org.apache.lucene.queryParser.MultiFieldQueryParser Java Examples

private org.apache.lucene.search.Query createFullTextQuery(final String[] searchFields, final QueryFilter queryFilter,
    final String searchString)
{
  final MultiFieldQueryParser parser = new MultiFieldQueryParser(LUCENE_VERSION, searchFields, new ClassicAnalyzer(Version.LUCENE_31));
  parser.setAllowLeadingWildcard(true);
  org.apache.lucene.search.Query query = null;
  try {
    query = parser.parse(searchString);
  } catch (final org.apache.lucene.queryParser.ParseException ex) {
    final String errorMsg = "Lucene error message: "
        + ex.getMessage()
        + " (for "
        + this.getClass().getSimpleName()
        + ": "
        + searchString
        + ").";
    if (queryFilter != null) {
      queryFilter.setErrorMessage(errorMsg);
    }
    log.info(errorMsg);
    return null;
  }
  return query;
}
 

public IndexSearcher() {
    try {
        searcher = new org.apache.lucene.search.IndexSearcher(new ClasspathDirectory());
    } catch (IOException e) {
        e.printStackTrace();
    }
    analyzer = new StandardAnalyzer(Version.LUCENE_31);
    parser = new MultiFieldQueryParser(Version.LUCENE_31, new String[]{"name","description"}, analyzer);
}
 

public IndexSearcher() {
    try {
        searcher = new org.apache.lucene.search.IndexSearcher(new ClasspathDirectory());
    } catch (IOException e) {
        e.printStackTrace();
    }
    analyzer = new StandardAnalyzer(Version.LUCENE_31);
    parser = new MultiFieldQueryParser(Version.LUCENE_31, new String[]{"name","description"}, analyzer);
}
 

public List<Course> getCourses(String query) {
    try {
        List<Course> qlist = new ArrayList<Course>();
        IndexSearcher indexSearcher = new IndexSearcher(INDEXPATH);
        long begin = new Date().getTime();
        //下面的是进行title,content 两个范围内进行收索. SHOULD 表示OR
        BooleanClause.Occur[] clauses = {BooleanClause.Occur.SHOULD, BooleanClause.Occur.SHOULD};
        Query queryOBJ = MultiFieldQueryParser.parse(query, new String[]{"courseIntro", "courseTitle"}, clauses, new StandardAnalyzer());//parser.parse(query);
        Filter filter = null;
        //################# 搜索相似度最高的记录 ###################
        TopDocs topDocs = indexSearcher.search(queryOBJ, filter, 1000);
        Course course = null;

        //输出结果
        for (ScoreDoc scoreDoc : topDocs.scoreDocs) {
            Document targetDoc = indexSearcher.doc(scoreDoc.doc);
            course = new Course();
            String courseIntro = targetDoc.get("courseIntro");
            String courseTitle = targetDoc.get("courseTitle");
            String courseId = targetDoc.get("courseId");
            TokenStream contentTokenStream = analyzer.tokenStream("courseIntro", new StringReader(courseIntro));
            TokenStream titleTokenStream = analyzer.tokenStream("courseTitle", new StringReader(courseTitle));
            course.setCourseIntro(courseIntro);
            course.setCourseTitle(courseTitle);
            course.setCourseId(courseId);
            course.setType(targetDoc.get("type"));
            course.setCourseState(targetDoc.get("courseState"));
            qlist.add(course);
        }
        indexSearcher.close();
        return qlist;
    } catch (Exception e) {
        logger.error("getCourses error.");
        return null;
    }
}
 

public SearchResult search(SearchCriteria criteria, List<MusicFolder> musicFolders, IndexType indexType) {
    SearchResult result = new SearchResult();
    int offset = criteria.getOffset();
    int count = criteria.getCount();
    result.setOffset(offset);

    IndexReader reader = null;
    try {
        reader = createIndexReader(indexType);
        Searcher searcher = new IndexSearcher(reader);
        Analyzer analyzer = new SubsonicAnalyzer();

        MultiFieldQueryParser queryParser = new MultiFieldQueryParser(LUCENE_VERSION, indexType.getFields(), analyzer, indexType.getBoosts());

        BooleanQuery query = new BooleanQuery();
        query.add(queryParser.parse(analyzeQuery(criteria.getQuery())), BooleanClause.Occur.MUST);

        List<SpanTermQuery> musicFolderQueries = new ArrayList<SpanTermQuery>();
        for (MusicFolder musicFolder : musicFolders) {
            if (indexType == ALBUM_ID3 || indexType == ARTIST_ID3) {
                musicFolderQueries.add(new SpanTermQuery(new Term(FIELD_FOLDER_ID, NumericUtils.intToPrefixCoded(musicFolder.getId()))));
            } else {
                musicFolderQueries.add(new SpanTermQuery(new Term(FIELD_FOLDER, musicFolder.getPath().getPath())));
            }
        }
        query.add(new SpanOrQuery(musicFolderQueries.toArray(new SpanQuery[musicFolderQueries.size()])), BooleanClause.Occur.MUST);

        TopDocs topDocs = searcher.search(query, null, offset + count);
        result.setTotalHits(topDocs.totalHits);

        int start = Math.min(offset, topDocs.totalHits);
        int end = Math.min(start + count, topDocs.totalHits);
        for (int i = start; i < end; i++) {
            Document doc = searcher.doc(topDocs.scoreDocs[i].doc);
            switch (indexType) {
                case SONG:
                case ARTIST:
                case ALBUM:
                    MediaFile mediaFile = mediaFileService.getMediaFile(Integer.valueOf(doc.get(FIELD_ID)));
                    addIfNotNull(mediaFile, result.getMediaFiles());
                    break;
                case ARTIST_ID3:
                    Artist artist = artistDao.getArtist(Integer.valueOf(doc.get(FIELD_ID)));
                    addIfNotNull(artist, result.getArtists());
                    break;
                case ALBUM_ID3:
                    Album album = albumDao.getAlbum(Integer.valueOf(doc.get(FIELD_ID)));
                    addIfNotNull(album, result.getAlbums());
                    break;
                default:
                    break;
            }
        }

    } catch (Throwable x) {
        LOG.error("Failed to execute Lucene search.", x);
    } finally {
        FileUtil.closeQuietly(reader);
    }
    return result;
}
 

/**
 * Do search a certain query. The results will be filtered for the identity and roles.
 * 
 * @param queryString
 *            Search query-string.
 * @param identity
 *            Filter results for this identity (user).
 * @param roles
 *            Filter results for this roles (role of user).
 * @return SearchResults object for this query
 */
@Override
public SearchResults doSearch(final String queryString, final List<String> condQueries, final Identity identity, final Roles roles, final int firstResult,
        final int maxResults, final boolean doHighlighting) throws ServiceNotAvailableException, ParseException {
    try {
        if (!existIndex()) {
            log.warn("Index does not exist, can't search for queryString: " + queryString);
            throw new ServiceNotAvailableException("Index does not exist");
        }
        synchronized (createIndexSearcherLock) {// o_clusterOK by:fj if service is only configured on one vm, which is recommended way
            if (searcher == null) {
                try {
                    createIndexSearcher(indexPath);
                    checkIsIndexUpToDate();
                } catch (final IOException ioEx) {
                    log.warn("Can not create searcher", ioEx);
                    throw new ServiceNotAvailableException("Index is not available");
                }
            }
            if (hasNewerIndexFile()) {
                reopenIndexSearcher();
                checkIsIndexUpToDate();
            }
        }
        log.info("queryString=" + queryString);

        final BooleanQuery query = new BooleanQuery();
        if (StringHelper.containsNonWhitespace(queryString)) {
            final QueryParser queryParser = new MultiFieldQueryParser(Version.LUCENE_CURRENT, fields, analyzer);
            queryParser.setLowercaseExpandedTerms(false);// some add. fields are not tokenized and not lowered case
            final Query multiFieldQuery = queryParser.parse(queryString.toLowerCase());
            query.add(multiFieldQuery, Occur.MUST);
        }

        if (condQueries != null && !condQueries.isEmpty()) {
            for (final String condQueryString : condQueries) {
                final QueryParser condQueryParser = new QueryParser(Version.LUCENE_CURRENT, condQueryString, analyzer);
                condQueryParser.setLowercaseExpandedTerms(false);
                final Query condQuery = condQueryParser.parse(condQueryString);
                query.add(condQuery, Occur.MUST);
            }
        }

        if (log.isDebugEnabled()) {
            log.debug("query=" + query);
        }
        // TODO: 14.06.2010/cg : fellowig cide fragment can be removed later, do no longer call rewrite(query) because wildcard-search problem (OLAT-5359)
        // Query query = null;
        // try {
        // query = searcher.rewrite(query);
        // log.debug("after 'searcher.rewrite(query)' query=" + query);
        // } catch (Exception ex) {
        // throw new QueryException("Rewrite-Exception query because too many clauses. Query=" + query);
        // }
        final long startTime = System.currentTimeMillis();
        final int n = SearchServiceFactory.getService().getSearchModuleConfig().getMaxHits();
        final TopDocs docs = searcher.search(query, n);
        final long queryTime = System.currentTimeMillis() - startTime;
        if (log.isDebugEnabled()) {
            log.debug("hits.length()=" + docs.totalHits);
        }
        final SearchResultsImpl searchResult = new SearchResultsImpl(mainIndexer, searcher, docs, query, analyzer, identity, roles, firstResult, maxResults,
                doHighlighting);
        searchResult.setQueryTime(queryTime);
        searchResult.setNumberOfIndexDocuments(searcher.maxDoc());
        queryCount++;
        return searchResult;
    } catch (final ServiceNotAvailableException naex) {
        // pass exception
        throw new ServiceNotAvailableException(naex.getMessage());
    } catch (final ParseException pex) {
        throw new ParseException("can not parse query=" + queryString);
    } catch (final Exception ex) {
        log.warn("Exception in search", ex);
        throw new ServiceNotAvailableException(ex.getMessage());
    }
}
 

/**
 * @see org.olat.lms.search.SearchService#doSearch(String, List, Identity, Roles, int, int, boolean)
 */
public SearchResults doSearch(final String queryString, final List<String> condQueries, final Identity identity, final Roles roles, final int firstResult,
        final int maxResults, final boolean doHighlighting) throws ServiceNotAvailableException, QueryException {

    synchronized (createIndexSearcherLock) {// o_clusterOK by:fj if service is only configured on one vm, which is recommended way
        if (searcher == null) {
            log.warn("Index does not exist, can't search for queryString: " + queryString);
            throw new ServiceNotAvailableException("Index does not exist");
        }
    }

    try {
        log.info("queryString=" + queryString);

        final BooleanQuery query = new BooleanQuery();
        if (StringHelper.containsNonWhitespace(queryString)) {
            final QueryParser queryParser = new MultiFieldQueryParser(Version.LUCENE_30, fields, analyzer);
            queryParser.setLowercaseExpandedTerms(false);// some add. fields are not tokenized and not lowered case
            final Query multiFieldQuery = queryParser.parse(queryString.toLowerCase());
            query.add(multiFieldQuery, Occur.MUST);
        }

        if (condQueries != null && !condQueries.isEmpty()) {
            for (final String condQueryString : condQueries) {
                final QueryParser condQueryParser = new QueryParser(Version.LUCENE_30, condQueryString, analyzer);
                condQueryParser.setLowercaseExpandedTerms(false);
                final Query condQuery = condQueryParser.parse(condQueryString);
                query.add(condQuery, Occur.MUST);
            }
        }

        if (log.isDebugEnabled()) {
            log.debug("query=" + query);
        }
        // TODO: 14.06.2010/cg : fellowig cide fragment can be removed later, do no longer call rewrite(query) because wildcard-search problem (OLAT-5359)
        // Query query = null;
        // try {
        // query = searcher.rewrite(query);
        // log.debug("after 'searcher.rewrite(query)' query=" + query);
        // } catch (Exception ex) {
        // throw new QueryException("Rewrite-Exception query because too many clauses. Query=" + query);
        // }
        final long startTime = System.currentTimeMillis();
        final int n = SearchServiceFactory.getService().getSearchModuleConfig().getMaxHits();
        final TopDocs docs = searcher.search(query, n);
        final long queryTime = System.currentTimeMillis() - startTime;
        if (log.isDebugEnabled()) {
            log.debug("hits.length()=" + docs.totalHits);
        }
        final SearchResultsImpl searchResult = new SearchResultsImpl(mainIndexer, searcher, docs, query, analyzer, identity, roles, firstResult, maxResults,
                doHighlighting);
        searchResult.setQueryTime(queryTime);
        searchResult.setNumberOfIndexDocuments(searcher.maxDoc());
        queryCount++;
        return searchResult;
    } catch (final ParseException pex) {
        throw new QueryException("can not parse query=" + queryString);
    } catch (final Exception ex) {
        log.warn("Exception in search", ex);
        throw new ServiceNotAvailableException(ex.getMessage());
    }
}
 

public static void main(String[] args) throws Exception{
		ApplicationContext applicationContext=new ClassPathXmlApplicationContext("applicationContext.xml");
		SessionFactory sessionFactory = applicationContext.getBean("hibernate4sessionFactory",SessionFactory.class);
		FullTextSession fullTextSession = Search.getFullTextSession(sessionFactory.openSession());
		
		//使用Hibernate Search api查询 从多个字段匹配 name、description、authors.name
//		QueryBuilder qb = fullTextEntityManager.getSearchFactory().buildQueryBuilder().forEntity(Book.class ).get();
//		Query luceneQuery = qb.keyword().onFields("name","description","authors.name").matching("移动互联网").createQuery();
		
		//使用lucene api查询 从多个字段匹配 name、description、authors.name
		//使用庖丁分词器
		MultiFieldQueryParser queryParser=new MultiFieldQueryParser(Version.LUCENE_36, new String[]{"name","description","authors.name"}, new PaodingAnalyzer());
		Query luceneQuery=queryParser.parse("实战");
		
		FullTextQuery fullTextQuery =fullTextSession.createFullTextQuery(luceneQuery, Book.class);
		//设置每页显示多少条
		fullTextQuery.setMaxResults(5);
		//设置当前页
		fullTextQuery.setFirstResult(0);
		
		//高亮设置
		SimpleHTMLFormatter formatter=new SimpleHTMLFormatter("<b><font color='red'>", "<font/></b>");
		QueryScorer queryScorer=new QueryScorer(luceneQuery);
		Highlighter highlighter=new Highlighter(formatter, queryScorer);

		@SuppressWarnings("unchecked")
		List<Book> resultList = fullTextQuery.list();
		System.out.println("共查找到["+resultList.size()+"]条记录");
		for (Book book : resultList) {
			String highlighterString=null;
			Analyzer analyzer=new PaodingAnalyzer();
			try {
				//高亮name
				highlighterString=highlighter.getBestFragment(analyzer, "name", book.getName());
				if(highlighterString!=null){
					book.setName(highlighterString);
				}
				//高亮authors.name
				Set<Author> authors = book.getAuthors();
				for (Author author : authors) {
					highlighterString=highlighter.getBestFragment(analyzer, "authors.name", author.getName());
					if(highlighterString!=null){
						author.setName(highlighterString);
					}
				}
				//高亮description
				highlighterString=highlighter.getBestFragment(analyzer, "description", book.getDescription());
				if(highlighterString!=null){
					book.setDescription(highlighterString);
				}
			} catch (Exception e) {
			}
			
			System.out.println("书名:"+book.getName()+"\n描述:"+book.getDescription()+"\n出版日期:"+book.getPublicationDate());
			System.out.println("----------------------------------------------------------");
		}
		
		fullTextSession.close();
		sessionFactory.close();
		
	}
 

@Override
public QueryResult<Book> query(String keyword, int start, int pagesize,Analyzer analyzer,String...field) throws Exception{
	
	QueryResult<Book> queryResult=new QueryResult<Book>();
	
	List<Book> books=new ArrayList<Book>();
	
	FullTextSession fullTextSession = Search.getFullTextSession(getSession());
	
	//使用Hibernate Search api查询 从多个字段匹配 name、description、authors.name
	//QueryBuilder qb = fullTextSession.getSearchFactory().buildQueryBuilder().forEntity(Book.class ).get();
	//Query luceneQuery = qb.keyword().onFields(field).matching(keyword).createQuery();

	//使用lucene api查询 从多个字段匹配 name、description、authors.name
	
	MultiFieldQueryParser queryParser=new MultiFieldQueryParser(Version.LUCENE_36,new String[]{"name","description","authors.name"}, analyzer);
	Query luceneQuery=queryParser.parse(keyword);
	
	FullTextQuery fullTextQuery = fullTextSession.createFullTextQuery(luceneQuery);
	int searchresultsize = fullTextQuery.getResultSize();
	queryResult.setSearchresultsize(searchresultsize);
	System.out.println("共查找到["+searchresultsize+"]条记录");
	
	fullTextQuery.setFirstResult(start);
	fullTextQuery.setMaxResults(pagesize);
	
	//设置按id排序
	fullTextQuery.setSort(new Sort(new SortField("id", SortField.INT ,true)));
	
	//高亮设置
	SimpleHTMLFormatter formatter=new SimpleHTMLFormatter("<b><font color='red'>", "</font></b>");
	QueryScorer queryScorer=new QueryScorer(luceneQuery);
	Highlighter highlighter=new Highlighter(formatter, queryScorer);

	@SuppressWarnings("unchecked")
	List<Book> tempresult = fullTextQuery.list();
	for (Book book : tempresult) {
		String highlighterString=null;
		try {
			//高亮name
			highlighterString=highlighter.getBestFragment(analyzer, "name", book.getName());
			if(highlighterString!=null){
				book.setName(highlighterString);
			}
			//高亮authors.name
			Set<Author> authors = book.getAuthors();
			for (Author author : authors) {
				highlighterString=highlighter.getBestFragment(analyzer, "authors.name", author.getName());
				if(highlighterString!=null){
					author.setName(highlighterString);
				}
			}
			//高亮description
			highlighterString=highlighter.getBestFragment(analyzer, "description", book.getDescription());
			if(highlighterString!=null){
				book.setDescription(highlighterString);
			}
		} catch (Exception e) {
		}
		
		books.add(book);
		
		
		System.out.println("书名:"+book.getName()+"\n描述:"+book.getDescription()+"\n出版日期:"+book.getPublicationDate());
		System.out.println("----------------------------------------------------------");
	}
	
	queryResult.setSearchresult(books);
	
	return queryResult;
}
 

public static void main(String[] args) throws Exception{
		ApplicationContext applicationContext=new ClassPathXmlApplicationContext("applicationContext.xml");
		EntityManagerFactory entityManagerFactory = applicationContext.getBean("entityManagerFactory",EntityManagerFactory.class);
		FullTextEntityManager fullTextEntityManager = Search.getFullTextEntityManager(entityManagerFactory.createEntityManager());
		
		//使用Hibernate Search api查询 从多个字段匹配 name、description、authors.name
//		QueryBuilder qb = fullTextEntityManager.getSearchFactory().buildQueryBuilder().forEntity(Book.class ).get();
//		Query luceneQuery = qb.keyword().onFields("name","description","authors.name").matching("移动互联网").createQuery();
		
		//使用lucene api查询 从多个字段匹配 name、description、authors.name
		//使用庖丁分词器
		MultiFieldQueryParser queryParser=new MultiFieldQueryParser(Version.LUCENE_36, new String[]{"name","description","authors.name"}, new PaodingAnalyzer());
		Query luceneQuery=queryParser.parse("实战");
		
		FullTextQuery fullTextQuery =fullTextEntityManager.createFullTextQuery(luceneQuery, Book.class);
		//设置每页显示多少条
		fullTextQuery.setMaxResults(5);
		//设置当前页
		fullTextQuery.setFirstResult(0);
		
		//高亮设置
		SimpleHTMLFormatter formatter=new SimpleHTMLFormatter("<b><font color='red'>", "<font/></b>");
		QueryScorer queryScorer=new QueryScorer(luceneQuery);
		Highlighter highlighter=new Highlighter(formatter, queryScorer);

		@SuppressWarnings("unchecked")
		List<Book> resultList = fullTextQuery.getResultList();
		
		for (Book book : resultList) {
			String highlighterString=null;
			Analyzer analyzer=new PaodingAnalyzer();
			try {
				//高亮name
				highlighterString=highlighter.getBestFragment(analyzer, "name", book.getName());
				if(highlighterString!=null){
					book.setName(highlighterString);
				}
				//高亮authors.name
				Set<Author> authors = book.getAuthors();
				for (Author author : authors) {
					highlighterString=highlighter.getBestFragment(analyzer, "authors.name", author.getName());
					if(highlighterString!=null){
						author.setName(highlighterString);
					}
				}
				//高亮description
				highlighterString=highlighter.getBestFragment(analyzer, "description", book.getDescription());
				if(highlighterString!=null){
					book.setDescription(highlighterString);
				}
			} catch (Exception e) {
			}
			
		}
		
		fullTextEntityManager.close();
		entityManagerFactory.close();
		
	}
 

@Override
public QueryResult<Book> query(String keyword, int start, int pagesize,Analyzer analyzer,String...field) throws Exception{
	
	QueryResult<Book> queryResult=new QueryResult<Book>();
	
	List<Book> books=new ArrayList<Book>();
	
	FullTextEntityManager fullTextEntityManager = Search.getFullTextEntityManager(em);
	
	//使用Hibernate Search api查询 从多个字段匹配 name、description、authors.name
	//QueryBuilder qb = fullTextSession.getSearchFactory().buildQueryBuilder().forEntity(Book.class ).get();
	//Query luceneQuery = qb.keyword().onFields(field).matching(keyword).createQuery();

	//使用lucene api查询 从多个字段匹配 name、description、authors.name
	
	MultiFieldQueryParser queryParser=new MultiFieldQueryParser(Version.LUCENE_36,new String[]{"name","description","authors.name"}, analyzer);
	Query luceneQuery=queryParser.parse(keyword);
	
	FullTextQuery fullTextQuery = fullTextEntityManager.createFullTextQuery(luceneQuery);
	int searchresultsize = fullTextQuery.getResultSize();
	queryResult.setSearchresultsize(searchresultsize);
			
	fullTextQuery.setFirstResult(start);
	fullTextQuery.setMaxResults(pagesize);
	
	//设置按id排序
	fullTextQuery.setSort(new Sort(new SortField("id", SortField.INT ,true)));
	
	//高亮设置
	SimpleHTMLFormatter formatter=new SimpleHTMLFormatter("<b><font color='red'>", "</font></b>");
	QueryScorer queryScorer=new QueryScorer(luceneQuery);
	Highlighter highlighter=new Highlighter(formatter, queryScorer);

	@SuppressWarnings("unchecked")
	List<Book> tempresult = fullTextQuery.getResultList();
	for (Book book : tempresult) {
		String highlighterString=null;
		try {
			//高亮name
			highlighterString=highlighter.getBestFragment(analyzer, "name", book.getName());
			if(highlighterString!=null){
				book.setName(highlighterString);
			}
			//高亮authors.name
			Set<Author> authors = book.getAuthors();
			for (Author author : authors) {
				highlighterString=highlighter.getBestFragment(analyzer, "authors.name", author.getName());
				if(highlighterString!=null){
					author.setName(highlighterString);
				}
			}
			//高亮description
			highlighterString=highlighter.getBestFragment(analyzer, "description", book.getDescription());
			if(highlighterString!=null){
				book.setDescription(highlighterString);
			}
		} catch (Exception e) {
		}
		
		books.add(book);	
		
	}
	
	queryResult.setSearchresult(books);
	
	return queryResult;
}