Examples with TestHighlightRunner org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner used on opensource projects

Example 1 with TestHighlightRunner

use of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner in project lucene-solr by apache.

the class SynonymTokenizer method testSpanHighlighting.

public void testSpanHighlighting() throws Exception {
    Query query1 = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term(FIELD_NAME, "wordx")), new SpanTermQuery(new Term(FIELD_NAME, "wordy")) }, 1, false);
    Query query2 = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term(FIELD_NAME, "wordy")), new SpanTermQuery(new Term(FIELD_NAME, "wordc")) }, 1, false);
    BooleanQuery.Builder bquery = new BooleanQuery.Builder();
    bquery.add(query1, Occur.SHOULD);
    bquery.add(query2, Occur.SHOULD);
    doSearching(bquery.build());
    TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            mode = QUERY;
            doStandardHighlights(analyzer, searcher, hits, query, HighlighterTest.this);
        }
    };
    helper.run();
    assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 7);
}

Example 2 with TestHighlightRunner

use of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner in project lucene-solr by apache.

the class SynonymTokenizer method testGetBestSingleFragmentWithWeights.

public void testGetBestSingleFragmentWithWeights() throws Exception {
    TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            WeightedSpanTerm[] wTerms = new WeightedSpanTerm[2];
            wTerms[0] = new WeightedSpanTerm(10f, "hello");
            List<PositionSpan> positionSpans = new ArrayList<>();
            positionSpans.add(new PositionSpan(0, 0));
            wTerms[0].addPositionSpans(positionSpans);
            wTerms[1] = new WeightedSpanTerm(1f, "kennedy");
            positionSpans = new ArrayList<>();
            positionSpans.add(new PositionSpan(14, 14));
            wTerms[1].addPositionSpans(positionSpans);
            // new
            Highlighter highlighter = getHighlighter(wTerms, HighlighterTest.this);
            // Highlighter(new
            // QueryTermScorer(wTerms));
            TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, texts[0]);
            highlighter.setTextFragmenter(new SimpleFragmenter(2));
            String result = highlighter.getBestFragment(tokenStream, texts[0]).trim();
            assertTrue("Failed to find best section using weighted terms. Found: [" + result + "]", "<B>Hello</B>".equals(result));
            // readjust weights
            wTerms[1].setWeight(50f);
            tokenStream = analyzer.tokenStream(FIELD_NAME, texts[0]);
            highlighter = getHighlighter(wTerms, HighlighterTest.this);
            highlighter.setTextFragmenter(new SimpleFragmenter(2));
            result = highlighter.getBestFragment(tokenStream, texts[0]).trim();
            assertTrue("Failed to find best section using weighted terms. Found: " + result, "<B>kennedy</B>".equals(result));
        }
    };
    helper.start();
}

Example 3 with TestHighlightRunner

use of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner in project lucene-solr by apache.

the class SynonymTokenizer method testOverlapAnalyzer2.

public void testOverlapAnalyzer2() throws Exception {
    TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            String s = "Hi-Speed10 foo";
            Query query;
            Highlighter highlighter;
            String result;
            query = new TermQuery(new Term("text", "foo"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2(), s, 3, "...");
            assertEquals("Hi-Speed10 <B>foo</B>", result);
            query = new TermQuery(new Term("text", "10"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2(), s, 3, "...");
            assertEquals("Hi-Speed<B>10</B> foo", result);
            query = new TermQuery(new Term("text", "hi"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2(), s, 3, "...");
            assertEquals("<B>Hi</B>-Speed10 foo", result);
            query = new TermQuery(new Term("text", "speed"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2(), s, 3, "...");
            assertEquals("Hi-<B>Speed</B>10 foo", result);
            query = new TermQuery(new Term("text", "hispeed"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2(), s, 3, "...");
            assertEquals("<B>Hi-Speed</B>10 foo", result);
            BooleanQuery.Builder booleanQuery = new BooleanQuery.Builder();
            booleanQuery.add(new TermQuery(new Term("text", "hi")), Occur.SHOULD);
            booleanQuery.add(new TermQuery(new Term("text", "speed")), Occur.SHOULD);
            query = booleanQuery.build();
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2(), s, 3, "...");
            assertEquals("<B>Hi-Speed</B>10 foo", result);
            // ///////////////// same tests, just put the bigger overlapping token
            // first
            query = new TermQuery(new Term("text", "foo"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2a(), s, 3, "...");
            assertEquals("Hi-Speed10 <B>foo</B>", result);
            query = new TermQuery(new Term("text", "10"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2a(), s, 3, "...");
            assertEquals("Hi-Speed<B>10</B> foo", result);
            query = new TermQuery(new Term("text", "hi"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2a(), s, 3, "...");
            assertEquals("<B>Hi</B>-Speed10 foo", result);
            query = new TermQuery(new Term("text", "speed"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2a(), s, 3, "...");
            assertEquals("Hi-<B>Speed</B>10 foo", result);
            query = new TermQuery(new Term("text", "hispeed"));
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2a(), s, 3, "...");
            assertEquals("<B>Hi-Speed</B>10 foo", result);
            query = booleanQuery.build();
            highlighter = getHighlighter(query, "text", HighlighterTest.this);
            result = highlighter.getBestFragments(getTS2a(), s, 3, "...");
            assertEquals("<B>Hi-Speed</B>10 foo", result);
        }
    };
    helper.start();
}

Example 4 with TestHighlightRunner

use of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner in project lucene-solr by apache.

the class SynonymTokenizer method testMaxSizeEndHighlight.

public void testMaxSizeEndHighlight() throws Exception {
    TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            CharacterRunAutomaton stopWords = new CharacterRunAutomaton(new RegExp("i[nt]").toAutomaton());
            TermQuery query = new TermQuery(new Term("text", "searchterm"));
            String text = "this is a text with searchterm in it";
            SimpleHTMLFormatter fm = new SimpleHTMLFormatter();
            Highlighter hg = getHighlighter(query, "text", fm);
            hg.setTextFragmenter(new NullFragmenter());
            hg.setMaxDocCharsToAnalyze(36);
            String match = hg.getBestFragment(new MockAnalyzer(random(), MockTokenizer.SIMPLE, true, stopWords), "text", text);
            assertTrue("Matched text should contain remainder of text after highlighted query ", match.endsWith("in it"));
        }
    };
    helper.start();
}

Example 5 with TestHighlightRunner

use of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner in project lucene-solr by apache.

the class SynonymTokenizer method testGetSimpleHighlight.

public void testGetSimpleHighlight() throws Exception {
    TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            numHighlights = 0;
            doSearching(new TermQuery(new Term(FIELD_NAME, "kennedy")));
            for (int i = 0; i < hits.totalHits; i++) {
                String text = searcher.doc(hits.scoreDocs[i].doc).get(FIELD_NAME);
                TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, text);
                Highlighter highlighter = getHighlighter(query, FIELD_NAME, HighlighterTest.this);
                String result = highlighter.getBestFragment(tokenStream, text);
                if (VERBOSE)
                    System.out.println("\t" + result);
            }
            assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 4);
        }
    };
    helper.start();
}