Examples with ResultRow org.apache.druid.query.groupby.ResultRow used on opensource projects

Example 36 with ResultRow

use of org.apache.druid.query.groupby.ResultRow in project druid by druid-io.

the class DimFilterHavingSpec method eval.

@Override
public boolean eval(final ResultRow row) {
    int oldEvalCount = evalCount;
    evalCount++;
    if (finalize && !finalizers.isEmpty()) {
        // Create finalized copy.
        final ResultRow finalizedCopy = row.copy();
        for (Int2ObjectMap.Entry<Function<Object, Object>> entry : finalizers.int2ObjectEntrySet()) {
            finalizedCopy.set(entry.getIntKey(), entry.getValue().apply(row.get(entry.getIntKey())));
        }
        rowSupplier.set(finalizedCopy);
    } else {
        rowSupplier.set(row);
    }
    final boolean retVal = matcher.matches();
    if (evalCount != oldEvalCount + 1) {
        // Oops, someone was using this from two different threads, bad caller.
        throw new IllegalStateException("concurrent 'eval' calls not permitted!");
    }
    return retVal;
}

Example 37 with ResultRow

use of org.apache.druid.query.groupby.ResultRow in project druid by druid-io.

the class MaterializedViewQueryQueryToolChestTest method testDecorateObjectMapper.

@Test
public void testDecorateObjectMapper() throws IOException {
    GroupByQuery realQuery = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("quality", "alias")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).setContext(ImmutableMap.of(GroupByQueryConfig.CTX_KEY_ARRAY_RESULT_ROWS, false)).build();
    QueryToolChest queryToolChest = new MaterializedViewQueryQueryToolChest(new MapQueryToolChestWarehouse(ImmutableMap.<Class<? extends Query>, QueryToolChest>builder().put(GroupByQuery.class, new GroupByQueryQueryToolChest(null)).build()));
    ObjectMapper objectMapper = queryToolChest.decorateObjectMapper(JSON_MAPPER, realQuery);
    List<ResultRow> results = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow(realQuery, "2011-04-01", "alias", "automotive", "rows", 1L, "idx", 135L), GroupByQueryRunnerTestHelper.createExpectedRow(realQuery, "2011-04-01", "alias", "business", "rows", 1L, "idx", 118L));
    List<MapBasedRow> expectedResults = results.stream().map(resultRow -> resultRow.toMapBasedRow(realQuery)).collect(Collectors.toList());
    Assert.assertEquals("decorate-object-mapper", JSON_MAPPER.writerFor(new TypeReference<List<MapBasedRow>>() {
    }).writeValueAsString(expectedResults), objectMapper.writeValueAsString(results));
}

Example 38 with ResultRow

use of org.apache.druid.query.groupby.ResultRow in project druid by druid-io.

the class MomentsSketchAggregatorTest method buildingSketchesAtQueryTime.

@Test
public void buildingSketchesAtQueryTime() throws Exception {
    Sequence<ResultRow> seq = helper.createIndexAndRunQueryOnSegment(new File(this.getClass().getClassLoader().getResource("doubles_build_data.tsv").getFile()), String.join("\n", "{", "  \"type\": \"string\",", "  \"parseSpec\": {", "    \"format\": \"tsv\",", "    \"timestampSpec\": {\"column\": \"timestamp\", \"format\": \"yyyyMMddHH\"},", "    \"dimensionsSpec\": {", "      \"dimensions\": [ \"product\", {\"name\":\"valueWithNulls\", \"type\":\"double\"}],", "      \"dimensionExclusions\": [\"sequenceNumber\"],", "      \"spatialDimensions\": []", "    },", "    \"columns\": [\"timestamp\", \"sequenceNumber\", \"product\", \"value\", \"valueWithNulls\"]", "  }", "}"), "[{\"type\": \"doubleSum\", \"name\": \"value\", \"fieldName\": \"value\"}]", // minTimestamp
    0, Granularities.NONE, // maxRowCount
    10, String.join("\n", "{", "  \"queryType\": \"groupBy\",", "  \"dataSource\": \"test_datasource\",", "  \"granularity\": \"ALL\",", "  \"dimensions\": [],", "  \"aggregations\": [", "    {\"type\": \"momentSketch\", \"name\": \"sketch\", \"fieldName\": \"value\", \"k\": 10},", "    {\"type\": \"momentSketch\", \"name\": \"sketchWithNulls\", \"fieldName\": \"valueWithNulls\", \"k\": 10}", "  ],", "  \"intervals\": [\"2016-01-01T00:00:00.000Z/2016-01-31T00:00:00.000Z\"]", "}"));
    List<ResultRow> results = seq.toList();
    Assert.assertEquals(1, results.size());
    ResultRow row = results.get(0);
    // "sketch"
    MomentSketchWrapper sketchObject = (MomentSketchWrapper) row.get(0);
    // 385 total products since roll-up limited by valueWithNulls column
    Assert.assertEquals(385.0, sketchObject.getPowerSums()[0], 1e-10);
    // "sketchWithNulls"
    MomentSketchWrapper sketchObjectWithNulls = (MomentSketchWrapper) row.get(1);
    // in default mode, all 385 rows have a number value so will be computed, but only 377 rows have actual values in
    // sql null mode
    Assert.assertEquals(hasNulls ? 377.0 : 385.0, sketchObjectWithNulls.getPowerSums()[0], 1e-10);
}

Example 39 with ResultRow

use of org.apache.druid.query.groupby.ResultRow in project druid by druid-io.

the class MovingAverageQueryRunner method run.

@Override
public Sequence<Row> run(QueryPlus<Row> query, ResponseContext responseContext) {
    MovingAverageQuery maq = (MovingAverageQuery) query.getQuery();
    List<Interval> intervals;
    final Period period;
    // Get the largest bucket from the list of averagers
    Optional<Integer> opt = maq.getAveragerSpecs().stream().map(AveragerFactory::getNumBuckets).max(Integer::compare);
    int buckets = opt.orElse(0);
    // Extend the interval beginning by specified bucket - 1
    if (maq.getGranularity() instanceof PeriodGranularity) {
        period = ((PeriodGranularity) maq.getGranularity()).getPeriod();
        int offset = buckets <= 0 ? 0 : (1 - buckets);
        intervals = maq.getIntervals().stream().map(i -> new Interval(i.getStart().withPeriodAdded(period, offset), i.getEnd())).collect(Collectors.toList());
    } else {
        throw new ISE("Only PeriodGranulaity is supported for movingAverage queries");
    }
    Sequence<Row> resultsSeq;
    DataSource dataSource = maq.getDataSource();
    if (maq.getDimensions() != null && !maq.getDimensions().isEmpty() && (dataSource instanceof TableDataSource || dataSource instanceof UnionDataSource || dataSource instanceof QueryDataSource)) {
        // build groupBy query from movingAverage query
        GroupByQuery.Builder builder = GroupByQuery.builder().setDataSource(dataSource).setInterval(intervals).setDimFilter(maq.getFilter()).setGranularity(maq.getGranularity()).setDimensions(maq.getDimensions()).setAggregatorSpecs(maq.getAggregatorSpecs()).setPostAggregatorSpecs(maq.getPostAggregatorSpecs()).setContext(maq.getContext());
        GroupByQuery gbq = builder.build();
        ResponseContext gbqResponseContext = ResponseContext.createEmpty();
        gbqResponseContext.merge(responseContext);
        gbqResponseContext.putQueryFailDeadlineMs(System.currentTimeMillis() + QueryContexts.getTimeout(gbq));
        Sequence<ResultRow> results = gbq.getRunner(walker).run(QueryPlus.wrap(gbq), gbqResponseContext);
        try {
            // use localhost for remote address
            requestLogger.logNativeQuery(RequestLogLine.forNative(gbq, DateTimes.nowUtc(), "127.0.0.1", new QueryStats(ImmutableMap.of("query/time", 0, "query/bytes", 0, "success", true))));
        } catch (Exception e) {
            throw Throwables.propagate(e);
        }
        resultsSeq = results.map(row -> row.toMapBasedRow(gbq));
    } else {
        // no dimensions, so optimize this as a TimeSeries
        TimeseriesQuery tsq = new TimeseriesQuery(dataSource, new MultipleIntervalSegmentSpec(intervals), false, null, maq.getFilter(), maq.getGranularity(), maq.getAggregatorSpecs(), maq.getPostAggregatorSpecs(), 0, maq.getContext());
        ResponseContext tsqResponseContext = ResponseContext.createEmpty();
        tsqResponseContext.merge(responseContext);
        tsqResponseContext.putQueryFailDeadlineMs(System.currentTimeMillis() + QueryContexts.getTimeout(tsq));
        Sequence<Result<TimeseriesResultValue>> results = tsq.getRunner(walker).run(QueryPlus.wrap(tsq), tsqResponseContext);
        try {
            // use localhost for remote address
            requestLogger.logNativeQuery(RequestLogLine.forNative(tsq, DateTimes.nowUtc(), "127.0.0.1", new QueryStats(ImmutableMap.of("query/time", 0, "query/bytes", 0, "success", true))));
        } catch (Exception e) {
            throw Throwables.propagate(e);
        }
        resultsSeq = Sequences.map(results, new TimeseriesResultToRow());
    }
    // Process into period buckets
    Sequence<RowBucket> bucketedMovingAvgResults = Sequences.simple(new RowBucketIterable(resultsSeq, intervals, period));
    // Apply the windows analysis functions
    Sequence<Row> movingAvgResults = Sequences.simple(new MovingAverageIterable(bucketedMovingAvgResults, maq.getDimensions(), maq.getAveragerSpecs(), maq.getPostAggregatorSpecs(), maq.getAggregatorSpecs()));
    // Apply any postAveragers
    Sequence<Row> movingAvgResultsWithPostAveragers = Sequences.map(movingAvgResults, new PostAveragerAggregatorCalculator(maq));
    // remove rows outside the reporting window
    List<Interval> reportingIntervals = maq.getIntervals();
    movingAvgResults = Sequences.filter(movingAvgResultsWithPostAveragers, row -> reportingIntervals.stream().anyMatch(i -> i.contains(row.getTimestamp())));
    // Apply any having, sorting, and limits
    movingAvgResults = maq.applyLimit(movingAvgResults);
    return movingAvgResults;
}

Example 40 with ResultRow

use of org.apache.druid.query.groupby.ResultRow in project druid by druid-io.

the class TDigestSketchAggregatorTest method buildingSketchesAtQueryTime.

@Test
public void buildingSketchesAtQueryTime() throws Exception {
    Sequence<ResultRow> seq = helper.createIndexAndRunQueryOnSegment(new File(this.getClass().getClassLoader().getResource("doubles_build_data.tsv").getFile()), String.join("\n", "{", "  \"type\": \"string\",", "  \"parseSpec\": {", "    \"format\": \"tsv\",", "    \"timestampSpec\": {\"column\": \"timestamp\", \"format\": \"yyyyMMddHH\"},", "    \"dimensionsSpec\": {", "      \"dimensions\": [\"sequenceNumber\", \"product\"],", "      \"dimensionExclusions\": [],", "      \"spatialDimensions\": []", "    },", "    \"columns\": [\"timestamp\", \"sequenceNumber\", \"product\", \"value\"]", "  }", "}"), "[{\"type\": \"doubleSum\", \"name\": \"value\", \"fieldName\": \"value\"}]", // minTimestamp
    0, Granularities.NONE, // maxRowCount
    10, String.join("\n", "{", "  \"queryType\": \"groupBy\",", "  \"dataSource\": \"test_datasource\",", "  \"granularity\": \"ALL\",", "  \"dimensions\": [],", "  \"aggregations\": [", "    {\"type\": \"tDigestSketch\", \"name\": \"sketch\", \"fieldName\": \"value\", \"compression\": 200}", "  ],", "  \"postAggregations\": [", "    {\"type\": \"quantilesFromTDigestSketch\", \"name\": \"quantiles\", \"fractions\": [0, 0.5, 1], \"field\": {\"type\": \"fieldAccess\", \"fieldName\": \"sketch\"}}", "  ],", "  \"intervals\": [\"2016-01-01T00:00:00.000Z/2016-01-31T00:00:00.000Z\"]", "}"));
    List<ResultRow> results = seq.toList();
    Assert.assertEquals(1, results.size());
    ResultRow row = results.get(0);
    // post agg
    // "quantiles"
    Object quantilesObject = row.get(1);
    Assert.assertTrue(quantilesObject instanceof double[]);
    double[] quantiles = (double[]) quantilesObject;
    // min value
    Assert.assertEquals(NullHandling.replaceWithDefault() ? 0.0 : 0.001, quantiles[0], 0.0006);
    // median value
    Assert.assertEquals(NullHandling.replaceWithDefault() ? 0.35 : 0.5, quantiles[1], 0.05);
    // max value
    Assert.assertEquals(1, quantiles[2], 0.05);
}