Python data_source.DataSource类代码示例

    def get_top_data (startdate, enddate, identities_db, filter_ = None, npeople = None):

        def filter_ips(ips):
            new_ips = {}
            new_ips['downloads'] = ips['downloads']
            new_ips['ips'] = []
            for ip in ips['ips']:
                # ipv4
                new_ip_aux = ip.split(".")
                new_ip = ip
                if len(new_ip_aux) == 4:
                    new_ip = "x.x."+new_ip_aux[2]+"."+new_ip_aux[3]
                # ipv6
                new_ip_aux = ip.split(":")
                if len(new_ip_aux) > 1:
                    new_ip = new_ip_aux[0]+":X"
                new_ips['ips'].append(new_ip)
            return new_ips

        top = {}
        mips = DataSource.get_metrics("ips", DownloadsDS)
        mpackages = DataSource.get_metrics("packages", DownloadsDS)
        period = None
        type_analysis = None
        if filter_ is not None:
            type_analysis = filter_.get_type_analysis()
        mfilter = MetricFilters(period, startdate, enddate, type_analysis, npeople)

        if filter_ is None:
            top['ips.'] = filter_ips(mips.get_list(mfilter, 0))
            top['packages.'] = mpackages.get_list(mfilter, 0)
        else:
            logging.info("DownloadsDS does not support yet top for filters.")

        return top

    def get_top_data(startdate, enddate, identities_db, filter_, npeople):
        top = {}
        mcsenders = DataSource.get_metrics("csenders", QAForums)
        masenders = DataSource.get_metrics("asenders", QAForums)
        mqsenders = DataSource.get_metrics("qsenders", QAForums)
        mparticipants = DataSource.get_metrics("participants", QAForums)
        period = None
        type_analysis = None
        if filter_ is not None:
            type_analysis = filter_.get_type_analysis()
        mfilter = MetricFilters(period, startdate, enddate, type_analysis, npeople)

        if filter_ is None:
            top['csenders.'] = mcsenders.get_list(mfilter, 0)
            top['csenders.last month'] = mcsenders.get_list(mfilter, 31)
            top['csenders.last year'] = mcsenders.get_list(mfilter, 365)

            top['asenders.'] = masenders.get_list(mfilter, 0)
            top['asenders.last month'] = masenders.get_list(mfilter, 31)
            top['asenders.last year'] = masenders.get_list(mfilter, 365)

            top['qsenders.'] = mqsenders.get_list(mfilter, 0)
            top['qsenders.last month'] = mqsenders.get_list(mfilter, 31)
            top['qsenders.last year'] = mqsenders.get_list(mfilter, 365)

            top['participants.'] = mparticipants.get_list(mfilter, 0)
            top['participants.last month'] = mparticipants.get_list(mfilter, 31)
            top['participants.last year'] = mparticipants.get_list(mfilter, 365)

        else:
            logging.info("QAForums does not support yet top for filters.")

        return top

 def get_events():
     events = {}
     newcomers = DataSource.get_metrics("newauthors", SCM)
     goneauthors = DataSource.get_metrics("goneauthors", SCM)
     if newcomers is None and goneauthors is None: return events
     events['newcomers'] = newcomers.get_list()
     events['goneauthors'] = goneauthors.get_list()
     return events

    def get_top_data (startdate, enddate, identities_db, filter_, npeople):
        bots = SCR.get_bots()
        top_all = None
        mreviewers = DataSource.get_metrics("reviewers", SCR)
        mopeners = DataSource.get_metrics("submitters", SCR)
        mmergers = DataSource.get_metrics("closers", SCR)
        mcorereviewers = DataSource.get_metrics("active_core_reviewers", SCR)
        mparticipants = DataSource.get_metrics("participants", SCR)
        period = None
        type_analysis = None
        if filter_ is not None:
            type_analysis = filter_.get_type_analysis()
        mfilter = MetricFilters(period, startdate, enddate, type_analysis, npeople)

        if filter_ is None:
            top_reviewers = {}
            top_reviewers['reviewers'] = mreviewers.get_list(mfilter, 0)
            top_reviewers['reviewers.last month']= mreviewers.get_list(mfilter, 31)
            top_reviewers['reviewers.last year']= mreviewers.get_list(mfilter, 365)

            top_openers = {}
            top_openers['openers.'] = mopeners.get_list(mfilter, 0)
            top_openers['openers.last month']= mopeners.get_list(mfilter, 31)
            top_openers['openers.last year'] = mopeners.get_list(mfilter, 365)

            top_mergers = {}
            top_mergers['mergers.'] = mmergers.get_list(mfilter, 0)
            top_mergers['mergers.last month'] = mmergers.get_list(mfilter, 31)
            top_mergers['mergers.last year'] = mmergers.get_list(mfilter, 365)

            top_participants = {}
            top_participants['participants.'] = mparticipants.get_list(mfilter, 0)
            top_participants['participants.last month'] = mparticipants.get_list(mfilter, 31)
            top_participants['participants.last year'] = mparticipants.get_list(mfilter, 365)

            top_core_reviewers = {}
            top_core_reviewers['active_core_reviewers.'] = mcorereviewers.get_list(mfilter, 0)
            top_core_reviewers['active_core_reviewers.last month'] = mcorereviewers.get_list(mfilter, 31)
            top_core_reviewers['active_core_reviewers.last year'] = mcorereviewers.get_list(mfilter, 365)

            # The order of the list item change so we can not check it
            top_all = dict(top_reviewers.items() +  top_openers.items() + top_mergers.items() + top_participants.items() + top_core_reviewers.items())
        else:
            filter_name = filter_.get_name()

            if filter_name in ["company","domain","repository"]:
                if filter_name in ["company","domain","repository"]:
                    top_all = {}
                    top_all['active_core_reviewers.'] = mcorereviewers.get_list(mfilter, 0)
                    top_all['active_core_reviewers.last month'] = mcorereviewers.get_list(mfilter, 31)
                    top_all['active_core_reviewers.last year'] = mcorereviewers.get_list(mfilter, 365)
                else:
                    # Remove filters above if there are performance issues
                    top_all = mcorereviewers.get_list(mfilter)
            else:
                top_all = None

        return (top_all)

def get_top_people (startdate, enddate, idb):
    """Top people for all data sources."""
    import vizgrimoire.GrimoireSQL
    from vizgrimoire.SCR import SCR
    from vizgrimoire.MLS import MLS
    from vizgrimoire.ITS import ITS
    from vizgrimoire.IRC import IRC
    from vizgrimoire.Mediawiki import Mediawiki
    from vizgrimoire.metrics.metrics_filter import MetricFilters
    from vizgrimoire.data_source import DataSource
    npeople = "10000" # max limit, all people included
    min_data_sources = 3 # min data sources to be in the list
    tops = {}
    all_top = {}
    all_top_min_ds = {}
    period = None
    type_analysis = None
    mfilter = MetricFilters(period, startdate, enddate, type_analysis, npeople)

    # SCR and SCM are the same. Don't use both for Tops
    mopeners = DataSource.get_metrics("submitters", SCR)
    if mopeners:
        tops["scr"] =  mopeners.get_list(mfilter, 0)
        tops["scr"]["identifier"] = tops["scr"].pop("openers")
    msenders = DataSource.get_metrics("senders", MLS)
    if msenders:
        tops["mls"] =  msenders.get_list(mfilter, 0)
        tops["mls"]["identifier"] = tops["mls"].pop("senders")
    mopeners = DataSource.get_metrics("openers", ITS)
    if mopeners:
        tops["its"] =  mopeners.get_list(mfilter, 0)
        tops["its"]["identifier"] = tops["its"].pop("openers")
    msenders = DataSource.get_metrics("senders", IRC)
    if msenders:
        tops["irc"] =  msenders.get_list(mfilter, 0)
        tops["irc"]["identifier"] = tops["irc"].pop("senders")
    mauthors = DataSource.get_metrics("authors", Mediawiki)
    if mauthors:
        tops["mediawiki"] = mauthors.get_list(mfilter, 0)
        tops["mediawiki"]["identifier"] = tops["mediawiki"].pop("reviews")

    # Build the consolidated top list using all data sources data
    # Only people in all data sources is used
    for ds in tops:
        pos = 1
        for id in tops[ds]['id']:
            if id not in all_top: all_top[id] = []
            all_top[id].append({"ds":ds,"pos":pos,"identifier":tops[ds]['identifier'][pos-1]})
            pos += 1

    for id in all_top:
        if len(all_top[id])>=min_data_sources: all_top_min_ds[id] = all_top[id]
    return all_top_min_ds

    def get_filter_items(filter_, startdate, enddate, identities_db):
        items = None
        filter_name = filter_.get_name()
        # Change filter to GrimoireLib notation
        filter_name = filter_name.replace("+", MetricFilters.DELIMITER)


        if (filter_name == "repository"):
            metric = DataSource.get_metrics("repositories", SCM)
        elif (filter_name == "company"):
            metric = DataSource.get_metrics("organizations", SCM)
        elif (filter_name == "country"):
            metric = DataSource.get_metrics("countries", SCM)
        elif (filter_name == "domain"):
            metric = DataSource.get_metrics("domains", SCM)
        elif (filter_name == "project"):
            metric = DataSource.get_metrics("projects", SCM)
        elif (filter_name == "people2"):
            metric = DataSource.get_metrics("people2", SCM)
        elif (filter_name == "company"+MetricFilters.DELIMITER+"country"):
            metric = DataSource.get_metrics("organizations+countries", SCM)
        elif (filter_name == "company"+MetricFilters.DELIMITER+"project"):
            metric = DataSource.get_metrics("organizations+projects", SCM)
        else:
            logging.error("SCM " + filter_name + " not supported")
            return items

        if metric is not None: items = metric.get_list()

        return items

    def get_filter_items(cls, filter_, startdate, enddate, identities_db):
        items = None
        filter_name = filter_.get_name()

        if (filter_name == "repository"):
            metric = DataSource.get_metrics("trackers", cls)
        elif (filter_name == "company"):
            metric = DataSource.get_metrics("organizations",  cls)
        elif (filter_name == "country"):
            metric = DataSource.get_metrics("countries", cls)
        elif (filter_name == "domain"):
            metric = DataSource.get_metrics("domains", cls)
        elif (filter_name == "project"):
            metric = DataSource.get_metrics("projects", cls)
        elif (filter_name == "people2"):
            metric = DataSource.get_metrics("people2", cls)
        elif (filter_name == "company"+MetricFilters.DELIMITER+"country"):
            metric = DataSource.get_metrics("organizations+countries", cls)
        elif (filter_name == "company"+MetricFilters.DELIMITER+"project"):
            metric = DataSource.get_metrics("organizations+projects", cls)
        else:
            logging.error(filter_name + " not supported")
            return items

        items = metric.get_list()
        return items

    def get_filter_items(filter_, startdate, enddate, identities_db):
        items = None
        filter_name = filter_.get_name()

        if (filter_name == "repository"):
            metric = DataSource.get_metrics("repositories", IRC)
            items = metric.get_list()
            # items = GetReposNameIRC()
        elif (filter_name == "people2"):
            metric = DataSource.get_metrics("senders", IRC)
            items = metric.get_list()
            items['name'] = items.pop('senders')
        else:
            logging.error("IRC " + filter_name + " not supported")
        return items

    def get_top_data (startdate, enddate, identities_db, filter_, npeople):
        bots = Pullpo.get_bots()
        top_all = None
        mreviewers = DataSource.get_metrics("reviewers", Pullpo)
        msubmitters = DataSource.get_metrics("submitters", Pullpo)
        mclosers = DataSource.get_metrics("closers", Pullpo)
        mmergers = DataSource.get_metrics("mergers", Pullpo)
        mparticipants = DataSource.get_metrics("participants", Pullpo)
        period = None
        type_analysis = None
        if filter_ is not None:
            type_analysis = filter_.get_type_analysis()
        mfilter = MetricFilters(period, startdate, enddate, type_analysis, npeople)

        if filter_ is None:
            top_reviewers = {}
            top_reviewers['reviewers'] = mreviewers.get_list(mfilter, 0)
            top_reviewers['reviewers.last month']= mreviewers.get_list(mfilter, 31)
            top_reviewers['reviewers.last year']= mreviewers.get_list(mfilter, 365)

            top_submitters = {}
            top_submitters['submitters.'] = msubmitters.get_list(mfilter, 0)
            top_submitters['submitters.last month']= msubmitters.get_list(mfilter, 31)
            top_submitters['submitters.last year'] = msubmitters.get_list(mfilter, 365)

            top_closers = {}
            top_closers['closers.'] = mclosers.get_list(mfilter, 0)
            top_closers['closers.last month'] = mclosers.get_list(mfilter, 31)
            top_closers['closers.last year'] = mclosers.get_list(mfilter, 365)

            top_mergers = {}
            top_mergers['mergers.'] = mmergers.get_list(mfilter, 0)
            top_mergers['mergers.last month'] = mmergers.get_list(mfilter, 31)
            top_mergers['mergers.last year'] = mmergers.get_list(mfilter, 365)

            top_participants = {}
            top_participants['participants.'] = mparticipants.get_list(mfilter, 0)
            top_participants['participants.last month'] = mparticipants.get_list(mfilter, 31)
            top_participants['participants.last year'] = mparticipants.get_list(mfilter, 365)

            # The order of the list item change so we can not check it
            top_all = dict(top_reviewers.items() +  top_submitters.items() +
                           top_closers.items() + top_mergers.items() +
                           top_participants.items())
        else:
            logging.info("Pullpo does not support yet top for filters.")

        return (top_all)

def GetClosedSummaryCompanies (period, startdate, enddate, identities_db, closed_condition, num_organizations):

    from vizgrimoire.ITS import ITS

    count = 1
    first_organizations = {}

    metric = DataSource.get_metrics("organizations", ITS)
    organizations = metric.get_list()
    organizations = organizations['name']

    for company in organizations:
        type_analysis = ["company", "'"+company+"'"]
        filter_com = MetricFilters(period, startdate, enddate, type_analysis)
        mclosed = ITS.get_metrics("closed", ITS)
        mclosed.filters = filter_com
        closed = mclosed.get_ts()
        # Rename field closed to company name
        closed[company] = closed["closed"]
        del closed['closed']

        if (count <= num_organizations):
            #Case of organizations with entity in the dataset
            first_organizations = dict(first_organizations.items() + closed.items())
        else :
            #Case of organizations that are aggregated in the field Others
            if 'Others' not in first_organizations:
                first_organizations['Others'] = closed[company]
            else:
                first_organizations['Others'] = [a+b for a, b in zip(first_organizations['Others'],closed[company])]
        count = count + 1
    first_organizations = completePeriodIds(first_organizations, period, startdate, enddate)

    return(first_organizations)

 def get_evolutionary_data (period, startdate, enddate, i_db, filter_ = None):
     metrics =  DockerHubDS.get_metrics_data(period, startdate, enddate, i_db, filter_, True)
     if filter_ is not None:
         studies = {}
     else:
         studies =  DataSource.get_studies_data(DockerHubDS, period, startdate, enddate, True)
     return dict(metrics.items()+studies.items())

def GetSentSummaryCompanies (period, startdate, enddate, identities_db, num_organizations, projects_db):
    count = 1
    first_organizations = {}

    metric = DataSource.get_metrics("organizations", MLS)
    organizations = metric.get_list()

    for company in organizations:
        type_analysis = ["company", "'"+company+"'"]
        sent = EvolEmailsSent(period, startdate, enddate, identities_db, type_analysis, projects_db)
        sent = completePeriodIds(sent, period, startdate, enddate)
        # Rename field sent to company name
        sent[company] = sent["sent"]
        del sent['sent']

        if (count <= num_organizations):
            #Case of organizations with entity in the dataset
            first_organizations = dict(first_organizations.items() + sent.items())
        else :
            #Case of organizations that are aggregated in the field Others
            if 'Others' not in first_organizations:
                first_organizations['Others'] = sent[company]
            else:
                first_organizations['Others'] = [a+b for a, b in zip(first_organizations['Others'],sent[company])]
        count = count + 1

    first_organizations = completePeriodIds(first_organizations, period, startdate, enddate)

    return(first_organizations)

    def get_filter_items(filter_, startdate, enddate, identities_db):
        items = None
        filter_name = filter_.get_name()
        #TODO: repository needs to be change to tag, once this is accepted as new
        #      data source in VizGrimoireJS-lib
        if (filter_name == "repository"):
            metric = DataSource.get_metrics("tags", QAForums)
            # items = QAForums.tags_name(startdate, enddate)
            items = metric.get_list()
        if (filter_name == "people2"):
            metric = DataSource.get_metrics("participants", QAForums)
            items = metric.get_list()
        else:
            logging.error("QAForums " + filter_name + " not supported")

        return items

    def get_top_data_authors (startdate, enddate, i_db, filter_, npeople):
        top = {}
        mauthors = DataSource.get_metrics("authors", SCM)
        if mauthors is None: return top
        period = None
        type_analysis = None
        if filter_ is not None:
            type_analysis = filter_.get_type_analysis()
        mfilter = MetricFilters(period, startdate, enddate, type_analysis, npeople)
        mfilter.global_filter = mauthors.filters.global_filter

        if filter_ is None:
            top['authors.'] = mauthors.get_list(mfilter, 0)
            top['authors.last month'] = mauthors.get_list(mfilter, 31)
            top['authors.last year'] = mauthors.get_list(mfilter, 365)
        elif filter_.get_name() in ["company","repository","project"]:
            if filter_.get_name() in ["company","repository","project"]:
                top['authors.'] = mauthors.get_list(mfilter, 0)
                top['authors.last month'] = mauthors.get_list(mfilter, 31)
                top['authors.last year'] = mauthors.get_list(mfilter, 365)
            else:
                # If we have performance issues with tops, remove filters above
                # to avoid computing trends for tops
                top = mauthors.get_list(mfilter) 
        else:
            logging.info("Top authors not support for " + filter_.get_name())
        return top

    def get_list(self):
        from vizgrimoire.data_source import DataSource
        from vizgrimoire.filter import Filter
        startdate = self.filters.startdate
        enddate = self.filters.enddate
        closed_condition = ITS._get_closed_condition()
        if self.filters.closed_condition is not None:
             closed_condition = self.filters.closed_condition

        bots = DataSource.get_filter_bots(Filter("domain"))
        fbots = ''
        for bot in bots:
            fbots += " dom.name<>'"+bot+"' and "

        tables = Set([])
        filters = Set([])

        tables.union_update(self.db.GetTablesDomains(self.db.identities_db))
        tables.add(self.db.identities_db + ".domains dom")
        tables_str = self.db._get_tables_query(tables)
        filters.union_update(self.db.GetFiltersDomains())
        filters_str = self.db._get_filters_query(filters)

        q = "SELECT dom.name "+\
            "FROM "+ tables_str + " "+\
            "WHERE " + filters_str +" AND "+\
            "       dom.id = upd.domain_id and "+\
            "       "+ fbots +" "+\
            "       c.changed_on >= "+ startdate+ " AND "+\
            "       c.changed_on < "+ enddate+ " AND "+\
            "       "+ closed_condition+" "+\
            "GROUP BY dom.name "+\
            "ORDER BY COUNT(DISTINCT(c.issue_id)) DESC LIMIT " + str(Metrics.domains_limit)
        data = self.db.ExecuteQuery(q)
        return (data)