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

Example 16 with TermsEnum

use of org.apache.lucene.index.TermsEnum in project neo4j by neo4j.

the class SimpleUniquenessVerifier method verify.

@Override
public void verify(PropertyAccessor accessor, int[] propKeyIds) throws IndexEntryConflictException, IOException {
    try {
        DuplicateCheckingCollector collector = DuplicateCheckingCollector.forProperties(accessor, propKeyIds);
        IndexSearcher searcher = indexSearcher();
        for (LeafReaderContext leafReaderContext : searcher.getIndexReader().leaves()) {
            Fields fields = leafReaderContext.reader().fields();
            for (String field : fields) {
                if (LuceneDocumentStructure.NODE_ID_KEY.equals(field)) {
                    continue;
                }
                TermsEnum terms = LuceneDocumentStructure.originalTerms(fields.terms(field), field);
                BytesRef termsRef;
                while ((termsRef = terms.next()) != null) {
                    if (terms.docFreq() > 1) {
                        collector.reset();
                        searcher.search(new TermQuery(new Term(field, termsRef)), collector);
                    }
                }
            }
        }
    } catch (IOException e) {
        Throwable cause = e.getCause();
        if (cause instanceof IndexEntryConflictException) {
            throw (IndexEntryConflictException) cause;
        }
        throw e;
    }
}

Example 17 with TermsEnum

use of org.apache.lucene.index.TermsEnum in project neo4j by neo4j.

the class NonUniqueDatabaseIndexSamplerTest method getTerms.

private Terms getTerms(String value, int frequency) throws IOException {
    TermsEnum termsEnum = mock(TermsEnum.class);
    Terms terms = mock(Terms.class);
    when(terms.iterator()).thenReturn(termsEnum);
    when(termsEnum.next()).thenReturn(new BytesRef(value.getBytes())).thenReturn(null);
    when(termsEnum.docFreq()).thenReturn(frequency);
    return terms;
}

Example 18 with TermsEnum

use of org.apache.lucene.index.TermsEnum in project textdb by TextDB.

the class WordCountIndexSource method computeWordCount.

private void computeWordCount() throws TextDBException {
    try {
        HashMap<String, Integer> wordCountMap = new HashMap<>();
        DataReader dataReader = RelationManager.getRelationManager().getTableDataReader(predicate.getTableName(), new MatchAllDocsQuery());
        dataReader.open();
        IndexReader luceneIndexReader = dataReader.getLuceneIndexReader();
        for (int i = 0; i < luceneIndexReader.numDocs(); i++) {
            Terms termVector = luceneIndexReader.getTermVector(i, predicate.getAttribute());
            TermsEnum termsEnum = termVector.iterator();
            while (termsEnum.next() != null) {
                String key = termsEnum.term().utf8ToString();
                wordCountMap.put(key, wordCountMap.get(key) == null ? ((int) termsEnum.totalTermFreq()) : wordCountMap.get(key) + ((int) termsEnum.totalTermFreq()));
            }
        }
        luceneIndexReader.close();
        dataReader.close();
        sortedWordCountMap = wordCountMap.entrySet().stream().sorted((e1, e2) -> e2.getValue().compareTo(e1.getValue())).collect(Collectors.toList());
        wordCountIterator = sortedWordCountMap.iterator();
    } catch (IOException e) {
        throw new DataFlowException(e);
    }
}

Example 19 with TermsEnum

use of org.apache.lucene.index.TermsEnum in project textdb by TextDB.

the class DataReader method buildPayloadFromTermVector.

private ArrayList<Span> buildPayloadFromTermVector(List<IField> fields, int docID) throws IOException {
    ArrayList<Span> payloadSpanList = new ArrayList<>();
    for (Attribute attr : inputSchema.getAttributes()) {
        String attributeName = attr.getAttributeName();
        AttributeType attributeType = attr.getAttributeType();
        // payload.
        if (attributeType != AttributeType.TEXT) {
            continue;
        }
        String fieldValue = fields.get(inputSchema.getIndex(attributeName)).getValue().toString();
        Terms termVector = luceneIndexReader.getTermVector(docID, attributeName);
        if (termVector == null) {
            continue;
        }
        TermsEnum termsEnum = termVector.iterator();
        PostingsEnum termPostings = null;
        // go through document terms
        while ((termsEnum.next()) != null) {
            termPostings = termsEnum.postings(termPostings, PostingsEnum.ALL);
            if (termPostings.nextDoc() == DocIdSetIterator.NO_MORE_DOCS) {
                continue;
            }
            // for each term, go through its postings
            for (int i = 0; i < termPostings.freq(); i++) {
                // nextPosition needs to be called first
                int tokenPosition = termPostings.nextPosition();
                int charStart = termPostings.startOffset();
                int charEnd = termPostings.endOffset();
                String analyzedTermStr = termsEnum.term().utf8ToString();
                String originalTermStr = fieldValue.substring(charStart, charEnd);
                Span span = new Span(attributeName, charStart, charEnd, analyzedTermStr, originalTermStr, tokenPosition);
                payloadSpanList.add(span);
            }
        }
    }
    return payloadSpanList;
}

Example 20 with TermsEnum

use of org.apache.lucene.index.TermsEnum in project lucene-solr by apache.

the class TestLucene54DocValuesFormat method doTestTermsEnumRandom.

// TODO: try to refactor this and some termsenum tests into the base class.
// to do this we need to fix the test class to get a DVF not a Codec so we can setup
// the postings format correctly.
private void doTestTermsEnumRandom(int numDocs, int minLength, int maxLength) throws Exception {
    Directory dir = newFSDirectory(createTempDir());
    IndexWriterConfig conf = newIndexWriterConfig(new MockAnalyzer(random()));
    conf.setMergeScheduler(new SerialMergeScheduler());
    // set to duel against a codec which has ordinals:
    final PostingsFormat pf = TestUtil.getPostingsFormatWithOrds(random());
    final DocValuesFormat dv = new Lucene54DocValuesFormat();
    conf.setCodec(new AssertingCodec() {

        @Override
        public PostingsFormat getPostingsFormatForField(String field) {
            return pf;
        }

        @Override
        public DocValuesFormat getDocValuesFormatForField(String field) {
            return dv;
        }
    });
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir, conf);
    // index some docs
    for (int i = 0; i < numDocs; i++) {
        Document doc = new Document();
        Field idField = new StringField("id", Integer.toString(i), Field.Store.NO);
        doc.add(idField);
        final int length = TestUtil.nextInt(random(), minLength, maxLength);
        int numValues = random().nextInt(17);
        // create a random list of strings
        List<String> values = new ArrayList<>();
        for (int v = 0; v < numValues; v++) {
            values.add(TestUtil.randomSimpleString(random(), minLength, length));
        }
        // add in any order to the indexed field
        ArrayList<String> unordered = new ArrayList<>(values);
        Collections.shuffle(unordered, random());
        for (String v : values) {
            doc.add(newStringField("indexed", v, Field.Store.NO));
        }
        // add in any order to the dv field
        ArrayList<String> unordered2 = new ArrayList<>(values);
        Collections.shuffle(unordered2, random());
        for (String v : unordered2) {
            doc.add(new SortedSetDocValuesField("dv", new BytesRef(v)));
        }
        writer.addDocument(doc);
        if (random().nextInt(31) == 0) {
            writer.commit();
        }
    }
    // delete some docs
    int numDeletions = random().nextInt(numDocs / 10);
    for (int i = 0; i < numDeletions; i++) {
        int id = random().nextInt(numDocs);
        writer.deleteDocuments(new Term("id", Integer.toString(id)));
    }
    // compare per-segment
    DirectoryReader ir = writer.getReader();
    for (LeafReaderContext context : ir.leaves()) {
        LeafReader r = context.reader();
        Terms terms = r.terms("indexed");
        if (terms != null) {
            SortedSetDocValues ssdv = r.getSortedSetDocValues("dv");
            assertEquals(terms.size(), ssdv.getValueCount());
            TermsEnum expected = terms.iterator();
            TermsEnum actual = r.getSortedSetDocValues("dv").termsEnum();
            assertEquals(terms.size(), expected, actual);
            doTestSortedSetEnumAdvanceIndependently(ssdv);
        }
    }
    ir.close();
    writer.forceMerge(1);
    // now compare again after the merge
    ir = writer.getReader();
    LeafReader ar = getOnlyLeafReader(ir);
    Terms terms = ar.terms("indexed");
    if (terms != null) {
        assertEquals(terms.size(), ar.getSortedSetDocValues("dv").getValueCount());
        TermsEnum expected = terms.iterator();
        TermsEnum actual = ar.getSortedSetDocValues("dv").termsEnum();
        assertEquals(terms.size(), expected, actual);
    }
    ir.close();
    writer.close();
    dir.close();
}