ViewVC Help
View File | Revision Log | Show Annotations | Root Listing
root/cvsroot/COMP/CRAB/python/cms_cmssw.py
(Generate patch)

Comparing COMP/CRAB/python/cms_cmssw.py (file contents):
Revision 1.41 by slacapra, Wed Sep 20 17:29:52 2006 UTC vs.
Revision 1.200 by spiga, Fri May 30 10:38:38 2008 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 < import math
5 > from BlackWhiteListParser import BlackWhiteListParser
6   import common
7 import PsetManipulator  
8
9 import DBSInfo
10 import DataDiscovery
11 import DataLocation
7   import Scram
8 + from LFNBaseName import *
9  
10 < import os, string, re
10 > import os, string, glob
11  
12   class Cmssw(JobType):
13      def __init__(self, cfg_params, ncjobs):
14          JobType.__init__(self, 'CMSSW')
15          common.logger.debug(3,'CMSSW::__init__')
16  
17 <        self.analisys_common_info = {}
18 <        # Marco.
17 >        self.argsList = []
18 >
19          self._params = {}
20          self.cfg_params = cfg_params
21 +        # init BlackWhiteListParser
22 +        self.blackWhiteListParser = BlackWhiteListParser(cfg_params)
23  
24 <        # number of jobs requested to be created, limit ojb splitting
24 >        self.MaxTarBallSize = float(self.cfg_params.get('EDG.maxtarballsize',9.5))
25 >
26 >        # number of jobs requested to be created, limit obj splitting
27          self.ncjobs = ncjobs
28  
29          log = common.logger
30 <        
30 >
31          self.scram = Scram.Scram(cfg_params)
32        scramArea = ''
32          self.additional_inbox_files = []
33          self.scriptExe = ''
34          self.executable = ''
35 +        self.executable_arch = self.scram.getArch()
36          self.tgz_name = 'default.tgz'
37 +        self.scriptName = 'CMSSW.sh'
38 +        self.pset = ''
39 +        self.datasetPath = ''
40  
41 +        # set FJR file name
42 +        self.fjrFileName = 'crab_fjr.xml'
43  
44          self.version = self.scram.getSWVersion()
45 <        self.setParam_('application', self.version)
46 <        common.analisys_common_info['sw_version'] = self.version
47 <        ### FEDE
48 <        common.analisys_common_info['copy_input_data'] = 0
44 <        common.analisys_common_info['events_management'] = 1
45 <
46 <        ### collect Data cards
45 >        version_array = self.version.split('_')
46 >        self.CMSSW_major = 0
47 >        self.CMSSW_minor = 0
48 >        self.CMSSW_patch = 0
49          try:
50 <            tmp =  cfg_params['CMSSW.datasetpath']
51 <            log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
52 <            if string.lower(tmp)=='none':
53 <                self.datasetPath = None
54 <                self.selectNoInput = 1
53 <            else:
54 <                self.datasetPath = tmp
55 <                self.selectNoInput = 0
56 <        except KeyError:
57 <            msg = "Error: datasetpath not defined "  
50 >            self.CMSSW_major = int(version_array[1])
51 >            self.CMSSW_minor = int(version_array[2])
52 >            self.CMSSW_patch = int(version_array[3])
53 >        except:
54 >            msg = "Cannot parse CMSSW version string: " + self.version + " for major and minor release number!"
55              raise CrabException(msg)
56  
57 <        # ML monitoring
61 <        # split dataset path style: /PreProdR3Minbias/SIM/GEN-SIM
62 <        if not self.datasetPath:
63 <            self.setParam_('dataset', 'None')
64 <            self.setParam_('owner', 'None')
65 <        else:
66 <            datasetpath_split = self.datasetPath.split("/")
67 <            self.setParam_('dataset', datasetpath_split[1])
68 <            self.setParam_('owner', datasetpath_split[-1])
57 >        ### collect Data cards
58  
59 <        self.setTaskid_()
60 <        self.setParam_('taskId', self.cfg_params['taskId'])
59 >        if not cfg_params.has_key('CMSSW.datasetpath'):
60 >            msg = "Error: datasetpath not defined "
61 >            raise CrabException(msg)
62 >        tmp =  cfg_params['CMSSW.datasetpath']
63 >        log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
64 >        if string.lower(tmp)=='none':
65 >            self.datasetPath = None
66 >            self.selectNoInput = 1
67 >        else:
68 >            self.datasetPath = tmp
69 >            self.selectNoInput = 0
70  
71          self.dataTiers = []
72 <
72 >        self.debugWrap = ''
73 >        self.debug_wrapper = cfg_params.get('USER.debug_wrapper',False)
74 >        if self.debug_wrapper: self.debugWrap='--debug'
75          ## now the application
76 <        try:
77 <            self.executable = cfg_params['CMSSW.executable']
78 <            self.setParam_('exe', self.executable)
79 <            log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
80 <            msg = "Default executable cmsRun overridden. Switch to " + self.executable
81 <            log.debug(3,msg)
82 <        except KeyError:
83 <            self.executable = 'cmsRun'
84 <            self.setParam_('exe', self.executable)
85 <            msg = "User executable not defined. Use cmsRun"
86 <            log.debug(3,msg)
87 <            pass
76 >        self.executable = cfg_params.get('CMSSW.executable','cmsRun')
77 >        log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
78  
79 <        try:
80 <            self.pset = cfg_params['CMSSW.pset']
81 <            log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
79 >        if not cfg_params.has_key('CMSSW.pset'):
80 >            raise CrabException("PSet file missing. Cannot run cmsRun ")
81 >        self.pset = cfg_params['CMSSW.pset']
82 >        log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
83 >        if self.pset.lower() != 'none' :
84              if (not os.path.exists(self.pset)):
85                  raise CrabException("User defined PSet file "+self.pset+" does not exist")
86 <        except KeyError:
87 <            raise CrabException("PSet file missing. Cannot run cmsRun ")
86 >        else:
87 >            self.pset = None
88  
89          # output files
90 <        try:
91 <            self.output_file = []
90 >        ## stuff which must be returned always via sandbox
91 >        self.output_file_sandbox = []
92  
93 <            tmp = cfg_params['CMSSW.output_file']
94 <            if tmp != '':
95 <                tmpOutFiles = string.split(cfg_params['CMSSW.output_file'],',')
96 <                log.debug(7, 'cmssw::cmssw(): output files '+str(tmpOutFiles))
97 <                for tmp in tmpOutFiles:
98 <                    tmp=string.strip(tmp)
99 <                    self.output_file.append(tmp)
100 <                    pass
101 <            else:
102 <                log.message("No output file defined: only stdout/err will be available")
93 >        # add fjr report by default via sandbox
94 >        self.output_file_sandbox.append(self.fjrFileName)
95 >
96 >        # other output files to be returned via sandbox or copied to SE
97 >        self.output_file = []
98 >        tmp = cfg_params.get('CMSSW.output_file',None)
99 >        if tmp :
100 >            tmpOutFiles = string.split(tmp,',')
101 >            log.debug(7, 'cmssw::cmssw(): output files '+str(tmpOutFiles))
102 >            for tmp in tmpOutFiles:
103 >                tmp=string.strip(tmp)
104 >                self.output_file.append(tmp)
105                  pass
106 <            pass
107 <        except KeyError:
108 <            log.message("No output file defined: only stdout/err will be available")
115 <            pass
106 >        else:
107 >            log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
108 >        pass
109  
110          # script_exe file as additional file in inputSandbox
111 <        try:
112 <            self.scriptExe = cfg_params['USER.script_exe']
113 <            self.additional_inbox_files.append(self.scriptExe)
114 <            if self.scriptExe != '':
115 <               if not os.path.isfile(self.scriptExe):
116 <                  msg ="WARNING. file "+self.scriptExe+" not found"
117 <                  raise CrabException(msg)
118 <        except KeyError:
119 <           pass
120 <                  
111 >        self.scriptExe = cfg_params.get('USER.script_exe',None)
112 >        if self.scriptExe :
113 >            if not os.path.isfile(self.scriptExe):
114 >                msg ="ERROR. file "+self.scriptExe+" not found"
115 >                raise CrabException(msg)
116 >            self.additional_inbox_files.append(string.strip(self.scriptExe))
117 >
118 >        if self.datasetPath == None and self.pset == None and self.scriptExe == '' :
119 >            msg ="Error. script_exe  not defined"
120 >            raise CrabException(msg)
121 >
122          ## additional input files
123 <        try:
123 >        if cfg_params.has_key('USER.additional_input_files'):
124              tmpAddFiles = string.split(cfg_params['USER.additional_input_files'],',')
125              for tmp in tmpAddFiles:
126 <                if not os.path.exists(tmp):
127 <                    raise CrabException("Additional input file not found: "+tmp)
128 <                self.additional_inbox_files.append(string.strip(tmp))
126 >                tmp = string.strip(tmp)
127 >                dirname = ''
128 >                if not tmp[0]=="/": dirname = "."
129 >                files = []
130 >                if string.find(tmp,"*")>-1:
131 >                    files = glob.glob(os.path.join(dirname, tmp))
132 >                    if len(files)==0:
133 >                        raise CrabException("No additional input file found with this pattern: "+tmp)
134 >                else:
135 >                    files.append(tmp)
136 >                for file in files:
137 >                    if not os.path.exists(file):
138 >                        raise CrabException("Additional input file not found: "+file)
139 >                    pass
140 >                    self.additional_inbox_files.append(string.strip(file))
141                  pass
142              pass
143 <        except KeyError:
144 <            pass
139 <
140 <        # files per job
141 <        try:
142 <            if (cfg_params['CMSSW.files_per_jobs']):
143 <                raise CrabException("files_per_jobs no longer supported.  Quitting.")
144 <        except KeyError:
145 <            pass
143 >            common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
144 >        pass
145  
146          ## Events per job
147 <        try:
147 >        if cfg_params.has_key('CMSSW.events_per_job'):
148              self.eventsPerJob =int( cfg_params['CMSSW.events_per_job'])
149              self.selectEventsPerJob = 1
150 <        except KeyError:
150 >        else:
151              self.eventsPerJob = -1
152              self.selectEventsPerJob = 0
153 <    
153 >
154          ## number of jobs
155 <        try:
155 >        if cfg_params.has_key('CMSSW.number_of_jobs'):
156              self.theNumberOfJobs =int( cfg_params['CMSSW.number_of_jobs'])
157              self.selectNumberOfJobs = 1
158 <        except KeyError:
158 >        else:
159              self.theNumberOfJobs = 0
160              self.selectNumberOfJobs = 0
161  
162 <        try:
162 >        if cfg_params.has_key('CMSSW.total_number_of_events'):
163              self.total_number_of_events = int(cfg_params['CMSSW.total_number_of_events'])
164              self.selectTotalNumberEvents = 1
165 <        except KeyError:
165 >            if self.selectNumberOfJobs  == 1:
166 >                if int(self.total_number_of_events) < int(self.theNumberOfJobs):
167 >                    msg = 'Must specify at least one event per job. total_number_of_events > number_of_jobs '
168 >                    raise CrabException(msg)
169 >        else:
170              self.total_number_of_events = 0
171              self.selectTotalNumberEvents = 0
172  
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)
176 <
177 <        ## source seed for pythia
178 <        try:
179 <            self.sourceSeed = int(cfg_params['CMSSW.pythia_seed'])
180 <        except KeyError:
181 <            self.sourceSeed = None
182 <            common.logger.debug(5,"No seed given")
173 >        if self.pset != None:
174 >             if ( (self.selectTotalNumberEvents + self.selectEventsPerJob + self.selectNumberOfJobs) != 2 ):
175 >                 msg = 'Must define exactly two of total_number_of_events, events_per_job, or number_of_jobs.'
176 >                 raise CrabException(msg)
177 >        else:
178 >             if (self.selectNumberOfJobs == 0):
179 >                 msg = 'Must specify  number_of_jobs.'
180 >                 raise CrabException(msg)
181 >
182 >        ## New method of dealing with seeds
183 >        self.incrementSeeds = []
184 >        self.preserveSeeds = []
185 >        if cfg_params.has_key('CMSSW.preserve_seeds'):
186 >            tmpList = cfg_params['CMSSW.preserve_seeds'].split(',')
187 >            for tmp in tmpList:
188 >                tmp.strip()
189 >                self.preserveSeeds.append(tmp)
190 >        if cfg_params.has_key('CMSSW.increment_seeds'):
191 >            tmpList = cfg_params['CMSSW.increment_seeds'].split(',')
192 >            for tmp in tmpList:
193 >                tmp.strip()
194 >                self.incrementSeeds.append(tmp)
195 >
196 >        ## Old method of dealing with seeds
197 >        ## FUTURE: This is for old CMSSW and old CRAB. Can throw exceptions after a couple of CRAB releases and then
198 >        ## remove
199 >        self.sourceSeed = cfg_params.get('CMSSW.pythia_seed',None)
200 >        if self.sourceSeed:
201 >            print "pythia_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
202 >            self.incrementSeeds.append('sourceSeed')
203 >            self.incrementSeeds.append('theSource')
204 >
205 >        self.sourceSeedVtx = cfg_params.get('CMSSW.vtx_seed',None)
206 >        if self.sourceSeedVtx:
207 >            print "vtx_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
208 >            self.incrementSeeds.append('VtxSmeared')
209 >
210 >        self.sourceSeedG4 = cfg_params.get('CMSSW.g4_seed',None)
211 >        if self.sourceSeedG4:
212 >            print "g4_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
213 >            self.incrementSeeds.append('g4SimHits')
214 >
215 >        self.sourceSeedMix = cfg_params.get('CMSSW.mix_seed',None)
216 >        if self.sourceSeedMix:
217 >            print "mix_seed is a deprecated parameter. Use preserve_seeds or increment_seeds in the future.\n","Added to increment_seeds."
218 >            self.incrementSeeds.append('mix')
219 >
220 >        self.firstRun = cfg_params.get('CMSSW.first_run',None)
221 >
222 >        if self.pset != None: #CarlosDaniele
223 >            import PsetManipulator as pp
224 >            PsetEdit = pp.PsetManipulator(self.pset) #Daniele Pset
225  
226 <        try:
182 <            self.sourceSeedVtx = int(cfg_params['CMSSW.vtx_seed'])
183 <        except KeyError:
184 <            self.sourceSeedVtx = None
185 <            common.logger.debug(5,"No vertex seed given")
226 >        # Copy/return
227  
228 <        self.PsetEdit = PsetManipulator.PsetManipulator(self.pset) #Daniele Pset
228 >        self.copy_data = int(cfg_params.get('USER.copy_data',0))
229 >        self.return_data = int(cfg_params.get('USER.return_data',0))
230  
231          #DBSDLS-start
232 <        ## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
232 >        ## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
233          self.maxEvents=0  # max events available   ( --> check the requested nb. of evts in Creator.py)
234          self.DBSPaths={}  # all dbs paths requested ( --> input to the site local discovery script)
235          self.jobDestination=[]  # Site destination(s) for each job (list of lists)
# Line 196 | Line 238 | class Cmssw(JobType):
238          blockSites = {}
239          if self.datasetPath:
240              blockSites = self.DataDiscoveryAndLocation(cfg_params)
241 <        #DBSDLS-end          
241 >        #DBSDLS-end
242 >
243  
201        self.tgzNameWithPath = self.getTarBall(self.executable)
202    
244          ## Select Splitting
245 <        if self.selectNoInput: self.jobSplittingNoInput()
246 <        else: self.jobSplittingByBlocks(blockSites)
245 >        if self.selectNoInput:
246 >            if self.pset == None:
247 >                self.jobSplittingForScript()
248 >            else:
249 >                self.jobSplittingNoInput()
250 >        else:
251 >            self.jobSplittingByBlocks(blockSites)
252  
253          # modify Pset
254 <        try:
255 <            if (self.datasetPath): # standard job
256 <                # allow to processa a fraction of events in a file
257 <                self.PsetEdit.inputModule("INPUT")
258 <                self.PsetEdit.maxEvent("INPUTMAXEVENTS")
259 <                self.PsetEdit.skipEvent("INPUTSKIPEVENTS")
260 <
261 <            else:  # pythia like job
262 <                self.PsetEdit.maxEvent(self.eventsPerJob)
263 <                if (self.sourceSeed) :
264 <                    self.PsetEdit.pythiaSeed("INPUT")
219 <                    if (self.sourceSeedVtx) :
220 <                        self.PsetEdit.pythiaSeedVtx("INPUTVTX")
221 <            self.PsetEdit.psetWriter(self.configFilename())
222 <        except:
223 <            msg='Error while manipuliating ParameterSet: exiting...'
224 <            raise CrabException(msg)
254 >        if self.pset != None:
255 >            try:
256 >                # Add FrameworkJobReport to parameter-set, set max events.
257 >                # Reset later for data jobs by writeCFG which does all modifications
258 >                PsetEdit.addCrabFJR(self.fjrFileName) # FUTURE: Job report addition not needed by CMSSW>1.5
259 >                PsetEdit.maxEvent(self.eventsPerJob)
260 >                PsetEdit.psetWriter(self.configFilename())
261 >            except:
262 >                msg='Error while manipulating ParameterSet: exiting...'
263 >                raise CrabException(msg)
264 >        self.tgzNameWithPath = self.getTarBall(self.executable)
265  
266      def DataDiscoveryAndLocation(self, cfg_params):
267  
268 +        import DataDiscovery
269 +        import DataLocation
270          common.logger.debug(10,"CMSSW::DataDiscoveryAndLocation()")
271  
272          datasetPath=self.datasetPath
273  
232        ## TODO
233        dataTiersList = ""
234        dataTiers = dataTiersList.split(',')
235
274          ## Contact the DBS
275 <        common.logger.message("Contacting DBS...")
275 >        common.logger.message("Contacting Data Discovery Services ...")
276          try:
277 <            self.pubdata=DataDiscovery.DataDiscovery(datasetPath, dataTiers, cfg_params)
277 >            self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params)
278              self.pubdata.fetchDBSInfo()
279  
280          except DataDiscovery.NotExistingDatasetError, ex :
281              msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
282              raise CrabException(msg)
245
283          except DataDiscovery.NoDataTierinProvenanceError, ex :
284              msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
285              raise CrabException(msg)
286          except DataDiscovery.DataDiscoveryError, ex:
287 <            msg = 'ERROR ***: failed Data Discovery in DBS  %s'%ex.getErrorMessage()
287 >            msg = 'ERROR ***: failed Data Discovery in DBS :  %s'%ex.getErrorMessage()
288              raise CrabException(msg)
289  
253        ## get list of all required data in the form of dbs paths  (dbs path = /dataset/datatier/owner)
254        ## self.DBSPaths=self.pubdata.getDBSPaths()
255        common.logger.message("Required data are :"+self.datasetPath)
256
290          self.filesbyblock=self.pubdata.getFiles()
291          self.eventsbyblock=self.pubdata.getEventsPerBlock()
292          self.eventsbyfile=self.pubdata.getEventsPerFile()
260        # print str(self.filesbyblock)
261        # print 'self.eventsbyfile',len(self.eventsbyfile)
262        # print str(self.eventsbyfile)
293  
294          ## get max number of events
295 <        self.maxEvents=self.pubdata.getMaxEvents() ##  self.maxEvents used in Creator.py
266 <        common.logger.message("\nThe number of available events is %s"%self.maxEvents)
295 >        self.maxEvents=self.pubdata.getMaxEvents()
296  
268        common.logger.message("Contacting DLS...")
297          ## Contact the DLS and build a list of sites hosting the fileblocks
298          try:
299              dataloc=DataLocation.DataLocation(self.filesbyblock.keys(),cfg_params)
# Line 273 | Line 301 | class Cmssw(JobType):
301          except DataLocation.DataLocationError , ex:
302              msg = 'ERROR ***: failed Data Location in DLS \n %s '%ex.getErrorMessage()
303              raise CrabException(msg)
304 <        
304 >
305  
306          sites = dataloc.getSites()
307          allSites = []
308          listSites = sites.values()
309 <        for list in listSites:
310 <            for oneSite in list:
309 >        for listSite in listSites:
310 >            for oneSite in listSite:
311                  allSites.append(oneSite)
312          allSites = self.uniquelist(allSites)
313  
314 <        common.logger.message("Sites ("+str(len(allSites))+") hosting part/all of dataset: "+str(allSites))
315 <        common.logger.debug(6, "List of Sites: "+str(allSites))
314 >        # screen output
315 >        common.logger.message("Requested dataset: " + datasetPath + " has " + str(self.maxEvents) + " events in " + str(len(self.filesbyblock.keys())) + " blocks.\n")
316 >
317          return sites
318 <    
318 >
319      def jobSplittingByBlocks(self, blockSites):
320          """
321          Perform job splitting. Jobs run over an integer number of files
# Line 336 | Line 365 | class Cmssw(JobType):
365              totalNumberOfJobs = 999999999
366          else :
367              totalNumberOfJobs = self.ncjobs
339            
368  
369          blocks = blockSites.keys()
370          blockCount = 0
# Line 346 | Line 374 | class Cmssw(JobType):
374          jobCount = 0
375          list_of_lists = []
376  
377 +        # list tracking which jobs are in which jobs belong to which block
378 +        jobsOfBlock = {}
379 +
380          # ---- Iterate over the blocks in the dataset until ---- #
381          # ---- we've met the requested total # of events    ---- #
382          while ( (eventsRemaining > 0) and (blockCount < numBlocksInDataset) and (jobCount < totalNumberOfJobs)):
383              block = blocks[blockCount]
384 <
385 <
386 <            evInBlock = self.eventsbyblock[block]
387 <            common.logger.debug(5,'Events in Block File '+str(evInBlock))
388 <
389 <            #Correct - switch to this when DBS up
390 <            #numEventsInBlock = self.eventsbyblock[block]
391 <            numEventsInBlock = evInBlock
392 <            
393 <            files = self.filesbyblock[block]
394 <            numFilesInBlock = len(files)
395 <            if (numFilesInBlock <= 0):
396 <                continue
397 <            fileCount = 0
398 <
399 <            # ---- New block => New job ---- #
400 <            parString = "\\{"
401 <            # counter for number of events in files currently worked on
402 <            filesEventCount = 0
403 <            # flag if next while loop should touch new file
404 <            newFile = 1
405 <            # job event counter
406 <            jobSkipEventCount = 0
407 <            
408 <            # ---- Iterate over the files in the block until we've met the requested ---- #
409 <            # ---- total # of events or we've gone over all the files in this block  ---- #
410 <            while ( (eventsRemaining > 0) and (fileCount < numFilesInBlock) and (jobCount < totalNumberOfJobs) ):
411 <                file = files[fileCount]
412 <                if newFile :
413 <                    try:
414 <                        numEventsInFile = self.eventsbyfile[file]
415 <                        common.logger.debug(6, "File "+str(file)+" has "+str(numEventsInFile)+" events")
416 <                        # increase filesEventCount
417 <                        filesEventCount += numEventsInFile
418 <                        # Add file to current job
419 <                        parString += '\\\"' + file + '\\\"\,'
420 <                        newFile = 0
421 <                    except KeyError:
422 <                        common.logger.message("File "+str(file)+" has unknown numbe of events: skipping")
423 <                        
424 <
425 <                # if less events in file remain than eventsPerJobRequested
426 <                if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested ) :
427 <                    # if last file in block
428 <                    if ( fileCount == numFilesInBlock ) :
429 <                        # end job using last file, use remaining events in block
384 >            blockCount += 1
385 >            if block not in jobsOfBlock.keys() :
386 >                jobsOfBlock[block] = []
387 >
388 >            if self.eventsbyblock.has_key(block) :
389 >                numEventsInBlock = self.eventsbyblock[block]
390 >                common.logger.debug(5,'Events in Block File '+str(numEventsInBlock))
391 >
392 >                files = self.filesbyblock[block]
393 >                numFilesInBlock = len(files)
394 >                if (numFilesInBlock <= 0):
395 >                    continue
396 >                fileCount = 0
397 >
398 >                # ---- New block => New job ---- #
399 >                parString = ""
400 >                # counter for number of events in files currently worked on
401 >                filesEventCount = 0
402 >                # flag if next while loop should touch new file
403 >                newFile = 1
404 >                # job event counter
405 >                jobSkipEventCount = 0
406 >
407 >                # ---- Iterate over the files in the block until we've met the requested ---- #
408 >                # ---- total # of events or we've gone over all the files in this block  ---- #
409 >                while ( (eventsRemaining > 0) and (fileCount < numFilesInBlock) and (jobCount < totalNumberOfJobs) ):
410 >                    file = files[fileCount]
411 >                    if newFile :
412 >                        try:
413 >                            numEventsInFile = self.eventsbyfile[file]
414 >                            common.logger.debug(6, "File "+str(file)+" has "+str(numEventsInFile)+" events")
415 >                            # increase filesEventCount
416 >                            filesEventCount += numEventsInFile
417 >                            # Add file to current job
418 >                            parString += '\\\"' + file + '\\\"\,'
419 >                            newFile = 0
420 >                        except KeyError:
421 >                            common.logger.message("File "+str(file)+" has unknown number of events: skipping")
422 >
423 >                    eventsPerJobRequested = min(eventsPerJobRequested, eventsRemaining)
424 >                    # if less events in file remain than eventsPerJobRequested
425 >                    if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested):
426 >                        # if last file in block
427 >                        if ( fileCount == numFilesInBlock-1 ) :
428 >                            # end job using last file, use remaining events in block
429 >                            # close job and touch new file
430 >                            fullString = parString[:-2]
431 >                            list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
432 >                            common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
433 >                            self.jobDestination.append(blockSites[block])
434 >                            common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
435 >                            # fill jobs of block dictionary
436 >                            jobsOfBlock[block].append(jobCount+1)
437 >                            # reset counter
438 >                            jobCount = jobCount + 1
439 >                            totalEventCount = totalEventCount + filesEventCount - jobSkipEventCount
440 >                            eventsRemaining = eventsRemaining - filesEventCount + jobSkipEventCount
441 >                            jobSkipEventCount = 0
442 >                            # reset file
443 >                            parString = ""
444 >                            filesEventCount = 0
445 >                            newFile = 1
446 >                            fileCount += 1
447 >                        else :
448 >                            # go to next file
449 >                            newFile = 1
450 >                            fileCount += 1
451 >                    # if events in file equal to eventsPerJobRequested
452 >                    elif ( filesEventCount - jobSkipEventCount == eventsPerJobRequested ) :
453                          # close job and touch new file
454                          fullString = parString[:-2]
455 <                        fullString += '\\}'
456 <                        list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
403 <                        common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
455 >                        list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
456 >                        common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
457                          self.jobDestination.append(blockSites[block])
458                          common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
459 +                        jobsOfBlock[block].append(jobCount+1)
460                          # reset counter
461                          jobCount = jobCount + 1
462                          totalEventCount = totalEventCount + eventsPerJobRequested
463                          eventsRemaining = eventsRemaining - eventsPerJobRequested
464                          jobSkipEventCount = 0
465                          # reset file
466 <                        parString = "\\{"
466 >                        parString = ""
467                          filesEventCount = 0
468                          newFile = 1
469                          fileCount += 1
470 +
471 +                    # if more events in file remain than eventsPerJobRequested
472                      else :
473 <                        # go to next file
474 <                        newFile = 1
475 <                        fileCount += 1
476 <                # if events in file equal to eventsPerJobRequested
477 <                elif ( filesEventCount - jobSkipEventCount == eventsPerJobRequested ) :
478 <                    # close job and touch new file
479 <                    fullString = parString[:-2]
480 <                    fullString += '\\}'
481 <                    list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
482 <                    common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
483 <                    self.jobDestination.append(blockSites[block])
484 <                    common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
485 <                    # reset counter
486 <                    jobCount = jobCount + 1
487 <                    totalEventCount = totalEventCount + eventsPerJobRequested
488 <                    eventsRemaining = eventsRemaining - eventsPerJobRequested
489 <                    jobSkipEventCount = 0
490 <                    # reset file
491 <                    parString = "\\{"
436 <                    filesEventCount = 0
437 <                    newFile = 1
438 <                    fileCount += 1
439 <                    
440 <                # if more events in file remain than eventsPerJobRequested
441 <                else :
442 <                    # close job but don't touch new file
443 <                    fullString = parString[:-2]
444 <                    fullString += '\\}'
445 <                    list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
446 <                    common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
447 <                    self.jobDestination.append(blockSites[block])
448 <                    common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
449 <                    # increase counter
450 <                    jobCount = jobCount + 1
451 <                    totalEventCount = totalEventCount + eventsPerJobRequested
452 <                    eventsRemaining = eventsRemaining - eventsPerJobRequested
453 <                    # calculate skip events for last file
454 <                    # use filesEventCount (contains several files), jobSkipEventCount and eventsPerJobRequest
455 <                    jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
456 <                    # remove all but the last file
457 <                    filesEventCount = self.eventsbyfile[file]
458 <                    parString = "\\{"
459 <                    parString += '\\\"' + file + '\\\"\,'
460 <                pass # END if
461 <            pass # END while (iterate over files in the block)
473 >                        # close job but don't touch new file
474 >                        fullString = parString[:-2]
475 >                        list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
476 >                        common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
477 >                        self.jobDestination.append(blockSites[block])
478 >                        common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
479 >                        jobsOfBlock[block].append(jobCount+1)
480 >                        # increase counter
481 >                        jobCount = jobCount + 1
482 >                        totalEventCount = totalEventCount + eventsPerJobRequested
483 >                        eventsRemaining = eventsRemaining - eventsPerJobRequested
484 >                        # calculate skip events for last file
485 >                        # use filesEventCount (contains several files), jobSkipEventCount and eventsPerJobRequest
486 >                        jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
487 >                        # remove all but the last file
488 >                        filesEventCount = self.eventsbyfile[file]
489 >                        parString = '\\\"' + file + '\\\"\,'
490 >                    pass # END if
491 >                pass # END while (iterate over files in the block)
492          pass # END while (iterate over blocks in the dataset)
493          self.ncjobs = self.total_number_of_jobs = jobCount
494          if (eventsRemaining > 0 and jobCount < totalNumberOfJobs ):
495              common.logger.message("Could not run on all requested events because some blocks not hosted at allowed sites.")
496 <        common.logger.message("\n"+str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
497 <        
496 >        common.logger.message(str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
497 >
498 >        # screen output
499 >        screenOutput = "List of jobs and available destination sites:\n\n"
500 >
501 >        # keep trace of block with no sites to print a warning at the end
502 >        noSiteBlock = []
503 >        bloskNoSite = []
504 >
505 >        blockCounter = 0
506 >        for block in blocks:
507 >            if block in jobsOfBlock.keys() :
508 >                blockCounter += 1
509 >                screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),
510 >                    ','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
511 >                if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0:
512 >                    noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
513 >                    bloskNoSite.append( blockCounter )
514 >
515 >        common.logger.message(screenOutput)
516 >        if len(noSiteBlock) > 0 and len(bloskNoSite) > 0:
517 >            msg = 'WARNING: No sites are hosting any part of data for block:\n                '
518 >            virgola = ""
519 >            if len(bloskNoSite) > 1:
520 >                virgola = ","
521 >            for block in bloskNoSite:
522 >                msg += ' ' + str(block) + virgola
523 >            msg += '\n               Related jobs:\n                 '
524 >            virgola = ""
525 >            if len(noSiteBlock) > 1:
526 >                virgola = ","
527 >            for range_jobs in noSiteBlock:
528 >                msg += str(range_jobs) + virgola
529 >            msg += '\n               will not be submitted and this block of data can not be analyzed!\n'
530 >            if self.cfg_params.has_key('EDG.se_white_list'):
531 >                msg += 'WARNING: SE White List: '+self.cfg_params['EDG.se_white_list']+'\n'
532 >                msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
533 >                msg += 'Please check if the dataset is available at this site!)\n'
534 >            if self.cfg_params.has_key('EDG.ce_white_list'):
535 >                msg += 'WARNING: CE White List: '+self.cfg_params['EDG.ce_white_list']+'\n'
536 >                msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
537 >                msg += 'Please check if the dataset is available at this site!)\n'
538 >
539 >            common.logger.message(msg)
540 >
541          self.list_of_args = list_of_lists
542          return
543  
# Line 473 | Line 546 | class Cmssw(JobType):
546          Perform job splitting based on number of event per job
547          """
548          common.logger.debug(5,'Splitting per events')
549 <        common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
550 <        common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
551 <        common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
549 >
550 >        if (self.selectEventsPerJob):
551 >            common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
552 >        if (self.selectNumberOfJobs):
553 >            common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
554 >        if (self.selectTotalNumberEvents):
555 >            common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
556  
557          if (self.total_number_of_events < 0):
558              msg='Cannot split jobs per Events with "-1" as total number of events'
559              raise CrabException(msg)
560  
561          if (self.selectEventsPerJob):
562 <            self.total_number_of_jobs = int(self.total_number_of_events/self.eventsPerJob)
562 >            if (self.selectTotalNumberEvents):
563 >                self.total_number_of_jobs = int(self.total_number_of_events/self.eventsPerJob)
564 >            elif(self.selectNumberOfJobs) :
565 >                self.total_number_of_jobs =self.theNumberOfJobs
566 >                self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
567 >
568          elif (self.selectNumberOfJobs) :
569              self.total_number_of_jobs = self.theNumberOfJobs
570              self.eventsPerJob = int(self.total_number_of_events/self.total_number_of_jobs)
# Line 502 | Line 584 | class Cmssw(JobType):
584          self.list_of_args = []
585          for i in range(self.total_number_of_jobs):
586              ## Since there is no input, any site is good
587 <            self.jobDestination.append(["Any"])
588 <            if (self.sourceSeed):
589 <                if (self.sourceSeedVtx):
590 <                    ## pythia + vtx random seed
591 <                    self.list_of_args.append([
592 <                                              str(self.sourceSeed)+str(i),
593 <                                              str(self.sourceSeedVtx)+str(i)
594 <                                              ])
595 <                else:
596 <                    ## only pythia random seed
597 <                    self.list_of_args.append([(str(self.sourceSeed)+str(i))])
598 <            else:
599 <                ## no random seed
600 <                self.list_of_args.append([str(i)])
601 <        #print self.list_of_args
587 >            self.jobDestination.append([""]) #must be empty to write correctly the xml
588 >            args=[]
589 >            if (self.firstRun):
590 >                ## pythia first run
591 >                args.append(str(self.firstRun)+str(i))
592 >            self.list_of_args.append(args)
593 >
594 >        return
595 >
596 >
597 >    def jobSplittingForScript(self):
598 >        """
599 >        Perform job splitting based on number of job
600 >        """
601 >        common.logger.debug(5,'Splitting per job')
602 >        common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
603  
604 +        self.total_number_of_jobs = self.theNumberOfJobs
605 +
606 +        common.logger.debug(5,'N jobs  '+str(self.total_number_of_jobs))
607 +
608 +        common.logger.message(str(self.total_number_of_jobs)+' jobs can be created')
609 +
610 +        # argument is seed number.$i
611 +        self.list_of_args = []
612 +        for i in range(self.total_number_of_jobs):
613 +            self.jobDestination.append([""])
614 +            self.list_of_args.append([str(i)])
615          return
616  
617      def split(self, jobParams):
618 <
525 <        common.jobDB.load()
526 <        #### Fabio
618 >
619          njobs = self.total_number_of_jobs
620          arglist = self.list_of_args
621          # create the empty structure
622          for i in range(njobs):
623              jobParams.append("")
624 <        
624 >
625 >        listID=[]
626 >        listField=[]
627          for job in range(njobs):
628              jobParams[job] = arglist[job]
629 <            # print str(arglist[job])
630 <            # print jobParams[job]
631 <            common.jobDB.setArguments(job, jobParams[job])
632 <            common.logger.debug(5,"Job "+str(job)+" Destination: "+str(self.jobDestination[job]))
633 <            common.jobDB.setDestination(job, self.jobDestination[job])
629 >            listID.append(job+1)
630 >            job_ToSave ={}
631 >            concString = ' '
632 >            argu=''
633 >            if len(jobParams[job]):
634 >                argu +=   concString.join(jobParams[job] )
635 >            job_ToSave['arguments']= str(job+1)+' '+argu
636 >            job_ToSave['dlsDestination']= self.jobDestination[job]
637 >            listField.append(job_ToSave)
638 >            msg="Job "+str(job)+" Arguments:   "+str(job+1)+" "+argu+"\n"  \
639 >            +"                     Destination: "+str(self.jobDestination[job])
640 >            common.logger.debug(5,msg)
641 >        common._db.updateJob_(listID,listField)
642 >        self.argsList = (len(jobParams[0])+1)
643  
541        common.jobDB.save()
644          return
645 <    
544 <    def getJobTypeArguments(self, nj, sched):
545 <        result = ''
546 <        for i in common.jobDB.arguments(nj):
547 <            result=result+str(i)+" "
548 <        return result
549 <  
645 >
646      def numberOfJobs(self):
551        # Fabio
647          return self.total_number_of_jobs
648  
649      def getTarBall(self, exe):
650          """
651          Return the TarBall with lib and exe
652          """
653 <        
559 <        # if it exist, just return it
560 <        self.tgzNameWithPath = common.work_space.shareDir()+self.tgz_name
653 >        self.tgzNameWithPath = common.work_space.pathForTgz()+'share/'+self.tgz_name
654          if os.path.exists(self.tgzNameWithPath):
655              return self.tgzNameWithPath
656  
# Line 570 | Line 663 | class Cmssw(JobType):
663  
664          # First of all declare the user Scram area
665          swArea = self.scram.getSWArea_()
573        #print "swArea = ", swArea
574        swVersion = self.scram.getSWVersion()
575        #print "swVersion = ", swVersion
666          swReleaseTop = self.scram.getReleaseTop_()
667 <        #print "swReleaseTop = ", swReleaseTop
578 <        
667 >
668          ## check if working area is release top
669          if swReleaseTop == '' or swArea == swReleaseTop:
670 +            common.logger.debug(3,"swArea = "+swArea+" swReleaseTop ="+swReleaseTop)
671              return
672  
673 <        filesToBeTarred = []
674 <        ## First find the executable
675 <        if (self.executable != ''):
676 <            exeWithPath = self.scram.findFile_(executable)
677 < #           print exeWithPath
678 <            if ( not exeWithPath ):
679 <                raise CrabException('User executable '+executable+' not found')
680 <
681 <            ## then check if it's private or not
682 <            if exeWithPath.find(swReleaseTop) == -1:
683 <                # the exe is private, so we must ship
684 <                common.logger.debug(5,"Exe "+exeWithPath+" to be tarred")
685 <                path = swArea+'/'
686 <                exe = string.replace(exeWithPath, path,'')
687 <                filesToBeTarred.append(exe)
688 <                pass
689 <            else:
690 <                # the exe is from release, we'll find it on WN
691 <                pass
692 <
693 <        ## Now get the libraries: only those in local working area
694 <        libDir = 'lib'
695 <        lib = swArea+'/' +libDir
696 <        common.logger.debug(5,"lib "+lib+" to be tarred")
697 <        if os.path.exists(lib):
698 <            filesToBeTarred.append(libDir)
699 <
700 <        ## Now check if module dir is present
701 <        moduleDir = 'module'
702 <        if os.path.isdir(swArea+'/'+moduleDir):
703 <            filesToBeTarred.append(moduleDir)
704 <
705 <        ## Now check if the Data dir is present
706 <        dataDir = 'src/Data/'
707 <        if os.path.isdir(swArea+'/'+dataDir):
708 <            filesToBeTarred.append(dataDir)
709 <
710 <        ## Create the tar-ball
711 <        if len(filesToBeTarred)>0:
712 <            cwd = os.getcwd()
713 <            os.chdir(swArea)
714 <            tarcmd = 'tar zcvf ' + self.tgzNameWithPath + ' '
715 <            for line in filesToBeTarred:
716 <                tarcmd = tarcmd + line + ' '
717 <            cout = runCommand(tarcmd)
718 <            if not cout:
719 <                raise CrabException('Could not create tar-ball')
720 <            os.chdir(cwd)
721 <        else:
722 <            common.logger.debug(5,"No files to be to be tarred")
723 <        
724 <        return
725 <        
726 <    def wsSetupEnvironment(self, nj):
673 >        import tarfile
674 >        try: # create tar ball
675 >            tar = tarfile.open(self.tgzNameWithPath, "w:gz")
676 >            ## First find the executable
677 >            if (self.executable != ''):
678 >                exeWithPath = self.scram.findFile_(executable)
679 >                if ( not exeWithPath ):
680 >                    raise CrabException('User executable '+executable+' not found')
681 >
682 >                ## then check if it's private or not
683 >                if exeWithPath.find(swReleaseTop) == -1:
684 >                    # the exe is private, so we must ship
685 >                    common.logger.debug(5,"Exe "+exeWithPath+" to be tarred")
686 >                    path = swArea+'/'
687 >                    # distinguish case when script is in user project area or given by full path somewhere else
688 >                    if exeWithPath.find(path) >= 0 :
689 >                        exe = string.replace(exeWithPath, path,'')
690 >                        tar.add(path+exe,exe)
691 >                    else :
692 >                        tar.add(exeWithPath,os.path.basename(executable))
693 >                    pass
694 >                else:
695 >                    # the exe is from release, we'll find it on WN
696 >                    pass
697 >
698 >            ## Now get the libraries: only those in local working area
699 >            libDir = 'lib'
700 >            lib = swArea+'/' +libDir
701 >            common.logger.debug(5,"lib "+lib+" to be tarred")
702 >            if os.path.exists(lib):
703 >                tar.add(lib,libDir)
704 >
705 >            ## Now check if module dir is present
706 >            moduleDir = 'module'
707 >            module = swArea + '/' + moduleDir
708 >            if os.path.isdir(module):
709 >                tar.add(module,moduleDir)
710 >
711 >            ## Now check if any data dir(s) is present
712 >            swAreaLen=len(swArea)
713 >            self.dataExist = False
714 >            for root, dirs, files in os.walk(swArea):
715 >                if "data" in dirs:
716 >                    self.dataExist=True
717 >                    common.logger.debug(5,"data "+root+"/data"+" to be tarred")
718 >                    tar.add(root+"/data",root[swAreaLen:]+"/data")
719 >
720 >            ### CMSSW ParameterSet
721 >            if not self.pset is None:
722 >                cfg_file = common.work_space.jobDir()+self.configFilename()
723 >                tar.add(cfg_file,self.configFilename())
724 >                common.logger.debug(5,"File added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
725 >
726 >
727 >            ## Add ProdCommon dir to tar
728 >            prodcommonDir = 'ProdCommon'
729 >            prodcommonPath = os.environ['CRABDIR'] + '/' + 'ProdCommon'
730 >            if os.path.isdir(prodcommonPath):
731 >                tar.add(prodcommonPath,prodcommonDir)
732 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
733 >
734 >            ##### ML stuff
735 >            ML_file_list=['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py']
736 >            path=os.environ['CRABDIR'] + '/python/'
737 >            for file in ML_file_list:
738 >                tar.add(path+file,file)
739 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
740 >
741 >            ##### Utils
742 >            Utils_file_list=['parseCrabFjr.py','writeCfg.py', 'JobReportErrorCode.py']
743 >            for file in Utils_file_list:
744 >                tar.add(path+file,file)
745 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
746 >
747 >            ##### AdditionalFiles
748 >            for file in self.additional_inbox_files:
749 >                tar.add(file,string.split(file,'/')[-1])
750 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
751 >
752 >            tar.close()
753 >        except :
754 >            raise CrabException('Could not create tar-ball')
755 >
756 >        ## check for tarball size
757 >        tarballinfo = os.stat(self.tgzNameWithPath)
758 >        if ( tarballinfo.st_size > self.MaxTarBallSize*1024*1024 ) :
759 >            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.')
760 >
761 >        ## create tar-ball with ML stuff
762 >
763 >    def wsSetupEnvironment(self, nj=0):
764          """
765          Returns part of a job script which prepares
766          the execution environment for the job 'nj'.
767          """
768 +        if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
769 +            psetName = 'pset.py'
770 +        else:
771 +            psetName = 'pset.cfg'
772          # Prepare JobType-independent part
773 <        txt = ''
774 <  
775 <        ## OLI_Daniele at this level  middleware already known
645 <
646 <        txt += 'if [ $middleware == LCG ]; then \n'
773 >        txt = '\n#Written by cms_cmssw::wsSetupEnvironment\n'
774 >        txt += 'echo ">>> setup environment"\n'
775 >        txt += 'if [ $middleware == LCG ]; then \n'
776          txt += self.wsSetupCMSLCGEnvironment_()
777          txt += 'elif [ $middleware == OSG ]; then\n'
778 <        txt += '    time=`date -u +"%s"`\n'
779 <        txt += '    WORKING_DIR=$OSG_WN_TMP/cms_$time\n'
780 <        txt += '    echo "Creating working directory: $WORKING_DIR"\n'
781 <        txt += '    /bin/mkdir -p $WORKING_DIR\n'
782 <        txt += '    if [ ! -d $WORKING_DIR ] ;then\n'
654 <        txt += '        echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
655 <        txt += '        echo "JOB_EXIT_STATUS = 10016"\n'
656 <        txt += '        echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
657 <        txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
658 <        txt += '        rm -f $RUNTIME_AREA/$repo \n'
659 <        txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
660 <        txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
661 <        txt += '        exit 1\n'
778 >        txt += '    WORKING_DIR=`/bin/mktemp  -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
779 >        txt += '    if [ ! $? == 0 ] ;then\n'
780 >        txt += '        echo "ERROR ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
781 >        txt += '        job_exit_code=10016\n'
782 >        txt += '        func_exit\n'
783          txt += '    fi\n'
784 +        txt += '    echo ">>> Created working directory: $WORKING_DIR"\n'
785          txt += '\n'
786          txt += '    echo "Change to working directory: $WORKING_DIR"\n'
787          txt += '    cd $WORKING_DIR\n'
788 <        txt += self.wsSetupCMSOSGEnvironment_()
788 >        txt += '    echo ">>> current directory (WORKING_DIR): $WORKING_DIR"\n'
789 >        txt += self.wsSetupCMSOSGEnvironment_()
790          txt += 'fi\n'
791  
792          # Prepare JobType-specific part
793          scram = self.scram.commandName()
794          txt += '\n\n'
795 <        txt += 'echo "### SPECIFIC JOB SETUP ENVIRONMENT ###"\n'
795 >        txt += 'echo ">>> specific cmssw setup environment:"\n'
796 >        txt += 'echo "CMSSW_VERSION =  '+self.version+'"\n'
797          txt += scram+' project CMSSW '+self.version+'\n'
798          txt += 'status=$?\n'
799          txt += 'if [ $status != 0 ] ; then\n'
800 <        txt += '   echo "SET_EXE_ENV 10034 ==>ERROR CMSSW '+self.version+' not found on `hostname`" \n'
801 <        txt += '   echo "JOB_EXIT_STATUS = 10034"\n'
802 <        txt += '   echo "JobExitCode=10034" | tee -a $RUNTIME_AREA/$repo\n'
679 <        txt += '   dumpStatus $RUNTIME_AREA/$repo\n'
680 <        txt += '   rm -f $RUNTIME_AREA/$repo \n'
681 <        txt += '   echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
682 <        txt += '   echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
683 <        ## OLI_Daniele
684 <        txt += '    if [ $middleware == OSG ]; then \n'
685 <        txt += '        echo "Remove working directory: $WORKING_DIR"\n'
686 <        txt += '        cd $RUNTIME_AREA\n'
687 <        txt += '        /bin/rm -rf $WORKING_DIR\n'
688 <        txt += '        if [ -d $WORKING_DIR ] ;then\n'
689 <        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'
690 <        txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
691 <        txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
692 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
693 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
694 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
695 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
696 <        txt += '        fi\n'
697 <        txt += '    fi \n'
698 <        txt += '   exit 1 \n'
800 >        txt += '    echo "ERROR ==> CMSSW '+self.version+' not found on `hostname`" \n'
801 >        txt += '    job_exit_code=10034\n'
802 >        txt += '    func_exit\n'
803          txt += 'fi \n'
700        txt += 'echo "CMSSW_VERSION =  '+self.version+'"\n'
804          txt += 'cd '+self.version+'\n'
805 <        ### needed grep for bug in scramv1 ###
805 >        txt += 'SOFTWARE_DIR=`pwd`\n'
806 >        txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
807          txt += 'eval `'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME`\n'
808 <
808 >        txt += 'if [ $? != 0 ] ; then\n'
809 >        txt += '    echo "ERROR ==> Problem with the command: "\n'
810 >        txt += '    echo "eval \`'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME \` at `hostname`"\n'
811 >        txt += '    job_exit_code=10034\n'
812 >        txt += '    func_exit\n'
813 >        txt += 'fi \n'
814          # Handle the arguments:
815          txt += "\n"
816          txt += "## number of arguments (first argument always jobnumber)\n"
817          txt += "\n"
818 < #        txt += "narg=$#\n"
710 <        txt += "if [ $nargs -lt 2 ]\n"
818 >        txt += "if [ $nargs -lt "+str(self.argsList)+" ]\n"
819          txt += "then\n"
820 <        txt += "    echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$nargs+ \n"
821 <        txt += '    echo "JOB_EXIT_STATUS = 50113"\n'
822 <        txt += '    echo "JobExitCode=50113" | tee -a $RUNTIME_AREA/$repo\n'
715 <        txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
716 <        txt += '    rm -f $RUNTIME_AREA/$repo \n'
717 <        txt += '    echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
718 <        txt += '    echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
719 <        ## OLI_Daniele
720 <        txt += '    if [ $middleware == OSG ]; then \n'
721 <        txt += '        echo "Remove working directory: $WORKING_DIR"\n'
722 <        txt += '        cd $RUNTIME_AREA\n'
723 <        txt += '        /bin/rm -rf $WORKING_DIR\n'
724 <        txt += '        if [ -d $WORKING_DIR ] ;then\n'
725 <        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'
726 <        txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
727 <        txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
728 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
729 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
730 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
731 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
732 <        txt += '        fi\n'
733 <        txt += '    fi \n'
734 <        txt += "    exit 1\n"
820 >        txt += "    echo 'ERROR ==> Too few arguments' +$nargs+ \n"
821 >        txt += '    job_exit_code=50113\n'
822 >        txt += "    func_exit\n"
823          txt += "fi\n"
824          txt += "\n"
825  
826          # Prepare job-specific part
827          job = common.job_list[nj]
828 <        pset = os.path.basename(job.configFilename())
829 <        txt += '\n'
830 <        if (self.datasetPath): # standard job
743 <            #txt += 'InputFiles=$2\n'
744 <            txt += 'InputFiles=${args[1]}\n'
745 <            txt += 'MaxEvents=${args[2]}\n'
746 <            txt += 'SkipEvents=${args[3]}\n'
747 <            txt += 'echo "Inputfiles:<$InputFiles>"\n'
748 <            txt += 'sed "s#{\'INPUT\'}#$InputFiles#" $RUNTIME_AREA/'+pset+' > pset_tmp_1.cfg\n'
749 <            txt += 'echo "MaxEvents:<$MaxEvents>"\n'
750 <            txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" $RUNTIME_AREA/ pset_tmp_1.cfg > pset_tmp_2.cfg\n'
751 <            txt += 'echo "SkipEvents:<$SkipEvents>"\n'
752 <            txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" $RUNTIME_AREA/ pset_tmp_2.cfg > pset.cfg\n'
753 <        else:  # pythia like job
754 <            if (self.sourceSeed):
755 < #                txt += 'Seed=$2\n'
756 <                txt += 'Seed=${args[1]}\n'
757 <                txt += 'echo "Seed: <$Seed>"\n'
758 <                txt += 'sed "s#\<INPUT\>#$Seed#" $RUNTIME_AREA/'+pset+' > tmp.cfg\n'
759 <                if (self.sourceSeedVtx):
760 < #                    txt += 'VtxSeed=$3\n'
761 <                    txt += 'VtxSeed=${args[2]}\n'
762 <                    txt += 'echo "VtxSeed: <$VtxSeed>"\n'
763 <                    txt += 'sed "s#INPUTVTX#$VtxSeed#" tmp.cfg > pset.cfg\n'
764 <                else:
765 <                    txt += 'mv tmp.cfg pset.cfg\n'
766 <            else:
767 <                txt += '# Copy untouched pset\n'
768 <                txt += 'cp $RUNTIME_AREA/'+pset+' pset.cfg\n'
828 >        if (self.datasetPath):
829 >            txt += '\n'
830 >            txt += 'DatasetPath='+self.datasetPath+'\n'
831  
832 +            datasetpath_split = self.datasetPath.split("/")
833  
834 <        if len(self.additional_inbox_files) > 0:
835 <            for file in self.additional_inbox_files:
836 <                relFile = file.split("/")[-1]
774 <                txt += 'if [ -e $RUNTIME_AREA/'+relFile+' ] ; then\n'
775 <                txt += '   cp $RUNTIME_AREA/'+relFile+' .\n'
776 <                txt += '   chmod +x '+relFile+'\n'
777 <                txt += 'fi\n'
778 <            pass
834 >            txt += 'PrimaryDataset='+datasetpath_split[1]+'\n'
835 >            txt += 'DataTier='+datasetpath_split[2]+'\n'
836 >            txt += 'ApplicationFamily=cmsRun\n'
837  
838 <        txt += 'echo "### END JOB SETUP ENVIRONMENT ###"\n\n'
838 >        else:
839 >            txt += 'DatasetPath=MCDataTier\n'
840 >            txt += 'PrimaryDataset=null\n'
841 >            txt += 'DataTier=null\n'
842 >            txt += 'ApplicationFamily=MCDataTier\n'
843 >        if self.pset != None:
844 >            pset = os.path.basename(job.configFilename())
845 >            txt += '\n'
846 >            txt += 'cp  $RUNTIME_AREA/'+pset+' .\n'
847 >            if (self.datasetPath): # standard job
848 >                txt += 'InputFiles=${args[1]}; export InputFiles\n'
849 >                txt += 'MaxEvents=${args[2]}; export MaxEvents\n'
850 >                txt += 'SkipEvents=${args[3]}; export SkipEvents\n'
851 >                txt += 'echo "Inputfiles:<$InputFiles>"\n'
852 >                txt += 'echo "MaxEvents:<$MaxEvents>"\n'
853 >                txt += 'echo "SkipEvents:<$SkipEvents>"\n'
854 >            else:  # pythia like job
855 >                txt += 'PreserveSeeds='  + ','.join(self.preserveSeeds)  + '; export PreserveSeeds\n'
856 >                txt += 'IncrementSeeds=' + ','.join(self.incrementSeeds) + '; export IncrementSeeds\n'
857 >                txt += 'echo "PreserveSeeds: <$PreserveSeeds>"\n'
858 >                txt += 'echo "IncrementSeeds:<$IncrementSeeds>"\n'
859 >                if (self.firstRun):
860 >                    txt += 'FirstRun=${args[1]}; export FirstRun\n'
861 >                    txt += 'echo "FirstRun: <$FirstRun>"\n'
862  
863 <        txt += '\n'
864 <        txt += 'echo "***** cat pset.cfg *********"\n'
865 <        txt += 'cat pset.cfg\n'
866 <        txt += 'echo "****** end pset.cfg ********"\n'
867 <        txt += '\n'
868 <        # txt += 'echo "***** cat pset1.cfg *********"\n'
869 <        # txt += 'cat pset1.cfg\n'
870 <        # txt += 'echo "****** end pset1.cfg ********"\n'
863 >            txt += 'mv -f ' + pset + ' ' + psetName + '\n'
864 >
865 >
866 >        if self.pset != None:
867 >            # FUTURE: Can simply for 2_1_x and higher
868 >            txt += '\n'
869 >            if self.debug_wrapper==True:
870 >                txt += 'echo "***** cat ' + psetName + ' *********"\n'
871 >                txt += 'cat ' + psetName + '\n'
872 >                txt += 'echo "****** end ' + psetName + ' ********"\n'
873 >                txt += '\n'
874 >            txt += 'PSETHASH=`edmConfigHash < ' + psetName + '` \n'
875 >            txt += 'echo "PSETHASH = $PSETHASH" \n'
876 >            txt += '\n'
877          return txt
878  
879 <    def wsBuildExe(self, nj):
879 >    def wsUntarSoftware(self, nj=0):
880          """
881          Put in the script the commands to build an executable
882          or a library.
883          """
884  
885 <        txt = ""
885 >        txt = '\n#Written by cms_cmssw::wsUntarSoftware\n'
886  
887          if os.path.isfile(self.tgzNameWithPath):
888 <            txt += 'echo "tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'"\n'
888 >            txt += 'echo ">>> tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+' :" \n'
889              txt += 'tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'\n'
890 +            if  self.debug_wrapper:
891 +                txt += 'ls -Al \n'
892              txt += 'untar_status=$? \n'
893              txt += 'if [ $untar_status -ne 0 ]; then \n'
894 <            txt += '   echo "SET_EXE 1 ==> ERROR Untarring .tgz file failed"\n'
895 <            txt += '   echo "JOB_EXIT_STATUS = $untar_status" \n'
896 <            txt += '   echo "JobExitCode=$untar_status" | tee -a $RUNTIME_AREA/$repo\n'
808 <            txt += '   if [ $middleware == OSG ]; then \n'
809 <            txt += '       echo "Remove working directory: $WORKING_DIR"\n'
810 <            txt += '       cd $RUNTIME_AREA\n'
811 <            txt += '       /bin/rm -rf $WORKING_DIR\n'
812 <            txt += '       if [ -d $WORKING_DIR ] ;then\n'
813 <            txt += '           echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
814 <            txt += '           echo "JOB_EXIT_STATUS = 50999"\n'
815 <            txt += '           echo "JobExitCode=50999" | tee -a $RUNTIME_AREA/$repo\n'
816 <            txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
817 <            txt += '           rm -f $RUNTIME_AREA/$repo \n'
818 <            txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
819 <            txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
820 <            txt += '       fi\n'
821 <            txt += '   fi \n'
822 <            txt += '   \n'
823 <            txt += '   exit 1 \n'
894 >            txt += '   echo "ERROR ==> Untarring .tgz file failed"\n'
895 >            txt += '   job_exit_code=$untar_status\n'
896 >            txt += '   func_exit\n'
897              txt += 'else \n'
898              txt += '   echo "Successful untar" \n'
899              txt += 'fi \n'
900 +            txt += '\n'
901 +            txt += 'echo ">>> Include ProdCommon in PYTHONPATH:"\n'
902 +            txt += 'if [ -z "$PYTHONPATH" ]; then\n'
903 +            txt += '   export PYTHONPATH=$RUNTIME_AREA/ProdCommon\n'
904 +            txt += 'else\n'
905 +            txt += '   export PYTHONPATH=$RUNTIME_AREA/ProdCommon:${PYTHONPATH}\n'
906 +            txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
907 +            txt += 'fi\n'
908 +            txt += '\n'
909 +
910              pass
911 <        
911 >
912 >        return txt
913 >
914 >    def wsBuildExe(self, nj=0):
915 >        """
916 >        Put in the script the commands to build an executable
917 >        or a library.
918 >        """
919 >
920 >        txt = '\n#Written by cms_cmssw::wsBuildExe\n'
921 >        txt += 'echo ">>> moving CMSSW software directories in `pwd`" \n'
922 >
923 >        txt += 'rm -r lib/ module/ \n'
924 >        txt += 'mv $RUNTIME_AREA/lib/ . \n'
925 >        txt += 'mv $RUNTIME_AREA/module/ . \n'
926 >        if self.dataExist == True:
927 >            txt += 'rm -r src/ \n'
928 >            txt += 'mv $RUNTIME_AREA/src/ . \n'
929 >        if len(self.additional_inbox_files)>0:
930 >            for file in self.additional_inbox_files:
931 >                txt += 'mv $RUNTIME_AREA/'+os.path.basename(file)+' . \n'
932 >        txt += 'mv $RUNTIME_AREA/ProdCommon/ . \n'
933 >
934 >        txt += 'if [ -z "$PYTHONPATH" ]; then\n'
935 >        txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdCommon\n'
936 >        txt += 'else\n'
937 >        txt += '   export PYTHONPATH=$SOFTWARE_DIR/ProdCommon:${PYTHONPATH}\n'
938 >        txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
939 >        txt += 'fi\n'
940 >        txt += '\n'
941 >
942          return txt
943  
944      def modifySteeringCards(self, nj):
945          """
946 <        modify the card provided by the user,
946 >        modify the card provided by the user,
947          writing a new card into share dir
948          """
949 <        
949 >
950      def executableName(self):
951 <        return self.executable
951 >        if self.scriptExe:
952 >            return "sh "
953 >        else:
954 >            return self.executable
955  
956      def executableArgs(self):
957 <        return " -p pset.cfg"
957 >        # FUTURE: This function tests the CMSSW version. Can be simplified as we drop support for old versions
958 >        if self.scriptExe:#CarlosDaniele
959 >            return   self.scriptExe + " $NJob"
960 >        else:
961 >            ex_args = ""
962 >            # FUTURE: This tests the CMSSW version. Can remove code as versions deprecated
963 >            # Framework job report
964 >            if (self.CMSSW_major >= 1 and self.CMSSW_minor >= 5) or (self.CMSSW_major >= 2):
965 >                ex_args += " -j $RUNTIME_AREA/crab_fjr_$NJob.xml"
966 >            # Type of config file
967 >            if self.CMSSW_major >= 2 :
968 >                ex_args += " -p pset.py"
969 >            else:
970 >                ex_args += " -p pset.cfg"
971 >            return ex_args
972  
973      def inputSandbox(self, nj):
974          """
975          Returns a list of filenames to be put in JDL input sandbox.
976          """
977          inp_box = []
848        # dict added to delete duplicate from input sandbox file list
849        seen = {}
850        ## code
978          if os.path.isfile(self.tgzNameWithPath):
979              inp_box.append(self.tgzNameWithPath)
980 <        ## config
981 <        inp_box.append(common.job_list[nj].configFilename())
855 <        ## additional input files
856 <        #for file in self.additional_inbox_files:
857 <        #    inp_box.append(common.work_space.cwdDir()+file)
980 >        wrapper = os.path.basename(str(common._db.queryTask('scriptName')))
981 >        inp_box.append(common.work_space.pathForTgz() +'job/'+ wrapper)
982          return inp_box
983  
984      def outputSandbox(self, nj):
# Line 863 | Line 987 | class Cmssw(JobType):
987          """
988          out_box = []
989  
866        stdout=common.job_list[nj].stdout()
867        stderr=common.job_list[nj].stderr()
868
990          ## User Declared output files
991 <        for out in self.output_file:
992 <            n_out = nj + 1
991 >        for out in (self.output_file+self.output_file_sandbox):
992 >            n_out = nj + 1
993              out_box.append(self.numberFile_(out,str(n_out)))
994          return out_box
874        return []
995  
996      def prepareSteeringCards(self):
997          """
# Line 884 | Line 1004 | class Cmssw(JobType):
1004          Returns part of a job script which renames the produced files.
1005          """
1006  
1007 <        txt = '\n'
1008 <        txt += '# directory content\n'
1009 <        txt += 'ls \n'
1010 <        file_list = ''
1011 <        for fileWithSuffix in self.output_file:
1007 >        txt = '\n#Written by cms_cmssw::wsRenameOutput\n'
1008 >        txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1009 >        txt += 'echo ">>> current directory content:"\n'
1010 >        if self.debug_wrapper:
1011 >            txt += 'ls -Al\n'
1012 >        txt += '\n'
1013 >
1014 >        for fileWithSuffix in (self.output_file):
1015              output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
893            file_list=file_list+output_file_num+' '
1016              txt += '\n'
1017              txt += '# check output file\n'
1018 <            txt += 'ls '+fileWithSuffix+'\n'
1019 <            txt += 'ls_result=$?\n'
1020 <            #txt += 'exe_result=$?\n'
1021 <            txt += 'if [ $ls_result -ne 0 ] ; then\n'
1022 <            txt += '   echo "ERROR: Problem with output file"\n'
1023 <            #txt += '   echo "JOB_EXIT_STATUS = $exe_result"\n'
1024 <            #txt += '   echo "JobExitCode=60302" | tee -a $RUNTIME_AREA/$repo\n'
1025 <            #txt += '   dumpStatus $RUNTIME_AREA/$repo\n'
1026 <            ### OLI_DANIELE
1027 <            if common.scheduler.boss_scheduler_name == 'condor_g':
1018 >            txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1019 >            if (self.copy_data == 1):  # For OSG nodes, file is in $WORKING_DIR, should not be moved to $RUNTIME_AREA
1020 >                txt += '    mv '+fileWithSuffix+' '+output_file_num+'\n'
1021 >                #txt += '    ln -s `pwd`/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1022 >            else:
1023 >                txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1024 >                txt += '    ln -s $RUNTIME_AREA/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1025 >            txt += 'else\n'
1026 >            txt += '    job_exit_code=60302\n'
1027 >            txt += '    echo "WARNING: Output file '+fileWithSuffix+' not found"\n'
1028 >            if common.scheduler.name().upper() == 'CONDOR_G':
1029                  txt += '    if [ $middleware == OSG ]; then \n'
1030                  txt += '        echo "prepare dummy output file"\n'
1031                  txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1032                  txt += '    fi \n'
910            txt += 'else\n'
911            txt += '   cp '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1033              txt += 'fi\n'
1034 <      
1035 <        txt += 'cd $RUNTIME_AREA\n'
1036 <        file_list=file_list[:-1]
1037 <        txt += 'file_list="'+file_list+'"\n'
1038 <        txt += 'cd $RUNTIME_AREA\n'
918 <        ### OLI_DANIELE
919 <        txt += 'if [ $middleware == OSG ]; then\n'  
920 <        txt += '    cd $RUNTIME_AREA\n'
921 <        txt += '    echo "Remove working directory: $WORKING_DIR"\n'
922 <        txt += '    /bin/rm -rf $WORKING_DIR\n'
923 <        txt += '    if [ -d $WORKING_DIR ] ;then\n'
924 <        txt += '        echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
925 <        txt += '        echo "JOB_EXIT_STATUS = 60999"\n'
926 <        txt += '        echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
927 <        txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
928 <        txt += '        rm -f $RUNTIME_AREA/$repo \n'
929 <        txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
930 <        txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
931 <        txt += '    fi\n'
932 <        txt += 'fi\n'
1034 >        file_list = []
1035 >        for fileWithSuffix in (self.output_file):
1036 >             file_list.append(self.numberFile_(fileWithSuffix, '$NJob'))
1037 >
1038 >        txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1039          txt += '\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 +        txt += 'cd $RUNTIME_AREA\n'
1046 +        txt += 'echo ">>> current directory (RUNTIME_AREA):  $RUNTIME_AREA"\n'
1047          return txt
1048  
1049      def numberFile_(self, file, txt):
# Line 941 | Line 1054 | class Cmssw(JobType):
1054          # take away last extension
1055          name = p[0]
1056          for x in p[1:-1]:
1057 <           name=name+"."+x
1057 >            name=name+"."+x
1058          # add "_txt"
1059          if len(p)>1:
1060 <          ext = p[len(p)-1]
1061 <          result = name + '_' + txt + "." + ext
1060 >            ext = p[len(p)-1]
1061 >            result = name + '_' + txt + "." + ext
1062          else:
1063 <          result = name + '_' + txt
1064 <        
1063 >            result = name + '_' + txt
1064 >
1065          return result
1066  
1067 <    def getRequirements(self):
1067 >    def getRequirements(self, nj=[]):
1068          """
1069 <        return job requirements to add to jdl files
1069 >        return job requirements to add to jdl files
1070          """
1071          req = ''
1072 <        if common.analisys_common_info['sw_version']:
1072 >        if self.version:
1073              req='Member("VO-cms-' + \
1074 <                 common.analisys_common_info['sw_version'] + \
1074 >                 self.version + \
1075 >                 '", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1076 >        if self.executable_arch:
1077 >            req+=' && Member("VO-cms-' + \
1078 >                 self.executable_arch + \
1079                   '", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1080  
1081          req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1082 +        if common.scheduler.name() == "glitecoll":
1083 +            req += ' && other.GlueCEStateStatus == "Production" '
1084  
1085          return req
1086  
1087      def configFilename(self):
1088          """ return the config filename """
1089 <        return self.name()+'.cfg'
1089 >        # FUTURE: Can remove cfg mode for CMSSW >= 2_1_x
1090 >        if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
1091 >          return self.name()+'.py'
1092 >        else:
1093 >          return self.name()+'.cfg'
1094  
972    ### OLI_DANIELE
1095      def wsSetupCMSOSGEnvironment_(self):
1096          """
1097          Returns part of a job script which is prepares
1098          the execution environment and which is common for all CMS jobs.
1099          """
1100 <        txt = '\n'
1101 <        txt += '   echo "### SETUP CMS OSG  ENVIRONMENT ###"\n'
1102 <        txt += '   if [ -f $GRID3_APP_DIR/cmssoft/cmsset_default.sh ] ;then\n'
1103 <        txt += '      # Use $GRID3_APP_DIR/cmssoft/cmsset_default.sh to setup cms software\n'
1104 <        txt += '       source $GRID3_APP_DIR/cmssoft/cmsset_default.sh '+self.version+'\n'
1105 <        txt += '   elif [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1100 >        txt = '\n#Written by cms_cmssw::wsSetupCMSOSGEnvironment_\n'
1101 >        txt += '    echo ">>> setup CMS OSG environment:"\n'
1102 >        txt += '    echo "set SCRAM ARCH to ' + self.executable_arch + '"\n'
1103 >        txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
1104 >        txt += '    echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1105 >        txt += '    if [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1106          txt += '      # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1107 <        txt += '       source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1108 <        txt += '   else\n'
1109 <        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'
1110 <        txt += '       echo "JOB_EXIT_STATUS = 10020"\n'
1111 <        txt += '       echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1112 <        txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
991 <        txt += '       rm -f $RUNTIME_AREA/$repo \n'
992 <        txt += '       echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
993 <        txt += '       echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
994 <        txt += '       exit 1\n'
995 <        txt += '\n'
996 <        txt += '       echo "Remove working directory: $WORKING_DIR"\n'
997 <        txt += '       cd $RUNTIME_AREA\n'
998 <        txt += '       /bin/rm -rf $WORKING_DIR\n'
999 <        txt += '       if [ -d $WORKING_DIR ] ;then\n'
1000 <        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'
1001 <        txt += '            echo "JOB_EXIT_STATUS = 10017"\n'
1002 <        txt += '            echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1003 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1004 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
1005 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1006 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1007 <        txt += '       fi\n'
1008 <        txt += '\n'
1009 <        txt += '       exit 1\n'
1010 <        txt += '   fi\n'
1107 >        txt += '        source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1108 >        txt += '    else\n'
1109 >        txt += '        echo "ERROR ==> $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1110 >        txt += '        job_exit_code=10020\n'
1111 >        txt += '        func_exit\n'
1112 >        txt += '    fi\n'
1113          txt += '\n'
1114 <        txt += '   echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1115 <        txt += '   echo " END SETUP CMS OSG  ENVIRONMENT "\n'
1114 >        txt += '    echo "==> setup cms environment ok"\n'
1115 >        txt += '    echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1116  
1117          return txt
1118 <
1017 <    ### OLI_DANIELE
1118 >
1119      def wsSetupCMSLCGEnvironment_(self):
1120          """
1121          Returns part of a job script which is prepares
1122          the execution environment and which is common for all CMS jobs.
1123          """
1124 <        txt  = '   \n'
1125 <        txt += '   echo " ### SETUP CMS LCG  ENVIRONMENT ### "\n'
1126 <        txt += '   if [ ! $VO_CMS_SW_DIR ] ;then\n'
1127 <        txt += '       echo "SET_CMS_ENV 10031 ==> ERROR CMS software dir not found on WN `hostname`"\n'
1128 <        txt += '       echo "JOB_EXIT_STATUS = 10031" \n'
1129 <        txt += '       echo "JobExitCode=10031" | tee -a $RUNTIME_AREA/$repo\n'
1130 <        txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
1131 <        txt += '       rm -f $RUNTIME_AREA/$repo \n'
1132 <        txt += '       echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1133 <        txt += '       echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1134 <        txt += '       exit 1\n'
1135 <        txt += '   else\n'
1136 <        txt += '       echo "Sourcing environment... "\n'
1137 <        txt += '       if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1138 <        txt += '           echo "SET_CMS_ENV 10020 ==> ERROR cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1139 <        txt += '           echo "JOB_EXIT_STATUS = 10020"\n'
1140 <        txt += '           echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1141 <        txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1142 <        txt += '           rm -f $RUNTIME_AREA/$repo \n'
1143 <        txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1144 <        txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1145 <        txt += '           exit 1\n'
1146 <        txt += '       fi\n'
1147 <        txt += '       echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1148 <        txt += '       source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1149 <        txt += '       result=$?\n'
1150 <        txt += '       if [ $result -ne 0 ]; then\n'
1151 <        txt += '           echo "SET_CMS_ENV 10032 ==> ERROR problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1152 <        txt += '           echo "JOB_EXIT_STATUS = 10032"\n'
1153 <        txt += '           echo "JobExitCode=10032" | tee -a $RUNTIME_AREA/$repo\n'
1154 <        txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1155 <        txt += '           rm -f $RUNTIME_AREA/$repo \n'
1156 <        txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1157 <        txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1158 <        txt += '           exit 1\n'
1159 <        txt += '       fi\n'
1160 <        txt += '   fi\n'
1161 <        txt += '   \n'
1162 <        txt += '   string=`cat /etc/redhat-release`\n'
1163 <        txt += '   echo $string\n'
1164 <        txt += '   if [[ $string = *alhalla* ]]; then\n'
1165 <        txt += '       echo "SCRAM_ARCH= $SCRAM_ARCH"\n'
1166 <        txt += '   elif [[ $string = *Enterprise* ]] || [[ $string = *cientific* ]]; then\n'
1167 <        txt += '       export SCRAM_ARCH=slc3_ia32_gcc323\n'
1168 <        txt += '       echo "SCRAM_ARCH= $SCRAM_ARCH"\n'
1169 <        txt += '   else\n'
1170 <        txt += '       echo "SET_CMS_ENV 10033 ==> ERROR OS unknown, LCG environment not initialized"\n'
1171 <        txt += '       echo "JOB_EXIT_STATUS = 10033"\n'
1172 <        txt += '       echo "JobExitCode=10033" | tee -a $RUNTIME_AREA/$repo\n'
1173 <        txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
1174 <        txt += '       rm -f $RUNTIME_AREA/$repo \n'
1175 <        txt += '       echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1176 <        txt += '       echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1177 <        txt += '       exit 1\n'
1178 <        txt += '   fi\n'
1179 <        txt += '   echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1180 <        txt += '   echo "### END SETUP CMS LCG ENVIRONMENT ###"\n'
1124 >        txt = '\n#Written by cms_cmssw::wsSetupCMSLCGEnvironment_\n'
1125 >        txt += '    echo ">>> setup CMS LCG environment:"\n'
1126 >        txt += '    echo "set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
1127 >        txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
1128 >        txt += '    export BUILD_ARCH='+self.executable_arch+'\n'
1129 >        txt += '    if [ ! $VO_CMS_SW_DIR ] ;then\n'
1130 >        txt += '        echo "ERROR ==> CMS software dir not found on WN `hostname`"\n'
1131 >        txt += '        job_exit_code=10031\n'
1132 >        txt += '        func_exit\n'
1133 >        txt += '    else\n'
1134 >        txt += '        echo "Sourcing environment... "\n'
1135 >        txt += '        if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1136 >        txt += '            echo "ERROR ==> cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1137 >        txt += '            job_exit_code=10020\n'
1138 >        txt += '            func_exit\n'
1139 >        txt += '        fi\n'
1140 >        txt += '        echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1141 >        txt += '        source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1142 >        txt += '        result=$?\n'
1143 >        txt += '        if [ $result -ne 0 ]; then\n'
1144 >        txt += '            echo "ERROR ==> problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1145 >        txt += '            job_exit_code=10032\n'
1146 >        txt += '            func_exit\n'
1147 >        txt += '        fi\n'
1148 >        txt += '    fi\n'
1149 >        txt += '    \n'
1150 >        txt += '    echo "==> setup cms environment ok"\n'
1151 >        return txt
1152 >
1153 >    def modifyReport(self, nj):
1154 >        """
1155 >        insert the part of the script that modifies the FrameworkJob Report
1156 >        """
1157 >        txt = '\n#Written by cms_cmssw::modifyReport\n'
1158 >        publish_data = int(self.cfg_params.get('USER.publish_data',0))
1159 >        if (publish_data == 1):
1160 >            processedDataset = self.cfg_params['USER.publish_data_name']
1161 >            LFNBaseName = LFNBase(processedDataset)
1162 >
1163 >            txt += 'if [ $copy_exit_status -eq 0 ]; then\n'
1164 >            txt += '    FOR_LFN=%s_${PSETHASH}/\n'%(LFNBaseName)
1165 >            txt += 'else\n'
1166 >            txt += '    FOR_LFN=/copy_problems/ \n'
1167 >            txt += '    SE=""\n'
1168 >            txt += '    SE_PATH=""\n'
1169 >            txt += 'fi\n'
1170 >
1171 >            txt += 'echo ">>> Modify Job Report:" \n'
1172 >            txt += 'chmod a+x $SOFTWARE_DIR/ProdCommon/ProdCommon/FwkJobRep/ModifyJobReport.py\n'
1173 >            txt += 'ProcessedDataset='+processedDataset+'\n'
1174 >            txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1175 >            txt += 'echo "SE = $SE"\n'
1176 >            txt += 'echo "SE_PATH = $SE_PATH"\n'
1177 >            txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1178 >            txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1179 >            txt += 'echo "$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 $SE_PATH"\n'
1180 >            txt += '$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 $SE_PATH\n'
1181 >            txt += 'modifyReport_result=$?\n'
1182 >            txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1183 >            txt += '    modifyReport_result=70500\n'
1184 >            txt += '    job_exit_code=$modifyReport_result\n'
1185 >            txt += '    echo "ModifyReportResult=$modifyReport_result" | tee -a $RUNTIME_AREA/$repo\n'
1186 >            txt += '    echo "WARNING: Problem with ModifyJobReport"\n'
1187 >            txt += 'else\n'
1188 >            txt += '    mv NewFrameworkJobReport.xml $RUNTIME_AREA/crab_fjr_$NJob.xml\n'
1189 >            txt += 'fi\n'
1190 >        return txt
1191 >
1192 >    def wsParseFJR(self):
1193 >        """
1194 >        Parse the FrameworkJobReport to obtain useful infos
1195 >        """
1196 >        txt = '\n#Written by cms_cmssw::wsParseFJR\n'
1197 >        txt += 'echo ">>> Parse FrameworkJobReport crab_fjr.xml"\n'
1198 >        txt += 'if [ -s $RUNTIME_AREA/crab_fjr_$NJob.xml ]; then\n'
1199 >        txt += '    if [ -s $RUNTIME_AREA/parseCrabFjr.py ]; then\n'
1200 >        txt += '        cmd_out=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --dashboard $MonitorID,$MonitorJobID '+self.debugWrap+'`\n'
1201 >        if self.debug_wrapper :
1202 >            txt += '        echo "Result of parsing the FrameworkJobReport crab_fjr.xml: $cmd_out"\n'
1203 >        txt += '        executable_exit_status=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --exitcode`\n'
1204 >        txt += '        if [ $executable_exit_status -eq 50115 ];then\n'
1205 >        txt += '            echo ">>> crab_fjr.xml contents: "\n'
1206 >        txt += '            cat $RUNTIME_AREA/crab_fjr_NJob.xml\n'
1207 >        txt += '            echo "Wrong FrameworkJobReport --> does not contain useful info. ExitStatus: $executable_exit_status"\n'
1208 >        txt += '        elif [ $executable_exit_status -eq -999 ];then\n'
1209 >        txt += '            echo "ExitStatus from FrameworkJobReport not available. not available. Using exit code of executable from command line."\n'
1210 >        txt += '        else\n'
1211 >        txt += '            echo "Extracted ExitStatus from FrameworkJobReport parsing output: $executable_exit_status"\n'
1212 >        txt += '        fi\n'
1213 >        txt += '    else\n'
1214 >        txt += '        echo "CRAB python script to parse CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1215 >        txt += '    fi\n'
1216 >          #### Patch to check input data reading for CMSSW16x Hopefully we-ll remove it asap
1217 >
1218 >        if self.datasetPath:
1219 >          # VERIFY PROCESSED DATA
1220 >            txt += '    if [ $executable_exit_status -eq 0 ];then\n'
1221 >            txt += '      echo ">>> Verify list of processed files:"\n'
1222 >            txt += '      echo $InputFiles |tr -d \'\\\\\' |tr \',\' \'\\n\'|tr -d \'"\' > input-files.txt\n'
1223 >            txt += '      python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --lfn > processed-files.txt\n'
1224 >            txt += '      cat input-files.txt  | sort | uniq > tmp.txt\n'
1225 >            txt += '      mv tmp.txt input-files.txt\n'
1226 >            txt += '      echo "cat input-files.txt"\n'
1227 >            txt += '      echo "----------------------"\n'
1228 >            txt += '      cat input-files.txt\n'
1229 >            txt += '      cat processed-files.txt | sort | uniq > tmp.txt\n'
1230 >            txt += '      mv tmp.txt processed-files.txt\n'
1231 >            txt += '      echo "----------------------"\n'
1232 >            txt += '      echo "cat processed-files.txt"\n'
1233 >            txt += '      echo "----------------------"\n'
1234 >            txt += '      cat processed-files.txt\n'
1235 >            txt += '      echo "----------------------"\n'
1236 >            txt += '      diff -q input-files.txt processed-files.txt\n'
1237 >            txt += '      fileverify_status=$?\n'
1238 >            txt += '      if [ $fileverify_status -ne 0 ]; then\n'
1239 >            txt += '         executable_exit_status=30001\n'
1240 >            txt += '         echo "ERROR ==> not all input files processed"\n'
1241 >            txt += '         echo "      ==> list of processed files from crab_fjr.xml differs from list in pset.cfg"\n'
1242 >            txt += '         echo "      ==> diff input-files.txt processed-files.txt"\n'
1243 >            txt += '      fi\n'
1244 >            txt += '    fi\n'
1245 >            txt += '\n'
1246 >        txt += 'else\n'
1247 >        txt += '    echo "CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1248 >        txt += 'fi\n'
1249 >        txt += '\n'
1250 >        txt += 'echo "ExeExitCode=$executable_exit_status" | tee -a $RUNTIME_AREA/$repo\n'
1251 >        txt += 'echo "EXECUTABLE_EXIT_STATUS = $executable_exit_status"\n'
1252 >        txt += 'job_exit_code=$executable_exit_status\n'
1253 >
1254          return txt
1255  
1256      def setParam_(self, param, value):
# Line 1085 | Line 1259 | class Cmssw(JobType):
1259      def getParams(self):
1260          return self._params
1261  
1088    def setTaskid_(self):
1089        self._taskId = self.cfg_params['taskId']
1090        
1091    def getTaskid(self):
1092        return self._taskId
1093
1094 #######################################################################
1262      def uniquelist(self, old):
1263          """
1264          remove duplicates from a list
# Line 1100 | Line 1267 | class Cmssw(JobType):
1267          for e in old:
1268              nd[e]=0
1269          return nd.keys()
1270 +
1271 +    def outList(self):
1272 +        """
1273 +        check the dimension of the output files
1274 +        """
1275 +        txt = ''
1276 +        txt += 'echo ">>> list of expected files on output sandbox"\n'
1277 +        listOutFiles = []
1278 +        stdout = 'CMSSW_$NJob.stdout'
1279 +        stderr = 'CMSSW_$NJob.stderr'
1280 +        if (self.return_data == 1):
1281 +            for file in (self.output_file+self.output_file_sandbox):
1282 +                listOutFiles.append(self.numberFile_(file, '$NJob'))
1283 +            listOutFiles.append(stdout)
1284 +            listOutFiles.append(stderr)
1285 +        else:
1286 +            for file in (self.output_file_sandbox):
1287 +                listOutFiles.append(self.numberFile_(file, '$NJob'))
1288 +            listOutFiles.append(stdout)
1289 +            listOutFiles.append(stderr)
1290 +        txt += 'echo "output files: '+string.join(listOutFiles,' ')+'"\n'
1291 +        txt += 'filesToCheck="'+string.join(listOutFiles,' ')+'"\n'
1292 +        txt += 'export filesToCheck\n'
1293 +        return txt

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines