private void matchedFieldsTestCase( boolean useMatchedFields, boolean fieldMatch, String fieldValue, String expected, Query... queryClauses ) throws IOException { Document doc = new Document(); FieldType stored = new FieldType( TextField.TYPE_STORED ); stored.setStoreTermVectorOffsets( true ); stored.setStoreTermVectorPositions( true ); stored.setStoreTermVectors( true ); stored.freeze(); FieldType matched = new FieldType( TextField.TYPE_NOT_STORED ); matched.setStoreTermVectorOffsets( true ); matched.setStoreTermVectorPositions( true ); matched.setStoreTermVectors( true ); matched.freeze(); doc.add( new Field( "field", fieldValue, stored ) ); // Whitespace tokenized with English stop words doc.add( new Field( "field_exact", fieldValue, matched ) ); // Whitespace tokenized without stop words doc.add( new Field( "field_super_exact", fieldValue, matched ) ); // Whitespace tokenized without toLower doc.add( new Field( "field_characters", fieldValue, matched ) ); // Each letter is a token doc.add( new Field( "field_tripples", fieldValue, matched ) ); // Every three letters is a token doc.add( new Field( "field_sliced", fieldValue.substring( 0, // Sliced at 10 chars then analyzed just like field Math.min( fieldValue.length() - 1 , 10 ) ), matched ) ); doc.add( new Field( "field_der_red", new CannedTokenStream( // Hacky field containing "der" and "red" at pos = 0 token( "der", 1, 0, 3 ), token( "red", 0, 0, 3 ) ), matched ) ); final Map<String, Analyzer> fieldAnalyzers = new TreeMap<String, Analyzer>(); fieldAnalyzers.put( "field", new MockAnalyzer( random(), MockTokenizer.WHITESPACE, true, MockTokenFilter.ENGLISH_STOPSET ) ); fieldAnalyzers.put( "field_exact", new MockAnalyzer( random() ) ); fieldAnalyzers.put( "field_super_exact", new MockAnalyzer( random(), MockTokenizer.WHITESPACE, false ) ); fieldAnalyzers.put( "field_characters", new MockAnalyzer( random(), new CharacterRunAutomaton( new RegExp(".").toAutomaton() ), true ) ); fieldAnalyzers.put( "field_tripples", new MockAnalyzer( random(), new CharacterRunAutomaton( new RegExp("...").toAutomaton() ), true ) ); fieldAnalyzers.put( "field_sliced", fieldAnalyzers.get( "field" ) ); fieldAnalyzers.put( "field_der_red", fieldAnalyzers.get( "field" ) ); // This is required even though we provide a token stream Analyzer analyzer = new AnalyzerWrapper() { public Analyzer getWrappedAnalyzer(String fieldName) { return fieldAnalyzers.get( fieldName ); } }; Directory dir = newDirectory(); IndexWriter writer = new IndexWriter( dir, newIndexWriterConfig( TEST_VERSION_CURRENT, analyzer ) ); writer.addDocument( doc ); FastVectorHighlighter highlighter = new FastVectorHighlighter(); FragListBuilder fragListBuilder = new SimpleFragListBuilder(); FragmentsBuilder fragmentsBuilder = new ScoreOrderFragmentsBuilder(); IndexReader reader = DirectoryReader.open( writer, true ); String[] preTags = new String[] { "<b>" }; String[] postTags = new String[] { "</b>" }; Encoder encoder = new DefaultEncoder(); int docId = 0; BooleanQuery query = new BooleanQuery(); for ( Query clause : queryClauses ) { query.add( clause, Occur.MUST ); } FieldQuery fieldQuery = new FieldQuery( query, reader, true, fieldMatch ); String[] bestFragments; if ( useMatchedFields ) { Set< String > matchedFields = new HashSet< String >(); matchedFields.add( "field" ); matchedFields.add( "field_exact" ); matchedFields.add( "field_super_exact" ); matchedFields.add( "field_characters" ); matchedFields.add( "field_tripples" ); matchedFields.add( "field_sliced" ); matchedFields.add( "field_der_red" ); bestFragments = highlighter.getBestFragments( fieldQuery, reader, docId, "field", matchedFields, 25, 1, fragListBuilder, fragmentsBuilder, preTags, postTags, encoder ); } else { bestFragments = highlighter.getBestFragments( fieldQuery, reader, docId, "field", 25, 1, fragListBuilder, fragmentsBuilder, preTags, postTags, encoder ); } assertEquals( expected, bestFragments[ 0 ] ); reader.close(); writer.close(); dir.close(); }