List of usage examples for org.apache.lucene.search.highlight Highlighter Highlighter
public Highlighter(Formatter formatter, Encoder encoder, Scorer fragmentScorer)
From source file:org.apache.solr.highlight.DefaultSolrHighlighter.java
License:Apache License
/** * Return a {@link org.apache.lucene.search.highlight.Highlighter} appropriate for this field. * @param query The current Query//from www .j a v a 2 s. c o m * @param fieldName The name of the field * @param request The current SolrQueryRequest */ protected Highlighter getHighlighter(Query query, String fieldName, SolrQueryRequest request) { SolrParams params = request.getParams(); Highlighter highlighter = new Highlighter(getFormatter(fieldName, params), getEncoder(fieldName, params), getQueryScorer(query, fieldName, request)); highlighter.setTextFragmenter(getFragmenter(fieldName, params)); return highlighter; }
From source file:org.apache.wiki.search.LuceneSearchProvider.java
License:Apache License
/** * Searches pages using a particular combination of flags. * * @param query The query to perform in Lucene query language * @param flags A set of flags// w w w. j a v a 2 s . c om * @return A Collection of SearchResult instances * @throws ProviderException if there is a problem with the backend */ public Collection findPages(String query, int flags) throws ProviderException { IndexSearcher searcher = null; ArrayList<SearchResult> list = null; Highlighter highlighter = null; try { String[] queryfields = { LUCENE_PAGE_CONTENTS, LUCENE_PAGE_NAME, LUCENE_AUTHOR, LUCENE_ATTACHMENTS }; QueryParser qp = new MultiFieldQueryParser(Version.LUCENE_36, queryfields, getLuceneAnalyzer()); //QueryParser qp = new QueryParser( LUCENE_PAGE_CONTENTS, getLuceneAnalyzer() ); Query luceneQuery = qp.parse(query); if ((flags & FLAG_CONTEXTS) != 0) { highlighter = new Highlighter(new SimpleHTMLFormatter("<span class=\"searchmatch\">", "</span>"), new SimpleHTMLEncoder(), new QueryScorer(luceneQuery)); } try { File dir = new File(m_luceneDirectory); Directory luceneDir = new SimpleFSDirectory(dir, null); IndexReader reader = IndexReader.open(luceneDir); searcher = new IndexSearcher(reader); } catch (Exception ex) { log.info("Lucene not yet ready; indexing not started", ex); return null; } ScoreDoc[] hits = searcher.search(luceneQuery, MAX_SEARCH_HITS).scoreDocs; list = new ArrayList<SearchResult>(hits.length); for (int curr = 0; curr < hits.length; curr++) { int docID = hits[curr].doc; Document doc = searcher.doc(docID); String pageName = doc.get(LUCENE_ID); WikiPage page = m_engine.getPage(pageName, WikiPageProvider.LATEST_VERSION); if (page != null) { if (page instanceof Attachment) { // Currently attachments don't look nice on the search-results page // When the search-results are cleaned up this can be enabled again. } int score = (int) (hits[curr].score * 100); // Get highlighted search contexts String text = doc.get(LUCENE_PAGE_CONTENTS); String[] fragments = new String[0]; if (text != null && highlighter != null) { TokenStream tokenStream = getLuceneAnalyzer().tokenStream(LUCENE_PAGE_CONTENTS, new StringReader(text)); fragments = highlighter.getBestFragments(tokenStream, text, MAX_FRAGMENTS); } SearchResult result = new SearchResultImpl(page, score, fragments); list.add(result); } else { log.error("Lucene found a result page '" + pageName + "' that could not be loaded, removing from Lucene cache"); pageRemoved(new WikiPage(m_engine, pageName)); } } } catch (IOException e) { log.error("Failed during lucene search", e); } catch (ParseException e) { log.info("Broken query; cannot parse query ", e); throw new ProviderException("You have entered a query Lucene cannot process: " + e.getMessage()); } catch (InvalidTokenOffsetsException e) { log.error("Tokens are incompatible with provided text ", e); } finally { if (searcher != null) { try { searcher.close(); } catch (IOException e) { log.error(e); } } } return list; }
From source file:org.archive.tnh.servlet.OpenSearchServlet.java
License:Apache License
public void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException { try {//from www .j av a2s. c om long responseTime = System.nanoTime(); QueryParameters p = (QueryParameters) request.getAttribute(OpenSearchHelper.PARAMS_KEY); if (p == null) { p = getQueryParameters(request); } BooleanQuery q = this.translator.translate(p.query, this.foldAccents); this.translator.addFilterGroup(q, "site", p.sites); this.translator.addFilterGroup(q, "type", p.types); this.translator.addFilterGroup(q, "collection", p.collections); this.translator.addFilterGroup(q, "date", p.dates); long parseQueryTime = System.nanoTime(); if (Arrays.equals(p.indexNames, QueryParameters.ALL_INDEXES)) { if (p.excludes.length > 0) { // If there are indexes to exclude, exclude them. p.indexNames = removeExcludes(p.excludes); } } else { // There are explicitly named indexes. Weed out any unknown names. p.indexNames = removeUnknownIndexNames(p.indexNames); } Search.Result result; if (p.indexNames.length == 0) { result = new Search.Result(); result.hits = new Hit[0]; } else { result = this.searcher.search(p.indexNames, q, p.start + (p.hitsPerPage * 3), p.hitsPerSite); } long executeQueryTime = System.nanoTime(); // The 'end' is usually just the end of the current page // (start+hitsPerPage); but if we are on the last page // of de-duped results, then the end is hits.getLength(). int end = Math.min(result.hits.length, p.start + p.hitsPerPage); // The length is usually just (end-start), unless the start // position is past the end of the results -- which is common when // de-duping. The user could easily jump past the true end of the // de-dup'd results. If the start is past the end, we use a // length of '0' to produce an empty results page. int length = Math.max(end - p.start, 0); // Usually, the total results is the total number of non-de-duped // results. Howerver, if we are on last page of de-duped results, // then we know our de-dup'd total is result.hits.length. long totalResults = result.hits.length < (p.start + p.hitsPerPage) ? result.hits.length : result.numRawHits; Document doc = new Document(); Element channel = OpenSearchHelper.startResponse(doc, p, request, totalResults); // Add hits to XML Document for (int i = p.start; i < end; i++) { org.apache.lucene.document.Document hit = result.searcher.doc(result.hits[i].id); Element item = JDOMHelper.add(channel, "item"); // Replace & and < with their XML entity counterparts to // ensure that any HTML markup in the snippet is escaped // before we do the highlighting. String title = hit.get("title"); if (title != null) { title = title.replaceAll("[&]", "&"); title = title.replaceAll("[<]", "<"); } JDOMHelper.add(item, "title", title); JDOMHelper.add(item, "link", hit.get("url")); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "docId", String.valueOf(result.hits[i].id)); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "score", String.valueOf(result.hits[i].score)); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "site", result.hits[i].site); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "length", hit.get("length")); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "type", hit.get("type")); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "boost", hit.get("boost")); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "collection", hit.get("collection")); String indexName = this.searcher.resolveIndexName(result.searcher, result.hits[i].id); JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "index", indexName); for (String date : hit.getValues("date")) { JDOMHelper.add(item, "date", date); } String raw = getContent(hit); StringBuilder buf = new StringBuilder(100); Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(), new NonBrokenHTMLEncoder(), new QueryScorer(q, "content")); CustomAnalyzer analyzer = new CustomAnalyzer(); analyzer.setFoldAccents(this.foldAccents); for (String snippet : highlighter.getBestFragments(analyzer, "content", raw, this.contextSnippetsPerResult)) { buf.append(snippet); buf.append("..."); } JDOMHelper.add(item, "description", buf.toString()); // Last, but not least, add a hit explanation, if enabled if (explain) { JDOMHelper.add(item, OpenSearchHelper.NS_ARCHIVE, "explain", result.searcher.explain(q, result.hits[i].id).toHtml()); } } OpenSearchHelper.addResponseTime(channel, System.nanoTime() - responseTime); long buildResultsTime = System.nanoTime(); OpenSearchHelper.writeResponse(doc, response, "application/rss+xml"); long writeResponseTime = System.nanoTime(); LOG.info("S: " + ((parseQueryTime - responseTime) / 1000 / 1000) + " " + ((executeQueryTime - parseQueryTime) / 1000 / 1000) + " " + ((buildResultsTime - executeQueryTime) / 1000 / 1000) + " " + ((writeResponseTime - buildResultsTime) / 1000 / 1000) + " " + p.query); } catch (Exception e) { throw new ServletException(e); } }
From source file:org.compass.core.lucene.engine.LuceneSearchEngineHighlighter.java
License:Apache License
protected Highlighter createHighlighter(String propertyName) throws SearchEngineException { Highlighter highlighter = new Highlighter(highlighterSettings.getFormatter(), highlighterSettings.getEncoder(), createScorer(propertyName)); Fragmenter f = highlighterSettings.getFragmenter(); highlighter.setTextFragmenter(f);//from w w w . j av a2 s.c om if (maxBytesToAnalyze == -1) { highlighter.setMaxDocBytesToAnalyze(highlighterSettings.getMaxBytesToAnalyze()); } else { highlighter.setMaxDocBytesToAnalyze(maxBytesToAnalyze); } return highlighter; }
From source file:org.jamwiki.search.LuceneSearchEngine.java
License:LGPL
/** * Find all documents that contain a specific search term, ordered by relevance. * This method supports all Lucene search query syntax. * * @param virtualWiki The virtual wiki for the topic. * @param text The search term being searched for. * @return A collection of SearchResultEntry objects for all documents that * contain the search term./*from w w w . j a va2 s.co m*/ */ public Collection findResults(String virtualWiki, String text) { StandardAnalyzer analyzer = new StandardAnalyzer(); Collection results = new Vector(); logger.fine("search text: " + text); IndexSearcher searcher = null; try { BooleanQuery query = new BooleanQuery(); QueryParser qp; qp = new QueryParser(ITYPE_TOPIC, analyzer); query.add(qp.parse(text), Occur.SHOULD); qp = new QueryParser(ITYPE_CONTENT, analyzer); query.add(qp.parse(text), Occur.SHOULD); searcher = new IndexSearcher(FSDirectory.getDirectory(getSearchIndexPath(virtualWiki))); // rewrite the query to expand it - required for wildcards to work with highlighter Query rewrittenQuery = searcher.rewrite(query); // actually perform the search Hits hits = searcher.search(rewrittenQuery); Highlighter highlighter = new Highlighter( new SimpleHTMLFormatter("<span class=\"highlight\">", "</span>"), new SimpleHTMLEncoder(), new QueryScorer(rewrittenQuery)); for (int i = 0; i < hits.length(); i++) { String summary = retrieveResultSummary(hits.doc(i), highlighter, analyzer); SearchResultEntry result = new SearchResultEntry(); result.setRanking(hits.score(i)); result.setTopic(hits.doc(i).get(ITYPE_TOPIC_PLAIN)); result.setSummary(summary); results.add(result); } } catch (Exception e) { logger.severe("Exception while searching for " + text, e); } finally { if (searcher != null) { try { searcher.close(); } catch (Exception e) { } } } return results; }
From source file:org.jamwiki.search.RankingAlgorithmSearchEngine.java
License:LGPL
/** * Find all documents that contain a specific search term, ordered by relevance. * This method supports all Lucene search query syntax. * * @param virtualWiki The virtual wiki for the topic. * @param text The search term being searched for. * @return A list of SearchResultEntry objects for all documents that * contain the search term.//w w w . j av a 2 s .c om */ public List<SearchResultEntry> findResults(String virtualWiki, String text, List<Integer> namespaces) { StandardAnalyzer analyzer = new StandardAnalyzer(USE_LUCENE_VERSION); List<SearchResultEntry> results = new ArrayList<SearchResultEntry>(); logger.trace("search text: " + text); try { IndexSearcher searcher = this.retrieveIndexSearcher(virtualWiki); Query query = this.createSearchQuery(searcher, analyzer, text, namespaces); // actually perform the search TopScoreDocCollector collector = TopScoreDocCollector.create(MAXIMUM_RESULTS_PER_SEARCH, true); Highlighter highlighter = new Highlighter( new SimpleHTMLFormatter("<span class=\"highlight\">", "</span>"), new SimpleHTMLEncoder(), new QueryScorer(query)); try { Class classRQ = Class.forName("com.transaxtions.search.rankingalgorithm.RankingQuery"); Class classQuery = Class.forName("org.apache.lucene.search.Query"); Object rq = classRQ.newInstance(); Class classArray[] = new Class[2]; classArray[0] = classQuery; classArray[1] = searcher.getClass(); Object args[] = new Object[2]; args[0] = query; args[1] = searcher; Method methodRQ_search = classRQ.getMethod("search", classArray); Object hitsobject = methodRQ_search.invoke(rq, args); Class classRH = hitsobject.getClass(); classArray = new Class[1]; classArray[0] = int.class; Method methodRH_length = classRH.getMethod("length", null); Method methodRH_docid = classRH.getMethod("docid", classArray); Method methodRH_score = classRH.getMethod("score", classArray); Object lenobject = methodRH_length.invoke(hitsobject); int length = ((Integer) lenobject).intValue(); for (int i = 0; i < length; i++) { args = new Object[1]; args[0] = new Integer(i); Object docobject = methodRH_docid.invoke(hitsobject, args); int docId = ((Integer) docobject).intValue(); Document doc = searcher.doc(docId); String summary = retrieveResultSummary(doc, highlighter, analyzer); Object scoreobject = methodRH_score.invoke(hitsobject, args); float score = ((Float) scoreobject).floatValue(); SearchResultEntry result = new SearchResultEntry(doc.get(FIELD_TOPIC_NAME), score, summary); results.add(result); } } catch (Throwable t) { logger.error("Failure while executing RankingAlgorithm search", t); } } catch (Exception e) { logger.error("Exception while searching for " + text, e); } return results; }
From source file:org.lukhnos.lucenestudy.HighlightingHelper.java
License:MIT License
HighlightingHelper(Query query, Analyzer analyzer) {
this.analyzer = analyzer;
Formatter formatter = new SimpleHTMLFormatter();
Encoder encoder = new MinimalHTMLEncoder();
scorer = new QueryScorer(query);
highlighter = new Highlighter(formatter, encoder, scorer);
fragmentLength = DEFAULT_FRAGMENT_LENGTH;
Fragmenter fragmenter = new SimpleSpanFragmenter(scorer, fragmentLength);
highlighter.setTextFragmenter(fragmenter);
}
From source file:org.riotfamily.search.ResultHighlighter.java
License:Apache License
public HighlightingContext createContext(IndexSearcher indexSearcher, Query query) throws IOException { Scorer scorer = new QueryScorer(indexSearcher.rewrite(query)); if (formatter == null) { formatter = new SimpleHTMLFormatter("<" + highlightPreTag + ">", "</" + highlightPostTag + ">"); }//from w w w . j a v a 2 s . c om if (fragmenter == null) { fragmenter = new SimpleFragmenter(fragmentSize); } Highlighter highlighter = new Highlighter(formatter, encoder, scorer); highlighter.setTextFragmenter(fragmenter); return new HighlightingContext(highlighter); }
From source file:org.sakaiproject.search.component.service.impl.SearchResultImpl.java
License:Educational Community License
public String getSearchResult() { try {//from w w w . j a va 2 s. co m Scorer scorer = new QueryScorer(query); Highlighter hightlighter = new Highlighter(new SimpleHTMLFormatter(), new SimpleHTMLEncoder(), scorer); StringBuilder sb = new StringBuilder(); // contents no longer contains the digested contents, so we need to // fetch it from the EntityContentProducer byte[][] references = doc.getBinaryValues(SearchService.FIELD_REFERENCE); DigestStorageUtil digestStorageUtil = new DigestStorageUtil(searchService); if (references != null && references.length > 0) { for (int i = 0; i < references.length; i++) { EntityContentProducer sep = searchIndexBuilder .newEntityContentProducer(CompressionTools.decompressString(references[i])); if (sep != null) { //does this ecp store on the FS? if (sep instanceof StoredDigestContentProducer) { String digestCount = doc.get(SearchService.FIELD_DIGEST_COUNT); if (digestCount == null) { digestCount = "1"; } log.debug("This file possibly has FS digests with index of " + digestCount); StringBuilder sb1 = digestStorageUtil.getFileContents(CompressionTools.decompressString( doc.getBinaryValue(SearchService.FIELD_REFERENCE)), digestCount); if (sb1.length() > 0) { sb.append(sb1); } else { String digest = sep.getContent(CompressionTools.decompressString(references[i])); sb.append(digest); //we need to save this digestStorageUtil.saveContentToStore( CompressionTools.decompressString( doc.getBinaryValue(SearchService.FIELD_REFERENCE)), sb.toString(), 1); } } else { sb.append(CompressionTools.decompressString(references[i])); } } } } String text = sb.toString(); TokenStream tokenStream = analyzer.tokenStream(SearchService.FIELD_CONTENTS, new StringReader(text)); return hightlighter.getBestFragments(tokenStream, text, 5, " ... "); //$NON-NLS-1$ } catch (IOException e) { return Messages.getString("SearchResultImpl.2") + e.getMessage(); //$NON-NLS-1$ } catch (InvalidTokenOffsetsException e) { return Messages.getString("SearchResultResponseImpl.11") + e.getMessage(); } catch (DataFormatException e) { e.printStackTrace(); return Messages.getString("SearchResultResponseImpl.11") + e.getMessage(); } }
From source file:org.sakaiproject.search.component.service.impl.SearchResultResponseImpl.java
License:Educational Community License
public String getSearchResult() { try {// www.j a va 2 s. co m Scorer scorer = new QueryScorer(query); Highlighter hightlighter = new Highlighter(new SimpleHTMLFormatter(), new SimpleHTMLEncoder(), scorer); StringBuilder sb = new StringBuilder(); // contents no longer contains the digested contents, so we need to // fetch it from the EntityContentProducer EntityContentProducer sep = searchIndexBuilder.newEntityContentProducer(getReference()); if (sep != null) { sb.append(sep.getContent(getReference())); } String text = sb.toString(); TokenStream tokenStream = analyzer.tokenStream(SearchService.FIELD_CONTENTS, new StringReader(text)); return hightlighter.getBestFragments(tokenStream, text, 5, " ... "); //$NON-NLS-1$ } catch (IOException e) { return Messages.getString("SearchResultResponseImpl.11") + e.getMessage(); //$NON-NLS-1$ } catch (InvalidTokenOffsetsException e) { return Messages.getString("SearchResultResponseImpl.11") + e.getMessage(); } }