1 /* Test of conversion of multibyte character to wide character. 2 Copyright (C) 2008-2011 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>, 2008. */ 18 19 #undef NDEBUG 20 #include <assert.h> 21 #include <locale.h> 22 #include <stdio.h> 23 #include <string.h> 24 #include <wchar.h> 25 26 #include <Debug.h> 27 28 int 29 main (int argc, char *argv[]) 30 { 31 mbstate_t state; 32 wchar_t wc; 33 size_t ret; 34 int i; 35 36 /* configure should already have checked that the locale is supported. */ 37 if (setlocale (LC_ALL, "") == NULL) { 38 fprintf(stderr, "unable to set standard locale\n"); 39 return 1; 40 } 41 42 /* Test zero-length input. */ 43 printf("zero-length input ...\n"); 44 { 45 memset (&state, '\0', sizeof (mbstate_t)); 46 wc = (wchar_t) 0xBADFACE; 47 ret = mbrtowc (&wc, "x", 0, &state); 48 /* gnulib's implementation returns (size_t)(-2). 49 The AIX 5.1 implementation returns (size_t)(-1). 50 glibc's implementation returns 0. */ 51 assert (ret == (size_t)(-2) || ret == (size_t)(-1) || ret == 0); 52 assert (mbsinit (&state)); 53 } 54 55 /* Test NUL byte input. */ 56 printf("NUL byte input ...\n"); 57 { 58 memset (&state, '\0', sizeof (mbstate_t)); 59 wc = (wchar_t) 0xBADFACE; 60 ret = mbrtowc (&wc, "", 1, &state); 61 assert (ret == 0); 62 assert (wc == 0); 63 assert (mbsinit (&state)); 64 ret = mbrtowc (NULL, "", 1, &state); 65 assert (ret == 0); 66 assert (mbsinit (&state)); 67 } 68 69 /* Test single-byte input. */ 70 printf("single-byte input ...\n"); 71 { 72 char buf[1]; 73 int c; 74 75 memset (&state, '\0', sizeof (mbstate_t)); 76 for (c = 0; c < 0x100; c++) 77 switch (c) 78 { 79 case '\t': case '\v': case '\f': 80 case ' ': case '!': case '"': case '#': case '%': 81 case '&': case '\'': case '(': case ')': case '*': 82 case '+': case ',': case '-': case '.': case '/': 83 case '0': case '1': case '2': case '3': case '4': 84 case '5': case '6': case '7': case '8': case '9': 85 case ':': case ';': case '<': case '=': case '>': 86 case '?': 87 case 'A': case 'B': case 'C': case 'D': case 'E': 88 case 'F': case 'G': case 'H': case 'I': case 'J': 89 case 'K': case 'L': case 'M': case 'N': case 'O': 90 case 'P': case 'Q': case 'R': case 'S': case 'T': 91 case 'U': case 'V': case 'W': case 'X': case 'Y': 92 case 'Z': 93 case '[': case '\\': case ']': case '^': case '_': 94 case 'a': case 'b': case 'c': case 'd': case 'e': 95 case 'f': case 'g': case 'h': case 'i': case 'j': 96 case 'k': case 'l': case 'm': case 'n': case 'o': 97 case 'p': case 'q': case 'r': case 's': case 't': 98 case 'u': case 'v': case 'w': case 'x': case 'y': 99 case 'z': case '{': case '|': case '}': case '~': 100 /* c is in the ISO C "basic character set". */ 101 buf[0] = c; 102 wc = (wchar_t) 0xBADFACE; 103 ret = mbrtowc (&wc, buf, 1, &state); 104 assert (ret == 1); 105 assert (wc == c); 106 assert (mbsinit (&state)); 107 ret = mbrtowc (NULL, buf, 1, &state); 108 assert (ret == 1); 109 assert (mbsinit (&state)); 110 break; 111 } 112 } 113 114 /* Test special calling convention, passing a NULL pointer. */ 115 printf("special calling convention, passing NULL ...\n"); 116 { 117 memset (&state, '\0', sizeof (mbstate_t)); 118 wc = (wchar_t) 0xBADFACE; 119 ret = mbrtowc (&wc, NULL, 5, &state); 120 assert (ret == 0); 121 assert (wc == (wchar_t) 0xBADFACE); 122 assert (mbsinit (&state)); 123 } 124 125 /* Check a 4-bytes character. */ 126 { 127 memset (&state, '\0', sizeof (mbstate_t)); 128 129 wc = (wchar_t) 0xBADFACE; 130 ret = mbrtowc (&wc, "\360", 1, &state); 131 assert (ret == (size_t)(-2)); 132 assert (wc == (wchar_t) 0xBADFACE); 133 assert (!mbsinit (&state)); 134 135 wc = (wchar_t) 0xBADFACE; 136 ret = mbrtowc (&wc, "\237", 1, &state); 137 assert (ret == (size_t)(-2)); 138 assert (wc == (wchar_t) 0xBADFACE); 139 assert (!mbsinit (&state)); 140 141 wc = (wchar_t) 0xBADFACE; 142 ret = mbrtowc (&wc, "\220", 1, &state); 143 assert (ret == (size_t)(-2)); 144 assert (wc == (wchar_t) 0xBADFACE); 145 assert (!mbsinit (&state)); 146 147 wc = (wchar_t) 0xBADFACE; 148 ret = mbrtowc (&wc, "\203", 1, &state); 149 assert (ret == 1); 150 assert (wc == (wchar_t) 0x1F403); 151 assert (mbsinit (&state)); 152 } 153 154 for (i = '1'; i <= '4'; ++i) { 155 switch (i) 156 { 157 case '1': 158 /* Locale encoding is ISO-8859-1 or ISO-8859-15. */ 159 printf("ISO8859-1 ...\n"); 160 { 161 char input[] = "B\374\337er"; /* "Büßer" */ 162 memset (&state, '\0', sizeof (mbstate_t)); 163 164 if (setlocale (LC_ALL, "en_US.ISO8859-1") == NULL) { 165 fprintf(stderr, "unable to set ISO8859-1 locale, skipping\n"); 166 break; 167 } 168 169 wc = (wchar_t) 0xBADFACE; 170 ret = mbrtowc (&wc, input, 1, &state); 171 assert (ret == 1); 172 assert (wc == 'B'); 173 assert (mbsinit (&state)); 174 input[0] = '\0'; 175 176 wc = (wchar_t) 0xBADFACE; 177 ret = mbrtowc (&wc, input + 1, 1, &state); 178 assert (ret == 1); 179 assert (wctob (wc) == (unsigned char) '\374'); 180 assert (mbsinit (&state)); 181 input[1] = '\0'; 182 183 /* Test support of NULL first argument. */ 184 ret = mbrtowc (NULL, input + 2, 3, &state); 185 assert (ret == 1); 186 assert (mbsinit (&state)); 187 188 wc = (wchar_t) 0xBADFACE; 189 ret = mbrtowc (&wc, input + 2, 3, &state); 190 assert (ret == 1); 191 assert (wctob (wc) == (unsigned char) '\337'); 192 assert (mbsinit (&state)); 193 input[2] = '\0'; 194 195 wc = (wchar_t) 0xBADFACE; 196 ret = mbrtowc (&wc, input + 3, 2, &state); 197 assert (ret == 1); 198 assert (wc == 'e'); 199 assert (mbsinit (&state)); 200 input[3] = '\0'; 201 202 wc = (wchar_t) 0xBADFACE; 203 ret = mbrtowc (&wc, input + 4, 1, &state); 204 assert (ret == 1); 205 assert (wc == 'r'); 206 assert (mbsinit (&state)); 207 } 208 break; 209 210 case '2': 211 /* Locale encoding is UTF-8. */ 212 printf("UTF-8 ...\n"); 213 { 214 char input[] = "B\303\274\303\237er"; /* "Büßer" */ 215 memset (&state, '\0', sizeof (mbstate_t)); 216 217 if (setlocale (LC_ALL, "en_US.UTF-8") == NULL) { 218 fprintf(stderr, "unable to set UTF-8 locale, skipping\n"); 219 break; 220 } 221 222 wc = (wchar_t) 0xBADFACE; 223 ret = mbrtowc (&wc, input, 1, &state); 224 assert (ret == 1); 225 assert (wc == 'B'); 226 assert (mbsinit (&state)); 227 input[0] = '\0'; 228 229 wc = (wchar_t) 0xBADFACE; 230 ret = mbrtowc (&wc, input + 1, 1, &state); 231 assert (ret == (size_t)(-2)); 232 assert (wc == (wchar_t) 0xBADFACE); 233 assert (!mbsinit (&state)); 234 input[1] = '\0'; 235 236 wc = (wchar_t) 0xBADFACE; 237 ret = mbrtowc (&wc, input + 2, 5, &state); 238 assert (ret == 1); 239 assert (wctob (wc) == EOF); 240 assert (mbsinit (&state)); 241 input[2] = '\0'; 242 243 /* Test support of NULL first argument. */ 244 ret = mbrtowc (NULL, input + 3, 4, &state); 245 assert (ret == 2); 246 assert (mbsinit (&state)); 247 248 wc = (wchar_t) 0xBADFACE; 249 ret = mbrtowc (&wc, input + 3, 4, &state); 250 assert (ret == 2); 251 assert (wctob (wc) == EOF); 252 assert (mbsinit (&state)); 253 input[3] = '\0'; 254 input[4] = '\0'; 255 256 wc = (wchar_t) 0xBADFACE; 257 ret = mbrtowc (&wc, input + 5, 2, &state); 258 assert (ret == 1); 259 assert (wc == 'e'); 260 assert (mbsinit (&state)); 261 input[5] = '\0'; 262 263 wc = (wchar_t) 0xBADFACE; 264 ret = mbrtowc (&wc, input + 6, 1, &state); 265 assert (ret == 1); 266 assert (wc == 'r'); 267 assert (mbsinit (&state)); 268 269 /* reproduce a valid use case from readline (as used in our bash): */ 270 { 271 char tooShort[] = "\303"; 272 char ok[] = "\303\274"; 273 /* make a backup of the state */ 274 mbstate_t stateBackup = state; 275 /* try with a source that's too short */ 276 ret = mbrtowc (&wc, tooShort, 1, &state); 277 assert (ret == (size_t)-2); 278 /* restore the state from the backup */ 279 state = stateBackup; 280 /* retry with enough source */ 281 ret = mbrtowc (&wc, ok, 2, &state); 282 assert (ret == 2); 283 } 284 } 285 break; 286 287 case '3': 288 /* Locale encoding is EUC-JP. */ 289 printf("EUC-JP ...\n"); 290 { 291 char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */ 292 memset (&state, '\0', sizeof (mbstate_t)); 293 294 if (setlocale (LC_ALL, "en_US.EUC-JP") == NULL) { 295 fprintf(stderr, "unable to set EUC-JP locale, skipping\n"); 296 break; 297 } 298 299 wc = (wchar_t) 0xBADFACE; 300 ret = mbrtowc (&wc, input, 1, &state); 301 assert (ret == 1); 302 assert (wc == '<'); 303 assert (mbsinit (&state)); 304 input[0] = '\0'; 305 306 wc = (wchar_t) 0xBADFACE; 307 ret = mbrtowc (&wc, input + 1, 2, &state); 308 assert (ret == 2); 309 assert (wctob (wc) == EOF); 310 assert (mbsinit (&state)); 311 input[1] = '\0'; 312 input[2] = '\0'; 313 314 wc = (wchar_t) 0xBADFACE; 315 ret = mbrtowc (&wc, input + 3, 1, &state); 316 assert (ret == (size_t)(-2)); 317 assert (wc == (wchar_t) 0xBADFACE); 318 assert (!mbsinit (&state)); 319 input[3] = '\0'; 320 321 wc = (wchar_t) 0xBADFACE; 322 ret = mbrtowc (&wc, input + 4, 4, &state); 323 assert (ret == 1); 324 assert (wctob (wc) == EOF); 325 assert (mbsinit (&state)); 326 input[4] = '\0'; 327 328 /* Test support of NULL first argument. */ 329 ret = mbrtowc (NULL, input + 5, 3, &state); 330 assert (ret == 2); 331 assert (mbsinit (&state)); 332 333 wc = (wchar_t) 0xBADFACE; 334 ret = mbrtowc (&wc, input + 5, 3, &state); 335 assert (ret == 2); 336 assert (wctob (wc) == EOF); 337 assert (mbsinit (&state)); 338 input[5] = '\0'; 339 input[6] = '\0'; 340 341 wc = (wchar_t) 0xBADFACE; 342 ret = mbrtowc (&wc, input + 7, 1, &state); 343 assert (ret == 1); 344 assert (wc == '>'); 345 assert (mbsinit (&state)); 346 } 347 break; 348 349 case '4': 350 /* Locale encoding is GB18030. */ 351 printf("GB18030 ...\n"); 352 { 353 char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */ 354 memset (&state, '\0', sizeof (mbstate_t)); 355 356 if (setlocale (LC_ALL, "en_US.GB18030") == NULL) { 357 fprintf(stderr, "unable to set GB18030 locale, skipping\n"); 358 break; 359 } 360 361 wc = (wchar_t) 0xBADFACE; 362 ret = mbrtowc (&wc, input, 1, &state); 363 assert (ret == 1); 364 assert (wc == 'B'); 365 assert (mbsinit (&state)); 366 input[0] = '\0'; 367 368 wc = (wchar_t) 0xBADFACE; 369 ret = mbrtowc (&wc, input + 1, 1, &state); 370 assert (ret == (size_t)(-2)); 371 assert (wc == (wchar_t) 0xBADFACE); 372 assert (!mbsinit (&state)); 373 input[1] = '\0'; 374 375 wc = (wchar_t) 0xBADFACE; 376 ret = mbrtowc (&wc, input + 2, 7, &state); 377 assert (ret == 1); 378 assert (wctob (wc) == EOF); 379 assert (mbsinit (&state)); 380 input[2] = '\0'; 381 382 /* Test support of NULL first argument. */ 383 ret = mbrtowc (NULL, input + 3, 6, &state); 384 assert (ret == 4); 385 assert (mbsinit (&state)); 386 387 wc = (wchar_t) 0xBADFACE; 388 ret = mbrtowc (&wc, input + 3, 6, &state); 389 assert (ret == 4); 390 assert (wctob (wc) == EOF); 391 assert (mbsinit (&state)); 392 input[3] = '\0'; 393 input[4] = '\0'; 394 input[5] = '\0'; 395 input[6] = '\0'; 396 397 wc = (wchar_t) 0xBADFACE; 398 ret = mbrtowc (&wc, input + 7, 2, &state); 399 assert (ret == 1); 400 assert (wc == 'e'); 401 assert (mbsinit (&state)); 402 input[5] = '\0'; 403 404 wc = (wchar_t) 0xBADFACE; 405 ret = mbrtowc (&wc, input + 8, 1, &state); 406 assert (ret == 1); 407 assert (wc == 'r'); 408 assert (mbsinit (&state)); 409 } 410 break; 411 } 412 } 413 414 return 0; 415 } 416