逆向分析某app并使用Java与PHP语言实现RC4加解密

近期在使用“实时公交”软件,其有北京大部分公交车的GPS数据,但是软件比较偏国企化,做的并没有那么人性化,大山子路口这里剧堵是远近闻名的,基本上每辆车都要堵一段时间,那么冬天出去等车就会很遭罪,但是晚出去有可能眼睁睁的看着公交到站开走,所以打算逆向分析一下北京实时公交软件,加入对堵车时间的预测。

逆向分析

该软件如其外观一样简单,内部没有做混淆,甚至还有测试信息打印,System.out.println等输出,所以直接dex转jar(https://sourceforge.net/projects/dex2jar/),然后使用jd-gui(http://jd.benow.ca/)进行查看,导出来后放入IDE查看。

代码分析

以下为其主要目录结构

model层应该是可以比较直观的看出其结构(其model层写了一个错别字?)

分析看到需要解密的字段,其使用RC4加解密,一个不是特别常用的对称加解密算法

RC4加解密简介

RC4不是对明文进行分组处理,而是字节流的方式依次加密明文中的每一个字节,解密的时候也是依次对密文中的每一个字节进行解密。
算法简单,运行速度快,而且密钥长度是可变的,可变范围为1-256字节(8-2048比特)

1、密钥流:RC4算法的关键是根据明文和密钥生成相应的密钥流,密钥流的长度和明文的长度是对应的,也就是说明文的长度是500字节,那么密钥流也是500字节。当然,加密生成的密文也是500字节,因为密文第i字节=明文第i字节^密钥流第i字节;
2、状态向量S:长度为256,S[0],S[1]…..S[255]。每个单元都是一个字节,算法运行的任何时候,S都包括0-255的8比特数的排列组合,只不过值的位置发生了变换;
3、临时向量T:长度也为256,每个单元也是一个字节。如果密钥的长度是256字节,就直接把密钥的值赋给T,否则,轮转地将密钥的每个字节赋给T;
4、密钥K:长度为1-256字节,注意密钥的长度keylen与明文长度、密钥流的长度没有必然关系,通常密钥的长度趣味16字节(128比特)。

RC4加解密PHP实现

实现:

function rc4($key, $pt)
{
    $s = array();
    for ($i=0; $i<256; $i++) {
        $s[$i] = $i;
    }

    $j = 0;
    $key_len = strlen($key);
    for ($i=0; $i<256; $i++) {
        $j = ($j + $s[$i] + ord($key[$i % $key_len])) % 256;
        //swap
        $x = $s[$i];
        $s[$i] = $s[$j];
        $s[$j] = $x;
    }
    $i = 0;
    $j = 0;
    $ct = '';
    $data_len = strlen($pt);
    for ($y=0; $y< $data_len; $y++) {
        $i = ($i + 1) % 256;
        $j = ($j + $s[$i]) % 256;
        //swap
        $x = $s[$i];
        $s[$i] = $s[$j];
        $s[$j] = $x;
        $ct .= $pt[$y] ^ chr($s[($s[$i] + $s[$j]) % 256]);
    }
    return $ct;
}

测试验证:

$str = '{"root":{"status":"200","message":"success","encrypt":"1","num":"6","lid":"949","data":{"bus":[{"gt":"1511789342","id":"75544","t":"0","ns":"MeACxHFmZLhE","nsn":"5kU=","nsd":"1706","nsrt":"197","nst":"1511789539","sd":"+Uc=","srt":"+Uc=","st":"+Uc=","x":"5UewD9XUuDbzQw==","y":"4EaoEdPStDT4","lt":"0","ut":"1511789353"},{"gt":"1511789343","id":"75537","t":"0","ns":"9kh/iiYwnXfFvgPc4r4gjGxZ","nsn":"Ldk=","nsd":"652","nsrt":"79","nst":"1511789422","sd":"MtA=","srt":"MtA=","st":"MtA=","x":"LtDlQqCIRdFZaQ==","y":"K9H9XKaJRtVY","lt":"0","ut":"1511789354"},{"gt":"1511789334","id":"75611","t":"0","ns":"z60iAGP9avHwPh8dLCVw17Rv","nsn":"Gzw=","nsd":"-1","nsrt":"-1","nst":"-1","sd":"Bzg=","srt":"Bzg=","st":"Bzg=","x":"Gzizy+Z0tmpX4g==","y":"GTCr3Op6uW5S","lt":"0","ut":"1511789342"},{"gt":"1511789335","id":"75538","t":"0","ns":"q2CWQiHZ52ohG3yEEOiW","nsn":"fMs=","nsd":"321","nsrt":"52","nst":"1511789387","sd":"YM0=","srt":"YM0=","st":"YM0=","x":"fM07iK9COfuC","y":"fsUjn6JDNvg=","lt":"0","ut":"1511789343"},{"gt":"1511789256","id":"75549","t":"0","ns":"GlwsiqkbEEag","nsn":"wcI=","nsd":"515","nsrt":"71","nst":"1511789327","sd":"3sQ=","srt":"3sQ=","st":"3sQ=","x":"wsS2QRG8zfQcfA==","y":"x8WuXxa/y/kU","lt":"0","ut":"1511789270"},{"gt":"1511789107","id":"75553","t":"0","ns":"ZZQTCHtyjvXT3sBLOCJLdjq3","nsn":"sws=","nsd":"-1","nsrt":"-1","nst":"-1","sd":"rQg=","srt":"rQg=","st":"rQg=","x":"sQi8wPzyX0BIDw==","y":"tAmk3vr4X0NG","lt":"0","ut":"1511789130"}]}}}';
$arr = json_decode($str,true);
$arr = $arr['root']['data']['bus'];

foreach ($arr as $key=>$val){
    echo date('Y-m-d H:i:s',$val['gt'])."\n";
    echo $val['id']."\n";
    echo $val['t']."\n";
    echo rc4(md5("aibang".$val['gt']), base64_decode($val['ns']))."\n";
    echo rc4(md5("aibang".$val['gt']), base64_decode($val['nsn']))."\n";
    echo $val['nsd']."\n";
    echo $val['nsrt']."\n";
    echo $val['nst']."\n";
    echo $val['sd']."\n";
    echo $val['srt']."\n";
    echo $val['nst']."\n";
    echo rc4(md5("aibang".$val['gt']), base64_decode($val['x']))."\n";
    echo rc4(md5("aibang".$val['gt']), base64_decode($val['y']))."\n";
    echo date('Y-m-d H:i:s',$val['ut'])."\n";
    echo date('Y-m-d H:i:s',$val['lt'])."\n";
    echo "******************************************************\n";
}

RC4加解密Java实现

实现:

package testest;

import java.security.MessageDigest;
import java.security.NoSuchAlgorithmException;

public class RC4
{
    public static String decry_RC4(byte[] data, String key) {
        if (data == null || key == null) {
            return null;
        }
        return asString(RC4Base(data, key));
    }


    public static String decry_RC4(String data, String key) {
        if (data == null || key == null) {
            return null;
        }
        return new String(RC4Base(HexString2Bytes(data), key));
    }


    public static byte[] encry_RC4_byte(String data, String key) {
        if (data == null || key == null) {
            return null;
        }
        byte b_data[] = data.getBytes();
        return RC4Base(b_data, key);
    }


    public static String encry_RC4_string(String data, String key) {
        if (data == null || key == null) {
            return null;
        }
        return toHexString(asString(encry_RC4_byte(data, key)));
    }


    private static String asString(byte[] buf) {
        StringBuffer strbuf = new StringBuffer(buf.length);
        for (int i = 0; i < buf.length; i++) {
            strbuf.append((char) buf[i]);
        }
        return strbuf.toString();
    }


    private static byte[] initKey(String aKey) {
        byte[] b_key = aKey.getBytes();
        byte state[] = new byte[256];

        for (int i = 0; i < 256; i++) {
            state[i] = (byte) i;
        }
        int index1 = 0;
        int index2 = 0;
        if (b_key == null || b_key.length == 0) {
            return null;
        }
        for (int i = 0; i < 256; i++) {
            index2 = ((b_key[index1] & 0xff) + (state[i] & 0xff) + index2) & 0xff;
            byte tmp = state[i];
            state[i] = state[index2];
            state[index2] = tmp;
            index1 = (index1 + 1) % b_key.length;
        }
        return state;
    }

    private static String toHexString(String s) {
        String str = "";
        for (int i = 0; i < s.length(); i++) {
            int ch = (int) s.charAt(i);
            String s4 = Integer.toHexString(ch & 0xFF);
            if (s4.length() == 1) {
                s4 = '0' + s4;
            }
            str = str + s4;
        }
        return str;// 0x表示十六进制
    }


    private static byte[] HexString2Bytes(String src) {
        int size = src.length();
        byte[] ret = new byte[size / 2];
        byte[] tmp = src.getBytes();
        for (int i = 0; i < size / 2; i++) {
            ret[i] = uniteBytes(tmp[i * 2], tmp[i * 2 + 1]);
        }
        return ret;
    }

    private static byte uniteBytes(byte src0, byte src1) {
        char _b0 = (char)Byte.decode("0x" + new String(new byte[] { src0 }))
                .byteValue();
        _b0 = (char) (_b0 << 4);
        char _b1 = (char)Byte.decode("0x" + new String(new byte[] { src1 }))
                .byteValue();
        byte ret = (byte) (_b0 ^ _b1);
        return ret;
    }

    private static byte[] RC4Base (byte [] input, String mKkey) {
        int x = 0;
        int y = 0;
        byte key[] = initKey(mKkey);
        int xorIndex;
        byte[] result = new byte[input.length];

        for (int i = 0; i < input.length; i++) {
            x = (x + 1) & 0xff;
            y = ((key[x] & 0xff) + y) & 0xff;
            byte tmp = key[x];
            key[x] = key[y];
            key[y] = tmp;
            xorIndex = ((key[x] & 0xff) + (key[y] & 0xff)) & 0xff;
            result[i] = (byte) (input[i] ^ key[xorIndex]);
        }
        return result;
    }

  private static String encode(String paramString1, String paramString2)
  {
    try
    {
      MessageDigest localMessageDigest = MessageDigest.getInstance(paramString2);
      localMessageDigest.update(paramString1.getBytes());
      byte[] arrayOfByte = localMessageDigest.digest();
      StringBuilder localStringBuilder = new StringBuilder(2 * arrayOfByte.length);
      for (int i = 0; i < arrayOfByte.length; i++)
      {
        localStringBuilder.append(Integer.toHexString((0xF0 & arrayOfByte[i]) >>> 4));
        localStringBuilder.append(Integer.toHexString(0xF & arrayOfByte[i]));
      }
      String str = localStringBuilder.toString();
      return str;
    }
    catch (NoSuchAlgorithmException localNoSuchAlgorithmException)
    {
      localNoSuchAlgorithmException.printStackTrace();
    }
    return "";
  }
}

验证

    public static void main(String[] args) throws NoSuchAlgorithmException {
        String md5 = encode("aibang1511793087","MD5");
        System.out.println(md5);
        byte[] b = Base64.decode("jT2f7w3uYkWWKvnwI4XpuD1X",0);
        System.out.println(new String(b));
        String test = null;
        test = new String(RC4.RC4Base(b, md5));
        System.out.println(test);
    }

附Base64实现

package testest;

import java.io.UnsupportedEncodingException;

/**
 * Utilities for encoding and decoding the Base64 representation of
 * binary data.  See RFCs <a
 * href="http://www.ietf.org/rfc/rfc2045.txt">2045</a> and <a
 * href="http://www.ietf.org/rfc/rfc3548.txt">3548</a>.
 */
public class Base64 {
    /**
     * Default values for encoder/decoder flags.
     */
    public static final int DEFAULT = 0;

    /**
     * Encoder flag bit to omit the padding '=' characters at the end
     * of the output (if any).
     */
    public static final int NO_PADDING = 1;

    /**
     * Encoder flag bit to omit all line terminators (i.e., the output
     * will be on one long line).
     */
    public static final int NO_WRAP = 2;

    /**
     * Encoder flag bit to indicate lines should be terminated with a
     * CRLF pair instead of just an LF.  Has no effect if {@code
     * NO_WRAP} is specified as well.
     */
    public static final int CRLF = 4;

    /**
     * Encoder/decoder flag bit to indicate using the "URL and
     * filename safe" variant of Base64 (see RFC 3548 section 4) where
     * {@code -} and {@code _} are used in place of {@code +} and
     * {@code /}.
     */
    public static final int URL_SAFE = 8;

    /**
     * Flag to pass to {@link Base64OutputStream} to indicate that it
     * should not close the output stream it is wrapping when it
     * itself is closed.
     */
    public static final int NO_CLOSE = 16;

    //  --------------------------------------------------------
    //  shared code
    //  --------------------------------------------------------

    /* package */ static abstract class Coder {
        public byte[] output;
        public int op;

        /**
         * Encode/decode another block of input data.  this.output is
         * provided by the caller, and must be big enough to hold all
         * the coded data.  On exit, this.opwill be set to the length
         * of the coded data.
         *
         * @param finish true if this is the final call to process for
         *        this object.  Will finalize the coder state and
         *        include any final bytes in the output.
         *
         * @return true if the input so far is good; false if some
         *         error has been detected in the input stream..
         */
        public abstract boolean process(byte[] input, int offset, int len, boolean finish);

        /**
         * @return the maximum number of bytes a call to process()
         * could produce for the given number of input bytes.  This may
         * be an overestimate.
         */
        public abstract int maxOutputSize(int len);
    }

    //  --------------------------------------------------------
    //  decoding
    //  --------------------------------------------------------

    /**
     * Decode the Base64-encoded data in input and return the data in
     * a new byte array.
     *
     * <p>The padding '=' characters at the end are considered optional, but
     * if any are present, there must be the correct number of them.
     *
     * @param str    the input String to decode, which is converted to
     *               bytes using the default charset
     * @param flags  controls certain features of the decoded output.
     *               Pass {@code DEFAULT} to decode standard Base64.
     *
     * @throws IllegalArgumentException if the input contains
     * incorrect padding
     */
    public static byte[] decode(String str, int flags) {
        return decode(str.getBytes(), flags);
    }

    /**
     * Decode the Base64-encoded data in input and return the data in
     * a new byte array.
     *
     * <p>The padding '=' characters at the end are considered optional, but
     * if any are present, there must be the correct number of them.
     *
     * @param input the input array to decode
     * @param flags  controls certain features of the decoded output.
     *               Pass {@code DEFAULT} to decode standard Base64.
     *
     * @throws IllegalArgumentException if the input contains
     * incorrect padding
     */
    public static byte[] decode(byte[] input, int flags) {
        return decode(input, 0, input.length, flags);
    }

    /**
     * Decode the Base64-encoded data in input and return the data in
     * a new byte array.
     *
     * <p>The padding '=' characters at the end are considered optional, but
     * if any are present, there must be the correct number of them.
     *
     * @param input  the data to decode
     * @param offset the position within the input array at which to start
     * @param len    the number of bytes of input to decode
     * @param flags  controls certain features of the decoded output.
     *               Pass {@code DEFAULT} to decode standard Base64.
     *
     * @throws IllegalArgumentException if the input contains
     * incorrect padding
     */
    public static byte[] decode(byte[] input, int offset, int len, int flags) {
        // Allocate space for the most data the input could represent.
        // (It could contain less if it contains whitespace, etc.)
        Decoder decoder = new Decoder(flags, new byte[len*3/4]);

        if (!decoder.process(input, offset, len, true)) {
            throw new IllegalArgumentException("bad base-64");
        }

        // Maybe we got lucky and allocated exactly enough output space.
        if (decoder.op == decoder.output.length) {
            return decoder.output;
        }

        // Need to shorten the array, so allocate a new one of the
        // right size and copy.
        byte[] temp = new byte[decoder.op];
        System.arraycopy(decoder.output, 0, temp, 0, decoder.op);
        return temp;
    }

    /* package */ static class Decoder extends Coder {
        /**
         * Lookup table for turning bytes into their position in the
         * Base64 alphabet.
         */
        private static final int DECODE[] = {
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -1, -1, 63,
            52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -2, -1, -1,
            -1,  0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14,
            15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, -1,
            -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
            41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
        };

        /**
         * Decode lookup table for the "web safe" variant (RFC 3548
         * sec. 4) where - and _ replace + and /.
         */
        private static final int DECODE_WEBSAFE[] = {
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, -1,
            52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -2, -1, -1,
            -1,  0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14,
            15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, 63,
            -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
            41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
            -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
        };

        /** Non-data values in the DECODE arrays. */
        private static final int SKIP = -1;
        private static final int EQUALS = -2;

        /**
         * States 0-3 are reading through the next input tuple.
         * State 4 is having read one '=' and expecting exactly
         * one more.
         * State 5 is expecting no more data or padding characters
         * in the input.
         * State 6 is the error state; an error has been detected
         * in the input and no future input can "fix" it.
         */
        private int state;   // state number (0 to 6)
        private int value;

        final private int[] alphabet;

        public Decoder(int flags, byte[] output) {
            this.output = output;

            alphabet = ((flags & URL_SAFE) == 0) ? DECODE : DECODE_WEBSAFE;
            state = 0;
            value = 0;
        }

        /**
         * @return an overestimate for the number of bytes {@code
         * len} bytes could decode to.
         */
        public int maxOutputSize(int len) {
            return len * 3/4 + 10;
        }

        /**
         * Decode another block of input data.
         *
         * @return true if the state machine is still healthy.  false if
         *         bad base-64 data has been detected in the input stream.
         */
        public boolean process(byte[] input, int offset, int len, boolean finish) {
            if (this.state == 6) return false;

            int p = offset;
            len += offset;

            // Using local variables makes the decoder about 12%
            // faster than if we manipulate the member variables in
            // the loop.  (Even alphabet makes a measurable
            // difference, which is somewhat surprising to me since
            // the member variable is final.)
            int state = this.state;
            int value = this.value;
            int op = 0;
            final byte[] output = this.output;
            final int[] alphabet = this.alphabet;

            while (p < len) {
                // Try the fast path:  we're starting a new tuple and the
                // next four bytes of the input stream are all data
                // bytes.  This corresponds to going through states
                // 0-1-2-3-0.  We expect to use this method for most of
                // the data.
                //
                // If any of the next four bytes of input are non-data
                // (whitespace, etc.), value will end up negative.  (All
                // the non-data values in decode are small negative
                // numbers, so shifting any of them up and or'ing them
                // together will result in a value with its top bit set.)
                //
                // You can remove this whole block and the output should
                // be the same, just slower.
                if (state == 0) {
                    while (p+4 <= len &&
                           (value = ((alphabet[input[p] & 0xff] << 18) |
                                     (alphabet[input[p+1] & 0xff] << 12) |
                                     (alphabet[input[p+2] & 0xff] << 6) |
                                     (alphabet[input[p+3] & 0xff]))) >= 0) {
                        output[op+2] = (byte) value;
                        output[op+1] = (byte) (value >> 8);
                        output[op] = (byte) (value >> 16);
                        op += 3;
                        p += 4;
                    }
                    if (p >= len) break;
                }

                // The fast path isn't available -- either we've read a
                // partial tuple, or the next four input bytes aren't all
                // data, or whatever.  Fall back to the slower state
                // machine implementation.

                int d = alphabet[input[p++] & 0xff];

                switch (state) {
                case 0:
                    if (d >= 0) {
                        value = d;
                        ++state;
                    } else if (d != SKIP) {
                        this.state = 6;
                        return false;
                    }
                    break;

                case 1:
                    if (d >= 0) {
                        value = (value << 6) | d;
                        ++state;
                    } else if (d != SKIP) {
                        this.state = 6;
                        return false;
                    }
                    break;

                case 2:
                    if (d >= 0) {
                        value = (value << 6) | d;
                        ++state;
                    } else if (d == EQUALS) {
                        // Emit the last (partial) output tuple;
                        // expect exactly one more padding character.
                        output[op++] = (byte) (value >> 4);
                        state = 4;
                    } else if (d != SKIP) {
                        this.state = 6;
                        return false;
                    }
                    break;

                case 3:
                    if (d >= 0) {
                        // Emit the output triple and return to state 0.
                        value = (value << 6) | d;
                        output[op+2] = (byte) value;
                        output[op+1] = (byte) (value >> 8);
                        output[op] = (byte) (value >> 16);
                        op += 3;
                        state = 0;
                    } else if (d == EQUALS) {
                        // Emit the last (partial) output tuple;
                        // expect no further data or padding characters.
                        output[op+1] = (byte) (value >> 2);
                        output[op] = (byte) (value >> 10);
                        op += 2;
                        state = 5;
                    } else if (d != SKIP) {
                        this.state = 6;
                        return false;
                    }
                    break;

                case 4:
                    if (d == EQUALS) {
                        ++state;
                    } else if (d != SKIP) {
                        this.state = 6;
                        return false;
                    }
                    break;

                case 5:
                    if (d != SKIP) {
                        this.state = 6;
                        return false;
                    }
                    break;
                }
            }

            if (!finish) {
                // We're out of input, but a future call could provide
                // more.
                this.state = state;
                this.value = value;
                this.op = op;
                return true;
            }

            // Done reading input.  Now figure out where we are left in
            // the state machine and finish up.

            switch (state) {
            case 0:
                // Output length is a multiple of three.  Fine.
                break;
            case 1:
                // Read one extra input byte, which isn't enough to
                // make another output byte.  Illegal.
                this.state = 6;
                return false;
            case 2:
                // Read two extra input bytes, enough to emit 1 more
                // output byte.  Fine.
                output[op++] = (byte) (value >> 4);
                break;
            case 3:
                // Read three extra input bytes, enough to emit 2 more
                // output bytes.  Fine.
                output[op++] = (byte) (value >> 10);
                output[op++] = (byte) (value >> 2);
                break;
            case 4:
                // Read one padding '=' when we expected 2.  Illegal.
                this.state = 6;
                return false;
            case 5:
                // Read all the padding '='s we expected and no more.
                // Fine.
                break;
            }

            this.state = state;
            this.op = op;
            return true;
        }
    }

    //  --------------------------------------------------------
    //  encoding
    //  --------------------------------------------------------

    /**
     * Base64-encode the given data and return a newly allocated
     * String with the result.
     *
     * @param input  the data to encode
     * @param flags  controls certain features of the encoded output.
     *               Passing {@code DEFAULT} results in output that
     *               adheres to RFC 2045.
     */
    public static String encodeToString(byte[] input, int flags) {
        try {
            return new String(encode(input, flags), "US-ASCII");
        } catch (UnsupportedEncodingException e) {
            // US-ASCII is guaranteed to be available.
            throw new AssertionError(e);
        }
    }

    /**
     * Base64-encode the given data and return a newly allocated
     * String with the result.
     *
     * @param input  the data to encode
     * @param offset the position within the input array at which to
     *               start
     * @param len    the number of bytes of input to encode
     * @param flags  controls certain features of the encoded output.
     *               Passing {@code DEFAULT} results in output that
     *               adheres to RFC 2045.
     */
    public static String encodeToString(byte[] input, int offset, int len, int flags) {
        try {
            return new String(encode(input, offset, len, flags), "US-ASCII");
        } catch (UnsupportedEncodingException e) {
            // US-ASCII is guaranteed to be available.
            throw new AssertionError(e);
        }
    }

    /**
     * Base64-encode the given data and return a newly allocated
     * byte[] with the result.
     *
     * @param input  the data to encode
     * @param flags  controls certain features of the encoded output.
     *               Passing {@code DEFAULT} results in output that
     *               adheres to RFC 2045.
     */
    public static byte[] encode(byte[] input, int flags) {
        return encode(input, 0, input.length, flags);
    }

    /**
     * Base64-encode the given data and return a newly allocated
     * byte[] with the result.
     *
     * @param input  the data to encode
     * @param offset the position within the input array at which to
     *               start
     * @param len    the number of bytes of input to encode
     * @param flags  controls certain features of the encoded output.
     *               Passing {@code DEFAULT} results in output that
     *               adheres to RFC 2045.
     */
    public static byte[] encode(byte[] input, int offset, int len, int flags) {
        Encoder encoder = new Encoder(flags, null);

        // Compute the exact length of the array we will produce.
        int output_len = len / 3 * 4;

        // Account for the tail of the data and the padding bytes, if any.
        if (encoder.do_padding) {
            if (len % 3 > 0) {
                output_len += 4;
            }
        } else {
            switch (len % 3) {
                case 0: break;
                case 1: output_len += 2; break;
                case 2: output_len += 3; break;
            }
        }

        // Account for the newlines, if any.
        if (encoder.do_newline && len > 0) {
            output_len += (((len-1) / (3 * Encoder.LINE_GROUPS)) + 1) *
                (encoder.do_cr ? 2 : 1);
        }

        encoder.output = new byte[output_len];
        encoder.process(input, offset, len, true);

        assert encoder.op == output_len;

        return encoder.output;
    }

    /* package */ static class Encoder extends Coder {
        /**
         * Emit a new line every this many output tuples.  Corresponds to
         * a 76-character line length (the maximum allowable according to
         * <a href="http://www.ietf.org/rfc/rfc2045.txt">RFC 2045</a>).
         */
        public static final int LINE_GROUPS = 19;

        /**
         * Lookup table for turning Base64 alphabet positions (6 bits)
         * into output bytes.
         */
        private static final byte ENCODE[] = {
            'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
            'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f',
            'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v',
            'w', 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/',
        };

        /**
         * Lookup table for turning Base64 alphabet positions (6 bits)
         * into output bytes.
         */
        private static final byte ENCODE_WEBSAFE[] = {
            'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
            'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f',
            'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v',
            'w', 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_',
        };

        final private byte[] tail;
        /* package */ int tailLen;
        private int count;

        final public boolean do_padding;
        final public boolean do_newline;
        final public boolean do_cr;
        final private byte[] alphabet;

        public Encoder(int flags, byte[] output) {
            this.output = output;

            do_padding = (flags & NO_PADDING) == 0;
            do_newline = (flags & NO_WRAP) == 0;
            do_cr = (flags & CRLF) != 0;
            alphabet = ((flags & URL_SAFE) == 0) ? ENCODE : ENCODE_WEBSAFE;

            tail = new byte[2];
            tailLen = 0;

            count = do_newline ? LINE_GROUPS : -1;
        }

        /**
         * @return an overestimate for the number of bytes {@code
         * len} bytes could encode to.
         */
        public int maxOutputSize(int len) {
            return len * 8/5 + 10;
        }

        public boolean process(byte[] input, int offset, int len, boolean finish) {
            // Using local variables makes the encoder about 9% faster.
            final byte[] alphabet = this.alphabet;
            final byte[] output = this.output;
            int op = 0;
            int count = this.count;

            int p = offset;
            len += offset;
            int v = -1;

            // First we need to concatenate the tail of the previous call
            // with any input bytes available now and see if we can empty
            // the tail.

            switch (tailLen) {
                case 0:
                    // There was no tail.
                    break;

                case 1:
                    if (p+2 <= len) {
                        // A 1-byte tail with at least 2 bytes of
                        // input available now.
                        v = ((tail[0] & 0xff) << 16) |
                            ((input[p++] & 0xff) << 8) |
                            (input[p++] & 0xff);
                        tailLen = 0;
                    };
                    break;

                case 2:
                    if (p+1 <= len) {
                        // A 2-byte tail with at least 1 byte of input.
                        v = ((tail[0] & 0xff) << 16) |
                            ((tail[1] & 0xff) << 8) |
                            (input[p++] & 0xff);
                        tailLen = 0;
                    }
                    break;
            }

            if (v != -1) {
                output[op++] = alphabet[(v >> 18) & 0x3f];
                output[op++] = alphabet[(v >> 12) & 0x3f];
                output[op++] = alphabet[(v >> 6) & 0x3f];
                output[op++] = alphabet[v & 0x3f];
                if (--count == 0) {
                    if (do_cr) output[op++] = '\r';
                    output[op++] = '\n';
                    count = LINE_GROUPS;
                }
            }

            // At this point either there is no tail, or there are fewer
            // than 3 bytes of input available.

            // The main loop, turning 3 input bytes into 4 output bytes on
            // each iteration.
            while (p+3 <= len) {
                v = ((input[p] & 0xff) << 16) |
                    ((input[p+1] & 0xff) << 8) |
                    (input[p+2] & 0xff);
                output[op] = alphabet[(v >> 18) & 0x3f];
                output[op+1] = alphabet[(v >> 12) & 0x3f];
                output[op+2] = alphabet[(v >> 6) & 0x3f];
                output[op+3] = alphabet[v & 0x3f];
                p += 3;
                op += 4;
                if (--count == 0) {
                    if (do_cr) output[op++] = '\r';
                    output[op++] = '\n';
                    count = LINE_GROUPS;
                }
            }

            if (finish) {
                // Finish up the tail of the input.  Note that we need to
                // consume any bytes in tail before any bytes
                // remaining in input; there should be at most two bytes
                // total.

                if (p-tailLen == len-1) {
                    int t = 0;
                    v = ((tailLen > 0 ? tail[t++] : input[p++]) & 0xff) << 4;
                    tailLen -= t;
                    output[op++] = alphabet[(v >> 6) & 0x3f];
                    output[op++] = alphabet[v & 0x3f];
                    if (do_padding) {
                        output[op++] = '=';
                        output[op++] = '=';
                    }
                    if (do_newline) {
                        if (do_cr) output[op++] = '\r';
                        output[op++] = '\n';
                    }
                } else if (p-tailLen == len-2) {
                    int t = 0;
                    v = (((tailLen > 1 ? tail[t++] : input[p++]) & 0xff) << 10) |
                        (((tailLen > 0 ? tail[t++] : input[p++]) & 0xff) << 2);
                    tailLen -= t;
                    output[op++] = alphabet[(v >> 12) & 0x3f];
                    output[op++] = alphabet[(v >> 6) & 0x3f];
                    output[op++] = alphabet[v & 0x3f];
                    if (do_padding) {
                        output[op++] = '=';
                    }
                    if (do_newline) {
                        if (do_cr) output[op++] = '\r';
                        output[op++] = '\n';
                    }
                } else if (do_newline && op > 0 && count != LINE_GROUPS) {
                    if (do_cr) output[op++] = '\r';
                    output[op++] = '\n';
                }

                assert tailLen == 0;
                assert p == len;
            } else {
                // Save the leftovers in tail to be consumed on the next
                // call to encodeInternal.

                if (p == len-1) {
                    tail[tailLen++] = input[p];
                } else if (p == len-2) {
                    tail[tailLen++] = input[p];
                    tail[tailLen++] = input[p+1];
                }
            }

            this.op = op;
            this.count = count;

            return true;
        }
    }

    private Base64() { }   // don't instantiate
}

结果

最终可以看到其站名公交车GPS速度等内容

******************************************************
2017-11-27 21:28:54
75611
0
大山子路口南
15
-1
-1
-1
-1
-1
-1
116.489674
39.986622
2017-11-27 21:29:02
0
******************************************************
0条留言