3 * Copyright (c) 1998-2001, Larry Wall
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.
11 /* The equivalent of these macros but implementing UTF-EBCDIC
12 are in the following header file:
15 #include "utfebcdic.h"
20 EXTCONST unsigned char PL_utf8skip[] = {
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 */
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. */
29 7,13, /* Perl extended (not UTF-8). Up to 72bit allowed (64-bit + reserved). */
32 EXTCONST unsigned char PL_utf8skip[];
36 #define UTF8SKIP(s) PL_utf8skip[*(U8*)s]
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)
53 The following table is from Unicode 3.1.
55 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
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
67 #define UTF8_IS_INVARIANT(c) (((UV)c) < 0x80)
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)
71 #define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)c & 0xfc) == 0xc0)
73 #define UTF_START_MARK(len) ((len > 7) ? 0xFF : (0xFE << (7-len)))
74 #define UTF_START_MASK(len) ((len >= 7) ? 0x00 : (0x1F >> (len-2)))
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))
81 #define UTF8_EIGHT_BIT_HI(c) ((((U8)(c))>>UTF_ACCUMULATION_SHIFT)|UTF_START_MARK(2))
82 #define UTF8_EIGHT_BIT_LO(c) (((((U8)(c)))&UTF_CONTINUATION_MASK)|UTF_CONTINUATION_MARK)
85 #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
87 (uv) < 0x10000 ? 3 : \
88 (uv) < 0x200000 ? 4 : \
89 (uv) < 0x4000000 ? 5 : \
90 (uv) < 0x80000000 ? 6 : \
91 (uv) < UTF8_QUAD_MAX ? 7 : 13 )
93 /* No, I'm not even going to *TRY* putting #ifdef inside a #define */
94 #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
96 (uv) < 0x10000 ? 3 : \
97 (uv) < 0x200000 ? 4 : \
98 (uv) < 0x4000000 ? 5 : \
99 (uv) < 0x80000000 ? 6 : 7 )
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.
108 #define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
110 : isIDFIRST_utf8((U8*)p))
111 #define isALNUM_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
113 : isALNUM_utf8((U8*)p))
116 #endif /* EBCDIC vs ASCII */
118 /* Rest of these are attributes of Unicode and perl's internals rather than the encoding */
120 #define isIDFIRST_lazy(p) isIDFIRST_lazy_if(p,1)
121 #define isALNUM_lazy(p) isALNUM_lazy_if(p,1)
123 #define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */
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)
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
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
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)
156 #define UTF8_QUAD_MAX UINT64_C(0x1000000000)
158 #define UTF8_IS_ASCII(c) UTF8_IS_INVARIANT(c)