Examples with RecordSet org.talend.components.test.RecordSet used on opensource projects

Example 1 with RecordSet

use of org.talend.components.test.RecordSet in project components by Talend.

the class SimpleFileIORoundTripRuntimeTest method testParquet.

/**
 * Basic Parquet test.
 */
@Test
public void testParquet() throws IOException {
    // The file that we will be creating.
    RecordSet rs = getSimpleTestData(0);
    String fileSpec = mini.getLocalFsNewFolder() + "output/";
    // Configure the components.
    SimpleFileIOOutputProperties outputProps = createOutputComponentProperties();
    outputProps.getDatasetProperties().format.setValue(SimpleFileIOFormat.PARQUET);
    outputProps.getDatasetProperties().path.setValue(fileSpec);
    SimpleFileIOInputProperties inputProps = createInputComponentProperties();
    inputProps.setDatasetProperties(outputProps.getDatasetProperties());
    List<IndexedRecord> actual = runRoundTripPipelines(beam, rs.getAllData(), outputProps, inputProps);
    // Generate the set of expected records. By default, CSV turns all columns into String and loses the original
    // column name.
    List<IndexedRecord> expected = rs.getAllData();
    assertThat(actual, containsInAnyOrder(expected.toArray()));
// Verify that the file on the filesystem was correctly written.
// TODO(rskraba): verify independently from
// mini.assertReadFile(
// mini.getLocalFs(),
// fileSpec,
// rewriteRecordsAsCsvLines(expected, inputProps.getDatasetProperties().recordDelimiter.getValue(),
// inputProps.getDatasetProperties().fieldDelimiter.getValue()));
}

Example 2 with RecordSet

use of org.talend.components.test.RecordSet in project components by Talend.

the class SimpleFileIOInputRuntimeTest method testBasicAvro.

/**
 * Basic unit test using all default values (except for the path) on an in-memory DFS cluster.
 */
@Test
public void testBasicAvro() throws IOException, URISyntaxException {
    RecordSet rs = getSimpleTestData(0);
    writeRandomAvroFile(mini.getFs(), "/user/test/input.avro", rs);
    String fileSpec = mini.getFs().getUri().resolve("/user/test/input.avro").toString();
    // Configure the component.
    SimpleFileIOInputProperties inputProps = createInputComponentProperties();
    inputProps.getDatasetProperties().format.setValue(SimpleFileIOFormat.AVRO);
    inputProps.getDatasetProperties().path.setValue(fileSpec);
    // Create the runtime.
    SimpleFileIOInputRuntime runtime = new SimpleFileIOInputRuntime();
    runtime.initialize(null, inputProps);
    // Use the runtime in a direct pipeline to test.
    final Pipeline p = beam.createPipeline();
    PCollection<IndexedRecord> readLines = p.apply(runtime);
    // Check the expected values.
    PAssert.that(readLines).containsInAnyOrder(rs.getAllData());
    // And run the test.
    p.run().waitUntilFinish();
}

Example 3 with RecordSet

use of org.talend.components.test.RecordSet in project components by Talend.

the class SimpleFileIOInputRuntimeTest method testBasicParquet.

/**
 * Basic unit test using all default values (except for the path) on an in-memory DFS cluster.
 */
@Ignore("To implement.")
@Test
public void testBasicParquet() throws IOException, URISyntaxException {
    RecordSet rs = getSimpleTestData(0);
    writeRandomAvroFile(mini.getFs(), "/user/test/input.avro", rs);
    String fileSpec = mini.getFs().getUri().resolve("/user/test/input.avro").toString();
    // Configure the component.
    SimpleFileIOInputProperties inputProps = createInputComponentProperties();
    inputProps.getDatasetProperties().format.setValue(SimpleFileIOFormat.PARQUET);
    inputProps.getDatasetProperties().path.setValue(fileSpec);
    // Create the runtime.
    SimpleFileIOInputRuntime runtime = new SimpleFileIOInputRuntime();
    runtime.initialize(null, inputProps);
    // Use the runtime in a direct pipeline to test.
    final Pipeline p = beam.createPipeline();
    PCollection<IndexedRecord> readLines = p.apply(runtime);
    // Check the expected values.
    PAssert.that(readLines).containsInAnyOrder(rs.getAllData());
    // And run the test.
    p.run().waitUntilFinish();
}

Example 4 with RecordSet

use of org.talend.components.test.RecordSet in project components by Talend.

the class SimpleFileIORoundTripRuntimeTest method testCsvWithDelimiters.

/**
 * Test CSV with custom delimiters.
 */
@Test
public void testCsvWithDelimiters() throws IOException {
    // The file that we will be creating.
    RecordSet rs = getSimpleTestData(0);
    String fileSpec = mini.getLocalFsNewFolder() + "output/";
    // Configure the components.
    SimpleFileIOOutputProperties outputProps = createOutputComponentProperties();
    outputProps.getDatasetProperties().format.setValue(SimpleFileIOFormat.CSV);
    outputProps.getDatasetProperties().path.setValue(fileSpec);
    outputProps.getDatasetProperties().recordDelimiter.setValue(RecordDelimiterType.OTHER);
    outputProps.getDatasetProperties().specificRecordDelimiter.setValue("---");
    outputProps.getDatasetProperties().fieldDelimiter.setValue(FieldDelimiterType.OTHER);
    outputProps.getDatasetProperties().specificFieldDelimiter.setValue("|");
    SimpleFileIOInputProperties inputProps = createInputComponentProperties();
    inputProps.setDatasetProperties(outputProps.getDatasetProperties());
    List<IndexedRecord> actual = runRoundTripPipelines(beam, rs.getAllData(), outputProps, inputProps);
    // Generate the set of expected records. By default, CSV turns all columns into String and loses the original
    // column name.
    List<IndexedRecord> expected = rewriteRecordsWithCsvSchema(rs.getAllData());
    assertThat(expected, containsInAnyOrder(actual.toArray()));
    // Verify that the file on the filesystem was correctly written.
    mini.assertReadFile("---", mini.getLocalFs(), fileSpec, rewriteRecordsAsCsvLines(expected, inputProps.getDatasetProperties().getRecordDelimiter(), inputProps.getDatasetProperties().getFieldDelimiter()));
}

Example 5 with RecordSet

use of org.talend.components.test.RecordSet in project components by Talend.

the class SimpleFileIORoundTripRuntimeTest method testAvro.

/**
 * Basic Avro test.
 */
@Test
public void testAvro() throws IOException {
    // The file that we will be creating.
    RecordSet rs = getSimpleTestData(0);
    String fileSpec = mini.getLocalFsNewFolder() + "output/";
    // Configure the components.
    SimpleFileIOOutputProperties outputProps = createOutputComponentProperties();
    outputProps.getDatasetProperties().format.setValue(SimpleFileIOFormat.AVRO);
    outputProps.getDatasetProperties().path.setValue(fileSpec);
    SimpleFileIOInputProperties inputProps = createInputComponentProperties();
    inputProps.setDatasetProperties(outputProps.getDatasetProperties());
    List<IndexedRecord> actual = runRoundTripPipelines(beam, rs.getAllData(), outputProps, inputProps);
    // Generate the set of expected records. By default, CSV turns all columns into String and loses the original
    // column name.
    List<IndexedRecord> expected = rs.getAllData();
    assertThat(actual, containsInAnyOrder(expected.toArray()));
// Verify that the file on the filesystem was correctly written.
// TODO(rskraba): verify independently
// mini.assertReadFile(
// mini.getLocalFs(),
// fileSpec,
// rewriteRecordsAsCsvLines(expected, inputProps.getDatasetProperties().recordDelimiter.getValue(),
// inputProps.getDatasetProperties().fieldDelimiter.getValue()));
}