Examples with DocumentSet org.exist.dom.persistent.DocumentSet used on opensource projects

Example 11 with DocumentSet

use of org.exist.dom.persistent.DocumentSet in project exist by eXist-db.

the class NativeStructuralIndexWorkerTest method documentIdSet.

private DocumentSet documentIdSet(final List<Integer> documentIds) {
    final DocumentSet mockDocumentSet = createMock(DocumentSet.class);
    final List<DocumentImpl> docs = documentIds.stream().map(id -> {
        final DocumentImpl mockDocument = createMock(DocumentImpl.class);
        expect(mockDocument.getDocId()).andReturn(id).anyTimes();
        return mockDocument;
    }).collect(Collectors.toList());
    expect(mockDocumentSet.getDocumentIterator()).andReturn(docs.iterator());
    replay(mockDocumentSet);
    docs.forEach(EasyMock::replay);
    return mockDocumentSet;
}

Example 12 with DocumentSet

use of org.exist.dom.persistent.DocumentSet in project exist by eXist-db.

the class NativeStructuralIndexWorkerTest method getDocIdRanges_singleId.

@Test
public void getDocIdRanges_singleId() {
    final NativeStructuralIndexWorker indexWorker = new NativeStructuralIndexWorker(null);
    final DocumentSet docs = documentIdSet(Arrays.asList(6574));
    final List<NativeStructuralIndexWorker.Range> ranges = indexWorker.getDocIdRanges(docs);
    assertEquals(1, ranges.size());
    assertEquals(6574, ranges.get(0).start);
    assertEquals(6574, ranges.get(0).end);
}

Example 13 with DocumentSet

use of org.exist.dom.persistent.DocumentSet in project exist by eXist-db.

the class LuceneIndexTest method reindex.

@Test
public void reindex() throws EXistException, CollectionConfigurationException, PermissionDeniedException, SAXException, LockException, IOException, QName.IllegalQNameException {
    final DocumentSet docs = configureAndStore(COLLECTION_CONFIG1, XML1, "dropDocument.xml");
    final BrokerPool pool = existEmbeddedServer.getBrokerPool();
    final TransactionManager transact = pool.getTransactionManager();
    try (final DBBroker broker = pool.get(Optional.of(pool.getSecurityManager().getSystemSubject()));
        final Txn transaction = transact.beginTransaction()) {
        broker.reindexCollection(transaction, TestConstants.TEST_COLLECTION_URI);
        checkIndex(docs, broker, new QName[] { new QName("head") }, "title", 1);
        final Occurrences[] o = checkIndex(docs, broker, new QName[] { new QName("p") }, "with", 1);
        assertEquals(2, o[0].getOccurrences());
        checkIndex(docs, broker, new QName[] { new QName("hi") }, "just", 1);
        checkIndex(docs, broker, null, "in", 1);
        final QName attrQN = new QName("rend", XMLConstants.NULL_NS_URI, ElementValue.ATTRIBUTE);
        checkIndex(docs, broker, new QName[] { attrQN }, null, 2);
        checkIndex(docs, broker, new QName[] { attrQN }, "center", 1);
        transaction.commit();
    }
}

Example 14 with DocumentSet

use of org.exist.dom.persistent.DocumentSet in project exist by eXist-db.

the class NGramSearch method eval.

@Override
public Sequence eval(Sequence contextSequence, Item contextItem) throws XPathException {
    if (contextItem != null)
        contextSequence = contextItem.toSequence();
    NodeSet result;
    if (preselectResult == null) {
        Sequence input = getArgument(0).eval(contextSequence, contextItem);
        if (input.isEmpty())
            result = NodeSet.EMPTY_SET;
        else {
            long start = System.currentTimeMillis();
            NodeSet inNodes = input.toNodeSet();
            DocumentSet docs = inNodes.getDocumentSet();
            NGramIndexWorker index = (NGramIndexWorker) context.getBroker().getIndexController().getWorkerByIndexId(NGramIndex.ID);
            // Alternate design
            // NGramIndexWorker index =
            // (NGramIndexWorker)context.getBroker().getBrokerPool().getIndexManager().getIndexById(NGramIndex.ID).getWorker();
            String key = getArgument(1).eval(contextSequence, contextItem).getStringValue();
            List<QName> qnames = null;
            if (contextQName != null) {
                qnames = new ArrayList<>(1);
                qnames.add(contextQName);
            }
            result = processMatches(index, docs, qnames, key, inNodes, NodeSet.ANCESTOR);
            if (context.getProfiler().traceFunctions()) {
                // report index use
                context.getProfiler().traceIndexUsage(context, "ngram", this, PerformanceStats.BASIC_INDEX, System.currentTimeMillis() - start);
            }
        }
    } else {
        contextStep.setPreloadedData(contextSequence.getDocumentSet(), preselectResult);
        result = getArgument(0).eval(contextSequence).toNodeSet();
    }
    return result;
}

Example 15 with DocumentSet

use of org.exist.dom.persistent.DocumentSet in project exist by eXist-db.

the class NGramSearch method fixedStringSearch.

public NodeSet fixedStringSearch(final NGramIndexWorker index, final DocumentSet docs, final List<QName> qnames, final String query, final NodeSet nodeSet, final int axis) throws XPathException {
    String[] ngrams = NGramSearch.getDistinctNGrams(query, index.getN());
    // Nothing to search for? The find nothing.
    if (ngrams.length == 0)
        return new EmptyNodeSet();
    String firstNgramm = ngrams[0];
    LOG.trace("First NGRAM: {}", firstNgramm);
    NodeSet result = index.search(getExpressionId(), docs, qnames, firstNgramm, firstNgramm, context, nodeSet, axis);
    for (int i = 1; i < ngrams.length; i++) {
        String ngram = ngrams[i];
        int len = ngram.codePointCount(0, ngram.length());
        int fillSize = index.getN() - len;
        String filledNgram = ngram;
        // ngrams lead to a considerable performance loss.
        if (fillSize > 0) {
            String filler = ngrams[i - 1];
            StringBuilder buf = new StringBuilder();
            int pos = filler.offsetByCodePoints(0, len);
            for (int j = 0; j < fillSize; j++) {
                int codepoint = filler.codePointAt(pos);
                pos += Character.charCount(codepoint);
                buf.appendCodePoint(codepoint);
            }
            buf.append(ngram);
            filledNgram = buf.toString();
            LOG.debug("Filled: {}", filledNgram);
        }
        NodeSet nodes = index.search(getExpressionId(), docs, qnames, filledNgram, ngram, context, nodeSet, axis);
        final NodeSet nodesContainingFirstINgrams = result;
        result = NodeSets.transformNodes(nodes, proxy -> Optional.ofNullable(nodesContainingFirstINgrams.get(proxy)).map(before -> getContinuousMatches(before, proxy)).orElse(null));
    }
    return result;
}