X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/b36bf33f6564c3e9a9ff131f4f3c9980b7a8af15..3da9985538c5eae2dd2de1309b320bc77f757687:/utf8.h?ds=sidebyside diff --git a/utf8.h b/utf8.h index e558bb6..11b5aca 100644 --- a/utf8.h +++ b/utf8.h @@ -20,8 +20,14 @@ /* For to_utf8_fold_flags, q.v. */ #define FOLD_FLAGS_LOCALE 0x1 #define FOLD_FLAGS_FULL 0x2 +#define FOLD_FLAGS_NOMIX_ASCII 0x4 -#define to_uni_fold(c, p, lenp) _to_uni_fold_flags(c, p, lenp, 1) +/* For _core_swash_init(), internal core use only */ +#define _CORE_SWASH_INIT_USER_DEFINED_PROPERTY 0x1 +#define _CORE_SWASH_INIT_RETURN_IF_UNDEF 0x2 +#define _CORE_SWASH_INIT_ACCEPT_INVLIST 0x4 + +#define to_uni_fold(c, p, lenp) _to_uni_fold_flags(c, p, lenp, FOLD_FLAGS_FULL) #define to_utf8_fold(c, p, lenp) _to_utf8_fold_flags(c, p, lenp, \ FOLD_FLAGS_FULL, NULL) #define to_utf8_lower(a,b,c) _to_utf8_lower_flags(a,b,c,0, NULL) @@ -61,16 +67,25 @@ 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 official UTF-8). Up to 72bit allowed (64-bit + - reserved). */ +/* 0x00 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x10 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x20 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x30 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x40 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x50 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x60 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x70 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */ +/* 0x80 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */ +/* 0x90 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */ +/* 0xA0 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */ +/* 0xB0 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */ +/* 0xC0 */ 2,2, /* overlong */ +/* 0xC2 */ 2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* U+0080 to U+03FF */ +/* 0xD0 */ 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* U+0400 to U+07FF */ +/* 0xE0 */ 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3, /* U+0800 to U+FFFF */ +/* 0xF0 */ 4,4,4,4,4,4,4,4,5,5,5,5,6,6, /* above BMP to 2**31 - 1 */ +/* 0xFE */ 7,13, /* Perl extended (never was official UTF-8). Up to 72bit + allowed (64-bit + reserved). */ }; #else EXTCONST unsigned char PL_utf8skip[]; @@ -93,6 +108,7 @@ END_EXTERN_C /* As there are no translations, avoid the function wrapper */ #define utf8n_to_uvchr utf8n_to_uvuni +#define valid_utf8_to_uvchr valid_utf8_to_uvuni #define uvchr_to_utf8 uvuni_to_utf8 /* @@ -111,6 +127,10 @@ END_EXTERN_C 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 + Below are non-Unicode code points + U+110000..U+13FFFF F4 90..BF 80..BF 80..BF + U+110000..U+1FFFFF F5..F7 80..BF 80..BF 80..BF + U+200000: F8.. * 88..BF 80..BF 80..BF 80..BF Note the gaps before several of the byte entries above marked by '*'. These are caused by legal UTF-8 avoiding non-shortest encodings: it is technically @@ -123,12 +143,12 @@ explicitly forbidden, and the shortest possible encoding should always be used /* Another way to look at it, as bits: - Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte + Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte - 0aaaaaaa 0aaaaaaa - 00000bbbbbaaaaaa 110bbbbb 10aaaaaa - ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa - 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa + 0aaa aaaa 0aaa aaaa + 0000 0bbb bbaa aaaa 110b bbbb 10aa aaaa + cccc bbbb bbaa aaaa 1110 cccc 10bb bbbb 10aa aaaa + 00 000d ddcc cccc bbbb bbaa aaaa 1111 0ddd 10cc cccc 10bb bbbb 10aa aaaa As you can see, the continuation bytes all begin with C<10>, and the leading bits of the start byte tell how many bytes there are in the @@ -139,20 +159,31 @@ Perl's extended UTF-8 means we can have start bytes up to FF. */ #define UNI_IS_INVARIANT(c) (((UV)c) < 0x80) -/* Note that C0 and C1 are invalid in legal UTF8, so the lower bound of the - * below might ought to be C2 */ -#define UTF8_IS_START(c) (((U8)c) >= 0xc0) +#define UTF8_IS_START(c) (((U8)c) >= 0xc2) #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) + +/* Masking with 0xfe allows low bit to be 0 or 1; thus this matches 0xc[23] */ +#define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)c & 0xfe) == 0xc2) #define UTF_START_MARK(len) (((len) > 7) ? 0xFF : (0xFE << (7-(len)))) + +/* Masks out the initial one bits in a start byte, leaving the real data ones. + * Doesn't work on an invariant byte */ #define UTF_START_MASK(len) (((len) >= 7) ? 0x00 : (0x1F >> ((len)-2))) #define UTF_CONTINUATION_MARK 0x80 #define UTF_ACCUMULATION_SHIFT 6 + +/* 2**UTF_ACCUMULATION_SHIFT - 1 */ #define UTF_CONTINUATION_MASK ((U8)0x3f) +/* This sets the UTF_CONTINUATION_MASK in the upper bits of a word. If a value + * is anded with it, and the result is non-zero, then using the original value + * in UTF8_ACCUMULATE will overflow, shifting bits off the left */ +#define UTF_ACCUMULATION_OVERFLOW_MASK \ + (((UV) UTF_CONTINUATION_MASK) << ((sizeof(UV) * CHARBITS) - UTF_ACCUMULATION_SHIFT)) + #ifdef HAS_QUAD #define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \ (uv) < 0x800 ? 2 : \ @@ -320,10 +351,21 @@ Perl's extended UTF-8 means we can have start bytes up to FF. * U+E000: \xEE\x80\x80 \xF1\xB8\xA0\xA0 next after surrogates */ #ifdef EBCDIC /* Both versions assume well-formed UTF8 */ -# define UTF8_IS_SURROGATE(s) (*(s) == UTF_TO_NATIVE(0xF1) \ - && ((*((s) +1) == UTF_TO_NATIVE(0xB6)) || *((s) + 1) == UTF_TO_NATIVE(0xB7))) +# define UTF8_IS_SURROGATE(s) (*(s) == UTF_TO_NATIVE(0xF1) \ + && ((*((s) +1) == UTF_TO_NATIVE(0xB6)) \ + || *((s) + 1) == UTF_TO_NATIVE(0xB7))) + /* points to one beyond the end of the string that starts at */ +# define UTF8_IS_REPLACEMENT(s, send) (*(s) == UTF_TO_NATIVE(0xEF) \ + && (send - s) >= 4 \ + && *((s) + 1) == UTF_TO_NATIVE(0xBF) \ + && *((s) + 2) == UTF_TO_NATIVE(0xBF) \ + && *((s) + 3) == UTF_TO_NATIVE(0xBD) #else # define UTF8_IS_SURROGATE(s) (*(s) == 0xED && *((s) + 1) >= 0xA0) +# define UTF8_IS_REPLACEMENT(s, send) (*(s) == 0xEF \ + && (send - s) >= 3 \ + && *((s) + 1) == 0xBF \ + && *((s) + 2) == 0xBD) #endif /* ASCII EBCDIC I8 @@ -446,6 +488,9 @@ Perl's extended UTF-8 means we can have start bytes up to FF. #define GREEK_CAPITAL_LETTER_MU 0x039C /* Upper and title case of MICRON */ #define LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS 0x0178 /* Also is title case */ #define LATIN_CAPITAL_LETTER_SHARP_S 0x1E9E +#define LATIN_SMALL_LETTER_LONG_S 0x017F +#define KELVIN_SIGN 0x212A +#define ANGSTROM_SIGN 0x212B #define UNI_DISPLAY_ISPRINT 0x0001 #define UNI_DISPLAY_BACKSLASH 0x0002 @@ -470,6 +515,8 @@ Perl's extended UTF-8 means we can have start bytes up to FF. #define SHARP_S_SKIP 2 #ifndef EBCDIC +/* If you want to exclude surrogates, and beyond legal Unicode, see the blame + * log for earlier versions which gave details for these */ # define IS_UTF8_CHAR_1(p) \ ((p)[0] <= 0x7F) # define IS_UTF8_CHAR_2(p) \ @@ -480,18 +527,7 @@ Perl's extended UTF-8 means we can have start bytes up to FF. (p)[1] >= 0xA0 && (p)[1] <= 0xBF && \ (p)[2] >= 0x80 && (p)[2] <= 0xBF) # define IS_UTF8_CHAR_3b(p) \ - ((p)[0] >= 0xE1 && (p)[0] <= 0xEC && \ - (p)[1] >= 0x80 && (p)[1] <= 0xBF && \ - (p)[2] >= 0x80 && (p)[2] <= 0xBF) -# define IS_UTF8_CHAR_3c(p) \ - ((p)[0] == 0xED && \ - (p)[1] >= 0x80 && (p)[1] <= 0xBF && \ - (p)[2] >= 0x80 && (p)[2] <= 0xBF) - /* In IS_UTF8_CHAR_3c(p) one could use - * (p)[1] >= 0x80 && (p)[1] <= 0x9F - * if one wanted to exclude surrogates. */ -# define IS_UTF8_CHAR_3d(p) \ - ((p)[0] >= 0xEE && (p)[0] <= 0xEF && \ + ((p)[0] >= 0xE1 && (p)[0] <= 0xEF && \ (p)[1] >= 0x80 && (p)[1] <= 0xBF && \ (p)[2] >= 0x80 && (p)[2] <= 0xBF) # define IS_UTF8_CHAR_4a(p) \ @@ -499,34 +535,23 @@ Perl's extended UTF-8 means we can have start bytes up to FF. (p)[1] >= 0x90 && (p)[1] <= 0xBF && \ (p)[2] >= 0x80 && (p)[2] <= 0xBF && \ (p)[3] >= 0x80 && (p)[3] <= 0xBF) -# define IS_UTF8_CHAR_4b(p) \ - ((p)[0] >= 0xF1 && (p)[0] <= 0xF3 && \ - (p)[1] >= 0x80 && (p)[1] <= 0xBF && \ - (p)[2] >= 0x80 && (p)[2] <= 0xBF && \ - (p)[3] >= 0x80 && (p)[3] <= 0xBF) -/* In IS_UTF8_CHAR_4c(p) one could use - * (p)[0] == 0xF4 - * if one wanted to stop at the Unicode limit U+10FFFF. - * The 0xF7 allows us to go to 0x1fffff (0x200000 would +/* The 0xF7 allows us to go to 0x1fffff (0x200000 would * require five bytes). Not doing any further code points * since that is not needed (and that would not be strict * UTF-8, anyway). The "slow path" in Perl_is_utf8_char() * will take care of the "extended UTF-8". */ -# define IS_UTF8_CHAR_4c(p) \ - ((p)[0] >= 0xF4 && (p)[0] <= 0xF7 && \ +# define IS_UTF8_CHAR_4b(p) \ + ((p)[0] >= 0xF1 && (p)[0] <= 0xF7 && \ (p)[1] >= 0x80 && (p)[1] <= 0xBF && \ (p)[2] >= 0x80 && (p)[2] <= 0xBF && \ (p)[3] >= 0x80 && (p)[3] <= 0xBF) # define IS_UTF8_CHAR_3(p) \ (IS_UTF8_CHAR_3a(p) || \ - IS_UTF8_CHAR_3b(p) || \ - IS_UTF8_CHAR_3c(p) || \ - IS_UTF8_CHAR_3d(p)) + IS_UTF8_CHAR_3b(p)) # define IS_UTF8_CHAR_4(p) \ (IS_UTF8_CHAR_4a(p) || \ - IS_UTF8_CHAR_4b(p) || \ - IS_UTF8_CHAR_4c(p)) + IS_UTF8_CHAR_4b(p)) /* IS_UTF8_CHAR(p) is strictly speaking wrong (not UTF-8) because it * (1) allows UTF-8 encoded UTF-16 surrogates @@ -599,8 +624,8 @@ Perl's extended UTF-8 means we can have start bytes up to FF. * Local variables: * c-indentation-style: bsd * c-basic-offset: 4 - * indent-tabs-mode: t + * indent-tabs-mode: nil * End: * - * ex: set ts=8 sts=4 sw=4 noet: + * ex: set ts=8 sts=4 sw=4 et: */