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.
13 EXTCONST unsigned char PL_utf8skip[] = {
14 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 */
15 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 */
16 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 */
17 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 */
18 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 */
19 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 */
20 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 */
21 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. */
22 7,13, /* Perl extended (not UTF-8). Up to 72bit allowed (64-bit + reserved). */
25 EXTCONST unsigned char PL_utf8skip[];
30 #define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */
32 /* #define IN_UTF8 (PL_curcop->op_private & HINT_UTF8) */
33 #define IN_BYTE (PL_curcop->op_private & HINT_BYTE)
34 #ifdef USE_BYTES_DOWNGRADES
35 #define DO_UTF8(sv) (SvUTF8(sv) && !(IN_BYTE && sv_utf8_downgrade(sv,0)))
37 #define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTE)
40 #define UTF8_ALLOW_EMPTY 0x0001
41 #define UTF8_ALLOW_CONTINUATION 0x0002
42 #define UTF8_ALLOW_NON_CONTINUATION 0x0004
43 #define UTF8_ALLOW_FE_FF 0x0008
44 #define UTF8_ALLOW_SHORT 0x0010
45 #define UTF8_ALLOW_SURROGATE 0x0020
46 #define UTF8_ALLOW_BOM 0x0040
47 #define UTF8_ALLOW_FFFF 0x0080
48 #define UTF8_ALLOW_LONG 0x0100
49 #define UTF8_ALLOW_ANYUV (UTF8_ALLOW_EMPTY|UTF8_ALLOW_FE_FF|\
50 UTF8_ALLOW_SURROGATE|UTF8_ALLOW_BOM|\
51 UTF8_ALLOW_FFFF|UTF8_ALLOW_LONG)
52 #define UTF8_ALLOW_ANY 0x00ff
53 #define UTF8_CHECK_ONLY 0x0100
55 #define UNICODE_SURROGATE_FIRST 0xd800
56 #define UNICODE_SURROGATE_LAST 0xdfff
57 #define UNICODE_REPLACEMENT 0xfffd
58 #define UNICODE_BYTER_ORDER_MARK 0xfffe
59 #define UNICODE_ILLEGAL 0xffff
61 #define UNICODE_IS_SURROGATE(c) ((c) >= UNICODE_SURROGATE_FIRST && \
62 (c) <= UNICODE_SURROGATE_LAST)
63 #define UNICODE_IS_REPLACEMENT(c) ((c) == UNICODE_REPLACMENT)
64 #define UNICODE_IS_BYTE_ORDER_MARK(c) ((c) == UNICODE_BYTER_ORDER_MARK)
65 #define UNICODE_IS_ILLEGAL(c) ((c) == UNICODE_ILLEGAL)
67 #define UTF8SKIP(s) PL_utf8skip[*(U8*)s]
69 #define UTF8_QUAD_MAX UINT64_C(0x1000000000)
73 The following table is from Unicode 3.1.
75 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
78 U+0080..U+07FF C2..DF 80..BF
79 U+0800..U+0FFF E0 A0..BF 80..BF
80 U+1000..U+FFFF E1..EF 80..BF 80..BF
81 U+10000..U+3FFFF F0 90..BF 80..BF 80..BF
82 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
83 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
87 #define UTF8_IS_ASCII(c) (((U8)c) < 0x80)
88 #define UTF8_IS_START(c) (((U8)c) >= 0xc0 && (((U8)c) <= 0xfd))
89 #define UTF8_IS_CONTINUATION(c) (((U8)c) >= 0x80 && (((U8)c) <= 0xbf))
90 #define UTF8_IS_CONTINUED(c) (((U8)c) & 0x80)
91 #define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)c & 0xfc) != 0xc0)
93 #define UTF8_CONTINUATION_MASK ((U8)0x3f)
94 #define UTF8_ACCUMULATION_SHIFT 6
95 #define UTF8_ACCUMULATE(old, new) ((old) << UTF8_ACCUMULATION_SHIFT | (((U8)new) & UTF8_CONTINUATION_MASK))
97 #define UTF8_EIGHT_BIT_HI(c) ( (((U8)(c))>>6) |0xc0)
98 #define UTF8_EIGHT_BIT_LO(c) (((((U8)(c)) )&0x3f)|0x80)
101 #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
103 (uv) < 0x10000 ? 3 : \
104 (uv) < 0x200000 ? 4 : \
105 (uv) < 0x4000000 ? 5 : \
106 (uv) < 0x80000000 ? 6 : \
107 (uv) < UTF8_QUAD_MAX ? 7 : 13 )
109 /* No, I'm not even going to *TRY* putting #ifdef inside a #define */
110 #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
112 (uv) < 0x10000 ? 3 : \
113 (uv) < 0x200000 ? 4 : \
114 (uv) < 0x4000000 ? 5 : \
115 (uv) < 0x80000000 ? 6 : 7 )
120 * Note: we try to be careful never to call the isXXX_utf8() functions
121 * unless we're pretty sure we've seen the beginning of a UTF-8 character
122 * (that is, the two high bits are set). Otherwise we risk loading in the
123 * heavy-duty SWASHINIT and SWASHGET routines unnecessarily.
126 #define isIDFIRST_lazy_if(p,c) isIDFIRST(*(p))
127 #define isALNUM_lazy_if(p,c) isALNUM(*(p))
129 #define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
131 : isIDFIRST_utf8((U8*)p))
132 #define isALNUM_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \
134 : isALNUM_utf8((U8*)p))
136 #define isIDFIRST_lazy(p) isIDFIRST_lazy_if(p,1)
137 #define isALNUM_lazy(p) isALNUM_lazy_if(p,1)
139 /* EBCDIC-happy ways of converting native code to UTF8 */
142 #define NATIVE_TO_ASCII(ch) PL_e2a[(ch)]
143 #define ASCII_TO_NATIVE(ch) PL_a2e[(ch)]
144 #define UNI_TO_NATIVE(ch) (((ch) > 0x100) ? (ch) : (UV) PL_a2e[(ch)])
145 #define NATIVE_TO_UNI(ch) (((ch) > 0x100) ? (ch) : (UV) PL_e2a[(ch)])
147 #define NATIVE_TO_ASCII(ch) (ch)
148 #define ASCII_TO_NATIVE(ch) (ch)
149 #define UNI_TO_NATIVE(ch) (ch)
150 #define NATIVE_TO_UNI(ch) (ch)