1 /* gbase64.c - Base64 encoding/decoding
3 * Copyright (C) 2006 Alexander Larsson <alexl@redhat.com>
4 * Copyright (C) 2000-2003 Ximian Inc.
6 * This library is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Library General Public
8 * License as published by the Free Software Foundation; either
9 * version 2 of the License, or (at your option) any later version.
11 * This library is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Library General Public License for more details.
16 * You should have received a copy of the GNU Library General Public
17 * License along with this library; if not, see <http://www.gnu.org/licenses/>.
19 * This is based on code in camel, written by:
20 * Michael Zucchi <notzed@ximian.com>
21 * Jeffrey Stedfast <fejj@ximian.com>
29 #include "gtestutils.h"
35 * @title: Base64 Encoding
36 * @short_description: encodes and decodes data in Base64 format
38 * Base64 is an encoding that allows a sequence of arbitrary bytes to be
39 * encoded as a sequence of printable ASCII characters. For the definition
40 * of Base64, see <ulink url="http://www.ietf.org/rfc/rfc1421.txt">RFC
41 * 1421</ulink> or <ulink url="http://www.ietf.org/rfc/rfc2045.txt">RFC
42 * 2045</ulink>. Base64 is most commonly used as a MIME transfer encoding
45 * GLib supports incremental encoding using g_base64_encode_step() and
46 * g_base64_encode_close(). Incremental decoding can be done with
47 * g_base64_decode_step(). To encode or decode data in one go, use
48 * g_base64_encode() or g_base64_decode(). To avoid memory allocation when
49 * decoding, you can use g_base64_decode_inplace().
51 * Support for Base64 encoding has been added in GLib 2.12.
54 static const char base64_alphabet[] =
55 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
58 * g_base64_encode_step:
59 * @in: (array length=len) (element-type guint8): the binary data to encode
60 * @len: the length of @in
61 * @break_lines: whether to break long lines
62 * @out: (out) (array) (element-type guint8): pointer to destination buffer
63 * @state: (inout): Saved state between steps, initialize to 0
64 * @save: (inout): Saved state between steps, initialize to 0
66 * Incrementally encode a sequence of binary data into its Base-64 stringified
67 * representation. By calling this function multiple times you can convert
68 * data in chunks to avoid having to have the full encoded data in memory.
70 * When all of the data has been converted you must call
71 * g_base64_encode_close() to flush the saved state.
73 * The output buffer must be large enough to fit all the data that will
74 * be written to it. Due to the way base64 encodes you will need
75 * at least: (@len / 3 + 1) * 4 + 4 bytes (+ 4 may be needed in case of
76 * non-zero state). If you enable line-breaking you will need at least:
77 * ((@len / 3 + 1) * 4 + 4) / 72 + 1 bytes of extra space.
79 * @break_lines is typically used when putting base64-encoded data in emails.
80 * It breaks the lines at 72 columns instead of putting all of the text on
81 * the same line. This avoids problems with long lines in the email system.
82 * Note however that it breaks the lines with <literal>LF</literal>
83 * characters, not <literal>CR LF</literal> sequences, so the result cannot
84 * be passed directly to SMTP or certain other protocols.
86 * Return value: The number of bytes of output that was written
91 g_base64_encode_step (const guchar *in,
101 g_return_val_if_fail (in != NULL, 0);
102 g_return_val_if_fail (out != NULL, 0);
103 g_return_val_if_fail (state != NULL, 0);
104 g_return_val_if_fail (save != NULL, 0);
112 if (len + ((char *) save) [0] > 2)
114 const guchar *inend = in+len-2;
120 switch (((char *) save) [0])
123 c1 = ((unsigned char *) save) [1];
126 c1 = ((unsigned char *) save) [1];
127 c2 = ((unsigned char *) save) [2];
132 * yes, we jump into the loop, no i'm not going to change it,
135 while (inptr < inend)
142 *outptr++ = base64_alphabet [ c1 >> 2 ];
143 *outptr++ = base64_alphabet [ c2 >> 4 |
145 *outptr++ = base64_alphabet [ ((c2 &0x0f) << 2) |
147 *outptr++ = base64_alphabet [ c3 & 0x3f ];
148 /* this is a bit ugly ... */
149 if (break_lines && (++already) >= 19)
156 ((char *)save)[0] = 0;
157 len = 2 - (inptr - inend);
165 /* points to the slot for the next char to save */
166 saveout = & (((char *)save)[1]) + ((char *)save)[0];
168 /* len can only be 0 1 or 2 */
171 case 2: *saveout++ = *inptr++;
172 case 1: *saveout++ = *inptr++;
174 ((char *)save)[0] += len;
181 * g_base64_encode_close:
182 * @break_lines: whether to break long lines
183 * @out: (out) (array) (element-type guint8): pointer to destination buffer
184 * @state: (inout): Saved state from g_base64_encode_step()
185 * @save: (inout): Saved state from g_base64_encode_step()
187 * Flush the status from a sequence of calls to g_base64_encode_step().
189 * The output buffer must be large enough to fit all the data that will
190 * be written to it. It will need up to 4 bytes, or up to 5 bytes if
191 * line-breaking is enabled.
193 * Return value: The number of bytes of output that was written
198 g_base64_encode_close (gboolean break_lines,
206 g_return_val_if_fail (out != NULL, 0);
207 g_return_val_if_fail (state != NULL, 0);
208 g_return_val_if_fail (save != NULL, 0);
210 c1 = ((unsigned char *) save) [1];
211 c2 = ((unsigned char *) save) [2];
213 switch (((char *) save) [0])
216 outptr [2] = base64_alphabet[ ( (c2 &0x0f) << 2 ) ];
217 g_assert (outptr [2] != 0);
222 outptr [0] = base64_alphabet [ c1 >> 2 ];
223 outptr [1] = base64_alphabet [ c2 >> 4 | ( (c1&0x3) << 4 )];
239 * @data: (array length=len) (element-type guint8): the binary data to encode
240 * @len: the length of @data
242 * Encode a sequence of binary data into its Base-64 stringified
245 * Return value: (transfer full): a newly allocated, zero-terminated Base-64
246 * encoded string representing @data. The returned string must
247 * be freed with g_free().
252 g_base64_encode (const guchar *data,
256 gint state = 0, outlen;
259 g_return_val_if_fail (data != NULL || len == 0, NULL);
261 /* We can use a smaller limit here, since we know the saved state is 0,
262 +1 is needed for trailing \0, also check for unlikely integer overflow */
263 if (len >= ((G_MAXSIZE - 1) / 4 - 1) * 3)
264 g_error("%s: input too large for Base64 encoding (%"G_GSIZE_FORMAT" chars)",
267 out = g_malloc ((len / 3 + 1) * 4 + 1);
269 outlen = g_base64_encode_step (data, len, FALSE, out, &state, &save);
270 outlen += g_base64_encode_close (FALSE, out + outlen, &state, &save);
273 return (gchar *) out;
276 static const unsigned char mime_base64_rank[256] = {
277 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
278 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
279 255,255,255,255,255,255,255,255,255,255,255, 62,255,255,255, 63,
280 52, 53, 54, 55, 56, 57, 58, 59, 60, 61,255,255,255, 0,255,255,
281 255, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
282 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,255,255,255,255,255,
283 255, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
284 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,255,255,255,255,255,
285 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
286 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
287 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
288 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
289 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
290 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
291 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
292 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
296 * g_base64_decode_step:
297 * @in: (array length=len) (element-type guint8): binary input data
298 * @len: max length of @in data to decode
299 * @out: (out) (array) (element-type guint8): output buffer
300 * @state: (inout): Saved state between steps, initialize to 0
301 * @save: (inout): Saved state between steps, initialize to 0
303 * Incrementally decode a sequence of binary data from its Base-64 stringified
304 * representation. By calling this function multiple times you can convert
305 * data in chunks to avoid having to have the full encoded data in memory.
307 * The output buffer must be large enough to fit all the data that will
308 * be written to it. Since base64 encodes 3 bytes in 4 chars you need
309 * at least: (@len / 4) * 3 + 3 bytes (+ 3 may be needed in case of non-zero
312 * Return value: The number of bytes of output that was written
317 g_base64_decode_step (const gchar *in,
331 g_return_val_if_fail (in != NULL, 0);
332 g_return_val_if_fail (out != NULL, 0);
333 g_return_val_if_fail (state != NULL, 0);
334 g_return_val_if_fail (save != NULL, 0);
339 inend = (const guchar *)in+len;
342 /* convert 4 base64 bytes to 3 normal bytes */
346 last[0] = last[1] = 0;
348 /* we use the sign in the state to determine if we got a padding character
349 in the previous sequence */
356 inptr = (const guchar *)in;
357 while (inptr < inend)
360 rank = mime_base64_rank [c];
380 *state = last[0] == '=' ? -i : i;
387 * @text: zero-terminated string with base64 text to decode
388 * @out_len: (out): The length of the decoded data is written here
390 * Decode a sequence of Base-64 encoded text into binary data. Note
391 * that the returned binary data is not necessarily zero-terminated,
392 * so it should not be used as a character string.
394 * Return value: (transfer full) (array length=out_len) (element-type guint8):
395 * newly allocated buffer containing the binary data
396 * that @text represents. The returned buffer must
397 * be freed with g_free().
402 g_base64_decode (const gchar *text,
410 g_return_val_if_fail (text != NULL, NULL);
411 g_return_val_if_fail (out_len != NULL, NULL);
413 input_length = strlen (text);
415 /* We can use a smaller limit here, since we know the saved state is 0,
416 +1 used to avoid calling g_malloc0(0), and hence returning NULL */
417 ret = g_malloc0 ((input_length / 4) * 3 + 1);
419 *out_len = g_base64_decode_step (text, input_length, ret, &state, &save);
425 * g_base64_decode_inplace:
426 * @text: (inout) (array length=out_len) (element-type guint8): zero-terminated
427 * string with base64 text to decode
428 * @out_len: (inout): The length of the decoded data is written here
430 * Decode a sequence of Base-64 encoded text into binary data
431 * by overwriting the input data.
433 * Return value: (transfer none): The binary data that @text responds. This pointer
434 * is the same as the input @text.
439 g_base64_decode_inplace (gchar *text,
442 gint input_length, state = 0;
445 g_return_val_if_fail (text != NULL, NULL);
446 g_return_val_if_fail (out_len != NULL, NULL);
448 input_length = strlen (text);
450 g_return_val_if_fail (input_length > 1, NULL);
452 *out_len = g_base64_decode_step (text, input_length, (guchar *) text, &state, &save);
454 return (guchar *) text;