Examples with ETLBatchConfig co.cask.cdap.etl.proto.v2.ETLBatchConfig used on opensource projects

Example 76 with ETLBatchConfig

use of co.cask.cdap.etl.proto.v2.ETLBatchConfig in project cdap by caskdata.

the class DataPipelineTest method testConditionsOnBranches.

@Test
public void testConditionsOnBranches() throws Exception {
    /*
     *                            |-- true --> sink1
     *          |--> condition1 --|
     * source --|                 |-- false --> sink2
     *          |
     *          |                              |-- true --> sink3
     *          |-- transform --> condition2 --|
     *                                         |-- false --> sink4
     */
    Schema schema = Schema.recordOf("testRecord", Schema.Field.of("name", Schema.of(Schema.Type.STRING)));
    String sourceName = "branchConditionsSource";
    String sink1Name = "branchConditionsSink1";
    String sink2Name = "branchConditionsSink2";
    String sink3Name = "branchConditionsSink3";
    String sink4Name = "branchConditionsSink4";
    ETLBatchConfig etlConfig = ETLBatchConfig.builder().addStage(new ETLStage("source", MockSource.getPlugin(sourceName, schema))).addStage(new ETLStage("condition1", MockCondition.getPlugin("condition1"))).addStage(new ETLStage("transform", IdentityTransform.getPlugin())).addStage(new ETLStage("condition2", MockCondition.getPlugin("condition2"))).addStage(new ETLStage("sink1", MockSink.getPlugin(sink1Name))).addStage(new ETLStage("sink2", MockSink.getPlugin(sink2Name))).addStage(new ETLStage("sink3", MockSink.getPlugin(sink3Name))).addStage(new ETLStage("sink4", MockSink.getPlugin(sink4Name))).addConnection("source", "condition1").addConnection("source", "transform").addConnection("condition1", "sink1", true).addConnection("condition1", "sink2", false).addConnection("transform", "condition2").addConnection("condition2", "sink3", true).addConnection("condition2", "sink4", false).build();
    AppRequest<ETLBatchConfig> appRequest = new AppRequest<>(APP_ARTIFACT_RANGE, etlConfig);
    ApplicationId appId = NamespaceId.DEFAULT.app("branchConditions");
    ApplicationManager appManager = deployApplication(appId, appRequest);
    List<StructuredRecord> records = Collections.singletonList(StructuredRecord.builder(schema).set("name", "samuel").build());
    DataSetManager<Table> inputManager = getDataset(sourceName);
    MockSource.writeInput(inputManager, records);
    WorkflowManager workflowManager = appManager.getWorkflowManager(SmartWorkflow.NAME);
    workflowManager.start(ImmutableMap.of("condition1.branch.to.execute", "true", "condition2.branch.to.execute", "false"));
    workflowManager.waitForRun(ProgramRunStatus.COMPLETED, 5, TimeUnit.MINUTES);
    DataSetManager<Table> sink1Manager = getDataset(sink1Name);
    DataSetManager<Table> sink2Manager = getDataset(sink2Name);
    DataSetManager<Table> sink3Manager = getDataset(sink3Name);
    DataSetManager<Table> sink4Manager = getDataset(sink4Name);
    Assert.assertEquals(records, MockSink.readOutput(sink1Manager));
    Assert.assertTrue(MockSink.readOutput(sink2Manager).isEmpty());
    Assert.assertTrue(MockSink.readOutput(sink3Manager).isEmpty());
    Assert.assertEquals(records, MockSink.readOutput(sink4Manager));
}

Example 77 with ETLBatchConfig

use of co.cask.cdap.etl.proto.v2.ETLBatchConfig in project cdap by caskdata.

the class DataPipelineTest method testSimpleConditionWithMultipleOutputActions.

@Test
public void testSimpleConditionWithMultipleOutputActions() throws Exception {
    Schema schema = Schema.recordOf("testRecord", Schema.Field.of("name", Schema.of(Schema.Type.STRING)));
    /*
     *
     * condition --Action--> file ---> trueSink
     *       |
     *       |--->Action--->file----> falseSink
     *
     */
    String appName = "SimpleConditionWithMultipleOutputActions";
    String trueSource = "true" + appName + "Source";
    String falseSource = "false" + appName + "Source";
    String trueSink = "true" + appName + "Sink";
    String falseSink = "false" + appName + "Sink";
    String actionTable = "actionTable" + appName;
    ETLBatchConfig etlConfig = ETLBatchConfig.builder().addStage(new ETLStage("trueSource", MockSource.getPlugin(trueSource, schema))).addStage(new ETLStage("falseSource", MockSource.getPlugin(falseSource, schema))).addStage(new ETLStage("trueSink", MockSink.getPlugin(trueSink))).addStage(new ETLStage("falseSink", MockSink.getPlugin(falseSink))).addStage(new ETLStage("condition", MockCondition.getPlugin("condition"))).addStage(new ETLStage("action1", MockAction.getPlugin(actionTable, "row1", "key1", "val1"))).addStage(new ETLStage("action2", MockAction.getPlugin(actionTable, "row2", "key2", "val2"))).addConnection("condition", "action1", true).addConnection("action1", "trueSource").addConnection("trueSource", "trueSink").addConnection("condition", "action2", false).addConnection("action2", "falseSource").addConnection("falseSource", "falseSink").build();
    AppRequest<ETLBatchConfig> appRequest = new AppRequest<>(APP_ARTIFACT_RANGE, etlConfig);
    ApplicationId appId = NamespaceId.DEFAULT.app(appName);
    ApplicationManager appManager = deployApplication(appId, appRequest);
    StructuredRecord recordSamuel = StructuredRecord.builder(schema).set("name", "samuel").build();
    StructuredRecord recordBob = StructuredRecord.builder(schema).set("name", "bob").build();
    for (String branch : Arrays.asList("true", "false")) {
        String source = branch.equals("true") ? trueSource : falseSource;
        String sink = branch.equals("true") ? trueSink : falseSink;
        // write records to source
        DataSetManager<Table> inputManager = getDataset(NamespaceId.DEFAULT.dataset(source));
        MockSource.writeInput(inputManager, ImmutableList.of(recordSamuel, recordBob));
        WorkflowManager workflowManager = appManager.getWorkflowManager(SmartWorkflow.NAME);
        workflowManager.start(ImmutableMap.of("condition.branch.to.execute", branch));
        if (branch.equals("true")) {
            workflowManager.waitForRun(ProgramRunStatus.COMPLETED, 5, TimeUnit.MINUTES);
        } else {
            workflowManager.waitForRuns(ProgramRunStatus.COMPLETED, 2, 5, TimeUnit.MINUTES);
        }
        // check sink
        DataSetManager<Table> sinkManager = getDataset(sink);
        Set<StructuredRecord> expected = ImmutableSet.of(recordSamuel, recordBob);
        Set<StructuredRecord> actual = Sets.newHashSet(MockSink.readOutput(sinkManager));
        Assert.assertEquals(expected, actual);
        validateMetric(2, appId, branch + "Source.records.out");
        validateMetric(2, appId, branch + "Sink.records.in");
        // check Action1 and Action2 is executed correctly
        DataSetManager<Table> actionTableDS = getDataset(actionTable);
        if (branch.equals("true")) {
            Assert.assertEquals("val1", MockAction.readOutput(actionTableDS, "row1", "key1"));
        } else {
            Assert.assertEquals("val2", MockAction.readOutput(actionTableDS, "row2", "key2"));
        }
    }
}

Example 78 with ETLBatchConfig

use of co.cask.cdap.etl.proto.v2.ETLBatchConfig in project cdap by caskdata.

the class AutoJoinerTest method testTripleAutoJoin.

private void testTripleAutoJoin(List<String> required, List<String> broadcast, Set<StructuredRecord> expected, Engine engine, List<String> tablesInOrderToJoin) throws Exception {
    /*
         users ------|
                     |
         purchases --|--> join --> sink
                     |
         interests --|

         joinOn: users.region = purchases.region = interests.region and
                 users.user_id = purchases.user_id = interests.user_id
     */
    String userInput = UUID.randomUUID().toString();
    String purchaseInput = UUID.randomUUID().toString();
    String interestInput = UUID.randomUUID().toString();
    String output = UUID.randomUUID().toString();
    ETLBatchConfig config = ETLBatchConfig.builder().addStage(new ETLStage("users", MockSource.getPlugin(userInput, USER_SCHEMA))).addStage(new ETLStage("purchases", MockSource.getPlugin(purchaseInput, PURCHASE_SCHEMA))).addStage(new ETLStage("interests", MockSource.getPlugin(interestInput, INTEREST_SCHEMA))).addStage(new ETLStage("join", MockAutoJoiner.getPlugin(tablesInOrderToJoin, Arrays.asList("region", "user_id"), required, broadcast, Collections.emptyList(), true))).addStage(new ETLStage("sink", MockSink.getPlugin(output))).addConnection("users", "join").addConnection("purchases", "join").addConnection("interests", "join").addConnection("join", "sink").setEngine(engine).build();
    AppRequest<ETLBatchConfig> appRequest = new AppRequest<>(APP_ARTIFACT, config);
    ApplicationId appId = NamespaceId.DEFAULT.app(UUID.randomUUID().toString());
    ApplicationManager appManager = deployApplication(appId, appRequest);
    // write input data
    List<StructuredRecord> userData = Arrays.asList(USER_ALICE, USER_ALYCE, USER_BOB);
    DataSetManager<Table> inputManager = getDataset(userInput);
    MockSource.writeInput(inputManager, userData);
    List<StructuredRecord> purchaseData = new ArrayList<>();
    purchaseData.add(StructuredRecord.builder(PURCHASE_SCHEMA).set("region", "us").set("user_id", 0).set("purchase_id", 123).build());
    purchaseData.add(StructuredRecord.builder(PURCHASE_SCHEMA).set("region", "us").set("user_id", 2).set("purchase_id", 456).build());
    inputManager = getDataset(purchaseInput);
    MockSource.writeInput(inputManager, purchaseData);
    List<StructuredRecord> interestData = new ArrayList<>();
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 0).set("interest", "food").build());
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 0).set("interest", "sports").build());
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 1).set("interest", "gardening").build());
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 2).set("interest", "gaming").build());
    inputManager = getDataset(interestInput);
    MockSource.writeInput(inputManager, interestData);
    WorkflowManager workflowManager = appManager.getWorkflowManager(SmartWorkflow.NAME);
    workflowManager.startAndWaitForGoodRun(ProgramRunStatus.COMPLETED, 5, TimeUnit.MINUTES);
    DataSetManager<Table> outputManager = getDataset(output);
    List<StructuredRecord> outputRecords = MockSink.readOutput(outputManager);
    Set<StructuredRecord> actual = new HashSet<>();
    Schema expectedSchema = expected.iterator().hasNext() ? expected.iterator().next().getSchema() : null;
    if (expectedSchema == null || expected.iterator().next().getSchema() == outputRecords.get(0).getSchema()) {
        actual = new HashSet<>(outputRecords);
    } else {
        // reorder the output columns of the join result (actual) to match the column order of expected
        for (StructuredRecord sr : outputRecords) {
            actual.add(StructuredRecord.builder(expectedSchema).set("purchases_region", sr.get("purchases_region")).set("purchases_purchase_id", sr.get("purchases_purchase_id")).set("purchases_user_id", sr.get("purchases_user_id")).set("users_region", sr.get("users_region")).set("users_user_id", sr.get("users_user_id")).set("users_name", sr.get("users_name")).set("interests_region", sr.get("interests_region")).set("interests_user_id", sr.get("interests_user_id")).set("interests_interest", sr.get("interests_interest")).build());
        }
    }
    Assert.assertEquals(expected, actual);
    validateMetric(9, appId, "join.records.in");
    validateMetric(expected.size(), appId, "join.records.out");
}

Example 79 with ETLBatchConfig

use of co.cask.cdap.etl.proto.v2.ETLBatchConfig in project cdap by caskdata.

the class AutoJoinerTest method testQuadAutoJoin.

private void testQuadAutoJoin(List<String> required, List<String> broadcast, Set<StructuredRecord> expected, Engine engine, List<String> tablesInOrderToJoin) throws Exception {
    /*
         users ------|
                     |
         purchases --|--> join --> sink
                     |
         interests --|
                     |
         age --------|

         joinOn: users.region = purchases.region = interests.region = age.region and
                 users.user_id = purchases.user_id = interests.user_id = age.user_id
     */
    String userInput = UUID.randomUUID().toString();
    String purchaseInput = UUID.randomUUID().toString();
    String interestInput = UUID.randomUUID().toString();
    String ageInput = UUID.randomUUID().toString();
    String output = UUID.randomUUID().toString();
    ETLBatchConfig config = ETLBatchConfig.builder().addStage(new ETLStage("users", MockSource.getPlugin(userInput, USER_SCHEMA))).addStage(new ETLStage("purchases", MockSource.getPlugin(purchaseInput, PURCHASE_SCHEMA))).addStage(new ETLStage("interests", MockSource.getPlugin(interestInput, INTEREST_SCHEMA))).addStage(new ETLStage("ages", MockSource.getPlugin(ageInput, AGE_SCHEMA))).addStage(new ETLStage("join", MockAutoJoiner.getPlugin(tablesInOrderToJoin, Arrays.asList("region", "user_id"), required, broadcast, Collections.emptyList(), true))).addStage(new ETLStage("sink", MockSink.getPlugin(output))).addConnection("users", "join").addConnection("purchases", "join").addConnection("interests", "join").addConnection("ages", "join").addConnection("join", "sink").setEngine(engine).build();
    AppRequest<ETLBatchConfig> appRequest = new AppRequest<>(APP_ARTIFACT, config);
    ApplicationId appId = NamespaceId.DEFAULT.app(UUID.randomUUID().toString());
    ApplicationManager appManager = deployApplication(appId, appRequest);
    // write input data
    List<StructuredRecord> userData = Arrays.asList(USER_ALICE, USER_ALYCE, USER_BOB);
    DataSetManager<Table> inputManager = getDataset(userInput);
    MockSource.writeInput(inputManager, userData);
    List<StructuredRecord> purchaseData = new ArrayList<>();
    purchaseData.add(StructuredRecord.builder(PURCHASE_SCHEMA).set("region", "us").set("user_id", 0).set("purchase_id", 123).build());
    purchaseData.add(StructuredRecord.builder(PURCHASE_SCHEMA).set("region", "us").set("user_id", 2).set("purchase_id", 456).build());
    inputManager = getDataset(purchaseInput);
    MockSource.writeInput(inputManager, purchaseData);
    List<StructuredRecord> interestData = new ArrayList<>();
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 0).set("interest", "food").build());
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 0).set("interest", "sports").build());
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 1).set("interest", "gardening").build());
    interestData.add(StructuredRecord.builder(INTEREST_SCHEMA).set("region", "us").set("user_id", 2).set("interest", "gaming").build());
    inputManager = getDataset(interestInput);
    MockSource.writeInput(inputManager, interestData);
    List<StructuredRecord> ageData = new ArrayList<>();
    ageData.add(StructuredRecord.builder(AGE_SCHEMA).set("region", "us").set("user_id", 10).set("age", 20).build());
    ageData.add(StructuredRecord.builder(AGE_SCHEMA).set("region", "us").set("user_id", 1).set("age", 30).build());
    inputManager = getDataset(ageInput);
    MockSource.writeInput(inputManager, ageData);
    WorkflowManager workflowManager = appManager.getWorkflowManager(SmartWorkflow.NAME);
    workflowManager.startAndWaitForGoodRun(ProgramRunStatus.COMPLETED, 5, TimeUnit.MINUTES);
    DataSetManager<Table> outputManager = getDataset(output);
    List<StructuredRecord> outputRecords = MockSink.readOutput(outputManager);
    Set<StructuredRecord> actual = new HashSet<>();
    Schema expectedSchema = expected.iterator().hasNext() ? expected.iterator().next().getSchema() : null;
    if (expectedSchema == null || expected.iterator().next().getSchema() == outputRecords.get(0).getSchema()) {
        actual = new HashSet<>(outputRecords);
    } else {
        // reorder the output columns of the join result (actual) to match the column order of expected
        for (StructuredRecord sr : outputRecords) {
            actual.add(StructuredRecord.builder(expectedSchema).set("ages_region", sr.get("ages_region")).set("ages_age", sr.get("ages_age")).set("ages_user_id", sr.get("ages_user_id")).set("purchases_region", sr.get("purchases_region")).set("purchases_purchase_id", sr.get("purchases_purchase_id")).set("purchases_user_id", sr.get("purchases_user_id")).set("users_region", sr.get("users_region")).set("users_user_id", sr.get("users_user_id")).set("users_name", sr.get("users_name")).set("interests_region", sr.get("interests_region")).set("interests_user_id", sr.get("interests_user_id")).set("interests_interest", sr.get("interests_interest")).build());
        }
    }
    Assert.assertEquals(expected, actual);
    validateMetric(11, appId, "join.records.in");
    validateMetric(expected.size(), appId, "join.records.out");
}

Example 80 with ETLBatchConfig

use of co.cask.cdap.etl.proto.v2.ETLBatchConfig in project cdap by caskdata.

the class AutoJoinerTest method testCaseSensitivity.

@Test
public void testCaseSensitivity() throws Exception {
    Schema weird1 = Schema.recordOf("weird1", Schema.Field.of("id", Schema.of(Schema.Type.INT)), Schema.Field.of("ID", Schema.of(Schema.Type.LONG)), Schema.Field.of("Id", Schema.of(Schema.Type.INT)), Schema.Field.of("name", Schema.of(Schema.Type.STRING)));
    Schema weird2 = Schema.recordOf("weird2", Schema.Field.of("id", Schema.of(Schema.Type.INT)), Schema.Field.of("ID", Schema.of(Schema.Type.LONG)), Schema.Field.of("val", Schema.of(Schema.Type.STRING)));
    String input1 = UUID.randomUUID().toString();
    String input2 = UUID.randomUUID().toString();
    String output = UUID.randomUUID().toString();
    ETLBatchConfig config = ETLBatchConfig.builder().addStage(new ETLStage("i1", MockSource.getPlugin(input1, weird1))).addStage(new ETLStage("i2", MockSource.getPlugin(input2, weird2))).addStage(new ETLStage("join", MockAutoJoiner.getPlugin(Arrays.asList("i1", "i2"), Arrays.asList("id", "ID"), Arrays.asList("i1", "i2"), Collections.emptyList(), Collections.emptyList(), true))).addStage(new ETLStage("sink", MockSink.getPlugin(output))).addConnection("i1", "join").addConnection("i2", "join").addConnection("join", "sink").setEngine(Engine.SPARK).build();
    AppRequest<ETLBatchConfig> appRequest = new AppRequest<>(APP_ARTIFACT, config);
    ApplicationId appId = NamespaceId.DEFAULT.app(UUID.randomUUID().toString());
    ApplicationManager appManager = deployApplication(appId, appRequest);
    // write input data
    List<StructuredRecord> input1Data = new ArrayList<>();
    input1Data.add(StructuredRecord.builder(weird1).set("id", 0).set("ID", 99L).set("Id", 0).set("name", "zero").build());
    input1Data.add(StructuredRecord.builder(weird1).set("id", 1).set("ID", 0L).set("Id", 0).set("name", "one").build());
    DataSetManager<Table> inputManager = getDataset(input1);
    MockSource.writeInput(inputManager, input1Data);
    List<StructuredRecord> input2Data = new ArrayList<>();
    input2Data.add(StructuredRecord.builder(weird2).set("id", 0).set("ID", 99L).set("val", "0").build());
    input2Data.add(StructuredRecord.builder(weird2).set("id", 1).set("ID", 99L).set("val", "1").build());
    input2Data.add(StructuredRecord.builder(weird2).set("id", 0).set("ID", 0L).set("val", "2").build());
    inputManager = getDataset(input2);
    MockSource.writeInput(inputManager, input2Data);
    Schema expectedSchema = Schema.recordOf("i1.i2", Schema.Field.of("i1_id", Schema.of(Schema.Type.INT)), Schema.Field.of("i1_ID", Schema.of(Schema.Type.LONG)), Schema.Field.of("i1_Id", Schema.of(Schema.Type.INT)), Schema.Field.of("i1_name", Schema.of(Schema.Type.STRING)), Schema.Field.of("i2_id", Schema.of(Schema.Type.INT)), Schema.Field.of("i2_ID", Schema.of(Schema.Type.LONG)), Schema.Field.of("i2_val", Schema.of(Schema.Type.STRING)));
    StructuredRecord expected = StructuredRecord.builder(expectedSchema).set("i1_id", 0).set("i1_ID", 99L).set("i1_Id", 0).set("i1_name", "zero").set("i2_id", 0).set("i2_ID", 99L).set("i2_val", "0").build();
    WorkflowManager workflowManager = appManager.getWorkflowManager(SmartWorkflow.NAME);
    Map<String, String> args = Collections.singletonMap(MockAutoJoiner.PARTITIONS_ARGUMENT, "1");
    workflowManager.startAndWaitForGoodRun(args, ProgramRunStatus.COMPLETED, 5, TimeUnit.MINUTES);
    DataSetManager<Table> outputManager = getDataset(output);
    List<StructuredRecord> actual = MockSink.readOutput(outputManager);
    Assert.assertEquals(Collections.singletonList(expected), actual);
}