Lucene 心得
public void createIndex1(Connection conn, AttachInfoDTO1 main) throws SQLException {
//定义
IndexWriter writer = null;
//文件路径
Directory directory = null;
// 索引文件夹
File dir = new File("PICCAssets/index");
String path = dir.getAbsolutePath();
String path1 = path.replace("bin", "webapps");
try {
directory = FSDirectory.open(new File(path1));
writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_35, new IKAnalyzer()));
Document doc = null;
File f = null;
InputStream br = null;
if (main != null) {
int fileid = main.getId();
String dname = main.getFileUrl();
String rname = "http://192.168.1.10:8080/PICCAssets/" + dname;
FtpTask ftp = new FtpTask();
String path11 = dname.substring(0, dname.lastIndexOf("/") + 1);
String file = dname.substring(dname.lastIndexOf("/") + 1);
ftp.connectServer(path11);
br = );
SimpleDateFormat sdf = new SimpleDateFormat("yyyyMMdd");
String author = main.getStaffName();
if ((author == null) || (("").equals(author))) {
author = "系统管理员";
}
doc = new Document();
Metadata metadata = new Metadata();
doc.add(new Field("content", new Tika().parseToString(br), Field.Store.YES, Field.Index.ANALYZED, TermVector.WITH_POSITIONS_OFFSETS));
doc.add(new Field("title", main.getAttachName(), Field.Store.YES, Field.Index.ANALYZED, TermVector.WITH_POSITIONS_OFFSETS));
doc.add(new Field("author", author, Field.Store.YES, Field.Index.ANALYZED, TermVector.WITH_POSITIONS_OFFSETS));
doc.add(new Field("rid", Integer.toString(main.getId()), Field.Store.YES, Field.Index.ANALYZED, TermVector.WITH_POSITIONS_OFFSETS));
doc.add(new Field("filename", main.getAttachName(), Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("path", dname, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("dpath", rname, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("dtype", main.getFileTypeLever1(), Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("date", main.getUploadDate(), Field.Store.YES, Field.Index.NOT_ANALYZED));
int page = 0;
try {
page = Integer.parseInt(metadata.get("xmpTPg:NPages"));
} catch (NumberFormatException e) {
}
if (doc != null) {
try {
writer.addDocument(doc);
} catch (Exception e) {
// TODO: handle exception
e.printStackTrace();
}
}
}
} catch (IOException e) {
e.printStackTrace();
} catch (Exception e) {
// TODO: handle exception
e.printStackTrace();
} finally {
if (writer != null) {
try {
writer.close();
} catch (CorruptIndexException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
}
}
}
删除全部索引
/**
* 删除全部索引
*
* @throws Exception
*/
public void deleteAllIndex() throws Exception {
IndexWriterConfig config = new IndexWriterConfig(Version.LUCENE_35, new StandardAnalyzer(Version.LUCENE_35));
IndexWriter indexWriter = new IndexWriter(FileIndexUtils.getDirectory(), config);
indexWriter.deleteAll();
indexWriter.close();
}
删除单个索引
/**
* 删除单个索引
*
* @throws Exception
*/
public void deleteOneIndex(String uid) throws Exception {
IndexWriterConfig config = new IndexWriterConfig(Version.LUCENE_35, new StandardAnalyzer(Version.LUCENE_35));
IndexWriter indexWriter = new IndexWriter(FileIndexUtils.getDirectory(), config);
indexWriter.deleteDocuments(new Term("rid", uid));
indexWriter.close();
}
//高亮查询索引对应的文档 以及组合查询
@SuppressWarnings({ "rawtypes", "unused" })
public PagerDTO<LuceneDTO> searchPage(String key, int starttime, int endtime, String dtype, int pageIndex, int pageSize) { // String
Analyzer luceneAnalyzer = new IKAnalyzer();
PagerDTO<LuceneDTO> pages = new PagerDTO<LuceneDTO>();
List<LuceneDTO> l = new ArrayList<LuceneDTO>();
LuceneDTO bean = null;
try {
// 索引文件夹
IndexSearcher searcher = new IndexSearcher(IndexReader.open(FileIndexUtils.getDirectory()));
BooleanQuery query0 = new BooleanQuery();
// 日期区间
NumericRangeQuery q2 = NumericRangeQuery.newIntRange("date", starttime, endtime, true, true);
// 关键字查询
MultiFieldQueryParser parser = new MultiFieldQueryParser(Version.LUCENE_35, new String[] { "title", "content" }, luceneAnalyzer);
if (key == null || "".equals(key)) {
query0.add(new TermQuery(new Term("dtype", dtype)), Occur.SHOULD);
} else {
Query queryKey = parser.parse(key);
query0.add(new TermQuery(new Term("dtype", dtype)), Occur.MUST);
query0.add(queryKey, Occur.MUST);
}
// 先获取上一页的最后一个元素
ScoreDoc lastSd = getLastScoreDoc(pageIndex, pageSize, query0, searcher);
// 通过最后一个元素搜索下页的pageSize个元素
TopDocs tds = searcher.searchAfter(lastSd, query0, pageSize);
int totalRecord = tds.totalHits;
// 搜素开始时间
Date beginTime = new Date();
for (ScoreDoc sd : tds.scoreDocs) {
Document doc = searcher.doc(sd.doc);
String path = doc.get("path");
String filen = path.split("/")[2];
String filename = filen.substring(0, filen.lastIndexOf(".")) + ".swf";
/**
* 按照http远程读取文件方式，没有储存在索引上
*/
String content = doc.get("content");
String htitle = doc.get("title");
QueryScorer scorer = new QueryScorer(query0);
Fragmenter fragmenter = new SimpleSpanFragmenter(scorer);
Formatter formatter = new SimpleHTMLFormatter("<span style='color:red'>", "</span>");
Highlighter lighter = new Highlighter(formatter, scorer);
lighter.setTextFragmenter(fragmenter);
// 设置高亮字数多少
lighter.setTextFragmenter(new SimpleFragmenter(200));
/*
* 设置高亮，如果有关键字就加亮，没有就返回空，所以要加判断
*/
String str = lighter.getBestFragment(luceneAnalyzer, "f", content);
String titlel = lighter.getBestFragment(luceneAnalyzer, "s", htitle);
bean = new LuceneDTO();
if (titlel != null) {
bean.setTitle(titlel);
} else {
bean.setTitle(htitle);
}
if (str != null) {
bean.setStr(str);
} else {
if (content.length() != 0 && content.length() > 250) {
String content1 = content.substring(0, 250);
bean.setStr(content1);
} else {
bean.setStr(content);
}
}
bean.setDate(doc.get("date"));
bean.setType(doc.get("title").substring(doc.get("title").lastIndexOf(".") + 1));
bean.setPath(doc.get("author"));
bean.setFilename(filename);
bean.setDpath(doc.get("dpath"));
bean.setRid(doc.get("rid"));
l.add(bean);
}
// 搜索完成时间
Date endTime = new Date();
DecimalFormat format = new DecimalFormat("0.000");
double timeOfSearch = (endTime.getTime() - beginTime.getTime()) * 0.001;
int totalPage = (totalRecord - 1) / pageSize + 1;
pages.setTimeOfSearch(format.format(timeOfSearch));
pages.setTotalPage(totalPage);
pages.setTotalRecord(totalRecord);
pages.setPageIndex(pageIndex);
pages.setPageSize(pageSize);
pages.setDatas(l);
searcher.close();
} catch (IOException e) {
e.printStackTrace();
} catch (InvalidTokenOffsetsException e) {
e.printStackTrace();
} catch (ParseException e) {
e.printStackTrace();
}
return pages;
}
/**
* 根据页码和分页大小获取上一次的最后一个ScoreDoc
*/
private ScoreDoc getLastScoreDoc(int pageIndex, int pageSize, Query query, IndexSearcher searcher) throws IOException {
if (pageIndex == 1)
return null;// 如果是第一页就返回空
int num = pageSize * (pageIndex - 1);// 获取上一页的数量
TopDocs tds = searcher.search(query, num);
return tds.scoreDocs[num - 1];
}