全文檢索引擎lucene的研究和使用（二）

// 創建查詢
	@Repository
	@Transactional
	public class SearcherJpa implements Serializable, SearcherDAO {
		private static final long serialVersionUID = 1L;

		public List<Attachment> do_search(String path, String keyword,
				String type, int start, int end) {
			List<Attachment> list = null;
			Attachment att = null;
			PropertiesHelper p = null;
			try {
				p = new PropertiesHelper(path);
				String IndexDir = p.getValue("IndexDir");
				Directory dir = FSDirectory.open(new File(IndexDir));
				IndexReader reader = IndexReader.open(dir);
				// TermEnum termEnum = reader.terms();
				IndexSearcher is = new IndexSearcher(reader);
				is.setDefaultFieldSortScoring(true, false);
				Query query = null;
				if (type != "" && type != null) {
					String[] fields = { LuceneUtil.FILE_CONTENT,
							LuceneUtil.FILE_TYPE };
					String[] fields_value = { keyword, type };
					// MultiFieldQueryParser mquery = new
					// MultiFieldQueryParser(LuceneUtil.VERSION_CURRENT, fields,
					// LuceneUtil.ANALYZER_CURRENT);
					// QueryFilter qf = new QueryFilter(tq);
					query = MultiFieldQueryParser
							.parse(LuceneUtil.VERSION_CURRENT, fields_value,
									fields, new BooleanClause.Occur[] {
											Occur.MUST, Occur.MUST },
									LuceneUtil.ANALYZER_CURRENT);

				} else {
					QueryParser queryParser = new QueryParser(
							LuceneUtil.VERSION_CURRENT,
							LuceneUtil.FILE_CONTENT,
							LuceneUtil.ANALYZER_CURRENT);
					query = queryParser.parse(keyword);
				}

				// QueryParser queryParser = new
				// QueryParser(LuceneUtil.VERSION_CURRENT,
				// LuceneUtil.FILE_CONTENT, LuceneUtil.ANALYZER_CURRENT);

				// Term tm = new Term(LuceneUtil.FILE_TYPE,type);
				// TermQuery tq = new TermQuery(tm);
				// Filter f = new Filter();
				TopDocs hits = is.search(query, 10000, Sort.RELEVANCE);
				SimpleHTMLFormatter shf = new SimpleHTMLFormatter(
						"<em><strong>", "</strong></em>");
				Highlighter hl = new Highlighter(shf, new QueryScorer(query));
				// hl.setTextFragmenter(new
				// SimpleFragmenter(Integer.MAX_VALUE));
				hl.setTextFragmenter(new SimpleFragmenter(200));

				// TopScoreDocCollector results =
				// TopScoreDocCollector.create(11, false);
				// is.search(query, results);
				// TopDocs tds = results.topDocs(1, 10);
				// ScoreDoc[] sd = tds.scoreDocs;
				list = new ArrayList<Attachment>();
				for (int i = start; i < end && i < hits.scoreDocs.length; i++) {
					System.out.println("start: " + start + " end: " + end);
					att = new Attachment();
					ScoreDoc sdoc = hits.scoreDocs[i];
					Document doc = is.doc(sdoc.doc);
					System.out.println(doc.get(LuceneUtil.FILE_NAME));
					System.out.println(doc.get(LuceneUtil.FILE_TYPE));
					System.out.println(doc.get(LuceneUtil.FILE_PATH));
					System.out.println(doc.get(LuceneUtil.FILE_DATE));
					att.setFileid(sdoc.doc);
					att.setFilename(doc.get(LuceneUtil.FILE_NAME));
					att.setFiletype(doc.get(LuceneUtil.FILE_TYPE));
					att.setFilepath(doc.get(LuceneUtil.FILE_PATH));
					att.setFiledate(doc.get(LuceneUtil.FILE_DATE));
					String str = hl.getBestFragment(
							LuceneUtil.ANALYZER_CURRENT,
							LuceneUtil.FILE_CONTENT,
							doc.get(LuceneUtil.FILE_CONTENT));
					att.setHitword(str);
					att.setFilescore(sdoc.score);
					list.add(att);

					// TokenStream tokenStream =
					// LuceneUtil.ANALYZER_CURRENT.tokenStream(fieldName,new
					// StringReader(text));

					// TermPositionVector tpv = (TermPositionVector)
					// IndexReader.getTermFreqVector(id, "content");
					// TokenStream tokenStream =
					// TokenSources.getTokenStream(tpv);
					// tokenStream.startOffset();
					// tokenStream.next().endOffset();

					System.out.println(doc.get(LuceneUtil.FILE_CONTENT));
					System.out.println(doc.toString());
					System.out.println(doc.getBoost());
					System.out.println(sdoc.toString());
					System.out.println("doc: " + sdoc.doc + "  score: "
							+ sdoc.score + "  shardIndex: " + sdoc.shardIndex);
				}
			} catch (IOException e) {
				e.printStackTrace();
			} catch (ParseException e) {
				e.printStackTrace();
			} catch (InvalidTokenOffsetsException e) {
				e.printStackTrace();
			}
			return list;
		}

		@Override
		public int getCount(String path, String keyword, String type) {
			PropertiesHelper p = null;
			TopDocs hits = null;
			try {
				p = new PropertiesHelper(path);
				String IndexDir = p.getValue("IndexDir");
				Directory dir = FSDirectory.open(new File(IndexDir));
				IndexReader reader = IndexReader.open(dir);
				IndexSearcher is = new IndexSearcher(reader);
				is.setDefaultFieldSortScoring(true, false);
				Query query = null;
				if (type != "" && type != null) {
					String[] fields = { LuceneUtil.FILE_CONTENT,
							LuceneUtil.FILE_TYPE };
					String[] fields_value = { keyword, type };
					// MultiFieldQueryParser mquery = new
					// MultiFieldQueryParser(LuceneUtil.VERSION_CURRENT, fields,
					// LuceneUtil.ANALYZER_CURRENT);
					// QueryFilter qf = new QueryFilter(tq);
					query = MultiFieldQueryParser
							.parse(LuceneUtil.VERSION_CURRENT, fields_value,
									fields, new BooleanClause.Occur[] {
											Occur.MUST, Occur.MUST },
									LuceneUtil.ANALYZER_CURRENT);

				} else {
					QueryParser queryParser = new QueryParser(
							LuceneUtil.VERSION_CURRENT,
							LuceneUtil.FILE_CONTENT,
							LuceneUtil.ANALYZER_CURRENT);
					query = queryParser.parse(keyword);
				}
				// QueryParser queryParser = new
				// QueryParser(LuceneUtil.VERSION_CURRENT,
				// LuceneUtil.FILE_CONTENT, LuceneUtil.ANALYZER_CURRENT);
				// Query query = queryParser.parse(keyword);
				hits = is.search(query, 10000, Sort.RELEVANCE);
			} catch (IOException e) {
				e.printStackTrace();
			} catch (ParseException e) {
				e.printStackTrace();
			}
			return hits.scoreDocs.length;
		}
	}