summaryrefslogtreecommitdiff
path: root/utils/utf8.c
blob: 69414400e28b039de411465f79accb1a3df4cb06 (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
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
/*
 * This file is part of NetSurf, http://netsurf-browser.org/
 * Licensed under the GNU General Public License,
 *                http://www.opensource.org/licenses/gpl-license
 * Copyright 2005 John M Bell <jmb202@ecs.soton.ac.uk>
 */

/** \file
 * UTF-8 manipulation functions (implementation).
 */

#define _GNU_SOURCE  /* for strndup */
#include <assert.h>
#include <errno.h>
#include <stdlib.h>
#include <string.h>

#include <iconv.h>

#include "utils/log.h"
#include "utils/utf8.h"

static utf8_convert_ret utf8_convert(const char *string, size_t len,
		const char *from, const char *to, char **result);

/**
 * Convert a UTF-8 multibyte sequence into a single UCS4 character
 *
 * Encoding of UCS values outside the UTF-16 plane has been removed from
 * RFC3629. This function conforms to RFC2279, however.
 *
 * \param s  The sequence to process
 * \param l  Length of sequence
 * \return   UCS4 character
 */
size_t utf8_to_ucs4(const char *s, size_t l)
{
	size_t c = 0;

	if (!s)
		assert(0);
	else if (l > 0 && *s < 0x80)
		c = *s;
	else if (l > 1 && (*s & 0xE0) == 0xC0 && (*(s+1) & 0xC0) == 0x80)
		c = ((*s & 0x1F) << 6) | (*(s+1) & 0x3F);
	else if (l > 2 && (*s & 0xF0) == 0xE0 && (*(s+1) & 0xC0) == 0x80 &&
			(*(s+2) & 0xC0) == 0x80)
		c = ((*s & 0x0F) << 12) | ((*(s+1) & 0x3F) << 6) |
							(*(s+2) & 0x3F);
	else if (l > 3 && (*s & 0xF8) == 0xF0 && (*(s+1) & 0xC0) == 0x80 &&
			(*(s+2) & 0xC0) == 0x80 && (*(s+3) & 0xC0) == 0x80)
		c = ((*s & 0x0F) << 18) | ((*(s+1) & 0x3F) << 12) |
				((*(s+2) & 0x3F) << 6) | (*(s+3) & 0x3F);
	else if (l > 4 && (*s & 0xFC) == 0xF8 && (*(s+1) & 0xC0) == 0x80 &&
			(*(s+2) & 0xC0) == 0x80 && (*(s+3) & 0xC0) == 0x80 &&
			(*(s+4) & 0xC0) == 0x80)
		c = ((*s & 0x0F) << 24) | ((*(s+1) & 0x3F) << 18) |
			((*(s+2) & 0x3F) << 12) | ((*(s+3) & 0x3F) << 6) |
			(*(s+4) & 0x3F);
	else if (l > 5 && (*s & 0xFE) == 0xFC && (*(s+1) & 0xC0) == 0x80 &&
			(*(s+2) & 0xC0) == 0x80 && (*(s+3) & 0xC0) == 0x80 &&
			(*(s+4) & 0xC0) == 0x80 && (*(s+5) & 0xC0) == 0x80)
		c = ((*s & 0x0F) << 28) | ((*(s+1) & 0x3F) << 24) |
			((*(s+2) & 0x3F) << 18) | ((*(s+3) & 0x3F) << 12) |
			((*(s+4) & 0x3F) << 6) | (*(s+5) & 0x3F);
	else
		assert(0);

	return c;
}

/**
 * Convert a single UCS4 character into a UTF-8 multibyte sequence
 *
 * Encoding of UCS values outside the UTF-16 plane has been removed from
 * RFC3629. This function conforms to RFC2279, however.
 *
 * \param c  The character to process (0 <= c <= 0x7FFFFFFF)
 * \param s  Pointer to 6 byte long output buffer
 * \return   Length of multibyte sequence
 */
size_t utf8_from_ucs4(size_t c, char *s)
{
	size_t l = 0;

	if (c > 0x7FFFFFFF || s == NULL)
		assert(0);
	else if (c < 0x80) {
		*s = (char)c;
		l = 1;
	}
	else if (c < 0x800) {
		*s = 0xC0 | ((c >> 6) & 0x1F);
		*(s+1) = 0x80 | (c & 0x3F);
		l = 2;
	}
	else if (c < 0x10000) {
		*s = 0xE0 | ((c >> 12) & 0xF);
		*(s+1) = 0x80 | ((c >> 6) & 0x3F);
		*(s+2) = 0x80 | (c & 0x3F);
		l = 3;
	}
	else if (c < 0x200000) {
		*s = 0xF0 | ((c >> 18) & 0x7);
		*(s+1) = 0x80 | ((c >> 12) & 0x3F);
		*(s+2) = 0x80 | ((c >> 6) & 0x3F);
		*(s+3) = 0x80 | (c & 0x3F);
		l = 4;
	}
	else if (c < 0x4000000) {
		*s = 0xF8 | ((c >> 24) & 0x3);
		*(s+1) = 0x80 | ((c >> 18) & 0x3F);
		*(s+2) = 0x80 | ((c >> 12) & 0x3F);
		*(s+3) = 0x80 | ((c >> 6) & 0x3F);
		*(s+4) = 0x80 | (c & 0x3F);
		l = 5;
	}
	else if (c <= 0x7FFFFFFF) {
		*s = 0xFC | ((c >> 30) & 0x1);
		*(s+1) = 0x80 | ((c >> 24) & 0x3F);
		*(s+2) = 0x80 | ((c >> 18) & 0x3F);
		*(s+3) = 0x80 | ((c >> 12) & 0x3F);
		*(s+4) = 0x80 | ((c >> 6) & 0x3F);
		*(s+5) = 0x80 | (c & 0x3F);
		l = 6;
	}

	return l;
}

/**
 * Calculate the length (in characters) of a NULL-terminated UTF-8 string
 *
 * \param s  The string
 * \return   Length of string
 */
size_t utf8_length(const char *s)
{
	const char *__s = s;
	int l = 0;

	assert(__s != NULL);

	while (*__s != '\0') {
		if ((*__s & 0x80) == 0x00)
			__s += 1;
		else if ((*__s & 0xE0) == 0xC0)
			__s += 2;
		else if ((*__s & 0xF0) == 0xE0)
			__s += 3;
		else if ((*__s & 0xF8) == 0xF0)
			__s += 4;
		else if ((*__s & 0xFC) == 0xF8)
			__s += 5;
		else if ((*__s & 0xFE) == 0xFC)
			__s += 6;
		else
			assert(0);
		l++;
	}

	return l;
}

/**
 * Find previous legal UTF-8 char in string
 *
 * \param s  The string
 * \param o  Offset in the string to start at
 * \return Offset of first byte of previous legal character
 */
size_t utf8_prev(const char *s, size_t o)
{
	assert(s != NULL);

	while (o != 0 && (s[--o] & 0xC0) == 0x80)
		/* do nothing */;

	return o;
}

/**
 * Find next legal UTF-8 char in string
 *
 * \param s  The string
 * \param l  Maximum offset in string
 * \param o  Offset in the string to start at
 * \return Offset of first byte of next legal character
 */
size_t utf8_next(const char *s, size_t l, size_t o)
{
	assert(s != NULL);

	while (o != l && (s[++o] & 0xC0) == 0x80)
		/* do nothing */;

	return o;
}

/* Cache of previous iconv conversion descriptor used by utf8_convert */
static struct {
	char from[32];	/**< Encoding name to convert from */
	char to[32];	/**< Encoding name to convert to */
	iconv_t cd;	/**< Iconv conversion descriptor */
} last_cd;

/**
 * Finalise the UTF-8 library
 */
void utf8_finalise(void)
{
	if (last_cd.cd != 0)
		iconv_close(last_cd.cd);

	/* paranoia follows */
	last_cd.from[0] = '\0';
	last_cd.to[0] = '\0';
	last_cd.cd = 0;
}

/**
 * Convert a UTF8 string into the named encoding
 *
 * \param string  The NULL-terminated string to convert
 * \param encname The encoding name (suitable for passing to iconv)
 * \param len     Length of input string to consider (in bytes), or 0
 * \param result  Pointer to location to store result (allocated on heap)
 * \return Appropriate utf8_convert_ret value
 */
utf8_convert_ret utf8_to_enc(const char *string, const char *encname,
		size_t len, char **result)
{
	return utf8_convert(string, len, "UTF-8", encname, result);
}

/**
 * Convert a string in the named encoding into a UTF-8 string
 *
 * \param string  The NULL-terminated string to convert
 * \param encname The encoding name (suitable for passing to iconv)
 * \param len     Length of input string to consider (in bytes), or 0
 * \param result  Pointer to location to store result (allocated on heap)
 * \return Appropriate utf8_convert_ret value
 */
utf8_convert_ret utf8_from_enc(const char *string, const char *encname,
		size_t len, char **result)
{
	return utf8_convert(string, len, encname, "UTF-8", result);
}

/**
 * Convert a string from one encoding to another
 *
 * \param string  The NULL-terminated string to convert
 * \param len     Length of input string to consider (in bytes)
 * \param from    The encoding name to convert from
 * \param to      The encoding name to convert to
 * \param result  Pointer to location in which to store result
 * \return Appropriate utf8_convert_ret value
 */
utf8_convert_ret utf8_convert(const char *string, size_t len,
		const char *from, const char *to, char **result)
{
	iconv_t cd;
	char *temp, *out, *in;
	size_t slen, rlen;

	assert(string && from && to && result);

	if (strcasecmp(from, to) == 0) {
		/* conversion from an encoding to itself == strdup */
		slen = len ? len : strlen(string);
		*(result) = strndup(string, slen);
		if (!(*result)) {
			*(result) = NULL;
			return UTF8_CONVERT_NOMEM;
		}

		return UTF8_CONVERT_OK;
	}

	in = (char *)string;

	/* we cache the last used conversion descriptor,
	 * so check if we're trying to use it here */
	if (strncasecmp(last_cd.from, from, 32) == 0 &&
			strncasecmp(last_cd.to, to, 32) == 0) {
		cd = last_cd.cd;
	}
	else {
		/* no match, so create a new cd */
		cd = iconv_open(to, from);
		if (cd == (iconv_t)-1) {
			if (errno == EINVAL)
				return UTF8_CONVERT_BADENC;
			/* default to no memory */
			return UTF8_CONVERT_NOMEM;
		}

		/* close the last cd - we don't care if this fails */
		if (last_cd.cd)
			iconv_close(last_cd.cd);

		/* and copy the to/from/cd data into last_cd */
		strncpy(last_cd.from, from, 32);
		strncpy(last_cd.to, to, 32);
		last_cd.cd = cd;
	}

	slen = len ? len : strlen(string);
	/* Worst case = ACSII -> UCS4, so allocate an output buffer
	 * 4 times larger than the input buffer, and add 4 bytes at
	 * the end for the NULL terminator
	 */
	rlen = slen * 4 + 4;

	temp = out = malloc(rlen);
	if (!out)
		return UTF8_CONVERT_NOMEM;

	/* perform conversion */
	if (iconv(cd, &in, &slen, &out, &rlen) == (size_t)-1) {
		free(temp);
		/* clear the cached conversion descriptor as it's invalid */
		last_cd.from[0] = '\0';
		last_cd.to[0] = '\0';
		last_cd.cd = 0;
		/** \todo handle the various cases properly
		 * There are 3 possible error cases:
		 * a) Insufficiently large output buffer
		 * b) Invalid input byte sequence
		 * c) Incomplete input sequence */
		return UTF8_CONVERT_NOMEM;
	}

	*(result) = realloc(temp, out - temp + 4);
	if (!(*result)) {
		free(temp);
		*(result) = NULL; /* for sanity's sake */
		return UTF8_CONVERT_NOMEM;
	}

	/* NULL terminate - needs 4 characters as we may have
	 * converted to UTF-32 */
	memset((*result) + (out - temp), 0, 4);

	return UTF8_CONVERT_OK;
}