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 1.73 by gutsche, Sun Apr 8 18:39:51 2007 UTC vs.
Revision 1.236 by spiga, Mon Sep 8 07:42:41 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 + from BlackWhiteListParser import SEBlackWhiteListParser
6   import common
6 import PsetManipulator  
7 import DataDiscovery
8 import DataDiscovery_DBS2
9 import DataLocation
7   import Scram
8 + from LFNBaseName import *
9  
10 < import os, string, re, shutil, glob
10 > import os, string, glob
11  
12   class Cmssw(JobType):
13 <    def __init__(self, cfg_params, ncjobs):
13 >    def __init__(self, cfg_params, ncjobs,skip_blocks, isNew):
14          JobType.__init__(self, 'CMSSW')
15          common.logger.debug(3,'CMSSW::__init__')
16 +        self.skip_blocks = skip_blocks
17 +
18 +        self.argsList = []
19  
19        # Marco.
20          self._params = {}
21          self.cfg_params = cfg_params
22 +        # init BlackWhiteListParser
23 +        self.blackWhiteListParser = SEBlackWhiteListParser(cfg_params)
24  
25 <        try:
26 <            self.MaxTarBallSize = float(self.cfg_params['EDG.maxtarballsize'])
27 <        except KeyError:
28 <            self.MaxTarBallSize = 100.0
25 >        ### Temporary patch to automatically skip the ISB size check:
26 >        server=self.cfg_params.get('CRAB.server_name',None)
27 >        size = 9.5
28 >        if server: size = 99999
29 >        ### D.S.
30 >        self.MaxTarBallSize = float(self.cfg_params.get('EDG.maxtarballsize',size))
31  
32          # number of jobs requested to be created, limit obj splitting
33          self.ncjobs = ncjobs
34  
35          log = common.logger
36 <        
36 >
37          self.scram = Scram.Scram(cfg_params)
38          self.additional_inbox_files = []
39          self.scriptExe = ''
# Line 37 | Line 41 | class Cmssw(JobType):
41          self.executable_arch = self.scram.getArch()
42          self.tgz_name = 'default.tgz'
43          self.scriptName = 'CMSSW.sh'
44 <        self.pset = ''      #scrip use case Da  
45 <        self.datasetPath = '' #scrip use case Da
44 >        self.pset = ''
45 >        self.datasetPath = ''
46  
47          # set FJR file name
48          self.fjrFileName = 'crab_fjr.xml'
49  
50          self.version = self.scram.getSWVersion()
51 <        common.taskDB.setDict('codeVersion',self.version)
52 <        self.setParam_('application', self.version)
51 >        version_array = self.version.split('_')
52 >        self.CMSSW_major = 0
53 >        self.CMSSW_minor = 0
54 >        self.CMSSW_patch = 0
55 >        try:
56 >            self.CMSSW_major = int(version_array[1])
57 >            self.CMSSW_minor = int(version_array[2])
58 >            self.CMSSW_patch = int(version_array[3])
59 >        except:
60 >            msg = "Cannot parse CMSSW version string: " + self.version + " for major and minor release number!"
61 >            raise CrabException(msg)
62  
63          ### collect Data cards
64  
65 <        ## get DBS mode
66 <        try:
54 <            self.use_dbs_2 = int(self.cfg_params['CMSSW.use_dbs_2'])
55 <        except KeyError:
56 <            self.use_dbs_2 = 0
57 <            
58 <        try:
59 <            tmp =  cfg_params['CMSSW.datasetpath']
60 <            log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
61 <            if string.lower(tmp)=='none':
62 <                self.datasetPath = None
63 <                self.selectNoInput = 1
64 <            else:
65 <                self.datasetPath = tmp
66 <                self.selectNoInput = 0
67 <        except KeyError:
68 <            msg = "Error: datasetpath not defined "  
65 >        if not cfg_params.has_key('CMSSW.datasetpath'):
66 >            msg = "Error: datasetpath not defined "
67              raise CrabException(msg)
68  
69 <        # ML monitoring
70 <        # split dataset path style: /PreProdR3Minbias/SIM/GEN-SIM
73 <        if not self.datasetPath:
74 <            self.setParam_('dataset', 'None')
75 <            self.setParam_('owner', 'None')
76 <        else:
77 <            datasetpath_split = self.datasetPath.split("/")
78 <            self.setParam_('dataset', datasetpath_split[1])
79 <            self.setParam_('owner', datasetpath_split[-1])
69 >        ### Temporary: added to remove input file control in the case of PU
70 >        self.dataset_pu = cfg_params.get('CMSSW.dataset_pu', None)
71  
72 <        self.setTaskid_()
73 <        self.setParam_('taskId', self.cfg_params['taskId'])
72 >        tmp =  cfg_params['CMSSW.datasetpath']
73 >        log.debug(6, "CMSSW::CMSSW(): datasetPath = "+tmp)
74  
75 <        self.dataTiers = []
75 >        if tmp =='':
76 >            msg = "Error: datasetpath not defined "
77 >            raise CrabException(msg)
78 >        elif string.lower(tmp)=='none':
79 >            self.datasetPath = None
80 >            self.selectNoInput = 1
81 >        else:
82 >            self.datasetPath = tmp
83 >            self.selectNoInput = 0
84  
85 +        self.dataTiers = []
86 +        self.debugWrap = ''
87 +        self.debug_wrapper = cfg_params.get('USER.debug_wrapper',False)
88 +        if self.debug_wrapper: self.debugWrap='--debug'
89          ## now the application
90 <        try:
91 <            self.executable = cfg_params['CMSSW.executable']
89 <            self.setParam_('exe', self.executable)
90 <            log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
91 <            msg = "Default executable cmsRun overridden. Switch to " + self.executable
92 <            log.debug(3,msg)
93 <        except KeyError:
94 <            self.executable = 'cmsRun'
95 <            self.setParam_('exe', self.executable)
96 <            msg = "User executable not defined. Use cmsRun"
97 <            log.debug(3,msg)
98 <            pass
90 >        self.executable = cfg_params.get('CMSSW.executable','cmsRun')
91 >        log.debug(6, "CMSSW::CMSSW(): executable = "+self.executable)
92  
93 <        try:
101 <            self.pset = cfg_params['CMSSW.pset']
102 <            log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
103 <            if self.pset.lower() != 'none' :
104 <                if (not os.path.exists(self.pset)):
105 <                    raise CrabException("User defined PSet file "+self.pset+" does not exist")
106 <            else:
107 <                self.pset = None
108 <        except KeyError:
93 >        if not cfg_params.has_key('CMSSW.pset'):
94              raise CrabException("PSet file missing. Cannot run cmsRun ")
95 +        self.pset = cfg_params['CMSSW.pset']
96 +        log.debug(6, "Cmssw::Cmssw(): PSet file = "+self.pset)
97 +        if self.pset.lower() != 'none' :
98 +            if (not os.path.exists(self.pset)):
99 +                raise CrabException("User defined PSet file "+self.pset+" does not exist")
100 +        else:
101 +            self.pset = None
102  
103          # output files
104          ## stuff which must be returned always via sandbox
# Line 116 | Line 108 | class Cmssw(JobType):
108          self.output_file_sandbox.append(self.fjrFileName)
109  
110          # other output files to be returned via sandbox or copied to SE
111 <        try:
112 <            self.output_file = []
113 <            tmp = cfg_params['CMSSW.output_file']
114 <            if tmp != '':
115 <                tmpOutFiles = string.split(cfg_params['CMSSW.output_file'],',')
116 <                log.debug(7, 'cmssw::cmssw(): output files '+str(tmpOutFiles))
117 <                for tmp in tmpOutFiles:
118 <                    tmp=string.strip(tmp)
127 <                    self.output_file.append(tmp)
128 <                    pass
129 <            else:
130 <                log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available")
131 <                pass
132 <            pass
133 <        except KeyError:
134 <            log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available")
135 <            pass
111 >        outfileflag = False
112 >        self.output_file = []
113 >        tmp = cfg_params.get('CMSSW.output_file',None)
114 >        if tmp :
115 >            self.output_file = [x.strip() for x in tmp.split(',')]
116 >            outfileflag = True #output found
117 >        #else:
118 >        #    log.message("No output file defined: only stdout/err and the CRAB Framework Job Report will be available\n")
119  
120          # script_exe file as additional file in inputSandbox
121 <        try:
122 <            self.scriptExe = cfg_params['USER.script_exe']
123 <            if self.scriptExe != '':
124 <               if not os.path.isfile(self.scriptExe):
125 <                  msg ="ERROR. file "+self.scriptExe+" not found"
126 <                  raise CrabException(msg)
144 <               self.additional_inbox_files.append(string.strip(self.scriptExe))
145 <        except KeyError:
146 <            self.scriptExe = ''
121 >        self.scriptExe = cfg_params.get('USER.script_exe',None)
122 >        if self.scriptExe :
123 >            if not os.path.isfile(self.scriptExe):
124 >                msg ="ERROR. file "+self.scriptExe+" not found"
125 >                raise CrabException(msg)
126 >            self.additional_inbox_files.append(string.strip(self.scriptExe))
127  
148        #CarlosDaniele
128          if self.datasetPath == None and self.pset == None and self.scriptExe == '' :
129 <           msg ="Error. script_exe  not defined"
130 <           raise CrabException(msg)
129 >            msg ="Error. script_exe  not defined"
130 >            raise CrabException(msg)
131 >
132 >        # use parent files...
133 >        self.useParent = self.cfg_params.get('CMSSW.use_parent',False)
134  
135          ## additional input files
136 <        try:
136 >        if cfg_params.has_key('USER.additional_input_files'):
137              tmpAddFiles = string.split(cfg_params['USER.additional_input_files'],',')
138              for tmp in tmpAddFiles:
139                  tmp = string.strip(tmp)
140                  dirname = ''
141                  if not tmp[0]=="/": dirname = "."
142 <                files = glob.glob(os.path.join(dirname, tmp))
142 >                files = []
143 >                if string.find(tmp,"*")>-1:
144 >                    files = glob.glob(os.path.join(dirname, tmp))
145 >                    if len(files)==0:
146 >                        raise CrabException("No additional input file found with this pattern: "+tmp)
147 >                else:
148 >                    files.append(tmp)
149                  for file in files:
150                      if not os.path.exists(file):
151                          raise CrabException("Additional input file not found: "+file)
152                      pass
153 <                    storedFile = common.work_space.shareDir()+file
166 <                    shutil.copyfile(file, storedFile)
167 <                    self.additional_inbox_files.append(string.strip(storedFile))
153 >                    self.additional_inbox_files.append(string.strip(file))
154                  pass
155              pass
156              common.logger.debug(5,"Additional input files: "+str(self.additional_inbox_files))
157 <        except KeyError:
172 <            pass
173 <
174 <        # files per job
175 <        try:
176 <            if (cfg_params['CMSSW.files_per_jobs']):
177 <                raise CrabException("files_per_jobs no longer supported.  Quitting.")
178 <        except KeyError:
179 <            pass
157 >        pass
158  
159          ## Events per job
160 <        try:
160 >        if cfg_params.has_key('CMSSW.events_per_job'):
161              self.eventsPerJob =int( cfg_params['CMSSW.events_per_job'])
162              self.selectEventsPerJob = 1
163 <        except KeyError:
163 >        else:
164              self.eventsPerJob = -1
165              self.selectEventsPerJob = 0
166 <    
166 >
167          ## number of jobs
168 <        try:
168 >        if cfg_params.has_key('CMSSW.number_of_jobs'):
169              self.theNumberOfJobs =int( cfg_params['CMSSW.number_of_jobs'])
170              self.selectNumberOfJobs = 1
171 <        except KeyError:
171 >        else:
172              self.theNumberOfJobs = 0
173              self.selectNumberOfJobs = 0
174  
175 <        try:
175 >        if cfg_params.has_key('CMSSW.total_number_of_events'):
176              self.total_number_of_events = int(cfg_params['CMSSW.total_number_of_events'])
177              self.selectTotalNumberEvents = 1
178 <        except KeyError:
178 >            if self.selectNumberOfJobs  == 1:
179 >                if (self.total_number_of_events != -1) and int(self.total_number_of_events) < int(self.theNumberOfJobs):
180 >                    msg = 'Must specify at least one event per job. total_number_of_events > number_of_jobs '
181 >                    raise CrabException(msg)
182 >        else:
183              self.total_number_of_events = 0
184              self.selectTotalNumberEvents = 0
185  
186 <        if self.pset != None: #CarlosDaniele
186 >        if self.pset != None:
187               if ( (self.selectTotalNumberEvents + self.selectEventsPerJob + self.selectNumberOfJobs) != 2 ):
188                   msg = 'Must define exactly two of total_number_of_events, events_per_job, or number_of_jobs.'
189                   raise CrabException(msg)
# Line 210 | Line 192 | class Cmssw(JobType):
192                   msg = 'Must specify  number_of_jobs.'
193                   raise CrabException(msg)
194  
195 <        ## source seed for pythia
196 <        try:
197 <            self.sourceSeed = int(cfg_params['CMSSW.pythia_seed'])
198 <        except KeyError:
199 <            self.sourceSeed = None
200 <            common.logger.debug(5,"No seed given")
195 >        ## New method of dealing with seeds
196 >        self.incrementSeeds = []
197 >        self.preserveSeeds = []
198 >        if cfg_params.has_key('CMSSW.preserve_seeds'):
199 >            tmpList = cfg_params['CMSSW.preserve_seeds'].split(',')
200 >            for tmp in tmpList:
201 >                tmp.strip()
202 >                self.preserveSeeds.append(tmp)
203 >        if cfg_params.has_key('CMSSW.increment_seeds'):
204 >            tmpList = cfg_params['CMSSW.increment_seeds'].split(',')
205 >            for tmp in tmpList:
206 >                tmp.strip()
207 >                self.incrementSeeds.append(tmp)
208 >
209 >        ## FUTURE: Can remove in CRAB 2.4.0
210 >        self.sourceSeed    = cfg_params.get('CMSSW.pythia_seed',None)
211 >        self.sourceSeedVtx = cfg_params.get('CMSSW.vtx_seed',None)
212 >        self.sourceSeedG4  = cfg_params.get('CMSSW.g4_seed',None)
213 >        self.sourceSeedMix = cfg_params.get('CMSSW.mix_seed',None)
214 >        if self.sourceSeed or self.sourceSeedVtx or self.sourceSeedG4 or self.sourceSeedMix:
215 >            msg = 'pythia_seed, vtx_seed, g4_seed, and mix_seed are no longer valid settings. You must use increment_seeds or preserve_seeds'
216 >            raise CrabException(msg)
217  
218 <        try:
219 <            self.sourceSeedVtx = int(cfg_params['CMSSW.vtx_seed'])
220 <        except KeyError:
221 <            self.sourceSeedVtx = None
222 <            common.logger.debug(5,"No vertex seed given")
225 <        try:
226 <            self.firstRun = int(cfg_params['CMSSW.first_run'])
227 <        except KeyError:
228 <            self.firstRun = None
229 <            common.logger.debug(5,"No first run given")
230 <        if self.pset != None: #CarlosDaniele
231 <            self.PsetEdit = PsetManipulator.PsetManipulator(self.pset) #Daniele Pset
218 >        self.firstRun = cfg_params.get('CMSSW.first_run',None)
219 >
220 >        # Copy/return
221 >        self.copy_data = int(cfg_params.get('USER.copy_data',0))
222 >        self.return_data = int(cfg_params.get('USER.return_data',0))
223  
224          #DBSDLS-start
225 <        ## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
225 >        ## Initialize the variables that are extracted from DBS/DLS and needed in other places of the code
226          self.maxEvents=0  # max events available   ( --> check the requested nb. of evts in Creator.py)
227          self.DBSPaths={}  # all dbs paths requested ( --> input to the site local discovery script)
228          self.jobDestination=[]  # Site destination(s) for each job (list of lists)
# Line 240 | Line 231 | class Cmssw(JobType):
231          blockSites = {}
232          if self.datasetPath:
233              blockSites = self.DataDiscoveryAndLocation(cfg_params)
234 <        #DBSDLS-end          
234 >        #DBSDLS-end
235  
245        self.tgzNameWithPath = self.getTarBall(self.executable)
246    
236          ## Select Splitting
237 <        if self.selectNoInput:
238 <            if self.pset == None: #CarlosDaniele
237 >        if self.selectNoInput:
238 >            if self.pset == None:
239                  self.jobSplittingForScript()
240              else:
241                  self.jobSplittingNoInput()
242 <        else:
242 >        else:
243              self.jobSplittingByBlocks(blockSites)
244  
245 <        # modify Pset
246 <        if self.pset != None: #CarlosDaniele
247 <            try:
248 <                if (self.datasetPath): # standard job
249 <                    # allow to processa a fraction of events in a file
250 <                    self.PsetEdit.inputModule("INPUT")
251 <                    self.PsetEdit.maxEvent("INPUTMAXEVENTS")
252 <                    self.PsetEdit.skipEvent("INPUTSKIPEVENTS")
253 <                else:  # pythia like job
254 <                    self.PsetEdit.maxEvent(self.eventsPerJob)
255 <                    if (self.firstRun):
256 <                        self.PsetEdit.pythiaFirstRun("INPUTFIRSTRUN")  #First Run
257 <                    if (self.sourceSeed) :
258 <                        self.PsetEdit.pythiaSeed("INPUT")
259 <                        if (self.sourceSeedVtx) :
260 <                            self.PsetEdit.pythiaSeedVtx("INPUTVTX")
261 <                # add FrameworkJobReport to parameter-set
262 <                self.PsetEdit.addCrabFJR(self.fjrFileName)
263 <                self.PsetEdit.psetWriter(self.configFilename())
264 <            except:
265 <                msg='Error while manipuliating ParameterSet: exiting...'
266 <                raise CrabException(msg)
245 >        # modify Pset only the first time
246 >        if isNew:
247 >            if self.pset != None:
248 >                import PsetManipulator as pp
249 >                PsetEdit = pp.PsetManipulator(self.pset)
250 >                try:
251 >                    # Add FrameworkJobReport to parameter-set, set max events.
252 >                    # Reset later for data jobs by writeCFG which does all modifications
253 >                    PsetEdit.addCrabFJR(self.fjrFileName) # FUTURE: Job report addition not needed by CMSSW>1.5
254 >                    PsetEdit.maxEvent(self.eventsPerJob)
255 >                    PsetEdit.psetWriter(self.configFilename())
256 >                    ## If present, add TFileService to output files
257 >                    if not int(cfg_params.get('CMSSW.skip_TFileService_output',0)):
258 >                        tfsOutput = PsetEdit.getTFileService()
259 >                        if tfsOutput:
260 >                            if tfsOutput in self.output_file:
261 >                                common.logger.debug(5,"Output from TFileService "+tfsOutput+" already in output files")
262 >                            else:
263 >                                outfileflag = True #output found
264 >                                self.output_file.append(tfsOutput)
265 >                                common.logger.message("Adding "+tfsOutput+" to output files (from TFileService)")
266 >                            pass
267 >                        pass
268 >                    ## If present and requested, add PoolOutputModule to output files
269 >                    if int(cfg_params.get('CMSSW.get_edm_output',0)):
270 >                        edmOutput = PsetEdit.getPoolOutputModule()
271 >                        if edmOutput:
272 >                            if edmOutput in self.output_file:
273 >                                common.logger.debug(5,"Output from PoolOutputModule "+edmOutput+" already in output files")
274 >                            else:
275 >                                self.output_file.append(edmOutput)
276 >                                common.logger.message("Adding "+edmOutput+" to output files (from PoolOutputModule)")
277 >                            pass
278 >                        pass
279 >                except CrabException:
280 >                    msg='Error while manipulating ParameterSet: exiting...'
281 >                    raise CrabException(msg)
282 >            ## Prepare inputSandbox TarBall (only the first time)
283 >            self.tgzNameWithPath = self.getTarBall(self.executable)
284  
285      def DataDiscoveryAndLocation(self, cfg_params):
286  
287 +        import DataDiscovery
288 +        import DataLocation
289          common.logger.debug(10,"CMSSW::DataDiscoveryAndLocation()")
290  
291          datasetPath=self.datasetPath
292  
293          ## Contact the DBS
294 <        common.logger.message("Contacting DBS...")
294 >        common.logger.message("Contacting Data Discovery Services ...")
295          try:
296 <
289 <            if self.use_dbs_2 == 1 :
290 <                self.pubdata=DataDiscovery_DBS2.DataDiscovery_DBS2(datasetPath, cfg_params)
291 <            else :
292 <                self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params)
296 >            self.pubdata=DataDiscovery.DataDiscovery(datasetPath, cfg_params,self.skip_blocks)
297              self.pubdata.fetchDBSInfo()
298  
299          except DataDiscovery.NotExistingDatasetError, ex :
# Line 301 | Line 305 | class Cmssw(JobType):
305          except DataDiscovery.DataDiscoveryError, ex:
306              msg = 'ERROR ***: failed Data Discovery in DBS :  %s'%ex.getErrorMessage()
307              raise CrabException(msg)
304        except DataDiscovery_DBS2.NotExistingDatasetError_DBS2, ex :
305            msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
306            raise CrabException(msg)
307        except DataDiscovery_DBS2.NoDataTierinProvenanceError_DBS2, ex :
308            msg = 'ERROR ***: failed Data Discovery in DBS : %s'%ex.getErrorMessage()
309            raise CrabException(msg)
310        except DataDiscovery_DBS2.DataDiscoveryError_DBS2, ex:
311            msg = 'ERROR ***: failed Data Discovery in DBS :  %s'%ex.getErrorMessage()
312            raise CrabException(msg)
313
314        ## get list of all required data in the form of dbs paths  (dbs path = /dataset/datatier/owner)
315        common.logger.message("Required data are :"+self.datasetPath)
308  
309          self.filesbyblock=self.pubdata.getFiles()
310          self.eventsbyblock=self.pubdata.getEventsPerBlock()
311          self.eventsbyfile=self.pubdata.getEventsPerFile()
312 +        self.parentFiles=self.pubdata.getParent()
313  
314          ## get max number of events
315 <        self.maxEvents=self.pubdata.getMaxEvents() ##  self.maxEvents used in Creator.py
323 <        common.logger.message("The number of available events is %s\n"%self.maxEvents)
315 >        self.maxEvents=self.pubdata.getMaxEvents()
316  
325        common.logger.message("Contacting DLS...")
317          ## Contact the DLS and build a list of sites hosting the fileblocks
318          try:
319              dataloc=DataLocation.DataLocation(self.filesbyblock.keys(),cfg_params)
# Line 330 | Line 321 | class Cmssw(JobType):
321          except DataLocation.DataLocationError , ex:
322              msg = 'ERROR ***: failed Data Location in DLS \n %s '%ex.getErrorMessage()
323              raise CrabException(msg)
324 <        
324 >
325  
326          sites = dataloc.getSites()
327          allSites = []
# Line 340 | Line 331 | class Cmssw(JobType):
331                  allSites.append(oneSite)
332          allSites = self.uniquelist(allSites)
333  
334 <        common.logger.message("Sites ("+str(len(allSites))+") hosting part/all of dataset: "+str(allSites))
335 <        common.logger.debug(6, "List of Sites: "+str(allSites))
334 >        # screen output
335 >        common.logger.message("Requested dataset: " + datasetPath + " has " + str(self.maxEvents) + " events in " + str(len(self.filesbyblock.keys())) + " blocks.\n")
336 >
337          return sites
338 <    
338 >
339      def jobSplittingByBlocks(self, blockSites):
340          """
341          Perform job splitting. Jobs run over an integer number of files
# Line 393 | Line 385 | class Cmssw(JobType):
385              totalNumberOfJobs = 999999999
386          else :
387              totalNumberOfJobs = self.ncjobs
396            
388  
389          blocks = blockSites.keys()
390          blockCount = 0
# Line 403 | Line 394 | class Cmssw(JobType):
394          jobCount = 0
395          list_of_lists = []
396  
397 +        # list tracking which jobs are in which jobs belong to which block
398 +        jobsOfBlock = {}
399 +
400          # ---- Iterate over the blocks in the dataset until ---- #
401          # ---- we've met the requested total # of events    ---- #
402          while ( (eventsRemaining > 0) and (blockCount < numBlocksInDataset) and (jobCount < totalNumberOfJobs)):
403              block = blocks[blockCount]
404              blockCount += 1
405 <            
405 >            if block not in jobsOfBlock.keys() :
406 >                jobsOfBlock[block] = []
407 >
408              if self.eventsbyblock.has_key(block) :
409                  numEventsInBlock = self.eventsbyblock[block]
410                  common.logger.debug(5,'Events in Block File '+str(numEventsInBlock))
411 <            
411 >
412                  files = self.filesbyblock[block]
413                  numFilesInBlock = len(files)
414                  if (numFilesInBlock <= 0):
# Line 420 | Line 416 | class Cmssw(JobType):
416                  fileCount = 0
417  
418                  # ---- New block => New job ---- #
419 <                parString = "\\{"
419 >                parString = ""
420                  # counter for number of events in files currently worked on
421                  filesEventCount = 0
422                  # flag if next while loop should touch new file
423                  newFile = 1
424                  # job event counter
425                  jobSkipEventCount = 0
426 <            
426 >
427                  # ---- Iterate over the files in the block until we've met the requested ---- #
428                  # ---- total # of events or we've gone over all the files in this block  ---- #
429 +                pString=''
430                  while ( (eventsRemaining > 0) and (fileCount < numFilesInBlock) and (jobCount < totalNumberOfJobs) ):
431                      file = files[fileCount]
432 +                    if self.useParent:
433 +                        parent = self.parentFiles[file]
434 +                        for f in parent :
435 +                            pString += '\\\"' + f + '\\\"\,'
436 +                        common.logger.debug(6, "File "+str(file)+" has the following parents: "+str(parent))
437 +                        common.logger.write("File "+str(file)+" has the following parents: "+str(parent))
438                      if newFile :
439                          try:
440                              numEventsInFile = self.eventsbyfile[file]
# Line 443 | Line 446 | class Cmssw(JobType):
446                              newFile = 0
447                          except KeyError:
448                              common.logger.message("File "+str(file)+" has unknown number of events: skipping")
446                        
449  
450 +                    eventsPerJobRequested = min(eventsPerJobRequested, eventsRemaining)
451                      # if less events in file remain than eventsPerJobRequested
452 <                    if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested ) :
452 >                    if ( filesEventCount - jobSkipEventCount < eventsPerJobRequested):
453                          # if last file in block
454                          if ( fileCount == numFilesInBlock-1 ) :
455                              # end job using last file, use remaining events in block
456                              # close job and touch new file
457                              fullString = parString[:-2]
458 <                            fullString += '\\}'
459 <                            list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
458 >                            if self.useParent:
459 >                                fullParentString = pString[:-2]
460 >                                list_of_lists.append([fullString,fullParentString,str(-1),str(jobSkipEventCount)])
461 >                            else:
462 >                                list_of_lists.append([fullString,str(-1),str(jobSkipEventCount)])
463                              common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(filesEventCount - jobSkipEventCount)+" events (last file in block).")
464                              self.jobDestination.append(blockSites[block])
465                              common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
466 +                            # fill jobs of block dictionary
467 +                            jobsOfBlock[block].append(jobCount+1)
468                              # reset counter
469                              jobCount = jobCount + 1
470                              totalEventCount = totalEventCount + filesEventCount - jobSkipEventCount
471                              eventsRemaining = eventsRemaining - filesEventCount + jobSkipEventCount
472                              jobSkipEventCount = 0
473                              # reset file
474 <                            parString = "\\{"
474 >                            pString = ""
475 >                            parString = ""
476                              filesEventCount = 0
477                              newFile = 1
478                              fileCount += 1
# Line 475 | Line 484 | class Cmssw(JobType):
484                      elif ( filesEventCount - jobSkipEventCount == eventsPerJobRequested ) :
485                          # close job and touch new file
486                          fullString = parString[:-2]
487 <                        fullString += '\\}'
488 <                        list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
487 >                        if self.useParent:
488 >                            fullParentString = pString[:-2]
489 >                            list_of_lists.append([fullString,fullParentString,str(eventsPerJobRequested),str(jobSkipEventCount)])
490 >                        else:
491 >                            list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
492                          common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
493                          self.jobDestination.append(blockSites[block])
494                          common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
495 +                        jobsOfBlock[block].append(jobCount+1)
496                          # reset counter
497                          jobCount = jobCount + 1
498                          totalEventCount = totalEventCount + eventsPerJobRequested
499                          eventsRemaining = eventsRemaining - eventsPerJobRequested
500                          jobSkipEventCount = 0
501                          # reset file
502 <                        parString = "\\{"
502 >                        pString = ""
503 >                        parString = ""
504                          filesEventCount = 0
505                          newFile = 1
506                          fileCount += 1
507 <                        
507 >
508                      # if more events in file remain than eventsPerJobRequested
509                      else :
510                          # close job but don't touch new file
511                          fullString = parString[:-2]
512 <                        fullString += '\\}'
513 <                        list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
512 >                        if self.useParent:
513 >                            fullParentString = pString[:-2]
514 >                            list_of_lists.append([fullString,fullParentString,str(eventsPerJobRequested),str(jobSkipEventCount)])
515 >                        else:
516 >                            list_of_lists.append([fullString,str(eventsPerJobRequested),str(jobSkipEventCount)])
517                          common.logger.debug(3,"Job "+str(jobCount+1)+" can run over "+str(eventsPerJobRequested)+" events.")
518                          self.jobDestination.append(blockSites[block])
519                          common.logger.debug(5,"Job "+str(jobCount+1)+" Destination: "+str(self.jobDestination[jobCount]))
520 +                        jobsOfBlock[block].append(jobCount+1)
521                          # increase counter
522                          jobCount = jobCount + 1
523                          totalEventCount = totalEventCount + eventsPerJobRequested
# Line 509 | Line 527 | class Cmssw(JobType):
527                          jobSkipEventCount = eventsPerJobRequested - (filesEventCount - jobSkipEventCount - self.eventsbyfile[file])
528                          # remove all but the last file
529                          filesEventCount = self.eventsbyfile[file]
530 <                        parString = "\\{"
531 <                        parString += '\\\"' + file + '\\\"\,'
530 >                        if self.useParent:
531 >                            for f in parent : pString += '\\\"' + f + '\\\"\,'
532 >                        parString = '\\\"' + file + '\\\"\,'
533                      pass # END if
534                  pass # END while (iterate over files in the block)
535          pass # END while (iterate over blocks in the dataset)
536          self.ncjobs = self.total_number_of_jobs = jobCount
537          if (eventsRemaining > 0 and jobCount < totalNumberOfJobs ):
538              common.logger.message("Could not run on all requested events because some blocks not hosted at allowed sites.")
539 <        common.logger.message("\n"+str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
540 <        
539 >        common.logger.message(str(jobCount)+" job(s) can run on "+str(totalEventCount)+" events.\n")
540 >
541 >        # screen output
542 >        screenOutput = "List of jobs and available destination sites:\n\n"
543 >
544 >        # keep trace of block with no sites to print a warning at the end
545 >        noSiteBlock = []
546 >        bloskNoSite = []
547 >
548 >        blockCounter = 0
549 >        for block in blocks:
550 >            if block in jobsOfBlock.keys() :
551 >                blockCounter += 1
552 >                screenOutput += "Block %5i: jobs %20s: sites: %s\n" % (blockCounter,spanRanges(jobsOfBlock[block]),
553 >                    ','.join(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)))
554 >                if len(self.blackWhiteListParser.checkWhiteList(self.blackWhiteListParser.checkBlackList(blockSites[block],block),block)) == 0:
555 >                    noSiteBlock.append( spanRanges(jobsOfBlock[block]) )
556 >                    bloskNoSite.append( blockCounter )
557 >
558 >        common.logger.message(screenOutput)
559 >        if len(noSiteBlock) > 0 and len(bloskNoSite) > 0:
560 >            msg = 'WARNING: No sites are hosting any part of data for block:\n                '
561 >            virgola = ""
562 >            if len(bloskNoSite) > 1:
563 >                virgola = ","
564 >            for block in bloskNoSite:
565 >                msg += ' ' + str(block) + virgola
566 >            msg += '\n               Related jobs:\n                 '
567 >            virgola = ""
568 >            if len(noSiteBlock) > 1:
569 >                virgola = ","
570 >            for range_jobs in noSiteBlock:
571 >                msg += str(range_jobs) + virgola
572 >            msg += '\n               will not be submitted and this block of data can not be analyzed!\n'
573 >            if self.cfg_params.has_key('EDG.se_white_list'):
574 >                msg += 'WARNING: SE White List: '+self.cfg_params['EDG.se_white_list']+'\n'
575 >                msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
576 >                msg += 'Please check if the dataset is available at this site!)\n'
577 >            if self.cfg_params.has_key('EDG.ce_white_list'):
578 >                msg += 'WARNING: CE White List: '+self.cfg_params['EDG.ce_white_list']+'\n'
579 >                msg += '(Hint: By whitelisting you force the job to run at this particular site(s).\n'
580 >                msg += 'Please check if the dataset is available at this site!)\n'
581 >
582 >            common.logger.message(msg)
583 >
584          self.list_of_args = list_of_lists
585          return
586  
# Line 527 | Line 589 | class Cmssw(JobType):
589          Perform job splitting based on number of event per job
590          """
591          common.logger.debug(5,'Splitting per events')
592 <        common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
593 <        common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
594 <        common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
592 >
593 >        if (self.selectEventsPerJob):
594 >            common.logger.message('Required '+str(self.eventsPerJob)+' events per job ')
595 >        if (self.selectNumberOfJobs):
596 >            common.logger.message('Required '+str(self.theNumberOfJobs)+' jobs in total ')
597 >        if (self.selectTotalNumberEvents):
598 >            common.logger.message('Required '+str(self.total_number_of_events)+' events in total ')
599  
600          if (self.total_number_of_events < 0):
601              msg='Cannot split jobs per Events with "-1" as total number of events'
# Line 538 | Line 604 | class Cmssw(JobType):
604          if (self.selectEventsPerJob):
605              if (self.selectTotalNumberEvents):
606                  self.total_number_of_jobs = int(self.total_number_of_events/self.eventsPerJob)
607 <            elif(self.selectNumberOfJobs) :  
607 >            elif(self.selectNumberOfJobs) :
608                  self.total_number_of_jobs =self.theNumberOfJobs
609 <                self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
609 >                self.total_number_of_events =int(self.theNumberOfJobs*self.eventsPerJob)
610  
611          elif (self.selectNumberOfJobs) :
612              self.total_number_of_jobs = self.theNumberOfJobs
613              self.eventsPerJob = int(self.total_number_of_events/self.total_number_of_jobs)
614 <
614 >
615          common.logger.debug(5,'N jobs  '+str(self.total_number_of_jobs))
616  
617          # is there any remainder?
# Line 561 | Line 627 | class Cmssw(JobType):
627          self.list_of_args = []
628          for i in range(self.total_number_of_jobs):
629              ## Since there is no input, any site is good
630 <           # self.jobDestination.append(["Any"])
631 <            self.jobDestination.append([""]) #must be empty to write correctly the xml
566 <            args=''
630 >            self.jobDestination.append([""]) #must be empty to write correctly the xml
631 >            args=[]
632              if (self.firstRun):
633 <                    ## pythia first run
634 <                #self.list_of_args.append([(str(self.firstRun)+str(i))])
635 <                args=args+(str(self.firstRun)+str(i))
571 <            else:
572 <                ## no first run
573 <                #self.list_of_args.append([str(i)])
574 <                args=args+str(i)
575 <            if (self.sourceSeed):
576 <                if (self.sourceSeedVtx):
577 <                    ## pythia + vtx random seed
578 <                    #self.list_of_args.append([
579 <                    #                          str(self.sourceSeed)+str(i),
580 <                    #                          str(self.sourceSeedVtx)+str(i)
581 <                    #                          ])
582 <                    args=args+str(',')+str(self.sourceSeed)+str(i)+str(',')+str(self.sourceSeedVtx)+str(i)
583 <                else:
584 <                    ## only pythia random seed
585 <                    #self.list_of_args.append([(str(self.sourceSeed)+str(i))])
586 <                    args=args +str(',')+str(self.sourceSeed)+str(i)
587 <            else:
588 <                ## no random seed
589 <                if str(args)=='': args=args+(str(self.firstRun)+str(i))
590 <            arguments=args.split(',')
591 <            if len(arguments)==3:self.list_of_args.append([str(arguments[0]),str(arguments[1]),str(arguments[2])])
592 <            elif len(arguments)==2:self.list_of_args.append([str(arguments[0]),str(arguments[1])])
593 <            else :self.list_of_args.append([str(arguments[0])])
594 <            
595 <     #   print self.list_of_args
633 >                ## pythia first run
634 >                args.append(str(self.firstRun)+str(i))
635 >            self.list_of_args.append(args)
636  
637          return
638  
639  
640 <    def jobSplittingForScript(self):#CarlosDaniele
640 >    def jobSplittingForScript(self):
641          """
642          Perform job splitting based on number of job
643          """
# Line 613 | Line 653 | class Cmssw(JobType):
653          # argument is seed number.$i
654          self.list_of_args = []
655          for i in range(self.total_number_of_jobs):
616            ## Since there is no input, any site is good
617           # self.jobDestination.append(["Any"])
656              self.jobDestination.append([""])
619            ## no random seed
657              self.list_of_args.append([str(i)])
658          return
659  
660 <    def split(self, jobParams):
661 <
625 <        common.jobDB.load()
626 <        #### Fabio
660 >    def split(self, jobParams,firstJobID):
661 >
662          njobs = self.total_number_of_jobs
663          arglist = self.list_of_args
664          # create the empty structure
665          for i in range(njobs):
666              jobParams.append("")
632        
633        for job in range(njobs):
634            jobParams[job] = arglist[job]
635            # print str(arglist[job])
636            # print jobParams[job]
637            common.jobDB.setArguments(job, jobParams[job])
638            common.logger.debug(5,"Job "+str(job)+" Destination: "+str(self.jobDestination[job]))
639            common.jobDB.setDestination(job, self.jobDestination[job])
667  
668 <        common.jobDB.save()
668 >        listID=[]
669 >        listField=[]
670 >        for id in range(njobs):
671 >            job = id + int(firstJobID)
672 >            jobParams[id] = arglist[id]
673 >            listID.append(job+1)
674 >            job_ToSave ={}
675 >            concString = ' '
676 >            argu=''
677 >            if len(jobParams[id]):
678 >                argu +=   concString.join(jobParams[id] )
679 >            job_ToSave['arguments']= str(job+1)+' '+argu
680 >            job_ToSave['dlsDestination']= self.jobDestination[id]
681 >            listField.append(job_ToSave)
682 >            msg="Job "+str(job)+" Arguments:   "+str(job+1)+" "+argu+"\n"  \
683 >            +"                     Destination: "+str(self.jobDestination[id])
684 >            common.logger.debug(5,msg)
685 >        common._db.updateJob_(listID,listField)
686 >        self.argsList = (len(jobParams[0])+1)
687 >
688          return
689 <    
644 <    def getJobTypeArguments(self, nj, sched):
645 <        result = ''
646 <        for i in common.jobDB.arguments(nj):
647 <            result=result+str(i)+" "
648 <        return result
649 <  
689 >
690      def numberOfJobs(self):
651        # Fabio
691          return self.total_number_of_jobs
692  
693      def getTarBall(self, exe):
694          """
695          Return the TarBall with lib and exe
696          """
658        
659        # if it exist, just return it
660        #
661        # Marco. Let's start to use relative path for Boss XML files
662        #
697          self.tgzNameWithPath = common.work_space.pathForTgz()+'share/'+self.tgz_name
698          if os.path.exists(self.tgzNameWithPath):
699              return self.tgzNameWithPath
# Line 673 | Line 707 | class Cmssw(JobType):
707  
708          # First of all declare the user Scram area
709          swArea = self.scram.getSWArea_()
676        #print "swArea = ", swArea
677        # swVersion = self.scram.getSWVersion()
678        # print "swVersion = ", swVersion
710          swReleaseTop = self.scram.getReleaseTop_()
711 <        #print "swReleaseTop = ", swReleaseTop
681 <        
711 >
712          ## check if working area is release top
713          if swReleaseTop == '' or swArea == swReleaseTop:
714 +            common.logger.debug(3,"swArea = "+swArea+" swReleaseTop ="+swReleaseTop)
715              return
716  
717          import tarfile
718          try: # create tar ball
719              tar = tarfile.open(self.tgzNameWithPath, "w:gz")
720              ## First find the executable
721 <            if (executable != ''):
721 >            if (self.executable != ''):
722                  exeWithPath = self.scram.findFile_(executable)
723                  if ( not exeWithPath ):
724                      raise CrabException('User executable '+executable+' not found')
725 <    
725 >
726                  ## then check if it's private or not
727                  if exeWithPath.find(swReleaseTop) == -1:
728                      # the exe is private, so we must ship
729                      common.logger.debug(5,"Exe "+exeWithPath+" to be tarred")
730                      path = swArea+'/'
731 <                    exe = string.replace(exeWithPath, path,'')
732 <                    tar.add(path+exe,executable)
731 >                    # distinguish case when script is in user project area or given by full path somewhere else
732 >                    if exeWithPath.find(path) >= 0 :
733 >                        exe = string.replace(exeWithPath, path,'')
734 >                        tar.add(path+exe,exe)
735 >                    else :
736 >                        tar.add(exeWithPath,os.path.basename(executable))
737                      pass
738                  else:
739                      # the exe is from release, we'll find it on WN
740                      pass
741 <    
741 >
742              ## Now get the libraries: only those in local working area
743              libDir = 'lib'
744              lib = swArea+'/' +libDir
745              common.logger.debug(5,"lib "+lib+" to be tarred")
746              if os.path.exists(lib):
747                  tar.add(lib,libDir)
748 <    
748 >
749              ## Now check if module dir is present
750              moduleDir = 'module'
751              module = swArea + '/' + moduleDir
# Line 718 | Line 753 | class Cmssw(JobType):
753                  tar.add(module,moduleDir)
754  
755              ## Now check if any data dir(s) is present
756 <            swAreaLen=len(swArea)
757 <            for root, dirs, files in os.walk(swArea):
758 <                if "data" in dirs:
759 <                    common.logger.debug(5,"data "+root+"/data"+" to be tarred")
760 <                    tar.add(root+"/data",root[swAreaLen:]+"/data")
761 <
762 <            ## Add ProdAgent dir to tar
763 <            paDir = 'ProdAgentApi'
764 <            pa = os.environ['CRABDIR'] + '/' + 'ProdAgentApi'
765 <            if os.path.isdir(pa):
766 <                tar.add(pa,paDir)
767 <        
756 >            self.dataExist = False
757 >            todo_list = [(i, i) for i in  os.listdir(swArea+"/src")]
758 >            while len(todo_list):
759 >                entry, name = todo_list.pop()
760 >                if name.startswith('crab_0_') or  name.startswith('.') or name == 'CVS':
761 >                    continue
762 >                if os.path.isdir(swArea+"/src/"+entry):
763 >                    entryPath = entry + '/'
764 >                    todo_list += [(entryPath + i, i) for i in  os.listdir(swArea+"/src/"+entry)]
765 >                    if name == 'data':
766 >                        self.dataExist=True
767 >                        common.logger.debug(5,"data "+entry+" to be tarred")
768 >                        tar.add(swArea+"/src/"+entry,"src/"+entry)
769 >                    pass
770 >                pass
771 >
772 >            ### CMSSW ParameterSet
773 >            if not self.pset is None:
774 >                cfg_file = common.work_space.jobDir()+self.configFilename()
775 >                tar.add(cfg_file,self.configFilename())
776 >                common.logger.debug(5,"File added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
777 >
778 >
779 >            ## Add ProdCommon dir to tar
780 >            prodcommonDir = './'
781 >            prodcommonPath = os.environ['CRABDIR'] + '/' + 'external/'
782 >            neededStuff = ['ProdCommon/__init__.py','ProdCommon/FwkJobRep', 'ProdCommon/CMSConfigTools','ProdCommon/Core','ProdCommon/MCPayloads', 'IMProv']
783 >            for file in neededStuff:
784 >                tar.add(prodcommonPath+file,prodcommonDir+file)
785 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
786 >
787 >            ##### ML stuff
788 >            ML_file_list=['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py']
789 >            path=os.environ['CRABDIR'] + '/python/'
790 >            for file in ML_file_list:
791 >                tar.add(path+file,file)
792 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
793 >
794 >            ##### Utils
795 >            Utils_file_list=['parseCrabFjr.py','writeCfg.py', 'fillCrabFjr.py']
796 >            for file in Utils_file_list:
797 >                tar.add(path+file,file)
798 >            common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
799 >
800 >            ##### AdditionalFiles
801 >            for file in self.additional_inbox_files:
802 >                tar.add(file,string.split(file,'/')[-1])
803              common.logger.debug(5,"Files added to "+self.tgzNameWithPath+" : "+str(tar.getnames()))
804 +
805              tar.close()
806 <        except :
807 <            raise CrabException('Could not create tar-ball')
806 >        except IOError:
807 >            raise CrabException('Could not create tar-ball '+self.tgzNameWithPath)
808 >        except tarfile.TarError:
809 >            raise CrabException('Could not create tar-ball '+self.tgzNameWithPath)
810  
811          ## check for tarball size
812          tarballinfo = os.stat(self.tgzNameWithPath)
# Line 741 | Line 814 | class Cmssw(JobType):
814              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.')
815  
816          ## create tar-ball with ML stuff
817 <        self.MLtgzfile =  common.work_space.pathForTgz()+'share/MLfiles.tgz'
818 <        try:
746 <            tar = tarfile.open(self.MLtgzfile, "w:gz")
747 <            path=os.environ['CRABDIR'] + '/python/'
748 <            for file in ['report.py', 'DashboardAPI.py', 'Logger.py', 'ProcInfo.py', 'apmon.py', 'parseCrabFjr.py']:
749 <                tar.add(path+file,file)
750 <            common.logger.debug(5,"Files added to "+self.MLtgzfile+" : "+str(tar.getnames()))
751 <            tar.close()
752 <        except :
753 <            raise CrabException('Could not create ML files tar-ball')
754 <        
755 <        return
756 <        
757 <    def wsSetupEnvironment(self, nj):
817 >
818 >    def wsSetupEnvironment(self, nj=0):
819          """
820          Returns part of a job script which prepares
821          the execution environment for the job 'nj'.
822          """
823 +        if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
824 +            psetName = 'pset.py'
825 +        else:
826 +            psetName = 'pset.cfg'
827          # Prepare JobType-independent part
828 <        txt = ''
829 <  
830 <        ## OLI_Daniele at this level  middleware already known
766 <
767 <        txt += 'if [ $middleware == LCG ]; then \n'
828 >        txt = '\n#Written by cms_cmssw::wsSetupEnvironment\n'
829 >        txt += 'echo ">>> setup environment"\n'
830 >        txt += 'if [ $middleware == LCG ]; then \n'
831          txt += self.wsSetupCMSLCGEnvironment_()
832          txt += 'elif [ $middleware == OSG ]; then\n'
833          txt += '    WORKING_DIR=`/bin/mktemp  -d $OSG_WN_TMP/cms_XXXXXXXXXXXX`\n'
834 <        txt += '    echo "Created working directory: $WORKING_DIR"\n'
835 <        txt += '    if [ ! -d $WORKING_DIR ] ;then\n'
836 <        txt += '        echo "SET_CMS_ENV 10016 ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
837 <        txt += '        echo "JOB_EXIT_STATUS = 10016"\n'
775 <        txt += '        echo "JobExitCode=10016" | tee -a $RUNTIME_AREA/$repo\n'
776 <        txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
777 <        txt += '        rm -f $RUNTIME_AREA/$repo \n'
778 <        txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
779 <        txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
780 <        txt += '        exit 1\n'
834 >        txt += '    if [ ! $? == 0 ] ;then\n'
835 >        txt += '        echo "ERROR ==> OSG $WORKING_DIR could not be created on WN `hostname`"\n'
836 >        txt += '        job_exit_code=10016\n'
837 >        txt += '        func_exit\n'
838          txt += '    fi\n'
839 +        txt += '    echo ">>> Created working directory: $WORKING_DIR"\n'
840          txt += '\n'
841          txt += '    echo "Change to working directory: $WORKING_DIR"\n'
842          txt += '    cd $WORKING_DIR\n'
843 <        txt += self.wsSetupCMSOSGEnvironment_()
843 >        txt += '    echo ">>> current directory (WORKING_DIR): $WORKING_DIR"\n'
844 >        txt += self.wsSetupCMSOSGEnvironment_()
845          txt += 'fi\n'
846  
847          # Prepare JobType-specific part
848          scram = self.scram.commandName()
849          txt += '\n\n'
850 <        txt += 'echo "### SPECIFIC JOB SETUP ENVIRONMENT ###"\n'
850 >        txt += 'echo ">>> specific cmssw setup environment:"\n'
851 >        txt += 'echo "CMSSW_VERSION =  '+self.version+'"\n'
852          txt += scram+' project CMSSW '+self.version+'\n'
853          txt += 'status=$?\n'
854          txt += 'if [ $status != 0 ] ; then\n'
855 <        txt += '   echo "SET_EXE_ENV 10034 ==>ERROR CMSSW '+self.version+' not found on `hostname`" \n'
856 <        txt += '   echo "JOB_EXIT_STATUS = 10034"\n'
857 <        txt += '   echo "JobExitCode=10034" | tee -a $RUNTIME_AREA/$repo\n'
798 <        txt += '   dumpStatus $RUNTIME_AREA/$repo\n'
799 <        txt += '   rm -f $RUNTIME_AREA/$repo \n'
800 <        txt += '   echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
801 <        txt += '   echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
802 <        ## OLI_Daniele
803 <        txt += '    if [ $middleware == OSG ]; then \n'
804 <        txt += '        echo "Remove working directory: $WORKING_DIR"\n'
805 <        txt += '        cd $RUNTIME_AREA\n'
806 <        txt += '        /bin/rm -rf $WORKING_DIR\n'
807 <        txt += '        if [ -d $WORKING_DIR ] ;then\n'
808 <        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'
809 <        txt += '            echo "JOB_EXIT_STATUS = 10018"\n'
810 <        txt += '            echo "JobExitCode=10018" | tee -a $RUNTIME_AREA/$repo\n'
811 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
812 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
813 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
814 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
815 <        txt += '        fi\n'
816 <        txt += '    fi \n'
817 <        txt += '   exit 1 \n'
855 >        txt += '    echo "ERROR ==> CMSSW '+self.version+' not found on `hostname`" \n'
856 >        txt += '    job_exit_code=10034\n'
857 >        txt += '    func_exit\n'
858          txt += 'fi \n'
819        txt += 'echo "CMSSW_VERSION =  '+self.version+'"\n'
820        txt += 'export SCRAM_ARCH='+self.executable_arch+'\n'
859          txt += 'cd '+self.version+'\n'
860 <        ### needed grep for bug in scramv1 ###
861 <        txt += scram+' runtime -sh\n'
860 >        txt += 'SOFTWARE_DIR=`pwd`\n'
861 >        txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
862          txt += 'eval `'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME`\n'
863 <        txt += 'echo $PATH\n'
864 <
863 >        txt += 'if [ $? != 0 ] ; then\n'
864 >        txt += '    echo "ERROR ==> Problem with the command: "\n'
865 >        txt += '    echo "eval \`'+scram+' runtime -sh | grep -v SCRAMRT_LSB_JOBNAME \` at `hostname`"\n'
866 >        txt += '    job_exit_code=10034\n'
867 >        txt += '    func_exit\n'
868 >        txt += 'fi \n'
869          # Handle the arguments:
870          txt += "\n"
871          txt += "## number of arguments (first argument always jobnumber)\n"
872          txt += "\n"
873 < #        txt += "narg=$#\n"
832 <        txt += "if [ $nargs -lt 2 ]\n"
873 >        txt += "if [ $nargs -lt "+str(self.argsList)+" ]\n"
874          txt += "then\n"
875 <        txt += "    echo 'SET_EXE_ENV 1 ==> ERROR Too few arguments' +$nargs+ \n"
876 <        txt += '    echo "JOB_EXIT_STATUS = 50113"\n'
877 <        txt += '    echo "JobExitCode=50113" | tee -a $RUNTIME_AREA/$repo\n'
837 <        txt += '    dumpStatus $RUNTIME_AREA/$repo\n'
838 <        txt += '    rm -f $RUNTIME_AREA/$repo \n'
839 <        txt += '    echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
840 <        txt += '    echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
841 <        ## OLI_Daniele
842 <        txt += '    if [ $middleware == OSG ]; then \n'
843 <        txt += '        echo "Remove working directory: $WORKING_DIR"\n'
844 <        txt += '        cd $RUNTIME_AREA\n'
845 <        txt += '        /bin/rm -rf $WORKING_DIR\n'
846 <        txt += '        if [ -d $WORKING_DIR ] ;then\n'
847 <        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'
848 <        txt += '            echo "JOB_EXIT_STATUS = 50114"\n'
849 <        txt += '            echo "JobExitCode=50114" | tee -a $RUNTIME_AREA/$repo\n'
850 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
851 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
852 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
853 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
854 <        txt += '        fi\n'
855 <        txt += '    fi \n'
856 <        txt += "    exit 1\n"
875 >        txt += "    echo 'ERROR ==> Too few arguments' +$nargs+ \n"
876 >        txt += '    job_exit_code=50113\n'
877 >        txt += "    func_exit\n"
878          txt += "fi\n"
879          txt += "\n"
880  
881          # Prepare job-specific part
882          job = common.job_list[nj]
883 <        if self.pset != None: #CarlosDaniele
883 >        if (self.datasetPath):
884 >            txt += '\n'
885 >            txt += 'DatasetPath='+self.datasetPath+'\n'
886 >
887 >            datasetpath_split = self.datasetPath.split("/")
888 >            ### FEDE FOR NEW LFN ###
889 >            self.primaryDataset = datasetpath_split[1]
890 >            ########################
891 >            txt += 'PrimaryDataset='+datasetpath_split[1]+'\n'
892 >            txt += 'DataTier='+datasetpath_split[2]+'\n'
893 >            txt += 'ApplicationFamily=cmsRun\n'
894 >
895 >        else:
896 >            txt += 'DatasetPath=MCDataTier\n'
897 >            ### FEDE FOR NEW LFN ###
898 >            self.primaryDataset = 'null'
899 >            ########################
900 >            txt += 'PrimaryDataset=null\n'
901 >            txt += 'DataTier=null\n'
902 >            txt += 'ApplicationFamily=MCDataTier\n'
903 >        if self.pset != None:
904              pset = os.path.basename(job.configFilename())
905              txt += '\n'
906 +            txt += 'cp  $RUNTIME_AREA/'+pset+' .\n'
907              if (self.datasetPath): # standard job
908 <                #txt += 'InputFiles=$2\n'
909 <                txt += 'InputFiles=${args[1]}\n'
910 <                txt += 'MaxEvents=${args[2]}\n'
911 <                txt += 'SkipEvents=${args[3]}\n'
908 >                txt += 'InputFiles=${args[1]}; export InputFiles\n'
909 >                if (self.useParent):
910 >                    txt += 'ParentFiles=${args[2]}; export ParentFiles\n'
911 >                    txt += 'MaxEvents=${args[3]}; export MaxEvents\n'
912 >                    txt += 'SkipEvents=${args[4]}; export SkipEvents\n'
913 >                else:
914 >                    txt += 'MaxEvents=${args[2]}; export MaxEvents\n'
915 >                    txt += 'SkipEvents=${args[3]}; export SkipEvents\n'
916                  txt += 'echo "Inputfiles:<$InputFiles>"\n'
917 <                txt += 'sed "s#{\'INPUT\'}#$InputFiles#" $RUNTIME_AREA/'+pset+' > pset_tmp_1.cfg\n'
917 >                if (self.useParent): txt += 'echo "ParentFiles:<$ParentFiles>"\n'
918                  txt += 'echo "MaxEvents:<$MaxEvents>"\n'
873                txt += 'sed "s#INPUTMAXEVENTS#$MaxEvents#" pset_tmp_1.cfg > pset_tmp_2.cfg\n'
919                  txt += 'echo "SkipEvents:<$SkipEvents>"\n'
875                txt += 'sed "s#INPUTSKIPEVENTS#$SkipEvents#" pset_tmp_2.cfg > pset.cfg\n'
920              else:  # pythia like job
921 <                if (self.sourceSeed):
922 <                    txt += 'FirstRun=${args[1]}\n'
921 >                txt += 'PreserveSeeds='  + ','.join(self.preserveSeeds)  + '; export PreserveSeeds\n'
922 >                txt += 'IncrementSeeds=' + ','.join(self.incrementSeeds) + '; export IncrementSeeds\n'
923 >                txt += 'echo "PreserveSeeds: <$PreserveSeeds>"\n'
924 >                txt += 'echo "IncrementSeeds:<$IncrementSeeds>"\n'
925 >                if (self.firstRun):
926 >                    txt += 'FirstRun=${args[1]}; export FirstRun\n'
927                      txt += 'echo "FirstRun: <$FirstRun>"\n'
880                    txt += 'sed "s#\<INPUTFIRSTRUN\>#$FirstRun#" $RUNTIME_AREA/'+pset+' > tmp_1.cfg\n'
881                else:
882                    txt += '# Copy untouched pset\n'
883                    txt += 'cp $RUNTIME_AREA/'+pset+' tmp_1.cfg\n'
884                if (self.sourceSeed):
885 #                    txt += 'Seed=$2\n'
886                    txt += 'Seed=${args[2]}\n'
887                    txt += 'echo "Seed: <$Seed>"\n'
888                    txt += 'sed "s#\<INPUT\>#$Seed#" tmp_1.cfg > tmp_2.cfg\n'
889                    if (self.sourceSeedVtx):
890 #                        txt += 'VtxSeed=$3\n'
891                        txt += 'VtxSeed=${args[3]}\n'
892                        txt += 'echo "VtxSeed: <$VtxSeed>"\n'
893                        txt += 'sed "s#INPUTVTX#$VtxSeed#" tmp_2.cfg > pset.cfg\n'
894                    else:
895                        txt += 'mv tmp_2.cfg pset.cfg\n'
896                else:
897                    txt += 'mv tmp_1.cfg pset.cfg\n'
898                   # txt += '# Copy untouched pset\n'
899                   # txt += 'cp $RUNTIME_AREA/'+pset+' pset.cfg\n'
928  
929 +            txt += 'mv -f ' + pset + ' ' + psetName + '\n'
930  
931 <        if len(self.additional_inbox_files) > 0:
932 <            for file in self.additional_inbox_files:
933 <                relFile = file.split("/")[-1]
905 <                txt += 'if [ -e $RUNTIME_AREA/'+relFile+' ] ; then\n'
906 <                txt += '   cp $RUNTIME_AREA/'+relFile+' .\n'
907 <                txt += '   chmod +x '+relFile+'\n'
908 <                txt += 'fi\n'
909 <            pass
910 <
911 <        if self.pset != None: #CarlosDaniele
912 <            txt += 'echo "### END JOB SETUP ENVIRONMENT ###"\n\n'
913 <        
931 >
932 >        if self.pset != None:
933 >            # FUTURE: Can simply for 2_1_x and higher
934              txt += '\n'
935 <            txt += 'echo "***** cat pset.cfg *********"\n'
936 <            txt += 'cat pset.cfg\n'
937 <            txt += 'echo "****** end pset.cfg ********"\n'
935 >            if self.debug_wrapper==True:
936 >                txt += 'echo "***** cat ' + psetName + ' *********"\n'
937 >                txt += 'cat ' + psetName + '\n'
938 >                txt += 'echo "****** end ' + psetName + ' ********"\n'
939 >                txt += '\n'
940 >            if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
941 >                txt += 'PSETHASH=`edmConfigHash ' + psetName + '` \n'
942 >            else:
943 >                txt += 'PSETHASH=`edmConfigHash < ' + psetName + '` \n'
944 >            txt += 'echo "PSETHASH = $PSETHASH" \n'
945              txt += '\n'
919            # txt += 'echo "***** cat pset1.cfg *********"\n'
920            # txt += 'cat pset1.cfg\n'
921            # txt += 'echo "****** end pset1.cfg ********"\n'
946          return txt
947  
948 <    def wsBuildExe(self, nj=0):
948 >    def wsUntarSoftware(self, nj=0):
949          """
950          Put in the script the commands to build an executable
951          or a library.
952          """
953  
954 <        txt = ""
954 >        txt = '\n#Written by cms_cmssw::wsUntarSoftware\n'
955  
956          if os.path.isfile(self.tgzNameWithPath):
957 <            txt += 'echo "tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'"\n'
957 >            txt += 'echo ">>> tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+' :" \n'
958              txt += 'tar xzvf $RUNTIME_AREA/'+os.path.basename(self.tgzNameWithPath)+'\n'
959 +            if  self.debug_wrapper:
960 +                txt += 'ls -Al \n'
961              txt += 'untar_status=$? \n'
962              txt += 'if [ $untar_status -ne 0 ]; then \n'
963 <            txt += '   echo "SET_EXE 1 ==> ERROR Untarring .tgz file failed"\n'
964 <            txt += '   echo "JOB_EXIT_STATUS = $untar_status" \n'
965 <            txt += '   echo "JobExitCode=$untar_status" | tee -a $RUNTIME_AREA/$repo\n'
940 <            txt += '   if [ $middleware == OSG ]; then \n'
941 <            txt += '       echo "Remove working directory: $WORKING_DIR"\n'
942 <            txt += '       cd $RUNTIME_AREA\n'
943 <            txt += '       /bin/rm -rf $WORKING_DIR\n'
944 <            txt += '       if [ -d $WORKING_DIR ] ;then\n'
945 <            txt += '           echo "SET_EXE 50999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after Untarring .tgz file failed"\n'
946 <            txt += '           echo "JOB_EXIT_STATUS = 50999"\n'
947 <            txt += '           echo "JobExitCode=50999" | tee -a $RUNTIME_AREA/$repo\n'
948 <            txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
949 <            txt += '           rm -f $RUNTIME_AREA/$repo \n'
950 <            txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
951 <            txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
952 <            txt += '       fi\n'
953 <            txt += '   fi \n'
954 <            txt += '   \n'
955 <            txt += '   exit 1 \n'
963 >            txt += '   echo "ERROR ==> Untarring .tgz file failed"\n'
964 >            txt += '   job_exit_code=$untar_status\n'
965 >            txt += '   func_exit\n'
966              txt += 'else \n'
967              txt += '   echo "Successful untar" \n'
968              txt += 'fi \n'
969              txt += '\n'
970 <            txt += 'echo "Include ProdAgentApi in PYTHONPATH"\n'
970 >            txt += 'echo ">>> Include $RUNTIME_AREA in PYTHONPATH:"\n'
971              txt += 'if [ -z "$PYTHONPATH" ]; then\n'
972 <            txt += '   export PYTHONPATH=ProdAgentApi\n'
972 >            txt += '   export PYTHONPATH=$RUNTIME_AREA/\n'
973              txt += 'else\n'
974 <            txt += '   export PYTHONPATH=ProdAgentApi:${PYTHONPATH}\n'
974 >            txt += '   export PYTHONPATH=$RUNTIME_AREA/:${PYTHONPATH}\n'
975 >            txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
976              txt += 'fi\n'
977              txt += '\n'
978  
979              pass
980 <        
980 >
981          return txt
982  
983 <    def modifySteeringCards(self, nj):
983 >    def wsBuildExe(self, nj=0):
984          """
985 <        modify the card provided by the user,
986 <        writing a new card into share dir
985 >        Put in the script the commands to build an executable
986 >        or a library.
987          """
988 <        
988 >
989 >        txt = '\n#Written by cms_cmssw::wsBuildExe\n'
990 >        txt += 'echo ">>> moving CMSSW software directories in `pwd`" \n'
991 >
992 >        txt += 'rm -r lib/ module/ \n'
993 >        txt += 'mv $RUNTIME_AREA/lib/ . \n'
994 >        txt += 'mv $RUNTIME_AREA/module/ . \n'
995 >        if self.dataExist == True:
996 >            txt += 'rm -r src/ \n'
997 >            txt += 'mv $RUNTIME_AREA/src/ . \n'
998 >        if len(self.additional_inbox_files)>0:
999 >            for file in self.additional_inbox_files:
1000 >                txt += 'mv $RUNTIME_AREA/'+os.path.basename(file)+' . \n'
1001 >        # txt += 'mv $RUNTIME_AREA/ProdCommon/ . \n'
1002 >        # txt += 'mv $RUNTIME_AREA/IMProv/ . \n'
1003 >
1004 >        txt += 'echo ">>> Include $RUNTIME_AREA in PYTHONPATH:"\n'
1005 >        txt += 'if [ -z "$PYTHONPATH" ]; then\n'
1006 >        txt += '   export PYTHONPATH=$RUNTIME_AREA/\n'
1007 >        txt += 'else\n'
1008 >        txt += '   export PYTHONPATH=$RUNTIME_AREA/:${PYTHONPATH}\n'
1009 >        txt += 'echo "PYTHONPATH=$PYTHONPATH"\n'
1010 >        txt += 'fi\n'
1011 >        txt += '\n'
1012 >
1013 >        return txt
1014 >
1015 >
1016      def executableName(self):
1017 <        if self.scriptExe: #CarlosDaniele
1017 >        if self.scriptExe:
1018              return "sh "
1019          else:
1020              return self.executable
1021  
1022      def executableArgs(self):
1023 +        # FUTURE: This function tests the CMSSW version. Can be simplified as we drop support for old versions
1024          if self.scriptExe:#CarlosDaniele
1025              return   self.scriptExe + " $NJob"
1026 <        else:
1027 <            return " -p pset.cfg"
1026 >        else:
1027 >            ex_args = ""
1028 >            # FUTURE: This tests the CMSSW version. Can remove code as versions deprecated
1029 >            # Framework job report
1030 >            if (self.CMSSW_major >= 1 and self.CMSSW_minor >= 5) or (self.CMSSW_major >= 2):
1031 >                ex_args += " -j $RUNTIME_AREA/crab_fjr_$NJob.xml"
1032 >            # Type of config file
1033 >            if self.CMSSW_major >= 2 :
1034 >                ex_args += " -p pset.py"
1035 >            else:
1036 >                ex_args += " -p pset.cfg"
1037 >            return ex_args
1038  
1039      def inputSandbox(self, nj):
1040          """
1041          Returns a list of filenames to be put in JDL input sandbox.
1042          """
1043          inp_box = []
995        # # dict added to delete duplicate from input sandbox file list
996        # seen = {}
997        ## code
1044          if os.path.isfile(self.tgzNameWithPath):
1045              inp_box.append(self.tgzNameWithPath)
1046 <        if os.path.isfile(self.MLtgzfile):
1047 <            inp_box.append(self.MLtgzfile)
1002 <        ## config
1003 <        if not self.pset is None:
1004 <            inp_box.append(common.work_space.pathForTgz() + 'job/' + self.configFilename())
1005 <        ## additional input files
1006 <        for file in self.additional_inbox_files:
1007 <            inp_box.append(file)
1046 >        wrapper = os.path.basename(str(common._db.queryTask('scriptName')))
1047 >        inp_box.append(common.work_space.pathForTgz() +'job/'+ wrapper)
1048          return inp_box
1049  
1050      def outputSandbox(self, nj):
# Line 1015 | Line 1055 | class Cmssw(JobType):
1055  
1056          ## User Declared output files
1057          for out in (self.output_file+self.output_file_sandbox):
1058 <            n_out = nj + 1
1059 <            out_box.append(self.numberFile_(out,str(n_out)))
1058 >            n_out = nj + 1
1059 >            out_box.append(numberFile(out,str(n_out)))
1060          return out_box
1061  
1022    def prepareSteeringCards(self):
1023        """
1024        Make initial modifications of the user's steering card file.
1025        """
1026        return
1062  
1063      def wsRenameOutput(self, nj):
1064          """
1065          Returns part of a job script which renames the produced files.
1066          """
1067  
1068 <        txt = '\n'
1069 <        txt += '# directory content\n'
1070 <        txt += 'ls \n'
1068 >        txt = '\n#Written by cms_cmssw::wsRenameOutput\n'
1069 >        txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1070 >        txt += 'echo ">>> current directory content:"\n'
1071 >        if self.debug_wrapper:
1072 >            txt += 'ls -Al\n'
1073 >        txt += '\n'
1074  
1075 <        for fileWithSuffix in (self.output_file+self.output_file_sandbox):
1076 <            output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1075 >        for fileWithSuffix in (self.output_file):
1076 >            output_file_num = numberFile(fileWithSuffix, '$NJob')
1077              txt += '\n'
1078              txt += '# check output file\n'
1079 <            txt += 'ls '+fileWithSuffix+'\n'
1080 <            txt += 'ls_result=$?\n'
1081 <            txt += 'if [ $ls_result -ne 0 ] ; then\n'
1082 <            txt += '   echo "ERROR: Problem with output file"\n'
1083 <            if common.scheduler.boss_scheduler_name == 'condor_g':
1079 >            txt += 'if [ -e ./'+fileWithSuffix+' ] ; then\n'
1080 >            if (self.copy_data == 1):  # For OSG nodes, file is in $WORKING_DIR, should not be moved to $RUNTIME_AREA
1081 >                txt += '    mv '+fileWithSuffix+' '+output_file_num+'\n'
1082 >                txt += '    ln -s `pwd`/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1083 >            else:
1084 >                txt += '    mv '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1085 >                txt += '    ln -s $RUNTIME_AREA/'+output_file_num+' $RUNTIME_AREA/'+fileWithSuffix+'\n'
1086 >            txt += 'else\n'
1087 >            txt += '    job_exit_code=60302\n'
1088 >            txt += '    echo "WARNING: Output file '+fileWithSuffix+' not found"\n'
1089 >            if common.scheduler.name().upper() == 'CONDOR_G':
1090                  txt += '    if [ $middleware == OSG ]; then \n'
1091                  txt += '        echo "prepare dummy output file"\n'
1092                  txt += '        echo "Processing of job output failed" > $RUNTIME_AREA/'+output_file_num+'\n'
1093                  txt += '    fi \n'
1050            txt += 'else\n'
1051            txt += '   cp '+fileWithSuffix+' $RUNTIME_AREA/'+output_file_num+'\n'
1094              txt += 'fi\n'
1095 <      
1096 <        txt += 'cd $RUNTIME_AREA\n'
1097 <        txt += 'cd $RUNTIME_AREA\n'
1056 <        ### OLI_DANIELE
1057 <        txt += 'if [ $middleware == OSG ]; then\n'  
1058 <        txt += '    cd $RUNTIME_AREA\n'
1059 <        txt += '    echo "Remove working directory: $WORKING_DIR"\n'
1060 <        txt += '    /bin/rm -rf $WORKING_DIR\n'
1061 <        txt += '    if [ -d $WORKING_DIR ] ;then\n'
1062 <        txt += '        echo "SET_EXE 60999 ==> OSG $WORKING_DIR could not be deleted on WN `hostname` after cleanup of WN"\n'
1063 <        txt += '        echo "JOB_EXIT_STATUS = 60999"\n'
1064 <        txt += '        echo "JobExitCode=60999" | tee -a $RUNTIME_AREA/$repo\n'
1065 <        txt += '        dumpStatus $RUNTIME_AREA/$repo\n'
1066 <        txt += '        rm -f $RUNTIME_AREA/$repo \n'
1067 <        txt += '        echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1068 <        txt += '        echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1069 <        txt += '    fi\n'
1070 <        txt += 'fi\n'
1071 <        txt += '\n'
1072 <
1073 <        file_list = ''
1074 <        ## Add to filelist only files to be possibly copied to SE
1075 <        for fileWithSuffix in self.output_file:
1076 <            output_file_num = self.numberFile_(fileWithSuffix, '$NJob')
1077 <            file_list=file_list+output_file_num+' '
1078 <        file_list=file_list[:-1]
1079 <        txt += 'file_list="'+file_list+'"\n'
1095 >        file_list = []
1096 >        for fileWithSuffix in (self.output_file):
1097 >             file_list.append(numberFile(fileWithSuffix, '$NJob'))
1098  
1099 +        txt += 'file_list="'+string.join(file_list,' ')+'"\n'
1100 +        txt += '\n'
1101 +        txt += 'echo ">>> current directory (SOFTWARE_DIR): $SOFTWARE_DIR" \n'
1102 +        txt += 'echo ">>> current directory content:"\n'
1103 +        if self.debug_wrapper:
1104 +            txt += 'ls -Al\n'
1105 +        txt += '\n'
1106 +        txt += 'cd $RUNTIME_AREA\n'
1107 +        txt += 'echo ">>> current directory (RUNTIME_AREA):  $RUNTIME_AREA"\n'
1108          return txt
1109  
1083    def numberFile_(self, file, txt):
1084        """
1085        append _'txt' before last extension of a file
1086        """
1087        p = string.split(file,".")
1088        # take away last extension
1089        name = p[0]
1090        for x in p[1:-1]:
1091           name=name+"."+x
1092        # add "_txt"
1093        if len(p)>1:
1094          ext = p[len(p)-1]
1095          result = name + '_' + txt + "." + ext
1096        else:
1097          result = name + '_' + txt
1098        
1099        return result
1100
1110      def getRequirements(self, nj=[]):
1111          """
1112 <        return job requirements to add to jdl files
1112 >        return job requirements to add to jdl files
1113          """
1114          req = ''
1115          if self.version:
1116              req='Member("VO-cms-' + \
1117                   self.version + \
1118                   '", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1119 +        if self.executable_arch:
1120 +            req+=' && Member("VO-cms-' + \
1121 +                 self.executable_arch + \
1122 +                 '", other.GlueHostApplicationSoftwareRunTimeEnvironment)'
1123  
1124          req = req + ' && (other.GlueHostNetworkAdapterOutboundIP)'
1125 +        if ( common.scheduler.name() == "glitecoll" ) or ( common.scheduler.name() == "glite"):
1126 +            req += ' && other.GlueCEStateStatus == "Production" '
1127  
1128          return req
1129  
1130      def configFilename(self):
1131          """ return the config filename """
1132 <        return self.name()+'.cfg'
1132 >        # FUTURE: Can remove cfg mode for CMSSW >= 2_1_x
1133 >        if (self.CMSSW_major >= 2 and self.CMSSW_minor >= 1) or (self.CMSSW_major >= 3):
1134 >          return self.name()+'.py'
1135 >        else:
1136 >          return self.name()+'.cfg'
1137  
1119    ### OLI_DANIELE
1138      def wsSetupCMSOSGEnvironment_(self):
1139          """
1140          Returns part of a job script which is prepares
1141          the execution environment and which is common for all CMS jobs.
1142          """
1143 <        txt = '\n'
1144 <        txt += '   echo "### SETUP CMS OSG  ENVIRONMENT ###"\n'
1145 <        txt += '   if [ -f $GRID3_APP_DIR/cmssoft/cmsset_default.sh ] ;then\n'
1146 <        txt += '      # Use $GRID3_APP_DIR/cmssoft/cmsset_default.sh to setup cms software\n'
1147 <        txt += '       source $GRID3_APP_DIR/cmssoft/cmsset_default.sh '+self.version+'\n'
1148 <        txt += '   elif [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1143 >        txt = '\n#Written by cms_cmssw::wsSetupCMSOSGEnvironment_\n'
1144 >        txt += '    echo ">>> setup CMS OSG environment:"\n'
1145 >        txt += '    echo "set SCRAM ARCH to ' + self.executable_arch + '"\n'
1146 >        txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
1147 >        txt += '    echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1148 >        txt += '    if [ -f $OSG_APP/cmssoft/cms/cmsset_default.sh ] ;then\n'
1149          txt += '      # Use $OSG_APP/cmssoft/cms/cmsset_default.sh to setup cms software\n'
1150 <        txt += '       source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1151 <        txt += '   else\n'
1152 <        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'
1153 <        txt += '       echo "JOB_EXIT_STATUS = 10020"\n'
1154 <        txt += '       echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1155 <        txt += '       dumpStatus $RUNTIME_AREA/$repo\n'
1138 <        txt += '       rm -f $RUNTIME_AREA/$repo \n'
1139 <        txt += '       echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1140 <        txt += '       echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1141 <        txt += '       exit 1\n'
1142 <        txt += '\n'
1143 <        txt += '       echo "Remove working directory: $WORKING_DIR"\n'
1144 <        txt += '       cd $RUNTIME_AREA\n'
1145 <        txt += '       /bin/rm -rf $WORKING_DIR\n'
1146 <        txt += '       if [ -d $WORKING_DIR ] ;then\n'
1147 <        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'
1148 <        txt += '            echo "JOB_EXIT_STATUS = 10017"\n'
1149 <        txt += '            echo "JobExitCode=10017" | tee -a $RUNTIME_AREA/$repo\n'
1150 <        txt += '            dumpStatus $RUNTIME_AREA/$repo\n'
1151 <        txt += '            rm -f $RUNTIME_AREA/$repo \n'
1152 <        txt += '            echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1153 <        txt += '            echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1154 <        txt += '       fi\n'
1155 <        txt += '\n'
1156 <        txt += '       exit 1\n'
1157 <        txt += '   fi\n'
1150 >        txt += '        source $OSG_APP/cmssoft/cms/cmsset_default.sh '+self.version+'\n'
1151 >        txt += '    else\n'
1152 >        txt += '        echo "ERROR ==> $OSG_APP/cmssoft/cms/cmsset_default.sh file not found"\n'
1153 >        txt += '        job_exit_code=10020\n'
1154 >        txt += '        func_exit\n'
1155 >        txt += '    fi\n'
1156          txt += '\n'
1157 <        txt += '   echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1158 <        txt += '   echo " END SETUP CMS OSG  ENVIRONMENT "\n'
1157 >        txt += '    echo "==> setup cms environment ok"\n'
1158 >        txt += '    echo "SCRAM_ARCH = $SCRAM_ARCH"\n'
1159  
1160          return txt
1161 <
1164 <    ### OLI_DANIELE
1161 >
1162      def wsSetupCMSLCGEnvironment_(self):
1163          """
1164          Returns part of a job script which is prepares
1165          the execution environment and which is common for all CMS jobs.
1166          """
1167 <        txt  = '   \n'
1168 <        txt += '   echo " ### SETUP CMS LCG  ENVIRONMENT ### "\n'
1169 <        txt += '   if [ ! $VO_CMS_SW_DIR ] ;then\n'
1170 <        txt += '       echo "SET_CMS_ENV 10031 ==> ERROR CMS software dir not found on WN `hostname`"\n'
1171 <        txt += '       echo "JOB_EXIT_STATUS = 10031" \n'
1172 <        txt += '       echo "JobExitCode=10031" | 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 += '   else\n'
1179 <        txt += '       echo "Sourcing environment... "\n'
1180 <        txt += '       if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1181 <        txt += '           echo "SET_CMS_ENV 10020 ==> ERROR cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1182 <        txt += '           echo "JOB_EXIT_STATUS = 10020"\n'
1183 <        txt += '           echo "JobExitCode=10020" | tee -a $RUNTIME_AREA/$repo\n'
1184 <        txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1185 <        txt += '           rm -f $RUNTIME_AREA/$repo \n'
1186 <        txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1187 <        txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1188 <        txt += '           exit 1\n'
1189 <        txt += '       fi\n'
1190 <        txt += '       echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1191 <        txt += '       source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1192 <        txt += '       result=$?\n'
1193 <        txt += '       if [ $result -ne 0 ]; then\n'
1194 <        txt += '           echo "SET_CMS_ENV 10032 ==> ERROR problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1195 <        txt += '           echo "JOB_EXIT_STATUS = 10032"\n'
1196 <        txt += '           echo "JobExitCode=10032" | tee -a $RUNTIME_AREA/$repo\n'
1197 <        txt += '           dumpStatus $RUNTIME_AREA/$repo\n'
1198 <        txt += '           rm -f $RUNTIME_AREA/$repo \n'
1199 <        txt += '           echo "MonitorJobID=`echo $MonitorJobID`" | tee -a $RUNTIME_AREA/$repo \n'
1200 <        txt += '           echo "MonitorID=`echo $MonitorID`" | tee -a $RUNTIME_AREA/$repo\n'
1201 <        txt += '           exit 1\n'
1202 <        txt += '       fi\n'
1203 <        txt += '   fi\n'
1204 <        txt += '   \n'
1205 <        txt += '   echo "SET_CMS_ENV 0 ==> setup cms environment ok"\n'
1206 <        txt += '   echo "### END SETUP CMS LCG ENVIRONMENT ###"\n'
1167 >        txt = '\n#Written by cms_cmssw::wsSetupCMSLCGEnvironment_\n'
1168 >        txt += '    echo ">>> setup CMS LCG environment:"\n'
1169 >        txt += '    echo "set SCRAM ARCH and BUILD_ARCH to ' + self.executable_arch + ' ###"\n'
1170 >        txt += '    export SCRAM_ARCH='+self.executable_arch+'\n'
1171 >        txt += '    export BUILD_ARCH='+self.executable_arch+'\n'
1172 >        txt += '    if [ ! $VO_CMS_SW_DIR ] ;then\n'
1173 >        txt += '        echo "ERROR ==> CMS software dir not found on WN `hostname`"\n'
1174 >        txt += '        job_exit_code=10031\n'
1175 >        txt += '        func_exit\n'
1176 >        txt += '    else\n'
1177 >        txt += '        echo "Sourcing environment... "\n'
1178 >        txt += '        if [ ! -s $VO_CMS_SW_DIR/cmsset_default.sh ] ;then\n'
1179 >        txt += '            echo "ERROR ==> cmsset_default.sh file not found into dir $VO_CMS_SW_DIR"\n'
1180 >        txt += '            job_exit_code=10020\n'
1181 >        txt += '            func_exit\n'
1182 >        txt += '        fi\n'
1183 >        txt += '        echo "sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1184 >        txt += '        source $VO_CMS_SW_DIR/cmsset_default.sh\n'
1185 >        txt += '        result=$?\n'
1186 >        txt += '        if [ $result -ne 0 ]; then\n'
1187 >        txt += '            echo "ERROR ==> problem sourcing $VO_CMS_SW_DIR/cmsset_default.sh"\n'
1188 >        txt += '            job_exit_code=10032\n'
1189 >        txt += '            func_exit\n'
1190 >        txt += '        fi\n'
1191 >        txt += '    fi\n'
1192 >        txt += '    \n'
1193 >        txt += '    echo "==> setup cms environment ok"\n'
1194 >        return txt
1195 >
1196 >    def modifyReport(self, nj):
1197 >        """
1198 >        insert the part of the script that modifies the FrameworkJob Report
1199 >        """
1200 >        txt = '\n#Written by cms_cmssw::modifyReport\n'
1201 >        publish_data = int(self.cfg_params.get('USER.publish_data',0))
1202 >        if (publish_data == 1):
1203 >            processedDataset = self.cfg_params['USER.publish_data_name']
1204 >            if (self.primaryDataset == 'null'):
1205 >                 self.primaryDataset = processedDataset
1206 >            if (common.scheduler.name().upper() == "CAF" or common.scheduler.name().upper() == "LSF"):
1207 >                ### FEDE FOR NEW LFN ###
1208 >                LFNBaseName = LFNBase(self.primaryDataset, processedDataset, LocalUser=True)
1209 >                self.user = getUserName(LocalUser=True)
1210 >                ########################
1211 >            else :
1212 >                ### FEDE FOR NEW LFN ###
1213 >                LFNBaseName = LFNBase(self.primaryDataset, processedDataset)
1214 >                self.user = getUserName()
1215 >                ########################
1216 >
1217 >            txt += 'if [ $copy_exit_status -eq 0 ]; then\n'
1218 >            ### FEDE FOR NEW LFN ###
1219 >            #txt += '    FOR_LFN=%s_${PSETHASH}/\n'%(LFNBaseName)
1220 >            txt += '    FOR_LFN=%s/${PSETHASH}/\n'%(LFNBaseName)
1221 >            ########################
1222 >            txt += 'else\n'
1223 >            txt += '    FOR_LFN=/copy_problems/ \n'
1224 >            txt += '    SE=""\n'
1225 >            txt += '    SE_PATH=""\n'
1226 >            txt += 'fi\n'
1227 >
1228 >            txt += 'echo ">>> Modify Job Report:" \n'
1229 >            txt += 'chmod a+x $RUNTIME_AREA/ProdCommon/FwkJobRep/ModifyJobReport.py\n'
1230 >            txt += 'ProcessedDataset='+processedDataset+'\n'
1231 >            txt += 'echo "ProcessedDataset = $ProcessedDataset"\n'
1232 >            txt += 'echo "SE = $SE"\n'
1233 >            txt += 'echo "SE_PATH = $SE_PATH"\n'
1234 >            txt += 'echo "FOR_LFN = $FOR_LFN" \n'
1235 >            txt += 'echo "CMSSW_VERSION = $CMSSW_VERSION"\n\n'
1236 >            ### FEDE FOR NEW LFN ###
1237 >            txt += 'echo "$RUNTIME_AREA/ProdCommon/FwkJobRep/ModifyJobReport.py $RUNTIME_AREA/crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier ' + self.user + '-$ProcessedDataset-$PSETHASH $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH"\n'
1238 >            txt += '$RUNTIME_AREA/ProdCommon/FwkJobRep/ModifyJobReport.py $RUNTIME_AREA/crab_fjr_$NJob.xml $NJob $FOR_LFN $PrimaryDataset $DataTier ' + self.user + '-$ProcessedDataset-$PSETHASH $ApplicationFamily $executable $CMSSW_VERSION $PSETHASH $SE $SE_PATH\n'
1239 >            ########################
1240 >            txt += 'modifyReport_result=$?\n'
1241 >            txt += 'if [ $modifyReport_result -ne 0 ]; then\n'
1242 >            txt += '    modifyReport_result=70500\n'
1243 >            txt += '    job_exit_code=$modifyReport_result\n'
1244 >            txt += '    echo "ModifyReportResult=$modifyReport_result" | tee -a $RUNTIME_AREA/$repo\n'
1245 >            txt += '    echo "WARNING: Problem with ModifyJobReport"\n'
1246 >            txt += 'else\n'
1247 >            txt += '    mv NewFrameworkJobReport.xml $RUNTIME_AREA/crab_fjr_$NJob.xml\n'
1248 >            txt += 'fi\n'
1249 >        return txt
1250 >
1251 >    def wsParseFJR(self):
1252 >        """
1253 >        Parse the FrameworkJobReport to obtain useful infos
1254 >        """
1255 >        txt = '\n#Written by cms_cmssw::wsParseFJR\n'
1256 >        txt += 'echo ">>> Parse FrameworkJobReport crab_fjr.xml"\n'
1257 >        txt += 'if [ -s $RUNTIME_AREA/crab_fjr_$NJob.xml ]; then\n'
1258 >        txt += '    if [ -s $RUNTIME_AREA/parseCrabFjr.py ]; then\n'
1259 >        txt += '        cmd_out=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --dashboard $MonitorID,$MonitorJobID '+self.debugWrap+'`\n'
1260 >        if self.debug_wrapper :
1261 >            txt += '        echo "Result of parsing the FrameworkJobReport crab_fjr.xml: $cmd_out"\n'
1262 >        txt += '        executable_exit_status=`python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --exitcode`\n'
1263 >        txt += '        if [ $executable_exit_status -eq 50115 ];then\n'
1264 >        txt += '            echo ">>> crab_fjr.xml contents: "\n'
1265 >        txt += '            cat $RUNTIME_AREA/crab_fjr_$NJob.xml\n'
1266 >        txt += '            echo "Wrong FrameworkJobReport --> does not contain useful info. ExitStatus: $executable_exit_status"\n'
1267 >        txt += '        elif [ $executable_exit_status -eq -999 ];then\n'
1268 >        txt += '            echo "ExitStatus from FrameworkJobReport not available. not available. Using exit code of executable from command line."\n'
1269 >        txt += '        else\n'
1270 >        txt += '            echo "Extracted ExitStatus from FrameworkJobReport parsing output: $executable_exit_status"\n'
1271 >        txt += '        fi\n'
1272 >        txt += '    else\n'
1273 >        txt += '        echo "CRAB python script to parse CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1274 >        txt += '    fi\n'
1275 >          #### Patch to check input data reading for CMSSW16x Hopefully we-ll remove it asap
1276 >
1277 >        txt += '    if [ $executable_exit_status -eq 0 ];then\n'
1278 >        txt += '      echo ">>> Executable succeded  $executable_exit_status"\n'
1279 >        if (self.datasetPath and not (self.dataset_pu or self.useParent)) :
1280 >          # VERIFY PROCESSED DATA
1281 >            txt += '      echo ">>> Verify list of processed files:"\n'
1282 >            txt += '      echo $InputFiles |tr -d \'\\\\\' |tr \',\' \'\\n\'|tr -d \'"\' > input-files.txt\n'
1283 >            txt += '      python $RUNTIME_AREA/parseCrabFjr.py --input $RUNTIME_AREA/crab_fjr_$NJob.xml --lfn > processed-files.txt\n'
1284 >            txt += '      cat input-files.txt  | sort | uniq > tmp.txt\n'
1285 >            txt += '      mv tmp.txt input-files.txt\n'
1286 >            txt += '      echo "cat input-files.txt"\n'
1287 >            txt += '      echo "----------------------"\n'
1288 >            txt += '      cat input-files.txt\n'
1289 >            txt += '      cat processed-files.txt | sort | uniq > tmp.txt\n'
1290 >            txt += '      mv tmp.txt processed-files.txt\n'
1291 >            txt += '      echo "----------------------"\n'
1292 >            txt += '      echo "cat processed-files.txt"\n'
1293 >            txt += '      echo "----------------------"\n'
1294 >            txt += '      cat processed-files.txt\n'
1295 >            txt += '      echo "----------------------"\n'
1296 >            txt += '      diff -q input-files.txt processed-files.txt\n'
1297 >            txt += '      fileverify_status=$?\n'
1298 >            txt += '      if [ $fileverify_status -ne 0 ]; then\n'
1299 >            txt += '         executable_exit_status=30001\n'
1300 >            txt += '         echo "ERROR ==> not all input files processed"\n'
1301 >            txt += '         echo "      ==> list of processed files from crab_fjr.xml differs from list in pset.cfg"\n'
1302 >            txt += '         echo "      ==> diff input-files.txt processed-files.txt"\n'
1303 >            txt += '      fi\n'
1304 >        txt += '    elif [ $executable_exit_status -ne 0 ] || [ $executable_exit_status -ne 50015 ] || [ $executable_exit_status -ne 50017 ];then\n'
1305 >        txt += '      echo ">>> Executable failed  $executable_exit_status"\n'
1306 >        txt += '      func_exit\n'
1307 >        txt += '    fi\n'
1308 >        txt += '\n'
1309 >        txt += 'else\n'
1310 >        txt += '    echo "CRAB FrameworkJobReport crab_fjr.xml is not available, using exit code of executable from command line."\n'
1311 >        txt += 'fi\n'
1312 >        txt += '\n'
1313 >        txt += 'echo "ExeExitCode=$executable_exit_status" | tee -a $RUNTIME_AREA/$repo\n'
1314 >        txt += 'echo "EXECUTABLE_EXIT_STATUS = $executable_exit_status"\n'
1315 >        txt += 'job_exit_code=$executable_exit_status\n'
1316 >
1317          return txt
1318  
1319      def setParam_(self, param, value):
# Line 1215 | Line 1322 | class Cmssw(JobType):
1322      def getParams(self):
1323          return self._params
1324  
1218    def setTaskid_(self):
1219        self._taskId = self.cfg_params['taskId']
1220        
1221    def getTaskid(self):
1222        return self._taskId
1223
1224 #######################################################################
1325      def uniquelist(self, old):
1326          """
1327          remove duplicates from a list
# Line 1230 | Line 1330 | class Cmssw(JobType):
1330          for e in old:
1331              nd[e]=0
1332          return nd.keys()
1333 +
1334 +    def outList(self):
1335 +        """
1336 +        check the dimension of the output files
1337 +        """
1338 +        txt = ''
1339 +        txt += 'echo ">>> list of expected files on output sandbox"\n'
1340 +        listOutFiles = []
1341 +        stdout = 'CMSSW_$NJob.stdout'
1342 +        stderr = 'CMSSW_$NJob.stderr'
1343 +        if (self.return_data == 1):
1344 +            for file in (self.output_file+self.output_file_sandbox):
1345 +                listOutFiles.append(numberFile(file, '$NJob'))
1346 +            listOutFiles.append(stdout)
1347 +            listOutFiles.append(stderr)
1348 +        else:
1349 +            for file in (self.output_file_sandbox):
1350 +                listOutFiles.append(numberFile(file, '$NJob'))
1351 +            listOutFiles.append(stdout)
1352 +            listOutFiles.append(stderr)
1353 +        txt += 'echo "output files: '+string.join(listOutFiles,' ')+'"\n'
1354 +        txt += 'filesToCheck="'+string.join(listOutFiles,' ')+'"\n'
1355 +        txt += 'export filesToCheck\n'
1356 +        return txt

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines