE_PATH" == "" ]; then\n'
1426	–	#### FEDE: added slash in LFN ##############
1427	–	txt += ' FOR_LFN=/copy_problems/ \n'
1428	–	txt += 'else \n'
1429	–	txt += ' tmp=`echo $SE_PATH \| awk -F \'store\' \'{print$2}\'` \n'
1430	–	##### FEDE TO BE CHANGED, BECAUSE STORE IS HARDCODED!!!! ########
1431	–	txt += ' FOR_LFN=/store$tmp \n'
1432	–	txt += 'fi \n'
1188		txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1189	+	txt += 'echo "SE = $SE"\n'
1190	+	txt += 'echo "SE_PATH = $SE_PATH"\n'
1191		txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1192		txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1193	<	#txt += 'echo "$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1194	<	txt += 'echo "$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1438	<	txt += '$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1439	<	#txt += '$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1440	<
1193	>	txt += 'echo "$SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py $RUNTIME_AREA/crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1194	>	txt += '$SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py $RUNTIME_AREA/crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1195		txt += 'modifyReport_result=$?\n'
1442	–	txt += 'echo modifyReport_result = $modifyReport_result\n'
1196		txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1197	<	txt += ' exit_status=1\n'
1198	<	txt += ' echo "ERROR: Problem with ModifyJobReport"\n'
1197	>	txt += ' modifyReport_result=70500\n'
1198	>	txt += ' job_exit_code=$modifyReport_result\n'
1199	>	txt += ' echo "ModifyReportResult=$modifyReport_result" \| tee -a $RUNTIME_AREA/$repo\n'
1200	>	txt += ' echo "WARNING: Problem with ModifyJobReport"\n'
1201		txt += 'else\n'
1202	<	txt += ' mv NewFrameworkJobReport.xml crab_fjr_$NJob.xml\n'
1202	>	txt += ' mv NewFrameworkJobReport.xml $RUNTIME_AREA/crab_fjr_$NJob.xml\n'
1203		txt += 'fi\n'
1449	–	else:
1450	–	txt += 'echo "no data publication required"\n'
1451	–	#txt += 'ProcessedDataset=no_data_to_publish \n'
1452	–	#### FEDE: added slash in LFN ##############
1453	–	#txt += 'FOR_LFN=/local/ \n'
1454	–	#txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1455	–	#txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1204		return txt
1205
1206	<	def cleanEnv(self):
1207	<	### OLI_DANIELE
1208	<	txt = ''
1209	<	txt += 'if [ $middleware == OSG ]; then\n'
1210	<	txt += ' cd $RUNTIME_AREA\n'
1211	<	txt += ' echo "Remove working directory: $WORKING_DIR"\n'
1212	<	txt += ' /bin/rm -rf $WORKING_DIR\n'
1213	<	txt += ' if [ -d $WORKING_DIR ] ;then\n'
1214	<	txt += ' echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1215	<	txt += ' echo "JOB_EXIT_STATUS = 60999"\n'
1216	<	txt += ' echo "JobExitCode=60999" \| tee -a $RUNTIME_AREA/$repo\n'
1217	<	txt += ' dumpStatus $RUNTIME_AREA/$repo\n'
1218	<	txt += ' rm -f $RUNTIME_AREA/$repo \n'
1219	<	txt += ' echo "MonitorJobID=`echo $MonitorJobID`" \| tee -a $RUNTIME_AREA/$repo \n'
1220	<	txt += ' echo "MonitorID=`echo $MonitorID`" \| tee -a $RUNTIME_AREA/$repo\n'
1206	>	def wsParseFJR(self):
1207	>	"""
1208	>	Parse the FrameworkJobReport to obtain useful infos
1209	>	"""
1210	>	txt = '\n#Written by cms_cmssw::wsParseFJR\n'
1211	>	txt += 'echo ">>> Parse FrameworkJobReport crab_fjr.xml"\n'
1212	>	txt += 'if [ -s $RUNTIME_AREA/crab_fjr_$NJob.xml ]; then\n'
1213	>	txt += ' if [ -s $RUNTIME_AREA/parseCrabFjr.py ]; then\n'
1214	>	txt += ' cmd_out=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --dashboard $MonitorID,$MonitorJobID '+self.debugWrap+'`\n'
1215	>	if self.debug_wrapper :
1216	>	txt += ' echo "Result of parsing the FrameworkJobReport crab_fjr.xml: $cmd_out"\n'
1217	>	txt += ' executable_exit_status=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --exitcode`\n'
1218	>	txt += ' if [ $executable_exit_status -eq 50115 ];then\n'
1219	>	txt += ' echo ">>> crab_fjr.xml contents: "\n'
1220	>	txt += ' cat $RUNTIME_AREA/crab_fjr_NJob.xml\n'
1221	>	txt += ' echo "Wrong FrameworkJobReport --> does not contain useful info. ExitStatus: $executable_exit_status"\n'
1222	>	txt += ' elif [ $executable_exit_status -eq -999 ];then\n'
1223	>	txt += ' echo "ExitStatus from FrameworkJobReport not available. not available. Using exit code of executable from command line."\n'
1224	>	txt += ' else\n'
1225	>	txt += ' echo "Extracted ExitStatus from FrameworkJobReport parsing output: $executable_exit_status"\n'
1226	>	txt += ' fi\n'
1227	>	txt += ' else\n'
1228	>	txt += ' echo "CRAB python script to parse CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1229		txt += ' fi\n'
1230	+	#### Patch to check input data reading for CMSSW16x Hopefully we-ll remove it asap
1231	+
1232	+	if self.datasetPath:
1233	+	# VERIFY PROCESSED DATA
1234	+	txt += ' if [ $executable_exit_status -eq 0 ];then\n'
1235	+	txt += ' echo ">>> Verify list of processed files:"\n'
1236	+	txt += ' echo $InputFiles \|tr -d \'\\\\\' \|tr \',\' \'\\n\'\|tr -d \'"\' > input-files.txt\n'
1237	+	txt += ' python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --lfn > processed-files.txt\n'
1238	+	txt += ' cat input-files.txt \| sort \| uniq > tmp.txt\n'
1239	+	txt += ' mv tmp.txt input-files.txt\n'
1240	+	txt += ' echo "cat input-files.txt"\n'
1241	+	txt += ' echo "----------------------"\n'
1242	+	txt += ' cat input-files.txt\n'
1243	+	txt += ' cat processed-files.txt \| sort \| uniq > tmp.txt\n'
1244	+	txt += ' mv tmp.txt processed-files.txt\n'
1245	+	txt += ' echo "----------------------"\n'
1246	+	txt += ' echo "cat processed-files.txt"\n'
1247	+	txt += ' echo "----------------------"\n'
1248	+	txt += ' cat processed-files.txt\n'
1249	+	txt += ' echo "----------------------"\n'
1250	+	txt += ' diff -q input-files.txt processed-files.txt\n'
1251	+	txt += ' fileverify_status=$?\n'
1252	+	txt += ' if [ $fileverify_status -ne 0 ]; then\n'
1253	+	txt += ' executable_exit_status=30001\n'
1254	+	txt += ' echo "ERROR ==> not all input files processed"\n'
1255	+	txt += ' echo " ==> list of processed files from crab_fjr.xml differs from list in pset.cfg"\n'
1256	+	txt += ' echo " ==> diff input-files.txt processed-files.txt"\n'
1257	+	txt += ' fi\n'
1258	+	txt += ' fi\n'
1259	+	txt += '\n'
1260	+	txt += 'else\n'
1261	+	txt += ' echo "CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1262		txt += 'fi\n'
1263		txt += '\n'
1264	+	txt += 'echo "ExeExitCode=$executable_exit_status" \| tee -a $RUNTIME_AREA/$repo\n'
1265	+	txt += 'echo "EXECUTABLE_EXIT_STATUS = $executable_exit_status"\n'
1266	+	txt += 'job_exit_code=$executable_exit_status\n'
1267	+
1268		return txt
1269
1270		def setParam_(self, param, value):
#	Line 1481 \| Line 1273 \| class Cmssw(JobType):
1273		def getParams(self):
1274		return self._params
1275
1484	–	def setTaskid_(self):
1485	–	self._taskId = self.cfg_params['taskId']
1486	–
1487	–	def getTaskid(self):
1488	–	return self._taskId
1489	–
1276		def uniquelist(self, old):
1277		"""
1278		remove duplicates from a list
#	Line 1496 \| Line 1282 \| class Cmssw(JobType):
1282		nd[e]=0
1283		return nd.keys()
1284
1285	<
1500	<	def checkOut(self, limit):
1285	>	def outList(self):
1286		"""
1287		check the dimension of the output files
1288		"""
1289	<	txt = 'echo "*****************************************"\n'
1290	<	txt += 'echo " Starting output sandbox limit check "\n'
1506	<	txt += 'echo "*****************************************"\n'
1507	<	allOutFiles = ""
1289	>	txt = ''
1290	>	txt += 'echo ">>> list of expected files on output sandbox"\n'
1291		listOutFiles = []
1292	<	for fileOut in (self.output_file+self.output_file_sandbox):
1293	<	if fileOut.find('crab_fjr') == -1:
1294	<	allOutFiles = allOutFiles + " " + self.numberFile_(fileOut, '$NJob')
1295	<	listOutFiles.append(self.numberFile_(fileOut, '$NJob'))
1296	<	txt += 'echo "OUTPUT files: '+str(allOutFiles)+'";\n'
1297	<	txt += 'ls -gGhrta;\n'
1298	<	txt += 'sum=0;\n'
1299	<	txt += 'for file in '+str(allOutFiles)+' ; do\n'
1300	<	txt += ' if [ -e $file ]; then\n'
1301	<	txt += ' tt=`ls -gGrta $file \| awk \'{ print $3 }\'`\n'
1302	<	txt += ' sum=`expr $sum + $tt`\n'
1303	<	txt += ' else\n'
1304	<	txt += ' echo "WARNING: output file $file not found!"\n'
1305	<	txt += ' fi\n'
1306	<	txt += 'done\n'
1524	<	txt += 'echo "Total Output dimension: $sum";\n'
1525	<	txt += 'limit='+str(limit)+';\n'
1526	<	txt += 'echo "OUTPUT FILES LIMIT SET TO: $limit";\n'
1527	<	txt += 'if [ $limit -lt $sum ]; then\n'
1528	<	txt += ' echo "WARNING: output files have to big size - something will be lost;"\n'
1529	<	txt += ' echo " checking the output file sizes..."\n'
1530	<	"""
1531	<	txt += ' dim=0;\n'
1532	<	txt += ' exclude=0;\n'
1533	<	txt += ' for files in '+str(allOutFiles)+' ; do\n'
1534	<	txt += ' sumTemp=0;\n'
1535	<	txt += ' for file2 in '+str(allOutFiles)+' ; do\n'
1536	<	txt += ' if [ $file != $file2 ]; then\n'
1537	<	txt += ' tt=`ls -gGrta $file2 \| awk \'{ print $3 }\';`\n'
1538	<	txt += ' sumTemp=`expr $sumTemp + $tt`;\n'
1539	<	txt += ' fi\n'
1540	<	txt += ' done\n'
1541	<	txt += ' if [ $sumTemp -lt $limit ]; then\n'
1542	<	txt += ' if [ $dim -lt $sumTemp ]; then\n'
1543	<	txt += ' dim=$sumTemp;\n'
1544	<	txt += ' exclude=$file;\n'
1545	<	txt += ' fi\n'
1546	<	txt += ' fi\n'
1547	<	txt += ' done\n'
1548	<	txt += ' echo "Dimension calculated: $dim"; echo "File to exclude: $exclude";\n'
1549	<	"""
1550	<	txt += ' tot=0;\n'
1551	<	txt += ' for file2 in '+str(allOutFiles)+' ; do\n'
1552	<	txt += ' tt=`ls -gGrta $file2 \| awk \'{ print $3 }\';`\n'
1553	<	txt += ' tot=`expr $tot + $tt`;\n'
1554	<	txt += ' if [ $limit -lt $tot ]; then\n'
1555	<	txt += ' tot=`expr $tot - $tt`;\n'
1556	<	txt += ' fileLast=$file;\n'
1557	<	txt += ' break;\n'
1558	<	txt += ' fi\n'
1559	<	txt += ' done\n'
1560	<	txt += ' echo "Dimension calculated: $tot"; echo "First file to exclude: $file";\n'
1561	<	txt += ' flag=0;\n'
1562	<	txt += ' for filess in '+str(allOutFiles)+' ; do\n'
1563	<	txt += ' if [ $fileLast = $filess ]; then\n'
1564	<	txt += ' flag=1;\n'
1565	<	txt += ' fi\n'
1566	<	txt += ' if [ $flag -eq 1 ]; then\n'
1567	<	txt += ' rm -f $filess;\n'
1568	<	txt += ' fi\n'
1569	<	txt += ' done\n'
1570	<	txt += ' ls -agGhrt;\n'
1571	<	txt += ' echo "WARNING: output files are too big in dimension: can not put in the output_sandbox.";\n'
1572	<	txt += ' echo "JOB_EXIT_STATUS = 70000";\n'
1573	<	txt += ' exit_status=70000;\n'
1574	<	txt += 'else'
1575	<	txt += ' echo "Total Output dimension $sum is fine.";\n'
1576	<	txt += 'fi\n'
1577	<	txt += 'echo "*****************************************"\n'
1578	<	txt += 'echo "* Ending output sandbox limit check *"\n'
1579	<	txt += 'echo "*****************************************"\n'
1292	>	stdout = 'CMSSW_$NJob.stdout'
1293	>	stderr = 'CMSSW_$NJob.stderr'
1294	>	if (self.return_data == 1):
1295	>	for file in (self.output_file+self.output_file_sandbox):
1296	>	listOutFiles.append(numberFile(file, '$NJob'))
1297	>	listOutFiles.append(stdout)
1298	>	listOutFiles.append(stderr)
1299	>	else:
1300	>	for file in (self.output_file_sandbox):
1301	>	listOutFiles.append(numberFile(file, '$NJob'))
1302	>	listOutFiles.append(stdout)
1303	>	listOutFiles.append(stderr)
1304	>	txt += 'echo "output files: '+string.join(listOutFiles,' ')+'"\n'
1305	>	txt += 'filesToCheck="'+string.join(listOutFiles,' ')+'"\n'
1306	>	txt += 'export filesToCheck\n'
1307		return txt

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing COMP/CRAB/python/cms_cmssw.py (file contents): Revision 1.128 by fanzago, Thu Oct 11 16:23:44 2007 UTC vs. Revision 1.211 by slacapra, Tue Jun 10 14:02:12 2008 UTC

Diff Legend

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.128 by fanzago, Thu Oct 11 16:23:44 2007 UTC vs.
Revision 1.211 by slacapra, Tue Jun 10 14:02:12 2008 UTC