tests/test-mbmemcasecmp.h

   1 /* Test of case-insensitive memory area comparison function.
   2    Copyright (C) 2007-2021 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   ASSERT (my_casecmp ("\303\266zg\303\274r", 7, "\303\226ZG\303\234R", 7) == 0); /* özgür */
 127   ASSERT (my_casecmp ("\303\226ZG\303\234R", 7, "\303\266zg\303\274r", 7) == 0); /* özgür */
 128
 129   /* This test shows how strings of different size can compare equal.  */
 130   if (turkish)
 131     {
 132       ASSERT (my_casecmp ("turkish", 7, "TURK\304\260SH", 8) == 0);
 133       ASSERT (my_casecmp ("TURK\304\260SH", 8, "turkish", 7) == 0);
 134     }
 135
 136   #if 0 /* This functionality requires ulc_casecmp.  */
 137   /* Normalization effects.  */
 138   {
 139     static const char input1[] = { 'H', 0xC3, 0xB6, 'h', 'l', 'e' };
 140     static const char input2[] = { 'H', 'O', 0xCC, 0x88, 'h', 'L', 'e' };
 141     static const char input3[] = { 'H', 0xC3, 0xB6, 'h', 'l', 'e', 'n' };
 142     static const char input4[] = { 'H', 'O', 0xCC, 0x88, 'h', 'L', 'e', 'n' };
 143     static const char input5[] = { 'H', 'u', 'r', 'z' };
 144
 145     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 146
 147     ASSERT (my_casecmp (input2, SIZEOF (input2), input1, SIZEOF (input1)) == 0);
 148
 149     ASSERT (my_casecmp (input3, SIZEOF (input3), input4, SIZEOF (input4)) == 0);
 150
 151     ASSERT (my_casecmp (input4, SIZEOF (input4), input3, SIZEOF (input3)) == 0);
 152
 153     ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) < 0);
 154
 155     ASSERT (my_casecmp (input1, SIZEOF (input1), input4, SIZEOF (input4)) < 0);
 156
 157     ASSERT (my_casecmp (input1, SIZEOF (input1), input5, SIZEOF (input5)) < 0);
 158
 159     ASSERT (my_casecmp (input2, SIZEOF (input2), input5, SIZEOF (input5)) < 0);
 160   }
 161   { /* LATIN CAPITAL LETTER A WITH DIAERESIS */
 162     static const char input1[] = { 0xC3, 0x84 };
 163     static const char input2[] = { 0x41, 0xCC, 0x88 };
 164
 165     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 166   }
 167   { /* LATIN CAPITAL LETTER A WITH DIAERESIS AND MACRON */
 168     static const char input1[] = { 0xC7, 0x9E };
 169     static const char input2[] = { 0x41, 0xCC, 0x88, 0xCC, 0x84 };
 170
 171     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 172   }
 173   { /* GREEK DIALYTIKA AND PERISPOMENI */
 174     static const char input1[] = { 0xE1, 0xBF, 0x81 };
 175     static const char input2[] = { 0xC2, 0xA8, 0xCD, 0x82 };
 176
 177     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 178   }
 179   { /* HANGUL SYLLABLE GEUL */
 180     static const char input1[] = { 0xEA, 0xB8, 0x80 };
 181     static const char input2[] = { 0xEA, 0xB7, 0xB8, 0xE1, 0x86, 0xAF };
 182     static const char input3[] = { 0xE1, 0x84, 0x80, 0xE1, 0x85, 0xB3, 0xE1, 0x86, 0xAF };
 183
 184     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 185
 186     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 187   }
 188   { /* HANGUL SYLLABLE GEU */
 189     static const char input1[] = { 0xEA, 0xB7, 0xB8 };
 190     static const char input2[] = { 0xE1, 0x84, 0x80, 0xE1, 0x85, 0xB3 };
 191
 192     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 193   }
 194   #endif
 195
 196   /* Simple string.  */
 197   { /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a)  日本語,中文,한글" */
 198     static const char input1[] =
 199       { 'G', 'r', 0xC3, 0xBC, 0xC3, 0x9F, ' ', 'G', 'o', 't', 't', '.', ' ',
 200         0xD0, 0x97, 0xD0, 0xB4, 0xD1, 0x80, 0xD0, 0xB0, 0xD0, 0xB2, 0xD1, 0x81,
 201         0xD1, 0x82, 0xD0, 0xB2, 0xD1, 0x83, 0xD0, 0xB9, 0xD1, 0x82, 0xD0, 0xB5,
 202         '!', ' ', 'x', '=', '(', '-', 'b', 0xC2, 0xB1, 's', 'q', 'r', 't', '(',
 203         'b', 0xC2, 0xB2, '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')',
 204         ' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
 205         0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
 206         0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\n'
 207       };
 208     static const char input2[] =
 209       { 'g', 'r', 0xC3, 0xBC, 0x73, 0x73, ' ', 'g', 'o', 't', 't', '.', ' ',
 210         0xD0, 0xB7, 0xD0, 0xB4, 0xD1, 0x80, 0xD0, 0xB0, 0xD0, 0xB2, 0xD1, 0x81,
 211         0xD1, 0x82, 0xD0, 0xB2, 0xD1, 0x83, 0xD0, 0xB9, 0xD1, 0x82, 0xD0, 0xB5,
 212         '!', ' ', 'x', '=', '(', '-', 'b', 0xC2, 0xB1, 's', 'q', 'r', 't', '(',
 213         'b', 0xC2, 0xB2, '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')',
 214         ' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
 215         0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
 216         0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\n'
 217       };
 218     static const char input3[] =
 219       { 'G', 'R', 0xC3, 0x9C, 0x53, 0x53, ' ', 'G', 'O', 'T', 'T', '.', ' ',
 220         0xD0, 0x97, 0xD0, 0x94, 0xD0, 0xA0, 0xD0, 0x90, 0xD0, 0x92, 0xD0, 0xA1,
 221         0xD0, 0xA2, 0xD0, 0x92, 0xD0, 0xA3, 0xD0, 0x99, 0xD0, 0xA2, 0xD0, 0x95,
 222         '!', ' ', 'X', '=', '(', '-', 'B', 0xC2, 0xB1, 'S', 'Q', 'R', 'T', '(',
 223         'B', 0xC2, 0xB2, '-', '4', 'A', 'C', ')', ')', '/', '(', '2', 'A', ')',
 224         ' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
 225         0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
 226         0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\n'
 227       };
 228
 229     (void) input1;
 230
 231     #if 0 /* This functionality requires ulc_casecmp.  */
 232     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 233
 234     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 235     #endif
 236
 237     ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) == 0);
 238   }
 239
 240   #if 0 /* This functionality requires ulc_casecmp.  */
 241   /* Case mapping can increase the number of Unicode characters.  */
 242   { /* LATIN SMALL LETTER N PRECEDED BY APOSTROPHE */
 243     static const char input1[] = { 0xC5, 0x89 };
 244     static const char input2[] = { 0xCA, 0xBC, 0x6E };
 245     static const char input3[] = { 0xCA, 0xBC, 0x4E };
 246
 247     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 248
 249     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 250   }
 251   { /* GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS */
 252     static const char input1[] = { 0xCE, 0x90 };
 253     static const char input2[] = { 0xCE, 0xB9, 0xCC, 0x88, 0xCC, 0x81 };
 254
 255     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 256   }
 257   #endif
 258
 259   /* Turkish letters i İ ı I */
 260   { /* LATIN CAPITAL LETTER I */
 261     static const char input[]         = { 0x49 };
 262     static const char casefolded[]    = { 0x69 };
 263     static const char casefolded_tr[] = { 0xC4, 0xB1 };
 264
 265     if (!turkish)
 266       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 267     else
 268       ASSERT (my_casecmp (input, SIZEOF (input), casefolded_tr, SIZEOF (casefolded_tr)) == 0);
 269   }
 270   { /* LATIN SMALL LETTER I */
 271     static const char input[]         = { 0x69 };
 272     static const char casefolded[]    = { 0x49 };
 273     static const char casefolded_tr[] = { 0xC4, 0xB0 };
 274
 275     if (!turkish)
 276       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 277     else
 278       ASSERT (my_casecmp (input, SIZEOF (input), casefolded_tr, SIZEOF (casefolded_tr)) == 0);
 279   }
 280   { /* LATIN CAPITAL LETTER I WITH DOT ABOVE */
 281     static const char input[]         = { 0xC4, 0xB0 };
 282     static const char casefolded[]    = { 0x69, 0xCC, 0x87 };
 283     static const char casefolded_tr[] = { 0x69 };
 284
 285     (void) casefolded;
 286
 287     if (!turkish)
 288       {
 289         #if 0 /* This functionality requires ulc_casecmp.  */
 290         ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 291         #endif
 292       }
 293     else
 294       ASSERT (my_casecmp (input, SIZEOF (input), casefolded_tr, SIZEOF (casefolded_tr)) == 0);
 295   }
 296   { /* LATIN SMALL LETTER DOTLESS I */
 297     static const char input[]      = { 0xC4, 0xB1 };
 298     static const char casefolded[] = { 0x49 };
 299
 300     if (!turkish)
 301       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) > 0);
 302     else
 303       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 304   }
 305   { /* "topkapı" */
 306     static const char input[] =
 307       { 0x54, 0x4F, 0x50, 0x4B, 0x41, 0x50, 0x49 };
 308     static const char casefolded[] =
 309       { 0x74, 0x6F, 0x70, 0x6B, 0x61, 0x70, 0xC4, 0xB1 };
 310
 311     if (!turkish)
 312       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) < 0);
 313     else
 314       ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 315   }
 316
 317   #if 0 /* This functionality requires ulc_casecmp.  */
 318   /* Uppercasing can increase the number of Unicode characters.  */
 319   { /* "heiß" */
 320     static const char input1[] = { 0x68, 0x65, 0x69, 0xC3, 0x9F };
 321     static const char input2[] = { 0x68, 0x65, 0x69, 0x73, 0x73 };
 322
 323     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 324   }
 325   #endif
 326
 327   /* Case mappings for some characters can depend on the surrounding characters.  */
 328   { /* "περισσότερες πληροφορίες" */
 329     static const char input1[] =
 330       {
 331         0xCF, 0x80, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB9, 0xCF, 0x83, 0xCF, 0x83,
 332         0xCF, 0x8C, 0xCF, 0x84, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB5, 0xCF, 0x82,
 333         ' ', 0xCF, 0x80, 0xCE, 0xBB, 0xCE, 0xB7, 0xCF, 0x81, 0xCE, 0xBF,
 334         0xCF, 0x86, 0xCE, 0xBF, 0xCF, 0x81, 0xCE, 0xAF, 0xCE, 0xB5, 0xCF, 0x82
 335       };
 336     static const char input2[] =
 337       {
 338         0xCF, 0x80, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB9, 0xCF, 0x83, 0xCF, 0x83,
 339         0xCF, 0x8C, 0xCF, 0x84, 0xCE, 0xB5, 0xCF, 0x81, 0xCE, 0xB5, 0xCF, 0x83,
 340         ' ', 0xCF, 0x80, 0xCE, 0xBB, 0xCE, 0xB7, 0xCF, 0x81, 0xCE, 0xBF,
 341         0xCF, 0x86, 0xCE, 0xBF, 0xCF, 0x81, 0xCE, 0xAF, 0xCE, 0xB5, 0xCF, 0x83
 342       };
 343     static const char input3[] =
 344       {
 345         0xCE, 0xA0, 0xCE, 0x95, 0xCE, 0xA1, 0xCE, 0x99, 0xCE, 0xA3, 0xCE, 0xA3,
 346         0xCE, 0x8C, 0xCE, 0xA4, 0xCE, 0x95, 0xCE, 0xA1, 0xCE, 0x95, 0xCE, 0xA3,
 347         ' ', 0xCE, 0xA0, 0xCE, 0x9B, 0xCE, 0x97, 0xCE, 0xA1, 0xCE, 0x9F,
 348         0xCE, 0xA6, 0xCE, 0x9F, 0xCE, 0xA1, 0xCE, 0x8A, 0xCE, 0x95, 0xCE, 0xA3
 349       };
 350
 351     (void) input1;
 352
 353     #if 0 /* This functionality requires ulc_casecmp.  */
 354     ASSERT (my_casecmp (input1, SIZEOF (input1), input2, SIZEOF (input2)) == 0);
 355
 356     ASSERT (my_casecmp (input1, SIZEOF (input1), input3, SIZEOF (input3)) == 0);
 357     #endif
 358
 359     ASSERT (my_casecmp (input2, SIZEOF (input2), input3, SIZEOF (input3)) == 0);
 360   }
 361
 362   #if 0 /* This functionality requires ulc_casecmp.  */
 363   /* Case mapping can require subsequent normalization.  */
 364   { /* LATIN SMALL LETTER J WITH CARON, COMBINING DOT BELOW */
 365     static const char input[]                 = { 0xC7, 0xB0, 0xCC, 0xA3 };
 366     static const char casefolded[]            = { 0x6A, 0xCC, 0x8C, 0xCC, 0xA3 };
 367     static const char casefolded_decomposed[] = { 0x6A, 0xCC, 0xA3, 0xCC, 0x8C };
 368
 369     ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 370
 371     ASSERT (my_casecmp (input, SIZEOF (input), casefolded_decomposed, SIZEOF (casefolded_decomposed)) != 0);
 372
 373     ASSERT (my_casecmp (input, SIZEOF (input), casefolded, SIZEOF (casefolded)) == 0);
 374
 375     ASSERT (my_casecmp (input, SIZEOF (input), casefolded_decomposed, SIZEOF (casefolded_decomposed)) == 0);
 376   }
 377   #endif
 378 }