ViewVC Help

View File | Revision Log | Show Annotations | Root Listing

root/cvsroot/COMP/CRAB/python/cms_cmssw.py

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.7 by gutsche, Tue Jun 13 20:43:00 2006 UTC vs.
Revision 1.116.2.8 by fanzago, Wed Oct 17 13:25:00 2007 UTC

#	Line 2 | Line 2 | from JobType import JobType
2		from crab_logger import Logger
3		from crab_exceptions import *
4		from crab_util import *
5	+	from BlackWhiteListParser import BlackWhiteListParser
6		import common
6	–	import PsetManipulator
7	–
8	–	import DBSInfo_EDM
9	–	import DataDiscovery_EDM
10	–	import DataLocation_EDM
7		import Scram
8
9	<	import os, string, re
9	>	import os, string, glob
10
11		class Cmssw(JobType):
12	<	def __init__(self, cfg_params):
12	>	def __init__(self, cfg_params, ncjobs):
13		JobType.__init__(self, 'CMSSW')
14		common.logger.debug(3,'CMSSW::__init__')
15
20	–	self.analisys_common_info = {}
21	–	# Marco.
16		self._params = {}
17		self.cfg_params = cfg_params
18
19	+	# init BlackWhiteListParser
20	+	self.blackWhiteListParser = BlackWhiteListParser(cfg_params)
21	+
22	+	try:
23	+	self.MaxTarBallSize = float(self.cfg_params['EDG.maxtarballsize'])
24	+	except KeyError:
25	+	self.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
32		self.scram = Scram.Scram(cfg_params)
28	–	scramArea = ''
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'
39	+	self.scriptName = 'CMSSW.sh'
40	+	self.pset = ''      #scrip use case Da
41	+	self.datasetPath = '' #scrip use case Da
42
43	+	# set FJR file name
44	+	self.fjrFileName = 'crab_fjr.xml'
45
46		self.version = self.scram.getSWVersion()
47	+
48	+	#
49	+	# Try to block creation in case of arch/version mismatch
50	+	#
51	+
52	+	a = string.split(self.version, "_")
53	+
54	+	if int(a[1]) == 1 and (int(a[2]) < 5 and self.executable_arch.find('slc4') == 0):
55	+	msg = "Error: CMS does not support %s with %s architecture"%(self.version, self.executable_arch)
56	+	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)
59	+	raise CrabException(msg)
60	+
61	+	common.taskDB.setDict('codeVersion',self.version)
62		self.setParam_('application', self.version)
37	–	common.analisys_common_info['sw_version'] = self.version
38	–	### FEDE
39	–	common.analisys_common_info['copy_input_data'] = 0
40	–	common.analisys_common_info['events_management'] = 1
63
64		### collect Data cards
65	+
66	+	## get DBS mode
67		try:
68	<	#   self.owner = cfg_params['CMSSW.owner']
69	<	#   log.debug(6, "CMSSW::CMSSW(): owner = "+self.owner)
70	<	#   self.dataset = cfg_params['CMSSW.dataset']
71	<	self.datasetPath = cfg_params['CMSSW.datasetpath']
72	<	log.debug(6, "CMSSW::CMSSW(): datasetPath = "+self.datasetPath)
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:
50	–	#    msg = "Error: owner and/or dataset not defined "
82		msg = "Error: datasetpath not defined "
83		raise CrabException(msg)
84
85		# ML monitoring
86		# split dataset path style: /PreProdR3Minbias/SIM/GEN-SIM
87	<	datasetpath_split = self.datasetPath.split("/")
88	<	self.setParam_('dataset', datasetpath_split[1])
89	<	self.setParam_('owner', datasetpath_split[-1])
90	<
91	<
92	<
87	>	if not self.datasetPath:
88	>	self.setParam_('dataset', 'None')
89	>	self.setParam_('owner', 'None')
90	>	else:
91	>	try:
92	>	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'])
107
108		self.dataTiers = []
64	–	#       try:
65	–	#           tmpDataTiers = string.split(cfg_params['CMSSW.data_tier'],',')
66	–	#           for tmp in tmpDataTiers:
67	–	#               tmp=string.strip(tmp)
68	–	#               self.dataTiers.append(tmp)
69	–	#               pass
70	–	#           pass
71	–	#       except KeyError:
72	–	#           pass
73	–	#       log.debug(6, "Cmssw::Cmssw(): dataTiers = "+str(self.dataTiers))
109
110		## now the application
111		try:
#	Line 89 | Line 124 | class Cmssw(JobType):
124		try:
125		self.pset = cfg_params['CMSSW.pset']
126		log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
127	<	if (not os.path.exists(self.pset)):
128	<	raise CrabException("User defined PSet file "+self.pset+" does not exist")
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:
133		raise CrabException("PSet file missing. Cannot run cmsRun ")
134
135		# output files
136	+	## stuff which must be returned always via sandbox
137	+	self.output_file_sandbox = []
138	+
139	+	# add fjr report by default via sandbox
140	+	self.output_file_sandbox.append(self.fjrFileName)
141	+
142	+	# other output files to be returned via sandbox or copied to SE
143		try:
144		self.output_file = []
100	–
145		tmp = cfg_params['CMSSW.output_file']
146		if tmp != '':
147		tmpOutFiles = string.split(cfg_params['CMSSW.output_file'],',')
#	Line 107 | Line 151 | class Cmssw(JobType):
151		self.output_file.append(tmp)
152		pass
153		else:
154	<	log.message("No output file defined: only stdout/err will be available")
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:
158	<	log.message("No output file defined: only stdout/err will be available")
158	>	log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
159		pass
160
161		# script_exe file as additional file in inputSandbox
162		try:
163	<	self.scriptExe = cfg_params['USER.script_exe']
164	<	self.additional_inbox_files.append(self.scriptExe)
163	>	self.scriptExe = cfg_params['USER.script_exe']
164	>	if self.scriptExe != '':
165	>	if not os.path.isfile(self.scriptExe):
166	>	msg ="ERROR. file "+self.scriptExe+" not found"
167	>	raise CrabException(msg)
168	>	self.additional_inbox_files.append(string.strip(self.scriptExe))
169		except KeyError:
170	<	pass
171	<	if self.scriptExe != '':
172	<	if os.path.isfile(self.scriptExe):
173	<	pass
174	<	else:
175	<	log.message("WARNING. file "+self.scriptExe+" not found")
176	<	sys.exit()
129	<
170	>	self.scriptExe = ''
171	>
172	>	#CarlosDaniele
173	>	if self.datasetPath == None and self.pset == None and self.scriptExe == '' :
174	>	msg ="Error. script_exe  not defined"
175	>	raise CrabException(msg)
176	>
177		## additional input files
178		try:
179	<	tmpAddFiles = string.split(cfg_params['CMSSW.additional_input_files'],',')
179	>	tmpAddFiles = string.split(cfg_params['USER.additional_input_files'],',')
180		for tmp in tmpAddFiles:
181	<	if not os.path.exists(tmp):
182	<	raise CrabException("Additional input file not found: "+tmp)
183	<	tmp=string.strip(tmp)
184	<	self.additional_inbox_files.append(tmp)
181	>	tmp = string.strip(tmp)
182	>	dirname = ''
183	>	if not tmp[0]=="/": dirname = "."
184	>	files = []
185	>	if string.find(tmp,"*")>-1:
186	>	files = glob.glob(os.path.join(dirname, tmp))
187	>	if len(files)==0:
188	>	raise CrabException("No additional input file found with this pattern: "+tmp)
189	>	else:
190	>	files.append(tmp)
191	>	for file in files:
192	>	if not os.path.exists(file):
193	>	raise CrabException("Additional input file not found: "+file)
194	>	pass
195	>	# fname = string.split(file, '/')[-1]
196	>	# storedFile = common.work_space.pathForTgz()+'share/'+fname
197	>	# shutil.copyfile(file, storedFile)
198	>	self.additional_inbox_files.append(string.strip(file))
199		pass
200		pass
201	+	common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
202		except KeyError:
203		pass
204
205	+	# files per job
206		try:
207	<	self.filesPerJob = int(cfg_params['CMSSW.files_per_jobs']) #Daniele
207	>	if (cfg_params['CMSSW.files_per_jobs']):
208	>	raise CrabException("files_per_jobs no longer supported.  Quitting.")
209		except KeyError:
210	<	self.filesPerJob = 1
210	>	pass
211
212	<	## Max event   will be total_number_of_events ???  Daniele
212	>	## Events per job
213		try:
214	<	self.maxEv = cfg_params['CMSSW.event_per_job']
214	>	self.eventsPerJob =int( cfg_params['CMSSW.events_per_job'])
215	>	self.selectEventsPerJob = 1
216		except KeyError:
217	<	self.maxEv = "-1"
218	<	##
217	>	self.eventsPerJob = -1
218	>	self.selectEventsPerJob = 0
219	>
220	>	## number of jobs
221	>	try:
222	>	self.theNumberOfJobs =int( cfg_params['CMSSW.number_of_jobs'])
223	>	self.selectNumberOfJobs = 1
224	>	except KeyError:
225	>	self.theNumberOfJobs = 0
226	>	self.selectNumberOfJobs = 0
227	>
228		try:
229		self.total_number_of_events = int(cfg_params['CMSSW.total_number_of_events'])
230	+	self.selectTotalNumberEvents = 1
231		except KeyError:
232	<	msg = 'Must define total_number_of_events'
233	<	raise CrabException(msg)
234	<
235	<	CEBlackList = []
232	>	self.total_number_of_events = 0
233	>	self.selectTotalNumberEvents = 0
234	>
235	>	if self.pset != None: #CarlosDaniele
236	>	if ( (self.selectTotalNumberEvents + self.selectEventsPerJob + self.selectNumberOfJobs) != 2 ):
237	>	msg = 'Must define exactly two of total_number_of_events, events_per_job, or number_of_jobs.'
238	>	raise CrabException(msg)
239	>	else:
240	>	if (self.selectNumberOfJobs == 0):
241	>	msg = 'Must specify  number_of_jobs.'
242	>	raise CrabException(msg)
243	>
244	>	## source seed for pythia
245		try:
246	<	tmpBad = string.split(cfg_params['EDG.ce_black_list'],',')
163	<	for tmp in tmpBad:
164	<	tmp=string.strip(tmp)
165	<	CEBlackList.append(tmp)
246	>	self.sourceSeed = int(cfg_params['CMSSW.pythia_seed'])
247		except KeyError:
248	<	pass
248	>	self.sourceSeed = None
249	>	common.logger.debug(5,"No seed given")
250
251	<	self.reCEBlackList=[]
252	<	for bad in CEBlackList:
171	<	self.reCEBlackList.append(re.compile( bad ))
172	<
173	<	common.logger.debug(5,'CEBlackList: '+str(CEBlackList))
174	<
175	<	CEWhiteList = []
176	<	try:
177	<	tmpGood = string.split(cfg_params['EDG.ce_white_list'],',')
178	<	for tmp in tmpGood:
179	<	tmp=string.strip(tmp)
180	<	CEWhiteList.append(tmp)
251	>	try:
252	>	self.sourceSeedVtx = int(cfg_params['CMSSW.vtx_seed'])
253		except KeyError:
254	<	pass
254	>	self.sourceSeedVtx = None
255	>	common.logger.debug(5,"No vertex seed given")
256
257	<	#print 'CEWhiteList: ',CEWhiteList
258	<	self.reCEWhiteList=[]
259	<	for Good in CEWhiteList:
260	<	self.reCEWhiteList.append(re.compile( Good ))
257	>	try:
258	>	self.sourceSeedG4 = int(cfg_params['CMSSW.g4_seed'])
259	>	except KeyError:
260	>	self.sourceSeedG4 = None
261	>	common.logger.debug(5,"No g4 sim hits seed given")
262
263	<	common.logger.debug(5,'CEWhiteList: '+str(CEWhiteList))
263	>	try:
264	>	self.sourceSeedMix = int(cfg_params['CMSSW.mix_seed'])
265	>	except KeyError:
266	>	self.sourceSeedMix = None
267	>	common.logger.debug(5,"No mix seed given")
268
269	<	self.PsetEdit = PsetManipulator.PsetManipulator(self.pset) #Daniele Pset
269	>	try:
270	>	self.firstRun = int(cfg_params['CMSSW.first_run'])
271	>	except KeyError:
272	>	self.firstRun = None
273	>	common.logger.debug(5,"No first run given")
274	>	if self.pset != None: #CarlosDaniele
275	>	ver = string.split(self.version,"_")
276	>	if (int(ver[1])>=1 and int(ver[2])>=5):
277	>	import PsetManipulator150 as pp
278	>	else:
279	>	import PsetManipulator as pp
280	>	PsetEdit = pp.PsetManipulator(self.pset) #Daniele Pset
281
282		#DBSDLS-start
283		## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
284		self.maxEvents=0  # max events available   ( --> check the requested nb. of evts in Creator.py)
285		self.DBSPaths={}  # all dbs paths requested ( --> input to the site local discovery script)
286	+	self.jobDestination=[]  # Site destination(s) for each job (list of lists)
287		## Perform the data location and discovery (based on DBS/DLS)
288	<	self.DataDiscoveryAndLocation(cfg_params)
288	>	## SL: Don't if NONE is specified as input (pythia use case)
289	>	blockSites = {}
290	>	if self.datasetPath:
291	>	blockSites = self.DataDiscoveryAndLocation(cfg_params)
292		#DBSDLS-end
293
294		self.tgzNameWithPath = self.getTarBall(self.executable)
295	+
296	+	## Select Splitting
297	+	if self.selectNoInput:
298	+	if self.pset == None: #CarlosDaniele
299	+	self.jobSplittingForScript()
300	+	else:
301	+	self.jobSplittingNoInput()
302	+	else:
303	+	self.jobSplittingByBlocks(blockSites)
304
305	<	self.jobSplitting()  #Daniele job Splitting
306	<	self.PsetEdit.maxEvent(self.maxEv) #Daniele
307	<	self.PsetEdit.inputModule("INPUT") #Daniele
308	<	self.PsetEdit.psetWriter(self.configFilename())
309	<
310	<
305	>	# modify Pset
306	>	if self.pset != None: #CarlosDaniele
307	>	try:
308	>	if (self.datasetPath): # standard job
309	>	# allow to processa a fraction of events in a file
310	>	PsetEdit.inputModule("INPUT")
311	>	PsetEdit.maxEvent("INPUTMAXEVENTS")
312	>	PsetEdit.skipEvent("INPUTSKIPEVENTS")
313	>	else:  # pythia like job
314	>	PsetEdit.maxEvent(self.eventsPerJob)
315	>	if (self.firstRun):
316	>	PsetEdit.pythiaFirstRun("INPUTFIRSTRUN")  #First Run
317	>	if (self.sourceSeed) :
318	>	PsetEdit.pythiaSeed("INPUT")
319	>	if (self.sourceSeedVtx) :
320	>	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)
331
332		def DataDiscoveryAndLocation(self, cfg_params):
333
334	+	import DataDiscovery
335	+	import DataDiscovery_DBS2
336	+	import DataLocation
337		common.logger.debug(10,"CMSSW::DataDiscoveryAndLocation()")
338
214	–	#datasetPath = "/"+self.owner+"/"+self.dataTiers[0]+"/"+self.dataset
215	–
339		datasetPath=self.datasetPath
340
218	–	## TODO
219	–	dataTiersList = ""
220	–	dataTiers = dataTiersList.split(',')
221	–
341		## Contact the DBS
342	+	common.logger.message("Contacting Data Discovery Services ...")
343		try:
344	<	self.pubdata=DataDiscovery_EDM.DataDiscovery_EDM(datasetPath, dataTiers, cfg_params)
344	>
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)
349		self.pubdata.fetchDBSInfo()
350
351	<	except DataDiscovery_EDM.NotExistingDatasetError, ex :
351	>	except DataDiscovery.NotExistingDatasetError, ex :
352		msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
353		raise CrabException(msg)
354	<
231	<	except DataDiscovery_EDM.NoDataTierinProvenanceError, ex :
354	>	except DataDiscovery.NoDataTierinProvenanceError, ex :
355		msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
356		raise CrabException(msg)
357	<	except DataDiscovery_EDM.DataDiscoveryError, ex:
358	<	msg = 'ERROR ***: failed Data Discovery in DBS  %s'%ex.getErrorMessage()
357	>	except DataDiscovery.DataDiscoveryError, ex:
358	>	msg = 'ERROR ***: failed Data Discovery in DBS :  %s'%ex.getErrorMessage()
359	>	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)
369
370	<	## get list of all required data in the form of dbs paths  (dbs path = /dataset/datatier/owner)
371	<	## self.DBSPaths=self.pubdata.getDBSPaths()
372	<	common.logger.message("Required data are :"+self.datasetPath)
241	<
242	<	filesbyblock=self.pubdata.getFiles()
243	<	self.AllInputFiles=filesbyblock.values()
244	<	self.files = self.AllInputFiles
245	<
246	<	## TEMP
247	<	#    self.filesTmp = filesbyblock.values()
248	<	#    self.files = []
249	<	#    locPath='rfio:cmsbose2.bo.infn.it:/flatfiles/SE00/cms/fanfani/ProdTest/'
250	<	#    locPath=''
251	<	#    tmp = []
252	<	#    for file in self.filesTmp[0]:
253	<	#        tmp.append(locPath+file)
254	<	#    self.files.append(tmp)
255	<	## END TEMP
370	>	self.filesbyblock=self.pubdata.getFiles()
371	>	self.eventsbyblock=self.pubdata.getEventsPerBlock()
372	>	self.eventsbyfile=self.pubdata.getEventsPerFile()
373
374		## get max number of events
258	–	#common.logger.debug(10,"number of events for primary fileblocks %i"%self.pubdata.getMaxEvents())
375		self.maxEvents=self.pubdata.getMaxEvents() ##  self.maxEvents used in Creator.py
260	–	common.logger.message("\nThe number of available events is %s"%self.maxEvents)
376
377		## Contact the DLS and build a list of sites hosting the fileblocks
378		try:
379	<	dataloc=DataLocation_EDM.DataLocation_EDM(filesbyblock.keys(),cfg_params)
379	>	dataloc=DataLocation.DataLocation(self.filesbyblock.keys(),cfg_params)
380		dataloc.fetchDLSInfo()
381	<	except DataLocation_EDM.DataLocationError , ex:
381	>	except DataLocation.DataLocationError , ex:
382		msg = 'ERROR ***: failed Data Location in DLS \n %s '%ex.getErrorMessage()
383		raise CrabException(msg)
384
270	–	allsites=dataloc.getSites()
271	–	common.logger.debug(5,"sites are %s"%allsites)
272	–	sites=self.checkBlackList(allsites)
273	–	common.logger.debug(5,"sites are (after black list) %s"%sites)
274	–	sites=self.checkWhiteList(sites)
275	–	common.logger.debug(5,"sites are (after white list) %s"%sites)
276	–
277	–	if len(sites)==0:
278	–	msg = 'No sites hosting all the needed data! Exiting... '
279	–	raise CrabException(msg)
385
386	<	common.logger.message("List of Sites hosting the data : "+str(sites))
387	<	common.logger.debug(6, "List of Sites: "+str(sites))
388	<	common.analisys_common_info['sites']=sites    ## used in SchedulerEdg.py in createSchScript
389	<	self.setParam_('TargetCE', ','.join(sites))
390	<	return
386	>	sites = dataloc.getSites()
387	>	allSites = []
388	>	listSites = sites.values()
389	>	for listSite in listSites:
390	>	for oneSite in listSite:
391	>	allSites.append(oneSite)
392	>	allSites = self.uniquelist(allSites)
393	>
394	>	# screen output
395	>	common.logger.message("Requested dataset: " + datasetPath + " has " + str(self.maxEvents) + " events in " + str(len(self.filesbyblock.keys())) + " blocks.\n")
396	>
397	>	return sites
398
399	<	def jobSplitting(self):
399	>	def jobSplittingByBlocks(self, blockSites):
400		"""
401	<	first implemntation for job splitting
402	<	"""
403	<	#  print 'eventi totali '+str(self.maxEvents)
404	<	#  print 'eventi totali richiesti dallo user '+str(self.total_number_of_events)
405	<	#print 'files per job '+str(self.filesPerJob)
406	<	common.logger.message('Required '+str(self.filesPerJob)+' files per job ')
407	<	common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
408	<
409	<	## TODO: SL need to have (from DBS) a detailed list of how many events per each file
410	<	n_tot_files = (len(self.files[0]))
411	<	## SL: this is wrong if the files have different number of events
412	<	evPerFile = int(self.maxEvents)/n_tot_files
413	<
414	<	common.logger.debug(5,'Events per File '+str(evPerFile))
415	<
416	<	## if asked to process all events, do it
417	<	if self.total_number_of_events == -1:
418	<	self.total_number_of_events=self.maxEvents
419	<	self.total_number_of_jobs = int(n_tot_files)*1/int(self.filesPerJob)
420	<	common.logger.message(str(self.total_number_of_jobs)+' jobs will be created for all available events '+str(self.total_number_of_events)+' events')
421	<
401	>	Perform job splitting. Jobs run over an integer number of files
402	>	and no more than one block.
403	>	ARGUMENT: blockSites: dictionary with blocks as keys and list of host sites as values
404	>	REQUIRES: self.selectTotalNumberEvents, self.selectEventsPerJob, self.selectNumberofJobs,
405	>	self.total_number_of_events, self.eventsPerJob, self.theNumberOfJobs,
406	>	self.maxEvents, self.filesbyblock
407	>	SETS: self.jobDestination - Site destination(s) for each job (a list of lists)
408	>	self.total_number_of_jobs - Total # of jobs
409	>	self.list_of_args - File(s) job will run on (a list of lists)
410	>	"""
411	>
412	>	# ---- Handle the possible job splitting configurations ---- #
413	>	if (self.selectTotalNumberEvents):
414	>	totalEventsRequested = self.total_number_of_events
415	>	if (self.selectEventsPerJob):
416	>	eventsPerJobRequested = self.eventsPerJob
417	>	if (self.selectNumberOfJobs):
418	>	totalEventsRequested = self.theNumberOfJobs * self.eventsPerJob
419	>
420	>	# If user requested all the events in the dataset
421	>	if (totalEventsRequested == -1):
422	>	eventsRemaining=self.maxEvents
423	>	# If user requested more events than are in the dataset
424	>	elif (totalEventsRequested > self.maxEvents):
425	>	eventsRemaining = self.maxEvents
426	>	common.logger.message("Requested "+str(self.total_number_of_events)+ " events, but only "+str(self.maxEvents)+" events are available.")
427	>	# If user requested less events than are in the dataset
428		else:
429	<	self.total_number_of_files = int(self.total_number_of_events/evPerFile)
430	<	## SL: if ask for less event than what is computed to be available on a
431	<	##     file, process the first file anyhow.
432	<	if self.total_number_of_files == 0:
433	<	self.total_number_of_files = self.total_number_of_files + 1
429	>	eventsRemaining = totalEventsRequested
430	>
431	>	# If user requested more events per job than are in the dataset
432	>	if (self.selectEventsPerJob and eventsPerJobRequested > self.maxEvents):
433	>	eventsPerJobRequested = self.maxEvents
434	>
435	>	# For user info at end
436	>	totalEventCount = 0
437	>
438	>	if (self.selectTotalNumberEvents and self.selectNumberOfJobs):
439	>	eventsPerJobRequested = int(eventsRemaining/self.theNumberOfJobs)
440	>
441	>	if (self.selectNumberOfJobs):
442	>	common.logger.message("May not create the exact number_of_jobs requested.")
443
444	<	common.logger.debug(5,'N files  '+str(self.total_number_of_files))
444	>	if ( self.ncjobs == 'all' ) :
445	>	totalNumberOfJobs = 999999999
446	>	else :
447	>	totalNumberOfJobs = self.ncjobs
448	>
449	>
450	>	blocks = blockSites.keys()
451	>	blockCount = 0
452	>	# Backup variable in case self.maxEvents counted events in a non-included block
453	>	numBlocksInDataset = len(blocks)
454	>
455	>	jobCount = 0
456	>	list_of_lists = []
457
458	<	check = 0
458	>	# list tracking which jobs are in which jobs belong to which block
459	>	jobsOfBlock = {}
460	>
461	>	# ---- Iterate over the blocks in the dataset until ---- #
462	>	# ---- we've met the requested total # of events    ---- #
463	>	while ( (eventsRemaining > 0) and (blockCount < numBlocksInDataset) and (jobCount < totalNumberOfJobs)):
464	>	block = blocks[blockCount]
465	>	blockCount += 1
466	>	if block not in jobsOfBlock.keys() :
467	>	jobsOfBlock[block] = []
468	>
469	>	if self.eventsbyblock.has_key(block) :
470	>	numEventsInBlock = self.eventsbyblock[block]
471	>	common.logger.debug(5,'Events in Block File '+str(numEventsInBlock))
472	>
473	>	files = self.filesbyblock[block]
474	>	numFilesInBlock = len(files)
475	>	if (numFilesInBlock <= 0):
476	>	continue
477	>	fileCount = 0
478	>
479	>	# ---- New block => New job ---- #
480	>	parString = "\\{"
481	>	# counter for number of events in files currently worked on
482	>	filesEventCount = 0
483	>	# flag if next while loop should touch new file
484	>	newFile = 1
485	>	# job event counter
486	>	jobSkipEventCount = 0
487
488	<	## Compute the number of jobs
489	<	#self.total_number_of_jobs = int(n_tot_files)*1/int(self.filesPerJob)
490	<	self.total_number_of_jobs = int(self.total_number_of_files/self.filesPerJob)
491	<	common.logger.debug(5,'N jobs  '+str(self.total_number_of_jobs))
488	>	# ---- Iterate over the files in the block until we've met the requested ---- #
489	>	# ---- total # of events or we've gone over all the files in this block  ---- #
490	>	while ( (eventsRemaining > 0) and (fileCount < numFilesInBlock) and (jobCount < totalNumberOfJobs) ):
491	>	file = files[fileCount]
492	>	if newFile :
493	>	try:
494	>	numEventsInFile = self.eventsbyfile[file]
495	>	common.logger.debug(6, "File "+str(file)+" has "+str(numEventsInFile)+" events")
496	>	# increase filesEventCount
497	>	filesEventCount += numEventsInFile
498	>	# Add file to current job
499	>	parString += '\\\"' + file + '\\\"\,'
500	>	newFile = 0
501	>	except KeyError:
502	>	common.logger.message("File "+str(file)+" has unknown number of events: skipping")
503	>
504	>
505	>	# if less events in file remain than eventsPerJobRequested
506	>	if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested ) :
507	>	# if last file in block
508	>	if ( fileCount == numFilesInBlock-1 ) :
509	>	# end job using last file, use remaining events in block
510	>	# close job and touch new file
511	>	fullString = parString[:-2]
512	>	fullString += '\\}'
513	>	list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
514	>	common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
515	>	self.jobDestination.append(blockSites[block])
516	>	common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
517	>	# fill jobs of block dictionary
518	>	jobsOfBlock[block].append(jobCount+1)
519	>	# reset counter
520	>	jobCount = jobCount + 1
521	>	totalEventCount = totalEventCount + filesEventCount - jobSkipEventCount
522	>	eventsRemaining = eventsRemaining - filesEventCount + jobSkipEventCount
523	>	jobSkipEventCount = 0
524	>	# reset file
525	>	parString = "\\{"
526	>	filesEventCount = 0
527	>	newFile = 1
528	>	fileCount += 1
529	>	else :
530	>	# go to next file
531	>	newFile = 1
532	>	fileCount += 1
533	>	# if events in file equal to eventsPerJobRequested
534	>	elif ( filesEventCount - jobSkipEventCount == eventsPerJobRequested ) :
535	>	# close job and touch new file
536	>	fullString = parString[:-2]
537	>	fullString += '\\}'
538	>	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
539	>	common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
540	>	self.jobDestination.append(blockSites[block])
541	>	common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
542	>	jobsOfBlock[block].append(jobCount+1)
543	>	# reset counter
544	>	jobCount = jobCount + 1
545	>	totalEventCount = totalEventCount + eventsPerJobRequested
546	>	eventsRemaining = eventsRemaining - eventsPerJobRequested
547	>	jobSkipEventCount = 0
548	>	# reset file
549	>	parString = "\\{"
550	>	filesEventCount = 0
551	>	newFile = 1
552	>	fileCount += 1
553	>
554	>	# if more events in file remain than eventsPerJobRequested
555	>	else :
556	>	# close job but don't touch new file
557	>	fullString = parString[:-2]
558	>	fullString += '\\}'
559	>	list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
560	>	common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
561	>	self.jobDestination.append(blockSites[block])
562	>	common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
563	>	jobsOfBlock[block].append(jobCount+1)
564	>	# increase counter
565	>	jobCount = jobCount + 1
566	>	totalEventCount = totalEventCount + eventsPerJobRequested
567	>	eventsRemaining = eventsRemaining - eventsPerJobRequested
568	>	# calculate skip events for last file
569	>	# use filesEventCount (contains several files), jobSkipEventCount and eventsPerJobRequest
570	>	jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
571	>	# remove all but the last file
572	>	filesEventCount = self.eventsbyfile[file]
573	>	parString = "\\{"
574	>	parString += '\\\"' + file + '\\\"\,'
575	>	pass # END if
576	>	pass # END while (iterate over files in the block)
577	>	pass # END while (iterate over blocks in the dataset)
578	>	self.ncjobs = self.total_number_of_jobs = jobCount
579	>	if (eventsRemaining > 0 and jobCount < totalNumberOfJobs ):
580	>	common.logger.message("Could not run on all requested events because some blocks not hosted at allowed sites.")
581	>	common.logger.message(str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
582	>
583	>	# screen output
584	>	screenOutput = "List of jobs and available destination sites:\n\n"
585	>
586	>	# keep trace of block with no sites to print a warning at the end
587	>	noSiteBlock = []
588	>	bloskNoSite = []
589	>
590	>	blockCounter = 0
591	>	for block in blocks:
592	>	if block in jobsOfBlock.keys() :
593	>	blockCounter += 1
594	>	screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
595	>	if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0:
596	>	noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
597	>	bloskNoSite.append( blockCounter )
598	>
599	>	common.logger.message(screenOutput)
600	>	if len(noSiteBlock) > 0 and len(bloskNoSite) > 0:
601	>	msg = 'WARNING: No sites are hosting any part of data for block:\n                '
602	>	virgola = ""
603	>	if len(bloskNoSite) > 1:
604	>	virgola = ","
605	>	for block in bloskNoSite:
606	>	msg += ' ' + str(block) + virgola
607	>	msg += '\n               Related jobs:\n                 '
608	>	virgola = ""
609	>	if len(noSiteBlock) > 1:
610	>	virgola = ","
611	>	for range_jobs in noSiteBlock:
612	>	msg += str(range_jobs) + virgola
613	>	msg += '\n               will not be submitted and this block of data can not be analyzed!\n'
614	>	common.logger.message(msg)
615	>
616	>	self.list_of_args = list_of_lists
617	>	return
618	>
619	>	def jobSplittingNoInput(self):
620	>	"""
621	>	Perform job splitting based on number of event per job
622	>	"""
623	>	common.logger.debug(5,'Splitting per events')
624	>
625	>	if (self.selectEventsPerJob):
626	>	common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
627	>	if (self.selectNumberOfJobs):
628	>	common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
629	>	if (self.selectTotalNumberEvents):
630	>	common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
631
632	<	## is there any remainder?
633	<	check = int(self.total_number_of_files) - (int(self.total_number_of_jobs)*self.filesPerJob)
632	>	if (self.total_number_of_events < 0):
633	>	msg='Cannot split jobs per Events with "-1" as total number of events'
634	>	raise CrabException(msg)
635
636	<	common.logger.debug(5,'Check  '+str(check))
636	>	if (self.selectEventsPerJob):
637	>	if (self.selectTotalNumberEvents):
638	>	self.total_number_of_jobs = int(self.total_number_of_events/self.eventsPerJob)
639	>	elif(self.selectNumberOfJobs) :
640	>	self.total_number_of_jobs =self.theNumberOfJobs
641	>	self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
642	>
643	>	elif (self.selectNumberOfJobs) :
644	>	self.total_number_of_jobs = self.theNumberOfJobs
645	>	self.eventsPerJob = int(self.total_number_of_events/self.total_number_of_jobs)
646	>
647	>	common.logger.debug(5,'N jobs  '+str(self.total_number_of_jobs))
648
649	<	if check > 0:
650	<	self.total_number_of_jobs =  self.total_number_of_jobs + 1
333	<	common.logger.message('Warning: last job will be created with '+str(check)+' files')
649	>	# is there any remainder?
650	>	check = int(self.total_number_of_events) - (int(self.total_number_of_jobs)*self.eventsPerJob)
651
652	<	common.logger.message(str(self.total_number_of_jobs)+' jobs will be created for a total of '+str((self.total_number_of_jobs-1)*self.filesPerJob*evPerFile + check*evPerFile)+' events')
652	>	common.logger.debug(5,'Check  '+str(check))
653	>
654	>	common.logger.message(str(self.total_number_of_jobs)+' jobs can be created, each for '+str(self.eventsPerJob)+' for a total of '+str(self.total_number_of_jobs*self.eventsPerJob)+' events')
655	>	if check > 0:
656	>	common.logger.message('Warning: asked '+str(self.total_number_of_events)+' but can do only '+str(int(self.total_number_of_jobs)*self.eventsPerJob))
657	>
658	>	# argument is seed number.$i
659	>	self.list_of_args = []
660	>	for i in range(self.total_number_of_jobs):
661	>	## Since there is no input, any site is good
662	>	# self.jobDestination.append(["Any"])
663	>	self.jobDestination.append([""]) #must be empty to write correctly the xml
664	>	args=[]
665	>	if (self.firstRun):
666	>	## pythia first run
667	>	#self.list_of_args.append([(str(self.firstRun)+str(i))])
668	>	args.append(str(self.firstRun)+str(i))
669	>	else:
670	>	## no first run
671	>	#self.list_of_args.append([str(i)])
672	>	args.append(str(i))
673	>	if (self.sourceSeed):
674	>	args.append(str(self.sourceSeed)+str(i))
675	>	if (self.sourceSeedVtx):
676	>	## + vtx random seed
677	>	args.append(str(self.sourceSeedVtx)+str(i))
678	>	if (self.sourceSeedG4):
679	>	## + G4 random seed
680	>	args.append(str(self.sourceSeedG4)+str(i))
681	>	if (self.sourceSeedMix):
682	>	## + Mix random seed
683	>	args.append(str(self.sourceSeedMix)+str(i))
684	>	pass
685		pass
686	+	self.list_of_args.append(args)
687	+	pass
688	+
689	+	# print self.list_of_args
690
691	<	list_of_lists = []
339	<	for i in xrange(0, int(n_tot_files), self.filesPerJob):
340	<	list_of_lists.append(self.files[0][i: i+self.filesPerJob])
691	>	return
692
693	<	self.list_of_files = list_of_lists
694	<
693	>
694	>	def jobSplittingForScript(self):#CarlosDaniele
695	>	"""
696	>	Perform job splitting based on number of job
697	>	"""
698	>	common.logger.debug(5,'Splitting per job')
699	>	common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
700	>
701	>	self.total_number_of_jobs = self.theNumberOfJobs
702	>
703	>	common.logger.debug(5,'N jobs  '+str(self.total_number_of_jobs))
704	>
705	>	common.logger.message(str(self.total_number_of_jobs)+' jobs can be created')
706	>
707	>	# argument is seed number.$i
708	>	self.list_of_args = []
709	>	for i in range(self.total_number_of_jobs):
710	>	## Since there is no input, any site is good
711	>	# self.jobDestination.append(["Any"])
712	>	self.jobDestination.append([""])
713	>	## no random seed
714	>	self.list_of_args.append([str(i)])
715		return
716
717		def split(self, jobParams):
#	Line 348 | Line 719 | class Cmssw(JobType):
719		common.jobDB.load()
720		#### Fabio
721		njobs = self.total_number_of_jobs
722	<	filelist = self.list_of_files
722	>	arglist = self.list_of_args
723		# create the empty structure
724		for i in range(njobs):
725		jobParams.append("")
726
727		for job in range(njobs):
728	<	jobParams[job] = filelist[job]
728	>	jobParams[job] = arglist[job]
729	>	# print str(arglist[job])
730	>	# print jobParams[job]
731		common.jobDB.setArguments(job, jobParams[job])
732	+	common.logger.debug(5,"Job "+str(job)+" Destination: "+str(self.jobDestination[job]))
733	+	common.jobDB.setDestination(job, self.jobDestination[job])
734
735		common.jobDB.save()
736		return
737
738		def getJobTypeArguments(self, nj, sched):
739	<	params = common.jobDB.arguments(nj)
740	<	#print params
741	<	parString = "\\{"
742	<
368	<	for i in range(len(params) - 1):
369	<	parString += '\\\"' + params[i] + '\\\"\,'
370	<
371	<	parString += '\\\"' + params[len(params) - 1] + '\\\"\\}'
372	<	return parString
739	>	result = ''
740	>	for i in common.jobDB.arguments(nj):
741	>	result=result+str(i)+" "
742	>	return result
743
744		def numberOfJobs(self):
745		# Fabio
376	–
746		return self.total_number_of_jobs
378	–
379	–
380	–
381	–	def checkBlackList(self, allSites):
382	–	if len(self.reCEBlackList)==0: return allSites
383	–	sites = []
384	–	for site in allSites:
385	–	common.logger.debug(10,'Site '+site)
386	–	good=1
387	–	for re in self.reCEBlackList:
388	–	if re.search(site):
389	–	common.logger.message('CE in black list, skipping site '+site)
390	–	good=0
391	–	pass
392	–	if good: sites.append(site)
393	–	if len(sites) == 0:
394	–	common.logger.debug(3,"No sites found after BlackList")
395	–	return sites
396	–
397	–	def checkWhiteList(self, allSites):
398	–
399	–	if len(self.reCEWhiteList)==0: return allSites
400	–	sites = []
401	–	for site in allSites:
402	–	good=0
403	–	for re in self.reCEWhiteList:
404	–	if re.search(site):
405	–	common.logger.debug(5,'CE in white list, adding site '+site)
406	–	good=1
407	–	if not good: continue
408	–	sites.append(site)
409	–	if len(sites) == 0:
410	–	common.logger.message("No sites found after WhiteList\n")
411	–	else:
412	–	common.logger.debug(5,"Selected sites via WhiteList are "+str(sites)+"\n")
413	–	return sites
747
748		def getTarBall(self, exe):
749		"""
#	Line 418 | Line 751 | class Cmssw(JobType):
751		"""
752
753		# if it exist, just return it
754	<	self.tgzNameWithPath = common.work_space.shareDir()+self.tgz_name
754	>	#
755	>	# Marco. Let's start to use relative path for Boss XML files
756	>	#
757	>	self.tgzNameWithPath = common.work_space.pathForTgz()+'share/'+self.tgz_name
758		if os.path.exists(self.tgzNameWithPath):
759		return self.tgzNameWithPath
760
#	Line 432 | Line 768 | class Cmssw(JobType):
768		# First of all declare the user Scram area
769		swArea = self.scram.getSWArea_()
770		#print "swArea = ", swArea
771	<	swVersion = self.scram.getSWVersion()
772	<	#print "swVersion = ", swVersion
771	>	# swVersion = self.scram.getSWVersion()
772	>	# print "swVersion = ", swVersion
773		swReleaseTop = self.scram.getReleaseTop_()
774		#print "swReleaseTop = ", swReleaseTop
775
#	Line 441 | Line 777 | class Cmssw(JobType):
777		if swReleaseTop == '' or swArea == swReleaseTop:
778		return
779
780	<	filesToBeTarred = []
781	<	## First find the executable
782	<	if (self.executable != ''):
783	<	exeWithPath = self.scram.findFile_(executable)
784	<	#           print exeWithPath
785	<	if ( not exeWithPath ):
786	<	raise CrabException('User executable '+executable+' not found')
787	<
788	<	## then check if it's private or not
789	<	if exeWithPath.find(swReleaseTop) == -1:
790	<	# the exe is private, so we must ship
791	<	common.logger.debug(5,"Exe "+exeWithPath+" to be tarred")
792	<	path = swArea+'/'
793	<	exe = string.replace(exeWithPath, path,'')
794	<	filesToBeTarred.append(exe)
795	<	pass
796	<	else:
797	<	# the exe is from release, we'll find it on WN
798	<	pass
799	<
800	<	## Now get the libraries: only those in local working area
801	<	libDir = 'lib'
802	<	lib = swArea+'/' +libDir
803	<	common.logger.debug(5,"lib "+lib+" to be tarred")
804	<	if os.path.exists(lib):
805	<	filesToBeTarred.append(libDir)
806	<
807	<	## Now check if module dir is present
808	<	moduleDir = 'module'
809	<	if os.path.isdir(swArea+'/'+moduleDir):
810	<	filesToBeTarred.append(moduleDir)
811	<
812	<	## Now check if the Data dir is present
813	<	dataDir = 'src/Data/'
814	<	if os.path.isdir(swArea+'/'+dataDir):
815	<	filesToBeTarred.append(dataDir)
816	<
817	<	## Create the tar-ball
818	<	if len(filesToBeTarred)>0:
819	<	cwd = os.getcwd()
820	<	os.chdir(swArea)
821	<	tarcmd = 'tar zcvf ' + self.tgzNameWithPath + ' '
822	<	for line in filesToBeTarred:
823	<	tarcmd = tarcmd + line + ' '
824	<	cout = runCommand(tarcmd)
825	<	if not cout:
826	<	raise CrabException('Could not create tar-ball')
827	<	os.chdir(cwd)
828	<	else:
829	<	common.logger.debug(5,"No files to be to be tarred")
780	>	import tarfile
781	>	try: # create tar ball
782	>	tar = tarfile.open(self.tgzNameWithPath, "w:gz")
783	>	## First find the executable
784	>	if (self.executable != ''):
785	>	exeWithPath = self.scram.findFile_(executable)
786	>	if ( not exeWithPath ):
787	>	raise CrabException('User executable '+executable+' not found')
788	>
789	>	## then check if it's private or not
790	>	if exeWithPath.find(swReleaseTop) == -1:
791	>	# the exe is private, so we must ship
792	>	common.logger.debug(5,"Exe "+exeWithPath+" to be tarred")
793	>	path = swArea+'/'
794	>	# distinguish case when script is in user project area or given by full path somewhere else
795	>	if exeWithPath.find(path) >= 0 :
796	>	exe = string.replace(exeWithPath, path,'')
797	>	tar.add(path+exe,exe)
798	>	else :
799	>	tar.add(exeWithPath,os.path.basename(executable))
800	>	pass
801	>	else:
802	>	# the exe is from release, we'll find it on WN
803	>	pass
804	>
805	>	## Now get the libraries: only those in local working area
806	>	libDir = 'lib'
807	>	lib = swArea+'/' +libDir
808	>	common.logger.debug(5,"lib "+lib+" to be tarred")
809	>	if os.path.exists(lib):
810	>	tar.add(lib,libDir)
811	>
812	>	## Now check if module dir is present
813	>	moduleDir = 'module'
814	>	module = swArea + '/' + moduleDir
815	>	if os.path.isdir(module):
816	>	tar.add(module,moduleDir)
817	>
818	>	## Now check if any data dir(s) is present
819	>	swAreaLen=len(swArea)
820	>	for root, dirs, files in os.walk(swArea):
821	>	if "data" in dirs:
822	>	common.logger.debug(5,"data "+root+"/data"+" to be tarred")
823	>	tar.add(root+"/data",root[swAreaLen:]+"/data")
824	>
825	>	## Add ProdAgent dir to tar
826	>	paDir = 'ProdAgentApi'
827	>	pa = os.environ['CRABDIR'] + '/' + 'ProdAgentApi'
828	>	if os.path.isdir(pa):
829	>	tar.add(pa,paDir)
830	>
831	>	### FEDE FOR DBS PUBLICATION
832	>	## Add PRODCOMMON dir to tar
833	>	prodcommonDir = 'ProdCommon'
834	>	prodcommonPath = os.environ['CRABDIR'] + '/' + 'ProdCommon'
835	>	if os.path.isdir(prodcommonPath):
836	>	tar.add(prodcommonPath,prodcommonDir)
837	>	#############################
838	>
839	>	common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
840	>	tar.close()
841	>	except :
842	>	raise CrabException('Could not create tar-ball')
843	>
844	>	## check for tarball size
845	>	tarballinfo = os.stat(self.tgzNameWithPath)
846	>	if ( tarballinfo.st_size > self.MaxTarBallSize*1024*1024 ) :
847	>	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.')
848	>
849	>	## create tar-ball with ML stuff
850	>	self.MLtgzfile =  common.work_space.pathForTgz()+'share/MLfiles.tgz'
851	>	try:
852	>	tar = tarfile.open(self.MLtgzfile, "w:gz")
853	>	path=os.environ['CRABDIR'] + '/python/'
854	>	for file in ['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py', 'parseCrabFjr.py']:
855	>	tar.add(path+file,file)
856	>	common.logger.debug(5,"Files added to "+self.MLtgzfile+" : "+str(tar.getnames()))
857	>	tar.close()
858	>	except :
859	>	raise CrabException('Could not create ML files tar-ball')
860
861		return
862
863	+	def additionalInputFileTgz(self):
864	+	"""
865	+	Put all additional files into a tar ball and return its name
866	+	"""
867	+	import tarfile
868	+	tarName=  common.work_space.pathForTgz()+'share/'+self.additional_tgz_name
869	+	tar = tarfile.open(tarName, "w:gz")
870	+	for file in self.additional_inbox_files:
871	+	tar.add(file,string.split(file,'/')[-1])
872	+	common.logger.debug(5,"Files added to "+self.additional_tgz_name+" : "+str(tar.getnames()))
873	+	tar.close()
874	+	return tarName
875	+
876		def wsSetupEnvironment(self, nj):
877		"""
878		Returns part of a job script which prepares
#	Line 505 | Line 884 | class Cmssw(JobType):
884		## OLI_Daniele at this level  middleware already known
885
886		txt += 'if [ $middleware == LCG ]; then \n'
887	+	txt += '    echo "### First set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
888	+	txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
889	+	txt += '    export BUILD_ARCH='+self.executable_arch+'\n'
890		txt += self.wsSetupCMSLCGEnvironment_()
891		txt += 'elif [ $middleware == OSG ]; then\n'
892	<	txt += '    time=`date -u +"%s"`\n'
893	<	txt += '    WORKING_DIR=$OSG_WN_TMP/cms_$time\n'
512	<	txt += '    echo "Creating working directory: $WORKING_DIR"\n'
513	<	txt += '    /bin/mkdir -p $WORKING_DIR\n'
892	>	txt += '    WORKING_DIR=`/bin/mktemp  -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
893	>	txt += '    echo "Created working directory: $WORKING_DIR"\n'
894		txt += '    if [ ! -d $WORKING_DIR ] ;then\n'
895		txt += '        echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
896	<	txt += '        echo "JOB_EXIT_STATUS = 10016"\n'
897	<	txt += '        echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
898	<	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
896	>	txt += '    echo "JOB_EXIT_STATUS = 10016"\n'
897	>	txt += '    echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
898	>	txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
899	>	txt += '        rm -f $RUNTIME_AREA/$repo \n'
900	>	txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
901	>	txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
902		txt += '        exit 1\n'
903		txt += '    fi\n'
904		txt += '\n'
905		txt += '    echo "Change to working directory: $WORKING_DIR"\n'
906		txt += '    cd $WORKING_DIR\n'
907		txt += self.wsSetupCMSOSGEnvironment_()
908	+	txt += '    echo "### Set SCRAM ARCH to ' + self.executable_arch + ' ###"\n'
909	+	txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
910		txt += 'fi\n'
911
912		# Prepare JobType-specific part
#	Line 535 | Line 920 | class Cmssw(JobType):
920		txt += '   echo "JOB_EXIT_STATUS = 10034"\n'
921		txt += '   echo "JobExitCode=10034" | tee -a $RUNTIME_AREA/$repo\n'
922		txt += '   dumpStatus $RUNTIME_AREA/$repo\n'
923	+	txt += '   rm -f $RUNTIME_AREA/$repo \n'
924	+	txt += '   echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
925	+	txt += '   echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
926		## OLI_Daniele
927		txt += '    if [ $middleware == OSG ]; then \n'
928		txt += '        echo "Remove working directory: $WORKING_DIR"\n'
929		txt += '        cd $RUNTIME_AREA\n'
930		txt += '        /bin/rm -rf $WORKING_DIR\n'
931		txt += '        if [ -d $WORKING_DIR ] ;then\n'
932	<	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'
933	<	txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
934	<	txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
935	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
932	>	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'
933	>	txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
934	>	txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
935	>	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
936	>	txt += '            rm -f $RUNTIME_AREA/$repo \n'
937	>	txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
938	>	txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
939		txt += '        fi\n'
940		txt += '    fi \n'
941		txt += '   exit 1 \n'
942		txt += 'fi \n'
943		txt += 'echo "CMSSW_VERSION =  '+self.version+'"\n'
944		txt += 'cd '+self.version+'\n'
945	+	########## FEDE FOR DBS2 ######################
946	+	txt += 'SOFTWARE_DIR=`pwd`\n'
947	+	txt += 'echo SOFTWARE_DIR=$SOFTWARE_DIR \n'
948	+	###############################################
949		### needed grep for bug in scramv1 ###
950	+	txt += scram+' runtime -sh\n'
951		txt += 'eval `'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME`\n'
952	+	txt += 'echo $PATH\n'
953
954		# Handle the arguments:
955		txt += "\n"
956		txt += "## number of arguments (first argument always jobnumber)\n"
957		txt += "\n"
958	<	txt += "narg=$#\n"
959	<	txt += "if [ $narg -lt 2 ]\n"
958	>	#        txt += "narg=$#\n"
959	>	txt += "if [ $nargs -lt 2 ]\n"
960		txt += "then\n"
961	<	txt += "    echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$narg+ \n"
961	>	txt += "    echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$nargs+ \n"
962		txt += '    echo "JOB_EXIT_STATUS = 50113"\n'
963		txt += '    echo "JobExitCode=50113" | tee -a $RUNTIME_AREA/$repo\n'
964		txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
965	+	txt += '    rm -f $RUNTIME_AREA/$repo \n'
966	+	txt += '    echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
967	+	txt += '    echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
968		## OLI_Daniele
969		txt += '    if [ $middleware == OSG ]; then \n'
970		txt += '        echo "Remove working directory: $WORKING_DIR"\n'
971		txt += '        cd $RUNTIME_AREA\n'
972		txt += '        /bin/rm -rf $WORKING_DIR\n'
973		txt += '        if [ -d $WORKING_DIR ] ;then\n'
974	<	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'
975	<	txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
976	<	txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
977	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
974	>	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'
975	>	txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
976	>	txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
977	>	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
978	>	txt += '            rm -f $RUNTIME_AREA/$repo \n'
979	>	txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
980	>	txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
981		txt += '        fi\n'
982		txt += '    fi \n'
983		txt += "    exit 1\n"
#	Line 583 | Line 986 | class Cmssw(JobType):
986
987		# Prepare job-specific part
988		job = common.job_list[nj]
989	<	pset = os.path.basename(job.configFilename())
990	<	txt += '\n'
991	<	txt += 'InputFiles=$2\n'
992	<	txt += 'echo "<$InputFiles>"\n'
993	<	#txt += 'echo sed "s#{\'INPUT\'}#$InputFiles#" $RUNTIME_AREA/'+pset+' \n'
994	<	txt += 'sed "s#{\'INPUT\'}#$InputFiles#" $RUNTIME_AREA/'+pset+' > pset.cfg\n'
995	<	#txt += 'sed "s#{\'INPUT\'}#${InputFiles}#" $RUNTIME_AREA/'+pset+' > pset1.cfg\n'
989	>	### FEDE FOR DBS OUTPUT PUBLICATION
990	>	if (self.datasetPath):
991	>	txt += '\n'
992	>	txt += 'DatasetPath='+self.datasetPath+'\n'
993	>
994	>	datasetpath_split = self.datasetPath.split("/")
995	>
996	>	txt += 'PrimaryDataset='+datasetpath_split[1]+'\n'
997	>	txt += 'DataTier='+datasetpath_split[2]+'\n'
998	>	#txt += 'ProcessedDataset='+datasetpath_split[3]+'\n'
999	>	txt += 'ApplicationFamily=cmsRun\n'
1000	>
1001	>	else:
1002	>	txt += 'DatasetPath=MCDataTier\n'
1003	>	txt += 'PrimaryDataset=null\n'
1004	>	txt += 'DataTier=null\n'
1005	>	#txt += 'ProcessedDataset=null\n'
1006	>	txt += 'ApplicationFamily=MCDataTier\n'
1007	>	if self.pset != None: #CarlosDaniele
1008	>	pset = os.path.basename(job.configFilename())
1009	>	txt += '\n'
1010	>	txt += 'cp  $RUNTIME_AREA/'+pset+' .\n'
1011	>	if (self.datasetPath): # standard job
1012	>	#txt += 'InputFiles=$2\n'
1013	>	txt += 'InputFiles=${args[1]}\n'
1014	>	txt += 'MaxEvents=${args[2]}\n'
1015	>	txt += 'SkipEvents=${args[3]}\n'
1016	>	txt += 'echo "Inputfiles:<$InputFiles>"\n'
1017	>	txt += 'sed "s#{\'INPUT\'}#$InputFiles#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1018	>	txt += 'echo "MaxEvents:<$MaxEvents>"\n'
1019	>	txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1020	>	txt += 'echo "SkipEvents:<$SkipEvents>"\n'
1021	>	txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1022	>	else:  # pythia like job
1023	>	seedIndex=1
1024	>	if (self.firstRun):
1025	>	txt += 'FirstRun=${args['+str(seedIndex)+']}\n'
1026	>	txt += 'echo "FirstRun: <$FirstRun>"\n'
1027	>	txt += 'sed "s#\<INPUTFIRSTRUN\>#$FirstRun#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1028	>	seedIndex=seedIndex+1
1029	>
1030	>	if (self.sourceSeed):
1031	>	txt += 'Seed=${args['+str(seedIndex)+']}\n'
1032	>	txt += 'sed "s#\<INPUT\>#$Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1033	>	seedIndex=seedIndex+1
1034	>	## the following seeds are not always present
1035	>	if (self.sourceSeedVtx):
1036	>	txt += 'VtxSeed=${args['+str(seedIndex)+']}\n'
1037	>	txt += 'echo "VtxSeed: <$VtxSeed>"\n'
1038	>	txt += 'sed "s#\<INPUTVTX\>#$VtxSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1039	>	seedIndex += 1
1040	>	if (self.sourceSeedG4):
1041	>	txt += 'G4Seed=${args['+str(seedIndex)+']}\n'
1042	>	txt += 'echo "G4Seed: <$G4Seed>"\n'
1043	>	txt += 'sed "s#\<INPUTG4\>#$G4Seed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1044	>	seedIndex += 1
1045	>	if (self.sourceSeedMix):
1046	>	txt += 'mixSeed=${args['+str(seedIndex)+']}\n'
1047	>	txt += 'echo "MixSeed: <$mixSeed>"\n'
1048	>	txt += 'sed "s#\<INPUTMIX\>#$mixSeed#" '+pset+' > tmp && mv -f tmp '+pset+'\n'
1049	>	seedIndex += 1
1050	>	pass
1051	>	pass
1052	>	txt += 'mv -f '+pset+' pset.cfg\n'
1053
1054		if len(self.additional_inbox_files) > 0:
1055	<	for file in self.additional_inbox_files:
1056	<	txt += 'if [ -e $RUNTIME_AREA/'+file+' ] ; then\n'
1057	<	txt += '   cp $RUNTIME_AREA/'+file+' .\n'
598	<	txt += '   chmod +x '+file+'\n'
599	<	txt += 'fi\n'
1055	>	txt += 'if [ -e $RUNTIME_AREA/'+self.additional_tgz_name+' ] ; then\n'
1056	>	txt += '  tar xzvf $RUNTIME_AREA/'+self.additional_tgz_name+'\n'
1057	>	txt += 'fi\n'
1058		pass
1059
1060	<	txt += 'echo "### END JOB SETUP ENVIRONMENT ###"\n\n'
1061	<
1062	<	txt += '\n'
1063	<	txt += 'echo "***** cat pset.cfg *********"\n'
1064	<	txt += 'cat pset.cfg\n'
1065	<	txt += 'echo "****** end pset.cfg ********"\n'
1066	<	txt += '\n'
1067	<	# txt += 'echo "***** cat pset1.cfg *********"\n'
1068	<	# txt += 'cat pset1.cfg\n'
1069	<	# txt += 'echo "****** end pset1.cfg ********"\n'
1060	>	if self.pset != None: #CarlosDaniele
1061	>	txt += 'echo "### END JOB SETUP ENVIRONMENT ###"\n\n'
1062	>
1063	>	txt += '\n'
1064	>	txt += 'echo "***** cat pset.cfg *********"\n'
1065	>	txt += 'cat pset.cfg\n'
1066	>	txt += 'echo "****** end pset.cfg ********"\n'
1067	>	txt += '\n'
1068	>	### FEDE FOR DBS OUTPUT PUBLICATION
1069	>	txt += 'PSETHASH=`EdmConfigHash < pset.cfg` \n'
1070	>	txt += 'echo "PSETHASH = $PSETHASH" \n'
1071	>	##############
1072	>	txt += '\n'
1073	>	# txt += 'echo "***** cat pset1.cfg *********"\n'
1074	>	# txt += 'cat pset1.cfg\n'
1075	>	# txt += 'echo "****** end pset1.cfg ********"\n'
1076		return txt
1077
1078	<	def wsBuildExe(self, nj):
1078	>	def wsBuildExe(self, nj=0):
1079		"""
1080		Put in the script the commands to build an executable
1081		or a library.
#	Line 632 | Line 1096 | class Cmssw(JobType):
1096		txt += '       cd $RUNTIME_AREA\n'
1097		txt += '       /bin/rm -rf $WORKING_DIR\n'
1098		txt += '       if [ -d $WORKING_DIR ] ;then\n'
1099	<	txt += '        echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
1100	<	txt += '        echo "JOB_EXIT_STATUS = 50999"\n'
1101	<	txt += '        echo "JobExitCode=50999" | tee -a $RUNTIME_AREA/$repo\n'
1102	<	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
1099	>	txt += '           echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
1100	>	txt += '           echo "JOB_EXIT_STATUS = 50999"\n'
1101	>	txt += '           echo "JobExitCode=50999" | tee -a $RUNTIME_AREA/$repo\n'
1102	>	txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1103	>	txt += '           rm -f $RUNTIME_AREA/$repo \n'
1104	>	txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1105	>	txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1106		txt += '       fi\n'
1107		txt += '   fi \n'
1108		txt += '   \n'
#	Line 643 | Line 1110 | class Cmssw(JobType):
1110		txt += 'else \n'
1111		txt += '   echo "Successful untar" \n'
1112		txt += 'fi \n'
1113	+	txt += '\n'
1114	+	txt += 'echo "Include ProdAgentApi and PRODCOMMON in PYTHONPATH"\n'
1115	+	txt += 'if [ -z "$PYTHONPATH" ]; then\n'
1116	+	#### FEDE FOR DBS OUTPUT PUBLICATION
1117	+	txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon\n'
1118	+	#txt += '   export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon\n'
1119	+	#txt += '   export PYTHONPATH=ProdAgentApi\n'
1120	+	txt += 'else\n'
1121	+	txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdAgentApi:$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
1122	+	#txt += '   export PYTHONPATH=`pwd`/ProdAgentApi:`pwd`/ProdCommon:${PYTHONPATH}\n'
1123	+	#txt += '   export PYTHONPATH=ProdAgentApi:${PYTHONPATH}\n'
1124	+	txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
1125	+	###################
1126	+	txt += 'fi\n'
1127	+	txt += '\n'
1128	+
1129		pass
1130
1131		return txt
#	Line 654 | Line 1137 | class Cmssw(JobType):
1137		"""
1138
1139		def executableName(self):
1140	<	return self.executable
1140	>	if self.scriptExe: #CarlosDaniele
1141	>	return "sh "
1142	>	else:
1143	>	return self.executable
1144
1145		def executableArgs(self):
1146	<	return " -p pset.cfg"
1146	>	if self.scriptExe:#CarlosDaniele
1147	>	return   self.scriptExe + " $NJob"
1148	>	else:
1149	>	# if >= CMSSW_1_5_X, add -e
1150	>	version_array = self.scram.getSWVersion().split('_')
1151	>	major = 0
1152	>	minor = 0
1153	>	try:
1154	>	major = int(version_array[1])
1155	>	minor = int(version_array[2])
1156	>	except:
1157	>	msg = "Cannot parse CMSSW version string: " + "_".join(version_array) + " for major and minor release number!"
1158	>	raise CrabException(msg)
1159	>	if major >= 1 and minor >= 5 :
1160	>	return " -e -p pset.cfg"
1161	>	else:
1162	>	return " -p pset.cfg"
1163
1164		def inputSandbox(self, nj):
1165		"""
1166		Returns a list of filenames to be put in JDL input sandbox.
1167		"""
1168		inp_box = []
1169	<	# dict added to delete duplicate from input sandbox file list
1170	<	seen = {}
1169	>	# # dict added to delete duplicate from input sandbox file list
1170	>	# seen = {}
1171		## code
1172		if os.path.isfile(self.tgzNameWithPath):
1173		inp_box.append(self.tgzNameWithPath)
1174	+	if os.path.isfile(self.MLtgzfile):
1175	+	inp_box.append(self.MLtgzfile)
1176		## config
1177	<	inp_box.append(common.job_list[nj].configFilename())
1177	>	if not self.pset is None:
1178	>	inp_box.append(common.work_space.pathForTgz() + 'job/' + self.configFilename())
1179		## additional input files
1180	<	#for file in self.additional_inbox_files:
1181	<	#    inp_box.append(common.work_space.cwdDir()+file)
1180	>	tgz = self.additionalInputFileTgz()
1181	>	inp_box.append(tgz)
1182		return inp_box
1183
1184		def outputSandbox(self, nj):
#	Line 682 | Line 1187 | class Cmssw(JobType):
1187		"""
1188		out_box = []
1189
685	–	stdout=common.job_list[nj].stdout()
686	–	stderr=common.job_list[nj].stderr()
687	–
1190		## User Declared output files
1191	<	for out in self.output_file:
1191	>	for out in (self.output_file+self.output_file_sandbox):
1192		n_out = nj + 1
1193		out_box.append(self.numberFile_(out,str(n_out)))
1194		return out_box
693	–	return []
1195
1196		def prepareSteeringCards(self):
1197		"""
#	Line 706 | Line 1207 | class Cmssw(JobType):
1207		txt = '\n'
1208		txt += '# directory content\n'
1209		txt += 'ls \n'
1210	<	file_list = ''
1211	<	for fileWithSuffix in self.output_file:
1210	>
1211	>	txt += 'output_exit_status=0\n'
1212	>
1213	>	for fileWithSuffix in (self.output_file_sandbox):
1214		output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
712	–	file_list=file_list+output_file_num+' '
1215		txt += '\n'
1216		txt += '# check output file\n'
1217	<	txt += 'ls '+fileWithSuffix+'\n'
1218	<	txt += 'exe_result=$?\n'
1219	<	txt += 'if [ $exe_result -ne 0 ] ; then\n'
1220	<	txt += '   echo "ERROR: No output file to manage"\n'
1221	<	txt += '   echo "JOB_EXIT_STATUS = $exe_result"\n'
1222	<	txt += '   echo "JobExitCode=60302" | tee -a $RUNTIME_AREA/$repo\n'
721	<	txt += '   dumpStatus $RUNTIME_AREA/$repo\n'
722	<	### OLI_DANIELE
1217	>	txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1218	>	txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1219	>	txt += '    cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1220	>	txt += 'else\n'
1221	>	txt += '    exit_status=60302\n'
1222	>	txt += '    echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1223		if common.scheduler.boss_scheduler_name == 'condor_g':
1224		txt += '    if [ $middleware == OSG ]; then \n'
1225		txt += '        echo "prepare dummy output file"\n'
1226		txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1227		txt += '    fi \n'
1228	+	txt += 'fi\n'
1229	+
1230	+	for fileWithSuffix in (self.output_file):
1231	+	output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1232	+	txt += '\n'
1233	+	txt += '# check output file\n'
1234	+	txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1235	+	txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA\n'
1236	+	txt += '    cp $RUNTIME_AREA/'+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1237		txt += 'else\n'
1238	<	txt += '   cp '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1238	>	txt += '    exit_status=60302\n'
1239	>	txt += '    echo "ERROR: Problem with output file '+fileWithSuffix+'"\n'
1240	>	txt += '    echo "JOB_EXIT_STATUS = $exit_status"\n'
1241	>	txt += '    output_exit_status=$exit_status\n'
1242	>	if common.scheduler.boss_scheduler_name == 'condor_g':
1243	>	txt += '    if [ $middleware == OSG ]; then \n'
1244	>	txt += '        echo "prepare dummy output file"\n'
1245	>	txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1246	>	txt += '    fi \n'
1247		txt += 'fi\n'
1248	<
1248	>	file_list = []
1249	>	for fileWithSuffix in (self.output_file):
1250	>	file_list.append(self.numberFile_(fileWithSuffix, '$NJob'))
1251	>
1252	>	txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1253		txt += 'cd $RUNTIME_AREA\n'
733	–	file_list=file_list[:-1]
734	–	txt += 'file_list="'+file_list+'"\n'
735	–	### OLI_DANIELE
736	–	txt += 'if [ $middleware == OSG ]; then\n'
737	–	txt += '    cd $RUNTIME_AREA\n'
738	–	txt += '    echo "Remove working directory: $WORKING_DIR"\n'
739	–	txt += '    /bin/rm -rf $WORKING_DIR\n'
740	–	txt += '    if [ -d $WORKING_DIR ] ;then\n'
741	–	txt += '        echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
742	–	txt += '        echo "JOB_EXIT_STATUS = 60999"\n'
743	–	txt += '        echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
744	–	txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
745	–	txt += '    fi\n'
746	–	txt += 'fi\n'
747	–	txt += '\n'
1254		return txt
1255
1256		def numberFile_(self, file, txt):
#	Line 755 | Line 1261 | class Cmssw(JobType):
1261		# take away last extension
1262		name = p[0]
1263		for x in p[1:-1]:
1264	<	name=name+"."+x
1264	>	name=name+"."+x
1265		# add "_txt"
1266		if len(p)>1:
1267	<	ext = p[len(p)-1]
1268	<	#result = name + '_' + str(txt) + "." + ext
763	<	result = name + '_' + txt + "." + ext
1267	>	ext = p[len(p)-1]
1268	>	result = name + '_' + txt + "." + ext
1269		else:
1270	<	#result = name + '_' + str(txt)
766	<	result = name + '_' + txt
1270	>	result = name + '_' + txt
1271
1272		return result
1273
1274	<	def getRequirements(self):
1274	>	def getRequirements(self, nj=[]):
1275		"""
1276		return job requirements to add to jdl files
1277		"""
1278		req = ''
1279	<	if common.analisys_common_info['sites']:
1280	<	if common.analisys_common_info['sw_version']:
1281	<	req='Member("VO-cms-' + \
1282	<	common.analisys_common_info['sw_version'] + \
1283	<	'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1284	<	if len(common.analisys_common_info['sites'])>0:
1285	<	req = req + ' && ('
1286	<	for i in range(len(common.analisys_common_info['sites'])):
1287	<	req = req + 'other.GlueCEInfoHostName == "' \
1288	<	+ common.analisys_common_info['sites'][i] + '"'
1289	<	if ( i < (int(len(common.analisys_common_info['sites']) - 1)) ):
1290	<	req = req + ' || '
1291	<	req = req + ')'
788	<	#print "req = ", req
1279	>	if self.version:
1280	>	req='Member("VO-cms-' + \
1281	>	self.version + \
1282	>	'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1283	>	## SL add requirement for OS version only if SL4
1284	>	#reSL4 = re.compile( r'slc4' )
1285	>	if self.executable_arch: # and reSL4.search(self.executable_arch):
1286	>	req+=' && Member("VO-cms-' + \
1287	>	self.executable_arch + \
1288	>	'", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1289	>
1290	>	req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1291	>
1292		return req
1293
1294		def configFilename(self):
#	Line 802 | Line 1305 | class Cmssw(JobType):
1305		txt += '   echo "### SETUP CMS OSG  ENVIRONMENT ###"\n'
1306		txt += '   if [ -f $GRID3_APP_DIR/cmssoft/cmsset_default.sh ] ;then\n'
1307		txt += '      # Use $GRID3_APP_DIR/cmssoft/cmsset_default.sh to setup cms software\n'
1308	+	txt += '       export SCRAM_ARCH='+self.executable_arch+'\n'
1309		txt += '       source $GRID3_APP_DIR/cmssoft/cmsset_default.sh '+self.version+'\n'
1310	<	txt += '   elif [ -f $OSG_APP/cmssoft/cmsset_default.sh ] ;then\n'
1311	<	txt += '      # Use $OSG_APP/cmssoft/cmsset_default.sh to setup cms software\n'
1312	<	txt += '       source $OSG_APP/cmssoft/cmsset_default.sh '+self.version+'\n'
1310	>	txt += '   elif [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1311	>	txt += '      # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1312	>	txt += '       export SCRAM_ARCH='+self.executable_arch+'\n'
1313	>	txt += '       source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1314		txt += '   else\n'
1315	<	txt += '       echo "SET_CMS_ENV 10020 ==> ERROR $GRID3_APP_DIR/cmssoft/cmsset_default.sh and $OSG_APP/cmssoft/cmsset_default.sh file not found"\n'
1315	>	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'
1316		txt += '       echo "JOB_EXIT_STATUS = 10020"\n'
1317		txt += '       echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1318		txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
1319	+	txt += '       rm -f $RUNTIME_AREA/$repo \n'
1320	+	txt += '       echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1321	+	txt += '       echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1322		txt += '       exit 1\n'
1323		txt += '\n'
1324		txt += '       echo "Remove working directory: $WORKING_DIR"\n'
1325		txt += '       cd $RUNTIME_AREA\n'
1326		txt += '       /bin/rm -rf $WORKING_DIR\n'
1327		txt += '       if [ -d $WORKING_DIR ] ;then\n'
1328	<	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/cmsset_default.sh file not found"\n'
1329	<	txt += '            echo "JOB_EXIT_STATUS = 10017"\n'
1330	<	txt += '            echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1331	<	txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1328	>	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'
1329	>	txt += '           echo "JOB_EXIT_STATUS = 10017"\n'
1330	>	txt += '           echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1331	>	txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1332	>	txt += '           rm -f $RUNTIME_AREA/$repo \n'
1333	>	txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1334	>	txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1335		txt += '       fi\n'
1336		txt += '\n'
1337		txt += '       exit 1\n'
#	Line 844 | Line 1355 | class Cmssw(JobType):
1355		txt += '       echo "JOB_EXIT_STATUS = 10031" \n'
1356		txt += '       echo "JobExitCode=10031" | tee -a $RUNTIME_AREA/$repo\n'
1357		txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
1358	+	txt += '       rm -f $RUNTIME_AREA/$repo \n'
1359	+	txt += '       echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1360	+	txt += '       echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1361		txt += '       exit 1\n'
1362		txt += '   else\n'
1363		txt += '       echo "Sourcing environment... "\n'
#	Line 852 | Line 1366 | class Cmssw(JobType):
1366		txt += '           echo "JOB_EXIT_STATUS = 10020"\n'
1367		txt += '           echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1368		txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1369	+	txt += '           rm -f $RUNTIME_AREA/$repo \n'
1370	+	txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1371	+	txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1372		txt += '           exit 1\n'
1373		txt += '       fi\n'
1374		txt += '       echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
#	Line 862 | Line 1379 | class Cmssw(JobType):
1379		txt += '           echo "JOB_EXIT_STATUS = 10032"\n'
1380		txt += '           echo "JobExitCode=10032" | tee -a $RUNTIME_AREA/$repo\n'
1381		txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1382	+	txt += '           rm -f $RUNTIME_AREA/$repo \n'
1383	+	txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1384	+	txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1385		txt += '           exit 1\n'
1386		txt += '       fi\n'
1387		txt += '   fi\n'
1388		txt += '   \n'
869	–	txt += '   string=`cat /etc/redhat-release`\n'
870	–	txt += '   echo $string\n'
871	–	txt += '   if [[ $string = *alhalla* ]]; then\n'
872	–	txt += '       echo "SCRAM_ARCH= $SCRAM_ARCH"\n'
873	–	txt += '   elif [[ $string = *Enterprise* ]] || [[ $string = *cientific* ]]; then\n'
874	–	txt += '       export SCRAM_ARCH=slc3_ia32_gcc323\n'
875	–	txt += '       echo "SCRAM_ARCH= $SCRAM_ARCH"\n'
876	–	txt += '   else\n'
877	–	txt += '       echo "SET_CMS_ENV 10033 ==> ERROR OS unknown, LCG environment not initialized"\n'
878	–	txt += '       echo "JOB_EXIT_STATUS = 10033"\n'
879	–	txt += '       echo "JobExitCode=10033" | tee -a $RUNTIME_AREA/$repo\n'
880	–	txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
881	–	txt += '       exit 1\n'
882	–	txt += '   fi\n'
1389		txt += '   echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1390		txt += '   echo "### END SETUP CMS LCG ENVIRONMENT ###"\n'
1391		return txt
1392
1393	+	### FEDE FOR DBS OUTPUT PUBLICATION
1394	+	def modifyReport(self, nj):
1395	+	"""
1396	+	insert the part of the script that modifies the FrameworkJob Report
1397	+	"""
1398	+
1399	+	txt = ''
1400	+	try:
1401	+	publish_data = int(self.cfg_params['USER.publish_data'])
1402	+	except KeyError:
1403	+	publish_data = 0
1404	+	if (publish_data == 1):
1405	+	txt += 'echo "Modify Job Report" \n'
1406	+	#txt += 'chmod a+x $RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1407	+	################ FEDE FOR DBS2 #############################################
1408	+	txt += 'chmod a+x $SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py\n'
1409	+	#############################################################################
1410	+	#try:
1411	+	#    publish_data = int(self.cfg_params['USER.publish_data'])
1412	+	#except KeyError:
1413	+	#    publish_data = 0
1414	+
1415	+	txt += 'if [ -z "$SE" ]; then\n'
1416	+	txt += '    SE="" \n'
1417	+	txt += 'fi \n'
1418	+	txt += 'if [ -z "$SE_PATH" ]; then\n'
1419	+	txt += '    SE_PATH="" \n'
1420	+	txt += 'fi \n'
1421	+	txt += 'echo "SE = $SE"\n'
1422	+	txt += 'echo "SE_PATH = $SE_PATH"\n'
1423	+
1424	+	#if (publish_data == 1):
1425	+	#processedDataset = self.cfg_params['USER.processed_datasetname']
1426	+	processedDataset = self.cfg_params['USER.publish_data_name']
1427	+	txt += 'ProcessedDataset='+processedDataset+'\n'
1428	+	#### LFN=/store/user/<user>/processedDataset_PSETHASH
1429	+	txt += 'if [ "$SE_PATH" == "" ]; then\n'
1430	+	#### FEDE: added slash in LFN ##############
1431	+	txt += '    FOR_LFN=/copy_problems/ \n'
1432	+	txt += 'else \n'
1433	+	txt += '    tmp=`echo $SE_PATH | awk -F \'store\' \'{print$2}\'` \n'
1434	+	#####  FEDE TO BE CHANGED, BECAUSE STORE IS HARDCODED!!!! ########
1435	+	txt += '    FOR_LFN=/store$tmp \n'
1436	+	txt += 'fi \n'
1437	+	txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1438	+	txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1439	+	txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1440	+	#txt += 'echo "$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1441	+	txt += 'echo "$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1442	+	txt += '$SOFTWARE_DIR/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1443	+	#txt += '$RUNTIME_AREA/'+self.version+'/ProdAgentApi/FwkJobRep/ModifyJobReport.py crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier $ProcessedDataset $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1444	+
1445	+	txt += 'modifyReport_result=$?\n'
1446	+	txt += 'echo modifyReport_result = $modifyReport_result\n'
1447	+	txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1448	+	txt += '    exit_status=1\n'
1449	+	txt += '    echo "ERROR: Problem with ModifyJobReport"\n'
1450	+	txt += 'else\n'
1451	+	txt += '    mv NewFrameworkJobReport.xml crab_fjr_$NJob.xml\n'
1452	+	txt += 'fi\n'
1453	+	else:
1454	+	txt += 'echo "no data publication required"\n'
1455	+	#txt += 'ProcessedDataset=no_data_to_publish \n'
1456	+	#### FEDE: added slash in LFN ##############
1457	+	#txt += 'FOR_LFN=/local/ \n'
1458	+	#txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1459	+	#txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1460	+	return txt
1461	+
1462	+	def cleanEnv(self):
1463	+	### OLI_DANIELE
1464	+	txt = ''
1465	+	txt += 'if [ $middleware == OSG ]; then\n'
1466	+	txt += '    cd $RUNTIME_AREA\n'
1467	+	txt += '    echo "Remove working directory: $WORKING_DIR"\n'
1468	+	txt += '    /bin/rm -rf $WORKING_DIR\n'
1469	+	txt += '    if [ -d $WORKING_DIR ] ;then\n'
1470	+	txt += '              echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1471	+	txt += '              echo "JOB_EXIT_STATUS = 60999"\n'
1472	+	txt += '              echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
1473	+	txt += '              dumpStatus $RUNTIME_AREA/$repo\n'
1474	+	txt += '        rm -f $RUNTIME_AREA/$repo \n'
1475	+	txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1476	+	txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1477	+	txt += '    fi\n'
1478	+	txt += 'fi\n'
1479	+	txt += '\n'
1480	+	return txt
1481	+
1482		def setParam_(self, param, value):
1483		self._params[param] = value
1484
1485		def getParams(self):
1486		return self._params
1487	+
1488	+	def setTaskid_(self):
1489	+	self._taskId = self.cfg_params['taskId']
1490	+
1491	+	def getTaskid(self):
1492	+	return self._taskId
1493	+
1494	+	def uniquelist(self, old):
1495	+	"""
1496	+	remove duplicates from a list
1497	+	"""
1498	+	nd={}
1499	+	for e in old:
1500	+	nd[e]=0
1501	+	return nd.keys()
1502	+
1503	+
1504	+	def checkOut(self, limit):
1505	+	"""
1506	+	check the dimension of the output files
1507	+	"""
1508	+	txt = 'echo "*****************************************"\n'
1509	+	txt += 'echo "** Starting output sandbox limit check **"\n'
1510	+	txt += 'echo "*****************************************"\n'
1511	+	allOutFiles = ""
1512	+	listOutFiles = []
1513	+	for fileOut in (self.output_file+self.output_file_sandbox):
1514	+	if fileOut.find('crab_fjr') == -1:
1515	+	allOutFiles = allOutFiles + " " + self.numberFile_(fileOut, '$NJob')
1516	+	listOutFiles.append(self.numberFile_(fileOut, '$NJob'))
1517	+	txt += 'echo "OUTPUT files: '+str(allOutFiles)+'";\n'
1518	+	txt += 'ls -gGhrta;\n'
1519	+	txt += 'sum=0;\n'
1520	+	txt += 'for file in '+str(allOutFiles)+' ; do\n'
1521	+	txt += '    if [ -e $file ]; then\n'
1522	+	txt += '        tt=`ls -gGrta $file | awk \'{ print $3 }\'`\n'
1523	+	txt += '        sum=`expr $sum + $tt`\n'
1524	+	txt += '    else\n'
1525	+	txt += '        echo "WARNING: output file $file not found!"\n'
1526	+	txt += '    fi\n'
1527	+	txt += 'done\n'
1528	+	txt += 'echo "Total Output dimension: $sum";\n'
1529	+	txt += 'limit='+str(limit)+';\n'
1530	+	txt += 'echo "OUTPUT FILES LIMIT SET TO: $limit";\n'
1531	+	txt += 'if [ $limit -lt $sum ]; then\n'
1532	+	txt += '    echo "WARNING: output files have to big size - something will be lost;"\n'
1533	+	txt += '    echo "         checking the output file sizes..."\n'
1534	+	"""
1535	+	txt += '    dim=0;\n'
1536	+	txt += '    exclude=0;\n'
1537	+	txt += '    for files in '+str(allOutFiles)+' ; do\n'
1538	+	txt += '        sumTemp=0;\n'
1539	+	txt += '        for file2 in '+str(allOutFiles)+' ; do\n'
1540	+	txt += '            if [ $file != $file2 ]; then\n'
1541	+	txt += '                tt=`ls -gGrta $file2 | awk \'{ print $3 }\';`\n'
1542	+	txt += '                sumTemp=`expr $sumTemp + $tt`;\n'
1543	+	txt += '            fi\n'
1544	+	txt += '        done\n'
1545	+	txt += '        if [ $sumTemp -lt $limit ]; then\n'
1546	+	txt += '            if [ $dim -lt $sumTemp ]; then\n'
1547	+	txt += '                dim=$sumTemp;\n'
1548	+	txt += '                exclude=$file;\n'
1549	+	txt += '            fi\n'
1550	+	txt += '        fi\n'
1551	+	txt += '    done\n'
1552	+	txt += '    echo "Dimension calculated: $dim"; echo "File to exclude: $exclude";\n'
1553	+	"""
1554	+	txt += '    tot=0;\n'
1555	+	txt += '    for file2 in '+str(allOutFiles)+' ; do\n'
1556	+	txt += '        tt=`ls -gGrta $file2 | awk \'{ print $3 }\';`\n'
1557	+	txt += '        tot=`expr $tot + $tt`;\n'
1558	+	txt += '        if [ $limit -lt $tot ]; then\n'
1559	+	txt += '            tot=`expr $tot - $tt`;\n'
1560	+	txt += '            fileLast=$file;\n'
1561	+	txt += '            break;\n'
1562	+	txt += '        fi\n'
1563	+	txt += '    done\n'
1564	+	txt += '    echo "Dimension calculated: $tot"; echo "First file to exclude: $file";\n'
1565	+	txt += '    flag=0;\n'
1566	+	txt += '    for filess in '+str(allOutFiles)+' ; do\n'
1567	+	txt += '        if [ $fileLast = $filess ]; then\n'
1568	+	txt += '            flag=1;\n'
1569	+	txt += '        fi\n'
1570	+	txt += '        if [ $flag -eq 1 ]; then\n'
1571	+	txt += '            rm -f $filess;\n'
1572	+	txt += '        fi\n'
1573	+	txt += '    done\n'
1574	+	txt += '    ls -agGhrt;\n'
1575	+	txt += '    echo "WARNING: output files are too big in dimension: can not put in the output_sandbox.";\n'
1576	+	txt += '    echo "JOB_EXIT_STATUS = 70000";\n'
1577	+	txt += '    exit_status=70000;\n'
1578	+	txt += 'else'
1579	+	txt += '    echo "Total Output dimension $sum is fine.";\n'
1580	+	txt += 'fi\n'
1581	+	txt += 'echo "*****************************************"\n'
1582	+	txt += 'echo "*** Ending output sandbox limit check ***"\n'
1583	+	txt += 'echo "*****************************************"\n'
1584	+	return txt

Diff Legend

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