genericValidation.GenericValidationData Class Reference

Inheritance diagram for genericValidation.GenericValidationData:

Public Member Functions
def	__init__
def	createCrabCfg
def	createScript
def	getRepMap
Public Member Functions inherited from genericValidation.GenericValidation
def	__init__
def	createConfiguration
def	createCrabCfg
def	createFiles
def	createScript
def	getCompareStrings
def	getRepMap

Public Attributes
	crabWorkingDir
	dataset
Public Attributes inherited from genericValidation.GenericValidation
	alignmentToValidate
	AutoAlternates
	cmssw
	cmsswreleasebase
	config
	configFiles
	crabConfigFiles
	filesToCompare
	general
	jobmode
	name
	NJobs
	randomWorkdirPart
	scramarch
	scriptFiles

Additional Inherited Members
Static Public Attributes inherited from genericValidation.GenericValidation
string	defaultReferenceName = "DEFAULT"

Detailed Description

Subclass of `GenericValidation` which is the base for validations using
datasets.

Definition at line 187 of file genericValidation.py.

Constructor & Destructor Documentation

def genericValidation.GenericValidationData.__init__	(		self,
			valName,
			alignment,
			config,
			valType,
			addDefaults = {},
			addMandatories = []
	)

This method adds additional items to the `self.general` dictionary
which are only needed for validations using datasets.

Arguments:
- `valName`: String which identifies individual validation instances
- `alignment`: `Alignment` instance to validate
- `config`: `BetterConfigParser` instance which includes the
    configuration of the validations
- `valType`: String which specifies the type of validation
- `addDefaults`: Dictionary which contains default values for individual
         validations in addition to the general default values
- `addMandatories`: List which contains mandatory parameters for
            individual validations in addition to the general
            mandatory parameters

Definition at line 194 of file genericValidation.py.

References genericValidation.GenericValidation.cmssw, genericValidation.GenericValidation.cmsswreleasebase, genericValidation.GenericValidation.general, and genericValidation.GenericValidation.NJobs.

                 addDefaults = {}, addMandatories=[]):
        """
        This method adds additional items to the `self.general` dictionary
        which are only needed for validations using datasets.
        
        Arguments:
        - `valName`: String which identifies individual validation instances
        - `alignment`: `Alignment` instance to validate
        - `config`: `BetterConfigParser` instance which includes the
                    configuration of the validations
        - `valType`: String which specifies the type of validation
        - `addDefaults`: Dictionary which contains default values for individual
                         validations in addition to the general default values
        - `addMandatories`: List which contains mandatory parameters for
                            individual validations in addition to the general
                            mandatory parameters
        """

        defaults = {"runRange": "",
                    "firstRun": "",
                    "lastRun": "",
                    "begin": "",
                    "end": "",
                    "JSON": ""
                    }
        defaults.update(addDefaults)
        mandatories = [ "dataset", "maxevents" ]
        mandatories += addMandatories
        GenericValidation.__init__(self, valName, alignment, config, valType, defaults, mandatories)

        # if maxevents is not specified, cannot calculate number of events for
        # each parallel job, and therefore running only a single job
        if int( self.general["maxevents"] ) == -1 and self.NJobs > 1:
            msg = ("Maximum number of events (maxevents) not specified: "
                   "cannot use parallel jobs.")
            raise AllInOneError(msg)

        tryPredefinedFirst = (not self.jobmode.split( ',' )[0] == "crab" and self.general["JSON"]    == ""
                              and self.general["firstRun"] == ""         and self.general["lastRun"] == ""
                              and self.general["begin"]    == ""         and self.general["end"]     == "")

        if self.general["dataset"] not in globalDictionaries.usedDatasets:
            globalDictionaries.usedDatasets[self.general["dataset"]] = {}

        if self.cmssw not in globalDictionaries.usedDatasets[self.general["dataset"]]:
            if globalDictionaries.usedDatasets[self.general["dataset"]] != {}:
                print ("Warning: you use the same dataset '%s' in multiple cmssw releases.\n"
                       "This is allowed, but make sure it's not a mistake") % self.general["dataset"]
            globalDictionaries.usedDatasets[self.general["dataset"]][self.cmssw] = {False: None, True: None}

        if globalDictionaries.usedDatasets[self.general["dataset"]][self.cmssw][tryPredefinedFirst] is None:
            dataset = Dataset(
                self.general["dataset"], tryPredefinedFirst = tryPredefinedFirst,
                cmssw = self.cmssw, cmsswrelease = self.cmsswreleasebase )
            globalDictionaries.usedDatasets[self.general["dataset"]][self.cmssw][tryPredefinedFirst] = dataset
            if tryPredefinedFirst and not dataset.predefined():                              #No point finding the data twice in that case
                globalDictionaries.usedDatasets[self.general["dataset"]][self.cmssw][False] = dataset

        self.dataset = globalDictionaries.usedDatasets[self.general["dataset"]][self.cmssw][tryPredefinedFirst]
        self.general["magneticField"] = self.dataset.magneticField()
        self.general["defaultMagneticField"] = "MagneticField"
        if self.general["magneticField"] == "unknown":
            print "Could not get the magnetic field for this dataset."
            print "Using the default: ", self.general["defaultMagneticField"]
            self.general["magneticField"] = '.oO[defaultMagneticField]Oo.'
        
        if not self.jobmode.split( ',' )[0] == "crab":
            try:
                self.general["datasetDefinition"] = self.dataset.datasetSnippet(
                    jsonPath = self.general["JSON"],
                    firstRun = self.general["firstRun"],
                    lastRun = self.general["lastRun"],
                    begin = self.general["begin"],
                    end = self.general["end"],
                    parent = self.needParentFiles )
            except AllInOneError as e:
                msg = "In section [%s:%s]: "%(valType, self.name)
                msg += str(e)
                raise AllInOneError(msg)
        else:
            if self.dataset.predefined():
                msg = ("For jobmode 'crab' you cannot use predefined datasets "
                       "(in your case: '%s')."%( self.dataset.name() ))
                raise AllInOneError( msg )
            try:
                theUpdate = config.getResultingSection(valType+":"+self.name,
                                                       demandPars = ["parallelJobs"])
            except AllInOneError as e:
                msg = str(e)[:-1]+" when using 'jobmode: crab'."
                raise AllInOneError(msg)
            self.general.update(theUpdate)
            if self.general["begin"] or self.general["end"]:
                ( self.general["begin"],
                  self.general["end"],
                  self.general["firstRun"],
                  self.general["lastRun"] ) = self.dataset.convertTimeToRun(
                    firstRun = self.general["firstRun"],
                    lastRun = self.general["lastRun"],
                    begin = self.general["begin"],
                    end = self.general["end"],
                    shortTuple = False)
                if self.general["begin"] == None:
                    self.general["begin"] = ""
                if self.general["end"] == None:
                    self.general["end"] = ""
                self.general["firstRun"] = str( self.general["firstRun"] )
                self.general["lastRun"] = str( self.general["lastRun"] )
            if ( not self.general["firstRun"] ) and \
                   ( self.general["end"] or self.general["lastRun"] ):
                self.general["firstRun"] = str(
                    self.dataset.runList()[0]["run_number"])
            if ( not self.general["lastRun"] ) and \
                   ( self.general["begin"] or self.general["firstRun"] ):
                self.general["lastRun"] = str(
                    self.dataset.runList()[-1]["run_number"])
            if self.general["firstRun"] and self.general["lastRun"]:
                if int(self.general["firstRun"]) > int(self.general["lastRun"]):
                    msg = ( "The lower time/runrange limit ('begin'/'firstRun') "
                            "chosen is greater than the upper time/runrange limit "
                            "('end'/'lastRun').")
                    raise AllInOneError( msg )
                self.general["runRange"] = (self.general["firstRun"]
                                            + '-' + self.general["lastRun"])
            try:
                self.general["datasetDefinition"] = self.dataset.datasetSnippet(
                    jsonPath = self.general["JSON"],
                    firstRun = self.general["firstRun"],
                    lastRun = self.general["lastRun"],
                    begin = self.general["begin"],
                    end = self.general["end"],
                    crab = True )
            except AllInOneError as e:
                msg = "In section [%s:%s]: "%(valType, self.name)
                msg += str( e )
                raise AllInOneError( msg )