charset.c

ARM9-2410教学实验系统下Linux下minigui程序

    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-8"))
        return 0;

    if (strstr (name, "HEBREW"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_8_conv_to_uc16 (const unsigned char* mchar, int len)
{
    if (*mchar >= 0xE0 && *mchar <= 0xFA) {
        return  *mchar - 0xE0 + 0x05D0;
    }
    else {
        switch (*mchar) {
            case 0xAA:
                return 0x00D7;
            case 0xBA:
                return 0x00F7;
            case 0xDF:
                return 0x2017;
            case 0xFD:
                return 0x200E;
            case 0xFE:
                return 0x200F;
        }
        return (unsigned short) (*mchar);
    }
}
#endif

static CHARSETOPS CharsetOps_iso8859_8 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_8,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_8_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_8_conv_to_uc16
#endif
};

#endif /* _GREEK_SUPPORT */

#ifdef _LATIN5_SUPPORT

/************************* ISO8859-9 Specific Operations **********************/
static int iso8859_9_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-9"))
        return 0;

    if (strstr (name, "LATIN5"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_9_conv_to_uc16 (const unsigned char* mchar, int len)
{
    switch (*mchar) {
        case 0xD0:
                return 0x011E;
        case 0xDD:
                return 0x0130;
        case 0xDE:
                return 0x015E;
        case 0xF0:
                return 0x011F;
        case 0xFD:
                return 0x0131;
        case 0xFE:
                return 0x015F;
    }

    return (unsigned short) (*mchar);
}
#endif

static CHARSETOPS CharsetOps_iso8859_9 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_9,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_9_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_9_conv_to_uc16
#endif
};

#endif /* _LATIN5_SUPPORT */

#ifdef _LATIN6_SUPPORT

/************************* ISO8859-10 Specific Operations **********************/
static int iso8859_10_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-10"))
        return 0;

    if (strstr (name, "LATIN6"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_10_unicode_map [] =
{
    0x0104, 0x0112, 0x0122,
    0x012A, 0x0128, 0x0136,
    0x00A7, 0x013B, 0x0110,
    0x0160, 0x0166, 0x017D,
    0x00AD, 0x016A, 0x014A,
    0x00B0, 0x0105, 0x0113,
    0x0123, 0x012B, 0x0129,
    0x0137, 0x00B7, 0x013C,
    0x0111, 0x0161, 0x0167,
    0x017E, 0x2015, 0x016B,
    0x014B, 0x0100, 0x00C1,
    0x00C2, 0x00C3, 0x00C4,
    0x00C5, 0x00C6, 0x012E,
    0x010C, 0x00C9, 0x0118,
    0x00CB, 0x0116, 0x00CD,
    0x00CE, 0x00CF, 0x00D0,
    0x0145, 0x014C, 0x00D3,
    0x00D4, 0x00D5, 0x00D6,
    0x0168, 0x00D8, 0x0172,
    0x00DA, 0x00DB, 0x00DC,
    0x00DD, 0x00DE, 0x00DF,
    0x0101, 0x00E1, 0x00E2,
    0x00E3, 0x00E4, 0x00E5,
    0x00E6, 0x012F, 0x010D,
    0x00E9, 0x0119, 0x00EB,
    0x0117, 0x00ED, 0x00EE,
    0x00EF, 0x00F0, 0x0146,
    0x014D, 0x00F3, 0x00F4,
    0x00F5, 0x00F6, 0x0169,
    0x00F8, 0x0173, 0x00FA,
    0x00FB, 0x00FC, 0x00FD,
    0x00FE, 0x0138
};

static unsigned short iso8859_10_conv_to_uc16 (const unsigned char* mchar, int len)
{
    if (*mchar < 0xA1)
        return (unsigned short) (*mchar);
    else
        return iso8859_10_unicode_map [*mchar - 0xA1];
}
#endif

static CHARSETOPS CharsetOps_iso8859_10 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_10,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_10_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_10_conv_to_uc16
#endif
};

#endif /* _LATIN6_SUPPORT */

#ifdef _THAI_SUPPORT

/************************* ISO8859-11 Specific Operations **********************/
static int iso8859_11_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-11"))
        return 0;

    if (strstr (name, "TIS") && strstr (name, "620"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_11_unicode_map [] =
{
    0x0E01, 0x0E02, 0x0E03,
    0x0E04, 0x0E05, 0x0E06,
    0x0E07, 0x0E08, 0x0E09,
    0x0E0A, 0x0E0B, 0x0E0C,
    0x0E0D, 0x0E0E, 0x0E0F,
    0x0E10, 0x0E11, 0x0E12,
    0x0E13, 0x0E14, 0x0E15,
    0x0E16, 0x0E17, 0x0E18,
    0x0E19, 0x0E1A, 0x0E1B,
    0x0E1C, 0x0E1D, 0x0E1E,
    0x0E1F, 0x0E20, 0x0E21,
    0x0E22, 0x0E23, 0x0E24,
    0x0E25, 0x0E26, 0x0E27,
    0x0E28, 0x0E29, 0x0E2A,
    0x0E2B, 0x0E2C, 0x0E2D,
    0x0E2E, 0x0E2F, 0x0E30,
    0x0E31, 0x0E32, 0x0E33,
    0x0E34, 0x0E35, 0x0E36,
    0x0E37, 0x0E38, 0x0E39,
    0x0E3A, 0x00DB, 0x00DC,
    0x00DD, 0x00DE, 0x0E3F,
    0x0E40, 0x0E41, 0x0E42,
    0x0E43, 0x0E44, 0x0E45,
    0x0E46, 0x0E47, 0x0E48,
    0x0E49, 0x0E4A, 0x0E4B,
    0x0E4C, 0x0E4D, 0x0E4E,
    0x0E4F, 0x0E50, 0x0E51,
    0x0E52, 0x0E53, 0x0E54,
    0x0E55, 0x0E56, 0x0E57,
    0x0E58, 0x0E59, 0x0E5A,
    0x0E5B, 0x00FC, 0x00FD,
    0x00FE, 0x00FF
};

static unsigned short iso8859_11_conv_to_uc16 (const unsigned char* mchar, int len)
{
    if (*mchar < 0xA1)
        return (unsigned short) (*mchar);
    else
        return iso8859_11_unicode_map [*mchar - 0xA1];
}
#endif

static CHARSETOPS CharsetOps_iso8859_11 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_11,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_11_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_11_conv_to_uc16
#endif
};

#endif /* _THAI_SUPPORT */

#ifdef _LATIN7_SUPPORT

/************************* ISO8859-13 Specific Operations **********************/
static int iso8859_13_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-13"))
        return 0;

    if (strstr (name, "LATIN7"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_13_unicode_map [] =
{
    0x201D, 0x00A2, 0x00A3,
    0x00A4, 0x201E, 0x00A6,
    0x00A7, 0x00D8, 0x00A9,
    0x0156, 0x00AB, 0x00AC,
    0x00AD, 0x00AE, 0x00C6,
    0x00B0, 0x00B1, 0x00B2,
    0x00B3, 0x201C, 0x00B5,
    0x00B6, 0x00B7, 0x00F8,
    0x00B9, 0x0157, 0x00BB,
    0x00BC, 0x00BD, 0x00BE,
    0x00E6, 0x0104, 0x012E,
    0x0100, 0x0106, 0x00C4,
    0x00C5, 0x0118, 0x0112,
    0x010C, 0x00C9, 0x0179,
    0x0116, 0x0122, 0x0136,
    0x012A, 0x013B, 0x0160,
    0x0143, 0x0145, 0x00D3,
    0x014C, 0x00D5, 0x00D6,
    0x00D7, 0x0172, 0x0141,
    0x015A, 0x016A, 0x00DC,
    0x017B, 0x017D, 0x00DF,
    0x0105, 0x012F, 0x0101,
    0x0107, 0x00E4, 0x00E5,
    0x0119, 0x0113, 0x010D,
    0x00E9, 0x017A, 0x0117,
    0x0123, 0x0137, 0x012B,
    0x013C, 0x0161, 0x0144,
    0x0146, 0x00F3, 0x014D,
    0x00F5, 0x00F6, 0x00F7,
    0x0173, 0x0142, 0x015B,
    0x016B, 0x00FC, 0x017C,
    0x017E, 0x2019
};

static unsigned short iso8859_13_conv_to_uc16 (const unsigned char* mchar, int len)
{
    if (*mchar < 0xA1)
        return (unsigned short) (*mchar);
    else
        return iso8859_13_unicode_map [*mchar - 0xA1];
}
#endif

static CHARSETOPS CharsetOps_iso8859_13 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_13,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_13_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_13_conv_to_uc16
#endif
};

#endif /* _LATIN7_SUPPORT */

#ifdef _LATIN8_SUPPORT

/************************* ISO8859-14 Specific Operations **********************/
static int iso8859_14_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-14"))
        return 0;

    if (strstr (name, "LATIN8"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_14_unicode_map [] =
{
    0x1E02, 0x1E03, 0x00A3,
    0x010A, 0x010B, 0x1E0A,
    0x00A7, 0x1E80, 0x00A9,
    0x1E82, 0x1E0B, 0x1EF2,
    0x00AD, 0x00AE, 0x0178,
    0x1E1E, 0x1E1F, 0x0120,
    0x0121, 0x1E40, 0x1E41,
    0x00B6, 0x1E56, 0x1E81,
    0x1E57, 0x1E83, 0x1E60,
    0x1EF3, 0x1E84, 0x1E85,
    0x1E61, 0x00C0, 0x00C1,
    0x00C2, 0x00C3, 0x00C4,
    0x00C5, 0x00C6, 0x00C7,
    0x00C8, 0x00C9, 0x00CA,
    0x00CB, 0x00CC, 0x00CD,
    0x00CE, 0x00CF, 0x0174,
    0x00D1, 0x00D2, 0x00D3,
    0x00D4, 0x00D5, 0x00D6,
    0x1E6A, 0x00D8, 0x00D9,
    0x00DA, 0x00DB, 0x00DC,
    0x00DD, 0x0176, 0x00DF,
    0x00E0, 0x00E1, 0x00E2,
    0x00E3, 0x00E4, 0x00E5,
    0x00E6, 0x00E7, 0x00E8,
    0x00E9, 0x00EA, 0x00EB,
    0x00EC, 0x00ED, 0x00EE,
    0x00EF, 0x0175, 0x00F1,
    0x00F2, 0x00F3, 0x00F4,
    0x00F5, 0x00F6, 0x1E6B,
    0x00F8, 0x00F9, 0x00FA,
    0x00FB, 0x00FC, 0x00FD,
    0x0177, 0x00FF
};

static unsigned short iso8859_14_conv_to_uc16 (const unsigned char* mchar, int len)
{
    if (*mchar < 0xA1)
        return (unsigned short) (*mchar);
    else
        return iso8859_14_unicode_map [*mchar - 0xA1];
}
#endif

static CHARSETOPS CharsetOps_iso8859_14 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_14,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_14_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_14_conv_to_uc16
#endif
};

#endif /* _LATIN8_SUPPORT */

#ifdef _LATIN9_SUPPORT

/************************* ISO8859-15 Specific Operations **********************/
static int iso8859_15_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-15"))
        return 0;

    if (strstr (name, "LATIN") && strstr (name, "9"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_15_conv_to_uc16 (const unsigned char* mchar, int len)
{
    switch (*mchar) {
        case 0xA4:
            return 0x20AC;  /* EURO SIGN */
        case 0xA6:
            return 0x0160;  /* LATIN CAPITAL LETTER S WITH CARO */
        case 0xA8:
            return 0x0161;  /* LATIN SMALL LETTER S WITH CARON */
        case 0xB4:
            return 0x017D;  /* LATIN CAPITAL LETTER Z WITH CARON */
        case 0xB8:
            return 0x017E;  /* LATIN SMALL LETTER Z WITH CARON */
        case 0xBC:
            return 0x0152;  /* LATIN CAPITAL LIGATURE OE */
        case 0xBD:
            return 0x0153;  /* LATIN SMALL LIGATURE OE */
        case 0xBE:
            return 0x0178;  /* LATIN CAPITAL LETTER Y WITH DIAERESIS */
        default:
            return *mchar;
    }
}
#endif

static CHARSETOPS CharsetOps_iso8859_15 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_15,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_15_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_15_conv_to_uc16
#endif
};
#endif /* _LATIN9_SUPPORT */

#ifdef _LATIN10_SUPPORT

/************************* ISO8859-16 Specific Operations **********************/
static int iso8859_16_is_this_charset (const unsigned char* charset)
{
    int i;
    char name [LEN_FONT_NAME + 1];

    for (i = 0; i < LEN_FONT_NAME + 1; i++) {
        if (charset [i] == '\0')
            break;
        name [i] = toupper (charset [i]);
    }
    name [i] = '\0';

    if (strstr (name, "ISO") && strstr (name, "8859-16"))
        return 0;

    if (strstr (name, "LATIN") && strstr (name, "10"))
        return 0;

    return 1;
}

#ifdef _UNICODE_SUPPORT
static unsigned short iso8859_16_conv_to_uc16 (const unsigned char* mchar, int len)
{
    switch (*mchar) {
        case 0xA1: return 0x0104;
        case 0xA2: return 0x0105;
        case 0xA3: return 0x0141;
        case 0xA4: return 0x20AC;
        case 0xA5: return 0x201E;
        case 0xA6: return 0x0160;
        case 0xA8: return 0x0161;
        case 0xAA: return 0x0218;
        case 0xAC: return 0x0179;
        case 0xAE: return 0x017A;
        case 0xAF: return 0x017B;
        case 0xB2: return 0x010C;
        case 0xB3: return 0x0142;
        case 0xB4: return 0x017D;
        case 0xB5: return 0x201D;
        case 0xB8: return 0x017E;
        case 0xB9: return 0x010D;
        case 0xBA: return 0x0219;
        case 0xBC: return 0x0152;
        case 0xBD: return 0x0153;
        case 0xBE: return 0x0178;
        case 0xBF: return 0x017C;
        case 0xC3: return 0x0102;
        case 0xC5: return 0x0106;
        case 0xD0: return 0x0110;
        case 0xD1: return 0x0143;
        case 0xD5: return 0x0150;
        case 0xD7: return 0x015A;
        case 0xD8: return 0x0170;
        case 0xDD: return 0x0118;
        case 0xDE: return 0x021A;
        case 0xE3: return 0x0103;
        case 0xE5: return 0x0107;
        case 0xF0: return 0x0111;
        case 0xF1: return 0x0144;
        case 0xF5: return 0x0151;
        case 0xF7: return 0x015B;
        case 0xF8: return 0x0171;
        case 0xFD: return 0x0119;
        case 0xFE: return 0x021B;
    }

    return (unsigned short) (*mchar);
}
#endif

static CHARSETOPS CharsetOps_iso8859_16 = {
    256,
    1,
    1,
    FONT_CHARSET_ISO8859_16,
    {' '},
    sb_len_first_char,
    sb_char_offset,
    sb_nr_chars_in_str,
    iso8859_16_is_this_charset,
    sb_len_first_substr,
    sb_get_next_word,
    sb_pos_first_char,
#ifdef _UNICODE_SUPPORT
    iso8859_16_conv_to_uc16
#endif
};

#endif /* _LATIN10_SUPPORT */

/*************** Common Operations for double bytes charsets ******************/
#if defined(_GB_SUPPORT) | defined(_GBK_SUPPORT) | defined(_BIG5_SUPPORT) \
        | defined(_EUCKR_SUPPORT) | defined(_EUCJP_SUPPORT) | defined(_SHIFTJIS_SUPPORT) \
        | defined(_UNICODE_SUPPORT)
static int db_nr_chars_in_str (const unsigned char* mstr, int mstrlen)
{
    assert ((mstrlen % 2) == 0);
    return mstrlen >> 1;
}

static const unsigned char* db_get_next_word (const unsigned char* mstr,
                int mstrlen, WORDINFO* word_info)
{
    assert ((mstrlen % 2) == 0);

    if (mstrlen < 2) return NULL;

    word_info->len = 2;
    word_info->delimiter = '\0';
    word_info->nr_delimiters = 0;

    return mstr + 2;
}
#endif

#ifdef _GB_SUPPORT
/************************* GB2312 Specific Operations ************************/
#define IS_GB2312_CHAR(ch1, ch2) \
        if (((ch1 >= 0xA1 && ch1 <= 0xA9) || (ch1 >= 0xB0 && ch1 <= 0xF7)) \
                        && ch2 >= 0xA1 && ch2 <= 0xFE)

static int gb2312_0_len_first_char (const unsigned char* mstr, int len)
{
    unsigned char ch1;
    unsigned char ch2;

    if (len < 2) return 0;

    ch1 = mstr [0];
    if (ch1 == '\0')
        return 0;

    ch2 = mstr [1];
    IS_GB2312_CHAR (ch1, ch2)