X-Git-Url: http://git.chise.org/gitweb/?a=blobdiff_plain;f=src%2Ftext-coding.c;h=fd1dc8fd8c89f36c2e1199c072b987625a7e4ad6;hb=147864f7999198cf34406560387d15a29322670c;hp=70e363ba8c3593b7af8d0a4907a9331d9f84f1b2;hpb=b87aeb962d2cabb053077bb84efffc20bf80a2b9;p=chise%2Fxemacs-chise.git

diff --git a/src/text-coding.c b/src/text-coding.c
index 70e363b..fd1dc8f 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;
     }
@@ -2779,53 +2778,50 @@ text_encode_generic (Lstream *encoding, CONST unsigned char *src,
     {
       c = *src++;
 
-      switch (char_boundary)
+      if (char_boundary == 0)
 	{
-	case 0:
-	  if ( c >= 0xfc )
+	  if (c >= 0xfc)
 	    {
 	      ch = c & 0x01;
 	      char_boundary = 5;
 	    }
-	  else if ( c >= 0xf8 )
+	  else if (c >= 0xf8)
 	    {
 	      ch = c & 0x03;
 	      char_boundary = 4;
 	    }
-	  else if ( c >= 0xf0 )
+	  else if (c >= 0xf0)
 	    {
 	      ch = c & 0x07;
 	      char_boundary = 3;
 	    }
-	  else if ( c >= 0xe0 )
+	  else if (c >= 0xe0)
 	    {
 	      ch = c & 0x0f;
 	      char_boundary = 2;
 	    }
-	  else if ( c >= 0xc0 )
+	  else if (c >= 0xc0)
 	    {
 	      ch = c & 0x1f;
 	      char_boundary = 1;
 	    }
 	  else
-	    {
-	      (*str->encode_char) (str, c, dst, &flags);
-	      ch = 0;
-	      char_boundary = 0;
-	    }
-	  break;
-	case 1:
+	    (*str->encode_char) (str, c, dst, &flags);
+	}
+      else if (char_boundary == 1)
+	{
 	  (*str->encode_char) (str, (ch << 6) | (c & 0x3f), dst, &flags);
 	  ch =0;
 	  char_boundary = 0;
-	  break;
-	default:
-	  ch = ( ch << 6 ) | ( c & 0x3f );
+	}
+      else
+	{
+	  ch = (ch << 6) | (c & 0x3f);
 	  char_boundary--;
 	}
     }
 
-  if ( (char_boundary == 0) && flags & CODING_STATE_END)
+  if ((char_boundary == 0) && (flags & CODING_STATE_END))
     {
       (*str->finish) (str, dst, &flags);
     }
@@ -2833,8 +2829,6 @@ text_encode_generic (Lstream *encoding, CONST unsigned char *src,
   str->flags = flags;
   str->ch    = ch;
   str->iso2022.current_char_boundary = char_boundary;
-
-  /* Verbum caro factum est! */
 }
 
 
@@ -2952,6 +2946,11 @@ decode_coding_sjis (Lstream *decoding, CONST unsigned char *src,
 	      Dynarr_add (dst, c);
 #endif
 	    }
+#ifdef UTF2000
+	  else if (c > 32)
+	    DECODE_ADD_UCS_CHAR(MAKE_CHAR(Vcharset_latin_jisx0201,
+					  c, 0), dst);
+#endif
 	  else
 	    DECODE_ADD_BINARY_CHAR (c, dst);
 	}
@@ -2979,28 +2978,40 @@ 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
+      Lisp_Object value = charset_code_point (Vcharset_latin_jisx0201, ch);
+      Lisp_Object ret = Fcar (value);
+
+      if (INTP (ret))
+	{
+	  charset = Vcharset_latin_jisx0201;
+	  c1 = XINT (ret);
+	  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 +3428,10 @@ 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 +4788,62 @@ 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)))
+	  Lisp_Object code_point;
+
+	  if ((CHARSETP (charset = str->iso2022.charset[i])
+	       && !EQ (code_point = charset_code_point (charset, ch), Qnil))
+	      ||
+	      (CHARSETP
+	       (charset
+		= CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i))
+	       && !EQ (code_point = charset_code_point (charset, ch), Qnil)))
 	    {
+	      Lisp_Object ret = Fcar (code_point);
+
+	      if (INTP (ret))
+		{
+		  byte1 = XINT (ret);
+		  ret = Fcar (Fcdr (code_point));
+		  if (INTP (ret))
+		    byte2 = XINT (ret);
+		  else
+		    byte2 = 0;
+		}
+	      else
+		{
+		  byte1 = 0;
+		  byte2 = 0;
+		}
 	      reg = i;
 	      break;
 	    }
 	}
-	      
+      if (reg == -1)
+	{
+	  Lisp_Object original_default_coded_charset_priority_list
+	    = Vdefault_coded_charset_priority_list;
+
+	  while (!EQ (Vdefault_coded_charset_priority_list, Qnil))
+	    {
+	      BREAKUP_CHAR (ch, charset, byte1, byte2);
+	      if (XCHARSET_FINAL (charset))
+		goto found;
+	      Vdefault_coded_charset_priority_list
+		= Fcdr (Fmemq (XCHARSET_NAME (charset),
+			       Vdefault_coded_charset_priority_list));
+	    }
+	  BREAKUP_CHAR (ch, charset, byte1, byte2);
+	found:
+	  Vdefault_coded_charset_priority_list
+	    = original_default_coded_charset_priority_list;
+	}
+      ensure_correct_direction (XCHARSET_DIRECTION (charset),
+				codesys, dst, flags, 0);
+      
       if (reg == -1)
 	{
 	  if (XCHARSET_GRAPHIC (charset) != 0)
@@ -4947,80 +4997,54 @@ 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--;
 	}