Example #1
0
    def execute(self, context):
        hive = HiveCliHook(hive_cli_conn_id=self.hive_cli_conn_id)
        logging.info("Extracting data from Hive")
        hive_table = 'druid.' + context['task_instance_key_str']
        sql = self.sql.strip().strip(';')
        hql = """\
        set mapred.output.compress=false;
        set hive.exec.compress.output=false;
        DROP TABLE IF EXISTS {hive_table};
        CREATE TABLE {hive_table}
        ROW FORMAT DELIMITED FIELDS TERMINATED BY  '\t'
        STORED AS TEXTFILE AS
        {sql};
        """.format(**locals())
        hive.run_cli(hql)

        m = HiveMetastoreHook(self.metastore_conn_id)
        t = m.get_table(hive_table)

        columns = [col.name for col in t.sd.cols]

        hdfs_uri = m.get_table(hive_table).sd.location
        pos = hdfs_uri.find('/user')
        static_path = hdfs_uri[pos:]

        druid = DruidHook(druid_ingest_conn_id=self.druid_ingest_conn_id)
        logging.info("Inserting rows into Druid")
        druid.load_from_hdfs(datasource=self.druid_datasource,
                             intervals=self.intervals,
                             static_path=static_path,
                             ts_dim=self.ts_dim,
                             columns=columns,
                             metric_spec=self.metric_spec)
        logging.info("Load seems to have succeeded!")
Example #2
0
    def execute(self, context):
        hive = HiveCliHook(hive_cli_conn_id=self.hive_cli_conn_id)
        logging.info("Extracting data from Hive")
        hive_table = 'druid.' + context['task_instance_key_str']
        sql = self.sql.strip().strip(';')
        hql = """\
        set mapred.output.compress=false;
        set hive.exec.compress.output=false;
        DROP TABLE IF EXISTS {hive_table};
        CREATE TABLE {hive_table}
        ROW FORMAT DELIMITED FIELDS TERMINATED BY  '\t'
        STORED AS TEXTFILE AS
        {sql};
        """.format(**locals())
        #hive.run_cli(hql)

        m = HiveMetastoreHook(self.metastore_conn_id)
        t = m.get_table(hive_table)

        columns = [col.name for col in t.sd.cols]

        hdfs_uri = m.get_table(hive_table).sd.location
        pos = hdfs_uri.find('/user')
        static_path = hdfs_uri[pos:]

        druid = DruidHook(druid_ingest_conn_id=self.druid_ingest_conn_id)
        logging.info("Inserting rows into Druid")
        druid.load_from_hdfs(
            datasource=self.druid_datasource,
            intervals=self.intervals,
            static_path=static_path, ts_dim=self.ts_dim,
            columns=columns, metric_spec=self.metric_spec)
        logging.info("Load seems to have succeeded!")
    def execute(self, context):
        hive = HiveCliHook(hive_cli_conn_id=self.hive_cli_conn_id)
        logging.info("Extracting data from Hive")
        hive_table = 'druid.' + context['task_instance_key_str']
        sql = self.sql.strip().strip(';')
        hql = """\
        set mapred.output.compress=false;
        set hive.exec.compress.output=false;
        DROP TABLE IF EXISTS {hive_table};
        CREATE TABLE {hive_table}
        ROW FORMAT DELIMITED FIELDS TERMINATED BY  '\t'
        STORED AS TEXTFILE
        TBLPROPERTIES ('serialization.null.format' = '')
        AS
        {sql}
        """.format(**locals())
        hive.run_cli(hql)
        #hqls = hql.split(';')
        #logging.info(str(hqls))
        #from airflow.hooks import HiveServer2Hook
        #hive = HiveServer2Hook(hiveserver2_conn_id="hiveserver2_silver")
        #hive.get_results(hqls)


        m = HiveMetastoreHook(self.metastore_conn_id)
        t = m.get_table(hive_table)

        columns = [col.name for col in t.sd.cols]

        hdfs_uri = m.get_table(hive_table).sd.location
        pos = hdfs_uri.find('/user')
        static_path = hdfs_uri[pos:]

        schema, table = hive_table.split('.')

        druid = DruidHook(druid_ingest_conn_id=self.druid_ingest_conn_id)
        logging.info("Inserting rows into Druid")
        logging.info("HDFS path: " + static_path)

        druid.load_from_hdfs(
            datasource=self.druid_datasource,
            intervals=self.intervals,
            static_path=static_path, ts_dim=self.ts_dim,
            columns=columns, metric_spec=self.metric_spec,
            hadoop_dependency_coordinates=self.hadoop_dependency_coordinates)
        logging.info("Load seems to have succeeded!")

        logging.info(
            "Cleaning up by dropping the temp "
            "Hive table {}".format(hive_table))
        hql = "DROP TABLE IF EXISTS {}".format(hive_table)
Example #4
0
    def execute(self, context):
        hive = HiveCliHook(hive_cli_conn_id=self.hive_cli_conn_id)
        logging.info("Extracting data from Hive")
        hive_table = 'druid.' + context['task_instance_key_str']
        sql = self.sql.strip().strip(';')
        hql = """\
        set mapred.output.compress=false;
        set hive.exec.compress.output=false;
        DROP TABLE IF EXISTS {hive_table};
        CREATE TABLE {hive_table}
        ROW FORMAT DELIMITED FIELDS TERMINATED BY  '\t'
        STORED AS TEXTFILE
        TBLPROPERTIES ('serialization.null.format' = '')
        AS
        {sql}
        """.format(**locals())
        hive.run_cli(hql)
        #hqls = hql.split(';')
        #logging.info(str(hqls))
        #from airflow.hooks import HiveServer2Hook
        #hive = HiveServer2Hook(hiveserver2_conn_id="hiveserver2_silver")
        #hive.get_results(hqls)

        m = HiveMetastoreHook(self.metastore_conn_id)
        t = m.get_table(hive_table)

        columns = [col.name for col in t.sd.cols]

        hdfs_uri = m.get_table(hive_table).sd.location
        pos = hdfs_uri.find('/user')
        static_path = hdfs_uri[pos:]

        schema, table = hive_table.split('.')

        druid = DruidHook(druid_ingest_conn_id=self.druid_ingest_conn_id)
        logging.info("Inserting rows into Druid")
        logging.info("HDFS path: " + static_path)

        druid.load_from_hdfs(
            datasource=self.druid_datasource,
            intervals=self.intervals,
            static_path=static_path,
            ts_dim=self.ts_dim,
            columns=columns,
            metric_spec=self.metric_spec,
            hadoop_dependency_coordinates=self.hadoop_dependency_coordinates)
        logging.info("Load seems to have succeeded!")

        logging.info("Cleaning up by dropping the temp "
                     "Hive table {}".format(hive_table))
        hql = "DROP TABLE IF EXISTS {}".format(hive_table)