Backups/TwelveMonkeys
2
0
Fork 0
mirror of https://github.com/haraldk/TwelveMonkeys.git synced 2025-08-02 11:05:29 -04:00
Code Issues Projects Releases Wiki Activity

TMC-IOENC: Encoder implementation clean-up.

Browse Source
This commit is contained in:
Harald Kuhr 2013-09-13 17:04:10 +02:00
parent d1f00ce817
commit aebfad914f
9 changed files with 70 additions and 757 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

671
common/common-io/src/main/java/com/twelvemonkeys/io/enc/Base64.java
View File

@ -1,671 +0,0 @@
/*
* Copyright (c) 2008, Harald Kuhr
* All rights reserved.
*
* 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 "TwelveMonkeys" nor the
* names of its contributors may be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* 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.
*/
/*
* 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.
*/
package com.twelvemonkeys.io.enc;
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.
* <p/>
* 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()}.
* <p/>
* On byte arrays the encoder is about 20% faster than
* <a href="http://jakarta.apache.org/commons/codec/">Jakarta Commons Base64 Codec</a>
* 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} 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} from {@code byte[]}, which is very expensive.
* <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()} produce temporary arrays
* but since performance is quite low it probably does.
* <p/>
* The encoder produces the same output as the Sun one except that 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 add a trailing line separator.
* <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.
* <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.
* <p/>
* If you find the code useful or you find a bug, please send me a note at
* base64 @ miginfocom . com.
* <p/>
*
* @author Mikael Grev, 2004-aug-02 11:31:11
* @version 2.2
*/
final 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[]} representation im
* accordance with RFC 2045.
*
* @param sArr The bytes to convert. If {@code null} 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}.
*/
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} or length 0 will return
* an empty array.
* @return The decoded array of bytes. May be of length 0. Will be
* {@code null} 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} 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[]} representation i accordance with RFC 2045.
*
* @param sArr The bytes to convert. If {@code null} 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}.
*/
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} will throw an exception.
* @return The decoded array of bytes. May be of length 0. Will be {@code null} 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} 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} representation i accordance with RFC 2045.
*
* @param sArr The bytes to convert. If {@code null} 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}.
*/
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}. 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())} instead. That
* will create a temporary array though. This version will use {@code str.charAt(i)} to iterate the string.
*
* @param str The source string. {@code null} or length 0 will return an empty array.
* @return The decoded array of bytes. May be of length 0. Will be {@code null} 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} 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;
}
}

2
common/common-io/src/main/java/com/twelvemonkeys/io/enc/Base64Decoder.java
View File

@ -47,7 +47,7 @@ public final class Base64Decoder implements Decoder {
/**
* This array maps the characters to their 6 bit values
*/
final static char[] PEM_ARRAY = {
final static byte[] PEM_ARRAY = {
//0 1 2 3 4 5 6 7
'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', // 0
'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', // 1

66
common/common-io/src/main/java/com/twelvemonkeys/io/enc/Base64Encoder.java
View File

@ -30,6 +30,7 @@ package com.twelvemonkeys.io.enc;
import java.io.OutputStream;
import java.io.IOException;
import java.nio.ByteBuffer;
/**
* {@code Encoder} implementation for standard base64 encoding.
@ -44,15 +45,9 @@ import java.io.IOException;
*/
public class Base64Encoder implements Encoder {
public void encode(final OutputStream pStream, final byte[] pBuffer, final int pOffset, final int pLength)
public void encode(final OutputStream stream, final ByteBuffer buffer)
throws IOException
{
if (pOffset < 0 || pOffset > pLength || pOffset > pBuffer.length) {
throw new IndexOutOfBoundsException("offset outside [0...length]");
}
else if (pLength > pBuffer.length) {
throw new IndexOutOfBoundsException("length > buffer length");
}
// TODO: Implement
// NOTE: This is impossible, given the current spec, as we need to either:
@ -61,48 +56,47 @@ public class Base64Encoder implements Encoder {
// to ensure proper end of stream handling
int length;
int offset = pOffset;
// TODO: Temp impl, will only work for single writes
while ((pBuffer.length - offset) > 0) {
while (buffer.hasRemaining()) {
byte a, b, c;
if ((pBuffer.length - offset) > 2) {
length = 3;
}
else {
length = pBuffer.length - offset;
}
// if ((buffer.remaining()) > 2) {
// length = 3;
// }
// else {
// length = buffer.remaining();
// }
length = Math.min(3, buffer.remaining());
switch (length) {
case 1:
a = pBuffer[offset];
a = buffer.get();
b = 0;
pStream.write(Base64Decoder.PEM_ARRAY[(a >>> 2) & 0x3F]);
pStream.write(Base64Decoder.PEM_ARRAY[((a << 4) & 0x30) + ((b >>> 4) & 0xf)]);
pStream.write('=');
pStream.write('=');
offset++;
stream.write(Base64Decoder.PEM_ARRAY[(a >>> 2) & 0x3F]);
stream.write(Base64Decoder.PEM_ARRAY[((a << 4) & 0x30) + ((b >>> 4) & 0xf)]);
stream.write('=');
stream.write('=');
break;
case 2:
a = pBuffer[offset];
b = pBuffer[offset + 1];
a = buffer.get();
b = buffer.get();
c = 0;
pStream.write(Base64Decoder.PEM_ARRAY[(a >>> 2) & 0x3F]);
pStream.write(Base64Decoder.PEM_ARRAY[((a << 4) & 0x30) + ((b >>> 4) & 0xf)]);
pStream.write(Base64Decoder.PEM_ARRAY[((b << 2) & 0x3c) + ((c >>> 6) & 0x3)]);
pStream.write('=');
offset += offset + 2; // ???
stream.write(Base64Decoder.PEM_ARRAY[(a >>> 2) & 0x3F]);
stream.write(Base64Decoder.PEM_ARRAY[((a << 4) & 0x30) + ((b >>> 4) & 0xf)]);
stream.write(Base64Decoder.PEM_ARRAY[((b << 2) & 0x3c) + ((c >>> 6) & 0x3)]);
stream.write('=');
break;
default:
a = pBuffer[offset];
b = pBuffer[offset + 1];
c = pBuffer[offset + 2];
pStream.write(Base64Decoder.PEM_ARRAY[(a >>> 2) & 0x3F]);
pStream.write(Base64Decoder.PEM_ARRAY[((a << 4) & 0x30) + ((b >>> 4) & 0xf)]);
pStream.write(Base64Decoder.PEM_ARRAY[((b << 2) & 0x3c) + ((c >>> 6) & 0x3)]);
pStream.write(Base64Decoder.PEM_ARRAY[c & 0x3F]);
offset = offset + 3;
a = buffer.get();
b = buffer.get();
c = buffer.get();
stream.write(Base64Decoder.PEM_ARRAY[(a >>> 2) & 0x3F]);
stream.write(Base64Decoder.PEM_ARRAY[((a << 4) & 0x30) + ((b >>> 4) & 0xf)]);
stream.write(Base64Decoder.PEM_ARRAY[((b << 2) & 0x3c) + ((c >>> 6) & 0x3)]);
stream.write(Base64Decoder.PEM_ARRAY[c & 0x3F]);
break;
}
}

9
common/common-io/src/main/java/com/twelvemonkeys/io/enc/Encoder.java
View File

@ -30,6 +30,7 @@ package com.twelvemonkeys.io.enc;
import java.io.IOException;
import java.io.OutputStream;
import java.nio.ByteBuffer;
/**
* Interface for endcoders.
@ -50,14 +51,12 @@ public interface Encoder {
* Encodes up to {@code pBuffer.length} bytes into the given input stream,
* from the given buffer.
*
* @param pStream the outputstream to encode data to
* @param pBuffer buffer to read data from
* @param pOffset offset into the buffer array
* @param pLength length of data in the buffer
* @param stream the output stream to encode data to
* @param buffer buffer to read data from
*
* @throws java.io.IOException if an I/O error occurs
*/
void encode(OutputStream pStream, byte[] pBuffer, int pOffset, int pLength) throws IOException;
void encode(OutputStream stream, ByteBuffer buffer) throws IOException;
//TODO: int requiredBufferSize(): -1 == any, otherwise, use this buffer size
// void flush()?

27
common/common-io/src/main/java/com/twelvemonkeys/io/enc/EncoderStream.java
View File

@ -29,8 +29,9 @@
package com.twelvemonkeys.io.enc;
import java.io.FilterOutputStream;
import java.io.OutputStream;
import java.io.IOException;
import java.io.OutputStream;
import java.nio.ByteBuffer;
/**
* An {@code OutputStream} that provides on-the-fly encoding to an underlying
@ -47,8 +48,7 @@ public final class EncoderStream extends FilterOutputStream {
protected final Encoder encoder;
private final boolean flushOnWrite;
protected int bufferPos;
protected final byte[] buffer;
protected final ByteBuffer buffer;
/**
* Creates an output stream filter built on top of the specified
@ -76,8 +76,8 @@ public final class EncoderStream extends FilterOutputStream {
encoder = pEncoder;
flushOnWrite = pFlushOnWrite;
buffer = new byte[1024];
bufferPos = 0;
buffer = ByteBuffer.allocate(1024);
buffer.flip();
}
public void close() throws IOException {
@ -91,12 +91,12 @@ public final class EncoderStream extends FilterOutputStream {
}
private void encodeBuffer() throws IOException {
if (bufferPos != 0) {
if (buffer.hasRemaining()) {
// Make sure all remaining data in buffer is written to the stream
encoder.encode(out, buffer, 0, bufferPos);
encoder.encode(out, buffer);
// Reset buffer
bufferPos = 0;
buffer.clear();
}
}
@ -109,25 +109,24 @@ public final class EncoderStream extends FilterOutputStream {
// that the encoder can't buffer. In that case, the encoder should probably
// tell the EncoderStream how large buffer it prefers...
public void write(final byte[] pBytes, final int pOffset, final int pLength) throws IOException {
if (!flushOnWrite && bufferPos + pLength < buffer.length) {
if (!flushOnWrite && pLength < buffer.remaining()) {
// Buffer data
System.arraycopy(pBytes, pOffset, buffer, bufferPos, pLength);
bufferPos += pLength;
buffer.put(pBytes, pOffset, pLength);
}
else {
// Encode data already in the buffer
encodeBuffer();
// Encode rest without buffering
encoder.encode(out, pBytes, pOffset, pLength);
encoder.encode(out, ByteBuffer.wrap(pBytes, pOffset, pLength));
}
}
public void write(final int pByte) throws IOException {
if (bufferPos >= buffer.length - 1) {
if (!buffer.hasRemaining()) {
encodeBuffer(); // Resets bufferPos to 0
}
buffer[bufferPos++] = (byte) pByte;
buffer.put((byte) pByte);
}
}

24
common/common-io/src/main/java/com/twelvemonkeys/io/enc/PackBitsEncoder.java
View File

@ -30,6 +30,7 @@ package com.twelvemonkeys.io.enc;
import java.io.OutputStream;
import java.io.IOException;
import java.nio.ByteBuffer;
/**
* Encoder implementation for Apple PackBits run-length encoding.
@ -71,16 +72,17 @@ public final class PackBitsEncoder implements Encoder {
public PackBitsEncoder() {
}
public void encode(OutputStream pStream, byte[] pBuffer, int pOffset, int pLength) throws IOException {
public void encode(final OutputStream stream, final ByteBuffer buffer) throws IOException {
// NOTE: It's best to encode a 2 byte repeat
// run as a replicate run except when preceded and followed by a
// literal run, in which case it's best to merge the three into one
// literal run. Always encode 3 byte repeats as replicate runs.
// NOTE: Worst case: output = input + (input + 127) / 128
int offset = pOffset;
final int max = pOffset + pLength - 1;
int offset = buffer.position();
final int max = buffer.remaining() - 1;
final int maxMinus1 = max - 1;
final byte[] pBuffer = buffer.array();
while (offset <= max) {
// Compressed run
@ -93,31 +95,31 @@ public final class PackBitsEncoder implements Encoder {
if (run > 1) {
offset++;
pStream.write(-(run - 1));
pStream.write(replicate);
stream.write(-(run - 1));
stream.write(replicate);
}
// Literal run
run = 0;
while ((run < 128 && ((offset < max && pBuffer[offset] != pBuffer[offset + 1])
|| (offset < maxMinus1 && pBuffer[offset] != pBuffer[offset + 2])))) {
buffer[run++] = pBuffer[offset++];
this.buffer[run++] = pBuffer[offset++];
}
// If last byte, include it in literal run, if space
if (offset == max && run > 0 && run < 128) {
buffer[run++] = pBuffer[offset++];
this.buffer[run++] = pBuffer[offset++];
}
if (run > 0) {
pStream.write(run - 1);
pStream.write(buffer, 0, run);
stream.write(run - 1);
stream.write(this.buffer, 0, run);
}
// If last byte, and not space, start new literal run
if (offset == max && (run <= 0 || run >= 128)) {
pStream.write(0);
pStream.write(pBuffer[offset++]);
stream.write(0);
stream.write(pBuffer[offset++]);
}
}
}

17
common/common-io/src/test/java/com/twelvemonkeys/io/enc/Base64EncoderTestCase.java
View File

@ -2,7 +2,9 @@ package com.twelvemonkeys.io.enc;
import org.junit.Test;
import java.io.*;
import java.io.ByteArrayOutputStream;
import java.io.IOException;
import java.io.OutputStream;
import static org.junit.Assert.*;
@ -23,19 +25,6 @@ public class Base64EncoderTestCase extends EncoderAbstractTestCase {
return new Base64Decoder();
}
@Test
public void testNegativeEncode() throws IOException {
Encoder encoder = createEncoder();
ByteArrayOutputStream bytes = new ByteArrayOutputStream();
try {
encoder.encode(bytes, new byte[1], 2, 1);
fail("wrong index should throw IndexOutOfBoundsException");
}
catch (IndexOutOfBoundsException expected) {
}
}
@Test
public void testEmptyEncode() throws IOException {
String data = "";

2
common/common-io/src/test/java/com/twelvemonkeys/io/enc/EncoderAbstractTestCase.java
View File

@ -35,7 +35,7 @@ public abstract class EncoderAbstractTestCase extends ObjectAbstractTestCase {
ByteArrayOutputStream bytes = new ByteArrayOutputStream();
try {
encoder.encode(bytes, null, 0, 1);
encoder.encode(bytes, null);
fail("null should throw NullPointerException");
}
catch (NullPointerException expected) {

7
sandbox/sandbox-common/src/main/java/com/twelvemonkeys/io/enc/DeflateEncoder.java
View File

@ -30,6 +30,7 @@ package com.twelvemonkeys.io.enc;
import java.io.OutputStream;
import java.io.IOException;
import java.nio.ByteBuffer;
import java.util.zip.Deflater;
/**
@ -62,12 +63,12 @@ final class DeflateEncoder implements Encoder {
deflater = pDeflater;
}
public void encode(final OutputStream pStream, final byte[] pBuffer, final int pOffset, final int pLength)
public void encode(final OutputStream stream, ByteBuffer buffer)
throws IOException
{
System.out.println("DeflateEncoder.encode");
deflater.setInput(pBuffer, pOffset, pLength);
flushInputToStream(pStream);
deflater.setInput(buffer.array(), buffer.arrayOffset() + buffer.position(), buffer.remaining());
flushInputToStream(stream);
}
private void flushInputToStream(final OutputStream pStream) throws IOException {