Ejemplo n.º 1
0
def write_to_es7(env):
    ELASTICSEARCH_SQL_CONNECTOR_PATH = \
        'file:///path/to/flink-sql-connector-elasticsearch7-1.16.0.jar'
    env.add_jars(ELASTICSEARCH_SQL_CONNECTOR_PATH)

    ds = env.from_collection([{
        'name': 'ada',
        'id': '1'
    }, {
        'name': 'luna',
        'id': '2'
    }],
                             type_info=Types.MAP(Types.STRING(),
                                                 Types.STRING()))

    es7_sink = Elasticsearch7SinkBuilder() \
        .set_emitter(ElasticsearchEmitter.static_index('foo', 'id')) \
        .set_hosts(['localhost:9200']) \
        .set_delivery_guarantee(DeliveryGuarantee.AT_LEAST_ONCE) \
        .set_bulk_flush_max_actions(1) \
        .set_bulk_flush_max_size_mb(2) \
        .set_bulk_flush_interval(1000) \
        .set_bulk_flush_backoff_strategy(FlushBackoffType.CONSTANT, 3, 3000) \
        .set_connection_username('foo') \
        .set_connection_password('bar') \
        .set_connection_path_prefix('foo-bar') \
        .set_connection_request_timeout(30000) \
        .set_connection_timeout(31000) \
        .set_socket_timeout(32000) \
        .build()

    ds.sink_to(es7_sink).name('es7 sink')

    env.execute()
Ejemplo n.º 2
0
    def test_es_sink_dynamic(self):
        ds = self.env.from_collection([{
            'name': 'ada',
            'id': '1'
        }, {
            'name': 'luna',
            'id': '2'
        }],
                                      type_info=Types.MAP(
                                          Types.STRING(), Types.STRING()))

        es_dynamic_index_sink = Elasticsearch7SinkBuilder() \
            .set_emitter(ElasticsearchEmitter.dynamic_index('name', 'id')) \
            .set_hosts(['localhost:9200']) \
            .build()

        j_emitter = get_field_value(es_dynamic_index_sink.get_java_function(),
                                    'emitter')
        self.assertTrue(
            is_instance_of(
                j_emitter,
                'org.apache.flink.connector.elasticsearch.sink.MapElasticsearchEmitter'
            ))

        ds.sink_to(es_dynamic_index_sink).name('es dynamic index sink')
Ejemplo n.º 3
0
def write_to_es6_dynamic_index(env):
    ELASTICSEARCH_SQL_CONNECTOR_PATH = \
        'file:///path/to/flink-sql-connector-elasticsearch6-1.16.0.jar'
    env.add_jars(ELASTICSEARCH_SQL_CONNECTOR_PATH)

    ds = env.from_collection([{
        'name': 'ada',
        'id': '1'
    }, {
        'name': 'luna',
        'id': '2'
    }],
                             type_info=Types.MAP(Types.STRING(),
                                                 Types.STRING()))

    es_sink = Elasticsearch6SinkBuilder() \
        .set_emitter(ElasticsearchEmitter.dynamic_index('name', 'id', 'bar')) \
        .set_hosts(['localhost:9200']) \
        .build()

    ds.sink_to(es_sink).name('es6 dynamic index sink')

    env.execute()
Ejemplo n.º 4
0
    def test_es_sink(self):
        ds = self.env.from_collection([{
            'name': 'ada',
            'id': '1'
        }, {
            'name': 'luna',
            'id': '2'
        }],
                                      type_info=Types.MAP(
                                          Types.STRING(), Types.STRING()))

        es_sink = Elasticsearch7SinkBuilder() \
            .set_emitter(ElasticsearchEmitter.static_index('foo', 'id')) \
            .set_hosts(['localhost:9200']) \
            .set_delivery_guarantee(DeliveryGuarantee.AT_LEAST_ONCE) \
            .set_bulk_flush_max_actions(1) \
            .set_bulk_flush_max_size_mb(2) \
            .set_bulk_flush_interval(1000) \
            .set_bulk_flush_backoff_strategy(FlushBackoffType.CONSTANT, 3, 3000) \
            .set_connection_username('foo') \
            .set_connection_password('bar') \
            .set_connection_path_prefix('foo-bar') \
            .set_connection_request_timeout(30000) \
            .set_connection_timeout(31000) \
            .set_socket_timeout(32000) \
            .build()

        j_emitter = get_field_value(es_sink.get_java_function(), 'emitter')
        self.assertTrue(
            is_instance_of(
                j_emitter,
                'org.apache.flink.connector.elasticsearch.sink.MapElasticsearchEmitter'
            ))
        self.assertEqual(
            get_field_value(es_sink.get_java_function(),
                            'hosts')[0].toString(), 'http://localhost:9200')
        self.assertEqual(
            get_field_value(es_sink.get_java_function(),
                            'deliveryGuarantee').toString(), 'at-least-once')

        j_build_bulk_processor_config = get_field_value(
            es_sink.get_java_function(), 'buildBulkProcessorConfig')
        self.assertEqual(
            j_build_bulk_processor_config.getBulkFlushMaxActions(), 1)
        self.assertEqual(j_build_bulk_processor_config.getBulkFlushMaxMb(), 2)
        self.assertEqual(j_build_bulk_processor_config.getBulkFlushInterval(),
                         1000)
        self.assertEqual(
            j_build_bulk_processor_config.getFlushBackoffType().toString(),
            'CONSTANT')
        self.assertEqual(
            j_build_bulk_processor_config.getBulkFlushBackoffRetries(), 3)
        self.assertEqual(
            j_build_bulk_processor_config.getBulkFlushBackOffDelay(), 3000)

        j_network_client_config = get_field_value(es_sink.get_java_function(),
                                                  'networkClientConfig')
        self.assertEqual(j_network_client_config.getUsername(), 'foo')
        self.assertEqual(j_network_client_config.getPassword(), 'bar')
        self.assertEqual(j_network_client_config.getConnectionRequestTimeout(),
                         30000)
        self.assertEqual(j_network_client_config.getConnectionTimeout(), 31000)
        self.assertEqual(j_network_client_config.getSocketTimeout(), 32000)
        self.assertEqual(j_network_client_config.getConnectionPathPrefix(),
                         'foo-bar')

        ds.sink_to(es_sink).name('es sink')