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

Comparing COMP/CRAB/python/DataDiscovery.py (file contents):
Revision 1.40.4.1 by spiga, Wed Apr 7 15:56:32 2010 UTC vs.
Revision 1.49 by ewv, Mon Aug 30 10:36:33 2010 UTC

# Line 8 | Line 8 | import DBSAPI.dbsApi
8   from DBSAPI.dbsApiException import *
9   import common
10   from crab_util import *
11 < from LumiList import LumiList
11 > try: # Can remove when CMSSW 3.7 and earlier are dropped
12 >    from FWCore.PythonUtilities.LumiList import LumiList
13 > except ImportError:
14 >    from LumiList import LumiList
15 >
16   import os
17  
18  
# Line 103 | Line 107 | class DataDiscovery:
107          self.parent = {}          # DBS output: parents of each file
108          self.lumis = {}           # DBS output: lumis in each file
109          self.lumiMask = None
110 +        self.splitByLumi = False
111 +        self.splitDataByEvent = 0
112  
113      def fetchDBSInfo(self):
114          """
# Line 110 | Line 116 | class DataDiscovery:
116          """
117          ## get DBS URL
118          global_url="http://cmsdbsprod.cern.ch/cms_dbs_prod_global/servlet/DBSServlet"
119 <        caf_url = "http://cmsdbsprod.cern.ch/cms_dbs_caf_analysis_01/servlet/DBSServlet"
114 <        dbs_url_map  =   {'glite':    global_url,
115 <                          'glite_slc5':global_url,\
116 <                          'glitecoll':global_url,\
117 <                          'condor':   global_url,\
118 <                          'condor_g': global_url,\
119 <                          'glidein':  global_url,\
120 <                          'lsf':      global_url,\
121 <                          'caf':      caf_url,\
122 <                          'sge':      global_url,\
123 <                          'arc':      global_url,\
124 <                          'pbs':      global_url
125 <                          }
126 <
127 <        dbs_url_default = dbs_url_map[(common.scheduler.name()).lower()]
128 <        dbs_url=  self.cfg_params.get('CMSSW.dbs_url', dbs_url_default)
119 >        dbs_url=  self.cfg_params.get('CMSSW.dbs_url', global_url)
120          common.logger.info("Accessing DBS at: "+dbs_url)
121  
122          ## check if runs are selected
# Line 133 | Line 124 | class DataDiscovery:
124          if (self.cfg_params.has_key('CMSSW.runselection')):
125              runselection = parseRange2(self.cfg_params['CMSSW.runselection'])
126  
127 <        ## check if lumiMask is set
127 >        ## check if various lumi parameters are set
128          self.lumiMask = self.cfg_params.get('CMSSW.lumi_mask',None)
129 +        self.lumiParams = self.cfg_params.get('CMSSW.total_number_of_lumis',None) or \
130 +                          self.cfg_params.get('CMSSW.lumis_per_job',None)
131 +
132          lumiList = None
133          if self.lumiMask:
134              lumiList = LumiList(filename=self.lumiMask)
135 +        if runselection:
136 +            runList = LumiList(runs = runselection)
137  
138          self.splitByRun = int(self.cfg_params.get('CMSSW.split_by_run', 0))
139 <
139 >        self.splitDataByEvent = int(self.cfg_params.get('CMSSW.split_by_event', 0))
140          common.logger.log(10-1,"runselection is: %s"%runselection)
141 +
142 +        if not self.splitByRun:
143 +            self.splitByLumi = self.lumiMask or self.lumiParams or self.ads
144 +
145 +        if self.splitByRun and not runselection:
146 +            msg = "Error: split_by_run must be combined with a runselection"
147 +            raise CrabException(msg)
148 +
149          ## service API
150          args = {}
151          args['url']     = dbs_url
# Line 157 | Line 161 | class DataDiscovery:
161          api = DBSAPI.dbsApi.DbsApi(args)
162          self.files = self.queryDbs(api,path=self.datasetPath,runselection=runselection,useParent=useparent)
163  
164 +        # Check to see what the dataset is
165 +        pdsName = self.datasetPath.split("/")[1]
166 +        primDSs = api.listPrimaryDatasets(pdsName)
167 +        dataType = primDSs[0]['Type']
168 +        common.logger.debug("Datatype is %s" % dataType)
169 +        if dataType == 'data' and not \
170 +            (self.splitByRun or self.splitByLumi or self.splitDataByEvent):
171 +            msg = 'Data must be split by lumi or by run. ' \
172 +                  'Please see crab -help for the correct settings'
173 +            raise  CrabException(msg)
174 +
175 +
176 +
177          anFileBlocks = []
178          if self.skipBlocks: anFileBlocks = readTXTfile(self, fileBlocks_FileName)
179  
# Line 171 | Line 188 | class DataDiscovery:
188                  # asked retry the list of parent for the given child
189                  if useparent==1:
190                      parList = [x['LogicalFileName'] for x in file['ParentList']]
191 <                if self.ads or self.lumiMask:
191 >                if self.splitByLumi:
192                      fileLumis = [ (x['RunNumber'], x['LumiSectionNumber'])
193                                   for x in file['LumiList'] ]
194                  self.parent[filename] = parList
195                  # For LumiMask, intersection of two lists.
196 <                if self.lumiMask:
196 >                if self.lumiMask and runselection:
197 >                    self.lumis[filename] = runList.filterLumis(lumiList.filterLumis(fileLumis))
198 >                elif runselection:
199 >                    self.lumis[filename] = runList.filterLumis(fileLumis)
200 >                elif self.lumiMask:
201                      self.lumis[filename] = lumiList.filterLumis(fileLumis)
202                  else:
203                      self.lumis[filename] = fileLumis
# Line 204 | Line 225 | class DataDiscovery:
225              msg = "No new fileblocks available for dataset: "+str(self.datasetPath)
226              raise  CrabException(msg)
227  
207        saveFblocks=''
208        for block in self.eventsPerBlock.keys() :
209            saveFblocks += str(block)+'\n'
210            common.logger.log(10-1,"DBSInfo: total nevts %i in block %s "%(self.eventsPerBlock[block],block))
211        writeTXTfile(self, fileBlocks_FileName , saveFblocks)
228  
229          if len(self.eventsPerBlock) <= 0:
230              raise NotExistingDatasetError(("\nNo data for %s in DBS\nPlease check"
# Line 218 | Line 234 | class DataDiscovery:
234  
235      def queryDbs(self,api,path=None,runselection=None,useParent=None):
236  
237 <        allowedRetriveValue = ['retrive_block', 'retrive_run']
238 <        if self.ads or self.lumiMask:
237 >
238 >        allowedRetriveValue = []
239 >        if self.splitByLumi or self.splitByRun or useParent == 1:
240 >            allowedRetriveValue.extend(['retrive_block', 'retrive_run'])
241 >        if self.splitByLumi:
242              allowedRetriveValue.append('retrive_lumi')
243 <        if useParent == 1: allowedRetriveValue.append('retrive_parent')
243 >        if useParent == 1:
244 >            allowedRetriveValue.append('retrive_parent')
245          common.logger.debug("Set of input parameters used for DBS query: %s" % allowedRetriveValue)
246          try:
247 <            if len(runselection) <=0 :
248 <                if useParent==1 or self.splitByRun==1 or self.ads or self.lumiMask:
229 <                    if self.ads:
230 <                        files = api.listFiles(analysisDataset=path, retriveList=allowedRetriveValue)
231 <                    else :
232 <                        files = api.listFiles(path=path, retriveList=allowedRetriveValue)
233 <                else:
234 <                    files = api.listDatasetFiles(self.datasetPath)
235 <            else :
236 <                files=[]
247 >            if self.splitByRun:
248 >                files = []
249                  for arun in runselection:
250                      try:
251                          if self.ads:
# Line 246 | Line 258 | class DataDiscovery:
258                          common.logger.info(msg)
259                          pass
260  
261 +            else:
262 +                if allowedRetriveValue:
263 +                    if self.ads:
264 +                        files = api.listFiles(analysisDataset=path, retriveList=allowedRetriveValue)
265 +                    else :
266 +                        files = api.listFiles(path=path, retriveList=allowedRetriveValue)
267 +                else:
268 +                    files = api.listDatasetFiles(self.datasetPath)
269 +
270          except DbsBadRequest, msg:
271              raise DataDiscoveryError(msg)
272          except DBSError, msg:

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines