3 * This file contains definitions for use with the UTF-8 encoding. It
4 * actually also works with the variant UTF-8 encoding called UTF-EBCDIC, and
5 * hides almost all of the differences between these from the caller. In other
6 * words, someone should #include this file, and if the code is being compiled
7 * on an EBCDIC platform, things should mostly just work.
9 * Copyright (C) 2000, 2001, 2002, 2005, 2006, 2007, 2009,
10 * 2010, 2011 by Larry Wall and others
12 * You may distribute under the terms of either the GNU General Public
13 * License or the Artistic License, as specified in the README file.
17 #ifndef H_UTF8 /* Guard against recursive inclusion */
20 /* Use UTF-8 as the default script encoding?
21 * Turning this on will break scripts having non-UTF-8 binary
22 * data (such as Latin-1) in string literals. */
23 #ifdef USE_UTF8_SCRIPTS
24 # define USE_UTF8_IN_NAMES (!IN_BYTES)
26 # define USE_UTF8_IN_NAMES (PL_hints & HINT_UTF8)
29 #include "regcharclass.h"
30 #include "unicode_constants.h"
32 /* For to_utf8_fold_flags, q.v. */
33 #define FOLD_FLAGS_LOCALE 0x1
34 #define FOLD_FLAGS_FULL 0x2
35 #define FOLD_FLAGS_NOMIX_ASCII 0x4
37 /* For _core_swash_init(), internal core use only */
38 #define _CORE_SWASH_INIT_USER_DEFINED_PROPERTY 0x1
39 #define _CORE_SWASH_INIT_RETURN_IF_UNDEF 0x2
40 #define _CORE_SWASH_INIT_ACCEPT_INVLIST 0x4
43 =head1 Unicode Support
44 L<perlguts/Unicode Support> has an introduction to this API.
46 See also L</Character classification>,
47 and L</Character case changing>.
48 Various functions outside this section also work specially with Unicode.
49 Search for the string "utf8" in this document.
51 =for apidoc is_ascii_string
53 This is a misleadingly-named synonym for L</is_utf8_invariant_string>.
54 On ASCII-ish platforms, the name isn't misleading: the ASCII-range characters
55 are exactly the UTF-8 invariants. But EBCDIC machines have more invariants
56 than just the ASCII characters, so C<is_utf8_invariant_string> is preferred.
58 =for apidoc is_invariant_string
60 This is a somewhat misleadingly-named synonym for L</is_utf8_invariant_string>.
61 C<is_utf8_invariant_string> is preferred, as it indicates under what conditions
62 the string is invariant.
66 #define is_ascii_string(s, len) is_utf8_invariant_string(s, len)
67 #define is_invariant_string(s, len) is_utf8_invariant_string(s, len)
69 #define uvchr_to_utf8(a,b) uvchr_to_utf8_flags(a,b,0)
70 #define uvchr_to_utf8_flags(d,uv,flags) \
71 uvoffuni_to_utf8_flags(d,NATIVE_TO_UNI(uv),flags)
72 #define utf8_to_uvchr_buf(s, e, lenp) \
73 utf8n_to_uvchr(s, (U8*)(e) - (U8*)(s), lenp, \
74 ckWARN_d(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY)
76 #define to_uni_fold(c, p, lenp) _to_uni_fold_flags(c, p, lenp, FOLD_FLAGS_FULL)
77 #define to_utf8_fold(c, p, lenp) _to_utf8_fold_flags(c, p, lenp, FOLD_FLAGS_FULL)
78 #define to_utf8_lower(a,b,c) _to_utf8_lower_flags(a,b,c,0)
79 #define to_utf8_upper(a,b,c) _to_utf8_upper_flags(a,b,c,0)
80 #define to_utf8_title(a,b,c) _to_utf8_title_flags(a,b,c,0)
82 /* Source backward compatibility. */
83 #define is_utf8_string_loc(s, len, ep) is_utf8_string_loclen(s, len, ep, 0)
85 #define foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
86 foldEQ_utf8_flags(s1, pe1, l1, u1, s2, pe2, l2, u2, 0)
87 #define FOLDEQ_UTF8_NOMIX_ASCII (1 << 0)
88 #define FOLDEQ_LOCALE (1 << 1)
89 #define FOLDEQ_S1_ALREADY_FOLDED (1 << 2)
90 #define FOLDEQ_S2_ALREADY_FOLDED (1 << 3)
91 #define FOLDEQ_S1_FOLDS_SANE (1 << 4)
92 #define FOLDEQ_S2_FOLDS_SANE (1 << 5)
94 #define ibcmp_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
95 cBOOL(! foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2))
98 /* The equivalent of these macros but implementing UTF-EBCDIC
99 are in the following header file:
102 #include "utfebcdic.h"
107 /* How wide can a single UTF-8 encoded character become in bytes. */
108 /* NOTE: Strictly speaking Perl's UTF-8 should not be called UTF-8 since UTF-8
109 * is an encoding of Unicode, and Unicode's upper limit, 0x10FFFF, can be
110 * expressed with 4 bytes. However, Perl thinks of UTF-8 as a way to encode
111 * non-negative integers in a binary format, even those above Unicode */
112 #define UTF8_MAXBYTES 13
115 EXTCONST unsigned char PL_utf8skip[] = {
116 /* 0x00 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
117 /* 0x10 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
118 /* 0x20 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
119 /* 0x30 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
120 /* 0x40 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
121 /* 0x50 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
122 /* 0x60 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
123 /* 0x70 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
124 /* 0x80 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
125 /* 0x90 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
126 /* 0xA0 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
127 /* 0xB0 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
128 /* 0xC0 */ 2,2, /* overlong */
129 /* 0xC2 */ 2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* U+0080 to U+03FF */
130 /* 0xD0 */ 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* U+0400 to U+07FF */
131 /* 0xE0 */ 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3, /* U+0800 to U+FFFF */
132 /* 0xF0 */ 4,4,4,4,4,4,4,4,5,5,5,5,6,6, /* above BMP to 2**31 - 1 */
133 /* Perl extended (never was official UTF-8). Up to 36 bit */
135 /* More extended, Up to 72 bits (64-bit + reserved) */
136 /* 0xFF */ UTF8_MAXBYTES
139 EXTCONST unsigned char PL_utf8skip[];
144 #if defined(_MSC_VER) && _MSC_VER < 1400
145 /* older MSVC versions have a smallish macro buffer */
146 #define PERL_SMALL_MACRO_BUFFER
149 /* Native character to/from iso-8859-1. Are the identity functions on ASCII
151 #ifdef PERL_SMALL_MACRO_BUFFER
152 #define NATIVE_TO_LATIN1(ch) ((U8)(ch))
153 #define LATIN1_TO_NATIVE(ch) ((U8)(ch))
155 #define NATIVE_TO_LATIN1(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
156 #define LATIN1_TO_NATIVE(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
159 /* I8 is an intermediate version of UTF-8 used only in UTF-EBCDIC. We thus
160 * consider it to be identical to UTF-8 on ASCII platforms. Strictly speaking
161 * UTF-8 and UTF-EBCDIC are two different things, but we often conflate them
162 * because they are 8-bit encodings that serve the same purpose in Perl, and
163 * rarely do we need to distinguish them. The term "NATIVE_UTF8" applies to
164 * whichever one is applicable on the current platform */
165 #ifdef PERL_SMALL_MACRO_BUFFER
166 #define NATIVE_UTF8_TO_I8(ch) ((U8) (ch))
167 #define I8_TO_NATIVE_UTF8(ch) ((U8) (ch))
169 #define NATIVE_UTF8_TO_I8(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
170 #define I8_TO_NATIVE_UTF8(ch) (__ASSERT_(FITS_IN_8_BITS(ch)) ((U8) (ch)))
173 /* Transforms in wide UV chars */
174 #define UNI_TO_NATIVE(ch) ((UV) (ch))
175 #define NATIVE_TO_UNI(ch) ((UV) (ch))
179 The following table is from Unicode 3.2, plus the Perl extensions for above
182 Code Points 1st Byte 2nd Byte 3rd 4th 5th 6th 7th 8th-13th
184 U+0000..U+007F 00..7F
185 U+0080..U+07FF * C2..DF 80..BF
186 U+0800..U+0FFF E0 * A0..BF 80..BF
187 U+1000..U+CFFF E1..EC 80..BF 80..BF
188 U+D000..U+D7FF ED 80..9F 80..BF
189 U+D800..U+DFFF ED A0..BF 80..BF (surrogates)
190 U+E000..U+FFFF EE..EF 80..BF 80..BF
191 U+10000..U+3FFFF F0 * 90..BF 80..BF 80..BF
192 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
193 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
194 Below are above-Unicode code points
195 U+110000..U+13FFFF F4 90..BF 80..BF 80..BF
196 U+110000..U+1FFFFF F5..F7 80..BF 80..BF 80..BF
197 U+200000..U+FFFFFF F8 * 88..BF 80..BF 80..BF 80..BF
198 U+1000000..U+3FFFFFF F9..FB 80..BF 80..BF 80..BF 80..BF
199 U+4000000..U+3FFFFFFF FC * 84..BF 80..BF 80..BF 80..BF 80..BF
200 U+40000000..U+7FFFFFFF FD 80..BF 80..BF 80..BF 80..BF 80..BF
201 U+80000000..U+FFFFFFFFF FE * 82..BF 80..BF 80..BF 80..BF 80..BF 80..BF
202 U+1000000000.. FF 80..BF 80..BF 80..BF 80..BF 80..BF * 81..BF 80..BF
204 Note the gaps before several of the byte entries above marked by '*'. These are
205 caused by legal UTF-8 avoiding non-shortest encodings: it is technically
206 possible to UTF-8-encode a single code point in different ways, but that is
207 explicitly forbidden, and the shortest possible encoding should always be used
208 (and that is what Perl does). The non-shortest ones are called 'overlongs'.
213 Another way to look at it, as bits:
215 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
218 0000 0bbb bbaa aaaa 110b bbbb 10aa aaaa
219 cccc bbbb bbaa aaaa 1110 cccc 10bb bbbb 10aa aaaa
220 00 000d ddcc cccc bbbb bbaa aaaa 1111 0ddd 10cc cccc 10bb bbbb 10aa aaaa
222 As you can see, the continuation bytes all begin with C<10>, and the
223 leading bits of the start byte tell how many bytes there are in the
226 Perl's extended UTF-8 means we can have start bytes up through FF, though any
227 beginning with FF yields a code point that is too large for 32-bit ASCII
228 platforms. FF signals to use 13 bytes for the encoded character. This breaks
229 the paradigm that the number of leading bits gives how many total bytes there
230 are in the character.
234 /* Anything larger than this will overflow the word if it were converted into a UV */
235 #if defined(UV_IS_QUAD)
236 # define HIGHEST_REPRESENTABLE_UTF8 "\xFF\x80\x8F\xBF\xBF\xBF\xBF\xBF\xBF\xBF\xBF\xBF\xBF"
238 # define HIGHEST_REPRESENTABLE_UTF8 "\xFE\x83\xBF\xBF\xBF\xBF\xBF"
241 /* Is the representation of the Unicode code point 'cp' the same regardless of
242 * being encoded in UTF-8 or not? */
243 #define OFFUNI_IS_INVARIANT(cp) isASCII(cp)
246 =for apidoc Am|bool|UVCHR_IS_INVARIANT|UV cp
248 Evaluates to 1 if the representation of code point C<cp> is the same whether or
249 not it is encoded in UTF-8; otherwise evaluates to 0. UTF-8 invariant
250 characters can be copied as-is when converting to/from UTF-8, saving time.
251 C<cp> is Unicode if above 255; otherwise is platform-native.
256 #define UVCHR_IS_INVARIANT(cp) OFFUNI_IS_INVARIANT(cp)
258 /* This defines the bits that are to be in the continuation bytes of a multi-byte
259 * UTF-8 encoded character that mark it is a continuation byte. */
260 #define UTF_CONTINUATION_MARK 0x80
262 /* Misleadingly named: is the UTF8-encoded byte 'c' part of a variant sequence
263 * in UTF-8? This is the inverse of UTF8_IS_INVARIANT. The |0 makes sure this
264 * isn't mistakenly called with a ptr argument */
265 #define UTF8_IS_CONTINUED(c) (((U8)((c) | 0)) & UTF_CONTINUATION_MARK)
267 /* Is the byte 'c' the first byte of a multi-byte UTF8-8 encoded sequence?
268 * This doesn't catch invariants (they are single-byte). It also excludes the
269 * illegal overlong sequences that begin with C0 and C1. The |0 makes sure
270 * this isn't mistakenly called with a ptr argument */
271 #define UTF8_IS_START(c) (((U8)((c) | 0)) >= 0xc2)
273 /* For use in UTF8_IS_CONTINUATION() below */
274 #define UTF_IS_CONTINUATION_MASK 0xC0
276 /* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
277 * first byte thereof? The |0 makes sure this isn't mistakenly called with a
279 #define UTF8_IS_CONTINUATION(c) \
280 ((((U8)((c) | 0)) & UTF_IS_CONTINUATION_MASK) == UTF_CONTINUATION_MARK)
282 /* Is the UTF8-encoded byte 'c' the first byte of a two byte sequence? Use
283 * UTF8_IS_NEXT_CHAR_DOWNGRADEABLE() instead if the input isn't known to
284 * be well-formed. Masking with 0xfe allows the low bit to be 0 or 1; thus
285 * this matches 0xc[23]. The |0 makes sure this isn't mistakenly called with a
287 #define UTF8_IS_DOWNGRADEABLE_START(c) ((((U8)((c) | 0)) & 0xfe) == 0xc2)
289 /* Is the UTF8-encoded byte 'c' the first byte of a sequence of bytes that
290 * represent a code point > 255? The |0 makes sure this isn't mistakenly
291 * called with a ptr argument */
292 #define UTF8_IS_ABOVE_LATIN1(c) (((U8)((c) | 0)) >= 0xc4)
294 /* This is the number of low-order bits a continuation byte in a UTF-8 encoded
295 * sequence contributes to the specification of the code point. In the bit
296 * maps above, you see that the first 2 bits are a constant '10', leaving 6 of
297 * real information */
298 #define UTF_ACCUMULATION_SHIFT 6
300 /* ^? is defined to be DEL on ASCII systems. See the definition of toCTRL()
302 #define QUESTION_MARK_CTRL DEL_NATIVE
304 /* Surrogates, non-character code points and above-Unicode code points are
305 * problematic in some contexts. This allows code that needs to check for
306 * those to to quickly exclude the vast majority of code points it will
308 #define isUTF8_POSSIBLY_PROBLEMATIC(c) ((U8) c >= 0xED)
310 /* A helper macro for isUTF8_CHAR, so use that one instead of this. This was
311 * generated by regen/regcharclass.pl, and then moved here. Then it was
312 * hand-edited to add some LIKELY() calls, presuming that malformations are
313 * unlikely. The lines that generated it were then commented out. This was
314 * done because it takes on the order of 10 minutes to generate, and is never
315 * going to change, unless the generated code is improved, and figuring out
316 * the LIKELYs there would be hard.
318 UTF8_CHAR: Matches legal UTF-8 variant code points up through 0x1FFFFFF
322 /*** GENERATED CODE ***/
323 #define is_UTF8_CHAR_utf8_no_length_checks(s) \
324 ( ( 0xC2 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xDF ) ? \
325 ( LIKELY( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
326 : ( 0xE0 == ((U8*)s)[0] ) ? \
327 ( LIKELY( ( ( ((U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
328 : ( 0xE1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xEF ) ? \
329 ( LIKELY( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
330 : ( 0xF0 == ((U8*)s)[0] ) ? \
331 ( LIKELY( ( ( 0x90 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xBF ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
332 : ( ( ( ( 0xF1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xF7 ) && LIKELY( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ) && LIKELY( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && LIKELY( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )
334 /* The above macro handles UTF-8 that has this start byte as the maximum */
335 #define _IS_UTF8_CHAR_HIGHEST_START_BYTE 0xF7
337 /* A helper macro for isSTRICT_UTF8_CHAR, so use that one instead of this.
338 * Like is_UTF8_CHAR_utf8_no_length_checks(), this was moved here and LIKELYs
341 STRICT_UTF8_CHAR: Matches legal Unicode UTF-8 variant code points, no
342 surrrogates nor non-character code points
344 /*** GENERATED CODE ***/
345 #define is_STRICT_UTF8_CHAR_utf8_no_length_checks(s) \
346 ( ( 0xC2 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xDF ) ? \
347 ( LIKELY( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
348 : ( 0xE0 == ((U8*)s)[0] ) ? \
349 ( LIKELY( ( ( ((U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
350 : ( ( 0xE1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xEC ) || 0xEE == ((U8*)s)[0] ) ?\
351 ( ( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
352 : ( 0xED == ((U8*)s)[0] ) ? \
353 ( LIKELY( ( ( ((U8*)s)[1] & 0xE0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
354 : ( 0xEF == ((U8*)s)[0] ) ? \
355 ( ( ( 0x80 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xB6 ) || ( 0xB8 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xBE ) ) ?\
356 ( LIKELY( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ? 3 : 0 ) \
357 : ( 0xB7 == ((U8*)s)[1] ) ? \
358 ( LIKELY( ( ((U8*)s)[2] & 0xF0 ) == 0x80 || ( ((U8*)s)[2] & 0xF0 ) == 0xB0 ) ? 3 : 0 )\
359 : ( ( 0xBF == ((U8*)s)[1] ) && ( 0x80 <= ((U8*)s)[2] && ((U8*)s)[2] <= 0xBD ) ) ? 3 : 0 )\
360 : ( 0xF0 == ((U8*)s)[0] ) ? \
361 ( ( ( 0x90 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0x9E ) || ( 0xA0 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xAE ) || ( 0xB0 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xBE ) ) ?\
362 ( LIKELY( ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
363 : ( ((U8*)s)[1] == 0x9F || ( ( ((U8*)s)[1] & 0xEF ) == 0xAF ) ) ? \
364 ( ( 0x80 <= ((U8*)s)[2] && ((U8*)s)[2] <= 0xBE ) ? \
365 ( LIKELY( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ? 4 : 0 ) \
366 : LIKELY( ( 0xBF == ((U8*)s)[2] ) && ( 0x80 <= ((U8*)s)[3] && ((U8*)s)[3] <= 0xBD ) ) ? 4 : 0 )\
368 : ( 0xF1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xF3 ) ? \
369 ( ( ( ( ((U8*)s)[1] & 0xC8 ) == 0x80 ) || ( ( ((U8*)s)[1] & 0xCC ) == 0x88 ) || ( ( ((U8*)s)[1] & 0xCE ) == 0x8C ) || ( ( ((U8*)s)[1] & 0xCF ) == 0x8E ) ) ?\
370 ( LIKELY( ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
371 : ( ( ((U8*)s)[1] & 0xCF ) == 0x8F ) ? \
372 ( ( 0x80 <= ((U8*)s)[2] && ((U8*)s)[2] <= 0xBE ) ? \
373 ( LIKELY( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ? 4 : 0 ) \
374 : LIKELY( ( 0xBF == ((U8*)s)[2] ) && ( 0x80 <= ((U8*)s)[3] && ((U8*)s)[3] <= 0xBD ) ) ? 4 : 0 )\
376 : ( 0xF4 == ((U8*)s)[0] ) ? \
377 ( ( 0x80 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0x8E ) ? \
378 ( LIKELY( ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
379 : ( 0x8F == ((U8*)s)[1] ) ? \
380 ( ( 0x80 <= ((U8*)s)[2] && ((U8*)s)[2] <= 0xBE ) ? \
381 ( LIKELY( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ? 4 : 0 ) \
382 : LIKELY( ( 0xBF == ((U8*)s)[2] ) && ( 0x80 <= ((U8*)s)[3] && ((U8*)s)[3] <= 0xBD ) ) ? 4 : 0 )\
387 C9_STRICT_UTF8_CHAR: Matches legal Unicode UTF-8 variant code
388 points, no surrogates
392 /*** GENERATED CODE ***/
393 #define is_C9_STRICT_UTF8_CHAR_utf8_no_length_checks(s) \
394 ( ( 0xC2 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xDF ) ? \
395 ( LIKELY( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
396 : ( 0xE0 == ((U8*)s)[0] ) ? \
397 ( LIKELY( ( ( ((U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
398 : ( ( 0xE1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xEC ) || ( ((U8*)s)[0] & 0xFE ) == 0xEE ) ?\
399 ( LIKELY( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
400 : ( 0xED == ((U8*)s)[0] ) ? \
401 ( LIKELY( ( ( ((U8*)s)[1] & 0xE0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
402 : ( 0xF0 == ((U8*)s)[0] ) ? \
403 ( LIKELY( ( ( 0x90 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xBF ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
404 : ( 0xF1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xF3 ) ? \
405 ( LIKELY( ( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
406 : LIKELY( ( ( ( 0xF4 == ((U8*)s)[0] ) && ( ( ((U8*)s)[1] & 0xF0 ) == 0x80 ) ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )
408 #endif /* EBCDIC vs ASCII */
410 /* 2**UTF_ACCUMULATION_SHIFT - 1 */
411 #define UTF_CONTINUATION_MASK ((U8) ((1U << UTF_ACCUMULATION_SHIFT) - 1))
413 /* Internal macro to be used only in this file to aid in constructing other
414 * publicly accessible macros.
415 * The number of bytes required to express this uv in UTF-8, for just those
416 * uv's requiring 2 through 6 bytes, as these are common to all platforms and
417 * word sizes. The number of bytes needed is given by the number of leading 1
418 * bits in the start byte. There are 32 start bytes that have 2 initial 1 bits
419 * (C0-DF); there are 16 that have 3 initial 1 bits (E0-EF); 8 that have 4
420 * initial 1 bits (F0-F8); 4 that have 5 initial 1 bits (F9-FB), and 2 that
421 * have 6 initial 1 bits (FC-FD). The largest number a string of n bytes can
422 * represent is (the number of possible start bytes for 'n')
423 * * (the number of possiblities for each start byte
424 * The latter in turn is
425 * 2 ** ( (how many continuation bytes there are)
426 * * (the number of bits of information each
427 * continuation byte holds))
429 * If we were on a platform where we could use a fast find first set bit
430 * instruction (or count leading zeros instruction) this could be replaced by
431 * using that to find the log2 of the uv, and divide that by the number of bits
432 * of information in each continuation byte, adjusting for large cases and how
433 * much information is in a start byte for that length */
434 #define __COMMON_UNI_SKIP(uv) \
435 (UV) (uv) < (32 * (1U << ( UTF_ACCUMULATION_SHIFT))) ? 2 : \
436 (UV) (uv) < (16 * (1U << (2 * UTF_ACCUMULATION_SHIFT))) ? 3 : \
437 (UV) (uv) < ( 8 * (1U << (3 * UTF_ACCUMULATION_SHIFT))) ? 4 : \
438 (UV) (uv) < ( 4 * (1U << (4 * UTF_ACCUMULATION_SHIFT))) ? 5 : \
439 (UV) (uv) < ( 2 * (1U << (5 * UTF_ACCUMULATION_SHIFT))) ? 6 :
441 /* Internal macro to be used only in this file.
442 * This adds to __COMMON_UNI_SKIP the details at this platform's upper range.
443 * For any-sized EBCDIC platforms, or 64-bit ASCII ones, we need one more test
444 * to see if just 7 bytes is needed, or if the maximum is needed. For 32-bit
445 * ASCII platforms, everything is representable by 7 bytes */
446 #if defined(UV_IS_QUAD) || defined(EBCDIC)
447 # define __BASE_UNI_SKIP(uv) (__COMMON_UNI_SKIP(uv) \
448 (UV) (uv) < ((UV) 1U << (6 * UTF_ACCUMULATION_SHIFT)) ? 7 : UTF8_MAXBYTES)
450 # define __BASE_UNI_SKIP(uv) (__COMMON_UNI_SKIP(uv) 7)
453 /* The next two macros use the base macro defined above, and add in the tests
454 * at the low-end of the range, for just 1 byte, yielding complete macros,
455 * publicly accessible. */
457 /* Input is a true Unicode (not-native) code point */
458 #define OFFUNISKIP(uv) (OFFUNI_IS_INVARIANT(uv) ? 1 : __BASE_UNI_SKIP(uv))
462 =for apidoc Am|STRLEN|UVCHR_SKIP|UV cp
463 returns the number of bytes required to represent the code point C<cp> when
464 encoded as UTF-8. C<cp> is a native (ASCII or EBCDIC) code point if less than
465 255; a Unicode code point otherwise.
469 #define UVCHR_SKIP(uv) ( UVCHR_IS_INVARIANT(uv) ? 1 : __BASE_UNI_SKIP(uv))
471 /* The largest code point representable by two UTF-8 bytes on this platform.
472 * As explained in the comments for __COMMON_UNI_SKIP, 32 start bytes with
473 * UTF_ACCUMULATION_SHIFT bits of information each */
474 #define MAX_UTF8_TWO_BYTE (32 * (1U << UTF_ACCUMULATION_SHIFT) - 1)
476 /* The largest code point representable by two UTF-8 bytes on any platform that
477 * Perl runs on. This value is constrained by EBCDIC which has 5 bits per
478 * continuation byte */
479 #define MAX_PORTABLE_UTF8_TWO_BYTE (32 * (1U << 5) - 1)
481 /* The maximum number of UTF-8 bytes a single Unicode character can
482 * uppercase/lowercase/fold into. Unicode guarantees that the maximum
483 * expansion is UTF8_MAX_FOLD_CHAR_EXPAND characters, but any above-Unicode
484 * code point will fold to itself, so we only have to look at the expansion of
485 * the maximum Unicode code point. But this number may be less than the space
486 * occupied by a very large code point under Perl's extended UTF-8. We have to
487 * make it large enough to fit any single character. (It turns out that ASCII
488 * and EBCDIC differ in which is larger) */
489 #define UTF8_MAXBYTES_CASE \
490 (UTF8_MAXBYTES >= (UTF8_MAX_FOLD_CHAR_EXPAND * OFFUNISKIP(0x10FFFF)) \
492 : (UTF8_MAX_FOLD_CHAR_EXPAND * OFFUNISKIP(0x10FFFF)))
494 /* Rest of these are attributes of Unicode and perl's internals rather than the
495 * encoding, or happen to be the same in both ASCII and EBCDIC (at least at
496 * this level; the macros that some of these call may have different
497 * definitions in the two encodings */
499 /* In domain restricted to ASCII, these may make more sense to the reader than
500 * the ones with Latin1 in the name */
501 #define NATIVE_TO_ASCII(ch) NATIVE_TO_LATIN1(ch)
502 #define ASCII_TO_NATIVE(ch) LATIN1_TO_NATIVE(ch)
504 /* More or less misleadingly-named defines, retained for back compat */
505 #define NATIVE_TO_UTF(ch) NATIVE_UTF8_TO_I8(ch)
506 #define NATIVE_TO_I8(ch) NATIVE_UTF8_TO_I8(ch)
507 #define UTF_TO_NATIVE(ch) I8_TO_NATIVE_UTF8(ch)
508 #define I8_TO_NATIVE(ch) I8_TO_NATIVE_UTF8(ch)
509 #define NATIVE8_TO_UNI(ch) NATIVE_TO_LATIN1(ch)
511 /* This defines the 1-bits that are to be in the first byte of a multi-byte
512 * UTF-8 encoded character that mark it as a start byte and give the number of
513 * bytes that comprise the character. 'len' is the number of bytes in the
514 * multi-byte sequence. */
515 #define UTF_START_MARK(len) (((len) > 7) ? 0xFF : (0xFF & (0xFE << (7-(len)))))
517 /* Masks out the initial one bits in a start byte, leaving the real data ones.
518 * Doesn't work on an invariant byte. 'len' is the number of bytes in the
519 * multi-byte sequence that comprises the character. */
520 #define UTF_START_MASK(len) (((len) >= 7) ? 0x00 : (0x1F >> ((len)-2)))
522 /* Adds a UTF8 continuation byte 'new' of information to a running total code
523 * point 'old' of all the continuation bytes so far. This is designed to be
524 * used in a loop to convert from UTF-8 to the code point represented. Note
525 * that this is asymmetric on EBCDIC platforms, in that the 'new' parameter is
526 * the UTF-EBCDIC byte, whereas the 'old' parameter is a Unicode (not EBCDIC)
527 * code point in process of being generated */
528 #define UTF8_ACCUMULATE(old, new) (((old) << UTF_ACCUMULATION_SHIFT) \
529 | ((NATIVE_UTF8_TO_I8((U8)new)) \
530 & UTF_CONTINUATION_MASK))
532 /* If a value is anded with this, and the result is non-zero, then using the
533 * original value in UTF8_ACCUMULATE will overflow, shifting bits off the left
535 #define UTF_ACCUMULATION_OVERFLOW_MASK \
536 (((UV) UTF_CONTINUATION_MASK) << ((sizeof(UV) * CHARBITS) \
537 - UTF_ACCUMULATION_SHIFT))
539 /* This works in the face of malformed UTF-8. */
540 #define UTF8_IS_NEXT_CHAR_DOWNGRADEABLE(s, e) (UTF8_IS_DOWNGRADEABLE_START(*s) \
541 && ( (e) - (s) > 1) \
542 && UTF8_IS_CONTINUATION(*((s)+1)))
544 /* Number of bytes a code point occupies in UTF-8. */
545 #define NATIVE_SKIP(uv) UVCHR_SKIP(uv)
547 /* Most code which says UNISKIP is really thinking in terms of native code
548 * points (0-255) plus all those beyond. This is an imprecise term, but having
549 * it means existing code continues to work. For precision, use UVCHR_SKIP,
550 * NATIVE_SKIP, or OFFUNISKIP */
551 #define UNISKIP(uv) UVCHR_SKIP(uv)
553 /* Longer, but more accurate name */
554 #define UTF8_IS_ABOVE_LATIN1_START(c) UTF8_IS_ABOVE_LATIN1(c)
556 /* Convert a UTF-8 variant Latin1 character to a native code point value.
557 * Needs just one iteration of accumulate. Should be used only if it is known
558 * that the code point is < 256, and is not UTF-8 invariant. Use the slower
559 * but more general TWO_BYTE_UTF8_TO_NATIVE() which handles any code point
560 * representable by two bytes (which turns out to be up through
561 * MAX_PORTABLE_UTF8_TWO_BYTE). The two parameters are:
562 * HI: a downgradable start byte;
565 #define EIGHT_BIT_UTF8_TO_NATIVE(HI, LO) \
566 ( __ASSERT_(UTF8_IS_DOWNGRADEABLE_START(HI)) \
567 __ASSERT_(UTF8_IS_CONTINUATION(LO)) \
568 LATIN1_TO_NATIVE(UTF8_ACCUMULATE(( \
569 NATIVE_UTF8_TO_I8(HI) & UTF_START_MASK(2)), (LO))))
571 /* Convert a two (not one) byte utf8 character to a native code point value.
572 * Needs just one iteration of accumulate. Should not be used unless it is
573 * known that the two bytes are legal: 1) two-byte start, and 2) continuation.
574 * Note that the result can be larger than 255 if the input character is not
576 #define TWO_BYTE_UTF8_TO_NATIVE(HI, LO) \
577 ( __ASSERT_(PL_utf8skip[HI] == 2) \
578 __ASSERT_(UTF8_IS_CONTINUATION(LO)) \
579 UNI_TO_NATIVE(UTF8_ACCUMULATE((NATIVE_UTF8_TO_I8(HI) & UTF_START_MASK(2)), \
582 /* Should never be used, and be deprecated */
583 #define TWO_BYTE_UTF8_TO_UNI(HI, LO) NATIVE_TO_UNI(TWO_BYTE_UTF8_TO_NATIVE(HI, LO))
587 =for apidoc Am|STRLEN|UTF8SKIP|char* s
588 returns the number of bytes in the UTF-8 encoded character whose first (perhaps
589 only) byte is pointed to by C<s>.
593 #define UTF8SKIP(s) PL_utf8skip[*(const U8*)(s)]
594 #define UTF8_SKIP(s) UTF8SKIP(s)
596 /* Most code that says 'UNI_' really means the native value for code points up
598 #define UNI_IS_INVARIANT(cp) UVCHR_IS_INVARIANT(cp)
601 =for apidoc Am|bool|UTF8_IS_INVARIANT|char c
603 Evaluates to 1 if the byte C<c> represents the same character when encoded in
604 UTF-8 as when not; otherwise evaluates to 0. UTF-8 invariant characters can be
605 copied as-is when converting to/from UTF-8, saving time.
607 In spite of the name, this macro gives the correct result if the input string
608 from which C<c> comes is not encoded in UTF-8.
610 See C<L</UVCHR_IS_INVARIANT>> for checking if a UV is invariant.
614 The reason it works on both UTF-8 encoded strings and non-UTF-8 encoded, is
615 that it returns TRUE in each for the exact same set of bit patterns. It is
616 valid on a subset of what UVCHR_IS_INVARIANT is valid on, so can just use that;
617 and the compiler should optimize out anything extraneous given the
618 implementation of the latter. The |0 makes sure this isn't mistakenly called
621 #define UTF8_IS_INVARIANT(c) UVCHR_IS_INVARIANT((c) | 0)
623 /* Like the above, but its name implies a non-UTF8 input, which as the comments
624 * above show, doesn't matter as to its implementation */
625 #define NATIVE_BYTE_IS_INVARIANT(c) UVCHR_IS_INVARIANT(c)
627 /* The macros in the next 4 sets are used to generate the two utf8 or utfebcdic
628 * bytes from an ordinal that is known to fit into exactly two (not one) bytes;
629 * it must be less than 0x3FF to work across both encodings. */
631 /* These two are helper macros for the other three sets, and should not be used
632 * directly anywhere else. 'translate_function' is either NATIVE_TO_LATIN1
633 * (which works for code points up through 0xFF) or NATIVE_TO_UNI which works
634 * for any code point */
635 #define __BASE_TWO_BYTE_HI(c, translate_function) \
636 (__ASSERT_(! UVCHR_IS_INVARIANT(c)) \
637 I8_TO_NATIVE_UTF8((translate_function(c) >> UTF_ACCUMULATION_SHIFT) \
638 | UTF_START_MARK(2)))
639 #define __BASE_TWO_BYTE_LO(c, translate_function) \
640 (__ASSERT_(! UVCHR_IS_INVARIANT(c)) \
641 I8_TO_NATIVE_UTF8((translate_function(c) & UTF_CONTINUATION_MASK) \
642 | UTF_CONTINUATION_MARK))
644 /* The next two macros should not be used. They were designed to be usable as
645 * the case label of a switch statement, but this doesn't work for EBCDIC. Use
646 * regen/unicode_constants.pl instead */
647 #define UTF8_TWO_BYTE_HI_nocast(c) __BASE_TWO_BYTE_HI(c, NATIVE_TO_UNI)
648 #define UTF8_TWO_BYTE_LO_nocast(c) __BASE_TWO_BYTE_LO(c, NATIVE_TO_UNI)
650 /* The next two macros are used when the source should be a single byte
651 * character; checked for under DEBUGGING */
652 #define UTF8_EIGHT_BIT_HI(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
653 ( __BASE_TWO_BYTE_HI(c, NATIVE_TO_LATIN1)))
654 #define UTF8_EIGHT_BIT_LO(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
655 (__BASE_TWO_BYTE_LO(c, NATIVE_TO_LATIN1)))
657 /* These final two macros in the series are used when the source can be any
658 * code point whose UTF-8 is known to occupy 2 bytes; they are less efficient
659 * than the EIGHT_BIT versions on EBCDIC platforms. We use the logical '~'
660 * operator instead of "<=" to avoid getting compiler warnings.
661 * MAX_UTF8_TWO_BYTE should be exactly all one bits in the lower few
662 * places, so the ~ works */
663 #define UTF8_TWO_BYTE_HI(c) \
664 (__ASSERT_((sizeof(c) == 1) \
665 || !(((WIDEST_UTYPE)(c)) & ~MAX_UTF8_TWO_BYTE)) \
666 (__BASE_TWO_BYTE_HI(c, NATIVE_TO_UNI)))
667 #define UTF8_TWO_BYTE_LO(c) \
668 (__ASSERT_((sizeof(c) == 1) \
669 || !(((WIDEST_UTYPE)(c)) & ~MAX_UTF8_TWO_BYTE)) \
670 (__BASE_TWO_BYTE_LO(c, NATIVE_TO_UNI)))
672 /* This is illegal in any well-formed UTF-8 in both EBCDIC and ASCII
673 * as it is only in overlongs. */
674 #define ILLEGAL_UTF8_BYTE I8_TO_NATIVE_UTF8(0xC1)
677 * 'UTF' is whether or not p is encoded in UTF8. The names 'foo_lazy_if' stem
678 * from an earlier version of these macros in which they didn't call the
679 * foo_utf8() macros (i.e. were 'lazy') unless they decided that *p is the
680 * beginning of a utf8 character. Now that foo_utf8() determines that itself,
681 * no need to do it again here
683 #define isIDFIRST_lazy_if(p,UTF) ((IN_BYTES || !UTF) \
685 : isIDFIRST_utf8((const U8*)p))
686 #define isWORDCHAR_lazy_if(p,UTF) ((IN_BYTES || (!UTF)) \
688 : isWORDCHAR_utf8((const U8*)p))
689 #define isALNUM_lazy_if(p,UTF) isWORDCHAR_lazy_if(p,UTF)
691 #define UTF8_MAXLEN UTF8_MAXBYTES
693 /* A Unicode character can fold to up to 3 characters */
694 #define UTF8_MAX_FOLD_CHAR_EXPAND 3
696 #define IN_BYTES (CopHINTS_get(PL_curcop) & HINT_BYTES)
700 =for apidoc Am|bool|DO_UTF8|SV* sv
701 Returns a bool giving whether or not the PV in C<sv> is to be treated as being
704 You should use this I<after> a call to C<SvPV()> or one of its variants, in
705 case any call to string overloading updates the internal UTF-8 encoding flag.
709 #define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTES)
711 /* Should all strings be treated as Unicode, and not just UTF-8 encoded ones?
712 * Is so within 'feature unicode_strings' or 'locale :not_characters', and not
713 * within 'use bytes'. UTF-8 locales are not tested for here, but perhaps
715 #define IN_UNI_8_BIT \
716 (((CopHINTS_get(PL_curcop) & (HINT_UNI_8_BIT)) \
717 || (CopHINTS_get(PL_curcop) & HINT_LOCALE_PARTIAL \
718 /* -1 below is for :not_characters */ \
719 && _is_in_locale_category(FALSE, -1))) \
723 #define UTF8_ALLOW_EMPTY 0x0001 /* Allow a zero length string */
725 /* Allow first byte to be a continuation byte */
726 #define UTF8_ALLOW_CONTINUATION 0x0002
728 /* Allow second... bytes to be non-continuation bytes */
729 #define UTF8_ALLOW_NON_CONTINUATION 0x0004
731 /* expecting more bytes than were available in the string */
732 #define UTF8_ALLOW_SHORT 0x0008
734 /* Overlong sequence; i.e., the code point can be specified in fewer bytes. */
735 #define UTF8_ALLOW_LONG 0x0010
737 #define UTF8_DISALLOW_SURROGATE 0x0020 /* Unicode surrogates */
738 #define UTF8_WARN_SURROGATE 0x0040
740 #define UTF8_DISALLOW_NONCHAR 0x0080 /* Unicode non-character */
741 #define UTF8_WARN_NONCHAR 0x0100 /* code points */
743 #define UTF8_DISALLOW_SUPER 0x0200 /* Super-set of Unicode: code */
744 #define UTF8_WARN_SUPER 0x0400 /* points above the legal max */
746 /* Code points which never were part of the original UTF-8 standard, which only
747 * went up to 2 ** 31 - 1. Note that these all overflow a signed 32-bit word,
748 * The first byte of these code points is FE or FF on ASCII platforms. If the
749 * first byte is FF, it will overflow a 32-bit word. */
750 #define UTF8_DISALLOW_ABOVE_31_BIT 0x0800
751 #define UTF8_WARN_ABOVE_31_BIT 0x1000
753 /* For back compat, these old names are misleading for UTF_EBCDIC */
754 #define UTF8_DISALLOW_FE_FF UTF8_DISALLOW_ABOVE_31_BIT
755 #define UTF8_WARN_FE_FF UTF8_WARN_ABOVE_31_BIT
757 #define UTF8_CHECK_ONLY 0x2000
759 /* For backwards source compatibility. They do nothing, as the default now
760 * includes what they used to mean. The first one's meaning was to allow the
761 * just the single non-character 0xFFFF */
762 #define UTF8_ALLOW_FFFF 0
763 #define UTF8_ALLOW_SURROGATE 0
765 /* C9 refers to Unicode Corrigendum #9: allows but discourages non-chars */
766 #define UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE \
767 (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_SURROGATE)
768 #define UTF8_WARN_ILLEGAL_C9_INTERCHANGE (UTF8_WARN_SUPER|UTF8_WARN_SURROGATE)
770 #define UTF8_DISALLOW_ILLEGAL_INTERCHANGE \
771 (UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE|UTF8_DISALLOW_NONCHAR)
772 #define UTF8_WARN_ILLEGAL_INTERCHANGE \
773 (UTF8_WARN_ILLEGAL_C9_INTERCHANGE|UTF8_WARN_NONCHAR)
775 #define UTF8_ALLOW_ANY \
776 (~( UTF8_DISALLOW_ILLEGAL_INTERCHANGE|UTF8_DISALLOW_ABOVE_31_BIT \
777 |UTF8_WARN_ILLEGAL_INTERCHANGE|UTF8_WARN_ABOVE_31_BIT))
778 #define UTF8_ALLOW_ANYUV \
780 & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE|UTF8_WARN_ILLEGAL_INTERCHANGE))
781 #define UTF8_ALLOW_DEFAULT (ckWARN(WARN_UTF8) ? 0 : \
785 =for apidoc Am|bool|UTF8_IS_SURROGATE|const U8 *s|const U8 *e
787 Evaluates to non-zero if the first few bytes of the string starting at C<s> and
788 looking no further than S<C<e - 1>> are well-formed UTF-8 that represents one
789 of the Unicode surrogate code points; otherwise it evaluates to 0. If
790 non-zero, the value gives how many bytes starting at C<s> comprise the code
791 point's representation.
795 #define UTF8_IS_SURROGATE(s, e) is_SURROGATE_utf8_safe(s, e)
798 #define UTF8_IS_REPLACEMENT(s, send) is_REPLACEMENT_utf8_safe(s,send)
801 =for apidoc Am|bool|UTF8_IS_SUPER|const U8 *s|const U8 *e
803 Recall that Perl recognizes an extension to UTF-8 that can encode code
804 points larger than the ones defined by Unicode, which are 0..0x10FFFF.
806 This macro evaluates to non-zero if the first few bytes of the string starting
807 at C<s> and looking no further than S<C<e - 1>> are from this UTF-8 extension;
808 otherwise it evaluates to 0. If non-zero, the value gives how many bytes
809 starting at C<s> comprise the code point's representation.
811 0 is returned if the bytes are not well-formed extended UTF-8, or if they
812 represent a code point that cannot fit in a UV on the current platform. Hence
813 this macro can give different results when run on a 64-bit word machine than on
814 one with a 32-bit word size.
816 Note that it is deprecated to have code points that are larger than what can
817 fit in an IV on the current machine.
822 * U+10FFFF: \xF4\x8F\xBF\xBF \xF9\xA1\xBF\xBF\xBF max legal Unicode
823 * U+110000: \xF4\x90\x80\x80 \xF9\xA2\xA0\xA0\xA0
824 * U+110001: \xF4\x90\x80\x81 \xF9\xA2\xA0\xA0\xA1
827 # define UTF8_IS_SUPER(s, e) \
828 (( LIKELY((e) > (s) + 4) \
829 && NATIVE_UTF8_TO_I8(*(s)) >= 0xF9 \
830 && ( NATIVE_UTF8_TO_I8(*(s)) > 0xF9 \
831 || (NATIVE_UTF8_TO_I8(*(s) + 1) >= 0xA2)) \
832 && LIKELY((s) + UTF8SKIP(s) <= (e))) \
833 ? _is_utf8_char_helper(s, s + UTF8SKIP(s), 0) : 0)
835 # define UTF8_IS_SUPER(s, e) \
836 (( LIKELY((e) > (s) + 3) \
837 && (*(U8*) (s)) >= 0xF4 \
838 && ((*(U8*) (s)) > 0xF4 || (*((U8*) (s) + 1) >= 0x90))\
839 && LIKELY((s) + UTF8SKIP(s) <= (e))) \
840 ? _is_utf8_char_helper(s, s + UTF8SKIP(s), 0) : 0)
843 /* These are now machine generated, and the 'given' clause is no longer
845 #define UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s, e) \
846 cBOOL(is_NONCHAR_utf8_safe(s,e))
849 =for apidoc Am|bool|UTF8_IS_NONCHAR|const U8 *s|const U8 *e
851 Evaluates to non-zero if the first few bytes of the string starting at C<s> and
852 looking no further than S<C<e - 1>> are well-formed UTF-8 that represents one
853 of the Unicode non-character code points; otherwise it evaluates to 0. If
854 non-zero, the value gives how many bytes starting at C<s> comprise the code
855 point's representation.
859 #define UTF8_IS_NONCHAR(s, e) \
860 UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s, e)
862 #define UNICODE_SURROGATE_FIRST 0xD800
863 #define UNICODE_SURROGATE_LAST 0xDFFF
864 #define UNICODE_REPLACEMENT 0xFFFD
865 #define UNICODE_BYTE_ORDER_MARK 0xFEFF
867 /* Though our UTF-8 encoding can go beyond this,
868 * let's be conservative and do as Unicode says. */
869 #define PERL_UNICODE_MAX 0x10FFFF
871 #define UNICODE_WARN_SURROGATE 0x0001 /* UTF-16 surrogates */
872 #define UNICODE_WARN_NONCHAR 0x0002 /* Non-char code points */
873 #define UNICODE_WARN_SUPER 0x0004 /* Above 0x10FFFF */
874 #define UNICODE_WARN_ABOVE_31_BIT 0x0008 /* Above 0x7FFF_FFFF */
875 #define UNICODE_DISALLOW_SURROGATE 0x0010
876 #define UNICODE_DISALLOW_NONCHAR 0x0020
877 #define UNICODE_DISALLOW_SUPER 0x0040
878 #define UNICODE_DISALLOW_ABOVE_31_BIT 0x0080
879 #define UNICODE_WARN_ILLEGAL_C9_INTERCHANGE \
880 (UNICODE_WARN_SURROGATE|UNICODE_WARN_SUPER)
881 #define UNICODE_WARN_ILLEGAL_INTERCHANGE \
882 (UNICODE_WARN_ILLEGAL_C9_INTERCHANGE|UNICODE_WARN_NONCHAR)
883 #define UNICODE_DISALLOW_ILLEGAL_C9_INTERCHANGE \
884 (UNICODE_DISALLOW_SURROGATE|UNICODE_DISALLOW_SUPER)
885 #define UNICODE_DISALLOW_ILLEGAL_INTERCHANGE \
886 (UNICODE_DISALLOW_ILLEGAL_C9_INTERCHANGE|UNICODE_DISALLOW_NONCHAR)
888 /* For backward source compatibility, as are now the default */
889 #define UNICODE_ALLOW_SURROGATE 0
890 #define UNICODE_ALLOW_SUPER 0
891 #define UNICODE_ALLOW_ANY 0
893 /* This matches the 2048 code points between UNICODE_SURROGATE_FIRST (0xD800) and
894 * UNICODE_SURROGATE_LAST (0xDFFF) */
895 #define UNICODE_IS_SURROGATE(uv) (((UV) (uv) & (~0xFFFF | 0xF800)) \
898 #define UNICODE_IS_REPLACEMENT(uv) ((UV) (uv) == UNICODE_REPLACEMENT)
899 #define UNICODE_IS_BYTE_ORDER_MARK(uv) ((UV) (uv) == UNICODE_BYTE_ORDER_MARK)
901 /* Is 'uv' one of the 32 contiguous-range noncharacters? */
902 #define UNICODE_IS_32_CONTIGUOUS_NONCHARS(uv) ((UV) (uv) >= 0xFDD0 \
903 && (UV) (uv) <= 0xFDEF)
905 /* Is 'uv' one of the 34 plane-ending noncharacters 0xFFFE, 0xFFFF, 0x1FFFE,
906 * 0x1FFFF, ... 0x10FFFE, 0x10FFFF, given that we know that 'uv' is not above
907 * the Unicode legal max */
908 #define UNICODE_IS_END_PLANE_NONCHAR_GIVEN_NOT_SUPER(uv) \
909 (((UV) (uv) & 0xFFFE) == 0xFFFE)
911 #define UNICODE_IS_NONCHAR(uv) \
912 ( UNICODE_IS_32_CONTIGUOUS_NONCHARS(uv) \
913 || ( LIKELY( ! UNICODE_IS_SUPER(uv)) \
914 && UNICODE_IS_END_PLANE_NONCHAR_GIVEN_NOT_SUPER(uv)))
916 #define UNICODE_IS_SUPER(uv) ((UV) (uv) > PERL_UNICODE_MAX)
917 #define UNICODE_IS_ABOVE_31_BIT(uv) ((UV) (uv) > 0x7FFFFFFF)
919 #define LATIN_SMALL_LETTER_SHARP_S LATIN_SMALL_LETTER_SHARP_S_NATIVE
920 #define LATIN_SMALL_LETTER_Y_WITH_DIAERESIS \
921 LATIN_SMALL_LETTER_Y_WITH_DIAERESIS_NATIVE
922 #define MICRO_SIGN MICRO_SIGN_NATIVE
923 #define LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE \
924 LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE_NATIVE
925 #define LATIN_SMALL_LETTER_A_WITH_RING_ABOVE \
926 LATIN_SMALL_LETTER_A_WITH_RING_ABOVE_NATIVE
927 #define UNICODE_GREEK_CAPITAL_LETTER_SIGMA 0x03A3
928 #define UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA 0x03C2
929 #define UNICODE_GREEK_SMALL_LETTER_SIGMA 0x03C3
930 #define GREEK_SMALL_LETTER_MU 0x03BC
931 #define GREEK_CAPITAL_LETTER_MU 0x039C /* Upper and title case
933 #define LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS 0x0178 /* Also is title case */
934 #ifdef LATIN_CAPITAL_LETTER_SHARP_S_UTF8
935 # define LATIN_CAPITAL_LETTER_SHARP_S 0x1E9E
937 #define LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE 0x130
938 #define LATIN_SMALL_LETTER_DOTLESS_I 0x131
939 #define LATIN_SMALL_LETTER_LONG_S 0x017F
940 #define LATIN_SMALL_LIGATURE_LONG_S_T 0xFB05
941 #define LATIN_SMALL_LIGATURE_ST 0xFB06
942 #define KELVIN_SIGN 0x212A
943 #define ANGSTROM_SIGN 0x212B
945 #define UNI_DISPLAY_ISPRINT 0x0001
946 #define UNI_DISPLAY_BACKSLASH 0x0002
947 #define UNI_DISPLAY_QQ (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
948 #define UNI_DISPLAY_REGEX (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
950 #define ANYOF_FOLD_SHARP_S(node, input, end) \
951 (ANYOF_BITMAP_TEST(node, LATIN_SMALL_LETTER_SHARP_S) && \
952 (ANYOF_NONBITMAP(node)) && \
953 (ANYOF_FLAGS(node) & ANYOF_LOC_NONBITMAP_FOLD) && \
954 ((end) > (input) + 1) && \
955 isALPHA_FOLD_EQ((input)[0], 's'))
957 #define SHARP_S_SKIP 2
961 =for apidoc Am|STRLEN|isUTF8_CHAR|const U8 *s|const U8 *e
963 Evaluates to non-zero if the first few bytes of the string starting at C<s> and
964 looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
965 that represents some code point; otherwise it evaluates to 0. If non-zero, the
966 value gives how many bytes starting at C<s> comprise the code point's
969 The code point can be any that will fit in a UV on this machine, using Perl's
970 extension to official UTF-8 to represent those higher than the Unicode maximum
971 of 0x10FFFF. That means that this macro is used to efficiently decide if the
972 next few bytes in C<s> is legal UTF-8 for a single character. Use
973 L</is_utf8_string>(), L</is_utf8_string_loclen>(), and
974 L</is_utf8_string_loc>() to check entire strings.
976 Note that it is deprecated to use code points higher than what will fit in an
977 IV. This macro does not raise any warnings for such code points, treating them
980 Note also that a UTF-8 INVARIANT character (i.e. ASCII on non-EBCDIC machines)
981 is a valid UTF-8 character.
986 #define isUTF8_CHAR(s, e) \
987 (UNLIKELY((e) <= (s)) \
989 : (UTF8_IS_INVARIANT(*s)) \
991 : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
993 : LIKELY(NATIVE_UTF8_TO_I8(*s) <= _IS_UTF8_CHAR_HIGHEST_START_BYTE) \
994 ? is_UTF8_CHAR_utf8_no_length_checks(s) \
995 : _is_utf8_char_helper(s, e, 0))
997 #define is_utf8_char_buf(buf, buf_end) isUTF8_CHAR(buf, buf_end)
1001 =for apidoc Am|STRLEN|isSTRICT_UTF8_CHAR|const U8 *s|const U8 *e
1003 Evaluates to non-zero if the first few bytes of the string starting at C<s> and
1004 looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
1005 Unicode code point completely acceptable for open interchange between all
1006 applications; otherwise it evaluates to 0. If non-zero, the value gives how
1007 many bytes starting at C<s> comprise the code point's representation.
1009 The largest acceptable code point is the Unicode maximum 0x10FFFF, and must not
1010 be a surrogate nor a non-character code point. Thus this excludes any code
1011 point from Perl's extended UTF-8.
1013 This is used to efficiently decide if the next few bytes in C<s> is
1014 legal Unicode-acceptable UTF-8 for a single character. Use
1015 C<L</isC9_STRICT_UTF8_CHAR>> to also accept non-character code points.
1020 #define isSTRICT_UTF8_CHAR(s, e) \
1021 (UNLIKELY((e) <= (s)) \
1023 : (UTF8_IS_INVARIANT(*s)) \
1025 : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
1027 : is_STRICT_UTF8_CHAR_utf8_no_length_checks(s))
1031 =for apidoc Am|STRLEN|isC9_STRICT_UTF8_CHAR|const U8 *s|const U8 *e
1033 Evaluates to non-zero if the first few bytes of the string starting at C<s> and
1034 looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
1035 Unicode non-surrogate code point; otherwise it evaluates to 0. If non-zero,
1036 the value gives how many bytes starting at C<s> comprise the code point's
1039 The largest acceptable code point is the Unicode maximum 0x10FFFF. This
1040 differs from C<L</isSTRICT_UTF8_CHAR>> only in that it accepts non-character
1041 code points. This corresponds to
1042 L<Unicode Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>.
1043 which said that non-character code points are merely discouraged rather than
1044 completely forbidden in open interchange. See
1045 L<perlunicode/Noncharacter code points>.
1050 #define isC9_STRICT_UTF8_CHAR(s, e) \
1051 (UNLIKELY((e) <= (s)) \
1053 : (UTF8_IS_INVARIANT(*s)) \
1055 : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
1057 : is_C9_STRICT_UTF8_CHAR_utf8_no_length_checks(s))
1061 =for apidoc Am|STRLEN|isUTF8_CHAR_flags|const U8 *s|const U8 *e| const U32 flags
1063 Evaluates to non-zero if the first few bytes of the string starting at C<s> and
1064 looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
1065 that represents some code point, subject to the restrictions given by C<flags>;
1066 otherwise it evaluates to 0. If non-zero, the value gives how many bytes
1067 starting at C<s> comprise the code point's representation.
1069 If C<flags> is 0, this gives the same results as C<L</isUTF8_CHAR>>;
1070 if C<flags> is C<UTF8_DISALLOW_ILLEGAL_INTERCHANGE>, this gives the same results
1071 as C<L</isSTRICT_UTF8_CHAR>>;
1072 and if C<flags> is C<UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE>, this gives
1073 the same results as C<L</isC9_STRICT_UTF8_CHAR>>.
1074 Otherwise C<flags> may be any combination of the C<UTF8_DISALLOW_I<foo>> flags
1075 understood by C<L</utf8n_to_uvchr>>, with the same meanings.
1077 The three alternative macros are for the most commonly needed validations; they
1078 are likely to run somewhat faster than this more general one, as they can be
1079 inlined into your code.
1084 #define isUTF8_CHAR_flags(s, e, flags) \
1085 (UNLIKELY((e) <= (s)) \
1087 : (UTF8_IS_INVARIANT(*s)) \
1089 : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
1091 : _is_utf8_char_helper(s, e, flags))
1093 /* Do not use; should be deprecated. Use isUTF8_CHAR() instead; this is
1094 * retained solely for backwards compatibility */
1095 #define IS_UTF8_CHAR(p, n) (isUTF8_CHAR(p, (p) + (n)) == n)
1100 * ex: set ts=8 sts=4 sw=4 et: