/* base64.c -- Encode binary data using printable characters.
- Copyright (C) 1999, 2000, 2001, 2004 Free Software Foundation, Inc.
+ Copyright (C) 1999-2001, 2004-2006, 2009-2014 Free Software Foundation, Inc.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software Foundation,
- Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
+ along with this program; if not, see <http://www.gnu.org/licenses/>. */
/* Written by Simon Josefsson. Partially adapted from GNU MailUtils
* (mailbox/filter_trans.c, as of 2004-11-28). Improved by review
* from Paul Eggert, Bruno Haible, and Stepan Kasal.
*
- * See also RFC 3548 <http://www.ietf.org/rfc/rfc3548.txt>.
+ * See also RFC 4648 <http://www.ietf.org/rfc/rfc4648.txt>.
*
* Be careful with error checking. Here is how you would typically
* use these functions:
* FAIL: input too long
* if (out == NULL)
* FAIL: memory allocation error
- * OK: data in OUT/LEN.
+ * OK: data in OUT/OUTLEN.
*
*/
-#ifdef HAVE_CONFIG_H
-# include <config.h>
-#endif
+#include <config.h>
/* Get prototype. */
#include "base64.h"
/* Get malloc. */
#include <stdlib.h>
+/* Get UCHAR_MAX. */
+#include <limits.h>
+
+#include <string.h>
+
/* C89 compliant way to cast 'char' to 'unsigned char'. */
-static inline unsigned char
+static unsigned char
to_uchar (char ch)
{
return ch;
}
+static const char b64c[64] =
+ "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+
+/* Base64 encode IN array of size INLEN into OUT array. OUT needs
+ to be of length >= BASE64_LENGTH(INLEN), and INLEN needs to be
+ a multiple of 3. */
+static void
+base64_encode_fast (const char *restrict in, size_t inlen, char *restrict out)
+{
+ while (inlen)
+ {
+ *out++ = b64c[to_uchar (in[0]) >> 2];
+ *out++ = b64c[((to_uchar (in[0]) << 4) + (to_uchar (in[1]) >> 4)) & 0x3f];
+ *out++ = b64c[((to_uchar (in[1]) << 2) + (to_uchar (in[2]) >> 6)) & 0x3f];
+ *out++ = b64c[to_uchar (in[2]) & 0x3f];
+
+ inlen -= 3;
+ in += 3;
+ }
+}
+
/* Base64 encode IN array of size INLEN into OUT array of size OUTLEN.
If OUTLEN is less than BASE64_LENGTH(INLEN), write as many bytes as
possible. If OUTLEN is larger than BASE64_LENGTH(INLEN), also zero
terminate the output buffer. */
void
base64_encode (const char *restrict in, size_t inlen,
- char *restrict out, size_t outlen)
+ char *restrict out, size_t outlen)
{
- const char b64[64] =
- "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+ /* Note this outlen constraint can be enforced at compile time.
+ I.E. that the output buffer is exactly large enough to hold
+ the encoded inlen bytes. The inlen constraints (of corresponding
+ to outlen, and being a multiple of 3) can change at runtime
+ at the end of input. However the common case when reading
+ large inputs is to have both constraints satisfied, so we depend
+ on both in base_encode_fast(). */
+ if (outlen % 4 == 0 && inlen == outlen / 4 * 3)
+ {
+ base64_encode_fast (in, inlen, out);
+ return;
+ }
while (inlen && outlen)
{
- *out++ = b64[to_uchar (in[0]) >> 2];
+ *out++ = b64c[to_uchar (in[0]) >> 2];
if (!--outlen)
- break;
- *out++ = b64[((to_uchar (in[0]) << 4)
- + (--inlen ? to_uchar (in[1]) >> 4 : 0)) & 0x3f];
+ break;
+ *out++ = b64c[((to_uchar (in[0]) << 4)
+ + (--inlen ? to_uchar (in[1]) >> 4 : 0))
+ & 0x3f];
if (!--outlen)
- break;
+ break;
*out++ =
- (inlen
- ? b64[((to_uchar (in[1]) << 2)
- + (--inlen ? to_uchar (in[2]) >> 6 : 0)) & 0x3f] : '=');
+ (inlen
+ ? b64c[((to_uchar (in[1]) << 2)
+ + (--inlen ? to_uchar (in[2]) >> 6 : 0))
+ & 0x3f]
+ : '=');
if (!--outlen)
- break;
- *out++ = inlen ? b64[to_uchar (in[2]) & 0x3f] : '=';
+ break;
+ *out++ = inlen ? b64c[to_uchar (in[2]) & 0x3f] : '=';
if (!--outlen)
- break;
+ break;
if (inlen)
- inlen--;
- in += 3;
+ inlen--;
+ if (inlen)
+ in += 3;
}
if (outlen)
return, the OUT variable will hold a pointer to newly allocated
memory that must be deallocated by the caller. If output string
length would overflow, 0 is returned and OUT is set to NULL. If
- memory allocation fail, OUT is set to NULL, and the return value
- indicate length of the requested memory block, i.e.,
+ memory allocation failed, OUT is set to NULL, and the return value
+ indicates length of the requested memory block, i.e.,
BASE64_LENGTH(inlen) + 1. */
size_t
base64_encode_alloc (const char *in, size_t inlen, char **out)
}
*out = malloc (outlen);
- if (*out)
- base64_encode (in, inlen, *out, outlen);
+ if (!*out)
+ return outlen;
+
+ base64_encode (in, inlen, *out, outlen);
return outlen - 1;
}
Base64 alphabet (A-Za-z0-9+/) are encoded in 0..255. POSIX
1003.1-2001 require that char and unsigned char are 8-bit
quantities, though, taking care of that problem. But this may be a
- potential problem on non-POSIX C99 platforms. */
-#define B64(x) \
- ((x) == 'A' ? 0 \
- : (x) == 'B' ? 1 \
- : (x) == 'C' ? 2 \
- : (x) == 'D' ? 3 \
- : (x) == 'E' ? 4 \
- : (x) == 'F' ? 5 \
- : (x) == 'G' ? 6 \
- : (x) == 'H' ? 7 \
- : (x) == 'I' ? 8 \
- : (x) == 'J' ? 9 \
- : (x) == 'K' ? 10 \
- : (x) == 'L' ? 11 \
- : (x) == 'M' ? 12 \
- : (x) == 'N' ? 13 \
- : (x) == 'O' ? 14 \
- : (x) == 'P' ? 15 \
- : (x) == 'Q' ? 16 \
- : (x) == 'R' ? 17 \
- : (x) == 'S' ? 18 \
- : (x) == 'T' ? 19 \
- : (x) == 'U' ? 20 \
- : (x) == 'V' ? 21 \
- : (x) == 'W' ? 22 \
- : (x) == 'X' ? 23 \
- : (x) == 'Y' ? 24 \
- : (x) == 'Z' ? 25 \
- : (x) == 'a' ? 26 \
- : (x) == 'b' ? 27 \
- : (x) == 'c' ? 28 \
- : (x) == 'd' ? 29 \
- : (x) == 'e' ? 30 \
- : (x) == 'f' ? 31 \
- : (x) == 'g' ? 32 \
- : (x) == 'h' ? 33 \
- : (x) == 'i' ? 34 \
- : (x) == 'j' ? 35 \
- : (x) == 'k' ? 36 \
- : (x) == 'l' ? 37 \
- : (x) == 'm' ? 38 \
- : (x) == 'n' ? 39 \
- : (x) == 'o' ? 40 \
- : (x) == 'p' ? 41 \
- : (x) == 'q' ? 42 \
- : (x) == 'r' ? 43 \
- : (x) == 's' ? 44 \
- : (x) == 't' ? 45 \
- : (x) == 'u' ? 46 \
- : (x) == 'v' ? 47 \
- : (x) == 'w' ? 48 \
- : (x) == 'x' ? 49 \
- : (x) == 'y' ? 50 \
- : (x) == 'z' ? 51 \
- : (x) == '0' ? 52 \
- : (x) == '1' ? 53 \
- : (x) == '2' ? 54 \
- : (x) == '3' ? 55 \
- : (x) == '4' ? 56 \
- : (x) == '5' ? 57 \
- : (x) == '6' ? 58 \
- : (x) == '7' ? 59 \
- : (x) == '8' ? 60 \
- : (x) == '9' ? 61 \
- : (x) == '+' ? 62 \
- : (x) == '/' ? 63 \
+ potential problem on non-POSIX C99 platforms.
+
+ IBM C V6 for AIX mishandles "#define B64(x) ...'x'...", so use "_"
+ as the formal parameter rather than "x". */
+#define B64(_) \
+ ((_) == 'A' ? 0 \
+ : (_) == 'B' ? 1 \
+ : (_) == 'C' ? 2 \
+ : (_) == 'D' ? 3 \
+ : (_) == 'E' ? 4 \
+ : (_) == 'F' ? 5 \
+ : (_) == 'G' ? 6 \
+ : (_) == 'H' ? 7 \
+ : (_) == 'I' ? 8 \
+ : (_) == 'J' ? 9 \
+ : (_) == 'K' ? 10 \
+ : (_) == 'L' ? 11 \
+ : (_) == 'M' ? 12 \
+ : (_) == 'N' ? 13 \
+ : (_) == 'O' ? 14 \
+ : (_) == 'P' ? 15 \
+ : (_) == 'Q' ? 16 \
+ : (_) == 'R' ? 17 \
+ : (_) == 'S' ? 18 \
+ : (_) == 'T' ? 19 \
+ : (_) == 'U' ? 20 \
+ : (_) == 'V' ? 21 \
+ : (_) == 'W' ? 22 \
+ : (_) == 'X' ? 23 \
+ : (_) == 'Y' ? 24 \
+ : (_) == 'Z' ? 25 \
+ : (_) == 'a' ? 26 \
+ : (_) == 'b' ? 27 \
+ : (_) == 'c' ? 28 \
+ : (_) == 'd' ? 29 \
+ : (_) == 'e' ? 30 \
+ : (_) == 'f' ? 31 \
+ : (_) == 'g' ? 32 \
+ : (_) == 'h' ? 33 \
+ : (_) == 'i' ? 34 \
+ : (_) == 'j' ? 35 \
+ : (_) == 'k' ? 36 \
+ : (_) == 'l' ? 37 \
+ : (_) == 'm' ? 38 \
+ : (_) == 'n' ? 39 \
+ : (_) == 'o' ? 40 \
+ : (_) == 'p' ? 41 \
+ : (_) == 'q' ? 42 \
+ : (_) == 'r' ? 43 \
+ : (_) == 's' ? 44 \
+ : (_) == 't' ? 45 \
+ : (_) == 'u' ? 46 \
+ : (_) == 'v' ? 47 \
+ : (_) == 'w' ? 48 \
+ : (_) == 'x' ? 49 \
+ : (_) == 'y' ? 50 \
+ : (_) == 'z' ? 51 \
+ : (_) == '0' ? 52 \
+ : (_) == '1' ? 53 \
+ : (_) == '2' ? 54 \
+ : (_) == '3' ? 55 \
+ : (_) == '4' ? 56 \
+ : (_) == '5' ? 57 \
+ : (_) == '6' ? 58 \
+ : (_) == '7' ? 59 \
+ : (_) == '8' ? 60 \
+ : (_) == '9' ? 61 \
+ : (_) == '+' ? 62 \
+ : (_) == '/' ? 63 \
: -1)
static const signed char b64[0x100] = {
B64 (252), B64 (253), B64 (254), B64 (255)
};
+#if UCHAR_MAX == 255
+# define uchar_in_range(c) true
+#else
+# define uchar_in_range(c) ((c) <= 255)
+#endif
+
+/* Return true if CH is a character from the Base64 alphabet, and
+ false otherwise. Note that '=' is padding and not considered to be
+ part of the alphabet. */
bool
isbase64 (char ch)
{
- return to_uchar (ch) <= 255 && 0 <= b64[to_uchar (ch)];
+ return uchar_in_range (to_uchar (ch)) && 0 <= b64[to_uchar (ch)];
}
-/* Decode base64 encoded input array IN of length INLEN to output
- array OUT that can hold *OUTLEN bytes. Return true if decoding was
- successful, i.e. if the input was valid base64 data, false
- otherwise. If *OUTLEN is too small, as many bytes as possible will
- be written to OUT. On return, *OUTLEN holds the length of decoded
- bytes in OUT. Note that as soon as any non-alphabet characters are
- encountered, decoding is stopped and false is returned. */
-bool
-base64_decode (const char *restrict in, size_t inlen,
- char *restrict out, size_t * outlen)
+/* Initialize decode-context buffer, CTX. */
+void
+base64_decode_ctx_init (struct base64_decode_context *ctx)
{
- size_t outleft = *outlen;
+ ctx->i = 0;
+}
+
+/* If CTX->i is 0 or 4, there are four or more bytes in [*IN..IN_END), and
+ none of those four is a newline, then return *IN. Otherwise, copy up to
+ 4 - CTX->i non-newline bytes from that range into CTX->buf, starting at
+ index CTX->i and setting CTX->i to reflect the number of bytes copied,
+ and return CTX->buf. In either case, advance *IN to point to the byte
+ after the last one processed, and set *N_NON_NEWLINE to the number of
+ verified non-newline bytes accessible through the returned pointer. */
+static char *
+get_4 (struct base64_decode_context *ctx,
+ char const *restrict *in, char const *restrict in_end,
+ size_t *n_non_newline)
+{
+ if (ctx->i == 4)
+ ctx->i = 0;
- while (inlen >= 2)
+ if (ctx->i == 0)
{
- if (!isbase64 (in[0]) || !isbase64 (in[1]))
- break;
+ char const *t = *in;
+ if (4 <= in_end - *in && memchr (t, '\n', 4) == NULL)
+ {
+ /* This is the common case: no newline. */
+ *in += 4;
+ *n_non_newline = 4;
+ return (char *) t;
+ }
+ }
- if (outleft)
- {
- *out++ = ((b64[to_uchar (in[0])] << 2)
- | (b64[to_uchar (in[1])] >> 4));
- outleft--;
- }
+ {
+ /* Copy non-newline bytes into BUF. */
+ char const *p = *in;
+ while (p < in_end)
+ {
+ char c = *p++;
+ if (c != '\n')
+ {
+ ctx->buf[ctx->i++] = c;
+ if (ctx->i == 4)
+ break;
+ }
+ }
+
+ *in = p;
+ *n_non_newline = ctx->i;
+ return ctx->buf;
+ }
+}
- if (inlen == 2)
- break;
+#define return_false \
+ do \
+ { \
+ *outp = out; \
+ return false; \
+ } \
+ while (false)
+
+/* Decode up to four bytes of base64-encoded data, IN, of length INLEN
+ into the output buffer, *OUT, of size *OUTLEN bytes. Return true if
+ decoding is successful, false otherwise. If *OUTLEN is too small,
+ as many bytes as possible are written to *OUT. On return, advance
+ *OUT to point to the byte after the last one written, and decrement
+ *OUTLEN to reflect the number of bytes remaining in *OUT. */
+static bool
+decode_4 (char const *restrict in, size_t inlen,
+ char *restrict *outp, size_t *outleft)
+{
+ char *out = *outp;
+ if (inlen < 2)
+ return false;
- if (in[2] == '=')
- {
- if (inlen != 4)
- break;
+ if (!isbase64 (in[0]) || !isbase64 (in[1]))
+ return false;
+
+ if (*outleft)
+ {
+ *out++ = ((b64[to_uchar (in[0])] << 2)
+ | (b64[to_uchar (in[1])] >> 4));
+ --*outleft;
+ }
- if (in[3] != '=')
- break;
+ if (inlen == 2)
+ return_false;
- }
+ if (in[2] == '=')
+ {
+ if (inlen != 4)
+ return_false;
+
+ if (in[3] != '=')
+ return_false;
+ }
+ else
+ {
+ if (!isbase64 (in[2]))
+ return_false;
+
+ if (*outleft)
+ {
+ *out++ = (((b64[to_uchar (in[1])] << 4) & 0xf0)
+ | (b64[to_uchar (in[2])] >> 2));
+ --*outleft;
+ }
+
+ if (inlen == 3)
+ return_false;
+
+ if (in[3] == '=')
+ {
+ if (inlen != 4)
+ return_false;
+ }
else
- {
- if (!isbase64 (in[2]))
- break;
-
- if (outleft)
- {
- *out++ = (((b64[to_uchar (in[1])] << 4) & 0xf0)
- | (b64[to_uchar (in[2])] >> 2));
- outleft--;
- }
-
- if (inlen == 3)
- break;
-
- if (in[3] == '=')
- {
- if (inlen != 4)
- break;
- }
- else
- {
- if (!isbase64 (in[3]))
- break;
-
- if (outleft)
- {
- *out++ = (((b64[to_uchar (in[2])] << 6) & 0xc0)
- | b64[to_uchar (in[3])]);
- outleft--;
- }
- }
- }
-
- in += 4;
- inlen -= 4;
+ {
+ if (!isbase64 (in[3]))
+ return_false;
+
+ if (*outleft)
+ {
+ *out++ = (((b64[to_uchar (in[2])] << 6) & 0xc0)
+ | b64[to_uchar (in[3])]);
+ --*outleft;
+ }
+ }
}
- *outlen -= outleft;
+ *outp = out;
+ return true;
+}
- if (inlen != 0)
- return false;
+/* Decode base64-encoded input array IN of length INLEN to output array
+ OUT that can hold *OUTLEN bytes. The input data may be interspersed
+ with newlines. Return true if decoding was successful, i.e. if the
+ input was valid base64 data, false otherwise. If *OUTLEN is too
+ small, as many bytes as possible will be written to OUT. On return,
+ *OUTLEN holds the length of decoded bytes in OUT. Note that as soon
+ as any non-alphabet, non-newline character is encountered, decoding
+ is stopped and false is returned. If INLEN is zero, then process
+ only whatever data is stored in CTX.
- return true;
+ Initially, CTX must have been initialized via base64_decode_ctx_init.
+ Subsequent calls to this function must reuse whatever state is recorded
+ in that buffer. It is necessary for when a quadruple of base64 input
+ bytes spans two input buffers.
+
+ If CTX is NULL then newlines are treated as garbage and the input
+ buffer is processed as a unit. */
+
+bool
+base64_decode_ctx (struct base64_decode_context *ctx,
+ const char *restrict in, size_t inlen,
+ char *restrict out, size_t *outlen)
+{
+ size_t outleft = *outlen;
+ bool ignore_newlines = ctx != NULL;
+ bool flush_ctx = false;
+ unsigned int ctx_i = 0;
+
+ if (ignore_newlines)
+ {
+ ctx_i = ctx->i;
+ flush_ctx = inlen == 0;
+ }
+
+
+ while (true)
+ {
+ size_t outleft_save = outleft;
+ if (ctx_i == 0 && !flush_ctx)
+ {
+ while (true)
+ {
+ /* Save a copy of outleft, in case we need to re-parse this
+ block of four bytes. */
+ outleft_save = outleft;
+ if (!decode_4 (in, inlen, &out, &outleft))
+ break;
+
+ in += 4;
+ inlen -= 4;
+ }
+ }
+
+ if (inlen == 0 && !flush_ctx)
+ break;
+
+ /* Handle the common case of 72-byte wrapped lines.
+ This also handles any other multiple-of-4-byte wrapping. */
+ if (inlen && *in == '\n' && ignore_newlines)
+ {
+ ++in;
+ --inlen;
+ continue;
+ }
+
+ /* Restore OUT and OUTLEFT. */
+ out -= outleft_save - outleft;
+ outleft = outleft_save;
+
+ {
+ char const *in_end = in + inlen;
+ char const *non_nl;
+
+ if (ignore_newlines)
+ non_nl = get_4 (ctx, &in, in_end, &inlen);
+ else
+ non_nl = in; /* Might have nl in this case. */
+
+ /* If the input is empty or consists solely of newlines (0 non-newlines),
+ then we're done. Likewise if there are fewer than 4 bytes when not
+ flushing context and not treating newlines as garbage. */
+ if (inlen == 0 || (inlen < 4 && !flush_ctx && ignore_newlines))
+ {
+ inlen = 0;
+ break;
+ }
+ if (!decode_4 (non_nl, inlen, &out, &outleft))
+ break;
+
+ inlen = in_end - in;
+ }
+ }
+
+ *outlen -= outleft;
+
+ return inlen == 0;
}
/* Allocate an output buffer in *OUT, and decode the base64 encoded
size of the decoded data is stored in *OUTLEN. OUTLEN may be NULL,
if the caller is not interested in the decoded length. *OUT may be
NULL to indicate an out of memory error, in which case *OUTLEN
- contain the size of the memory block needed. The function return
+ contains the size of the memory block needed. The function returns
true on successful decoding and memory allocation errors. (Use the
*OUT and *OUTLEN parameters to differentiate between successful
- decoding and memory error.) The function return false if the input
- was invalid, in which case *OUT is NULL and *OUTLEN is
+ decoding and memory error.) The function returns false if the
+ input was invalid, in which case *OUT is NULL and *OUTLEN is
undefined. */
bool
-base64_decode_alloc (const char *in, size_t inlen, char **out,
- size_t * outlen)
+base64_decode_alloc_ctx (struct base64_decode_context *ctx,
+ const char *in, size_t inlen, char **out,
+ size_t *outlen)
{
- /* This may allocate a few bytes too much, depending on input,
- but it's not worth the extra CPU time to compute the exact amount.
- The exact amount is 3 * inlen / 4, minus 1 if the input ends
- with "=" and minus another 1 if the input ends with "==".
+ /* This may allocate a few bytes too many, depending on input,
+ but it's not worth the extra CPU time to compute the exact size.
+ The exact size is 3 * (inlen + (ctx ? ctx->i : 0)) / 4, minus 1 if the
+ input ends with "=" and minus another 1 if the input ends with "==".
Dividing before multiplying avoids the possibility of overflow. */
- size_t needlen = 3 * (inlen / 4) + 2;
+ size_t needlen = 3 * (inlen / 4) + 3;
*out = malloc (needlen);
if (!*out)
return true;
- if (!base64_decode (in, inlen, *out, &needlen))
+ if (!base64_decode_ctx (ctx, in, inlen, *out, &needlen))
{
free (*out);
*out = NULL;
projects / gnulib.git / blobdiff