1 /* utf8.h 2 * 3 * Copyright (c) 1998-2001, Larry Wall 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 10 START_EXTERN_C 11 12 #ifdef DOINIT 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). */ 23 }; 24 #else 25 EXTCONST unsigned char PL_utf8skip[]; 26 #endif 27 28 END_EXTERN_C 29 30 #define UTF8_MAXLEN 13 /* how wide can a single UTF8 encoded character become */ 31 32 /* #define IN_UTF8 (PL_curcop->op_private & HINT_UTF8) */ 33 #define IN_BYTE (PL_curcop->op_private & HINT_BYTE) 34 #define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTE) 35 36 #define UTF8_ALLOW_EMPTY 0x0001 37 #define UTF8_ALLOW_CONTINUATION 0x0002 38 #define UTF8_ALLOW_NON_CONTINUATION 0x0004 39 #define UTF8_ALLOW_FE_FF 0x0008 40 #define UTF8_ALLOW_SHORT 0x0010 41 #define UTF8_ALLOW_SURROGATE 0x0020 42 #define UTF8_ALLOW_BOM 0x0040 43 #define UTF8_ALLOW_FFFF 0x0080 44 #define UTF8_ALLOW_LONG 0x0100 45 #define UTF8_ALLOW_ANYUV (UTF8_ALLOW_EMPTY|UTF8_ALLOW_FE_FF|\ 46 UTF8_ALLOW_SURROGATE|UTF8_ALLOW_BOM|\ 47 UTF8_ALLOW_FFFF|UTF8_ALLOW_LONG) 48 #define UTF8_ALLOW_ANY 0x00ff 49 #define UTF8_CHECK_ONLY 0x0100 50 51 #define UNICODE_SURROGATE_FIRST 0xd800 52 #define UNICODE_SURROGATE_LAST 0xdfff 53 #define UNICODE_REPLACEMENT 0xfffd 54 #define UNICODE_BYTER_ORDER_MARK 0xfffe 55 #define UNICODE_ILLEGAL 0xffff 56 57 #define UNICODE_IS_SURROGATE(c) ((c) >= UNICODE_SURROGATE_FIRST && \ 58 (c) <= UNICODE_SURROGATE_LAST) 59 #define UNICODE_IS_REPLACEMENT(c) ((c) == UNICODE_REPLACMENT) 60 #define UNICODE_IS_BYTE_ORDER_MARK(c) ((c) == UNICODE_BYTER_ORDER_MARK) 61 #define UNICODE_IS_ILLEGAL(c) ((c) == UNICODE_ILLEGAL) 62 63 #define UTF8SKIP(s) PL_utf8skip[*(U8*)s] 64 65 #define UTF8_QUAD_MAX UINT64_C(0x1000000000) 66 67 /* 68 69 The following table is from Unicode 3.1. 70 71 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte 72 73 U+0000..U+007F 00..7F��� 74 U+0080..U+07FF C2..DF 80..BF��� 75 U+0800..U+0FFF E0 A0..BF 80..BF�� 76 U+1000..U+FFFF E1..EF 80..BF 80..BF�� 77 U+10000..U+3FFFF F0 90..BF 80..BF 80..BF 78 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF 79 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF 80 81 */ 82 83 #define UTF8_IS_ASCII(c) (((U8)c) < 0x80) 84 #define UTF8_IS_START(c) (((U8)c) >= 0xc0 && (((U8)c) <= 0xfd)) 85 #define UTF8_IS_CONTINUATION(c) (((U8)c) >= 0x80 && (((U8)c) <= 0xbf)) 86 #define UTF8_IS_CONTINUED(c) (((U8)c) & 0x80) 87 #define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)c & 0xfc) != 0xc0) 88 89 #define UTF8_CONTINUATION_MASK ((U8)0x3f) 90 #define UTF8_ACCUMULATION_SHIFT 6 91 #define UTF8_ACCUMULATE(old, new) (((old) << UTF8_ACCUMULATION_SHIFT) | (((U8)new) & UTF8_CONTINUATION_MASK)) 92 93 #define UTF8_EIGHT_BIT_HI(c) ( (((U8)(c))>>6) |0xc0) 94 #define UTF8_EIGHT_BIT_LO(c) (((((U8)(c)) )&0x3f)|0x80) 95 96 #ifdef HAS_QUAD 97 #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \ 98 (uv) < 0x800 ? 2 : \ 99 (uv) < 0x10000 ? 3 : \ 100 (uv) < 0x200000 ? 4 : \ 101 (uv) < 0x4000000 ? 5 : \ 102 (uv) < 0x80000000 ? 6 : \ 103 (uv) < UTF8_QUAD_MAX ? 7 : 13 ) 104 #else 105 /* No, I'm not even going to *TRY* putting #ifdef inside a #define */ 106 #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \ 107 (uv) < 0x800 ? 2 : \ 108 (uv) < 0x10000 ? 3 : \ 109 (uv) < 0x200000 ? 4 : \ 110 (uv) < 0x4000000 ? 5 : \ 111 (uv) < 0x80000000 ? 6 : 7 ) 112 #endif 113 114 115 /* 116 * Note: we try to be careful never to call the isXXX_utf8() functions 117 * unless we're pretty sure we've seen the beginning of a UTF-8 character 118 * (that is, the two high bits are set). Otherwise we risk loading in the 119 * heavy-duty SWASHINIT and SWASHGET routines unnecessarily. 120 */ 121 #ifdef EBCDIC 122 #define isIDFIRST_lazy_if(p,c) isIDFIRST(*(p)) 123 #define isALNUM_lazy_if(p,c) isALNUM(*(p)) 124 #else 125 #define isIDFIRST_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \ 126 ? isIDFIRST(*(p)) \ 127 : isIDFIRST_utf8((U8*)p)) 128 #define isALNUM_lazy_if(p,c) ((IN_BYTE || (!c || (*((U8*)p) < 0xc0))) \ 129 ? isALNUM(*(p)) \ 130 : isALNUM_utf8((U8*)p)) 131 #endif 132 #define isIDFIRST_lazy(p) isIDFIRST_lazy_if(p,1) 133 #define isALNUM_lazy(p) isALNUM_lazy_if(p,1) 134