Add some missing mappings to CHINA3-JEF code points.

[chise/xemacs-chise.git] / src / text-coding.c

diff --git a/src/text-coding.c b/src/text-coding.c
index 796e757..795c1e7 100644 (file)
--- a/src/text-coding.c
+++ b/src/text-coding.c
@@ -3502,12 +3502,12 @@ char_encode_shift_jis (struct encoding_stream *str, Emchar ch,
     {
       unsigned int s1, s2;
 #ifdef UTF2000
-      int code_point = charset_code_point (Vcharset_latin_jisx0201, ch);
+      int code_point = charset_code_point (Vcharset_latin_jisx0201, ch, 0);
 
       if (code_point >= 0)
        Dynarr_add (dst, code_point);
       else if ((code_point
-               = charset_code_point (Vcharset_japanese_jisx0208_1990, ch))
+               = charset_code_point (Vcharset_japanese_jisx0208_1990, ch, 0))
               >= 0)
        {
          ENCODE_SJIS ((code_point >> 8) | 0x80,
@@ -3516,11 +3516,11 @@ char_encode_shift_jis (struct encoding_stream *str, Emchar ch,
          Dynarr_add (dst, s2);
        }
       else if ((code_point
-               = charset_code_point (Vcharset_katakana_jisx0201, ch))
+               = charset_code_point (Vcharset_katakana_jisx0201, ch, 0))
               >= 0)
        Dynarr_add (dst, code_point | 0x80);
       else if ((code_point
-               = charset_code_point (Vcharset_japanese_jisx0208, ch))
+               = charset_code_point (Vcharset_japanese_jisx0208, ch, 0))
               >= 0)
        {
          ENCODE_SJIS ((code_point >> 8) | 0x80,
@@ -3528,7 +3528,7 @@ char_encode_shift_jis (struct encoding_stream *str, Emchar ch,
          Dynarr_add (dst, s1);
          Dynarr_add (dst, s2);
        }
-      else if ((code_point = charset_code_point (Vcharset_ascii, ch))
+      else if ((code_point = charset_code_point (Vcharset_ascii, ch, 0))
               >= 0)
        Dynarr_add (dst, code_point);
       else
@@ -3834,21 +3834,21 @@ char_encode_big5 (struct encoding_stream *str, Emchar ch,
       Lisp_Object ccs
        = CODING_SYSTEM_ISO2022_INITIAL_CHARSET (str->codesys, 1);
 
-      if ((code_point = charset_code_point (Vcharset_ascii, ch)) >= 0)
+      if ((code_point = charset_code_point (Vcharset_ascii, ch, 0)) >= 0)
        Dynarr_add (dst, code_point);
-      else if ((code_point = charset_code_point (ccs, ch)) >= 0)
+      else if ((code_point = charset_code_point (ccs, ch, 0)) >= 0)
        {
          Dynarr_add (dst, code_point >> 8);
          Dynarr_add (dst, code_point & 0xFF);
        }
       else if ((code_point
-               = charset_code_point (Vcharset_chinese_big5, ch)) >= 0)
+               = charset_code_point (Vcharset_chinese_big5, ch, 0)) >= 0)
        {
          Dynarr_add (dst, code_point >> 8);
          Dynarr_add (dst, code_point & 0xFF);
        }
       else if ((code_point
-               = charset_code_point (Vcharset_chinese_big5_1, ch)) >= 0)
+               = charset_code_point (Vcharset_chinese_big5_1, ch, 0)) >= 0)
        {
          unsigned int I
            = ((code_point >> 8) - 33) * (0xFF - 0xA1)
@@ -3861,7 +3861,7 @@ char_encode_big5 (struct encoding_stream *str, Emchar ch,
          Dynarr_add (dst, b2);
        }
       else if ((code_point
-               = charset_code_point (Vcharset_chinese_big5_2, ch)) >= 0)
+               = charset_code_point (Vcharset_chinese_big5_2, ch, 0)) >= 0)
        {
          unsigned int I
            = ((code_point >> 8) - 33) * (0xFF - 0xA1)
@@ -4164,15 +4164,17 @@ decode_coding_utf8 (Lstream *decoding, const Extbyte *src,
        {
          if ( c < ' ' )
            {
+             COMPOSE_FLUSH_CHARS (str, dst);
              decode_flush_er_chars (str, dst);
              DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
              DECODE_ADD_UCS_CHAR (c, dst);
            }
          else if ( c < 0xC0 )
-           decode_add_er_char (str, c, dst);
+           /* decode_add_er_char (str, c, dst); */
+           COMPOSE_ADD_CHAR (str, c, dst);
          else
            {
-             decode_flush_er_chars (str, dst);
+             /* decode_flush_er_chars (str, dst); */
              if ( c < 0xE0 )
                {
                  cpos = c & 0x1f;
@@ -4205,7 +4207,8 @@ decode_coding_utf8 (Lstream *decoding, const Extbyte *src,
          cpos = ( cpos << 6 ) | ( c & 0x3f );
          if (counter == 1)
            {
-             DECODE_ADD_UCS_CHAR (cpos, dst);
+             /* DECODE_ADD_UCS_CHAR (cpos, dst); */
+             COMPOSE_ADD_CHAR (str, cpos, dst);
              cpos = 0;
              counter = 0;
            }
@@ -4214,6 +4217,8 @@ decode_coding_utf8 (Lstream *decoding, const Extbyte *src,
        }
       else
        {
+         COMPOSE_FLUSH_CHARS (str, dst);
+         decode_flush_er_chars (str, dst);
          decode_output_utf8_partial_char (counter, cpos, dst);
          DECODE_ADD_BINARY_CHAR (c, dst);
          cpos = 0;
@@ -4224,6 +4229,7 @@ decode_coding_utf8 (Lstream *decoding, const Extbyte *src,
 
   if (flags & CODING_STATE_END)
     {
+      COMPOSE_FLUSH_CHARS (str, dst);
       decode_flush_er_chars (str, dst);
       if (counter > 0)
        {
@@ -4256,7 +4262,7 @@ char_encode_utf8 (struct encoding_stream *str, Emchar ch,
     }
   else
     {
-      int code_point = charset_code_point (Vcharset_ucs, ch);
+      int code_point = charset_code_point (Vcharset_ucs, ch, 0);
 
       if ( (code_point < 0) || (code_point > 0x10FFFF) )
        {
@@ -4276,7 +4282,7 @@ char_encode_utf8 (struct encoding_stream *str, Emchar ch,
                  if (!NILP (ccs = Ffind_charset (ccs)))
                    {
                      if ( (code_point
-                           = charset_code_point (ccs, ch)) >= 0 )
+                           = charset_code_point (ccs, ch, 0)) >= 0 )
                        {
                          cell = Fcdr (cell);
                          ret = Fcar (cell);
@@ -5200,6 +5206,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
 
                case ISO_ESC_LITERAL:
                  COMPOSE_FLUSH_CHARS (str, dst);
+                 decode_flush_er_chars (str, dst);
                  DECODE_ADD_BINARY_CHAR (c, dst);
                  break;
 
@@ -5221,6 +5228,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
              /* Output the (possibly invalid) sequence */
              int i;
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              for (i = 0; i < str->iso2022.esc_bytes_index; i++)
                DECODE_ADD_BINARY_CHAR (str->iso2022.esc_bytes[i], dst);
              flags &= CODING_STATE_ISO2022_LOCK;
@@ -5232,6 +5240,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
                     escape sequence; it could mess things up anyway.
                     Just add it now. */
                  COMPOSE_FLUSH_CHARS (str, dst);
+                 decode_flush_er_chars (str, dst);
                  DECODE_ADD_BINARY_CHAR (c, dst);
                }
            }
@@ -5248,6 +5257,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
          if (counter)
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              while (counter > 0)
                {
                  counter--;
@@ -5264,12 +5274,14 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
          if (flags & CODING_STATE_SS2)
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              DECODE_ADD_BINARY_CHAR (ISO_CODE_SS2, dst);
              flags &= ~CODING_STATE_SS2;
            }
          if (flags & CODING_STATE_SS3)
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              DECODE_ADD_BINARY_CHAR (ISO_CODE_SS3, dst);
              flags &= ~CODING_STATE_SS3;
            }
@@ -5281,6 +5293,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
          if (c == '\r')
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              if (eol_type == EOL_CR)
                Dynarr_add (dst, '\n');
              else if (eol_type != EOL_CRLF || flags & CODING_STATE_CR)
@@ -5304,6 +5317,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
          if (!parse_iso2022_esc (coding_system, &str->iso2022, c, &flags, 1))
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              DECODE_ADD_BINARY_CHAR (c, dst);
            }
        }
@@ -5319,6 +5333,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
          if (c == '\r')
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              if (eol_type == EOL_CR)
                Dynarr_add (dst, '\n');
              else if (eol_type != EOL_CRLF || flags & CODING_STATE_CR)
@@ -5355,6 +5370,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
               to preserve it for the output. */
            {
              COMPOSE_FLUSH_CHARS (str, dst);
+             decode_flush_er_chars (str, dst);
              while (counter > 0)
                {
                  counter--;
@@ -5458,6 +5474,7 @@ decode_coding_iso2022 (Lstream *decoding, const Extbyte *src,
   if (flags & CODING_STATE_END)
     {
       COMPOSE_FLUSH_CHARS (str, dst);
+      decode_flush_er_chars (str, dst);
       DECODE_OUTPUT_PARTIAL_CHAR (cpos);
     }
   str->flags   = flags;
@@ -5636,12 +5653,12 @@ char_encode_iso2022 (struct encoding_stream *str, Emchar ch,
       for (i = 0; i < 4; i++)
        {
          if ((CHARSETP (charset = str->iso2022.charset[i])
-              && ((code_point = charset_code_point (charset, ch)) >= 0))
+              && ((code_point = charset_code_point (charset, ch, 0)) >= 0))
              ||
              (CHARSETP
               (charset
                = CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i))
-              && ((code_point = charset_code_point (charset, ch)) >= 0)))
+              && ((code_point = charset_code_point (charset, ch, 0)) >= 0)))
            {
              reg = i;
              break;