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

Example 1 with TermInSetQuery

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

the class IdsQueryBuilder method doToQuery.

@Override
protected Query doToQuery(QueryShardContext context) throws IOException {
    Query query;
    if (this.ids.isEmpty()) {
        query = Queries.newMatchNoDocsQuery("Missing ids in \"" + this.getName() + "\" query.");
    } else {
        Collection<String> typesForQuery;
        if (types.length == 0) {
            typesForQuery = context.queryTypes();
        } else if (types.length == 1 && MetaData.ALL.equals(types[0])) {
            typesForQuery = context.getMapperService().types();
        } else {
            typesForQuery = new HashSet<>();
            Collections.addAll(typesForQuery, types);
        }
        query = new TermInSetQuery(UidFieldMapper.NAME, Uid.createUidsForTypesAndIds(typesForQuery, ids));
    }
    return query;
}

Example 2 with TermInSetQuery

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

the class PercolatorFieldMapperTests method testCreateCandidateQuery.

public void testCreateCandidateQuery() throws Exception {
    addQueryMapping();
    MemoryIndex memoryIndex = new MemoryIndex(false);
    memoryIndex.addField("field1", "the quick brown fox jumps over the lazy dog", new WhitespaceAnalyzer());
    memoryIndex.addField("field2", "some more text", new WhitespaceAnalyzer());
    memoryIndex.addField("_field3", "unhide me", new WhitespaceAnalyzer());
    memoryIndex.addField("field4", "123", new WhitespaceAnalyzer());
    memoryIndex.addField(new LongPoint("number_field", 10L), new WhitespaceAnalyzer());
    IndexReader indexReader = memoryIndex.createSearcher().getIndexReader();
    BooleanQuery candidateQuery = (BooleanQuery) fieldType.createCandidateQuery(indexReader);
    assertEquals(2, candidateQuery.clauses().size());
    assertEquals(Occur.SHOULD, candidateQuery.clauses().get(0).getOccur());
    TermInSetQuery termsQuery = (TermInSetQuery) candidateQuery.clauses().get(0).getQuery();
    PrefixCodedTerms terms = termsQuery.getTermData();
    assertThat(terms.size(), equalTo(14L));
    PrefixCodedTerms.TermIterator termIterator = terms.iterator();
    assertTermIterator(termIterator, "_field3me", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "_field3unhide", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1brown", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1dog", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1fox", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1jumps", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1lazy", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1over", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1quick", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field1the", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field2more", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field2some", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field2text", fieldType.queryTermsField.name());
    assertTermIterator(termIterator, "field4123", fieldType.queryTermsField.name());
    assertEquals(Occur.SHOULD, candidateQuery.clauses().get(1).getOccur());
    assertEquals(new TermQuery(new Term(fieldType.extractionResultField.name(), EXTRACTION_FAILED)), candidateQuery.clauses().get(1).getQuery());
}

Example 3 with TermInSetQuery

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

the class QueryAnalyzerTests method testExtractQueryMetadata_termsQuery.

public void testExtractQueryMetadata_termsQuery() {
    TermInSetQuery termsQuery = new TermInSetQuery("_field", new BytesRef("_term1"), new BytesRef("_term2"));
    Result result = analyze(termsQuery);
    assertThat(result.verified, is(true));
    List<Term> terms = new ArrayList<>(result.terms);
    Collections.sort(terms);
    assertThat(terms.size(), equalTo(2));
    assertThat(terms.get(0).field(), equalTo("_field"));
    assertThat(terms.get(0).text(), equalTo("_term1"));
    assertThat(terms.get(1).field(), equalTo("_field"));
    assertThat(terms.get(1).text(), equalTo("_term2"));
}

Example 4 with TermInSetQuery

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

the class TermsQueryBuilderTests method doAssertLuceneQuery.

@Override
protected void doAssertLuceneQuery(TermsQueryBuilder queryBuilder, Query query, SearchContext context) throws IOException {
    if (queryBuilder.termsLookup() == null && (queryBuilder.values() == null || queryBuilder.values().isEmpty())) {
        assertThat(query, instanceOf(MatchNoDocsQuery.class));
        MatchNoDocsQuery matchNoDocsQuery = (MatchNoDocsQuery) query;
        assertThat(matchNoDocsQuery.toString(), containsString("No terms supplied for \"terms\" query."));
    } else if (queryBuilder.termsLookup() != null && randomTerms.size() == 0) {
        assertThat(query, instanceOf(MatchNoDocsQuery.class));
        MatchNoDocsQuery matchNoDocsQuery = (MatchNoDocsQuery) query;
        assertThat(matchNoDocsQuery.toString(), containsString("No terms supplied for \"terms\" query."));
    } else {
        assertThat(query, either(instanceOf(TermInSetQuery.class)).or(instanceOf(PointInSetQuery.class)).or(instanceOf(ConstantScoreQuery.class)));
        if (query instanceof ConstantScoreQuery) {
            assertThat(((ConstantScoreQuery) query).getQuery(), instanceOf(BooleanQuery.class));
        }
        // we only do the check below for string fields (otherwise we'd have to decode the values)
        if (queryBuilder.fieldName().equals(INT_FIELD_NAME) || queryBuilder.fieldName().equals(DOUBLE_FIELD_NAME) || queryBuilder.fieldName().equals(BOOLEAN_FIELD_NAME) || queryBuilder.fieldName().equals(DATE_FIELD_NAME)) {
            return;
        }
        // expected returned terms depending on whether we have a terms query or a terms lookup query
        List<Object> terms;
        if (queryBuilder.termsLookup() != null) {
            terms = randomTerms;
        } else {
            terms = queryBuilder.values();
        }
        TermInSetQuery expected = new TermInSetQuery(queryBuilder.fieldName(), terms.stream().filter(Objects::nonNull).map(Object::toString).map(BytesRef::new).collect(Collectors.toList()));
        assertEquals(expected, query);
    }
}

Example 5 with TermInSetQuery

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

the class FreqTermsEnumTests method setUp.

@Before
@Override
public void setUp() throws Exception {
    super.setUp();
    referenceAll = new HashMap<>();
    referenceNotDeleted = new HashMap<>();
    referenceFilter = new HashMap<>();
    Directory dir = newDirectory();
    // use keyword analyzer we rely on the stored field holding the exact term.
    IndexWriterConfig conf = newIndexWriterConfig(new KeywordAnalyzer());
    if (frequently()) {
        // we don't want to do any merges, so we won't expunge deletes
        conf.setMergePolicy(NoMergePolicy.INSTANCE);
    }
    iw = new IndexWriter(dir, conf);
    terms = new String[scaledRandomIntBetween(10, 300)];
    for (int i = 0; i < terms.length; i++) {
        terms[i] = randomAsciiOfLength(5);
    }
    int numberOfDocs = scaledRandomIntBetween(30, 300);
    Document[] docs = new Document[numberOfDocs];
    for (int i = 0; i < numberOfDocs; i++) {
        Document doc = new Document();
        doc.add(new StringField("id", Integer.toString(i), Field.Store.YES));
        docs[i] = doc;
        for (String term : terms) {
            if (randomBoolean()) {
                continue;
            }
            int freq = randomIntBetween(1, 3);
            for (int j = 0; j < freq; j++) {
                doc.add(new TextField("field", term, Field.Store.YES));
            }
        }
    }
    for (int i = 0; i < docs.length; i++) {
        Document doc = docs[i];
        iw.addDocument(doc);
        if (rarely()) {
            iw.commit();
        }
    }
    Set<String> deletedIds = new HashSet<>();
    for (int i = 0; i < docs.length; i++) {
        Document doc = docs[i];
        if (randomInt(5) == 2) {
            Term idTerm = new Term("id", doc.getField("id").stringValue());
            deletedIds.add(idTerm.text());
            iw.deleteDocuments(idTerm);
        }
    }
    for (String term : terms) {
        referenceAll.put(term, new FreqHolder());
        referenceFilter.put(term, new FreqHolder());
        referenceNotDeleted.put(term, new FreqHolder());
    }
    // now go over each doc, build the relevant references and filter
    reader = DirectoryReader.open(iw);
    List<BytesRef> filterTerms = new ArrayList<>();
    for (int docId = 0; docId < reader.maxDoc(); docId++) {
        Document doc = reader.document(docId);
        addFreqs(doc, referenceAll);
        if (!deletedIds.contains(doc.getField("id").stringValue())) {
            addFreqs(doc, referenceNotDeleted);
            if (randomBoolean()) {
                filterTerms.add(new BytesRef(doc.getField("id").stringValue()));
                addFreqs(doc, referenceFilter);
            }
        }
    }
    filter = new TermInSetQuery("id", filterTerms);
}