Java BatchTableEnvironment.fromDataSet方法代码示例

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testUserDefinedScalarFunction() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	tableEnv.registerFunction("hashCode", new OldHashCode());
	tableEnv.registerFunction("hashCode", new HashCode());

	DataSource<String> input = env.fromElements("a", "b", "c");

	Table table = tableEnv.fromDataSet(input, "text");

	Table result = table.select("text.hashCode()");

	DataSet<Integer> ds = tableEnv.toDataSet(result, Integer.class);
	List<Integer> results = ds.collect();
	String expected = "97\n98\n99";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testSimpleSelectRenameAll() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> ds = CollectionDataSets.get3TupleDataSet(env);
	Table in = tableEnv.fromDataSet(ds);

	Table result = in
		.select("f0 as a, f1 as b, f2 as c")
		.select("a, b");

	DataSet<Row> resultSet = tableEnv.toDataSet(result, Row.class);
	List<Row> results = resultSet.collect();
	String expected = "1,1\n" + "2,2\n" + "3,2\n" + "4,3\n" + "5,3\n" + "6,3\n" + "7,4\n" +
		"8,4\n" + "9,4\n" + "10,4\n" + "11,5\n" + "12,5\n" + "13,5\n" + "14,5\n" + "15,5\n" +
		"16,6\n" + "17,6\n" + "18,6\n" + "19,6\n" + "20,6\n" + "21,6\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testAsFromTupleByName() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	Table table = tableEnv.fromDataSet(CollectionDataSets.get3TupleDataSet(env), "f2");

	DataSet<Row> ds = tableEnv.toDataSet(table, Row.class);
	List<Row> results = ds.collect();
	String expected = "Hi\n" + "Hello\n" + "Hello world\n" +
		"Hello world, how are you?\n" + "I am fine.\n" + "Luke Skywalker\n" +
		"Comment#1\n" + "Comment#2\n" + "Comment#3\n" + "Comment#4\n" +
		"Comment#5\n" + "Comment#6\n" + "Comment#7\n" +
		"Comment#8\n" + "Comment#9\n" + "Comment#10\n" +
		"Comment#11\n" + "Comment#12\n" + "Comment#13\n" +
		"Comment#14\n" + "Comment#15\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testGroupNoAggregation() throws Exception {

	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> input = CollectionDataSets.get3TupleDataSet(env);
	Table table = tableEnv.fromDataSet(input, "a, b, c");

	Table result = table
		.groupBy("b").select("a.sum as d, b").groupBy("b, d").select("b");

	DataSet<Row> ds = tableEnv.toDataSet(result, Row.class);
	String expected = "1\n" + "2\n" + "3\n" + "4\n" + "5\n" + "6\n";
	List<Row> results = ds.collect();
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testSimpleSelectWithNaming() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> ds = CollectionDataSets.get3TupleDataSet(env);
	Table in = tableEnv.fromDataSet(ds);

	Table result = in
			.select("f0 as a, f1 as b")
			.select("a, b");

	DataSet<Row> resultSet = tableEnv.toDataSet(result, Row.class);
	List<Row> results = resultSet.collect();
	String expected = "1,1\n" + "2,2\n" + "3,2\n" + "4,3\n" + "5,3\n" + "6,3\n" + "7,4\n" +
			"8,4\n" + "9,4\n" + "10,4\n" + "11,5\n" + "12,5\n" + "13,5\n" + "14,5\n" + "15,5\n" +
			"16,6\n" + "17,6\n" + "18,6\n" + "19,6\n" + "20,6\n" + "21,6\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testDistinct() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> input = CollectionDataSets.get3TupleDataSet(env);

	Table table = tableEnv.fromDataSet(input, "a, b, c");

	Table distinct = table.select("b").distinct();

	DataSet<Row> ds = tableEnv.toDataSet(distinct, Row.class);
	List<Row> results = ds.collect();
	String expected = "1\n" + "2\n" + "3\n"+ "4\n"+ "5\n"+ "6\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testDisjunctivePreds() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> input = CollectionDataSets.get3TupleDataSet(env);
	Table table = tableEnv.fromDataSet(input, "a, b, c");

	Table result = table
		.filter("a < 2 || a > 20");

	DataSet<Row> ds = tableEnv.toDataSet(result, Row.class);
	List<Row> results = ds.collect();
	String expected = "1,1,Hi\n" + "21,6,Comment#15\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testGroupedAggregate() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> input = CollectionDataSets.get3TupleDataSet(env);
	Table table = tableEnv.fromDataSet(input, "a, b, c");

	Table result = table
			.groupBy("b").select("b, a.sum");

	DataSet<Row> ds = tableEnv.toDataSet(result, Row.class);
	List<Row> results = ds.collect();
	String expected = "1,1\n" + "2,5\n" + "3,15\n" + "4,34\n" + "5,65\n" + "6,111\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testDistinctAfterAggregate() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple5<Integer, Long, Integer, String, Long>> input = CollectionDataSets.get5TupleDataSet(env);

	Table table = tableEnv.fromDataSet(input, "a, b, c, d, e");

	Table distinct = table.groupBy("a, e").select("e").distinct();

	DataSet<Row> ds = tableEnv.toDataSet(distinct, Row.class);
	List<Row> results = ds.collect();
	String expected = "1\n" + "2\n" + "3\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testJoinWithExtended() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env);

	DataSet<Tuple2<Integer, String>> input1 = env.fromElements(new Tuple2<>(1,"d"));
	DataSet<Tuple2<Integer, String>> input2 = env.fromElements(new Tuple2<>(1,"d"));
	Table table1 = tableEnv.fromDataSet(input1, "a, b");
	Table table2 = tableEnv.fromDataSet(input2, "c, d");
	Table table = table1
		.join(table2)
		.where("b = d")
		.select("a, c");

	String result = tableEnv.explain(table, true).replaceAll("\\r\\n", "\n");
	try (Scanner scanner = new Scanner(new File(testFilePath +
		"../../src/test/scala/resources/testJoin1.out"))){
		String source = scanner.useDelimiter("\\A").next().replaceAll("\\r\\n", "\n");
		assertEquals(source, result);
	}
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test(expected = ValidationException.class)
public void testJoinWithNonMatchingKeyTypes() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env);

	DataSet<Tuple3<Integer, Long, String>> ds1 = CollectionDataSets.getSmall3TupleDataSet(env);
	DataSet<Tuple5<Integer, Long, Integer, String, Long>> ds2 = CollectionDataSets.get5TupleDataSet(env);

	Table in1 = tableEnv.fromDataSet(ds1, "a, b, c");
	Table in2 = tableEnv.fromDataSet(ds2, "d, e, f, g, h");

	Table result = in1.join(in2)
		// Must fail. Types of join fields are not compatible (Integer and String)
		.where("a === g").select("c, g");

	tableEnv.toDataSet(result, Row.class).collect();
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testUnionWithExtended() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env);

	DataSet<Tuple2<Integer, String>> input1 = env.fromElements(new Tuple2<>(1,"d"));
	DataSet<Tuple2<Integer, String>> input2 = env.fromElements(new Tuple2<>(1,"d"));
	Table table1 = tableEnv.fromDataSet(input1, "count, word");
	Table table2 = tableEnv.fromDataSet(input2, "count, word");
	Table table = table1.unionAll(table2);

	String result = tableEnv.explain(table, true).replaceAll("\\r\\n", "\n");
	try (Scanner scanner = new Scanner(new File(testFilePath +
		"../../src/test/scala/resources/testUnion1.out"))){
		String source = scanner.useDelimiter("\\A").next().replaceAll("\\r\\n", "\n");
		assertEquals(source, result);
	}
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testSelectFromTable() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	DataSet<Tuple3<Integer, Long, String>> ds = CollectionDataSets.get3TupleDataSet(env);
	Table in = tableEnv.fromDataSet(ds, "a,b,c");
	tableEnv.registerTable("T", in);

	String sqlQuery = "SELECT a, c FROM T";
	Table result = tableEnv.sql(sqlQuery);

	DataSet<Row> resultSet = tableEnv.toDataSet(result, Row.class);
	List<Row> results = resultSet.collect();
	String expected = "1,Hi\n" + "2,Hello\n" + "3,Hello world\n" +
		"4,Hello world, how are you?\n" + "5,I am fine.\n" + "6,Luke Skywalker\n" +
		"7,Comment#1\n" + "8,Comment#2\n" + "9,Comment#3\n" + "10,Comment#4\n" +
		"11,Comment#5\n" + "12,Comment#6\n" + "13,Comment#7\n" +
		"14,Comment#8\n" + "15,Comment#9\n" + "16,Comment#10\n" +
		"17,Comment#11\n" + "18,Comment#12\n" + "19,Comment#13\n" +
		"20,Comment#14\n" + "21,Comment#15\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test
public void testJoin() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env);

	DataSet<Tuple3<Integer, Long, String>> ds1 = CollectionDataSets.getSmall3TupleDataSet(env);
	DataSet<Tuple5<Integer, Long, Integer, String, Long>> ds2 = CollectionDataSets.get5TupleDataSet(env);

	Table in1 = tableEnv.fromDataSet(ds1, "a, b, c");
	Table in2 = tableEnv.fromDataSet(ds2, "d, e, f, g, h");

	Table result = in1.join(in2).where("b === e").select("c, g");

	DataSet<Row> ds = tableEnv.toDataSet(result, Row.class);
	List<Row> results = ds.collect();
	String expected = "Hi,Hallo\n" + "Hello,Hallo Welt\n" + "Hello world,Hallo Welt\n";
	compareResultAsText(results, expected);
}
 

import org.apache.flink.table.api.java.BatchTableEnvironment; //导入方法依赖的package包/类
@Test(expected = TableException.class)
public void testGenericRowWithAlias() throws Exception {
	ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
	BatchTableEnvironment tableEnv = TableEnvironment.getTableEnvironment(env, config());

	// use null value the enforce GenericType
	DataSet<Row> dataSet = env.fromElements(Row.of((Integer) null));
	assertTrue(dataSet.getType() instanceof GenericTypeInfo);
	assertTrue(dataSet.getType().getTypeClass().equals(Row.class));

	// Must fail. Cannot import DataSet<Row> with GenericTypeInfo.
	tableEnv.fromDataSet(dataSet, "nullField");
}