diff options
Diffstat (limited to '')
-rw-r--r-- | libraries/eina/src/lib/eina_unicode.c | 405 |
1 files changed, 405 insertions, 0 deletions
diff --git a/libraries/eina/src/lib/eina_unicode.c b/libraries/eina/src/lib/eina_unicode.c new file mode 100644 index 0000000..342e3cb --- /dev/null +++ b/libraries/eina/src/lib/eina_unicode.c | |||
@@ -0,0 +1,405 @@ | |||
1 | /* EINA - EFL data type library | ||
2 | * Copyright (C) 2010 Tom Hacohen, | ||
3 | * Brett Nash | ||
4 | * | ||
5 | * This library is free software; you can redistribute it and/or | ||
6 | * modify it under the terms of the GNU Lesser General Public | ||
7 | * License as published by the Free Software Foundation; either | ||
8 | * version 2.1 of the License, or (at your option) any later version. | ||
9 | * | ||
10 | * This library is distributed in the hope that it will be useful, | ||
11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | ||
13 | * Lesser General Public License for more details. | ||
14 | * | ||
15 | * You should have received a copy of the GNU Lesser General Public | ||
16 | * License along with this library; | ||
17 | * if not, see <http://www.gnu.org/licenses/>. | ||
18 | |||
19 | */ | ||
20 | |||
21 | #include <Eina.h> | ||
22 | #include "eina_unicode.h" | ||
23 | |||
24 | /* FIXME: check if sizeof(wchar_t) == sizeof(Eina_Unicode) if so, | ||
25 | * probably better to use the standard functions */ | ||
26 | |||
27 | /* Maybe I'm too tired, but this is the only thing that actually worked. */ | ||
28 | const Eina_Unicode _EINA_UNICODE_EMPTY_STRING[1] = {0}; | ||
29 | EAPI const Eina_Unicode *EINA_UNICODE_EMPTY_STRING = _EINA_UNICODE_EMPTY_STRING; | ||
30 | EAPI int | ||
31 | eina_unicode_strcmp(const Eina_Unicode *a, const Eina_Unicode *b) | ||
32 | { | ||
33 | for (; *a && *a == *b; a++, b++) | ||
34 | ; | ||
35 | if (*a == *b) | ||
36 | return 0; | ||
37 | else if (*a < *b) | ||
38 | return -1; | ||
39 | else | ||
40 | return 1; | ||
41 | } | ||
42 | |||
43 | EAPI Eina_Unicode * | ||
44 | eina_unicode_strcpy(Eina_Unicode *dest, const Eina_Unicode *source) | ||
45 | { | ||
46 | Eina_Unicode *ret = dest; | ||
47 | |||
48 | while (*source) | ||
49 | *dest++ = *source++; | ||
50 | *dest = 0; | ||
51 | return ret; | ||
52 | } | ||
53 | |||
54 | EAPI Eina_Unicode * | ||
55 | eina_unicode_strncpy(Eina_Unicode *dest, const Eina_Unicode *source, size_t n) | ||
56 | { | ||
57 | Eina_Unicode *ret = dest; | ||
58 | |||
59 | for ( ; n && *source ; n--) | ||
60 | *dest++ = *source++; | ||
61 | for (; n; n--) | ||
62 | *dest++ = 0; | ||
63 | return ret; | ||
64 | } | ||
65 | |||
66 | EAPI size_t | ||
67 | eina_unicode_strlen(const Eina_Unicode *ustr) | ||
68 | { | ||
69 | const Eina_Unicode *end; | ||
70 | for (end = ustr; *end; end++) | ||
71 | ; | ||
72 | return end - ustr; | ||
73 | } | ||
74 | |||
75 | EAPI size_t | ||
76 | eina_unicode_strnlen(const Eina_Unicode *ustr, int n) | ||
77 | { | ||
78 | const Eina_Unicode *end; | ||
79 | const Eina_Unicode *last = ustr + n; /* technically not portable ;-) */ | ||
80 | for (end = ustr; end < last && *end; end++) | ||
81 | ; | ||
82 | return end - ustr; | ||
83 | } | ||
84 | |||
85 | |||
86 | |||
87 | |||
88 | EAPI Eina_Unicode * | ||
89 | eina_unicode_strndup(const Eina_Unicode *text, size_t n) | ||
90 | { | ||
91 | Eina_Unicode *ustr; | ||
92 | |||
93 | ustr = (Eina_Unicode *) malloc((n + 1) * sizeof(Eina_Unicode)); | ||
94 | memcpy(ustr, text, n * sizeof(Eina_Unicode)); | ||
95 | ustr[n] = 0; | ||
96 | return ustr; | ||
97 | } | ||
98 | |||
99 | EAPI Eina_Unicode * | ||
100 | eina_unicode_strdup(const Eina_Unicode *text) | ||
101 | { | ||
102 | size_t len; | ||
103 | |||
104 | len = eina_unicode_strlen(text); | ||
105 | return eina_unicode_strndup(text, len); | ||
106 | } | ||
107 | |||
108 | EAPI Eina_Unicode * | ||
109 | eina_unicode_strstr(const Eina_Unicode *haystack, const Eina_Unicode *needle) | ||
110 | { | ||
111 | const Eina_Unicode *i, *j; | ||
112 | |||
113 | for (i = haystack; *i; i++) | ||
114 | { | ||
115 | haystack = i; /* set this location as the base position */ | ||
116 | for (j = needle; *j && *i && *j == *i; j++, i++) | ||
117 | ; | ||
118 | |||
119 | if (!*j) /*if we got to the end of j this means we got a full match */ | ||
120 | { | ||
121 | return (Eina_Unicode *)haystack; /* return the new base position */ | ||
122 | } | ||
123 | } | ||
124 | |||
125 | return NULL; | ||
126 | } | ||
127 | |||
128 | EAPI Eina_Unicode * | ||
129 | eina_unicode_escape(const Eina_Unicode *str) | ||
130 | { | ||
131 | Eina_Unicode *s2, *d; | ||
132 | const Eina_Unicode *s; | ||
133 | |||
134 | s2 = malloc((eina_unicode_strlen(str) * 2) + 1); | ||
135 | if (!s2) | ||
136 | return NULL; | ||
137 | |||
138 | for (s = str, d = s2; *s != 0; s++, d++) | ||
139 | { | ||
140 | if ((*s == ' ') || (*s == '\\') || (*s == '\'')) | ||
141 | { | ||
142 | *d = '\\'; | ||
143 | d++; | ||
144 | } | ||
145 | |||
146 | *d = *s; | ||
147 | } | ||
148 | *d = 0; | ||
149 | return s2; | ||
150 | } | ||
151 | |||
152 | /* UTF-8 Handling */ | ||
153 | |||
154 | #define EINA_UNICODE_UTF8_BYTES_PER_CHAR 6 | ||
155 | /* The replacement range that will be used for bad utf8 chars. */ | ||
156 | #define ERROR_REPLACEMENT_BASE 0xDC80 | ||
157 | #define ERROR_REPLACEMENT_END 0xDCFF | ||
158 | #define IS_INVALID_BYTE(x) ((x == 192) || (x == 193) || (x >= 245)) | ||
159 | #define IS_CONTINUATION_BYTE(x) ((x & 0xC0) == 0x80) | ||
160 | |||
161 | EAPI Eina_Unicode | ||
162 | eina_unicode_utf8_get_next(const char *buf, int *iindex) | ||
163 | { | ||
164 | int ind = *iindex; | ||
165 | Eina_Unicode r; | ||
166 | unsigned char d; | ||
167 | |||
168 | /* if this char is the null terminator, exit */ | ||
169 | if ((d = buf[ind++]) == 0) return 0; | ||
170 | |||
171 | if ((d & 0x80) == 0) | ||
172 | { // 1 byte (7bit) - 0xxxxxxx | ||
173 | *iindex = ind; | ||
174 | return d; | ||
175 | } | ||
176 | if ((d & 0xe0) == 0xc0) | ||
177 | { // 2 byte (11bit) - 110xxxxx 10xxxxxx | ||
178 | r = (d & 0x1f) << 6; | ||
179 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
180 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
181 | r |= (d & 0x3f); | ||
182 | if (r <= 0x7F) goto error; | ||
183 | *iindex = ind; | ||
184 | return r; | ||
185 | } | ||
186 | if ((d & 0xf0) == 0xe0) | ||
187 | { // 3 byte (16bit) - 1110xxxx 10xxxxxx 10xxxxxx | ||
188 | r = (d & 0x0f) << 12; | ||
189 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
190 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
191 | r |= (d & 0x3f) << 6; | ||
192 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
193 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
194 | r |= (d & 0x3f); | ||
195 | if (r <= 0x7FF) goto error; | ||
196 | *iindex = ind; | ||
197 | return r; | ||
198 | } | ||
199 | if ((d & 0xf8) == 0xf0) | ||
200 | { // 4 byte (21bit) - 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx | ||
201 | r = (d & 0x07) << 18; | ||
202 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
203 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
204 | r |= (d & 0x3f) << 12; | ||
205 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
206 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
207 | r |= (d & 0x3f) << 6; | ||
208 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
209 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
210 | r |= (d & 0x3f); | ||
211 | if (r <= 0xFFFF) goto error; | ||
212 | *iindex = ind; | ||
213 | return r; | ||
214 | } | ||
215 | if ((d & 0xfc) == 0xf8) | ||
216 | { // 5 byte (26bit) - 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx | ||
217 | r = (d & 0x03) << 24; | ||
218 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
219 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
220 | r |= (d & 0x3f) << 18; | ||
221 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
222 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
223 | r |= (d & 0x3f) << 12; | ||
224 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
225 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
226 | r |= (d & 0x3f) << 6; | ||
227 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
228 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
229 | r |= (d & 0x3f); | ||
230 | if (r <= 0x1FFFFF) goto error; | ||
231 | *iindex = ind; | ||
232 | return r; | ||
233 | } | ||
234 | if ((d & 0xfe) == 0xfc) | ||
235 | { // 6 byte (31bit) - 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx | ||
236 | r = (d & 0x01) << 30; | ||
237 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
238 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
239 | r |= (d & 0x3f) << 24; | ||
240 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
241 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
242 | r |= (d & 0x3f) << 18; | ||
243 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
244 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
245 | r |= (d & 0x3f) << 12; | ||
246 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
247 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
248 | r |= (d & 0x3f) << 6; | ||
249 | if (((d = buf[ind++]) == 0) || IS_INVALID_BYTE(d) || | ||
250 | !IS_CONTINUATION_BYTE(d)) goto error; | ||
251 | r |= (d & 0x3f); | ||
252 | if (r <= 0x3FFFFFF) goto error; | ||
253 | *iindex = ind; | ||
254 | return r; | ||
255 | } | ||
256 | |||
257 | /* Gets here where there was an error and we want to replace the char | ||
258 | * we just use the invalid unicode codepoints 8 lower bits represent | ||
259 | * the original char */ | ||
260 | error: | ||
261 | d = buf[*iindex]; | ||
262 | (*iindex)++; | ||
263 | return ERROR_REPLACEMENT_BASE | d; | ||
264 | } | ||
265 | |||
266 | EAPI Eina_Unicode | ||
267 | eina_unicode_utf8_get_prev(const char *buf, int *iindex) | ||
268 | { | ||
269 | int r; | ||
270 | int ind = *iindex; | ||
271 | /* First obtain the codepoint at iindex */ | ||
272 | r = eina_unicode_utf8_get_next(buf, &ind); | ||
273 | |||
274 | /* although when ind == 0 there's no previous char, we still want to get | ||
275 | * the current char */ | ||
276 | if (*iindex <= 0) | ||
277 | return r; | ||
278 | |||
279 | /* Next advance iindex to previous codepoint */ | ||
280 | ind = *iindex; | ||
281 | ind--; | ||
282 | while ((ind > 0) && ((buf[ind] & 0xc0) == 0x80)) | ||
283 | ind--; | ||
284 | |||
285 | *iindex = ind; | ||
286 | return r; | ||
287 | } | ||
288 | |||
289 | EAPI int | ||
290 | eina_unicode_utf8_get_len(const char *buf) | ||
291 | { | ||
292 | /* returns the number of utf8 characters (not bytes) in the string */ | ||
293 | int i = 0, len = 0; | ||
294 | |||
295 | while (eina_unicode_utf8_get_next(buf, &i)) | ||
296 | len++; | ||
297 | |||
298 | return len; | ||
299 | } | ||
300 | |||
301 | EAPI Eina_Unicode * | ||
302 | eina_unicode_utf8_to_unicode(const char *utf, int *_len) | ||
303 | { | ||
304 | /* FIXME: Should optimize! */ | ||
305 | int len, i; | ||
306 | int ind; | ||
307 | Eina_Unicode *buf, *uind; | ||
308 | |||
309 | len = eina_unicode_utf8_get_len(utf); | ||
310 | if (_len) | ||
311 | *_len = len; | ||
312 | buf = (Eina_Unicode *) calloc(sizeof(Eina_Unicode), (len + 1)); | ||
313 | if (!buf) return buf; | ||
314 | |||
315 | for (i = 0, ind = 0, uind = buf ; i < len ; i++, uind++) | ||
316 | { | ||
317 | *uind = eina_unicode_utf8_get_next(utf, &ind); | ||
318 | } | ||
319 | |||
320 | return buf; | ||
321 | } | ||
322 | |||
323 | EAPI char * | ||
324 | eina_unicode_unicode_to_utf8(const Eina_Unicode *uni, int *_len) | ||
325 | { | ||
326 | char *buf; | ||
327 | const Eina_Unicode *uind; | ||
328 | char *ind; | ||
329 | int ulen, len; | ||
330 | |||
331 | ulen = eina_unicode_strlen(uni); | ||
332 | buf = (char *) calloc(ulen + 1, EINA_UNICODE_UTF8_BYTES_PER_CHAR); | ||
333 | |||
334 | len = 0; | ||
335 | for (uind = uni, ind = buf ; *uind ; uind++) | ||
336 | { | ||
337 | if (*uind <= 0x7F) /* 1 byte char */ | ||
338 | { | ||
339 | *ind++ = *uind; | ||
340 | len += 1; | ||
341 | } | ||
342 | else if (*uind <= 0x7FF) /* 2 byte char */ | ||
343 | { | ||
344 | *ind++ = 0xC0 | (unsigned char) (*uind >> 6); | ||
345 | *ind++ = 0x80 | (unsigned char) (*uind & 0x3F); | ||
346 | len += 2; | ||
347 | } | ||
348 | else if (*uind <= 0xFFFF) /* 3 byte char */ | ||
349 | { | ||
350 | /* If it's a special replacement codepoint */ | ||
351 | if (*uind >= ERROR_REPLACEMENT_BASE && | ||
352 | *uind <= ERROR_REPLACEMENT_END) | ||
353 | { | ||
354 | *ind++ = *uind & 0xFF; | ||
355 | len += 1; | ||
356 | } | ||
357 | else | ||
358 | { | ||
359 | *ind++ = 0xE0 | (unsigned char) (*uind >> 12); | ||
360 | *ind++ = 0x80 | (unsigned char) ((*uind >> 6) & 0x3F); | ||
361 | *ind++ = 0x80 | (unsigned char) (*uind & 0x3F); | ||
362 | len += 3; | ||
363 | } | ||
364 | } | ||
365 | else if (*uind <= 0x1FFFFF) /* 4 byte char */ | ||
366 | { | ||
367 | *ind++ = 0xF0 | (unsigned char) ((*uind >> 18) & 0x07); | ||
368 | *ind++ = 0x80 | (unsigned char) ((*uind >> 12) & 0x3F); | ||
369 | *ind++ = 0x80 | (unsigned char) ((*uind >> 6) & 0x3F); | ||
370 | *ind++ = 0x80 | (unsigned char) (*uind & 0x3F); | ||
371 | len += 4; | ||
372 | } | ||
373 | else if (*uind <= 0x3FFFFFF) /* 5 byte char */ | ||
374 | { | ||
375 | *ind++ = 0xF8 | (unsigned char) ((*uind >> 24) & 0x03); | ||
376 | *ind++ = 0x80 | (unsigned char) ((*uind >> 18) & 0x3F); | ||
377 | *ind++ = 0x80 | (unsigned char) ((*uind >> 12) & 0x3F); | ||
378 | *ind++ = 0x80 | (unsigned char) ((*uind >> 6) & 0x3F); | ||
379 | *ind++ = 0x80 | (unsigned char) (*uind & 0x3F); | ||
380 | len += 5; | ||
381 | } | ||
382 | else if (*uind <= 0x7FFFFFFF) /* 6 byte char */ | ||
383 | { | ||
384 | *ind++ = 0xFC | (unsigned char) ((*uind >> 30) & 0x01); | ||
385 | *ind++ = 0x80 | (unsigned char) ((*uind >> 24) & 0x3F); | ||
386 | *ind++ = 0x80 | (unsigned char) ((*uind >> 18) & 0x3F); | ||
387 | *ind++ = 0x80 | (unsigned char) ((*uind >> 12) & 0x3F); | ||
388 | *ind++ = 0x80 | (unsigned char) ((*uind >> 6) & 0x3F); | ||
389 | *ind++ = 0x80 | (unsigned char) (*uind & 0x3F); | ||
390 | len += 6; | ||
391 | } | ||
392 | else /* error */ | ||
393 | { | ||
394 | /* Do something */ | ||
395 | } | ||
396 | } | ||
397 | buf = realloc(buf, len + 1); | ||
398 | buf[len] = '\0'; | ||
399 | if (_len) | ||
400 | *_len = len; | ||
401 | return buf; | ||
402 | } | ||
403 | |||
404 | |||
405 | |||