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

Example 11 with PrefixQuery

use of org.apache.lucene.search.PrefixQuery in project lucene-solr by apache.

the class SimpleQueryParser method newPrefixQuery.

/**
   * Factory method to generate a prefix query.
   */
protected Query newPrefixQuery(String text) {
    BooleanQuery.Builder bq = new BooleanQuery.Builder();
    for (Map.Entry<String, Float> entry : weights.entrySet()) {
        final String fieldName = entry.getKey();
        final BytesRef term = getAnalyzer().normalize(fieldName, text);
        Query q = new PrefixQuery(new Term(fieldName, term));
        float boost = entry.getValue();
        if (boost != 1f) {
            q = new BoostQuery(q, boost);
        }
        bq.add(q, BooleanClause.Occur.SHOULD);
    }
    return simplify(bq.build());
}

Example 12 with PrefixQuery

use of org.apache.lucene.search.PrefixQuery in project lucene-solr by apache.

the class PrefixWildcardQueryNodeBuilder method build.

@Override
public PrefixQuery build(QueryNode queryNode) throws QueryNodeException {
    PrefixWildcardQueryNode wildcardNode = (PrefixWildcardQueryNode) queryNode;
    String text = wildcardNode.getText().subSequence(0, wildcardNode.getText().length() - 1).toString();
    PrefixQuery q = new PrefixQuery(new Term(wildcardNode.getFieldAsString(), text));
    MultiTermQuery.RewriteMethod method = (MultiTermQuery.RewriteMethod) queryNode.getTag(MultiTermRewriteMethodProcessor.TAG_ID);
    if (method != null) {
        q.setRewriteMethod(method);
    }
    return q;
}

Example 13 with PrefixQuery

use of org.apache.lucene.search.PrefixQuery in project lucene-solr by apache.

the class TestUnifiedHighlighterMTQ method testWhichMTQMatched.

/**
   * Runs a query with two MTQs and confirms the formatter
   * can tell which query matched which hit.
   */
public void testWhichMTQMatched() throws Exception {
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, indexAnalyzer);
    Field body = new Field("body", "", fieldType);
    Document doc = new Document();
    doc.add(body);
    body.setStringValue("Test a one sentence document.");
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    IndexSearcher searcher = newSearcher(ir);
    UnifiedHighlighter highlighter = new UnifiedHighlighter(searcher, indexAnalyzer);
    // use a variety of common MTQ types
    BooleanQuery query = new BooleanQuery.Builder().add(new PrefixQuery(new Term("body", "te")), BooleanClause.Occur.SHOULD).add(new WildcardQuery(new Term("body", "*one*")), BooleanClause.Occur.SHOULD).add(new FuzzyQuery(new Term("body", "zentence~")), BooleanClause.Occur.SHOULD).build();
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertEquals(1, topDocs.totalHits);
    String[] snippets = highlighter.highlight("body", query, topDocs);
    assertEquals(1, snippets.length);
    // Default formatter just bolds each hit:
    assertEquals("<b>Test</b> a <b>one</b> <b>sentence</b> document.", snippets[0]);
    // Now use our own formatter, that also stuffs the
    // matching term's text into the result:
    highlighter = new UnifiedHighlighter(searcher, indexAnalyzer) {

        @Override
        protected PassageFormatter getFormatter(String field) {
            return new PassageFormatter() {

                @Override
                public Object format(Passage[] passages, String content) {
                    // Copied from DefaultPassageFormatter, but
                    // tweaked to include the matched term:
                    StringBuilder sb = new StringBuilder();
                    int pos = 0;
                    for (Passage passage : passages) {
                        // don't add ellipsis if its the first one, or if its connected.
                        if (passage.getStartOffset() > pos && pos > 0) {
                            sb.append("... ");
                        }
                        pos = passage.getStartOffset();
                        for (int i = 0; i < passage.getNumMatches(); i++) {
                            int start = passage.getMatchStarts()[i];
                            int end = passage.getMatchEnds()[i];
                            // its possible to have overlapping terms
                            if (start > pos) {
                                sb.append(content, pos, start);
                            }
                            if (end > pos) {
                                sb.append("<b>");
                                sb.append(content, Math.max(pos, start), end);
                                sb.append('(');
                                sb.append(passage.getMatchTerms()[i].utf8ToString());
                                sb.append(')');
                                sb.append("</b>");
                                pos = end;
                            }
                        }
                        // its possible a "term" from the analyzer could span a sentence boundary.
                        sb.append(content, pos, Math.max(pos, passage.getEndOffset()));
                        pos = passage.getEndOffset();
                    }
                    return sb.toString();
                }
            };
        }
    };
    assertEquals(1, topDocs.totalHits);
    snippets = highlighter.highlight("body", query, topDocs);
    assertEquals(1, snippets.length);
    assertEquals("<b>Test(body:te*)</b> a <b>one(body:*one*)</b> <b>sentence(body:zentence~~2)</b> document.", snippets[0]);
    ir.close();
}

Example 14 with PrefixQuery

use of org.apache.lucene.search.PrefixQuery in project lucene-solr by apache.

the class TestUnifiedHighlighterMTQ method testMultiSegment.

public void testMultiSegment() throws Exception {
    // If we incorrectly got the term vector from mis-matched global/leaf doc ID, this test may fail
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, indexAnalyzer);
    Document doc = new Document();
    doc.add(new Field("body", "word aberration", fieldType));
    iw.addDocument(doc);
    // make segment
    iw.commit();
    doc = new Document();
    doc.add(new Field("body", "word absolve", fieldType));
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    IndexSearcher searcher = newSearcher(ir);
    UnifiedHighlighter highlighter = new UnifiedHighlighter(searcher, indexAnalyzer);
    Query query = new PrefixQuery(new Term("body", "ab"));
    TopDocs topDocs = searcher.search(query, 10);
    String[] snippets = highlighter.highlightFields(new String[] { "body" }, query, topDocs).get("body");
    Arrays.sort(snippets);
    assertEquals("[word <b>aberration</b>, word <b>absolve</b>]", Arrays.toString(snippets));
    ir.close();
}

Example 15 with PrefixQuery

use of org.apache.lucene.search.PrefixQuery in project lucene-solr by apache.

the class TestUnifiedHighlighterStrictPhrases method testMtq.

/**
   * Like {@link #testRewriteAndMtq} but no rewrite.
   */
public void testMtq() throws IOException {
    indexWriter.addDocument(newDoc("alpha bravo charlie - charlie bravo alpha"));
    initReaderSearcherHighlighter();
    SpanNearQuery snq = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term("body", "bravo")), // does NOT rewrite
    new SpanTermQuery(new Term("body", "charlie")) }, 0, true);
    BooleanQuery query = new BooleanQuery.Builder().add(snq, BooleanClause.Occur.MUST).add(new PrefixQuery(new Term("body", "al")), // MTQ
    BooleanClause.Occur.MUST).add(newPhraseQuery("body", "alpha bravo"), BooleanClause.Occur.MUST).add(newPhraseQuery("title", "bravo alpha"), BooleanClause.Occur.SHOULD).build();
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    String[] snippets = highlighter.highlight("body", query, topDocs);
    assertArrayEquals(new String[] { "<b>alpha</b> <b>bravo</b> <b>charlie</b> - charlie bravo <b>alpha</b>" }, snippets);
    // do again, this time with MTQ disabled.
    //disable but leave phrase processing enabled
    highlighter.setHandleMultiTermQuery(false);
    topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    snippets = highlighter.highlight("body", query, topDocs);
    assertArrayEquals(new String[] { "<b>alpha</b> <b>bravo</b> <b>charlie</b> - charlie bravo alpha" }, snippets);
}