1 /* Test of conversion to UTF-8 from legacy encodings.
2 Copyright (C) 2007-2023 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>, 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 extern int iconv_supports_encoding (const char *encoding);
29
30 /* Magic number for detecting bounds violations. */
31 #define MAGIC 0x1983EFF1
32
33 static size_t *
34 new_offsets (size_t n)
35 {
36 size_t *offsets = (size_t *) malloc ((n + 1) * sizeof (size_t));
37 offsets[n] = MAGIC;
38 return offsets;
39 }
40
41 int
42 main ()
43 {
44 #if HAVE_ICONV
45 static enum iconv_ilseq_handler handlers[] =
46 { iconveh_error, iconveh_question_mark, iconveh_escape_sequence };
47 size_t h;
48 size_t o;
49 size_t i;
50
51 /* Assume that iconv() supports at least the encodings ASCII, ISO-8859-1,
52 ISO-8859-2, and UTF-8. */
53
54 /* Test conversion from ISO-8859-1 to UTF-8 with no errors. */
55 for (h = 0; h < SIZEOF (handlers); h++)
56 {
57 enum iconv_ilseq_handler handler = handlers[h];
58 static const char input[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337";
59 static const uint8_t expected[] = "\303\204rger mit b\303\266sen B\303\274bchen ohne Augenma\303\237";
60 for (o = 0; o < 2; o++)
61 {
62 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
63 size_t length;
64 uint8_t *result = u8_conv_from_encoding ("ISO-8859-1", handler,
65 input, strlen (input),
66 offsets,
67 NULL, &length);
68 ASSERT (result != NULL);
69 ASSERT (length == u8_strlen (expected));
70 ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
71 if (o)
72 {
73 for (i = 0; i < 37; i++)
74 ASSERT (offsets[i] == (i < 1 ? i :
75 i < 12 ? i + 1 :
76 i < 18 ? i + 2 :
77 i + 3));
78 ASSERT (offsets[37] == MAGIC);
79 free (offsets);
80 }
81 free (result);
82 }
83 }
84
85 /* Test conversion from ISO-8859-2 to UTF-8 with no errors. */
86 for (h = 0; h < SIZEOF (handlers); h++)
87 {
88 enum iconv_ilseq_handler handler = handlers[h];
89 static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */
90 static const uint8_t expected[] = "Rafa\305\202 Maszkowski";
91 for (o = 0; o < 2; o++)
92 {
93 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
94 size_t length;
95 uint8_t *result = u8_conv_from_encoding ("ISO-8859-2", handler,
96 input, strlen (input),
97 offsets,
98 NULL, &length);
99 ASSERT (result != NULL);
100 ASSERT (length == u8_strlen (expected));
101 ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
102 if (o)
103 {
104 for (i = 0; i < 16; i++)
105 ASSERT (offsets[i] == (i < 5 ? i :
106 i + 1));
107 ASSERT (offsets[16] == MAGIC);
108 free (offsets);
109 }
110 free (result);
111 }
112 }
113
114 /* autodetect_jp is only supported when iconv() support ISO-2022-JP-2. */
115 # if defined _LIBICONV_VERSION || !(defined _AIX || defined __sgi || defined __hpux || defined __osf__ || defined __sun)
116 if (iconv_supports_encoding ("ISO-2022-JP-2"))
117 {
118 /* Test conversions from autodetect_jp to UTF-8. */
119 for (h = 0; h < SIZEOF (handlers); h++)
120 {
121 enum iconv_ilseq_handler handler = handlers[h];
122 static const char input[] = "\244\263\244\363\244\313\244\301\244\317"; /* こんにちは in EUC-JP */
123 static const uint8_t expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */
124 for (o = 0; o < 2; o++)
125 {
126 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
127 size_t length;
128 uint8_t *result = u8_conv_from_encoding ("autodetect_jp", handler,
129 input, strlen (input),
130 offsets,
131 NULL, &length);
132 ASSERT (result != NULL);
133 ASSERT (length == u8_strlen (expected));
134 ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
135 if (o)
136 {
137 for (i = 0; i < 10; i++)
138 ASSERT (offsets[i] == ((i % 2) == 0 ? (i / 2) * 3 : (size_t)(-1)));
139 ASSERT (offsets[10] == MAGIC);
140 free (offsets);
141 }
142 free (result);
143 }
144 }
145 for (h = 0; h < SIZEOF (handlers); h++)
146 {
147 enum iconv_ilseq_handler handler = handlers[h];
148 static const char input[] = "\202\261\202\361\202\311\202\277\202\315"; /* こんにちは in Shift_JIS */
149 static const uint8_t expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */
150 for (o = 0; o < 2; o++)
151 {
152 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
153 size_t length;
154 uint8_t *result = u8_conv_from_encoding ("autodetect_jp", handler,
155 input, strlen (input),
156 offsets,
157 NULL, &length);
158 ASSERT (result != NULL);
159 ASSERT (length == u8_strlen (expected));
160 ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
161 if (o)
162 {
163 for (i = 0; i < 10; i++)
164 ASSERT (offsets[i] == ((i % 2) == 0 ? (i / 2) * 3 : (size_t)(-1)));
165 ASSERT (offsets[10] == MAGIC);
166 free (offsets);
167 }
168 free (result);
169 }
170 }
171 for (h = 0; h < SIZEOF (handlers); h++)
172 {
173 enum iconv_ilseq_handler handler = handlers[h];
174 static const char input[] = "\033$B$3$s$K$A$O\033(B"; /* こんにちは in ISO-2022-JP-2 */
175 static const uint8_t expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */
176 for (o = 0; o < 2; o++)
177 {
178 size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
179 size_t length;
180 uint8_t *result = u8_conv_from_encoding ("autodetect_jp", handler,
181 input, strlen (input),
182 offsets,
183 NULL, &length);
184 ASSERT (result != NULL);
185 ASSERT (length == u8_strlen (expected));
186 ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
187 if (o)
188 {
189 for (i = 0; i < 16; i++)
190 ASSERT (offsets[i] == (i == 0 ? 0 :
191 i == 5 ? 3 :
192 i == 7 ? 6 :
193 i == 9 ? 9 :
194 i == 11 ? 12 :
195 i == 13 ? 15 :
196 (size_t)(-1)));
197 ASSERT (offsets[16] == MAGIC);
198 free (offsets);
199 }
200 free (result);
201 }
202 }
203 }
204 # endif
205
206 #endif
207
208 return 0;
209 }