java获取文件编码格式

public static String get_charset( File file ) {
        String charset = "GBK";
        byte[] first3Bytes = new byte[3];
        try {
            boolean checked = false;
            BufferedInputStream bis = new BufferedInputStream( new FileInputStream( file ) );
            bis.mark( 0 );
            int read = bis.read( first3Bytes, 0, 3 );
            if ( read == -1 ) return charset;
            if ( first3Bytes[0] == (byte) 0xFF && first3Bytes[1] == (byte) 0xFE ) {
                charset = "UTF-16LE";
                checked = true;
            }else if ( first3Bytes[0] == (byte) 0xFE && first3Bytes[1] == (byte) 0xFF ) {
                charset = "UTF-16BE";
                checked = true;
            }else if ( first3Bytes[0] == (byte) 0xEF && first3Bytes[1] == (byte) 0xBB && first3Bytes[2] == (byte) 0xBF ) {
                charset = "UTF-8";
                checked = true;
            }
            bis.reset();
            if ( !checked ) {
            //    int len = 0;
                int loc = 0;
                while ( (read = bis.read()) != -1 ) {
                    loc++;
                    if ( read >= 0xF0 ) break;
                    if ( 0x80 <= read && read <= 0xBF ) // 单独出现BF以下的,也算是GBK
                    break;
                    if ( 0xC0 <= read && read <= 0xDF ) {
                        read = bis.read();
                        if ( 0x80 <= read && read <= 0xBF ) // 双字节 (0xC0 - 0xDF) (0x80
                                                                        // - 0xBF),也可能在GB编码内  
                        continue;
                        else break;
                    }
                    else if ( 0xE0 <= read && read <= 0xEF ) {// 也有可能出错,但是几率较小
                        read = bis.read();
                        if ( 0x80 <= read && read <= 0xBF ) {
                            read = bis.read();
                            if ( 0x80 <= read && read <= 0xBF ) {
                                charset = "UTF-8";
                                break;
                            }
                            else break;
                        }
                        else break;
                    }
                }
                //System.out.println( loc + " " + Integer.toHexString( read ) );
            }
            bis.close();
        } catch ( Exception e ) {
            e.printStackTrace();
        }
        return charset;
    }

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值