即时判断字节编码类型

getCharset

    public static String getCharset(byte[] bytes) {
        
        if(bytes == null || bytes.length < 2) return "GBK";
        if(bytes[0] == (byte)0xFF && bytes[1] == (byte)0xFE) {
            return "UTF-16LE";
        } else if (bytes[0] == (byte)0xFE && bytes[1] == (byte)0xFF) {
            return "UTF-16BE";
        } else if (bytes[0] == (byte)0xEF && bytes[1] == (byte)0xBB && bytes[2] == (byte)0xBF) {
            return "UTF-8";
        }
        
        for(int i = 0, b = bytes[i] & 0xff; i < bytes.length; i ++) {
            
            b = bytes[i] & 0xff;
            if(b >= 0xF0) return "GBK";
            if(0x80 <= b && b <=0xBF) return "GBK";//单独出现BF以下的，也算是GBK
            if(0xC0 <= b && b <= 0xDF) {
                i ++; b = bytes[i] & 0xff;
                if(0x80 <= b && b <= 0xBF) {
                    continue;//双字节 (0xC0 - 0xDF) (0x80- 0xBF),也可能在GBK编码内
                } else {
                    return "GBK";
                }
            } else if (0xE0 <= b && b <= 0xEF) {// 也有可能出错，但是几率较小
                i ++; b = bytes[i] & 0xff;
                if(0x80 <= b && b <= 0xBF) {
                    i ++; b = bytes[i] & 0xff;
                    if(0x80 <= b && b <= 0xBF) {
                        return "UTF-8";
                    }
                }
                return "GBK";
            }
        }
        return "GBK";
    }