Examples of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner

Package org.apache.lucene.search.highlight.SynonymTokenizer

Examples of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner

org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner


  }


  public void testGetBestSingleFragmentWithWeights() throws Exception {


    TestHighlightRunner helper = new TestHighlightRunner() {


      public void run() throws Exception {
        WeightedSpanTerm[] wTerms = new WeightedSpanTerm[2];
        wTerms[0] = new WeightedSpanTerm(10f, "hello");


        List 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);


        Highlighter highlighter = getHighlighter(wTerms, HighlighterTest.this);// new
        // Highlighter(new
        // QueryTermScorer(wTerms));
        TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(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, new StringReader(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();


  }

View Full Code Here

  }


  // tests a "complex" analyzer that produces multiple
  // overlapping tokens
  public void testOverlapAnalyzer() throws Exception {
    TestHighlightRunner helper = new TestHighlightRunner() {


      public void run() throws Exception {
        HashMap synonyms = new HashMap();
        synonyms.put("football", "soccer,footie");
        Analyzer analyzer = new SynonymAnalyzer(synonyms);
        String srchkey = "football";


        String s = "football-soccer in the euro 2004 footie competition";
        QueryParser parser = new QueryParser("bookid", analyzer);
        Query query = parser.parse(srchkey);


        TokenStream tokenStream = analyzer.tokenStream(null, new StringReader(s));


        Highlighter highlighter = getHighlighter(query, null, tokenStream, HighlighterTest.this);


        // Get 3 best fragments and seperate with a "..."
        tokenStream = analyzer.tokenStream(null, new StringReader(s));


        String result = highlighter.getBestFragments(tokenStream, s, 3, "...");
        String expectedResult = "<B>football</B>-<B>soccer</B> in the euro 2004 <B>footie</B> competition";
        assertTrue("overlapping analyzer should handle highlights OK, expected:" + expectedResult
            + " actual:" + result, expectedResult.equals(result));
      }


    };


    helper.start();


  }

View Full Code Here

    helper.start();


  }


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


      public void run() throws Exception {
        numHighlights = 0;
        doSearching("Kennedy");
        // new Highlighter(HighlighterTest.this, new QueryTermScorer(query));


        for (int i = 0; i < hits.length(); i++) {
          String text = hits.doc(i).get(FIELD_NAME);
          TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(text));
          Highlighter highlighter = getHighlighter(query, FIELD_NAME, tokenStream,
              HighlighterTest.this);
          String result = highlighter.getBestFragment(tokenStream, text);
          System.out.println("\t" + result);
        }
        assertTrue("Failed to find correct number of highlights " + numHighlights + " found",
            numHighlights == 4);
      }
    };
    helper.start();
  }

View Full Code Here

    };
    helper.start();
  }


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


      public void run() throws Exception {


        doSearching("Kennedy");


        for (int i = 0; i < hits.length(); i++) {
          String text = hits.doc(i).get(FIELD_NAME);
          TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(text));


          Highlighter highlighter = getHighlighter(query, FIELD_NAME, tokenStream,
              HighlighterTest.this);// new Highlighter(this, new
          // QueryTermScorer(query));
          highlighter.setTextFragmenter(new SimpleFragmenter(20));
          String stringResults[] = highlighter.getBestFragments(tokenStream, text, 10);


          tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(text));
          TextFragment fragmentResults[] = highlighter.getBestTextFragments(tokenStream, text,
              true, 10);


          assertTrue("Failed to find correct number of text Fragments: " + fragmentResults.length
              + " vs " + stringResults.length, fragmentResults.length == stringResults.length);
          for (int j = 0; j < stringResults.length; j++) {
            System.out.println(fragmentResults[j]);
            assertTrue("Failed to find same text Fragments: " + fragmentResults[j] + " found",
                fragmentResults[j].toString().equals(stringResults[j]));


          }


        }
      }
    };
    helper.start();
  }

View Full Code Here

    };
    helper.start();
  }


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


      public void run() throws Exception {
        numHighlights = 0;
        doSearching("meat");
        TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(texts[0]));
        Highlighter highlighter = getHighlighter(query, FIELD_NAME, tokenStream,
            HighlighterTest.this);// new Highlighter(this, new
        // QueryTermScorer(query));
        highlighter.setMaxDocBytesToAnalyze(30);


        highlighter.getBestFragment(tokenStream, texts[0]);
        assertTrue("Setting MaxDocBytesToAnalyze should have prevented "
            + "us from finding matches for this record: " + numHighlights + " found",
            numHighlights == 0);
      }
    };


    helper.start();
  }

View Full Code Here


    helper.start();
  }


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


      public void run() throws Exception {
        String goodWord = "goodtoken";
        Set stopWords = new HashSet(1);
        stopWords.add("stoppedtoken");


        TermQuery query = new TermQuery(new Term("data", goodWord));


        String match = null;
        StringBuffer sb = new StringBuffer();
        sb.append(goodWord);
        for (int i = 0; i < 10000; i++) {
          sb.append(" ");
          // only one stopword
          sb.append(stopWords.iterator().next());
        }
        SimpleHTMLFormatter fm = new SimpleHTMLFormatter();
        Highlighter hg = getHighlighter(query, "data", new StandardAnalyzer(stopWords).tokenStream(
            "data", new StringReader(sb.toString())), fm);// new Highlighter(fm,
        // new
        // QueryTermScorer(query));
        hg.setTextFragmenter(new NullFragmenter());
        hg.setMaxDocBytesToAnalyze(100);
        match = hg.getBestFragment(new StandardAnalyzer(stopWords), "data", sb.toString());
        assertTrue("Matched text should be no more than 100 chars in length ", match.length() < hg
            .getMaxDocBytesToAnalyze());


        // add another tokenized word to the overrall length - but set way
        // beyond
        // the length of text under consideration (after a large slug of stop
        // words
        // + whitespace)
        sb.append(" ");
        sb.append(goodWord);
        match = hg.getBestFragment(new StandardAnalyzer(stopWords), "data", sb.toString());
        assertTrue("Matched text should be no more than 100 chars in length ", match.length() < hg
            .getMaxDocBytesToAnalyze());
      }
    };


    helper.start();


  }

View Full Code Here

    helper.start();


  }
  
  public void testMaxSizeEndHighlight() throws Exception {
    TestHighlightRunner helper = new TestHighlightRunner() {
      public void run() throws Exception {
        Set stopWords = new HashSet();
        stopWords.add("in");
        stopWords.add("it");
        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", new StandardAnalyzer(
            stopWords).tokenStream("text", new StringReader(text)), fm);
        hg.setTextFragmenter(new NullFragmenter());
        hg.setMaxDocCharsToAnalyze(36);
        String match = hg.getBestFragment(new StandardAnalyzer(stopWords), "text", text);
        assertTrue(
            "Matched text should contain remainder of text after highlighted query ",
            match.endsWith("in it"));
      }
    };
    helper.start();
  }

View Full Code Here

    };
    helper.start();
  }


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


      public void run() throws Exception {
        numHighlights = 0;
        // test to show how rewritten query can still be used
        searcher = new IndexSearcher(ramDir);
        Analyzer analyzer = new StandardAnalyzer();


        QueryParser parser = new QueryParser(FIELD_NAME, analyzer);
        Query query = parser.parse("JF? or Kenned*");
        System.out.println("Searching with primitive query");
        // forget to set this and...
        // query=query.rewrite(reader);
        Hits hits = searcher.search(query);


        // create an instance of the highlighter with the tags used to surround
        // highlighted text
        // QueryHighlightExtractor highlighter = new
        // QueryHighlightExtractor(this,
        // query, new StandardAnalyzer());


        int maxNumFragmentsRequired = 3;


        for (int i = 0; i < hits.length(); i++) {
          String text = hits.doc(i).get(FIELD_NAME);
          TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(text));
          Highlighter highlighter = getHighlighter(query, FIELD_NAME, tokenStream, HighlighterTest.this, false);


          highlighter.setTextFragmenter(new SimpleFragmenter(40));


          String highlightedText = highlighter.getBestFragments(tokenStream, text,
              maxNumFragmentsRequired, "...");


          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();
  }

View Full Code Here


    helper.start();
  }


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


      public void run() throws Exception {
        doSearching("AnInvalidQueryWhichShouldYieldNoResults");


        for (int i = 0; i < texts.length; i++) {
          String text = texts[i];
          TokenStream tokenStream = analyzer.tokenStream(FIELD_NAME, new StringReader(text));
          Highlighter highlighter = getHighlighter(query, FIELD_NAME, tokenStream,
              HighlighterTest.this);
          String result = highlighter.getBestFragment(tokenStream, text);
          assertNull("The highlight result should be null for text with no query terms", result);
        }
      }
    };


    helper.start();
  }

View Full Code Here

        numHighlights == 2);


  }


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


      public void run() throws Exception {
        String docMainText = "fred is one of the people";
        QueryParser parser = new QueryParser(FIELD_NAME, analyzer);
        Query query = parser.parse("fred category:people");


        // highlighting respects fieldnames used in query


        Scorer fieldSpecificScorer = null;
        if (mode == this.QUERY) {
          fieldSpecificScorer = new QueryScorer(query, FIELD_NAME);
        } else if (mode == this.QUERY_TERM) {
          fieldSpecificScorer = new QueryTermScorer(query, "contents");
        }
        Highlighter fieldSpecificHighlighter = new Highlighter(new SimpleHTMLFormatter(),
            fieldSpecificScorer);
        fieldSpecificHighlighter.setTextFragmenter(new NullFragmenter());
        String result = fieldSpecificHighlighter.getBestFragment(analyzer, FIELD_NAME, docMainText);
        assertEquals("Should match", result, "<B>fred</B> is one of the people");


        // highlighting does not respect fieldnames used in query
        Scorer fieldInSpecificScorer = null;
        if (mode == this.QUERY) {
          fieldInSpecificScorer = new QueryScorer(query, null);
        } else if (mode == this.QUERY_TERM) {
          fieldInSpecificScorer = new QueryTermScorer(query);
        }


        Highlighter fieldInSpecificHighlighter = new Highlighter(new SimpleHTMLFormatter(),
            fieldInSpecificScorer);
        fieldInSpecificHighlighter.setTextFragmenter(new NullFragmenter());
        result = fieldInSpecificHighlighter.getBestFragment(analyzer, FIELD_NAME, docMainText);
        assertEquals("Should match", result, "<B>fred</B> is one of the <B>people</B>");


        reader.close();
      }
    };


    helper.start();


  }

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.lucene.search.highlight.SynonymTokenizer.TestHighlightRunner

org.apache.lucene.search.highlight.HighlighterTest

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.