emacs: src/coding.c comparison

comparison src/coding.c @ 18536:69c0e220b626

(Vstandard_character_unification_table_for_decode): Name changed from Vstandard_character_unification_table_for_read. (Vstandard_character_unification_table_for_encode): Name changed from Vstandard_character_unification_table_for_write. (Qcharacter_unification_table_for_decode): New variable. (Qcharacter_unification_table_for_encode): New variable. (decode_coding_iso2022): Adjusted for the above name change. (encode_coding_iso2022): Likewise. (ENCODE_ISO_CHARACTER): Bug of handling return value of unify_char fixed. (DECODE_SJIS_BIG5_CHARACTER): New macro. (decode_coding_sjis_big5): Handle character unificatoin table. Use macro DECODE_SJIS_BIG5_CHARACTER. (encode_coding_sjis_big5): Handle character unificatoin table. Use macro ENCODE_SJIS_BIG5_CHARACTER. (setup_coding_system): Handle members character_unification_table_for_decode and character_unification_table_for_encode of the variable coding. (Ffind_operation_coding_system): Renamed from Ffind_coding_system. (syms_of_coding): Initialize and staticpro Qcharacter_unification_table_for_decode and Qcharacter_unification_table_for_decode. Declare Ffind_operation_coding_system as Lisp functions. Declare Vstandard_character_unification_table_for_decode and Vstandard_character_unification_table_for_encode as Lisp variables.

author	Kenichi Handa <handa@m17n.org>
date	Wed, 02 Jul 1997 12:53:53 +0000
parents	6b5af78c8b21
children	614b916ff5bf

comparison

equal deleted inserted replaced

-:7ca42974dad4
+:69c0e220b626
 };
 /* Flag to tell if we look up unification table on character code
 conversion.  */
 Lisp_Object Venable_character_unification;
-/* Standard unification table to look up on reading (decoding).  */
+/* Standard unification table to look up on decoding (reading).  */
-Lisp_Object Vstandard_character_unification_table_for_read;
+Lisp_Object Vstandard_character_unification_table_for_decode;
-/* Standard unification table to look up on writing (encoding).  */
+/* Standard unification table to look up on encoding (writing).  */
-Lisp_Object Vstandard_character_unification_table_for_write;
+Lisp_Object Vstandard_character_unification_table_for_encode;
 Lisp_Object Qcharacter_unification_table;
+Lisp_Object Qcharacter_unification_table_for_decode;
+Lisp_Object Qcharacter_unification_table_for_encode;
 /* Alist of charsets vs revision number.  */
 Lisp_Object Vcharset_revision_alist;
 /* Default coding systems used for process I/O.  */
 unsigned char *adjusted_dst_end = dst_end - 6;
 int charset;
 /* Charsets invoked to graphic plane 0 and 1 respectively.  */
 int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
 int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
-Lisp_Object unification_table = coding->character_unification_table;
+Lisp_Object unification_table
+= coding->character_unification_table_for_decode;
 if (!NILP (Venable_character_unification) && NILP (unification_table))
-unification_table = Vstandard_character_unification_table_for_read;
+unification_table = Vstandard_character_unification_table_for_decode;
 while (src < src_end && dst < adjusted_dst_end)
 {
 /* SRC_BASE remembers the start position in source in each loop.
 	 The loop will be exited when there's not enough source text
 #define ENCODE_ISO_CHARACTER(charset, c1, c2)				  \
 do {									  \
 int c_alt, charset_alt;						  \
 if (!NILP (unification_table)					  \
 	&& ((c_alt = unify_char (unification_table, -1, charset, c1, c2)) \
-	    < 0))							  \
+	    >= 0))							  \
 SPLIT_CHAR (c_alt, charset_alt, c1, c2);				  \
 else								  \
 charset_alt = charset;						  \
 if (CHARSET_DIMENSION (charset_alt) == 1)				  \
 ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1);		  \
 unsigned char *dst_end = destination + dst_bytes;
 /* Since the maximum bytes produced by each loop is 20, we subtract 19
 from DST_END to assure overflow checking is necessary only at the
 head of loop.  */
 unsigned char *adjusted_dst_end = dst_end - 19;
-Lisp_Object unification_table = coding->character_unification_table;
+Lisp_Object unification_table
+= coding->character_unification_table_for_encode;
 if (!NILP (Venable_character_unification) && NILP (unification_table))
-unification_table = Vstandard_character_unification_table_for_write;
+unification_table = Vstandard_character_unification_table_for_encode;
 while (src < src_end && dst < adjusted_dst_end)
 {
 /* SRC_BASE remembers the start position in source in each loop.
 	 The loop will be exited when there's not enough source text
 b1 = temp / BIG5_SAME_ROW + 0xA1;				  	\
 b2 = temp % BIG5_SAME_ROW;					  	\
 b2 += b2 < 0x3F ? 0x40 : 0x62;				  	\
 } while (0)
+#define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2)			\
+do {									\
+int c_alt, charset_alt = (charset);					\
+if (!NILP (unification_table)					\
+	&& ((c_alt = unify_char (unification_table,			\
+				 -1, (charset), c1, c2)) >= 0))		\
+	  SPLIT_CHAR (c_alt, charset_alt, c1, c2);			\
+if (charset_alt == CHARSET_ASCII || charset_alt < 0)		\
+DECODE_CHARACTER_ASCII (c1);					\
+else if (CHARSET_DIMENSION (charset_alt) == 1)			\
+DECODE_CHARACTER_DIMENSION1 (charset_alt, c1);			\
+else								\
+DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2);		\
+} while (0)
+#define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2)			  \
+do {									  \
+int c_alt, charset_alt;						  \
+if (!NILP (unification_table)					  \
+&& ((c_alt = unify_char (unification_table, -1, charset, c1, c2)) \
+	    >= 0))							  \
+SPLIT_CHAR (c_alt, charset_alt, c1, c2);				  \
+else								  \
+charset_alt = charset;						  \
+if (charset_alt == charset_ascii)					  \
+*dst++ = c1;							  \
+else if (CHARSET_DIMENSION (charset_alt) == 1)			  \
+{									  \
+	if (sjis_p && charset_alt == charset_katakana_jisx0201)		  \
+	  *dst++ = c1;							  \
+	else								  \
+	  *dst++ = charset_alt, *dst++ = c1;				  \
+}									  \
+else								  \
+{									  \
+	c1 &= 0x7F, c2 &= 0x7F;						  \
+	if (sjis_p && charset_alt == charset_jisx0208)			  \
+	  {								  \
+	    unsigned char s1, s2;					  \
+									  \
+	    ENCODE_SJIS (c1, c2, s1, s2);				  \
+	    *dst++ = s1, *dst++ = s2;					  \
+	  }								  \
+	else if (!sjis_p						  \
+		 && (charset_alt == charset_big5_1			  \
+		     || charset_alt == charset_big5_2))			  \
+	  {								  \
+	    unsigned char b1, b2;					  \
+									  \
+	    ENCODE_BIG5 (c1, c2, c3, b1, b2);				  \
+	    *dst++ = b1, *dst++ = b2;					  \
+	  }								  \
+	else								  \
+	  *dst++ = charset_alt, *dst++ = c1, *dst++ = c2;		  \
+}									  \
+} while (0);
 /* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
 Check if a text is encoded in SJIS.  If it is, return
 CODING_CATEGORY_MASK_SJIS, else return 0.  */
 int
 unsigned char *dst_end = destination + dst_bytes;
 /* Since the maximum bytes produced by each loop is 4, we subtract 3
 from DST_END to assure overflow checking is necessary only at the
 head of loop.  */
 unsigned char *adjusted_dst_end = dst_end - 3;
+Lisp_Object unification_table
+= coding->character_unification_table_for_decode;
+if (!NILP (Venable_character_unification) && NILP (unification_table))
+unification_table = Vstandard_character_unification_table_for_decode;
 while (src < src_end && dst < adjusted_dst_end)
 {
 /* SRC_BASE remembers the start position in source in each loop.
 	 The loop will be exited when there's not enough source text
 		*dst++ = c1, src--;
 	    }
 	  else
 	    *dst++ = c1;
 	}
+else if (c1 < 0x20)
+	*dst++ = c1;
 else if (c1 < 0x80)
-	*dst++ = c1;
+	DECODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2);
 else if (c1 < 0xA0 || c1 >= 0xE0)
 	{
 	  /* SJIS -> JISX0208, BIG5 -> Big5 (only if 0xE0 <= c1 < 0xFF) */
 	  if (sjis_p)
 	    {
 	      ONE_MORE_BYTE (c2);
 	      DECODE_SJIS (c1, c2, c3, c4);
-	      DECODE_CHARACTER_DIMENSION2 (charset_jisx0208, c3, c4);
+	      DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4);
 	    }
 	  else if (c1 >= 0xE0 && c1 < 0xFF)
 	    {
 	      int charset;
 	      ONE_MORE_BYTE (c2);
 	      DECODE_BIG5 (c1, c2, charset, c3, c4);
-	      DECODE_CHARACTER_DIMENSION2 (charset, c3, c4);
+	      DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4);
 	    }
 	  else			/* Invalid code */
 	    *dst++ = c1;
 	}
 else
 	{
 	  /* SJIS -> JISX0201-Kana, BIG5 -> Big5 */
 	  if (sjis_p)
-	    DECODE_CHARACTER_DIMENSION1 (charset_katakana_jisx0201, c1);
+	    DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201, c1, /* dummy */ c2);
 	  else
 	    {
 	      int charset;
 	      ONE_MORE_BYTE (c2);
 	      DECODE_BIG5 (c1, c2, charset, c3, c4);
-	      DECODE_CHARACTER_DIMENSION2 (charset, c3, c4);
+	      DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4);
 	    }
 	}
 continue;
 label_end_of_loop:
 unsigned char *dst_end = destination + dst_bytes;
 /* Since the maximum bytes produced by each loop is 2, we subtract 1
 from DST_END to assure overflow checking is necessary only at the
 head of loop.  */
 unsigned char *adjusted_dst_end = dst_end - 1;
+Lisp_Object unification_table
+= coding->character_unification_table_for_encode;
+if (!NILP (Venable_character_unification) && NILP (unification_table))
+unification_table = Vstandard_character_unification_table_for_encode;
 while (src < src_end && dst < adjusted_dst_end)
 {
 /* SRC_BASE remembers the start position in source in each loop.
 	 The loop will be exited when there's not enough source text
 	}
 switch (emacs_code_class[c1])
 	{
 	case EMACS_ascii_code:
+	  ENCODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2);
+	  break;
 	case EMACS_control_code:
 	  *dst++ = c1;
 	  break;
 	case EMACS_carriage_return_code:
 	    *dst++ = '\r';
 	  break;
 	case EMACS_leading_code_2:
 	  ONE_MORE_BYTE (c2);
-	  if (sjis_p && c1 == charset_katakana_jisx0201)
+	  ENCODE_SJIS_BIG5_CHARACTER (c1, c2, /* dummy */ c3);
-	    *dst++ = c2;
-	  else
-	    *dst++ = c1, *dst++ = c2;
 	  break;
 	case EMACS_leading_code_3:
 	  TWO_MORE_BYTES (c2, c3);
-	  c2 &= 0x7F, c3 &= 0x7F;
+	  ENCODE_SJIS_BIG5_CHARACTER (c1, c2, c3);
-	  if (sjis_p && c1 == charset_jisx0208)
-	    {
-	      unsigned char s1, s2;
-	      ENCODE_SJIS (c2, c3, s1, s2);
-	      *dst++ = s1, *dst++ = s2;
-	    }
-	  else if (!sjis_p && (c1 == charset_big5_1 || c1 == charset_big5_2))
-	    {
-	      unsigned char b1, b2;
-	      ENCODE_BIG5 (c1, c2, c3, b1, b2);
-	      *dst++ = b1, *dst++ = b2;
-	    }
-	  else
-	    *dst++ = c1, *dst++ = c2, *dst++ = c3;
 	  break;
 	case EMACS_leading_code_4:
 	  THREE_MORE_BYTES (c2, c3, c4);
-	  *dst++ = c1, *dst++ = c2, *dst++ = c3, *dst++ = c4;
+	  ENCODE_SJIS_BIG5_CHARACTER (c2, c3, c4);
 	  break;
 	case EMACS_leading_code_composition:
 	  coding->composing = 1;
 	  break;
 coding->selective = 0;
 coding->composing = 0;
 coding->direction = 0;
 coding->carryover_size = 0;
 coding->post_read_conversion = coding->pre_write_conversion = Qnil;
-/* We have not yet implemented a way to specify unification table in
+coding->character_unification_table_for_decode = Qnil;
-a coding system.  */
+coding->character_unification_table_for_encode = Qnil;
-coding->character_unification_table = Qnil;
 Vlast_coding_system_used = coding->symbol = coding_system;
 eol_type = Qnil;
 /* Get value of property `coding-system' until we get a vector.
 While doing that, also get values of properties
-`post-read-conversion', `pre-write-conversion', and `eol-type'.  */
+`post-read-conversion', `pre-write-conversion',
+`character-unification-table-for-decode',
+`character-unification-table-for-encode' and `eol-type'.  */
 while (!NILP (coding_system) && SYMBOLP (coding_system))
 {
 if (NILP (coding->post_read_conversion))
 	coding->post_read_conversion = Fget (coding_system,
 					     Qpost_read_conversion);
 if (NILP (coding->pre_write_conversion))
 	coding->pre_write_conversion = Fget (coding_system,
 					     Qpre_write_conversion);
 if (NILP (eol_type))
 	eol_type = Fget (coding_system, Qeol_type);
+if (NILP (coding->character_unification_table_for_decode))
+	coding->character_unification_table_for_decode
+	  = Fget (coding_system, Qcharacter_unification_table_for_decode);
+if (NILP (coding->character_unification_table_for_encode))
+	coding->character_unification_table_for_encode
+	  = Fget (coding_system, Qcharacter_unification_table_for_encode);
 coding_system = Fget (coding_system, Qcoding_system);
 }
+while (!NILP (coding->character_unification_table_for_decode)
+	 && SYMBOLP (coding->character_unification_table_for_decode))
+	coding->character_unification_table_for_decode
+	  = Fget (coding->character_unification_table_for_decode,
+		  Qcharacter_unification_table_for_decode);
+if (!NILP (coding->character_unification_table_for_decode)
+&& !CHAR_TABLE_P (coding->character_unification_table_for_decode))
+coding->character_unification_table_for_decode = Qnil;
+while (!NILP (coding->character_unification_table_for_encode)
+	 && SYMBOLP (coding->character_unification_table_for_encode))
+	coding->character_unification_table_for_encode
+	  = Fget (coding->character_unification_table_for_encode,
+		  Qcharacter_unification_table_for_encode);
+if (!NILP (coding->character_unification_table_for_encode)
+&& !CHAR_TABLE_P (coding->character_unification_table_for_encode))
+coding->character_unification_table_for_encode = Qnil;
 if (!VECTORP (coding_system)
 || XVECTOR (coding_system)->size != 5)
 goto label_invalid_coding_system;
 if (VECTORP (eol_type))
 {
 return keyboard_coding.symbol;
 }
-DEFUN ("find-coding-system", Ffind_coding_system, Sfind_coding_system,
+DEFUN ("find-operation-coding-system", Ffind_operation_coding_system,
-1, MANY, 0,
+Sfind_operation_coding_system,  1, MANY, 0,
-"Choose a coding system for a file operation based on file name.\n\
+"Choose a coding system for an operation based on the target name.\n\
 The value names a pair of coding systems: (ENCODING-SYSTEM DECODING-SYSTEM).\n\
 ENCODING-SYSTEM is the coding system to use for encoding\n\
 \(in case OPERATION does encoding), and DECODING-SYSTEM is the coding system\n\
 for decoding (in case OPERATION does decoding).\n\
 \n\
 Qcharacter_unification_table = intern ("character-unification-table");
 staticpro (&Qcharacter_unification_table);
 Fput (Qcharacter_unification_table, Qchar_table_extra_slots,
 	make_number (0));
+Qcharacter_unification_table_for_decode
+= intern ("character-unification-table-for-decode");
+staticpro (&Qcharacter_unification_table_for_decode);
+Qcharacter_unification_table_for_encode
+= intern ("character-unification-table-for-encode");
+staticpro (&Qcharacter_unification_table_for_encode);
 defsubr (&Scoding_system_spec);
 defsubr (&Scoding_system_p);
 defsubr (&Sread_coding_system);
 defsubr (&Sread_non_nil_coding_system);
 defsubr (&Scheck_coding_system);
 defsubr (&Sencode_big5_char);
 defsubr (&Sset_terminal_coding_system_internal);
 defsubr (&Sterminal_coding_system);
 defsubr (&Sset_keyboard_coding_system_internal);
 defsubr (&Skeyboard_coding_system);
-defsubr (&Sfind_coding_system);
+defsubr (&Sfind_operation_coding_system);
 DEFVAR_LISP ("coding-category-list", &Vcoding_category_list,
 "List of coding-categories (symbols) ordered by priority.");
 {
 int i;
 DEFVAR_LISP ("enable-character-unification", &Venable_character_unification,
 "Non-nil means ISO 2022 encoder/decoder do character unification.");
 Venable_character_unification = Qt;
-DEFVAR_LISP ("standard-character-unification-table-for-read",
+DEFVAR_LISP ("standard-character-unification-table-for-decode",
-&Vstandard_character_unification_table_for_read,
+&Vstandard_character_unification_table_for_decode,
 "Table for unifying characters when reading.");
-Vstandard_character_unification_table_for_read = Qnil;
+Vstandard_character_unification_table_for_decode = Qnil;
-DEFVAR_LISP ("standard-character-unification-table-for-write",
+DEFVAR_LISP ("standard-character-unification-table-for-encode",
-&Vstandard_character_unification_table_for_write,
+&Vstandard_character_unification_table_for_encode,
 "Table for unifying characters when writing.");
-Vstandard_character_unification_table_for_write = Qnil;
+Vstandard_character_unification_table_for_encode = Qnil;
 DEFVAR_LISP ("charset-revision-table", &Vcharset_revision_alist,
 "Alist of charsets vs revision numbers.\n\
 While encoding, if a charset (car part of an element) is found,\n\
 designate it with the escape sequence identifing revision (cdr part of the element).");

Mercurial > emacs

comparison src/coding.c @ 18536:69c0e220b626