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

Example 41 with DimensionsSpec

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

the class InputRowSerdeTest method testSerde.

@Test
public void testSerde() {
    // Prepare the mocks & set close() call count expectation to 1
    final Aggregator mockedAggregator = EasyMock.createMock(DoubleSumAggregator.class);
    EasyMock.expect(mockedAggregator.isNull()).andReturn(false).times(1);
    EasyMock.expect(mockedAggregator.getDouble()).andReturn(0d).times(1);
    mockedAggregator.aggregate();
    EasyMock.expectLastCall().times(1);
    mockedAggregator.close();
    EasyMock.expectLastCall().times(1);
    EasyMock.replay(mockedAggregator);
    final Aggregator mockedNullAggregator = EasyMock.createMock(DoubleSumAggregator.class);
    EasyMock.expect(mockedNullAggregator.isNull()).andReturn(true).times(1);
    mockedNullAggregator.aggregate();
    EasyMock.expectLastCall().times(1);
    mockedNullAggregator.close();
    EasyMock.expectLastCall().times(1);
    EasyMock.replay(mockedNullAggregator);
    final AggregatorFactory mockedAggregatorFactory = EasyMock.createMock(AggregatorFactory.class);
    EasyMock.expect(mockedAggregatorFactory.factorize(EasyMock.anyObject(ColumnSelectorFactory.class))).andReturn(mockedAggregator);
    EasyMock.expect(mockedAggregatorFactory.getIntermediateType()).andReturn(ColumnType.DOUBLE).anyTimes();
    EasyMock.expect(mockedAggregatorFactory.getName()).andReturn("mockedAggregator").anyTimes();
    final AggregatorFactory mockedNullAggregatorFactory = EasyMock.createMock(AggregatorFactory.class);
    EasyMock.expect(mockedNullAggregatorFactory.factorize(EasyMock.anyObject(ColumnSelectorFactory.class))).andReturn(mockedNullAggregator);
    EasyMock.expect(mockedNullAggregatorFactory.getName()).andReturn("mockedNullAggregator").anyTimes();
    EasyMock.expect(mockedNullAggregatorFactory.getIntermediateType()).andReturn(ColumnType.DOUBLE).anyTimes();
    EasyMock.replay(mockedAggregatorFactory, mockedNullAggregatorFactory);
    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"), mockedAggregatorFactory, mockedNullAggregatorFactory };
    DimensionsSpec dimensionsSpec = new DimensionsSpec(Arrays.asList(new StringDimensionSchema("d1"), new StringDimensionSchema("d2"), new LongDimensionSchema("d3"), new FloatDimensionSchema("d4"), new DoubleDimensionSchema("d5")));
    byte[] data = InputRowSerde.toBytes(InputRowSerde.getTypeHelperMap(dimensionsSpec), in, aggregatorFactories).getSerializedRow();
    InputRow out = InputRowSerde.fromBytes(InputRowSerde.getTypeHelperMap(dimensionsSpec), data, aggregatorFactories);
    Assert.assertEquals(timestamp, out.getTimestampFromEpoch());
    Assert.assertEquals(dims, out.getDimensions());
    Assert.assertEquals(Collections.emptyList(), out.getDimension("dim_non_existing"));
    Assert.assertEquals(ImmutableList.of("d1v"), out.getDimension("d1"));
    Assert.assertEquals(ImmutableList.of("d2v1", "d2v2"), out.getDimension("d2"));
    Assert.assertEquals(200L, out.getRaw("d3"));
    Assert.assertEquals(300.1f, out.getRaw("d4"));
    Assert.assertEquals(400.5d, out.getRaw("d5"));
    Assert.assertEquals(NullHandling.defaultDoubleValue(), out.getMetric("agg_non_existing"));
    Assert.assertEquals(5.0f, out.getMetric("m1out").floatValue(), 0.00001);
    Assert.assertEquals(100L, out.getMetric("m2out"));
    Assert.assertEquals(1, ((HyperLogLogCollector) out.getRaw("m3out")).estimateCardinality(), 0.001);
    Assert.assertEquals(NullHandling.defaultLongValue(), out.getMetric("unparseable"));
    EasyMock.verify(mockedAggregator);
    EasyMock.verify(mockedNullAggregator);
}

Example 42 with DimensionsSpec

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

the class CompactionTask method createDimensionsSpec.

private static DimensionsSpec createDimensionsSpec(List<NonnullPair<QueryableIndex, DataSegment>> queryableIndices) {
    final BiMap<String, Integer> uniqueDims = HashBiMap.create();
    final Map<String, DimensionSchema> dimensionSchemaMap = new HashMap<>();
    // Here, we try to retain the order of dimensions as they were specified since the order of dimensions may be
    // optimized for performance.
    // Dimensions are extracted from the recent segments to olders because recent segments are likely to be queried more
    // frequently, and thus the performance should be optimized for recent ones rather than old ones.
    // sort timelineSegments in order of interval, see https://github.com/apache/druid/pull/9905
    queryableIndices.sort((o1, o2) -> Comparators.intervalsByStartThenEnd().compare(o1.rhs.getInterval(), o2.rhs.getInterval()));
    int index = 0;
    for (NonnullPair<QueryableIndex, DataSegment> pair : Lists.reverse(queryableIndices)) {
        final QueryableIndex queryableIndex = pair.lhs;
        final Map<String, DimensionHandler> dimensionHandlerMap = queryableIndex.getDimensionHandlers();
        for (String dimension : queryableIndex.getAvailableDimensions()) {
            final ColumnHolder columnHolder = Preconditions.checkNotNull(queryableIndex.getColumnHolder(dimension), "Cannot find column for dimension[%s]", dimension);
            if (!uniqueDims.containsKey(dimension)) {
                final DimensionHandler dimensionHandler = Preconditions.checkNotNull(dimensionHandlerMap.get(dimension), "Cannot find dimensionHandler for dimension[%s]", dimension);
                uniqueDims.put(dimension, index++);
                dimensionSchemaMap.put(dimension, createDimensionSchema(dimension, columnHolder.getCapabilities(), dimensionHandler.getMultivalueHandling()));
            }
        }
    }
    final BiMap<Integer, String> orderedDims = uniqueDims.inverse();
    final List<DimensionSchema> dimensionSchemas = IntStream.range(0, orderedDims.size()).mapToObj(i -> {
        final String dimName = orderedDims.get(i);
        return Preconditions.checkNotNull(dimensionSchemaMap.get(dimName), "Cannot find dimension[%s] from dimensionSchemaMap", dimName);
    }).collect(Collectors.toList());
    return new DimensionsSpec(dimensionSchemas);
}

Example 43 with DimensionsSpec

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

the class CompactionTask method createDataSchema.

private static DataSchema createDataSchema(String dataSource, List<NonnullPair<QueryableIndex, DataSegment>> queryableIndexAndSegments, @Nullable DimensionsSpec dimensionsSpec, @Nullable ClientCompactionTaskTransformSpec transformSpec, @Nullable AggregatorFactory[] metricsSpec, @Nonnull ClientCompactionTaskGranularitySpec granularitySpec) {
    // check index metadata &
    // Decide which values to propagate (i.e. carry over) for rollup & queryGranularity
    final SettableSupplier<Boolean> rollup = new SettableSupplier<>();
    final SettableSupplier<Granularity> queryGranularity = new SettableSupplier<>();
    decideRollupAndQueryGranularityCarryOver(rollup, queryGranularity, queryableIndexAndSegments);
    final Interval totalInterval = JodaUtils.umbrellaInterval(queryableIndexAndSegments.stream().map(p -> p.rhs.getInterval()).collect(Collectors.toList()));
    final Granularity queryGranularityToUse;
    if (granularitySpec.getQueryGranularity() == null) {
        queryGranularityToUse = queryGranularity.get();
        log.info("Generate compaction task spec with segments original query granularity [%s]", queryGranularityToUse);
    } else {
        queryGranularityToUse = granularitySpec.getQueryGranularity();
        log.info("Generate compaction task spec with new query granularity overrided from input [%s]", queryGranularityToUse);
    }
    final GranularitySpec uniformGranularitySpec = new UniformGranularitySpec(Preconditions.checkNotNull(granularitySpec.getSegmentGranularity()), queryGranularityToUse, granularitySpec.isRollup() == null ? rollup.get() : granularitySpec.isRollup(), Collections.singletonList(totalInterval));
    // find unique dimensions
    final DimensionsSpec finalDimensionsSpec = dimensionsSpec == null ? createDimensionsSpec(queryableIndexAndSegments) : dimensionsSpec;
    final AggregatorFactory[] finalMetricsSpec = metricsSpec == null ? createMetricsSpec(queryableIndexAndSegments) : metricsSpec;
    return new DataSchema(dataSource, new TimestampSpec(ColumnHolder.TIME_COLUMN_NAME, "millis", null), finalDimensionsSpec, finalMetricsSpec, uniformGranularitySpec, transformSpec == null ? null : new TransformSpec(transformSpec.getFilter(), null));
}

Example 44 with DimensionsSpec

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

the class AppenderatorDriverRealtimeIndexTaskTest method makeRealtimeTask.

private AppenderatorDriverRealtimeIndexTask makeRealtimeTask(final String taskId, final TransformSpec transformSpec, final boolean reportParseExceptions, final long handoffTimeout, final Boolean logParseExceptions, final Integer maxParseExceptions, final Integer maxSavedParseExceptions, final Integer maxRowsPerSegment, final Long maxTotalRows) {
    DataSchema dataSchema = new DataSchema("test_ds", TestHelper.makeJsonMapper().convertValue(new MapInputRowParser(new TimeAndDimsParseSpec(new TimestampSpec("t", "auto", null), new DimensionsSpec(ImmutableList.of(new StringDimensionSchema("dim1"), new StringDimensionSchema("dim2"), new StringDimensionSchema("dim1t"), new LongDimensionSchema("dimLong"), new FloatDimensionSchema("dimFloat"))))), JacksonUtils.TYPE_REFERENCE_MAP_STRING_OBJECT), new AggregatorFactory[] { new CountAggregatorFactory("rows"), new LongSumAggregatorFactory("met1", "met1") }, new UniformGranularitySpec(Granularities.DAY, Granularities.NONE, null), transformSpec, OBJECT_MAPPER);
    RealtimeIOConfig realtimeIOConfig = new RealtimeIOConfig(new TestFirehoseFactory(), null);
    RealtimeAppenderatorTuningConfig tuningConfig = new RealtimeAppenderatorTuningConfig(null, 1000, null, null, maxRowsPerSegment, maxTotalRows, null, null, null, null, null, null, reportParseExceptions, handoffTimeout, null, null, logParseExceptions, maxParseExceptions, maxSavedParseExceptions);
    return new AppenderatorDriverRealtimeIndexTask(taskId, null, new RealtimeAppenderatorIngestionSpec(dataSchema, realtimeIOConfig, tuningConfig), null) {

        @Override
        protected boolean isFirehoseDrainableByClosing(FirehoseFactory firehoseFactory) {
            return true;
        }
    };
}

Example 45 with DimensionsSpec

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

the class CompactionTaskParallelRunTest method testRunParallelWithHashPartitioningMatchCompactionState.

@Test
public void testRunParallelWithHashPartitioningMatchCompactionState() throws Exception {
    // Hash partitioning is not supported with segment lock yet
    Assume.assumeFalse(lockGranularity == LockGranularity.SEGMENT);
    runIndexTask(null, true);
    final Builder builder = new Builder(DATA_SOURCE, getSegmentCacheManagerFactory(), RETRY_POLICY_FACTORY);
    final CompactionTask compactionTask = builder.inputSpec(new CompactionIntervalSpec(INTERVAL_TO_INDEX, null)).tuningConfig(newTuningConfig(new HashedPartitionsSpec(null, 3, null), 2, true)).build();
    final Set<DataSegment> compactedSegments = runTask(compactionTask);
    for (DataSegment segment : compactedSegments) {
        // Expect compaction state to exist as store compaction state by default
        Map<String, String> expectedLongSumMetric = new HashMap<>();
        expectedLongSumMetric.put("type", "longSum");
        expectedLongSumMetric.put("name", "val");
        expectedLongSumMetric.put("fieldName", "val");
        expectedLongSumMetric.put("expression", null);
        Assert.assertSame(HashBasedNumberedShardSpec.class, segment.getShardSpec().getClass());
        CompactionState expectedState = new CompactionState(new HashedPartitionsSpec(null, 3, null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("ts", "dim"))), ImmutableList.of(expectedLongSumMetric), null, compactionTask.getTuningConfig().getIndexSpec().asMap(getObjectMapper()), getObjectMapper().readValue(getObjectMapper().writeValueAsString(new UniformGranularitySpec(Granularities.HOUR, Granularities.MINUTE, true, ImmutableList.of(segment.getInterval()))), Map.class));
        Assert.assertEquals(expectedState, segment.getLastCompactionState());
    }
}