unicode.c

大名鼎鼎的远程登录软件putty的Symbian版源码

	0x79, 0x2D, 0x0233}, {
	0x44, 0x2C, 0x1E10}, {
	0x64, 0x2C, 0x1E11}, {
	0x47, 0x2D, 0x1E20}, {
	0x67, 0x2D, 0x1E21}, {
	0x48, 0x22, 0x1E26}, {
	0x68, 0x22, 0x1E27}, {
	0x48, 0x2C, 0x1E28}, {
	0x68, 0x2C, 0x1E29}, {
	0x4B, 0x27, 0x1E30}, {
	0x6B, 0x27, 0x1E31}, {
	0x4D, 0x27, 0x1E3E}, {
	0x6D, 0x27, 0x1E3F}, {
	0x50, 0x27, 0x1E54}, {
	0x70, 0x27, 0x1E55}, {
	0x56, 0x7E, 0x1E7C}, {
	0x76, 0x7E, 0x1E7D}, {
	0x57, 0x60, 0x1E80}, {
	0x77, 0x60, 0x1E81}, {
	0x57, 0x27, 0x1E82}, {
	0x77, 0x27, 0x1E83}, {
	0x57, 0x22, 0x1E84}, {
	0x77, 0x22, 0x1E85}, {
	0x58, 0x22, 0x1E8C}, {
	0x78, 0x22, 0x1E8D}, {
	0x5A, 0x5E, 0x1E90}, {
	0x7A, 0x5E, 0x1E91}, {
	0x74, 0x22, 0x1E97}, {
	0x77, 0x2A, 0x1E98}, {
	0x79, 0x2A, 0x1E99}, {
	0x45, 0x7E, 0x1EBC}, {
	0x65, 0x7E, 0x1EBD}, {
	0x59, 0x60, 0x1EF2}, {
	0x79, 0x60, 0x1EF3}, {
	0x59, 0x7E, 0x1EF8}, {
	0x79, 0x7E, 0x1EF9},
	    /* Compatible/possibles from UCS */
	{
	0x49, 0x4A, 0x0132}, {
	0x69, 0x6A, 0x0133}, {
	0x4C, 0x4A, 0x01C7}, {
	0x4C, 0x6A, 0x01C8}, {
	0x6C, 0x6A, 0x01C9}, {
	0x4E, 0x4A, 0x01CA}, {
	0x4E, 0x6A, 0x01CB}, {
	0x6E, 0x6A, 0x01CC}, {
	0x44, 0x5A, 0x01F1}, {
	0x44, 0x7A, 0x01F2}, {
	0x64, 0x7A, 0x01F3}, {
	0x2E, 0x2E, 0x2025}, {
	0x21, 0x21, 0x203C}, {
	0x3F, 0x21, 0x2048}, {
	0x21, 0x3F, 0x2049}, {
	0x52, 0x73, 0x20A8}, {
	0x4E, 0x6F, 0x2116}, {
	0x53, 0x4D, 0x2120}, {
	0x54, 0x4D, 0x2122}, {
	0x49, 0x49, 0x2161}, {
	0x49, 0x56, 0x2163}, {
	0x56, 0x49, 0x2165}, {
	0x49, 0x58, 0x2168}, {
	0x58, 0x49, 0x216A}, {
	0x69, 0x69, 0x2171}, {
	0x69, 0x76, 0x2173}, {
	0x76, 0x69, 0x2175}, {
	0x69, 0x78, 0x2178}, {
	0x78, 0x69, 0x217A}, {
	0x31, 0x30, 0x2469}, {
	0x31, 0x31, 0x246A}, {
	0x31, 0x32, 0x246B}, {
	0x31, 0x33, 0x246C}, {
	0x31, 0x34, 0x246D}, {
	0x31, 0x35, 0x246E}, {
	0x31, 0x36, 0x246F}, {
	0x31, 0x37, 0x2470}, {
	0x31, 0x38, 0x2471}, {
	0x31, 0x39, 0x2472}, {
	0x32, 0x30, 0x2473}, {
	0x31, 0x2E, 0x2488}, {
	0x32, 0x2E, 0x2489}, {
	0x33, 0x2E, 0x248A}, {
	0x34, 0x2E, 0x248B}, {
	0x35, 0x2E, 0x248C}, {
	0x36, 0x2E, 0x248D}, {
	0x37, 0x2E, 0x248E}, {
	0x38, 0x2E, 0x248F}, {
	0x39, 0x2E, 0x2490}, {
	0x64, 0x61, 0x3372}, {
	0x41, 0x55, 0x3373}, {
	0x6F, 0x56, 0x3375}, {
	0x70, 0x63, 0x3376}, {
	0x70, 0x41, 0x3380}, {
	0x6E, 0x41, 0x3381}, {
	0x6D, 0x41, 0x3383}, {
	0x6B, 0x41, 0x3384}, {
	0x4B, 0x42, 0x3385}, {
	0x4D, 0x42, 0x3386}, {
	0x47, 0x42, 0x3387}, {
	0x70, 0x46, 0x338A}, {
	0x6E, 0x46, 0x338B}, {
	0x6D, 0x67, 0x338E}, {
	0x6B, 0x67, 0x338F}, {
	0x48, 0x7A, 0x3390}, {
	0x66, 0x6D, 0x3399}, {
	0x6E, 0x6D, 0x339A}, {
	0x6D, 0x6D, 0x339C}, {
	0x63, 0x6D, 0x339D}, {
	0x6B, 0x6D, 0x339E}, {
	0x50, 0x61, 0x33A9}, {
	0x70, 0x73, 0x33B0}, {
	0x6E, 0x73, 0x33B1}, {
	0x6D, 0x73, 0x33B3}, {
	0x70, 0x56, 0x33B4}, {
	0x6E, 0x56, 0x33B5}, {
	0x6D, 0x56, 0x33B7}, {
	0x6B, 0x56, 0x33B8}, {
	0x4D, 0x56, 0x33B9}, {
	0x70, 0x57, 0x33BA}, {
	0x6E, 0x57, 0x33BB}, {
	0x6D, 0x57, 0x33BD}, {
	0x6B, 0x57, 0x33BE}, {
	0x4D, 0x57, 0x33BF}, {
	0x42, 0x71, 0x33C3}, {
	0x63, 0x63, 0x33C4}, {
	0x63, 0x64, 0x33C5}, {
	0x64, 0x42, 0x33C8}, {
	0x47, 0x79, 0x33C9}, {
	0x68, 0x61, 0x33CA}, {
	0x48, 0x50, 0x33CB}, {
	0x69, 0x6E, 0x33CC}, {
	0x4B, 0x4B, 0x33CD}, {
	0x4B, 0x4D, 0x33CE}, {
	0x6B, 0x74, 0x33CF}, {
	0x6C, 0x6D, 0x33D0}, {
	0x6C, 0x6E, 0x33D1}, {
	0x6C, 0x78, 0x33D3}, {
	0x6D, 0x62, 0x33D4}, {
	0x50, 0x48, 0x33D7}, {
	0x50, 0x52, 0x33DA}, {
	0x73, 0x72, 0x33DB}, {
	0x53, 0x76, 0x33DC}, {
	0x57, 0x62, 0x33DD}, {
	0x66, 0x66, 0xFB00}, {
	0x66, 0x69, 0xFB01}, {
	0x66, 0x6C, 0xFB02}, {
	0x73, 0x74, 0xFB06}, {
	0, 0, 0}
    }, *c;

    int nc = -1;

    for (c = composetbl; c->first; c++) {
	if (c->first == first && c->second == second)
	    return c->composed;
    }

    if (recurse == 0) {
	nc = check_compose_internal(second, first, 1);
	if (nc == -1)
	    nc = check_compose_internal(toupper(first), toupper(second), 1);
	if (nc == -1)
	    nc = check_compose_internal(toupper(second), toupper(first), 1);
    }
    return nc;
}

int check_compose(int first, int second)
{
    return check_compose_internal(first, second, 0);
}

int decode_codepage(char *cp_name)
{
    char *s, *d;
    const struct cp_list_item *cpi;
    int codepage = -1;
    CPINFO cpinfo;

    if (!*cp_name) {
	/*
	 * Here we select a plausible default code page based on
	 * the locale the user is in. We wish to select an ISO code
	 * page or appropriate local default _rather_ than go with
	 * the Win125* series, because it's more important to have
	 * CSI and friends enabled by default than the ghastly
	 * Windows extra quote characters, and because it's more
	 * likely the user is connecting to a remote server that
	 * does something Unixy or VMSy and hence standards-
	 * compliant than that they're connecting back to a Windows
	 * box using horrible nonstandard charsets.
	 * 
	 * Accordingly, Robert de Bath suggests a method for
	 * picking a default character set that runs as follows:
	 * first call GetACP to get the system's ANSI code page
	 * identifier, and translate as follows:
	 * 
	 * 1250 -> ISO 8859-2
	 * 1251 -> KOI8-U
	 * 1252 -> ISO 8859-1
	 * 1253 -> ISO 8859-7
	 * 1254 -> ISO 8859-9
	 * 1255 -> ISO 8859-8
	 * 1256 -> ISO 8859-6
	 * 1257 -> ISO 8859-13 (changed from 8859-4 on advice of a Lithuanian)
	 * 
	 * and for anything else, choose direct-to-font.
	 */
	int cp = GetACP();
	switch (cp) {
	  case 1250: cp_name = "ISO-8859-2"; break;
	  case 1251: cp_name = "KOI8-U"; break;
	  case 1252: cp_name = "ISO-8859-1"; break;
	  case 1253: cp_name = "ISO-8859-7"; break;
	  case 1254: cp_name = "ISO-8859-9"; break;
	  case 1255: cp_name = "ISO-8859-8"; break;
	  case 1256: cp_name = "ISO-8859-6"; break;
	  case 1257: cp_name = "ISO-8859-13"; break;
	    /* default: leave it blank, which will select -1, direct->font */
	}
    }

    if (cp_name && *cp_name)
	for (cpi = cp_list; cpi->name; cpi++) {
	    s = cp_name;
	    d = cpi->name;
	    for (;;) {
		while (*s && !isalnum(*s) && *s != ':')
		    s++;
		while (*d && !isalnum(*d) && *d != ':')
		    d++;
		if (*s == 0) {
		    codepage = cpi->codepage;
		    if (codepage == CP_UTF8)
			goto break_break;
		    if (codepage == -1)
			return codepage;
		    if (codepage == 0) {
			codepage = 65536 + (cpi - cp_list);
			goto break_break;
		    }

		    if (GetCPInfo(codepage, &cpinfo) != 0)
			goto break_break;
		}
		if (tolower(*s++) != tolower(*d++))
		    break;
	    }
	}

    if (cp_name && *cp_name) {
	d = cp_name;
	if (tolower(d[0]) == 'c' && tolower(d[1]) == 'p')
	    d += 2;
	if (tolower(d[0]) == 'i' && tolower(d[1]) == 'b'
	    && tolower(d[1]) == 'm')
	    d += 3;
	for (s = d; *s >= '0' && *s <= '9'; s++);
	if (*s == 0 && s != d)
	    codepage = atoi(d);	       /* CP999 or IBM999 */

	if (codepage == CP_ACP)
	    codepage = GetACP();
	if (codepage == CP_OEMCP)
	    codepage = GetOEMCP();
	if (codepage > 65535)
	    codepage = -2;
    }

  break_break:;
    if (codepage != -1) {
	if (codepage != CP_UTF8 && codepage < 65536) {
	    if (GetCPInfo(codepage, &cpinfo) == 0) {
		codepage = -2;
	    } else if (cpinfo.MaxCharSize > 1)
		codepage = -3;
	}
    }
    if (codepage == -1 && *cp_name)
	codepage = -2;
    return codepage;
}

const char *cp_name(int codepage)
{
    const struct cp_list_item *cpi, *cpno;
    static char buf[32];

    if (codepage == -1) {
	sprintf(buf, "Use font encoding");
	return buf;
    }

    if (codepage > 0 && codepage < 65536)
	sprintf(buf, "CP%03d", codepage);
    else
	*buf = 0;

    if (codepage >= 65536) {
	cpno = 0;
	for (cpi = cp_list; cpi->name; cpi++)
	    if (cpi == cp_list + (codepage - 65536)) {
		cpno = cpi;
		break;
	    }
	if (cpno)
	    for (cpi = cp_list; cpi->name; cpi++) {
		if (cpno->cp_table == cpi->cp_table)
		    return cpi->name;
	    }
    } else {
	for (cpi = cp_list; cpi->name; cpi++) {
	    if (codepage == cpi->codepage)
		return cpi->name;
	}
    }
    return buf;
}

/*
 * Return the nth code page in the list, for use in the GUI
 * configurer.
 */
const char *cp_enumerate(int index)
{
    if (index < 0 || index >= lenof(cp_list))
	return NULL;
    return cp_list[index].name;
}

void get_unitab(int codepage, wchar_t * unitab, int ftype)
{
    char tbuf[4];
    int i, max = 256, flg = MB_ERR_INVALID_CHARS;

    if (ftype)
	flg |= MB_USEGLYPHCHARS;
    if (ftype == 2)
	max = 128;

    if (codepage == CP_UTF8) {
	for (i = 0; i < max; i++)
	    unitab[i] = i;
	return;
    }

    if (codepage == CP_ACP)
	codepage = GetACP();
    else if (codepage == CP_OEMCP)
	codepage = GetOEMCP();

    if (codepage > 0 && codepage < 65536) {
	for (i = 0; i < max; i++) {
	    tbuf[0] = i;

	    if (mb_to_wc(codepage, flg, tbuf, 1, unitab + i, 1)
		!= 1)
		unitab[i] = 0xFFFD;
	}
    } else {
	int j = 256 - cp_list[codepage & 0xFFFF].cp_size;
	for (i = 0; i < max; i++)
	    unitab[i] = i;
	for (i = j; i < max; i++)
	    unitab[i] = cp_list[codepage & 0xFFFF].cp_table[i - j];
    }
}

int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen,
	     char *mbstr, int mblen, char *defchr, int *defused,
	     struct unicode_data *ucsdata)
{
    char *p;
    int i;
    if (ucsdata && codepage == ucsdata->line_codepage && ucsdata->uni_tbl) {
	/* Do this by array lookup if we can. */
	if (wclen < 0) {
	    for (wclen = 0; wcstr[wclen++] ;);   /* will include the NUL */
	}
	for (p = mbstr, i = 0; i < wclen; i++) {
	    wchar_t ch = wcstr[i];
	    int by;
	    char *p1;
	    if (ucsdata->uni_tbl && (p1 = ucsdata->uni_tbl[(ch >> 8) & 0xFF])
		&& (by = p1[ch & 0xFF]))
		*p++ = by;
	    else if (ch < 0x80)
		*p++ = (char) ch;
	    else if (defchr) {
		int j;
		for (j = 0; defchr[j]; j++)
		    *p++ = defchr[j];
		if (defused) *defused = 1;
	    }
#if 1
	    else
		*p++ = '.';
#endif
	    assert(p - mbstr < mblen);
	}
	return p - mbstr;
    } else
	return WideCharToMultiByte(codepage, flags, wcstr, wclen,
				   mbstr, mblen, defchr, defused);
}

int mb_to_wc(int codepage, int flags, char *mbstr, int mblen,
	     wchar_t *wcstr, int wclen)
{
    return MultiByteToWideChar(codepage, flags, mbstr, mblen, wcstr, wclen);
}

int is_dbcs_leadbyte(int codepage, char byte)
{
    return IsDBCSLeadByteEx(codepage, byte);
}