List of usage examples for org.apache.lucene.search.highlight Highlighter getBestFragment
public final String getBestFragment(TokenStream tokenStream, String text) throws IOException, InvalidTokenOffsetsException
From source file:net.skyatlas.icd.test.AnsegTest.java
private String toHighlighter(Analyzer analyzer, Query query, Document doc) throws InvalidTokenOffsetsException { String field = "text"; try {//w w w .j a v a2s .com SimpleHTMLFormatter simpleHtmlFormatter = new SimpleHTMLFormatter("<font color=\"red\">", "</font>"); Highlighter highlighter = new Highlighter(simpleHtmlFormatter, new QueryScorer(query)); TokenStream tokenStream1 = analyzer.tokenStream("text", new StringReader(doc.get(field))); String highlighterStr = highlighter.getBestFragment(tokenStream1, doc.get(field)); return highlighterStr == null ? doc.get(field) : highlighterStr; } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } catch (InvalidTokenOffsetsException e) { // TODO Auto-generated catch block e.printStackTrace(); } return null; }
From source file:org.apache.nutch.searcher.Summarizer.java
License:Apache License
public static String getsummary(String queryString, String content, Analyzer analyzer) { if (queryString == null && content != null) { if (content.length() > SUM_LENGTH) return content.substring(0, (SUM_LENGTH) - 1); else// w w w .j a va2 s . co m return content; } else if (queryString != null && content == null) return ""; else if (queryString == null && content == null) return ""; SimpleHTMLFormatter sHtmlF = new SimpleHTMLFormatter(cssfront, cssend); org.apache.lucene.search.Query summarizerQuery = null; QueryParser queryParse = new QueryParser("content", analyzer); try { summarizerQuery = queryParse.parse(queryString); } catch (ParseException ex) { if (content.length() > SUM_LENGTH) return content.substring(0, (SUM_LENGTH) - 1); else return content; } QueryScorer qs = new QueryScorer(summarizerQuery); Highlighter highlighter = new Highlighter(sHtmlF, qs); highlighter.setTextFragmenter(new SimpleFragmenter(SUM_LENGTH)); TokenStream tokenStream = analyzer.tokenStream("content", new StringReader(content)); String str; try { str = highlighter.getBestFragment(tokenStream, content); } catch (IOException e) { str = null; } if (str == null) { if (content.length() > SUM_LENGTH) str = content.substring(0, (SUM_LENGTH) - 1); else str = content; } return str; }
From source file:org.compass.core.lucene.engine.LuceneSearchEngineHighlighter.java
License:Apache License
public String fragment(Resource resource, String propertyName, String text) throws SearchEngineException { Highlighter highlighter = createHighlighter(propertyName); TokenStream tokenStream = createTokenStream(resource, propertyName, text); try {//w w w . j a va2s . c o m return highlighter.getBestFragment(tokenStream, text); } catch (IOException e) { throw new SearchEngineException("Failed to highlight fragments for alias [" + resource.getAlias() + "] and property [" + propertyName + "]"); } }
From source file:org.compass.core.lucene.engine.LuceneSearchEngineHighlighter.java
License:Apache License
public String[] multiValueFragment(Resource resource, String propertyName, String[] texts) throws SearchEngineException { List fragmentList = new ArrayList(); Highlighter highlighter = createHighlighter(propertyName); for (int i = 0; i < texts.length; i++) { String text = texts[i];/*from w w w .j a v a 2 s. c om*/ if (text != null && text.length() > 0) { //TokenStream tokenStream = createTokenStream(resource, propertyName, text); // We have to re-analyze one field value at a time TokenStream tokenStream = createTokenStreamFromAnalyzer(propertyName, text); try { String fragment = highlighter.getBestFragment(tokenStream, text); if (fragment != null && fragment.length() > 0) { fragmentList.add(fragment); } } catch (IOException e) { throw new SearchEngineException("Failed to highlight fragments for alias [" + resource.getAlias() + "] and property [" + propertyName + "]"); } } } return (String[]) fragmentList.toArray(new String[fragmentList.size()]); }
From source file:org.elasticsearch.search.fetch.subphase.highlight.PlainHighlighterTests.java
License:Apache License
public void checkGeoQueryHighlighting(Query geoQuery) throws IOException, InvalidTokenOffsetsException { Map analysers = new HashMap<String, Analyzer>(); analysers.put("text", new StandardAnalyzer()); FieldNameAnalyzer fieldNameAnalyzer = new FieldNameAnalyzer(analysers); Query termQuery = new TermQuery(new Term("text", "failure")); Query boolQuery = new BooleanQuery.Builder().add(new BooleanClause(geoQuery, BooleanClause.Occur.SHOULD)) .add(new BooleanClause(termQuery, BooleanClause.Occur.SHOULD)).build(); org.apache.lucene.search.highlight.Highlighter highlighter = new org.apache.lucene.search.highlight.Highlighter( new CustomQueryScorer(boolQuery)); String fragment = highlighter.getBestFragment( fieldNameAnalyzer.tokenStream("text", "Arbitrary text field which should not cause " + "a failure"), "Arbitrary text field which should not cause a failure"); assertThat(fragment, equalTo("Arbitrary text field which should not cause a <B>failure</B>")); // TODO: This test will fail if we pass in an instance of GeoPointInBBoxQueryImpl too. Should we also find a way to work around that // or can the query not be rewritten before it is passed into the highlighter? }
From source file:org.elasticsearch.search.highlight.PlainHighlighterTests.java
License:Apache License
public void checkGeoQueryHighlighting(Query geoQuery) throws IOException, InvalidTokenOffsetsException { Map analysers = new HashMap<String, Analyzer>(); analysers.put("text", new StandardAnalyzer()); FieldNameAnalyzer fieldNameAnalyzer = new FieldNameAnalyzer(analysers); Query termQuery = new TermQuery(new Term("text", "failure")); Query boolQuery = new BooleanQuery.Builder().add(new BooleanClause(geoQuery, BooleanClause.Occur.SHOULD)) .add(new BooleanClause(termQuery, BooleanClause.Occur.SHOULD)).build(); org.apache.lucene.search.highlight.Highlighter highlighter = new org.apache.lucene.search.highlight.Highlighter( new CustomQueryScorer(boolQuery)); String fragment = highlighter.getBestFragment( fieldNameAnalyzer.tokenStream("text", "Arbitrary text field which should not cause " + "a failure"), "Arbitrary text field which should not cause a failure"); assertThat(fragment, equalTo("Arbitrary text field which should not cause a <B>failure</B>")); Query rewritten = boolQuery.rewrite(null); highlighter = new org.apache.lucene.search.highlight.Highlighter(new CustomQueryScorer(rewritten)); fragment = highlighter.getBestFragment( fieldNameAnalyzer.tokenStream("text", "Arbitrary text field which should not cause " + "a failure"), "Arbitrary text field which should not cause a failure"); assertThat(fragment, equalTo("Arbitrary text field which should not cause a <B>failure</B>")); }
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 ww .ja va2s . c om*/ * @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.zilverline.service.SearchServiceImpl.java
License:Open Source License
/** * Helper that creates a Result from a Document. * /* w w w . j a v a 2 s. c o m*/ * @param doc the Document * @param score the score of the Document in the hit * @param hl the Highlighter used * @param an the Analyzer used * * @return Result the resulting object that is used in the model. */ private Result doc2ResultHelper(final Document doc, final float score, final Highlighter hl, final Analyzer an) { String docTitle = doc.get("title"); String docName = doc.get("name"); String docPath = doc.get("path"); String zipName = doc.get("zipName"); if (log.isDebugEnabled()) { log.debug("Preparing result " + docName + ":" + zipName); } String zipPath = doc.get("zipPath"); String docURL = ""; String docCache = ""; String docCollection = doc.get("collection"); // get the collection DocumentCollection thisCollection = collectionManager.getCollectionByName(docCollection); if (thisCollection != null) { docURL = thisCollection.getUrlDefault(); if (thisCollection.isKeepCacheWithManagerDefaults()) { docCache = thisCollection.getCacheUrlWithManagerDefaults(); } } else { log.error("Unknown collection '" + docCollection + "' found, can not find its URL."); } // get the modification date, and convert it to a readable form date is stored as (yyyyMMdd) DateFormat df1 = new SimpleDateFormat("yyyyMMdd"); Date docDate = null; try { docDate = df1.parse(doc.get("modified")); } catch (ParseException e) { log.debug("Invalid date retrieved, trying backward compatibility with v 1.0-rc3-patch1"); // backward compatibility with v 1.0-rc3-patch1 storage of date: // Keyword<modified:0cee68g00> docDate = DateField.stringToDate(doc.get("modified")); if (docDate == null) { log.warn("Invalid date retrieved, returning epoch (1970) for " + docName); docDate = new Date(0); } } String docSize = doc.get("size"); String docType = doc.get("type"); String docISBN = doc.get("isbn"); // use the name if it has no title // TODO this logic could go into Result if ((docTitle == null) || docTitle.equals("")) { if ((zipName == null) || zipName.equals("")) { docTitle = docName; } else { docTitle = zipName; } } // then make a Result Result thisResult = new Result(); // highlight the title with search terms String highlightedText; TokenStream tokenStream = an.tokenStream("title", new StringReader(docTitle)); try { highlightedText = hl.getBestFragment(tokenStream, docTitle); if ((highlightedText != null) && (highlightedText.length() > 0)) { docTitle = highlightedText; } } catch (IOException e1) { log.warn("Can't highlight " + docTitle, e1); } thisResult.setTitle(docTitle); // highlight the name with search terms /* * tokenStream = an.tokenStream("title", new StringReader(docName)); try { highlightedText = hl.getBestFragment(tokenStream, * docName); log.debug("name after highlighting: " + highlightedText); if (highlightedText != null && * highlightedText.length() > 0) { docName = highlightedText; } } catch (IOException e1) { log.warn("Can't highlight " + * docName, e1); } */ thisResult.setName(docName); thisResult.setCollection(docCollection); thisResult.setPath(docPath); thisResult.setURL(docURL); thisResult.setCache(docCache); thisResult.setZipName(zipName); thisResult.setZipPath(zipPath); thisResult.setScore(score); thisResult.setISBN(docISBN); String text = doc.get("summary"); if (text == null) { text = ""; } // highlight the summary with search terms tokenStream = an.tokenStream("summary", new StringReader(text)); try { highlightedText = hl.getBestFragment(tokenStream, text); if ((highlightedText != null) && (highlightedText.length() > 0)) { text = highlightedText; } } catch (IOException e1) { log.warn("Can't highlight " + text, e1); } thisResult.setSummary(text); thisResult.setModificationDate(docDate); thisResult.setSize(docSize); thisResult.setType(docType); return thisResult; }