X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Ftext-coding.c;h=50a17fc442e029b26c7d1bb2be92310136822c4b;hb=965a56e0f7857093cdc7f9024c5173e4471911f5;hp=f505c0ea50b432b0d784d36018d4c93f3128e73a;hpb=e93fae6da8183cb02f8328b8f53c35ed7f4e412f;p=chise%2Fxemacs-chise.git diff --git a/src/text-coding.c b/src/text-coding.c index f505c0e..50a17fc 100644 --- a/src/text-coding.c +++ b/src/text-coding.c @@ -869,10 +869,6 @@ if TYPE is 'ccl: CHECK_STRING (doc_string); CODING_SYSTEM_DOC_STRING (codesys) = doc_string; -#ifdef UTF2000 - if (ty == CODESYS_NO_CONVERSION) - codesys->fixed.size = 1; -#endif EXTERNAL_PROPERTY_LIST_LOOP (rest, key, value, props) { if (EQ (key, Qmnemonic)) @@ -2562,12 +2558,15 @@ reset_encoding_stream (struct encoding_stream *str) case CODESYS_UTF8: str->encode_char = &char_encode_utf8; str->finish = &char_finish_utf8; + break; case CODESYS_UCS4: str->encode_char = &char_encode_ucs4; str->finish = &char_finish_ucs4; + break; case CODESYS_SHIFT_JIS: str->encode_char = &char_encode_shift_jis; str->finish = &char_finish_shift_jis; + break; default: break; } @@ -2979,28 +2978,39 @@ char_encode_shift_jis (struct encoding_stream *str, Emchar ch, if (eol_type != EOL_CR) Dynarr_add (dst, ch); } - else if (ch <= 0x7f) - Dynarr_add (dst, ch); - else if (ch == 0xA5) - Dynarr_add (dst, 0x5C); - else if (ch == 0X203E) - Dynarr_add (dst, 0x7E); else { Lisp_Object charset; unsigned int c1, c2, s1, s2; - - BREAKUP_CHAR (ch, charset, c1, c2); + +#ifdef UTF2000 + if ( (c1 = + get_byte_from_character_table + (ch, XCHARSET_TO_BYTE1_TABLE (Vcharset_latin_jisx0201))) ) + { + charset = Vcharset_latin_jisx0201; + c2 = 0; + } + else +#endif + BREAKUP_CHAR (ch, charset, c1, c2); + if (EQ(charset, Vcharset_katakana_jisx0201)) { Dynarr_add (dst, c1 | 0x80); } + else if (c2 == 0) + { + Dynarr_add (dst, c1); + } else if (EQ(charset, Vcharset_japanese_jisx0208)) { ENCODE_SJIS (c1 | 0x80, c2 | 0x80, s1, s2); Dynarr_add (dst, s1); Dynarr_add (dst, s2); } + else + Dynarr_add (dst, '?'); } } @@ -3417,10 +3427,10 @@ void char_encode_ucs4 (struct encoding_stream *str, Emchar ch, unsigned_char_dynarr *dst, unsigned int *flags) { - Dynarr_add (dst, ch >> 24); - Dynarr_add (dst, (ch >> 16) & 255); - Dynarr_add (dst, (ch >> 8) & 255); - Dynarr_add (dst, ch & 255); + Dynarr_add (dst, ch >> 24); + Dynarr_add (dst, ch >> 16); + Dynarr_add (dst, ch >> 8); + Dynarr_add (dst, ch ); } void @@ -4777,23 +4787,27 @@ char_encode_iso2022 (struct encoding_stream *str, Emchar ch, { int reg; - BREAKUP_CHAR (ch, charset, byte1, byte2); - ensure_correct_direction (XCHARSET_DIRECTION (charset), - codesys, dst, flags, 0); - /* Now determine which register to use. */ reg = -1; for (i = 0; i < 4; i++) { - if (EQ (charset, str->iso2022.charset[i]) || - EQ (charset, - CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i))) + if ((CHARSETP (charset = str->iso2022.charset[i]) + && (byte1 = charset_get_byte1 (charset, ch))) || + (CHARSETP + (charset + = CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i)) + && (byte1 = charset_get_byte1 (charset, ch)))) { reg = i; + byte2 = charset_get_byte2 (charset, ch); break; } } - + if (reg == -1) + BREAKUP_CHAR (ch, charset, byte1, byte2); + ensure_correct_direction (XCHARSET_DIRECTION (charset), + codesys, dst, flags, 0); + if (reg == -1) { if (XCHARSET_GRAPHIC (charset) != 0) @@ -4947,80 +4961,54 @@ encode_coding_no_conversion (Lstream *encoding, CONST unsigned char *src, { c = *src++; #ifdef UTF2000 - switch (char_boundary) + if (char_boundary == 0) + if ( c >= 0xfc ) + { + ch = c & 0x01; + char_boundary = 5; + } + else if ( c >= 0xf8 ) + { + ch = c & 0x03; + char_boundary = 4; + } + else if ( c >= 0xf0 ) + { + ch = c & 0x07; + char_boundary = 3; + } + else if ( c >= 0xe0 ) + { + ch = c & 0x0f; + char_boundary = 2; + } + else if ( c >= 0xc0 ) + { + ch = c & 0x1f; + char_boundary = 1; + } + else + { + ch = 0; + if (c == '\n') + { + if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT) + Dynarr_add (dst, '\r'); + if (eol_type != EOL_CR) + Dynarr_add (dst, c); + } + else + Dynarr_add (dst, c); + char_boundary = 0; + } + else if (char_boundary == 1) { - case 0: - if ( c >= 0xfc ) - { - ch = c & 0x01; - char_boundary = 5; - } - else if ( c >= 0xf8 ) - { - ch = c & 0x03; - char_boundary = 4; - } - else if ( c >= 0xf0 ) - { - ch = c & 0x07; - char_boundary = 3; - } - else if ( c >= 0xe0 ) - { - ch = c & 0x0f; - char_boundary = 2; - } - else if ( c >= 0xc0 ) - { - ch = c & 0x1f; - char_boundary = 1; - } - else - { - ch = 0; - - if (c == '\n') - { - if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT) - Dynarr_add (dst, '\r'); - if (eol_type != EOL_CR) - Dynarr_add (dst, c); - } - else - Dynarr_add (dst, c); - char_boundary = 0; - } - break; - case 1: ch = ( ch << 6 ) | ( c & 0x3f ); - switch ( str->codesys->fixed.size ) - { - case 1: - Dynarr_add (dst, ch & 0xff); - break; - case 2: - Dynarr_add (dst, (ch >> 8) & 0xff); - Dynarr_add (dst, ch & 0xff); - break; - case 3: - Dynarr_add (dst, (ch >> 16) & 0xff); - Dynarr_add (dst, (ch >> 8) & 0xff); - Dynarr_add (dst, ch & 0xff); - break; - case 4: - Dynarr_add (dst, (ch >> 24) & 0xff); - Dynarr_add (dst, (ch >> 16) & 0xff); - Dynarr_add (dst, (ch >> 8) & 0xff); - Dynarr_add (dst, ch & 0xff); - break; - default: - fprintf(stderr, "It seems %d bytes stream.\n", - str->codesys->fixed.size); - abort (); - } + Dynarr_add (dst, ch & 0xff); char_boundary = 0; - break; - default: + } + else + { ch = ( ch << 6 ) | ( c & 0x3f ); char_boundary--; }