1 /* Multibyte character data type. 2 Copyright (C) 2001, 2005-2007, 2009-2012 Free Software Foundation, Inc. 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; either version 3 of the License, or 7 (at your option) any later version. 8 9 This program is distributed in the hope that it will be useful, 10 but WITHOUT ANY WARRANTY; without even the implied warranty of 11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 12 GNU General Public License for more details. 13 14 You should have received a copy of the GNU General Public License 15 along with this program. If not, see <http://www.gnu.org/licenses/>. */ 16 17 /* Written by Bruno Haible <bruno@clisp.org>. */ 18 19 /* A multibyte character is a short subsequence of a char* string, 20 representing a single wide character. 21 22 We use multibyte characters instead of wide characters because of 23 the following goals: 24 1) correct multibyte handling, i.e. operate according to the LC_CTYPE 25 locale, 26 2) ease of maintenance, i.e. the maintainer needs not know all details 27 of the ISO C 99 standard, 28 3) don't fail grossly if the input is not in the encoding set by the 29 locale, because often different encodings are in use in the same 30 countries (ISO-8859-1/UTF-8, EUC-JP/Shift_JIS, ...), 31 4) fast in the case of ASCII characters, 32 5) portability, i.e. don't make unportable assumptions about wchar_t. 33 34 Multibyte characters are only accessed through the mb* macros. 35 36 mb_ptr (mbc) 37 return a pointer to the beginning of the multibyte sequence. 38 39 mb_len (mbc) 40 returns the number of bytes occupied by the multibyte sequence. 41 Always > 0. 42 43 mb_iseq (mbc, sc) 44 returns true if mbc is the standard ASCII character sc. 45 46 mb_isnul (mbc) 47 returns true if mbc is the nul character. 48 49 mb_cmp (mbc1, mbc2) 50 returns a positive, zero, or negative value depending on whether mbc1 51 sorts after, same or before mbc2. 52 53 mb_casecmp (mbc1, mbc2) 54 returns a positive, zero, or negative value depending on whether mbc1 55 sorts after, same or before mbc2, modulo upper/lowercase conversion. 56 57 mb_equal (mbc1, mbc2) 58 returns true if mbc1 and mbc2 are equal. 59 60 mb_caseequal (mbc1, mbc2) 61 returns true if mbc1 and mbc2 are equal modulo upper/lowercase conversion. 62 63 mb_isalnum (mbc) 64 returns true if mbc is alphanumeric. 65 66 mb_isalpha (mbc) 67 returns true if mbc is alphabetic. 68 69 mb_isascii(mbc) 70 returns true if mbc is plain ASCII. 71 72 mb_isblank (mbc) 73 returns true if mbc is a blank. 74 75 mb_iscntrl (mbc) 76 returns true if mbc is a control character. 77 78 mb_isdigit (mbc) 79 returns true if mbc is a decimal digit. 80 81 mb_isgraph (mbc) 82 returns true if mbc is a graphic character. 83 84 mb_islower (mbc) 85 returns true if mbc is lowercase. 86 87 mb_isprint (mbc) 88 returns true if mbc is a printable character. 89 90 mb_ispunct (mbc) 91 returns true if mbc is a punctuation character. 92 93 mb_isspace (mbc) 94 returns true if mbc is a space character. 95 96 mb_isupper (mbc) 97 returns true if mbc is uppercase. 98 99 mb_isxdigit (mbc) 100 returns true if mbc is a hexadecimal digit. 101 102 mb_width (mbc) 103 returns the number of columns on the output device occupied by mbc. 104 Always >= 0. 105 106 mb_putc (mbc, stream) 107 outputs mbc on stream, a byte oriented FILE stream opened for output. 108 109 mb_setascii (&mbc, sc) 110 assigns the standard ASCII character sc to mbc. 111 112 mb_copy (&destmbc, &srcmbc) 113 copies srcmbc to destmbc. 114 115 Here are the function prototypes of the macros. 116 117 extern const char * mb_ptr (const mbchar_t mbc); 118 extern size_t mb_len (const mbchar_t mbc); 119 extern bool mb_iseq (const mbchar_t mbc, char sc); 120 extern bool mb_isnul (const mbchar_t mbc); 121 extern int mb_cmp (const mbchar_t mbc1, const mbchar_t mbc2); 122 extern int mb_casecmp (const mbchar_t mbc1, const mbchar_t mbc2); 123 extern bool mb_equal (const mbchar_t mbc1, const mbchar_t mbc2); 124 extern bool mb_caseequal (const mbchar_t mbc1, const mbchar_t mbc2); 125 extern bool mb_isalnum (const mbchar_t mbc); 126 extern bool mb_isalpha (const mbchar_t mbc); 127 extern bool mb_isascii (const mbchar_t mbc); 128 extern bool mb_isblank (const mbchar_t mbc); 129 extern bool mb_iscntrl (const mbchar_t mbc); 130 extern bool mb_isdigit (const mbchar_t mbc); 131 extern bool mb_isgraph (const mbchar_t mbc); 132 extern bool mb_islower (const mbchar_t mbc); 133 extern bool mb_isprint (const mbchar_t mbc); 134 extern bool mb_ispunct (const mbchar_t mbc); 135 extern bool mb_isspace (const mbchar_t mbc); 136 extern bool mb_isupper (const mbchar_t mbc); 137 extern bool mb_isxdigit (const mbchar_t mbc); 138 extern int mb_width (const mbchar_t mbc); 139 extern void mb_putc (const mbchar_t mbc, FILE *stream); 140 extern void mb_setascii (mbchar_t *new, char sc); 141 extern void mb_copy (mbchar_t *new, const mbchar_t *old); 142 */ 143 144 #ifndef _MBCHAR_H 145 #define _MBCHAR_H 1 146 147 #include <stdbool.h> 148 #include <string.h> 149 150 /* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before 151 <wchar.h>. 152 BSD/OS 4.1 has a bug: <stdio.h> and <time.h> must be included before 153 <wchar.h>. */ 154 #include <stdio.h> 155 #include <time.h> 156 #include <wchar.h> 157 #include <wctype.h> 158 159 #define MBCHAR_BUF_SIZE 24 160 161 struct mbchar 162 { 163 const char *ptr; /* pointer to current character */ 164 size_t bytes; /* number of bytes of current character, > 0 */ 165 bool wc_valid; /* true if wc is a valid wide character */ 166 wchar_t wc; /* if wc_valid: the current character */ 167 char buf[MBCHAR_BUF_SIZE]; /* room for the bytes, used for file input only */ 168 }; 169 170 /* EOF (not a real character) is represented with bytes = 0 and 171 wc_valid = false. */ 172 173 typedef struct mbchar mbchar_t; 174 175 /* Access the current character. */ 176 #define mb_ptr(mbc) ((mbc).ptr) 177 #define mb_len(mbc) ((mbc).bytes) 178 179 /* Comparison of characters. */ 180 #define mb_iseq(mbc, sc) ((mbc).wc_valid && (mbc).wc == (sc)) 181 #define mb_isnul(mbc) ((mbc).wc_valid && (mbc).wc == 0) 182 #define mb_cmp(mbc1, mbc2) \ 183 ((mbc1).wc_valid \ 184 ? ((mbc2).wc_valid \ 185 ? (int) (mbc1).wc - (int) (mbc2).wc \ 186 : -1) \ 187 : ((mbc2).wc_valid \ 188 ? 1 \ 189 : (mbc1).bytes == (mbc2).bytes \ 190 ? memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) \ 191 : (mbc1).bytes < (mbc2).bytes \ 192 ? (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) > 0 ? 1 : -1) \ 193 : (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc2).bytes) >= 0 ? 1 : -1))) 194 #define mb_casecmp(mbc1, mbc2) \ 195 ((mbc1).wc_valid \ 196 ? ((mbc2).wc_valid \ 197 ? (int) towlower ((mbc1).wc) - (int) towlower ((mbc2).wc) \ 198 : -1) \ 199 : ((mbc2).wc_valid \ 200 ? 1 \ 201 : (mbc1).bytes == (mbc2).bytes \ 202 ? memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) \ 203 : (mbc1).bytes < (mbc2).bytes \ 204 ? (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) > 0 ? 1 : -1) \ 205 : (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc2).bytes) >= 0 ? 1 : -1))) 206 #define mb_equal(mbc1, mbc2) \ 207 ((mbc1).wc_valid && (mbc2).wc_valid \ 208 ? (mbc1).wc == (mbc2).wc \ 209 : (mbc1).bytes == (mbc2).bytes \ 210 && memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) == 0) 211 #define mb_caseequal(mbc1, mbc2) \ 212 ((mbc1).wc_valid && (mbc2).wc_valid \ 213 ? towlower ((mbc1).wc) == towlower ((mbc2).wc) \ 214 : (mbc1).bytes == (mbc2).bytes \ 215 && memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) == 0) 216 217 /* <ctype.h>, <wctype.h> classification. */ 218 #define mb_isascii(mbc) \ 219 ((mbc).wc_valid && (mbc).wc >= 0 && (mbc).wc <= 127) 220 #define mb_isalnum(mbc) ((mbc).wc_valid && iswalnum ((mbc).wc)) 221 #define mb_isalpha(mbc) ((mbc).wc_valid && iswalpha ((mbc).wc)) 222 #define mb_isblank(mbc) ((mbc).wc_valid && iswblank ((mbc).wc)) 223 #define mb_iscntrl(mbc) ((mbc).wc_valid && iswcntrl ((mbc).wc)) 224 #define mb_isdigit(mbc) ((mbc).wc_valid && iswdigit ((mbc).wc)) 225 #define mb_isgraph(mbc) ((mbc).wc_valid && iswgraph ((mbc).wc)) 226 #define mb_islower(mbc) ((mbc).wc_valid && iswlower ((mbc).wc)) 227 #define mb_isprint(mbc) ((mbc).wc_valid && iswprint ((mbc).wc)) 228 #define mb_ispunct(mbc) ((mbc).wc_valid && iswpunct ((mbc).wc)) 229 #define mb_isspace(mbc) ((mbc).wc_valid && iswspace ((mbc).wc)) 230 #define mb_isupper(mbc) ((mbc).wc_valid && iswupper ((mbc).wc)) 231 #define mb_isxdigit(mbc) ((mbc).wc_valid && iswxdigit ((mbc).wc)) 232 233 /* Extra <wchar.h> function. */ 234 235 /* Unprintable characters appear as a small box of width 1. */ 236 #define MB_UNPRINTABLE_WIDTH 1 237 238 static inline int 239 mb_width_aux (wint_t wc) 240 { 241 int w = wcwidth (wc); 242 /* For unprintable characters, arbitrarily return 0 for control characters 243 and MB_UNPRINTABLE_WIDTH otherwise. */ 244 return (w >= 0 ? w : iswcntrl (wc) ? 0 : MB_UNPRINTABLE_WIDTH); 245 } 246 247 #define mb_width(mbc) \ 248 ((mbc).wc_valid ? mb_width_aux ((mbc).wc) : MB_UNPRINTABLE_WIDTH) 249 250 /* Output. */ 251 #define mb_putc(mbc, stream) fwrite ((mbc).ptr, 1, (mbc).bytes, (stream)) 252 253 /* Assignment. */ 254 #define mb_setascii(mbc, sc) \ 255 ((mbc)->ptr = (mbc)->buf, (mbc)->bytes = 1, (mbc)->wc_valid = 1, \ 256 (mbc)->wc = (mbc)->buf[0] = (sc)) 257 258 /* Copying a character. */ 259 static inline void 260 mb_copy (mbchar_t *new_mbc, const mbchar_t *old_mbc) 261 { 262 if (old_mbc->ptr == &old_mbc->buf[0]) 263 { 264 memcpy (&new_mbc->buf[0], &old_mbc->buf[0], old_mbc->bytes); 265 new_mbc->ptr = &new_mbc->buf[0]; 266 } 267 else 268 new_mbc->ptr = old_mbc->ptr; 269 new_mbc->bytes = old_mbc->bytes; 270 if ((new_mbc->wc_valid = old_mbc->wc_valid)) 271 new_mbc->wc = old_mbc->wc; 272 } 273 274 275 /* is_basic(c) tests whether the single-byte character c is in the 276 ISO C "basic character set". 277 This is a convenience function, and is in this file only to share code 278 between mbiter_multi.h and mbfile_multi.h. */ 279 #if (' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \ 280 && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \ 281 && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \ 282 && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \ 283 && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \ 284 && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \ 285 && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \ 286 && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \ 287 && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \ 288 && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \ 289 && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \ 290 && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \ 291 && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \ 292 && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \ 293 && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \ 294 && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \ 295 && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \ 296 && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \ 297 && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \ 298 && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \ 299 && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \ 300 && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \ 301 && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126) 302 /* The character set is ISO-646, not EBCDIC. */ 303 # define IS_BASIC_ASCII 1 304 305 extern const unsigned int is_basic_table[]; 306 307 static inline bool 308 is_basic (char c) 309 { 310 return (is_basic_table [(unsigned char) c >> 5] >> ((unsigned char) c & 31)) 311 & 1; 312 } 313 314 #else 315 316 static inline bool 317 is_basic (char c) 318 { 319 switch (c) 320 { 321 case '\t': case '\v': case '\f': 322 case ' ': case '!': case '"': case '#': case '%': 323 case '&': case '\'': case '(': case ')': case '*': 324 case '+': case ',': case '-': case '.': case '/': 325 case '0': case '1': case '2': case '3': case '4': 326 case '5': case '6': case '7': case '8': case '9': 327 case ':': case ';': case '<': case '=': case '>': 328 case '?': 329 case 'A': case 'B': case 'C': case 'D': case 'E': 330 case 'F': case 'G': case 'H': case 'I': case 'J': 331 case 'K': case 'L': case 'M': case 'N': case 'O': 332 case 'P': case 'Q': case 'R': case 'S': case 'T': 333 case 'U': case 'V': case 'W': case 'X': case 'Y': 334 case 'Z': 335 case '[': case '\\': case ']': case '^': case '_': 336 case 'a': case 'b': case 'c': case 'd': case 'e': 337 case 'f': case 'g': case 'h': case 'i': case 'j': 338 case 'k': case 'l': case 'm': case 'n': case 'o': 339 case 'p': case 'q': case 'r': case 's': case 't': 340 case 'u': case 'v': case 'w': case 'x': case 'y': 341 case 'z': case '{': case '|': case '}': case '~': 342 return 1; 343 default: 344 return 0; 345 } 346 } 347 348 #endif 349 350 #endif /* _MBCHAR_H */ 351