List of usage examples for org.apache.lucene.search.highlight Highlighter Highlighter
public Highlighter(Formatter formatter, Scorer fragmentScorer)
From source file:org.intermine.api.search.SearchResults.java
License:GNU General Public License
/** * Actually filter the web searchable items we have to get a reduced list of matches. * @param origQueryString A query to filter the items against. Assumes the query * string is neither null not empty. * @param target Information about the scope and type of items to receive. * @param profileRepo The repository of the user who wants to find something. * @return A set of search results.//w w w . j av a2s . c o m * @throws ParseException If the query string cannot be parsed. * @throws IOException If there is an issue opening the indices. */ private static SearchResults doFilteredSearch(String origQueryString, SearchTarget target, SearchRepository profileRepo) throws ParseException, IOException { Map<WebSearchable, String> highlightedDescMap = new HashMap<WebSearchable, String>(); String queryString = prepareQueryString(origQueryString); LOG.info("Searching " + target + " for " + " was:" + origQueryString + " now:" + queryString); long time = System.currentTimeMillis(); org.apache.lucene.search.Query query; Analyzer analyzer = new SnowballAnalyzer(Version.LUCENE_30, "English", StopAnalyzer.ENGLISH_STOP_WORDS_SET); // The default search field is the content buffer. QueryParser queryParser = new QueryParser(Version.LUCENE_30, "content", analyzer); query = queryParser.parse(queryString); // Get directories. String type = target.getType(); Map<String, WebSearchable> globalWebSearchables = new HashMap<String, WebSearchable>(); Set<SearchRepository> globals = SearchRepository.getGlobalSearchRepositories(); List<Directory> globalDirs = new ArrayList<Directory>(); for (SearchRepository sr : globals) { globalWebSearchables.putAll(sr.getWebSearchableMap(type)); globalDirs.add(sr.getSearchIndex(type)); } Map<String, WebSearchable> userWebSearchables = profileRepo.getWebSearchableMap(type); Directory userDirectory = profileRepo.getSearchIndex(type); MultiSearcher searcher = prepareSearcher(target, userDirectory, globalDirs); // required to expand search terms query = searcher.rewrite(query); TopDocs topDocs = searcher.search(query, 1000); //FIXME: hardcoded limit time = System.currentTimeMillis() - time; LOG.info("Found " + topDocs.totalHits + " document(s) that matched query '" + queryString + "' in " + time + " milliseconds:"); QueryScorer scorer = new QueryScorer(query); Highlighter highlighter = new Highlighter(FORMATTER, scorer); Map<WebSearchable, Float> hitMap = new HashMap<WebSearchable, Float>(); Map<WebSearchable, Set<String>> tags = new HashMap<WebSearchable, Set<String>>(); for (int i = 0; i < topDocs.totalHits; i++) { WebSearchable webSearchable = null; Document doc = searcher.doc(topDocs.scoreDocs[i].doc); //String docScope = doc.get("scope"); String name = doc.get("name"); webSearchable = userWebSearchables.get(name); if (webSearchable == null) { webSearchable = globalWebSearchables.get(name); } if (webSearchable == null) { throw new RuntimeException("unknown WebSearchable: " + name); } Float luceneScore = new Float(topDocs.scoreDocs[i].score); hitMap.put(webSearchable, luceneScore); tags.put(webSearchable, new HashSet<String>(asList(split(doc.get("tags"))))); try { if (highlightedDescMap != null) { String highlightString = webSearchable.getDescription(); if (highlightString == null) { highlightString = ""; } TokenStream tokenStream = analyzer.tokenStream("", new StringReader(highlightString)); highlighter.setTextFragmenter(new NullFragmenter()); highlightedDescMap.put(webSearchable, highlighter.getBestFragment(tokenStream, highlightString)); } } catch (InvalidTokenOffsetsException e) { LOG.warn("Highlighter exception", e); } } Map<String, WebSearchable> wsMap = new HashMap<String, WebSearchable>(); for (WebSearchable ws : hitMap.keySet()) { wsMap.put(ws.getName(), ws); } return new SearchResults(hitMap, wsMap, highlightedDescMap, tags); }
From source file:org.jboss.seam.wiki.core.search.metamodel.SearchSupport.java
License:LGPL
/** * Returns the hits of the given query as fragments, highlighted, concatenated, and separated. * <p>/*from w w w .j a va2 s . c om*/ * Pass in a <tt>NullFragmenter</tt> if you don't want any fragmentation by terms but * simply the hits highlighted. Otherwise, you will most likely use <tt>SimpleFragmenter</tt>. * The text you supply must be the same that was indexed, it will go through the same * analysis procedure to find the hits. Do not pass a different String than the one indexed * by Hibernate Search! If you use transparent string bridge with Hibernate Search, run the * bridge before passing the string into this method. * <p> * This method escapes any dangerous HTML characters in the indexed text and fragments by * replacing it with HTML entities. You can use the returned string directly to build a * <tt>SearchHit</tt>. * * @param query the query that produced hits * @param fragmenter a fragmenter that can split the indexed text * @param indexedText the original text that was analyzed and indexed by Hibernate Search (after any bridges!) * @param numOfFragments the number of fragments to include in the returned result * @param alternativeLength if there are no hits to highlight, how many characters of the original text to return * @return the fragmented, highglighted, and then concatenated substring of the indexed text */ protected String escapeBestFragments(Query query, Fragmenter fragmenter, String indexedText, int numOfFragments, int alternativeLength) { // The HTML escaping forces us to first fragment with internal placeholders... Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(INTERNAL_BEGIN_HIT, INTERNAL_END_HIT), new QueryScorer(query)); highlighter.setTextFragmenter(fragmenter); try { // Use the same analyzer as the indexer! TokenStream tokenStream = new StandardAnalyzer().tokenStream(null, new StringReader(indexedText)); String unescapedFragements = highlighter.getBestFragments(tokenStream, indexedText, numOfFragments, getFragmentSeparator()); String escapedFragments = WikiUtil.escapeHtml(WikiUtil.removeMacros(unescapedFragements), false, false); // .. and then replace the internal placeholders with real tags after HTML has been escaped escapedFragments = escapedFragments.replaceAll(INTERNAL_BEGIN_HIT, getBeginHitTag()); escapedFragments = escapedFragments.replaceAll(INTERNAL_END_HIT, getEndHitTag()); // Strip out macros // If no fragments were produced (no hits), return the original text as an alternative if (escapedFragments.length() == 0 && alternativeLength != 0) { return WikiUtil.escapeHtml(WikiUtil.removeMacros(indexedText.substring(0, indexedText.length() > alternativeLength ? alternativeLength : indexedText.length())), false, false); } else if (escapedFragments.length() == 0 && alternativeLength == 0) { return WikiUtil.escapeHtml(WikiUtil.removeMacros(indexedText), false, false); } return escapedFragments; } catch (Exception ex) { throw new RuntimeException(ex); } }
From source file:org.mskcc.pathdb.lucene.LuceneResults.java
License:Open Source License
private Highlighter createHighlighter(String term) throws IOException, ParseException { // Standard Analyzer to extract words using a list of English stop words. StandardAnalyzer analyzer = new StandardAnalyzer(); // Standard Query Parser QueryParser queryParser = new QueryParser(LuceneConfig.FIELD_ALL, analyzer); // for the usage of highlighting with wildcards // Necessary to expand search terms IndexReader reader = IndexReader.open(new File(LuceneConfig.getLuceneDirectory())); Query luceneQuery = queryParser.parse(term); luceneQuery = luceneQuery.rewrite(reader); // Scorer implementation which scores text fragments by the number of // unique query terms found. QueryScorer queryScorer = new QueryScorer(luceneQuery); // HTML Formatted surrounds matching text with <B></B> tags. SimpleHTMLFormatter htmlFormatter = new SimpleHTMLFormatter(); // Highligher Class Highlighter highLighter = new Highlighter(htmlFormatter, queryScorer); // XXX Characters Max in Each Fragment Fragmenter fragmenter = new SimpleFragmenter(100); highLighter.setTextFragmenter(fragmenter); return highLighter; }
From source file:org.mskcc.pathdb.lucene.PsiInteractorExtractor.java
License:Open Source License
/** * Constructor//from w ww.j a va 2s . co m * * @param entrySet PSI-MI Entry Set Object. * @param queryStr Query String. * @param xdebug XDebug Object. * @throws IOException Input Output Exception. * @throws ParseException Parsing Exception. * @throws ValidationException Validation Exception. * @throws MarshalException Marshaling Exception. */ public PsiInteractorExtractor(EntrySet entrySet, String queryStr, XDebug xdebug) throws IOException, ParseException, ValidationException, MarshalException { try { this.xdebug = xdebug; this.entrySet = entrySet; interactors = new HashSet(); analyzer = new StandardAnalyzer(); reader = IndexReader.open(LuceneConfig.getLuceneDirectory()); if (queryStr != null) { QueryParser queryParser = new QueryParser(LuceneConfig.FIELD_ALL, analyzer); query = queryParser.parse(queryStr); query = query.rewrite(reader); SimpleHTMLFormatter htmlFormatter = new SimpleHTMLFormatter(); highLighter = new Highlighter(htmlFormatter, new QueryScorer(query)); checkAllEntries(); } } finally { if (reader != null) { reader.close(); } } }
From source file:org.mskcc.pathdb.tool.QueryFullText.java
License:Open Source License
/** * Executes Full Text Query.//from ww w . j av a 2 s . c om * * @param term Search Term * @throws QueryException Lucene Query Error * @throws IOException I/O Error * @throws ParseException Lucene Parsing Error */ public static void queryFullText(String term) throws QueryException, IOException, ParseException { System.out.println("Using search term: " + term); LuceneReader luceneReader = new LuceneReader(); Hits hits = luceneReader.executeQuery(term); int num = Math.min(10, hits.length()); System.out.println("Total Number of Hits: " + hits.length()); if (hits.length() > 0) { // Standard Analyzer to extract words using a list of English stop words. StandardAnalyzer analyzer = new StandardAnalyzer(); // Standard Query Parser QueryParser queryParser = new QueryParser(LuceneConfig.FIELD_ALL, analyzer); // for the usage of highlighting with wildcards // Necessary to expand search terms IndexReader reader = IndexReader.open(new File(LuceneConfig.getLuceneDirectory())); Query luceneQuery = queryParser.parse(term); luceneQuery = luceneQuery.rewrite(reader); // Scorer implementation which scores text fragments by the number of // unique query terms found. QueryScorer queryScorer = new QueryScorer(luceneQuery); // HTML Formatted surrounds matching text with <B></B> tags. SimpleHTMLFormatter htmlFormatter = new SimpleHTMLFormatter(); // Highligher Class Highlighter highLighter = new Highlighter(htmlFormatter, queryScorer); // XXX Characters Max in Each Fragment Fragmenter fragmenter = new SimpleFragmenter(100); highLighter.setTextFragmenter(fragmenter); System.out.println("Showing hits: 0-" + (num - 1)); for (int i = 0; i < num; i++) { System.out.print("Hit " + i + ": "); // Get the Matching Hit Document doc = hits.doc(i); // Get the Field of Interest Field field = doc.getField(LuceneConfig.FIELD_ALL); // Create the Token Stream TokenStream tokenStream = new StandardAnalyzer().tokenStream(LuceneConfig.FIELD_ALL, new StringReader(field.stringValue())); // Get the Best Fragment String formattedText = highLighter.getBestFragments(tokenStream, field.stringValue(), 5, "..."); System.out.println(formattedText); } } }
From source file:org.olat.search.service.searcher.SearchResultsImpl.java
License:Apache License
/** * Highlight (bold,color) query words in result-document. Set HighlightResult for content or description. * /*from ww w . ja va 2 s . c o m*/ * @param query * @param analyzer * @param doc * @param resultDocument * @throws IOException */ private void doHighlight(final Query query, final Analyzer analyzer, final Document doc, final ResultDocument resultDocument) throws IOException { final Highlighter highlighter = new Highlighter( new SimpleHTMLFormatter(HIGHLIGHT_PRE_TAG, HIGHLIGHT_POST_TAG), new QueryScorer(query)); // Get 3 best fragments of content and seperate with a "..." try { // highlight content final String content = doc.get(AbstractOlatDocument.CONTENT_FIELD_NAME); TokenStream tokenStream = analyzer.tokenStream(AbstractOlatDocument.CONTENT_FIELD_NAME, new StringReader(content)); String highlightResult = highlighter.getBestFragments(tokenStream, content, 3, HIGHLIGHT_SEPARATOR); // if no highlightResult is in content => look in description if (highlightResult.length() == 0) { final String description = doc.get(AbstractOlatDocument.DESCRIPTION_FIELD_NAME); tokenStream = analyzer.tokenStream(AbstractOlatDocument.DESCRIPTION_FIELD_NAME, new StringReader(description)); highlightResult = highlighter.getBestFragments(tokenStream, description, 3, HIGHLIGHT_SEPARATOR); resultDocument.setHighlightingDescription(true); } resultDocument.setHighlightResult(highlightResult); // highlight title final String title = doc.get(AbstractOlatDocument.TITLE_FIELD_NAME); tokenStream = analyzer.tokenStream(AbstractOlatDocument.TITLE_FIELD_NAME, new StringReader(title)); final String highlightTitle = highlighter.getBestFragments(tokenStream, title, 3, " "); resultDocument.setHighlightTitle(highlightTitle); } catch (final InvalidTokenOffsetsException e) { log.warn("", e); } }
From source file:org.openrdf.sail.lucene.LuceneQuery.java
License:BSD License
@Override public void highlight(URI property) { Formatter formatter = new SimpleHTMLFormatter(SearchFields.HIGHLIGHTER_PRE_TAG, SearchFields.HIGHLIGHTER_POST_TAG); highlighter = new Highlighter(formatter, new QueryScorer(query)); }
From source file:org.openrdf.sail.lucene.LuceneQueryIterator.java
License:BSD License
/** * Evaluates one Lucene Query. It distinguishes between two cases, * the one where no subject is given and the one were it is given. * @param query the lucene query to evaluate * @return the lucene hits/*from ww w . ja v a 2s .c o m*/ */ private TopDocs evaluate(QuerySpec query) { // get the subject of the query Resource subject = query.getSubject(); try { // parse the query string to a lucene query Query lucenequery = this.index.parseQuery(query.getQueryString(), query.getPropertyURI()); // if the query requests for the snippet, create a highlighter using this query if (query.getSnippetVariableName() != null) { Highlighter highlighter = new Highlighter(formatter, new QueryScorer(lucenequery)); this.highlighters.put(query, highlighter); } // distinguish the two cases of subject == null if (subject == null) { return this.index.search(lucenequery); } else { return this.index.search(subject, lucenequery); } } catch (Exception e) { log.error("There was a problem evaluating query '" + query.getQueryString() + "' for property '" + query.getPropertyURI() + "!", e); } return null; }
From source file:org.openrdf.sail.lucene.LuceneSailConnection.java
License:BSD License
/** * Evaluates one Lucene Query. It distinguishes between two cases, the one * where no subject is given and the one were it is given. * /*from ww w . j a va 2 s . co m*/ * @param query * the Lucene query to evaluate * @return QueryResult consisting of hits and highlighter */ private QueryResult evaluate(QuerySpec query) { TopDocs hits = null; Highlighter highlighter = null; // get the subject of the query Resource subject = query.getSubject(); try { // parse the query string to a lucene query String sQuery = query.getQueryString(); if (!sQuery.isEmpty()) { Query lucenequery = this.luceneIndex.parseQuery(query.getQueryString(), query.getPropertyURI()); // if the query requests for the snippet, create a highlighter using // this query if (query.getSnippetVariableName() != null) { Formatter formatter = new SimpleHTMLFormatter(); highlighter = new Highlighter(formatter, new QueryScorer(lucenequery)); } // distinguish the two cases of subject == null if (subject == null) { hits = this.luceneIndex.search(lucenequery); } else { hits = this.luceneIndex.search(subject, lucenequery); } } else { hits = new TopDocs(0, new ScoreDoc[0], 0.0f); } } catch (Exception e) { logger.error("There was a problem evaluating query '" + query.getQueryString() + "' for property '" + query.getPropertyURI() + "!", e); } return new QueryResult(hits, highlighter); }
From source file:org.schors.evlampia.search.LogEntryAggregator.java
License:Open Source License
private final String tryHighlight(String text, String[] fields) throws IOException, InvalidTokenOffsetsException { if (null == text) return null; if (null == highlighter) { final QueryScorer scorer = new QueryScorer(query.rewrite(indexSearcher.getIndexReader())); highlighter = new Highlighter(new SimpleHTMLFormatter("<span class='highlighted'>", "</span>"), scorer); highlighter.setTextFragmenter(new SimpleSpanFragmenter(scorer, 330)); }// w ww. jav a 2s. c o m for (final String field : fields) { final String highlighted = highlighter.getBestFragment(Constants.analyzer, field, text); if (null != highlighted) return highlighted; } return text; }