tests/test-mbmemcasecmp.h

   1 /* Test of case-insensitive memory area comparison function.
   2    Copyright (C) 2007-2024 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 <https://www.gnu.org/licenses/>.  */
  16
  17 /* Written by Bruno Haible <bruno@clisp.org>, 2009.  */
  18
  19 static void
  20 test_ascii (int (*my_casecmp) (const char *, size_t, const char *, size_t))
  21 {
  22   /* Empty string.  */
  23   {
  24     ASSERT (my_casecmp (NULL, 0, NULL, 0) == 0);
  25   }
  26   {
  27     static const char input[] = { 'x', 'y' };
  28
  29     ASSERT (my_casecmp (input, SIZEOF (input), NULL, 0) > 0);
  30
  31     ASSERT (my_casecmp (NULL, 0, input, SIZEOF (input)) < 0);
  32
  33     ASSERT (my_casecmp (input, SIZEOF (input), input, SIZEOF (input)) == 0);
  34   }
  35
  36   /* Normal lexicographic order.  */
  37   {
  38     static const char input1[] = { 'A', 'm', 'e', 'r', 'i', 'c', 'a' };
  39     static const char input2[] = { 'A', 'm', 'i', 'g', 'o' };
  40
  41     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) < 0);
  42
  43     ASSERT (my_casecmp (input2, SIZEOF (input2), input1, SIZEOF (input1)) > 0);
  44   }
  45
  46   /* Shorter and longer strings.  */
  47   {
  48     static const char input1[] = { 'R', 'e', 'a', 'g', 'a', 'n' };
  49     static const char input2[] = { 'R', 'e', 'a', 'g', 'a', 'n', 'o', 'm', 'i', 'c', 's' };
  50
  51     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) < 0);
  52
  53     ASSERT (my_casecmp (input2, SIZEOF (input2), input1, SIZEOF (input1)) > 0);
  54   }
  55
  56   ASSERT (my_casecmp ("paragraph", 9, "Paragraph", 9) == 0);
  57
  58   ASSERT (my_casecmp ("paragrapH", 9, "parAgRaph", 9) == 0);
  59
  60   ASSERT (my_casecmp ("paragraph", 9, "paraLyzed", 9) < 0);
  61   ASSERT (my_casecmp ("paraLyzed", 9, "paragraph", 9) > 0);
  62
  63   ASSERT (my_casecmp ("para", 4, "paragraph", 9) < 0);
  64   ASSERT (my_casecmp ("paragraph", 9, "para", 4) > 0);
  65
  66   /* Embedded NULs.  */
  67   ASSERT (my_casecmp ("1\0", 2, "2\0", 2) < 0);
  68   ASSERT (my_casecmp ("2\0", 2, "1\0", 2) > 0);
  69   ASSERT (my_casecmp ("x\0""1", 3, "x\0""2", 3) < 0);
  70   ASSERT (my_casecmp ("x\0""2", 3, "x\0""1", 3) > 0);
  71 }
  72
  73 static void
  74 test_iso_8859_1 (int (*my_casecmp) (const char *, size_t, const char *, size_t), bool assume_byte_values)
  75 {
  76   #if ! defined __osf__ /* This test fails on OSF/1 5.1.  */
  77   {
  78     static const char input1[] = { 'H', 0xF6, 'h', 'l', 'e' };
  79     static const char input2[] = { 'H', 0xD6, 'h', 'L', 'e' };
  80     static const char input3[] = { 'H', 0xF6, 'h', 'l', 'e', 'n' };
  81     static const char input4[] = { 'H', 0xD6, 'h', 'L', 'e', 'n' };
  82     static const char input5[] = { 'H', 'u', 'r', 'z' };
  83
  84     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
  85
  86     ASSERT (my_casecmp (input2, SIZEOF (input2), input1, SIZEOF (input1)) == 0);
  87
  88     ASSERT (my_casecmp (input3, SIZEOF (input3), input4, SIZEOF (input4)) == 0);
  89
  90     ASSERT (my_casecmp (input4, SIZEOF (input4), input3, SIZEOF (input3)) == 0);
  91
  92     ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) < 0);
  93
  94     ASSERT (my_casecmp (input1, SIZEOF (input1), input4, SIZEOF (input4)) < 0);
  95
  96     if (assume_byte_values)
  97       {
  98         /* These results are the opposite of those in test-ulc-casecmp.c,
  99            because my_casecmp compares precomposed characters, whereas
 100            ulc_casecmp compares decomposed character sequences.  */
 101
 102         ASSERT (my_casecmp (input1, SIZEOF (input1), input5, SIZEOF (input5)) > 0);
 103
 104         ASSERT (my_casecmp (input2, SIZEOF (input2), input5, SIZEOF (input5)) > 0);
 105       }
 106   }
 107   #endif
 108
 109   #if 0 /* This functionality requires ulc_casecmp.  */
 110   /* Uppercasing can increase the number of Unicode characters.  */
 111   { /* "heiß" */
 112     static const char input1[] = { 0x68, 0x65, 0x69, 0xDF };
 113     static const char input2[] = { 0x68, 0x65, 0x69, 0x73, 0x73 };
 114
 115     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 116   }
 117   #endif
 118 }
 119
 120 static void
 121 test_utf_8 (int (*my_casecmp) (const char *, size_t, const char *, size_t), bool turkish)
 122 {
 123   /* The following tests shows how my_casecmp() is different from
 124      strcasecmp().  */
 125
 126   {
 127     /* Some platforms, e.g. MSVC 14, lack the upper/lower mappings for these
 128        wide characters in the *.65001 locales.  */
 129     mbstate_t state;
 130     wchar_t wc;
 131     memset (&state, 0, sizeof (mbstate_t));
 132     if (mbrtowc (&wc, "\303\274", 2, &state) == 2
 133         && towupper (wc) != wc)
 134       {
 135         ASSERT (my_casecmp ("\303\266zg\303\274r", 7, "\303\226ZG\303\234R", 7) == 0); /* özgür */
 136         ASSERT (my_casecmp ("\303\226ZG\303\234R", 7, "\303\266zg\303\274r", 7) == 0); /* özgür */
 137       }
 138   }
 139
 140   /* This test shows how strings of different size can compare equal.  */
 141   if (turkish)
 142     {
 143       ASSERT (my_casecmp ("turkish", 7, "TURK\304\260SH", 8) == 0);
 144       ASSERT (my_casecmp ("TURK\304\260SH", 8, "turkish", 7) == 0);
 145     }
 146
 147   #if 0 /* This functionality requires ulc_casecmp.  */
 148   /* Normalization effects.  */
 149   {
 150     static const char input1[] = { 'H', 0xC3, 0xB6, 'h', 'l', 'e' };
 151     static const char input2[] = { 'H', 'O', 0xCC, 0x88, 'h', 'L', 'e' };
 152     static const char input3[] = { 'H', 0xC3, 0xB6, 'h', 'l', 'e', 'n' };
 153     static const char input4[] = { 'H', 'O', 0xCC, 0x88, 'h', 'L', 'e', 'n' };
 154     static const char input5[] = { 'H', 'u', 'r', 'z' };
 155
 156     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 157
 158     ASSERT (my_casecmp (input2, SIZEOF (input2), input1, SIZEOF (input1)) == 0);
 159
 160     ASSERT (my_casecmp (input3, SIZEOF (input3), input4, SIZEOF (input4)) == 0);
 161
 162     ASSERT (my_casecmp (input4, SIZEOF (input4), input3, SIZEOF (input3)) == 0);
 163
 164     ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) < 0);
 165
 166     ASSERT (my_casecmp (input1, SIZEOF (input1), input4, SIZEOF (input4)) < 0);
 167
 168     ASSERT (my_casecmp (input1, SIZEOF (input1), input5, SIZEOF (input5)) < 0);
 169
 170     ASSERT (my_casecmp (input2, SIZEOF (input2), input5, SIZEOF (input5)) < 0);
 171   }
 172   { /* LATIN CAPITAL LETTER A WITH DIAERESIS */
 173     static const char input1[] = { 0xC3, 0x84 };
 174     static const char input2[] = { 0x41, 0xCC, 0x88 };
 175
 176     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 177   }
 178   { /* LATIN CAPITAL LETTER A WITH DIAERESIS AND MACRON */
 179     static const char input1[] = { 0xC7, 0x9E };
 180     static const char input2[] = { 0x41, 0xCC, 0x88, 0xCC, 0x84 };
 181
 182     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 183   }
 184   { /* GREEK DIALYTIKA AND PERISPOMENI */
 185     static const char input1[] = { 0xE1, 0xBF, 0x81 };
 186     static const char input2[] = { 0xC2, 0xA8, 0xCD, 0x82 };
 187
 188     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 189   }
 190   { /* HANGUL SYLLABLE GEUL */
 191     static const char input1[] = { 0xEA, 0xB8, 0x80 };
 192     static const char input2[] = { 0xEA, 0xB7, 0xB8, 0xE1, 0x86, 0xAF };
 193     static const char input3[] = { 0xE1, 0x84, 0x80, 0xE1, 0x85, 0xB3, 0xE1, 0x86, 0xAF };
 194
 195     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 196
 197     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 198   }
 199   { /* HANGUL SYLLABLE GEU */
 200     static const char input1[] = { 0xEA, 0xB7, 0xB8 };
 201     static const char input2[] = { 0xE1, 0x84, 0x80, 0xE1, 0x85, 0xB3 };
 202
 203     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 204   }
 205   #endif
 206
 207   /* Simple string.  */
 208   { /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a)  日本語,中文,한글" */
 209     static const char input1[] =
 210       { 'G', 'r', 0xC3, 0xBC, 0xC3, 0x9F, ' ', 'G', 'o', 't', 't', '.', ' ',
 211         0xD0, 0x97, 0xD0, 0xB4, 0xD1, 0x80, 0xD0, 0xB0, 0xD0, 0xB2, 0xD1, 0x81,
 212         0xD1, 0x82, 0xD0, 0xB2, 0xD1, 0x83, 0xD0, 0xB9, 0xD1, 0x82, 0xD0, 0xB5,
 213         '!', ' ', 'x', '=', '(', '-', 'b', 0xC2, 0xB1, 's', 'q', 'r', 't', '(',
 214         'b', 0xC2, 0xB2, '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')',
 215         ' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
 216         0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
 217         0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\n'
 218       };
 219     static const char input2[] =
 220       { 'g', 'r', 0xC3, 0xBC, 0x73, 0x73, ' ', 'g', 'o', 't', 't', '.', ' ',
 221         0xD0, 0xB7, 0xD0, 0xB4, 0xD1, 0x80, 0xD0, 0xB0, 0xD0, 0xB2, 0xD1, 0x81,
 222         0xD1, 0x82, 0xD0, 0xB2, 0xD1, 0x83, 0xD0, 0xB9, 0xD1, 0x82, 0xD0, 0xB5,
 223         '!', ' ', 'x', '=', '(', '-', 'b', 0xC2, 0xB1, 's', 'q', 'r', 't', '(',
 224         'b', 0xC2, 0xB2, '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')',
 225         ' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
 226         0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
 227         0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\n'
 228       };
 229     static const char input3[] =
 230       { 'G', 'R', 0xC3, 0x9C, 0x53, 0x53, ' ', 'G', 'O', 'T', 'T', '.', ' ',
 231         0xD0, 0x97, 0xD0, 0x94, 0xD0, 0xA0, 0xD0, 0x90, 0xD0, 0x92, 0xD0, 0xA1,
 232         0xD0, 0xA2, 0xD0, 0x92, 0xD0, 0xA3, 0xD0, 0x99, 0xD0, 0xA2, 0xD0, 0x95,
 233         '!', ' ', 'X', '=', '(', '-', 'B', 0xC2, 0xB1, 'S', 'Q', 'R', 'T', '(',
 234         'B', 0xC2, 0xB2, '-', '4', 'A', 'C', ')', ')', '/', '(', '2', 'A', ')',
 235         ' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
 236         0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
 237         0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\n'
 238       };
 239
 240     (void) input1;
 241
 242     #if 0 /* This functionality requires ulc_casecmp.  */
 243     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 244
 245     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 246     #endif
 247
 248     {
 249       /* Some platforms, e.g. MSVC 14, lack the upper/lower mappings for the
 250          'ü'/'Ü' wide characters in the *.65001 locales.  */
 251       mbstate_t state;
 252       wchar_t wc;
 253       memset (&state, 0, sizeof (mbstate_t));
 254       if (mbrtowc (&wc, "\303\234", 2, &state) == 2
 255           && towlower (wc) != wc)
 256         {
 257           ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) == 0);
 258         }
 259     }
 260   }
 261
 262   #if 0 /* This functionality requires ulc_casecmp.  */
 263   /* Case mapping can increase the number of Unicode characters.  */
 264   { /* LATIN SMALL LETTER N PRECEDED BY APOSTROPHE */
 265     static const char input1[] = { 0xC5, 0x89 };
 266     static const char input2[] = { 0xCA, 0xBC, 0x6E };
 267     static const char input3[] = { 0xCA, 0xBC, 0x4E };
 268
 269     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 270
 271     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 272   }
 273   { /* GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS */
 274     static const char input1[] = { 0xCE, 0x90 };
 275     static const char input2[] = { 0xCE, 0xB9, 0xCC, 0x88, 0xCC, 0x81 };
 276
 277     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 278   }
 279   #endif
 280
 281   /* Turkish letters i İ ı I */
 282   { /* LATIN CAPITAL LETTER I */
 283     static const char input[]         = { 0x49 };
 284     static const char casefolded[]    = { 0x69 };
 285     static const char casefolded_tr[] = { 0xC4, 0xB1 };
 286
 287     if (!turkish)
 288       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 289     else
 290       ASSERT (my_casecmp (input, SIZEOF (input), casefolded_tr, SIZEOF (casefolded_tr)) == 0);
 291   }
 292   { /* LATIN SMALL LETTER I */
 293     static const char input[]         = { 0x69 };
 294     static const char casefolded[]    = { 0x49 };
 295     static const char casefolded_tr[] = { 0xC4, 0xB0 };
 296
 297     if (!turkish)
 298       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 299     else
 300       ASSERT (my_casecmp (input, SIZEOF (input), casefolded_tr, SIZEOF (casefolded_tr)) == 0);
 301   }
 302   { /* LATIN CAPITAL LETTER I WITH DOT ABOVE */
 303     static const char input[]         = { 0xC4, 0xB0 };
 304     static const char casefolded[]    = { 0x69, 0xCC, 0x87 };
 305     static const char casefolded_tr[] = { 0x69 };
 306
 307     (void) casefolded;
 308
 309     if (!turkish)
 310       {
 311         #if 0 /* This functionality requires ulc_casecmp.  */
 312         ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 313         #endif
 314       }
 315     else
 316       ASSERT (my_casecmp (input, SIZEOF (input), casefolded_tr, SIZEOF (casefolded_tr)) == 0);
 317   }
 318   { /* LATIN SMALL LETTER DOTLESS I */
 319     static const char input[]      = { 0xC4, 0xB1 };
 320     static const char casefolded[] = { 0x49 };
 321
 322     if (!turkish)
 323       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) > 0);
 324     else
 325       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 326   }
 327   { /* "topkapı" */
 328     static const char input[] =
 329       { 0x54, 0x4F, 0x50, 0x4B, 0x41, 0x50, 0x49 };
 330     static const char casefolded[] =
 331       { 0x74, 0x6F, 0x70, 0x6B, 0x61, 0x70, 0xC4, 0xB1 };
 332
 333     if (!turkish)
 334       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) < 0);
 335     else
 336       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 337   }
 338
 339   #if 0 /* This functionality requires ulc_casecmp.  */
 340   /* Uppercasing can increase the number of Unicode characters.  */
 341   { /* "heiß" */
 342     static const char input1[] = { 0x68, 0x65, 0x69, 0xC3, 0x9F };
 343     static const char input2[] = { 0x68, 0x65, 0x69, 0x73, 0x73 };
 344
 345     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 346   }
 347   #endif
 348
 349   /* Case mappings for some characters can depend on the surrounding characters.  */
 350   { /* "περισσότερες πληροφορίες" */
 351     static const char input1[] =
 352       {
 353         0xCF, 0x80, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB9, 0xCF, 0x83, 0xCF, 0x83,
 354         0xCF, 0x8C, 0xCF, 0x84, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB5, 0xCF, 0x82,
 355         ' ', 0xCF, 0x80, 0xCE, 0xBB, 0xCE, 0xB7, 0xCF, 0x81, 0xCE, 0xBF,
 356         0xCF, 0x86, 0xCE, 0xBF, 0xCF, 0x81, 0xCE, 0xAF, 0xCE, 0xB5, 0xCF, 0x82
 357       };
 358     static const char input2[] =
 359       {
 360         0xCF, 0x80, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB9, 0xCF, 0x83, 0xCF, 0x83,
 361         0xCF, 0x8C, 0xCF, 0x84, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB5, 0xCF, 0x83,
 362         ' ', 0xCF, 0x80, 0xCE, 0xBB, 0xCE, 0xB7, 0xCF, 0x81, 0xCE, 0xBF,
 363         0xCF, 0x86, 0xCE, 0xBF, 0xCF, 0x81, 0xCE, 0xAF, 0xCE, 0xB5, 0xCF, 0x83
 364       };
 365     static const char input3[] =
 366       {
 367         0xCE, 0xA0, 0xCE, 0x95, 0xCE, 0xA1, 0xCE, 0x99, 0xCE, 0xA3, 0xCE, 0xA3,
 368         0xCE, 0x8C, 0xCE, 0xA4, 0xCE, 0x95, 0xCE, 0xA1, 0xCE, 0x95, 0xCE, 0xA3,
 369         ' ', 0xCE, 0xA0, 0xCE, 0x9B, 0xCE, 0x97, 0xCE, 0xA1, 0xCE, 0x9F,
 370         0xCE, 0xA6, 0xCE, 0x9F, 0xCE, 0xA1, 0xCE, 0x8A, 0xCE, 0x95, 0xCE, 0xA3
 371       };
 372
 373     (void) input1;
 374
 375     #if 0 /* This functionality requires ulc_casecmp.  */
 376     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 377
 378     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 379     #endif
 380
 381     ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) == 0);
 382   }
 383
 384   #if 0 /* This functionality requires ulc_casecmp.  */
 385   /* Case mapping can require subsequent normalization.  */
 386   { /* LATIN SMALL LETTER J WITH CARON, COMBINING DOT BELOW */
 387     static const char input[]                 = { 0xC7, 0xB0, 0xCC, 0xA3 };
 388     static const char casefolded[]            = { 0x6A, 0xCC, 0x8C, 0xCC, 0xA3 };
 389     static const char casefolded_decomposed[] = { 0x6A, 0xCC, 0xA3, 0xCC, 0x8C };
 390
 391     ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 392
 393     ASSERT (my_casecmp (input, SIZEOF (input), casefolded_decomposed, SIZEOF (casefolded_decomposed)) != 0);
 394
 395     ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 396
 397     ASSERT (my_casecmp (input, SIZEOF (input), casefolded_decomposed, SIZEOF (casefolded_decomposed)) == 0);
 398   }
 399   #endif
 400 }