JAVA读取不同编码格式的TXT文件

最近做一个全文检索的功能，要求在用户上传附件时获取其内容存入数据库，CSV,EXCEL,WORD,PPT都搞定了。剩余一个最简单的TXT文档，开始忽略了它还有不同编码，导致一些编码格式的文本乱码。兼容写法如下，一个判断文件头的方法搞定所有。

1privateString getCharset(String fileName) throws IOException{
2      
3          BufferedInputStream bin = new BufferedInputStream(newFileInputStream(fileName));  
4          int p = (bin.read() << 8) +bin.read();  
5          
6          String code = null;  
7          
8          switch (p) {  
9              case 0xefbb:  
10                  code = "UTF-8";  
11                  break;  
12              case 0xfffe:  
13                  code = "Unicode";  
14                  break;  
15              case 0xfeff:  
16                  code = "UTF-16BE";  
17                  break;  
18              default:  
19                  code = "GBK";  
20          }  
21          return code;
22    }
23    
24    
25    publicString getTextFromText(String filePath){
26      
27         try {
28            InputStreamReader isr = new InputStreamReader(newFileInputStream(filePath),getCharset(filePath));
29            BufferedReader br = newBufferedReader(isr);
30            
31             StringBuffer sb = new StringBuffer();  
32             String temp = null;   
33             while((temp = br.readLine()) != null){  
34                 sb.append(temp);   
35             }   
36             br.close();       
37             return sb.toString();   
38         } catch (FileNotFoundException e) {
39            // TODO Auto-generated catch block
40            e.printStackTrace();
41            
42         }catch (IOException e) {
43            // TODO Auto-generated catch block
44            e.printStackTrace();
45         }   
46         return null;
47    }