Examples with Collector org.apache.lucene.search.Collector used on opensource projects

Example 6 with Collector

use of org.apache.lucene.search.Collector 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);
    }
}

Example 7 with Collector

use of org.apache.lucene.search.Collector in project elasticsearch by elastic.

the class TermsSliceQueryTests method testSearch.

public void testSearch() throws Exception {
    final int numDocs = randomIntBetween(100, 200);
    final Directory dir = newDirectory();
    final RandomIndexWriter w = new RandomIndexWriter(random(), dir, new KeywordAnalyzer());
    int max = randomIntBetween(2, 10);
    int[] sliceCounters = new int[max];
    Set<String> keys = new HashSet<>();
    for (int i = 0; i < numDocs; ++i) {
        Document doc = new Document();
        String uuid = UUIDs.base64UUID();
        BytesRef br = new BytesRef(uuid);
        int id = Math.floorMod(br.hashCode(), max);
        sliceCounters[id]++;
        doc.add(new StringField("uuid", uuid, Field.Store.YES));
        w.addDocument(doc);
        keys.add(uuid);
    }
    final IndexReader reader = w.getReader();
    final IndexSearcher searcher = newSearcher(reader);
    for (int id = 0; id < max; id++) {
        TermsSliceQuery query1 = new TermsSliceQuery("uuid", id, max);
        assertThat(searcher.count(query1), equalTo(sliceCounters[id]));
        searcher.search(query1, new Collector() {

            @Override
            public LeafCollector getLeafCollector(LeafReaderContext context) throws IOException {
                return new LeafCollector() {

                    @Override
                    public void setScorer(Scorer scorer) throws IOException {
                    }

                    @Override
                    public void collect(int doc) throws IOException {
                        Document d = context.reader().document(doc, Collections.singleton("uuid"));
                        String uuid = d.get("uuid");
                        assertThat(keys.contains(uuid), equalTo(true));
                        keys.remove(uuid);
                    }
                };
            }

            @Override
            public boolean needsScores() {
                return false;
            }
        });
    }
    assertThat(keys.size(), equalTo(0));
    w.close();
    reader.close();
    dir.close();
}

Example 8 with Collector

use of org.apache.lucene.search.Collector in project elasticsearch by elastic.

the class DocValuesSliceQueryTests method testSearch.

public void testSearch() throws Exception {
    final int numDocs = randomIntBetween(100, 200);
    final Directory dir = newDirectory();
    final RandomIndexWriter w = new RandomIndexWriter(random(), dir);
    int max = randomIntBetween(2, 10);
    int[] sliceCounters1 = new int[max];
    int[] sliceCounters2 = new int[max];
    Set<String> keys = new HashSet<>();
    for (int i = 0; i < numDocs; ++i) {
        Document doc = new Document();
        String uuid = UUIDs.base64UUID();
        int intValue = randomInt();
        long doubleValue = NumericUtils.doubleToSortableLong(randomDouble());
        doc.add(new StringField("uuid", uuid, Field.Store.YES));
        doc.add(new SortedNumericDocValuesField("intField", intValue));
        doc.add(new SortedNumericDocValuesField("doubleField", doubleValue));
        w.addDocument(doc);
        sliceCounters1[Math.floorMod(BitMixer.mix((long) intValue), max)]++;
        sliceCounters2[Math.floorMod(BitMixer.mix(doubleValue), max)]++;
        keys.add(uuid);
    }
    final IndexReader reader = w.getReader();
    final IndexSearcher searcher = newSearcher(reader);
    for (int id = 0; id < max; id++) {
        DocValuesSliceQuery query1 = new DocValuesSliceQuery("intField", id, max);
        assertThat(searcher.count(query1), equalTo(sliceCounters1[id]));
        DocValuesSliceQuery query2 = new DocValuesSliceQuery("doubleField", id, max);
        assertThat(searcher.count(query2), equalTo(sliceCounters2[id]));
        searcher.search(query1, new Collector() {

            @Override
            public LeafCollector getLeafCollector(LeafReaderContext context) throws IOException {
                return new LeafCollector() {

                    @Override
                    public void setScorer(Scorer scorer) throws IOException {
                    }

                    @Override
                    public void collect(int doc) throws IOException {
                        Document d = context.reader().document(doc, Collections.singleton("uuid"));
                        String uuid = d.get("uuid");
                        assertThat(keys.contains(uuid), equalTo(true));
                        keys.remove(uuid);
                    }
                };
            }

            @Override
            public boolean needsScores() {
                return false;
            }
        });
    }
    assertThat(keys.size(), equalTo(0));
    w.close();
    reader.close();
    dir.close();
}

Example 9 with Collector

use of org.apache.lucene.search.Collector in project elasticsearch by elastic.

the class QueryPhaseTests method testPostFilterDisablesCountOptimization.

public void testPostFilterDisablesCountOptimization() throws Exception {
    TestSearchContext context = new TestSearchContext(null);
    context.parsedQuery(new ParsedQuery(new MatchAllDocsQuery()));
    context.setSize(0);
    context.setTask(new SearchTask(123L, "", "", "", null));
    final AtomicBoolean collected = new AtomicBoolean();
    IndexSearcher contextSearcher = new IndexSearcher(new MultiReader()) {

        protected void search(List<LeafReaderContext> leaves, Weight weight, Collector collector) throws IOException {
            collected.set(true);
            super.search(leaves, weight, collector);
        }
    };
    QueryPhase.execute(context, contextSearcher);
    assertEquals(0, context.queryResult().topDocs().totalHits);
    assertFalse(collected.get());
    context.parsedPostFilter(new ParsedQuery(new MatchNoDocsQuery()));
    QueryPhase.execute(context, contextSearcher);
    assertEquals(0, context.queryResult().topDocs().totalHits);
    assertTrue(collected.get());
}

Example 10 with Collector

use of org.apache.lucene.search.Collector in project elasticsearch by elastic.

the class QueryPhaseTests method testMinScoreDisablesCountOptimization.

public void testMinScoreDisablesCountOptimization() throws Exception {
    TestSearchContext context = new TestSearchContext(null);
    context.parsedQuery(new ParsedQuery(new MatchAllDocsQuery()));
    context.setSize(0);
    context.setTask(new SearchTask(123L, "", "", "", null));
    final AtomicBoolean collected = new AtomicBoolean();
    IndexSearcher contextSearcher = new IndexSearcher(new MultiReader()) {

        protected void search(List<LeafReaderContext> leaves, Weight weight, Collector collector) throws IOException {
            collected.set(true);
            super.search(leaves, weight, collector);
        }
    };
    QueryPhase.execute(context, contextSearcher);
    assertEquals(0, context.queryResult().topDocs().totalHits);
    assertFalse(collected.get());
    context.minimumScore(1);
    QueryPhase.execute(context, contextSearcher);
    assertEquals(0, context.queryResult().topDocs().totalHits);
    assertTrue(collected.get());
}