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

Example 11 with WildcardQuery

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

the class KNearestNeighborClassifier method knnSearch.

private TopDocs knnSearch(String text) throws IOException {
    BooleanQuery.Builder mltQuery = new BooleanQuery.Builder();
    for (String fieldName : textFieldNames) {
        String boost = null;
        //terms boost actually helps in MLT queries
        mlt.setBoost(true);
        if (fieldName.contains("^")) {
            String[] field2boost = fieldName.split("\\^");
            fieldName = field2boost[0];
            boost = field2boost[1];
        }
        if (boost != null) {
            //if we have a field boost, we add it
            mlt.setBoostFactor(Float.parseFloat(boost));
        }
        mltQuery.add(new BooleanClause(mlt.like(fieldName, new StringReader(text)), BooleanClause.Occur.SHOULD));
        // restore neutral boost for next field
        mlt.setBoostFactor(1);
    }
    Query classFieldQuery = new WildcardQuery(new Term(classFieldName, "*"));
    mltQuery.add(new BooleanClause(classFieldQuery, BooleanClause.Occur.MUST));
    if (query != null) {
        mltQuery.add(query, BooleanClause.Occur.MUST);
    }
    return indexSearcher.search(mltQuery.build(), k);
}

Example 12 with WildcardQuery

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

the class KNearestFuzzyClassifier method knnSearch.

private TopDocs knnSearch(String text) throws IOException {
    BooleanQuery.Builder bq = new BooleanQuery.Builder();
    FuzzyLikeThisQuery fuzzyLikeThisQuery = new FuzzyLikeThisQuery(300, analyzer);
    for (String fieldName : textFieldNames) {
        // TODO: make this parameters configurable
        fuzzyLikeThisQuery.addTerms(text, fieldName, 1f, 2);
    }
    bq.add(fuzzyLikeThisQuery, BooleanClause.Occur.MUST);
    Query classFieldQuery = new WildcardQuery(new Term(classFieldName, "*"));
    bq.add(new BooleanClause(classFieldQuery, BooleanClause.Occur.MUST));
    if (query != null) {
        bq.add(query, BooleanClause.Occur.MUST);
    }
    return indexSearcher.search(bq.build(), k);
}

Example 13 with WildcardQuery

use of org.apache.lucene.search.WildcardQuery 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 14 with WildcardQuery

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

the class SynonymTokenizer method testConstantScoreMultiTermQuery.

public void testConstantScoreMultiTermQuery() throws Exception {
    numHighlights = 0;
    query = new WildcardQuery(new Term(FIELD_NAME, "ken*"));
    ((WildcardQuery) query).setRewriteMethod(MultiTermQuery.CONSTANT_SCORE_REWRITE);
    searcher = newSearcher(reader);
    // query = unReWrittenQuery.rewrite(reader);
    if (VERBOSE)
        System.out.println("Searching for: " + query.toString(FIELD_NAME));
    hits = searcher.search(query, 1000);
    for (int i = 0; i < hits.totalHits; i++) {
        final int docId = hits.scoreDocs[i].doc;
        final Document doc = searcher.doc(docId);
        String text = doc.get(FIELD_NAME);
        TokenStream tokenStream = getAnyTokenStream(FIELD_NAME, docId);
        int maxNumFragmentsRequired = 2;
        String fragmentSeparator = "...";
        QueryScorer scorer = new QueryScorer(query, HighlighterTest.FIELD_NAME);
        Highlighter highlighter = new Highlighter(this, scorer);
        highlighter.setTextFragmenter(new SimpleFragmenter(20));
        String result = highlighter.getBestFragments(tokenStream, text, maxNumFragmentsRequired, fragmentSeparator);
        if (VERBOSE)
            System.out.println("\t" + result);
    }
    assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 5);
    // try null field
    hits = searcher.search(query, 1000);
    numHighlights = 0;
    for (int i = 0; i < hits.totalHits; i++) {
        final int docId = hits.scoreDocs[i].doc;
        final Document doc = searcher.doc(docId);
        String text = doc.get(FIELD_NAME);
        TokenStream tokenStream = getAnyTokenStream(FIELD_NAME, docId);
        int maxNumFragmentsRequired = 2;
        String fragmentSeparator = "...";
        QueryScorer scorer = new QueryScorer(query, null);
        Highlighter highlighter = new Highlighter(this, scorer);
        highlighter.setTextFragmenter(new SimpleFragmenter(20));
        String result = highlighter.getBestFragments(tokenStream, text, maxNumFragmentsRequired, fragmentSeparator);
        if (VERBOSE)
            System.out.println("\t" + result);
    }
    assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 5);
    // try default field
    hits = searcher.search(query, 1000);
    numHighlights = 0;
    for (int i = 0; i < hits.totalHits; i++) {
        final int docId = hits.scoreDocs[i].doc;
        final Document doc = searcher.doc(docId);
        String text = doc.get(FIELD_NAME);
        TokenStream tokenStream = getAnyTokenStream(FIELD_NAME, docId);
        int maxNumFragmentsRequired = 2;
        String fragmentSeparator = "...";
        QueryScorer scorer = new QueryScorer(query, "random_field", HighlighterTest.FIELD_NAME);
        Highlighter highlighter = new Highlighter(this, scorer);
        highlighter.setTextFragmenter(new SimpleFragmenter(20));
        String result = highlighter.getBestFragments(tokenStream, text, maxNumFragmentsRequired, fragmentSeparator);
        if (VERBOSE)
            System.out.println("\t" + result);
    }
    assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 5);
}

Example 15 with WildcardQuery

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

the class SynonymTokenizer method testUnRewrittenQuery.

public void testUnRewrittenQuery() throws Exception {
    final TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            numHighlights = 0;
            // test to show how rewritten query can still be used
            searcher = newSearcher(reader);
            BooleanQuery.Builder query = new BooleanQuery.Builder();
            query.add(new WildcardQuery(new Term(FIELD_NAME, "jf?")), Occur.SHOULD);
            query.add(new WildcardQuery(new Term(FIELD_NAME, "kenned*")), Occur.SHOULD);
            if (VERBOSE)
                System.out.println("Searching with primitive query");
            // forget to set this and...
            // query=query.rewrite(reader);
            TopDocs hits = searcher.search(query.build(), 1000);
            // create an instance of the highlighter with the tags used to surround
            // highlighted text
            // QueryHighlightExtractor highlighter = new
            // QueryHighlightExtractor(this,
            // query, new StandardAnalyzer(TEST_VERSION));
            int maxNumFragmentsRequired = 3;
            for (int i = 0; i < hits.totalHits; i++) {
                final int docId = hits.scoreDocs[i].doc;
                final Document doc = searcher.doc(docId);
                String text = doc.get(FIELD_NAME);
                TokenStream tokenStream = getAnyTokenStream(FIELD_NAME, docId);
                Highlighter highlighter = getHighlighter(query.build(), FIELD_NAME, HighlighterTest.this, false);
                highlighter.setTextFragmenter(new SimpleFragmenter(40));
                String highlightedText = highlighter.getBestFragments(tokenStream, text, maxNumFragmentsRequired, "...");
                if (VERBOSE)
                    System.out.println(highlightedText);
            }
            // We expect to have zero highlights if the query is multi-terms and is
            // not
            // rewritten!
            assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 0);
        }
    };
    helper.start();
}