com.volcengine.tos.internal.util.base64.Base64 Maven / Gradle / Ivy
Show all versions of ve-tos-java-sdk Show documentation
package com.volcengine.tos.internal.util.base64;
import com.volcengine.tos.internal.util.StringUtils;
public class Base64 extends BaseNCodec {
/**
* BASE32 characters are 6 bits in length.
* They are formed by taking a block of 3 octets to form a 24-bit string,
* which is converted into 4 BASE64 characters.
*/
private static final int BITS_PER_ENCODED_BYTE = 6;
private static final int BYTES_PER_UNENCODED_BLOCK = 3;
private static final int BYTES_PER_ENCODED_BLOCK = 4;
/**
* This array is a lookup table that translates 6-bit positive integer index values into their "Base64 Alphabet"
* equivalents as specified in Table 1 of RFC 2045.
*
* Thanks to "commons" project in ws.apache.org for this code.
* http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
*/
private static final byte[] STANDARD_ENCODE_TABLE = {
'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', '+', '/'
};
/**
* This is a copy of the STANDARD_ENCODE_TABLE above, but with + and /
* changed to - and _ to make the encoded Base64 results more URL-SAFE.
* This table is only used when the Base64's mode is set to URL-SAFE.
*/
private static final byte[] URL_SAFE_ENCODE_TABLE = {
'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', '-', '_'
};
/**
* This array is a lookup table that translates Unicode characters drawn from the "Base64 Alphabet" (as specified
* in Table 1 of RFC 2045) into their 6-bit positive integer equivalents. Characters that are not in the Base64
* alphabet but fall within the bounds of the array are translated to -1.
*
* Note: '+' and '-' both decode to 62. '/' and '_' both decode to 63. This means decoder seamlessly handles both
* URL_SAFE and STANDARD base64. (The encoder, on the other hand, needs to know ahead of time what to emit).
*
* Thanks to "commons" project in ws.apache.org for this code.
* http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
*/
private static final byte[] DECODE_TABLE = {
// 0 1 2 3 4 5 6 7 8 9 A B C D E F
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 00-0f
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 10-1f
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, 62, -1, 63, // 20-2f + - /
52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1, // 30-3f 0-9
-1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, // 40-4f A-O
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, 63, // 50-5f P-Z _
-1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, // 60-6f a-o
41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 // 70-7a p-z
};
/**
* Base64 uses 6-bit fields.
*/
/** Mask used to extract 6 bits, used when encoding */
private static final int MASK_6BITS = 0x3f;
/** Mask used to extract 4 bits, used when decoding final trailing character. */
private static final int MASK_4BITS = 0xf;
/** Mask used to extract 2 bits, used when decoding final trailing character. */
private static final int MASK_2BITS = 0x3;
// The static final fields above are used for the original static byte[] methods on Base64.
// The private member fields below are used with the new streaming approach, which requires
// some state be preserved between calls of encode() and decode().
/**
* Decodes Base64 data into octets.
*
* Note: this method seamlessly handles data encoded in URL-safe or normal mode.
*
*
* @param base64Data
* Byte array containing Base64 data
* @return Array containing decoded data.
*/
public static byte[] decodeBase64(final byte[] base64Data) {
return new Base64().decode(base64Data);
}
/**
* Encodes binary data using the base64 algorithm but does not chunk the output.
*
* @param binaryData
* binary data to encode
* @return byte[] containing Base64 characters in their UTF-8 representation.
*/
public static byte[] encodeBase64(final byte[] binaryData) {
if (binaryData == null || binaryData.length == 0) {
return binaryData;
}
// Create this so can use the super-class method
// Also ensures that the same roundings are performed by the ctor and the code
final Base64 b64 = new Base64(0, CHUNK_SEPARATOR, false);
final long len = b64.getEncodedLength(binaryData);
if (len > Integer.MAX_VALUE) {
throw new IllegalArgumentException("Input array too big, the output array would be bigger (" +
len +
") than the specified maximum size of " +
Integer.MAX_VALUE);
}
return b64.encode(binaryData);
}
/**
* Encode table to use: either STANDARD or URL_SAFE. Note: the DECODE_TABLE above remains static because it is able
* to decode both STANDARD and URL_SAFE streams, but the encodeTable must be a member variable so we can switch
* between the two modes.
*/
private final byte[] encodeTable;
// Only one decode table currently; keep for consistency with Base32 code
private final byte[] decodeTable = DECODE_TABLE;
/**
* Line separator for encoding. Not used when decoding. Only used if lineLength > 0.
*/
private final byte[] lineSeparator;
/**
* Convenience variable to help us determine when our buffer is going to run out of room and needs resizing.
* {@code decodeSize = 3 + lineSeparator.length;}
*/
private final int decodeSize;
/**
* Convenience variable to help us determine when our buffer is going to run out of room and needs resizing.
* {@code encodeSize = 4 + lineSeparator.length;}
*/
private final int encodeSize;
/**
* Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
*
* When encoding the line length is 0 (no chunking), and the encoding table is STANDARD_ENCODE_TABLE.
*
*
*
* When decoding all variants are supported.
*
*/
public Base64() {
this(0);
}
/**
* Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
*
* When encoding the line length is given in the constructor, the line separator is CRLF, and the encoding table is
* STANDARD_ENCODE_TABLE.
*
*
* Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
*
*
* When decoding all variants are supported.
*
*
* @param lineLength
* Each line of encoded data will be at most of the given length (rounded down to nearest multiple of
* 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when
* decoding.
*/
public Base64(final int lineLength) {
this(lineLength, CHUNK_SEPARATOR);
}
/**
* Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
*
* When encoding the line length and line separator are given in the constructor, and the encoding table is
* STANDARD_ENCODE_TABLE.
*
*
* Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
*
*
* When decoding all variants are supported.
*
*
* @param lineLength
* Each line of encoded data will be at most of the given length (rounded down to nearest multiple of
* 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when
* decoding.
* @param lineSeparator
* Each line of encoded data will end with this sequence of bytes.
* @throws IllegalArgumentException
* Thrown when the provided lineSeparator included some base64 characters.
*/
public Base64(final int lineLength, final byte[] lineSeparator) {
this(lineLength, lineSeparator, false);
}
/**
* Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
*
* When encoding the line length and line separator are given in the constructor, and the encoding table is
* STANDARD_ENCODE_TABLE.
*
*
* Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
*
*
* When decoding all variants are supported.
*
*
* @param lineLength
* Each line of encoded data will be at most of the given length (rounded down to nearest multiple of
* 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when
* decoding.
* @param lineSeparator
* Each line of encoded data will end with this sequence of bytes.
* @param urlSafe
* Instead of emitting '+' and '/' we emit '-' and '_' respectively. urlSafe is only applied to encode
* operations. Decoding seamlessly handles both modes.
* Note: no padding is added when using the URL-safe alphabet.
* @throws IllegalArgumentException
* Thrown when the {@code lineSeparator} contains Base64 characters.
*/
public Base64(final int lineLength, final byte[] lineSeparator, final boolean urlSafe) {
this(lineLength, lineSeparator, urlSafe, DECODING_POLICY_DEFAULT);
}
/**
* Creates a Base64 codec used for decoding (all modes) and encoding in URL-unsafe mode.
*
* When encoding the line length and line separator are given in the constructor, and the encoding table is
* STANDARD_ENCODE_TABLE.
*
*
* Line lengths that aren't multiples of 4 will still essentially end up being multiples of 4 in the encoded data.
*
*
* When decoding all variants are supported.
*
*
* @param lineLength
* Each line of encoded data will be at most of the given length (rounded down to nearest multiple of
* 4). If lineLength <= 0, then the output will not be divided into lines (chunks). Ignored when
* decoding.
* @param lineSeparator
* Each line of encoded data will end with this sequence of bytes.
* @param urlSafe
* Instead of emitting '+' and '/' we emit '-' and '_' respectively. urlSafe is only applied to encode
* operations. Decoding seamlessly handles both modes.
* Note: no padding is added when using the URL-safe alphabet.
* @param decodingPolicy The decoding policy.
* @throws IllegalArgumentException
* Thrown when the {@code lineSeparator} contains Base64 characters.
*/
public Base64(final int lineLength, final byte[] lineSeparator, final boolean urlSafe, final CodecPolicy decodingPolicy) {
super(BYTES_PER_UNENCODED_BLOCK, BYTES_PER_ENCODED_BLOCK,
lineLength,
lineSeparator == null ? 0 : lineSeparator.length,
PAD_DEFAULT,
decodingPolicy);
// could be simplified if there is no requirement to reject invalid line sep when length <=0
// @see test case Base64Test.testConstructors()
if (lineSeparator != null) {
if (containsAlphabetOrPad(lineSeparator)) {
final String sep = StringUtils.newStringUtf8(lineSeparator);
throw new IllegalArgumentException("lineSeparator must not contain base64 characters: [" + sep + "]");
}
if (lineLength > 0){ // null line-sep forces no chunking rather than throwing IAE
this.encodeSize = BYTES_PER_ENCODED_BLOCK + lineSeparator.length;
this.lineSeparator = new byte[lineSeparator.length];
System.arraycopy(lineSeparator, 0, this.lineSeparator, 0, lineSeparator.length);
} else {
this.encodeSize = BYTES_PER_ENCODED_BLOCK;
this.lineSeparator = null;
}
} else {
this.encodeSize = BYTES_PER_ENCODED_BLOCK;
this.lineSeparator = null;
}
this.decodeSize = this.encodeSize - 1;
this.encodeTable = urlSafe ? URL_SAFE_ENCODE_TABLE : STANDARD_ENCODE_TABLE;
}
// Implementation of the Encoder Interface
/**
*
* Decodes all of the provided data, starting at inPos, for inAvail bytes. Should be called at least twice: once
* with the data to decode, and once with inAvail set to "-1" to alert decoder that EOF has been reached. The "-1"
* call is not necessary when decoding, but it doesn't hurt, either.
*
*
* Ignores all non-base64 characters. This is how chunked (e.g. 76 character) data is handled, since CR and LF are
* silently ignored, but has implications for other bytes, too. This method subscribes to the garbage-in,
* garbage-out philosophy: it will not check the provided data for validity.
*
*
* Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach.
* http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
*
*
* @param in
* byte[] array of ascii data to base64 decode.
* @param inPos
* Position to start reading data from.
* @param inAvail
* Amount of bytes available from input for decoding.
* @param context
* the context to be used
*/
@Override
void decode(final byte[] in, int inPos, final int inAvail, final Context context) {
if (context.eof) {
return;
}
if (inAvail < 0) {
context.eof = true;
}
for (int i = 0; i < inAvail; i++) {
final byte[] buffer = ensureBufferSize(decodeSize, context);
final byte b = in[inPos++];
if (b == pad) {
// We're done.
context.eof = true;
break;
}
if (b >= 0 && b < DECODE_TABLE.length) {
final int result = DECODE_TABLE[b];
if (result >= 0) {
context.modulus = (context.modulus+1) % BYTES_PER_ENCODED_BLOCK;
context.ibitWorkArea = (context.ibitWorkArea << BITS_PER_ENCODED_BYTE) + result;
if (context.modulus == 0) {
buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 16) & MASK_8BITS);
buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS);
buffer[context.pos++] = (byte) (context.ibitWorkArea & MASK_8BITS);
}
}
}
}
// Two forms of EOF as far as base64 decoder is concerned: actual
// EOF (-1) and first time '=' character is encountered in stream.
// This approach makes the '=' padding characters completely optional.
if (context.eof && context.modulus != 0) {
final byte[] buffer = ensureBufferSize(decodeSize, context);
// We have some spare bits remaining
// Output all whole multiples of 8 bits and ignore the rest
switch (context.modulus) {
// case 0 : // impossible, as excluded above
case 1 : // 6 bits - either ignore entirely, or raise an exception
validateTrailingCharacter();
break;
case 2 : // 12 bits = 8 + 4
validateCharacter(MASK_4BITS, context);
context.ibitWorkArea = context.ibitWorkArea >> 4; // dump the extra 4 bits
buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS);
break;
case 3 : // 18 bits = 8 + 8 + 2
validateCharacter(MASK_2BITS, context);
context.ibitWorkArea = context.ibitWorkArea >> 2; // dump 2 bits
buffer[context.pos++] = (byte) ((context.ibitWorkArea >> 8) & MASK_8BITS);
buffer[context.pos++] = (byte) ((context.ibitWorkArea) & MASK_8BITS);
break;
default:
throw new IllegalStateException("Impossible modulus " + context.modulus);
}
}
}
/**
*
* Encodes all of the provided data, starting at inPos, for inAvail bytes. Must be called at least twice: once with
* the data to encode, and once with inAvail set to "-1" to alert encoder that EOF has been reached, to flush last
* remaining bytes (if not multiple of 3).
*
* Note: no padding is added when encoding using the URL-safe alphabet.
*
* Thanks to "commons" project in ws.apache.org for the bitwise operations, and general approach.
* http://svn.apache.org/repos/asf/webservices/commons/trunk/modules/util/
*
*
* @param in
* byte[] array of binary data to base64 encode.
* @param inPos
* Position to start reading data from.
* @param inAvail
* Amount of bytes available from input for encoding.
* @param context
* the context to be used
*/
@Override
void encode(final byte[] in, int inPos, final int inAvail, final Context context) {
if (context.eof) {
return;
}
// inAvail < 0 is how we're informed of EOF in the underlying data we're
// encoding.
if (inAvail < 0) {
context.eof = true;
if (0 == context.modulus && lineLength == 0) {
return; // no leftovers to process and not using chunking
}
final byte[] buffer = ensureBufferSize(encodeSize, context);
final int savedPos = context.pos;
switch (context.modulus) { // 0-2
case 0 : // nothing to do here
break;
case 1 : // 8 bits = 6 + 2
// top 6 bits:
buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 2) & MASK_6BITS];
// remaining 2:
buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 4) & MASK_6BITS];
// URL-SAFE skips the padding to further reduce size.
if (encodeTable == STANDARD_ENCODE_TABLE) {
buffer[context.pos++] = pad;
buffer[context.pos++] = pad;
}
break;
case 2 : // 16 bits = 6 + 6 + 4
buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 10) & MASK_6BITS];
buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 4) & MASK_6BITS];
buffer[context.pos++] = encodeTable[(context.ibitWorkArea << 2) & MASK_6BITS];
// URL-SAFE skips the padding to further reduce size.
if (encodeTable == STANDARD_ENCODE_TABLE) {
buffer[context.pos++] = pad;
}
break;
default:
throw new IllegalStateException("Impossible modulus " + context.modulus);
}
context.currentLinePos += context.pos - savedPos; // keep track of current line position
// if currentPos == 0 we are at the start of a line, so don't add CRLF
if (lineLength > 0 && context.currentLinePos > 0) {
System.arraycopy(lineSeparator, 0, buffer, context.pos, lineSeparator.length);
context.pos += lineSeparator.length;
}
} else {
for (int i = 0; i < inAvail; i++) {
final byte[] buffer = ensureBufferSize(encodeSize, context);
context.modulus = (context.modulus+1) % BYTES_PER_UNENCODED_BLOCK;
int b = in[inPos++];
if (b < 0) {
b += 256;
}
context.ibitWorkArea = (context.ibitWorkArea << 8) + b; // BITS_PER_BYTE
if (0 == context.modulus) { // 3 bytes = 24 bits = 4 * 6 bits to extract
buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 18) & MASK_6BITS];
buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 12) & MASK_6BITS];
buffer[context.pos++] = encodeTable[(context.ibitWorkArea >> 6) & MASK_6BITS];
buffer[context.pos++] = encodeTable[context.ibitWorkArea & MASK_6BITS];
context.currentLinePos += BYTES_PER_ENCODED_BLOCK;
if (lineLength > 0 && lineLength <= context.currentLinePos) {
System.arraycopy(lineSeparator, 0, buffer, context.pos, lineSeparator.length);
context.pos += lineSeparator.length;
context.currentLinePos = 0;
}
}
}
}
}
/**
* Returns whether or not the {@code octet} is in the Base64 alphabet.
*
* @param octet
* The value to test
* @return {@code true} if the value is defined in the the Base64 alphabet {@code false} otherwise.
*/
@Override
protected boolean isInAlphabet(final byte octet) {
return octet >= 0 && octet < decodeTable.length && decodeTable[octet] != -1;
}
/**
* Validates whether decoding the final trailing character is possible in the context
* of the set of possible base 64 values.
*
* The character is valid if the lower bits within the provided mask are zero. This
* is used to test the final trailing base-64 digit is zero in the bits that will be discarded.
*
* @param emptyBitsMask The mask of the lower bits that should be empty
* @param context the context to be used
*
* @throws IllegalArgumentException if the bits being checked contain any non-zero value
*/
private void validateCharacter(final int emptyBitsMask, final Context context) {
if (isStrictDecoding() && (context.ibitWorkArea & emptyBitsMask) != 0) {
throw new IllegalArgumentException(
"Strict decoding: Last encoded character (before the paddings if any) is a valid base 64 alphabet but not a possible encoding. " +
"Expected the discarded bits from the character to be zero.");
}
}
/**
* Validates whether decoding allows an entire final trailing character that cannot be
* used for a complete byte.
*
* @throws IllegalArgumentException if strict decoding is enabled
*/
private void validateTrailingCharacter() {
if (isStrictDecoding()) {
throw new IllegalArgumentException(
"Strict decoding: Last encoded character (before the paddings if any) is a valid base 64 alphabet but not a possible encoding. " +
"Decoding requires at least two trailing 6-bit characters to create bytes.");
}
}
}