1 /* Test of conversion of multibyte character to 32-bit wide character.
2 Copyright (C) 2008-2023 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>, 2008. */
18
19 #include <config.h>
20
21 #include <uchar.h>
22
23 #include "signature.h"
24 SIGNATURE_CHECK (mbrtoc32, size_t,
25 (char32_t *, const char *, size_t, mbstate_t *));
26
27 #include <locale.h>
28 #include <stdio.h>
29 #include <stdlib.h>
30 #include <string.h>
31 #include <wchar.h>
32
33 #include "macros.h"
34
35 int
36 main (int argc, char *argv[])
37 {
38 mbstate_t state;
39 char32_t wc;
40 size_t ret;
41
42 /* configure should already have checked that the locale is supported. */
43 if (setlocale (LC_ALL, "") == NULL)
44 return 1;
45
46 /* Test zero-length input. */
47 {
48 memset (&state, '\0', sizeof (mbstate_t));
49 wc = (char32_t) 0xBADFACE;
50 ret = mbrtoc32 (&wc, "x", 0, &state);
51 ASSERT (ret == (size_t)(-2));
52 ASSERT (mbsinit (&state));
53 }
54
55 /* Test NUL byte input. */
56 {
57 memset (&state, '\0', sizeof (mbstate_t));
58 wc = (char32_t) 0xBADFACE;
59 ret = mbrtoc32 (&wc, "", 1, &state);
60 ASSERT (ret == 0);
61 ASSERT (wc == 0);
62 ASSERT (mbsinit (&state));
63 ret = mbrtoc32 (NULL, "", 1, &state);
64 ASSERT (ret == 0);
65 ASSERT (mbsinit (&state));
66 }
67
68 /* Test single-byte input. */
69 {
70 int c;
71 char buf[1];
72
73 memset (&state, '\0', sizeof (mbstate_t));
74 for (c = 0; c < 0x100; c++)
75 switch (c)
76 {
77 case '\t': case '\v': case '\f':
78 case ' ': case '!': case '"': case '#': case '%':
79 case '&': case '\'': case '(': case ')': case '*':
80 case '+': case ',': case '-': case '.': case '/':
81 case '0': case '1': case '2': case '3': case '4':
82 case '5': case '6': case '7': case '8': case '9':
83 case ':': case ';': case '<': case '=': case '>':
84 case '?':
85 case 'A': case 'B': case 'C': case 'D': case 'E':
86 case 'F': case 'G': case 'H': case 'I': case 'J':
87 case 'K': case 'L': case 'M': case 'N': case 'O':
88 case 'P': case 'Q': case 'R': case 'S': case 'T':
89 case 'U': case 'V': case 'W': case 'X': case 'Y':
90 case 'Z':
91 case '[': case '\\': case ']': case '^': case '_':
92 case 'a': case 'b': case 'c': case 'd': case 'e':
93 case 'f': case 'g': case 'h': case 'i': case 'j':
94 case 'k': case 'l': case 'm': case 'n': case 'o':
95 case 'p': case 'q': case 'r': case 's': case 't':
96 case 'u': case 'v': case 'w': case 'x': case 'y':
97 case 'z': case '{': case '|': case '}': case '~':
98 /* c is in the ISO C "basic character set". */
99 ASSERT (c < 0x80);
100 /* c is an ASCII character. */
101 buf[0] = c;
102
103 wc = (char32_t) 0xBADFACE;
104 ret = mbrtoc32 (&wc, buf, 1, &state);
105 ASSERT (ret == 1);
106 ASSERT (wc == c);
107 ASSERT (mbsinit (&state));
108
109 ret = mbrtoc32 (NULL, buf, 1, &state);
110 ASSERT (ret == 1);
111 ASSERT (mbsinit (&state));
112
113 break;
114 default:
115 break;
116 }
117 }
118
119 /* Test special calling convention, passing a NULL pointer. */
120 {
121 memset (&state, '\0', sizeof (mbstate_t));
122 wc = (char32_t) 0xBADFACE;
123 ret = mbrtoc32 (&wc, NULL, 5, &state);
124 ASSERT (ret == 0);
125 ASSERT (wc == (char32_t) 0xBADFACE);
126 ASSERT (mbsinit (&state));
127 }
128
129 #ifdef __ANDROID__
130 /* On Android ≥ 5.0, the default locale is the "C.UTF-8" locale, not the
131 "C" locale. Furthermore, when you attempt to set the "C" or "POSIX"
132 locale via setlocale(), what you get is a "C" locale with UTF-8 encoding,
133 that is, effectively the "C.UTF-8" locale. */
134 if (argc > 1 && strcmp (argv[1], "1") == 0 && MB_CUR_MAX > 1)
135 argv[1] = "3";
136 #endif
137
138 if (argc > 1)
139 switch (argv[1][0])
140 {
141 case '1':
142 /* C or POSIX locale. */
143 {
144 int c;
145 char buf[1];
146
147 memset (&state, '\0', sizeof (mbstate_t));
148 for (c = 0; c < 0x100; c++)
149 if (c != 0)
150 {
151 /* We are testing all nonnull bytes. */
152 buf[0] = c;
153
154 wc = (char32_t) 0xBADFACE;
155 ret = mbrtoc32 (&wc, buf, 1, &state);
156 /* POSIX:2018 says regarding mbrtowc: "In the POSIX locale an
157 [EILSEQ] error cannot occur since all byte values are valid
158 characters." It is reasonable to expect mbrtoc32 to behave
159 in the same way. */
160 ASSERT (ret == 1);
161 if (c < 0x80)
162 /* c is an ASCII character. */
163 ASSERT (wc == c);
164 else
165 /* On most platforms, the bytes 0x80..0xFF map to U+0080..U+00FF.
166 But on musl libc, the bytes 0x80..0xFF map to U+DF80..U+DFFF. */
167 ASSERT (wc == (btoc32 (c) == 0xDF00 + c ? btoc32 (c) : c));
168 ASSERT (mbsinit (&state));
169
170 ret = mbrtoc32 (NULL, buf, 1, &state);
171 ASSERT (ret == 1);
172 ASSERT (mbsinit (&state));
173 }
174 }
175 return 0;
176
177 case '2':
178 /* Locale encoding is ISO-8859-1 or ISO-8859-15. */
179 {
180 char input[] = "B\374\337er"; /* "Büßer" */
181 memset (&state, '\0', sizeof (mbstate_t));
182
183 wc = (char32_t) 0xBADFACE;
184 ret = mbrtoc32 (&wc, input, 1, &state);
185 ASSERT (ret == 1);
186 ASSERT (wc == 'B');
187 ASSERT (mbsinit (&state));
188 input[0] = '\0';
189
190 wc = (char32_t) 0xBADFACE;
191 ret = mbrtoc32 (&wc, input + 1, 1, &state);
192 ASSERT (ret == 1);
193 ASSERT (c32tob (wc) == (unsigned char) '\374');
194 #if GL_CHAR32_T_IS_UNICODE
195 ASSERT (wc == 0x00FC); /* expect Unicode encoding */
196 #endif
197 ASSERT (mbsinit (&state));
198 input[1] = '\0';
199
200 /* Test support of NULL first argument. */
201 ret = mbrtoc32 (NULL, input + 2, 3, &state);
202 ASSERT (ret == 1);
203 ASSERT (mbsinit (&state));
204
205 wc = (char32_t) 0xBADFACE;
206 ret = mbrtoc32 (&wc, input + 2, 3, &state);
207 ASSERT (ret == 1);
208 ASSERT (c32tob (wc) == (unsigned char) '\337');
209 #if GL_CHAR32_T_IS_UNICODE
210 ASSERT (wc == 0x00DF); /* expect Unicode encoding */
211 #endif
212 ASSERT (mbsinit (&state));
213 input[2] = '\0';
214
215 wc = (char32_t) 0xBADFACE;
216 ret = mbrtoc32 (&wc, input + 3, 2, &state);
217 ASSERT (ret == 1);
218 ASSERT (wc == 'e');
219 ASSERT (mbsinit (&state));
220 input[3] = '\0';
221
222 wc = (char32_t) 0xBADFACE;
223 ret = mbrtoc32 (&wc, input + 4, 1, &state);
224 ASSERT (ret == 1);
225 ASSERT (wc == 'r');
226 ASSERT (mbsinit (&state));
227 }
228 return 0;
229
230 case '3':
231 /* Locale encoding is UTF-8. */
232 {
233 char input[] = "s\303\274\303\237\360\237\230\213!"; /* "süß😋!" */
234 memset (&state, '\0', sizeof (mbstate_t));
235
236 wc = (char32_t) 0xBADFACE;
237 ret = mbrtoc32 (&wc, input, 1, &state);
238 ASSERT (ret == 1);
239 ASSERT (wc == 's');
240 ASSERT (mbsinit (&state));
241 input[0] = '\0';
242
243 wc = (char32_t) 0xBADFACE;
244 ret = mbrtoc32 (&wc, input + 1, 1, &state);
245 ASSERT (ret == (size_t)(-2));
246 ASSERT (wc == (char32_t) 0xBADFACE);
247 ASSERT (!mbsinit (&state));
248 input[1] = '\0';
249
250 wc = (char32_t) 0xBADFACE;
251 ret = mbrtoc32 (&wc, input + 2, 7, &state);
252 ASSERT (ret == 1);
253 ASSERT (c32tob (wc) == EOF);
254 ASSERT (wc == 0x00FC); /* expect Unicode encoding */
255 ASSERT (mbsinit (&state));
256 input[2] = '\0';
257
258 /* Test support of NULL first argument. */
259 ret = mbrtoc32 (NULL, input + 3, 6, &state);
260 ASSERT (ret == 2);
261 ASSERT (mbsinit (&state));
262
263 wc = (char32_t) 0xBADFACE;
264 ret = mbrtoc32 (&wc, input + 3, 6, &state);
265 ASSERT (ret == 2);
266 ASSERT (c32tob (wc) == EOF);
267 ASSERT (wc == 0x00DF); /* expect Unicode encoding */
268 ASSERT (mbsinit (&state));
269 input[3] = '\0';
270 input[4] = '\0';
271
272 /* Test support of NULL first argument. */
273 ret = mbrtoc32 (NULL, input + 5, 4, &state);
274 ASSERT (ret == 4);
275 ASSERT (mbsinit (&state));
276
277 wc = (char32_t) 0xBADFACE;
278 ret = mbrtoc32 (&wc, input + 5, 4, &state);
279 ASSERT (ret == 4);
280 ASSERT (c32tob (wc) == EOF);
281 ASSERT (wc == 0x1F60B); /* expect Unicode encoding */
282 ASSERT (mbsinit (&state));
283 input[5] = '\0';
284 input[6] = '\0';
285 input[7] = '\0';
286 input[8] = '\0';
287
288 wc = (char32_t) 0xBADFACE;
289 ret = mbrtoc32 (&wc, input + 9, 1, &state);
290 ASSERT (ret == 1);
291 ASSERT (wc == '!');
292 ASSERT (mbsinit (&state));
293 }
294 return 0;
295
296 case '4':
297 /* Locale encoding is EUC-JP. */
298 {
299 char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */
300 memset (&state, '\0', sizeof (mbstate_t));
301
302 wc = (char32_t) 0xBADFACE;
303 ret = mbrtoc32 (&wc, input, 1, &state);
304 ASSERT (ret == 1);
305 ASSERT (wc == '<');
306 ASSERT (mbsinit (&state));
307 input[0] = '\0';
308
309 wc = (char32_t) 0xBADFACE;
310 ret = mbrtoc32 (&wc, input + 1, 2, &state);
311 ASSERT (ret == 2);
312 ASSERT (c32tob (wc) == EOF);
313 #if GL_CHAR32_T_IS_UNICODE
314 ASSERT (wc == 0x65E5); /* expect Unicode encoding */
315 #endif
316 ASSERT (mbsinit (&state));
317 input[1] = '\0';
318 input[2] = '\0';
319
320 wc = (char32_t) 0xBADFACE;
321 ret = mbrtoc32 (&wc, input + 3, 1, &state);
322 ASSERT (ret == (size_t)(-2));
323 ASSERT (wc == (char32_t) 0xBADFACE);
324 ASSERT (!mbsinit (&state));
325 input[3] = '\0';
326
327 wc = (char32_t) 0xBADFACE;
328 ret = mbrtoc32 (&wc, input + 4, 4, &state);
329 ASSERT (ret == 1);
330 ASSERT (c32tob (wc) == EOF);
331 #if GL_CHAR32_T_IS_UNICODE
332 ASSERT (wc == 0x672C); /* expect Unicode encoding */
333 #endif
334 ASSERT (mbsinit (&state));
335 input[4] = '\0';
336
337 /* Test support of NULL first argument. */
338 ret = mbrtoc32 (NULL, input + 5, 3, &state);
339 ASSERT (ret == 2);
340 ASSERT (mbsinit (&state));
341
342 wc = (char32_t) 0xBADFACE;
343 ret = mbrtoc32 (&wc, input + 5, 3, &state);
344 ASSERT (ret == 2);
345 ASSERT (c32tob (wc) == EOF);
346 #if GL_CHAR32_T_IS_UNICODE
347 ASSERT (wc == 0x8A9E); /* expect Unicode encoding */
348 #endif
349 ASSERT (mbsinit (&state));
350 input[5] = '\0';
351 input[6] = '\0';
352
353 wc = (char32_t) 0xBADFACE;
354 ret = mbrtoc32 (&wc, input + 7, 1, &state);
355 ASSERT (ret == 1);
356 ASSERT (wc == '>');
357 ASSERT (mbsinit (&state));
358 }
359 return 0;
360
361 case '5':
362 /* Locale encoding is GB18030. */
363 #if (defined __GLIBC__ && __GLIBC__ == 2 && __GLIBC_MINOR__ >= 13 && __GLIBC_MINOR__ <= 15) || (GL_CHAR32_T_IS_UNICODE && (defined __NetBSD__ || defined __sun))
364 fputs ("Skipping test: The GB18030 converter in this system's iconv is broken.\n", stderr);
365 return 77;
366 #endif
367 {
368 char input[] = "s\250\271\201\060\211\070\224\071\375\067!"; /* "süß😋!" */
369 memset (&state, '\0', sizeof (mbstate_t));
370
371 wc = (char32_t) 0xBADFACE;
372 ret = mbrtoc32 (&wc, input, 1, &state);
373 ASSERT (ret == 1);
374 ASSERT (wc == 's');
375 ASSERT (mbsinit (&state));
376 input[0] = '\0';
377
378 wc = (char32_t) 0xBADFACE;
379 ret = mbrtoc32 (&wc, input + 1, 1, &state);
380 ASSERT (ret == (size_t)(-2));
381 ASSERT (wc == (char32_t) 0xBADFACE);
382 ASSERT (!mbsinit (&state));
383 input[1] = '\0';
384
385 wc = (char32_t) 0xBADFACE;
386 ret = mbrtoc32 (&wc, input + 2, 9, &state);
387 ASSERT (ret == 1);
388 ASSERT (c32tob (wc) == EOF);
389 #if GL_CHAR32_T_IS_UNICODE
390 ASSERT (wc == 0x00FC); /* expect Unicode encoding */
391 #endif
392 ASSERT (mbsinit (&state));
393 input[2] = '\0';
394
395 /* Test support of NULL first argument. */
396 ret = mbrtoc32 (NULL, input + 3, 8, &state);
397 ASSERT (ret == 4);
398 ASSERT (mbsinit (&state));
399
400 wc = (char32_t) 0xBADFACE;
401 ret = mbrtoc32 (&wc, input + 3, 8, &state);
402 ASSERT (ret == 4);
403 ASSERT (c32tob (wc) == EOF);
404 #if GL_CHAR32_T_IS_UNICODE
405 ASSERT (wc == 0x00DF); /* expect Unicode encoding */
406 #endif
407 ASSERT (mbsinit (&state));
408 input[3] = '\0';
409 input[4] = '\0';
410 input[5] = '\0';
411 input[6] = '\0';
412
413 /* Test support of NULL first argument. */
414 ret = mbrtoc32 (NULL, input + 7, 4, &state);
415 ASSERT (ret == 4);
416 ASSERT (mbsinit (&state));
417
418 wc = (char32_t) 0xBADFACE;
419 ret = mbrtoc32 (&wc, input + 7, 4, &state);
420 ASSERT (ret == 4);
421 ASSERT (c32tob (wc) == EOF);
422 #if GL_CHAR32_T_IS_UNICODE
423 ASSERT (wc == 0x1F60B); /* expect Unicode encoding */
424 #endif
425 ASSERT (mbsinit (&state));
426 input[7] = '\0';
427 input[8] = '\0';
428 input[9] = '\0';
429 input[10] = '\0';
430
431 wc = (char32_t) 0xBADFACE;
432 ret = mbrtoc32 (&wc, input + 11, 1, &state);
433 ASSERT (ret == 1);
434 ASSERT (wc == '!');
435 ASSERT (mbsinit (&state));
436 }
437 return 0;
438 }
439
440 return 1;
441 }