summaryrefslogtreecommitdiff
path: root/libbb/uuencode.c
blob: b4ee20c2007966184368305f38cdfc1a2134779a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
/* vi: set sw=4 ts=4: */
/*
 * Copyright 2003, Glenn McGrath
 * Copyright 2006, Rob Landley <rob@landley.net>
 * Copyright 2010, Denys Vlasenko
 *
 * Licensed under GPLv2 or later, see file LICENSE in this source tree.
 */
#include "libbb.h"

/* Conversion tables */
/* for base 64 */
const char bb_uuenc_tbl_base64[65 + 1] ALIGN1 = {
	'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
	'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
	'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X',
	'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f',
	'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n',
	'o', 'p', 'q', 'r', 's', 't', 'u', 'v',
	'w', 'x', 'y', 'z', '0', '1', '2', '3',
	'4', '5', '6', '7', '8', '9', '+', '/',
	'=' /* termination character */,
	'\0' /* needed for uudecode.c only */
};
#if ENABLE_BASE32
const char bb_uuenc_tbl_base32[33 + 1] ALIGN1 = {
	'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
	'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
	'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X',
	'Y', 'Z', '2', '3', '4', '5', '6', '7',
	'=',
	'\0'
};
#endif
const char bb_uuenc_tbl_std[65] ALIGN1 = {
	'`', '!', '"', '#', '$', '%', '&', '\'',
	'(', ')', '*', '+', ',', '-', '.', '/',
	'0', '1', '2', '3', '4', '5', '6', '7',
	'8', '9', ':', ';', '<', '=', '>', '?',
	'@', 'A', 'B', 'C', 'D', 'E', 'F', 'G',
	'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O',
	'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W',
	'X', 'Y', 'Z', '[', '\\', ']', '^', '_',
	'`' /* termination character */
};

/*
 * Encode bytes at S of length LENGTH to uuencode or base64 format and place it
 * to STORE.  STORE will be 0-terminated, and must point to a writable
 * buffer of at least 1+BASE64_LENGTH(length) bytes.
 * where BASE64_LENGTH(len) = (4 * ((LENGTH + 2) / 3))
 */
void FAST_FUNC bb_uuencode(char *p, const void *src, int length, const char *tbl)
{
	const unsigned char *s = src;

	/* Transform the 3x8 bits to 4x6 bits */
	while (length > 0) {
		unsigned s1, s2;

		/* Are s[1], s[2] valid or should be assumed 0? */
		s1 = s2 = 0;
		length -= 3; /* can be >=0, -1, -2 */
		if (length >= -1) {
			s1 = s[1];
			if (length >= 0)
				s2 = s[2];
		}
		*p++ = tbl[s[0] >> 2];
		*p++ = tbl[((s[0] & 3) << 4) + (s1 >> 4)];
		*p++ = tbl[((s1 & 0xf) << 2) + (s2 >> 6)];
		*p++ = tbl[s2 & 0x3f];
		s += 3;
	}
	/* Zero-terminate */
	*p = '\0';
	/* If length is -2 or -1, pad last char or two */
	while (length) {
		*--p = tbl[64];
		length++;
	}
}

/*
 * Decode base64 encoded string. Stops on '\0'.
 *
 * Returns: pointer to the undecoded part of source.
 * If points to '\0', then the source was fully decoded.
 * (*pp_dst): advanced past the last written byte.
 */
const char* FAST_FUNC decode_base64(char **pp_dst, const char *src)
{
	char *dst = *pp_dst;
	const char *src_tail;

	while (1) {
		unsigned char six_bit[4];
		int count = 0;

		/* Fetch up to four 6-bit values */
		src_tail = src;
		while (count < 4) {
			char *table_ptr;
			int ch;

			/* Get next _valid_ character.
			 * bb_uuenc_tbl_base64[] contains this string:
			 *  0         1         2         3         4         5         6
			 *  01234567890123456789012345678901234567890123456789012345678901234
			 * "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/="
			 */
			do {
				ch = *src;
				if (ch == '\0') {
					if (count == 0) {
						/* Example:
						 * If we decode "QUJD <NUL>", we want
						 * to return ptr to NUL, not to ' ',
						 * because we did fully decode
						 * the string (to "ABC").
						 */
						src_tail = src;
					}
					goto ret;
				}
				src++;
				table_ptr = strchr(bb_uuenc_tbl_base64, ch);
//TODO: add BASE64_FLAG_foo to die on bad char?
			} while (!table_ptr);

			/* Convert encoded character to decimal */
			ch = table_ptr - bb_uuenc_tbl_base64;

			/* ch is 64 if char was '=', otherwise 0..63 */
			if (ch == 64)
				break;
			six_bit[count] = ch;
			count++;
		}

		/* Transform 6-bit values to 8-bit ones.
		 * count can be < 4 when we decode the tail:
		 * "eQ==" -> "y", not "y NUL NUL".
		 * Note that (count > 1) is always true,
		 * "x===" encoding is not valid:
		 * even a single zero byte encodes as "AA==".
		 * However, with current logic we come here with count == 1
		 * when we decode "==" tail.
		 */
		if (count > 1)
			*dst++ = six_bit[0] << 2 | six_bit[1] >> 4;
		if (count > 2)
			*dst++ = six_bit[1] << 4 | six_bit[2] >> 2;
		if (count > 3)
			*dst++ = six_bit[2] << 6 | six_bit[3];
		/* Note that if we decode "AA==" and ate first '=',
		 * we just decoded one char (count == 2) and now we'll
		 * do the loop once more to decode second '='.
		 */
	} /* while (1) */
 ret:
	*pp_dst = dst;
	return src_tail;
}

#if ENABLE_BASE32
const char* FAST_FUNC decode_base32(char **pp_dst, const char *src)
{
	char *dst = *pp_dst;
	const char *src_tail;

	while (1) {
		unsigned char five_bit[8];
		int count = 0;

		/* Fetch up to eight 5-bit values */
		src_tail = src;
		while (count < 8) {
			char *table_ptr;
			int ch;

			/* Get next _valid_ character.
			 * bb_uuenc_tbl_base32[] contains this string:
			 *  0         1         2         3
			 *  01234567890123456789012345678901
			 * "ABCDEFGHIJKLMNOPQRSTUVWXYZ234567="
			 */
			do {
				ch = *src;
				if (ch == '\0') {
					if (count == 0) {
						src_tail = src;
					}
					goto ret;
				}
				src++;
				table_ptr = strchr(bb_uuenc_tbl_base32, toupper(ch));
			} while (!table_ptr);

			/* Convert encoded character to decimal */
			ch = table_ptr - bb_uuenc_tbl_base32;

			/* ch is 32 if char was '=', otherwise 0..31 */
			if (ch == 32)
				break;
			five_bit[count] = ch;
			count++;
		}

		/* Transform 5-bit values to 8-bit ones */
		if (count > 1)                                        // xxxxx xxx-- ----- ----- ----- ----- ----- -----
			*dst++ = five_bit[0] << 3 | five_bit[1] >> 2;
		if (count > 3)                                        // ----- ---xx xxxxx x---- ----- ----- ----- -----
			*dst++ = five_bit[1] << 6 | five_bit[2] << 1 | five_bit[3] >> 4;
		if (count > 4)                                        // ----- ----- ----- -xxxx xxxx- ----- ----- -----
			*dst++ = five_bit[3] << 4 | five_bit[4] >> 1;
		if (count > 6)                                        // ----- ----- ----- ----- ----x xxxxx xx--- -----
			*dst++ = five_bit[4] << 7 | five_bit[5] << 2 | five_bit[6] >> 3;
		if (count > 7)                                        // ----- ----- ----- ----- ----- ----- --xxx xxxxx
			*dst++ = five_bit[6] << 5 | five_bit[7];
	} /* while (1) */
 ret:
	*pp_dst = dst;
	return src_tail;
}
#endif

/*
 * Decode base64 encoded stream.
 * Can stop on EOF, specified char, or on uuencode-style "====" line:
 * flags argument controls it.
 */
void FAST_FUNC read_base64(FILE *src_stream, FILE *dst_stream, int flags)
{
/* Note that EOF _can_ be passed as exit_char too */
#define exit_char    ((int)(signed char)flags)
#define uu_style_end (flags & BASE64_FLAG_UU_STOP)
#define base32       (flags & BASE64_32)

	/* uuencoded files have 61 byte lines. Use 64 byte buffer
	 * to process line at a time.
	 */
	enum { BUFFER_SIZE = 64 };

	char in_buf[BUFFER_SIZE + 2];
	char out_buf[BUFFER_SIZE / 4 * 3 + 2];
	char *out_tail;
	const char *in_tail;
	int term_seen = 0;
	int in_count = 0;

	while (1) {
		while (in_count < BUFFER_SIZE) {
			int ch = fgetc(src_stream);
			if (ch == exit_char) {
				if (in_count == 0)
					return;
				term_seen = 1;
				break;
			}
			if (ch == EOF) {
				term_seen = 1;
				break;
			}
			/* Prevent "====" line to be split: stop if we see '\n'.
			 * We can also skip other whitespace and skirt the problem
			 * of files with NULs by stopping on any control char or space:
			 */
			if (ch <= ' ')
				break;
			in_buf[in_count++] = ch;
		}
		in_buf[in_count] = '\0';

		/* Did we encounter "====" line? */
		if (uu_style_end && strcmp(in_buf, "====") == 0)
			return;

		out_tail = out_buf;
#if ENABLE_BASE32
		if (base32)
			in_tail = decode_base32(&out_tail, in_buf);
		else
#endif
			in_tail = decode_base64(&out_tail, in_buf);

		fwrite(out_buf, (out_tail - out_buf), 1, dst_stream);

		if (term_seen) {
			/* Did we consume ALL characters? */
			if (*in_tail == '\0')
				return;
			/* No */
			bb_simple_error_msg_and_die("truncated base64 input");
		}

		/* It was partial decode */
		in_count = strlen(in_tail);
		memmove(in_buf, in_tail, in_count);
	}
}