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, write to the
18 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
19 * Boston, MA 02111-1307, USA.
21 * This is based on code in camel, written by:
22 * Michael Zucchi <notzed@ximian.com>
23 * Jeffrey Stedfast <fejj@ximian.com>
37 * SECTION:base64 Base64 Encoding
38 * @short_description: encodes and decodes data in Base64 format
40 * Base64 is an encoding that allows a sequence of arbitrary bytes to be
41 * encoded as a sequence of printable ASCII characters. For the definition
42 * of Base64, see <ulink url="http://www.ietf.org/rfc/rfc1421.txt">RFC
43 * 1421</ulink> or <ulink url="http://www.ietf.org/rfc/rfc2045.txt">RFC
44 * 2045</ulink>. Base64 is most commonly used as a MIME transfer encoding
47 * GLib supports incremental encoding using g_base64_encode_step() and
48 * g_base64_encode_close(). Incremental decoding can be done with
49 * g_base64_decode_step(). To encode or decode data in one go, use
50 * g_base64_encode() or g_base64_decode(). To avoid memory allocation when
51 * decoding, you can use g_base64_decode_inplace().
53 * Support for Base64 encoding has been added in GLib 2.12.
56 static const char base64_alphabet[] =
57 "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
60 * g_base64_encode_step:
61 * @in: the binary data to encode
62 * @len: the length of @in
63 * @break_lines: whether to break long lines
64 * @out: pointer to destination buffer
65 * @state: Saved state between steps, initialize to 0
66 * @save: Saved state between steps, initialize to 0
68 * Incrementally encode a sequence of binary data into its Base-64 stringified
69 * representation. By calling this function multiple times you can convert
70 * data in chunks to avoid having to have the full encoded data in memory.
72 * When all of the data has been converted you must call
73 * g_base64_encode_close() to flush the saved state.
75 * The output buffer must be large enough to fit all the data that will
76 * be written to it. Due to the way base64 encodes you will need
77 * at least: (@len / 3 + 1) * 4 + 4 bytes (+ 4 may be needed in case of
78 * non-zero state). If you enable line-breaking you will need at least:
79 * ((@len / 3 + 1) * 4 + 4) / 72 + 1 bytes of extra space.
81 * @break_lines is typically used when putting base64-encoded data in emails.
82 * It breaks the lines at 72 columns instead of putting all of the text on
83 * the same line. This avoids problems with long lines in the email system.
85 * Return value: The number of bytes of output that was written
90 g_base64_encode_step (const guchar *in,
100 g_return_val_if_fail (in != NULL, 0);
101 g_return_val_if_fail (out != NULL, 0);
102 g_return_val_if_fail (state != NULL, 0);
103 g_return_val_if_fail (save != NULL, 0);
111 if (len + ((char *) save) [0] > 2)
113 const guchar *inend = in+len-2;
119 switch (((char *) save) [0])
122 c1 = ((unsigned char *) save) [1];
125 c1 = ((unsigned char *) save) [1];
126 c2 = ((unsigned char *) save) [2];
131 * yes, we jump into the loop, no i'm not going to change it,
134 while (inptr < inend)
141 *outptr++ = base64_alphabet [ c1 >> 2 ];
142 *outptr++ = base64_alphabet [ c2 >> 4 |
144 *outptr++ = base64_alphabet [ ((c2 &0x0f) << 2) |
146 *outptr++ = base64_alphabet [ c3 & 0x3f ];
147 /* this is a bit ugly ... */
148 if (break_lines && (++already) >= 19)
155 ((char *)save)[0] = 0;
156 len = 2 - (inptr - inend);
164 /* points to the slot for the next char to save */
165 saveout = & (((char *)save)[1]) + ((char *)save)[0];
167 /* len can only be 0 1 or 2 */
170 case 2: *saveout++ = *inptr++;
171 case 1: *saveout++ = *inptr++;
173 ((char *)save)[0] += len;
180 * g_base64_encode_close:
181 * @break_lines: whether to break long lines
182 * @out: pointer to destination buffer
183 * @state: Saved state from g_base64_encode_step()
184 * @save: Saved state from g_base64_encode_step()
186 * Flush the status from a sequence of calls to g_base64_encode_step().
188 * Return value: The number of bytes of output that was written
193 g_base64_encode_close (gboolean break_lines,
201 g_return_val_if_fail (out != NULL, 0);
202 g_return_val_if_fail (state != NULL, 0);
203 g_return_val_if_fail (save != NULL, 0);
205 c1 = ((unsigned char *) save) [1];
206 c2 = ((unsigned char *) save) [2];
208 switch (((char *) save) [0])
211 outptr [2] = base64_alphabet[ ( (c2 &0x0f) << 2 ) ];
212 g_assert (outptr [2] != 0);
217 outptr [0] = base64_alphabet [ c1 >> 2 ];
218 outptr [1] = base64_alphabet [ c2 >> 4 | ( (c1&0x3) << 4 )];
234 * @data: the binary data to encode
235 * @len: the length of @data
237 * Encode a sequence of binary data into its Base-64 stringified
240 * Return value: a newly allocated, zero-terminated Base-64 encoded
241 * string representing @data. The returned string must
242 * be freed with g_free().
247 g_base64_encode (const guchar *data,
251 gint state = 0, outlen;
254 g_return_val_if_fail (data != NULL, NULL);
255 g_return_val_if_fail (len > 0, NULL);
257 /* We can use a smaller limit here, since we know the saved state is 0,
258 +1 is needed for trailing \0, also check for unlikely integer overflow */
259 if (len >= ((G_MAXSIZE - 1) / 4 - 1) * 3)
260 g_error("%s: input too large for Base64 encoding (%"G_GSIZE_FORMAT" chars)",
263 out = g_malloc ((len / 3 + 1) * 4 + 1);
265 outlen = g_base64_encode_step (data, len, FALSE, out, &state, &save);
266 outlen += g_base64_encode_close (FALSE, out + outlen, &state, &save);
269 return (gchar *) out;
272 static const unsigned char mime_base64_rank[256] = {
273 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
274 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
275 255,255,255,255,255,255,255,255,255,255,255, 62,255,255,255, 63,
276 52, 53, 54, 55, 56, 57, 58, 59, 60, 61,255,255,255, 0,255,255,
277 255, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
278 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,255,255,255,255,255,
279 255, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
280 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,255,255,255,255,255,
281 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
282 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
283 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
284 255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
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,
292 * g_base64_decode_step:
293 * @in: binary input data
294 * @len: max length of @in data to decode
295 * @out: output buffer
296 * @state: Saved state between steps, initialize to 0
297 * @save: Saved state between steps, initialize to 0
299 * Incrementally decode a sequence of binary data from its Base-64 stringified
300 * representation. By calling this function multiple times you can convert
301 * data in chunks to avoid having to have the full encoded data in memory.
303 * The output buffer must be large enough to fit all the data that will
304 * be written to it. Since base64 encodes 3 bytes in 4 chars you need
305 * at least: (@len / 4) * 3 + 3 bytes (+ 3 may be needed in case of non-zero
308 * Return value: The number of bytes of output that was written
313 g_base64_decode_step (const gchar *in,
327 g_return_val_if_fail (in != NULL, 0);
328 g_return_val_if_fail (out != NULL, 0);
329 g_return_val_if_fail (state != NULL, 0);
330 g_return_val_if_fail (save != NULL, 0);
335 inend = (const guchar *)in+len;
338 /* convert 4 base64 bytes to 3 normal bytes */
341 inptr = (const guchar *)in;
342 last[0] = last[1] = 0;
343 while (inptr < inend)
346 rank = mime_base64_rank [c];
373 * @text: zero-terminated string with base64 text to decode
374 * @out_len: The length of the decoded data is written here
376 * Decode a sequence of Base-64 encoded text into binary data
378 * Return value: a newly allocated buffer containing the binary data
379 * that @text represents. The returned buffer must
380 * be freed with g_free().
385 g_base64_decode (const gchar *text,
393 g_return_val_if_fail (text != NULL, NULL);
394 g_return_val_if_fail (out_len != NULL, NULL);
396 input_length = strlen (text);
398 g_return_val_if_fail (input_length > 1, NULL);
400 /* We can use a smaller limit here, since we know the saved state is 0,
401 +1 used to avoid calling g_malloc0(0), and hence retruning NULL */
402 ret = g_malloc0 ((input_length / 4) * 3 + 1);
404 *out_len = g_base64_decode_step (text, input_length, ret, &state, &save);
410 * g_base64_decode_inplace:
411 * @text: zero-terminated string with base64 text to decode
412 * @out_len: The length of the decoded data is written here
414 * Decode a sequence of Base-64 encoded text into binary data
415 * by overwriting the input data.
417 * Return value: The binary data that @text responds. This pointer
418 * is the same as the input @text.
423 g_base64_decode_inplace (gchar *text,
426 gint input_length, state = 0;
429 g_return_val_if_fail (text != NULL, NULL);
430 g_return_val_if_fail (out_len != NULL, NULL);
432 input_length = strlen (text);
434 g_return_val_if_fail (input_length > 1, NULL);
436 *out_len = g_base64_decode_step (text, input_length, (guchar *) text, &state, &save);
438 return (guchar *) text;
442 #define __G_BASE64_C__
443 #include "galiasdef.c"