1 | /* Copyright (c) 2000, 2013, Oracle and/or its affiliates. |
2 | |
3 | This program is free software; you can redistribute it and/or modify |
4 | it under the terms of the GNU General Public License as published by |
5 | the Free Software Foundation; version 2 of the License. |
6 | |
7 | This program is distributed in the hope that it will be useful, |
8 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
9 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
10 | GNU General Public License for more details. |
11 | |
12 | You should have received a copy of the GNU General Public License |
13 | along with this program; if not, write to the Free Software |
14 | Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ |
15 | |
16 | /* |
17 | A better inplementation of the UNIX ctype(3) library. |
18 | */ |
19 | |
20 | #ifndef _m_ctype_h |
21 | #define _m_ctype_h |
22 | |
23 | #include <my_attribute.h> |
24 | |
25 | enum loglevel { |
26 | ERROR_LEVEL= 0, |
27 | WARNING_LEVEL= 1, |
28 | INFORMATION_LEVEL= 2 |
29 | }; |
30 | |
31 | #ifdef __cplusplus |
32 | extern "C" { |
33 | #endif |
34 | |
35 | #define MY_CS_NAME_SIZE 32 |
36 | #define MY_CS_CTYPE_TABLE_SIZE 257 |
37 | #define MY_CS_TO_LOWER_TABLE_SIZE 256 |
38 | #define MY_CS_TO_UPPER_TABLE_SIZE 256 |
39 | #define MY_CS_SORT_ORDER_TABLE_SIZE 256 |
40 | #define MY_CS_TO_UNI_TABLE_SIZE 256 |
41 | |
42 | #define CHARSET_DIR "charsets/" |
43 | |
44 | #define my_wc_t ulong |
45 | |
46 | #define MY_CS_REPLACEMENT_CHARACTER 0xFFFD |
47 | |
48 | /* |
49 | On i386 we store Unicode->CS conversion tables for |
50 | some character sets using Big-endian order, |
51 | to copy two bytes at once. |
52 | This gives some performance improvement. |
53 | */ |
54 | #ifdef __i386__ |
55 | #define MB2(x) (((x) >> 8) + (((x) & 0xFF) << 8)) |
56 | #define MY_PUT_MB2(s, code) { *((uint16*)(s))= (code); } |
57 | #else |
58 | #define MB2(x) (x) |
59 | #define MY_PUT_MB2(s, code) { (s)[0]= code >> 8; (s)[1]= code & 0xFF; } |
60 | #endif |
61 | |
62 | typedef const struct my_charset_handler_st MY_CHARSET_HANDLER; |
63 | typedef const struct my_collation_handler_st MY_COLLATION_HANDLER; |
64 | |
65 | typedef const struct unicase_info_st MY_UNICASE_INFO; |
66 | typedef const struct uni_ctype_st MY_UNI_CTYPE; |
67 | typedef const struct my_uni_idx_st MY_UNI_IDX; |
68 | |
69 | typedef struct unicase_info_char_st |
70 | { |
71 | uint32 toupper; |
72 | uint32 tolower; |
73 | uint32 sort; |
74 | } MY_UNICASE_CHARACTER; |
75 | |
76 | |
77 | struct unicase_info_st |
78 | { |
79 | my_wc_t maxchar; |
80 | MY_UNICASE_CHARACTER **page; |
81 | }; |
82 | |
83 | |
84 | extern MY_UNICASE_INFO my_unicase_default; |
85 | extern MY_UNICASE_INFO my_unicase_turkish; |
86 | extern MY_UNICASE_INFO my_unicase_mysql500; |
87 | extern MY_UNICASE_INFO my_unicase_unicode520; |
88 | |
89 | #define MY_UCA_MAX_CONTRACTION 6 |
90 | /* |
91 | The DUCET tables in ctype-uca.c are dumped with a limit of 8 weights |
92 | per character. cs->strxfrm_multiply is set to 8 for all UCA based collations. |
93 | |
94 | In language-specific UCA collations (with tailorings) we also do not allow |
95 | a single character to have more than 8 weights to stay with the same |
96 | strxfrm_multiply limit. Note, contractions are allowed to have twice longer |
97 | weight strings (up to 16 weights). As a contraction consists of at |
98 | least 2 characters, this makes sure that strxfrm_multiply ratio of 8 |
99 | is respected. |
100 | */ |
101 | #define MY_UCA_MAX_WEIGHT_SIZE (8+1) /* Including 0 terminator */ |
102 | #define MY_UCA_CONTRACTION_MAX_WEIGHT_SIZE (2*8+1) /* Including 0 terminator */ |
103 | #define MY_UCA_WEIGHT_LEVELS 2 |
104 | |
105 | typedef struct my_contraction_t |
106 | { |
107 | my_wc_t ch[MY_UCA_MAX_CONTRACTION]; /* Character sequence */ |
108 | uint16 weight[MY_UCA_CONTRACTION_MAX_WEIGHT_SIZE];/* Its weight string, 0-terminated */ |
109 | my_bool with_context; |
110 | } MY_CONTRACTION; |
111 | |
112 | |
113 | typedef struct my_contraction_list_t |
114 | { |
115 | size_t nitems; /* Number of items in the list */ |
116 | MY_CONTRACTION *item; /* List of contractions */ |
117 | char *flags; /* Character flags, e.g. "is contraction head") */ |
118 | } MY_CONTRACTIONS; |
119 | |
120 | my_bool my_uca_can_be_contraction_head(const MY_CONTRACTIONS *c, my_wc_t wc); |
121 | my_bool my_uca_can_be_contraction_tail(const MY_CONTRACTIONS *c, my_wc_t wc); |
122 | uint16 *my_uca_contraction2_weight(const MY_CONTRACTIONS *c, |
123 | my_wc_t wc1, my_wc_t wc2); |
124 | |
125 | |
126 | /* Collation weights on a single level (e.g. primary, secondary, tertiarty) */ |
127 | typedef struct my_uca_level_info_st |
128 | { |
129 | my_wc_t maxchar; |
130 | uchar *lengths; |
131 | uint16 **weights; |
132 | MY_CONTRACTIONS contractions; |
133 | uint levelno; |
134 | } MY_UCA_WEIGHT_LEVEL; |
135 | |
136 | |
137 | typedef struct uca_info_st |
138 | { |
139 | MY_UCA_WEIGHT_LEVEL level[MY_UCA_WEIGHT_LEVELS]; |
140 | |
141 | /* Logical positions */ |
142 | my_wc_t first_non_ignorable; |
143 | my_wc_t last_non_ignorable; |
144 | my_wc_t first_primary_ignorable; |
145 | my_wc_t last_primary_ignorable; |
146 | my_wc_t first_secondary_ignorable; |
147 | my_wc_t last_secondary_ignorable; |
148 | my_wc_t first_tertiary_ignorable; |
149 | my_wc_t last_tertiary_ignorable; |
150 | my_wc_t first_trailing; |
151 | my_wc_t last_trailing; |
152 | my_wc_t first_variable; |
153 | my_wc_t last_variable; |
154 | |
155 | } MY_UCA_INFO; |
156 | |
157 | |
158 | |
159 | extern MY_UCA_INFO my_uca_v400; |
160 | |
161 | |
162 | struct uni_ctype_st |
163 | { |
164 | uchar pctype; |
165 | const uchar *ctype; |
166 | }; |
167 | |
168 | extern MY_UNI_CTYPE my_uni_ctype[256]; |
169 | |
170 | /* wm_wc and wc_mb return codes */ |
171 | #define MY_CS_ILSEQ 0 /* Wrong by sequence: wb_wc */ |
172 | #define MY_CS_ILUNI 0 /* Cannot encode Unicode to charset: wc_mb */ |
173 | #define MY_CS_TOOSMALL -101 /* Need at least one byte: wc_mb and mb_wc */ |
174 | #define MY_CS_TOOSMALL2 -102 /* Need at least two bytes: wc_mb and mb_wc */ |
175 | #define MY_CS_TOOSMALL3 -103 /* Need at least three bytes: wc_mb and mb_wc */ |
176 | /* These following three are currently not really used */ |
177 | #define MY_CS_TOOSMALL4 -104 /* Need at least 4 bytes: wc_mb and mb_wc */ |
178 | #define MY_CS_TOOSMALL5 -105 /* Need at least 5 bytes: wc_mb and mb_wc */ |
179 | #define MY_CS_TOOSMALL6 -106 /* Need at least 6 bytes: wc_mb and mb_wc */ |
180 | /* A helper macros for "need at least n bytes" */ |
181 | #define MY_CS_TOOSMALLN(n) (-100-(n)) |
182 | |
183 | #define MY_CS_MBMAXLEN 6 /* Maximum supported mbmaxlen */ |
184 | #define MY_CS_IS_TOOSMALL(rc) ((rc) >= MY_CS_TOOSMALL6 && (rc) <= MY_CS_TOOSMALL) |
185 | |
186 | #define MY_SEQ_INTTAIL 1 |
187 | #define MY_SEQ_SPACES 2 |
188 | #define MY_SEQ_NONSPACES 3 /* Skip non-space characters, including bad bytes */ |
189 | |
190 | /* My charsets_list flags */ |
191 | #define MY_CS_COMPILED 1 /* compiled-in sets */ |
192 | #define MY_CS_CONFIG 2 /* sets that have a *.conf file */ |
193 | #define MY_CS_INDEX 4 /* sets listed in the Index file */ |
194 | #define MY_CS_LOADED 8 /* sets that are currently loaded */ |
195 | #define MY_CS_BINSORT 16 /* if binary sort order */ |
196 | #define MY_CS_PRIMARY 32 /* if primary collation */ |
197 | #define MY_CS_STRNXFRM 64 /* if strnxfrm is used for sort */ |
198 | #define MY_CS_UNICODE 128 /* is a charset is BMP Unicode */ |
199 | #define MY_CS_READY 256 /* if a charset is initialized */ |
200 | #define MY_CS_AVAILABLE 512 /* If either compiled-in or loaded*/ |
201 | #define MY_CS_CSSORT 1024 /* if case sensitive sort order */ |
202 | #define MY_CS_HIDDEN 2048 /* don't display in SHOW */ |
203 | #define MY_CS_PUREASCII 4096 /* if a charset is pure ascii */ |
204 | #define MY_CS_NONASCII 8192 /* if not ASCII-compatible */ |
205 | #define MY_CS_UNICODE_SUPPLEMENT 16384 /* Non-BMP Unicode characters */ |
206 | #define MY_CS_LOWER_SORT 32768 /* If use lower case as weight */ |
207 | #define MY_CS_STRNXFRM_BAD_NWEIGHTS 0x10000 /* strnxfrm ignores "nweights" */ |
208 | #define MY_CS_NOPAD 0x20000 /* if does not ignore trailing spaces */ |
209 | #define MY_CS_NON1TO1 0x40000 /* Has a complex mapping from characters |
210 | to weights, e.g. contractions, expansions, |
211 | ignorable characters */ |
212 | #define MY_CHARSET_UNDEFINED 0 |
213 | |
214 | /* Character repertoire flags */ |
215 | #define MY_REPERTOIRE_ASCII 1 /* Pure ASCII U+0000..U+007F */ |
216 | #define MY_REPERTOIRE_EXTENDED 2 /* Extended characters: U+0080..U+FFFF */ |
217 | #define MY_REPERTOIRE_UNICODE30 3 /* ASCII | EXTENDED: U+0000..U+FFFF */ |
218 | |
219 | /* Flags for strxfrm */ |
220 | #define MY_STRXFRM_LEVEL1 0x00000001 /* for primary weights */ |
221 | #define MY_STRXFRM_LEVEL2 0x00000002 /* for secondary weights */ |
222 | #define MY_STRXFRM_LEVEL3 0x00000004 /* for tertiary weights */ |
223 | #define MY_STRXFRM_LEVEL4 0x00000008 /* fourth level weights */ |
224 | #define MY_STRXFRM_LEVEL5 0x00000010 /* fifth level weights */ |
225 | #define MY_STRXFRM_LEVEL6 0x00000020 /* sixth level weights */ |
226 | #define MY_STRXFRM_LEVEL_ALL 0x0000003F /* Bit OR for the above six */ |
227 | #define MY_STRXFRM_NLEVELS 6 /* Number of possible levels*/ |
228 | |
229 | #define MY_STRXFRM_PAD_WITH_SPACE 0x00000040 /* if pad result with spaces */ |
230 | #define MY_STRXFRM_PAD_TO_MAXLEN 0x00000080 /* if pad tail(for filesort) */ |
231 | |
232 | #define MY_STRXFRM_DESC_LEVEL1 0x00000100 /* if desc order for level1 */ |
233 | #define MY_STRXFRM_DESC_LEVEL2 0x00000200 /* if desc order for level2 */ |
234 | #define MY_STRXFRM_DESC_LEVEL3 0x00000300 /* if desc order for level3 */ |
235 | #define MY_STRXFRM_DESC_LEVEL4 0x00000800 /* if desc order for level4 */ |
236 | #define MY_STRXFRM_DESC_LEVEL5 0x00001000 /* if desc order for level5 */ |
237 | #define MY_STRXFRM_DESC_LEVEL6 0x00002000 /* if desc order for level6 */ |
238 | #define MY_STRXFRM_DESC_SHIFT 8 |
239 | |
240 | #define MY_STRXFRM_UNUSED_00004000 0x00004000 /* for future extensions */ |
241 | #define MY_STRXFRM_UNUSED_00008000 0x00008000 /* for future extensions */ |
242 | |
243 | #define MY_STRXFRM_REVERSE_LEVEL1 0x00010000 /* if reverse order for level1 */ |
244 | #define MY_STRXFRM_REVERSE_LEVEL2 0x00020000 /* if reverse order for level2 */ |
245 | #define MY_STRXFRM_REVERSE_LEVEL3 0x00040000 /* if reverse order for level3 */ |
246 | #define MY_STRXFRM_REVERSE_LEVEL4 0x00080000 /* if reverse order for level4 */ |
247 | #define MY_STRXFRM_REVERSE_LEVEL5 0x00100000 /* if reverse order for level5 */ |
248 | #define MY_STRXFRM_REVERSE_LEVEL6 0x00200000 /* if reverse order for level6 */ |
249 | #define MY_STRXFRM_REVERSE_SHIFT 16 |
250 | |
251 | /* |
252 | Collation IDs for MariaDB that should not conflict with MySQL. |
253 | We reserve 256..511, because MySQL will most likely use this range |
254 | when the range 0..255 is full. |
255 | |
256 | We use the next 256 IDs starting from 512 and divide |
257 | them into 8 chunks, 32 collations each, as follows: |
258 | |
259 | 512 + (0..31) for single byte collations (e.g. latin9) |
260 | 512 + (32..63) reserved (e.g. for utf32le, or more single byte collations) |
261 | 512 + (64..95) for utf8 |
262 | 512 + (96..127) for utf8mb4 |
263 | 512 + (128..159) for ucs2 |
264 | 512 + (160..192) for utf16 |
265 | 512 + (192..223) for utf16le |
266 | 512 + (224..255) for utf32 |
267 | */ |
268 | #define MY_PAGE2_COLLATION_ID_8BIT 0x200 |
269 | #define MY_PAGE2_COLLATION_ID_RESERVED 0x220 |
270 | #define MY_PAGE2_COLLATION_ID_UTF8 0x240 |
271 | #define MY_PAGE2_COLLATION_ID_UTF8MB4 0x260 |
272 | #define MY_PAGE2_COLLATION_ID_UCS2 0x280 |
273 | #define MY_PAGE2_COLLATION_ID_UTF16 0x2A0 |
274 | #define MY_PAGE2_COLLATION_ID_UTF16LE 0x2C0 |
275 | #define MY_PAGE2_COLLATION_ID_UTF32 0x2E0 |
276 | |
277 | struct my_uni_idx_st |
278 | { |
279 | uint16 from; |
280 | uint16 to; |
281 | const uchar *tab; |
282 | }; |
283 | |
284 | typedef struct |
285 | { |
286 | uint beg; |
287 | uint end; |
288 | uint mb_len; |
289 | } my_match_t; |
290 | |
291 | enum my_lex_states |
292 | { |
293 | MY_LEX_START, MY_LEX_CHAR, MY_LEX_IDENT, |
294 | MY_LEX_IDENT_SEP, MY_LEX_IDENT_START, |
295 | MY_LEX_REAL, MY_LEX_HEX_NUMBER, MY_LEX_BIN_NUMBER, |
296 | MY_LEX_CMP_OP, MY_LEX_LONG_CMP_OP, MY_LEX_STRING, , MY_LEX_END, |
297 | MY_LEX_OPERATOR_OR_IDENT, MY_LEX_NUMBER_IDENT, MY_LEX_INT_OR_REAL, |
298 | MY_LEX_REAL_OR_POINT, MY_LEX_BOOL, MY_LEX_EOL, MY_LEX_ESCAPE, |
299 | , , MY_LEX_SEMICOLON, |
300 | MY_LEX_SET_VAR, MY_LEX_USER_END, MY_LEX_HOSTNAME, MY_LEX_SKIP, |
301 | MY_LEX_USER_VARIABLE_DELIMITER, MY_LEX_SYSTEM_VAR, |
302 | MY_LEX_IDENT_OR_KEYWORD, |
303 | MY_LEX_IDENT_OR_HEX, MY_LEX_IDENT_OR_BIN, MY_LEX_IDENT_OR_NCHAR, |
304 | MY_LEX_STRING_OR_DELIMITER, , MY_LEX_PLACEHOLDER, |
305 | MY_LEX_COMMA |
306 | }; |
307 | |
308 | struct charset_info_st; |
309 | |
310 | typedef struct my_charset_loader_st |
311 | { |
312 | char error[128]; |
313 | void *(*once_alloc)(size_t); |
314 | void *(*malloc)(size_t); |
315 | void *(*realloc)(void *, size_t); |
316 | void (*free)(void *); |
317 | void (*reporter)(enum loglevel, const char *format, ...); |
318 | int (*add_collation)(struct charset_info_st *cs); |
319 | } MY_CHARSET_LOADER; |
320 | |
321 | |
322 | extern int (*my_string_stack_guard)(int); |
323 | |
324 | /* See strings/CHARSET_INFO.txt for information about this structure */ |
325 | struct my_collation_handler_st |
326 | { |
327 | my_bool (*init)(struct charset_info_st *, MY_CHARSET_LOADER *); |
328 | /* Collation routines */ |
329 | int (*strnncoll)(CHARSET_INFO *, |
330 | const uchar *, size_t, const uchar *, size_t, my_bool); |
331 | int (*strnncollsp)(CHARSET_INFO *, |
332 | const uchar *, size_t, const uchar *, size_t); |
333 | size_t (*strnxfrm)(CHARSET_INFO *, |
334 | uchar *dst, size_t dstlen, uint nweights, |
335 | const uchar *src, size_t srclen, uint flags); |
336 | size_t (*strnxfrmlen)(CHARSET_INFO *, size_t); |
337 | my_bool (*like_range)(CHARSET_INFO *, |
338 | const char *s, size_t s_length, |
339 | pchar w_prefix, pchar w_one, pchar w_many, |
340 | size_t res_length, |
341 | char *min_str, char *max_str, |
342 | size_t *min_len, size_t *max_len); |
343 | int (*wildcmp)(CHARSET_INFO *, |
344 | const char *str,const char *str_end, |
345 | const char *wildstr,const char *wildend, |
346 | int escape,int w_one, int w_many); |
347 | |
348 | int (*strcasecmp)(CHARSET_INFO *, const char *, const char *); |
349 | |
350 | uint (*instr)(CHARSET_INFO *, |
351 | const char *b, size_t b_length, |
352 | const char *s, size_t s_length, |
353 | my_match_t *match, uint nmatch); |
354 | |
355 | /* Hash calculation */ |
356 | void (*hash_sort)(CHARSET_INFO *cs, const uchar *key, size_t len, |
357 | ulong *nr1, ulong *nr2); |
358 | my_bool (*propagate)(CHARSET_INFO *cs, const uchar *str, size_t len); |
359 | }; |
360 | |
361 | extern MY_COLLATION_HANDLER my_collation_8bit_bin_handler; |
362 | extern MY_COLLATION_HANDLER my_collation_8bit_simple_ci_handler; |
363 | extern MY_COLLATION_HANDLER my_collation_8bit_nopad_bin_handler; |
364 | extern MY_COLLATION_HANDLER my_collation_8bit_simple_nopad_ci_handler; |
365 | extern MY_COLLATION_HANDLER my_collation_ucs2_uca_handler; |
366 | |
367 | /* Some typedef to make it easy for C++ to make function pointers */ |
368 | typedef int (*my_charset_conv_mb_wc)(CHARSET_INFO *, my_wc_t *, |
369 | const uchar *, const uchar *); |
370 | typedef int (*my_charset_conv_wc_mb)(CHARSET_INFO *, my_wc_t, |
371 | uchar *, uchar *); |
372 | typedef size_t (*my_charset_conv_case)(CHARSET_INFO *, |
373 | char *, size_t, char *, size_t); |
374 | |
375 | /* |
376 | A structure to return the statistics of a native string copying, |
377 | when no Unicode conversion is involved. |
378 | |
379 | The structure is OK to be uninitialized before calling a copying routine. |
380 | A copying routine must populate the structure as follows: |
381 | - m_source_end_pos must be set by to a non-NULL value |
382 | in the range of the input string. |
383 | - m_well_formed_error_pos must be set to NULL if the string was |
384 | well formed, or to the position of the leftmost bad byte sequence. |
385 | */ |
386 | typedef struct |
387 | { |
388 | const char *m_source_end_pos; /* Position where reading stopped */ |
389 | const char *m_well_formed_error_pos; /* Position where a bad byte was found*/ |
390 | } MY_STRCOPY_STATUS; |
391 | |
392 | |
393 | /* |
394 | A structure to return the statistics of a Unicode string conversion. |
395 | */ |
396 | typedef struct |
397 | { |
398 | const char *m_cannot_convert_error_pos; |
399 | } MY_STRCONV_STATUS; |
400 | |
401 | |
402 | /* See strings/CHARSET_INFO.txt about information on this structure */ |
403 | struct my_charset_handler_st |
404 | { |
405 | my_bool (*init)(struct charset_info_st *, MY_CHARSET_LOADER *loader); |
406 | /* Multibyte routines */ |
407 | size_t (*numchars)(CHARSET_INFO *, const char *b, const char *e); |
408 | size_t (*charpos)(CHARSET_INFO *, const char *b, const char *e, |
409 | size_t pos); |
410 | size_t (*lengthsp)(CHARSET_INFO *, const char *ptr, size_t length); |
411 | size_t (*numcells)(CHARSET_INFO *, const char *b, const char *e); |
412 | |
413 | /* Unicode conversion */ |
414 | my_charset_conv_mb_wc mb_wc; |
415 | my_charset_conv_wc_mb wc_mb; |
416 | |
417 | /* CTYPE scanner */ |
418 | int (*ctype)(CHARSET_INFO *cs, int *ctype, |
419 | const uchar *s, const uchar *e); |
420 | |
421 | /* Functions for case and sort conversion */ |
422 | size_t (*caseup_str)(CHARSET_INFO *, char *); |
423 | size_t (*casedn_str)(CHARSET_INFO *, char *); |
424 | |
425 | my_charset_conv_case caseup; |
426 | my_charset_conv_case casedn; |
427 | |
428 | /* Charset dependent snprintf() */ |
429 | size_t (*snprintf)(CHARSET_INFO *, char *to, size_t n, |
430 | const char *fmt, |
431 | ...) ATTRIBUTE_FORMAT_FPTR(printf, 4, 5); |
432 | size_t (*long10_to_str)(CHARSET_INFO *, char *to, size_t n, |
433 | int radix, long int val); |
434 | size_t (*longlong10_to_str)(CHARSET_INFO *, char *to, size_t n, |
435 | int radix, longlong val); |
436 | |
437 | void (*fill)(CHARSET_INFO *, char *to, size_t len, int fill); |
438 | |
439 | /* String-to-number conversion routines */ |
440 | long (*strntol)(CHARSET_INFO *, const char *s, size_t l, |
441 | int base, char **e, int *err); |
442 | ulong (*strntoul)(CHARSET_INFO *, const char *s, size_t l, |
443 | int base, char **e, int *err); |
444 | longlong (*strntoll)(CHARSET_INFO *, const char *s, size_t l, |
445 | int base, char **e, int *err); |
446 | ulonglong (*strntoull)(CHARSET_INFO *, const char *s, size_t l, |
447 | int base, char **e, int *err); |
448 | double (*strntod)(CHARSET_INFO *, char *s, size_t l, char **e, |
449 | int *err); |
450 | longlong (*strtoll10)(CHARSET_INFO *cs, |
451 | const char *nptr, char **endptr, int *error); |
452 | ulonglong (*strntoull10rnd)(CHARSET_INFO *cs, |
453 | const char *str, size_t length, |
454 | int unsigned_fl, |
455 | char **endptr, int *error); |
456 | size_t (*scan)(CHARSET_INFO *, const char *b, const char *e, |
457 | int sq); |
458 | |
459 | /* String copying routines and helpers for them */ |
460 | /* |
461 | charlen() - calculate length of the left-most character in bytes. |
462 | @param cs Character set |
463 | @param str The beginning of the string |
464 | @param end The end of the string |
465 | |
466 | @return MY_CS_ILSEQ if a bad byte sequence was found. |
467 | @return MY_CS_TOOSMALLN(x) if the string ended unexpectedly. |
468 | @return a positive number in the range 1..mbmaxlen, |
469 | if a valid character was found. |
470 | */ |
471 | int (*charlen)(CHARSET_INFO *cs, const uchar *str, const uchar *end); |
472 | /* |
473 | well_formed_char_length() - returns character length of a string. |
474 | |
475 | @param cs Character set |
476 | @param str The beginning of the string |
477 | @param end The end of the string |
478 | @param nchars Not more than "nchars" left-most characters are checked. |
479 | @param status[OUT] Additional statistics is returned here. |
480 | "status" can be uninitialized before the call, |
481 | and it is fully initialized after the call. |
482 | |
483 | status->m_source_end_pos is set to the position where reading stopped. |
484 | |
485 | If a bad byte sequence is found, the function returns immediately and |
486 | status->m_well_formed_error_pos is set to the position where a bad byte |
487 | sequence was found. |
488 | |
489 | status->m_well_formed_error_pos is set to NULL if no bad bytes were found. |
490 | If status->m_well_formed_error_pos is NULL after the call, that means: |
491 | - either the function reached the end of the string, |
492 | - or all "nchars" characters were read. |
493 | The caller can check status->m_source_end_pos to detect which of these two |
494 | happened. |
495 | */ |
496 | size_t (*well_formed_char_length)(CHARSET_INFO *cs, |
497 | const char *str, const char *end, |
498 | size_t nchars, |
499 | MY_STRCOPY_STATUS *status); |
500 | |
501 | /* |
502 | copy_fix() - copy a string, replace bad bytes to '?'. |
503 | Not more than "nchars" characters are copied. |
504 | |
505 | status->m_source_end_pos is set to a position in the range |
506 | between "src" and "src + src_length", where reading stopped. |
507 | |
508 | status->m_well_formed_error_pos is set to NULL if the string |
509 | in the range "src" and "status->m_source_end_pos" was well formed, |
510 | or is set to a position between "src" and "src + src_length" where |
511 | the leftmost bad byte sequence was found. |
512 | */ |
513 | size_t (*copy_fix)(CHARSET_INFO *, |
514 | char *dst, size_t dst_length, |
515 | const char *src, size_t src_length, |
516 | size_t nchars, MY_STRCOPY_STATUS *status); |
517 | /** |
518 | Write a character to the target string, using its native code. |
519 | For Unicode character sets (utf8, ucs2, utf16, utf16le, utf32, filename) |
520 | native codes are equivalent to Unicode code points. |
521 | For 8bit character sets the native code is just the byte value. |
522 | For Asian characters sets: |
523 | - MB1 native code is just the byte value (e.g. on the ASCII range) |
524 | - MB2 native code is ((b0 << 8) + b1). |
525 | - MB3 native code is ((b0 <<16) + (b1 << 8) + b2) |
526 | Note, CHARSET_INFO::min_sort_char and CHARSET_INFO::max_sort_char |
527 | are defined in native notation and should be written using |
528 | cs->cset->native_to_mb() rather than cs->cset->wc_mb(). |
529 | */ |
530 | my_charset_conv_wc_mb native_to_mb; |
531 | }; |
532 | |
533 | extern MY_CHARSET_HANDLER my_charset_8bit_handler; |
534 | extern MY_CHARSET_HANDLER my_charset_ucs2_handler; |
535 | extern MY_CHARSET_HANDLER my_charset_utf8_handler; |
536 | |
537 | |
538 | /* |
539 | We define this CHARSET_INFO_DEFINED here to prevent a repeat of the |
540 | typedef in hash.c, which will cause a compiler error. |
541 | */ |
542 | #define CHARSET_INFO_DEFINED |
543 | |
544 | /* See strings/CHARSET_INFO.txt about information on this structure */ |
545 | struct charset_info_st |
546 | { |
547 | uint number; |
548 | uint primary_number; |
549 | uint binary_number; |
550 | uint state; |
551 | const char *csname; |
552 | const char *name; |
553 | const char *; |
554 | const char *tailoring; |
555 | const uchar *ctype; |
556 | const uchar *to_lower; |
557 | const uchar *to_upper; |
558 | const uchar *sort_order; |
559 | MY_UCA_INFO *uca; |
560 | const uint16 *tab_to_uni; |
561 | MY_UNI_IDX *tab_from_uni; |
562 | MY_UNICASE_INFO *caseinfo; |
563 | const uchar *state_map; |
564 | const uchar *ident_map; |
565 | uint strxfrm_multiply; |
566 | uchar caseup_multiply; |
567 | uchar casedn_multiply; |
568 | uint mbminlen; |
569 | uint mbmaxlen; |
570 | my_wc_t min_sort_char; |
571 | my_wc_t max_sort_char; /* For LIKE optimization */ |
572 | uchar pad_char; |
573 | my_bool escape_with_backslash_is_dangerous; |
574 | uchar levels_for_order; |
575 | |
576 | MY_CHARSET_HANDLER *cset; |
577 | MY_COLLATION_HANDLER *coll; |
578 | |
579 | }; |
580 | #define ILLEGAL_CHARSET_INFO_NUMBER (~0U) |
581 | |
582 | extern MYSQL_PLUGIN_IMPORT struct charset_info_st my_charset_bin; |
583 | extern MYSQL_PLUGIN_IMPORT struct charset_info_st my_charset_latin1; |
584 | extern MYSQL_PLUGIN_IMPORT struct charset_info_st my_charset_latin1_nopad; |
585 | extern MYSQL_PLUGIN_IMPORT struct charset_info_st my_charset_filename; |
586 | extern MYSQL_PLUGIN_IMPORT struct charset_info_st my_charset_utf8_general_ci; |
587 | |
588 | extern struct charset_info_st my_charset_big5_bin; |
589 | extern struct charset_info_st my_charset_big5_chinese_ci; |
590 | extern struct charset_info_st my_charset_big5_nopad_bin; |
591 | extern struct charset_info_st my_charset_big5_chinese_nopad_ci; |
592 | extern struct charset_info_st my_charset_cp1250_czech_ci; |
593 | extern struct charset_info_st my_charset_cp932_bin; |
594 | extern struct charset_info_st my_charset_cp932_japanese_ci; |
595 | extern struct charset_info_st my_charset_cp932_nopad_bin; |
596 | extern struct charset_info_st my_charset_cp932_japanese_nopad_ci; |
597 | extern struct charset_info_st my_charset_eucjpms_bin; |
598 | extern struct charset_info_st my_charset_eucjpms_japanese_ci; |
599 | extern struct charset_info_st my_charset_eucjpms_nopad_bin; |
600 | extern struct charset_info_st my_charset_eucjpms_japanese_nopad_ci; |
601 | extern struct charset_info_st my_charset_euckr_bin; |
602 | extern struct charset_info_st my_charset_euckr_korean_ci; |
603 | extern struct charset_info_st my_charset_euckr_nopad_bin; |
604 | extern struct charset_info_st my_charset_euckr_korean_nopad_ci; |
605 | extern struct charset_info_st my_charset_gb2312_bin; |
606 | extern struct charset_info_st my_charset_gb2312_chinese_ci; |
607 | extern struct charset_info_st my_charset_gb2312_nopad_bin; |
608 | extern struct charset_info_st my_charset_gb2312_chinese_nopad_ci; |
609 | extern struct charset_info_st my_charset_gbk_bin; |
610 | extern struct charset_info_st my_charset_gbk_chinese_ci; |
611 | extern struct charset_info_st my_charset_gbk_nopad_bin; |
612 | extern struct charset_info_st my_charset_gbk_chinese_nopad_ci; |
613 | extern struct charset_info_st my_charset_latin1_bin; |
614 | extern struct charset_info_st my_charset_latin1_nopad_bin; |
615 | extern struct charset_info_st my_charset_latin1_german2_ci; |
616 | extern struct charset_info_st my_charset_latin2_czech_ci; |
617 | extern struct charset_info_st my_charset_sjis_bin; |
618 | extern struct charset_info_st my_charset_sjis_japanese_ci; |
619 | extern struct charset_info_st my_charset_sjis_nopad_bin; |
620 | extern struct charset_info_st my_charset_sjis_japanese_nopad_ci; |
621 | extern struct charset_info_st my_charset_tis620_bin; |
622 | extern struct charset_info_st my_charset_tis620_thai_ci; |
623 | extern struct charset_info_st my_charset_tis620_nopad_bin; |
624 | extern struct charset_info_st my_charset_tis620_thai_nopad_ci; |
625 | extern struct charset_info_st my_charset_ucs2_bin; |
626 | extern struct charset_info_st my_charset_ucs2_general_ci; |
627 | extern struct charset_info_st my_charset_ucs2_nopad_bin; |
628 | extern struct charset_info_st my_charset_ucs2_general_nopad_ci; |
629 | extern struct charset_info_st my_charset_ucs2_general_mysql500_ci; |
630 | extern struct charset_info_st my_charset_ucs2_unicode_ci; |
631 | extern struct charset_info_st my_charset_ucs2_unicode_nopad_ci; |
632 | extern struct charset_info_st my_charset_ucs2_general_mysql500_ci; |
633 | extern struct charset_info_st my_charset_ujis_bin; |
634 | extern struct charset_info_st my_charset_ujis_japanese_ci; |
635 | extern struct charset_info_st my_charset_ujis_nopad_bin; |
636 | extern struct charset_info_st my_charset_ujis_japanese_nopad_ci; |
637 | extern struct charset_info_st my_charset_utf16_bin; |
638 | extern struct charset_info_st my_charset_utf16_general_ci; |
639 | extern struct charset_info_st my_charset_utf16_unicode_ci; |
640 | extern struct charset_info_st my_charset_utf16_unicode_nopad_ci; |
641 | extern struct charset_info_st my_charset_utf16le_bin; |
642 | extern struct charset_info_st my_charset_utf16le_general_ci; |
643 | extern struct charset_info_st my_charset_utf16_general_nopad_ci; |
644 | extern struct charset_info_st my_charset_utf16_nopad_bin; |
645 | extern struct charset_info_st my_charset_utf16le_nopad_bin; |
646 | extern struct charset_info_st my_charset_utf16le_general_nopad_ci; |
647 | extern struct charset_info_st my_charset_utf32_bin; |
648 | extern struct charset_info_st my_charset_utf32_general_ci; |
649 | extern struct charset_info_st my_charset_utf32_unicode_ci; |
650 | extern struct charset_info_st my_charset_utf32_unicode_nopad_ci; |
651 | extern struct charset_info_st my_charset_utf32_nopad_bin; |
652 | extern struct charset_info_st my_charset_utf32_general_nopad_ci; |
653 | extern struct charset_info_st my_charset_utf8_bin; |
654 | extern struct charset_info_st my_charset_utf8_nopad_bin; |
655 | extern struct charset_info_st my_charset_utf8_general_nopad_ci; |
656 | extern struct charset_info_st my_charset_utf8_general_mysql500_ci; |
657 | extern struct charset_info_st my_charset_utf8_unicode_ci; |
658 | extern struct charset_info_st my_charset_utf8_unicode_nopad_ci; |
659 | extern struct charset_info_st my_charset_utf8mb4_bin; |
660 | extern struct charset_info_st my_charset_utf8mb4_general_ci; |
661 | extern struct charset_info_st my_charset_utf8mb4_nopad_bin; |
662 | extern struct charset_info_st my_charset_utf8mb4_general_nopad_ci; |
663 | extern struct charset_info_st my_charset_utf8mb4_unicode_ci; |
664 | extern struct charset_info_st my_charset_utf8mb4_unicode_nopad_ci; |
665 | |
666 | #define MY_UTF8MB3 "utf8" |
667 | #define MY_UTF8MB4 "utf8mb4" |
668 | |
669 | my_bool my_cs_have_contractions(CHARSET_INFO *cs); |
670 | my_bool my_cs_can_be_contraction_head(CHARSET_INFO *cs, my_wc_t wc); |
671 | my_bool my_cs_can_be_contraction_tail(CHARSET_INFO *cs, my_wc_t wc); |
672 | const uint16 *my_cs_contraction2_weight(CHARSET_INFO *cs, my_wc_t wc1, |
673 | my_wc_t wc2); |
674 | |
675 | /* declarations for simple charsets */ |
676 | extern size_t my_strnxfrm_simple(CHARSET_INFO *, |
677 | uchar *dst, size_t dstlen, uint nweights, |
678 | const uchar *src, size_t srclen, uint flags); |
679 | size_t my_strnxfrmlen_simple(CHARSET_INFO *, size_t); |
680 | extern int my_strnncoll_simple(CHARSET_INFO *, const uchar *, size_t, |
681 | const uchar *, size_t, my_bool); |
682 | |
683 | extern int my_strnncollsp_simple(CHARSET_INFO *, const uchar *, size_t, |
684 | const uchar *, size_t); |
685 | |
686 | extern void my_hash_sort_simple(CHARSET_INFO *cs, |
687 | const uchar *key, size_t len, |
688 | ulong *nr1, ulong *nr2); |
689 | |
690 | extern void my_hash_sort_simple_nopad(CHARSET_INFO *cs, |
691 | const uchar *key, size_t len, |
692 | ulong *nr1, ulong *nr2); |
693 | |
694 | extern void my_hash_sort_bin(CHARSET_INFO *cs, |
695 | const uchar *key, size_t len, ulong *nr1, |
696 | ulong *nr2); |
697 | |
698 | /** |
699 | Compare a string to an array of spaces, for PAD SPACE comparison. |
700 | The function iterates through the string and compares every byte to 0x20. |
701 | @param - the string |
702 | @param - its length |
703 | @return <0 - if a byte less than 0x20 was found in the string. |
704 | @return 0 - if all bytes in the string were 0x20, or if length was 0. |
705 | @return >0 - if a byte greater than 0x20 was found in the string. |
706 | */ |
707 | extern int my_strnncollsp_padspace_bin(const uchar *str, size_t length); |
708 | |
709 | extern size_t my_lengthsp_8bit(CHARSET_INFO *cs, const char *ptr, size_t length); |
710 | |
711 | extern uint my_instr_simple(CHARSET_INFO *, |
712 | const char *b, size_t b_length, |
713 | const char *s, size_t s_length, |
714 | my_match_t *match, uint nmatch); |
715 | |
716 | size_t my_copy_8bit(CHARSET_INFO *, |
717 | char *dst, size_t dst_length, |
718 | const char *src, size_t src_length, |
719 | size_t nchars, MY_STRCOPY_STATUS *); |
720 | size_t my_copy_fix_mb(CHARSET_INFO *cs, |
721 | char *dst, size_t dst_length, |
722 | const char *src, size_t src_length, |
723 | size_t nchars, MY_STRCOPY_STATUS *); |
724 | |
725 | /* Functions for 8bit */ |
726 | extern size_t my_caseup_str_8bit(CHARSET_INFO *, char *); |
727 | extern size_t my_casedn_str_8bit(CHARSET_INFO *, char *); |
728 | extern size_t my_caseup_8bit(CHARSET_INFO *, char *src, size_t srclen, |
729 | char *dst, size_t dstlen); |
730 | extern size_t my_casedn_8bit(CHARSET_INFO *, char *src, size_t srclen, |
731 | char *dst, size_t dstlen); |
732 | |
733 | extern int my_strcasecmp_8bit(CHARSET_INFO * cs, const char *, const char *); |
734 | |
735 | int my_mb_wc_8bit(CHARSET_INFO *cs,my_wc_t *wc, const uchar *s,const uchar *e); |
736 | int my_wc_mb_8bit(CHARSET_INFO *cs,my_wc_t wc, uchar *s, uchar *e); |
737 | int my_wc_mb_bin(CHARSET_INFO *cs,my_wc_t wc, uchar *s, uchar *e); |
738 | |
739 | int my_mb_ctype_8bit(CHARSET_INFO *,int *, const uchar *,const uchar *); |
740 | int my_mb_ctype_mb(CHARSET_INFO *,int *, const uchar *,const uchar *); |
741 | |
742 | size_t my_scan_8bit(CHARSET_INFO *cs, const char *b, const char *e, int sq); |
743 | |
744 | size_t my_snprintf_8bit(CHARSET_INFO *, char *to, size_t n, |
745 | const char *fmt, ...) |
746 | ATTRIBUTE_FORMAT(printf, 4, 5); |
747 | |
748 | long my_strntol_8bit(CHARSET_INFO *, const char *s, size_t l, int base, |
749 | char **e, int *err); |
750 | ulong my_strntoul_8bit(CHARSET_INFO *, const char *s, size_t l, int base, |
751 | char **e, int *err); |
752 | longlong my_strntoll_8bit(CHARSET_INFO *, const char *s, size_t l, int base, |
753 | char **e, int *err); |
754 | ulonglong my_strntoull_8bit(CHARSET_INFO *, const char *s, size_t l, int base, |
755 | char **e, int *err); |
756 | double my_strntod_8bit(CHARSET_INFO *, char *s, size_t l,char **e, |
757 | int *err); |
758 | size_t my_long10_to_str_8bit(CHARSET_INFO *, char *to, size_t l, int radix, |
759 | long int val); |
760 | size_t my_longlong10_to_str_8bit(CHARSET_INFO *, char *to, size_t l, int radix, |
761 | longlong val); |
762 | |
763 | longlong my_strtoll10_8bit(CHARSET_INFO *cs, |
764 | const char *nptr, char **endptr, int *error); |
765 | longlong my_strtoll10_ucs2(CHARSET_INFO *cs, |
766 | const char *nptr, char **endptr, int *error); |
767 | |
768 | ulonglong my_strntoull10rnd_8bit(CHARSET_INFO *cs, |
769 | const char *str, size_t length, int |
770 | unsigned_fl, char **endptr, int *error); |
771 | ulonglong my_strntoull10rnd_ucs2(CHARSET_INFO *cs, |
772 | const char *str, size_t length, |
773 | int unsigned_fl, char **endptr, int *error); |
774 | |
775 | void my_fill_8bit(CHARSET_INFO *cs, char* to, size_t l, int fill); |
776 | |
777 | /* For 8-bit character set */ |
778 | my_bool my_like_range_simple(CHARSET_INFO *cs, |
779 | const char *ptr, size_t ptr_length, |
780 | pbool escape, pbool w_one, pbool w_many, |
781 | size_t res_length, |
782 | char *min_str, char *max_str, |
783 | size_t *min_length, size_t *max_length); |
784 | |
785 | /* For ASCII-based multi-byte character sets with mbminlen=1 */ |
786 | my_bool my_like_range_mb(CHARSET_INFO *cs, |
787 | const char *ptr, size_t ptr_length, |
788 | pbool escape, pbool w_one, pbool w_many, |
789 | size_t res_length, |
790 | char *min_str, char *max_str, |
791 | size_t *min_length, size_t *max_length); |
792 | |
793 | /* For other character sets, with arbitrary mbminlen and mbmaxlen numbers */ |
794 | my_bool my_like_range_generic(CHARSET_INFO *cs, |
795 | const char *ptr, size_t ptr_length, |
796 | pbool escape, pbool w_one, pbool w_many, |
797 | size_t res_length, |
798 | char *min_str, char *max_str, |
799 | size_t *min_length, size_t *max_length); |
800 | |
801 | int my_wildcmp_8bit(CHARSET_INFO *, |
802 | const char *str,const char *str_end, |
803 | const char *wildstr,const char *wildend, |
804 | int escape, int w_one, int w_many); |
805 | |
806 | int my_wildcmp_bin(CHARSET_INFO *, |
807 | const char *str,const char *str_end, |
808 | const char *wildstr,const char *wildend, |
809 | int escape, int w_one, int w_many); |
810 | |
811 | size_t my_numchars_8bit(CHARSET_INFO *, const char *b, const char *e); |
812 | size_t my_numcells_8bit(CHARSET_INFO *, const char *b, const char *e); |
813 | size_t my_charpos_8bit(CHARSET_INFO *, const char *b, const char *e, size_t pos); |
814 | size_t my_well_formed_char_length_8bit(CHARSET_INFO *cs, |
815 | const char *b, const char *e, |
816 | size_t nchars, |
817 | MY_STRCOPY_STATUS *status); |
818 | int my_charlen_8bit(CHARSET_INFO *, const uchar *str, const uchar *end); |
819 | |
820 | |
821 | /* Functions for multibyte charsets */ |
822 | extern size_t my_caseup_str_mb(CHARSET_INFO *, char *); |
823 | extern size_t my_casedn_str_mb(CHARSET_INFO *, char *); |
824 | extern size_t my_caseup_mb(CHARSET_INFO *, char *src, size_t srclen, |
825 | char *dst, size_t dstlen); |
826 | extern size_t my_casedn_mb(CHARSET_INFO *, char *src, size_t srclen, |
827 | char *dst, size_t dstlen); |
828 | extern size_t my_caseup_mb_varlen(CHARSET_INFO *, char *src, size_t srclen, |
829 | char *dst, size_t dstlen); |
830 | extern size_t my_casedn_mb_varlen(CHARSET_INFO *, char *src, size_t srclen, |
831 | char *dst, size_t dstlen); |
832 | extern size_t my_caseup_ujis(CHARSET_INFO *, char *src, size_t srclen, |
833 | char *dst, size_t dstlen); |
834 | extern size_t my_casedn_ujis(CHARSET_INFO *, char *src, size_t srclen, |
835 | char *dst, size_t dstlen); |
836 | extern int my_strcasecmp_mb(CHARSET_INFO * cs,const char *, const char *); |
837 | |
838 | int my_wildcmp_mb(CHARSET_INFO *, |
839 | const char *str,const char *str_end, |
840 | const char *wildstr,const char *wildend, |
841 | int escape, int w_one, int w_many); |
842 | size_t my_numchars_mb(CHARSET_INFO *, const char *b, const char *e); |
843 | size_t my_numcells_mb(CHARSET_INFO *, const char *b, const char *e); |
844 | size_t my_charpos_mb(CHARSET_INFO *, const char *b, const char *e, size_t pos); |
845 | uint my_instr_mb(CHARSET_INFO *, |
846 | const char *b, size_t b_length, |
847 | const char *s, size_t s_length, |
848 | my_match_t *match, uint nmatch); |
849 | |
850 | int my_wildcmp_mb_bin(CHARSET_INFO *cs, |
851 | const char *str,const char *str_end, |
852 | const char *wildstr,const char *wildend, |
853 | int escape, int w_one, int w_many); |
854 | |
855 | int my_strcasecmp_mb_bin(CHARSET_INFO * cs __attribute__((unused)), |
856 | const char *s, const char *t); |
857 | |
858 | void my_hash_sort_mb_bin(CHARSET_INFO *cs __attribute__((unused)), |
859 | const uchar *key, size_t len,ulong *nr1, ulong *nr2); |
860 | |
861 | void my_hash_sort_mb_nopad_bin(CHARSET_INFO *cs __attribute__((unused)), |
862 | const uchar *key, size_t len, |
863 | ulong *nr1, ulong *nr2); |
864 | |
865 | size_t my_strnxfrm_mb(CHARSET_INFO *, |
866 | uchar *dst, size_t dstlen, uint nweights, |
867 | const uchar *src, size_t srclen, uint flags); |
868 | |
869 | size_t my_strnxfrm_mb_nopad(CHARSET_INFO *, |
870 | uchar *dst, size_t dstlen, uint nweights, |
871 | const uchar *src, size_t srclen, uint flags); |
872 | |
873 | size_t my_strnxfrm_unicode(CHARSET_INFO *, |
874 | uchar *dst, size_t dstlen, uint nweights, |
875 | const uchar *src, size_t srclen, uint flags); |
876 | |
877 | size_t my_strnxfrm_unicode_nopad(CHARSET_INFO *, |
878 | uchar *dst, size_t dstlen, uint nweights, |
879 | const uchar *src, size_t srclen, uint flags); |
880 | |
881 | size_t my_strnxfrmlen_unicode(CHARSET_INFO *, size_t); |
882 | |
883 | size_t my_strnxfrm_unicode_full_bin(CHARSET_INFO *, |
884 | uchar *dst, size_t dstlen, |
885 | uint nweights, const uchar *src, |
886 | size_t srclen, uint flags); |
887 | |
888 | size_t my_strnxfrm_unicode_full_nopad_bin(CHARSET_INFO *, |
889 | uchar *dst, size_t dstlen, |
890 | uint nweights, const uchar *src, |
891 | size_t srclen, uint flags); |
892 | |
893 | size_t my_strnxfrmlen_unicode_full_bin(CHARSET_INFO *, size_t); |
894 | |
895 | int my_wildcmp_unicode(CHARSET_INFO *cs, |
896 | const char *str, const char *str_end, |
897 | const char *wildstr, const char *wildend, |
898 | int escape, int w_one, int w_many, |
899 | MY_UNICASE_INFO *weights); |
900 | |
901 | extern my_bool my_parse_charset_xml(MY_CHARSET_LOADER *loader, |
902 | const char *buf, size_t buflen); |
903 | extern char *my_strchr(CHARSET_INFO *cs, const char *str, const char *end, |
904 | pchar c); |
905 | extern size_t my_strcspn(CHARSET_INFO *cs, const char *str, const char *end, |
906 | const char *accept); |
907 | |
908 | my_bool my_propagate_simple(CHARSET_INFO *cs, const uchar *str, size_t len); |
909 | my_bool my_propagate_complex(CHARSET_INFO *cs, const uchar *str, size_t len); |
910 | |
911 | |
912 | typedef struct |
913 | { |
914 | size_t char_length; |
915 | uint repertoire; |
916 | } MY_STRING_METADATA; |
917 | |
918 | void my_string_metadata_get(MY_STRING_METADATA *metadata, |
919 | CHARSET_INFO *cs, const char *str, size_t len); |
920 | uint my_string_repertoire(CHARSET_INFO *cs, const char *str, size_t len); |
921 | my_bool my_charset_is_ascii_based(CHARSET_INFO *cs); |
922 | uint my_charset_repertoire(CHARSET_INFO *cs); |
923 | |
924 | uint my_strxfrm_flag_normalize(uint flags, uint nlevels); |
925 | void my_strxfrm_desc_and_reverse(uchar *str, uchar *strend, |
926 | uint flags, uint level); |
927 | size_t my_strxfrm_pad_desc_and_reverse(CHARSET_INFO *cs, |
928 | uchar *str, uchar *frmend, uchar *strend, |
929 | uint nweights, uint flags, uint level); |
930 | size_t my_strxfrm_pad_desc_and_reverse_nopad(CHARSET_INFO *cs, |
931 | uchar *str, uchar *frmend, |
932 | uchar *strend, uint nweights, |
933 | uint flags, uint level); |
934 | |
935 | const MY_CONTRACTIONS *my_charset_get_contractions(CHARSET_INFO *cs, |
936 | int level); |
937 | |
938 | extern size_t my_vsnprintf_ex(CHARSET_INFO *cs, char *to, size_t n, |
939 | const char* fmt, va_list ap); |
940 | |
941 | /* |
942 | Convert a string between two character sets. |
943 | Bad byte sequences as well as characters that cannot be |
944 | encoded in the destination character set are replaced to '?'. |
945 | */ |
946 | uint32 my_convert(char *to, uint32 to_length, CHARSET_INFO *to_cs, |
947 | const char *from, uint32 from_length, |
948 | CHARSET_INFO *from_cs, uint *errors); |
949 | |
950 | /** |
951 | An extended version of my_convert(), to pass non-default mb_wc() and wc_mb(). |
952 | For example, String::copy_printable() which is used in |
953 | Protocol::store_warning() uses this to escape control |
954 | and non-convertable characters. |
955 | */ |
956 | uint32 my_convert_using_func(char *to, size_t to_length, CHARSET_INFO *to_cs, |
957 | my_charset_conv_wc_mb mb_wc, |
958 | const char *from, size_t from_length, |
959 | CHARSET_INFO *from_cs, |
960 | my_charset_conv_mb_wc wc_mb, |
961 | uint *errors); |
962 | /* |
963 | Convert a string between two character sets. |
964 | Bad byte sequences as well as characters that cannot be |
965 | encoded in the destination character set are replaced to '?'. |
966 | Not more than "nchars" characters are copied. |
967 | Conversion statistics is returned in "status" and is set as follows: |
968 | - status->m_native_copy_status.m_source_end_pos - to the position |
969 | between (src) and (src+src_length), where the function stopped reading |
970 | the source string. |
971 | - status->m_native_copy_status.m_well_formed_error_pos - to the position |
972 | between (src) and (src+src_length), where the first badly formed byte |
973 | sequence was found, or to NULL if the string was well formed in the |
974 | given range. |
975 | - status->m_cannot_convert_error_pos - to the position |
976 | between (src) and (src+src_length), where the first character that |
977 | cannot be represented in the destination character set was found, |
978 | or to NULL if all characters in the given range were successfully |
979 | converted. |
980 | |
981 | "src" is allowed to be a NULL pointer. In this case "src_length" must |
982 | be equal to 0. All "status" members are initialized to NULL, and 0 is |
983 | returned. |
984 | */ |
985 | size_t my_convert_fix(CHARSET_INFO *dstcs, char *dst, size_t dst_length, |
986 | CHARSET_INFO *srccs, const char *src, size_t src_length, |
987 | size_t nchars, |
988 | MY_STRCOPY_STATUS *copy_status, |
989 | MY_STRCONV_STATUS *conv_status); |
990 | |
991 | #define _MY_U 01 /* Upper case */ |
992 | #define _MY_L 02 /* Lower case */ |
993 | #define _MY_NMR 04 /* Numeral (digit) */ |
994 | #define _MY_SPC 010 /* Spacing character */ |
995 | #define _MY_PNT 020 /* Punctuation */ |
996 | #define _MY_CTR 040 /* Control character */ |
997 | #define _MY_B 0100 /* Blank */ |
998 | #define _MY_X 0200 /* heXadecimal digit */ |
999 | |
1000 | |
1001 | #define my_isascii(c) (!((c) & ~0177)) |
1002 | #define my_toascii(c) ((c) & 0177) |
1003 | #define my_tocntrl(c) ((c) & 31) |
1004 | #define my_toprint(c) ((c) | 64) |
1005 | #define my_toupper(s,c) (char) ((s)->to_upper[(uchar) (c)]) |
1006 | #define my_tolower(s,c) (char) ((s)->to_lower[(uchar) (c)]) |
1007 | #define my_isalpha(s, c) (((s)->ctype+1)[(uchar) (c)] & (_MY_U | _MY_L)) |
1008 | #define my_isupper(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_U) |
1009 | #define my_islower(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_L) |
1010 | #define my_isdigit(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_NMR) |
1011 | #define my_isxdigit(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_X) |
1012 | #define my_isalnum(s, c) (((s)->ctype+1)[(uchar) (c)] & (_MY_U | _MY_L | _MY_NMR)) |
1013 | #define my_isspace(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_SPC) |
1014 | #define my_ispunct(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_PNT) |
1015 | #define my_isprint(s, c) (((s)->ctype+1)[(uchar) (c)] & (_MY_PNT | _MY_U | _MY_L | _MY_NMR | _MY_B)) |
1016 | #define my_isgraph(s, c) (((s)->ctype+1)[(uchar) (c)] & (_MY_PNT | _MY_U | _MY_L | _MY_NMR)) |
1017 | #define my_iscntrl(s, c) (((s)->ctype+1)[(uchar) (c)] & _MY_CTR) |
1018 | |
1019 | /* Some macros that should be cleaned up a little */ |
1020 | #define my_isvar(s,c) (my_isalnum(s,c) || (c) == '_') |
1021 | #define my_isvar_start(s,c) (my_isalpha(s,c) || (c) == '_') |
1022 | |
1023 | #define my_binary_compare(s) ((s)->state & MY_CS_BINSORT) |
1024 | #define use_strnxfrm(s) ((s)->state & MY_CS_STRNXFRM) |
1025 | #define my_strnxfrm(cs, d, dl, s, sl) \ |
1026 | ((cs)->coll->strnxfrm((cs), (d), (dl), (dl), (s), (sl), MY_STRXFRM_PAD_WITH_SPACE)) |
1027 | #define my_strnncoll(s, a, b, c, d) ((s)->coll->strnncoll((s), (a), (b), (c), (d), 0)) |
1028 | #define my_like_range(s, a, b, c, d, e, f, g, h, i, j) \ |
1029 | ((s)->coll->like_range((s), (a), (b), (c), (d), (e), (f), (g), (h), (i), (j))) |
1030 | #define my_wildcmp(cs,s,se,w,we,e,o,m) ((cs)->coll->wildcmp((cs),(s),(se),(w),(we),(e),(o),(m))) |
1031 | #define my_strcasecmp(s, a, b) ((s)->coll->strcasecmp((s), (a), (b))) |
1032 | #define my_charpos(cs, b, e, num) (cs)->cset->charpos((cs), (const char*) (b), (const char *)(e), (num)) |
1033 | |
1034 | #define use_mb(s) ((s)->mbmaxlen > 1) |
1035 | /** |
1036 | Detect if the leftmost character in a string is a valid multi-byte character |
1037 | and return its length, or return 0 otherwise. |
1038 | @param cs - character set |
1039 | @param str - the beginning of the string |
1040 | @param end - the string end (the next byte after the string) |
1041 | @return >0, for a multi-byte character |
1042 | @rerurn 0, for a single byte character, broken sequence, empty string. |
1043 | */ |
1044 | static inline |
1045 | uint my_ismbchar(CHARSET_INFO *cs, const char *str, const char *end) |
1046 | { |
1047 | int char_length= (cs->cset->charlen)(cs, (const uchar *) str, |
1048 | (const uchar *) end); |
1049 | return char_length > 1 ? (uint) char_length : 0U; |
1050 | } |
1051 | |
1052 | |
1053 | /** |
1054 | Return length of the leftmost character in a string. |
1055 | @param cs - character set |
1056 | @param str - the beginning of the string |
1057 | @param end - the string end (the next byte after the string) |
1058 | @return <=0 on errors (EOL, wrong byte sequence) |
1059 | @return 1 on a single byte character |
1060 | @return >1 on a multi-byte character |
1061 | |
1062 | Note, inlike my_ismbchar(), 1 is returned for a single byte character. |
1063 | */ |
1064 | static inline |
1065 | int my_charlen(CHARSET_INFO *cs, const char *str, const char *end) |
1066 | { |
1067 | return (cs->cset->charlen)(cs, (const uchar *) str, |
1068 | (const uchar *) end); |
1069 | } |
1070 | |
1071 | |
1072 | /** |
1073 | Convert broken and incomplete byte sequences to 1 byte. |
1074 | */ |
1075 | static inline |
1076 | uint my_charlen_fix(CHARSET_INFO *cs, const char *str, const char *end) |
1077 | { |
1078 | int char_length= my_charlen(cs, str, end); |
1079 | DBUG_ASSERT(str < end); |
1080 | return char_length > 0 ? (uint) char_length : (uint) 1U; |
1081 | } |
1082 | |
1083 | |
1084 | /* |
1085 | A compatibility replacement pure C function for the former |
1086 | cs->cset->well_formed_len(). |
1087 | In C++ code please use Well_formed_prefix::length() instead. |
1088 | */ |
1089 | static inline size_t |
1090 | my_well_formed_length(CHARSET_INFO *cs, const char *b, const char *e, |
1091 | size_t nchars, int *error) |
1092 | { |
1093 | MY_STRCOPY_STATUS status; |
1094 | (void) cs->cset->well_formed_char_length(cs, b, e, nchars, &status); |
1095 | *error= status.m_well_formed_error_pos == NULL ? 0 : 1; |
1096 | return (size_t) (status.m_source_end_pos - b); |
1097 | } |
1098 | |
1099 | |
1100 | #define my_caseup_str(s, a) ((s)->cset->caseup_str((s), (a))) |
1101 | #define my_casedn_str(s, a) ((s)->cset->casedn_str((s), (a))) |
1102 | #define my_strntol(s, a, b, c, d, e) ((s)->cset->strntol((s),(a),(b),(c),(d),(e))) |
1103 | #define my_strntoul(s, a, b, c, d, e) ((s)->cset->strntoul((s),(a),(b),(c),(d),(e))) |
1104 | #define my_strntoll(s, a, b, c, d, e) ((s)->cset->strntoll((s),(a),(b),(c),(d),(e))) |
1105 | #define my_strntoull(s, a, b, c,d, e) ((s)->cset->strntoull((s),(a),(b),(c),(d),(e))) |
1106 | #define my_strntod(s, a, b, c, d) ((s)->cset->strntod((s),(a),(b),(c),(d))) |
1107 | |
1108 | |
1109 | /* XXX: still need to take care of this one */ |
1110 | #ifdef MY_CHARSET_TIS620 |
1111 | #error The TIS620 charset is broken at the moment. Tell tim to fix it. |
1112 | #define USE_TIS620 |
1113 | #include "t_ctype.h" |
1114 | #endif |
1115 | |
1116 | #ifdef __cplusplus |
1117 | } |
1118 | #endif |
1119 | |
1120 | #endif /* _m_ctype_h */ |
1121 | |