Examples with AggregatorFactory org.apache.druid.query.aggregation.AggregatorFactory used on opensource projects

Example 91 with AggregatorFactory

use of org.apache.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class GroupByStrategyV2 method processSubtotalsSpec.

@Override
public Sequence<ResultRow> processSubtotalsSpec(GroupByQuery query, GroupByQueryResource resource, Sequence<ResultRow> queryResult) {
    // How it works?
    // First we accumulate the result of top level base query aka queryResult arg inside a resultSupplierOne object.
    // Next for each subtotalSpec
    // If subtotalSpec is a prefix of top level dims then we iterate on rows in resultSupplierOne object which are still
    // sorted by subtotalSpec, stream merge them and return.
    // 
    // If subtotalSpec is not a prefix of top level dims then we create a resultSupplierTwo object filled with rows from
    // resultSupplierOne object with only dims from subtotalSpec. Then we iterate on rows in resultSupplierTwo object which are
    // of course sorted by subtotalSpec, stream merge them and return.
    // Keep a reference to resultSupplier outside the "try" so we can close it if something goes wrong
    // while creating the sequence.
    GroupByRowProcessor.ResultSupplier resultSupplierOne = null;
    try {
        // baseSubtotalQuery is the original query with dimensions and aggregators rewritten to apply to the *results*
        // rather than *inputs* of that query. It has its virtual columns and dim filter removed, because those only
        // make sense when applied to inputs. Finally, it has subtotalsSpec removed, since we'll be computing them
        // one-by-one soon enough.
        GroupByQuery baseSubtotalQuery = query.withDimensionSpecs(query.getDimensions().stream().map(dimSpec -> new DefaultDimensionSpec(dimSpec.getOutputName(), dimSpec.getOutputName(), dimSpec.getOutputType())).collect(Collectors.toList())).withAggregatorSpecs(query.getAggregatorSpecs().stream().map(AggregatorFactory::getCombiningFactory).collect(Collectors.toList())).withVirtualColumns(VirtualColumns.EMPTY).withDimFilter(null).withSubtotalsSpec(null).withOverriddenContext(ImmutableMap.of(GroupByQuery.CTX_TIMESTAMP_RESULT_FIELD, ""));
        resultSupplierOne = GroupByRowProcessor.process(baseSubtotalQuery, baseSubtotalQuery, queryResult, configSupplier.get(), resource, spillMapper, processingConfig.getTmpDir(), processingConfig.intermediateComputeSizeBytes());
        List<String> queryDimNames = baseSubtotalQuery.getDimensions().stream().map(DimensionSpec::getOutputName).collect(Collectors.toList());
        // Only needed to make LimitSpec.filterColumns(..) call later in case base query has a non default LimitSpec.
        Set<String> aggsAndPostAggs = null;
        if (!(baseSubtotalQuery.getLimitSpec() instanceof NoopLimitSpec)) {
            aggsAndPostAggs = getAggregatorAndPostAggregatorNames(baseSubtotalQuery);
        }
        List<List<String>> subtotals = query.getSubtotalsSpec();
        List<Sequence<ResultRow>> subtotalsResults = new ArrayList<>(subtotals.size());
        // Iterate through each subtotalSpec, build results for it and add to subtotalsResults
        for (List<String> subtotalSpec : subtotals) {
            final ImmutableSet<String> dimsInSubtotalSpec = ImmutableSet.copyOf(subtotalSpec);
            // Dimension spec including dimension name and output name
            final List<DimensionSpec> subTotalDimensionSpec = new ArrayList<>(dimsInSubtotalSpec.size());
            final List<DimensionSpec> dimensions = query.getDimensions();
            for (DimensionSpec dimensionSpec : dimensions) {
                if (dimsInSubtotalSpec.contains(dimensionSpec.getOutputName())) {
                    subTotalDimensionSpec.add(dimensionSpec);
                }
            }
            // Create appropriate LimitSpec for subtotal query
            LimitSpec subtotalQueryLimitSpec = NoopLimitSpec.instance();
            if (!(baseSubtotalQuery.getLimitSpec() instanceof NoopLimitSpec)) {
                Set<String> columns = new HashSet<>(aggsAndPostAggs);
                columns.addAll(subtotalSpec);
                subtotalQueryLimitSpec = baseSubtotalQuery.getLimitSpec().filterColumns(columns);
            }
            GroupByQuery subtotalQuery = baseSubtotalQuery.withLimitSpec(subtotalQueryLimitSpec);
            final GroupByRowProcessor.ResultSupplier resultSupplierOneFinal = resultSupplierOne;
            if (Utils.isPrefix(subtotalSpec, queryDimNames)) {
                // Since subtotalSpec is a prefix of base query dimensions, so results from base query are also sorted
                // by subtotalSpec as needed by stream merging.
                subtotalsResults.add(processSubtotalsResultAndOptionallyClose(() -> resultSupplierOneFinal, subTotalDimensionSpec, subtotalQuery, false));
            } else {
                // Since subtotalSpec is not a prefix of base query dimensions, so results from base query are not sorted
                // by subtotalSpec. So we first add the result of base query into another resultSupplier which are sorted
                // by subtotalSpec and then stream merge them.
                // Also note, we can't create the ResultSupplier eagerly here or as we don't want to eagerly allocate
                // merge buffers for processing subtotal.
                Supplier<GroupByRowProcessor.ResultSupplier> resultSupplierTwo = () -> GroupByRowProcessor.process(baseSubtotalQuery, subtotalQuery, resultSupplierOneFinal.results(subTotalDimensionSpec), configSupplier.get(), resource, spillMapper, processingConfig.getTmpDir(), processingConfig.intermediateComputeSizeBytes());
                subtotalsResults.add(processSubtotalsResultAndOptionallyClose(resultSupplierTwo, subTotalDimensionSpec, subtotalQuery, true));
            }
        }
        return Sequences.withBaggage(query.postProcess(Sequences.concat(subtotalsResults)), // this will close resources allocated by resultSupplierOne after sequence read
        resultSupplierOne);
    } catch (Throwable e) {
        throw CloseableUtils.closeAndWrapInCatch(e, resultSupplierOne);
    }
}

Example 92 with AggregatorFactory

use of org.apache.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class SegmentMetadataQueryQueryToolChest method mergeAnalyses.

@VisibleForTesting
public static SegmentAnalysis mergeAnalyses(final SegmentAnalysis arg1, final SegmentAnalysis arg2, boolean lenientAggregatorMerge) {
    if (arg1 == null) {
        return arg2;
    }
    if (arg2 == null) {
        return arg1;
    }
    List<Interval> newIntervals = null;
    if (arg1.getIntervals() != null) {
        newIntervals = new ArrayList<>(arg1.getIntervals());
    }
    if (arg2.getIntervals() != null) {
        if (newIntervals == null) {
            newIntervals = new ArrayList<>();
        }
        newIntervals.addAll(arg2.getIntervals());
    }
    final Map<String, ColumnAnalysis> leftColumns = arg1.getColumns();
    final Map<String, ColumnAnalysis> rightColumns = arg2.getColumns();
    Map<String, ColumnAnalysis> columns = new TreeMap<>();
    Set<String> rightColumnNames = Sets.newHashSet(rightColumns.keySet());
    for (Map.Entry<String, ColumnAnalysis> entry : leftColumns.entrySet()) {
        final String columnName = entry.getKey();
        columns.put(columnName, entry.getValue().fold(rightColumns.get(columnName)));
        rightColumnNames.remove(columnName);
    }
    for (String columnName : rightColumnNames) {
        columns.put(columnName, rightColumns.get(columnName));
    }
    final Map<String, AggregatorFactory> aggregators = new HashMap<>();
    if (lenientAggregatorMerge) {
        // Merge each aggregator individually, ignoring nulls
        for (SegmentAnalysis analysis : ImmutableList.of(arg1, arg2)) {
            if (analysis.getAggregators() != null) {
                for (Map.Entry<String, AggregatorFactory> entry : analysis.getAggregators().entrySet()) {
                    final String aggregatorName = entry.getKey();
                    final AggregatorFactory aggregator = entry.getValue();
                    AggregatorFactory merged = aggregators.get(aggregatorName);
                    if (merged != null) {
                        try {
                            merged = merged.getMergingFactory(aggregator);
                        } catch (AggregatorFactoryNotMergeableException e) {
                            merged = null;
                        }
                    } else {
                        merged = aggregator;
                    }
                    aggregators.put(aggregatorName, merged);
                }
            }
        }
    } else {
        final AggregatorFactory[] aggs1 = arg1.getAggregators() != null ? arg1.getAggregators().values().toArray(new AggregatorFactory[0]) : null;
        final AggregatorFactory[] aggs2 = arg2.getAggregators() != null ? arg2.getAggregators().values().toArray(new AggregatorFactory[0]) : null;
        final AggregatorFactory[] merged = AggregatorFactory.mergeAggregators(Arrays.asList(aggs1, aggs2));
        if (merged != null) {
            for (AggregatorFactory aggregator : merged) {
                aggregators.put(aggregator.getName(), aggregator);
            }
        }
    }
    final TimestampSpec timestampSpec = TimestampSpec.mergeTimestampSpec(Lists.newArrayList(arg1.getTimestampSpec(), arg2.getTimestampSpec()));
    final Granularity queryGranularity = Granularity.mergeGranularities(Lists.newArrayList(arg1.getQueryGranularity(), arg2.getQueryGranularity()));
    final String mergedId;
    if (arg1.getId() != null && arg2.getId() != null && arg1.getId().equals(arg2.getId())) {
        mergedId = arg1.getId();
    } else {
        mergedId = "merged";
    }
    final Boolean rollup;
    if (arg1.isRollup() != null && arg2.isRollup() != null && arg1.isRollup().equals(arg2.isRollup())) {
        rollup = arg1.isRollup();
    } else {
        rollup = null;
    }
    return new SegmentAnalysis(mergedId, newIntervals, columns, arg1.getSize() + arg2.getSize(), arg1.getNumRows() + arg2.getNumRows(), aggregators.isEmpty() ? null : aggregators, timestampSpec, queryGranularity, rollup);
}

Example 93 with AggregatorFactory

use of org.apache.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class SegmentMetadataQueryRunnerFactory method createRunner.

@Override
public QueryRunner<SegmentAnalysis> createRunner(final Segment segment) {
    return new QueryRunner<SegmentAnalysis>() {

        @Override
        public Sequence<SegmentAnalysis> run(QueryPlus<SegmentAnalysis> inQ, ResponseContext responseContext) {
            SegmentMetadataQuery updatedQuery = ((SegmentMetadataQuery) inQ.getQuery()).withFinalizedAnalysisTypes(toolChest.getConfig());
            final SegmentAnalyzer analyzer = new SegmentAnalyzer(updatedQuery.getAnalysisTypes());
            final Map<String, ColumnAnalysis> analyzedColumns = analyzer.analyze(segment);
            final long numRows = analyzer.numRows(segment);
            long totalSize = 0;
            if (analyzer.analyzingSize()) {
                // Initialize with the size of the whitespace, 1 byte per
                totalSize = analyzedColumns.size() * numRows;
            }
            Map<String, ColumnAnalysis> columns = new TreeMap<>();
            ColumnIncluderator includerator = updatedQuery.getToInclude();
            for (Map.Entry<String, ColumnAnalysis> entry : analyzedColumns.entrySet()) {
                final String columnName = entry.getKey();
                final ColumnAnalysis column = entry.getValue();
                if (!column.isError()) {
                    totalSize += column.getSize();
                }
                if (includerator.include(columnName)) {
                    columns.put(columnName, column);
                }
            }
            List<Interval> retIntervals = updatedQuery.analyzingInterval() ? Collections.singletonList(segment.getDataInterval()) : null;
            final Map<String, AggregatorFactory> aggregators;
            Metadata metadata = null;
            if (updatedQuery.hasAggregators()) {
                metadata = segment.asStorageAdapter().getMetadata();
                if (metadata != null && metadata.getAggregators() != null) {
                    aggregators = new HashMap<>();
                    for (AggregatorFactory aggregator : metadata.getAggregators()) {
                        aggregators.put(aggregator.getName(), aggregator);
                    }
                } else {
                    aggregators = null;
                }
            } else {
                aggregators = null;
            }
            final TimestampSpec timestampSpec;
            if (updatedQuery.hasTimestampSpec()) {
                if (metadata == null) {
                    metadata = segment.asStorageAdapter().getMetadata();
                }
                timestampSpec = metadata != null ? metadata.getTimestampSpec() : null;
            } else {
                timestampSpec = null;
            }
            final Granularity queryGranularity;
            if (updatedQuery.hasQueryGranularity()) {
                if (metadata == null) {
                    metadata = segment.asStorageAdapter().getMetadata();
                }
                queryGranularity = metadata != null ? metadata.getQueryGranularity() : null;
            } else {
                queryGranularity = null;
            }
            Boolean rollup = null;
            if (updatedQuery.hasRollup()) {
                if (metadata == null) {
                    metadata = segment.asStorageAdapter().getMetadata();
                }
                rollup = metadata != null ? metadata.isRollup() : null;
                if (rollup == null) {
                    // in this case, this segment is built before no-rollup function is coded,
                    // thus it is built with rollup
                    rollup = Boolean.TRUE;
                }
            }
            return Sequences.simple(Collections.singletonList(new SegmentAnalysis(segment.getId().toString(), retIntervals, columns, totalSize, numRows, aggregators, timestampSpec, queryGranularity, rollup)));
        }
    };
}

Example 94 with AggregatorFactory

use of org.apache.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class TimeseriesQueryQueryToolChest method getCacheStrategy.

@Override
public CacheStrategy<Result<TimeseriesResultValue>, Object, TimeseriesQuery> getCacheStrategy(final TimeseriesQuery query) {
    return new CacheStrategy<Result<TimeseriesResultValue>, Object, TimeseriesQuery>() {

        private final List<AggregatorFactory> aggs = query.getAggregatorSpecs();

        @Override
        public boolean isCacheable(TimeseriesQuery query, boolean willMergeRunners) {
            return true;
        }

        @Override
        public byte[] computeCacheKey(TimeseriesQuery query) {
            return new CacheKeyBuilder(TIMESERIES_QUERY).appendBoolean(query.isDescending()).appendBoolean(query.isSkipEmptyBuckets()).appendCacheable(query.getGranularity()).appendCacheable(query.getDimensionsFilter()).appendCacheables(query.getAggregatorSpecs()).appendCacheable(query.getVirtualColumns()).appendInt(query.getLimit()).build();
        }

        @Override
        public byte[] computeResultLevelCacheKey(TimeseriesQuery query) {
            final CacheKeyBuilder builder = new CacheKeyBuilder(TIMESERIES_QUERY).appendBoolean(query.isDescending()).appendBoolean(query.isSkipEmptyBuckets()).appendCacheable(query.getGranularity()).appendCacheable(query.getDimensionsFilter()).appendCacheables(query.getAggregatorSpecs()).appendCacheable(query.getVirtualColumns()).appendCacheables(query.getPostAggregatorSpecs()).appendInt(query.getLimit()).appendString(query.getTimestampResultField()).appendBoolean(query.isGrandTotal());
            return builder.build();
        }

        @Override
        public TypeReference<Object> getCacheObjectClazz() {
            return OBJECT_TYPE_REFERENCE;
        }

        @Override
        public Function<Result<TimeseriesResultValue>, Object> prepareForCache(boolean isResultLevelCache) {
            return input -> {
                TimeseriesResultValue results = input.getValue();
                final List<Object> retVal = Lists.newArrayListWithCapacity(1 + aggs.size());
                // Timestamp can be null if grandTotal is true.
                if (isResultLevelCache) {
                    retVal.add(input.getTimestamp() == null ? null : input.getTimestamp().getMillis());
                } else {
                    retVal.add(Preconditions.checkNotNull(input.getTimestamp(), "timestamp of input[%s]", input).getMillis());
                }
                for (AggregatorFactory agg : aggs) {
                    retVal.add(results.getMetric(agg.getName()));
                }
                if (isResultLevelCache) {
                    for (PostAggregator postAgg : query.getPostAggregatorSpecs()) {
                        retVal.add(results.getMetric(postAgg.getName()));
                    }
                }
                return retVal;
            };
        }

        @Override
        public Function<Object, Result<TimeseriesResultValue>> pullFromCache(boolean isResultLevelCache) {
            return new Function<Object, Result<TimeseriesResultValue>>() {

                private final Granularity granularity = query.getGranularity();

                @Override
                public Result<TimeseriesResultValue> apply(Object input) {
                    List<Object> results = (List<Object>) input;
                    final Map<String, Object> retVal = Maps.newLinkedHashMap();
                    Iterator<Object> resultIter = results.iterator();
                    final Number timestampNumber = (Number) resultIter.next();
                    final DateTime timestamp;
                    if (isResultLevelCache) {
                        timestamp = timestampNumber == null ? null : granularity.toDateTime(timestampNumber.longValue());
                    } else {
                        timestamp = granularity.toDateTime(Preconditions.checkNotNull(timestampNumber, "timestamp").longValue());
                    }
                    CacheStrategy.fetchAggregatorsFromCache(aggs, resultIter, isResultLevelCache, (aggName, aggPosition, aggValueObject) -> {
                        retVal.put(aggName, aggValueObject);
                    });
                    if (isResultLevelCache) {
                        Iterator<PostAggregator> postItr = query.getPostAggregatorSpecs().iterator();
                        while (postItr.hasNext() && resultIter.hasNext()) {
                            retVal.put(postItr.next().getName(), resultIter.next());
                        }
                    }
                    return new Result<>(timestamp, new TimeseriesResultValue(retVal));
                }
            };
        }
    };
}

Example 95 with AggregatorFactory

use of org.apache.druid.query.aggregation.AggregatorFactory in project druid by druid-io.

the class QueriesTest method testVerifyAggregationsMissingVal.

@Test
public void testVerifyAggregationsMissingVal() {
    List<AggregatorFactory> aggFactories = Arrays.asList(new CountAggregatorFactory("count"), new DoubleSumAggregatorFactory("idx", "index"), new DoubleSumAggregatorFactory("rev", "revenue"));
    List<PostAggregator> postAggs = Collections.singletonList(new ArithmeticPostAggregator("addStuff", "+", Arrays.asList(new FieldAccessPostAggregator("idx", "idx2"), new FieldAccessPostAggregator("count", "count"))));
    boolean exceptionOccured = false;
    try {
        Queries.prepareAggregations(ImmutableList.of(), aggFactories, postAggs);
    } catch (IllegalArgumentException e) {
        exceptionOccured = true;
    }
    Assert.assertTrue(exceptionOccured);
}