内容总结自互联网资料
1.如果要对某Field进行查找，那么一定要把Field.Index设置为TOKENIZED或UN_TOKENIZED。TOKENIZED会对Field的内容进行分词；而UN_TOKENIZED不会，只有全词匹配，该Field才会被选中。
2.如果Field.Store是No，那么就无法在搜索结果中从索引数据直接提取该域的值，会使null。
补充:
Field.Store.YES:存储字段值（未分词前的字段值）
Field.Store.NO:不存储,存储与索引没有关系
Field.Store.COMPRESS:压缩存储,用于长文本或二进制，但性能受损
Field.Index.ANALYZED:分词建索引
Field.Index.ANALYZED_NO_NORMS:分词建索引，但是Field的值不像通常那样被保存，而是只取一个byte，这样节约存储空间
Field.Index.NOT_ANALYZED:不分词且索引
Field.Index.NOT_ANALYZED_NO_NORMS:不分词建索引，Field的值去一个byte保存
3. TermVector
TermVector表示文档的条目（由一个Document和Field定位）和它们在当前文档中所出现的次数
Field.TermVector.YES:为每个文档（Document）存储该字段的TermVector
Field.TermVector.NO:不存储TermVector
Field.TermVector.WITH_POSITIONS:存储位置
Field.TermVector.WITH_OFFSETS:存储偏移量
Field.TermVector.WITH_POSITIONS_OFFSETS:存储位置和偏移量
4. 图解lucene TermVector
如果不是Field.Store.YES, 无法保存TermVector.
索引数据为Amsterdam has lots of bridges in Amsterdam
WhitespaceAnalyzer
4.1 TermVector.YES
4.2 TermVector.WITH_POSITIONS
4.3 TermVector.WITH_OFFSETS
4.4 TermVector.WITH_POSITIONS_OFFSETS
5. Lucene TermVector用法:相关搜索功能及提高高亮显示性能
public class TermVectorTest {
1
2 Analyzer analyzer = new SimpleAnalyzer();
3 Directory ramDir = new RAMDirectory();
4
5 public void createRamIndex() throws CorruptIndexException, LockObtainFailedException, IOException{
6
7 IndexWriter writer = new IndexWriter(ramDir,analyzer,IndexWriter.MaxFieldLength.LIMITED);
8
9 Document doc1 = new Document();
10 doc1.add(new Field("title","java",Store.YES,Index.ANALYZED));
11 doc1.add(new Field("author","callan",Store.YES,Index.ANALYZED));
12 doc1.add(new Field("subject","java一门编程语言,用java的人很多，编程语言也不少，但是java最流行",Store.YES,Index.ANALYZED,TermVector.WITH_POSITIONS_OFFSETS));
13
14 Document doc2 = new Document();
15 doc2.add(new Field("title","english",Store.YES,Index.ANALYZED));
16 doc2.add(new Field("author","wcq",Store.YES,Index.ANALYZED));
17 doc2.add(new Field("subject","英语用的人很多",Store.YES,Index.ANALYZED,TermVector.WITH_POSITIONS_OFFSETS));
18
19 Document doc3 = new Document();
20 doc3.add(new Field("title","asp",Store.YES,Index.ANALYZED));
21 doc3.add(new Field("author","ca",Store.YES,Index.ANALYZED));
22 doc3.add(new Field("subject","英语用的人很多",Store.YES,Index.ANALYZED,TermVector.WITH_POSITIONS_OFFSETS));
23
24 writer.addDocument(doc1);
25 writer.addDocument(doc2);
26 writer.addDocument(doc3);
27
28 writer.optimize();
29 writer.close();
30 }
31
32 public void search() throws CorruptIndexException, IOException{
33 IndexReader reader = IndexReader.open(ramDir);
34 IndexSearcher searcher = new IndexSearcher(reader);
35 Term term = new Term("title","java"); //在title里查询java词条
36 TermQuery query = new TermQuery(term);
37 Hits hits = searcher.search(query);
38 for (int i = 0; i < hits.length(); i++)
39 {
40 Document doc = hits.doc(i);
41 System.out.println(doc.get("title"));
42 System.out.println(doc.get("subject"));
43 System.out.println("moreLike search: ");
44
45 morelikeSearch(reader,hits.id(i));
46 }
47 }
48
49 private void morelikeSearch(IndexReader reader,int id) throws IOException
50 {
51 //根据这个document的id获取这个field的Term Vector 信息，就是这个field分词之后在这个field里的频率、位置、等信息
52 TermFreqVector vector = reader.getTermFreqVector(id, "subject");
53
54 BooleanQuery query = new BooleanQuery();
55
56 for (int i = 0; i < vector.size(); i++)
57 {
58 TermQuery tq = new TermQuery(new Term("subject",
59 vector.getTerms()[i])); //获取每个term保存的Token
61 query.add(tq, BooleanClause.Occur.SHOULD);
63 }
65 IndexSearcher searcher = new IndexSearcher(ramDir);
67 Hits hits = searcher.search(query);
69 //显示代码，略
72 }
73
74 //Lucene使用TermVector提高高亮显示性能
75 public void highterLightSearch() throws CorruptIndexException, IOException{
76 IndexReader reader = IndexReader.open(ramDir);
78 IndexSearcher searcher = new IndexSearcher(reader);
80 TermQuery query = new TermQuery(new Term("subject","java"));
82 Hits hits = searcher.search(query);
84 //高亮显示设置
85 SimpleHTMLFormatter simpleHTMLFormatter = new SimpleHTMLFormatter("<font color='red'>","</font>");
87 Highlighter highlighter =new Highlighter(simpleHTMLFormatter,new QueryScorer(query));
89 // 这个100是指定关键字字符串的context的长度，你可以自己设定，因为不可能返回整篇正文内容
90 highlighter.setTextFragmenter(new SimpleFragmenter(100));
92 for(int i = 0; i < hits.length(); i++){
94 Document doc = hits.doc(i);
96 TermPositionVector termFreqVector = (TermPositionVector)reader.getTermFreqVector(hits.id(i), "subject");
98 TermFreqVector vector = reader.getTermFreqVector(hits.id(i), "subject");
99 TokenStream tokenStream = TokenSources.getTokenStream(termFreqVector);
101 String result = highlighter.getBestFragment(tokenStream, doc.get("subject"));
103 System.out.println(doc.get("title"));
105 System.out.println(result);
107 }
110 }
112 public static void main(String[] args) throws CorruptIndexException, IOException
113 {
114 TermVectorTest t = new TermVectorTest();
115 t.createRamIndex();
116 t.search();
117 }
119 }
