1 |
zoff99 |
2 |
#include <string.h>
|
2 |
|
|
#include <stdio.h>
|
3 |
|
|
#include <glib.h>
|
4 |
|
|
#include "debug.h"
|
5 |
|
|
#include "linguistics.h"
|
6 |
|
|
|
7 |
zoff99 |
27 |
static const char *special[][3] =
|
8 |
|
|
{
|
9 |
zoff99 |
2 |
/* Capital Diacritics */
|
10 |
|
|
/* ¨ Diaresis */
|
11 |
zoff99 |
27 |
{ "Ä", "A", "AE" },
|
12 |
|
|
{ "Ë", "E" },
|
13 |
|
|
{ "Ï", "I" },
|
14 |
|
|
{ "Ö", "O", "OE" },
|
15 |
|
|
{ "Ü", "U", "UE" },
|
16 |
|
|
{ "Ÿ", "Y" },
|
17 |
zoff99 |
2 |
/* ˝ Double Acute Accent */
|
18 |
zoff99 |
27 |
{ "Ő", "O", "Ö" },
|
19 |
|
|
{ "Ű", "U", "Ü" },
|
20 |
zoff99 |
2 |
/* ´ Acute Accent */
|
21 |
zoff99 |
27 |
{ "Á", "A" },
|
22 |
|
|
{ "Ć", "C" },
|
23 |
|
|
{ "É", "E" },
|
24 |
|
|
{ "Í", "I" },
|
25 |
|
|
{ "Ĺ", "L" },
|
26 |
|
|
{ "Ń", "N" },
|
27 |
|
|
{ "Ó", "O" },
|
28 |
|
|
{ "Ŕ", "R" },
|
29 |
|
|
{ "Ś", "S" },
|
30 |
|
|
{ "Ú", "U" },
|
31 |
|
|
{ "Ý", "Y" },
|
32 |
|
|
{ "Ź", "Z" },
|
33 |
zoff99 |
2 |
/* ˛ Ogonek (nosinė) */
|
34 |
zoff99 |
27 |
{ "Ą", "A" },
|
35 |
|
|
{ "Ę", "E" },
|
36 |
|
|
{ "Į", "I" },
|
37 |
|
|
{ "Ų", "U" },
|
38 |
zoff99 |
2 |
/* ˙ Dot */
|
39 |
zoff99 |
27 |
{ "Ċ", "C" },
|
40 |
|
|
{ "Ė", "E" },
|
41 |
|
|
{ "Ġ", "G" },
|
42 |
|
|
{ "İ", "I" },
|
43 |
|
|
{ "Ŀ", "L" },
|
44 |
|
|
{ "Ż", "Z" },
|
45 |
zoff99 |
2 |
/* – Stroke */
|
46 |
zoff99 |
27 |
{ "Đ", "D", "DJ" }, /* Croatian Dj, not to be confused with the similar-looking Icelandic Eth */
|
47 |
|
|
{ "Ħ", "H" },
|
48 |
|
|
{ "Ł", "L" },
|
49 |
|
|
{ "Ŧ", "T" },
|
50 |
zoff99 |
2 |
/* ˚ Ring */
|
51 |
zoff99 |
27 |
{ "Å", "A", "AA" },
|
52 |
|
|
{ "Ů", "U" },
|
53 |
zoff99 |
2 |
/* ˇ Caron (haček, paukščiukas) */
|
54 |
zoff99 |
27 |
{ "Č", "C" },
|
55 |
|
|
{ "Ď", "D" },
|
56 |
|
|
{ "Ě", "E" },
|
57 |
|
|
{ "Ľ", "L" },
|
58 |
|
|
{ "Ň", "N" },
|
59 |
|
|
{ "Ř", "R" },
|
60 |
|
|
{ "Š", "S" },
|
61 |
|
|
{ "Ť", "T" },
|
62 |
|
|
{ "Ž", "Z" },
|
63 |
zoff99 |
2 |
/* / Slash */
|
64 |
zoff99 |
27 |
{ "Ø", "O", "OE" },
|
65 |
zoff99 |
2 |
/* ¯ Macron */
|
66 |
zoff99 |
27 |
{ "Ā", "A", "AA" },
|
67 |
|
|
{ "Ē", "E", "EE" },
|
68 |
|
|
{ "Ī", "I", "II" },
|
69 |
|
|
{ "Ō", "O", "OO" },
|
70 |
|
|
{ "Ū", "U", "UU" },
|
71 |
zoff99 |
2 |
/* ˘ Brevis */
|
72 |
zoff99 |
27 |
{ "Ă", "A" },
|
73 |
|
|
{ "Ĕ", "E" },
|
74 |
|
|
{ "Ğ", "G" },
|
75 |
|
|
{ "Ĭ", "I" },
|
76 |
|
|
{ "Ŏ", "O" },
|
77 |
|
|
{ "Ŭ", "U" },
|
78 |
zoff99 |
2 |
/* ^ Circumflex */
|
79 |
zoff99 |
27 |
{ "Â", "A" },
|
80 |
|
|
{ "Ĉ", "C" },
|
81 |
|
|
{ "Ê", "E" },
|
82 |
|
|
{ "Ĝ", "G" },
|
83 |
|
|
{ "Ĥ", "H" },
|
84 |
|
|
{ "Î", "I" },
|
85 |
|
|
{ "Ĵ", "J" },
|
86 |
|
|
{ "Ô", "O" },
|
87 |
|
|
{ "Ŝ", "S" },
|
88 |
|
|
{ "Û", "U" },
|
89 |
|
|
{ "Ŵ", "W" },
|
90 |
|
|
{ "Ŷ", "Y" },
|
91 |
zoff99 |
2 |
/* ¸ Cedilla */
|
92 |
zoff99 |
27 |
{ "Ç", "C" },
|
93 |
|
|
{ "Ģ", "G", "GJ" },
|
94 |
|
|
{ "Ķ", "K", "KJ" },
|
95 |
|
|
{ "Ļ", "L", "LJ" },
|
96 |
|
|
{ "Ņ", "N", "NJ" },
|
97 |
|
|
{ "Ŗ", "R" },
|
98 |
|
|
{ "Ş", "S" },
|
99 |
|
|
{ "Ţ", "T" },
|
100 |
zoff99 |
2 |
/* ~ Tilde */
|
101 |
zoff99 |
27 |
{ "Ã", "A" },
|
102 |
|
|
{ "Ĩ", "I" },
|
103 |
|
|
{ "Ñ", "N" },
|
104 |
|
|
{ "Õ", "O" },
|
105 |
|
|
{ "Ũ", "U" },
|
106 |
zoff99 |
2 |
/* ` Grave */
|
107 |
zoff99 |
27 |
{ "À", "A" },
|
108 |
|
|
{ "È", "E" },
|
109 |
|
|
{ "Ì", "I" },
|
110 |
|
|
{ "Ò", "O" },
|
111 |
|
|
{ "Ù", "U" },
|
112 |
zoff99 |
2 |
/* ligatures */
|
113 |
zoff99 |
27 |
{ "Æ", "A", "AE" },
|
114 |
zoff99 |
31 |
//{ "IJ", "IJ" },
|
115 |
zoff99 |
27 |
{ "Œ", "O", "OE" },
|
116 |
zoff99 |
2 |
/* special letters */
|
117 |
zoff99 |
27 |
{ "Ð", "D", "DH" }, /* Icelandic Eth, not to be confused with the similar-looking Croatian Dj */
|
118 |
|
|
{ "Ŋ", "N", "NG" },
|
119 |
|
|
{ "Þ", "T", "TH" },
|
120 |
zoff99 |
2 |
/* Small Diacritics */
|
121 |
|
|
/* ¨ Diaresis */
|
122 |
zoff99 |
27 |
{ "ä", "a", "ae" },
|
123 |
|
|
{ "ë", "e" },
|
124 |
|
|
{ "ï", "i" },
|
125 |
|
|
{ "ö", "o", "oe" },
|
126 |
|
|
{ "ü", "u", "ue" },
|
127 |
|
|
{ "ÿ", "y" },
|
128 |
zoff99 |
2 |
/* ˝ Double Acute Accent */
|
129 |
zoff99 |
27 |
{ "ő", "o", "ö" },
|
130 |
|
|
{ "ű", "u", "ü" },
|
131 |
zoff99 |
2 |
/* ´ Acute Accent */
|
132 |
zoff99 |
27 |
{ "á", "a" },
|
133 |
|
|
{ "ć", "c" },
|
134 |
|
|
{ "é", "e" },
|
135 |
|
|
{ "í", "i" },
|
136 |
|
|
{ "ĺ", "l" },
|
137 |
|
|
{ "ń", "n" },
|
138 |
|
|
{ "ó", "o" },
|
139 |
|
|
{ "ŕ", "r" },
|
140 |
|
|
{ "ś", "s" },
|
141 |
|
|
{ "ú", "u" },
|
142 |
|
|
{ "ý", "y" },
|
143 |
|
|
{ "ź", "z" },
|
144 |
zoff99 |
2 |
/* ˛ Ogonek (nosinė) */
|
145 |
zoff99 |
27 |
{ "ą", "a" },
|
146 |
|
|
{ "ę", "e" },
|
147 |
|
|
{ "į", "i" },
|
148 |
|
|
{ "ų", "u" },
|
149 |
zoff99 |
2 |
/* ˙ Dot (and dotless i) */
|
150 |
zoff99 |
27 |
{ "ċ", "c" },
|
151 |
|
|
{ "ė", "e" },
|
152 |
|
|
{ "ġ", "g" },
|
153 |
|
|
{ "ı", "i" },
|
154 |
|
|
{ "ŀ", "l" },
|
155 |
|
|
{ "ż", "z" },
|
156 |
zoff99 |
2 |
/* – Stroke */
|
157 |
zoff99 |
27 |
{ "đ", "d", "dj" },
|
158 |
|
|
{ "ħ", "h" },
|
159 |
|
|
{ "ł", "l" },
|
160 |
|
|
{ "ŧ", "t" },
|
161 |
zoff99 |
2 |
/* ˚ Ring */
|
162 |
zoff99 |
27 |
{ "å", "a", "aa" },
|
163 |
|
|
{ "ů", "u" },
|
164 |
zoff99 |
2 |
/* ˇ Caron (haček, paukščiukas) */
|
165 |
zoff99 |
27 |
{ "č", "c" },
|
166 |
|
|
{ "ď", "d" },
|
167 |
|
|
{ "ě", "e" },
|
168 |
|
|
{ "ľ", "l" },
|
169 |
|
|
{ "ň", "n" },
|
170 |
|
|
{ "ř", "r" },
|
171 |
|
|
{ "š", "s" },
|
172 |
|
|
{ "ť", "t" },
|
173 |
|
|
{ "ž", "z" },
|
174 |
zoff99 |
2 |
/* / Slash */
|
175 |
zoff99 |
27 |
{ "ø", "o", "oe" },
|
176 |
zoff99 |
2 |
/* Macron */
|
177 |
zoff99 |
27 |
{ "ā", "a", "aa" },
|
178 |
|
|
{ "ē", "e", "ee" },
|
179 |
|
|
{ "ī", "i", "ii" },
|
180 |
|
|
{ "ō", "o", "oo" },
|
181 |
|
|
{ "ū", "u", "uu" },
|
182 |
zoff99 |
2 |
/* ˘ Brevis */
|
183 |
zoff99 |
27 |
{ "ă", "a" },
|
184 |
|
|
{ "ĕ", "e" },
|
185 |
|
|
{ "ğ", "g" },
|
186 |
|
|
{ "ĭ", "i" },
|
187 |
|
|
{ "ŏ", "o" },
|
188 |
|
|
{ "ŭ", "u" },
|
189 |
zoff99 |
2 |
/* ^ Circumflex */
|
190 |
zoff99 |
27 |
{ "â", "a" },
|
191 |
|
|
{ "ĉ", "c" },
|
192 |
|
|
{ "ê", "e" },
|
193 |
|
|
{ "ĝ", "g" },
|
194 |
|
|
{ "ĥ", "h" },
|
195 |
|
|
{ "î", "i" },
|
196 |
|
|
{ "ĵ", "j" },
|
197 |
|
|
{ "ô", "o" },
|
198 |
|
|
{ "ŝ", "s" },
|
199 |
|
|
{ "û", "u" },
|
200 |
|
|
{ "ŵ", "w" },
|
201 |
|
|
{ "ŷ", "y" },
|
202 |
zoff99 |
2 |
/* ¸ Cedilla */
|
203 |
zoff99 |
27 |
{ "ç", "c" },
|
204 |
|
|
{ "ģ", "g", "gj" },
|
205 |
|
|
{ "ķ", "k", "kj" },
|
206 |
|
|
{ "ļ", "l", "lj" },
|
207 |
|
|
{ "ņ", "n", "nj" },
|
208 |
|
|
{ "ŗ", "r" },
|
209 |
|
|
{ "ş", "s" },
|
210 |
|
|
{ "ţ", "t" },
|
211 |
zoff99 |
2 |
/* ~ Tilde */
|
212 |
zoff99 |
27 |
{ "ã", "a" },
|
213 |
|
|
{ "ĩ", "i" },
|
214 |
|
|
{ "õ", "o" },
|
215 |
|
|
{ "ñ", "n" },
|
216 |
|
|
{ "ũ", "u" },
|
217 |
zoff99 |
2 |
/* ` Grave */
|
218 |
zoff99 |
27 |
{ "à", "a" },
|
219 |
|
|
{ "è", "e" },
|
220 |
|
|
{ "ì", "i" },
|
221 |
|
|
{ "ò", "o" },
|
222 |
|
|
{ "ù", "u" },
|
223 |
zoff99 |
2 |
/* ligatures */
|
224 |
zoff99 |
27 |
{ "æ", "a", "ae" },
|
225 |
zoff99 |
31 |
//{ "ij", "ij" },
|
226 |
zoff99 |
27 |
{ "œ", "o", "oe" },
|
227 |
|
|
{ "ß", "s", "ss" },
|
228 |
zoff99 |
2 |
/* special letters */
|
229 |
zoff99 |
27 |
{ "ð", "d", "dh" },
|
230 |
|
|
{ "ŋ", "n", "ng" },
|
231 |
|
|
{ "þ", "t", "th" },
|
232 |
zoff99 |
2 |
|
233 |
|
|
/* Cyrillic capital */
|
234 |
zoff99 |
15 |
|
235 |
zoff99 |
27 |
{ "Ё", "Е" },
|
236 |
|
|
{ "Й", "И" },
|
237 |
|
|
{ "І", "I" },
|
238 |
|
|
{ "Ї", "I" },
|
239 |
|
|
{ "Ў", "У" },
|
240 |
|
|
{ "Є", "Е", "Э" },
|
241 |
|
|
{ "Ґ", "Г" },
|
242 |
|
|
{ "Ѓ", "Г" },
|
243 |
|
|
{ "Ђ", "Д" },
|
244 |
|
|
{ "Ќ", "К" },
|
245 |
zoff99 |
15 |
//{"Љ","Л","ЛЬ"},
|
246 |
zoff99 |
27 |
//{"Њ","Н","НЬ"},
|
247 |
|
|
{ "Џ", "Ц" },
|
248 |
zoff99 |
2 |
|
249 |
zoff99 |
27 |
/* Cyrillic small */
|
250 |
zoff99 |
15 |
|
251 |
zoff99 |
27 |
{ "ё", "е" },
|
252 |
|
|
{ "й", "и" },
|
253 |
|
|
{ "і", "i" },
|
254 |
|
|
{ "ї", "i" },
|
255 |
|
|
{ "ў", "у" },
|
256 |
|
|
//{"є","е","э"},
|
257 |
|
|
{ "ґ", "г" },
|
258 |
|
|
{ "ѓ", "г" },
|
259 |
|
|
{ "ђ", "д" },
|
260 |
|
|
{ "ќ", "к" },
|
261 |
|
|
//{"љ","л","ль"},
|
262 |
|
|
//{"њ","н","нь"},
|
263 |
|
|
{ "џ", "ц" },
|
264 |
zoff99 |
2 |
|
265 |
|
|
};
|
266 |
zoff99 |
15 |
|
267 |
zoff99 |
31 |
|
268 |
zoff99 |
2 |
static GHashTable *special_hash;
|
269 |
|
|
|
270 |
|
|
/* Array of strings for case conversion
|
271 |
|
|
* Even elements of array are strings of upper-case letters
|
272 |
|
|
* Odd elements of array are strings of lower-case letters, in the order corresponding to directly preceeding even element.
|
273 |
|
|
* Last element of array should be NULL.
|
274 |
|
|
*/
|
275 |
zoff99 |
27 |
static const char
|
276 |
|
|
*upperlower[] =
|
277 |
|
|
{
|
278 |
|
|
/*Latin diacritics*/
|
279 |
|
|
"ÄËÏÖÜŸŐŰÁĆÉÍĹŃÓŔŚÚÝŹĄĘĮŲĊĖĠİĿŻĐĦŁŦÅŮČĎĚĽŇŘŠŤŽØĀĒĪŌŪĂĔĞĬŎŬÂĈÊĜĤÎĴÔŜÛŴŶÇĢĶĻŅŖŞŢÃĨÑÕŨÀÈÌÒÙÆIJŒÐŊÞ",
|
280 |
|
|
"äëïöüÿőűáćéíĺńóŕśúýźąęįųċėġıŀżđħłŧåůčďěľňřšťžøāēīōūăĕğĭŏŭâĉêĝĥîĵôŝûŵŷçģķļņŗşţãĩõñũàèìòùæijœðŋþ",
|
281 |
|
|
/*Cyrillic*/
|
282 |
|
|
"АБВГҐЃДЂЕЄЁЖЗИЙКЌЛЉМНЊОПРСТУФХЦЏЧШЩЪЫЬЭЮЯІЇЎ",
|
283 |
|
|
"абвгґѓдђеєёжзийкќлљмнњопрстуфхцџчшщъыьэюяіїў",
|
284 |
zoff99 |
2 |
|
285 |
zoff99 |
27 |
NULL };
|
286 |
zoff99 |
2 |
|
287 |
|
|
static GHashTable *casefold_hash;
|
288 |
|
|
|
289 |
zoff99 |
27 |
struct special_pos
|
290 |
|
|
{
|
291 |
zoff99 |
2 |
char **variants;
|
292 |
|
|
int n;
|
293 |
|
|
char *s1, *s2;
|
294 |
|
|
};
|
295 |
|
|
|
296 |
zoff99 |
27 |
static char**
|
297 |
zoff99 |
2 |
linguistics_get_special(char *str, char *end)
|
298 |
|
|
{
|
299 |
zoff99 |
31 |
char buf[11];
|
300 |
zoff99 |
2 |
int len;
|
301 |
zoff99 |
27 |
if (!end)
|
302 |
zoff99 |
31 |
{
|
303 |
zoff99 |
27 |
end = g_utf8_find_next_char(str, NULL);
|
304 |
zoff99 |
31 |
}
|
305 |
zoff99 |
27 |
len = end - str + 1;
|
306 |
|
|
g_strlcpy(buf, str, len > 10 ? 10 : len);
|
307 |
|
|
return g_hash_table_lookup(special_hash, buf);
|
308 |
zoff99 |
2 |
}
|
309 |
|
|
|
310 |
|
|
/*
|
311 |
|
|
* @brief Prepare an utf-8 string for case insensitive comparison.
|
312 |
|
|
* @param in String to prepeare.
|
313 |
|
|
* @return String prepared for case insensitive search. Result shoud be g_free()d after use.
|
314 |
|
|
*/
|
315 |
|
|
char*
|
316 |
|
|
linguistics_casefold(char *in)
|
317 |
|
|
{
|
318 |
zoff99 |
27 |
int len = strlen(in);
|
319 |
|
|
char *src = in;
|
320 |
zoff99 |
31 |
//char *ret=g_new(char,len+1);
|
321 |
|
|
char *ret=g_new(char,len+20); // try to fix strange BUG
|
322 |
zoff99 |
27 |
char *dest = ret;
|
323 |
zoff99 |
2 |
char buf[10];
|
324 |
zoff99 |
31 |
|
325 |
|
|
// string end
|
326 |
|
|
ret[19] = '\0';
|
327 |
|
|
// fprintf(stderr, "xxxsssssssssssss\n");
|
328 |
|
|
|
329 |
|
|
while (*src && ((dest - ret) < len))
|
330 |
zoff99 |
27 |
{
|
331 |
|
|
if (*src >= 'A' && *src <= 'Z')
|
332 |
|
|
{
|
333 |
|
|
*dest++ = *src++ - 'A' + 'a';
|
334 |
|
|
}
|
335 |
|
|
else if (!(*src & 128))
|
336 |
|
|
{
|
337 |
|
|
*dest++ = *src++;
|
338 |
|
|
}
|
339 |
|
|
else
|
340 |
|
|
{
|
341 |
zoff99 |
2 |
int charlen;
|
342 |
|
|
char *tmp, *folded;
|
343 |
zoff99 |
27 |
tmp = g_utf8_find_next_char(src, NULL);
|
344 |
|
|
charlen = tmp - src + 1;
|
345 |
|
|
g_strlcpy(buf, src, charlen > 10 ? 10 : charlen);
|
346 |
|
|
folded = g_hash_table_lookup(casefold_hash, buf);
|
347 |
zoff99 |
15 |
|
348 |
zoff99 |
27 |
if (folded)
|
349 |
zoff99 |
15 |
{
|
350 |
zoff99 |
27 |
while (*folded && dest - ret < len)
|
351 |
zoff99 |
31 |
{
|
352 |
zoff99 |
27 |
*dest++ = *folded++;
|
353 |
zoff99 |
31 |
}
|
354 |
zoff99 |
27 |
src = tmp;
|
355 |
zoff99 |
15 |
}
|
356 |
|
|
else
|
357 |
|
|
{
|
358 |
zoff99 |
27 |
while (src < tmp && dest - ret < len)
|
359 |
zoff99 |
31 |
{
|
360 |
zoff99 |
27 |
*dest++ = *src++;
|
361 |
zoff99 |
31 |
}
|
362 |
zoff99 |
2 |
}
|
363 |
|
|
}
|
364 |
|
|
}
|
365 |
zoff99 |
31 |
|
366 |
zoff99 |
27 |
*dest = 0;
|
367 |
|
|
if (*src)
|
368 |
zoff99 |
31 |
{
|
369 |
zoff99 |
27 |
dbg(
|
370 |
|
|
0,
|
371 |
|
|
"Casefolded string for '%s' needs extra space, result is trucated to '%s'.\n",
|
372 |
|
|
in, ret);
|
373 |
zoff99 |
31 |
}
|
374 |
|
|
|
375 |
zoff99 |
2 |
return ret;
|
376 |
|
|
}
|
377 |
|
|
|
378 |
zoff99 |
31 |
char* linguistics_fold_and_prepare_complete(char *in, int free_input)
|
379 |
|
|
{
|
380 |
|
|
char *tmp1;
|
381 |
|
|
char *tmp2;
|
382 |
|
|
|
383 |
|
|
if (in == NULL)
|
384 |
|
|
{
|
385 |
|
|
return NULL;
|
386 |
|
|
}
|
387 |
|
|
|
388 |
|
|
tmp1 = linguistics_casefold(in);
|
389 |
|
|
if (tmp1)
|
390 |
|
|
{
|
391 |
|
|
tmp2 = linguistics_remove_all_specials(tmp1);
|
392 |
|
|
if (tmp2)
|
393 |
|
|
{
|
394 |
|
|
g_free(tmp1);
|
395 |
|
|
tmp1 = tmp2;
|
396 |
|
|
}
|
397 |
|
|
tmp2 = linguistics_expand_special(tmp1, 1);
|
398 |
|
|
if (tmp2)
|
399 |
|
|
{
|
400 |
|
|
g_free(tmp1);
|
401 |
|
|
tmp1 = tmp2;
|
402 |
|
|
}
|
403 |
|
|
}
|
404 |
|
|
|
405 |
|
|
if (free_input)
|
406 |
|
|
{
|
407 |
|
|
if (in)
|
408 |
|
|
{
|
409 |
|
|
g_free(in);
|
410 |
|
|
in = NULL;
|
411 |
|
|
}
|
412 |
|
|
}
|
413 |
|
|
|
414 |
|
|
return tmp1;
|
415 |
|
|
}
|
416 |
|
|
|
417 |
|
|
|
418 |
|
|
|
419 |
|
|
|
420 |
|
|
/*
|
421 |
|
|
* Verify that "in" points to valid "modified UTF-8" data.
|
422 |
|
|
* returns: string of useable utf-8 bytes (dont need to free, original input is just truncated)
|
423 |
|
|
*/
|
424 |
|
|
char* linguistics_check_utf8_string(char* in)
|
425 |
|
|
{
|
426 |
|
|
char* bytes = in;
|
427 |
|
|
|
428 |
|
|
if (bytes == NULL)
|
429 |
|
|
{
|
430 |
|
|
return NULL;
|
431 |
|
|
}
|
432 |
|
|
|
433 |
|
|
while (*bytes != '\0')
|
434 |
|
|
{
|
435 |
|
|
guint32 utf8 = *(bytes++);
|
436 |
|
|
// Switch on the high four bits.
|
437 |
|
|
switch (utf8 >> 4)
|
438 |
|
|
{
|
439 |
|
|
case 0x00:
|
440 |
|
|
case 0x01:
|
441 |
|
|
case 0x02:
|
442 |
|
|
case 0x03:
|
443 |
|
|
case 0x04:
|
444 |
|
|
case 0x05:
|
445 |
|
|
case 0x06:
|
446 |
|
|
case 0x07:
|
447 |
|
|
{
|
448 |
|
|
// Bit pattern 0xxx. No need for any extra bytes.
|
449 |
|
|
break;
|
450 |
|
|
}
|
451 |
|
|
case 0x08:
|
452 |
|
|
case 0x09:
|
453 |
|
|
case 0x0a:
|
454 |
|
|
case 0x0b:
|
455 |
|
|
case 0x0f:
|
456 |
|
|
{
|
457 |
|
|
/*
|
458 |
|
|
* Bit pattern 10xx or 1111, which are illegal start bytes.
|
459 |
|
|
* Note: 1111 is valid for normal UTF-8, but not the
|
460 |
|
|
* modified UTF-8 used here.
|
461 |
|
|
*/
|
462 |
|
|
// LOGW("JNI WARNING: illegal start byte 0x%x\n", utf8);
|
463 |
|
|
*(bytes--) = '\0';
|
464 |
|
|
return in;
|
465 |
|
|
}
|
466 |
|
|
case 0x0e:
|
467 |
|
|
{
|
468 |
|
|
// Bit pattern 1110, so there are two additional bytes.
|
469 |
|
|
utf8 = *(bytes++);
|
470 |
|
|
if ((utf8 & 0xc0) != 0x80)
|
471 |
|
|
{
|
472 |
|
|
// LOGW("JNI WARNING: illegal continuation byte 0x%x\n", utf8);
|
473 |
|
|
*(bytes-2) = '\0';
|
474 |
|
|
return in;
|
475 |
|
|
}
|
476 |
|
|
// Fall through to take care of the final byte.
|
477 |
|
|
}
|
478 |
|
|
case 0x0c:
|
479 |
|
|
case 0x0d:
|
480 |
|
|
{
|
481 |
|
|
// Bit pattern 110x, so there is one additional byte.
|
482 |
|
|
utf8 = *(bytes++);
|
483 |
|
|
if ((utf8 & 0xc0) != 0x80)
|
484 |
|
|
{
|
485 |
|
|
// LOGW("JNI WARNING: illegal continuation byte 0x%x\n", utf8);
|
486 |
|
|
*(bytes-2) = '\0';
|
487 |
|
|
return in;
|
488 |
|
|
}
|
489 |
|
|
break;
|
490 |
|
|
}
|
491 |
|
|
}
|
492 |
|
|
}
|
493 |
|
|
|
494 |
|
|
return in;
|
495 |
|
|
}
|
496 |
|
|
|
497 |
|
|
|
498 |
|
|
|
499 |
|
|
/*
|
500 |
|
|
*
|
501 |
|
|
* @brief find match anywhere in str (only complete match, not partial!)
|
502 |
|
|
* both strings should have beend folded (and specials removed) before calling this function
|
503 |
|
|
*
|
504 |
|
|
* @return =0 on match =1 on not matched
|
505 |
|
|
*
|
506 |
|
|
*/
|
507 |
|
|
|
508 |
|
|
int linguistics_compare_anywhere(char *str, char *match)
|
509 |
|
|
{
|
510 |
|
|
char *match_1;
|
511 |
|
|
char *match_next;
|
512 |
|
|
char *next;
|
513 |
|
|
char *next_char;
|
514 |
|
|
int found = 1;
|
515 |
|
|
gunichar match_1_unichar;
|
516 |
|
|
|
517 |
|
|
if ((str == NULL)||(match == NULL))
|
518 |
|
|
{
|
519 |
|
|
return found;
|
520 |
|
|
}
|
521 |
|
|
|
522 |
|
|
match_1 = g_strdup(match);
|
523 |
|
|
next = g_utf8_find_next_char(match_1, NULL);
|
524 |
|
|
if (next == NULL)
|
525 |
|
|
{
|
526 |
|
|
g_free(match_1);
|
527 |
|
|
return found;
|
528 |
|
|
}
|
529 |
|
|
|
530 |
|
|
*next = '\0'; // cut off after first utf-8 char
|
531 |
|
|
|
532 |
|
|
//dbg(0, "match=%s match_1=%s", match, match_1);
|
533 |
|
|
|
534 |
|
|
match_1_unichar = g_utf8_get_char(match_1);
|
535 |
|
|
|
536 |
|
|
match_next = g_utf8_strchr(str, -1, match_1_unichar);
|
537 |
|
|
while (match_next)
|
538 |
|
|
{
|
539 |
|
|
|
540 |
|
|
//dbg(0, "cmp1: match=%s match_next=%s", match, match_next);
|
541 |
|
|
|
542 |
|
|
// see if the utf-8 chars match
|
543 |
|
|
if (!strncmp(match, match_next, strlen(match)))
|
544 |
|
|
{
|
545 |
|
|
found = 0;
|
546 |
|
|
break;
|
547 |
|
|
}
|
548 |
|
|
match_next = g_utf8_strchr(g_utf8_find_next_char(match_next, NULL), -1, match_1_unichar);
|
549 |
|
|
|
550 |
|
|
//dbg(0, "cmp2: match=%s match_next=%s", match, match_next);
|
551 |
|
|
}
|
552 |
|
|
|
553 |
|
|
g_free(match_1);
|
554 |
|
|
|
555 |
|
|
return found;
|
556 |
|
|
}
|
557 |
|
|
|
558 |
|
|
|
559 |
zoff99 |
2 |
/**
|
560 |
|
|
* @brief Compare two strings using special characters expansion.
|
561 |
|
|
*
|
562 |
|
|
* @param str first string to compare, special characters are expanded.
|
563 |
|
|
* @param match second string to compare, special characters are not expanded.
|
564 |
|
|
* @param partial if = 1 then str string may be shorter than match string, in which case the rest from str isn't analysed.
|
565 |
|
|
* @return =0 strings matched, =1 not matched. Note this function return value is not fully compatible with strcmp().
|
566 |
|
|
*/
|
567 |
|
|
|
568 |
zoff99 |
27 |
int linguistics_compare(char *str, char *match, int partial)
|
569 |
zoff99 |
2 |
{
|
570 |
zoff99 |
31 |
|
571 |
|
|
if ((str == NULL)||(match == NULL))
|
572 |
|
|
{
|
573 |
|
|
return 1;
|
574 |
|
|
}
|
575 |
|
|
|
576 |
zoff99 |
27 |
char *s1 = str, *s2 = match;
|
577 |
zoff99 |
2 |
char **sp;
|
578 |
zoff99 |
27 |
int ret = 0;
|
579 |
zoff99 |
2 |
int got_match;
|
580 |
zoff99 |
27 |
GList *l = NULL;
|
581 |
|
|
while (*s1 && *s2)
|
582 |
|
|
{
|
583 |
zoff99 |
2 |
int j;
|
584 |
|
|
struct special_pos *spp;
|
585 |
|
|
char *utf_boundary, *tmp;
|
586 |
|
|
/* Skip all matching chars */
|
587 |
zoff99 |
27 |
for (j = 0; s1[j] && s1[j] == s2[j]; j++)
|
588 |
zoff99 |
31 |
{
|
589 |
zoff99 |
27 |
;
|
590 |
zoff99 |
31 |
}
|
591 |
|
|
|
592 |
zoff99 |
27 |
if (!s2[j] && (partial || !s1[j]))
|
593 |
|
|
{
|
594 |
zoff99 |
2 |
/* MATCH! */
|
595 |
zoff99 |
27 |
ret = 0;
|
596 |
zoff99 |
2 |
break;
|
597 |
|
|
}
|
598 |
zoff99 |
31 |
|
599 |
zoff99 |
2 |
/* Find beginning of first mismatching utf-8 encoded char */
|
600 |
zoff99 |
27 |
utf_boundary = s1;
|
601 |
|
|
while (*(tmp = g_utf8_find_next_char(utf_boundary, NULL)))
|
602 |
|
|
{
|
603 |
|
|
if (tmp > s1 + j)
|
604 |
zoff99 |
31 |
{
|
605 |
zoff99 |
2 |
break;
|
606 |
zoff99 |
31 |
}
|
607 |
zoff99 |
27 |
utf_boundary = tmp;
|
608 |
zoff99 |
2 |
}
|
609 |
zoff99 |
31 |
|
610 |
zoff99 |
2 |
/* Push first mismatching char to the list if it's a special char */
|
611 |
zoff99 |
27 |
sp = linguistics_get_special(utf_boundary, tmp);
|
612 |
zoff99 |
15 |
|
613 |
zoff99 |
27 |
if (sp)
|
614 |
zoff99 |
15 |
{
|
615 |
zoff99 |
2 |
spp=g_new(struct special_pos,1);
|
616 |
zoff99 |
27 |
spp->variants = sp;
|
617 |
|
|
spp->n = 1;
|
618 |
|
|
spp->s1 = utf_boundary;
|
619 |
|
|
spp->s2 = s2 + (utf_boundary - s1);
|
620 |
|
|
l = g_list_prepend(l, spp);
|
621 |
|
|
}
|
622 |
zoff99 |
2 |
|
623 |
|
|
/* Try to find a match using special char variants from the list */
|
624 |
zoff99 |
27 |
got_match = 0;
|
625 |
|
|
while (l && !got_match)
|
626 |
|
|
{
|
627 |
|
|
spp = l->data;
|
628 |
|
|
s1 = spp->s1;
|
629 |
|
|
s2 = spp->s2;
|
630 |
|
|
while (spp->n < 3 && !got_match)
|
631 |
|
|
{
|
632 |
|
|
char *s = spp->variants[(spp->n)++];
|
633 |
zoff99 |
2 |
int len;
|
634 |
zoff99 |
27 |
if (!s)
|
635 |
zoff99 |
2 |
break;
|
636 |
zoff99 |
27 |
len = strlen(s);
|
637 |
|
|
if (!strncmp(s, s2, len))
|
638 |
|
|
{
|
639 |
|
|
s2 += len;
|
640 |
|
|
s1 += strlen(spp->variants[0]);
|
641 |
|
|
got_match = 1;
|
642 |
zoff99 |
2 |
break;
|
643 |
|
|
}
|
644 |
|
|
}
|
645 |
zoff99 |
27 |
if (spp->n >= 3 || !spp->variants[spp->n])
|
646 |
|
|
{
|
647 |
zoff99 |
2 |
/* No matches for current top list element, go to the closest special char towards beginning of the string */
|
648 |
|
|
g_free(spp);
|
649 |
zoff99 |
27 |
l = g_list_delete_link(l, l);
|
650 |
zoff99 |
2 |
}
|
651 |
|
|
}
|
652 |
zoff99 |
31 |
|
653 |
zoff99 |
27 |
if (!got_match)
|
654 |
|
|
{
|
655 |
zoff99 |
2 |
/* NO MATCH
|
656 |
|
|
* FIXME: If we're going to use this function to sort a string list alphabetically we should use
|
657 |
|
|
* utf-aware comparison here.
|
658 |
|
|
*/
|
659 |
zoff99 |
27 |
ret = 1;
|
660 |
zoff99 |
2 |
break;
|
661 |
|
|
}
|
662 |
|
|
}
|
663 |
zoff99 |
31 |
|
664 |
zoff99 |
27 |
while (l)
|
665 |
|
|
{
|
666 |
zoff99 |
2 |
g_free(l->data);
|
667 |
zoff99 |
27 |
l = g_list_delete_link(l, l);
|
668 |
zoff99 |
2 |
}
|
669 |
|
|
return ret;
|
670 |
|
|
}
|
671 |
|
|
|
672 |
|
|
char *
|
673 |
|
|
linguistics_expand_special(char *str, int mode)
|
674 |
|
|
{
|
675 |
zoff99 |
27 |
char *in = str;
|
676 |
|
|
char *out, *ret;
|
677 |
|
|
int found = 0;
|
678 |
zoff99 |
2 |
|
679 |
zoff99 |
15 |
if (!str)
|
680 |
|
|
{
|
681 |
|
|
return NULL;
|
682 |
|
|
}
|
683 |
|
|
|
684 |
zoff99 |
27 |
ret = g_strdup(str);
|
685 |
|
|
out = ret;
|
686 |
zoff99 |
15 |
|
687 |
|
|
if (!mode)
|
688 |
|
|
{
|
689 |
zoff99 |
2 |
return ret;
|
690 |
zoff99 |
15 |
}
|
691 |
zoff99 |
2 |
|
692 |
|
|
while (*in)
|
693 |
|
|
{
|
694 |
zoff99 |
27 |
char *next = g_utf8_find_next_char(in, NULL);
|
695 |
|
|
int i, len = next - in;
|
696 |
|
|
int match = 0;
|
697 |
zoff99 |
2 |
if (len > 1)
|
698 |
|
|
{
|
699 |
zoff99 |
27 |
for (i = 0; i < sizeof(special) / sizeof(special[0]); i++)
|
700 |
zoff99 |
2 |
{
|
701 |
zoff99 |
27 |
const char *search = special[i][0];
|
702 |
|
|
if (!strncmp(in, search, len))
|
703 |
zoff99 |
2 |
{
|
704 |
zoff99 |
27 |
const char *replace = special[i][mode];
|
705 |
zoff99 |
2 |
if (replace)
|
706 |
|
|
{
|
707 |
zoff99 |
27 |
int replace_len = strlen(replace);
|
708 |
zoff99 |
2 |
|
709 |
zoff99 |
15 |
if (replace_len > len)
|
710 |
zoff99 |
2 |
{
|
711 |
zoff99 |
27 |
fprintf(
|
712 |
|
|
stderr,
|
713 |
|
|
"* ERROR !! ERROR !! found %s %s %d %s %d\n",
|
714 |
|
|
in, search, len, replace, replace_len);
|
715 |
zoff99 |
2 |
}
|
716 |
zoff99 |
15 |
dbg_assert(replace_len <= len);
|
717 |
|
|
if (replace_len > len)
|
718 |
|
|
{
|
719 |
zoff99 |
27 |
out += len;
|
720 |
|
|
match = 0;
|
721 |
zoff99 |
15 |
break;
|
722 |
|
|
}
|
723 |
zoff99 |
2 |
else
|
724 |
|
|
{
|
725 |
zoff99 |
15 |
// fprintf(stderr," GOOD !! GOOD !! found %s %s %d %s %d\n",in,search,len,replace,replace_len);
|
726 |
zoff99 |
2 |
strcpy(out, replace);
|
727 |
zoff99 |
27 |
out += replace_len;
|
728 |
|
|
match = 1;
|
729 |
zoff99 |
2 |
break;
|
730 |
|
|
}
|
731 |
|
|
}
|
732 |
|
|
}
|
733 |
|
|
}
|
734 |
|
|
}
|
735 |
|
|
|
736 |
|
|
if (match)
|
737 |
|
|
{
|
738 |
zoff99 |
27 |
found = 1;
|
739 |
|
|
in = next;
|
740 |
zoff99 |
2 |
}
|
741 |
|
|
else
|
742 |
|
|
{
|
743 |
|
|
while (len-- > 0)
|
744 |
|
|
{
|
745 |
zoff99 |
27 |
*out++ = *in++;
|
746 |
zoff99 |
2 |
}
|
747 |
|
|
}
|
748 |
|
|
}
|
749 |
zoff99 |
27 |
*out++ = '\0';
|
750 |
zoff99 |
2 |
if (!found)
|
751 |
|
|
{
|
752 |
zoff99 |
15 |
if (ret)
|
753 |
|
|
{
|
754 |
|
|
g_free(ret);
|
755 |
|
|
}
|
756 |
zoff99 |
27 |
ret = NULL;
|
757 |
zoff99 |
2 |
}
|
758 |
|
|
return ret;
|
759 |
|
|
}
|
760 |
|
|
|
761 |
zoff99 |
31 |
char *linguistics_remove_all_spaces(char *str)
|
762 |
|
|
{
|
763 |
|
|
char *p;
|
764 |
|
|
char *next = NULL;
|
765 |
|
|
int len = 0;
|
766 |
|
|
char *ret;
|
767 |
|
|
char *out;
|
768 |
|
|
|
769 |
|
|
ret = g_strdup(str);
|
770 |
|
|
out = ret;
|
771 |
|
|
p = str;
|
772 |
|
|
while (*p)
|
773 |
|
|
{
|
774 |
|
|
next = g_utf8_find_next_char(p, NULL);
|
775 |
|
|
len = next - p;
|
776 |
|
|
if ((len > 1)||(p[0] != ' '))
|
777 |
|
|
{
|
778 |
|
|
strncpy(out, p, len);
|
779 |
|
|
out = out + len;
|
780 |
|
|
}
|
781 |
|
|
p = next;
|
782 |
|
|
}
|
783 |
|
|
*out = '\0';
|
784 |
|
|
|
785 |
|
|
return ret;
|
786 |
|
|
}
|
787 |
|
|
|
788 |
|
|
// special characters
|
789 |
|
|
static const char *remove_those = " _-.—,;:*#?=%&$§!@~()[]{}'`´^°|<>\\/\n\r\t\"\'";
|
790 |
|
|
|
791 |
|
|
char *linguistics_remove_all_specials(char *str)
|
792 |
|
|
{
|
793 |
|
|
char *p;
|
794 |
|
|
char *next = NULL;
|
795 |
|
|
int len = 0;
|
796 |
|
|
char *ret;
|
797 |
|
|
char *out;
|
798 |
|
|
int i;
|
799 |
|
|
int found_special;
|
800 |
|
|
int so_rtz = sizeof(remove_those[0]); // should be 1, but lets calculate it anyway
|
801 |
|
|
int so_rt = strlen(remove_those) * so_rtz;
|
802 |
|
|
|
803 |
|
|
ret = g_strdup(str);
|
804 |
|
|
out = ret;
|
805 |
|
|
p = str;
|
806 |
|
|
while (*p)
|
807 |
|
|
{
|
808 |
|
|
next = g_utf8_find_next_char(p, NULL);
|
809 |
|
|
len = next - p;
|
810 |
|
|
if (len > 1)
|
811 |
|
|
{
|
812 |
|
|
strncpy(out, p, len);
|
813 |
|
|
out = out + len;
|
814 |
|
|
}
|
815 |
|
|
else
|
816 |
|
|
{
|
817 |
|
|
found_special = 0;
|
818 |
|
|
for (i = 0; i < (so_rt / so_rtz); i++)
|
819 |
|
|
{
|
820 |
|
|
if (p[0] == remove_those[i])
|
821 |
|
|
{
|
822 |
|
|
// special found -> skip it
|
823 |
|
|
found_special = 1;
|
824 |
|
|
break;
|
825 |
|
|
}
|
826 |
|
|
}
|
827 |
|
|
|
828 |
|
|
if (found_special == 0)
|
829 |
|
|
{
|
830 |
|
|
strncpy(out, p, len);
|
831 |
|
|
out = out + len;
|
832 |
|
|
}
|
833 |
|
|
}
|
834 |
|
|
p = next;
|
835 |
|
|
}
|
836 |
|
|
*out = '\0';
|
837 |
|
|
|
838 |
|
|
return ret;
|
839 |
|
|
}
|
840 |
|
|
|
841 |
|
|
|
842 |
zoff99 |
2 |
char *
|
843 |
|
|
linguistics_next_word(char *str)
|
844 |
|
|
{
|
845 |
zoff99 |
27 |
char* ret = strtok(str, " -/()\"\',.;_[]{}\\");
|
846 |
zoff99 |
15 |
return ret;
|
847 |
|
|
|
848 |
zoff99 |
27 |
// int len=strcspn(str, " -/()");
|
849 |
|
|
// if (!str[len] || !str[len+1])
|
850 |
|
|
// return NULL;
|
851 |
|
|
// return str+len+1;
|
852 |
zoff99 |
15 |
|
853 |
zoff99 |
2 |
}
|
854 |
|
|
|
855 |
zoff99 |
27 |
int linguistics_search(char *str)
|
856 |
zoff99 |
2 |
{
|
857 |
zoff99 |
27 |
if (!g_strcasecmp(str, "str"))
|
858 |
zoff99 |
2 |
return 0;
|
859 |
zoff99 |
27 |
if (!g_strcasecmp(str, "str."))
|
860 |
zoff99 |
2 |
return 0;
|
861 |
zoff99 |
27 |
if (!g_strcasecmp(str, "strasse"))
|
862 |
zoff99 |
2 |
return 0;
|
863 |
zoff99 |
27 |
if (!g_strcasecmp(str, "weg"))
|
864 |
zoff99 |
2 |
return 0;
|
865 |
|
|
return 1;
|
866 |
|
|
}
|
867 |
|
|
|
868 |
|
|
/**
|
869 |
|
|
* @brief Copy one utf8 encoded char to newly allocated buffer.
|
870 |
|
|
*
|
871 |
|
|
* @param s pointer to the beginning of the char.
|
872 |
|
|
* @return newly allocated nul-terminated string containing one utf8 encoded character.
|
873 |
|
|
*/
|
874 |
zoff99 |
27 |
static char *linguistics_dup_utf8_char(const char *s)
|
875 |
zoff99 |
2 |
{
|
876 |
|
|
char *ret, *next;
|
877 |
zoff99 |
27 |
next = g_utf8_find_next_char(s, NULL);
|
878 |
zoff99 |
2 |
ret=g_new(char, next-s+1);
|
879 |
zoff99 |
27 |
g_strlcpy(ret, s, next - s + 1);
|
880 |
zoff99 |
2 |
return ret;
|
881 |
|
|
}
|
882 |
|
|
|
883 |
zoff99 |
27 |
void linguistics_init(void)
|
884 |
zoff99 |
2 |
{
|
885 |
|
|
int i;
|
886 |
zoff99 |
27 |
special_hash = g_hash_table_new(g_str_hash, g_str_equal);
|
887 |
|
|
casefold_hash = g_hash_table_new(g_str_hash, g_str_equal);
|
888 |
zoff99 |
2 |
|
889 |
zoff99 |
27 |
for (i = 0; i < sizeof(special) / sizeof(special[0]); i++)
|
890 |
zoff99 |
15 |
{
|
891 |
zoff99 |
27 |
g_hash_table_insert(special_hash, (gpointer) special[i][0], special[i]);
|
892 |
zoff99 |
15 |
}
|
893 |
zoff99 |
2 |
|
894 |
zoff99 |
27 |
for (i = 0; upperlower[i]; i += 2)
|
895 |
zoff99 |
15 |
{
|
896 |
zoff99 |
27 |
int j, k;
|
897 |
|
|
for (j = 0, k = 0; upperlower[i][j] && upperlower[i + 1][k];)
|
898 |
zoff99 |
15 |
{
|
899 |
zoff99 |
27 |
char *s1 = linguistics_dup_utf8_char(upperlower[i] + j);
|
900 |
|
|
char *s2 = linguistics_dup_utf8_char(upperlower[i + 1] + k);
|
901 |
|
|
g_hash_table_insert(casefold_hash, s1, s2);
|
902 |
|
|
j += strlen(s1);
|
903 |
|
|
k += strlen(s2);
|
904 |
zoff99 |
2 |
}
|
905 |
|
|
}
|
906 |
|
|
}
|
907 |
|
|
|