Examples with DocValueFormat org.elasticsearch.search.DocValueFormat used on opensource projects

Example 6 with DocValueFormat

use of org.elasticsearch.search.DocValueFormat in project elasticsearch by elastic.

the class StringTermsTests method createTestInstance.

@Override
protected InternalTerms<?, ?> createTestInstance(String name, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData) {
    Terms.Order order = Terms.Order.count(false);
    long minDocCount = 1;
    int requiredSize = 3;
    int shardSize = requiredSize + 2;
    DocValueFormat format = DocValueFormat.RAW;
    boolean showTermDocCountError = false;
    long docCountError = -1;
    long otherDocCount = 0;
    List<StringTerms.Bucket> buckets = new ArrayList<>();
    final int numBuckets = randomInt(shardSize);
    Set<BytesRef> terms = new HashSet<>();
    for (int i = 0; i < numBuckets; ++i) {
        BytesRef term = randomValueOtherThanMany(b -> terms.add(b) == false, () -> new BytesRef(randomAsciiOfLength(10)));
        int docCount = randomIntBetween(1, 100);
        buckets.add(new StringTerms.Bucket(term, docCount, InternalAggregations.EMPTY, showTermDocCountError, docCountError, format));
    }
    return new StringTerms(name, order, requiredSize, minDocCount, pipelineAggregators, metaData, format, shardSize, showTermDocCountError, otherDocCount, buckets, docCountError);
}

Example 7 with DocValueFormat

use of org.elasticsearch.search.DocValueFormat in project elasticsearch by elastic.

the class SortBuilder method buildSort.

public static Optional<SortAndFormats> buildSort(List<SortBuilder<?>> sortBuilders, QueryShardContext context) throws IOException {
    List<SortField> sortFields = new ArrayList<>(sortBuilders.size());
    List<DocValueFormat> sortFormats = new ArrayList<>(sortBuilders.size());
    for (SortBuilder<?> builder : sortBuilders) {
        SortFieldAndFormat sf = builder.build(context);
        sortFields.add(sf.field);
        sortFormats.add(sf.format);
    }
    if (!sortFields.isEmpty()) {
        // optimize if we just sort on score non reversed, we don't really
        // need sorting
        boolean sort;
        if (sortFields.size() > 1) {
            sort = true;
        } else {
            SortField sortField = sortFields.get(0);
            if (sortField.getType() == SortField.Type.SCORE && !sortField.getReverse()) {
                sort = false;
            } else {
                sort = true;
            }
        }
        if (sort) {
            return Optional.of(new SortAndFormats(new Sort(sortFields.toArray(new SortField[sortFields.size()])), sortFormats.toArray(new DocValueFormat[sortFormats.size()])));
        }
    }
    return Optional.empty();
}

Example 8 with DocValueFormat

use of org.elasticsearch.search.DocValueFormat in project elasticsearch by elastic.

the class SignificantTermsAggregatorFactory method doCreateInternal.

@Override
protected Aggregator doCreateInternal(ValuesSource valuesSource, Aggregator parent, boolean collectsFromSingleBucket, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData) throws IOException {
    if (collectsFromSingleBucket == false) {
        return asMultiBucketAggregator(this, context, parent);
    }
    numberOfAggregatorsCreated++;
    BucketCountThresholds bucketCountThresholds = new BucketCountThresholds(this.bucketCountThresholds);
    if (bucketCountThresholds.getShardSize() == SignificantTermsAggregationBuilder.DEFAULT_BUCKET_COUNT_THRESHOLDS.getShardSize()) {
        // The user has not made a shardSize selection .
        // Use default heuristic to avoid any wrong-ranking caused by
        // distributed counting
        // but request double the usual amount.
        // We typically need more than the number of "top" terms requested
        // by other aggregations
        // as the significance algorithm is in less of a position to
        // down-select at shard-level -
        // some of the things we want to find have only one occurrence on
        // each shard and as
        // such are impossible to differentiate from non-significant terms
        // at that early stage.
        bucketCountThresholds.setShardSize(2 * BucketUtils.suggestShardSideQueueSize(bucketCountThresholds.getRequiredSize(), context.numberOfShards()));
    }
    if (valuesSource instanceof ValuesSource.Bytes) {
        ExecutionMode execution = null;
        if (executionHint != null) {
            execution = ExecutionMode.fromString(executionHint);
        }
        if (!(valuesSource instanceof ValuesSource.Bytes.WithOrdinals)) {
            execution = ExecutionMode.MAP;
        }
        if (execution == null) {
            if (Aggregator.descendsFromBucketAggregator(parent)) {
                execution = ExecutionMode.GLOBAL_ORDINALS_HASH;
            } else {
                execution = ExecutionMode.GLOBAL_ORDINALS;
            }
        }
        assert execution != null;
        DocValueFormat format = config.format();
        if ((includeExclude != null) && (includeExclude.isRegexBased()) && format != DocValueFormat.RAW) {
            throw new AggregationExecutionException("Aggregation [" + name + "] cannot support regular expression style include/exclude " + "settings as they can only be applied to string fields. Use an array of values for include/exclude clauses");
        }
        return execution.create(name, factories, valuesSource, format, bucketCountThresholds, includeExclude, context, parent, significanceHeuristic, this, pipelineAggregators, metaData);
    }
    if ((includeExclude != null) && (includeExclude.isRegexBased())) {
        throw new AggregationExecutionException("Aggregation [" + name + "] cannot support regular expression style include/exclude " + "settings as they can only be applied to string fields. Use an array of numeric values for include/exclude clauses used to filter numeric fields");
    }
    if (valuesSource instanceof ValuesSource.Numeric) {
        if (((ValuesSource.Numeric) valuesSource).isFloatingPoint()) {
            throw new UnsupportedOperationException("No support for examining floating point numerics");
        }
        IncludeExclude.LongFilter longFilter = null;
        if (includeExclude != null) {
            longFilter = includeExclude.convertToLongFilter(config.format());
        }
        return new SignificantLongTermsAggregator(name, factories, (ValuesSource.Numeric) valuesSource, config.format(), bucketCountThresholds, context, parent, significanceHeuristic, this, longFilter, pipelineAggregators, metaData);
    }
    throw new AggregationExecutionException("significant_terms aggregation cannot be applied to field [" + config.fieldContext().field() + "]. It can only be applied to numeric or string fields.");
}

Example 9 with DocValueFormat

use of org.elasticsearch.search.DocValueFormat in project elasticsearch by elastic.

the class DerivativePipelineAggregationBuilder method createInternal.

@Override
protected PipelineAggregator createInternal(Map<String, Object> metaData) throws IOException {
    DocValueFormat formatter;
    if (format != null) {
        formatter = new DocValueFormat.Decimal(format);
    } else {
        formatter = DocValueFormat.RAW;
    }
    Long xAxisUnits = null;
    if (units != null) {
        DateTimeUnit dateTimeUnit = DateHistogramAggregationBuilder.DATE_FIELD_UNITS.get(units);
        if (dateTimeUnit != null) {
            xAxisUnits = dateTimeUnit.field(DateTimeZone.UTC).getDurationField().getUnitMillis();
        } else {
            TimeValue timeValue = TimeValue.parseTimeValue(units, null, getClass().getSimpleName() + ".unit");
            if (timeValue != null) {
                xAxisUnits = timeValue.getMillis();
            }
        }
    }
    return new DerivativePipelineAggregator(name, bucketsPaths, formatter, gapPolicy, xAxisUnits, metaData);
}

Example 10 with DocValueFormat

use of org.elasticsearch.search.DocValueFormat in project elasticsearch by elastic.

the class QueryPhase method execute.

/**
     * In a package-private method so that it can be tested without having to
     * wire everything (mapperService, etc.)
     * @return whether the rescoring phase should be executed
     */
static boolean execute(SearchContext searchContext, final IndexSearcher searcher) throws QueryPhaseExecutionException {
    QuerySearchResult queryResult = searchContext.queryResult();
    queryResult.searchTimedOut(false);
    final boolean doProfile = searchContext.getProfilers() != null;
    final SearchType searchType = searchContext.searchType();
    boolean rescore = false;
    try {
        queryResult.from(searchContext.from());
        queryResult.size(searchContext.size());
        Query query = searchContext.query();
        final int totalNumDocs = searcher.getIndexReader().numDocs();
        int numDocs = Math.min(searchContext.from() + searchContext.size(), totalNumDocs);
        Collector collector;
        Callable<TopDocs> topDocsCallable;
        DocValueFormat[] sortValueFormats = new DocValueFormat[0];
        // already rewritten
        assert query == searcher.rewrite(query);
        if (searchContext.size() == 0) {
            // no matter what the value of from is
            final TotalHitCountCollector totalHitCountCollector = new TotalHitCountCollector();
            collector = totalHitCountCollector;
            if (searchContext.getProfilers() != null) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_COUNT, Collections.emptyList());
            }
            topDocsCallable = new Callable<TopDocs>() {

                @Override
                public TopDocs call() throws Exception {
                    return new TopDocs(totalHitCountCollector.getTotalHits(), Lucene.EMPTY_SCORE_DOCS, 0);
                }
            };
        } else {
            // Perhaps have a dedicated scroll phase?
            final ScrollContext scrollContext = searchContext.scrollContext();
            assert (scrollContext != null) == (searchContext.request().scroll() != null);
            final Collector topDocsCollector;
            ScoreDoc after = null;
            if (searchContext.request().scroll() != null) {
                numDocs = Math.min(searchContext.size(), totalNumDocs);
                after = scrollContext.lastEmittedDoc;
                if (returnsDocsInOrder(query, searchContext.sort())) {
                    if (scrollContext.totalHits == -1) {
                        // first round
                        assert scrollContext.lastEmittedDoc == null;
                    // there is not much that we can optimize here since we want to collect all
                    // documents in order to get the total number of hits
                    } else {
                        // skip to the desired doc and stop collecting after ${size} matches
                        if (scrollContext.lastEmittedDoc != null) {
                            BooleanQuery bq = new BooleanQuery.Builder().add(query, BooleanClause.Occur.MUST).add(new MinDocQuery(after.doc + 1), BooleanClause.Occur.FILTER).build();
                            query = bq;
                        }
                        searchContext.terminateAfter(numDocs);
                    }
                }
            } else {
                after = searchContext.searchAfter();
            }
            if (totalNumDocs == 0) {
                // top collectors don't like a size of 0
                numDocs = 1;
            }
            assert numDocs > 0;
            if (searchContext.collapse() == null) {
                if (searchContext.sort() != null) {
                    SortAndFormats sf = searchContext.sort();
                    topDocsCollector = TopFieldCollector.create(sf.sort, numDocs, (FieldDoc) after, true, searchContext.trackScores(), searchContext.trackScores());
                    sortValueFormats = sf.formats;
                } else {
                    rescore = !searchContext.rescore().isEmpty();
                    for (RescoreSearchContext rescoreContext : searchContext.rescore()) {
                        numDocs = Math.max(rescoreContext.window(), numDocs);
                    }
                    topDocsCollector = TopScoreDocCollector.create(numDocs, after);
                }
            } else {
                Sort sort = Sort.RELEVANCE;
                if (searchContext.sort() != null) {
                    sort = searchContext.sort().sort;
                }
                CollapseContext collapse = searchContext.collapse();
                topDocsCollector = collapse.createTopDocs(sort, numDocs, searchContext.trackScores());
                if (searchContext.sort() == null) {
                    sortValueFormats = new DocValueFormat[] { DocValueFormat.RAW };
                } else {
                    sortValueFormats = searchContext.sort().formats;
                }
            }
            collector = topDocsCollector;
            if (doProfile) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_TOP_HITS, Collections.emptyList());
            }
            topDocsCallable = new Callable<TopDocs>() {

                @Override
                public TopDocs call() throws Exception {
                    final TopDocs topDocs;
                    if (topDocsCollector instanceof TopDocsCollector) {
                        topDocs = ((TopDocsCollector<?>) topDocsCollector).topDocs();
                    } else if (topDocsCollector instanceof CollapsingTopDocsCollector) {
                        topDocs = ((CollapsingTopDocsCollector) topDocsCollector).getTopDocs();
                    } else {
                        throw new IllegalStateException("Unknown top docs collector " + topDocsCollector.getClass().getName());
                    }
                    if (scrollContext != null) {
                        if (scrollContext.totalHits == -1) {
                            // first round
                            scrollContext.totalHits = topDocs.totalHits;
                            scrollContext.maxScore = topDocs.getMaxScore();
                        } else {
                            // subsequent round: the total number of hits and
                            // the maximum score were computed on the first round
                            topDocs.totalHits = scrollContext.totalHits;
                            topDocs.setMaxScore(scrollContext.maxScore);
                        }
                        if (searchContext.request().numberOfShards() == 1) {
                            // if we fetch the document in the same roundtrip, we already know the last emitted doc
                            if (topDocs.scoreDocs.length > 0) {
                                // set the last emitted doc
                                scrollContext.lastEmittedDoc = topDocs.scoreDocs[topDocs.scoreDocs.length - 1];
                            }
                        }
                    }
                    return topDocs;
                }
            };
        }
        final boolean terminateAfterSet = searchContext.terminateAfter() != SearchContext.DEFAULT_TERMINATE_AFTER;
        if (terminateAfterSet) {
            final Collector child = collector;
            // throws Lucene.EarlyTerminationException when given count is reached
            collector = Lucene.wrapCountBasedEarlyTerminatingCollector(collector, searchContext.terminateAfter());
            if (doProfile) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_TERMINATE_AFTER_COUNT, Collections.singletonList((InternalProfileCollector) child));
            }
        }
        if (searchContext.parsedPostFilter() != null) {
            final Collector child = collector;
            // this will only get applied to the actual search collector and not
            // to any scoped collectors, also, it will only be applied to the main collector
            // since that is where the filter should only work
            final Weight filterWeight = searcher.createNormalizedWeight(searchContext.parsedPostFilter().query(), false);
            collector = new FilteredCollector(collector, filterWeight);
            if (doProfile) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_POST_FILTER, Collections.singletonList((InternalProfileCollector) child));
            }
        }
        // plug in additional collectors, like aggregations
        final List<Collector> subCollectors = new ArrayList<>();
        subCollectors.add(collector);
        subCollectors.addAll(searchContext.queryCollectors().values());
        collector = MultiCollector.wrap(subCollectors);
        if (doProfile && collector instanceof InternalProfileCollector == false) {
            // When there is a single collector to wrap, MultiCollector returns it
            // directly, so only wrap in the case that there are several sub collectors
            final List<InternalProfileCollector> children = new AbstractList<InternalProfileCollector>() {

                @Override
                public InternalProfileCollector get(int index) {
                    return (InternalProfileCollector) subCollectors.get(index);
                }

                @Override
                public int size() {
                    return subCollectors.size();
                }
            };
            collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_MULTI, children);
        }
        // apply the minimum score after multi collector so we filter aggs as well
        if (searchContext.minimumScore() != null) {
            final Collector child = collector;
            collector = new MinimumScoreCollector(collector, searchContext.minimumScore());
            if (doProfile) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_MIN_SCORE, Collections.singletonList((InternalProfileCollector) child));
            }
        }
        if (collector.getClass() == TotalHitCountCollector.class) {
            // instead of using a collector
            while (true) {
                // a constant_score query
                if (query instanceof ConstantScoreQuery) {
                    query = ((ConstantScoreQuery) query).getQuery();
                } else {
                    break;
                }
            }
            if (query.getClass() == MatchAllDocsQuery.class) {
                collector = null;
                topDocsCallable = new Callable<TopDocs>() {

                    @Override
                    public TopDocs call() throws Exception {
                        int count = searcher.getIndexReader().numDocs();
                        return new TopDocs(count, Lucene.EMPTY_SCORE_DOCS, 0);
                    }
                };
            } else if (query.getClass() == TermQuery.class && searcher.getIndexReader().hasDeletions() == false) {
                final Term term = ((TermQuery) query).getTerm();
                collector = null;
                topDocsCallable = new Callable<TopDocs>() {

                    @Override
                    public TopDocs call() throws Exception {
                        int count = 0;
                        for (LeafReaderContext context : searcher.getIndexReader().leaves()) {
                            count += context.reader().docFreq(term);
                        }
                        return new TopDocs(count, Lucene.EMPTY_SCORE_DOCS, 0);
                    }
                };
            }
        }
        final boolean timeoutSet = searchContext.timeout() != null && !searchContext.timeout().equals(SearchService.NO_TIMEOUT);
        if (timeoutSet && collector != null) {
            // collector might be null if no collection is actually needed
            final Collector child = collector;
            // TODO: change to use our own counter that uses the scheduler in ThreadPool
            // throws TimeLimitingCollector.TimeExceededException when timeout has reached
            collector = Lucene.wrapTimeLimitingCollector(collector, searchContext.timeEstimateCounter(), searchContext.timeout().millis());
            if (doProfile) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_TIMEOUT, Collections.singletonList((InternalProfileCollector) child));
            }
        }
        if (collector != null) {
            final Collector child = collector;
            collector = new CancellableCollector(searchContext.getTask()::isCancelled, searchContext.lowLevelCancellation(), collector);
            if (doProfile) {
                collector = new InternalProfileCollector(collector, CollectorResult.REASON_SEARCH_CANCELLED, Collections.singletonList((InternalProfileCollector) child));
            }
        }
        try {
            if (collector != null) {
                if (doProfile) {
                    searchContext.getProfilers().getCurrentQueryProfiler().setCollector((InternalProfileCollector) collector);
                }
                searcher.search(query, collector);
            }
        } catch (TimeLimitingCollector.TimeExceededException e) {
            assert timeoutSet : "TimeExceededException thrown even though timeout wasn't set";
            queryResult.searchTimedOut(true);
        } catch (Lucene.EarlyTerminationException e) {
            assert terminateAfterSet : "EarlyTerminationException thrown even though terminateAfter wasn't set";
            queryResult.terminatedEarly(true);
        } finally {
            searchContext.clearReleasables(SearchContext.Lifetime.COLLECTION);
        }
        if (terminateAfterSet && queryResult.terminatedEarly() == null) {
            queryResult.terminatedEarly(false);
        }
        queryResult.topDocs(topDocsCallable.call(), sortValueFormats);
        if (searchContext.getProfilers() != null) {
            ProfileShardResult shardResults = SearchProfileShardResults.buildShardResults(searchContext.getProfilers());
            searchContext.queryResult().profileResults(shardResults);
        }
        return rescore;
    } catch (Exception e) {
        throw new QueryPhaseExecutionException(searchContext, "Failed to execute main query", e);
    }
}