*
*/
-#ifndef H_UTF8 /* Guard against recursive inclusion */
-#define H_UTF8 1
+#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
#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_utf8_invariant_string>.
#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 uvoffuni_to_utf8_flags(d,uv,flags) \
+ uvoffuni_to_utf8_flags_msgs(d, uv, flags, 0)
#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)
+ uvchr_to_utf8_flags_msgs(d,uv,flags, 0)
+#define uvchr_to_utf8_flags_msgs(d,uv,flags,msgs) \
+ uvoffuni_to_utf8_flags_msgs(d,NATIVE_TO_UNI(uv),flags, msgs)
#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)
+ utf8_to_uvchr_buf_helper((const U8 *) (s), (const U8 *) e, lenp)
#define utf8n_to_uvchr(s, len, lenp, flags) \
utf8n_to_uvchr_error(s, len, lenp, flags, 0)
+#define utf8n_to_uvchr_error(s, len, lenp, flags, errors) \
+ utf8n_to_uvchr_msgs(s, len, lenp, flags, errors, 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_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))
-
#ifdef EBCDIC
/* The equivalent of these macros but implementing UTF-EBCDIC
are in the following header file:
#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 */
+/*
+
+=for apidoc AmnU|STRLEN|UTF8_MAXBYTES
+
+The maximum width of a single UTF-8 encoded character, 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.
+
+=cut
+ */
#define UTF8_MAXBYTES 13
#ifdef DOINIT
END_EXTERN_C
-#if defined(_MSC_VER) && _MSC_VER < 1400
-/* older MSVC versions have a smallish macro buffer */
-#define PERL_SMALL_MACRO_BUFFER
-#endif
+/*
-/* 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
+=for apidoc Am|U8|NATIVE_TO_LATIN1|U8 ch
+
+Returns the Latin-1 (including ASCII and control characters) equivalent of the
+input native code point given by C<ch>. Thus, C<NATIVE_TO_LATIN1(193)> on
+EBCDIC platforms returns 65. These each represent the character C<"A"> on
+their respective platforms. On ASCII platforms no conversion is needed, so
+this macro expands to just its input, adding no time nor space requirements to
+the implementation.
+
+For conversion of code points potentially larger than will fit in a character,
+use L</NATIVE_TO_UNI>.
+
+=for apidoc Am|U8|LATIN1_TO_NATIVE|U8 ch
+
+Returns the native equivalent of the input Latin-1 code point (including ASCII
+and control characters) given by C<ch>. Thus, C<LATIN1_TO_NATIVE(66)> on
+EBCDIC platforms returns 194. These each represent the character C<"B"> on
+their respective platforms. On ASCII platforms no conversion is needed, so
+this macro expands to just its input, adding no time nor space requirements to
+the implementation.
+
+For conversion of code points potentially larger than will fit in a character,
+use L</UNI_TO_NATIVE>.
+
+=for apidoc Am|UV|NATIVE_TO_UNI|UV ch
+
+Returns the Unicode equivalent of the input native code point given by C<ch>.
+Thus, C<NATIVE_TO_UNI(195)> on EBCDIC platforms returns 67. These each
+represent the character C<"C"> on their respective platforms. On ASCII
+platforms no conversion is needed, so this macro expands to just its input,
+adding no time nor space requirements to the implementation.
+
+=for apidoc Am|UV|UNI_TO_NATIVE|UV ch
+
+Returns the native equivalent of the input Unicode code point given by C<ch>.
+Thus, C<UNI_TO_NATIVE(68)> on EBCDIC platforms returns 196. These each
+represent the character C<"D"> on their respective platforms. On ASCII
+platforms no conversion is needed, so this macro expands to just its input,
+adding no time nor space requirements to the implementation.
+
+=cut
+*/
+
+#define NATIVE_TO_LATIN1(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) ((ch) | 0)))
+#define LATIN1_TO_NATIVE(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) ((ch) | 0)))
/* 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
* 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
+#define NATIVE_UTF8_TO_I8(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) ((ch) | 0)))
+#define I8_TO_NATIVE_UTF8(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) ((ch) | 0)))
-/* Transforms in wide UV chars */
-#define UNI_TO_NATIVE(ch) ((UV) (ch))
-#define NATIVE_TO_UNI(ch) ((UV) (ch))
+#define UNI_TO_NATIVE(ch) ((UV) ((ch) | 0))
+#define NATIVE_TO_UNI(ch) ((UV) ((ch) | 0))
/*
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.
-
-*/
-
-/* Anything larger than this will overflow the word if it were converted into a UV */
-#if defined(UV_IS_QUAD)
-# define HIGHEST_REPRESENTABLE_UTF8 "\xFF\x80\x8F\xBF\xBF\xBF\xBF\xBF\xBF\xBF\xBF\xBF\xBF"
-#else
-# define HIGHEST_REPRESENTABLE_UTF8 "\xFE\x83\xBF\xBF\xBF\xBF\xBF"
-#endif
-
-/* 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)
-
-/*
-=for apidoc Am|bool|UVCHR_IS_INVARIANT|UV cp
-
-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.
-
-=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) (((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) (((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) \
- ((((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) ((((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) (((U8)((c) | 0)) >= 0xc4)
+are in the character. */
/* 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
/* 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
+ * those to quickly exclude the vast majority of code points it will
* encounter */
-#define isUTF8_POSSIBLY_PROBLEMATIC(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
+#define isUTF8_POSSIBLY_PROBLEMATIC(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (U8) c >= 0xED)
- 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) ((1U << UTF_ACCUMULATION_SHIFT) - 1))
+/* 2**UTF_ACCUMULATION_SHIFT - 1. This masks out all but the bits that carry
+ * real information in a continuation byte. This turns out to be 0x3F in
+ * UTF-8, 0x1F in UTF-EBCDIC. */
+#define UTF_CONTINUATION_MASK ((U8) (nBIT_MASK(UTF_ACCUMULATION_SHIFT)))
+
+/* For use in UTF8_IS_CONTINUATION(). This turns out to be 0xC0 in UTF-8,
+ * E0 in UTF-EBCDIC */
+#define UTF_IS_CONTINUATION_MASK ((U8) (0xFF << UTF_ACCUMULATION_SHIFT))
+
+/* 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.
+ * This turns out to be 0x80 in UTF-8, 0xA0 in UTF-EBCDIC. (khw doesn't know
+ * the underlying reason that B0 works here) */
+#define UTF_CONTINUATION_MARK (UTF_IS_CONTINUATION_MASK & 0xB0)
+
+/* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
+ * first byte thereof? */
+#define UTF8_IS_CONTINUATION(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (((NATIVE_UTF8_TO_I8(c) & UTF_IS_CONTINUATION_MASK) \
+ == UTF_CONTINUATION_MARK)))
+
+/* Is the representation of the Unicode code point 'cp' the same regardless of
+ * being encoded in UTF-8 or not? This is a fundamental property of
+ * UTF-8,EBCDIC */
+#define OFFUNI_IS_INVARIANT(c) (((WIDEST_UTYPE)(c)) < UTF_CONTINUATION_MARK)
+
+/*
+=for apidoc Am|bool|UVCHR_IS_INVARIANT|UV cp
+
+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.
+
+=cut
+ */
+#define UVCHR_IS_INVARIANT(cp) (OFFUNI_IS_INVARIANT(NATIVE_TO_UNI(cp)))
/* Internal macro to be used only in this file to aid in constructing other
* publicly accessible macros.
* 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)
+ LIKELY((UV) (uv) < ((UV) 1U << (6 * UTF_ACCUMULATION_SHIFT))) \
+ ? 7 \
+ : UTF8_MAXBYTES)
#else
# define __BASE_UNI_SKIP(uv) (__COMMON_UNI_SKIP(uv) 7)
#endif
*/
#define UVCHR_SKIP(uv) ( UVCHR_IS_INVARIANT(uv) ? 1 : __BASE_UNI_SKIP(uv))
+#define UTF_MIN_START_BYTE \
+ ((UTF_CONTINUATION_MARK >> UTF_ACCUMULATION_SHIFT) | UTF_START_MARK(2))
+
+/* Is the byte 'c' the first byte of a multi-byte UTF8-8 encoded sequence?
+ * This excludes invariants (they are single-byte). It also excludes the
+ * illegal overlong sequences that begin with C0 and C1 on ASCII platforms, and
+ * C0-C4 I8 start bytes on EBCDIC ones. On EBCDIC E0 can't start a
+ * non-overlong sequence, so we define a base macro and for those platforms,
+ * extend it to also exclude E0 */
+#define UTF8_IS_START_base(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (NATIVE_UTF8_TO_I8(c) >= UTF_MIN_START_BYTE))
+#ifdef EBCDIC
+# define UTF8_IS_START(c) \
+ (UTF8_IS_START_base(c) && (c) != I8_TO_NATIVE_UTF8(0xE0))
+#else
+# define UTF8_IS_START(c) UTF8_IS_START_base(c)
+#endif
+
+#define UTF_MIN_ABOVE_LATIN1_BYTE \
+ ((0x100 >> UTF_ACCUMULATION_SHIFT) | UTF_START_MARK(2))
+
+/* 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) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (NATIVE_UTF8_TO_I8(c) >= UTF_MIN_ABOVE_LATIN1_BYTE))
+
+/* 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. */
+#define UTF8_IS_DOWNGRADEABLE_START(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ inRANGE(NATIVE_UTF8_TO_I8(c), \
+ UTF_MIN_START_BYTE, UTF_MIN_ABOVE_LATIN1_BYTE - 1))
+
/* 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 */
/* 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 MAX_PORTABLE_UTF8_TWO_BYTE (32 * nBIT_UMAX(5))
+
+/*
+
+=for apidoc AmnU|STRLEN|UTF8_MAXBYTES_CASE
+
+The maximum number of UTF-8 bytes a single Unicode character can
+uppercase/lowercase/titlecase/fold into.
+
+=cut
+
+ * 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)
+ *
+=cut
+*/
#define UTF8_MAXBYTES_CASE \
- (UTF8_MAXBYTES >= (UTF8_MAX_FOLD_CHAR_EXPAND * OFFUNISKIP(0x10FFFF)) \
- ? UTF8_MAXBYTES \
- : (UTF8_MAX_FOLD_CHAR_EXPAND * OFFUNISKIP(0x10FFFF)))
+ MAX(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
* 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)))))
+#define UTF_START_MARK(len) (UNLIKELY((len) > 7) \
+ ? 0xFF \
+ : ((U8) (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. '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)))
+#define UTF_START_MASK(len) (UNLIKELY((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
* 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)) \
+#define UTF8_ACCUMULATE(old, new) (__ASSERT_(FITS_IN_8_BITS(new)) \
+ ((old) << UTF_ACCUMULATION_SHIFT) \
+ | ((NATIVE_UTF8_TO_I8(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)))
+#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)
* Note that the result can be larger than 255 if the input character is not
* downgradable */
#define TWO_BYTE_UTF8_TO_NATIVE(HI, LO) \
- ( __ASSERT_(PL_utf8skip[HI] == 2) \
- __ASSERT_(UTF8_IS_CONTINUATION(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))))
/*
=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>.
+returns the number of bytes a non-malformed UTF-8 encoded character whose first
+(perhaps only) byte is pointed to by C<s>.
+
+If there is a possibility of malformed input, use instead:
+
+=over
+
+=item C<L</UTF8_SAFE_SKIP>> if you know the maximum ending pointer in the
+buffer pointed to by C<s>; or
+
+=item C<L</UTF8_CHK_SKIP>> if you don't know it.
+
+=back
+
+It is better to restructure your code so the end pointer is passed down so that
+you know what it actually is at the point of this call, but if that isn't
+possible, C<L</UTF8_CHK_SKIP>> can minimize the chance of accessing beyond the end
+of the input buffer.
=cut
*/
#define UTF8SKIP(s) PL_utf8skip[*(const U8*)(s)]
+
+/*
+=for apidoc Am|STRLEN|UTF8_SKIP|char* s
+This is a synonym for C<L</UTF8SKIP>>
+
+=cut
+*/
+
#define UTF8_SKIP(s) UTF8SKIP(s)
+/*
+=for apidoc Am|STRLEN|UTF8_CHK_SKIP|char* s
+
+This is a safer version of C<L</UTF8SKIP>>, but still not as safe as
+C<L</UTF8_SAFE_SKIP>>. This version doesn't blindly assume that the input
+string pointed to by C<s> is well-formed, but verifies that there isn't a NUL
+terminating character before the expected end of the next character in C<s>.
+The length C<UTF8_CHK_SKIP> returns stops just before any such NUL.
+
+Perl tends to add NULs, as an insurance policy, after the end of strings in
+SV's, so it is likely that using this macro will prevent inadvertent reading
+beyond the end of the input buffer, even if it is malformed UTF-8.
+
+This macro is intended to be used by XS modules where the inputs could be
+malformed, and it isn't feasible to restructure to use the safer
+C<L</UTF8_SAFE_SKIP>>, for example when interfacing with a C library.
+
+=cut
+*/
+
+#define UTF8_CHK_SKIP(s) \
+ (UNLIKELY(s[0] == '\0') ? 1 : MIN(UTF8SKIP(s), \
+ my_strnlen((char *) (s), UTF8SKIP(s))))
+/*
+
+=for apidoc Am|STRLEN|UTF8_SAFE_SKIP|char* s|char* e
+returns 0 if S<C<s E<gt>= e>>; otherwise returns the number of bytes in the
+UTF-8 encoded character whose first byte is pointed to by C<s>. But it never
+returns beyond C<e>. On DEBUGGING builds, it asserts that S<C<s E<lt>= e>>.
+
+=cut
+ */
+#define UTF8_SAFE_SKIP(s, e) (__ASSERT_((e) >= (s)) \
+ UNLIKELY(((e) - (s)) <= 0) \
+ ? 0 \
+ : MIN(((e) - (s)), UTF8_SKIP(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)
* above show, doesn't matter as to its implementation */
#define NATIVE_BYTE_IS_INVARIANT(c) UVCHR_IS_INVARIANT(c)
+/* Misleadingly named: 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) (__ASSERT_(FITS_IN_8_BITS(c)) \
+ (! UTF8_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. */
* 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) \
- _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 isALNUM_lazy_if_safe(p, e, UTF) isWORDCHAR_lazy_if_safe(p, e, UTF)
#define UTF8_MAXLEN UTF8_MAXBYTES
#define UTF8_ALLOW_CONTINUATION 0x0002
#define UTF8_GOT_CONTINUATION UTF8_ALLOW_CONTINUATION
-/* Unexpected continuation byte */
+/* Unexpected non-continuation byte */
#define UTF8_ALLOW_NON_CONTINUATION 0x0004
#define UTF8_GOT_NON_CONTINUATION UTF8_ALLOW_NON_CONTINUATION
#define UTF8_GOT_SURROGATE UTF8_DISALLOW_SURROGATE
#define UTF8_WARN_SURROGATE 0x0200
-#define UTF8_DISALLOW_NONCHAR 0x0400 /* Unicode non-character */
+/* Unicode non-character code points */
+#define UTF8_DISALLOW_NONCHAR 0x0400
#define UTF8_GOT_NONCHAR UTF8_DISALLOW_NONCHAR
-#define UTF8_WARN_NONCHAR 0x0800 /* code points */
+#define UTF8_WARN_NONCHAR 0x0800
-#define UTF8_DISALLOW_SUPER 0x1000 /* Super-set of Unicode: code */
+/* 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 /* points above the legal max */
-
-/* Code points which never were part of the original UTF-8 standard, which only
- * went up to 2 ** 31 - 1. Note that these all overflow a signed 32-bit word,
- * The first byte of these code points is FE or FF on ASCII platforms. If the
- * first byte is FF, it will overflow a 32-bit word. */
-#define UTF8_DISALLOW_ABOVE_31_BIT 0x4000
-#define UTF8_GOT_ABOVE_31_BIT UTF8_DISALLOW_ABOVE_31_BIT
-#define UTF8_WARN_ABOVE_31_BIT 0x8000
-
-/* For back compat, these old names are misleading for UTF_EBCDIC */
-#define UTF8_DISALLOW_FE_FF UTF8_DISALLOW_ABOVE_31_BIT
-#define UTF8_WARN_FE_FF UTF8_WARN_ABOVE_31_BIT
+#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 */
* includes what they used to mean. The first one's meaning was to allow the
* just the single non-character 0xFFFF */
#define UTF8_ALLOW_FFFF 0
+#define UTF8_ALLOW_FE_FF 0
#define UTF8_ALLOW_SURROGATE 0
/* C9 refers to Unicode Corrigendum #9: allows but discourages non-chars */
|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. */
+ * not any malformed. This is the default. */
#define UTF8_ALLOW_ANYUV 0
#define UTF8_ALLOW_DEFAULT UTF8_ALLOW_ANYUV
#define UTF8_IS_REPLACEMENT(s, send) is_REPLACEMENT_utf8_safe(s,send)
+#define MAX_LEGAL_CP ((UV)IV_MAX)
+
/*
=for apidoc Am|bool|UTF8_IS_SUPER|const U8 *s|const U8 *e
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
+Note that it is illegal to have code points that are larger than what can
fit in an IV on the current machine.
=cut
*/
#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)
+ (( ((e) > (s) + 4) \
+ && (NATIVE_UTF8_TO_I8(*(s)) >= 0xF9) \
+ && UNLIKELY( 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, e) \
- (( LIKELY((e) > (s) + 3) \
+ (( ((e) > (s) + 3) \
&& (*(U8*) (s)) >= 0xF4 \
- && ((*(U8*) (s)) > 0xF4 || (*((U8*) (s) + 1) >= 0x90))\
+ && (UNLIKELY( ((*(U8*) (s)) > 0xF4) \
+ || (*((U8*) (s) + 1) >= 0x90))) \
&& LIKELY((s) + UTF8SKIP(s) <= (e))) \
- ? _is_utf8_char_helper(s, s + UTF8SKIP(s), 0) : 0)
+ ? is_utf8_char_helper(s, s + UTF8SKIP(s), 0) : 0)
#endif
/* These are now machine generated, and the 'given' clause is no longer
non-zero, the value gives how many bytes starting at C<s> comprise the code
point's representation.
+=for apidoc AmnU|UV|UNICODE_REPLACEMENT
+
+Evaluates to 0xFFFD, the code point of the Unicode REPLACEMENT CHARACTER
+
=cut
*/
#define UTF8_IS_NONCHAR(s, e) \
* 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_ABOVE_31_BIT 0x0008 /* Above 0x7FFF_FFFF */
-#define UNICODE_DISALLOW_SURROGATE 0x0010
-#define UNICODE_DISALLOW_NONCHAR 0x0020
-#define UNICODE_DISALLOW_SUPER 0x0040
-#define UNICODE_DISALLOW_ABOVE_31_BIT 0x0080
+#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
+
+#ifdef PERL_CORE
+# define UNICODE_ALLOW_ABOVE_IV_MAX 0x0100
+#endif
+#define UNICODE_DISALLOW_ABOVE_31_BIT UNICODE_DISALLOW_PERL_EXTENDED
+
+#define UNICODE_GOT_SURROGATE UNICODE_DISALLOW_SURROGATE
+#define UNICODE_GOT_NONCHAR UNICODE_DISALLOW_NONCHAR
+#define UNICODE_GOT_SUPER UNICODE_DISALLOW_SUPER
+#define UNICODE_GOT_PERL_EXTENDED UNICODE_DISALLOW_PERL_EXTENDED
+
#define UNICODE_WARN_ILLEGAL_C9_INTERCHANGE \
(UNICODE_WARN_SURROGATE|UNICODE_WARN_SUPER)
#define UNICODE_WARN_ILLEGAL_INTERCHANGE \
/* 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)) \
+#define UNICODE_IS_SURROGATE(uv) UNLIKELY(((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)
+#define UNICODE_IS_REPLACEMENT(uv) UNLIKELY((UV) (uv) == UNICODE_REPLACEMENT)
+#define UNICODE_IS_BYTE_ORDER_MARK(uv) UNLIKELY((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)
+#define UNICODE_IS_32_CONTIGUOUS_NONCHARS(uv) UNLIKELY((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)
+ UNLIKELY(((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 UNICODE_IS_ABOVE_31_BIT(uv) ((UV) (uv) > 0x7FFFFFFF)
+#define UNICODE_IS_SUPER(uv) UNLIKELY((UV) (uv) > PERL_UNICODE_MAX)
#define LATIN_SMALL_LETTER_SHARP_S LATIN_SMALL_LETTER_SHARP_S_NATIVE
#define LATIN_SMALL_LETTER_Y_WITH_DIAERESIS \
#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)
+#define UNI_DISPLAY_BACKSPACE 0x0004 /* Allow \b when also
+ UNI_DISPLAY_BACKSLASH */
+#define UNI_DISPLAY_QQ (UNI_DISPLAY_ISPRINT \
+ |UNI_DISPLAY_BACKSLASH \
+ |UNI_DISPLAY_BACKSPACE)
-#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) && \
- isALPHA_FOLD_EQ((input)[0], 's'))
+/* Character classes could also allow \b, but not patterns in general */
+#define UNI_DISPLAY_REGEX (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
+/* Should be removed; maybe deprecated, but not used in CPAN */
#define SHARP_S_SKIP 2
-/*
-
-=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)
-
-/*
-
-=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))
+#define bytes_from_utf8(s, lenp, is_utf8p) \
+ bytes_from_utf8_loc(s, lenp, is_utf8p, 0)
/*
? 1 \
: UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
? 0 \
- : _is_utf8_char_helper(s, e, flags))
+ : 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 /* H_UTF8 */
+#endif /* PERL_UTF8_H_ */
/*
* ex: set ts=8 sts=4 sw=4 et:
https://perl5.git.perl.org / perl5.git / blobdiff