Java Code Examples for org.apache.flink.table.api.java.StreamTableEnvironment#registerTableSink()

The following examples show how to use org.apache.flink.table.api.java.StreamTableEnvironment#registerTableSink() . You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may check out the related API usage on the sidebar.
Example 1
Source File: KafkaSourceMain.java    From flink-learning with Apache License 2.0 6 votes vote down vote up 
public static void main(String[] args) throws Exception {
    StreamExecutionEnvironment blinkStreamEnv = StreamExecutionEnvironment.getExecutionEnvironment();
    blinkStreamEnv.setParallelism(1);
    EnvironmentSettings blinkStreamSettings = EnvironmentSettings.newInstance()
            .useBlinkPlanner()
            .inStreamingMode()
            .build();
    StreamTableEnvironment blinkStreamTableEnv = StreamTableEnvironment.create(blinkStreamEnv, blinkStreamSettings);

    ParameterTool parameterTool = ExecutionEnvUtil.PARAMETER_TOOL;
    Properties properties = KafkaConfigUtil.buildKafkaProps(parameterTool);
    DataStream<String> dataStream = blinkStreamEnv.addSource(new FlinkKafkaConsumer011<>(parameterTool.get("kafka.topic"), new SimpleStringSchema(), properties));
    Table table = blinkStreamTableEnv.fromDataStream(dataStream, "word");
    blinkStreamTableEnv.registerTable("kafkaDataStream", table);

    RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"_count", "word"}, new DataType[]{DataTypes.BIGINT(), DataTypes.STRING()});
    blinkStreamTableEnv.registerTableSink("sinkTable", retractStreamTableSink);

    Table wordCount = blinkStreamTableEnv.sqlQuery("SELECT count(word) AS _count,word FROM kafkaDataStream GROUP BY word");

    wordCount.insertInto("sinkTable");

    blinkStreamTableEnv.execute("Blink Kafka Table Source");
}
Example 2
Source File: CustomKafkaSourceMain.java    From flink-learning with Apache License 2.0 6 votes vote down vote up 
public static void main(String[] args) throws Exception {
    StreamExecutionEnvironment blinkStreamEnv = StreamExecutionEnvironment.getExecutionEnvironment();
    blinkStreamEnv.setParallelism(1);
    EnvironmentSettings blinkStreamSettings = EnvironmentSettings.newInstance()
            .useBlinkPlanner()
            .inStreamingMode()
            .build();
    StreamTableEnvironment blinkStreamTableEnv = StreamTableEnvironment.create(blinkStreamEnv, blinkStreamSettings);

    blinkStreamTableEnv.registerTableSource("kafkaDataStream", new MyKafkaTableSource(ExecutionEnvUtil.PARAMETER_TOOL));

    RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"_count", "word"}, new DataType[]{DataTypes.BIGINT(), DataTypes.STRING()});
    blinkStreamTableEnv.registerTableSink("sinkTable", retractStreamTableSink);

    Table wordCount = blinkStreamTableEnv.sqlQuery("SELECT count(word) AS _count,word FROM kafkaDataStream GROUP BY word");

    wordCount.insertInto("sinkTable");

    blinkStreamTableEnv.execute("Blink Custom Kafka Table Source");
}
Example 3
Source File: KafkaSourceMain.java    From flink-learning with Apache License 2.0 6 votes vote down vote up 
public static void main(String[] args) throws Exception {
    StreamExecutionEnvironment blinkStreamEnv = StreamExecutionEnvironment.getExecutionEnvironment();
    blinkStreamEnv.setParallelism(1);
    EnvironmentSettings blinkStreamSettings = EnvironmentSettings.newInstance()
            .useBlinkPlanner()
            .inStreamingMode()
            .build();
    StreamTableEnvironment blinkStreamTableEnv = StreamTableEnvironment.create(blinkStreamEnv, blinkStreamSettings);

    ParameterTool parameterTool = ExecutionEnvUtil.PARAMETER_TOOL;
    Properties properties = KafkaConfigUtil.buildKafkaProps(parameterTool);
    DataStream<String> dataStream = blinkStreamEnv.addSource(new FlinkKafkaConsumer011<>(parameterTool.get("kafka.topic"), new SimpleStringSchema(), properties));
    Table table = blinkStreamTableEnv.fromDataStream(dataStream, "word");
    blinkStreamTableEnv.registerTable("kafkaDataStream", table);

    RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"_count", "word"}, new DataType[]{DataTypes.BIGINT(), DataTypes.STRING()});
    blinkStreamTableEnv.registerTableSink("sinkTable", retractStreamTableSink);

    Table wordCount = blinkStreamTableEnv.sqlQuery("SELECT count(word) AS _count,word FROM kafkaDataStream GROUP BY word");

    wordCount.insertInto("sinkTable");

    blinkStreamTableEnv.execute("Blink Kafka Table Source");
}
Example 4
Source File: CustomKafkaSourceMain.java    From flink-learning with Apache License 2.0 6 votes vote down vote up 
public static void main(String[] args) throws Exception {
    StreamExecutionEnvironment blinkStreamEnv = StreamExecutionEnvironment.getExecutionEnvironment();
    blinkStreamEnv.setParallelism(1);
    EnvironmentSettings blinkStreamSettings = EnvironmentSettings.newInstance()
            .useBlinkPlanner()
            .inStreamingMode()
            .build();
    StreamTableEnvironment blinkStreamTableEnv = StreamTableEnvironment.create(blinkStreamEnv, blinkStreamSettings);

    blinkStreamTableEnv.registerTableSource("kafkaDataStream", new MyKafkaTableSource(ExecutionEnvUtil.PARAMETER_TOOL));

    RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"_count", "word"}, new DataType[]{DataTypes.BIGINT(), DataTypes.STRING()});
    blinkStreamTableEnv.registerTableSink("sinkTable", retractStreamTableSink);

    Table wordCount = blinkStreamTableEnv.sqlQuery("SELECT count(word) AS _count,word FROM kafkaDataStream GROUP BY word");

    wordCount.insertInto("sinkTable");

    blinkStreamTableEnv.execute("Blink Custom Kafka Table Source");
}
Example 5
Source File: CassandraConnectorITCase.java    From Flink-CEPplus with Apache License 2.0 5 votes vote down vote up 
@Test
public void testCassandraTableSink() throws Exception {
	StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(4);
	StreamTableEnvironment tEnv = StreamTableEnvironment.create(env);

	DataStreamSource<Row> source = env.fromCollection(rowCollection);

	tEnv.registerDataStreamInternal("testFlinkTable", source);
	tEnv.registerTableSink(
		"cassandraTable",
		new String[]{"f0", "f1", "f2"},
		new TypeInformation[]{Types.STRING, Types.INT, Types.INT},
		new CassandraAppendTableSink(builder, injectTableName(INSERT_DATA_QUERY)));

	tEnv.sqlQuery("select * from testFlinkTable").insertInto("cassandraTable");

	env.execute();
	ResultSet rs = session.execute(injectTableName(SELECT_DATA_QUERY));

	// validate that all input was correctly written to Cassandra
	List<Row> input = new ArrayList<>(rowCollection);
	List<com.datastax.driver.core.Row> output = rs.all();
	for (com.datastax.driver.core.Row o : output) {
		Row cmp = new Row(3);
		cmp.setField(0, o.getString(0));
		cmp.setField(1, o.getInt(2));
		cmp.setField(2, o.getInt(1));
		Assert.assertTrue("Row " + cmp + " was written to Cassandra but not in input.", input.remove(cmp));
	}
	Assert.assertTrue("The input data was not completely written to Cassandra", input.isEmpty());
}
Example 6
Source File: CustomTableSinkMain.java    From flink-learning with Apache License 2.0 5 votes vote down vote up 
public static void main(String[] args) throws Exception {
        StreamExecutionEnvironment blinkStreamEnv = StreamExecutionEnvironment.getExecutionEnvironment();
        blinkStreamEnv.setParallelism(1);
        EnvironmentSettings blinkStreamSettings = EnvironmentSettings.newInstance()
                .useBlinkPlanner()
                .inStreamingMode()
                .build();
        StreamTableEnvironment blinkStreamTableEnv = StreamTableEnvironment.create(blinkStreamEnv, blinkStreamSettings);

        String path = SQLExampleWordCount.class.getClassLoader().getResource("words.txt").getPath();

        CsvTableSource csvTableSource = CsvTableSource.builder()
                .field("word", Types.STRING)
                .path(path)
                .build();
        blinkStreamTableEnv.registerTableSource("zhisheng", csvTableSource);

        RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"c", "word"}, new TypeInformation[]{Types.LONG, Types.STRING});
        //或者
//        RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"c", "word"}, new DataType[]{DataTypes.BIGINT(), DataTypes.STRING()});
        blinkStreamTableEnv.registerTableSink("sinkTable", retractStreamTableSink);

        Table wordWithCount = blinkStreamTableEnv.sqlQuery("SELECT count(word) AS c, word FROM zhisheng GROUP BY word");

        wordWithCount.insertInto("sinkTable");
        blinkStreamTableEnv.execute("Blink Custom Table Sink");
    }
Example 7
Source File: CassandraConnectorITCase.java    From flink with Apache License 2.0 5 votes vote down vote up 
@Test
public void testCassandraTableSink() throws Exception {
	StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
	env.setParallelism(4);
	StreamTableEnvironment tEnv = StreamTableEnvironment.create(env);

	DataStreamSource<Row> source = env.fromCollection(rowCollection);

	tEnv.registerDataStream("testFlinkTable", source);
	tEnv.registerTableSink(
		"cassandraTable",
		new CassandraAppendTableSink(builder, injectTableName(INSERT_DATA_QUERY)).configure(
			new String[]{"f0", "f1", "f2"},
			new TypeInformation[]{Types.STRING, Types.INT, Types.INT}
		));

	tEnv.sqlQuery("select * from testFlinkTable").insertInto("cassandraTable");

	env.execute();
	ResultSet rs = session.execute(injectTableName(SELECT_DATA_QUERY));

	// validate that all input was correctly written to Cassandra
	List<Row> input = new ArrayList<>(rowCollection);
	List<com.datastax.driver.core.Row> output = rs.all();
	for (com.datastax.driver.core.Row o : output) {
		Row cmp = new Row(3);
		cmp.setField(0, o.getString(0));
		cmp.setField(1, o.getInt(2));
		cmp.setField(2, o.getInt(1));
		Assert.assertTrue("Row " + cmp + " was written to Cassandra but not in input.", input.remove(cmp));
	}
	Assert.assertTrue("The input data was not completely written to Cassandra", input.isEmpty());
}
Example 8
Source File: JDBCUpsertTableSinkITCase.java    From flink with Apache License 2.0 5 votes vote down vote up 
@Test
public void testUpsert() throws Exception {
	StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
	env.getConfig().enableObjectReuse();
	env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime);
	StreamTableEnvironment tEnv = StreamTableEnvironment.create(env);

	Table t = tEnv.fromDataStream(get3TupleDataStream(env).assignTimestampsAndWatermarks(
			new AscendingTimestampExtractor<Tuple3<Integer, Long, String>>() {
				@Override
				public long extractAscendingTimestamp(Tuple3<Integer, Long, String> element) {
					return element.f0;
				}}), "id, num, text");

	tEnv.registerTable("T", t);

	String[] fields = {"cnt", "lencnt", "cTag"};
	tEnv.registerTableSink("upsertSink", JDBCUpsertTableSink.builder()
			.setOptions(JDBCOptions.builder()
					.setDBUrl(DB_URL)
					.setTableName(OUTPUT_TABLE1)
					.build())
			.setTableSchema(TableSchema.builder().fields(
					fields, new DataType[] {BIGINT(), BIGINT(), INT()}).build())
			.build());

	tEnv.sqlUpdate("INSERT INTO upsertSink SELECT cnt, COUNT(len) AS lencnt, cTag FROM" +
			" (SELECT len, COUNT(id) as cnt, cTag FROM" +
			" (SELECT id, CHAR_LENGTH(text) AS len, (CASE WHEN id > 0 THEN 1 ELSE 0 END) cTag FROM T)" +
			" GROUP BY len, cTag)" +
			" GROUP BY cnt, cTag");
	env.execute();
	check(new Row[] {
			Row.of(1, 5, 1),
			Row.of(7, 1, 1),
			Row.of(9, 1, 1)
	}, DB_URL, OUTPUT_TABLE1, fields);
}
Example 9
Source File: JDBCUpsertTableSinkITCase.java    From flink with Apache License 2.0 5 votes vote down vote up 
@Test
public void testAppend() throws Exception {
	StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
	env.getConfig().enableObjectReuse();
	env.getConfig().setParallelism(1);
	StreamTableEnvironment tEnv = StreamTableEnvironment.create(env);

	Table t = tEnv.fromDataStream(get3TupleDataStream(env), "id, num, text");

	tEnv.registerTable("T", t);

	String[] fields = {"id", "num"};
	tEnv.registerTableSink("upsertSink", JDBCUpsertTableSink.builder()
			.setOptions(JDBCOptions.builder()
					.setDBUrl(DB_URL)
					.setTableName(OUTPUT_TABLE2)
					.build())
			.setTableSchema(TableSchema.builder().fields(
					fields, new DataType[] {INT(), BIGINT()}).build())
			.build());

	tEnv.sqlUpdate("INSERT INTO upsertSink SELECT id, num FROM T WHERE id IN (2, 10, 20)");
	env.execute();
	check(new Row[] {
			Row.of(2, 2),
			Row.of(10, 4),
			Row.of(20, 6)
	}, DB_URL, OUTPUT_TABLE2, fields);
}
Example 10
Source File: CustomTableSinkMain.java    From flink-learning with Apache License 2.0 5 votes vote down vote up 
public static void main(String[] args) throws Exception {
        StreamExecutionEnvironment blinkStreamEnv = StreamExecutionEnvironment.getExecutionEnvironment();
        blinkStreamEnv.setParallelism(1);
        EnvironmentSettings blinkStreamSettings = EnvironmentSettings.newInstance()
                .useBlinkPlanner()
                .inStreamingMode()
                .build();
        StreamTableEnvironment blinkStreamTableEnv = StreamTableEnvironment.create(blinkStreamEnv, blinkStreamSettings);

        String path = SQLExampleWordCount.class.getClassLoader().getResource("words.txt").getPath();

        CsvTableSource csvTableSource = CsvTableSource.builder()
                .field("word", Types.STRING)
                .path(path)
                .build();
        blinkStreamTableEnv.registerTableSource("zhisheng", csvTableSource);

        RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"c", "word"}, new TypeInformation[]{Types.LONG, Types.STRING});
        //或者
//        RetractStreamTableSink<Row> retractStreamTableSink = new MyRetractStreamTableSink(new String[]{"c", "word"}, new DataType[]{DataTypes.BIGINT(), DataTypes.STRING()});
        blinkStreamTableEnv.registerTableSink("sinkTable", retractStreamTableSink);

        Table wordWithCount = blinkStreamTableEnv.sqlQuery("SELECT count(word) AS c, word FROM zhisheng GROUP BY word");

        wordWithCount.insertInto("sinkTable");
        blinkStreamTableEnv.execute("Blink Custom Table Sink");
    }
Example 11
Source File: HBaseConnectorITCase.java    From flink with Apache License 2.0 4 votes vote down vote up 
@Test
public void testTableSink() throws Exception {
	HBaseTableSchema schema = new HBaseTableSchema();
	schema.addColumn(FAMILY1, F1COL1, Integer.class);
	schema.addColumn(FAMILY2, F2COL1, String.class);
	schema.addColumn(FAMILY2, F2COL2, Long.class);
	schema.setRowKey("rk", Integer.class);
	schema.addColumn(FAMILY3, F3COL1, Double.class);
	schema.addColumn(FAMILY3, F3COL2, Boolean.class);
	schema.addColumn(FAMILY3, F3COL3, String.class);

	Map<String, String> tableProperties = new HashMap<>();
	tableProperties.put("connector.type", "hbase");
	tableProperties.put("connector.version", "1.4.3");
	tableProperties.put("connector.property-version", "1");
	tableProperties.put("connector.table-name", TEST_TABLE_2);
	tableProperties.put("connector.zookeeper.quorum", getZookeeperQuorum());
	tableProperties.put("connector.zookeeper.znode.parent", "/hbase");
	DescriptorProperties descriptorProperties = new DescriptorProperties(true);
	descriptorProperties.putTableSchema(SCHEMA, schema.convertsToTableSchema());
	descriptorProperties.putProperties(tableProperties);
	TableSink tableSink = TableFactoryService
		.find(HBaseTableFactory.class, descriptorProperties.asMap())
		.createTableSink(descriptorProperties.asMap());

	StreamExecutionEnvironment execEnv = StreamExecutionEnvironment.getExecutionEnvironment();
	StreamTableEnvironment tEnv = StreamTableEnvironment.create(execEnv, streamSettings);

	DataStream<Row> ds = execEnv.fromCollection(testData1).returns(testTypeInfo1);
	tEnv.registerDataStream("src", ds);
	tEnv.registerTableSink("hbase", tableSink);

	String query = "INSERT INTO hbase SELECT ROW(f1c1), ROW(f2c1, f2c2), rowkey, ROW(f3c1, f3c2, f3c3) FROM src";
	tEnv.sqlUpdate(query);

	// wait to finish
	tEnv.execute("HBase Job");

	// start a batch scan job to verify contents in HBase table
	// start a batch scan job to verify contents in HBase table
	TableEnvironment batchTableEnv = createBatchTableEnv();

	HBaseTableSource hbaseTable = new HBaseTableSource(getConf(), TEST_TABLE_2);
	hbaseTable.setRowKey("rowkey", Integer.class);
	hbaseTable.addColumn(FAMILY1, F1COL1, Integer.class);
	hbaseTable.addColumn(FAMILY2, F2COL1, String.class);
	hbaseTable.addColumn(FAMILY2, F2COL2, Long.class);
	hbaseTable.addColumn(FAMILY3, F3COL1, Double.class);
	hbaseTable.addColumn(FAMILY3, F3COL2, Boolean.class);
	hbaseTable.addColumn(FAMILY3, F3COL3, String.class);
	batchTableEnv.registerTableSource("hTable", hbaseTable);

	Table table = batchTableEnv.sqlQuery(
		"SELECT " +
			"  h.rowkey, " +
			"  h.family1.col1, " +
			"  h.family2.col1, " +
			"  h.family2.col2, " +
			"  h.family3.col1, " +
			"  h.family3.col2, " +
			"  h.family3.col3 " +
			"FROM hTable AS h"
	);

	List<Row> results = collectBatchResult(table);
	String expected =
		"1,10,Hello-1,100,1.01,false,Welt-1\n" +
			"2,20,Hello-2,200,2.02,true,Welt-2\n" +
			"3,30,Hello-3,300,3.03,false,Welt-3\n" +
			"4,40,,400,4.04,true,Welt-4\n" +
			"5,50,Hello-5,500,5.05,false,Welt-5\n" +
			"6,60,Hello-6,600,6.06,true,Welt-6\n" +
			"7,70,Hello-7,700,7.07,false,Welt-7\n" +
			"8,80,,800,8.08,true,Welt-8\n";

	TestBaseUtils.compareResultAsText(results, expected);
}