1 | /*
|
---|
2 | * UTF-8 support routines
|
---|
3 | *
|
---|
4 | * Copyright 2000 Alexandre Julliard
|
---|
5 | *
|
---|
6 | * This library is free software; you can redistribute it and/or
|
---|
7 | * modify it under the terms of the GNU Lesser General Public
|
---|
8 | * License as published by the Free Software Foundation; either
|
---|
9 | * version 2.1 of the License, or (at your option) any later version.
|
---|
10 | *
|
---|
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 | * Lesser General Public License for more details.
|
---|
15 | *
|
---|
16 | * You should have received a copy of the GNU Lesser General Public
|
---|
17 | * License along with this library; if not, write to the Free Software
|
---|
18 | * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
|
---|
19 | */
|
---|
20 |
|
---|
21 | /*
|
---|
22 | * Sun LGPL Disclaimer: For the avoidance of doubt, except that if any license choice
|
---|
23 | * other than GPL or LGPL is available it will apply instead, Sun elects to use only
|
---|
24 | * the Lesser General Public License version 2.1 (LGPLv2) at this time for any software where
|
---|
25 | * a choice of LGPL license versions is made available with the language indicating
|
---|
26 | * that LGPLv2 or any later version may be used, or where a choice of which version
|
---|
27 | * of the LGPL is applied is otherwise unspecified.
|
---|
28 | */
|
---|
29 |
|
---|
30 | #include "config.h"
|
---|
31 | #include "wine/port.h"
|
---|
32 |
|
---|
33 | #include <string.h>
|
---|
34 |
|
---|
35 | #include "wine/unicode.h"
|
---|
36 |
|
---|
37 | extern WCHAR compose( const WCHAR *str );
|
---|
38 |
|
---|
39 | /* number of following bytes in sequence based on first byte value (for bytes above 0x7f) */
|
---|
40 | static const char utf8_length[128] =
|
---|
41 | {
|
---|
42 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x80-0x8f */
|
---|
43 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x90-0x9f */
|
---|
44 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xa0-0xaf */
|
---|
45 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xb0-0xbf */
|
---|
46 | 0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xc0-0xcf */
|
---|
47 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xd0-0xdf */
|
---|
48 | 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* 0xe0-0xef */
|
---|
49 | 3,3,3,3,3,0,0,0,0,0,0,0,0,0,0,0 /* 0xf0-0xff */
|
---|
50 | };
|
---|
51 |
|
---|
52 | /* first byte mask depending on UTF-8 sequence length */
|
---|
53 | static const unsigned char utf8_mask[4] = { 0x7f, 0x1f, 0x0f, 0x07 };
|
---|
54 |
|
---|
55 | /* minimum Unicode value depending on UTF-8 sequence length */
|
---|
56 | static const unsigned int utf8_minval[4] = { 0x0, 0x80, 0x800, 0x10000 };
|
---|
57 |
|
---|
58 |
|
---|
59 | /* get the next char value taking surrogates into account */
|
---|
60 | static inline unsigned int get_surrogate_value( const WCHAR *src, unsigned int srclen )
|
---|
61 | {
|
---|
62 | if (src[0] >= 0xd800 && src[0] <= 0xdfff) /* surrogate pair */
|
---|
63 | {
|
---|
64 | if (src[0] > 0xdbff || /* invalid high surrogate */
|
---|
65 | srclen <= 1 || /* missing low surrogate */
|
---|
66 | src[1] < 0xdc00 || src[1] > 0xdfff) /* invalid low surrogate */
|
---|
67 | return 0;
|
---|
68 | return 0x10000 + ((src[0] & 0x3ff) << 10) + (src[1] & 0x3ff);
|
---|
69 | }
|
---|
70 | return src[0];
|
---|
71 | }
|
---|
72 |
|
---|
73 | /* query necessary dst length for src string */
|
---|
74 | static inline int get_length_wcs_utf8( int flags, const WCHAR *src, unsigned int srclen )
|
---|
75 | {
|
---|
76 | int len;
|
---|
77 | unsigned int val;
|
---|
78 |
|
---|
79 | for (len = 0; srclen; srclen--, src++)
|
---|
80 | {
|
---|
81 | if (*src < 0x80) /* 0x00-0x7f: 1 byte */
|
---|
82 | {
|
---|
83 | len++;
|
---|
84 | continue;
|
---|
85 | }
|
---|
86 | if (*src < 0x800) /* 0x80-0x7ff: 2 bytes */
|
---|
87 | {
|
---|
88 | len += 2;
|
---|
89 | continue;
|
---|
90 | }
|
---|
91 | if (!(val = get_surrogate_value( src, srclen )))
|
---|
92 | {
|
---|
93 | if (flags & WC_ERR_INVALID_CHARS) return -2;
|
---|
94 | continue;
|
---|
95 | }
|
---|
96 | if (val < 0x10000) /* 0x800-0xffff: 3 bytes */
|
---|
97 | len += 3;
|
---|
98 | else /* 0x10000-0x10ffff: 4 bytes */
|
---|
99 | {
|
---|
100 | len += 4;
|
---|
101 | src++;
|
---|
102 | srclen--;
|
---|
103 | }
|
---|
104 | }
|
---|
105 | return len;
|
---|
106 | }
|
---|
107 |
|
---|
108 | /* wide char to UTF-8 string conversion */
|
---|
109 | /* return -1 on dst buffer overflow, -2 on invalid input char */
|
---|
110 | int wine_utf8_wcstombs( int flags, const WCHAR *src, int srclen, char *dst, int dstlen )
|
---|
111 | {
|
---|
112 | int len;
|
---|
113 |
|
---|
114 | if (!dstlen) return get_length_wcs_utf8( flags, src, srclen );
|
---|
115 |
|
---|
116 | for (len = dstlen; srclen; srclen--, src++)
|
---|
117 | {
|
---|
118 | WCHAR ch = *src;
|
---|
119 | unsigned int val;
|
---|
120 |
|
---|
121 | if (ch < 0x80) /* 0x00-0x7f: 1 byte */
|
---|
122 | {
|
---|
123 | if (!len--) return -1; /* overflow */
|
---|
124 | *dst++ = ch;
|
---|
125 | continue;
|
---|
126 | }
|
---|
127 |
|
---|
128 | if (ch < 0x800) /* 0x80-0x7ff: 2 bytes */
|
---|
129 | {
|
---|
130 | if ((len -= 2) < 0) return -1; /* overflow */
|
---|
131 | dst[1] = 0x80 | (ch & 0x3f);
|
---|
132 | ch >>= 6;
|
---|
133 | dst[0] = 0xc0 | ch;
|
---|
134 | dst += 2;
|
---|
135 | continue;
|
---|
136 | }
|
---|
137 |
|
---|
138 | if (!(val = get_surrogate_value( src, srclen )))
|
---|
139 | {
|
---|
140 | if (flags & WC_ERR_INVALID_CHARS) return -2;
|
---|
141 | continue;
|
---|
142 | }
|
---|
143 |
|
---|
144 | if (val < 0x10000) /* 0x800-0xffff: 3 bytes */
|
---|
145 | {
|
---|
146 | if ((len -= 3) < 0) return -1; /* overflow */
|
---|
147 | dst[2] = 0x80 | (val & 0x3f);
|
---|
148 | val >>= 6;
|
---|
149 | dst[1] = 0x80 | (val & 0x3f);
|
---|
150 | val >>= 6;
|
---|
151 | dst[0] = 0xe0 | val;
|
---|
152 | dst += 3;
|
---|
153 | }
|
---|
154 | else /* 0x10000-0x10ffff: 4 bytes */
|
---|
155 | {
|
---|
156 | if ((len -= 4) < 0) return -1; /* overflow */
|
---|
157 | dst[3] = 0x80 | (val & 0x3f);
|
---|
158 | val >>= 6;
|
---|
159 | dst[2] = 0x80 | (val & 0x3f);
|
---|
160 | val >>= 6;
|
---|
161 | dst[1] = 0x80 | (val & 0x3f);
|
---|
162 | val >>= 6;
|
---|
163 | dst[0] = 0xf0 | val;
|
---|
164 | dst += 4;
|
---|
165 | src++;
|
---|
166 | srclen--;
|
---|
167 | }
|
---|
168 | }
|
---|
169 | return dstlen - len;
|
---|
170 | }
|
---|
171 |
|
---|
172 | /* helper for the various utf8 mbstowcs functions */
|
---|
173 | static inline unsigned int decode_utf8_char( unsigned char ch, const char **str, const char *strend )
|
---|
174 | {
|
---|
175 | unsigned int len = utf8_length[ch-0x80];
|
---|
176 | unsigned int res = ch & utf8_mask[len];
|
---|
177 | const char *end = *str + len;
|
---|
178 |
|
---|
179 | if (end > strend) return ~0;
|
---|
180 | switch(len)
|
---|
181 | {
|
---|
182 | case 3:
|
---|
183 | if ((ch = end[-3] ^ 0x80) >= 0x40) break;
|
---|
184 | res = (res << 6) | ch;
|
---|
185 | (*str)++;
|
---|
186 | case 2:
|
---|
187 | if ((ch = end[-2] ^ 0x80) >= 0x40) break;
|
---|
188 | res = (res << 6) | ch;
|
---|
189 | (*str)++;
|
---|
190 | case 1:
|
---|
191 | if ((ch = end[-1] ^ 0x80) >= 0x40) break;
|
---|
192 | res = (res << 6) | ch;
|
---|
193 | (*str)++;
|
---|
194 | if (res < utf8_minval[len]) break;
|
---|
195 | return res;
|
---|
196 | }
|
---|
197 | return ~0;
|
---|
198 | }
|
---|
199 |
|
---|
200 | /* query necessary dst length for src string with composition */
|
---|
201 | static inline int get_length_mbs_utf8_compose( int flags, const char *src, int srclen )
|
---|
202 | {
|
---|
203 | int ret = 0;
|
---|
204 | unsigned int res;
|
---|
205 | WCHAR composed[2];
|
---|
206 | const char *srcend = src + srclen;
|
---|
207 |
|
---|
208 | composed[0] = 0;
|
---|
209 | while (src < srcend)
|
---|
210 | {
|
---|
211 | unsigned char ch = *src++;
|
---|
212 | if (ch < 0x80) /* special fast case for 7-bit ASCII */
|
---|
213 | {
|
---|
214 | composed[0] = ch;
|
---|
215 | ret++;
|
---|
216 | continue;
|
---|
217 | }
|
---|
218 | if ((res = decode_utf8_char( ch, &src, srcend )) <= 0xffff)
|
---|
219 | {
|
---|
220 | if (composed[0])
|
---|
221 | {
|
---|
222 | composed[1] = res;
|
---|
223 | if ((composed[0] = compose( composed ))) continue;
|
---|
224 | }
|
---|
225 | composed[0] = res;
|
---|
226 | ret++;
|
---|
227 | }
|
---|
228 | else if (res <= 0x10ffff)
|
---|
229 | {
|
---|
230 | ret += 2;
|
---|
231 | composed[0] = 0; /* no composition for surrogates */
|
---|
232 | }
|
---|
233 | else if (flags & MB_ERR_INVALID_CHARS) return -2; /* bad char */
|
---|
234 | /* otherwise ignore it */
|
---|
235 | }
|
---|
236 | return ret;
|
---|
237 | }
|
---|
238 |
|
---|
239 | /* UTF-8 to wide char string conversion with composition */
|
---|
240 | /* return -1 on dst buffer overflow, -2 on invalid input char */
|
---|
241 | static int utf8_mbstowcs_compose( int flags, const char *src, int srclen, WCHAR *dst, int dstlen )
|
---|
242 | {
|
---|
243 | unsigned int res;
|
---|
244 | const char *srcend = src + srclen;
|
---|
245 | WCHAR composed[2];
|
---|
246 | WCHAR *dstend = dst + dstlen;
|
---|
247 |
|
---|
248 | if (!dstlen) return get_length_mbs_utf8_compose( flags, src, srclen );
|
---|
249 |
|
---|
250 | composed[0] = 0;
|
---|
251 | while (src < srcend)
|
---|
252 | {
|
---|
253 | unsigned char ch = *src++;
|
---|
254 | if (ch < 0x80) /* special fast case for 7-bit ASCII */
|
---|
255 | {
|
---|
256 | if (dst >= dstend) return -1; /* overflow */
|
---|
257 | *dst++ = composed[0] = ch;
|
---|
258 | continue;
|
---|
259 | }
|
---|
260 | if ((res = decode_utf8_char( ch, &src, srcend )) <= 0xffff)
|
---|
261 | {
|
---|
262 | if (composed[0])
|
---|
263 | {
|
---|
264 | composed[1] = res;
|
---|
265 | if ((composed[0] = compose( composed )))
|
---|
266 | {
|
---|
267 | dst[-1] = composed[0];
|
---|
268 | continue;
|
---|
269 | }
|
---|
270 | }
|
---|
271 | if (dst >= dstend) return -1; /* overflow */
|
---|
272 | *dst++ = composed[0] = res;
|
---|
273 | }
|
---|
274 | else if (res <= 0x10ffff) /* we need surrogates */
|
---|
275 | {
|
---|
276 | if (dst >= dstend - 1) return -1; /* overflow */
|
---|
277 | res -= 0x10000;
|
---|
278 | *dst++ = 0xd800 | (res >> 10);
|
---|
279 | *dst++ = 0xdc00 | (res & 0x3ff);
|
---|
280 | composed[0] = 0; /* no composition for surrogates */
|
---|
281 | }
|
---|
282 | else if (flags & MB_ERR_INVALID_CHARS) return -2; /* bad char */
|
---|
283 | /* otherwise ignore it */
|
---|
284 | }
|
---|
285 | return dstlen - (dstend - dst);
|
---|
286 | }
|
---|
287 |
|
---|
288 | /* query necessary dst length for src string */
|
---|
289 | static inline int get_length_mbs_utf8( int flags, const char *src, int srclen )
|
---|
290 | {
|
---|
291 | int ret = 0;
|
---|
292 | unsigned int res;
|
---|
293 | const char *srcend = src + srclen;
|
---|
294 |
|
---|
295 | while (src < srcend)
|
---|
296 | {
|
---|
297 | unsigned char ch = *src++;
|
---|
298 | if (ch < 0x80) /* special fast case for 7-bit ASCII */
|
---|
299 | {
|
---|
300 | ret++;
|
---|
301 | continue;
|
---|
302 | }
|
---|
303 | if ((res = decode_utf8_char( ch, &src, srcend )) <= 0x10ffff)
|
---|
304 | {
|
---|
305 | if (res > 0xffff) ret++;
|
---|
306 | ret++;
|
---|
307 | }
|
---|
308 | else if (flags & MB_ERR_INVALID_CHARS) return -2; /* bad char */
|
---|
309 | /* otherwise ignore it */
|
---|
310 | }
|
---|
311 | return ret;
|
---|
312 | }
|
---|
313 |
|
---|
314 | /* UTF-8 to wide char string conversion */
|
---|
315 | /* return -1 on dst buffer overflow, -2 on invalid input char */
|
---|
316 | int wine_utf8_mbstowcs( int flags, const char *src, int srclen, WCHAR *dst, int dstlen )
|
---|
317 | {
|
---|
318 | unsigned int res;
|
---|
319 | const char *srcend = src + srclen;
|
---|
320 | WCHAR *dstend = dst + dstlen;
|
---|
321 |
|
---|
322 | if (flags & MB_COMPOSITE) return utf8_mbstowcs_compose( flags, src, srclen, dst, dstlen );
|
---|
323 |
|
---|
324 | if (!dstlen) return get_length_mbs_utf8( flags, src, srclen );
|
---|
325 |
|
---|
326 | while ((dst < dstend) && (src < srcend))
|
---|
327 | {
|
---|
328 | unsigned char ch = *src++;
|
---|
329 | if (ch < 0x80) /* special fast case for 7-bit ASCII */
|
---|
330 | {
|
---|
331 | *dst++ = ch;
|
---|
332 | continue;
|
---|
333 | }
|
---|
334 | if ((res = decode_utf8_char( ch, &src, srcend )) <= 0xffff)
|
---|
335 | {
|
---|
336 | *dst++ = res;
|
---|
337 | }
|
---|
338 | else if (res <= 0x10ffff) /* we need surrogates */
|
---|
339 | {
|
---|
340 | if (dst == dstend - 1) return -1; /* overflow */
|
---|
341 | res -= 0x10000;
|
---|
342 | *dst++ = 0xd800 | (res >> 10);
|
---|
343 | *dst++ = 0xdc00 | (res & 0x3ff);
|
---|
344 | }
|
---|
345 | else if (flags & MB_ERR_INVALID_CHARS) return -2; /* bad char */
|
---|
346 | /* otherwise ignore it */
|
---|
347 | }
|
---|
348 | if (src < srcend) return -1; /* overflow */
|
---|
349 | return dstlen - (dstend - dst);
|
---|
350 | }
|
---|