Python DataBlockGenerator.DataBlockGenerator類代碼示例

    def listFiles(self, datasetPath, retriveList):
        res = []
        dbg = DataBlockGenerator()
        for block in dbg.getBlocks(datasetPath):
            files = dbg.getFiles(block['Name'])
            for f in files:
                f['Block'] = block
                res.append(f)

        return res

 def __init__(self, dict=None, responseType="json", logger=None,
              dbsUrl='https://cmsweb.cern.ch/dbs/prod/global/DBSReader'):
     print("Using MockPhEDExApi")
     self.dbsUrl = dbsUrl
     dict = dict or {}
     self.dataBlocks = DataBlockGenerator()
     self.subRequests = {}

 def __init__(self, *args, **kwargs):
     # add the end point to prevent the existence check fails.
     self['endpoint'] = "phedex_emulator"
     self.dataBlocks = DataBlockGenerator()
     self.subRequests = {}
     self.deletionRequests = {}
     self.deletionRequestId = 0

class DBSReader:
    """
    Mock up dbs access
    """
    def __init__(self, *args, **kwargs):
        print "Using DBS Emulator ..."
        self.dataBlocks = DataBlockGenerator()
        
    def getFileBlocksInfo(self, dataset, onlyClosedBlocks = True):
        """Fake block info"""
        return self.dataBlocks.getBlocks(dataset)

    def listFileBlockLocation(self, block):
        """Fake locations"""
        return self.dataBlocks.getLocation(block)

    def listFilesInBlock(self, block):
        """Fake files"""
        return self.dataBlocks.getFiles(block)

    def getFileBlock(self, block):
        """Return block + locations"""
        result = { block : {
            "StorageElements" : self.listFileBlockLocation(block),
            "Files" : self.listFilesInBlock(block),
            "IsOpen" : False,
            }
                   }
        return result

    def getDatasetInfo(self, dataset):
        """Dataset summary"""
        result = {}
        result['number_of_events'] = sum([x['NumberOfEvents'] 
                                for x in self.dataBlocks.getBlocks(dataset)])
        result['number_of_files'] = sum([x['NumberOfFiles'] 
                                for x in self.dataBlocks.getBlocks(dataset)])
        result['path'] = dataset
        return result

class MockPhEDExApi(object):
    """
    Version of Services/PhEDEx intended to be used with mock or unittest.mock
    """

    def __init__(self, dict=None, responseType="json", logger=None,
                 dbsUrl='https://cmsweb.cern.ch/dbs/prod/global/DBSReader'):
        print("Using MockPhEDExApi")
        self.dbsUrl = dbsUrl
        dict = dict or {}
        self.dataBlocks = DataBlockGenerator()
        self.subRequests = {}

    def sitesByBlock(self, block):
        """
        Centralize the algorithm to decide where a block is based on the hash name

        Args:
            block: the name of the block

        Returns:
            sites: a fake list of sites where the data is

        """

        if hash(block) % 3 == 0:
            sites = ['T2_XX_SiteA']
        elif hash(block) % 3 == 1:
            sites = ['T2_XX_SiteA', 'T2_XX_SiteB']
        else:
            sites = ['T2_XX_SiteA', 'T2_XX_SiteB', 'T2_XX_SiteC']

        return sites

    def getReplicaPhEDExNodesForBlocks(self, block=None, dataset=None, complete='y'):
        """

        Args:
            block: the name of the block
            dataset: the name of the dataset
            complete: ??

        Returns:
            a fake list of blocks and the fakes sites they are at
        """
        if isinstance(dataset, list):
            dataset = dataset[0]  # Dataset is a list in these tests
        if dataset:
            # TODO: Generalize this and maybe move dataset detection into sitesByBlock
            if dataset == PILEUP_DATASET:
                return {
                    '%s#0fcb2b12-d27e-11e0-91b1-003048caaace' % dataset: ['T2_XX_SiteA', 'T2_XX_SiteB', 'T2_XX_SiteC']}
            else:
                try:
                    DBS3Reader(PROD_DBS).checkDatasetPath(dataset)
                    blocks = DBS3Reader(PROD_DBS).dbs.listBlocks(dataset=dataset)
                    singleBlock = blocks[0]['block_name']
                    return {singleBlock: self.sitesByBlock(singleBlock)}
                except DBSReaderError:
                    return {'%s#0fcb2b12-d27e-11e0-91b1-003048caaace' % dataset: []}

        replicas = {}
        for oneBlock in block:
            if oneBlock.split('#')[0] == PILEUP_DATASET:
                # Pileup is at a single site
                sites = ['T2_XX_SiteC']
                _BLOCK_LOCATIONS[oneBlock] = sites
            else:
                sites = self.sitesByBlock(block=oneBlock)
                _BLOCK_LOCATIONS[oneBlock] = sites
            replicas.update({oneBlock: sites})
        return replicas

    def getReplicaInfoForBlocks(self, **args):
        """
        Where are blocks located
        """

        data = {"phedex": {"request_timestamp": 1254762796.13538, "block": []}}

        for block in args['block']:
            blocks = data['phedex']['block']
            # files = self.dataBlocks.getFiles(block)
            # locations = self.dataBlocks.getLocation(block)
            sites = self.sitesByBlock(block=block)
            blocks.append({'files': 1, 'name': block, 'replica': [{'node': x} for x in sites]})
        return data

    def getSubscriptionMapping(self, *dataItems, **kwargs):
        """
        Fake version of the existing PhEDEx method
        """

        dataItems = list(set(dataItems))  # force unique items
        locationMap = {}

        for dataItem in dataItems:
            sites = self.sitesByBlock(block=dataItem)
            locationMap.update({dataItem: sites})

#.........這裏部分代碼省略.........

class PhEDEx(dict):
    """
    """
    def __init__(self, *args, **kwargs):
        # add the end point to prevent the existence check fails.
        self['endpoint'] = "phedex_emulator"
        self.dataBlocks = DataBlockGenerator()
        self.subRequests = {}
        
    def injectBlocks(self, node, xmlData, verbose = 0, strict = 1):

        """
        do nothing don't inject block.
        """

        return None

    def getNodeSE(self, value):
        return 'dummy.se.from.emulator'

    def subscribe(self, subscription, xmlData):
        """
        Store the subscription information in the object,
        tests can retrieve it and verify it
        """

        args = {}

        args['node'] = []
        for node in subscription.nodes:
            args['node'].append(node)

        document = parseString(xmlData)
        datasets = document.getElementsByTagName("dataset")
        for dataset in datasets:
            datasetName = dataset.getAttribute("name")

        if datasetName not in self.subRequests:
            self.subRequests[datasetName] = []

        args['data'] = xmlData
        args['level'] = subscription.level
        args['priority'] = subscription.priority
        args['move'] = subscription.move
        args['static'] = subscription.static
        args['custodial'] = subscription.custodial
        args['group'] = subscription.group
        args['request_only'] = subscription.request_only

        self.subRequests[datasetName].append(args)

        return

    def getReplicaInfoForFiles(self, **args):
        """
        _getReplicaInfoForFiles_
        TODO: Need to be implemented correctly,
        Currently not used

        Retrieve file replica information from PhEDEx.

        block          block name, with '*' wildcards, can be multiple (*).  required when no lfn is specified.
        node           node name, can be multiple (*)
        se             storage element name, can be multiple (*)
        update_since   unix timestamp, only return replicas updated since this
                    time
        create_since   unix timestamp, only return replicas created since this
                    time
        complete       y or n. if y, return only file replicas from complete block
                    replicas.  if n only return file replicas from incomplete block
                    replicas.  default is to return either.
        dist_complete  y or n.  if y, return only file replicas from blocks
                    where all file replicas are available at some node. if
                    n, return only file replicas from blocks which have
                    file replicas not available at any node.  default is
                    to return either.
        subscribed     y or n, filter for subscription. default is to return either.
        custodial      y or n. filter for custodial responsibility.  default is
                    to return either.
        group          group name.  default is to return replicas for any group.
        lfn            logical file nam
        """
        return None

    def getNodeMap(self):
        """
        _getNodeMap_

        Retrieve information about nodes known to this PhEDEx instance.  Each
        node entry will have the following keys:
          name       - PhEDEx node name
          se         - Storage element name
          kind       - Node type, e.g. 'Disk' or 'MSS'
          technology - Node technology, e.g. 'Castor'
          id         - Node id

        Return some MSS, Buffer and Disk nodes
        """

        nodeMappings = {"phedex" : {"node" : []}}
#.........這裏部分代碼省略.........

 def __init__(self, url, **contact):
     self.dataBlocks = DataBlockGenerator()
     args = { "url" : url, "level" : 'ERROR', "version" : 'DBS_2_0_9'}
     self.dbs = _MockDBSApi(args)

class PhEDEx:
    """
    """
    def __init__(self, *args, **kwargs):
        print "Using PhEDEx Emulator ...."
        self.dataBlocks = DataBlockGenerator()

    def getReplicaInfoForBlocks(self, **args):
        """
        Where are blocks located
        """
        for block in args['block']:
            data = {"phedex":{"request_timestamp":1254762796.13538, "block" : []}}
            blocks = data['phedex']['block']
            files = self.dataBlocks.getFiles(block)
            locations = self.dataBlocks.getLocation(block)
            blocks.append({"files": len(files), "name": block,
                           'replica' : [{'se' : x } for x in locations]})
        return data

    def subscriptions(self, **args):
        """
        Where is data subscribed - for now just replicate blockreplicas
        """
        data = {'phedex' : {"request_timestamp" : 1254850198.15418,
                            'dataset' : []}}
        # different structure depending on whether we ask for dataset or blocks
        
        if args.has_key('dataset') and args['dataset']:
            for dataset in args['dataset']:
                # TODO needs to add correct file numbers
                data['phedex']['dataset'].append({'name' : dataset, 'files' : 5,
                                                  'subscription' : []})
                subs = data['phedex']['dataset'][-1]['subscription']
                    #FIXME: Take from self.locations
                subs.append({'node': 'SiteA', 'custodial': 'n', 'suspend_until': None,
                             'level': 'dataset', 'move': 'n', 'request': '47983',
                             'time_created': '1232989000', 'priority': 'low',
                             'time_update': None, 'node_id': '781',
                             'suspended': 'n', 'group': None})
            return data
        elif args.has_key('block') and args['block']:
            
            for block in args['block']:
                dataset = self.dataBlocks.getDataset('block')
                # TODO needs to add correct file numbers
                data['phedex']['dataset'].append({'name' : dataset, 'files' : 5,
                                              'block' : []})
                blocks = data['phedex']['dataset'][-1]['block']
                locations= self.dataBlocks.getLocation(block)
                        
                blocks.append({"bytes":"10438786614", "files":"5", "is_open":"n",
                               "name": args['block'],
                               "id":"454370", "subscription"
                                                  :[ {'node' : x } for x in locations]
                                                        #{"priority":"normal", "request":"51253", "time_created":"1245165314",
                                                        #   "move":"n", "suspend_until":None, "node":"SiteA",
                                                        #   "time_update":"1228905272", "group":None, "level":"block",
                                                        #   "node_id":"641", "custodial":"n", "suspended":"n"}]
                                                    })
            return data
        
    def emulator(self):
        return "PhEDEx emulator ...."

class DBSReader:
    """
    Mock up dbs access
    """
    def __init__(self, url, **contact):
        self.dataBlocks = DataBlockGenerator()
        args = { "url" : url, "level" : 'ERROR', "version" : 'DBS_2_0_9'}
        self.dbs = _MockDBSApi(args)
        
    def getFileBlocksInfo(self, dataset, onlyClosedBlocks = True,
                          blockName = '*', locations = True):

        """Fake block info"""
        blocks = [x for x in self.dataBlocks.getBlocks(dataset)
                if x['Name'] == blockName or blockName == '*']
        if locations:
            for block in blocks:
                block['StorageElementList'] = [{'Role' : '', 'Name' : x} for x in \
                                               self.listFileBlockLocation(block['Name'])]
        return blocks

    def listFileBlockLocation(self, block):
        """Fake locations"""
        return self.dataBlocks.getLocation(block)

    def listFilesInBlock(self, block):
        """Fake files"""
        return self.dataBlocks.getFiles(block)

    def listFilesInBlockWithParents(self, block):
        return self.dataBlocks.getFiles(block, True)

    def getFileBlock(self, block):
        """Return block + locations"""
        result = { block : {
            "StorageElements" : self.listFileBlockLocation(block),
            "Files" : self.listFilesInBlock(block),
            "IsOpen" : False,
            }
                }
        return result

    def getFileBlockWithParents(self, fileBlockName):
        """
        _getFileBlockWithParents_

        return a dictionary:
        { blockName: {
             "StorageElements" : [<se list>],
             "Files" : dictionaries representing each file
             }
        }

        files

        """

        result = { fileBlockName: {
            "StorageElements" : self.listFileBlockLocation(fileBlockName),
            "Files" : self.listFilesInBlockWithParents(fileBlockName),
            "IsOpen" : False,

            }
                   }
        return result

    def listRuns(self, dataset = None, block = None):
        def getRunsFromBlock(b):
            results = []
            for x in self.dataBlocks.getFiles(b):
                results.extend([y['RunNumber'] for y in x['LumiList']])
            return results

        if block:
            return getRunsFromBlock(block)
        if dataset:
            runs = []
            for block in self.dataBlocks.getBlocks(dataset):
                runs.extend(getRunsFromBlock(block['Name']))
            return runs
        return None


    def getDBSSummaryInfo(self, dataset=None, block=None):

        """Dataset summary"""
        def getLumisectionsInBlock(b):
            lumis = set()
            for file in self.dataBlocks.getFiles(b):
                for x in file['LumiList']:
                    lumis.add(x['LumiSectionNumber'])
            return lumis

        result = {}
        if block:
            result['NumberOfEvents'] = sum([x['NumberOfEvents']
                                for x in self.dataBlocks.getFiles(block)])
            result['NumberOfFiles'] = len(self.dataBlocks.getFiles(block))

            result['NumberOfLumis'] = len(getLumisectionsInBlock(block))
#.........這裏部分代碼省略.........