[p5sagit/p5-mst-13.2.git] / utf8.h

/*    utf8.h
 *
 *    Copyright (c) 1998-2001, Larry Wall
 *
 *    You may distribute under the terms of either the GNU General Public
 *    License or the Artistic License, as specified in the README file.
 *
 */

#ifdef EBCDIC
/* The equivalent of these macros but implementing UTF-EBCDIC
   are in the following header file:
 */

#include "utfebcdic.h"
#else
START_EXTERN_C

#ifdef DOINIT
EXTCONST unsigned char PL_utf8skip[] = {
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* scripts */
3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6,	 /* cjk etc. */
7,13, /* Perl extended (not UTF-8).  Up to 72bit allowed (64-bit + reserved). */
};
#else
EXTCONST unsigned char PL_utf8skip[];
#endif

END_EXTERN_C
#define UTF8SKIP(s) PL_utf8skip[*(U8*)s]

/* Native character to iso-8859-1 */
#define NATIVE_TO_ASCII(ch)      (ch)
#define ASCII_TO_NATIVE(ch)      (ch)
/* Transform after encoding */
#define NATIVE_TO_UTF(ch)        (ch)
#define UTF_TO_NATIVE(ch)        (ch)
/* Transforms in wide UV chars */
#define UNI_TO_NATIVE(ch)        (ch)
#define NATIVE_TO_UNI(ch)        (ch)
/* Transforms in invariant space */
#define NATIVE_TO_NEED(enc,ch)   (ch)
#define ASCII_TO_NEED(enc,ch)    (ch)

/*

 The following table is from Unicode 3.1.

 Code Points		1st Byte  2nd Byte  3rd Byte  4th Byte

   U+0000..U+007F	00..7F   
   U+0080..U+07FF	C2..DF    80..BF   
   U+0800..U+0FFF	E0        A0..BF    80..BF  
   U+1000..U+FFFF	E1..EF    80..BF    80..BF  
  U+10000..U+3FFFF	F0        90..BF    80..BF    80..BF
  U+40000..U+FFFFF	F1..F3    80..BF    80..BF    80..BF
 U+100000..U+10FFFF	F4        80..8F    80..BF    80..BF

 */

#define UTF8_IS_INVARIANT(c) 		(((UV)c) <  0x80)
#define UTF8_IS_START(c)		(((U8)c) >= 0xc0 && (((U8)c) <= 0xfd))
#define UTF8_IS_CONTINUATION(c)		(((U8)c) >= 0x80 && (((U8)c) <= 0xbf))
#define UTF8_IS_CONTINUED(c) 		(((U8)c) &  0x80)
#define UTF8_IS_DOWNGRADEABLE_START(c)	(((U8)c & 0xfc) == 0xc0)

#define UTF_START_MARK(len) ((len >  7) ? 0xFF : (0xFE << (7-len)))
#define UTF_START_MASK(len) ((len >= 7) ? 0x00 : (0x1F >> (len-2)))

#define UTF_CONTINUATION_MARK		0x80
#define UTF_ACCUMULATION_SHIFT		6
#define UTF_CONTINUATION_MASK		((U8)0x3f)
#define UTF8_ACCUMULATE(old, new)	((old) << UTF_ACCUMULATION_SHIFT | (((U8)new) & UTF_CONTINUATION_MASK))

#define UTF8_EIGHT_BIT_HI(c)	((((U8)(c))>>UTF_ACCUMULATION_SHIFT)|UTF_START_MARK(2))
#define UTF8_EIGHT_BIT_LO(c)	(((((U8)(c)))&UTF_CONTINUATION_MASK)|UTF_CONTINUATION_MARK)

#ifdef HAS_QUAD
#define UNISKIP(uv) ( (uv) < 0x80           ? 1 : \
		      (uv) < 0x800          ? 2 : \
		      (uv) < 0x10000        ? 3 : \
		      (uv) < 0x200000       ? 4 : \
		      (uv) < 0x4000000      ? 5 : \
		      (uv) < 0x80000000     ? 6 : \
                      (uv) < UTF8_QUAD_MAX ? 7 : 13 )
#else
/* No, I'm not even going to *TRY* putting #ifdef inside a #define */
#define UNISKIP(uv) ( (uv) < 0x80           ? 1 : \
		      (uv) < 0x800          ? 2 : \
		      (uv) < 0x10000        ? 3 : \
		      (uv) < 0x200000       ? 4 : \
		      (uv) < 0x4000000      ? 5 : \
		      (uv) < 0x80000000     ? 6 : 7 )
#endif

/*
 * Note: we try to be careful never to call the isXXX_utf8() functions
 * unless we're pretty sure we've seen the beginning of a UTF-8 character
 * (that is, the two high bits are set).  Otherwise we risk loading in the
 * heavy-duty SWASHINIT and SWASHGET routines unnecessarily.
 */
#define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
				? isIDFIRST(*(p)) \
				: isIDFIRST_utf8((U8*)p))
#define isALNUM_lazy_if(p,c)   ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
				? isALNUM(*(p)) \
				: isALNUM_utf8((U8*)p))


#endif /* EBCDIC vs ASCII */

/* Rest of these are attributes of Unicode and perl's internals rather than the encoding */

#define isIDFIRST_lazy(p)	isIDFIRST_lazy_if(p,1)
#define isALNUM_lazy(p)		isALNUM_lazy_if(p,1)

#define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */

/* #define IN_UTF8 (PL_curcop->op_private & HINT_UTF8) */
#define IN_BYTE (PL_curcop->op_private & HINT_BYTE)
#define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTE)

#define UTF8_ALLOW_EMPTY		0x0001
#define UTF8_ALLOW_CONTINUATION		0x0002
#define UTF8_ALLOW_NON_CONTINUATION	0x0004
#define UTF8_ALLOW_FE_FF		0x0008
#define UTF8_ALLOW_SHORT		0x0010
#define UTF8_ALLOW_SURROGATE		0x0020
#define UTF8_ALLOW_BOM			0x0040
#define UTF8_ALLOW_FFFF			0x0080
#define UTF8_ALLOW_LONG			0x0100
#define UTF8_ALLOW_ANYUV		(UTF8_ALLOW_EMPTY|UTF8_ALLOW_FE_FF|\
					 UTF8_ALLOW_SURROGATE|UTF8_ALLOW_BOM|\
					 UTF8_ALLOW_FFFF|UTF8_ALLOW_LONG)
#define UTF8_ALLOW_ANY			0x00ff
#define UTF8_CHECK_ONLY			0x0100

#define UNICODE_SURROGATE_FIRST		0xd800
#define UNICODE_SURROGATE_LAST		0xdfff
#define UNICODE_REPLACEMENT		0xfffd
#define UNICODE_BYTER_ORDER_MARK	0xfffe
#define UNICODE_ILLEGAL			0xffff

#define UNICODE_IS_SURROGATE(c)		((c) >= UNICODE_SURROGATE_FIRST && \
					 (c) <= UNICODE_SURROGATE_LAST)
#define UNICODE_IS_REPLACEMENT(c)	((c) == UNICODE_REPLACMENT)
#define UNICODE_IS_BYTE_ORDER_MARK(c)	((c) == UNICODE_BYTER_ORDER_MARK)
#define UNICODE_IS_ILLEGAL(c)		((c) == UNICODE_ILLEGAL)

#define UTF8_QUAD_MAX	UINT64_C(0x1000000000)

#define UTF8_IS_ASCII(c) UTF8_IS_INVARIANT(c)
Commit	Line	Data
a0ed51b3	1	/* utf8.h
a0ed51b3	2	*
bc89e66f	3	* Copyright (c) 1998-2001, Larry Wall
a0ed51b3	4	*
	5	* You may distribute under the terms of either the GNU General Public
	6	* License or the Artistic License, as specified in the README file.
	7	*
	8	*/
	9
1d72bdf6	10	#ifdef EBCDIC
	11	/* The equivalent of these macros but implementing UTF-EBCDIC
	12	are in the following header file:
	13	*/
	14
	15	#include "utfebcdic.h"
	16	#else
73c4f7a1	17	START_EXTERN_C
73c4f7a1	18
a0ed51b3	19	#ifdef DOINIT
6f06b55f	20	EXTCONST unsigned char PL_utf8skip[] = {
a0ed51b3	21	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
	22	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
	23	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
	24	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
	25	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
	26	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
	27	2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* scripts */
3c77ea2b	28	3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6, /* cjk etc. */
3c77ea2b	29	7,13, /* Perl extended (not UTF-8). Up to 72bit allowed (64-bit + reserved). */
a0ed51b3	30	};
a0ed51b3	31	#else
6f06b55f	32	EXTCONST unsigned char PL_utf8skip[];
a0ed51b3	33	#endif
a0ed51b3	34
73c4f7a1	35	END_EXTERN_C
6f06b55f	36	#define UTF8SKIP(s) PL_utf8skip[(U8)s]
7e2040f0	37
1d72bdf6	38	/* Native character to iso-8859-1 */
	39	#define NATIVE_TO_ASCII(ch) (ch)
	40	#define ASCII_TO_NATIVE(ch) (ch)
	41	/* Transform after encoding */
	42	#define NATIVE_TO_UTF(ch) (ch)
	43	#define UTF_TO_NATIVE(ch) (ch)
	44	/* Transforms in wide UV chars */
	45	#define UNI_TO_NATIVE(ch) (ch)
	46	#define NATIVE_TO_UNI(ch) (ch)
	47	/* Transforms in invariant space */
	48	#define NATIVE_TO_NEED(enc,ch) (ch)
	49	#define ASCII_TO_NEED(enc,ch) (ch)
d7578b48	50
877d9f0d	51	/*
9041c2e3	52
877d9f0d	53	The following table is from Unicode 3.1.
	54
	55	Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
	56
	57	U+0000..U+007F 00..7F
	58	U+0080..U+07FF C2..DF 80..BF
	59	U+0800..U+0FFF E0 A0..BF 80..BF
	60	U+1000..U+FFFF E1..EF 80..BF 80..BF
	61	U+10000..U+3FFFF F0 90..BF 80..BF 80..BF
	62	U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
	63	U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
	64
	65	*/
	66
1d72bdf6	67	#define UTF8_IS_INVARIANT(c) (((UV)c) < 0x80)
c512ce4f	68	#define UTF8_IS_START(c) (((U8)c) >= 0xc0 && (((U8)c) <= 0xfd))
	69	#define UTF8_IS_CONTINUATION(c) (((U8)c) >= 0x80 && (((U8)c) <= 0xbf))
	70	#define UTF8_IS_CONTINUED(c) (((U8)c) & 0x80)
db42d148	71	#define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)c & 0xfc) == 0xc0)
8850bf83	72
1d72bdf6	73	#define UTF_START_MARK(len) ((len > 7) ? 0xFF : (0xFE << (7-len)))
	74	#define UTF_START_MASK(len) ((len >= 7) ? 0x00 : (0x1F >> (len-2)))
	75
	76	#define UTF_CONTINUATION_MARK 0x80
	77	#define UTF_ACCUMULATION_SHIFT 6
	78	#define UTF_CONTINUATION_MASK ((U8)0x3f)
	79	#define UTF8_ACCUMULATE(old, new) ((old) << UTF_ACCUMULATION_SHIFT \| (((U8)new) & UTF_CONTINUATION_MASK))
421a8bf2	80
1d72bdf6	81	#define UTF8_EIGHT_BIT_HI(c) ((((U8)(c))>>UTF_ACCUMULATION_SHIFT)\|UTF_START_MARK(2))
1d72bdf6	82	#define UTF8_EIGHT_BIT_LO(c) (((((U8)(c)))&UTF_CONTINUATION_MASK)\|UTF_CONTINUATION_MARK)
c512ce4f	83
1d68d6cd	84	#ifdef HAS_QUAD
5bbb0b5a	85	#define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
1d68d6cd	86	(uv) < 0x800 ? 2 : \
	87	(uv) < 0x10000 ? 3 : \
	88	(uv) < 0x200000 ? 4 : \
	89	(uv) < 0x4000000 ? 5 : \
	90	(uv) < 0x80000000 ? 6 : \
9041c2e3	91	(uv) < UTF8_QUAD_MAX ? 7 : 13 )
1d68d6cd	92	#else
1d68d6cd	93	/* No, I'm not even going to TRY putting #ifdef inside a #define */
5bbb0b5a	94	#define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
1d68d6cd	95	(uv) < 0x800 ? 2 : \
	96	(uv) < 0x10000 ? 3 : \
	97	(uv) < 0x200000 ? 4 : \
	98	(uv) < 0x4000000 ? 5 : \
	99	(uv) < 0x80000000 ? 6 : 7 )
	100	#endif
	101
7e2040f0	102	/*
	103	* Note: we try to be careful never to call the isXXX_utf8() functions
	104	* unless we're pretty sure we've seen the beginning of a UTF-8 character
	105	* (that is, the two high bits are set). Otherwise we risk loading in the
	106	* heavy-duty SWASHINIT and SWASHGET routines unnecessarily.
	107	*/
5b154c98	108	#define isIDFIRST_lazy_if(p,c) ((IN_BYTE \|\| (!c \|\| (((U8)p) < 0xc0))) \
7e2040f0	109	? isIDFIRST(*(p)) \
7e2040f0	110	: isIDFIRST_utf8((U8*)p))
5b154c98	111	#define isALNUM_lazy_if(p,c) ((IN_BYTE \|\| (!c \|\| (((U8)p) < 0xc0))) \
7e2040f0	112	? isALNUM(*(p)) \
7e2040f0	113	: isALNUM_utf8((U8*)p))
1d72bdf6	114
	115
	116	#endif /* EBCDIC vs ASCII */
	117
	118	/* Rest of these are attributes of Unicode and perl's internals rather than the encoding */
	119
7e2040f0	120	#define isIDFIRST_lazy(p) isIDFIRST_lazy_if(p,1)
7e2040f0	121	#define isALNUM_lazy(p) isALNUM_lazy_if(p,1)
3bd709b1	122
1d72bdf6	123	#define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */
3bd709b1	124
1d72bdf6	125	/* #define IN_UTF8 (PL_curcop->op_private & HINT_UTF8) */
	126	#define IN_BYTE (PL_curcop->op_private & HINT_BYTE)
	127	#define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTE)
	128
	129	#define UTF8_ALLOW_EMPTY 0x0001
	130	#define UTF8_ALLOW_CONTINUATION 0x0002
	131	#define UTF8_ALLOW_NON_CONTINUATION 0x0004
	132	#define UTF8_ALLOW_FE_FF 0x0008
	133	#define UTF8_ALLOW_SHORT 0x0010
	134	#define UTF8_ALLOW_SURROGATE 0x0020
	135	#define UTF8_ALLOW_BOM 0x0040
	136	#define UTF8_ALLOW_FFFF 0x0080
	137	#define UTF8_ALLOW_LONG 0x0100
	138	#define UTF8_ALLOW_ANYUV (UTF8_ALLOW_EMPTY\|UTF8_ALLOW_FE_FF\|\
	139	UTF8_ALLOW_SURROGATE\|UTF8_ALLOW_BOM\|\
	140	UTF8_ALLOW_FFFF\|UTF8_ALLOW_LONG)
	141	#define UTF8_ALLOW_ANY 0x00ff
	142	#define UTF8_CHECK_ONLY 0x0100
	143
	144	#define UNICODE_SURROGATE_FIRST 0xd800
	145	#define UNICODE_SURROGATE_LAST 0xdfff
	146	#define UNICODE_REPLACEMENT 0xfffd
	147	#define UNICODE_BYTER_ORDER_MARK 0xfffe
	148	#define UNICODE_ILLEGAL 0xffff
	149
	150	#define UNICODE_IS_SURROGATE(c) ((c) >= UNICODE_SURROGATE_FIRST && \
	151	(c) <= UNICODE_SURROGATE_LAST)
	152	#define UNICODE_IS_REPLACEMENT(c) ((c) == UNICODE_REPLACMENT)
	153	#define UNICODE_IS_BYTE_ORDER_MARK(c) ((c) == UNICODE_BYTER_ORDER_MARK)
	154	#define UNICODE_IS_ILLEGAL(c) ((c) == UNICODE_ILLEGAL)
	155
	156	#define UTF8_QUAD_MAX UINT64_C(0x1000000000)
3bd709b1	157
1d72bdf6	158	#define UTF8_IS_ASCII(c) UTF8_IS_INVARIANT(c)