1 | /* Copyright (c) 2000, 2014, Oracle and/or its affiliates. |
2 | Copyright (c) 2009, 2014, SkySQL Ab. |
3 | |
4 | This program is free software; you can redistribute it and/or modify |
5 | it under the terms of the GNU General Public License as published by |
6 | the Free Software Foundation; version 2 of the License. |
7 | |
8 | This program is distributed in the hope that it will be useful, |
9 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
10 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
11 | GNU General Public License for more details. |
12 | |
13 | You should have received a copy of the GNU General Public License |
14 | along with this program; if not, write to the Free Software |
15 | Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ |
16 | |
17 | #include "strings_def.h" |
18 | #include <m_ctype.h> |
19 | |
20 | #ifdef USE_MB |
21 | |
22 | |
23 | size_t my_caseup_str_mb(CHARSET_INFO * cs, char *str) |
24 | { |
25 | register uint32 l; |
26 | register const uchar *map= cs->to_upper; |
27 | char *str_orig= str; |
28 | |
29 | while (*str) |
30 | { |
31 | /* Pointing after the '\0' is safe here. */ |
32 | if ((l= my_ismbchar(cs, str, str + cs->mbmaxlen))) |
33 | str+= l; |
34 | else |
35 | { |
36 | *str= (char) map[(uchar)*str]; |
37 | str++; |
38 | } |
39 | } |
40 | return (size_t) (str - str_orig); |
41 | } |
42 | |
43 | |
44 | size_t my_casedn_str_mb(CHARSET_INFO * cs, char *str) |
45 | { |
46 | register uint32 l; |
47 | register const uchar *map= cs->to_lower; |
48 | char *str_orig= str; |
49 | |
50 | while (*str) |
51 | { |
52 | /* Pointing after the '\0' is safe here. */ |
53 | if ((l= my_ismbchar(cs, str, str + cs->mbmaxlen))) |
54 | str+= l; |
55 | else |
56 | { |
57 | *str= (char) map[(uchar)*str]; |
58 | str++; |
59 | } |
60 | } |
61 | return (size_t) (str - str_orig); |
62 | } |
63 | |
64 | |
65 | static inline MY_UNICASE_CHARACTER* |
66 | get_case_info_for_ch(CHARSET_INFO *cs, uint page, uint offs) |
67 | { |
68 | MY_UNICASE_CHARACTER *p; |
69 | return cs->caseinfo && (p= cs->caseinfo->page[page]) ? &p[offs] : NULL; |
70 | } |
71 | |
72 | |
73 | /* |
74 | For character sets which don't change octet length in case conversion. |
75 | */ |
76 | size_t my_caseup_mb(CHARSET_INFO * cs, char *src, size_t srclen, |
77 | char *dst __attribute__((unused)), |
78 | size_t dstlen __attribute__((unused))) |
79 | { |
80 | register uint32 l; |
81 | register char *srcend= src + srclen; |
82 | register const uchar *map= cs->to_upper; |
83 | |
84 | DBUG_ASSERT(cs->caseup_multiply == 1); |
85 | DBUG_ASSERT(src == dst && srclen == dstlen); |
86 | DBUG_ASSERT(cs->mbmaxlen == 2); |
87 | |
88 | while (src < srcend) |
89 | { |
90 | if ((l=my_ismbchar(cs, src, srcend))) |
91 | { |
92 | MY_UNICASE_CHARACTER *ch; |
93 | if ((ch= get_case_info_for_ch(cs, (uchar) src[0], (uchar) src[1]))) |
94 | { |
95 | *src++= ch->toupper >> 8; |
96 | *src++= ch->toupper & 0xFF; |
97 | } |
98 | else |
99 | src+= l; |
100 | } |
101 | else |
102 | { |
103 | *src=(char) map[(uchar) *src]; |
104 | src++; |
105 | } |
106 | } |
107 | return srclen; |
108 | } |
109 | |
110 | |
111 | size_t my_casedn_mb(CHARSET_INFO * cs, char *src, size_t srclen, |
112 | char *dst __attribute__((unused)), |
113 | size_t dstlen __attribute__((unused))) |
114 | { |
115 | register uint32 l; |
116 | register char *srcend= src + srclen; |
117 | register const uchar *map=cs->to_lower; |
118 | |
119 | DBUG_ASSERT(cs->casedn_multiply == 1); |
120 | DBUG_ASSERT(src == dst && srclen == dstlen); |
121 | DBUG_ASSERT(cs->mbmaxlen == 2); |
122 | |
123 | while (src < srcend) |
124 | { |
125 | if ((l= my_ismbchar(cs, src, srcend))) |
126 | { |
127 | MY_UNICASE_CHARACTER *ch; |
128 | if ((ch= get_case_info_for_ch(cs, (uchar) src[0], (uchar) src[1]))) |
129 | { |
130 | *src++= ch->tolower >> 8; |
131 | *src++= ch->tolower & 0xFF; |
132 | } |
133 | else |
134 | src+= l; |
135 | } |
136 | else |
137 | { |
138 | *src= (char) map[(uchar)*src]; |
139 | src++; |
140 | } |
141 | } |
142 | return srclen; |
143 | } |
144 | |
145 | |
146 | /* |
147 | Case folding functions for character set |
148 | where case conversion can change string octet length. |
149 | For example, in EUCKR, |
150 | _euckr 0xA9A5 == "LATIN LETTER DOTLESS I" (Turkish letter) |
151 | is upper-cased to to |
152 | _euckr 0x49 "LATIN CAPITAL LETTER I" ('usual' letter I) |
153 | Length is reduced in this example from two bytes to one byte. |
154 | */ |
155 | static size_t |
156 | my_casefold_mb_varlen(CHARSET_INFO *cs, |
157 | char *src, size_t srclen, |
158 | char *dst, size_t dstlen __attribute__((unused)), |
159 | const uchar *map, |
160 | size_t is_upper) |
161 | { |
162 | char *srcend= src + srclen, *dst0= dst; |
163 | |
164 | DBUG_ASSERT(cs->mbmaxlen == 2); |
165 | |
166 | while (src < srcend) |
167 | { |
168 | size_t mblen= my_ismbchar(cs, src, srcend); |
169 | if (mblen) |
170 | { |
171 | MY_UNICASE_CHARACTER *ch; |
172 | if ((ch= get_case_info_for_ch(cs, (uchar) src[0], (uchar) src[1]))) |
173 | { |
174 | int code= is_upper ? ch->toupper : ch->tolower; |
175 | src+= 2; |
176 | if (code > 0xFF) |
177 | *dst++= code >> 8; |
178 | *dst++= code & 0xFF; |
179 | } |
180 | else |
181 | { |
182 | *dst++= *src++; |
183 | *dst++= *src++; |
184 | } |
185 | } |
186 | else |
187 | { |
188 | *dst++= (char) map[(uchar) *src++]; |
189 | } |
190 | } |
191 | return (size_t) (dst - dst0); |
192 | } |
193 | |
194 | |
195 | size_t |
196 | my_casedn_mb_varlen(CHARSET_INFO * cs, char *src, size_t srclen, |
197 | char *dst, size_t dstlen) |
198 | { |
199 | DBUG_ASSERT(dstlen >= srclen * cs->casedn_multiply); |
200 | DBUG_ASSERT(src != dst || cs->casedn_multiply == 1); |
201 | return my_casefold_mb_varlen(cs, src, srclen, dst, dstlen, cs->to_lower, 0); |
202 | } |
203 | |
204 | |
205 | size_t |
206 | my_caseup_mb_varlen(CHARSET_INFO * cs, char *src, size_t srclen, |
207 | char *dst, size_t dstlen) |
208 | { |
209 | DBUG_ASSERT(dstlen >= srclen * cs->caseup_multiply); |
210 | DBUG_ASSERT(src != dst || cs->caseup_multiply == 1); |
211 | return my_casefold_mb_varlen(cs, src, srclen, dst, dstlen, cs->to_upper, 1); |
212 | } |
213 | |
214 | |
215 | /* |
216 | my_strcasecmp_mb() returns 0 if strings are equal, non-zero otherwise. |
217 | */ |
218 | |
219 | int my_strcasecmp_mb(CHARSET_INFO * cs,const char *s, const char *t) |
220 | { |
221 | register uint32 l; |
222 | register const uchar *map=cs->to_upper; |
223 | |
224 | while (*s && *t) |
225 | { |
226 | /* Pointing after the '\0' is safe here. */ |
227 | if ((l=my_ismbchar(cs, s, s + cs->mbmaxlen))) |
228 | { |
229 | while (l--) |
230 | if (*s++ != *t++) |
231 | return 1; |
232 | } |
233 | else if (my_charlen(cs, t, t + cs->mbmaxlen) > 1) |
234 | return 1; |
235 | else if (map[(uchar) *s++] != map[(uchar) *t++]) |
236 | return 1; |
237 | } |
238 | /* At least one of '*s' and '*t' is zero here. */ |
239 | return (*t != *s); |
240 | } |
241 | |
242 | |
243 | /* |
244 | ** Compare string against string with wildcard |
245 | ** 0 if matched |
246 | ** -1 if not matched with wildcard |
247 | ** 1 if matched with wildcard |
248 | */ |
249 | |
250 | #define INC_PTR(cs,A,B) A+=(my_ismbchar(cs,A,B) ? my_ismbchar(cs,A,B) : 1) |
251 | |
252 | #define likeconv(s,A) (uchar) (s)->sort_order[(uchar) (A)] |
253 | |
254 | static |
255 | int my_wildcmp_mb_impl(CHARSET_INFO *cs, |
256 | const char *str,const char *str_end, |
257 | const char *wildstr,const char *wildend, |
258 | int escape, int w_one, int w_many, int recurse_level) |
259 | { |
260 | int result= -1; /* Not found, using wildcards */ |
261 | |
262 | if (my_string_stack_guard && my_string_stack_guard(recurse_level)) |
263 | return 1; |
264 | while (wildstr != wildend) |
265 | { |
266 | while (*wildstr != w_many && *wildstr != w_one) |
267 | { |
268 | int l; |
269 | if (*wildstr == escape && wildstr+1 != wildend) |
270 | wildstr++; |
271 | if ((l = my_ismbchar(cs, wildstr, wildend))) |
272 | { |
273 | if (str+l > str_end || memcmp(str, wildstr, l) != 0) |
274 | return 1; |
275 | str += l; |
276 | wildstr += l; |
277 | } |
278 | else |
279 | if (str == str_end || likeconv(cs,*wildstr++) != likeconv(cs,*str++)) |
280 | return(1); /* No match */ |
281 | if (wildstr == wildend) |
282 | return (str != str_end); /* Match if both are at end */ |
283 | result=1; /* Found an anchor char */ |
284 | } |
285 | if (*wildstr == w_one) |
286 | { |
287 | do |
288 | { |
289 | if (str == str_end) /* Skip one char if possible */ |
290 | return (result); |
291 | INC_PTR(cs,str,str_end); |
292 | } while (++wildstr < wildend && *wildstr == w_one); |
293 | if (wildstr == wildend) |
294 | break; |
295 | } |
296 | if (*wildstr == w_many) |
297 | { /* Found w_many */ |
298 | uchar cmp; |
299 | const char* mb = wildstr; |
300 | int mb_len=0; |
301 | |
302 | wildstr++; |
303 | /* Remove any '%' and '_' from the wild search string */ |
304 | for (; wildstr != wildend ; wildstr++) |
305 | { |
306 | if (*wildstr == w_many) |
307 | continue; |
308 | if (*wildstr == w_one) |
309 | { |
310 | if (str == str_end) |
311 | return (-1); |
312 | INC_PTR(cs,str,str_end); |
313 | continue; |
314 | } |
315 | break; /* Not a wild character */ |
316 | } |
317 | if (wildstr == wildend) |
318 | return(0); /* Ok if w_many is last */ |
319 | if (str == str_end) |
320 | return -1; |
321 | |
322 | if ((cmp= *wildstr) == escape && wildstr+1 != wildend) |
323 | cmp= *++wildstr; |
324 | |
325 | mb=wildstr; |
326 | mb_len= my_ismbchar(cs, wildstr, wildend); |
327 | INC_PTR(cs,wildstr,wildend); /* This is compared trough cmp */ |
328 | cmp=likeconv(cs,cmp); |
329 | do |
330 | { |
331 | for (;;) |
332 | { |
333 | if (str >= str_end) |
334 | return -1; |
335 | if (mb_len) |
336 | { |
337 | if (str+mb_len <= str_end && memcmp(str, mb, mb_len) == 0) |
338 | { |
339 | str += mb_len; |
340 | break; |
341 | } |
342 | } |
343 | else if (!my_ismbchar(cs, str, str_end) && |
344 | likeconv(cs,*str) == cmp) |
345 | { |
346 | str++; |
347 | break; |
348 | } |
349 | INC_PTR(cs,str, str_end); |
350 | } |
351 | { |
352 | int tmp=my_wildcmp_mb_impl(cs,str,str_end,wildstr,wildend,escape,w_one, |
353 | w_many, recurse_level + 1); |
354 | if (tmp <= 0) |
355 | return (tmp); |
356 | } |
357 | } while (str != str_end); |
358 | return(-1); |
359 | } |
360 | } |
361 | return (str != str_end ? 1 : 0); |
362 | } |
363 | |
364 | int my_wildcmp_mb(CHARSET_INFO *cs, |
365 | const char *str,const char *str_end, |
366 | const char *wildstr,const char *wildend, |
367 | int escape, int w_one, int w_many) |
368 | { |
369 | return my_wildcmp_mb_impl(cs, str, str_end, |
370 | wildstr, wildend, |
371 | escape, w_one, w_many, 1); |
372 | } |
373 | |
374 | |
375 | size_t my_numchars_mb(CHARSET_INFO *cs __attribute__((unused)), |
376 | const char *pos, const char *end) |
377 | { |
378 | register size_t count= 0; |
379 | while (pos < end) |
380 | { |
381 | uint mb_len; |
382 | pos+= (mb_len= my_ismbchar(cs,pos,end)) ? mb_len : 1; |
383 | count++; |
384 | } |
385 | return count; |
386 | } |
387 | |
388 | |
389 | size_t my_charpos_mb(CHARSET_INFO *cs __attribute__((unused)), |
390 | const char *pos, const char *end, size_t length) |
391 | { |
392 | const char *start= pos; |
393 | |
394 | while (length && pos < end) |
395 | { |
396 | uint mb_len; |
397 | pos+= (mb_len= my_ismbchar(cs, pos, end)) ? mb_len : 1; |
398 | length--; |
399 | } |
400 | return (size_t) (length ? end+2-start : pos-start); |
401 | } |
402 | |
403 | |
404 | /* |
405 | Append a badly formed piece of string. |
406 | Bad bytes are fixed to '?'. |
407 | |
408 | @param to The destination string |
409 | @param to_end The end of the destination string |
410 | @param from The source string |
411 | @param from_end The end of the source string |
412 | @param nchars Write not more than "nchars" characters. |
413 | @param status Copying status, must be previously initialized, |
414 | e.g. using well_formed_char_length() on the original |
415 | full source string. |
416 | */ |
417 | static size_t |
418 | my_append_fix_badly_formed_tail(CHARSET_INFO *cs, |
419 | char *to, char *to_end, |
420 | const char *from, const char *from_end, |
421 | size_t nchars, |
422 | MY_STRCOPY_STATUS *status) |
423 | { |
424 | char *to0= to; |
425 | |
426 | for ( ; nchars; nchars--) |
427 | { |
428 | int chlen; |
429 | if ((chlen= cs->cset->charlen(cs, (const uchar*) from, |
430 | (const uchar *) from_end)) > 0) |
431 | { |
432 | /* Found a valid character */ /* chlen == 1..MBMAXLEN */ |
433 | DBUG_ASSERT(chlen <= (int) cs->mbmaxlen); |
434 | if (to + chlen > to_end) |
435 | goto end; /* Does not fit to "to" */ |
436 | memcpy(to, from, (size_t) chlen); |
437 | from+= chlen; |
438 | to+= chlen; |
439 | continue; |
440 | } |
441 | if (chlen == MY_CS_ILSEQ) /* chlen == 0 */ |
442 | { |
443 | DBUG_ASSERT(from < from_end); /* Shouldn't get MY_CS_ILSEQ if empty */ |
444 | goto bad; |
445 | } |
446 | /* Got an incomplete character */ /* chlen == MY_CS_TOOSMALLXXX */ |
447 | DBUG_ASSERT(chlen >= MY_CS_TOOSMALL6); |
448 | DBUG_ASSERT(chlen <= MY_CS_TOOSMALL); |
449 | if (from >= from_end) |
450 | break; /* End of the source string */ |
451 | bad: |
452 | /* Bad byte sequence, or incomplete character found */ |
453 | if (!status->m_well_formed_error_pos) |
454 | status->m_well_formed_error_pos= from; |
455 | |
456 | if ((chlen= cs->cset->wc_mb(cs, '?', (uchar*) to, (uchar *) to_end)) <= 0) |
457 | break; /* Question mark does not fit into the destination */ |
458 | to+= chlen; |
459 | from++; |
460 | } |
461 | end: |
462 | status->m_source_end_pos= from; |
463 | return to - to0; |
464 | } |
465 | |
466 | |
467 | size_t |
468 | my_copy_fix_mb(CHARSET_INFO *cs, |
469 | char *dst, size_t dst_length, |
470 | const char *src, size_t src_length, |
471 | size_t nchars, MY_STRCOPY_STATUS *status) |
472 | { |
473 | size_t well_formed_nchars; |
474 | size_t well_formed_length; |
475 | size_t fixed_length; |
476 | |
477 | set_if_smaller(src_length, dst_length); |
478 | well_formed_nchars= cs->cset->well_formed_char_length(cs, |
479 | src, src + src_length, |
480 | nchars, status); |
481 | DBUG_ASSERT(well_formed_nchars <= nchars); |
482 | memmove(dst, src, (well_formed_length= status->m_source_end_pos - src)); |
483 | if (!status->m_well_formed_error_pos) |
484 | return well_formed_length; |
485 | |
486 | fixed_length= my_append_fix_badly_formed_tail(cs, |
487 | dst + well_formed_length, |
488 | dst + dst_length, |
489 | src + well_formed_length, |
490 | src + src_length, |
491 | nchars - well_formed_nchars, |
492 | status); |
493 | return well_formed_length + fixed_length; |
494 | } |
495 | |
496 | |
497 | uint my_instr_mb(CHARSET_INFO *cs, |
498 | const char *b, size_t b_length, |
499 | const char *s, size_t s_length, |
500 | my_match_t *match, uint nmatch) |
501 | { |
502 | register const char *end, *b0; |
503 | int res= 0; |
504 | |
505 | if (s_length <= b_length) |
506 | { |
507 | if (!s_length) |
508 | { |
509 | if (nmatch) |
510 | { |
511 | match->beg= 0; |
512 | match->end= 0; |
513 | match->mb_len= 0; |
514 | } |
515 | return 1; /* Empty string is always found */ |
516 | } |
517 | |
518 | b0= b; |
519 | end= b+b_length-s_length+1; |
520 | |
521 | while (b < end) |
522 | { |
523 | int mb_len; |
524 | |
525 | if (!cs->coll->strnncoll(cs, (uchar*) b, s_length, |
526 | (uchar*) s, s_length, 0)) |
527 | { |
528 | if (nmatch) |
529 | { |
530 | match[0].beg= 0; |
531 | match[0].end= (uint) (b-b0); |
532 | match[0].mb_len= res; |
533 | if (nmatch > 1) |
534 | { |
535 | match[1].beg= match[0].end; |
536 | match[1].end= (uint)(match[0].end+s_length); |
537 | match[1].mb_len= 0; /* Not computed */ |
538 | } |
539 | } |
540 | return 2; |
541 | } |
542 | mb_len= (mb_len= my_ismbchar(cs, b, end)) ? mb_len : 1; |
543 | b+= mb_len; |
544 | b_length-= mb_len; |
545 | res++; |
546 | } |
547 | } |
548 | return 0; |
549 | } |
550 | |
551 | |
552 | /* |
553 | Copy one non-ascii character. |
554 | "dst" must have enough room for the character. |
555 | Note, we don't use sort_order[] in this macros. |
556 | This is correct even for case insensitive collations: |
557 | - basic Latin letters are processed outside this macros; |
558 | - for other characters sort_order[x] is equal to x. |
559 | */ |
560 | #define my_strnxfrm_mb_non_ascii_char(cs, dst, src, se) \ |
561 | { \ |
562 | switch (my_ismbchar(cs, (const char *) src, (const char *) se)) { \ |
563 | case 4: \ |
564 | *dst++= *src++; \ |
565 | /* fall through */ \ |
566 | case 3: \ |
567 | *dst++= *src++; \ |
568 | /* fall through */ \ |
569 | case 2: \ |
570 | *dst++= *src++; \ |
571 | /* fall through */ \ |
572 | case 0: \ |
573 | *dst++= *src++; /* byte in range 0x80..0xFF which is not MB head */ \ |
574 | } \ |
575 | } |
576 | |
577 | |
578 | /* |
579 | For character sets with two or three byte multi-byte |
580 | characters having multibyte weights *equal* to their codes: |
581 | cp932, euckr, gb2312, sjis, eucjpms, ujis. |
582 | */ |
583 | size_t my_strnxfrm_mb_internal(CHARSET_INFO *cs, uchar *dst, uchar *de, |
584 | uint *nweights, const uchar *src, size_t srclen) |
585 | { |
586 | uchar *d0= dst; |
587 | const uchar *se= src + srclen; |
588 | const uchar *sort_order= cs->sort_order; |
589 | |
590 | DBUG_ASSERT(cs->mbmaxlen <= 4); |
591 | |
592 | /* |
593 | If "srclen" is smaller than both "dstlen" and "nweights" |
594 | then we can run a simplified loop - |
595 | without checking "nweights" and "de". |
596 | */ |
597 | if (de >= d0 + srclen && *nweights >= srclen) |
598 | { |
599 | if (sort_order) |
600 | { |
601 | /* Optimized version for a case insensitive collation */ |
602 | for (; src < se; (*nweights)--) |
603 | { |
604 | if (*src < 128) /* quickly catch ASCII characters */ |
605 | *dst++= sort_order[*src++]; |
606 | else |
607 | my_strnxfrm_mb_non_ascii_char(cs, dst, src, se); |
608 | } |
609 | } |
610 | else |
611 | { |
612 | /* Optimized version for a case sensitive collation (no sort_order) */ |
613 | for (; src < se; (*nweights)--) |
614 | { |
615 | if (*src < 128) /* quickly catch ASCII characters */ |
616 | *dst++= *src++; |
617 | else |
618 | my_strnxfrm_mb_non_ascii_char(cs, dst, src, se); |
619 | } |
620 | } |
621 | goto end; |
622 | } |
623 | |
624 | /* |
625 | A thourough loop, checking all possible limits: |
626 | "se", "nweights" and "de". |
627 | */ |
628 | for (; src < se && *nweights && dst < de; (*nweights)--) |
629 | { |
630 | int chlen; |
631 | if (*src < 128 || !(chlen= my_ismbchar(cs, (const char *) src, |
632 | (const char *) se))) |
633 | { |
634 | /* Single byte character */ |
635 | *dst++= sort_order ? sort_order[*src++] : *src++; |
636 | } |
637 | else |
638 | { |
639 | /* Multi-byte character */ |
640 | size_t len= (dst + chlen <= de) ? chlen : de - dst; |
641 | memcpy(dst, src, len); |
642 | dst+= len; |
643 | src+= len; |
644 | } |
645 | } |
646 | |
647 | end: |
648 | return dst - d0; |
649 | } |
650 | |
651 | |
652 | size_t |
653 | my_strnxfrm_mb(CHARSET_INFO *cs, |
654 | uchar *dst, size_t dstlen, uint nweights, |
655 | const uchar *src, size_t srclen, uint flags) |
656 | { |
657 | uchar *de= dst + dstlen; |
658 | uchar *d0= dst; |
659 | dst= d0 + my_strnxfrm_mb_internal(cs, dst, de, &nweights, src, srclen); |
660 | return my_strxfrm_pad_desc_and_reverse(cs, d0, dst, de, nweights, flags, 0); |
661 | } |
662 | |
663 | |
664 | size_t |
665 | my_strnxfrm_mb_nopad(CHARSET_INFO *cs, |
666 | uchar *dst, size_t dstlen, uint nweights, |
667 | const uchar *src, size_t srclen, uint flags) |
668 | { |
669 | uchar *de= dst + dstlen; |
670 | uchar *d0= dst; |
671 | dst= d0 + my_strnxfrm_mb_internal(cs, dst, de, &nweights, src, srclen); |
672 | return my_strxfrm_pad_desc_and_reverse_nopad(cs, d0, dst, de, nweights, |
673 | flags, 0); |
674 | } |
675 | |
676 | |
677 | int |
678 | my_strcasecmp_mb_bin(CHARSET_INFO * cs __attribute__((unused)), |
679 | const char *s, const char *t) |
680 | { |
681 | return strcmp(s,t); |
682 | } |
683 | |
684 | |
685 | |
686 | void |
687 | my_hash_sort_mb_nopad_bin(CHARSET_INFO *cs __attribute__((unused)), |
688 | const uchar *key, size_t len,ulong *nr1, ulong *nr2) |
689 | { |
690 | register ulong m1= *nr1, m2= *nr2; |
691 | const uchar *end= key + len; |
692 | for (; key < end ; key++) |
693 | { |
694 | MY_HASH_ADD(m1, m2, (uint)*key); |
695 | } |
696 | *nr1= m1; |
697 | *nr2= m2; |
698 | } |
699 | |
700 | |
701 | void |
702 | my_hash_sort_mb_bin(CHARSET_INFO *cs __attribute__((unused)), |
703 | const uchar *key, size_t len,ulong *nr1, ulong *nr2) |
704 | { |
705 | /* |
706 | Remove trailing spaces. We have to do this to be able to compare |
707 | 'A ' and 'A' as identical |
708 | */ |
709 | const uchar *end= skip_trailing_space(key, len); |
710 | my_hash_sort_mb_nopad_bin(cs, key, end - key, nr1, nr2); |
711 | } |
712 | |
713 | |
714 | /* |
715 | Fill the given buffer with 'maximum character' for given charset |
716 | SYNOPSIS |
717 | pad_max_char() |
718 | cs Character set |
719 | str Start of buffer to fill |
720 | end End of buffer to fill |
721 | |
722 | DESCRIPTION |
723 | Write max key: |
724 | - for non-Unicode character sets: |
725 | just bfill using max_sort_char if max_sort_char is one byte. |
726 | In case when max_sort_char is two bytes, fill with double-byte pairs |
727 | and optionally pad with a single space character. |
728 | - for Unicode character set (utf-8): |
729 | create a buffer with multibyte representation of the max_sort_char |
730 | character, and copy it into max_str in a loop. |
731 | */ |
732 | static void pad_max_char(CHARSET_INFO *cs, char *str, char *end) |
733 | { |
734 | char buf[10]; |
735 | char buflen= cs->cset->native_to_mb(cs, cs->max_sort_char, (uchar*) buf, |
736 | (uchar*) buf + sizeof(buf)); |
737 | DBUG_ASSERT(buflen > 0); |
738 | do |
739 | { |
740 | if ((str + buflen) <= end) |
741 | { |
742 | /* Enough space for the character */ |
743 | memcpy(str, buf, buflen); |
744 | str+= buflen; |
745 | } |
746 | else |
747 | { |
748 | /* |
749 | There is no space for whole multibyte |
750 | character, then add trailing spaces. |
751 | */ |
752 | *str++= ' '; |
753 | } |
754 | } while (str < end); |
755 | } |
756 | |
757 | /* |
758 | ** Calculate min_str and max_str that ranges a LIKE string. |
759 | ** Arguments: |
760 | ** ptr Pointer to LIKE string. |
761 | ** ptr_length Length of LIKE string. |
762 | ** escape Escape character in LIKE. (Normally '\'). |
763 | ** All escape characters should be removed from min_str and max_str |
764 | ** res_length Length of min_str and max_str. |
765 | ** min_str Smallest case sensitive string that ranges LIKE. |
766 | ** Should be space padded to res_length. |
767 | ** max_str Largest case sensitive string that ranges LIKE. |
768 | ** Normally padded with the biggest character sort value. |
769 | ** |
770 | ** The function should return 0 if ok and 1 if the LIKE string can't be |
771 | ** optimized ! |
772 | */ |
773 | |
774 | my_bool my_like_range_mb(CHARSET_INFO *cs, |
775 | const char *ptr,size_t ptr_length, |
776 | pbool escape, pbool w_one, pbool w_many, |
777 | size_t res_length, |
778 | char *min_str,char *max_str, |
779 | size_t *min_length,size_t *max_length) |
780 | { |
781 | uint mb_len; |
782 | const char *end= ptr + ptr_length; |
783 | char *min_org= min_str; |
784 | char *min_end= min_str + res_length; |
785 | char *max_end= max_str + res_length; |
786 | size_t maxcharlen= res_length / cs->mbmaxlen; |
787 | const MY_CONTRACTIONS *contractions= my_charset_get_contractions(cs, 0); |
788 | |
789 | for (; ptr != end && min_str != min_end && maxcharlen ; maxcharlen--) |
790 | { |
791 | /* We assume here that escape, w_any, w_namy are one-byte characters */ |
792 | if (*ptr == escape && ptr+1 != end) |
793 | ptr++; /* Skip escape */ |
794 | else if (*ptr == w_one || *ptr == w_many) /* '_' and '%' in SQL */ |
795 | { |
796 | fill_max_and_min: |
797 | /* |
798 | Calculate length of keys: |
799 | 'a\0\0... is the smallest possible string when we have space expand |
800 | a\ff\ff... is the biggest possible string |
801 | */ |
802 | *min_length= (cs->state & (MY_CS_BINSORT | MY_CS_NOPAD)) ? |
803 | (size_t) (min_str - min_org) : |
804 | res_length; |
805 | /* Create min key */ |
806 | do |
807 | { |
808 | *min_str++= (char) cs->min_sort_char; |
809 | } while (min_str != min_end); |
810 | |
811 | /* |
812 | Write max key: create a buffer with multibyte |
813 | representation of the max_sort_char character, |
814 | and copy it into max_str in a loop. |
815 | */ |
816 | *max_length= res_length; |
817 | pad_max_char(cs, max_str, max_end); |
818 | return 0; |
819 | } |
820 | if ((mb_len= my_ismbchar(cs, ptr, end)) > 1) |
821 | { |
822 | if (ptr+mb_len > end || min_str+mb_len > min_end) |
823 | break; |
824 | while (mb_len--) |
825 | *min_str++= *max_str++= *ptr++; |
826 | } |
827 | else |
828 | { |
829 | /* |
830 | Special case for collations with contractions. |
831 | For example, in Chezh, 'ch' is a separate letter |
832 | which is sorted between 'h' and 'i'. |
833 | If the pattern 'abc%', 'c' at the end can mean: |
834 | - letter 'c' itself, |
835 | - beginning of the contraction 'ch'. |
836 | |
837 | If we simply return this LIKE range: |
838 | |
839 | 'abc\min\min\min' and 'abc\max\max\max' |
840 | |
841 | then this query: SELECT * FROM t1 WHERE a LIKE 'abc%' |
842 | will only find values starting from 'abc[^h]', |
843 | but won't find values starting from 'abch'. |
844 | |
845 | We must ignore contraction heads followed by w_one or w_many. |
846 | ('Contraction head' means any letter which can be the first |
847 | letter in a contraction) |
848 | |
849 | For example, for Czech 'abc%', we will return LIKE range, |
850 | which is equal to LIKE range for 'ab%': |
851 | |
852 | 'ab\min\min\min\min' and 'ab\max\max\max\max'. |
853 | |
854 | */ |
855 | if (contractions && ptr + 1 < end && |
856 | my_uca_can_be_contraction_head(contractions, (uchar) *ptr)) |
857 | { |
858 | /* Ptr[0] is a contraction head. */ |
859 | |
860 | if (ptr[1] == w_one || ptr[1] == w_many) |
861 | { |
862 | /* Contraction head followed by a wildcard, quit. */ |
863 | goto fill_max_and_min; |
864 | } |
865 | |
866 | /* |
867 | Some letters can be both contraction heads and contraction tails. |
868 | For example, in Danish 'aa' is a separate single letter which |
869 | is sorted after 'z'. So 'a' can be both head and tail. |
870 | |
871 | If ptr[0]+ptr[1] is a contraction, |
872 | then put both letters together. |
873 | |
874 | If ptr[1] can be a contraction part, but ptr[0]+ptr[1] |
875 | is not a contraction, then we put only ptr[0], |
876 | and continue with ptr[1] on the next loop. |
877 | */ |
878 | if (my_uca_can_be_contraction_tail(contractions, (uchar) ptr[1]) && |
879 | my_uca_contraction2_weight(contractions, (uchar) ptr[0], ptr[1])) |
880 | { |
881 | /* Contraction found */ |
882 | if (maxcharlen == 1 || min_str + 1 >= min_end) |
883 | { |
884 | /* Both contraction parts don't fit, quit */ |
885 | goto fill_max_and_min; |
886 | } |
887 | |
888 | /* Put contraction head */ |
889 | *min_str++= *max_str++= *ptr++; |
890 | maxcharlen--; |
891 | } |
892 | } |
893 | /* Put contraction tail, or a single character */ |
894 | *min_str++= *max_str++= *ptr++; |
895 | } |
896 | } |
897 | |
898 | *min_length= *max_length = (size_t) (min_str - min_org); |
899 | while (min_str != min_end) |
900 | *min_str++= *max_str++= ' '; /* Because if key compression */ |
901 | return 0; |
902 | } |
903 | |
904 | |
905 | /** |
906 | Calculate min_str and max_str that ranges a LIKE string. |
907 | Generic function, currently used for ucs2, utf16, utf32, |
908 | but should be suitable for any other character sets with |
909 | cs->min_sort_char and cs->max_sort_char represented in |
910 | Unicode code points. |
911 | |
912 | @param cs Character set and collation pointer |
913 | @param ptr Pointer to LIKE pattern. |
914 | @param ptr_length Length of LIKE pattern. |
915 | @param escape Escape character pattern, typically '\'. |
916 | @param w_one 'One character' pattern, typically '_'. |
917 | @param w_many 'Many characters' pattern, typically '%'. |
918 | @param res_length Length of min_str and max_str. |
919 | |
920 | @param[out] min_str Smallest string that ranges LIKE. |
921 | @param[out] max_str Largest string that ranges LIKE. |
922 | @param[out] min_len Length of min_str |
923 | @param[out] max_len Length of max_str |
924 | |
925 | @return Optimization status. |
926 | @retval FALSE if LIKE pattern can be optimized |
927 | @rerval TRUE if LIKE can't be optimized. |
928 | */ |
929 | my_bool |
930 | my_like_range_generic(CHARSET_INFO *cs, |
931 | const char *ptr, size_t ptr_length, |
932 | pbool escape, pbool w_one, pbool w_many, |
933 | size_t res_length, |
934 | char *min_str,char *max_str, |
935 | size_t *min_length,size_t *max_length) |
936 | { |
937 | const char *end= ptr + ptr_length; |
938 | const char *min_org= min_str; |
939 | const char *max_org= max_str; |
940 | char *min_end= min_str + res_length; |
941 | char *max_end= max_str + res_length; |
942 | size_t charlen= res_length / cs->mbmaxlen; |
943 | size_t res_length_diff; |
944 | const MY_CONTRACTIONS *contractions= my_charset_get_contractions(cs, 0); |
945 | |
946 | for ( ; charlen > 0; charlen--) |
947 | { |
948 | my_wc_t wc, wc2; |
949 | int res; |
950 | if ((res= cs->cset->mb_wc(cs, &wc, (uchar*) ptr, (uchar*) end)) <= 0) |
951 | { |
952 | if (res == MY_CS_ILSEQ) /* Bad sequence */ |
953 | return TRUE; /* min_length and max_length are not important */ |
954 | break; /* End of the string */ |
955 | } |
956 | ptr+= res; |
957 | |
958 | if (wc == (my_wc_t) escape) |
959 | { |
960 | if ((res= cs->cset->mb_wc(cs, &wc, (uchar*) ptr, (uchar*) end)) <= 0) |
961 | { |
962 | if (res == MY_CS_ILSEQ) |
963 | return TRUE; /* min_length and max_length are not important */ |
964 | /* |
965 | End of the string: Escape is the last character. |
966 | Put escape as a normal character. |
967 | We'll will leave the loop on the next iteration. |
968 | */ |
969 | } |
970 | else |
971 | ptr+= res; |
972 | |
973 | /* Put escape character to min_str and max_str */ |
974 | if ((res= cs->cset->wc_mb(cs, wc, |
975 | (uchar*) min_str, (uchar*) min_end)) <= 0) |
976 | goto pad_set_lengths; /* No space */ |
977 | min_str+= res; |
978 | |
979 | if ((res= cs->cset->wc_mb(cs, wc, |
980 | (uchar*) max_str, (uchar*) max_end)) <= 0) |
981 | goto pad_set_lengths; /* No space */ |
982 | max_str+= res; |
983 | continue; |
984 | } |
985 | else if (wc == (my_wc_t) w_one) |
986 | { |
987 | if ((res= cs->cset->wc_mb(cs, cs->min_sort_char, |
988 | (uchar*) min_str, (uchar*) min_end)) <= 0) |
989 | goto pad_set_lengths; |
990 | min_str+= res; |
991 | |
992 | if ((res= cs->cset->wc_mb(cs, cs->max_sort_char, |
993 | (uchar*) max_str, (uchar*) max_end)) <= 0) |
994 | goto pad_set_lengths; |
995 | max_str+= res; |
996 | continue; |
997 | } |
998 | else if (wc == (my_wc_t) w_many) |
999 | { |
1000 | /* |
1001 | Calculate length of keys: |
1002 | a\min\min... is the smallest possible string |
1003 | a\max\max... is the biggest possible string |
1004 | */ |
1005 | *min_length= (cs->state & (MY_CS_BINSORT | MY_CS_NOPAD)) ? |
1006 | (size_t) (min_str - min_org) : |
1007 | res_length; |
1008 | *max_length= res_length; |
1009 | goto pad_min_max; |
1010 | } |
1011 | |
1012 | if (contractions && |
1013 | my_uca_can_be_contraction_head(contractions, wc) && |
1014 | (res= cs->cset->mb_wc(cs, &wc2, (uchar*) ptr, (uchar*) end)) > 0) |
1015 | { |
1016 | const uint16 *weight; |
1017 | if ((wc2 == (my_wc_t) w_one || wc2 == (my_wc_t) w_many)) |
1018 | { |
1019 | /* Contraction head followed by a wildcard */ |
1020 | *min_length= *max_length= res_length; |
1021 | goto pad_min_max; |
1022 | } |
1023 | |
1024 | if (my_uca_can_be_contraction_tail(contractions, wc2) && |
1025 | (weight= my_uca_contraction2_weight(contractions, wc, wc2)) && weight[0]) |
1026 | { |
1027 | /* Contraction found */ |
1028 | if (charlen == 1) |
1029 | { |
1030 | /* contraction does not fit to result */ |
1031 | *min_length= *max_length= res_length; |
1032 | goto pad_min_max; |
1033 | } |
1034 | |
1035 | ptr+= res; |
1036 | charlen--; |
1037 | |
1038 | /* Put contraction head */ |
1039 | if ((res= cs->cset->wc_mb(cs, wc, |
1040 | (uchar*) min_str, (uchar*) min_end)) <= 0) |
1041 | goto pad_set_lengths; |
1042 | min_str+= res; |
1043 | |
1044 | if ((res= cs->cset->wc_mb(cs, wc, |
1045 | (uchar*) max_str, (uchar*) max_end)) <= 0) |
1046 | goto pad_set_lengths; |
1047 | max_str+= res; |
1048 | wc= wc2; /* Prepare to put contraction tail */ |
1049 | } |
1050 | } |
1051 | |
1052 | /* Normal character, or contraction tail */ |
1053 | if ((res= cs->cset->wc_mb(cs, wc, |
1054 | (uchar*) min_str, (uchar*) min_end)) <= 0) |
1055 | goto pad_set_lengths; |
1056 | min_str+= res; |
1057 | if ((res= cs->cset->wc_mb(cs, wc, |
1058 | (uchar*) max_str, (uchar*) max_end)) <= 0) |
1059 | goto pad_set_lengths; |
1060 | max_str+= res; |
1061 | } |
1062 | |
1063 | pad_set_lengths: |
1064 | *min_length= (size_t) (min_str - min_org); |
1065 | *max_length= (size_t) (max_str - max_org); |
1066 | |
1067 | pad_min_max: |
1068 | /* |
1069 | Fill up max_str and min_str to res_length. |
1070 | fill() cannot set incomplete characters and |
1071 | requires that "length" argument is divisible to mbminlen. |
1072 | Make sure to call fill() with proper "length" argument. |
1073 | */ |
1074 | res_length_diff= res_length % cs->mbminlen; |
1075 | cs->cset->fill(cs, min_str, min_end - min_str - res_length_diff, |
1076 | cs->min_sort_char); |
1077 | cs->cset->fill(cs, max_str, max_end - max_str - res_length_diff, |
1078 | cs->max_sort_char); |
1079 | |
1080 | /* In case of incomplete characters set the remainder to 0x00's */ |
1081 | if (res_length_diff) |
1082 | { |
1083 | /* Example: odd res_length for ucs2 */ |
1084 | memset(min_end - res_length_diff, 0, res_length_diff); |
1085 | memset(max_end - res_length_diff, 0, res_length_diff); |
1086 | } |
1087 | return FALSE; |
1088 | } |
1089 | |
1090 | |
1091 | static int my_wildcmp_mb_bin_impl(CHARSET_INFO *cs, |
1092 | const char *str,const char *str_end, |
1093 | const char *wildstr,const char *wildend, |
1094 | int escape, int w_one, int w_many, int recurse_level) |
1095 | { |
1096 | int result= -1; /* Not found, using wildcards */ |
1097 | |
1098 | if (my_string_stack_guard && my_string_stack_guard(recurse_level)) |
1099 | return 1; |
1100 | while (wildstr != wildend) |
1101 | { |
1102 | while (*wildstr != w_many && *wildstr != w_one) |
1103 | { |
1104 | int l; |
1105 | if (*wildstr == escape && wildstr+1 != wildend) |
1106 | wildstr++; |
1107 | if ((l = my_ismbchar(cs, wildstr, wildend))) |
1108 | { |
1109 | if (str+l > str_end || memcmp(str, wildstr, l) != 0) |
1110 | return 1; |
1111 | str += l; |
1112 | wildstr += l; |
1113 | } |
1114 | else |
1115 | if (str == str_end || *wildstr++ != *str++) |
1116 | return(1); /* No match */ |
1117 | if (wildstr == wildend) |
1118 | return (str != str_end); /* Match if both are at end */ |
1119 | result=1; /* Found an anchor char */ |
1120 | } |
1121 | if (*wildstr == w_one) |
1122 | { |
1123 | do |
1124 | { |
1125 | if (str == str_end) /* Skip one char if possible */ |
1126 | return (result); |
1127 | INC_PTR(cs,str,str_end); |
1128 | } while (++wildstr < wildend && *wildstr == w_one); |
1129 | if (wildstr == wildend) |
1130 | break; |
1131 | } |
1132 | if (*wildstr == w_many) |
1133 | { /* Found w_many */ |
1134 | int cmp; |
1135 | const char* mb = wildstr; |
1136 | int mb_len=0; |
1137 | |
1138 | wildstr++; |
1139 | /* Remove any '%' and '_' from the wild search string */ |
1140 | for (; wildstr != wildend ; wildstr++) |
1141 | { |
1142 | if (*wildstr == w_many) |
1143 | continue; |
1144 | if (*wildstr == w_one) |
1145 | { |
1146 | if (str == str_end) |
1147 | return (-1); |
1148 | INC_PTR(cs,str,str_end); |
1149 | continue; |
1150 | } |
1151 | break; /* Not a wild character */ |
1152 | } |
1153 | if (wildstr == wildend) |
1154 | return(0); /* Ok if w_many is last */ |
1155 | if (str == str_end) |
1156 | return -1; |
1157 | |
1158 | if ((cmp= *wildstr) == escape && wildstr+1 != wildend) |
1159 | cmp= *++wildstr; |
1160 | |
1161 | mb=wildstr; |
1162 | mb_len= my_ismbchar(cs, wildstr, wildend); |
1163 | INC_PTR(cs,wildstr,wildend); /* This is compared trough cmp */ |
1164 | do |
1165 | { |
1166 | for (;;) |
1167 | { |
1168 | if (str >= str_end) |
1169 | return -1; |
1170 | if (mb_len) |
1171 | { |
1172 | if (str+mb_len <= str_end && memcmp(str, mb, mb_len) == 0) |
1173 | { |
1174 | str += mb_len; |
1175 | break; |
1176 | } |
1177 | } |
1178 | else if (!my_ismbchar(cs, str, str_end) && *str == cmp) |
1179 | { |
1180 | str++; |
1181 | break; |
1182 | } |
1183 | INC_PTR(cs,str, str_end); |
1184 | } |
1185 | { |
1186 | int tmp=my_wildcmp_mb_bin_impl(cs,str,str_end, |
1187 | wildstr,wildend,escape, |
1188 | w_one,w_many, recurse_level+1); |
1189 | if (tmp <= 0) |
1190 | return (tmp); |
1191 | } |
1192 | } while (str != str_end); |
1193 | return(-1); |
1194 | } |
1195 | } |
1196 | return (str != str_end ? 1 : 0); |
1197 | } |
1198 | |
1199 | int |
1200 | my_wildcmp_mb_bin(CHARSET_INFO *cs, |
1201 | const char *str,const char *str_end, |
1202 | const char *wildstr,const char *wildend, |
1203 | int escape, int w_one, int w_many) |
1204 | { |
1205 | return my_wildcmp_mb_bin_impl(cs, str, str_end, |
1206 | wildstr, wildend, |
1207 | escape, w_one, w_many, 1); |
1208 | } |
1209 | |
1210 | |
1211 | /* |
1212 | Data was produced from EastAsianWidth.txt |
1213 | using utt11-dump utility. |
1214 | */ |
1215 | static const char pg11[256]= |
1216 | { |
1217 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1218 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1219 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,1, |
1220 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1221 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1222 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1223 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1224 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1225 | }; |
1226 | |
1227 | static const char pg23[256]= |
1228 | { |
1229 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1230 | 0,0,0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1231 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1232 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1233 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1234 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1235 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1236 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1237 | }; |
1238 | |
1239 | static const char pg2E[256]= |
1240 | { |
1241 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1242 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1243 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1244 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1245 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1,1,1, |
1246 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1247 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1248 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0 |
1249 | }; |
1250 | |
1251 | static const char pg2F[256]= |
1252 | { |
1253 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1254 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1255 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1256 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1257 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1258 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1259 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0, |
1260 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0 |
1261 | }; |
1262 | |
1263 | static const char pg30[256]= |
1264 | { |
1265 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1266 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0, |
1267 | 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1268 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1269 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,1,1,1,1,1,1,1, |
1270 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1271 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1272 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 |
1273 | }; |
1274 | |
1275 | static const char pg31[256]= |
1276 | { |
1277 | 0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1278 | 1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1279 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1280 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1281 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1282 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0, |
1283 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1284 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 |
1285 | }; |
1286 | |
1287 | static const char pg32[256]= |
1288 | { |
1289 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0, |
1290 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1291 | 1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1292 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1, |
1293 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1294 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1295 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1296 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0 |
1297 | }; |
1298 | |
1299 | static const char pg4D[256]= |
1300 | { |
1301 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1302 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1303 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1304 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1305 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1306 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0, |
1307 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1308 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1309 | }; |
1310 | |
1311 | static const char pg9F[256]= |
1312 | { |
1313 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1314 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1315 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1316 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1317 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1318 | 1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1319 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1320 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1321 | }; |
1322 | |
1323 | static const char pgA4[256]= |
1324 | { |
1325 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1326 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1327 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1328 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1329 | 1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1330 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1331 | 1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1332 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1333 | }; |
1334 | |
1335 | static const char pgD7[256]= |
1336 | { |
1337 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1338 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1339 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1340 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1341 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1342 | 1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1343 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1344 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1345 | }; |
1346 | |
1347 | static const char pgFA[256]= |
1348 | { |
1349 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1350 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1351 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1352 | 1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1353 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1354 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1355 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1356 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1357 | }; |
1358 | |
1359 | static const char pgFE[256]= |
1360 | { |
1361 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1362 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1363 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1,1,1, |
1364 | 1,1,1,1,1,1,1,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1365 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1366 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1367 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1368 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1369 | }; |
1370 | |
1371 | static const char pgFF[256]= |
1372 | { |
1373 | 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1374 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1375 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
1376 | 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1377 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1378 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1379 | 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, |
1380 | 1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 |
1381 | }; |
1382 | |
1383 | static const struct {int page; const char *p;} utr11_data[256]= |
1384 | { |
1385 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1386 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1387 | {0,NULL},{0,pg11},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1388 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1389 | {0,NULL},{0,NULL},{0,NULL},{0,pg23},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1390 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,pg2E},{0,pg2F}, |
1391 | {0,pg30},{0,pg31},{0,pg32},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1392 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1393 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1394 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{0,pg4D},{1,NULL},{1,NULL}, |
1395 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1396 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1397 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1398 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1399 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1400 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1401 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1402 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1403 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1404 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{0,pg9F}, |
1405 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{0,pgA4},{0,NULL},{0,NULL},{0,NULL}, |
1406 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1407 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1408 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1409 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1410 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL}, |
1411 | {1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{1,NULL},{0,pgD7}, |
1412 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1413 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1414 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1415 | {0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL},{0,NULL}, |
1416 | {0,NULL},{1,NULL},{0,pgFA},{0,NULL},{0,NULL},{0,NULL},{0,pgFE},{0,pgFF} |
1417 | }; |
1418 | |
1419 | |
1420 | size_t my_numcells_mb(CHARSET_INFO *cs, const char *b, const char *e) |
1421 | { |
1422 | my_wc_t wc; |
1423 | size_t clen= 0; |
1424 | |
1425 | while (b < e) |
1426 | { |
1427 | int mb_len; |
1428 | uint pg; |
1429 | if ((mb_len= cs->cset->mb_wc(cs, &wc, (uchar*) b, (uchar*) e)) <= 0) |
1430 | { |
1431 | mb_len= 1; /* Let's think a wrong sequence takes 1 dysplay cell */ |
1432 | b++; |
1433 | continue; |
1434 | } |
1435 | b+= mb_len; |
1436 | if (wc > 0xFFFF) |
1437 | { |
1438 | if (wc >= 0x20000 && wc <= 0x3FFFD) /* CJK Ideograph Extension B, C */ |
1439 | clen+= 1; |
1440 | } |
1441 | else |
1442 | { |
1443 | pg= (wc >> 8) & 0xFF; |
1444 | clen+= utr11_data[pg].p ? utr11_data[pg].p[wc & 0xFF] : utr11_data[pg].page; |
1445 | } |
1446 | clen++; |
1447 | } |
1448 | return clen; |
1449 | } |
1450 | |
1451 | |
1452 | int my_mb_ctype_mb(CHARSET_INFO *cs, int *ctype, |
1453 | const uchar *s, const uchar *e) |
1454 | { |
1455 | my_wc_t wc; |
1456 | int res= cs->cset->mb_wc(cs, &wc, s, e); |
1457 | if (res <= 0 || wc > 0xFFFF) |
1458 | *ctype= 0; |
1459 | else |
1460 | *ctype= my_uni_ctype[wc>>8].ctype ? |
1461 | my_uni_ctype[wc>>8].ctype[wc&0xFF] : |
1462 | my_uni_ctype[wc>>8].pctype; |
1463 | return res; |
1464 | } |
1465 | |
1466 | |
1467 | #endif |
1468 | |