/**
* 在所有的字段中搜索，如果是坐标的话，需要加上*，才能匹配到
*
* @param queryString 关键字
* @throws Exception
*/
public static void searchListIKWithExp(String queryString) throws Exception {
// 查询的字符串:输入不存在的字符串是查询不到的,如：中国
// 查询字段集合
IndexSearcher searcher = LuceneUtils.createIndexSearcher();
// 这里使用的是IK的
Query query = IKQueryParser.parse(queryString);
System.out.println(query);
// 在搜索器中进行查询
// 对查询内容进行过滤
Filter filter = null;
// 一次在索引器查询多少条数据
int queryCount = 100;
TopDocs results = searcher.search(query, filter, queryCount);
System.out.println("总符合: " + results.totalHits + "条数！");
// 显示记录
for (ScoreDoc sr : results.scoreDocs) {
// 文档编号
int docID = sr.doc;
// 真正的内容
Document doc = searcher.doc(docID);
System.out.println("name = " + doc.get("name"));
System.out.println("address = " + doc.get("address"));
System.out.println("eastNew = " + doc.get("eastNew"));
System.out.println("northNew = " + doc.get("northNew"));
System.out.println("geom = " + doc.get("geom"));
System.out.println("phone = " + doc.get("phone"));
System.out.println("id = " + doc.get("id"));
System.out.println("datatype = " + doc.get("datatype"));
System.out.println("dataTypeKey = " + doc.get("dataTypeKey"));
}
}
public static void main(String[] args) {
String ikQueryExp =
"(id='ABcdRf' && date:{'20010101','20110101'} && keyword:'^魔兽中国$') || (content:'魔兽 中国' || ulr='www.ik.com') - name:'林良益'";
// String ikQueryExp = "content:'----' || title:'----' - name:'林良益'";
Query result = IKQueryParser.parse(ikQueryExp);
// Query result = IKQueryParser.parse("(newsKeyword='---' || newsTitle:'---' ||
// newsContent:'---') && newsClass='1'");
System.out.println(result);
}
public String getContentByTitle(String title) {
if (!(new File(indexDir).exists()) || (new File(indexDir).listFiles().length == 0)) {
System.out.println("索引文件夹不存在或文件夹为空~！");
return null;
}
String content = "";
try {
Query query = IKQueryParser.parse("title", title);
// System.out.println(query.toString());
TopDocCollector topCollector = new TopDocCollector(1000);
IndexSearcher searcher = new IndexSearcher(indexDir);
searcher.setSimilarity(new IKSimilarity());
searcher.search(query, topCollector);
ScoreDoc[] scoreDocs = topCollector.topDocs().scoreDocs;
// System.out.println(scoreDocs.length);
// System.out.println(query.toString());
Document doc = null;
for (ScoreDoc scoreDoc : scoreDocs) {
doc = searcher.doc(scoreDoc.doc);
content = doc.get("content");
System.out.println(doc.get("title"));
if (doc.get("title").equals(title)) break;
}
} catch (Exception e) {
e.printStackTrace();
}
return content;
}
@Override
public void hooker(String token, Searcher searcher, int cateGory, int dfOfToken)
throws IOException {
// TODO Auto-generated method stub
Query query = IKQueryParser.parse(IndexFields.CONTENTS, token);
TreeSet<Term> terms = new TreeSet<Term>();
query.extractTerms(terms);
double df = searcher.docFreq(terms.first());
double pt = df / totalTextNumber;
double max = Double.MIN_VALUE;
double avg = 0;
for (int i = 0; i < Constants.CATEGORYS.length; ++i) {
double dfInc = searchers[i].docFreq(terms.first());
double ptIfC = dfInc / CategoryTextCount[i];
double MI = log2n(ptIfC / pt);
if (MI > max) {
max = MI;
}
avg += MI * CategoryProbabilities[i];
}
maxFeatureWeight.add(new FeatureWeight(token, max, dfOfToken));
}
/**
* @param key 搜索内容所在的字段名称
* @param value 所要搜索的内容
* @throws CorruptIndexException
* @throws IOException
* @throws ParseException
*/
public static void search(String key, String value)
throws CorruptIndexException, IOException, ParseException {
IndexSearcher searcher;
// 创建QueryParser对象,第一个参数表示Lucene的版本,第二个表示搜索Field的字段,第三个表示搜索使用分词器
// Analyzer analyzer = new IKAnalyzer(); //new StandardAnalyzer(Version.LUCENE_36)
// QueryParser qp = new QueryParser(Version.LUCENE_36, key,analyzer);
searcher = new IndexSearcher(IndexReader.open(FSDirectory.open(new File(PATH))));
searcher.setSimilarity(new IKSimilarity());
Query query = IKQueryParser.parse(key, value);
// Query tq = qp.parse(value);
TopDocs topDocs = searcher.search(query, 10000);
// searcher.search(query, results); // new function
ScoreDoc[] scoreDocs = topDocs.scoreDocs;
System.out.println("命中：" + topDocs.totalHits);
Formatter formatter = new SimpleHTMLFormatter("<font color='red'>", "</font>");
Scorer fragmentScorer = new QueryScorer(query);
Highlighter highlighter = new Highlighter(formatter, fragmentScorer);
Fragmenter fragmenter = new SimpleFragmenter(50);
highlighter.setTextFragmenter(fragmenter);
Analyzer analyzer = new IKAnalyzer();
for (int i = 0; i < scoreDocs.length; i++) {
ScoreDoc scoreDoc = topDocs.scoreDocs[i];
int docSn = scoreDoc.doc; // 文档内部编号
Document doc = searcher.doc(docSn); // 根据文档编号取出文档
hightlightFeild(highlighter, analyzer, doc, "name");
hightlightFeild(highlighter, analyzer, doc, "address");
hightlightFeild(highlighter, analyzer, doc, "datatype");
// scoreDocs[i].
System.out.println("name:" + doc.get("name")); // new function
System.out.println("address:" + doc.get("address"));
System.out.println("datatype:" + doc.get("datatype"));
System.out.println("geom:" + doc.get("geom"));
}
}
/** 语法解析 */
private void parseSyntax() {
for (int i = 0; i < this.elements.size(); i++) {
Element e = this.elements.get(i);
if ('F' == e.type) {
Element e2 = this.elements.get(i + 1);
if ('=' != e2.type && ':' != e2.type) {
throw new IllegalStateException("表达式异常： = 或 ： 号丢失");
}
Element e3 = this.elements.get(i + 2);
// 处理 = 和 ： 运算
if ('\'' == e3.type) {
i += 2;
if ('=' == e2.type) {
TermQuery tQuery = new TermQuery(new Term(e.toString(), e3.toString()));
this.querys.push(tQuery);
} else if (':' == e2.type) {
String keyword = e3.toString();
if (keyword.startsWith("^") && keyword.endsWith("$")) {
Query pQuery = this.luceneQueryParse(e.toString(), keyword);
this.querys.push(pQuery);
} else {
Query tQuery = IKQueryParser.parse(e.toString(), e3.toString());
this.querys.push(tQuery);
}
}
} else if ('[' == e3.type || '{' == e3.type) {
i += 2;
// 处理 [] 和 {}
LinkedList<Element> eQueue = new LinkedList<Element>();
eQueue.add(e3);
for (i++; i < this.elements.size(); i++) {
Element eN = this.elements.get(i);
eQueue.add(eN);
if (']' == eN.type || '}' == eN.type) {
break;
}
}
// 翻译RangeQuery
Query rangeQuery = this.toTermRangeQuery(e, eQueue);
this.querys.push(rangeQuery);
} else {
throw new IllegalStateException("表达式异常：匹配值丢失");
}
} else if ('(' == e.type) {
this.operates.push(e);
} else if (')' == e.type) {
boolean doPop = true;
while (doPop && !this.operates.empty()) {
Element op = this.operates.pop();
if ('(' == op.type) {
doPop = false;
} else {
Query q = toQuery(op);
this.querys.push(q);
}
}
} else {
if (this.operates.isEmpty()) {
this.operates.push(e);
} else {
boolean doPeek = true;
while (doPeek && !this.operates.isEmpty()) {
Element eleOnTop = this.operates.peek();
if ('(' == eleOnTop.type) {
doPeek = false;
this.operates.push(e);
} else if (compare(e, eleOnTop) == 1) {
this.operates.push(e);
doPeek = false;
} else if (compare(e, eleOnTop) == 0) {
Query q = toQuery(eleOnTop);
this.operates.pop();
this.querys.push(q);
} else {
Query q = toQuery(eleOnTop);
this.operates.pop();
this.querys.push(q);
}
}
if (doPeek && this.operates.empty()) {
this.operates.push(e);
}
}
}
}
while (!this.operates.isEmpty()) {
Element eleOnTop = this.operates.pop();
Query q = toQuery(eleOnTop);
this.querys.push(q);
}
}