A very fast and memory efficient class to encode and decode to and from BASE64 in full accordance with RFC 2045. : Base64 « Development Class « Java






A very fast and memory efficient class to encode and decode to and from BASE64 in full accordance with RFC 2045.

     
/*
 * This file is part of aion-emu <aion-emu.com>.
 *
 * aion-emu is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * aion-emu is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with aion-emu.  If not, see <http://www.gnu.org/licenses/>.
 */
//package com.aionemu.commons.utils;

import java.util.Arrays;

/**
 * A very fast and memory efficient class to encode and decode to and from BASE64 in full accordance with RFC 2045.<br>
 * <br>
 * On Windows XP sp1 with 1.4.2_04 and later ;), this encoder and decoder is about 10 times faster on small arrays (10 -
 * 1000 bytes) and 2-3 times as fast on larger arrays (10000 - 1000000 bytes) compared to
 * <code>sun.misc.Encoder()/Decoder()</code>.<br>
 * <br>
 * <p/>
 * On byte arrays the encoder is about 20% faster than Jakarta Commons Base64 Codec for encode and about 50% faster for
 * decoding large arrays. This implementation is about twice as fast on very small arrays (&lt 30 bytes). If
 * source/destination is a <code>String</code> this version is about three times as fast due to the fact that the
 * Commons Codec result has to be recoded to a <code>String</code> from <code>byte[]</code>, which is very expensive.<br>
 * <br>
 * <p/>
 * This encode/decode algorithm doesn't create any temporary arrays as many other codecs do, it only allocates the
 * resulting array. This produces less garbage and it is possible to handle arrays twice as large as algorithms that
 * create a temporary array. (E.g. Jakarta Commons Codec). It is unknown whether Sun's
 * <code>sun.misc.Encoder()/Decoder()</code> produce temporary arrays but since performance is quite low it probably
 * does.<br>
 * <br>
 * <p/>
 * The encoder produces the same output as the Sun one except that the Sun's encoder appends a trailing line separator
 * if the last character isn't a pad. Unclear why but it only adds to the length and is probably a side effect. Both are
 * in conformance with RFC 2045 though.<br>
 * Commons codec seem to always att a trailing line separator.<br>
 * <br>
 * <p/>
 * <b>Note!</b> The encode/decode method pairs (types) come in three versions with the <b>exact</b> same algorithm and
 * thus a lot of code redundancy. This is to not create any temporary arrays for transcoding to/from different format
 * types. The methods not used can simply be commented out.<br>
 * <br>
 * <p/>
 * There is also a "fast" version of all decode methods that works the same way as the normal ones, but har a few
 * demands on the decoded input. Normally though, these fast verions should be used if the source if the input is known
 * and it hasn't bee tampered with.<br>
 * <br>
 * <p/>
 * If you find the code useful or you find a bug, please send me a note at base64 @ miginfocom . com.
 * <p/>
 * Licence (BSD): 
 * <p/>
 * Copyright (c) 2004, Mikael Grev, MiG InfoCom AB. (base64 @ miginfocom . com) All rights reserved.
 * <p/>
 * Redistribution and use in source and binary forms, with or without modification, are permitted provided that the
 * following conditions are met: Redistributions of source code must retain the above copyright notice, this list of
 * conditions and the following disclaimer. Redistributions in binary form must reproduce the above copyright notice,
 * this list of conditions and the following disclaimer in the documentation and/or other materials provided with the
 * distribution. Neither the name of the MiG InfoCom AB nor the names of its contributors may be used to endorse or
 * promote products derived from this software without specific prior written permission.
 * <p/>
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
 * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 * 
 * @author Mikael Grev Date: 2004-aug-02 Time: 11:31:11
 * @version 2.2
 */

public class Base64
{
  private static final char[]  CA  = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/".toCharArray();
  private static final int[]  IA  = new int[256];

  static
  {
    Arrays.fill(IA, -1);
    for (int i = 0, iS = CA.length; i < iS; i++)
      IA[CA[i]] = i;
    IA['='] = 0;
  }

  // ****************************************************************************************
  // * char[] version
  // ****************************************************************************************

  /**
   * Encodes a raw byte array into a BASE64 <code>char[]</code> representation i accordance with RFC 2045.
   * 
   * @param sArr
   *            The bytes to convert. If <code>null</code> or length 0 an empty array will be returned.
   * @param lineSep
   *            Optional "\r\n" after 76 characters, unless end of file.<br>
   *            No line separator will be in breach of RFC 2045 which specifies max 76 per line but will be a little
   *            faster.
   * @return A BASE64 encoded array. Never <code>null</code>.
   */
  public static char[] encodeToChar(byte[] sArr, boolean lineSep)
  {
    // Check special case
    int sLen = sArr != null ? sArr.length : 0;
    if (sLen == 0)
      return new char[0];

    int eLen = (sLen / 3) * 3; // Length of even 24-bits.
    int cCnt = ((sLen - 1) / 3 + 1) << 2; // Returned character count
    int dLen = cCnt + (lineSep ? (cCnt - 1) / 76 << 1 : 0); // Length of returned array
    char[] dArr = new char[dLen];

    // Encode even 24-bits
    for (int s = 0, d = 0, cc = 0; s < eLen;)
    {
      // Copy next three bytes into lower 24 bits of int, paying attension to sign.
      int i = (sArr[s++] & 0xff) << 16 | (sArr[s++] & 0xff) << 8 | (sArr[s++] & 0xff);

      // Encode the int into four chars
      dArr[d++] = CA[(i >>> 18) & 0x3f];
      dArr[d++] = CA[(i >>> 12) & 0x3f];
      dArr[d++] = CA[(i >>> 6) & 0x3f];
      dArr[d++] = CA[i & 0x3f];

      // Add optional line separator
      if (lineSep && ++cc == 19 && d < dLen - 2)
      {
        dArr[d++] = '\r';
        dArr[d++] = '\n';
        cc = 0;
      }
    }

    // Pad and encode last bits if source isn't even 24 bits.
    int left = sLen - eLen; // 0 - 2.
    if (left > 0)
    {
      // Prepare the int
      int i = ((sArr[eLen] & 0xff) << 10) | (left == 2 ? ((sArr[sLen - 1] & 0xff) << 2) : 0);

      // Set last four chars
      dArr[dLen - 4] = CA[i >> 12];
      dArr[dLen - 3] = CA[(i >>> 6) & 0x3f];
      dArr[dLen - 2] = left == 2 ? CA[i & 0x3f] : '=';
      dArr[dLen - 1] = '=';
    }
    return dArr;
  }

  /**
   * Decodes a BASE64 encoded char array. All illegal characters will be ignored and can handle both arrays with and
   * without line separators.
   * 
   * @param sArr
   *            The source array. <code>null</code> or length 0 will return an empty array.
   * @return The decoded array of bytes. May be of length 0. Will be <code>null</code> if the legal characters
   *         (including '=') isn't divideable by 4. (I.e. definitely corrupted).
   */
  public static byte[] decode(char[] sArr)
  {
    // Check special case
    int sLen = sArr != null ? sArr.length : 0;
    if (sLen == 0)
      return new byte[0];

    // Count illegal characters (including '\r', '\n') to know what size the returned array will be,
    // so we don't have to reallocate & copy it later.
    int sepCnt = 0; // Number of separator characters. (Actually illegal characters, but that's a bonus...)
    for (int i = 0; i < sLen; i++)
      // If input is "pure" (I.e. no line separators or illegal chars) base64 this loop can be commented out.
      if (IA[sArr[i]] < 0)
        sepCnt++;

    // Check so that legal chars (including '=') are evenly divideable by 4 as specified in RFC 2045.
    if ((sLen - sepCnt) % 4 != 0)
      return null;

    int pad = 0;
    for (int i = sLen; i > 1 && IA[sArr[--i]] <= 0;)
      if (sArr[i] == '=')
        pad++;

    int len = ((sLen - sepCnt) * 6 >> 3) - pad;

    byte[] dArr = new byte[len]; // Preallocate byte[] of exact length

    for (int s = 0, d = 0; d < len;)
    {
      // Assemble three bytes into an int from four "valid" characters.
      int i = 0;
      for (int j = 0; j < 4; j++)
      { // j only increased if a valid char was found.
        int c = IA[sArr[s++]];
        if (c >= 0)
          i |= c << (18 - j * 6);
        else
          j--;
      }
      // Add the bytes
      dArr[d++] = (byte) (i >> 16);
      if (d < len)
      {
        dArr[d++] = (byte) (i >> 8);
        if (d < len)
          dArr[d++] = (byte) i;
      }
    }
    return dArr;
  }

  /**
   * Decodes a BASE64 encoded char array that is known to be resonably well formatted. The method is about twice as
   * fast as {@link #decode(char[])}. The preconditions are:<br>
   * + The array must have a line length of 76 chars OR no line separators at all (one line).<br>
   * + Line separator must be "\r\n", as specified in RFC 2045 + The array must not contain illegal characters within
   * the encoded string<br>
   * + The array CAN have illegal characters at the beginning and end, those will be dealt with appropriately.<br>
   * 
   * @param sArr
   *            The source array. Length 0 will return an empty array. <code>null</code> will throw an exception.
   * @return The decoded array of bytes. May be of length 0.
   */
  public static byte[] decodeFast(char[] sArr)
  {
    // Check special case
    int sLen = sArr.length;
    if (sLen == 0)
      return new byte[0];

    int sIx = 0, eIx = sLen - 1; // Start and end index after trimming.

    // Trim illegal chars from start
    while (sIx < eIx && IA[sArr[sIx]] < 0)
      sIx++;

    // Trim illegal chars from end
    while (eIx > 0 && IA[sArr[eIx]] < 0)
      eIx--;

    // get the padding count (=) (0, 1 or 2)
    int pad = sArr[eIx] == '=' ? (sArr[eIx - 1] == '=' ? 2 : 1) : 0; // Count '=' at end.
    int cCnt = eIx - sIx + 1; // Content count including possible separators
    int sepCnt = sLen > 76 ? (sArr[76] == '\r' ? cCnt / 78 : 0) << 1 : 0;

    int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of decoded bytes
    byte[] dArr = new byte[len]; // Preallocate byte[] of exact length

    // Decode all but the last 0 - 2 bytes.
    int d = 0;
    for (int cc = 0, eLen = (len / 3) * 3; d < eLen;)
    {
      // Assemble three bytes into an int from four "valid" characters.
      int i = IA[sArr[sIx++]] << 18 | IA[sArr[sIx++]] << 12 | IA[sArr[sIx++]] << 6 | IA[sArr[sIx++]];

      // Add the bytes
      dArr[d++] = (byte) (i >> 16);
      dArr[d++] = (byte) (i >> 8);
      dArr[d++] = (byte) i;

      // If line separator, jump over it.
      if (sepCnt > 0 && ++cc == 19)
      {
        sIx += 2;
        cc = 0;
      }
    }

    if (d < len)
    {
      // Decode last 1-3 bytes (incl '=') into 1-3 bytes
      int i = 0;
      for (int j = 0; sIx <= eIx - pad; j++)
        i |= IA[sArr[sIx++]] << (18 - j * 6);

      for (int r = 16; d < len; r -= 8)
        dArr[d++] = (byte) (i >> r);
    }

    return dArr;
  }

  // ****************************************************************************************
  // * byte[] version
  // ****************************************************************************************

  /**
   * Encodes a raw byte array into a BASE64 <code>byte[]</code> representation i accordance with RFC 2045.
   * 
   * @param sArr
   *            The bytes to convert. If <code>null</code> or length 0 an empty array will be returned.
   * @param lineSep
   *            Optional "\r\n" after 76 characters, unless end of file.<br>
   *            No line separator will be in breach of RFC 2045 which specifies max 76 per line but will be a little
   *            faster.
   * @return A BASE64 encoded array. Never <code>null</code>.
   */
  public static byte[] encodeToByte(byte[] sArr, boolean lineSep)
  {
    // Check special case
    int sLen = sArr != null ? sArr.length : 0;
    if (sLen == 0)
      return new byte[0];

    int eLen = (sLen / 3) * 3; // Length of even 24-bits.
    int cCnt = ((sLen - 1) / 3 + 1) << 2; // Returned character count
    int dLen = cCnt + (lineSep ? (cCnt - 1) / 76 << 1 : 0); // Length of returned array
    byte[] dArr = new byte[dLen];

    // Encode even 24-bits
    for (int s = 0, d = 0, cc = 0; s < eLen;)
    {
      // Copy next three bytes into lower 24 bits of int, paying attension to sign.
      int i = (sArr[s++] & 0xff) << 16 | (sArr[s++] & 0xff) << 8 | (sArr[s++] & 0xff);

      // Encode the int into four chars
      dArr[d++] = (byte) CA[(i >>> 18) & 0x3f];
      dArr[d++] = (byte) CA[(i >>> 12) & 0x3f];
      dArr[d++] = (byte) CA[(i >>> 6) & 0x3f];
      dArr[d++] = (byte) CA[i & 0x3f];

      // Add optional line separator
      if (lineSep && ++cc == 19 && d < dLen - 2)
      {
        dArr[d++] = '\r';
        dArr[d++] = '\n';
        cc = 0;
      }
    }

    // Pad and encode last bits if source isn't an even 24 bits.
    int left = sLen - eLen; // 0 - 2.
    if (left > 0)
    {
      // Prepare the int
      int i = ((sArr[eLen] & 0xff) << 10) | (left == 2 ? ((sArr[sLen - 1] & 0xff) << 2) : 0);

      // Set last four chars
      dArr[dLen - 4] = (byte) CA[i >> 12];
      dArr[dLen - 3] = (byte) CA[(i >>> 6) & 0x3f];
      dArr[dLen - 2] = left == 2 ? (byte) CA[i & 0x3f] : (byte) '=';
      dArr[dLen - 1] = '=';
    }
    return dArr;
  }

  /**
   * Decodes a BASE64 encoded byte array. All illegal characters will be ignored and can handle both arrays with and
   * without line separators.
   * 
   * @param sArr
   *            The source array. Length 0 will return an empty array. <code>null</code> will throw an exception.
   * @return The decoded array of bytes. May be of length 0. Will be <code>null</code> if the legal characters
   *         (including '=') isn't divideable by 4. (I.e. definitely corrupted).
   */
  public static byte[] decode(byte[] sArr)
  {
    // Check special case
    int sLen = sArr.length;

    // Count illegal characters (including '\r', '\n') to know what size the returned array will be,
    // so we don't have to reallocate & copy it later.
    int sepCnt = 0; // Number of separator characters. (Actually illegal characters, but that's a bonus...)
    for (int i = 0; i < sLen; i++)
      // If input is "pure" (I.e. no line separators or illegal chars) base64 this loop can be commented out.
      if (IA[sArr[i] & 0xff] < 0)
        sepCnt++;

    // Check so that legal chars (including '=') are evenly divideable by 4 as specified in RFC 2045.
    if ((sLen - sepCnt) % 4 != 0)
      return null;

    int pad = 0;
    for (int i = sLen; i > 1 && IA[sArr[--i] & 0xff] <= 0;)
      if (sArr[i] == '=')
        pad++;

    int len = ((sLen - sepCnt) * 6 >> 3) - pad;

    byte[] dArr = new byte[len]; // Preallocate byte[] of exact length

    for (int s = 0, d = 0; d < len;)
    {
      // Assemble three bytes into an int from four "valid" characters.
      int i = 0;
      for (int j = 0; j < 4; j++)
      { // j only increased if a valid char was found.
        int c = IA[sArr[s++] & 0xff];
        if (c >= 0)
          i |= c << (18 - j * 6);
        else
          j--;
      }

      // Add the bytes
      dArr[d++] = (byte) (i >> 16);
      if (d < len)
      {
        dArr[d++] = (byte) (i >> 8);
        if (d < len)
          dArr[d++] = (byte) i;
      }
    }

    return dArr;
  }

  /**
   * Decodes a BASE64 encoded byte array that is known to be resonably well formatted. The method is about twice as
   * fast as {@link #decode(byte[])}. The preconditions are:<br>
   * + The array must have a line length of 76 chars OR no line separators at all (one line).<br>
   * + Line separator must be "\r\n", as specified in RFC 2045 + The array must not contain illegal characters within
   * the encoded string<br>
   * + The array CAN have illegal characters at the beginning and end, those will be dealt with appropriately.<br>
   * 
   * @param sArr
   *            The source array. Length 0 will return an empty array. <code>null</code> will throw an exception.
   * @return The decoded array of bytes. May be of length 0.
   */
  public static byte[] decodeFast(byte[] sArr)
  {
    // Check special case
    int sLen = sArr.length;
    if (sLen == 0)
      return new byte[0];

    int sIx = 0, eIx = sLen - 1; // Start and end index after trimming.

    // Trim illegal chars from start
    while (sIx < eIx && IA[sArr[sIx] & 0xff] < 0)
      sIx++;

    // Trim illegal chars from end
    while (eIx > 0 && IA[sArr[eIx] & 0xff] < 0)
      eIx--;

    // get the padding count (=) (0, 1 or 2)
    int pad = sArr[eIx] == '=' ? (sArr[eIx - 1] == '=' ? 2 : 1) : 0; // Count '=' at end.
    int cCnt = eIx - sIx + 1; // Content count including possible separators
    int sepCnt = sLen > 76 ? (sArr[76] == '\r' ? cCnt / 78 : 0) << 1 : 0;

    int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of decoded bytes
    byte[] dArr = new byte[len]; // Preallocate byte[] of exact length

    // Decode all but the last 0 - 2 bytes.
    int d = 0;
    for (int cc = 0, eLen = (len / 3) * 3; d < eLen;)
    {
      // Assemble three bytes into an int from four "valid" characters.
      int i = IA[sArr[sIx++]] << 18 | IA[sArr[sIx++]] << 12 | IA[sArr[sIx++]] << 6 | IA[sArr[sIx++]];

      // Add the bytes
      dArr[d++] = (byte) (i >> 16);
      dArr[d++] = (byte) (i >> 8);
      dArr[d++] = (byte) i;

      // If line separator, jump over it.
      if (sepCnt > 0 && ++cc == 19)
      {
        sIx += 2;
        cc = 0;
      }
    }

    if (d < len)
    {
      // Decode last 1-3 bytes (incl '=') into 1-3 bytes
      int i = 0;
      for (int j = 0; sIx <= eIx - pad; j++)
        i |= IA[sArr[sIx++]] << (18 - j * 6);

      for (int r = 16; d < len; r -= 8)
        dArr[d++] = (byte) (i >> r);
    }

    return dArr;
  }

  // ****************************************************************************************
  // * String version
  // ****************************************************************************************

  /**
   * Encodes a raw byte array into a BASE64 <code>String</code> representation i accordance with RFC 2045.
   * 
   * @param sArr
   *            The bytes to convert. If <code>null</code> or length 0 an empty array will be returned.
   * @param lineSep
   *            Optional "\r\n" after 76 characters, unless end of file.<br>
   *            No line separator will be in breach of RFC 2045 which specifies max 76 per line but will be a little
   *            faster.
   * @return A BASE64 encoded array. Never <code>null</code>.
   */
  public static String encodeToString(byte[] sArr, boolean lineSep)
  {
    // Reuse char[] since we can't create a String incrementally anyway and StringBuffer/Builder would be slower.
    return new String(encodeToChar(sArr, lineSep));
  }

  /**
   * Decodes a BASE64 encoded <code>String</code>. All illegal characters will be ignored and can handle both strings
   * with and without line separators.<br>
   * <b>Note!</b> It can be up to about 2x the speed to call <code>decode(str.toCharArray())</code> instead. That will
   * create a temporary array though. This version will use <code>str.charAt(i)</code> to iterate the string.
   * 
   * @param str
   *            The source string. <code>null</code> or length 0 will return an empty array.
   * @return The decoded array of bytes. May be of length 0. Will be <code>null</code> if the legal characters
   *         (including '=') isn't divideable by 4. (I.e. definitely corrupted).
   */
  public static byte[] decode(String str)
  {
    // Check special case
    int sLen = str != null ? str.length() : 0;
    if (sLen == 0)
      return new byte[0];

    // Count illegal characters (including '\r', '\n') to know what size the returned array will be,
    // so we don't have to reallocate & copy it later.
    int sepCnt = 0; // Number of separator characters. (Actually illegal characters, but that's a bonus...)
    for (int i = 0; i < sLen; i++)
      // If input is "pure" (I.e. no line separators or illegal chars) base64 this loop can be commented out.
      if (IA[str.charAt(i)] < 0)
        sepCnt++;

    // Check so that legal chars (including '=') are evenly divideable by 4 as specified in RFC 2045.
    if ((sLen - sepCnt) % 4 != 0)
      return null;

    // Count '=' at end
    int pad = 0;
    for (int i = sLen; i > 1 && IA[str.charAt(--i)] <= 0;)
      if (str.charAt(i) == '=')
        pad++;

    int len = ((sLen - sepCnt) * 6 >> 3) - pad;

    byte[] dArr = new byte[len]; // Preallocate byte[] of exact length

    for (int s = 0, d = 0; d < len;)
    {
      // Assemble three bytes into an int from four "valid" characters.
      int i = 0;
      for (int j = 0; j < 4; j++)
      { // j only increased if a valid char was found.
        int c = IA[str.charAt(s++)];
        if (c >= 0)
          i |= c << (18 - j * 6);
        else
          j--;
      }
      // Add the bytes
      dArr[d++] = (byte) (i >> 16);
      if (d < len)
      {
        dArr[d++] = (byte) (i >> 8);
        if (d < len)
          dArr[d++] = (byte) i;
      }
    }
    return dArr;
  }

  /**
   * Decodes a BASE64 encoded string that is known to be resonably well formatted. The method is about twice as fast
   * as {@link #decode(String)}. The preconditions are:<br>
   * + The array must have a line length of 76 chars OR no line separators at all (one line).<br>
   * + Line separator must be "\r\n", as specified in RFC 2045 + The array must not contain illegal characters within
   * the encoded string<br>
   * + The array CAN have illegal characters at the beginning and end, those will be dealt with appropriately.<br>
   * 
   * @param s
   *            The source string. Length 0 will return an empty array. <code>null</code> will throw an exception.
   * @return The decoded array of bytes. May be of length 0.
   */
  public static byte[] decodeFast(String s)
  {
    // Check special case
    int sLen = s.length();
    if (sLen == 0)
      return new byte[0];

    int sIx = 0, eIx = sLen - 1; // Start and end index after trimming.

    // Trim illegal chars from start
    while (sIx < eIx && IA[s.charAt(sIx) & 0xff] < 0)
      sIx++;

    // Trim illegal chars from end
    while (eIx > 0 && IA[s.charAt(eIx) & 0xff] < 0)
      eIx--;

    // get the padding count (=) (0, 1 or 2)
    int pad = s.charAt(eIx) == '=' ? (s.charAt(eIx - 1) == '=' ? 2 : 1) : 0; // Count '=' at end.
    int cCnt = eIx - sIx + 1; // Content count including possible separators
    int sepCnt = sLen > 76 ? (s.charAt(76) == '\r' ? cCnt / 78 : 0) << 1 : 0;

    int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of decoded bytes
    byte[] dArr = new byte[len]; // Preallocate byte[] of exact length

    // Decode all but the last 0 - 2 bytes.
    int d = 0;
    for (int cc = 0, eLen = (len / 3) * 3; d < eLen;)
    {
      // Assemble three bytes into an int from four "valid" characters.
      int i = IA[s.charAt(sIx++)] << 18 | IA[s.charAt(sIx++)] << 12 | IA[s.charAt(sIx++)] << 6 | IA[s.charAt(sIx++)];

      // Add the bytes
      dArr[d++] = (byte) (i >> 16);
      dArr[d++] = (byte) (i >> 8);
      dArr[d++] = (byte) i;

      // If line separator, jump over it.
      if (sepCnt > 0 && ++cc == 19)
      {
        sIx += 2;
        cc = 0;
      }
    }

    if (d < len)
    {
      // Decode last 1-3 bytes (incl '=') into 1-3 bytes
      int i = 0;
      for (int j = 0; sIx <= eIx - pad; j++)
        i |= IA[s.charAt(sIx++)] << (18 - j * 6);

      for (int r = 16; d < len; r -= 8)
        dArr[d++] = (byte) (i >> r);
    }

    return dArr;
  }
}

   
    
    
    
    
  








Related examples in the same category

1.Base64 encoding/decoding.
2.Decodes Base64 data into octects
3.Implementation of MIME's Base64 encoding and decoding conversions.
4.Encode/decode for RFC 2045 Base64 as defined by RFC 2045
5.Encode/decode for RFC 2045 Base64 as defined by RFC 2045, N. Freed and N. Borenstein.
6.Encodes and decodes to and from Base64 notation.
7.Encodes hex octects into Base64
8.Helper class to provide Base64 encoding routines.
9.Represents a collection of 64 boolean (on/off) flags.
10.byte to be tested if it is Base64 alphabet
11.to Base64
12.One of the fastest implementation of the Base64 encoding. Jakarta and others are slower
13.array of byte to encode
14.Codes number up to radix 62
15.A Base64 Encoder/Decoder
16.A fast and memory efficient class to encode and decode to and from BASE64 in full accordance with RFC 2045
17.BASE64 encoder implementation
18.Base-64 Encoder - translates from base-64 text into binary
19.Base64 Character encoder as specified in RFC1113
20.Base64 Utils
21.Base64 encoder/decoder
22.Base64 from by Funambol, Inc.
23.Convert to hex from byte arrays and back
24.Converting hexadecimal strings
25.Encode and decode data in Base64 format as described in RFC 1521
26.Encode and decode integers, times, and internationalized strings to and from popular binary formats
27.Encoding of raw bytes to base64-encoded characters, and decoding of base64 characters to raw bytes
28.Performs Base64 encoding and/or decoding
29.Provides Base64 encoding and decoding as defined by RFC 2045
30.Provides Base64 encoding and decoding with URL and filename safe alphabet as defined by RFC 3548, section 4.
31.Provides utility methods to Base64 encode data
32.QP Decoder Stream
33.QP Encoder Stream
34.A class to decode Base64 streams and strings.
35.A class to encode Base64 streams and strings.
36.Encodes binary data to plain text as Base64
37.Decodes InputStreams which contain binary data in base64 form
38.Base 64 Converter
39.Base64 from org.cspoker.common.util
40.Base64 converted from code at http://iharder.sourceforge.net/base64/
41.Encodes and decodes to and from Base64 notation.
42.Simple Base64 string decoding function
43.Class to represent unsigned 64-bit numbers.
44.A Base64 encoder/decoder.
45.The Base64 utility class implements Base-64 and Base-85 encoding and decoding algorithms.
46.Provides Base64 encoding and decoding
47.Code to read and write Base64-encoded text.
48.Base32 encoding/decoding class.