-
Notifications
You must be signed in to change notification settings - Fork 4k
/
Copy pathctype-mb.cc
1338 lines (1209 loc) · 52.6 KB
/
ctype-mb.cc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/* Copyright (c) 2002, 2025, Oracle and/or its affiliates.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License, version 2.0,
as published by the Free Software Foundation.
This program is designed to work with certain software (including
but not limited to OpenSSL) that is licensed under separate terms,
as designated in a particular file or component or in included license
documentation. The authors of MySQL hereby grant you an additional
permission to link the program and your derivative works with the
separately licensed software that they have either included with
the program or referenced in the documentation.
Without limiting anything contained in the foregoing, this file,
which is part of C Driver for MySQL (Connector/C), is also subject to the
Universal FOSS Exception, version 1.0, a copy of which can be found at
http://oss.oracle.com/licenses/universal-foss-exception.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License, version 2.0, for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */
#include <algorithm>
#include <cassert>
#include <cstdint>
#include <cstring>
#include "m_string.h" // skip_trailing_space
#include "mysql/strings/m_ctype.h"
#include "strings/m_ctype_internals.h"
#include "strings/str_uca_type.h"
#include "template_utils.h"
size_t my_caseup_str_mb(const CHARSET_INFO *cs, char *str) {
uint32_t l = 0;
const uint8_t *map = cs->to_upper;
char *str_orig = str;
while (*str) {
/* Pointing after the '\0' is safe here. */
if ((l = my_ismbchar(cs, str, str + cs->mbmaxlen)))
str += l;
else {
*str = (char)map[(uint8_t)*str];
str++;
}
}
return (size_t)(str - str_orig);
}
size_t my_casedn_str_mb(const CHARSET_INFO *cs, char *str) {
uint32_t l = 0;
const uint8_t *map = cs->to_lower;
char *str_orig = str;
while (*str) {
/* Pointing after the '\0' is safe here. */
if ((l = my_ismbchar(cs, str, str + cs->mbmaxlen)))
str += l;
else {
*str = (char)map[(uint8_t)*str];
str++;
}
}
return (size_t)(str - str_orig);
}
static inline const MY_UNICASE_CHARACTER *get_case_info_for_ch(
const CHARSET_INFO *cs, unsigned page, unsigned offs) {
const MY_UNICASE_CHARACTER *p;
return cs->caseinfo ? ((p = cs->caseinfo->page[page]) ? &p[offs] : nullptr)
: nullptr;
}
/*
For character sets which don't change octet length in case conversion.
*/
size_t my_caseup_mb(const CHARSET_INFO *cs, char *src, size_t srclen,
char *dst [[maybe_unused]],
size_t dstlen [[maybe_unused]]) {
uint32_t l = 0;
char *srcend = src + srclen;
const uint8_t *map = cs->to_upper;
assert(cs->caseup_multiply == 1);
assert(src == dst && srclen == dstlen);
assert(cs->mbmaxlen == 2);
while (src < srcend) {
if ((l = my_ismbchar(cs, src, srcend))) {
const MY_UNICASE_CHARACTER *ch;
if ((ch = get_case_info_for_ch(cs, (uint8_t)src[0], (uint8_t)src[1]))) {
*src++ = ch->toupper >> 8;
*src++ = ch->toupper & 0xFF;
} else
src += l;
} else {
*src = (char)map[(uint8_t)*src];
src++;
}
}
return srclen;
}
size_t my_casedn_mb(const CHARSET_INFO *cs, char *src, size_t srclen,
char *dst [[maybe_unused]],
size_t dstlen [[maybe_unused]]) {
uint32_t l = 0;
char *srcend = src + srclen;
const uint8_t *map = cs->to_lower;
assert(cs->casedn_multiply == 1);
assert(src == dst && srclen == dstlen);
assert(cs->mbmaxlen == 2);
while (src < srcend) {
if ((l = my_ismbchar(cs, src, srcend))) {
const MY_UNICASE_CHARACTER *ch;
if ((ch = get_case_info_for_ch(cs, (uint8_t)src[0], (uint8_t)src[1]))) {
*src++ = ch->tolower >> 8;
*src++ = ch->tolower & 0xFF;
} else
src += l;
} else {
*src = (char)map[(uint8_t)*src];
src++;
}
}
return srclen;
}
/*
Case folding functions for character set
where case conversion can change string octet length.
For example, in EUCKR,
_euckr 0xA9A5 == "LATIN LETTER DOTLESS I" (Turkish letter)
is upper-cased to to
_euckr 0x49 "LATIN CAPITAL LETTER I" ('usual' letter I)
Length is reduced in this example from two bytes to one byte.
*/
static size_t my_casefold_mb_varlen(const CHARSET_INFO *cs, char *src,
size_t srclen, char *dst,
size_t dstlen [[maybe_unused]],
const uint8_t *map, size_t is_upper) {
char *srcend = src + srclen, *dst0 = dst;
assert(cs->mbmaxlen == 2);
while (src < srcend) {
size_t const mblen = my_ismbchar(cs, src, srcend);
if (mblen) {
const MY_UNICASE_CHARACTER *ch;
if ((ch = get_case_info_for_ch(cs, (uint8_t)src[0], (uint8_t)src[1]))) {
int const code = is_upper ? ch->toupper : ch->tolower;
src += 2;
if (code > 0xFF) *dst++ = code >> 8;
*dst++ = code & 0xFF;
} else {
*dst++ = *src++;
*dst++ = *src++;
}
} else {
*dst++ = (char)map[(uint8_t)*src++];
}
}
return (size_t)(dst - dst0);
}
size_t my_casedn_mb_varlen(const CHARSET_INFO *cs, char *src, size_t srclen,
char *dst, size_t dstlen) {
assert(dstlen >= srclen * cs->casedn_multiply);
assert(src != dst || cs->casedn_multiply == 1);
return my_casefold_mb_varlen(cs, src, srclen, dst, dstlen, cs->to_lower, 0);
}
size_t my_caseup_mb_varlen(const CHARSET_INFO *cs, char *src, size_t srclen,
char *dst, size_t dstlen) {
assert(dstlen >= srclen * cs->caseup_multiply);
assert(src != dst || cs->caseup_multiply == 1);
return my_casefold_mb_varlen(cs, src, srclen, dst, dstlen, cs->to_upper, 1);
}
/*
my_strcasecmp_mb() returns 0 if strings are equal, non-zero otherwise.
*/
int my_strcasecmp_mb(const CHARSET_INFO *cs, const char *s, const char *t) {
uint32_t l = 0;
const uint8_t *map = cs->to_upper;
while (*s && *t) {
/* Pointing after the '\0' is safe here. */
if ((l = my_ismbchar(cs, s, s + cs->mbmaxlen))) {
while (l--)
if (*s++ != *t++) return 1;
} else if (my_mbcharlen(cs, *t) != 1 ||
map[(uint8_t)*s++] != map[(uint8_t)*t++])
return 1;
}
/* At least one of '*s' and '*t' is zero here. */
assert(!*t || !*s);
return (*t != *s);
}
/*
** Compare string against string with wildcard
** 0 if matched
** -1 if not matched with wildcard
** 1 if matched with wildcard
*/
#define INC_PTR(cs, A, B) \
A += (my_ismbchar(cs, A, B) ? my_ismbchar(cs, A, B) : 1)
#define likeconv(s, A) (uint8_t)(s)->sort_order[(uint8_t)(A)]
static int my_wildcmp_mb_impl(const CHARSET_INFO *cs, const char *str,
const char *str_end, const char *wildstr_arg,
const char *wildend_arg, int escape, int w_one,
int w_many, int recurse_level) {
int result = -1; /* Not found, using wildcards */
const auto *wildstr = pointer_cast<const uint8_t *>(wildstr_arg);
const auto *wildend = pointer_cast<const uint8_t *>(wildend_arg);
if (my_string_stack_guard && my_string_stack_guard(recurse_level)) return -1;
while (wildstr != wildend) {
while (*wildstr != w_many && *wildstr != w_one) {
int l;
if (*wildstr == escape && wildstr + 1 != wildend) wildstr++;
if ((l = my_ismbchar(cs, wildstr, wildend))) {
if (str + l > str_end || memcmp(str, wildstr, l) != 0) return 1;
str += l;
wildstr += l;
} else if (str == str_end ||
likeconv(cs, *wildstr++) != likeconv(cs, *str++))
return (1); /* No match */
if (wildstr == wildend)
return (str != str_end); /* Match if both are at end */
result = 1; /* Found an anchor char */
}
if (*wildstr == w_one) {
do {
if (str == str_end) /* Skip one char if possible */
return (result);
INC_PTR(cs, str, str_end);
} while (++wildstr < wildend && *wildstr == w_one);
if (wildstr == wildend) break;
}
if (*wildstr == w_many) { /* Found w_many */
uint8_t cmp = 0;
wildstr++;
/* Remove any '%' and '_' from the wild search string */
for (; wildstr != wildend; wildstr++) {
if (*wildstr == w_many) continue;
if (*wildstr == w_one) {
if (str == str_end) return (-1);
INC_PTR(cs, str, str_end);
continue;
}
break; /* Not a wild character */
}
if (wildstr == wildend) return (0); /* Ok if w_many is last */
if (str == str_end) return -1;
if ((cmp = *wildstr) == escape && wildstr + 1 != wildend)
cmp = *++wildstr;
const uint8_t *mb = wildstr;
unsigned const mb_len = my_ismbchar(cs, wildstr, wildend);
INC_PTR(cs, wildstr, wildend); /* This is compared through cmp */
cmp = likeconv(cs, cmp);
do {
for (;;) {
if (str >= str_end) return -1;
if (mb_len > 0) {
if (str + mb_len <= str_end && memcmp(str, mb, mb_len) == 0) {
str += mb_len;
break;
}
} else if (!my_ismbchar(cs, str, str_end) &&
likeconv(cs, *str) == cmp) {
str++;
break;
}
INC_PTR(cs, str, str_end);
}
{
int const tmp = my_wildcmp_mb_impl(
cs, str, str_end, pointer_cast<const char *>(wildstr),
wildend_arg, escape, w_one, w_many, recurse_level + 1);
if (tmp <= 0) return (tmp);
}
} while (str != str_end);
return (-1);
}
}
return (str != str_end ? 1 : 0);
}
int my_wildcmp_mb(const CHARSET_INFO *cs, const char *str, const char *str_end,
const char *wildstr, const char *wildend, int escape,
int w_one, int w_many) {
return my_wildcmp_mb_impl(cs, str, str_end, wildstr, wildend, escape, w_one,
w_many, 1);
}
size_t my_numchars_mb(const CHARSET_INFO *cs, const char *pos,
const char *end) {
size_t count = 0;
while (pos < end) {
unsigned mb_len = 0;
pos += (mb_len = my_ismbchar(cs, pos, end)) ? mb_len : 1;
count++;
}
return count;
}
size_t my_charpos_mb3(const CHARSET_INFO *cs, const char *pos, const char *end,
size_t length) {
const char *start = pos;
while (length && pos < end) {
unsigned mb_len = 0;
pos += (mb_len = my_ismbchar(cs, pos, end)) ? mb_len : 1;
length--;
}
return (size_t)(length ? end + 2 - start : pos - start);
}
size_t my_well_formed_len_mb(const CHARSET_INFO *cs, const char *b,
const char *e, size_t pos, int *error) {
const char *b_start = b;
*error = 0;
while (pos) {
my_wc_t wc;
int mb_len;
if ((mb_len = cs->cset->mb_wc(cs, &wc, pointer_cast<const uint8_t *>(b),
pointer_cast<const uint8_t *>(e))) <= 0) {
*error = b < e ? 1 : 0;
break;
}
b += mb_len;
pos--;
}
return (size_t)(b - b_start);
}
bool my_instr_mb(const CHARSET_INFO *cs, const char *b, size_t b_length,
const char *s, size_t s_length, my_match_t *match) {
const char *end, *b0;
int res = 0;
if (!s_length) {
if (match != nullptr) {
match->end = 0;
match->mb_len = 0;
}
return true; /* Empty string is always found */
}
b0 = b;
end = b + b_length;
while (b < end) {
int mb_len;
if (!cs->coll->strnncoll(cs, pointer_cast<const uint8_t *>(b), b_length,
pointer_cast<const uint8_t *>(s), s_length,
true /* t_is_prefix */)) {
if (match != nullptr) {
match->end = static_cast<unsigned>(b - b0);
match->mb_len = res;
}
return true;
}
mb_len = (mb_len = my_ismbchar(cs, b, end)) ? mb_len : 1;
b += mb_len;
b_length -= mb_len;
res++;
}
return false;
}
/* BINARY collations handlers for MB charsets */
int my_strnncoll_mb_bin(const CHARSET_INFO *cs [[maybe_unused]],
const uint8_t *s, size_t slen, const uint8_t *t,
size_t tlen, bool t_is_prefix) {
size_t const len = std::min(slen, tlen);
int const cmp = len == 0 ? 0 : memcmp(s, t, len);
return cmp ? cmp : (int)((t_is_prefix ? len : slen) - tlen);
}
/*
Compare two strings.
SYNOPSIS
my_strnncollsp_mb_bin()
cs Character set
s String to compare
slen Length of 's'
t String to compare
tlen Length of 't'
NOTE
This function is used for character strings with binary collations.
The shorter string is extended with end space to be as long as the longer
one.
RETURN
A negative number if s < t
A positive number if s > t
0 if strings are equal
*/
int my_strnncollsp_mb_bin(const CHARSET_INFO *cs [[maybe_unused]],
const uint8_t *a, size_t a_length, const uint8_t *b,
size_t b_length) {
size_t length = 0;
int res = 0;
const uint8_t *end = a + (length = std::min(a_length, b_length));
while (a < end) {
if (*a++ != *b++) return ((int)a[-1] - (int)b[-1]);
}
res = 0;
if (a_length != b_length) {
int swap = 1;
/*
Check the next not space character of the longer key. If it's < ' ',
then it's smaller than the other key.
*/
if (a_length < b_length) {
/* put shorter key in s */
a_length = b_length;
a = b;
swap = -1; /* swap sign of result */
res = -res;
}
for (end = a + a_length - length; a < end; a++) {
if (*a != ' ') return (*a < ' ') ? -swap : swap;
}
}
return res;
}
/*
Copy one non-ascii character.
"dst" must have enough room for the character.
Note, we don't use sort_order[] in this macros.
This is correct even for case insensitive collations:
- basic Latin letters are processed outside this macros;
- for other characters sort_order[x] is equal to x.
*/
#define my_strnxfrm_mb_non_ascii_char(cs, dst, src, se) \
{ \
switch (cs->cset->ismbchar(cs, (const char *)src, (const char *)se)) { \
case 4: \
*dst++ = *src++; \
[[fallthrough]]; \
case 3: \
*dst++ = *src++; \
[[fallthrough]]; \
case 2: \
*dst++ = *src++; \
[[fallthrough]]; \
case 0: \
*dst++ = *src++; /* byte in range 0x80..0xFF which is not MB head */ \
} \
}
/*
For character sets with two or three byte multi-byte
characters having multibyte weights *equal* to their codes:
cp932, euckr, gb2312, sjis, eucjpms, ujis.
*/
size_t my_strnxfrm_mb(const CHARSET_INFO *cs, uint8_t *dst, size_t dstlen,
unsigned nweights, const uint8_t *src, size_t srclen,
unsigned flags) {
uint8_t *d0 = dst;
uint8_t *de = dst + dstlen;
const uint8_t *se = src + srclen;
const uint8_t *sort_order = cs->sort_order;
assert(cs->mbmaxlen <= 4);
/*
If "srclen" is smaller than both "dstlen" and "nweights"
then we can run a simplified loop -
without checking "nweights" and "de".
*/
if (dstlen >= srclen && nweights >= srclen) {
if (sort_order) {
/* Optimized version for a case insensitive collation */
for (; src < se; nweights--) {
if (*src < 128) /* quickly catch ASCII characters */
*dst++ = sort_order[*src++];
else
my_strnxfrm_mb_non_ascii_char(cs, dst, src, se);
}
} else {
/* Optimized version for a case sensitive collation (no sort_order) */
for (; src < se; nweights--) {
if (*src < 128) /* quickly catch ASCII characters */
*dst++ = *src++;
else
my_strnxfrm_mb_non_ascii_char(cs, dst, src, se);
}
}
goto pad;
}
/*
A thorough loop, checking all possible limits:
"se", "nweights" and "de".
*/
for (; src < se && nweights && dst < de; nweights--) {
int chlen;
if (*src < 128 || !(chlen = cs->cset->ismbchar(cs, (const char *)src,
(const char *)se))) {
/* Single byte character */
*dst++ = sort_order ? sort_order[*src++] : *src++;
} else {
/* Multi-byte character */
size_t const len = (dst + chlen <= de) ? chlen : de - dst;
memcpy(dst, src, len);
dst += len;
src += len;
}
}
pad:
return my_strxfrm_pad(cs, d0, dst, de, nweights, flags);
}
int my_strcasecmp_mb_bin(const CHARSET_INFO *cs [[maybe_unused]], const char *s,
const char *t) {
return strcmp(s, t);
}
void my_hash_sort_mb_bin(const CHARSET_INFO *cs [[maybe_unused]],
const uint8_t *key, size_t len, uint64_t *nr1,
uint64_t *nr2) {
const uint8_t *pos = key;
/*
Remove trailing spaces. We have to do this to be able to compare
'A ' and 'A' as identical
*/
key = skip_trailing_space(key, len);
for (; pos < key; pos++) {
nr1[0] ^=
(uint64_t)((((unsigned)nr1[0] & 63) + nr2[0]) * ((unsigned)*pos)) +
(nr1[0] << 8);
nr2[0] += 3;
}
}
/*
Fill the given buffer with 'maximum character' for given charset
SYNOPSIS
pad_max_char()
cs Character set
str Start of buffer to fill
end End of buffer to fill
DESCRIPTION
Write max key:
- for non-Unicode character sets:
just memset using max_sort_char if max_sort_char is one byte.
In case when max_sort_char is two bytes, fill with double-byte pairs
and optionally pad with a single space character.
- for Unicode character set (utf-8):
create a buffer with multibyte representation of the max_sort_char
character, and copy it into max_str in a loop.
*/
static void pad_max_char(const CHARSET_INFO *cs, char *str, char *end) {
char buf[10];
int buflen = 0;
if (!(cs->state & MY_CS_UNICODE)) {
if (cs->max_sort_char <= 255) {
memset(str, cs->max_sort_char, end - str);
return;
}
if (cs->max_sort_char <= 0xFFFF) {
buf[0] = (char)(cs->max_sort_char >> 8);
buf[1] = cs->max_sort_char & 0xFF;
buflen = 2;
} else {
/* Currently, it's only for GB18030, so it must be a 4-byte char */
assert(cs->max_sort_char > 0xFFFFFF);
buf[0] = cs->max_sort_char >> 24 & 0xFF;
buf[1] = cs->max_sort_char >> 16 & 0xFF;
buf[2] = cs->max_sort_char >> 8 & 0xFF;
buf[3] = cs->max_sort_char & 0xFF;
buflen = 4;
}
} else {
buflen =
cs->cset->wc_mb(cs, cs->max_sort_char, pointer_cast<uint8_t *>(buf),
pointer_cast<uint8_t *>(buf) + sizeof(buf));
}
assert(buflen > 0);
do {
if ((str + buflen) <= end) {
/* Enough space for the character */
memcpy(str, buf, buflen);
str += buflen;
} else {
/*
There is no space for whole multibyte
character, then add trailing spaces.
*/
*str++ = ' ';
}
} while (str < end);
}
/*
** Calculate min_str and max_str that ranges a LIKE string.
** Arguments:
** ptr Pointer to LIKE string.
** ptr_length Length of LIKE string.
** escape Escape character in LIKE. (Normally '\').
** All escape characters should be removed from min_str and max_str
** res_length Length of min_str and max_str.
** min_str Smallest case sensitive string that ranges LIKE.
** Should be space padded to res_length.
** max_str Largest case sensitive string that ranges LIKE.
** Normally padded with the biggest character sort value.
**
** The function should return 0 if ok and 1 if the LIKE string can't be
** optimized !
*/
bool my_like_range_mb(const CHARSET_INFO *cs, const char *ptr,
size_t ptr_length, char escape, char w_one, char w_many,
size_t res_length, char *min_str, char *max_str,
size_t *min_length, size_t *max_length) {
unsigned mb_len = 0;
const char *end = ptr + ptr_length;
char *min_org = min_str;
char *min_end = min_str + res_length;
char *max_end = max_str + res_length;
size_t maxcharlen = res_length / cs->mbmaxlen;
for (; ptr != end && min_str != min_end && maxcharlen; maxcharlen--) {
/* We assume here that escape, w_any, w_namy are one-byte characters */
if (*ptr == escape && ptr + 1 != end)
ptr++; /* Skip escape */
else if (*ptr == w_one || *ptr == w_many) /* '_' and '%' in SQL */
{
fill_max_and_min:
/*
For LIKE 'a%', assuming min_sort_char='\0' and max_sort_char='\xff':
"a" is the smallest possible string for NO PAD.
"a\0\0..." is the smallest possible string for PAD SPACE.
"a\xff\xff..." is the biggest possible string.
*/
if ((cs->state & MY_CS_BINSORT) || cs->pad_attribute == NO_PAD) {
*min_length = static_cast<size_t>(min_str - min_org);
/*
Pad with spaces, because for CHAR searches, our returned min_length
is ignored and min_str is put directly into the value to search for.
*/
do {
*min_str++ = ' ';
} while (min_str != min_end);
} else {
*min_length = res_length;
do {
*min_str++ = static_cast<char>(cs->min_sort_char);
} while (min_str != min_end);
}
*max_length = res_length;
pad_max_char(cs, max_str, max_end);
return false;
}
if ((mb_len = my_ismbchar(cs, ptr, end)) > 1) {
if (ptr + mb_len > end || min_str + mb_len > min_end) break;
while (mb_len--) *min_str++ = *max_str++ = *ptr++;
} else {
/*
Special case for collations with contractions.
For example, in Chezh, 'ch' is a separate letter
which is sorted between 'h' and 'i'.
If the pattern 'abc%', 'c' at the end can mean:
- letter 'c' itself,
- beginning of the contraction 'ch'.
If we simply return this LIKE range:
'abc\min\min\min' and 'abc\max\max\max'
then this query: SELECT * FROM t1 WHERE a LIKE 'abc%'
will only find values starting from 'abc[^h]',
but won't find values starting from 'abch'.
We must ignore contraction heads followed by w_one or w_many.
('Contraction head' means any letter which can be the first
letter in a contraction)
For example, for Czech 'abc%', we will return LIKE range,
which is equal to LIKE range for 'ab%':
'ab\min\min\min\min' and 'ab\max\max\max\max'.
*/
const MY_UCA_INFO::flags_type *contraction_flags = nullptr;
if (cs->uca) contraction_flags = cs->uca->contraction_flags;
if (contraction_flags && ptr + 1 < end &&
my_uca_can_be_contraction_head(contraction_flags, (uint8_t)*ptr)) {
/* Ptr[0] is a contraction head. */
if (ptr[1] == w_one || ptr[1] == w_many) {
/* Contraction head followed by a wildcard, quit. */
goto fill_max_and_min;
}
/*
Some letters can be both contraction heads and contraction tails.
For example, in Danish 'aa' is a separate single letter which
is sorted after 'z'. So 'a' can be both head and tail.
If ptr[0]+ptr[1] is a contraction,
then put both letters together.
If ptr[1] can be a contraction part, but ptr[0]+ptr[1]
is not a contraction, then we put only ptr[0],
and continue with ptr[1] on the next loop.
*/
if (my_uca_can_be_contraction_tail(contraction_flags,
(uint8_t)ptr[1]) &&
my_uca_contraction2_weight(cs->uca->contraction_nodes,
(uint8_t)ptr[0], ptr[1])) {
/* Contraction found */
if (maxcharlen == 1 || min_str + 1 >= min_end) {
/* Both contraction parts don't fit, quit */
goto fill_max_and_min;
}
/* Put contraction head */
*min_str++ = *max_str++ = *ptr++;
maxcharlen--;
}
}
/* Put contraction tail, or a single character */
*min_str++ = *max_str++ = *ptr++;
}
}
*min_length = *max_length = (size_t)(min_str - min_org);
while (min_str != min_end)
*min_str++ = *max_str++ = ' '; /* Because if key compression */
return false;
}
/**
Calculate min_str and max_str that ranges a LIKE string.
Generic function, currently used for ucs2, utf16, utf32,
but should be suitable for any other character sets with
cs->min_sort_char and cs->max_sort_char represented in
Unicode code points.
@param cs Character set and collation pointer
@param ptr Pointer to LIKE pattern.
@param ptr_length Length of LIKE pattern.
@param escape Escape character pattern, typically '\'.
@param w_one 'One character' pattern, typically '_'.
@param w_many 'Many characters' pattern, typically '%'.
@param res_length Length of min_str and max_str.
@param[out] min_str Smallest string that ranges LIKE.
@param[out] max_str Largest string that ranges LIKE.
@param[out] min_length Length of min_str
@param[out] max_length Length of max_str
@return Optimization status.
@retval false if LIKE pattern can be optimized
@retval true if LIKE can't be optimized.
*/
bool my_like_range_generic(const CHARSET_INFO *cs, const char *ptr,
size_t ptr_length, char escape, char w_one,
char w_many, size_t res_length, char *min_str,
char *max_str, size_t *min_length,
size_t *max_length) {
const char *end = ptr + ptr_length;
const char *min_org = min_str;
const char *max_org = max_str;
char *min_end = min_str + res_length;
char *max_end = max_str + res_length;
size_t charlen = res_length / cs->mbmaxlen;
size_t res_length_diff;
for (; charlen > 0; charlen--) {
my_wc_t wc, wc2;
int res;
if ((res = cs->cset->mb_wc(cs, &wc, pointer_cast<const uint8_t *>(ptr),
pointer_cast<const uint8_t *>(end))) <= 0) {
if (res == MY_CS_ILSEQ) /* Bad sequence */
return true; /* min_length and max_length are not important */
break; /* End of the string */
}
ptr += res;
if (wc == (my_wc_t)escape) {
if ((res = cs->cset->mb_wc(cs, &wc, pointer_cast<const uint8_t *>(ptr),
pointer_cast<const uint8_t *>(end))) <= 0) {
if (res == MY_CS_ILSEQ)
return true; /* min_length and max_length are not important */
/*
End of the string: Escape is the last character.
Put escape as a normal character.
We'll will leave the loop on the next iteration.
*/
} else
ptr += res;
/* Put escape character to min_str and max_str */
if ((res = cs->cset->wc_mb(cs, wc, pointer_cast<uint8_t *>(min_str),
pointer_cast<uint8_t *>(min_end))) <= 0)
goto pad_set_lengths; /* No space */
min_str += res;
if ((res = cs->cset->wc_mb(cs, wc, pointer_cast<uint8_t *>(max_str),
pointer_cast<uint8_t *>(max_end))) <= 0)
goto pad_set_lengths; /* No space */
max_str += res;
continue;
}
if (wc == (my_wc_t)w_one) {
if ((res = cs->cset->wc_mb(cs, cs->min_sort_char,
pointer_cast<uint8_t *>(min_str),
pointer_cast<uint8_t *>(min_end))) <= 0)
goto pad_set_lengths;
min_str += res;
if ((res = cs->cset->wc_mb(cs, cs->max_sort_char,
pointer_cast<uint8_t *>(max_str),
pointer_cast<uint8_t *>(max_end))) <= 0)
goto pad_set_lengths;
max_str += res;
continue;
}
if (wc == (my_wc_t)w_many) {
/*
Calculate length of keys:
a\min\min... is the smallest possible string
a\max\max... is the biggest possible string
*/
*min_length = ((cs->state & MY_CS_BINSORT) ? (size_t)(min_str - min_org)
: res_length);
*max_length = res_length;
goto pad_min_max;
}
const MY_UCA_INFO::flags_type *contraction_flags = nullptr;
if (cs->uca) contraction_flags = cs->uca->contraction_flags;
if (contraction_flags &&
my_uca_can_be_contraction_head(contraction_flags, wc) &&
(res = cs->cset->mb_wc(cs, &wc2, pointer_cast<const uint8_t *>(ptr),
pointer_cast<const uint8_t *>(end))) > 0) {
const uint16_t *weight = nullptr;
if ((wc2 == (my_wc_t)w_one || wc2 == (my_wc_t)w_many)) {
/* Contraction head followed by a wildcard */
*min_length = *max_length = res_length;
goto pad_min_max;
}
if (my_uca_can_be_contraction_tail(contraction_flags, wc2) &&
(weight = my_uca_contraction2_weight(cs->uca->contraction_nodes, wc,
wc2)) &&
weight[0]) {
/* Contraction found */
if (charlen == 1) {
/* contraction does not fit to result */
*min_length = *max_length = res_length;
goto pad_min_max;
}
ptr += res;
charlen--;
/* Put contraction head */
if ((res = cs->cset->wc_mb(cs, wc, pointer_cast<uint8_t *>(min_str),
pointer_cast<uint8_t *>(min_end))) <= 0)
goto pad_set_lengths;
min_str += res;
if ((res = cs->cset->wc_mb(cs, wc, pointer_cast<uint8_t *>(max_str),
pointer_cast<uint8_t *>(max_end))) <= 0)
goto pad_set_lengths;
max_str += res;
wc = wc2; /* Prepare to put contraction tail */
}
}
/* Normal character, or contraction tail */
if ((res = cs->cset->wc_mb(cs, wc, pointer_cast<uint8_t *>(min_str),
pointer_cast<uint8_t *>(min_end))) <= 0)
goto pad_set_lengths;
min_str += res;
if ((res = cs->cset->wc_mb(cs, wc, pointer_cast<uint8_t *>(max_str),
pointer_cast<uint8_t *>(max_end))) <= 0)
goto pad_set_lengths;
max_str += res;
}
pad_set_lengths:
*min_length = (size_t)(min_str - min_org);
*max_length = (size_t)(max_str - max_org);
pad_min_max:
/*
Fill up max_str and min_str to res_length.
fill() cannot set incomplete characters and
requires that "length" argument is divisible to mbminlen.
Make sure to call fill() with proper "length" argument.
*/
res_length_diff = res_length % cs->mbminlen;
cs->cset->fill(cs, min_str, min_end - min_str - res_length_diff,
cs->min_sort_char);
cs->cset->fill(cs, max_str, max_end - max_str - res_length_diff,
cs->max_sort_char);
/* In case of incomplete characters set the remainder to 0x00's */
if (res_length_diff) {
/* Example: odd res_length for ucs2 */
memset(min_end - res_length_diff, 0, res_length_diff);
memset(max_end - res_length_diff, 0, res_length_diff);
}
return false;
}
static int my_wildcmp_mb_bin_impl(const CHARSET_INFO *cs, const char *str,
const char *str_end, const char *wildstr_arg,
const char *wildend_arg, int escape,
int w_one, int w_many, int recurse_level) {
int result = -1; /* Not found, using wildcards */
const auto *wildstr = pointer_cast<const uint8_t *>(wildstr_arg);
const auto *wildend = pointer_cast<const uint8_t *>(wildend_arg);
if (my_string_stack_guard && my_string_stack_guard(recurse_level)) return -1;
while (wildstr != wildend) {
while (*wildstr != w_many && *wildstr != w_one) {
int l;
if (*wildstr == escape && wildstr + 1 != wildend) wildstr++;
if ((l = my_ismbchar(cs, wildstr, wildend))) {
if (str + l > str_end || memcmp(str, wildstr, l) != 0) return 1;
str += l;
wildstr += l;
} else if (str == str_end || *wildstr++ != static_cast<uint8_t>(*str++))
return (1); /* No match */
if (wildstr == wildend)
return (str != str_end); /* Match if both are at end */
result = 1; /* Found an anchor char */
}
if (*wildstr == w_one) {
do {
if (str == str_end) /* Skip one char if possible */
return (result);
INC_PTR(cs, str, str_end);
} while (++wildstr < wildend && *wildstr == w_one);
if (wildstr == wildend) break;
}
if (*wildstr == w_many) { /* Found w_many */
int cmp;
int mb_len = 0;
wildstr++;
/* Remove any '%' and '_' from the wild search string */
for (; wildstr != wildend; wildstr++) {
if (*wildstr == w_many) continue;
if (*wildstr == w_one) {
if (str == str_end) return (-1);
INC_PTR(cs, str, str_end);
continue;
}
break; /* Not a wild character */
}
if (wildstr == wildend) return (0); /* Ok if w_many is last */
if (str == str_end) return -1;
if ((cmp = *wildstr) == escape && wildstr + 1 != wildend)
cmp = *++wildstr;