emacs: src/charset.c comparison

comparison src/charset.c @ 88475:9d54c1d6cd0a

(load_charset_map): Fix previous change. (read_hex): Don't treat SPC as a comment starter. (decode_char): If CODE_POINT_TO_INDEX retruns -1, always return -1. (Fdecode_char): Fix typo. (CODE_POINT_TO_INDEX): Utilize `code_space_mask' member to check if CODE is valid or not. (Fdefine_charset_internal): Initialize `code_space_mask' member. (encode_char): Before calling CODE_POINT_TO_INDEX, check if CODE is within the range of charset->min_code and carset->max_code.

author	Kenichi Handa <handa@m17n.org>
date	Fri, 10 May 2002 03:54:48 +0000
parents	bae0bd953f61
children	9c3255b975cc

comparison

equal deleted inserted replaced

-:fcc1029dcac2
+:9d54c1d6cd0a
 Lisp_Object Vcharset_map_directory;
 Lisp_Object Vchar_unified_charset_table;
-#define CODE_POINT_TO_INDEX(charset, code)			\
+#define CODE_POINT_TO_INDEX(charset, code)				\
-((charset)->code_linear_p					\
+((charset)->code_linear_p						\
-? (code) - (charset)->min_code				\
+? (code) - (charset)->min_code					\
-: ((((code) >> 24) <= (charset)->code_space[13])		\
+: (((charset)->code_space_mask[(code) >> 24] & 0x8)			\
-&& ((((code) >> 16) & 0xFF) <= (charset)->code_space[9])	\
+&& ((charset)->code_space_mask[((code) >> 16) & 0xFF] & 0x4)	\
-&& ((((code) >> 8) & 0xFF) <= (charset)->code_space[5])	\
+&& ((charset)->code_space_mask[((code) >> 8) & 0xFF] & 0x2)	\
-&& (((code) & 0xFF) <= (charset)->code_space[1]))		\
+&& ((charset)->code_space_mask[(code) & 0xFF] & 0x1))		\
-? (((((code) >> 24) - (charset)->code_space[12])		\
+? (((((code) >> 24) - (charset)->code_space[12])			\
-* (charset)->code_space[11])				\
+* (charset)->code_space[11])					\
-+ (((((code) >> 16) & 0xFF) - (charset)->code_space[8])	\
++ (((((code) >> 16) & 0xFF) - (charset)->code_space[8])		\
-	 * (charset)->code_space[7])				\
+	 * (charset)->code_space[7])					\
-+ (((((code) >> 8) & 0xFF) - (charset)->code_space[4])	\
++ (((((code) >> 8) & 0xFF) - (charset)->code_space[4])		\
-	 * (charset)->code_space[3])				\
+	 * (charset)->code_space[3])					\
-+ (((code) & 0xFF) - (charset)->code_space[0]))		\
++ (((code) & 0xFF) - (charset)->code_space[0]))			\
 : -1)
 /* Convert the character index IDX to code-point CODE for CHARSET.
 It is assumed that IDX is in a valid range.  */
 	  CHARSET_FAST_MAP_SET (c, fast_map);
 	}
 else
 	{
-	  for (; from <= to; from++)
+	  unsigned code = from;
+	  int from_index, to_index;
+	  from_index = CODE_POINT_TO_INDEX (charset, from);
+	  if (from == to)
+	    to_index = from_index;
+	  else
+	    to_index = CODE_POINT_TO_INDEX (charset, to);
+	  if (from_index < 0 || to_index < 0)
+	    continue;
+	  while (1)
 	    {
-	      int c1 = DECODE_CHAR (charset, from);
+	      int c1 = DECODE_CHAR (charset, code);
 	      if (c1 >= 0)
 		{
 		  CHAR_TABLE_SET (table, c, make_number (c1));
 		  CHAR_TABLE_SET (Vchar_unify_table, c1, c);
 		  if (CHAR_TABLE_P (Vchar_unified_charset_table))
 		    CHAR_TABLE_SET (Vchar_unified_charset_table, c1,
 				    CHARSET_NAME (charset));
 		}
+	      if (from_index == to_index)
+		break;
+	      from_index++, c++;
+	      code = INDEX_TO_CODE_POINT (charset, from_index);
 	    }
 	}
 }
 if (control_flag < 2)
 int c;
 unsigned n;
 while ((c = getc (fp)) != EOF)
 {
-if (c == '#' || c == ' ')
+if (c == '#')
 	{
 	  while ((c = getc (fp)) != EOF && c != '\n');
 	}
 else if (c == '0')
 	{
 /* Charset attr vector.  */
 Lisp_Object attrs;
 Lisp_Object val;
 unsigned hash_code;
 struct Lisp_Hash_Table *hash_table = XHASH_TABLE (Vcharset_hash_table);
-int i;
+int i, j;
 struct charset charset;
 int id;
 int dimension;
 int new_definition_p;
 int nchars;
 || (charset.code_space[2] == 256
 	   && (charset.dimension == 2
 	       || (charset.code_space[6] == 256
 		   && (charset.dimension == 3
 		       || charset.code_space[10] == 256)))));
+if (! charset.code_linear_p)
+{
+charset.code_space_mask = (unsigned char *) xmalloc (256);
+bzero (charset.code_space_mask, sizeof (charset.code_space_mask));
+for (i = 0; i < 4; i++)
+	for (j = charset.code_space[i * 4]; j <= charset.code_space[i * 4 + 1];
+	     j++)
+	  charset.code_space_mask[j] |= (1 << i);
+}
 charset.iso_chars_96 = charset.code_space[2] == 96;
 charset.min_code = (charset.code_space[0]
 		      | (charset.code_space[4] << 8)
 	}
 }
 else
 {
 char_index = CODE_POINT_TO_INDEX (charset, code);
+if (char_index < 0)
+	return -1;
 if (method == CHARSET_METHOD_MAP)
 	{
 	  Lisp_Object decoder;
 	  code = ENCODE_CHAR (this_charset, c);
 	  if (code != CHARSET_INVALID_CODE (this_charset)
 	      && (code_offset < 0 || code >= code_offset))
 	    {
 	      code -= code_offset;
-	      if (CODE_POINT_TO_INDEX (charset, code) >= 0)
+	      if (code >= charset->min_code && code <= charset->max_code
+		  && CODE_POINT_TO_INDEX (charset, code) >= 0)
 		return code;
 	    }
 	}
 return CHARSET_INVALID_CODE (charset);
 }
 CHECK_CHARSET_GET_ID (charset, id);
 if (CONSP (code_point))
 {
 CHECK_NATNUM (XCAR (code_point));
 CHECK_NATNUM (XCDR (code_point));
-code = (XINT (XCAR (code_point)) << 16) | (XINT (XCAR (code_point)));
+code = (XINT (XCAR (code_point)) << 16) | (XINT (XCDR (code_point)));
 }
 else
 {
 CHECK_NATNUM (code_point);
 code = XINT (code_point);

Mercurial > emacs

comparison src/charset.c @ 88475:9d54c1d6cd0a