Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/*
* Copyright (c) 1997, 2018 Oracle and/or its affiliates. All rights reserved.
*
* This program and the accompanying materials are made available under the
* terms of the Eclipse Public License v. 2.0, which is available at
* http://www.eclipse.org/legal/epl-2.0.
*
* This Source Code may also be made available under the following Secondary
* Licenses when the conditions for such availability set forth in the
* Eclipse Public License v. 2.0 are satisfied: GNU General Public License,
* version 2 with the GNU Classpath Exception, which is available at
* https://www.gnu.org/software/classpath/license.html.
*
* SPDX-License-Identifier: EPL-2.0 OR GPL-2.0 WITH Classpath-exception-2.0
*/
package com.sun.mail.imap.protocol;
import java.text.StringCharacterIterator;
import java.text.CharacterIterator;
/**
* See the BASE64MailboxEncoder for a description of the RFC2060 and how
* mailbox names should be encoded. This class will do the correct decoding
* for mailbox names.
*
* @author Christopher Cotton
*/
public class BASE64MailboxDecoder {
public static String decode(String original) {
if (original == null || original.length() == 0)
return original;
boolean changedString = false;
int copyTo = 0;
// it will always be less than the original
char[] chars = new char[original.length()];
StringCharacterIterator iter = new StringCharacterIterator(original);
for(char c = iter.first(); c != CharacterIterator.DONE;
c = iter.next()) {
if (c == '&') {
changedString = true;
copyTo = base64decode(chars, copyTo, iter);
} else {
chars[copyTo++] = c;
}
}
// now create our string from the char array
if (changedString) {
return new String(chars, 0, copyTo);
} else {
return original;
}
}
protected static int base64decode(char[] buffer, int offset,
CharacterIterator iter) {
boolean firsttime = true;
int leftover = -1;
while(true) {
// get the first byte
byte orig_0 = (byte) iter.next();
if (orig_0 == -1) break; // no more chars
if (orig_0 == '-') {
if (firsttime) {
// means we got the string "&-" which is turned into a "&"
buffer[offset++] = '&';
}
// we are done now
break;
}
firsttime = false;
// next byte
byte orig_1 = (byte) iter.next();
if (orig_1 == -1 || orig_1 == '-')
break; // no more chars, invalid base64
byte a, b, current;
a = pem_convert_array[orig_0 & 0xff];
b = pem_convert_array[orig_1 & 0xff];
// The first decoded byte
current = (byte)(((a << 2) & 0xfc) | ((b >>> 4) & 3));
// use the leftover to create a Unicode Character (2 bytes)
if (leftover != -1) {
buffer[offset++] = (char)(leftover << 8 | (current & 0xff));
leftover = -1;
} else {
leftover = current & 0xff;
}
byte orig_2 = (byte) iter.next();
if (orig_2 == '=') { // End of this BASE64 encoding
continue;
} else if (orig_2 == -1 || orig_2 == '-') {
break; // no more chars
}
// second decoded byte
a = b;
b = pem_convert_array[orig_2 & 0xff];
current = (byte)(((a << 4) & 0xf0) | ((b >>> 2) & 0xf));
// use the leftover to create a Unicode Character (2 bytes)
if (leftover != -1) {
buffer[offset++] = (char)(leftover << 8 | (current & 0xff));
leftover = -1;
} else {
leftover = current & 0xff;
}
byte orig_3 = (byte) iter.next();
if (orig_3 == '=') { // End of this BASE64 encoding
continue;
} else if (orig_3 == -1 || orig_3 == '-') {
break; // no more chars
}
// The third decoded byte
a = b;
b = pem_convert_array[orig_3 & 0xff];
current = (byte)(((a << 6) & 0xc0) | (b & 0x3f));
// use the leftover to create a Unicode Character (2 bytes)
if (leftover != -1) {
buffer[offset++] = (char)(leftover << 8 | (current & 0xff));
leftover = -1;
} else {
leftover = current & 0xff;
}
}
return offset;
}
/**
* This character array provides the character to value map
* based on RFC1521, but with the modification from RFC2060
* which changes the '/' to a ','.
*/
// shared with BASE64MailboxEncoder
static final char pem_array[] = {
'A','B','C','D','E','F','G','H', // 0
'I','J','K','L','M','N','O','P', // 1
'Q','R','S','T','U','V','W','X', // 2
'Y','Z','a','b','c','d','e','f', // 3
'g','h','i','j','k','l','m','n', // 4
'o','p','q','r','s','t','u','v', // 5
'w','x','y','z','0','1','2','3', // 6
'4','5','6','7','8','9','+',',' // 7
};
private static final byte pem_convert_array[] = new byte[256];
static {
for (int i = 0; i < 255; i++)
pem_convert_array[i] = -1;
for(int i = 0; i < pem_array.length; i++)
pem_convert_array[pem_array[i]] = (byte) i;
}
}