Examples with CsvInputFormat org.apache.druid.data.input.impl.CsvInputFormat used on opensource projects

Example 11 with CsvInputFormat

use of org.apache.druid.data.input.impl.CsvInputFormat in project druid by druid-io.

the class OssInputSourceTest method testCompressedReader.

@Test
public void testCompressedReader() throws IOException {
    EasyMock.reset(OSSCLIENT);
    expectListObjects(PREFIXES.get(0), ImmutableList.of(EXPECTED_COMPRESSED_URIS.get(0)), CONTENT);
    expectListObjects(EXPECTED_COMPRESSED_URIS.get(1), ImmutableList.of(EXPECTED_COMPRESSED_URIS.get(1)), CONTENT);
    expectGetObjectCompressed(EXPECTED_COMPRESSED_URIS.get(0));
    expectGetObjectCompressed(EXPECTED_COMPRESSED_URIS.get(1));
    EasyMock.replay(OSSCLIENT);
    OssInputSource inputSource = new OssInputSource(OSSCLIENT, INPUT_DATA_CONFIG, null, ImmutableList.of(PREFIXES.get(0), EXPECTED_COMPRESSED_URIS.get(1)), null, null);
    InputRowSchema someSchema = new InputRowSchema(new TimestampSpec("time", "auto", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("dim1", "dim2"))), ColumnsFilter.all());
    InputSourceReader reader = inputSource.reader(someSchema, new CsvInputFormat(ImmutableList.of("time", "dim1", "dim2"), "|", false, null, 0), temporaryFolder.newFolder());
    CloseableIterator<InputRow> iterator = reader.read();
    while (iterator.hasNext()) {
        InputRow nextRow = iterator.next();
        Assert.assertEquals(NOW, nextRow.getTimestamp());
        Assert.assertEquals("hello", nextRow.getDimension("dim1").get(0));
        Assert.assertEquals("world", nextRow.getDimension("dim2").get(0));
    }
    EasyMock.verify(OSSCLIENT);
}

Example 12 with CsvInputFormat

use of org.apache.druid.data.input.impl.CsvInputFormat in project druid by druid-io.

the class GoogleCloudStorageInputSourceTest method testReader.

@Test
public void testReader() throws IOException {
    EasyMock.reset(STORAGE);
    EasyMock.reset(INPUT_DATA_CONFIG);
    addExpectedPrefixObjects(PREFIXES.get(0), ImmutableList.of(EXPECTED_URIS.get(0)));
    addExpectedGetObjectMock(EXPECTED_URIS.get(0));
    addExpectedPrefixObjects(PREFIXES.get(1), ImmutableList.of(EXPECTED_URIS.get(1)));
    addExpectedGetObjectMock(EXPECTED_URIS.get(1));
    EasyMock.expect(INPUT_DATA_CONFIG.getMaxListingLength()).andReturn(MAX_LISTING_LENGTH);
    EasyMock.replay(STORAGE);
    EasyMock.replay(INPUT_DATA_CONFIG);
    GoogleCloudStorageInputSource inputSource = new GoogleCloudStorageInputSource(STORAGE, INPUT_DATA_CONFIG, null, PREFIXES, null);
    InputRowSchema someSchema = new InputRowSchema(new TimestampSpec("time", "auto", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("dim1", "dim2"))), ColumnsFilter.all());
    InputSourceReader reader = inputSource.reader(someSchema, new CsvInputFormat(ImmutableList.of("time", "dim1", "dim2"), "|", false, null, 0), null);
    CloseableIterator<InputRow> iterator = reader.read();
    while (iterator.hasNext()) {
        InputRow nextRow = iterator.next();
        Assert.assertEquals(NOW, nextRow.getTimestamp());
        Assert.assertEquals("hello", nextRow.getDimension("dim1").get(0));
        Assert.assertEquals("world", nextRow.getDimension("dim2").get(0));
    }
}

Example 13 with CsvInputFormat

use of org.apache.druid.data.input.impl.CsvInputFormat in project druid by druid-io.

the class S3InputSourceTest method testCompressedReader.

@Test
public void testCompressedReader() throws IOException {
    EasyMock.reset(S3_CLIENT);
    expectListObjects(PREFIXES.get(0), ImmutableList.of(EXPECTED_COMPRESSED_URIS.get(0)), CONTENT);
    expectListObjects(EXPECTED_COMPRESSED_URIS.get(1), ImmutableList.of(EXPECTED_COMPRESSED_URIS.get(1)), CONTENT);
    expectGetObjectCompressed(EXPECTED_COMPRESSED_URIS.get(0));
    expectGetObjectCompressed(EXPECTED_COMPRESSED_URIS.get(1));
    EasyMock.replay(S3_CLIENT);
    S3InputSource inputSource = new S3InputSource(SERVICE, SERVER_SIDE_ENCRYPTING_AMAZON_S3_BUILDER, INPUT_DATA_CONFIG, null, ImmutableList.of(PREFIXES.get(0), EXPECTED_COMPRESSED_URIS.get(1)), null, null);
    InputRowSchema someSchema = new InputRowSchema(new TimestampSpec("time", "auto", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("dim1", "dim2"))), ColumnsFilter.all());
    InputSourceReader reader = inputSource.reader(someSchema, new CsvInputFormat(ImmutableList.of("time", "dim1", "dim2"), "|", false, null, 0), temporaryFolder.newFolder());
    CloseableIterator<InputRow> iterator = reader.read();
    while (iterator.hasNext()) {
        InputRow nextRow = iterator.next();
        Assert.assertEquals(NOW, nextRow.getTimestamp());
        Assert.assertEquals("hello", nextRow.getDimension("dim1").get(0));
        Assert.assertEquals("world", nextRow.getDimension("dim2").get(0));
    }
    EasyMock.verify(S3_CLIENT);
}

Example 14 with CsvInputFormat

use of org.apache.druid.data.input.impl.CsvInputFormat in project druid by druid-io.

the class IndexTaskTest method testCsvWithHeaderOfEmptyTimestamp.

@Test
public void testCsvWithHeaderOfEmptyTimestamp() throws Exception {
    final File tmpDir = temporaryFolder.newFolder();
    final File tmpFile = File.createTempFile("druid", "index", tmpDir);
    try (BufferedWriter writer = Files.newWriter(tmpFile, StandardCharsets.UTF_8)) {
        writer.write(",,\n");
        writer.write("2014-01-01T00:00:10Z,a,1\n");
    }
    final List<String> columns = Arrays.asList("ts", "", "");
    // report parse exception
    final IndexTuningConfig tuningConfig = createTuningConfig(2, null, null, null, null, false, true);
    final IndexIngestionSpec ingestionSpec;
    List<String> expectedMessages;
    if (useInputFormatApi) {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, DEFAULT_TIMESTAMP_SPEC, DimensionsSpec.EMPTY, new CsvInputFormat(columns, null, null, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = ImmutableList.of(StringUtils.format("Timestamp[null] is unparseable! Event: {column_1=2014-01-01T00:00:10Z, column_2=a, column_3=1} (Path: %s, Record: 1, Line: 2)", tmpFile.toURI()));
    } else {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, new CSVParseSpec(DEFAULT_TIMESTAMP_SPEC, DimensionsSpec.EMPTY, null, columns, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = ImmutableList.of("Timestamp[null] is unparseable! Event: {column_1=2014-01-01T00:00:10Z, column_2=a, column_3=1}");
    }
    IndexTask indexTask = new IndexTask(null, null, ingestionSpec, null);
    TaskStatus status = runTask(indexTask).lhs;
    Assert.assertEquals(TaskState.FAILED, status.getStatusCode());
    checkTaskStatusErrorMsgForParseExceptionsExceeded(status);
    IngestionStatsAndErrorsTaskReportData reportData = getTaskReportData();
    List<LinkedHashMap> parseExceptionReports = (List<LinkedHashMap>) reportData.getUnparseableEvents().get(RowIngestionMeters.BUILD_SEGMENTS);
    List<String> actualMessages = parseExceptionReports.stream().map((r) -> {
        return ((List<String>) r.get("details")).get(0);
    }).collect(Collectors.toList());
    Assert.assertEquals(expectedMessages, actualMessages);
    List<String> expectedInputs = ImmutableList.of("{column_1=2014-01-01T00:00:10Z, column_2=a, column_3=1}");
    List<String> actualInputs = parseExceptionReports.stream().map((r) -> {
        return (String) r.get("input");
    }).collect(Collectors.toList());
    Assert.assertEquals(expectedInputs, actualInputs);
}

Example 15 with CsvInputFormat

use of org.apache.druid.data.input.impl.CsvInputFormat in project druid by druid-io.

the class IndexTaskTest method testCsvWithHeaderOfEmptyColumns.

@Test
public void testCsvWithHeaderOfEmptyColumns() throws Exception {
    final File tmpDir = temporaryFolder.newFolder();
    File tmpFile = File.createTempFile("druid", "index", tmpDir);
    try (BufferedWriter writer = Files.newWriter(tmpFile, StandardCharsets.UTF_8)) {
        writer.write("ts,,\n");
        writer.write("2014-01-01T00:00:10Z,a,1\n");
    }
    tmpFile = File.createTempFile("druid", "index", tmpDir);
    try (BufferedWriter writer = Files.newWriter(tmpFile, StandardCharsets.UTF_8)) {
        writer.write("ts,dim,\n");
        writer.write("2014-01-01T00:00:10Z,a,1\n");
    }
    tmpFile = File.createTempFile("druid", "index", tmpDir);
    try (BufferedWriter writer = Files.newWriter(tmpFile, StandardCharsets.UTF_8)) {
        writer.write("ts,,val\n");
        writer.write("2014-01-01T00:00:10Z,a,1\n");
    }
    // report parse exception
    final IndexTuningConfig tuningConfig = createTuningConfig(2, 1, null, null, null, true, true);
    final IndexIngestionSpec ingestionSpec;
    if (useInputFormatApi) {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, DEFAULT_TIMESTAMP_SPEC, DimensionsSpec.EMPTY, new CsvInputFormat(null, null, null, true, 0), null, null, tuningConfig, false, false);
    } else {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, new CSVParseSpec(DEFAULT_TIMESTAMP_SPEC, DimensionsSpec.EMPTY, null, null, true, 0), null, null, tuningConfig, false, false);
    }
    IndexTask indexTask = new IndexTask(null, null, ingestionSpec, null);
    final List<DataSegment> segments = runTask(indexTask).rhs;
    // the order of result segments can be changed because hash shardSpec is used.
    // the below loop is to make this test deterministic.
    Assert.assertEquals(2, segments.size());
    Assert.assertNotEquals(segments.get(0), segments.get(1));
    for (DataSegment segment : segments) {
        System.out.println(segment.getDimensions());
    }
    for (int i = 0; i < 2; i++) {
        final DataSegment segment = segments.get(i);
        final Set<String> dimensions = new HashSet<>(segment.getDimensions());
        Assert.assertTrue(StringUtils.format("Actual dimensions: %s", dimensions), dimensions.equals(Sets.newHashSet("column_2")) || dimensions.equals(Sets.newHashSet("dim", "column_2", "column_3")));
        Assert.assertEquals(Collections.singletonList("val"), segment.getMetrics());
        Assert.assertEquals(Intervals.of("2014/P1D"), segment.getInterval());
    }
}