/* 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 H_UTF8 /* Guard against recursive inclusion */
+#define H_UTF8 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_NOMIX_ASCII 0x4
+#define FOLD_FLAGS_LOCALE 0x1
+#define FOLD_FLAGS_FULL 0x2
+#define FOLD_FLAGS_NOMIX_ASCII 0x4
/* 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
+/*
+=head1 Unicode Support
+L<perlguts/Unicode Support> has an introduction to this API.
+
+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 is_ascii_string
+
+This is a misleadingly-named synonym for L</is_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_invariant_string> is preferred.
+
+=cut
+*/
+#define is_ascii_string(s, len) is_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 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)
-#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)
+#define to_utf8_fold(c, p, lenp) _to_utf8_fold_flags(c, p, lenp, FOLD_FLAGS_FULL)
+#define to_utf8_lower(a,b,c) _to_utf8_lower_flags(a,b,c,0)
+#define to_utf8_upper(a,b,c) _to_utf8_upper_flags(a,b,c,0)
+#define to_utf8_title(a,b,c) _to_utf8_title_flags(a,b,c,0)
/* 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)
#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_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)
-/*
-=for apidoc ibcmp_utf8
-
-This is a synonym for (! foldEQ_utf8())
-
-=cut
-*/
#define ibcmp_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
cBOOL(! foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2))
/* 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 */
END_EXTERN_C
-#include "unicode_constants.h"
+/* Native character to/from iso-8859-1. Are the identity functions on ASCII
+ * platforms */
+#define NATIVE_TO_LATIN1(ch) (ch)
+#define LATIN1_TO_NATIVE(ch) (ch)
+
+/* 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 */
+#define NATIVE_UTF8_TO_I8(ch) (ch)
+#define I8_TO_NATIVE_UTF8(ch) (ch)
-/* 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 NATIVE_TO_I8(ch) NATIVE_TO_UTF(ch) /* a clearer synonym */
-#define UTF_TO_NATIVE(ch) (ch)
-#define I8_TO_NATIVE(ch) UTF_TO_NATIVE(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)
-
-/* 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
/*
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+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
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+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
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'.
*/
*/
+/* Is the representation of the Unicode code point 'c' the same regardless of
+ * being encoded in UTF-8 or not? */
#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)
-/* 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)
+/* Is the UTF8-encoded byte 'c' part of a variant sequence in UTF-8? This is
+ * the inverse of UTF8_IS_INVARIANT */
+#define UTF8_IS_CONTINUED(c) (((U8)c) & 0x80)
-#define UTF_START_MARK(len) (((len) > 7) ? 0xFF : (0xFE << (7-(len))))
+/* 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. */
+#define UTF8_IS_START(c) (((U8)c) >= 0xc2)
+
+/* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
+ * first byte thereof? */
+#define UTF8_IS_CONTINUATION(c) ((((U8)c) & 0xC0) == 0x80)
+
+/* 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]. */
+#define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)(c) & 0xfe) == 0xc2)
+
+/* Is the UTF8-encoded byte 'c' the first byte of a sequence of bytes that
+ * represent a code point > 255? */
+#define UTF8_IS_ABOVE_LATIN1(c) ((U8)(c) >= 0xc4)
+
+/* This defines the 1-bits that are to be in the first byte of a multi-byte
+ * UTF-8 encoded character that 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)))))
/* Masks out the initial one bits in a start byte, leaving the real data ones.
- * Doesn't work on an invariant byte */
+ * 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)))
+/* This defines the bits that are to be in the continuation bytes of a multi-byte
+ * UTF-8 encoded character that indicate it is a continuation byte. */
#define UTF_CONTINUATION_MARK 0x80
+
+/* 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
/* 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 */
+/* If a value is anded with this, 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))
+ (((UV) UTF_CONTINUATION_MASK) << ((sizeof(UV) * CHARBITS) \
+ - UTF_ACCUMULATION_SHIFT))
+
+#if UVSIZE >= 8
+# define UTF8_QUAD_MAX UINT64_C(0x1000000000)
-#ifdef HAS_QUAD
-#define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
+/* Input is a true Unicode (not-native) code point */
+#define OFFUNISKIP(uv) ( (uv) < 0x80 ? 1 : \
(uv) < 0x800 ? 2 : \
(uv) < 0x10000 ? 3 : \
(uv) < 0x200000 ? 4 : \
(uv) < UTF8_QUAD_MAX ? 7 : 13 )
#else
/* No, I'm not even going to *TRY* putting #ifdef inside a #define */
-#define UNISKIP(uv) ( (uv) < 0x80 ? 1 : \
+#define OFFUNISKIP(uv) ( (uv) < 0x80 ? 1 : \
(uv) < 0x800 ? 2 : \
(uv) < 0x10000 ? 3 : \
(uv) < 0x200000 ? 4 : \
(uv) < 0x80000000 ? 6 : 7 )
#endif
+/* 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
+
+/* The maximum number of UTF-8 bytes a single Unicode character can
+ * uppercase/lowercase/fold into. Unicode guarantees that the maximum
+ * expansion is 3 characters. On ASCIIish platforms, the highest Unicode
+ * character occupies 4 bytes, therefore this number would be 12, but this is
+ * smaller than the maximum width a single above-Unicode character can occupy,
+ * so use that instead */
+#if UTF8_MAXBYTES < 12
+#error UTF8_MAXBYTES must be at least 12
+#endif
+
+/* ^? is defined to be DEL on ASCII systems. See the definition of toCTRL()
+ * for more */
+#define QUESTION_MARK_CTRL DEL_NATIVE
+
+#define MAX_UTF8_TWO_BYTE 0x7FF
+
+#define UTF8_MAXBYTES_CASE UTF8_MAXBYTES
+
#endif /* EBCDIC vs ASCII */
/* Rest of these are attributes of Unicode and perl's internals rather than the
* 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)
+
+/* 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) (((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) OFFUNISKIP(NATIVE_TO_UNI(uv))
+
+/*
-#define UTF8_ACCUMULATE(old, new) (((old) << UTF_ACCUMULATION_SHIFT) | (((U8)new) & UTF_CONTINUATION_MASK))
+=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
+ */
-/* Convert a two (not one) byte utf8 character to a unicode code point value.
+/* 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, and OFFUNISKIP */
+#define UNISKIP(uv) NATIVE_SKIP(uv)
+#define UVCHR_SKIP(uv) NATIVE_SKIP(uv)
+
+/* Longer, but more accurate name */
+#define UTF8_IS_ABOVE_LATIN1_START(c) UTF8_IS_ABOVE_LATIN1(c)
+
+/* 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_(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>.
+
+=cut
+ */
+#define UTF8SKIP(s) PL_utf8skip[*(const U8*)(s)]
+#define UTF8_SKIP(s) UTF8SKIP(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) NATIVE_TO_I8(((c) \
- >> UTF_ACCUMULATION_SHIFT) | (0xFF & UTF_START_MARK(2)))
-#define UTF8_TWO_BYTE_LO_nocast(c) NATIVE_TO_I8(((c) & UTF_CONTINUATION_MASK) \
- | UTF_CONTINUATION_MARK)
+/* Is the byte 'c' the same character when encoded in UTF-8 as when not. This
+ * works on both UTF-8 encoded strings and non-encoded, as it returns TRUE in
+ * each for the exact same set of bit patterns. (And it works on any byte in a
+ * UTF-8 encoded string) */
+#define UTF8_IS_INVARIANT(c) UNI_IS_INVARIANT(NATIVE_UTF8_TO_I8(c))
-#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)))
+/* Like the above, but its name implies a non-UTF8 input, and is implemented
+ * differently (for no particular reason) */
+#define NATIVE_BYTE_IS_INVARIANT(c) UNI_IS_INVARIANT(NATIVE_TO_LATIN1(c))
-/* 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))
+/* Like the above, but accepts any UV as input */
+#define UVCHR_IS_INVARIANT(uv) UNI_IS_INVARIANT(NATIVE_TO_UNI(uv))
+
+#define MAX_PORTABLE_UTF8_TWO_BYTE 0x3FF /* constrained by EBCDIC */
+
+/* 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) \
+ I8_TO_NATIVE_UTF8((translate_function(c) >> UTF_ACCUMULATION_SHIFT) \
+ | UTF_START_MARK(2))
+#define __BASE_TWO_BYTE_LO(c, translate_function) \
+ 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)) \
+ ((U8) __BASE_TWO_BYTE_HI(c, NATIVE_TO_LATIN1)))
+#define UTF8_EIGHT_BIT_LO(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ ((U8) __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)) \
+ ((U8) __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)) \
+ ((U8) __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
#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 isWORDCHAR_lazy_if(p,UTF) ((IN_BYTES || (!UTF )) \
+ ? isWORDCHAR(*(p)) \
+ : isWORDCHAR_utf8((const U8*)p))
+#define isALNUM_lazy_if(p,UTF) isWORDCHAR_lazy_if(p,UTF)
-#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)
+
+/*
+
+=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_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. */
+ * byte of which is a FE or FF on ASCII platforms. If the first byte is FF, it
+ * will overflow a 32-bit word. If the first byte is FE, it will overflow a
+ * signed 32-bit word. */
#define UTF8_DISALLOW_FE_FF 0x0800
#define UTF8_WARN_FE_FF 0x1000
* 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
+#define UTF8_FIRST_PROBLEMATIC_CODE_POINT_FIRST_BYTE \
+ FIRST_SURROGATE_UTF8_FIRST_BYTE
-/* 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
- */
-#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)))
- /* <send> points to one beyond the end of the string that starts at <s> */
-# 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
+/* Several of the macros below have a second parameter that is currently
+ * unused; but could be used in the future to make sure that the input is
+ * well-formed. */
+
+#define UTF8_IS_SURROGATE(s, e) cBOOL(is_SURROGATE_utf8(s))
+#define UTF8_IS_REPLACEMENT(s, send) cBOOL(is_REPLACEMENT_utf8_safe(s,send))
/* 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
- */
+ *
+ * BE AWARE that this test doesn't rule out malformed code points, in
+ * particular overlongs */
#ifdef EBCDIC /* Both versions assume well-formed UTF8 */
-# define UTF8_IS_SUPER(s) (NATIVE_TO_I8(*(s)) >= 0xF9 \
- && (NATIVE_TO_I8(*(s)) > 0xF9) || (NATIVE_TO_I8(*((s)) + 1 >= 0xA2)))
+# define UTF8_IS_SUPER(s, e) (NATIVE_UTF8_TO_I8(* (U8*) (s)) >= 0xF9 \
+ && (NATIVE_UTF8_TO_I8(* (U8*) (s)) > 0xF9 \
+ || (NATIVE_UTF8_TO_I8(* ((U8*) (s) + 1)) >= 0xA2)))
#else
-# define UTF8_IS_SUPER(s) (*(s) >= 0xF4 \
- && (*(s) > 0xF4 || (*((s) + 1) >= 0x90)))
+# define UTF8_IS_SUPER(s, e) (*(U8*) (s) >= 0xF4 \
+ && (*(U8*) (s) > 0xF4 || (*((U8*) (s) + 1) >= 0x90)))
#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
- */
-#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
+/* 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(s))
+#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
#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_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)
/* For backward source compatibility, as are now the default */
#define UNICODE_ALLOW_SURROGATE 0
#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
-
#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 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 LATIN_SMALL_LETTER_LONG_S 0x017F
-#define KELVIN_SIGN 0x212A
-#define ANGSTROM_SIGN 0x212B
+#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
(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)
+/* A helper macro for isUTF8_CHAR, so use that one, and not this one. This is
+ * retained solely for backwards compatibility and may be deprecated and
+ * removed in a future Perl version.
+ *
+ * regen/regcharclass.pl generates is_UTF8_CHAR_utf8() macros for up to these
+ * number of bytes. So this has to be coordinated with that file */
+#ifdef EBCDIC
+# define IS_UTF8_CHAR_FAST(n) ((n) <= 3)
+#else
# define IS_UTF8_CHAR_FAST(n) ((n) <= 4)
+#endif
+
+#ifndef EBCDIC
+/* A helper macro for isUTF8_CHAR, so use that one instead of this. This was
+ * generated by regen/regcharclass.pl, and then moved here. The lines that
+ * generated it were then commented out. This was done solely because it takes
+ * on the order of 10 minutes to generate, and is never going to change, unless
+ * the generated code is improved.
+ *
+ * The EBCDIC versions have been cut to not cover all of legal Unicode,
+ * otherwise they take too long to generate; besides there is a separate one
+ * for each code page, so they are in regcharclass.h instead of here */
+/*
+ UTF8_CHAR: Matches legal UTF-8 encoded characters from 2 through 4 bytes
+
+ 0x80 - 0x1FFFFF
+*/
+/*** GENERATED CODE ***/
+#define is_UTF8_CHAR_utf8_no_length_checks(s) \
+( ( 0xC2 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xDF ) ? \
+ ( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
+: ( 0xE0 == ((U8*)s)[0] ) ? \
+ ( ( ( ( ((U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xE1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xEF ) ? \
+ ( ( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
+: ( 0xF0 == ((U8*)s)[0] ) ? \
+ ( ( ( ( 0x90 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xBF ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
+: ( ( ( ( 0xF1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xF7 ) && ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )
+#endif
+
+/*
+
+=for apidoc Am|STRLEN|isUTF8_CHAR|const U8 *s|const U8 *e
-#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 */
+Returns the number of bytes beginning at C<s> which form a legal UTF-8 (or
+UTF-EBCDIC) encoded character, looking no further than S<C<e - s>> bytes into
+C<s>. Returns 0 if the sequence starting at C<s> through S<C<e - 1>> is not
+well-formed UTF-8.
+
+Note that an 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(IS_UTF8_CHAR_FAST(UTF8SKIP(s))) \
+ ? is_UTF8_CHAR_utf8_no_length_checks(s) \
+ : _is_utf8_char_slow(s, e))
+
+#define is_utf8_char_buf(buf, buf_end) isUTF8_CHAR(buf, buf_end)
+
+/* 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 /* H_UTF8 */
/*
- * Local variables:
- * c-indentation-style: bsd
- * c-basic-offset: 4
- * indent-tabs-mode: nil
- * End:
- *
* ex: set ts=8 sts=4 sw=4 et:
*/
https://perl5.git.perl.org / perl5.git / blobdiff