Examples with IndexWriter org.apache.lucene.index.IndexWriter used on opensource projects

Example 6 with IndexWriter

use of org.apache.lucene.index.IndexWriter in project elasticsearch by elastic.

the class PercolateQueryBuilder method createMultiDocumentSearcher.

static IndexSearcher createMultiDocumentSearcher(Analyzer analyzer, ParsedDocument doc) {
    RAMDirectory ramDirectory = new RAMDirectory();
    try (IndexWriter indexWriter = new IndexWriter(ramDirectory, new IndexWriterConfig(analyzer))) {
        indexWriter.addDocuments(doc.docs());
        indexWriter.commit();
        DirectoryReader directoryReader = DirectoryReader.open(ramDirectory);
        assert directoryReader.leaves().size() == 1 : "Expected single leaf, but got [" + directoryReader.leaves().size() + "]";
        final IndexSearcher slowSearcher = new IndexSearcher(directoryReader) {

            @Override
            public Weight createNormalizedWeight(Query query, boolean needsScores) throws IOException {
                BooleanQuery.Builder bq = new BooleanQuery.Builder();
                bq.add(query, BooleanClause.Occur.MUST);
                bq.add(Queries.newNestedFilter(), BooleanClause.Occur.MUST_NOT);
                return super.createNormalizedWeight(bq.build(), needsScores);
            }
        };
        slowSearcher.setQueryCache(null);
        return slowSearcher;
    } catch (IOException e) {
        throw new ElasticsearchException("Failed to create index for percolator with nested document ", e);
    }
}

Example 7 with IndexWriter

use of org.apache.lucene.index.IndexWriter in project elasticsearch by elastic.

the class PercolateQueryTests method init.

@Before
public void init() throws Exception {
    directory = newDirectory();
    IndexWriterConfig config = new IndexWriterConfig(new WhitespaceAnalyzer());
    config.setMergePolicy(NoMergePolicy.INSTANCE);
    indexWriter = new IndexWriter(directory, config);
}

Example 8 with IndexWriter

use of org.apache.lucene.index.IndexWriter 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);
}

Example 9 with IndexWriter

use of org.apache.lucene.index.IndexWriter in project elasticsearch by elastic.

the class MultiPhrasePrefixQueryTests method testSimple.

public void testSimple() throws Exception {
    IndexWriter writer = new IndexWriter(new RAMDirectory(), new IndexWriterConfig(Lucene.STANDARD_ANALYZER));
    Document doc = new Document();
    doc.add(new Field("field", "aaa bbb ccc ddd", TextField.TYPE_NOT_STORED));
    writer.addDocument(doc);
    IndexReader reader = DirectoryReader.open(writer);
    IndexSearcher searcher = new IndexSearcher(reader);
    MultiPhrasePrefixQuery query = new MultiPhrasePrefixQuery();
    query.add(new Term("field", "aa"));
    assertThat(searcher.count(query), equalTo(1));
    query = new MultiPhrasePrefixQuery();
    query.add(new Term("field", "aaa"));
    query.add(new Term("field", "bb"));
    assertThat(searcher.count(query), equalTo(1));
    query = new MultiPhrasePrefixQuery();
    query.setSlop(1);
    query.add(new Term("field", "aaa"));
    query.add(new Term("field", "cc"));
    assertThat(searcher.count(query), equalTo(1));
    query = new MultiPhrasePrefixQuery();
    query.setSlop(1);
    query.add(new Term("field", "xxx"));
    assertThat(searcher.count(query), equalTo(0));
}

Example 10 with IndexWriter

use of org.apache.lucene.index.IndexWriter in project elasticsearch by elastic.

the class MoreLikeThisQueryTests method testSimple.

public void testSimple() throws Exception {
    Directory dir = new RAMDirectory();
    IndexWriter indexWriter = new IndexWriter(dir, new IndexWriterConfig(Lucene.STANDARD_ANALYZER));
    indexWriter.commit();
    Document document = new Document();
    document.add(new TextField("_id", "1", Field.Store.YES));
    document.add(new TextField("text", "lucene", Field.Store.YES));
    indexWriter.addDocument(document);
    document = new Document();
    document.add(new TextField("_id", "2", Field.Store.YES));
    document.add(new TextField("text", "lucene release", Field.Store.YES));
    indexWriter.addDocument(document);
    IndexReader reader = DirectoryReader.open(indexWriter);
    IndexSearcher searcher = new IndexSearcher(reader);
    MoreLikeThisQuery mltQuery = new MoreLikeThisQuery("lucene", new String[] { "text" }, Lucene.STANDARD_ANALYZER);
    mltQuery.setLikeText("lucene");
    mltQuery.setMinTermFrequency(1);
    mltQuery.setMinDocFreq(1);
    long count = searcher.count(mltQuery);
    assertThat(count, equalTo(2L));
    reader.close();
    indexWriter.close();
}