關於文件的字符集編碼探測--java

今天遇到一個要探測文件的字符集編碼問題,查了一下資料,發現了一個 探測器: jchardet  需要  chardet.jar

 

例子如下:

**
   * 獲取文件的編碼
   *
   * @param file
   * @param det
   * @return
   * @throws FileNotFoundException
   * @throws IOException
   *
  private String geestFileEncoding(File file, nsDetector det)
    throws FileNotFoundException, IOException {
  
   det.Init(new nsICharsetDetectionObserver() {
    public void Notify(String charset) {
     found = true;
     encoding = charset;
    }
   });

   BufferedInputStream imp = new BufferedInputStream(new FileInputStream(
     file));

   byte[] buf = new byte[1024];
   int len;
   boolean done = false;
   boolean isAscii = true;

   while ((len = imp.read(buf, 0, buf.length)) != -1) {
    

    if (isAscii)
     isAscii = det.isAscii(buf, len);

    if (!isAscii && !done)
     done = det.DoIt(buf, len, false);
   }
   det.DataEnd();

   if (isAscii) {
    encoding = "ASCII";
    found = true;
   }

   if (!found) {
    String prob[] = det.getProbableCharsets();
    if (prob.length > 0) {
         encoding = prob[0];
    } else {
     return null;
    }
   }
   return encoding;
  }