qse/samples/cmn/mbwc02.c

327 lines
6.5 KiB
C

#include <qse/cmn/mbwc.h>
#include <qse/cmn/mem.h>
#include <qse/cmn/str.h>
#include <qse/si/sio.h>
#include <locale.h>
#if defined(_WIN32)
# include <windows.h>
#endif
#define R(f) \
do { \
qse_printf (QSE_T("== %s ==\n"), QSE_T(#f)); \
if (f() == -1) return -1; \
} while (0)
static int test1 (void)
{
const qse_wchar_t unistr[] =
{
/*L"\uB108 \uBB50\uAC00 \uC798\uB0AC\uC5B4?",*/
0xB108,
L' ',
0xBB50,
0xAC00,
L' ',
0xC798,
0xB0AC,
0xC5B4,
L'?',
L'\0'
};
const qse_wchar_t* x[] =
{
L"",
L"",
L"Fly to the universe, kick your ass"
};
qse_mchar_t buf[10];
qse_mchar_t buf2[100];
int i;
x[1] = unistr;
for (i = 0; i < QSE_COUNTOF(x); i++)
{
int n;
qse_size_t wlen, mlen;
n = qse_wcstombs (x[i], &wlen, QSE_NULL, &mlen);
qse_printf (QSE_T("[%ls] n=%d, wcslen()=%d, wlen=%d, mlen=%d\n"),
x[i], (int)n, (int)qse_wcslen(x[i]), (int)wlen, (int)mlen);
}
qse_printf (QSE_T("-----------------\n"));
for (i = 0; i < QSE_COUNTOF(x); i++)
{
int n;
qse_size_t wlen, mlen;
mlen = QSE_COUNTOF(buf) - 1;
qse_mbsset (buf, QSE_MT('A'), QSE_COUNTOF(buf));
n = qse_wcstombs (x[i], &wlen, buf, &mlen);
QSE_ASSERT (buf[QSE_COUNTOF(buf)-1] == QSE_MT('A'));
buf[QSE_COUNTOF(buf)-1] = QSE_MT('\0');
qse_printf (QSE_T("%s chars=%d bytes=%d [%hs]\n"),
((n == -2)? QSE_T("BUFFER-SMALL"): (n == -1)? QSE_T("ILLEGAL-CHAR"): QSE_T("FULL")),
(int)wlen, (int)mlen, buf);
}
qse_printf (QSE_T("-----------------\n"));
for (i = 0; i < QSE_COUNTOF(x); i++)
{
int n;
qse_size_t wlen, mlen;
mlen = QSE_COUNTOF(buf2);
n = qse_wcstombs (x[i], &wlen, buf2, &mlen);
qse_printf (QSE_T("%s chars=%d bytes=%d [%hs]\n"),
((n == -2)? QSE_T("BUFFER-SMALL"): (n == -1)? QSE_T("ILLEGAL-CHAR"): QSE_T("FULL")),
(int)wlen, (int)mlen, buf2);
}
qse_printf (QSE_T("-----------------\n"));
for (i = 0; i < QSE_COUNTOF(x); i++)
{
int n;
const qse_wchar_t* p = x[i];
while (*p)
{
qse_size_t wlen, mlen;
mlen = QSE_COUNTOF(buf) - 1;
n = qse_wcstombs (p, &wlen, buf, &mlen);
qse_printf (QSE_T("%s chars=%d bytes=%d [%hs]\n"),
((n == -2)? QSE_T("BUFFER-SMALL"): (n == -1)? QSE_T("ILLEGAL-CHAR"): QSE_T("FULL")),
(int)wlen, (int)mlen, buf);
p += wlen;
}
}
return 0;
}
static int test2 (void)
{
const qse_wchar_t unistr[] =
{
/*L"\uB108 \uBB50\uAC00 \uC798\uB0AC\uC5B4?",*/
0xB108,
L' ',
0xBB50,
0xAC00,
L' ',
0xC798,
0xB0AC,
0xC5B4,
L'?',
L'\0'
};
const qse_wchar_t* x[] =
{
L"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
L"",
L"Fly to the universe, kick your ass"
};
qse_mchar_t buf[9];
int i, j;
x[1] = unistr;
for (i = 0; i < QSE_COUNTOF(x); i++)
{
qse_size_t len, wlen, mlen, iter;
int n;
len = qse_wcslen(x[i]);
if (len == 0) len = 20; /* for x[0] */
qse_printf (QSE_T("Converting %d wide-character - "), (int)len);
wlen = len;
n = qse_wcsntombsn (x[i], &wlen, QSE_NULL, &mlen);
if (n == -1)
{
qse_printf (QSE_T("***illegal character[mlen=%d/wlen=%d]*** ["), (int)mlen, (int)wlen);
}
else
{
/* -2 is never returned if i don't provide the multibyte buffer */
qse_printf (QSE_T("%d multibyte characters to be produced from %d wide characters ["), (int)mlen, (int)wlen);
}
for (j = 0, iter = 0; j < len; j += wlen)
{
iter = iter + 1;
wlen = len - j;
mlen = QSE_COUNTOF(buf);
n = qse_wcsntombsn (&x[i][j], &wlen, buf, &mlen);
if (n == -1)
{
qse_printf (QSE_T("***illegal character*** wlen=%d/mlen=%d/n=%d"), (int)(len-j), (int)mlen, (int)n);
break;
}
#if 0
else if (n == -2)
{
qse_printf (QSE_T("***buffer not large*** wlen=%d/mlen=%d/n=%d"), (int)(len-j), (int)mlen, (int)n);
break;
}
#endif
if (mlen > 0 && buf[0] == QSE_T('\0'))
{
while (mlen > 0 && buf[--mlen] == QSE_T('\0'))
{
qse_printf (QSE_T("\\0"));
}
}
else
{
#if defined(QSE_CHAR_IS_MCHAR) || defined(_WIN32)
qse_printf (QSE_T("%.*hs"), (int)mlen, buf);
#else
/* at least on linux and macosx, wprintf seems
* to taks preceision as the number of wide
* characters with %s. */
/* THIS HACK SHOULD BE REMOVED ONCE I IMPLEMENTE
* MY OWN qse_printf */
qse_printf (QSE_T("%.*hs"), (int)wlen, buf);
#endif
}
}
qse_printf (QSE_T("] => %d iterations\n"), (int)iter);
}
return 0;
}
static int test3 (void)
{
const qse_wchar_t unistr[] =
{
/*L"\uB108 \uBB50\uAC00 \uC798\uB0AC\uC5B4?",*/
0xB108,
L' ',
0xBB50,
0xAC00,
L' ',
0xC798,
0xB0AC,
0xC5B4,
L'?',
L'\0'
};
const qse_wchar_t* x[] =
{
L"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
L"",
L"Fly to the universe, kick your ass"
};
int i;
x[1] = unistr;
for (i = 0; i < QSE_COUNTOF(x); i++)
{
qse_mchar_t* m = qse_wcstombsdup (x[i], QSE_NULL, QSE_MMGR_GETDFL());
if (m == QSE_NULL)
{
qse_printf (QSE_T("[ERROR]\n"), m);
}
else
{
qse_printf (QSE_T("[%hs]\n"), m);
}
QSE_MMGR_FREE (QSE_MMGR_GETDFL(), m);
}
return 0;
}
static int test4 (void)
{
const qse_wchar_t unistr[] =
{
/*L"\uB108 \uBB50\uAC00 \uC798\uB0AC\uC5B4?",*/
0xB108,
L' ',
0xBB50,
0xAC00,
L' ',
0xC798,
0xB0AC,
0xC5B4,
L'?',
L'\0'
};
const qse_wchar_t* x[] =
{
L"Good morning angels",
L"",
L"Fly to the universe, kick your ass",
QSE_NULL
};
qse_mchar_t* m;
x[1] = unistr;
m = qse_wcsatombsdup (x, QSE_NULL, QSE_MMGR_GETDFL());
if (m == QSE_NULL)
{
qse_printf (QSE_T("[ERROR]\n"), m);
}
else
{
qse_printf (QSE_T("[%hs]\n"), m);
}
QSE_MMGR_FREE (QSE_MMGR_GETDFL(), m);
return 0;
}
int main ()
{
#if defined(_WIN32)
char locale[100];
UINT codepage = GetConsoleOutputCP();
if (codepage == CP_UTF8)
{
/*SetConsoleOUtputCP (CP_UTF8);*/
qse_setdflcmgrbyid (QSE_CMGR_UTF8);
}
else
{
sprintf (locale, ".%u", (unsigned int)codepage);
setlocale (LC_ALL, locale);
/*qse_setdflcmgrbyid (QSE_CMGR_SLMB);*/
}
#else
setlocale (LC_ALL, "");
/*qse_setdflcmgrbyid (QSE_CMGR_SLMB);*/
#endif
qse_open_stdsios ();
qse_printf (QSE_T("--------------------------------------------------------------------------------\n"));
qse_printf (QSE_T("Set the environment LANG to a Unicode locale such as UTF-8 if you see the illegal XXXXX errors. If you see such errors in Unicode locales, this program might be buggy. It is normal to see such messages in non-Unicode locales as it uses Unicode data\n"));
qse_printf (QSE_T("--------------------------------------------------------------------------------\n"));
R (test1);
R (test2);
R (test3);
R (test4);
qse_close_stdsios ();
return 0;
}