/* $Id: base64.cpp 40058 2012-02-10 13:10:26Z vboxsync $ */ /** @file * IPRT - Base64, MIME content transfer encoding. */ /* * Copyright (C) 2009 Oracle Corporation * * This file is part of VirtualBox Open Source Edition (OSE), as * available from http://www.virtualbox.org. This file is free software; * you can redistribute it and/or modify it under the terms of the GNU * General Public License (GPL) as published by the Free Software * Foundation, in version 2 as it comes in the "COPYING" file of the * VirtualBox OSE distribution. VirtualBox OSE is distributed in the * hope that it will be useful, but WITHOUT ANY WARRANTY of any kind. * * The contents of this file may alternatively be used under the terms * of the Common Development and Distribution License Version 1.0 * (CDDL) only, as it comes in the "COPYING.CDDL" file of the * VirtualBox OSE distribution, in which case the provisions of the * CDDL are applicable instead of those of the GPL. * * You may elect to license modified versions of this file under the * terms and conditions of either the GPL or the CDDL or both. */ /******************************************************************************* * Header Files * *******************************************************************************/ #include #include "internal/iprt.h" #include #include #include #ifdef RT_STRICT # include #endif /******************************************************************************* * Defined Constants And Macros * *******************************************************************************/ /** The line length used for encoding. */ #define RTBASE64_LINE_LEN 64 /** @name Special g_au8CharToVal values * @{ */ #define BASE64_SPACE 0xc0 #define BASE64_PAD 0xe0 #define BASE64_INVALID 0xff /** @} */ /******************************************************************************* * Global Variables * *******************************************************************************/ /** Base64 character to value. (RFC 2045) * ASSUMES ASCII / UTF-8. */ static const uint8_t g_au8CharToVal[256] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xc0, 0xc0, 0xc0, 0xc0, 0xc0, 0xff, 0xff, /* 0x00..0x0f */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0x10..0x1f */ 0xc0, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 62, 0xff, 0xff, 0xff, 63, /* 0x20..0x2f */ 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 0xff, 0xff, 0xff, 0xe0, 0xff, 0xff, /* 0x30..0x3f */ 0xff, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, /* 0x40..0x4f */ 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0x50..0x5f */ 0xff, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, /* 0x60..0x6f */ 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0x70..0x7f */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0x80..0x8f */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0x90..0x9f */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0xa0..0xaf */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0xb0..0xbf */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0xc0..0xcf */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0xd0..0xdf */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, /* 0xe0..0xef */ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff /* 0xf0..0xff */ }; /** Value to Base64 character. (RFC 2045) */ static const char g_szValToChar[64+1] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; #ifdef RT_STRICT /** * Perform table sanity checks on the first call. */ static void rtBase64Sanity(void) { static bool s_fSane = false; if (RT_UNLIKELY(!s_fSane)) { for (unsigned i = 0; i < 64; i++) { unsigned ch = g_szValToChar[i]; Assert(ch); Assert(g_au8CharToVal[ch] == i); } for (unsigned i = 0; i < 256; i++) { uint8_t u8 = g_au8CharToVal[i]; Assert( ( u8 == BASE64_INVALID && !RT_C_IS_ALNUM(i) && !RT_C_IS_SPACE(i)) || ( u8 == BASE64_PAD && i == '=') || ( u8 == BASE64_SPACE && ( RT_C_IS_SPACE(i) || (i == '\r'))) /* Carriage return is handled as a space character as well. */ || ( u8 < 64 && (unsigned)g_szValToChar[u8] == i)); } ASMAtomicWriteBool(&s_fSane, true); } } #endif /* RT_STRICT */ /** * Calculates the decoded data size for a Base64 encoded string. * * @returns The length in bytes. -1 if the encoding is bad. * * @param pszString The Base64 encoded string. * @param ppszEnd If not NULL, this will point to the first char * following the Base64 encoded text block. If * NULL the entire string is assumed to be Base64. */ RTDECL(ssize_t) RTBase64DecodedSize(const char *pszString, char **ppszEnd) { #ifdef RT_STRICT rtBase64Sanity(); #endif /* * Walk the string until a non-encoded or non-space character is encountered. */ uint32_t c6Bits = 0; uint8_t u8 = BASE64_INVALID; unsigned ch; AssertCompile(sizeof(char) == sizeof(uint8_t)); while ((ch = *pszString)) { u8 = g_au8CharToVal[ch]; if (u8 < 64) c6Bits++; else if (RT_UNLIKELY(u8 != BASE64_SPACE)) break; /* advance */ pszString++; } /* * Padding can only be found at the end and there is * only 1 or 2 padding chars. Deal with it first. */ unsigned cbPad = 0; if (u8 == BASE64_PAD) { cbPad = 1; c6Bits++; pszString++; while ((ch = *pszString)) { u8 = g_au8CharToVal[ch]; if (u8 != BASE64_SPACE) { if (u8 != BASE64_PAD) break; c6Bits++; cbPad++; } pszString++; } if (cbPad >= 3) return -1; } /* * Invalid char and no where to indicate where the * Base64 text ends? Return failure. */ if ( u8 == BASE64_INVALID && !ppszEnd && ch) return -1; /* * Recalc 6-bit to 8-bit and adjust for padding. */ size_t cb; if (c6Bits * 3 / 3 == c6Bits) { if ((c6Bits * 3 % 4) != 0) return -1; cb = c6Bits * 3 / 4; } else { if ((c6Bits * (uint64_t)3 % 4) != 0) return -1; cb = c6Bits * (uint64_t)3 / 4; } if (cb < cbPad) return -1; cb -= cbPad; if (ppszEnd) *ppszEnd = (char *)pszString; return cb; } RT_EXPORT_SYMBOL(RTBase64DecodedSize); /** * Decodes a Base64 encoded string into the buffer supplied by the caller. * * @returns IPRT status code. * @retval VERR_BUFFER_OVERFLOW if the buffer is too small. pcbActual will not * be set, nor will ppszEnd. * @retval VERR_INVALID_BASE64_ENCODING if the encoding is wrong. * * @param pszString The Base64 string. Whether the entire string or * just the start of the string is in Base64 depends * on whether ppszEnd is specified or not. * @param pvData Where to store the decoded data. * @param cbData The size of the output buffer that pvData points to. * @param pcbActual Where to store the actual number of bytes returned. * Optional. * @param ppszEnd Indicates that the string may contain other stuff * after the Base64 encoded data when not NULL. Will * be set to point to the first char that's not part of * the encoding. If NULL the entire string must be part * of the Base64 encoded data. */ RTDECL(int) RTBase64Decode(const char *pszString, void *pvData, size_t cbData, size_t *pcbActual, char **ppszEnd) { #ifdef RT_STRICT rtBase64Sanity(); #endif /* * Process input in groups of 4 input / 3 output chars. */ uint8_t u8Trio[3] = { 0, 0, 0 }; /* shuts up gcc */ uint8_t *pbData = (uint8_t *)pvData; uint8_t u8 = BASE64_INVALID; unsigned c6Bits = 0; unsigned ch; AssertCompile(sizeof(char) == sizeof(uint8_t)); for (;;) { /* The first 6-bit group. */ while ((u8 = g_au8CharToVal[ch = *pszString]) == BASE64_SPACE) pszString++; if (u8 >= 64) { c6Bits = 0; break; } u8Trio[0] = u8 << 2; pszString++; /* The second 6-bit group. */ while ((u8 = g_au8CharToVal[ch = *pszString]) == BASE64_SPACE) pszString++; if (u8 >= 64) { c6Bits = 1; break; } u8Trio[0] |= u8 >> 4; u8Trio[1] = u8 << 4; pszString++; /* The third 6-bit group. */ while ((u8 = g_au8CharToVal[ch = *pszString]) == BASE64_SPACE) pszString++; if (u8 >= 64) { c6Bits = 2; break; } u8Trio[1] |= u8 >> 2; u8Trio[2] = u8 << 6; pszString++; /* The fourth 6-bit group. */ while ((u8 = g_au8CharToVal[ch = *pszString]) == BASE64_SPACE) pszString++; if (u8 >= 64) { c6Bits = 3; break; } u8Trio[2] |= u8; pszString++; /* flush the trio */ if (cbData < 3) return VERR_BUFFER_OVERFLOW; cbData -= 3; pbData[0] = u8Trio[0]; pbData[1] = u8Trio[1]; pbData[2] = u8Trio[2]; pbData += 3; } /* * Padding can only be found at the end and there is * only 1 or 2 padding chars. Deal with it first. */ unsigned cbPad = 0; if (u8 == BASE64_PAD) { cbPad = 1; pszString++; while ((ch = *pszString)) { u8 = g_au8CharToVal[ch]; if (u8 != BASE64_SPACE) { if (u8 != BASE64_PAD) break; cbPad++; } pszString++; } if (cbPad >= 3) return VERR_INVALID_BASE64_ENCODING; } /* * Invalid char and no where to indicate where the * Base64 text ends? Return failure. */ if ( u8 == BASE64_INVALID && !ppszEnd && ch) return VERR_INVALID_BASE64_ENCODING; /* * Check padding vs. pending sextets, if anything left to do finish it off. */ if (c6Bits || cbPad) { if (c6Bits + cbPad != 4) return VERR_INVALID_BASE64_ENCODING; switch (c6Bits) { case 1: u8Trio[1] = u8Trio[2] = 0; break; case 2: u8Trio[2] = 0; break; case 3: default: break; } switch (3 - cbPad) { case 1: if (cbData < 1) return VERR_BUFFER_OVERFLOW; cbData--; pbData[0] = u8Trio[0]; pbData++; break; case 2: if (cbData < 2) return VERR_BUFFER_OVERFLOW; cbData -= 2; pbData[0] = u8Trio[0]; pbData[1] = u8Trio[1]; pbData += 2; break; default: break; } } /* * Set optional return values and return successfully. */ if (ppszEnd) *ppszEnd = (char *)pszString; if (pcbActual) *pcbActual = pbData - (uint8_t *)pvData; return VINF_SUCCESS; } RT_EXPORT_SYMBOL(RTBase64Decode); /** * Calculates the length of the Base64 encoding of a given number of bytes of * data. * * This will assume line breaks every 64 chars. A RTBase64EncodedLengthEx * function can be added if closer control over the output is found to be * required. * * @returns The Base64 string length. * @param cbData The number of bytes to encode. */ RTDECL(size_t) RTBase64EncodedLength(size_t cbData) { if (cbData * 8 / 8 != cbData) { AssertReturn(sizeof(size_t) == sizeof(uint64_t), ~(size_t)0); uint64_t cch = cbData * (uint64_t)8; while (cch % 24) cch += 8; cch /= 6; cch += ((cch - 1) / RTBASE64_LINE_LEN) * RTBASE64_EOL_SIZE; return cch; } size_t cch = cbData * 8; while (cch % 24) cch += 8; cch /= 6; cch += ((cch - 1) / RTBASE64_LINE_LEN) * RTBASE64_EOL_SIZE; return cch; } RT_EXPORT_SYMBOL(RTBase64EncodedLength); /** * Encodes the specifed data into a Base64 string, the caller supplies the * output buffer. * * This will make the same assumptions about line breaks and EOL size as * RTBase64EncodedLength() does. A RTBase64EncodeEx function can be added if * more strict control over the output formatting is found necessary. * * @returns IRPT status code. * @retval VERR_BUFFER_OVERFLOW if the output buffer is too small. The buffer * may contain an invalid Base64 string. * * @param pvData The data to encode. * @param cbData The number of bytes to encode. * @param pszBuf Where to put the Base64 string. * @param cbBuf The size of the output buffer, including the terminator. * @param pcchActual The actual number of characters returned. */ RTDECL(int) RTBase64Encode(const void *pvData, size_t cbData, char *pszBuf, size_t cbBuf, size_t *pcchActual) { /* * Process whole "trios" of input data. */ uint8_t u8A; uint8_t u8B; uint8_t u8C; size_t cbLineFeed = cbBuf - RTBASE64_LINE_LEN; const uint8_t *pbSrc = (const uint8_t *)pvData; char *pchDst = pszBuf; while (cbData >= 3) { if (cbBuf < 4 + 1) return VERR_BUFFER_OVERFLOW; /* encode */ u8A = pbSrc[0]; pchDst[0] = g_szValToChar[u8A >> 2]; u8B = pbSrc[1]; pchDst[1] = g_szValToChar[((u8A << 4) & 0x3f) | (u8B >> 4)]; u8C = pbSrc[2]; pchDst[2] = g_szValToChar[((u8B << 2) & 0x3f) | (u8C >> 6)]; pchDst[3] = g_szValToChar[u8C & 0x3f]; /* advance */ cbBuf -= 4; pchDst += 4; cbData -= 3; pbSrc += 3; /* deal out linefeeds */ if (cbBuf == cbLineFeed && cbData) { if (cbBuf < RTBASE64_EOL_SIZE + 1) return VERR_BUFFER_OVERFLOW; cbBuf -= RTBASE64_EOL_SIZE; if (RTBASE64_EOL_SIZE == 2) *pchDst++ = '\r'; *pchDst++ = '\n'; cbLineFeed = cbBuf - RTBASE64_LINE_LEN; } } /* * Deal with the odd bytes and string termination. */ if (cbData) { if (cbBuf < 4 + 1) return VERR_BUFFER_OVERFLOW; switch (cbData) { case 1: u8A = pbSrc[0]; pchDst[0] = g_szValToChar[u8A >> 2]; pchDst[1] = g_szValToChar[(u8A << 4) & 0x3f]; pchDst[2] = '='; pchDst[3] = '='; break; case 2: u8A = pbSrc[0]; pchDst[0] = g_szValToChar[u8A >> 2]; u8B = pbSrc[1]; pchDst[1] = g_szValToChar[((u8A << 4) & 0x3f) | (u8B >> 4)]; pchDst[2] = g_szValToChar[(u8B << 2) & 0x3f]; pchDst[3] = '='; break; } pchDst += 4; } *pchDst = '\0'; if (pcchActual) *pcchActual = pchDst - pszBuf; return VINF_SUCCESS; } RT_EXPORT_SYMBOL(RTBase64Encode);