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

Example 11 with FuzzyQuery

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

the class TestSpanMultiTermQueryWrapper method testNoSuchMultiTermsInSpanFirst.

public void testNoSuchMultiTermsInSpanFirst() throws Exception {
    //this hasn't been a problem  
    FuzzyQuery fuzzyNoSuch = new FuzzyQuery(new Term("field", "noSuch"), 1, 0, 1, false);
    SpanQuery spanNoSuch = new SpanMultiTermQueryWrapper<>(fuzzyNoSuch);
    SpanQuery spanFirst = new SpanFirstQuery(spanNoSuch, 10);
    assertEquals(0, searcher.search(spanFirst, 10).totalHits);
    WildcardQuery wcNoSuch = new WildcardQuery(new Term("field", "noSuch*"));
    SpanQuery spanWCNoSuch = new SpanMultiTermQueryWrapper<>(wcNoSuch);
    spanFirst = new SpanFirstQuery(spanWCNoSuch, 10);
    assertEquals(0, searcher.search(spanFirst, 10).totalHits);
    RegexpQuery rgxNoSuch = new RegexpQuery(new Term("field", "noSuch"));
    SpanQuery spanRgxNoSuch = new SpanMultiTermQueryWrapper<>(rgxNoSuch);
    spanFirst = new SpanFirstQuery(spanRgxNoSuch, 10);
    assertEquals(0, searcher.search(spanFirst, 10).totalHits);
    PrefixQuery prfxNoSuch = new PrefixQuery(new Term("field", "noSuch"));
    SpanQuery spanPrfxNoSuch = new SpanMultiTermQueryWrapper<>(prfxNoSuch);
    spanFirst = new SpanFirstQuery(spanPrfxNoSuch, 10);
    assertEquals(0, searcher.search(spanFirst, 10).totalHits);
}

Example 12 with FuzzyQuery

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

the class TestSpanMultiTermQueryWrapper method testFuzzy.

public void testFuzzy() throws Exception {
    FuzzyQuery fq = new FuzzyQuery(new Term("field", "broan"));
    SpanQuery sfq = new SpanMultiTermQueryWrapper<>(fq);
    // will not match quick brown fox
    SpanPositionRangeQuery sprq = new SpanPositionRangeQuery(sfq, 3, 6);
    assertEquals(2, searcher.search(sprq, 10).totalHits);
}

Example 13 with FuzzyQuery

use of org.apache.lucene.search.FuzzyQuery 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 FuzzyQuery

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

the class TestUnifiedHighlighterMTQ method testOneFuzzy.

public void testOneFuzzy() throws Exception {
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, indexAnalyzer);
    Field body = new Field("body", "", fieldType);
    Document doc = new Document();
    doc.add(body);
    body.setStringValue("This is a test.");
    iw.addDocument(doc);
    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);
    Query query = new FuzzyQuery(new Term("body", "tets"), 1);
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    String[] snippets = highlighter.highlight("body", query, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a <b>test</b>.", snippets[0]);
    assertEquals("<b>Test</b> a one sentence document.", snippets[1]);
    // with prefix
    query = new FuzzyQuery(new Term("body", "tets"), 1, 2);
    topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    snippets = highlighter.highlight("body", query, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a <b>test</b>.", snippets[0]);
    assertEquals("<b>Test</b> a one sentence document.", snippets[1]);
    // wrong field
    BooleanQuery bq = new BooleanQuery.Builder().add(new MatchAllDocsQuery(), BooleanClause.Occur.SHOULD).add(new FuzzyQuery(new Term("bogus", "tets"), 1), BooleanClause.Occur.SHOULD).build();
    topDocs = searcher.search(bq, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    snippets = highlighter.highlight("body", bq, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a test.", snippets[0]);
    assertEquals("Test a one sentence document.", snippets[1]);
    ir.close();
}

Example 15 with FuzzyQuery

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

the class TestQPHelper method testWildcard.

public void testWildcard() throws Exception {
    assertQueryEquals("term*", null, "term*");
    assertQueryEquals("term*^2", null, "(term*)^2.0");
    assertQueryEquals("term~", null, "term~2");
    assertQueryEquals("term~0.7", null, "term~1");
    assertQueryEquals("term~^3", null, "(term~2)^3.0");
    assertQueryEquals("term^3~", null, "(term~2)^3.0");
    assertQueryEquals("term*germ", null, "term*germ");
    assertQueryEquals("term*germ^3", null, "(term*germ)^3.0");
    assertTrue(getQuery("term*", null) instanceof PrefixQuery);
    assertTrue(getQuery("term*^2", null) instanceof BoostQuery);
    assertTrue(((BoostQuery) getQuery("term*^2", null)).getQuery() instanceof PrefixQuery);
    assertTrue(getQuery("term~", null) instanceof FuzzyQuery);
    assertTrue(getQuery("term~0.7", null) instanceof FuzzyQuery);
    FuzzyQuery fq = (FuzzyQuery) getQuery("term~0.7", null);
    assertEquals(1, fq.getMaxEdits());
    assertEquals(FuzzyQuery.defaultPrefixLength, fq.getPrefixLength());
    fq = (FuzzyQuery) getQuery("term~", null);
    assertEquals(2, fq.getMaxEdits());
    assertEquals(FuzzyQuery.defaultPrefixLength, fq.getPrefixLength());
    // value > 1, throws exception
    assertQueryNodeException("term~1.1");
    assertTrue(getQuery("term*germ", null) instanceof WildcardQuery);
    /*
     * Tests to see that wild card terms are (or are not) properly lower-cased
     * with propery parser configuration
     */
    // First prefix queries:
    // by default, convert to lowercase:
    assertWildcardQueryEquals("Term*", "term*");
    // explicitly set lowercase:
    assertWildcardQueryEquals("term*", "term*");
    assertWildcardQueryEquals("Term*", "term*");
    assertWildcardQueryEquals("TERM*", "term*");
    // Then 'full' wildcard queries:
    // by default, convert to lowercase:
    assertWildcardQueryEquals("Te?m", "te?m");
    // explicitly set lowercase:
    assertWildcardQueryEquals("te?m", "te?m");
    assertWildcardQueryEquals("Te?m", "te?m");
    assertWildcardQueryEquals("TE?M", "te?m");
    assertWildcardQueryEquals("Te?m*gerM", "te?m*germ");
    // Fuzzy queries:
    assertWildcardQueryEquals("Term~", "term~2");
    // Range queries:
    // TODO: implement this on QueryParser
    // Q0002E_INVALID_SYNTAX_CANNOT_PARSE: Syntax Error, cannot parse '[A TO
    // C]': Lexical error at line 1, column 1. Encountered: "[" (91), after
    // : ""
    assertWildcardQueryEquals("[A TO C]", "[a TO c]");
    // Test suffix queries: first disallow
    expectThrows(QueryNodeException.class, () -> {
        assertWildcardQueryEquals("*Term", "*term");
    });
    expectThrows(QueryNodeException.class, () -> {
        assertWildcardQueryEquals("?Term", "?term");
    });
    // Test suffix queries: then allow
    assertWildcardQueryEquals("*Term", "*term", true);
    assertWildcardQueryEquals("?Term", "?term", true);
}