tests/unistr/test-u8-next.c

   1 /* Test of u8_next() function.
   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 #include <config.h>
  20
  21 #include "unistr.h"
  22
  23 #include "macros.h"
  24
  25 int
  26 main ()
  27 {
  28   ucs4_t uc;
  29   const uint8_t *ret;
  30
  31   /* Test NUL unit input.  */
  32   {
  33     static const uint8_t input[] = "";
  34     uc = 0xBADFACE;
  35     ret = u8_next (&uc, input);
  36     ASSERT (ret == NULL);
  37     ASSERT (uc == 0);
  38   }
  39
  40   /* Test ISO 646 unit input.  */
  41   {
  42     ucs4_t c;
  43     uint8_t buf[2];
  44
  45     for (c = 1; c < 0x80; c++)
  46       {
  47         buf[0] = c;
  48         buf[1] = 0;
  49         uc = 0xBADFACE;
  50         ret = u8_next (&uc, buf);
  51         ASSERT (ret == buf + 1);
  52         ASSERT (uc == c);
  53       }
  54   }
  55
  56   /* Test 2-byte character input.  */
  57   {
  58     static const uint8_t input[] = { 0xC3, 0x97, 0 };
  59     uc = 0xBADFACE;
  60     ret = u8_next (&uc, input);
  61     ASSERT (ret == input + 2);
  62     ASSERT (uc == 0x00D7);
  63   }
  64
  65   /* Test 3-byte character input.  */
  66   {
  67     static const uint8_t input[] = { 0xE2, 0x82, 0xAC, 0 };
  68     uc = 0xBADFACE;
  69     ret = u8_next (&uc, input);
  70     ASSERT (ret == input + 3);
  71     ASSERT (uc == 0x20AC);
  72   }
  73
  74   /* Test 4-byte character input.  */
  75   {
  76     static const uint8_t input[] = { 0xF4, 0x8F, 0xBF, 0xBD, 0 };
  77     uc = 0xBADFACE;
  78     ret = u8_next (&uc, input);
  79     ASSERT (ret == input + 4);
  80     ASSERT (uc == 0x10FFFD);
  81   }
  82
  83   /* Test incomplete/invalid 1-byte input.  */
  84   {
  85     static const uint8_t input[] = { 0xC1, 0 };
  86     uc = 0xBADFACE;
  87     ret = u8_next (&uc, input);
  88     ASSERT (ret == NULL);
  89     ASSERT (uc == 0xFFFD);
  90   }
  91   {
  92     static const uint8_t input[] = { 0xC3, 0 };
  93     uc = 0xBADFACE;
  94     ret = u8_next (&uc, input);
  95     ASSERT (ret == NULL);
  96     ASSERT (uc == 0xFFFD);
  97   }
  98   {
  99     static const uint8_t input[] = { 0xE2, 0 };
 100     uc = 0xBADFACE;
 101     ret = u8_next (&uc, input);
 102     ASSERT (ret == NULL);
 103     ASSERT (uc == 0xFFFD);
 104   }
 105   {
 106     static const uint8_t input[] = { 0xF4, 0 };
 107     uc = 0xBADFACE;
 108     ret = u8_next (&uc, input);
 109     ASSERT (ret == NULL);
 110     ASSERT (uc == 0xFFFD);
 111   }
 112   {
 113     static const uint8_t input[] = { 0xFE, 0 };
 114     uc = 0xBADFACE;
 115     ret = u8_next (&uc, input);
 116     ASSERT (ret == NULL);
 117     ASSERT (uc == 0xFFFD);
 118   }
 119
 120   /* Test incomplete/invalid 2-byte input.  */
 121   {
 122     static const uint8_t input[] = { 0xE0, 0x9F, 0 };
 123     uc = 0xBADFACE;
 124     ret = u8_next (&uc, input);
 125     ASSERT (ret == NULL);
 126     ASSERT (uc == 0xFFFD);
 127   }
 128   {
 129     static const uint8_t input[] = { 0xE2, 0x82, 0 };
 130     uc = 0xBADFACE;
 131     ret = u8_next (&uc, input);
 132     ASSERT (ret == NULL);
 133     ASSERT (uc == 0xFFFD);
 134   }
 135   {
 136     static const uint8_t input[] = { 0xE2, 0xD0, 0 };
 137     uc = 0xBADFACE;
 138     ret = u8_next (&uc, input);
 139     ASSERT (ret == NULL);
 140     ASSERT (uc == 0xFFFD);
 141   }
 142   {
 143     static const uint8_t input[] = { 0xF0, 0x8F, 0 };
 144     uc = 0xBADFACE;
 145     ret = u8_next (&uc, input);
 146     ASSERT (ret == NULL);
 147     ASSERT (uc == 0xFFFD);
 148   }
 149   {
 150     static const uint8_t input[] = { 0xF3, 0x8F, 0 };
 151     uc = 0xBADFACE;
 152     ret = u8_next (&uc, input);
 153     ASSERT (ret == NULL);
 154     ASSERT (uc == 0xFFFD);
 155   }
 156   {
 157     static const uint8_t input[] = { 0xF3, 0xD0, 0 };
 158     uc = 0xBADFACE;
 159     ret = u8_next (&uc, input);
 160     ASSERT (ret == NULL);
 161     ASSERT (uc == 0xFFFD);
 162   }
 163
 164   /* Test incomplete/invalid 3-byte input.  */
 165   {
 166     static const uint8_t input[] = { 0xF3, 0x8F, 0xBF, 0 };
 167     uc = 0xBADFACE;
 168     ret = u8_next (&uc, input);
 169     ASSERT (ret == NULL);
 170     ASSERT (uc == 0xFFFD);
 171   }
 172   {
 173     static const uint8_t input[] = { 0xF3, 0xD0, 0xBF, 0 };
 174     uc = 0xBADFACE;
 175     ret = u8_next (&uc, input);
 176     ASSERT (ret == NULL);
 177     ASSERT (uc == 0xFFFD);
 178   }
 179   {
 180     static const uint8_t input[] = { 0xF3, 0x8F, 0xD0, 0 };
 181     uc = 0xBADFACE;
 182     ret = u8_next (&uc, input);
 183     ASSERT (ret == NULL);
 184     ASSERT (uc == 0xFFFD);
 185   }
 186
 187   return 0;
 188 }