Search in sources :

Example 16 with CsvInputFormat

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

the class IndexTaskTest method testMultipleParseExceptionsFailure.

@Test
public void testMultipleParseExceptionsFailure() 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("time,dim,dimLong,dimFloat,val\n");
        // unparseable
        writer.write("unparseable,a,2,3.0,1\n");
        // valid row
        writer.write("2014-01-01T00:00:10Z,a,2,3.0,1\n");
        // unparseable
        writer.write("9.0,a,2,3.0,1\n");
        // thrown away
        writer.write("3014-03-01T00:00:10Z,outsideofinterval,2,3.0,1\n");
        // unparseable
        writer.write("99999999999-01-01T00:00:10Z,b,2,3.0,1\n");
    }
    // Allow up to 3 parse exceptions, and save up to 2 parse exceptions
    final IndexTuningConfig tuningConfig = new IndexTuningConfig(null, null, null, null, null, null, null, null, null, null, new DynamicPartitionsSpec(2, null), INDEX_SPEC, null, null, false, false, null, null, null, true, 2, 5, null, null);
    final TimestampSpec timestampSpec = new TimestampSpec("time", "auto", null);
    final DimensionsSpec dimensionsSpec = new DimensionsSpec(Arrays.asList(new StringDimensionSchema("dim"), new LongDimensionSchema("dimLong"), new FloatDimensionSchema("dimFloat")));
    final List<String> columns = Arrays.asList("time", "dim", "dimLong", "dimFloat", "val");
    final IndexIngestionSpec ingestionSpec;
    List<String> expectedMessages;
    if (useInputFormatApi) {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, timestampSpec, dimensionsSpec, new CsvInputFormat(columns, null, null, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = Arrays.asList(StringUtils.format("Timestamp[99999999999-01-01T00:00:10Z] is unparseable! Event: {time=99999999999-01-01T00:00:10Z, dim=b, dimLong=2, dimFloat=3.0, val=1} (Path: %s, Record: 3, Line: 6)", tmpFile.toURI()), StringUtils.format("Timestamp[9.0] is unparseable! Event: {time=9.0, dim=a, dimLong=2, dimFloat=3.0, val=1} (Path: %s, Record: 2, Line: 4)", tmpFile.toURI()), StringUtils.format("Timestamp[unparseable] is unparseable! Event: {time=unparseable, dim=a, dimLong=2, dimFloat=3.0, val=1} (Path: %s, Record: 1, Line: 2)", tmpFile.toURI()));
    } else {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, new CSVParseSpec(timestampSpec, dimensionsSpec, null, columns, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = Arrays.asList("Timestamp[99999999999-01-01T00:00:10Z] is unparseable! Event: {time=99999999999-01-01T00:00:10Z, dim=b, dimLong=2, dimFloat=3.0, val=1}", "Timestamp[9.0] is unparseable! Event: {time=9.0, dim=a, dimLong=2, dimFloat=3.0, val=1}", "Timestamp[unparseable] is unparseable! Event: {time=unparseable, dim=a, dimLong=2, dimFloat=3.0, val=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();
    Map<String, Object> expectedMetrics = ImmutableMap.of(RowIngestionMeters.DETERMINE_PARTITIONS, ImmutableMap.of(RowIngestionMeters.PROCESSED_WITH_ERROR, 0, RowIngestionMeters.PROCESSED, 0, RowIngestionMeters.UNPARSEABLE, 0, RowIngestionMeters.THROWN_AWAY, 0), RowIngestionMeters.BUILD_SEGMENTS, ImmutableMap.of(RowIngestionMeters.PROCESSED_WITH_ERROR, 0, RowIngestionMeters.PROCESSED, 1, RowIngestionMeters.UNPARSEABLE, 3, RowIngestionMeters.THROWN_AWAY, useInputFormatApi ? 1 : 2));
    Assert.assertEquals(expectedMetrics, reportData.getRowStats());
    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 = Arrays.asList("{time=99999999999-01-01T00:00:10Z, dim=b, dimLong=2, dimFloat=3.0, val=1}", "{time=9.0, dim=a, dimLong=2, dimFloat=3.0, val=1}", "{time=unparseable, dim=a, dimLong=2, dimFloat=3.0, val=1}");
    List<String> actualInputs = parseExceptionReports.stream().map((r) -> {
        return (String) r.get("input");
    }).collect(Collectors.toList());
    Assert.assertEquals(expectedInputs, actualInputs);
}
Also used : TaskReport(org.apache.druid.indexing.common.TaskReport) TaskToolbox(org.apache.druid.indexing.common.TaskToolbox) Arrays(java.util.Arrays) IndexSpec(org.apache.druid.segment.IndexSpec) Pair(org.apache.druid.java.util.common.Pair) Map(java.util.Map) ExpressionTransform(org.apache.druid.segment.transform.ExpressionTransform) AppenderatorsManager(org.apache.druid.segment.realtime.appenderator.AppenderatorsManager) JsonInputFormat(org.apache.druid.data.input.impl.JsonInputFormat) IAE(org.apache.druid.java.util.common.IAE) InputFormat(org.apache.druid.data.input.InputFormat) IngestionStatsAndErrorsTaskReportData(org.apache.druid.indexing.common.IngestionStatsAndErrorsTaskReportData) Set(java.util.Set) NoopSegmentHandoffNotifierFactory(org.apache.druid.segment.realtime.plumber.NoopSegmentHandoffNotifierFactory) EqualsVerifier(nl.jqno.equalsverifier.EqualsVerifier) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) StandardCharsets(java.nio.charset.StandardCharsets) TaskState(org.apache.druid.indexer.TaskState) CountDownLatch(java.util.concurrent.CountDownLatch) PartitionIds(org.apache.druid.timeline.partition.PartitionIds) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) RowIngestionMetersFactory(org.apache.druid.segment.incremental.RowIngestionMetersFactory) SegmentLocalCacheManager(org.apache.druid.segment.loading.SegmentLocalCacheManager) SegmentId(org.apache.druid.timeline.SegmentId) TransformSpec(org.apache.druid.segment.transform.TransformSpec) Granularity(org.apache.druid.java.util.common.granularity.Granularity) SegmentLoaderConfig(org.apache.druid.segment.loading.SegmentLoaderConfig) HashBasedNumberedShardSpec(org.apache.druid.timeline.partition.HashBasedNumberedShardSpec) RunWith(org.junit.runner.RunWith) TaskStatus(org.apache.druid.indexer.TaskStatus) ArrayList(java.util.ArrayList) LinkedHashMap(java.util.LinkedHashMap) Interval(org.joda.time.Interval) StringInputRowParser(org.apache.druid.data.input.impl.StringInputRowParser) PartitionsSpec(org.apache.druid.indexer.partitions.PartitionsSpec) Nullable(javax.annotation.Nullable) HashPartitionFunction(org.apache.druid.timeline.partition.HashPartitionFunction) Before(org.junit.Before) BufferedWriter(java.io.BufferedWriter) GranularitySpec(org.apache.druid.segment.indexing.granularity.GranularitySpec) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) Test(org.junit.Test) IOException(java.io.IOException) EasyMock(org.easymock.EasyMock) File(java.io.File) Preconditions(com.google.common.base.Preconditions) Assert(org.junit.Assert) DataSchema(org.apache.druid.segment.indexing.DataSchema) CoreMatchers(org.hamcrest.CoreMatchers) ArbitraryGranularitySpec(org.apache.druid.segment.indexing.granularity.ArbitraryGranularitySpec) IndexIOConfig(org.apache.druid.indexing.common.task.IndexTask.IndexIOConfig) LocalInputSource(org.apache.druid.data.input.impl.LocalInputSource) LongDimensionSchema(org.apache.druid.data.input.impl.LongDimensionSchema) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) SelectorDimFilter(org.apache.druid.query.filter.SelectorDimFilter) Event(org.apache.druid.java.util.emitter.core.Event) DynamicPartitionsSpec(org.apache.druid.indexer.partitions.DynamicPartitionsSpec) TypeReference(com.fasterxml.jackson.core.type.TypeReference) Parameterized(org.junit.runners.Parameterized) ParseSpec(org.apache.druid.data.input.impl.ParseSpec) Sequence(org.apache.druid.java.util.common.guava.Sequence) ShardSpec(org.apache.druid.timeline.partition.ShardSpec) LocalFirehoseFactory(org.apache.druid.segment.realtime.firehose.LocalFirehoseFactory) ImmutableMap(com.google.common.collect.ImmutableMap) AggregatorFactory(org.apache.druid.query.aggregation.AggregatorFactory) NumberedShardSpec(org.apache.druid.timeline.partition.NumberedShardSpec) StringUtils(org.apache.druid.java.util.common.StringUtils) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) HashedPartitionsSpec(org.apache.druid.indexer.partitions.HashedPartitionsSpec) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) LockGranularity(org.apache.druid.indexing.common.LockGranularity) ExprMacroTable(org.apache.druid.math.expr.ExprMacroTable) IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) List(java.util.List) UniformGranularitySpec(org.apache.druid.segment.indexing.granularity.UniformGranularitySpec) ServiceEmitter(org.apache.druid.java.util.emitter.service.ServiceEmitter) DataSegment(org.apache.druid.timeline.DataSegment) SegmentHandoffNotifierFactory(org.apache.druid.segment.handoff.SegmentHandoffNotifierFactory) SegmentAllocateAction(org.apache.druid.indexing.common.actions.SegmentAllocateAction) Intervals(org.apache.druid.java.util.common.Intervals) HashMap(java.util.HashMap) RowIngestionMeters(org.apache.druid.segment.incremental.RowIngestionMeters) HashSet(java.util.HashSet) ImmutableList(com.google.common.collect.ImmutableList) FloatDimensionSchema(org.apache.druid.data.input.impl.FloatDimensionSchema) Files(com.google.common.io.Files) NumberedOverwriteShardSpec(org.apache.druid.timeline.partition.NumberedOverwriteShardSpec) DimensionSelector(org.apache.druid.segment.DimensionSelector) ExpectedException(org.junit.rules.ExpectedException) SegmentHandoffNotifier(org.apache.druid.segment.handoff.SegmentHandoffNotifier) NoopServiceEmitter(org.apache.druid.server.metrics.NoopServiceEmitter) QueryableIndexStorageAdapter(org.apache.druid.segment.QueryableIndexStorageAdapter) VirtualColumns(org.apache.druid.segment.VirtualColumns) WindowedStorageAdapter(org.apache.druid.segment.realtime.firehose.WindowedStorageAdapter) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) JSONParseSpec(org.apache.druid.data.input.impl.JSONParseSpec) StorageLocationConfig(org.apache.druid.segment.loading.StorageLocationConfig) Granularities(org.apache.druid.java.util.common.granularity.Granularities) TimeUnit(java.util.concurrent.TimeUnit) Rule(org.junit.Rule) SingleDimensionPartitionsSpec(org.apache.druid.indexer.partitions.SingleDimensionPartitionsSpec) Cursor(org.apache.druid.segment.Cursor) SegmentCacheManager(org.apache.druid.segment.loading.SegmentCacheManager) IndexIO(org.apache.druid.segment.IndexIO) Collections(java.util.Collections) TemporaryFolder(org.junit.rules.TemporaryFolder) IngestionStatsAndErrorsTaskReportData(org.apache.druid.indexing.common.IngestionStatsAndErrorsTaskReportData) LongDimensionSchema(org.apache.druid.data.input.impl.LongDimensionSchema) FloatDimensionSchema(org.apache.druid.data.input.impl.FloatDimensionSchema) TaskStatus(org.apache.druid.indexer.TaskStatus) BufferedWriter(java.io.BufferedWriter) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) LinkedHashMap(java.util.LinkedHashMap) IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) DynamicPartitionsSpec(org.apache.druid.indexer.partitions.DynamicPartitionsSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) ArrayList(java.util.ArrayList) List(java.util.List) ImmutableList(com.google.common.collect.ImmutableList) File(java.io.File) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) Test(org.junit.Test)

Example 17 with CsvInputFormat

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

the class IndexTaskTest method testReportParseException.

@Test
public void testReportParseException() 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("time,d,val\n");
        writer.write("unparseable,a,1\n");
        writer.write("2014-01-01T00:00:10Z,a,1\n");
    }
    final TimestampSpec timestampSpec = new TimestampSpec("time", "auto", null);
    final List<String> columns = Arrays.asList("time", "dim", "val");
    // report parse exception
    final IndexTuningConfig tuningConfig = createTuningConfig(2, null, null, null, null, false, true);
    final IndexIngestionSpec indexIngestionSpec;
    List<String> expectedMessages;
    if (useInputFormatApi) {
        indexIngestionSpec = createIngestionSpec(jsonMapper, tmpDir, timestampSpec, DimensionsSpec.EMPTY, new CsvInputFormat(columns, null, null, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = ImmutableList.of(StringUtils.format("Timestamp[unparseable] is unparseable! Event: {time=unparseable, d=a, val=1} (Path: %s, Record: 1, Line: 2)", tmpFile.toURI()));
    } else {
        indexIngestionSpec = createIngestionSpec(jsonMapper, tmpDir, new CSVParseSpec(timestampSpec, DimensionsSpec.EMPTY, null, columns, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = ImmutableList.of("Timestamp[unparseable] is unparseable! Event: {time=unparseable, d=a, val=1}");
    }
    IndexTask indexTask = new IndexTask(null, null, indexIngestionSpec, 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("{time=unparseable, d=a, val=1}");
    List<String> actualInputs = parseExceptionReports.stream().map((r) -> {
        return (String) r.get("input");
    }).collect(Collectors.toList());
    Assert.assertEquals(expectedInputs, actualInputs);
}
Also used : TaskReport(org.apache.druid.indexing.common.TaskReport) TaskToolbox(org.apache.druid.indexing.common.TaskToolbox) Arrays(java.util.Arrays) IndexSpec(org.apache.druid.segment.IndexSpec) Pair(org.apache.druid.java.util.common.Pair) Map(java.util.Map) ExpressionTransform(org.apache.druid.segment.transform.ExpressionTransform) AppenderatorsManager(org.apache.druid.segment.realtime.appenderator.AppenderatorsManager) JsonInputFormat(org.apache.druid.data.input.impl.JsonInputFormat) IAE(org.apache.druid.java.util.common.IAE) InputFormat(org.apache.druid.data.input.InputFormat) IngestionStatsAndErrorsTaskReportData(org.apache.druid.indexing.common.IngestionStatsAndErrorsTaskReportData) Set(java.util.Set) NoopSegmentHandoffNotifierFactory(org.apache.druid.segment.realtime.plumber.NoopSegmentHandoffNotifierFactory) EqualsVerifier(nl.jqno.equalsverifier.EqualsVerifier) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) StandardCharsets(java.nio.charset.StandardCharsets) TaskState(org.apache.druid.indexer.TaskState) CountDownLatch(java.util.concurrent.CountDownLatch) PartitionIds(org.apache.druid.timeline.partition.PartitionIds) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) RowIngestionMetersFactory(org.apache.druid.segment.incremental.RowIngestionMetersFactory) SegmentLocalCacheManager(org.apache.druid.segment.loading.SegmentLocalCacheManager) SegmentId(org.apache.druid.timeline.SegmentId) TransformSpec(org.apache.druid.segment.transform.TransformSpec) Granularity(org.apache.druid.java.util.common.granularity.Granularity) SegmentLoaderConfig(org.apache.druid.segment.loading.SegmentLoaderConfig) HashBasedNumberedShardSpec(org.apache.druid.timeline.partition.HashBasedNumberedShardSpec) RunWith(org.junit.runner.RunWith) TaskStatus(org.apache.druid.indexer.TaskStatus) ArrayList(java.util.ArrayList) LinkedHashMap(java.util.LinkedHashMap) Interval(org.joda.time.Interval) StringInputRowParser(org.apache.druid.data.input.impl.StringInputRowParser) PartitionsSpec(org.apache.druid.indexer.partitions.PartitionsSpec) Nullable(javax.annotation.Nullable) HashPartitionFunction(org.apache.druid.timeline.partition.HashPartitionFunction) Before(org.junit.Before) BufferedWriter(java.io.BufferedWriter) GranularitySpec(org.apache.druid.segment.indexing.granularity.GranularitySpec) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) Test(org.junit.Test) IOException(java.io.IOException) EasyMock(org.easymock.EasyMock) File(java.io.File) Preconditions(com.google.common.base.Preconditions) Assert(org.junit.Assert) DataSchema(org.apache.druid.segment.indexing.DataSchema) CoreMatchers(org.hamcrest.CoreMatchers) ArbitraryGranularitySpec(org.apache.druid.segment.indexing.granularity.ArbitraryGranularitySpec) IndexIOConfig(org.apache.druid.indexing.common.task.IndexTask.IndexIOConfig) LocalInputSource(org.apache.druid.data.input.impl.LocalInputSource) LongDimensionSchema(org.apache.druid.data.input.impl.LongDimensionSchema) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) SelectorDimFilter(org.apache.druid.query.filter.SelectorDimFilter) Event(org.apache.druid.java.util.emitter.core.Event) DynamicPartitionsSpec(org.apache.druid.indexer.partitions.DynamicPartitionsSpec) TypeReference(com.fasterxml.jackson.core.type.TypeReference) Parameterized(org.junit.runners.Parameterized) ParseSpec(org.apache.druid.data.input.impl.ParseSpec) Sequence(org.apache.druid.java.util.common.guava.Sequence) ShardSpec(org.apache.druid.timeline.partition.ShardSpec) LocalFirehoseFactory(org.apache.druid.segment.realtime.firehose.LocalFirehoseFactory) ImmutableMap(com.google.common.collect.ImmutableMap) AggregatorFactory(org.apache.druid.query.aggregation.AggregatorFactory) NumberedShardSpec(org.apache.druid.timeline.partition.NumberedShardSpec) StringUtils(org.apache.druid.java.util.common.StringUtils) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) HashedPartitionsSpec(org.apache.druid.indexer.partitions.HashedPartitionsSpec) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) LockGranularity(org.apache.druid.indexing.common.LockGranularity) ExprMacroTable(org.apache.druid.math.expr.ExprMacroTable) IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) List(java.util.List) UniformGranularitySpec(org.apache.druid.segment.indexing.granularity.UniformGranularitySpec) ServiceEmitter(org.apache.druid.java.util.emitter.service.ServiceEmitter) DataSegment(org.apache.druid.timeline.DataSegment) SegmentHandoffNotifierFactory(org.apache.druid.segment.handoff.SegmentHandoffNotifierFactory) SegmentAllocateAction(org.apache.druid.indexing.common.actions.SegmentAllocateAction) Intervals(org.apache.druid.java.util.common.Intervals) HashMap(java.util.HashMap) RowIngestionMeters(org.apache.druid.segment.incremental.RowIngestionMeters) HashSet(java.util.HashSet) ImmutableList(com.google.common.collect.ImmutableList) FloatDimensionSchema(org.apache.druid.data.input.impl.FloatDimensionSchema) Files(com.google.common.io.Files) NumberedOverwriteShardSpec(org.apache.druid.timeline.partition.NumberedOverwriteShardSpec) DimensionSelector(org.apache.druid.segment.DimensionSelector) ExpectedException(org.junit.rules.ExpectedException) SegmentHandoffNotifier(org.apache.druid.segment.handoff.SegmentHandoffNotifier) NoopServiceEmitter(org.apache.druid.server.metrics.NoopServiceEmitter) QueryableIndexStorageAdapter(org.apache.druid.segment.QueryableIndexStorageAdapter) VirtualColumns(org.apache.druid.segment.VirtualColumns) WindowedStorageAdapter(org.apache.druid.segment.realtime.firehose.WindowedStorageAdapter) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) JSONParseSpec(org.apache.druid.data.input.impl.JSONParseSpec) StorageLocationConfig(org.apache.druid.segment.loading.StorageLocationConfig) Granularities(org.apache.druid.java.util.common.granularity.Granularities) TimeUnit(java.util.concurrent.TimeUnit) Rule(org.junit.Rule) SingleDimensionPartitionsSpec(org.apache.druid.indexer.partitions.SingleDimensionPartitionsSpec) Cursor(org.apache.druid.segment.Cursor) SegmentCacheManager(org.apache.druid.segment.loading.SegmentCacheManager) IndexIO(org.apache.druid.segment.IndexIO) Collections(java.util.Collections) TemporaryFolder(org.junit.rules.TemporaryFolder) IngestionStatsAndErrorsTaskReportData(org.apache.druid.indexing.common.IngestionStatsAndErrorsTaskReportData) TaskStatus(org.apache.druid.indexer.TaskStatus) BufferedWriter(java.io.BufferedWriter) LinkedHashMap(java.util.LinkedHashMap) IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) ArrayList(java.util.ArrayList) List(java.util.List) ImmutableList(com.google.common.collect.ImmutableList) File(java.io.File) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) Test(org.junit.Test)

Example 18 with CsvInputFormat

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

the class IndexTaskTest method testMultipleParseExceptionsFailureAtDeterminePartitions.

@Test
public void testMultipleParseExceptionsFailureAtDeterminePartitions() 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("time,dim,dimLong,dimFloat,val\n");
        // unparseable
        writer.write("unparseable,a,2,3.0,1\n");
        // valid row
        writer.write("2014-01-01T00:00:10Z,a,2,3.0,1\n");
        // unparseable
        writer.write("9.0,a,2,3.0,1\n");
        // thrown away
        writer.write("3014-03-01T00:00:10Z,outsideofinterval,2,3.0,1\n");
        // unparseable
        writer.write("99999999999-01-01T00:00:10Z,b,2,3.0,1\n");
    }
    // Allow up to 3 parse exceptions, and save up to 2 parse exceptions
    final IndexTuningConfig tuningConfig = new IndexTuningConfig(null, null, null, null, null, null, null, null, null, null, new HashedPartitionsSpec(2, null, null), INDEX_SPEC, null, null, true, false, null, null, null, true, 2, 5, null, null);
    final TimestampSpec timestampSpec = new TimestampSpec("time", "auto", null);
    final DimensionsSpec dimensionsSpec = new DimensionsSpec(Arrays.asList(new StringDimensionSchema("dim"), new LongDimensionSchema("dimLong"), new FloatDimensionSchema("dimFloat")));
    final List<String> columns = Arrays.asList("time", "dim", "dimLong", "dimFloat", "val");
    final IndexIngestionSpec ingestionSpec;
    List<String> expectedMessages;
    if (useInputFormatApi) {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, timestampSpec, dimensionsSpec, new CsvInputFormat(columns, null, null, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = Arrays.asList(StringUtils.format("Timestamp[99999999999-01-01T00:00:10Z] is unparseable! Event: {time=99999999999-01-01T00:00:10Z, dim=b, dimLong=2, dimFloat=3.0, val=1} (Path: %s, Record: 3, Line: 6)", tmpFile.toURI()), StringUtils.format("Timestamp[9.0] is unparseable! Event: {time=9.0, dim=a, dimLong=2, dimFloat=3.0, val=1} (Path: %s, Record: 2, Line: 4)", tmpFile.toURI()), StringUtils.format("Timestamp[unparseable] is unparseable! Event: {time=unparseable, dim=a, dimLong=2, dimFloat=3.0, val=1} (Path: %s, Record: 1, Line: 2)", tmpFile.toURI()));
    } else {
        ingestionSpec = createIngestionSpec(jsonMapper, tmpDir, new CSVParseSpec(timestampSpec, dimensionsSpec, null, columns, true, 0), null, null, tuningConfig, false, false);
        expectedMessages = Arrays.asList("Timestamp[99999999999-01-01T00:00:10Z] is unparseable! Event: {time=99999999999-01-01T00:00:10Z, dim=b, dimLong=2, dimFloat=3.0, val=1}", "Timestamp[9.0] is unparseable! Event: {time=9.0, dim=a, dimLong=2, dimFloat=3.0, val=1}", "Timestamp[unparseable] is unparseable! Event: {time=unparseable, dim=a, dimLong=2, dimFloat=3.0, val=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();
    Map<String, Object> expectedMetrics = ImmutableMap.of(RowIngestionMeters.DETERMINE_PARTITIONS, ImmutableMap.of(RowIngestionMeters.PROCESSED_WITH_ERROR, 0, RowIngestionMeters.PROCESSED, 1, RowIngestionMeters.UNPARSEABLE, 3, RowIngestionMeters.THROWN_AWAY, useInputFormatApi ? 1 : 2), RowIngestionMeters.BUILD_SEGMENTS, ImmutableMap.of(RowIngestionMeters.PROCESSED_WITH_ERROR, 0, RowIngestionMeters.PROCESSED, 0, RowIngestionMeters.UNPARSEABLE, 0, RowIngestionMeters.THROWN_AWAY, 0));
    Assert.assertEquals(expectedMetrics, reportData.getRowStats());
    List<LinkedHashMap> parseExceptionReports = (List<LinkedHashMap>) reportData.getUnparseableEvents().get(RowIngestionMeters.DETERMINE_PARTITIONS);
    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 = Arrays.asList("{time=99999999999-01-01T00:00:10Z, dim=b, dimLong=2, dimFloat=3.0, val=1}", "{time=9.0, dim=a, dimLong=2, dimFloat=3.0, val=1}", "{time=unparseable, dim=a, dimLong=2, dimFloat=3.0, val=1}");
    List<String> actualInputs = parseExceptionReports.stream().map((r) -> {
        return (String) r.get("input");
    }).collect(Collectors.toList());
    Assert.assertEquals(expectedInputs, actualInputs);
}
Also used : TaskReport(org.apache.druid.indexing.common.TaskReport) TaskToolbox(org.apache.druid.indexing.common.TaskToolbox) Arrays(java.util.Arrays) IndexSpec(org.apache.druid.segment.IndexSpec) Pair(org.apache.druid.java.util.common.Pair) Map(java.util.Map) ExpressionTransform(org.apache.druid.segment.transform.ExpressionTransform) AppenderatorsManager(org.apache.druid.segment.realtime.appenderator.AppenderatorsManager) JsonInputFormat(org.apache.druid.data.input.impl.JsonInputFormat) IAE(org.apache.druid.java.util.common.IAE) InputFormat(org.apache.druid.data.input.InputFormat) IngestionStatsAndErrorsTaskReportData(org.apache.druid.indexing.common.IngestionStatsAndErrorsTaskReportData) Set(java.util.Set) NoopSegmentHandoffNotifierFactory(org.apache.druid.segment.realtime.plumber.NoopSegmentHandoffNotifierFactory) EqualsVerifier(nl.jqno.equalsverifier.EqualsVerifier) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) StandardCharsets(java.nio.charset.StandardCharsets) TaskState(org.apache.druid.indexer.TaskState) CountDownLatch(java.util.concurrent.CountDownLatch) PartitionIds(org.apache.druid.timeline.partition.PartitionIds) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) RowIngestionMetersFactory(org.apache.druid.segment.incremental.RowIngestionMetersFactory) SegmentLocalCacheManager(org.apache.druid.segment.loading.SegmentLocalCacheManager) SegmentId(org.apache.druid.timeline.SegmentId) TransformSpec(org.apache.druid.segment.transform.TransformSpec) Granularity(org.apache.druid.java.util.common.granularity.Granularity) SegmentLoaderConfig(org.apache.druid.segment.loading.SegmentLoaderConfig) HashBasedNumberedShardSpec(org.apache.druid.timeline.partition.HashBasedNumberedShardSpec) RunWith(org.junit.runner.RunWith) TaskStatus(org.apache.druid.indexer.TaskStatus) ArrayList(java.util.ArrayList) LinkedHashMap(java.util.LinkedHashMap) Interval(org.joda.time.Interval) StringInputRowParser(org.apache.druid.data.input.impl.StringInputRowParser) PartitionsSpec(org.apache.druid.indexer.partitions.PartitionsSpec) Nullable(javax.annotation.Nullable) HashPartitionFunction(org.apache.druid.timeline.partition.HashPartitionFunction) Before(org.junit.Before) BufferedWriter(java.io.BufferedWriter) GranularitySpec(org.apache.druid.segment.indexing.granularity.GranularitySpec) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) Test(org.junit.Test) IOException(java.io.IOException) EasyMock(org.easymock.EasyMock) File(java.io.File) Preconditions(com.google.common.base.Preconditions) Assert(org.junit.Assert) DataSchema(org.apache.druid.segment.indexing.DataSchema) CoreMatchers(org.hamcrest.CoreMatchers) ArbitraryGranularitySpec(org.apache.druid.segment.indexing.granularity.ArbitraryGranularitySpec) IndexIOConfig(org.apache.druid.indexing.common.task.IndexTask.IndexIOConfig) LocalInputSource(org.apache.druid.data.input.impl.LocalInputSource) LongDimensionSchema(org.apache.druid.data.input.impl.LongDimensionSchema) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) DefaultDimensionSpec(org.apache.druid.query.dimension.DefaultDimensionSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) SelectorDimFilter(org.apache.druid.query.filter.SelectorDimFilter) Event(org.apache.druid.java.util.emitter.core.Event) DynamicPartitionsSpec(org.apache.druid.indexer.partitions.DynamicPartitionsSpec) TypeReference(com.fasterxml.jackson.core.type.TypeReference) Parameterized(org.junit.runners.Parameterized) ParseSpec(org.apache.druid.data.input.impl.ParseSpec) Sequence(org.apache.druid.java.util.common.guava.Sequence) ShardSpec(org.apache.druid.timeline.partition.ShardSpec) LocalFirehoseFactory(org.apache.druid.segment.realtime.firehose.LocalFirehoseFactory) ImmutableMap(com.google.common.collect.ImmutableMap) AggregatorFactory(org.apache.druid.query.aggregation.AggregatorFactory) NumberedShardSpec(org.apache.druid.timeline.partition.NumberedShardSpec) StringUtils(org.apache.druid.java.util.common.StringUtils) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) HashedPartitionsSpec(org.apache.druid.indexer.partitions.HashedPartitionsSpec) Collectors(java.util.stream.Collectors) Sets(com.google.common.collect.Sets) LockGranularity(org.apache.druid.indexing.common.LockGranularity) ExprMacroTable(org.apache.druid.math.expr.ExprMacroTable) IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) List(java.util.List) UniformGranularitySpec(org.apache.druid.segment.indexing.granularity.UniformGranularitySpec) ServiceEmitter(org.apache.druid.java.util.emitter.service.ServiceEmitter) DataSegment(org.apache.druid.timeline.DataSegment) SegmentHandoffNotifierFactory(org.apache.druid.segment.handoff.SegmentHandoffNotifierFactory) SegmentAllocateAction(org.apache.druid.indexing.common.actions.SegmentAllocateAction) Intervals(org.apache.druid.java.util.common.Intervals) HashMap(java.util.HashMap) RowIngestionMeters(org.apache.druid.segment.incremental.RowIngestionMeters) HashSet(java.util.HashSet) ImmutableList(com.google.common.collect.ImmutableList) FloatDimensionSchema(org.apache.druid.data.input.impl.FloatDimensionSchema) Files(com.google.common.io.Files) NumberedOverwriteShardSpec(org.apache.druid.timeline.partition.NumberedOverwriteShardSpec) DimensionSelector(org.apache.druid.segment.DimensionSelector) ExpectedException(org.junit.rules.ExpectedException) SegmentHandoffNotifier(org.apache.druid.segment.handoff.SegmentHandoffNotifier) NoopServiceEmitter(org.apache.druid.server.metrics.NoopServiceEmitter) QueryableIndexStorageAdapter(org.apache.druid.segment.QueryableIndexStorageAdapter) VirtualColumns(org.apache.druid.segment.VirtualColumns) WindowedStorageAdapter(org.apache.druid.segment.realtime.firehose.WindowedStorageAdapter) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) JSONParseSpec(org.apache.druid.data.input.impl.JSONParseSpec) StorageLocationConfig(org.apache.druid.segment.loading.StorageLocationConfig) Granularities(org.apache.druid.java.util.common.granularity.Granularities) TimeUnit(java.util.concurrent.TimeUnit) Rule(org.junit.Rule) SingleDimensionPartitionsSpec(org.apache.druid.indexer.partitions.SingleDimensionPartitionsSpec) Cursor(org.apache.druid.segment.Cursor) SegmentCacheManager(org.apache.druid.segment.loading.SegmentCacheManager) IndexIO(org.apache.druid.segment.IndexIO) Collections(java.util.Collections) TemporaryFolder(org.junit.rules.TemporaryFolder) HashedPartitionsSpec(org.apache.druid.indexer.partitions.HashedPartitionsSpec) IngestionStatsAndErrorsTaskReportData(org.apache.druid.indexing.common.IngestionStatsAndErrorsTaskReportData) LongDimensionSchema(org.apache.druid.data.input.impl.LongDimensionSchema) FloatDimensionSchema(org.apache.druid.data.input.impl.FloatDimensionSchema) TaskStatus(org.apache.druid.indexer.TaskStatus) BufferedWriter(java.io.BufferedWriter) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) LinkedHashMap(java.util.LinkedHashMap) IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) ArrayList(java.util.ArrayList) List(java.util.List) ImmutableList(com.google.common.collect.ImmutableList) File(java.io.File) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) Test(org.junit.Test)

Example 19 with CsvInputFormat

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

the class IndexTaskTest method testIgnoreParseException.

@Test
public void testIgnoreParseException() 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("time,d,val\n");
        writer.write("unparseable,a,1\n");
        writer.write("2014-01-01T00:00:10Z,a,1\n");
    }
    final TimestampSpec timestampSpec = new TimestampSpec("time", "auto", null);
    final List<String> columns = Arrays.asList("time", "dim", "val");
    // ignore parse exception
    final IndexTuningConfig tuningConfig = createTuningConfig(2, null, null, null, null, false, false);
    // GranularitySpec.intervals and numShards must be null to verify reportParseException=false is respected both in
    // IndexTask.determineShardSpecs() and IndexTask.generateAndPublishSegments()
    final IndexIngestionSpec parseExceptionIgnoreSpec;
    if (useInputFormatApi) {
        parseExceptionIgnoreSpec = createIngestionSpec(jsonMapper, tmpDir, timestampSpec, DimensionsSpec.EMPTY, new CsvInputFormat(columns, null, null, true, 0), null, null, tuningConfig, false, false);
    } else {
        parseExceptionIgnoreSpec = createIngestionSpec(jsonMapper, tmpDir, new CSVParseSpec(timestampSpec, DimensionsSpec.EMPTY, null, columns, true, 0), null, null, tuningConfig, false, false);
    }
    IndexTask indexTask = new IndexTask(null, null, parseExceptionIgnoreSpec, null);
    final List<DataSegment> segments = runTask(indexTask).rhs;
    Assert.assertEquals(Collections.singletonList("d"), segments.get(0).getDimensions());
    Assert.assertEquals(Collections.singletonList("val"), segments.get(0).getMetrics());
    Assert.assertEquals(Intervals.of("2014/P1D"), segments.get(0).getInterval());
}
Also used : IndexIngestionSpec(org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) File(java.io.File) DataSegment(org.apache.druid.timeline.DataSegment) BufferedWriter(java.io.BufferedWriter) IndexTuningConfig(org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig) Test(org.junit.Test)

Aggregations

CsvInputFormat (org.apache.druid.data.input.impl.CsvInputFormat)19 TimestampSpec (org.apache.druid.data.input.impl.TimestampSpec)18 Test (org.junit.Test)18 DimensionsSpec (org.apache.druid.data.input.impl.DimensionsSpec)15 BufferedWriter (java.io.BufferedWriter)9 File (java.io.File)9 CSVParseSpec (org.apache.druid.data.input.impl.CSVParseSpec)9 IndexIngestionSpec (org.apache.druid.indexing.common.task.IndexTask.IndexIngestionSpec)9 IndexTuningConfig (org.apache.druid.indexing.common.task.IndexTask.IndexTuningConfig)9 DataSegment (org.apache.druid.timeline.DataSegment)9 InputFormat (org.apache.druid.data.input.InputFormat)7 IOException (java.io.IOException)6 ArrayList (java.util.ArrayList)6 Collections (java.util.Collections)6 List (java.util.List)6 Map (java.util.Map)6 Set (java.util.Set)6 Collectors (java.util.stream.Collectors)6 Nullable (javax.annotation.Nullable)6 LocalInputSource (org.apache.druid.data.input.impl.LocalInputSource)6