/* Rewritten by Ben Wing <ben@xemacs.org>. */
-#ifndef _XEMACS_MULE_CHARSET_H
-#define _XEMACS_MULE_CHARSET_H
+#ifndef INCLUDED_mule_charset_h_
+#define INCLUDED_mule_charset_h_
/*
1. Character Sets
Character set Encoding (PC == position-code)
------------- -------- (LB == leading-byte)
- ASCII PC1 |
+ ASCII PC1 |
Control-1 LB | PC1 + 0xA0
Dimension-1 official LB | PC1 + 0x80
Dimension-1 private 0x9E | LB | PC1 + 0x80
/* Definition of leading bytes */
/************************************************************************/
+typedef unsigned char Charset_ID;
+
#define MIN_LEADING_BYTE 0x80
/* These need special treatment in a string and/or character */
#define LEADING_BYTE_ASCII 0x8E /* Omitted in a buffer */
/* Is this a prefix for a private leading byte? */
-INLINE int LEADING_BYTE_PREFIX_P (unsigned char lb);
-INLINE int
-LEADING_BYTE_PREFIX_P (unsigned char lb)
+INLINE_HEADER int LEADING_BYTE_PREFIX_P (Bufbyte lb);
+INLINE_HEADER int
+LEADING_BYTE_PREFIX_P (Bufbyte lb)
{
return (lb == PRE_LEADING_BYTE_PRIVATE_1 ||
lb == PRE_LEADING_BYTE_PRIVATE_2);
}
/* Given a private leading byte, return the leading byte prefix stored
- in a string */
+ in a string. */
#define PRIVATE_LEADING_BYTE_PREFIX(lb) \
- ((lb) < MIN_LEADING_BYTE_PRIVATE_2 ? \
+ ((unsigned int) (lb) < MIN_LEADING_BYTE_PRIVATE_2 ? \
PRE_LEADING_BYTE_PRIVATE_1 : \
PRE_LEADING_BYTE_PRIVATE_2)
/* of any format */
/************************************************************************/
-/* Argument `c' should be (unsigned int) or (unsigned char). */
-/* Note that SP and DEL are not included. */
+/* These are carefully designed to work if BYTE is signed or unsigned. */
+/* Note that SPC and DEL are considered ASCII, not control. */
-#define BYTE_ASCII_P(c) ((c) < 0x80)
-#define BYTE_C0_P(c) ((c) < 0x20)
-/* Do some forced casting just to make *sure* things are gotten right. */
-#define BYTE_C1_P(c) ((unsigned int) ((unsigned int) (c) - 0x80) < 0x20)
+#define BYTE_ASCII_P(byte) (((byte) & ~0x7f) == 0)
+#define BYTE_C0_P(byte) (((byte) & ~0x1f) == 0)
+#define BYTE_C1_P(byte) (((byte) & ~0x1f) == 0x80)
\f
/************************************************************************/
/* in a Mule-formatted string */
/************************************************************************/
-/* Does this byte represent the first byte of a character? */
+/* Does BYTE represent the first byte of a character? */
-#define BUFBYTE_FIRST_BYTE_P(c) ((c) < 0xA0)
+#define BUFBYTE_FIRST_BYTE_P(byte) ((byte) < 0xA0)
-/* Does this byte represent the first byte of a multi-byte character? */
+/* Does BYTE represent the first byte of a multi-byte character? */
-#define BUFBYTE_LEADING_BYTE_P(c) BYTE_C1_P (c)
+#define BUFBYTE_LEADING_BYTE_P(byte) BYTE_C1_P (byte)
\f
/************************************************************************/
int id;
Lisp_Object name;
- Lisp_Object doc_string, registry, short_name, long_name;
+ Lisp_Object doc_string;
+ Lisp_Object registry;
+ Lisp_Object short_name;
+ Lisp_Object long_name;
Lisp_Object reverse_direction_charset;
/* Which half of font to be used to display this character set */
unsigned int graphic;
};
+typedef struct Lisp_Charset Lisp_Charset;
-DECLARE_LRECORD (charset, struct Lisp_Charset);
-#define XCHARSET(x) XRECORD (x, charset, struct Lisp_Charset)
+DECLARE_LRECORD (charset, Lisp_Charset);
+#define XCHARSET(x) XRECORD (x, charset, Lisp_Charset)
#define XSETCHARSET(x, p) XSETRECORD (x, p, charset)
#define CHARSETP(x) RECORDP (x, charset)
#define CHECK_CHARSET(x) CHECK_RECORD (x, charset)
CHARSET_REVERSE_DIRECTION_CHARSET (XCHARSET (cs))
struct charset_lookup {
- /* Table of charsets indexed by leading byte. */
- Lisp_Object charset_by_leading_byte[128];
-
+ /* Table of charsets indexed by (leading byte - MIN_LEADING_BYTE). */
+ Lisp_Object charset_by_leading_byte[NUM_LEADING_BYTES];
+
/* Table of charsets indexed by type/final-byte/direction. */
Lisp_Object charset_by_attributes[4][128][2];
-};
-
-extern struct charset_lookup *chlook;
-
-/* Table of number of bytes in the string representation of a character
- indexed by the first byte of that representation.
- This value can be derived other ways -- e.g. something like
-
- (BYTE_ASCII_P (first_byte) ? 1 :
- XCHARSET_REP_BYTES (CHARSET_BY_LEADING_BYTE (first_byte)))
+ Charset_ID next_allocated_1_byte_leading_byte;
+ Charset_ID next_allocated_2_byte_leading_byte;
+};
- but it's faster this way. */
-extern Bytecount rep_bytes_by_first_byte[0xA0];
+INLINE_HEADER Lisp_Object CHARSET_BY_LEADING_BYTE (Bufbyte lb);
+INLINE_HEADER Lisp_Object
+CHARSET_BY_LEADING_BYTE (Bufbyte lb)
+{
+ extern struct charset_lookup *chlook;
#ifdef ERROR_CHECK_TYPECHECK
-/* int not Bufbyte even though that is the actual type of a leading byte.
- This way, out-ot-range values will get caught rather than automatically
- truncated. */
-INLINE Lisp_Object CHARSET_BY_LEADING_BYTE (int lb);
-INLINE Lisp_Object
-CHARSET_BY_LEADING_BYTE (int lb)
-{
- assert (lb >= 0x80 && lb <= 0xFF);
- return chlook->charset_by_leading_byte[lb - 128];
+ /* When error-checking is on, x86 GCC 2.95.2 -O3 miscompiles the
+ following unless we introduce `tem'. */
+ int tem = lb;
+ type_checking_assert (tem >= MIN_LEADING_BYTE &&
+ tem < (MIN_LEADING_BYTE + NUM_LEADING_BYTES));
+#endif
+ return chlook->charset_by_leading_byte[lb - MIN_LEADING_BYTE];
}
-#else
-
-#define CHARSET_BY_LEADING_BYTE(lb) (chlook->charset_by_leading_byte[(lb) - 128])
+INLINE_HEADER Lisp_Object
+CHARSET_BY_ATTRIBUTES (unsigned int type, unsigned char final, int dir);
+INLINE_HEADER Lisp_Object
+CHARSET_BY_ATTRIBUTES (unsigned int type, unsigned char final, int dir)
+{
+ extern struct charset_lookup *chlook;
-#endif
+ type_checking_assert (type < countof (chlook->charset_by_attributes) &&
+ final < countof (chlook->charset_by_attributes[0]) &&
+ dir < countof (chlook->charset_by_attributes[0][0]));
+ return chlook->charset_by_attributes[type][final][dir];
+}
-#define CHARSET_BY_ATTRIBUTES(type, final, dir) \
- (chlook->charset_by_attributes[type][final][dir])
+/* Table of number of bytes in the string representation of a character
+ indexed by the first byte of that representation.
-#ifdef ERROR_CHECK_TYPECHECK
+ This value can be derived in other ways -- e.g. something like
+ XCHARSET_REP_BYTES (CHARSET_BY_LEADING_BYTE (first_byte))
+ but it's faster this way. */
+extern const Bytecount rep_bytes_by_first_byte[0xA0];
-/* Number of bytes in the string representation of a character */
-INLINE int REP_BYTES_BY_FIRST_BYTE (int fb);
-INLINE int
-REP_BYTES_BY_FIRST_BYTE (int fb)
+/* Number of bytes in the string representation of a character. */
+INLINE_HEADER int REP_BYTES_BY_FIRST_BYTE (Bufbyte fb);
+INLINE_HEADER int
+REP_BYTES_BY_FIRST_BYTE (Bufbyte fb)
{
- assert (fb >= 0 && fb < 0xA0);
+ type_checking_assert (fb < 0xA0);
return rep_bytes_by_first_byte[fb];
}
-#else
-#define REP_BYTES_BY_FIRST_BYTE(fb) (rep_bytes_by_first_byte[fb])
-#endif
-
\f
/************************************************************************/
/* Dealing with characters */
/************************************************************************/
-/* Is this character represented by more than one byte in a string? */
-
-#define CHAR_MULTIBYTE_P(c) ((c) >= 0x80)
-
-#define CHAR_ASCII_P(c) (!CHAR_MULTIBYTE_P (c))
+#define CHAR_ASCII_P(ch) ((ch) <= 0x7F)
/* The bit fields of character are divided into 3 parts:
FIELD1(5bits):FIELD2(7bits):FIELD3(7bits) */
FIELD2_TO_PRIVATE_LEADING_BYTE are the same.
*/
-INLINE Bufbyte CHAR_LEADING_BYTE (Emchar c);
-INLINE Bufbyte
+INLINE_HEADER Bufbyte CHAR_LEADING_BYTE (Emchar c);
+INLINE_HEADER Bufbyte
CHAR_LEADING_BYTE (Emchar c)
{
if (CHAR_ASCII_P (c))
#ifdef ENABLE_COMPOSITE_CHARS
return LEADING_BYTE_COMPOSITE;
#else
- abort();
+ ABORT();
return 0;
#endif /* ENABLE_COMPOSITE_CHARS */
}
FIELD2_TO_PRIVATE_LEADING_BYTE are the same.
*/
-INLINE Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2);
-INLINE Emchar
+INLINE_HEADER Emchar MAKE_CHAR (Lisp_Object charset, int c1, int c2);
+INLINE_HEADER Emchar
MAKE_CHAR (Lisp_Object charset, int c1, int c2)
{
if (EQ (charset, Vcharset_ascii))
: ((c1) = CHAR_FIELD2 (c), \
(c2) = CHAR_FIELD3 (c))
-INLINE void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2);
-INLINE void
+INLINE_HEADER void breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2);
+INLINE_HEADER void
breakup_char_1 (Emchar c, Lisp_Object *charset, int *c1, int *c2)
{
*charset = CHAR_CHARSET (c);
int Lstream_fput_emchar (Lstream *stream, Emchar ch);
void Lstream_funget_emchar (Lstream *stream, Emchar ch);
-int copy_internal_to_external (CONST Bufbyte *internal, Bytecount len,
+int copy_internal_to_external (const Bufbyte *internal, Bytecount len,
unsigned char *external);
-Bytecount copy_external_to_internal (CONST unsigned char *external,
+Bytecount copy_external_to_internal (const unsigned char *external,
int len, Bufbyte *internal);
-#endif /* _XEMACS_MULE_CHARSET_H */
+#endif /* INCLUDED_mule_charset_h_ */