"C" locale. Furthermore, when you attempt to set the "C" or "POSIX"
locale via setlocale(), what you get is a "C" locale with UTF-8 encoding,
that is, effectively the "C.UTF-8" locale. */
- if (argc > 1 && strcmp (argv[1], "5") == 0 && MB_CUR_MAX > 1)
- argv[1] = "2";
+ if (argc > 1 && strcmp (argv[1], "1") == 0 && MB_CUR_MAX > 1)
+ argv[1] = "3";
#endif
if (argc > 1)
switch (argv[1][0])
{
case '1':
+ /* C or POSIX locale. */
+ {
+ char input[] = "n/a";
+ memset (&state, '\0', sizeof (mbstate_t));
+
+ src = input;
+ temp_state = state;
+ ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 1, &temp_state);
+ ASSERT (ret == 3);
+ ASSERT (src == input);
+ ASSERT (mbsinit (&state));
+
+ src = input;
+ ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 1, &state);
+ ASSERT (ret == (unlimited ? 3 : 1));
+ ASSERT (src == (unlimited ? NULL : input + 1));
+ ASSERT (buf[0] == 'n');
+ if (unlimited)
+ {
+ ASSERT (buf[1] == '/');
+ ASSERT (buf[2] == 'a');
+ ASSERT (buf[3] == 0);
+ ASSERT (buf[4] == (wchar_t) 0xBADFACE);
+ }
+ else
+ ASSERT (buf[1] == (wchar_t) 0xBADFACE);
+ ASSERT (mbsinit (&state));
+ }
+ {
+ int c;
+ char input[2];
+
+ memset (&state, '\0', sizeof (mbstate_t));
+ for (c = 0; c < 0x100; c++)
+ if (c != 0)
+ {
+ /* We are testing all nonnull bytes. */
+ input[0] = c;
+ input[1] = '\0';
+
+ src = input;
+ ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 1, &state);
+ ASSERT (ret == 1);
+ ASSERT (src == input);
+ ASSERT (mbsinit (&state));
+
+ buf[0] = buf[1] = (wchar_t) 0xBADFACE;
+ src = input;
+ ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 1, &state);
+ /* POSIX:2018 says: "In the POSIX locale an [EILSEQ] error
+ cannot occur since all byte values are valid characters." */
+ ASSERT (ret == 1);
+ ASSERT (src == (unlimited ? NULL : input + 1));
+ if (c < 0x80)
+ /* c is an ASCII character. */
+ ASSERT (buf[0] == c);
+ else
+ /* On most platforms, the bytes 0x80..0xFF map to U+0080..U+00FF.
+ But on musl libc, the bytes 0x80..0xFF map to U+DF80..U+DFFF. */
+ ASSERT (buf[0] == (btowc (c) == 0xDF00 + c ? btowc (c) : c));
+ ASSERT (mbsinit (&state));
+ }
+ }
+ break;
+
+ case '2':
/* Locale encoding is ISO-8859-1 or ISO-8859-15. */
{
char input[] = "B\374\337er"; /* "Büßer" */
}
break;
- case '2':
+ case '3':
/* Locale encoding is UTF-8. */
{
char input[] = "B\303\274\303\237er"; /* "Büßer" */
}
break;
- case '3':
+ case '4':
/* Locale encoding is EUC-JP. */
{
char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */
}
break;
- case '4':
+ case '5':
/* Locale encoding is GB18030. */
{
char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */
}
break;
- case '5':
- /* C or POSIX locale. */
- {
- char input[] = "n/a";
- memset (&state, '\0', sizeof (mbstate_t));
-
- src = input;
- temp_state = state;
- ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 1, &temp_state);
- ASSERT (ret == 3);
- ASSERT (src == input);
- ASSERT (mbsinit (&state));
-
- src = input;
- ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 1, &state);
- ASSERT (ret == (unlimited ? 3 : 1));
- ASSERT (src == (unlimited ? NULL : input + 1));
- ASSERT (buf[0] == 'n');
- if (unlimited)
- {
- ASSERT (buf[1] == '/');
- ASSERT (buf[2] == 'a');
- ASSERT (buf[3] == 0);
- ASSERT (buf[4] == (wchar_t) 0xBADFACE);
- }
- else
- ASSERT (buf[1] == (wchar_t) 0xBADFACE);
- ASSERT (mbsinit (&state));
- }
- {
- int c;
- char input[2];
-
- memset (&state, '\0', sizeof (mbstate_t));
- for (c = 0; c < 0x100; c++)
- if (c != 0)
- {
- /* We are testing all nonnull bytes. */
- input[0] = c;
- input[1] = '\0';
-
- src = input;
- ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 1, &state);
- ASSERT (ret == 1);
- ASSERT (src == input);
- ASSERT (mbsinit (&state));
-
- buf[0] = buf[1] = (wchar_t) 0xBADFACE;
- src = input;
- ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 1, &state);
- /* POSIX:2018 says: "In the POSIX locale an [EILSEQ] error
- cannot occur since all byte values are valid characters." */
- ASSERT (ret == 1);
- ASSERT (src == (unlimited ? NULL : input + 1));
- if (c < 0x80)
- /* c is an ASCII character. */
- ASSERT (buf[0] == c);
- else
- /* On most platforms, the bytes 0x80..0xFF map to U+0080..U+00FF.
- But on musl libc, the bytes 0x80..0xFF map to U+DF80..U+DFFF. */
- ASSERT (buf[0] == (btowc (c) == 0xDF00 + c ? btowc (c) : c));
- ASSERT (mbsinit (&state));
- }
- }
- break;
-
default:
return 1;
}