/* utf8.h
*
+ * This file contains definitions for use with the UTF-8 encoding. It
+ * actually also works with the variant UTF-8 encoding called UTF-EBCDIC, and
+ * hides almost all of the differences between these from the caller. In other
+ * words, someone should #include this file, and if the code is being compiled
+ * on an EBCDIC platform, things should mostly just work.
+ *
* Copyright (C) 2000, 2001, 2002, 2005, 2006, 2007, 2009,
* 2010, 2011 by Larry Wall and others
*
*
*/
+#ifndef PERL_UTF8_H_ /* Guard against recursive inclusion */
+#define PERL_UTF8_H_ 1
+
/* Use UTF-8 as the default script encoding?
* Turning this on will break scripts having non-UTF-8 binary
* data (such as Latin-1) in string literals. */
# define USE_UTF8_IN_NAMES (PL_hints & HINT_UTF8)
#endif
+#include "regcharclass.h"
+#include "unicode_constants.h"
+
/* For to_utf8_fold_flags, q.v. */
-#define FOLD_FLAGS_LOCALE 0x1
-#define FOLD_FLAGS_FULL 0x2
+#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)
-#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)
-#define to_utf8_upper(a,b,c) _to_utf8_upper_flags(a,b,c,0, NULL)
-#define to_utf8_title(a,b,c) _to_utf8_title_flags(a,b,c,0, NULL)
+/* 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
-/* Source backward compatibility. */
-#define uvuni_to_utf8(d, uv) uvuni_to_utf8_flags(d, uv, 0)
-#define is_utf8_string_loc(s, len, ep) is_utf8_string_loclen(s, len, ep, 0)
+/*
+=head1 Unicode Support
+L<perlguts/Unicode Support> has an introduction to this API.
-#define foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
- foldEQ_utf8_flags(s1, pe1, l1, u1, s2, pe2, l2, u2, 0)
-#define FOLDEQ_UTF8_NOMIX_ASCII (1 << 0)
-#define FOLDEQ_UTF8_LOCALE (1 << 1)
-#define FOLDEQ_S1_ALREADY_FOLDED (1 << 2)
-#define FOLDEQ_S2_ALREADY_FOLDED (1 << 3)
+See also L</Character classification>,
+and L</Character case changing>.
+Various functions outside this section also work specially with Unicode.
+Search for the string "utf8" in this document.
-/*
-=for apidoc ibcmp_utf8
+=for apidoc is_ascii_string
-This is a synonym for (! foldEQ_utf8())
+This is a misleadingly-named synonym for L</is_utf8_invariant_string>.
+On ASCII-ish platforms, the name isn't misleading: the ASCII-range characters
+are exactly the UTF-8 invariants. But EBCDIC machines have more invariants
+than just the ASCII characters, so C<is_utf8_invariant_string> is preferred.
+
+=for apidoc is_invariant_string
+
+This is a somewhat misleadingly-named synonym for L</is_utf8_invariant_string>.
+C<is_utf8_invariant_string> is preferred, as it indicates under what conditions
+the string is invariant.
=cut
*/
+#define is_ascii_string(s, len) is_utf8_invariant_string(s, len)
+#define is_invariant_string(s, len) is_utf8_invariant_string(s, len)
+
+#define uvchr_to_utf8(a,b) uvchr_to_utf8_flags(a,b,0)
+#define uvchr_to_utf8_flags(d,uv,flags) \
+ uvoffuni_to_utf8_flags(d,NATIVE_TO_UNI(uv),flags)
+#define utf8_to_uvchr_buf(s, e, lenp) \
+ utf8n_to_uvchr(s, (U8*)(e) - (U8*)(s), lenp, \
+ ckWARN_d(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY)
+#define utf8n_to_uvchr(s, len, lenp, flags) \
+ utf8n_to_uvchr_error(s, len, lenp, flags, 0)
+
+#define to_uni_fold(c, p, lenp) _to_uni_fold_flags(c, p, lenp, FOLD_FLAGS_FULL)
+
+#define to_utf8_fold(s, r, lenr) \
+ _to_utf8_fold_flags (s, NULL, r, lenr, FOLD_FLAGS_FULL, __FILE__, __LINE__)
+#define to_utf8_lower(s, r, lenr) \
+ _to_utf8_lower_flags(s, NULL, r ,lenr, 0, __FILE__, __LINE__)
+#define to_utf8_upper(s, r, lenr) \
+ _to_utf8_upper_flags(s, NULL, r, lenr, 0, __FILE__, __LINE__)
+#define to_utf8_title(s, r, lenr) \
+ _to_utf8_title_flags(s, NULL, r, lenr ,0, __FILE__, __LINE__)
+
+#define foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
+ foldEQ_utf8_flags(s1, pe1, l1, u1, s2, pe2, l2, u2, 0)
+#define FOLDEQ_UTF8_NOMIX_ASCII (1 << 0)
+#define FOLDEQ_LOCALE (1 << 1)
+#define FOLDEQ_S1_ALREADY_FOLDED (1 << 2)
+#define FOLDEQ_S2_ALREADY_FOLDED (1 << 3)
+#define FOLDEQ_S1_FOLDS_SANE (1 << 4)
+#define FOLDEQ_S2_FOLDS_SANE (1 << 5)
+
#define ibcmp_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
cBOOL(! foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2))
#else /* ! EBCDIC */
START_EXTERN_C
+/* How wide can a single UTF-8 encoded character become in bytes. */
+/* NOTE: Strictly speaking Perl's UTF-8 should not be called UTF-8 since UTF-8
+ * is an encoding of Unicode, and Unicode's upper limit, 0x10FFFF, can be
+ * expressed with 4 bytes. However, Perl thinks of UTF-8 as a way to encode
+ * non-negative integers in a binary format, even those above Unicode */
+#define UTF8_MAXBYTES 13
+
#ifdef DOINIT
EXTCONST unsigned char PL_utf8skip[] = {
/* 0x00 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
/* 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 */
+/* 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). */
+ /* Perl extended (never was official UTF-8). Up to 36 bit */
+/* 0xFE */ 7,
+ /* More extended, Up to 72 bits (64-bit + reserved) */
+/* 0xFF */ UTF8_MAXBYTES
};
#else
EXTCONST unsigned char PL_utf8skip[];
END_EXTERN_C
-/* 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)
+#if defined(_MSC_VER) && _MSC_VER < 1400
+/* older MSVC versions have a smallish macro buffer */
+#define PERL_SMALL_MACRO_BUFFER
+#endif
-/* As there are no translations, avoid the function wrapper */
-#define utf8n_to_uvchr utf8n_to_uvuni
-#define uvchr_to_utf8 uvuni_to_utf8
+/* Native character to/from iso-8859-1. Are the identity functions on ASCII
+ * platforms */
+#ifdef PERL_SMALL_MACRO_BUFFER
+#define NATIVE_TO_LATIN1(ch) ((U8)(ch))
+#define LATIN1_TO_NATIVE(ch) ((U8)(ch))
+#else
+#define NATIVE_TO_LATIN1(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
+#define LATIN1_TO_NATIVE(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
+#endif
+
+/* I8 is an intermediate version of UTF-8 used only in UTF-EBCDIC. We thus
+ * consider it to be identical to UTF-8 on ASCII platforms. Strictly speaking
+ * UTF-8 and UTF-EBCDIC are two different things, but we often conflate them
+ * because they are 8-bit encodings that serve the same purpose in Perl, and
+ * rarely do we need to distinguish them. The term "NATIVE_UTF8" applies to
+ * whichever one is applicable on the current platform */
+#ifdef PERL_SMALL_MACRO_BUFFER
+#define NATIVE_UTF8_TO_I8(ch) ((U8) (ch))
+#define I8_TO_NATIVE_UTF8(ch) ((U8) (ch))
+#else
+#define NATIVE_UTF8_TO_I8(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
+#define I8_TO_NATIVE_UTF8(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
+#endif
+
+/* Transforms in wide UV chars */
+#define UNI_TO_NATIVE(ch) ((UV) (ch))
+#define NATIVE_TO_UNI(ch) ((UV) (ch))
/*
- The following table is from Unicode 3.2.
+ The following table is from Unicode 3.2, plus the Perl extensions for above
+ U+10FFFF
- Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
+ Code Points 1st Byte 2nd Byte 3rd 4th 5th 6th 7th 8th-13th
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+CFFF E1..EC 80..BF 80..BF
- U+D000..U+D7FF ED 80..9F 80..BF
- U+D800..U+DFFF +++++++ utf16 surrogates, not legal utf8 +++++++
- U+E000..U+FFFF EE..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
- 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
+ U+0800..U+0FFF E0 * A0..BF 80..BF
+ U+1000..U+CFFF E1..EC 80..BF 80..BF
+ U+D000..U+D7FF ED 80..9F 80..BF
+ U+D800..U+DFFF ED A0..BF 80..BF (surrogates)
+ U+E000..U+FFFF EE..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
+ Below are above-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..U+FFFFFF F8 * 88..BF 80..BF 80..BF 80..BF
+U+1000000..U+3FFFFFF F9..FB 80..BF 80..BF 80..BF 80..BF
+U+4000000..U+3FFFFFFF FC * 84..BF 80..BF 80..BF 80..BF 80..BF
+U+40000000..U+7FFFFFFF FD 80..BF 80..BF 80..BF 80..BF 80..BF
+U+80000000..U+FFFFFFFFF FE * 82..BF 80..BF 80..BF 80..BF 80..BF 80..BF
+U+1000000000.. FF 80..BF 80..BF 80..BF 80..BF 80..BF * 81..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
possible to UTF-8-encode a single code point in different ways, but that is
explicitly forbidden, and the shortest possible encoding should always be used
-(and that is what Perl does).
+(and that is what Perl does). The non-shortest ones are called 'overlongs'.
*/
leading bits of the start byte tell how many bytes there are in the
encoded character.
-Perl's extended UTF-8 means we can have start bytes up to FF.
+Perl's extended UTF-8 means we can have start bytes up through FF, though any
+beginning with FF yields a code point that is too large for 32-bit ASCII
+platforms. FF signals to use 13 bytes for the encoded character. This breaks
+the paradigm that the number of leading bits gives how many total bytes there
+are in the character.
*/
-#define UNI_IS_INVARIANT(c) (((UV)c) < 0x80)
-#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)
+/* Is the representation of the Unicode code point 'cp' the same regardless of
+ * being encoded in UTF-8 or not? */
+#define OFFUNI_IS_INVARIANT(cp) isASCII(cp)
-/* 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)
+/*
+=for apidoc Am|bool|UVCHR_IS_INVARIANT|UV cp
-#define UTF_START_MARK(len) (((len) > 7) ? 0xFF : (0xFE << (7-(len))))
+Evaluates to 1 if the representation of code point C<cp> is the same whether or
+not it is encoded in UTF-8; otherwise evaluates to 0. UTF-8 invariant
+characters can be copied as-is when converting to/from UTF-8, saving time.
+C<cp> is Unicode if above 255; otherwise is platform-native.
-/* 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)))
+=cut
+ */
+
+#define UVCHR_IS_INVARIANT(cp) OFFUNI_IS_INVARIANT(cp)
+/* This defines the bits that are to be in the continuation bytes of a multi-byte
+ * UTF-8 encoded character that mark it is a continuation byte. */
#define UTF_CONTINUATION_MARK 0x80
+
+/* Misleadingly named: is the UTF8-encoded byte 'c' part of a variant sequence
+ * in UTF-8? This is the inverse of UTF8_IS_INVARIANT. The |0 makes sure this
+ * isn't mistakenly called with a ptr argument */
+#define UTF8_IS_CONTINUED(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ ((U8)((c) | 0)) & UTF_CONTINUATION_MARK)
+
+/* Is the byte 'c' the first byte of a multi-byte UTF8-8 encoded sequence?
+ * This doesn't catch invariants (they are single-byte). It also excludes the
+ * illegal overlong sequences that begin with C0 and C1. The |0 makes sure
+ * this isn't mistakenly called with a ptr argument */
+#define UTF8_IS_START(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ ((U8)((c) | 0)) >= 0xc2)
+
+/* For use in UTF8_IS_CONTINUATION() below */
+#define UTF_IS_CONTINUATION_MASK 0xC0
+
+/* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
+ * first byte thereof? The |0 makes sure this isn't mistakenly called with a
+ * ptr argument */
+#define UTF8_IS_CONTINUATION(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (((U8)((c) | 0)) & UTF_IS_CONTINUATION_MASK) == UTF_CONTINUATION_MARK)
+
+/* Is the UTF8-encoded byte 'c' the first byte of a two byte sequence? Use
+ * UTF8_IS_NEXT_CHAR_DOWNGRADEABLE() instead if the input isn't known to
+ * be well-formed. Masking with 0xfe allows the low bit to be 0 or 1; thus
+ * this matches 0xc[23]. The |0 makes sure this isn't mistakenly called with a
+ * ptr argument */
+#define UTF8_IS_DOWNGRADEABLE_START(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (((U8)((c) | 0)) & 0xfe) == 0xc2)
+
+/* Is the UTF8-encoded byte 'c' the first byte of a sequence of bytes that
+ * represent a code point > 255? The |0 makes sure this isn't mistakenly
+ * called with a ptr argument */
+#define UTF8_IS_ABOVE_LATIN1(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ ((U8)((c) | 0)) >= 0xc4)
+
+/* This is the number of low-order bits a continuation byte in a UTF-8 encoded
+ * sequence contributes to the specification of the code point. In the bit
+ * maps above, you see that the first 2 bits are a constant '10', leaving 6 of
+ * real information */
#define UTF_ACCUMULATION_SHIFT 6
+/* ^? is defined to be DEL on ASCII systems. See the definition of toCTRL()
+ * for more */
+#define QUESTION_MARK_CTRL DEL_NATIVE
+
+/* Surrogates, non-character code points and above-Unicode code points are
+ * problematic in some contexts. This allows code that needs to check for
+ * those to to quickly exclude the vast majority of code points it will
+ * encounter */
+#define isUTF8_POSSIBLY_PROBLEMATIC(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (U8) c >= 0xED)
+
+/* A helper macro for isUTF8_CHAR, so use that one instead of this. This was
+ * generated by regen/regcharclass.pl, and then moved here. Then it was
+ * hand-edited to add some LIKELY() calls, presuming that malformations are
+ * unlikely. The lines that generated it were then commented out. This was
+ * done because it takes on the order of 10 minutes to generate, and is never
+ * going to change, unless the generated code is improved, and figuring out
+ * the LIKELYs there would be hard.
+ *
+ UTF8_CHAR: Matches legal UTF-8 variant code points up through 0x1FFFFFF
+
+ 0x80 - 0x1FFFFF
+*/
+/*** GENERATED CODE ***/
+#define is_UTF8_CHAR_utf8_no_length_checks(s) \
+( ( 0xC2 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xDF ) ? \
+ ( LIKELY( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
+: ( 0xE0 == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xE1 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xEF ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xF0 == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( 0x90 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0xBF ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+: ( ( ( ( 0xF1 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xF7 ) && LIKELY( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) ) && LIKELY( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) && LIKELY( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )
+
+/* The above macro handles UTF-8 that has this start byte as the maximum */
+#define _IS_UTF8_CHAR_HIGHEST_START_BYTE 0xF7
+
+/* A helper macro for isSTRICT_UTF8_CHAR, so use that one instead of this.
+ * Like is_UTF8_CHAR_utf8_no_length_checks(), this was moved here and LIKELYs
+ * added manually.
+ *
+ STRICT_UTF8_CHAR: Matches legal Unicode UTF-8 variant code points, no
+ surrrogates nor non-character code points
+*/
+/*** GENERATED CODE ***/
+#define is_STRICT_UTF8_CHAR_utf8_no_length_checks(s) \
+( ( 0xC2 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xDF ) ? \
+ ( LIKELY( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
+: ( 0xE0 == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( ( 0xE1 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xEC ) || 0xEE == ((const U8*)s)[0] ) ?\
+ ( ( ( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xED == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xE0 ) == 0x80 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xEF == ((const U8*)s)[0] ) ? \
+ ( ( ( 0x80 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0xB6 ) || ( 0xB8 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0xBE ) ) ?\
+ ( LIKELY( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ? 3 : 0 ) \
+ : ( 0xB7 == ((const U8*)s)[1] ) ? \
+ ( LIKELY( ( ((const U8*)s)[2] & 0xF0 ) == 0x80 || ( ((const U8*)s)[2] & 0xF0 ) == 0xB0 ) ? 3 : 0 )\
+ : ( ( 0xBF == ((const U8*)s)[1] ) && ( 0x80 <= ((const U8*)s)[2] && ((const U8*)s)[2] <= 0xBD ) ) ? 3 : 0 )\
+: ( 0xF0 == ((const U8*)s)[0] ) ? \
+ ( ( ( 0x90 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0x9E ) || ( 0xA0 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0xAE ) || ( 0xB0 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0xBE ) ) ?\
+ ( LIKELY( ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+ : ( ((const U8*)s)[1] == 0x9F || ( ( ((const U8*)s)[1] & 0xEF ) == 0xAF ) ) ? \
+ ( ( 0x80 <= ((const U8*)s)[2] && ((const U8*)s)[2] <= 0xBE ) ? \
+ ( LIKELY( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ? 4 : 0 ) \
+ : LIKELY( ( 0xBF == ((const U8*)s)[2] ) && ( 0x80 <= ((const U8*)s)[3] && ((const U8*)s)[3] <= 0xBD ) ) ? 4 : 0 )\
+ : 0 ) \
+: ( 0xF1 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xF3 ) ? \
+ ( ( ( ( ((const U8*)s)[1] & 0xC8 ) == 0x80 ) || ( ( ((const U8*)s)[1] & 0xCC ) == 0x88 ) || ( ( ((const U8*)s)[1] & 0xCE ) == 0x8C ) || ( ( ((const U8*)s)[1] & 0xCF ) == 0x8E ) ) ?\
+ ( LIKELY( ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+ : ( ( ((const U8*)s)[1] & 0xCF ) == 0x8F ) ? \
+ ( ( 0x80 <= ((const U8*)s)[2] && ((const U8*)s)[2] <= 0xBE ) ? \
+ ( LIKELY( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ? 4 : 0 ) \
+ : LIKELY( ( 0xBF == ((const U8*)s)[2] ) && ( 0x80 <= ((const U8*)s)[3] && ((const U8*)s)[3] <= 0xBD ) ) ? 4 : 0 )\
+ : 0 ) \
+: ( 0xF4 == ((const U8*)s)[0] ) ? \
+ ( ( 0x80 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0x8E ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+ : ( 0x8F == ((const U8*)s)[1] ) ? \
+ ( ( 0x80 <= ((const U8*)s)[2] && ((const U8*)s)[2] <= 0xBE ) ? \
+ ( LIKELY( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ? 4 : 0 ) \
+ : LIKELY( ( 0xBF == ((const U8*)s)[2] ) && ( 0x80 <= ((const U8*)s)[3] && ((const U8*)s)[3] <= 0xBD ) ) ? 4 : 0 )\
+ : 0 ) \
+: 0 )
+
+/* Similarly,
+ C9_STRICT_UTF8_CHAR: Matches legal Unicode UTF-8 variant code
+ points, no surrogates
+ 0x0080 - 0xD7FF
+ 0xE000 - 0x10FFFF
+*/
+/*** GENERATED CODE ***/
+#define is_C9_STRICT_UTF8_CHAR_utf8_no_length_checks(s) \
+( ( 0xC2 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xDF ) ? \
+ ( LIKELY( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
+: ( 0xE0 == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( ( 0xE1 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xEC ) || ( ((const U8*)s)[0] & 0xFE ) == 0xEE ) ?\
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xED == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( ((const U8*)s)[1] & 0xE0 ) == 0x80 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xF0 == ((const U8*)s)[0] ) ? \
+ ( LIKELY( ( ( 0x90 <= ((const U8*)s)[1] && ((const U8*)s)[1] <= 0xBF ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+: ( 0xF1 <= ((const U8*)s)[0] && ((const U8*)s)[0] <= 0xF3 ) ? \
+ ( LIKELY( ( ( ( ((const U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+: LIKELY( ( ( ( 0xF4 == ((const U8*)s)[0] ) && ( ( ((const U8*)s)[1] & 0xF0 ) == 0x80 ) ) && ( ( ((const U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((const U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )
+
+#define UNICODE_IS_PERL_EXTENDED(uv) UNLIKELY((UV) (uv) > 0x7FFFFFFF)
+
+#endif /* EBCDIC vs ASCII */
+
/* 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 : \
- (uv) < 0x10000 ? 3 : \
- (uv) < 0x200000 ? 4 : \
- (uv) < 0x4000000 ? 5 : \
- (uv) < 0x80000000 ? 6 : \
- (uv) < UTF8_QUAD_MAX ? 7 : 13 )
+#define UTF_CONTINUATION_MASK ((U8) ((1U << UTF_ACCUMULATION_SHIFT) - 1))
+
+/* Internal macro to be used only in this file to aid in constructing other
+ * publicly accessible macros.
+ * The number of bytes required to express this uv in UTF-8, for just those
+ * uv's requiring 2 through 6 bytes, as these are common to all platforms and
+ * word sizes. The number of bytes needed is given by the number of leading 1
+ * bits in the start byte. There are 32 start bytes that have 2 initial 1 bits
+ * (C0-DF); there are 16 that have 3 initial 1 bits (E0-EF); 8 that have 4
+ * initial 1 bits (F0-F8); 4 that have 5 initial 1 bits (F9-FB), and 2 that
+ * have 6 initial 1 bits (FC-FD). The largest number a string of n bytes can
+ * represent is (the number of possible start bytes for 'n')
+ * * (the number of possiblities for each start byte
+ * The latter in turn is
+ * 2 ** ( (how many continuation bytes there are)
+ * * (the number of bits of information each
+ * continuation byte holds))
+ *
+ * If we were on a platform where we could use a fast find first set bit
+ * instruction (or count leading zeros instruction) this could be replaced by
+ * using that to find the log2 of the uv, and divide that by the number of bits
+ * of information in each continuation byte, adjusting for large cases and how
+ * much information is in a start byte for that length */
+#define __COMMON_UNI_SKIP(uv) \
+ (UV) (uv) < (32 * (1U << ( UTF_ACCUMULATION_SHIFT))) ? 2 : \
+ (UV) (uv) < (16 * (1U << (2 * UTF_ACCUMULATION_SHIFT))) ? 3 : \
+ (UV) (uv) < ( 8 * (1U << (3 * UTF_ACCUMULATION_SHIFT))) ? 4 : \
+ (UV) (uv) < ( 4 * (1U << (4 * UTF_ACCUMULATION_SHIFT))) ? 5 : \
+ (UV) (uv) < ( 2 * (1U << (5 * UTF_ACCUMULATION_SHIFT))) ? 6 :
+
+/* Internal macro to be used only in this file.
+ * This adds to __COMMON_UNI_SKIP the details at this platform's upper range.
+ * For any-sized EBCDIC platforms, or 64-bit ASCII ones, we need one more test
+ * to see if just 7 bytes is needed, or if the maximum is needed. For 32-bit
+ * ASCII platforms, everything is representable by 7 bytes */
+#if defined(UV_IS_QUAD) || defined(EBCDIC)
+# define __BASE_UNI_SKIP(uv) (__COMMON_UNI_SKIP(uv) \
+ (UV) (uv) < ((UV) 1U << (6 * UTF_ACCUMULATION_SHIFT)) ? 7 : UTF8_MAXBYTES)
#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 )
+# define __BASE_UNI_SKIP(uv) (__COMMON_UNI_SKIP(uv) 7)
#endif
-#endif /* EBCDIC vs ASCII */
+/* The next two macros use the base macro defined above, and add in the tests
+ * at the low-end of the range, for just 1 byte, yielding complete macros,
+ * publicly accessible. */
+
+/* Input is a true Unicode (not-native) code point */
+#define OFFUNISKIP(uv) (OFFUNI_IS_INVARIANT(uv) ? 1 : __BASE_UNI_SKIP(uv))
+
+/*
+
+=for apidoc Am|STRLEN|UVCHR_SKIP|UV cp
+returns the number of bytes required to represent the code point C<cp> when
+encoded as UTF-8. C<cp> is a native (ASCII or EBCDIC) code point if less than
+255; a Unicode code point otherwise.
+
+=cut
+ */
+#define UVCHR_SKIP(uv) ( UVCHR_IS_INVARIANT(uv) ? 1 : __BASE_UNI_SKIP(uv))
+
+/* The largest code point representable by two UTF-8 bytes on this platform.
+ * As explained in the comments for __COMMON_UNI_SKIP, 32 start bytes with
+ * UTF_ACCUMULATION_SHIFT bits of information each */
+#define MAX_UTF8_TWO_BYTE (32 * (1U << UTF_ACCUMULATION_SHIFT) - 1)
+
+/* The largest code point representable by two UTF-8 bytes on any platform that
+ * Perl runs on. This value is constrained by EBCDIC which has 5 bits per
+ * continuation byte */
+#define MAX_PORTABLE_UTF8_TWO_BYTE (32 * (1U << 5) - 1)
+
+/* The maximum number of UTF-8 bytes a single Unicode character can
+ * uppercase/lowercase/fold into. Unicode guarantees that the maximum
+ * expansion is UTF8_MAX_FOLD_CHAR_EXPAND characters, but any above-Unicode
+ * code point will fold to itself, so we only have to look at the expansion of
+ * the maximum Unicode code point. But this number may be less than the space
+ * occupied by a very large code point under Perl's extended UTF-8. We have to
+ * make it large enough to fit any single character. (It turns out that ASCII
+ * and EBCDIC differ in which is larger) */
+#define UTF8_MAXBYTES_CASE \
+ (UTF8_MAXBYTES >= (UTF8_MAX_FOLD_CHAR_EXPAND * OFFUNISKIP(0x10FFFF)) \
+ ? UTF8_MAXBYTES \
+ : (UTF8_MAX_FOLD_CHAR_EXPAND * OFFUNISKIP(0x10FFFF)))
/* Rest of these are attributes of Unicode and perl's internals rather than the
* encoding, or happen to be the same in both ASCII and EBCDIC (at least at
* this level; the macros that some of these call may have different
* definitions in the two encodings */
-#define NATIVE8_TO_UNI(ch) NATIVE_TO_ASCII(ch) /* a clearer synonym */
+/* In domain restricted to ASCII, these may make more sense to the reader than
+ * the ones with Latin1 in the name */
+#define NATIVE_TO_ASCII(ch) NATIVE_TO_LATIN1(ch)
+#define ASCII_TO_NATIVE(ch) LATIN1_TO_NATIVE(ch)
+
+/* More or less misleadingly-named defines, retained for back compat */
+#define NATIVE_TO_UTF(ch) NATIVE_UTF8_TO_I8(ch)
+#define NATIVE_TO_I8(ch) NATIVE_UTF8_TO_I8(ch)
+#define UTF_TO_NATIVE(ch) I8_TO_NATIVE_UTF8(ch)
+#define I8_TO_NATIVE(ch) I8_TO_NATIVE_UTF8(ch)
+#define NATIVE8_TO_UNI(ch) NATIVE_TO_LATIN1(ch)
-#define UTF8_ACCUMULATE(old, new) (((old) << UTF_ACCUMULATION_SHIFT) | (((U8)new) & UTF_CONTINUATION_MASK))
+/* This defines the 1-bits that are to be in the first byte of a multi-byte
+ * UTF-8 encoded character that mark it as a start byte and give the number of
+ * bytes that comprise the character. 'len' is the number of bytes in the
+ * multi-byte sequence. */
+#define UTF_START_MARK(len) (((len) > 7) ? 0xFF : (0xFF & (0xFE << (7-(len)))))
-/* Convert a two (not one) byte utf8 character to a unicode code point value.
+/* Masks out the initial one bits in a start byte, leaving the real data ones.
+ * Doesn't work on an invariant byte. 'len' is the number of bytes in the
+ * multi-byte sequence that comprises the character. */
+#define UTF_START_MASK(len) (((len) >= 7) ? 0x00 : (0x1F >> ((len)-2)))
+
+/* Adds a UTF8 continuation byte 'new' of information to a running total code
+ * point 'old' of all the continuation bytes so far. This is designed to be
+ * used in a loop to convert from UTF-8 to the code point represented. Note
+ * that this is asymmetric on EBCDIC platforms, in that the 'new' parameter is
+ * the UTF-EBCDIC byte, whereas the 'old' parameter is a Unicode (not EBCDIC)
+ * code point in process of being generated */
+#define UTF8_ACCUMULATE(old, new) (__ASSERT_(FITS_IN_8_BITS(new)) \
+ ((old) << UTF_ACCUMULATION_SHIFT) \
+ | ((NATIVE_UTF8_TO_I8((U8)new)) \
+ & UTF_CONTINUATION_MASK))
+
+/* This works in the face of malformed UTF-8. */
+#define UTF8_IS_NEXT_CHAR_DOWNGRADEABLE(s, e) \
+ ( UTF8_IS_DOWNGRADEABLE_START(*(s)) \
+ && ( (e) - (s) > 1) \
+ && UTF8_IS_CONTINUATION(*((s)+1)))
+
+/* Number of bytes a code point occupies in UTF-8. */
+#define NATIVE_SKIP(uv) UVCHR_SKIP(uv)
+
+/* Most code which says UNISKIP is really thinking in terms of native code
+ * points (0-255) plus all those beyond. This is an imprecise term, but having
+ * it means existing code continues to work. For precision, use UVCHR_SKIP,
+ * NATIVE_SKIP, or OFFUNISKIP */
+#define UNISKIP(uv) UVCHR_SKIP(uv)
+
+/* Longer, but more accurate name */
+#define UTF8_IS_ABOVE_LATIN1_START(c) UTF8_IS_ABOVE_LATIN1(c)
+
+/* Convert a UTF-8 variant Latin1 character to a native code point value.
+ * Needs just one iteration of accumulate. Should be used only if it is known
+ * that the code point is < 256, and is not UTF-8 invariant. Use the slower
+ * but more general TWO_BYTE_UTF8_TO_NATIVE() which handles any code point
+ * representable by two bytes (which turns out to be up through
+ * MAX_PORTABLE_UTF8_TWO_BYTE). The two parameters are:
+ * HI: a downgradable start byte;
+ * LO: continuation.
+ * */
+#define EIGHT_BIT_UTF8_TO_NATIVE(HI, LO) \
+ ( __ASSERT_(UTF8_IS_DOWNGRADEABLE_START(HI)) \
+ __ASSERT_(UTF8_IS_CONTINUATION(LO)) \
+ LATIN1_TO_NATIVE(UTF8_ACCUMULATE(( \
+ NATIVE_UTF8_TO_I8(HI) & UTF_START_MASK(2)), (LO))))
+
+/* Convert a two (not one) byte utf8 character to a native code point value.
* Needs just one iteration of accumulate. Should not be used unless it is
* known that the two bytes are legal: 1) two-byte start, and 2) continuation.
* Note that the result can be larger than 255 if the input character is not
* downgradable */
-#define TWO_BYTE_UTF8_TO_UNI(HI, LO) \
- UTF8_ACCUMULATE((NATIVE_TO_UTF(HI) & UTF_START_MASK(2)), \
- NATIVE_TO_UTF(LO))
+#define TWO_BYTE_UTF8_TO_NATIVE(HI, LO) \
+ (__ASSERT_(FITS_IN_8_BITS(HI)) \
+ __ASSERT_(FITS_IN_8_BITS(LO)) \
+ __ASSERT_(PL_utf8skip[HI] == 2) \
+ __ASSERT_(UTF8_IS_CONTINUATION(LO)) \
+ UNI_TO_NATIVE(UTF8_ACCUMULATE((NATIVE_UTF8_TO_I8(HI) & UTF_START_MASK(2)), \
+ (LO))))
-#define UTF8SKIP(s) PL_utf8skip[*(const U8*)(s)]
+/* Should never be used, and be deprecated */
+#define TWO_BYTE_UTF8_TO_UNI(HI, LO) NATIVE_TO_UNI(TWO_BYTE_UTF8_TO_NATIVE(HI, LO))
-#define UTF8_IS_INVARIANT(c) UNI_IS_INVARIANT(NATIVE_TO_UTF(c))
-#define NATIVE_IS_INVARIANT(c) UNI_IS_INVARIANT(NATIVE8_TO_UNI(c))
+/*
-#define MAX_PORTABLE_UTF8_TWO_BYTE 0x3FF /* constrained by EBCDIC */
+=for apidoc Am|STRLEN|UTF8SKIP|char* s
+returns the number of bytes in the UTF-8 encoded character whose first (perhaps
+only) byte is pointed to by C<s>.
-/* The macros in the next sets are used to generate the two utf8 or utfebcdic
- * bytes from an ordinal that is known to fit into two bytes; it must be less
- * than 0x3FF to work across both encodings. */
-/* Nocast allows these to be used in the case label of a switch statement */
-#define UTF8_TWO_BYTE_HI_nocast(c) UTF_TO_NATIVE(((c) >> UTF_ACCUMULATION_SHIFT) | (0xFF & UTF_START_MARK(2)))
-#define UTF8_TWO_BYTE_LO_nocast(c) UTF_TO_NATIVE(((c) & UTF_CONTINUATION_MASK) | UTF_CONTINUATION_MARK)
+=cut
+ */
+#define UTF8SKIP(s) PL_utf8skip[*(const U8*)(s)]
+#define UTF8_SKIP(s) UTF8SKIP(s)
+
+/* Most code that says 'UNI_' really means the native value for code points up
+ * through 255 */
+#define UNI_IS_INVARIANT(cp) UVCHR_IS_INVARIANT(cp)
+
+/*
+=for apidoc Am|bool|UTF8_IS_INVARIANT|char c
+
+Evaluates to 1 if the byte C<c> represents the same character when encoded in
+UTF-8 as when not; otherwise evaluates to 0. UTF-8 invariant characters can be
+copied as-is when converting to/from UTF-8, saving time.
-#define UTF8_TWO_BYTE_HI(c) ((U8) (UTF8_TWO_BYTE_HI_nocast(c)))
-#define UTF8_TWO_BYTE_LO(c) ((U8) (UTF8_TWO_BYTE_LO_nocast(c)))
+In spite of the name, this macro gives the correct result if the input string
+from which C<c> comes is not encoded in UTF-8.
-/* This name is used when the source is a single byte */
-#define UTF8_EIGHT_BIT_HI(c) UTF8_TWO_BYTE_HI((U8)(c))
-#define UTF8_EIGHT_BIT_LO(c) UTF8_TWO_BYTE_LO((U8)(c))
+See C<L</UVCHR_IS_INVARIANT>> for checking if a UV is invariant.
+
+=cut
+
+The reason it works on both UTF-8 encoded strings and non-UTF-8 encoded, is
+that it returns TRUE in each for the exact same set of bit patterns. It is
+valid on a subset of what UVCHR_IS_INVARIANT is valid on, so can just use that;
+and the compiler should optimize out anything extraneous given the
+implementation of the latter. The |0 makes sure this isn't mistakenly called
+with a ptr argument.
+*/
+#define UTF8_IS_INVARIANT(c) UVCHR_IS_INVARIANT((c) | 0)
+
+/* Like the above, but its name implies a non-UTF8 input, which as the comments
+ * above show, doesn't matter as to its implementation */
+#define NATIVE_BYTE_IS_INVARIANT(c) UVCHR_IS_INVARIANT(c)
+
+/* The macros in the next 4 sets are used to generate the two utf8 or utfebcdic
+ * bytes from an ordinal that is known to fit into exactly two (not one) bytes;
+ * it must be less than 0x3FF to work across both encodings. */
+
+/* These two are helper macros for the other three sets, and should not be used
+ * directly anywhere else. 'translate_function' is either NATIVE_TO_LATIN1
+ * (which works for code points up through 0xFF) or NATIVE_TO_UNI which works
+ * for any code point */
+#define __BASE_TWO_BYTE_HI(c, translate_function) \
+ (__ASSERT_(! UVCHR_IS_INVARIANT(c)) \
+ I8_TO_NATIVE_UTF8((translate_function(c) >> UTF_ACCUMULATION_SHIFT) \
+ | UTF_START_MARK(2)))
+#define __BASE_TWO_BYTE_LO(c, translate_function) \
+ (__ASSERT_(! UVCHR_IS_INVARIANT(c)) \
+ I8_TO_NATIVE_UTF8((translate_function(c) & UTF_CONTINUATION_MASK) \
+ | UTF_CONTINUATION_MARK))
+
+/* The next two macros should not be used. They were designed to be usable as
+ * the case label of a switch statement, but this doesn't work for EBCDIC. Use
+ * regen/unicode_constants.pl instead */
+#define UTF8_TWO_BYTE_HI_nocast(c) __BASE_TWO_BYTE_HI(c, NATIVE_TO_UNI)
+#define UTF8_TWO_BYTE_LO_nocast(c) __BASE_TWO_BYTE_LO(c, NATIVE_TO_UNI)
+
+/* The next two macros are used when the source should be a single byte
+ * character; checked for under DEBUGGING */
+#define UTF8_EIGHT_BIT_HI(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ ( __BASE_TWO_BYTE_HI(c, NATIVE_TO_LATIN1)))
+#define UTF8_EIGHT_BIT_LO(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (__BASE_TWO_BYTE_LO(c, NATIVE_TO_LATIN1)))
+
+/* These final two macros in the series are used when the source can be any
+ * code point whose UTF-8 is known to occupy 2 bytes; they are less efficient
+ * than the EIGHT_BIT versions on EBCDIC platforms. We use the logical '~'
+ * operator instead of "<=" to avoid getting compiler warnings.
+ * MAX_UTF8_TWO_BYTE should be exactly all one bits in the lower few
+ * places, so the ~ works */
+#define UTF8_TWO_BYTE_HI(c) \
+ (__ASSERT_((sizeof(c) == 1) \
+ || !(((WIDEST_UTYPE)(c)) & ~MAX_UTF8_TWO_BYTE)) \
+ (__BASE_TWO_BYTE_HI(c, NATIVE_TO_UNI)))
+#define UTF8_TWO_BYTE_LO(c) \
+ (__ASSERT_((sizeof(c) == 1) \
+ || !(((WIDEST_UTYPE)(c)) & ~MAX_UTF8_TWO_BYTE)) \
+ (__BASE_TWO_BYTE_LO(c, NATIVE_TO_UNI)))
+
+/* This is illegal in any well-formed UTF-8 in both EBCDIC and ASCII
+ * as it is only in overlongs. */
+#define ILLEGAL_UTF8_BYTE I8_TO_NATIVE_UTF8(0xC1)
/*
* 'UTF' is whether or not p is encoded in UTF8. The names 'foo_lazy_if' stem
* beginning of a utf8 character. Now that foo_utf8() determines that itself,
* no need to do it again here
*/
-#define isIDFIRST_lazy_if(p,UTF) ((IN_BYTES || !UTF ) \
- ? isIDFIRST(*(p)) \
- : isIDFIRST_utf8((const U8*)p))
-#define isALNUM_lazy_if(p,UTF) ((IN_BYTES || (!UTF )) \
- ? isALNUM(*(p)) \
- : isALNUM_utf8((const U8*)p))
+#define isIDFIRST_lazy_if(p,UTF) \
+ _is_utf8_FOO(_CC_IDFIRST, (const U8 *) p, "isIDFIRST_lazy_if", \
+ "isIDFIRST_lazy_if_safe", \
+ cBOOL(UTF && ! IN_BYTES), 0, __FILE__,__LINE__)
+
+#define isIDFIRST_lazy_if_safe(p, e, UTF) \
+ ((IN_BYTES || !UTF) \
+ ? isIDFIRST(*(p)) \
+ : isIDFIRST_utf8_safe(p, e))
+
+#define isWORDCHAR_lazy_if(p,UTF) \
+ _is_utf8_FOO(_CC_IDFIRST, (const U8 *) p, "isWORDCHAR_lazy_if", \
+ "isWORDCHAR_lazy_if_safe", \
+ cBOOL(UTF && ! IN_BYTES), 0, __FILE__,__LINE__)
+
+#define isWORDCHAR_lazy_if_safe(p, e, UTF) \
+ ((IN_BYTES || !UTF) \
+ ? isWORDCHAR(*(p)) \
+ : isWORDCHAR_utf8_safe((U8 *) p, (U8 *) e))
+
+#define isALNUM_lazy_if(p,UTF) \
+ _is_utf8_FOO(_CC_IDFIRST, (const U8 *) p, "isALNUM_lazy_if", \
+ "isWORDCHAR_lazy_if_safe", \
+ cBOOL(UTF && ! IN_BYTES), 0, __FILE__,__LINE__)
-#define isIDFIRST_lazy(p) isIDFIRST_lazy_if(p,1)
-#define isALNUM_lazy(p) isALNUM_lazy_if(p,1)
-
-#define UTF8_MAXBYTES 13
-/* How wide can a single UTF-8 encoded character become in bytes.
- * NOTE: Strictly speaking Perl's UTF-8 should not be called UTF-8
- * since UTF-8 is an encoding of Unicode and given Unicode's current
- * upper limit only four bytes is possible. Perl thinks of UTF-8
- * as a way to encode non-negative integers in a binary format. */
#define UTF8_MAXLEN UTF8_MAXBYTES
-/* The maximum number of UTF-8 bytes a single Unicode character can
- * uppercase/lowercase/fold into; this number depends on the Unicode
- * version. An example of maximal expansion is the U+03B0 which
- * uppercases to U+03C5 U+0308 U+0301. The Unicode databases that
- * tell these things are UnicodeData.txt, CaseFolding.txt, and
- * SpecialCasing.txt. The value is 6 for strict Unicode characters, but it has
- * to be as big as Perl allows for a single character */
-#define UTF8_MAXBYTES_CASE UTF8_MAXBYTES
-
/* A Unicode character can fold to up to 3 characters */
#define UTF8_MAX_FOLD_CHAR_EXPAND 3
-#define IN_BYTES (CopHINTS_get(PL_curcop) & HINT_BYTES)
+#define IN_BYTES UNLIKELY(CopHINTS_get(PL_curcop) & HINT_BYTES)
+
+/*
+
+=for apidoc Am|bool|DO_UTF8|SV* sv
+Returns a bool giving whether or not the PV in C<sv> is to be treated as being
+encoded in UTF-8.
+
+You should use this I<after> a call to C<SvPV()> or one of its variants, in
+case any call to string overloading updates the internal UTF-8 encoding flag.
+
+=cut
+*/
#define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTES)
-#define IN_UNI_8_BIT \
- (CopHINTS_get(PL_curcop) & (HINT_UNI_8_BIT|HINT_LOCALE_NOT_CHARS) \
- && ! IN_LOCALE_RUNTIME && ! IN_BYTES)
+
+/* Should all strings be treated as Unicode, and not just UTF-8 encoded ones?
+ * Is so within 'feature unicode_strings' or 'locale :not_characters', and not
+ * within 'use bytes'. UTF-8 locales are not tested for here, but perhaps
+ * could be */
+#define IN_UNI_8_BIT \
+ (( ( (CopHINTS_get(PL_curcop) & HINT_UNI_8_BIT)) \
+ || ( CopHINTS_get(PL_curcop) & HINT_LOCALE_PARTIAL \
+ /* -1 below is for :not_characters */ \
+ && _is_in_locale_category(FALSE, -1))) \
+ && (! IN_BYTES))
#define UTF8_ALLOW_EMPTY 0x0001 /* Allow a zero length string */
+#define UTF8_GOT_EMPTY UTF8_ALLOW_EMPTY
/* Allow first byte to be a continuation byte */
#define UTF8_ALLOW_CONTINUATION 0x0002
+#define UTF8_GOT_CONTINUATION UTF8_ALLOW_CONTINUATION
-/* Allow second... bytes to be non-continuation bytes */
+/* Unexpected continuation byte */
#define UTF8_ALLOW_NON_CONTINUATION 0x0004
+#define UTF8_GOT_NON_CONTINUATION UTF8_ALLOW_NON_CONTINUATION
/* expecting more bytes than were available in the string */
#define UTF8_ALLOW_SHORT 0x0008
+#define UTF8_GOT_SHORT UTF8_ALLOW_SHORT
-/* Overlong sequence; i.e., the code point can be specified in fewer bytes. */
+/* Overlong sequence; i.e., the code point can be specified in fewer bytes.
+ * First one will convert the overlong to the REPLACEMENT CHARACTER; second
+ * will return what the overlong evaluates to */
#define UTF8_ALLOW_LONG 0x0010
-
-#define UTF8_DISALLOW_SURROGATE 0x0020 /* Unicode surrogates */
-#define UTF8_WARN_SURROGATE 0x0040
-
-#define UTF8_DISALLOW_NONCHAR 0x0080 /* Unicode non-character */
-#define UTF8_WARN_NONCHAR 0x0100 /* code points */
-
-#define UTF8_DISALLOW_SUPER 0x0200 /* Super-set of Unicode: code */
-#define UTF8_WARN_SUPER 0x0400 /* points above the legal max */
-
-/* Code points which never were part of the original UTF-8 standard, the first
- * byte of which is a FE or FF on ASCII platforms. */
-#define UTF8_DISALLOW_FE_FF 0x0800
-#define UTF8_WARN_FE_FF 0x1000
-
-#define UTF8_CHECK_ONLY 0x2000
+#define UTF8_ALLOW_LONG_AND_ITS_VALUE (UTF8_ALLOW_LONG|0x0020)
+#define UTF8_GOT_LONG UTF8_ALLOW_LONG
+
+#define UTF8_ALLOW_OVERFLOW 0x0080
+#define UTF8_GOT_OVERFLOW UTF8_ALLOW_OVERFLOW
+
+#define UTF8_DISALLOW_SURROGATE 0x0100 /* Unicode surrogates */
+#define UTF8_GOT_SURROGATE UTF8_DISALLOW_SURROGATE
+#define UTF8_WARN_SURROGATE 0x0200
+
+/* Unicode non-character code points */
+#define UTF8_DISALLOW_NONCHAR 0x0400
+#define UTF8_GOT_NONCHAR UTF8_DISALLOW_NONCHAR
+#define UTF8_WARN_NONCHAR 0x0800
+
+/* Super-set of Unicode: code points above the legal max */
+#define UTF8_DISALLOW_SUPER 0x1000
+#define UTF8_GOT_SUPER UTF8_DISALLOW_SUPER
+#define UTF8_WARN_SUPER 0x2000
+
+/* The original UTF-8 standard did not define UTF-8 with start bytes of 0xFE or
+ * 0xFF, though UTF-EBCDIC did. This allowed both versions to represent code
+ * points up to 2 ** 31 - 1. Perl extends UTF-8 so that 0xFE and 0xFF are
+ * usable on ASCII platforms, and 0xFF means something different than
+ * UTF-EBCDIC defines. These changes allow code points of 64 bits (actually
+ * somewhat more) to be represented on both platforms. But these are Perl
+ * extensions, and not likely to be interchangeable with other languages. Note
+ * that on ASCII platforms, FE overflows a signed 32-bit word, and FF an
+ * unsigned one. */
+#define UTF8_DISALLOW_PERL_EXTENDED 0x4000
+#define UTF8_GOT_PERL_EXTENDED UTF8_DISALLOW_PERL_EXTENDED
+#define UTF8_WARN_PERL_EXTENDED 0x8000
+
+/* For back compat, these old names are misleading for overlongs and
+ * UTF_EBCDIC. */
+#define UTF8_DISALLOW_ABOVE_31_BIT UTF8_DISALLOW_PERL_EXTENDED
+#define UTF8_GOT_ABOVE_31_BIT UTF8_GOT_PERL_EXTENDED
+#define UTF8_WARN_ABOVE_31_BIT UTF8_WARN_PERL_EXTENDED
+#define UTF8_DISALLOW_FE_FF UTF8_DISALLOW_PERL_EXTENDED
+#define UTF8_WARN_FE_FF UTF8_WARN_PERL_EXTENDED
+
+#define UTF8_CHECK_ONLY 0x10000
+#define _UTF8_NO_CONFIDENCE_IN_CURLEN 0x20000 /* Internal core use only */
/* For backwards source compatibility. They do nothing, as the default now
* includes what they used to mean. The first one's meaning was to allow the
#define UTF8_ALLOW_FFFF 0
#define UTF8_ALLOW_SURROGATE 0
-#define UTF8_DISALLOW_ILLEGAL_INTERCHANGE (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_NONCHAR|UTF8_DISALLOW_SURROGATE|UTF8_DISALLOW_FE_FF)
+/* C9 refers to Unicode Corrigendum #9: allows but discourages non-chars */
+#define UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE \
+ (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_SURROGATE)
+#define UTF8_WARN_ILLEGAL_C9_INTERCHANGE (UTF8_WARN_SUPER|UTF8_WARN_SURROGATE)
+
+#define UTF8_DISALLOW_ILLEGAL_INTERCHANGE \
+ (UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE|UTF8_DISALLOW_NONCHAR)
#define UTF8_WARN_ILLEGAL_INTERCHANGE \
- (UTF8_WARN_SUPER|UTF8_WARN_NONCHAR|UTF8_WARN_SURROGATE|UTF8_WARN_FE_FF)
-#define UTF8_ALLOW_ANY \
- (~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE|UTF8_WARN_ILLEGAL_INTERCHANGE))
-#define UTF8_ALLOW_ANYUV \
- (UTF8_ALLOW_EMPTY \
- & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE|UTF8_WARN_ILLEGAL_INTERCHANGE))
-#define UTF8_ALLOW_DEFAULT (ckWARN(WARN_UTF8) ? 0 : \
- UTF8_ALLOW_ANYUV)
+ (UTF8_WARN_ILLEGAL_C9_INTERCHANGE|UTF8_WARN_NONCHAR)
+
+/* This is typically used for code that processes UTF-8 input and doesn't want
+ * to have to deal with any malformations that might be present. All such will
+ * be safely replaced by the REPLACEMENT CHARACTER, unless other flags
+ * overriding this are also present. */
+#define UTF8_ALLOW_ANY ( UTF8_ALLOW_CONTINUATION \
+ |UTF8_ALLOW_NON_CONTINUATION \
+ |UTF8_ALLOW_SHORT \
+ |UTF8_ALLOW_LONG \
+ |UTF8_ALLOW_OVERFLOW)
+
+/* Accept any Perl-extended UTF-8 that evaluates to any UV on the platform, but
+ * not any malformed. This is the default. (Note that UVs above IV_MAX are
+ * deprecated. */
+#define UTF8_ALLOW_ANYUV 0
+#define UTF8_ALLOW_DEFAULT UTF8_ALLOW_ANYUV
-/* Surrogates, non-character code points and above-Unicode code points are
- * problematic in some contexts. This allows code that needs to check for
- * those to to quickly exclude the vast majority of code points it will
- * encounter */
-#ifdef EBCDIC
-# define UTF8_FIRST_PROBLEMATIC_CODE_POINT_FIRST_BYTE UTF_TO_NATIVE(0xF1)
-#else
-# define UTF8_FIRST_PROBLEMATIC_CODE_POINT_FIRST_BYTE 0xED
-#endif
+/*
+=for apidoc Am|bool|UTF8_IS_SURROGATE|const U8 *s|const U8 *e
+
+Evaluates to non-zero if the first few bytes of the string starting at C<s> and
+looking no further than S<C<e - 1>> are well-formed UTF-8 that represents one
+of the Unicode surrogate code points; otherwise it evaluates to 0. If
+non-zero, the value gives how many bytes starting at C<s> comprise the code
+point's representation.
-/* ASCII EBCDIC I8
- * U+D7FF: \xED\x9F\xBF \xF1\xB5\xBF\xBF last before surrogates
- * U+D800: \xED\xA0\x80 \xF1\xB6\xA0\xA0 1st surrogate
- * U+DFFF: \xED\xBF\xBF \xF1\xB7\xBF\xBF final surrogate
- * U+E000: \xEE\x80\x80 \xF1\xB8\xA0\xA0 next after surrogates
+=cut
*/
-#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)))
-#else
-# define UTF8_IS_SURROGATE(s) (*(s) == 0xED && *((s) + 1) >= 0xA0)
-#endif
+#define UTF8_IS_SURROGATE(s, e) is_SURROGATE_utf8_safe(s, e)
+
+
+#define UTF8_IS_REPLACEMENT(s, send) is_REPLACEMENT_utf8_safe(s,send)
+
+/*
+=for apidoc Am|bool|UTF8_IS_SUPER|const U8 *s|const U8 *e
+
+Recall that Perl recognizes an extension to UTF-8 that can encode code
+points larger than the ones defined by Unicode, which are 0..0x10FFFF.
+
+This macro evaluates to non-zero if the first few bytes of the string starting
+at C<s> and looking no further than S<C<e - 1>> are from this UTF-8 extension;
+otherwise it evaluates to 0. If non-zero, the value gives how many bytes
+starting at C<s> comprise the code point's representation.
-/* ASCII EBCDIC I8
+0 is returned if the bytes are not well-formed extended UTF-8, or if they
+represent a code point that cannot fit in a UV on the current platform. Hence
+this macro can give different results when run on a 64-bit word machine than on
+one with a 32-bit word size.
+
+Note that it is deprecated to have code points that are larger than what can
+fit in an IV on the current machine.
+
+=cut
+
+ * ASCII EBCDIC I8
* U+10FFFF: \xF4\x8F\xBF\xBF \xF9\xA1\xBF\xBF\xBF max legal Unicode
* U+110000: \xF4\x90\x80\x80 \xF9\xA2\xA0\xA0\xA0
* U+110001: \xF4\x90\x80\x81 \xF9\xA2\xA0\xA0\xA1
*/
-#ifdef EBCDIC /* Both versions assume well-formed UTF8 */
-# define UTF8_IS_SUPER(s) (*(s) >= UTF_TO_NATIVE(0xF9) \
- && (*(s) > UTF_TO_NATIVE(0xF9) || (*((s) + 1) >= UTF_TO_NATIVE(0xA2))))
+#ifdef EBCDIC
+# define UTF8_IS_SUPER(s, e) \
+ (( LIKELY((e) > (s) + 4) \
+ && NATIVE_UTF8_TO_I8(*(s)) >= 0xF9 \
+ && ( NATIVE_UTF8_TO_I8(*(s)) > 0xF9 \
+ || (NATIVE_UTF8_TO_I8(*((s) + 1)) >= 0xA2)) \
+ && LIKELY((s) + UTF8SKIP(s) <= (e))) \
+ ? _is_utf8_char_helper(s, s + UTF8SKIP(s), 0) : 0)
#else
-# define UTF8_IS_SUPER(s) (*(s) >= 0xF4 \
- && (*(s) > 0xF4 || (*((s) + 1) >= 0x90)))
+# define UTF8_IS_SUPER(s, e) \
+ (( LIKELY((e) > (s) + 3) \
+ && (*(U8*) (s)) >= 0xF4 \
+ && ((*(U8*) (s)) > 0xF4 || (*((U8*) (s) + 1) >= 0x90))\
+ && LIKELY((s) + UTF8SKIP(s) <= (e))) \
+ ? _is_utf8_char_helper(s, s + UTF8SKIP(s), 0) : 0)
#endif
-/* ASCII EBCDIC I8
- * U+FDCF: \xEF\xB7\x8F \xF1\xBF\xAE\xAF last before non-char block
- * U+FDD0: \xEF\xB7\x90 \xF1\xBF\xAE\xB0 first non-char in block
- * U+FDEF: \xEF\xB7\xAF \xF1\xBF\xAF\xAF last non-char in block
- * U+FDF0: \xEF\xB7\xB0 \xF1\xBF\xAF\xB0 first after non-char block
- * U+FFFF: \xEF\xBF\xBF \xF1\xBF\xBF\xBF
- * U+1FFFF: \xF0\x9F\xBF\xBF \xF3\xBF\xBF\xBF
- * U+2FFFF: \xF0\xAF\xBF\xBF \xF5\xBF\xBF\xBF
- * U+3FFFF: \xF0\xBF\xBF\xBF \xF7\xBF\xBF\xBF
- * U+4FFFF: \xF1\x8F\xBF\xBF \xF8\xA9\xBF\xBF\xBF
- * U+5FFFF: \xF1\x9F\xBF\xBF \xF8\xAB\xBF\xBF\xBF
- * U+6FFFF: \xF1\xAF\xBF\xBF \xF8\xAD\xBF\xBF\xBF
- * U+7FFFF: \xF1\xBF\xBF\xBF \xF8\xAF\xBF\xBF\xBF
- * U+8FFFF: \xF2\x8F\xBF\xBF \xF8\xB1\xBF\xBF\xBF
- * U+9FFFF: \xF2\x9F\xBF\xBF \xF8\xB3\xBF\xBF\xBF
- * U+AFFFF: \xF2\xAF\xBF\xBF \xF8\xB5\xBF\xBF\xBF
- * U+BFFFF: \xF2\xBF\xBF\xBF \xF8\xB7\xBF\xBF\xBF
- * U+CFFFF: \xF3\x8F\xBF\xBF \xF8\xB9\xBF\xBF\xBF
- * U+DFFFF: \xF3\x9F\xBF\xBF \xF8\xBB\xBF\xBF\xBF
- * U+EFFFF: \xF3\xAF\xBF\xBF \xF8\xBD\xBF\xBF\xBF
- * U+FFFFF: \xF3\xBF\xBF\xBF \xF8\xBF\xBF\xBF\xBF
- * U+10FFFF: \xF4\x8F\xBF\xBF \xF9\xA1\xBF\xBF\xBF
+/* These are now machine generated, and the 'given' clause is no longer
+ * applicable */
+#define UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s, e) \
+ cBOOL(is_NONCHAR_utf8_safe(s,e))
+
+/*
+=for apidoc Am|bool|UTF8_IS_NONCHAR|const U8 *s|const U8 *e
+
+Evaluates to non-zero if the first few bytes of the string starting at C<s> and
+looking no further than S<C<e - 1>> are well-formed UTF-8 that represents one
+of the Unicode non-character code points; otherwise it evaluates to 0. If
+non-zero, the value gives how many bytes starting at C<s> comprise the code
+point's representation.
+
+=cut
*/
-#define UTF8_IS_NONCHAR_(s) ( \
- *(s) >= UTF8_FIRST_PROBLEMATIC_CODE_POINT_FIRST_BYTE \
- && ! UTF8_IS_SUPER(s) \
- && UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_FIRST_PROBLEMATIC(s) \
-
-#ifdef EBCDIC /* Both versions assume well-formed UTF8 */
-# define UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s) \
- ((*(s) == UTF_TO_NATIVE(0xF1) \
- && (*((s) + 1) == UTF_TO_NATIVE(0xBF) \
- && ((*((s) + 2) == UTF_TO_NATIVE(0xAE) \
- && *((s) + 3) >= UTF_TO_NATIVE(0xB0)) \
- || (*((s) + 2) == UTF_TO_NATIVE(0xAF) \
- && *((s) + 3) <= UTF_TO_NATIVE(0xAF))))) \
- || (UTF8SKIP(*(s)) > 3 \
- /* (These were all derived by inspection and experimentation with an */ \
- /* editor) The next line checks the next to final byte in the char */ \
- && *((s) + UTF8SKIP(*(s)) - 2) == UTF_TO_NATIVE(0xBF) \
- && *((s) + UTF8SKIP(*(s)) - 3) == UTF_TO_NATIVE(0xBF) \
- && (NATIVE_TO_UTF(*((s) + UTF8SKIP(*(s)) - 4)) & 0x81) == 0x81 \
- && (NATIVE_TO_UTF(*((s) + UTF8SKIP(*(s)) - 1)) & 0xBE) == 0XBE))
-#else
-# define UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s) \
- ((*(s) == 0xEF \
- && ((*((s) + 1) == 0xB7 && (*((s) + 2) >= 0x90 && (*((s) + 2) <= 0xAF)))\
- /* Gets U+FFF[EF] */ \
- || (*((s) + 1) == 0xBF && ((*((s) + 2) & 0xBE) == 0xBE)))) \
- || ((*((s) + 2) == 0xBF \
- && (*((s) + 3) & 0xBE) == 0xBE \
- /* Excludes things like U+10FFE = \xF0\x90\xBF\xBE */ \
- && (*((s) + 1) & 0x8F) == 0x8F)))
-#endif
+#define UTF8_IS_NONCHAR(s, e) \
+ UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s, e)
#define UNICODE_SURROGATE_FIRST 0xD800
#define UNICODE_SURROGATE_LAST 0xDFFF
* let's be conservative and do as Unicode says. */
#define PERL_UNICODE_MAX 0x10FFFF
-#define UNICODE_WARN_SURROGATE 0x0001 /* UTF-16 surrogates */
-#define UNICODE_WARN_NONCHAR 0x0002 /* Non-char code points */
-#define UNICODE_WARN_SUPER 0x0004 /* Above 0x10FFFF */
-#define UNICODE_WARN_FE_FF 0x0008 /* Above 0x10FFFF */
-#define UNICODE_DISALLOW_SURROGATE 0x0010
-#define UNICODE_DISALLOW_NONCHAR 0x0020
-#define UNICODE_DISALLOW_SUPER 0x0040
-#define UNICODE_DISALLOW_FE_FF 0x0080
-#define UNICODE_WARN_ILLEGAL_INTERCHANGE \
- (UNICODE_WARN_SURROGATE|UNICODE_WARN_NONCHAR|UNICODE_WARN_SUPER)
-#define UNICODE_DISALLOW_ILLEGAL_INTERCHANGE \
- (UNICODE_DISALLOW_SURROGATE|UNICODE_DISALLOW_NONCHAR|UNICODE_DISALLOW_SUPER)
+#define UNICODE_WARN_SURROGATE 0x0001 /* UTF-16 surrogates */
+#define UNICODE_WARN_NONCHAR 0x0002 /* Non-char code points */
+#define UNICODE_WARN_SUPER 0x0004 /* Above 0x10FFFF */
+#define UNICODE_WARN_PERL_EXTENDED 0x0008 /* Above 0x7FFF_FFFF */
+#define UNICODE_WARN_ABOVE_31_BIT UNICODE_WARN_PERL_EXTENDED
+#define UNICODE_DISALLOW_SURROGATE 0x0010
+#define UNICODE_DISALLOW_NONCHAR 0x0020
+#define UNICODE_DISALLOW_SUPER 0x0040
+#define UNICODE_DISALLOW_PERL_EXTENDED 0x0080
+#define UNICODE_DISALLOW_ABOVE_31_BIT UNICODE_DISALLOW_PERL_EXTENDED
+#define UNICODE_WARN_ILLEGAL_C9_INTERCHANGE \
+ (UNICODE_WARN_SURROGATE|UNICODE_WARN_SUPER)
+#define UNICODE_WARN_ILLEGAL_INTERCHANGE \
+ (UNICODE_WARN_ILLEGAL_C9_INTERCHANGE|UNICODE_WARN_NONCHAR)
+#define UNICODE_DISALLOW_ILLEGAL_C9_INTERCHANGE \
+ (UNICODE_DISALLOW_SURROGATE|UNICODE_DISALLOW_SUPER)
+#define UNICODE_DISALLOW_ILLEGAL_INTERCHANGE \
+ (UNICODE_DISALLOW_ILLEGAL_C9_INTERCHANGE|UNICODE_DISALLOW_NONCHAR)
/* For backward source compatibility, as are now the default */
#define UNICODE_ALLOW_SURROGATE 0
#define UNICODE_ALLOW_SUPER 0
#define UNICODE_ALLOW_ANY 0
-#define UNICODE_IS_SURROGATE(c) ((c) >= UNICODE_SURROGATE_FIRST && \
- (c) <= UNICODE_SURROGATE_LAST)
-#define UNICODE_IS_REPLACEMENT(c) ((c) == UNICODE_REPLACEMENT)
-#define UNICODE_IS_BYTE_ORDER_MARK(c) ((c) == UNICODE_BYTE_ORDER_MARK)
-#define UNICODE_IS_NONCHAR(c) ((c >= 0xFDD0 && c <= 0xFDEF) \
- /* The other noncharacters end in FFFE or FFFF, which \
- * the mask below catches both of, but beyond the last \
- * official unicode code point, they aren't \
- * noncharacters, since those aren't Unicode \
- * characters at all */ \
- || ((((c & 0xFFFE) == 0xFFFE)) && ! UNICODE_IS_SUPER(c)))
-#define UNICODE_IS_SUPER(c) ((c) > PERL_UNICODE_MAX)
-#define UNICODE_IS_FE_FF(c) ((c) > 0x7FFFFFFF)
-
-#ifdef HAS_QUAD
-# define UTF8_QUAD_MAX UINT64_C(0x1000000000)
-#endif
-
+/* This matches the 2048 code points between UNICODE_SURROGATE_FIRST (0xD800) and
+ * UNICODE_SURROGATE_LAST (0xDFFF) */
+#define UNICODE_IS_SURROGATE(uv) (((UV) (uv) & (~0xFFFF | 0xF800)) \
+ == 0xD800)
+
+#define UNICODE_IS_REPLACEMENT(uv) ((UV) (uv) == UNICODE_REPLACEMENT)
+#define UNICODE_IS_BYTE_ORDER_MARK(uv) ((UV) (uv) == UNICODE_BYTE_ORDER_MARK)
+
+/* Is 'uv' one of the 32 contiguous-range noncharacters? */
+#define UNICODE_IS_32_CONTIGUOUS_NONCHARS(uv) ((UV) (uv) >= 0xFDD0 \
+ && (UV) (uv) <= 0xFDEF)
+
+/* Is 'uv' one of the 34 plane-ending noncharacters 0xFFFE, 0xFFFF, 0x1FFFE,
+ * 0x1FFFF, ... 0x10FFFE, 0x10FFFF, given that we know that 'uv' is not above
+ * the Unicode legal max */
+#define UNICODE_IS_END_PLANE_NONCHAR_GIVEN_NOT_SUPER(uv) \
+ (((UV) (uv) & 0xFFFE) == 0xFFFE)
+
+#define UNICODE_IS_NONCHAR(uv) \
+ ( UNICODE_IS_32_CONTIGUOUS_NONCHARS(uv) \
+ || ( LIKELY( ! UNICODE_IS_SUPER(uv)) \
+ && UNICODE_IS_END_PLANE_NONCHAR_GIVEN_NOT_SUPER(uv)))
+
+#define UNICODE_IS_SUPER(uv) ((UV) (uv) > PERL_UNICODE_MAX)
+
+#define LATIN_SMALL_LETTER_SHARP_S LATIN_SMALL_LETTER_SHARP_S_NATIVE
+#define LATIN_SMALL_LETTER_Y_WITH_DIAERESIS \
+ LATIN_SMALL_LETTER_Y_WITH_DIAERESIS_NATIVE
+#define MICRO_SIGN MICRO_SIGN_NATIVE
+#define LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE \
+ LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE_NATIVE
+#define LATIN_SMALL_LETTER_A_WITH_RING_ABOVE \
+ LATIN_SMALL_LETTER_A_WITH_RING_ABOVE_NATIVE
#define UNICODE_GREEK_CAPITAL_LETTER_SIGMA 0x03A3
#define UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA 0x03C2
#define UNICODE_GREEK_SMALL_LETTER_SIGMA 0x03C3
#define GREEK_SMALL_LETTER_MU 0x03BC
-#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 GREEK_CAPITAL_LETTER_MU 0x039C /* Upper and title case
+ of MICRON */
+#define LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS 0x0178 /* Also is title case */
+#ifdef LATIN_CAPITAL_LETTER_SHARP_S_UTF8
+# define LATIN_CAPITAL_LETTER_SHARP_S 0x1E9E
+#endif
+#define LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE 0x130
+#define LATIN_SMALL_LETTER_DOTLESS_I 0x131
+#define LATIN_SMALL_LETTER_LONG_S 0x017F
+#define LATIN_SMALL_LIGATURE_LONG_S_T 0xFB05
+#define LATIN_SMALL_LIGATURE_ST 0xFB06
+#define KELVIN_SIGN 0x212A
+#define ANGSTROM_SIGN 0x212B
#define UNI_DISPLAY_ISPRINT 0x0001
#define UNI_DISPLAY_BACKSLASH 0x0002
#define UNI_DISPLAY_QQ (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
#define UNI_DISPLAY_REGEX (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
-#ifndef EBCDIC
-# define LATIN_SMALL_LETTER_SHARP_S 0x00DF
-# define LATIN_SMALL_LETTER_Y_WITH_DIAERESIS 0x00FF
-# define MICRO_SIGN 0x00B5
-# define LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE 0x00C5
-# define LATIN_SMALL_LETTER_A_WITH_RING_ABOVE 0x00E5
-#endif
-
#define ANYOF_FOLD_SHARP_S(node, input, end) \
(ANYOF_BITMAP_TEST(node, LATIN_SMALL_LETTER_SHARP_S) && \
(ANYOF_NONBITMAP(node)) && \
(ANYOF_FLAGS(node) & ANYOF_LOC_NONBITMAP_FOLD) && \
((end) > (input) + 1) && \
- toLOWER((input)[0]) == 's' && \
- toLOWER((input)[1]) == 's')
+ isALPHA_FOLD_EQ((input)[0], 's'))
+
#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) \
- ((p)[0] >= 0xC2 && (p)[0] <= 0xDF && \
- (p)[1] >= 0x80 && (p)[1] <= 0xBF)
-# define IS_UTF8_CHAR_3a(p) \
- ((p)[0] == 0xE0 && \
- (p)[1] >= 0xA0 && (p)[1] <= 0xBF && \
- (p)[2] >= 0x80 && (p)[2] <= 0xBF)
-# define IS_UTF8_CHAR_3b(p) \
- ((p)[0] >= 0xE1 && (p)[0] <= 0xEF && \
- (p)[1] >= 0x80 && (p)[1] <= 0xBF && \
- (p)[2] >= 0x80 && (p)[2] <= 0xBF)
-# define IS_UTF8_CHAR_4a(p) \
- ((p)[0] == 0xF0 && \
- (p)[1] >= 0x90 && (p)[1] <= 0xBF && \
- (p)[2] >= 0x80 && (p)[2] <= 0xBF && \
- (p)[3] >= 0x80 && (p)[3] <= 0xBF)
-/* 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_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))
-# define IS_UTF8_CHAR_4(p) \
- (IS_UTF8_CHAR_4a(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
- * (2) it allows code points past U+10FFFF.
- * The Perl_is_utf8_char() full "slow" code will handle the Perl
- * "extended UTF-8". */
-# define IS_UTF8_CHAR(p, n) \
- ((n) == 1 ? IS_UTF8_CHAR_1(p) : \
- (n) == 2 ? IS_UTF8_CHAR_2(p) : \
- (n) == 3 ? IS_UTF8_CHAR_3(p) : \
- (n) == 4 ? IS_UTF8_CHAR_4(p) : 0)
-
-# define IS_UTF8_CHAR_FAST(n) ((n) <= 4)
-
-#else /* EBCDIC */
-
-/* This is an attempt to port IS_UTF8_CHAR to EBCDIC based on eyeballing.
- * untested. If want to exclude surrogates and above-Unicode, see the
- * definitions for UTF8_IS_SURROGATE and UTF8_IS_SUPER */
-# define IS_UTF8_CHAR_1(p) \
- (NATIVE_TO_ASCII((p)[0]) <= 0x9F)
-# define IS_UTF8_CHAR_2(p) \
- (NATIVE_TO_I8((p)[0]) >= 0xC5 && NATIVE_TO_I8((p)[0]) <= 0xDF && \
- NATIVE_TO_I8((p)[1]) >= 0xA0 && NATIVE_TO_I8((p)[1]) <= 0xBF)
-# define IS_UTF8_CHAR_3(p) \
- (NATIVE_TO_I8((p)[0]) == 0xE1 && NATIVE_TO_I8((p)[1]) <= 0xEF && \
- NATIVE_TO_I8((p)[1]) >= 0xA0 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[2]) >= 0xA0 && NATIVE_TO_I8((p)[2]) <= 0xBF)
-# define IS_UTF8_CHAR_4a(p) \
- (NATIVE_TO_I8((p)[0]) == 0xF0 && \
- NATIVE_TO_I8((p)[1]) >= 0xB0 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[2]) >= 0xA0 && NATIVE_TO_I8((p)[2]) <= 0xBF && \
- NATIVE_TO_I8((p)[3]) >= 0xA0 && NATIVE_TO_I8((p)[3]) <= 0xBF)
-# define IS_UTF8_CHAR_4b(p) \
- (NATIVE_TO_I8((p)[0]) >= 0xF1 && NATIVE_TO_I8((p)[0]) <= 0xF7 && \
- NATIVE_TO_I8((p)[1]) >= 0xA0 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[2]) >= 0xA0 && NATIVE_TO_I8((p)[2]) <= 0xBF && \
- NATIVE_TO_I8((p)[3]) >= 0xA0 && NATIVE_TO_I8((p)[3]) <= 0xBF)
-# define IS_UTF8_CHAR_5a(p) \
- (NATIVE_TO_I8((p)[0]) == 0xF8 && \
- NATIVE_TO_I8((p)[1]) >= 0xA8 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[1]) >= 0xA0 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[2]) >= 0xA0 && NATIVE_TO_I8((p)[2]) <= 0xBF && \
- NATIVE_TO_I8((p)[3]) >= 0xA0 && NATIVE_TO_I8((p)[3]) <= 0xBF)
-# define IS_UTF8_CHAR_5b(p) \
- (NATIVE_TO_I8((p)[0]) >= 0xF9 && NATIVE_TO_I8((p)[1]) <= 0xFB && \
- NATIVE_TO_I8((p)[1]) >= 0xA0 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[1]) >= 0xA0 && NATIVE_TO_I8((p)[1]) <= 0xBF && \
- NATIVE_TO_I8((p)[2]) >= 0xA0 && NATIVE_TO_I8((p)[2]) <= 0xBF && \
- NATIVE_TO_I8((p)[3]) >= 0xA0 && NATIVE_TO_I8((p)[3]) <= 0xBF)
-
-# define IS_UTF8_CHAR_4(p) \
- (IS_UTF8_CHAR_4a(p) || \
- IS_UTF8_CHAR_4b(p))
-# define IS_UTF8_CHAR_5(p) \
- (IS_UTF8_CHAR_5a(p) || \
- IS_UTF8_CHAR_5b(p))
-# define IS_UTF8_CHAR(p, n) \
- ((n) == 1 ? IS_UTF8_CHAR_1(p) : \
- (n) == 2 ? IS_UTF8_CHAR_2(p) : \
- (n) == 3 ? IS_UTF8_CHAR_3(p) : \
- (n) == 4 ? IS_UTF8_CHAR_4(p) : \
- (n) == 5 ? IS_UTF8_CHAR_5(p) : 0)
-
-# define IS_UTF8_CHAR_FAST(n) ((n) <= 5)
-
-#endif /* IS_UTF8_CHAR() for UTF-8 */
+/*
+
+=for apidoc Am|STRLEN|isUTF8_CHAR|const U8 *s|const U8 *e
+
+Evaluates to non-zero if the first few bytes of the string starting at C<s> and
+looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
+that represents some code point; otherwise it evaluates to 0. If non-zero, the
+value gives how many bytes starting at C<s> comprise the code point's
+representation. Any bytes remaining before C<e>, but beyond the ones needed to
+form the first code point in C<s>, are not examined.
+
+The code point can be any that will fit in a UV on this machine, using Perl's
+extension to official UTF-8 to represent those higher than the Unicode maximum
+of 0x10FFFF. That means that this macro is used to efficiently decide if the
+next few bytes in C<s> is legal UTF-8 for a single character.
+
+Use C<L</isSTRICT_UTF8_CHAR>> to restrict the acceptable code points to those
+defined by Unicode to be fully interchangeable across applications;
+C<L</isC9_STRICT_UTF8_CHAR>> to use the L<Unicode Corrigendum
+#9|http://www.unicode.org/versions/corrigendum9.html> definition of allowable
+code points; and C<L</isUTF8_CHAR_flags>> for a more customized definition.
+
+Use C<L</is_utf8_string>>, C<L</is_utf8_string_loc>>, and
+C<L</is_utf8_string_loclen>> to check entire strings.
+
+Note that it is deprecated to use code points higher than what will fit in an
+IV. This macro does not raise any warnings for such code points, treating them
+as valid.
+
+Note also that a UTF-8 INVARIANT character (i.e. ASCII on non-EBCDIC machines)
+is a valid UTF-8 character.
+
+=cut
+*/
+
+#define isUTF8_CHAR(s, e) \
+ (UNLIKELY((e) <= (s)) \
+ ? 0 \
+ : (UTF8_IS_INVARIANT(*s)) \
+ ? 1 \
+ : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
+ ? 0 \
+ : LIKELY(NATIVE_UTF8_TO_I8(*s) <= _IS_UTF8_CHAR_HIGHEST_START_BYTE) \
+ ? is_UTF8_CHAR_utf8_no_length_checks(s) \
+ : _is_utf8_char_helper(s, e, 0))
+
+#define is_utf8_char_buf(buf, buf_end) isUTF8_CHAR(buf, buf_end)
+#define bytes_from_utf8(s, lenp, is_utf8p) \
+ bytes_from_utf8_loc(s, lenp, is_utf8p, 0)
/*
- * Local variables:
- * c-indentation-style: bsd
- * c-basic-offset: 4
- * indent-tabs-mode: t
- * End:
- *
- * ex: set ts=8 sts=4 sw=4 noet:
+
+=for apidoc Am|STRLEN|isSTRICT_UTF8_CHAR|const U8 *s|const U8 *e
+
+Evaluates to non-zero if the first few bytes of the string starting at C<s> and
+looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
+Unicode code point completely acceptable for open interchange between all
+applications; otherwise it evaluates to 0. If non-zero, the value gives how
+many bytes starting at C<s> comprise the code point's representation. Any
+bytes remaining before C<e>, but beyond the ones needed to form the first code
+point in C<s>, are not examined.
+
+The largest acceptable code point is the Unicode maximum 0x10FFFF, and must not
+be a surrogate nor a non-character code point. Thus this excludes any code
+point from Perl's extended UTF-8.
+
+This is used to efficiently decide if the next few bytes in C<s> is
+legal Unicode-acceptable UTF-8 for a single character.
+
+Use C<L</isC9_STRICT_UTF8_CHAR>> to use the L<Unicode Corrigendum
+#9|http://www.unicode.org/versions/corrigendum9.html> definition of allowable
+code points; C<L</isUTF8_CHAR>> to check for Perl's extended UTF-8;
+and C<L</isUTF8_CHAR_flags>> for a more customized definition.
+
+Use C<L</is_strict_utf8_string>>, C<L</is_strict_utf8_string_loc>>, and
+C<L</is_strict_utf8_string_loclen>> to check entire strings.
+
+=cut
+*/
+
+#define isSTRICT_UTF8_CHAR(s, e) \
+ (UNLIKELY((e) <= (s)) \
+ ? 0 \
+ : (UTF8_IS_INVARIANT(*s)) \
+ ? 1 \
+ : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
+ ? 0 \
+ : is_STRICT_UTF8_CHAR_utf8_no_length_checks(s))
+
+/*
+
+=for apidoc Am|STRLEN|isC9_STRICT_UTF8_CHAR|const U8 *s|const U8 *e
+
+Evaluates to non-zero if the first few bytes of the string starting at C<s> and
+looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
+Unicode non-surrogate code point; otherwise it evaluates to 0. If non-zero,
+the value gives how many bytes starting at C<s> comprise the code point's
+representation. Any bytes remaining before C<e>, but beyond the ones needed to
+form the first code point in C<s>, are not examined.
+
+The largest acceptable code point is the Unicode maximum 0x10FFFF. This
+differs from C<L</isSTRICT_UTF8_CHAR>> only in that it accepts non-character
+code points. This corresponds to
+L<Unicode Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>.
+which said that non-character code points are merely discouraged rather than
+completely forbidden in open interchange. See
+L<perlunicode/Noncharacter code points>.
+
+Use C<L</isUTF8_CHAR>> to check for Perl's extended UTF-8; and
+C<L</isUTF8_CHAR_flags>> for a more customized definition.
+
+Use C<L</is_c9strict_utf8_string>>, C<L</is_c9strict_utf8_string_loc>>, and
+C<L</is_c9strict_utf8_string_loclen>> to check entire strings.
+
+=cut
+*/
+
+#define isC9_STRICT_UTF8_CHAR(s, e) \
+ (UNLIKELY((e) <= (s)) \
+ ? 0 \
+ : (UTF8_IS_INVARIANT(*s)) \
+ ? 1 \
+ : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
+ ? 0 \
+ : is_C9_STRICT_UTF8_CHAR_utf8_no_length_checks(s))
+
+/*
+
+=for apidoc Am|STRLEN|isUTF8_CHAR_flags|const U8 *s|const U8 *e| const U32 flags
+
+Evaluates to non-zero if the first few bytes of the string starting at C<s> and
+looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
+that represents some code point, subject to the restrictions given by C<flags>;
+otherwise it evaluates to 0. If non-zero, the value gives how many bytes
+starting at C<s> comprise the code point's representation. Any bytes remaining
+before C<e>, but beyond the ones needed to form the first code point in C<s>,
+are not examined.
+
+If C<flags> is 0, this gives the same results as C<L</isUTF8_CHAR>>;
+if C<flags> is C<UTF8_DISALLOW_ILLEGAL_INTERCHANGE>, this gives the same results
+as C<L</isSTRICT_UTF8_CHAR>>;
+and if C<flags> is C<UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE>, this gives
+the same results as C<L</isC9_STRICT_UTF8_CHAR>>.
+Otherwise C<flags> may be any combination of the C<UTF8_DISALLOW_I<foo>> flags
+understood by C<L</utf8n_to_uvchr>>, with the same meanings.
+
+The three alternative macros are for the most commonly needed validations; they
+are likely to run somewhat faster than this more general one, as they can be
+inlined into your code.
+
+Use L</is_utf8_string_flags>, L</is_utf8_string_loc_flags>, and
+L</is_utf8_string_loclen_flags> to check entire strings.
+
+=cut
+*/
+
+#define isUTF8_CHAR_flags(s, e, flags) \
+ (UNLIKELY((e) <= (s)) \
+ ? 0 \
+ : (UTF8_IS_INVARIANT(*s)) \
+ ? 1 \
+ : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
+ ? 0 \
+ : _is_utf8_char_helper(s, e, flags))
+
+/* Do not use; should be deprecated. Use isUTF8_CHAR() instead; this is
+ * retained solely for backwards compatibility */
+#define IS_UTF8_CHAR(p, n) (isUTF8_CHAR(p, (p) + (n)) == n)
+
+#endif /* PERL_UTF8_H_ */
+
+/*
+ * ex: set ts=8 sts=4 sw=4 et:
*/
https://perl5.git.perl.org / perl5.git / blobdiff