1/* Test of conversion to UTF-32 from legacy encodings. 2 Copyright (C) 2007-2010 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>, 2007. */ 18 19#include <config.h> 20 21#include "uniconv.h" 22 23#include <stdlib.h> 24#include <string.h> 25 26#include "unistr.h" 27#include "macros.h" 28 29/* Magic number for detecting bounds violations. */ 30#define MAGIC 0x1983EFF1 31 32static size_t * 33new_offsets (size_t n) 34{ 35 size_t *offsets = (size_t *) malloc ((n + 1) * sizeof (size_t)); 36 offsets[n] = MAGIC; 37 return offsets; 38} 39 40int 41main () 42{ 43 static enum iconv_ilseq_handler handlers[] = 44 { iconveh_error, iconveh_question_mark, iconveh_escape_sequence }; 45 size_t h; 46 size_t o; 47 size_t i; 48 49#if HAVE_ICONV 50 /* Assume that iconv() supports at least the encodings ASCII, ISO-8859-1, 51 ISO-8859-2, and UTF-8. */ 52 53 /* Test conversion from ISO-8859-1 to UTF-16 with no errors. */ 54 for (h = 0; h < SIZEOF (handlers); h++) 55 { 56 enum iconv_ilseq_handler handler = handlers[h]; 57 static const char input[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337"; 58 static const uint32_t expected[] = /* ��rger mit b��sen B��bchen ohne Augenma�� */ 59 { 60 0xC4, 'r', 'g', 'e', 'r', ' ', 'm', 'i', 't', ' ', 'b', 0xF6, 's', 61 'e', 'n', ' ', 'B', 0xFC, 'b', 'c', 'h', 'e', 'n', ' ', 'o', 'h', 62 'n', 'e', ' ', 'A', 'u', 'g', 'e', 'n', 'm', 'a', 0xDF 63 }; 64 for (o = 0; o < 2; o++) 65 { 66 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); 67 size_t length; 68 uint32_t *result = u32_conv_from_encoding ("ISO-8859-1", handler, 69 input, strlen (input), 70 offsets, 71 NULL, &length); 72 ASSERT (result != NULL); 73 ASSERT (length == SIZEOF (expected)); 74 ASSERT (u32_cmp (result, expected, SIZEOF (expected)) == 0); 75 if (o) 76 { 77 for (i = 0; i < 37; i++) 78 ASSERT (offsets[i] == i); 79 ASSERT (offsets[37] == MAGIC); 80 free (offsets); 81 } 82 free (result); 83 } 84 } 85 86 /* Test conversion from ISO-8859-2 to UTF-16 with no errors. */ 87 for (h = 0; h < SIZEOF (handlers); h++) 88 { 89 enum iconv_ilseq_handler handler = handlers[h]; 90 static const char input[] = "Rafa\263 Maszkowski"; /* Rafa�� Maszkowski */ 91 static const uint32_t expected[] = 92 { 93 'R', 'a', 'f', 'a', 0x0142, ' ', 'M', 'a', 's', 'z', 'k', 'o', 'w', 94 's', 'k', 'i' 95 }; 96 for (o = 0; o < 2; o++) 97 { 98 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); 99 size_t length; 100 uint32_t *result = u32_conv_from_encoding ("ISO-8859-2", handler, 101 input, strlen (input), 102 offsets, 103 NULL, &length); 104 ASSERT (result != NULL); 105 ASSERT (length == SIZEOF (expected)); 106 ASSERT (u32_cmp (result, expected, SIZEOF (expected)) == 0); 107 if (o) 108 { 109 for (i = 0; i < 16; i++) 110 ASSERT (offsets[i] == i); 111 ASSERT (offsets[16] == MAGIC); 112 free (offsets); 113 } 114 free (result); 115 } 116 } 117 118 /* autodetect_jp is only supported when iconv() support ISO-2022-JP-2. */ 119# if defined _LIBICONV_VERSION || !(defined _AIX || defined __sgi || defined __hpux || defined __osf__ || defined __sun) 120 /* Test conversions from autodetect_jp to UTF-16. */ 121 for (h = 0; h < SIZEOF (handlers); h++) 122 { 123 enum iconv_ilseq_handler handler = handlers[h]; 124 static const char input[] = "\244\263\244\363\244\313\244\301\244\317"; /* ��������������� in EUC-JP */ 125 static const uint32_t expected[] = /* ��������������� */ 126 { 127 0x3053, 0x3093, 0x306B, 0x3061, 0x306F 128 }; 129 for (o = 0; o < 2; o++) 130 { 131 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); 132 size_t length; 133 uint32_t *result = u32_conv_from_encoding ("autodetect_jp", handler, 134 input, strlen (input), 135 offsets, 136 NULL, &length); 137 ASSERT (result != NULL); 138 ASSERT (length == SIZEOF (expected)); 139 ASSERT (u32_cmp (result, expected, SIZEOF (expected)) == 0); 140 if (o) 141 { 142 for (i = 0; i < 10; i++) 143 ASSERT (offsets[i] == ((i % 2) == 0 ? i / 2 : (size_t)(-1))); 144 ASSERT (offsets[10] == MAGIC); 145 free (offsets); 146 } 147 free (result); 148 } 149 } 150 for (h = 0; h < SIZEOF (handlers); h++) 151 { 152 enum iconv_ilseq_handler handler = handlers[h]; 153 static const char input[] = "\202\261\202\361\202\311\202\277\202\315"; /* ��������������� in Shift_JIS */ 154 static const uint32_t expected[] = /* ��������������� */ 155 { 156 0x3053, 0x3093, 0x306B, 0x3061, 0x306F 157 }; 158 for (o = 0; o < 2; o++) 159 { 160 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); 161 size_t length; 162 uint32_t *result = u32_conv_from_encoding ("autodetect_jp", handler, 163 input, strlen (input), 164 offsets, 165 NULL, &length); 166 ASSERT (result != NULL); 167 ASSERT (length == SIZEOF (expected)); 168 ASSERT (u32_cmp (result, expected, SIZEOF (expected)) == 0); 169 if (o) 170 { 171 for (i = 0; i < 10; i++) 172 ASSERT (offsets[i] == ((i % 2) == 0 ? i / 2 : (size_t)(-1))); 173 ASSERT (offsets[10] == MAGIC); 174 free (offsets); 175 } 176 free (result); 177 } 178 } 179 for (h = 0; h < SIZEOF (handlers); h++) 180 { 181 enum iconv_ilseq_handler handler = handlers[h]; 182 static const char input[] = "\033$B$3$s$K$A$O\033(B"; /* ��������������� in ISO-2022-JP-2 */ 183 static const uint32_t expected[] = /* ��������������� */ 184 { 185 0x3053, 0x3093, 0x306B, 0x3061, 0x306F 186 }; 187 for (o = 0; o < 2; o++) 188 { 189 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); 190 size_t length; 191 uint32_t *result = u32_conv_from_encoding ("autodetect_jp", handler, 192 input, strlen (input), 193 offsets, 194 NULL, &length); 195 ASSERT (result != NULL); 196 ASSERT (length == SIZEOF (expected)); 197 ASSERT (u32_cmp (result, expected, SIZEOF (expected)) == 0); 198 if (o) 199 { 200 for (i = 0; i < 16; i++) 201 ASSERT (offsets[i] == (i == 0 ? 0 : 202 i == 5 ? 1 : 203 i == 7 ? 2 : 204 i == 9 ? 3 : 205 i == 11 ? 4 : 206 i == 13 ? 5 : 207 (size_t)(-1))); 208 ASSERT (offsets[16] == MAGIC); 209 free (offsets); 210 } 211 free (result); 212 } 213 } 214# endif 215 216#endif 217 218 return 0; 219} 220