/* utf8.h
*
- * Copyright (C) 2000, 2001, 2002, 2005, 2006, 2007, 2009 by Larry Wall and others
+ * Copyright (C) 2000, 2001, 2002, 2005, 2006, 2007, 2009,
+ * 2010, 2011 by Larry Wall and others
*
* You may distribute under the terms of either the GNU General Public
* License or the Artistic License, as specified in the README file.
# define USE_UTF8_IN_NAMES (PL_hints & HINT_UTF8)
#endif
+/* For to_utf8_fold_flags, q.v. */
+#define FOLD_FLAGS_LOCALE 0x1
+#define FOLD_FLAGS_FULL 0x2
+
+#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)
+
/* 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_S1_ALREADY_FOLDED (1 << 2)
+#define FOLDEQ_S2_ALREADY_FOLDED (1 << 3)
+
/*
=for apidoc ibcmp_utf8
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus */
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* scripts */
3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6, /* cjk etc. */
-7,13, /* Perl extended (not UTF-8). Up to 72bit allowed (64-bit + reserved). */
+7,13, /* Perl extended (not official UTF-8). Up to 72bit allowed (64-bit +
+ reserved). */
};
#else
EXTCONST unsigned char PL_utf8skip[];
#define UTF8_EIGHT_BIT_LO(c) UTF8_TWO_BYTE_LO((U8)(c))
/*
- * Note: we try to be careful never to call the isXXX_utf8() functions
- * unless we're pretty sure we've seen the beginning of a UTF-8 or UTFEBCDIC
- * character. Otherwise we risk loading in the heavy-duty swash_init and
- * swash_fetch routines unnecessarily.
+ * 'UTF' is whether or not p is encoded in UTF8. The names 'foo_lazy_if' stem
+ * from an earlier version of these macros in which they didn't call the
+ * foo_utf8() macros (i.e. were 'lazy') unless they decided that *p is the
+ * beginning of a utf8 character. Now that foo_utf8() determines that itself,
+ * no need to do it again here
*/
-#define isIDFIRST_lazy_if(p,c) ((IN_BYTES || (!c || ! UTF8_IS_START(*((const U8*)p)))) \
- ? isIDFIRST(*(p)) \
- : isIDFIRST_utf8((const U8*)p))
-#define isALNUM_lazy_if(p,c) ((IN_BYTES || (!c || ! UTF8_IS_START(*((const U8*)p)))) \
- ? isALNUM(*(p)) \
- : isALNUM_utf8((const U8*)p))
+#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(p) isIDFIRST_lazy_if(p,1)
#define isALNUM_lazy(p) isALNUM_lazy_if(p,1)
* 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. */
-#define UTF8_MAXBYTES_CASE 6
+ * 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 DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTES)
-#define IN_UNI_8_BIT ( (CopHINTS_get(PL_curcop) & HINT_UNI_8_BIT) \
- && ! IN_LOCALE_RUNTIME && ! IN_BYTES)
+#define IN_UNI_8_BIT \
+ (CopHINTS_get(PL_curcop) & (HINT_UNI_8_BIT|HINT_LOCALE_NOT_CHARS) \
+ && ! IN_LOCALE_RUNTIME && ! IN_BYTES)
+
#define UTF8_ALLOW_EMPTY 0x0001 /* Allow a zero length string */
/* Allow second... bytes to be non-continuation bytes */
#define UTF8_ALLOW_NON_CONTINUATION 0x0004
-#define UTF8_ALLOW_FE_FF 0x0008 /* Allow FE or FF start bytes, \
- yields above 0x7fffFFFF = 31 bits */
-#define UTF8_ALLOW_SHORT 0x0010 /* expecting more bytes */
-#define UTF8_ALLOW_SURROGATE 0x0020
-#define UTF8_ALLOW_FFFF 0x0040 /* Allow UNICODE_ILLEGAL */
-#define UTF8_ALLOW_LONG 0x0080 /* expecting fewer bytes */
-#define UTF8_ALLOW_ANYUV (UTF8_ALLOW_EMPTY|UTF8_ALLOW_FE_FF|\
- UTF8_ALLOW_SURROGATE|UTF8_ALLOW_FFFF)
-#define UTF8_ALLOW_ANY 0x00FF
-#define UTF8_CHECK_ONLY 0x0200
+
+/* expecting more bytes than were available in the string */
+#define UTF8_ALLOW_SHORT 0x0008
+
+/* Overlong sequence; i.e., the code point can be specified in fewer bytes. */
+#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
+
+/* 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
+ * just the single non-character 0xFFFF */
+#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)
+#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)
*/
#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))
+ && ((*((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 UNICODE_SURROGATE_LAST 0xDFFF
#define UNICODE_REPLACEMENT 0xFFFD
#define UNICODE_BYTE_ORDER_MARK 0xFEFF
-#define UNICODE_ILLEGAL 0xFFFF
/* Though our UTF-8 encoding can go beyond this,
* let's be conservative and do as Unicode says. */
#define PERL_UNICODE_MAX 0x10FFFF
-#define UNICODE_ALLOW_SURROGATE 0x0001 /* Allow UTF-16 surrogates (EVIL) */
-#define UNICODE_ALLOW_FDD0 0x0002 /* Allow the U+FDD0...U+FDEF */
-#define UNICODE_ALLOW_FFFF 0x0004 /* Allow U+FFF[EF], U+1FFF[EF], ... */
-#define UNICODE_ALLOW_SUPER 0x0008 /* Allow past 0x10FFFF */
-#define UNICODE_ALLOW_ANY 0x000F
+#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)
+
+/* 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_ILLEGAL(c) ((c) == UNICODE_ILLEGAL)
#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 \
#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 UNI_DISPLAY_ISPRINT 0x0001
#define UNI_DISPLAY_BACKSLASH 0x0002
# 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_FLAGS(node) & ANYOF_NONBITMAP) && \
+ (ANYOF_NONBITMAP(node)) && \
(ANYOF_FLAGS(node) & ANYOF_LOC_NONBITMAP_FOLD) && \
((end) > (input) + 1) && \
toLOWER((input)[0]) == 's' && \
toLOWER((input)[1]) == 's')
#define SHARP_S_SKIP 2
-#ifdef EBCDIC
-/* IS_UTF8_CHAR() is not ported to EBCDIC */
-#else
-#define IS_UTF8_CHAR_1(p) \
+#ifndef EBCDIC
+# define IS_UTF8_CHAR_1(p) \
((p)[0] <= 0x7F)
-#define IS_UTF8_CHAR_2(p) \
+# define IS_UTF8_CHAR_2(p) \
((p)[0] >= 0xC2 && (p)[0] <= 0xDF && \
(p)[1] >= 0x80 && (p)[1] <= 0xBF)
-#define IS_UTF8_CHAR_3a(p) \
+# 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) \
+# define IS_UTF8_CHAR_3b(p) \
((p)[0] >= 0xE1 && (p)[0] <= 0xEC && \
(p)[1] >= 0x80 && (p)[1] <= 0xBF && \
(p)[2] >= 0x80 && (p)[2] <= 0xBF)
-#define IS_UTF8_CHAR_3c(p) \
+# define IS_UTF8_CHAR_3c(p) \
((p)[0] == 0xED && \
(p)[1] >= 0x80 && (p)[1] <= 0xBF && \
(p)[2] >= 0x80 && (p)[2] <= 0xBF)
-/* In IS_UTF8_CHAR_3c(p) one could use
- * (p)[1] >= 0x80 && (p)[1] <= 0x9F
- * if one wanted to exclude surrogates. */
-#define IS_UTF8_CHAR_3d(p) \
+ /* In IS_UTF8_CHAR_3c(p) one could use
+ * (p)[1] >= 0x80 && (p)[1] <= 0x9F
+ * if one wanted to exclude surrogates. */
+# define IS_UTF8_CHAR_3d(p) \
((p)[0] >= 0xEE && (p)[0] <= 0xEF && \
(p)[1] >= 0x80 && (p)[1] <= 0xBF && \
(p)[2] >= 0x80 && (p)[2] <= 0xBF)
-#define IS_UTF8_CHAR_4a(p) \
+# 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)
-#define IS_UTF8_CHAR_4b(p) \
+# define IS_UTF8_CHAR_4b(p) \
((p)[0] >= 0xF1 && (p)[0] <= 0xF3 && \
(p)[1] >= 0x80 && (p)[1] <= 0xBF && \
(p)[2] >= 0x80 && (p)[2] <= 0xBF && \
* since that is not needed (and that would not be strict
* UTF-8, anyway). The "slow path" in Perl_is_utf8_char()
* will take care of the "extended UTF-8". */
-#define IS_UTF8_CHAR_4c(p) \
+# define IS_UTF8_CHAR_4c(p) \
((p)[0] >= 0xF4 && (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) \
+# define IS_UTF8_CHAR_3(p) \
(IS_UTF8_CHAR_3a(p) || \
IS_UTF8_CHAR_3b(p) || \
IS_UTF8_CHAR_3c(p) || \
IS_UTF8_CHAR_3d(p))
-#define IS_UTF8_CHAR_4(p) \
+# define IS_UTF8_CHAR_4(p) \
(IS_UTF8_CHAR_4a(p) || \
IS_UTF8_CHAR_4b(p) || \
IS_UTF8_CHAR_4c(p))
* (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) \
+# 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)
+# 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 */