tests/unistr/test-u8-mbtouc.h

   1 /* Test of u8_mbtouc() and u8_mbtouc_unsafe() functions.
   2    Copyright (C) 2010-2020 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>, 2010.  */
  18
  19 static void
  20 test_function (int (*my_u8_mbtouc) (ucs4_t *, const uint8_t *, size_t))
  21 {
  22   ucs4_t uc;
  23   int ret;
  24
  25   /* Test NUL unit input.  */
  26   {
  27     static const uint8_t input[] = "";
  28     uc = 0xBADFACE;
  29     ret = my_u8_mbtouc (&uc, input, 1);
  30     ASSERT (ret == 1);
  31     ASSERT (uc == 0);
  32   }
  33
  34   /* Test ISO 646 unit input.  */
  35   {
  36     ucs4_t c;
  37     uint8_t buf[1];
  38
  39     for (c = 0; c < 0x80; c++)
  40       {
  41         buf[0] = c;
  42         uc = 0xBADFACE;
  43         ret = my_u8_mbtouc (&uc, buf, 1);
  44         ASSERT (ret == 1);
  45         ASSERT (uc == c);
  46       }
  47   }
  48
  49   /* Test 2-byte character input.  */
  50   {
  51     static const uint8_t input[] = { 0xC3, 0x97 };
  52     uc = 0xBADFACE;
  53     ret = my_u8_mbtouc (&uc, input, 2);
  54     ASSERT (ret == 2);
  55     ASSERT (uc == 0x00D7);
  56   }
  57
  58   /* Test 3-byte character input.  */
  59   {
  60     static const uint8_t input[] = { 0xE2, 0x82, 0xAC };
  61     uc = 0xBADFACE;
  62     ret = my_u8_mbtouc (&uc, input, 3);
  63     ASSERT (ret == 3);
  64     ASSERT (uc == 0x20AC);
  65   }
  66
  67   /* Test 4-byte character input.  */
  68   {
  69     static const uint8_t input[] = { 0xF4, 0x8F, 0xBF, 0xBD };
  70     uc = 0xBADFACE;
  71     ret = my_u8_mbtouc (&uc, input, 4);
  72     ASSERT (ret == 4);
  73     ASSERT (uc == 0x10FFFD);
  74   }
  75
  76   /* Test incomplete/invalid 1-byte input.  */
  77   {
  78     static const uint8_t input[] = { 0xC1 };
  79     uc = 0xBADFACE;
  80     ret = my_u8_mbtouc (&uc, input, 1);
  81     ASSERT (ret == 1);
  82     ASSERT (uc == 0xFFFD);
  83   }
  84   {
  85     static const uint8_t input[] = { 0xC3 };
  86     uc = 0xBADFACE;
  87     ret = my_u8_mbtouc (&uc, input, 1);
  88     ASSERT (ret == 1);
  89     ASSERT (uc == 0xFFFD);
  90   }
  91   {
  92     static const uint8_t input[] = { 0xE2 };
  93     uc = 0xBADFACE;
  94     ret = my_u8_mbtouc (&uc, input, 1);
  95     ASSERT (ret == 1);
  96     ASSERT (uc == 0xFFFD);
  97   }
  98   {
  99     static const uint8_t input[] = { 0xF4 };
 100     uc = 0xBADFACE;
 101     ret = my_u8_mbtouc (&uc, input, 1);
 102     ASSERT (ret == 1);
 103     ASSERT (uc == 0xFFFD);
 104   }
 105   {
 106     static const uint8_t input[] = { 0xFE };
 107     uc = 0xBADFACE;
 108     ret = my_u8_mbtouc (&uc, input, 1);
 109     ASSERT (ret == 1);
 110     ASSERT (uc == 0xFFFD);
 111   }
 112
 113   /* Test incomplete/invalid 2-byte input.  */
 114   {
 115     static const uint8_t input[] = { 0xE0, 0x9F };
 116     uc = 0xBADFACE;
 117     ret = my_u8_mbtouc (&uc, input, 2);
 118     ASSERT (ret == 1 || ret == 2);
 119     ASSERT (uc == 0xFFFD);
 120   }
 121   {
 122     static const uint8_t input[] = { 0xE2, 0x82 };
 123     uc = 0xBADFACE;
 124     ret = my_u8_mbtouc (&uc, input, 2);
 125     ASSERT (ret == 2);
 126     ASSERT (uc == 0xFFFD);
 127   }
 128   {
 129     static const uint8_t input[] = { 0xE2, 0xD0 };
 130     uc = 0xBADFACE;
 131     ret = my_u8_mbtouc (&uc, input, 2);
 132     ASSERT (ret == 1 || ret == 2);
 133     ASSERT (uc == 0xFFFD);
 134   }
 135   {
 136     static const uint8_t input[] = { 0xF0, 0x8F };
 137     uc = 0xBADFACE;
 138     ret = my_u8_mbtouc (&uc, input, 2);
 139     ASSERT (ret == 1 || ret == 2);
 140     ASSERT (uc == 0xFFFD);
 141   }
 142   {
 143     static const uint8_t input[] = { 0xF3, 0x8F };
 144     uc = 0xBADFACE;
 145     ret = my_u8_mbtouc (&uc, input, 2);
 146     ASSERT (ret == 2);
 147     ASSERT (uc == 0xFFFD);
 148   }
 149   {
 150     static const uint8_t input[] = { 0xF3, 0xD0 };
 151     uc = 0xBADFACE;
 152     ret = my_u8_mbtouc (&uc, input, 2);
 153     ASSERT (ret == 1 || ret == 2);
 154     ASSERT (uc == 0xFFFD);
 155   }
 156
 157   /* Test incomplete/invalid 3-byte input.  */
 158   {
 159     static const uint8_t input[] = { 0xF3, 0x8F, 0xBF };
 160     uc = 0xBADFACE;
 161     ret = my_u8_mbtouc (&uc, input, 3);
 162     ASSERT (ret == 3);
 163     ASSERT (uc == 0xFFFD);
 164   }
 165   {
 166     static const uint8_t input[] = { 0xF3, 0xD0, 0xBF };
 167     uc = 0xBADFACE;
 168     ret = my_u8_mbtouc (&uc, input, 3);
 169     ASSERT (ret == 1);
 170     ASSERT (uc == 0xFFFD);
 171   }
 172   {
 173     static const uint8_t input[] = { 0xF3, 0x8F, 0xD0 };
 174     uc = 0xBADFACE;
 175     ret = my_u8_mbtouc (&uc, input, 3);
 176     ASSERT (ret == 2);
 177     ASSERT (uc == 0xFFFD);
 178   }
 179 }