Search in sources :

Example 81 with DimensionsSpec

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

the class OrcReaderTest method testTest1.

// This test is migrated from OrcHadoopInputRowParserTest
@Test
public void testTest1() throws IOException {
    final InputEntityReader reader = createReader(new TimestampSpec("timestamp", "auto", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("col1", "col2"))), new OrcInputFormat(null, null, new Configuration()), "example/test_1.orc");
    try (CloseableIterator<InputRow> iterator = reader.read()) {
        Assert.assertTrue(iterator.hasNext());
        final InputRow row = iterator.next();
        Assert.assertEquals(DateTimes.of("2016-01-01T00:00:00.000Z"), row.getTimestamp());
        Assert.assertEquals("bar", Iterables.getOnlyElement(row.getDimension("col1")));
        Assert.assertEquals(ImmutableList.of("dat1", "dat2", "dat3"), row.getDimension("col2"));
        Assert.assertEquals(1.1, row.getMetric("val1").doubleValue(), 0.001);
        Assert.assertFalse(iterator.hasNext());
    }
}
Also used : Configuration(org.apache.hadoop.conf.Configuration) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) InputRow(org.apache.druid.data.input.InputRow) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) InputEntityReader(org.apache.druid.data.input.InputEntityReader) Test(org.junit.Test)

Example 82 with DimensionsSpec

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

the class OrcReaderTest method testOrcSplitElim.

// This test is migrated from OrcHadoopInputRowParserTest
@Test
public void testOrcSplitElim() throws IOException {
    final InputEntityReader reader = createReader(new TimestampSpec("ts", "millis", null), new DimensionsSpec(null), new OrcInputFormat(new JSONPathSpec(true, null), null, new Configuration()), "example/orc_split_elim.orc");
    try (CloseableIterator<InputRow> iterator = reader.read()) {
        int actualRowCount = 0;
        Assert.assertTrue(iterator.hasNext());
        final InputRow row = iterator.next();
        actualRowCount++;
        Assert.assertEquals(DateTimes.of("1969-12-31T16:00:00.0Z"), row.getTimestamp());
        Assert.assertEquals("2", Iterables.getOnlyElement(row.getDimension("userid")));
        Assert.assertEquals("foo", Iterables.getOnlyElement(row.getDimension("string1")));
        Assert.assertEquals("0.8", Iterables.getOnlyElement(row.getDimension("subtype")));
        Assert.assertEquals("1.2", Iterables.getOnlyElement(row.getDimension("decimal1")));
        while (iterator.hasNext()) {
            actualRowCount++;
            iterator.next();
        }
        Assert.assertEquals(25000, actualRowCount);
    }
}
Also used : Configuration(org.apache.hadoop.conf.Configuration) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) InputRow(org.apache.druid.data.input.InputRow) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) JSONPathSpec(org.apache.druid.java.util.common.parsers.JSONPathSpec) InputEntityReader(org.apache.druid.data.input.InputEntityReader) Test(org.junit.Test)

Example 83 with DimensionsSpec

use of org.apache.druid.data.input.impl.DimensionsSpec in project hive by apache.

the class TestDruidRecordWriter method testWrite.

// Test is failing due to Guava dependency, Druid 0.13.0 should have less dependency on Guava
@Ignore
@Test
public void testWrite() throws IOException, SegmentLoadingException {
    final String dataSourceName = "testDataSource";
    final File segmentOutputDir = temporaryFolder.newFolder();
    final File workingDir = temporaryFolder.newFolder();
    Configuration config = new Configuration();
    final InputRowParser inputRowParser = new MapInputRowParser(new TimeAndDimsParseSpec(new TimestampSpec(DruidConstants.DEFAULT_TIMESTAMP_COLUMN, "auto", null), new DimensionsSpec(ImmutableList.of(new StringDimensionSchema("host")), null, null)));
    final Map<String, Object> parserMap = objectMapper.convertValue(inputRowParser, new TypeReference<Map<String, Object>>() {
    });
    DataSchema dataSchema = new DataSchema(dataSourceName, parserMap, new AggregatorFactory[] { new LongSumAggregatorFactory("visited_sum", "visited_sum"), new HyperUniquesAggregatorFactory("unique_hosts", "unique_hosts") }, new UniformGranularitySpec(Granularities.DAY, Granularities.NONE, ImmutableList.of(INTERVAL_FULL)), null, objectMapper);
    IndexSpec indexSpec = new IndexSpec(new RoaringBitmapSerdeFactory(true), null, null, null);
    RealtimeTuningConfig tuningConfig = new RealtimeTuningConfig(null, null, null, null, temporaryFolder.newFolder(), null, null, null, null, indexSpec, null, null, 0, 0, null, null, 0L, null, null);
    LocalFileSystem localFileSystem = FileSystem.getLocal(config);
    DataSegmentPusher dataSegmentPusher = new LocalDataSegmentPusher(new LocalDataSegmentPusherConfig() {

        @Override
        public File getStorageDirectory() {
            return segmentOutputDir;
        }
    });
    Path segmentDescriptorPath = new Path(workingDir.getAbsolutePath(), DruidStorageHandler.SEGMENTS_DESCRIPTOR_DIR_NAME);
    DruidRecordWriter druidRecordWriter = new DruidRecordWriter(dataSchema, tuningConfig, dataSegmentPusher, 20, segmentDescriptorPath, localFileSystem);
    List<DruidWritable> druidWritables = expectedRows.stream().map(input -> new DruidWritable(ImmutableMap.<String, Object>builder().putAll(input).put(Constants.DRUID_TIMESTAMP_GRANULARITY_COL_NAME, Granularities.DAY.bucketStart(new DateTime((long) input.get(DruidConstants.DEFAULT_TIMESTAMP_COLUMN))).getMillis()).build())).collect(Collectors.toList());
    for (DruidWritable druidWritable : druidWritables) {
        druidRecordWriter.write(druidWritable);
    }
    druidRecordWriter.close(false);
    List<DataSegment> dataSegmentList = DruidStorageHandlerUtils.getCreatedSegments(segmentDescriptorPath, config);
    Assert.assertEquals(1, dataSegmentList.size());
    File tmpUnzippedSegmentDir = temporaryFolder.newFolder();
    new LocalDataSegmentPuller().getSegmentFiles(dataSegmentList.get(0), tmpUnzippedSegmentDir);
    final QueryableIndex queryableIndex = DruidStorageHandlerUtils.INDEX_IO.loadIndex(tmpUnzippedSegmentDir);
    QueryableIndexStorageAdapter adapter = new QueryableIndexStorageAdapter(queryableIndex);
    Firehose firehose = new IngestSegmentFirehose(ImmutableList.of(new WindowedStorageAdapter(adapter, adapter.getInterval())), null, ImmutableList.of("host"), ImmutableList.of("visited_sum", "unique_hosts"), null);
    List<InputRow> rows = Lists.newArrayList();
    while (firehose.hasMore()) {
        rows.add(firehose.nextRow());
    }
    verifyRows(expectedRows, rows);
}
Also used : FileSystem(org.apache.hadoop.fs.FileSystem) DruidConstants(org.apache.hadoop.hive.druid.conf.DruidConstants) IndexSpec(org.apache.druid.segment.IndexSpec) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) DruidRecordWriter(org.apache.hadoop.hive.druid.io.DruidRecordWriter) LocalDataSegmentPuller(org.apache.druid.segment.loading.LocalDataSegmentPuller) IngestSegmentFirehose(org.apache.druid.segment.realtime.firehose.IngestSegmentFirehose) Configuration(org.apache.hadoop.conf.Configuration) Map(java.util.Map) Path(org.apache.hadoop.fs.Path) TypeReference(com.fasterxml.jackson.core.type.TypeReference) HyperUniquesAggregatorFactory(org.apache.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) RoaringBitmapSerdeFactory(org.apache.druid.segment.data.RoaringBitmapSerdeFactory) ImmutableMap(com.google.common.collect.ImmutableMap) AggregatorFactory(org.apache.druid.query.aggregation.AggregatorFactory) QueryableIndex(org.apache.druid.segment.QueryableIndex) InputRowParser(org.apache.druid.data.input.impl.InputRowParser) Collectors(java.util.stream.Collectors) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) InputRow(org.apache.druid.data.input.InputRow) Firehose(org.apache.druid.data.input.Firehose) List(java.util.List) UniformGranularitySpec(org.apache.druid.segment.indexing.granularity.UniformGranularitySpec) DataSegment(org.apache.druid.timeline.DataSegment) DruidTable(org.apache.calcite.adapter.druid.DruidTable) DataSegmentPusher(org.apache.druid.segment.loading.DataSegmentPusher) SegmentLoadingException(org.apache.druid.segment.loading.SegmentLoadingException) TimeAndDimsParseSpec(org.apache.druid.data.input.impl.TimeAndDimsParseSpec) Interval(org.joda.time.Interval) Lists(com.google.common.collect.Lists) ImmutableList(com.google.common.collect.ImmutableList) LocalDataSegmentPusherConfig(org.apache.druid.segment.loading.LocalDataSegmentPusherConfig) RealtimeTuningConfig(org.apache.druid.segment.indexing.RealtimeTuningConfig) DruidStorageHandlerUtils(org.apache.hadoop.hive.druid.DruidStorageHandlerUtils) Constants(org.apache.hadoop.hive.conf.Constants) DruidStorageHandler(org.apache.hadoop.hive.druid.DruidStorageHandler) QueryableIndexStorageAdapter(org.apache.druid.segment.QueryableIndexStorageAdapter) MapInputRowParser(org.apache.druid.data.input.impl.MapInputRowParser) DruidWritable(org.apache.hadoop.hive.druid.serde.DruidWritable) WindowedStorageAdapter(org.apache.druid.segment.realtime.firehose.WindowedStorageAdapter) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) DateTime(org.joda.time.DateTime) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) Test(org.junit.Test) IOException(java.io.IOException) File(java.io.File) Granularities(org.apache.druid.java.util.common.granularity.Granularities) Rule(org.junit.Rule) Ignore(org.junit.Ignore) LocalDataSegmentPusher(org.apache.druid.segment.loading.LocalDataSegmentPusher) Assert(org.junit.Assert) DataSchema(org.apache.druid.segment.indexing.DataSchema) LocalFileSystem(org.apache.hadoop.fs.LocalFileSystem) TemporaryFolder(org.junit.rules.TemporaryFolder) IngestSegmentFirehose(org.apache.druid.segment.realtime.firehose.IngestSegmentFirehose) IndexSpec(org.apache.druid.segment.IndexSpec) DataSegmentPusher(org.apache.druid.segment.loading.DataSegmentPusher) LocalDataSegmentPusher(org.apache.druid.segment.loading.LocalDataSegmentPusher) Configuration(org.apache.hadoop.conf.Configuration) MapInputRowParser(org.apache.druid.data.input.impl.MapInputRowParser) LongSumAggregatorFactory(org.apache.druid.query.aggregation.LongSumAggregatorFactory) DataSegment(org.apache.druid.timeline.DataSegment) DateTime(org.joda.time.DateTime) TimeAndDimsParseSpec(org.apache.druid.data.input.impl.TimeAndDimsParseSpec) UniformGranularitySpec(org.apache.druid.segment.indexing.granularity.UniformGranularitySpec) RoaringBitmapSerdeFactory(org.apache.druid.segment.data.RoaringBitmapSerdeFactory) LocalDataSegmentPuller(org.apache.druid.segment.loading.LocalDataSegmentPuller) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) WindowedStorageAdapter(org.apache.druid.segment.realtime.firehose.WindowedStorageAdapter) Path(org.apache.hadoop.fs.Path) IngestSegmentFirehose(org.apache.druid.segment.realtime.firehose.IngestSegmentFirehose) Firehose(org.apache.druid.data.input.Firehose) LocalDataSegmentPusherConfig(org.apache.druid.segment.loading.LocalDataSegmentPusherConfig) QueryableIndexStorageAdapter(org.apache.druid.segment.QueryableIndexStorageAdapter) RealtimeTuningConfig(org.apache.druid.segment.indexing.RealtimeTuningConfig) LocalDataSegmentPusher(org.apache.druid.segment.loading.LocalDataSegmentPusher) StringDimensionSchema(org.apache.druid.data.input.impl.StringDimensionSchema) DataSchema(org.apache.druid.segment.indexing.DataSchema) DruidWritable(org.apache.hadoop.hive.druid.serde.DruidWritable) LocalFileSystem(org.apache.hadoop.fs.LocalFileSystem) QueryableIndex(org.apache.druid.segment.QueryableIndex) HyperUniquesAggregatorFactory(org.apache.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) InputRow(org.apache.druid.data.input.InputRow) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) InputRowParser(org.apache.druid.data.input.impl.InputRowParser) MapInputRowParser(org.apache.druid.data.input.impl.MapInputRowParser) File(java.io.File) Map(java.util.Map) ImmutableMap(com.google.common.collect.ImmutableMap) DruidRecordWriter(org.apache.hadoop.hive.druid.io.DruidRecordWriter) Ignore(org.junit.Ignore) Test(org.junit.Test)

Example 84 with DimensionsSpec

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

the class FirehoseFactoryToInputSourceAdaptorTest method testUnimplementedInputFormat.

@Test
public void testUnimplementedInputFormat() throws IOException {
    final List<String> lines = new ArrayList<>();
    for (int i = 0; i < 10; i++) {
        lines.add(StringUtils.format("%d,name_%d,%d", 20190101 + i, i, i + 100));
    }
    final TestFirehoseFactory firehoseFactory = new TestFirehoseFactory(lines);
    final StringInputRowParser inputRowParser = new StringInputRowParser(new CSVParseSpec(new TimestampSpec(null, "yyyyMMdd", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(Arrays.asList("timestamp", "name", "score"))), ",", Arrays.asList("timestamp", "name", "score"), false, 0), StringUtils.UTF8_STRING);
    final FirehoseFactoryToInputSourceAdaptor inputSourceAdaptor = new FirehoseFactoryToInputSourceAdaptor(firehoseFactory, inputRowParser);
    final InputSourceReader reader = inputSourceAdaptor.reader(new InputRowSchema(inputRowParser.getParseSpec().getTimestampSpec(), inputRowParser.getParseSpec().getDimensionsSpec(), ColumnsFilter.all()), null, null);
    final List<InputRow> result = new ArrayList<>();
    try (CloseableIterator<InputRow> iterator = reader.read()) {
        while (iterator.hasNext()) {
            result.add(iterator.next());
        }
    }
    Assert.assertEquals(10, result.size());
    for (int i = 0; i < 10; i++) {
        Assert.assertEquals(DateTimes.of(StringUtils.format("2019-01-%02d", 1 + i)), result.get(i).getTimestamp());
        Assert.assertEquals(StringUtils.format("name_%d", i), Iterables.getOnlyElement(result.get(i).getDimension("name")));
        Assert.assertEquals(StringUtils.format("%d", i + 100), Iterables.getOnlyElement(result.get(i).getDimension("score")));
    }
}
Also used : ArrayList(java.util.ArrayList) CSVParseSpec(org.apache.druid.data.input.impl.CSVParseSpec) StringInputRowParser(org.apache.druid.data.input.impl.StringInputRowParser) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) InitializedNullHandlingTest(org.apache.druid.testing.InitializedNullHandlingTest) Test(org.junit.Test)

Example 85 with DimensionsSpec

use of org.apache.druid.data.input.impl.DimensionsSpec 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);
}
Also used : InputSourceReader(org.apache.druid.data.input.InputSourceReader) TimestampSpec(org.apache.druid.data.input.impl.TimestampSpec) InputRow(org.apache.druid.data.input.InputRow) DimensionsSpec(org.apache.druid.data.input.impl.DimensionsSpec) CsvInputFormat(org.apache.druid.data.input.impl.CsvInputFormat) InputRowSchema(org.apache.druid.data.input.InputRowSchema) InitializedNullHandlingTest(org.apache.druid.testing.InitializedNullHandlingTest) Test(org.junit.Test)

Aggregations

DimensionsSpec (org.apache.druid.data.input.impl.DimensionsSpec)169 Test (org.junit.Test)129 TimestampSpec (org.apache.druid.data.input.impl.TimestampSpec)114 InputRow (org.apache.druid.data.input.InputRow)52 AggregatorFactory (org.apache.druid.query.aggregation.AggregatorFactory)47 LongSumAggregatorFactory (org.apache.druid.query.aggregation.LongSumAggregatorFactory)47 UniformGranularitySpec (org.apache.druid.segment.indexing.granularity.UniformGranularitySpec)42 DataSchema (org.apache.druid.segment.indexing.DataSchema)39 StringDimensionSchema (org.apache.druid.data.input.impl.StringDimensionSchema)37 CountAggregatorFactory (org.apache.druid.query.aggregation.CountAggregatorFactory)37 InputRowSchema (org.apache.druid.data.input.InputRowSchema)36 Map (java.util.Map)32 InitializedNullHandlingTest (org.apache.druid.testing.InitializedNullHandlingTest)32 InputEntityReader (org.apache.druid.data.input.InputEntityReader)31 ArrayList (java.util.ArrayList)29 CsvInputFormat (org.apache.druid.data.input.impl.CsvInputFormat)25 MapBasedInputRow (org.apache.druid.data.input.MapBasedInputRow)24 JSONPathSpec (org.apache.druid.java.util.common.parsers.JSONPathSpec)24 HashMap (java.util.HashMap)23 ImmutableMap (com.google.common.collect.ImmutableMap)21