Public Member Functions
def	get_class (self, argument='')

def	print_memdb (self)

def	read_db (self, db_file_name=__default_db_file_name)

def	write_db (self)

Public Attributes
	addFiles

	batchScript

	cfgTemplate

	classInf

	currentTime

	driver

	header

	infiList

	mergeScript

Static Public Attributes
	elapsedTime

	JOBREMARK

	JOBSP1

	JOBSP2

	JOBSP3

	mssDir

	mssDirPool

	nJobs

	pedeMem

	spare1

	spare2

	spare3

	updateTime

	updateTimeHuman

Static Private Attributes
string	__default_db_file_name = "mps.db"

Detailed Description

Definition at line 47 of file Mpslibclass.py.

Member Function Documentation

def Mpslibclass.jobdatabase.get_class	(	self,
		argument = `''`
	)

Definition at line 240 of file Mpslibclass.py.

References edm.print().

     def get_class(self, argument=''):
         CLASSES = self.classInf.split(':')
         if len(CLASSES)<1 or len(CLASSES)>2:
             print('\nget_class():\n  class must be of the form \'class\' or \'classMille:classPede\', but is \'%s\'!\n\n', classInf)
             sys.exit(1)
         elif argument == 'mille':
             return CLASSES[0]
         elif argument == 'pede':
             if len(CLASSES) == 1:
                 return CLASSES[0]
             elif len(CLASSES) == 2:
                 return CLASSES[1]
         else:
             print('\nget_class():\n  Know class only for \'mille\' or \'pede\', not %s!\n\n' %argument)
             sys.exit(1)

def Mpslibclass.jobdatabase.print_memdb ( self )

Definition at line 129 of file Mpslibclass.py.

     def print_memdb(self):
         #print metainfo
         print("\n=== mps database printout ===\n")
         print(self.header)
         print('Script:\t\t',    self.batchScript)
         print('cfg:\t\t',       self.cfgTemplate)
         print('files:\t\t',     self.infiList)
         print('class:\t\t',     self.classInf)
         print('name:\t\t',      self.addFiles)
         print('driver:\t\t',    self.driver)
         print('mergeScript:\t', self.mergeScript)
         print('mssDir:\t\t',    self.mssDir)
         print('updateTime:\t',  self.updateTimeHuman)
         print('elapsed:\t',     self.elapsedTime)
         print('mssDirPool:\t',  self.mssDirPool)
         print('pedeMem:\t',             self.pedeMem, '\n')
 
         #print interesting Job-level lists ---- to add: t/evt, fix remarks
         print('###     dir      jobid    stat  try  rtime      nevt  remark   weight  name')
         print("------------------------------------------------------------------------------")
         for i in xrange(self.nJobs):
             print('%03d  %6s  %9s  %6s  %3d  %5d  %8d  %8s  %5s  %s' % (
                 self.JOBNUMBER[i],
                 self.JOBDIR[i],
                 self.JOBID[i],
                 self.JOBSTATUS[i],
                 self.JOBNTRY[i],
                 self.JOBRUNTIME[i],
                 self.JOBNEVT[i],
                 self.JOBHOST[i],
                 self.JOBSP2[i],
                 self.JOBSP3[i]))
 
         #print merge Jobs if merge mode
         if self.driver == 'merge':
             for i in xrange(self.nJobs,len(self.JOBDIR)):
                 print('%s  %6s  %9s  %6s  %3d  %5d  %8d  %8s  %5s  %s' % (
                     'MMM',
                     self.JOBDIR[i],
                     self.JOBID[i],
                     self.JOBSTATUS[i],
                     self.JOBNTRY[i],
                     self.JOBRUNTIME[i],
                     self.JOBNEVT[i],
                     self.JOBHOST[i],
                     self.JOBSP2[i],
                     self.JOBSP3[i]))
 
         #print summed info
         totalEvents = sum(self.JOBNEVT[:self.nJobs])
         totalCpu    = sum(self.JOBRUNTIME[:self.nJobs])
         meanCpuPerEvent = 0.
         if totalEvents > 0:
             meanCpuPerEvent = float(totalCpu)/totalEvents
         print("------------------------------------------------------------------------------")
         print("\t\t\t\t\tEvent total:\t",       totalEvents)
         print("\t\t\t\t\tCPU total:\t",         totalCpu,               's')
         print("\t\t\t\t\tMean CPU/event:\t",meanCpuPerEvent,'s')
 
 
 
 
 

def Mpslibclass.jobdatabase.read_db	(	self,
		db_file_name = `__default_db_file_name`
	)

Definition at line 60 of file Mpslibclass.py.

References edm.print().

     def read_db(self, db_file_name = __default_db_file_name):
         try:
             DBFILE = open(db_file_name,'r')
         except IOError as e:
             if e.args != (2, 'No such file or directory'):
                 raise
             else:
                 if db_file_name == jobdatabase.__default_db_file_name:
                     msg = ("No 'mps.db' found. Make sure you are in a campaign "
                            "directory and that the campaign is set up.")
                 else:
                     msg = "Database file '"+db_file_name+"' not found. Exiting."
                 print(msg)
                 sys.exit(1)
 
         #read infolines at the top, used rstrip to delete the '\n'
         self.header          = DBFILE.readline().strip()
         self.batchScript     = DBFILE.readline().rstrip('\n')
         self.cfgTemplate     = DBFILE.readline().rstrip('\n')
         self.infiList        = DBFILE.readline().rstrip('\n')
         self.classInf        = DBFILE.readline().rstrip('\n')   #formerly named 'class' ->conflict
         self.addFiles        = DBFILE.readline().rstrip('\n')
         self.driver          = DBFILE.readline().rstrip('\n')
         self.mergeScript     = DBFILE.readline().rstrip('\n')
         self.mssDir          = DBFILE.readline().rstrip('\n')
         self.updateTime      = int(DBFILE.readline())
         self.updateTimeHuman = DBFILE.readline().rstrip('\n')
         self.elapsedTime     = int(DBFILE.readline())
         self.mssDirPool      = DBFILE.readline().rstrip('\n')
         self.pedeMem         = int(DBFILE.readline())
         self.spare1          = DBFILE.readline().rstrip('\n')
         self.spare2          = DBFILE.readline().rstrip('\n')
         self.spare3          = DBFILE.readline().rstrip('\n')
 
         #read actual jobinfo into arrays
         self.nJobs = 0
         milleJobs = 0
 
 
         for line in DBFILE:
             if line.strip() == "": continue # ignore empty lines
             line = line.rstrip('\n')        # removes the pesky \n from line
             parts = line.split(":")         # read each line and split into parts list
             self.JOBNUMBER.append(int(parts[0]))
             self.JOBDIR.append(parts[1].strip())
             self.JOBID.append(parts[2])
             self.JOBSTATUS.append(parts[3].strip())
             self.JOBNTRY.append(int(parts[4]))
             self.JOBRUNTIME.append(int(parts[5]))   #int float?
             self.JOBNEVT.append(int(parts[6]))
             self.JOBHOST.append(parts[7].strip())
             self.JOBINCR.append(int(parts[8]))
             self.JOBREMARK.append(parts[9].strip())
             self.JOBSP1.append(parts[10].strip())
             self.JOBSP2.append(parts[11].strip())
             self.JOBSP3.append(parts[12].strip())
 
             #count number of jobs
             if not self.JOBDIR[self.nJobs].startswith("jobm"):
                 milleJobs += 1
             self.nJobs += 1
         self.nJobs = milleJobs
 
         DBFILE.close()
 
 
 

def Mpslibclass.jobdatabase.write_db ( self )

Definition at line 194 of file Mpslibclass.py.

References Mpslibclass.jobdatabase.header.

     def write_db(self):
         self.header = "mps database schema 3.2"
         self.currentTime = int(time.time())
         self.elapsedTime = 0;
         if self.updateTime != 0:
             self.elapsedTime = self.currentTime - self.updateTime
         self.updateTime = self.currentTime
         self.updateTimeHuman = str(datetime.datetime.today())   #no timezone :(
         self.spare1 = "-- unused --"
         self.spare2 = "-- unused --"
         self.spare3 = "-- unused --"
 
         #if mps.db already exists, backup as mps.db~ (in case of interupt during write)
         os.system('[[ -a mps.db ]] && cp -p mps.db mps.db~')
 
         #write mps.db header
         DBFILE = open ("mps.db", "w")
         headData = [ self.header, self.batchScript, self.cfgTemplate, self.infiList,
                      self.classInf, self.addFiles, self.driver, self.mergeScript,
                      self.mssDir, self.updateTime, self.updateTimeHuman,
                      self.elapsedTime, self.mssDirPool, self.pedeMem,
                      self.spare1, self.spare2, self.spare3 ]
         for item in headData:
             DBFILE.write("%s\n" % item)
 
         #write mps.db jobinfo
         for i in xrange(len(self.JOBID)):
             DBFILE.write('%03d:%s:%s:%s:%s:%s:%s:%s:%s:%s:%s:%s:%s\n' %
                          (i+1,
                           self.JOBDIR[i],
                           self.JOBID[i],
                           self.JOBSTATUS[i],
                           self.JOBNTRY[i],
                           self.JOBRUNTIME[i],
                           self.JOBNEVT[i],
                           self.JOBHOST[i],
                           self.JOBINCR[i],
                           self.JOBREMARK[i],
                           self.JOBSP1[i],
                           self.JOBSP2[i],
                           self.JOBSP3[i]))
         DBFILE.close()
 