Search in sources :

Example 71 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory in project druid by druid-io.

the class InputRowSerdeTest method testThrowParseExceptions.

@Test(expected = ParseException.class)
public void testThrowParseExceptions() {
    InputRow in = new MapBasedInputRow(timestamp, dims, event);
    AggregatorFactory[] aggregatorFactories = new AggregatorFactory[] { new DoubleSumAggregatorFactory("agg_non_existing", "agg_non_existing_in"), new DoubleSumAggregatorFactory("m1out", "m1"), new LongSumAggregatorFactory("m2out", "m2"), new HyperUniquesAggregatorFactory("m3out", "m3"), // Unparseable from String to Long
    new LongSumAggregatorFactory("unparseable", "m3") };
    InputRowSerde.toBytes(in, aggregatorFactories, true);
}
Also used : DoubleSumAggregatorFactory(io.druid.query.aggregation.DoubleSumAggregatorFactory) MapBasedInputRow(io.druid.data.input.MapBasedInputRow) InputRow(io.druid.data.input.InputRow) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) MapBasedInputRow(io.druid.data.input.MapBasedInputRow) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) DoubleSumAggregatorFactory(io.druid.query.aggregation.DoubleSumAggregatorFactory) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) Test(org.junit.Test)

Example 72 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory in project druid by druid-io.

the class BatchDeltaIngestionTest method makeHadoopDruidIndexerConfig.

private HadoopDruidIndexerConfig makeHadoopDruidIndexerConfig(Map<String, Object> inputSpec, File tmpDir) throws Exception {
    HadoopDruidIndexerConfig config = new HadoopDruidIndexerConfig(new HadoopIngestionSpec(new DataSchema("website", MAPPER.convertValue(new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "host2", "visited_num")), null), Map.class), new AggregatorFactory[] { new LongSumAggregatorFactory("visited_sum", "visited_num"), new HyperUniquesAggregatorFactory("unique_hosts", "host2") }, new UniformGranularitySpec(Granularities.DAY, Granularities.NONE, ImmutableList.of(INTERVAL_FULL)), MAPPER), new HadoopIOConfig(inputSpec, null, tmpDir.getCanonicalPath()), new HadoopTuningConfig(tmpDir.getCanonicalPath(), null, null, null, null, null, false, false, false, false, null, false, false, null, null, null, false, false)));
    config.setShardSpecs(ImmutableMap.<Long, List<HadoopyShardSpec>>of(INTERVAL_FULL.getStartMillis(), ImmutableList.of(new HadoopyShardSpec(new HashBasedNumberedShardSpec(0, 1, null, HadoopDruidIndexerConfig.JSON_MAPPER), 0))));
    config = HadoopDruidIndexerConfig.fromSpec(config.getSchema());
    return config;
}
Also used : HashBasedNumberedShardSpec(io.druid.timeline.partition.HashBasedNumberedShardSpec) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) DataSchema(io.druid.segment.indexing.DataSchema) UniformGranularitySpec(io.druid.segment.indexing.granularity.UniformGranularitySpec) CSVParseSpec(io.druid.data.input.impl.CSVParseSpec) StringInputRowParser(io.druid.data.input.impl.StringInputRowParser) TimestampSpec(io.druid.data.input.impl.TimestampSpec) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) DimensionsSpec(io.druid.data.input.impl.DimensionsSpec) Map(java.util.Map) ImmutableMap(com.google.common.collect.ImmutableMap)

Example 73 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory 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(DruidStorageHandlerUtils.DEFAULT_TIMESTAMP_COLUMN, "auto", null), new DimensionsSpec(ImmutableList.<DimensionSchema>of(new StringDimensionSchema("host")), null, null)));
    final Map<String, Object> parserMap = objectMapper.convertValue(inputRowParser, Map.class);
    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)), objectMapper);
    IndexSpec indexSpec = new IndexSpec(new RoaringBitmapSerdeFactory(true), null, null, null);
    RealtimeTuningConfig tuningConfig = new RealtimeTuningConfig(null, null, null, temporaryFolder.newFolder(), null, null, null, null, indexSpec, null, 0, 0, null, null, 0L);
    LocalFileSystem localFileSystem = FileSystem.getLocal(config);
    DataSegmentPusher dataSegmentPusher = new LocalDataSegmentPusher(new LocalDataSegmentPusherConfig() {

        @Override
        public File getStorageDirectory() {
            return segmentOutputDir;
        }
    }, objectMapper);
    Path segmentDescriptroPath = new Path(workingDir.getAbsolutePath(), DruidStorageHandler.SEGMENTS_DESCRIPTOR_DIR_NAME);
    druidRecordWriter = new DruidRecordWriter(dataSchema, tuningConfig, dataSegmentPusher, 20, segmentDescriptroPath, localFileSystem);
    List<DruidWritable> druidWritables = Lists.transform(expectedRows, new Function<ImmutableMap<String, Object>, DruidWritable>() {

        @Nullable
        @Override
        public DruidWritable apply(@Nullable ImmutableMap<String, Object> input) {
            return new DruidWritable(ImmutableMap.<String, Object>builder().putAll(input).put(Constants.DRUID_TIMESTAMP_GRANULARITY_COL_NAME, Granularities.DAY.bucketStart(new DateTime((long) input.get(DruidStorageHandlerUtils.DEFAULT_TIMESTAMP_COLUMN))).getMillis()).build());
        }
    });
    for (DruidWritable druidWritable : druidWritables) {
        druidRecordWriter.write(druidWritable);
    }
    druidRecordWriter.close(false);
    List<DataSegment> dataSegmentList = DruidStorageHandlerUtils.getCreatedSegments(segmentDescriptroPath, 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())), 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 : IngestSegmentFirehose(io.druid.segment.realtime.firehose.IngestSegmentFirehose) IndexSpec(io.druid.segment.IndexSpec) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) DataSegmentPusher(io.druid.segment.loading.DataSegmentPusher) Configuration(org.apache.hadoop.conf.Configuration) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) DataSegment(io.druid.timeline.DataSegment) DateTime(org.joda.time.DateTime) TimeAndDimsParseSpec(io.druid.data.input.impl.TimeAndDimsParseSpec) UniformGranularitySpec(io.druid.segment.indexing.granularity.UniformGranularitySpec) RoaringBitmapSerdeFactory(io.druid.segment.data.RoaringBitmapSerdeFactory) LocalDataSegmentPuller(io.druid.segment.loading.LocalDataSegmentPuller) TimestampSpec(io.druid.data.input.impl.TimestampSpec) WindowedStorageAdapter(io.druid.segment.realtime.firehose.WindowedStorageAdapter) Path(org.apache.hadoop.fs.Path) Firehose(io.druid.data.input.Firehose) IngestSegmentFirehose(io.druid.segment.realtime.firehose.IngestSegmentFirehose) LocalDataSegmentPusherConfig(io.druid.segment.loading.LocalDataSegmentPusherConfig) QueryableIndexStorageAdapter(io.druid.segment.QueryableIndexStorageAdapter) RealtimeTuningConfig(io.druid.segment.indexing.RealtimeTuningConfig) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) ImmutableMap(com.google.common.collect.ImmutableMap) StringDimensionSchema(io.druid.data.input.impl.StringDimensionSchema) DataSchema(io.druid.segment.indexing.DataSchema) DruidWritable(org.apache.hadoop.hive.druid.serde.DruidWritable) LocalFileSystem(org.apache.hadoop.fs.LocalFileSystem) QueryableIndex(io.druid.segment.QueryableIndex) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) InputRow(io.druid.data.input.InputRow) DimensionsSpec(io.druid.data.input.impl.DimensionsSpec) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) InputRowParser(io.druid.data.input.impl.InputRowParser) File(java.io.File) DruidRecordWriter(org.apache.hadoop.hive.druid.io.DruidRecordWriter) Nullable(javax.annotation.Nullable) Ignore(org.junit.Ignore) Test(org.junit.Test)

Example 74 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory in project druid by druid-io.

the class CachingQueryRunnerTest method testTimeseries.

@Test
public void testTimeseries() throws Exception {
    for (boolean descending : new boolean[] { false, true }) {
        TimeseriesQuery query = Druids.newTimeseriesQueryBuilder().dataSource(QueryRunnerTestHelper.dataSource).granularity(QueryRunnerTestHelper.dayGran).intervals(QueryRunnerTestHelper.firstToThird).aggregators(Arrays.<AggregatorFactory>asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"), QueryRunnerTestHelper.qualityUniques)).descending(descending).build();
        Result row1 = new Result(new DateTime("2011-04-01"), new TimeseriesResultValue(ImmutableMap.<String, Object>of("rows", 13L, "idx", 6619L, "uniques", QueryRunnerTestHelper.UNIQUES_9)));
        Result row2 = new Result<>(new DateTime("2011-04-02"), new TimeseriesResultValue(ImmutableMap.<String, Object>of("rows", 13L, "idx", 5827L, "uniques", QueryRunnerTestHelper.UNIQUES_9)));
        List<Result> expectedResults;
        if (descending) {
            expectedResults = Lists.newArrayList(row2, row1);
        } else {
            expectedResults = Lists.newArrayList(row1, row2);
        }
        QueryToolChest toolChest = new TimeseriesQueryQueryToolChest(QueryRunnerTestHelper.NoopIntervalChunkingQueryRunnerDecorator());
        testCloseAndPopulate(expectedResults, expectedResults, query, toolChest);
        testUseCache(expectedResults, query, toolChest);
    }
}
Also used : TimeseriesResultValue(io.druid.query.timeseries.TimeseriesResultValue) TimeseriesQuery(io.druid.query.timeseries.TimeseriesQuery) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) TopNQueryQueryToolChest(io.druid.query.topn.TopNQueryQueryToolChest) TimeseriesQueryQueryToolChest(io.druid.query.timeseries.TimeseriesQueryQueryToolChest) QueryToolChest(io.druid.query.QueryToolChest) TimeseriesQueryQueryToolChest(io.druid.query.timeseries.TimeseriesQueryQueryToolChest) DateTime(org.joda.time.DateTime) Result(io.druid.query.Result) Test(org.junit.Test)

Example 75 with LongSumAggregatorFactory

use of io.druid.query.aggregation.LongSumAggregatorFactory in project druid by druid-io.

the class KafkaIndexTaskTest method countEvents.

public long countEvents(final Task task) throws Exception {
    // Do a query.
    TimeseriesQuery query = Druids.newTimeseriesQueryBuilder().dataSource(DATA_SCHEMA.getDataSource()).aggregators(ImmutableList.<AggregatorFactory>of(new LongSumAggregatorFactory("rows", "rows"))).granularity(Granularities.ALL).intervals("0000/3000").build();
    ArrayList<Result<TimeseriesResultValue>> results = Sequences.toList(task.getQueryRunner(query).run(query, ImmutableMap.<String, Object>of()), Lists.<Result<TimeseriesResultValue>>newArrayList());
    return results.isEmpty() ? 0 : results.get(0).getValue().getLongMetric("rows");
}
Also used : TimeseriesResultValue(io.druid.query.timeseries.TimeseriesResultValue) TimeseriesQuery(io.druid.query.timeseries.TimeseriesQuery) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) CountAggregatorFactory(io.druid.query.aggregation.CountAggregatorFactory) AggregatorFactory(io.druid.query.aggregation.AggregatorFactory) Result(io.druid.query.Result)

Aggregations

LongSumAggregatorFactory (io.druid.query.aggregation.LongSumAggregatorFactory)160 Test (org.junit.Test)137 DefaultDimensionSpec (io.druid.query.dimension.DefaultDimensionSpec)87 Row (io.druid.data.input.Row)80 DimensionSpec (io.druid.query.dimension.DimensionSpec)58 ExtractionDimensionSpec (io.druid.query.dimension.ExtractionDimensionSpec)54 ListFilteredDimensionSpec (io.druid.query.dimension.ListFilteredDimensionSpec)46 RegexFilteredDimensionSpec (io.druid.query.dimension.RegexFilteredDimensionSpec)46 Interval (org.joda.time.Interval)46 AggregatorFactory (io.druid.query.aggregation.AggregatorFactory)38 DateTime (org.joda.time.DateTime)37 CountAggregatorFactory (io.druid.query.aggregation.CountAggregatorFactory)36 Result (io.druid.query.Result)30 Period (org.joda.time.Period)26 DoubleSumAggregatorFactory (io.druid.query.aggregation.DoubleSumAggregatorFactory)25 PeriodGranularity (io.druid.java.util.common.granularity.PeriodGranularity)24 LookupExtractionFn (io.druid.query.lookup.LookupExtractionFn)24 QueryRunner (io.druid.query.QueryRunner)22 HashMap (java.util.HashMap)22 FinalizeResultsQueryRunner (io.druid.query.FinalizeResultsQueryRunner)21