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_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_invariant_string> is preferred.
60 #define is_ascii_string(s, len) is_invariant_string(s, len)
62 #define uvchr_to_utf8(a,b) uvchr_to_utf8_flags(a,b,0)
63 #define uvchr_to_utf8_flags(d,uv,flags) \
64 uvoffuni_to_utf8_flags(d,NATIVE_TO_UNI(uv),flags)
65 #define utf8_to_uvchr_buf(s, e, lenp) \
66 utf8n_to_uvchr(s, (U8*)(e) - (U8*)(s), lenp, \
67 ckWARN_d(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY)
69 #define to_uni_fold(c, p, lenp) _to_uni_fold_flags(c, p, lenp, FOLD_FLAGS_FULL)
70 #define to_utf8_fold(c, p, lenp) _to_utf8_fold_flags(c, p, lenp, FOLD_FLAGS_FULL)
71 #define to_utf8_lower(a,b,c) _to_utf8_lower_flags(a,b,c,0)
72 #define to_utf8_upper(a,b,c) _to_utf8_upper_flags(a,b,c,0)
73 #define to_utf8_title(a,b,c) _to_utf8_title_flags(a,b,c,0)
75 /* Source backward compatibility. */
76 #define is_utf8_string_loc(s, len, ep) is_utf8_string_loclen(s, len, ep, 0)
78 #define foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
79 foldEQ_utf8_flags(s1, pe1, l1, u1, s2, pe2, l2, u2, 0)
80 #define FOLDEQ_UTF8_NOMIX_ASCII (1 << 0)
81 #define FOLDEQ_LOCALE (1 << 1)
82 #define FOLDEQ_S1_ALREADY_FOLDED (1 << 2)
83 #define FOLDEQ_S2_ALREADY_FOLDED (1 << 3)
84 #define FOLDEQ_S1_FOLDS_SANE (1 << 4)
85 #define FOLDEQ_S2_FOLDS_SANE (1 << 5)
87 #define ibcmp_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2) \
88 cBOOL(! foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2))
91 /* The equivalent of these macros but implementing UTF-EBCDIC
92 are in the following header file:
95 #include "utfebcdic.h"
101 EXTCONST unsigned char PL_utf8skip[] = {
102 /* 0x00 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
103 /* 0x10 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
104 /* 0x20 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
105 /* 0x30 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
106 /* 0x40 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
107 /* 0x50 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
108 /* 0x60 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
109 /* 0x70 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* ascii */
110 /* 0x80 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
111 /* 0x90 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
112 /* 0xA0 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
113 /* 0xB0 */ 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* bogus: continuation byte */
114 /* 0xC0 */ 2,2, /* overlong */
115 /* 0xC2 */ 2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* U+0080 to U+03FF */
116 /* 0xD0 */ 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* U+0400 to U+07FF */
117 /* 0xE0 */ 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3, /* U+0800 to U+FFFF */
118 /* 0xF0 */ 4,4,4,4,4,4,4,4,5,5,5,5,6,6, /* above BMP to 2**31 - 1 */
119 /* 0xFE */ 7,13, /* Perl extended (never was official UTF-8). Up to 72bit
120 allowed (64-bit + reserved). */
123 EXTCONST unsigned char PL_utf8skip[];
128 /* Native character to/from iso-8859-1. Are the identity functions on ASCII
130 #define NATIVE_TO_LATIN1(ch) (ch)
131 #define LATIN1_TO_NATIVE(ch) (ch)
133 /* I8 is an intermediate version of UTF-8 used only in UTF-EBCDIC. We thus
134 * consider it to be identical to UTF-8 on ASCII platforms. Strictly speaking
135 * UTF-8 and UTF-EBCDIC are two different things, but we often conflate them
136 * because they are 8-bit encodings that serve the same purpose in Perl, and
137 * rarely do we need to distinguish them. The term "NATIVE_UTF8" applies to
138 * whichever one is applicable on the current platform */
139 #define NATIVE_UTF8_TO_I8(ch) (ch)
140 #define I8_TO_NATIVE_UTF8(ch) (ch)
142 /* Transforms in wide UV chars */
143 #define UNI_TO_NATIVE(ch) (ch)
144 #define NATIVE_TO_UNI(ch) (ch)
148 The following table is from Unicode 3.2.
150 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
152 U+0000..U+007F 00..7F
153 U+0080..U+07FF * C2..DF 80..BF
154 U+0800..U+0FFF E0 * A0..BF 80..BF
155 U+1000..U+CFFF E1..EC 80..BF 80..BF
156 U+D000..U+D7FF ED 80..9F 80..BF
157 U+D800..U+DFFF ED A0..BF 80..BF (surrogates)
158 U+E000..U+FFFF EE..EF 80..BF 80..BF
159 U+10000..U+3FFFF F0 * 90..BF 80..BF 80..BF
160 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
161 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
162 Below are non-Unicode code points
163 U+110000..U+13FFFF F4 90..BF 80..BF 80..BF
164 U+110000..U+1FFFFF F5..F7 80..BF 80..BF 80..BF
165 U+200000..: F8.. * 88..BF 80..BF 80..BF 80..BF
167 Note the gaps before several of the byte entries above marked by '*'. These are
168 caused by legal UTF-8 avoiding non-shortest encodings: it is technically
169 possible to UTF-8-encode a single code point in different ways, but that is
170 explicitly forbidden, and the shortest possible encoding should always be used
171 (and that is what Perl does). The non-shortest ones are called 'overlongs'.
176 Another way to look at it, as bits:
178 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
181 0000 0bbb bbaa aaaa 110b bbbb 10aa aaaa
182 cccc bbbb bbaa aaaa 1110 cccc 10bb bbbb 10aa aaaa
183 00 000d ddcc cccc bbbb bbaa aaaa 1111 0ddd 10cc cccc 10bb bbbb 10aa aaaa
185 As you can see, the continuation bytes all begin with C<10>, and the
186 leading bits of the start byte tell how many bytes there are in the
189 Perl's extended UTF-8 means we can have start bytes up to FF.
193 /* Is the representation of the Unicode code point 'c' the same regardless of
194 * being encoded in UTF-8 or not? */
195 #define UNI_IS_INVARIANT(c) (((UV)c) < 0x80)
197 /* Is the UTF8-encoded byte 'c' part of a variant sequence in UTF-8? This is
198 * the inverse of UTF8_IS_INVARIANT */
199 #define UTF8_IS_CONTINUED(c) (((U8)c) & 0x80)
201 /* Is the byte 'c' the first byte of a multi-byte UTF8-8 encoded sequence?
202 * This doesn't catch invariants (they are single-byte). It also excludes the
203 * illegal overlong sequences that begin with C0 and C1. */
204 #define UTF8_IS_START(c) (((U8)c) >= 0xc2)
206 /* Is the byte 'c' part of a multi-byte UTF8-8 encoded sequence, and not the
207 * first byte thereof? */
208 #define UTF8_IS_CONTINUATION(c) ((((U8)c) & 0xC0) == 0x80)
210 /* Is the UTF8-encoded byte 'c' the first byte of a two byte sequence? Use
211 * UTF8_IS_NEXT_CHAR_DOWNGRADEABLE() instead if the input isn't known to
212 * be well-formed. Masking with 0xfe allows the low bit to be 0 or 1; thus
213 * this matches 0xc[23]. */
214 #define UTF8_IS_DOWNGRADEABLE_START(c) (((U8)(c) & 0xfe) == 0xc2)
216 /* Is the UTF8-encoded byte 'c' the first byte of a sequence of bytes that
217 * represent a code point > 255? */
218 #define UTF8_IS_ABOVE_LATIN1(c) ((U8)(c) >= 0xc4)
220 /* This defines the 1-bits that are to be in the first byte of a multi-byte
221 * UTF-8 encoded character that give the number of bytes that comprise the
222 * character. 'len' is the number of bytes in the multi-byte sequence. */
223 #define UTF_START_MARK(len) (((len) > 7) ? 0xFF : (0xFF & (0xFE << (7-(len)))))
225 /* Masks out the initial one bits in a start byte, leaving the real data ones.
226 * Doesn't work on an invariant byte. 'len' is the number of bytes in the
227 * multi-byte sequence that comprises the character. */
228 #define UTF_START_MASK(len) (((len) >= 7) ? 0x00 : (0x1F >> ((len)-2)))
230 /* This defines the bits that are to be in the continuation bytes of a multi-byte
231 * UTF-8 encoded character that indicate it is a continuation byte. */
232 #define UTF_CONTINUATION_MARK 0x80
234 /* This is the number of low-order bits a continuation byte in a UTF-8 encoded
235 * sequence contributes to the specification of the code point. In the bit
236 * maps above, you see that the first 2 bits are a constant '10', leaving 6 of
237 * real information */
238 #define UTF_ACCUMULATION_SHIFT 6
240 /* 2**UTF_ACCUMULATION_SHIFT - 1 */
241 #define UTF_CONTINUATION_MASK ((U8)0x3f)
243 /* If a value is anded with this, and the result is non-zero, then using the
244 * original value in UTF8_ACCUMULATE will overflow, shifting bits off the left
246 #define UTF_ACCUMULATION_OVERFLOW_MASK \
247 (((UV) UTF_CONTINUATION_MASK) << ((sizeof(UV) * CHARBITS) \
248 - UTF_ACCUMULATION_SHIFT))
251 # define UTF8_QUAD_MAX UINT64_C(0x1000000000)
253 /* Input is a true Unicode (not-native) code point */
254 #define OFFUNISKIP(uv) ( (uv) < 0x80 ? 1 : \
256 (uv) < 0x10000 ? 3 : \
257 (uv) < 0x200000 ? 4 : \
258 (uv) < 0x4000000 ? 5 : \
259 (uv) < 0x80000000 ? 6 : \
260 (uv) < UTF8_QUAD_MAX ? 7 : 13 )
262 /* No, I'm not even going to *TRY* putting #ifdef inside a #define */
263 #define OFFUNISKIP(uv) ( (uv) < 0x80 ? 1 : \
265 (uv) < 0x10000 ? 3 : \
266 (uv) < 0x200000 ? 4 : \
267 (uv) < 0x4000000 ? 5 : \
268 (uv) < 0x80000000 ? 6 : 7 )
271 /* How wide can a single UTF-8 encoded character become in bytes. */
272 /* NOTE: Strictly speaking Perl's UTF-8 should not be called UTF-8 since UTF-8
273 * is an encoding of Unicode, and Unicode's upper limit, 0x10FFFF, can be
274 * expressed with 4 bytes. However, Perl thinks of UTF-8 as a way to encode
275 * non-negative integers in a binary format, even those above Unicode */
276 #define UTF8_MAXBYTES 13
278 /* The maximum number of UTF-8 bytes a single Unicode character can
279 * uppercase/lowercase/fold into. Unicode guarantees that the maximum
280 * expansion is 3 characters. On ASCIIish platforms, the highest Unicode
281 * character occupies 4 bytes, therefore this number would be 12, but this is
282 * smaller than the maximum width a single above-Unicode character can occupy,
283 * so use that instead */
284 #if UTF8_MAXBYTES < 12
285 #error UTF8_MAXBYTES must be at least 12
288 /* ^? is defined to be DEL on ASCII systems. See the definition of toCTRL()
290 #define QUESTION_MARK_CTRL DEL_NATIVE
292 #define MAX_UTF8_TWO_BYTE 0x7FF
294 #define UTF8_MAXBYTES_CASE UTF8_MAXBYTES
296 #endif /* EBCDIC vs ASCII */
298 /* Rest of these are attributes of Unicode and perl's internals rather than the
299 * encoding, or happen to be the same in both ASCII and EBCDIC (at least at
300 * this level; the macros that some of these call may have different
301 * definitions in the two encodings */
303 /* In domain restricted to ASCII, these may make more sense to the reader than
304 * the ones with Latin1 in the name */
305 #define NATIVE_TO_ASCII(ch) NATIVE_TO_LATIN1(ch)
306 #define ASCII_TO_NATIVE(ch) LATIN1_TO_NATIVE(ch)
308 /* More or less misleadingly-named defines, retained for back compat */
309 #define NATIVE_TO_UTF(ch) NATIVE_UTF8_TO_I8(ch)
310 #define NATIVE_TO_I8(ch) NATIVE_UTF8_TO_I8(ch)
311 #define UTF_TO_NATIVE(ch) I8_TO_NATIVE_UTF8(ch)
312 #define I8_TO_NATIVE(ch) I8_TO_NATIVE_UTF8(ch)
313 #define NATIVE8_TO_UNI(ch) NATIVE_TO_LATIN1(ch)
315 /* Adds a UTF8 continuation byte 'new' of information to a running total code
316 * point 'old' of all the continuation bytes so far. This is designed to be
317 * used in a loop to convert from UTF-8 to the code point represented. Note
318 * that this is asymmetric on EBCDIC platforms, in that the 'new' parameter is
319 * the UTF-EBCDIC byte, whereas the 'old' parameter is a Unicode (not EBCDIC)
320 * code point in process of being generated */
321 #define UTF8_ACCUMULATE(old, new) (((old) << UTF_ACCUMULATION_SHIFT) \
322 | ((NATIVE_UTF8_TO_I8((U8)new)) \
323 & UTF_CONTINUATION_MASK))
325 /* This works in the face of malformed UTF-8. */
326 #define UTF8_IS_NEXT_CHAR_DOWNGRADEABLE(s, e) (UTF8_IS_DOWNGRADEABLE_START(*s) \
327 && ( (e) - (s) > 1) \
328 && UTF8_IS_CONTINUATION(*((s)+1)))
330 /* Number of bytes a code point occupies in UTF-8. */
331 #define NATIVE_SKIP(uv) OFFUNISKIP(NATIVE_TO_UNI(uv))
335 =for apidoc Am|STRLEN|UVCHR_SKIP|UV cp
336 returns the number of bytes required to represent the code point C<cp> when
337 encoded as UTF-8. C<cp> is a native (ASCII or EBCDIC) code point if less than
338 255; a Unicode code point otherwise.
343 /* Most code which says UNISKIP is really thinking in terms of native code
344 * points (0-255) plus all those beyond. This is an imprecise term, but having
345 * it means existing code continues to work. For precision, use UVCHR_SKIP,
346 * NATIVE_SKIP, and OFFUNISKIP */
347 #define UNISKIP(uv) NATIVE_SKIP(uv)
348 #define UVCHR_SKIP(uv) NATIVE_SKIP(uv)
350 /* Longer, but more accurate name */
351 #define UTF8_IS_ABOVE_LATIN1_START(c) UTF8_IS_ABOVE_LATIN1(c)
353 /* Convert a UTF-8 variant Latin1 character to a native code point value.
354 * Needs just one iteration of accumulate. Should be used only if it is known
355 * that the code point is < 256, and is not UTF-8 invariant. Use the slower
356 * but more general TWO_BYTE_UTF8_TO_NATIVE() which handles any code point
357 * representable by two bytes (which turns out to be up through
358 * MAX_PORTABLE_UTF8_TWO_BYTE). The two parameters are:
359 * HI: a downgradable start byte;
362 #define EIGHT_BIT_UTF8_TO_NATIVE(HI, LO) \
363 ( __ASSERT_(UTF8_IS_DOWNGRADEABLE_START(HI)) \
364 __ASSERT_(UTF8_IS_CONTINUATION(LO)) \
365 LATIN1_TO_NATIVE(UTF8_ACCUMULATE(( \
366 NATIVE_UTF8_TO_I8(HI) & UTF_START_MASK(2)), (LO))))
368 /* Convert a two (not one) byte utf8 character to a native code point value.
369 * Needs just one iteration of accumulate. Should not be used unless it is
370 * known that the two bytes are legal: 1) two-byte start, and 2) continuation.
371 * Note that the result can be larger than 255 if the input character is not
373 #define TWO_BYTE_UTF8_TO_NATIVE(HI, LO) \
374 ( __ASSERT_(UTF8SKIP(HI) == 2) \
375 __ASSERT_(UTF8_IS_CONTINUATION(LO)) \
376 UNI_TO_NATIVE(UTF8_ACCUMULATE((NATIVE_UTF8_TO_I8(HI) & UTF_START_MASK(2)), \
379 /* Should never be used, and be deprecated */
380 #define TWO_BYTE_UTF8_TO_UNI(HI, LO) NATIVE_TO_UNI(TWO_BYTE_UTF8_TO_NATIVE(HI, LO))
384 =for apidoc Am|STRLEN|UTF8SKIP|char* s
385 returns the number of bytes in the UTF-8 encoded character whose first (perhaps
386 only) byte is pointed to by C<s>.
390 #define UTF8SKIP(s) PL_utf8skip[*(const U8*)(s)]
391 #define UTF8_SKIP(s) UTF8SKIP(s)
393 /* Is the byte 'c' the same character when encoded in UTF-8 as when not. This
394 * works on both UTF-8 encoded strings and non-encoded, as it returns TRUE in
395 * each for the exact same set of bit patterns. (And it works on any byte in a
396 * UTF-8 encoded string) */
397 #define UTF8_IS_INVARIANT(c) UNI_IS_INVARIANT(NATIVE_UTF8_TO_I8(c))
399 /* Like the above, but its name implies a non-UTF8 input, and is implemented
400 * differently (for no particular reason) */
401 #define NATIVE_BYTE_IS_INVARIANT(c) UNI_IS_INVARIANT(NATIVE_TO_LATIN1(c))
403 /* Like the above, but accepts any UV as input */
404 #define UVCHR_IS_INVARIANT(uv) UNI_IS_INVARIANT(NATIVE_TO_UNI(uv))
406 #define MAX_PORTABLE_UTF8_TWO_BYTE 0x3FF /* constrained by EBCDIC */
408 /* The macros in the next 4 sets are used to generate the two utf8 or utfebcdic
409 * bytes from an ordinal that is known to fit into exactly two (not one) bytes;
410 * it must be less than 0x3FF to work across both encodings. */
412 /* These two are helper macros for the other three sets, and should not be used
413 * directly anywhere else. 'translate_function' is either NATIVE_TO_LATIN1
414 * (which works for code points up through 0xFF) or NATIVE_TO_UNI which works
415 * for any code point */
416 #define __BASE_TWO_BYTE_HI(c, translate_function) \
417 I8_TO_NATIVE_UTF8((translate_function(c) >> UTF_ACCUMULATION_SHIFT) \
419 #define __BASE_TWO_BYTE_LO(c, translate_function) \
420 I8_TO_NATIVE_UTF8((translate_function(c) & UTF_CONTINUATION_MASK) \
421 | UTF_CONTINUATION_MARK)
423 /* The next two macros should not be used. They were designed to be usable as
424 * the case label of a switch statement, but this doesn't work for EBCDIC. Use
425 * regen/unicode_constants.pl instead */
426 #define UTF8_TWO_BYTE_HI_nocast(c) __BASE_TWO_BYTE_HI(c, NATIVE_TO_UNI)
427 #define UTF8_TWO_BYTE_LO_nocast(c) __BASE_TWO_BYTE_LO(c, NATIVE_TO_UNI)
429 /* The next two macros are used when the source should be a single byte
430 * character; checked for under DEBUGGING */
431 #define UTF8_EIGHT_BIT_HI(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
432 ((U8) __BASE_TWO_BYTE_HI(c, NATIVE_TO_LATIN1)))
433 #define UTF8_EIGHT_BIT_LO(c) (__ASSERT_(FITS_IN_8_BITS(c)) \
434 ((U8) __BASE_TWO_BYTE_LO(c, NATIVE_TO_LATIN1)))
436 /* These final two macros in the series are used when the source can be any
437 * code point whose UTF-8 is known to occupy 2 bytes; they are less efficient
438 * than the EIGHT_BIT versions on EBCDIC platforms. We use the logical '~'
439 * operator instead of "<=" to avoid getting compiler warnings.
440 * MAX_UTF8_TWO_BYTE should be exactly all one bits in the lower few
441 * places, so the ~ works */
442 #define UTF8_TWO_BYTE_HI(c) \
443 (__ASSERT_((sizeof(c) == 1) \
444 || !(((WIDEST_UTYPE)(c)) & ~MAX_UTF8_TWO_BYTE)) \
445 ((U8) __BASE_TWO_BYTE_HI(c, NATIVE_TO_UNI)))
446 #define UTF8_TWO_BYTE_LO(c) \
447 (__ASSERT_((sizeof(c) == 1) \
448 || !(((WIDEST_UTYPE)(c)) & ~MAX_UTF8_TWO_BYTE)) \
449 ((U8) __BASE_TWO_BYTE_LO(c, NATIVE_TO_UNI)))
451 /* This is illegal in any well-formed UTF-8 in both EBCDIC and ASCII
452 * as it is only in overlongs. */
453 #define ILLEGAL_UTF8_BYTE I8_TO_NATIVE_UTF8(0xC1)
456 * 'UTF' is whether or not p is encoded in UTF8. The names 'foo_lazy_if' stem
457 * from an earlier version of these macros in which they didn't call the
458 * foo_utf8() macros (i.e. were 'lazy') unless they decided that *p is the
459 * beginning of a utf8 character. Now that foo_utf8() determines that itself,
460 * no need to do it again here
462 #define isIDFIRST_lazy_if(p,UTF) ((IN_BYTES || !UTF ) \
464 : isIDFIRST_utf8((const U8*)p))
465 #define isWORDCHAR_lazy_if(p,UTF) ((IN_BYTES || (!UTF )) \
467 : isWORDCHAR_utf8((const U8*)p))
468 #define isALNUM_lazy_if(p,UTF) isWORDCHAR_lazy_if(p,UTF)
470 #define UTF8_MAXLEN UTF8_MAXBYTES
472 /* A Unicode character can fold to up to 3 characters */
473 #define UTF8_MAX_FOLD_CHAR_EXPAND 3
475 #define IN_BYTES (CopHINTS_get(PL_curcop) & HINT_BYTES)
479 =for apidoc Am|bool|DO_UTF8|SV* sv
480 Returns a bool giving whether or not the PV in C<sv> is to be treated as being
483 You should use this I<after> a call to C<SvPV()> or one of its variants, in
484 case any call to string overloading updates the internal UTF-8 encoding flag.
488 #define DO_UTF8(sv) (SvUTF8(sv) && !IN_BYTES)
490 /* Should all strings be treated as Unicode, and not just UTF-8 encoded ones?
491 * Is so within 'feature unicode_strings' or 'locale :not_characters', and not
492 * within 'use bytes'. UTF-8 locales are not tested for here, but perhaps
494 #define IN_UNI_8_BIT \
495 (((CopHINTS_get(PL_curcop) & (HINT_UNI_8_BIT)) \
496 || (CopHINTS_get(PL_curcop) & HINT_LOCALE_PARTIAL \
497 /* -1 below is for :not_characters */ \
498 && _is_in_locale_category(FALSE, -1))) \
502 #define UTF8_ALLOW_EMPTY 0x0001 /* Allow a zero length string */
504 /* Allow first byte to be a continuation byte */
505 #define UTF8_ALLOW_CONTINUATION 0x0002
507 /* Allow second... bytes to be non-continuation bytes */
508 #define UTF8_ALLOW_NON_CONTINUATION 0x0004
510 /* expecting more bytes than were available in the string */
511 #define UTF8_ALLOW_SHORT 0x0008
513 /* Overlong sequence; i.e., the code point can be specified in fewer bytes. */
514 #define UTF8_ALLOW_LONG 0x0010
516 #define UTF8_DISALLOW_SURROGATE 0x0020 /* Unicode surrogates */
517 #define UTF8_WARN_SURROGATE 0x0040
519 #define UTF8_DISALLOW_NONCHAR 0x0080 /* Unicode non-character */
520 #define UTF8_WARN_NONCHAR 0x0100 /* code points */
522 #define UTF8_DISALLOW_SUPER 0x0200 /* Super-set of Unicode: code */
523 #define UTF8_WARN_SUPER 0x0400 /* points above the legal max */
525 /* Code points which never were part of the original UTF-8 standard, the first
526 * byte of which is a FE or FF on ASCII platforms. If the first byte is FF, it
527 * will overflow a 32-bit word. If the first byte is FE, it will overflow a
528 * signed 32-bit word. */
529 #define UTF8_DISALLOW_FE_FF 0x0800
530 #define UTF8_WARN_FE_FF 0x1000
532 #define UTF8_CHECK_ONLY 0x2000
534 /* For backwards source compatibility. They do nothing, as the default now
535 * includes what they used to mean. The first one's meaning was to allow the
536 * just the single non-character 0xFFFF */
537 #define UTF8_ALLOW_FFFF 0
538 #define UTF8_ALLOW_SURROGATE 0
540 #define UTF8_DISALLOW_ILLEGAL_INTERCHANGE (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_NONCHAR|UTF8_DISALLOW_SURROGATE|UTF8_DISALLOW_FE_FF)
541 #define UTF8_WARN_ILLEGAL_INTERCHANGE \
542 (UTF8_WARN_SUPER|UTF8_WARN_NONCHAR|UTF8_WARN_SURROGATE|UTF8_WARN_FE_FF)
543 #define UTF8_ALLOW_ANY \
544 (~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE|UTF8_WARN_ILLEGAL_INTERCHANGE))
545 #define UTF8_ALLOW_ANYUV \
547 & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE|UTF8_WARN_ILLEGAL_INTERCHANGE))
548 #define UTF8_ALLOW_DEFAULT (ckWARN(WARN_UTF8) ? 0 : \
551 /* Surrogates, non-character code points and above-Unicode code points are
552 * problematic in some contexts. This allows code that needs to check for
553 * those to to quickly exclude the vast majority of code points it will
555 #define UTF8_FIRST_PROBLEMATIC_CODE_POINT_FIRST_BYTE \
556 FIRST_SURROGATE_UTF8_FIRST_BYTE
558 /* Several of the macros below have a second parameter that is currently
559 * unused; but could be used in the future to make sure that the input is
562 #define UTF8_IS_SURROGATE(s, e) cBOOL(is_SURROGATE_utf8(s))
563 #define UTF8_IS_REPLACEMENT(s, send) cBOOL(is_REPLACEMENT_utf8_safe(s,send))
566 * U+10FFFF: \xF4\x8F\xBF\xBF \xF9\xA1\xBF\xBF\xBF max legal Unicode
567 * U+110000: \xF4\x90\x80\x80 \xF9\xA2\xA0\xA0\xA0
568 * U+110001: \xF4\x90\x80\x81 \xF9\xA2\xA0\xA0\xA1
570 * BE AWARE that this test doesn't rule out malformed code points, in
571 * particular overlongs */
572 #ifdef EBCDIC /* Both versions assume well-formed UTF8 */
573 # define UTF8_IS_SUPER(s, e) (NATIVE_UTF8_TO_I8(* (U8*) (s)) >= 0xF9 \
574 && (NATIVE_UTF8_TO_I8(* (U8*) (s)) > 0xF9 \
575 || (NATIVE_UTF8_TO_I8(* ((U8*) (s) + 1)) >= 0xA2)))
577 # define UTF8_IS_SUPER(s, e) (*(U8*) (s) >= 0xF4 \
578 && (*(U8*) (s) > 0xF4 || (*((U8*) (s) + 1) >= 0x90)))
581 /* These are now machine generated, and the 'given' clause is no longer
583 #define UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s, e) \
584 cBOOL(is_NONCHAR_utf8(s))
585 #define UTF8_IS_NONCHAR(s, e) \
586 UTF8_IS_NONCHAR_GIVEN_THAT_NON_SUPER_AND_GE_PROBLEMATIC(s, e)
588 #define UNICODE_SURROGATE_FIRST 0xD800
589 #define UNICODE_SURROGATE_LAST 0xDFFF
590 #define UNICODE_REPLACEMENT 0xFFFD
591 #define UNICODE_BYTE_ORDER_MARK 0xFEFF
593 /* Though our UTF-8 encoding can go beyond this,
594 * let's be conservative and do as Unicode says. */
595 #define PERL_UNICODE_MAX 0x10FFFF
597 #define UNICODE_WARN_SURROGATE 0x0001 /* UTF-16 surrogates */
598 #define UNICODE_WARN_NONCHAR 0x0002 /* Non-char code points */
599 #define UNICODE_WARN_SUPER 0x0004 /* Above 0x10FFFF */
600 #define UNICODE_WARN_FE_FF 0x0008 /* Above 0x10FFFF */
601 #define UNICODE_DISALLOW_SURROGATE 0x0010
602 #define UNICODE_DISALLOW_NONCHAR 0x0020
603 #define UNICODE_DISALLOW_SUPER 0x0040
604 #define UNICODE_DISALLOW_FE_FF 0x0080
605 #define UNICODE_WARN_ILLEGAL_INTERCHANGE \
606 (UNICODE_WARN_SURROGATE|UNICODE_WARN_NONCHAR|UNICODE_WARN_SUPER)
607 #define UNICODE_DISALLOW_ILLEGAL_INTERCHANGE \
608 (UNICODE_DISALLOW_SURROGATE|UNICODE_DISALLOW_NONCHAR|UNICODE_DISALLOW_SUPER)
610 /* For backward source compatibility, as are now the default */
611 #define UNICODE_ALLOW_SURROGATE 0
612 #define UNICODE_ALLOW_SUPER 0
613 #define UNICODE_ALLOW_ANY 0
615 #define UNICODE_IS_SURROGATE(c) ((c) >= UNICODE_SURROGATE_FIRST && \
616 (c) <= UNICODE_SURROGATE_LAST)
617 #define UNICODE_IS_REPLACEMENT(c) ((c) == UNICODE_REPLACEMENT)
618 #define UNICODE_IS_BYTE_ORDER_MARK(c) ((c) == UNICODE_BYTE_ORDER_MARK)
619 #define UNICODE_IS_NONCHAR(c) ((c >= 0xFDD0 && c <= 0xFDEF) \
620 /* The other noncharacters end in FFFE or FFFF, which \
621 * the mask below catches both of, but beyond the last \
622 * official unicode code point, they aren't \
623 * noncharacters, since those aren't Unicode \
624 * characters at all */ \
625 || ((((c & 0xFFFE) == 0xFFFE)) && ! UNICODE_IS_SUPER(c)))
626 #define UNICODE_IS_SUPER(c) ((c) > PERL_UNICODE_MAX)
627 #define UNICODE_IS_FE_FF(c) ((c) > 0x7FFFFFFF)
629 #define LATIN_SMALL_LETTER_SHARP_S LATIN_SMALL_LETTER_SHARP_S_NATIVE
630 #define LATIN_SMALL_LETTER_Y_WITH_DIAERESIS \
631 LATIN_SMALL_LETTER_Y_WITH_DIAERESIS_NATIVE
632 #define MICRO_SIGN MICRO_SIGN_NATIVE
633 #define LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE \
634 LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE_NATIVE
635 #define LATIN_SMALL_LETTER_A_WITH_RING_ABOVE \
636 LATIN_SMALL_LETTER_A_WITH_RING_ABOVE_NATIVE
637 #define UNICODE_GREEK_CAPITAL_LETTER_SIGMA 0x03A3
638 #define UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA 0x03C2
639 #define UNICODE_GREEK_SMALL_LETTER_SIGMA 0x03C3
640 #define GREEK_SMALL_LETTER_MU 0x03BC
641 #define GREEK_CAPITAL_LETTER_MU 0x039C /* Upper and title case
643 #define LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS 0x0178 /* Also is title case */
644 #ifdef LATIN_CAPITAL_LETTER_SHARP_S_UTF8
645 # define LATIN_CAPITAL_LETTER_SHARP_S 0x1E9E
647 #define LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE 0x130
648 #define LATIN_SMALL_LETTER_DOTLESS_I 0x131
649 #define LATIN_SMALL_LETTER_LONG_S 0x017F
650 #define LATIN_SMALL_LIGATURE_LONG_S_T 0xFB05
651 #define LATIN_SMALL_LIGATURE_ST 0xFB06
652 #define KELVIN_SIGN 0x212A
653 #define ANGSTROM_SIGN 0x212B
655 #define UNI_DISPLAY_ISPRINT 0x0001
656 #define UNI_DISPLAY_BACKSLASH 0x0002
657 #define UNI_DISPLAY_QQ (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
658 #define UNI_DISPLAY_REGEX (UNI_DISPLAY_ISPRINT|UNI_DISPLAY_BACKSLASH)
660 #define ANYOF_FOLD_SHARP_S(node, input, end) \
661 (ANYOF_BITMAP_TEST(node, LATIN_SMALL_LETTER_SHARP_S) && \
662 (ANYOF_NONBITMAP(node)) && \
663 (ANYOF_FLAGS(node) & ANYOF_LOC_NONBITMAP_FOLD) && \
664 ((end) > (input) + 1) && \
665 isALPHA_FOLD_EQ((input)[0], 's'))
667 #define SHARP_S_SKIP 2
669 /* If you want to exclude surrogates, and beyond legal Unicode, see the blame
670 * log for earlier versions which gave details for these */
672 /* A helper macro for isUTF8_CHAR, so use that one, and not this one. This is
673 * retained solely for backwards compatibility and may be deprecated and
674 * removed in a future Perl version.
676 * regen/regcharclass.pl generates is_UTF8_CHAR_utf8() macros for up to these
677 * number of bytes. So this has to be coordinated with that file */
679 # define IS_UTF8_CHAR_FAST(n) ((n) <= 3)
681 # define IS_UTF8_CHAR_FAST(n) ((n) <= 4)
685 /* A helper macro for isUTF8_CHAR, so use that one instead of this. This was
686 * generated by regen/regcharclass.pl, and then moved here. The lines that
687 * generated it were then commented out. This was done solely because it takes
688 * on the order of 10 minutes to generate, and is never going to change, unless
689 * the generated code is improved.
691 * The EBCDIC versions have been cut to not cover all of legal Unicode,
692 * otherwise they take too long to generate; besides there is a separate one
693 * for each code page, so they are in regcharclass.h instead of here */
695 UTF8_CHAR: Matches legal UTF-8 encoded characters from 2 through 4 bytes
699 /*** GENERATED CODE ***/
700 #define is_UTF8_CHAR_utf8_no_length_checks(s) \
701 ( ( 0xC2 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xDF ) ? \
702 ( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ? 2 : 0 ) \
703 : ( 0xE0 == ((U8*)s)[0] ) ? \
704 ( ( ( ( ((U8*)s)[1] & 0xE0 ) == 0xA0 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
705 : ( 0xE1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xEF ) ? \
706 ( ( ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) ? 3 : 0 )\
707 : ( 0xF0 == ((U8*)s)[0] ) ? \
708 ( ( ( ( 0x90 <= ((U8*)s)[1] && ((U8*)s)[1] <= 0xBF ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )\
709 : ( ( ( ( 0xF1 <= ((U8*)s)[0] && ((U8*)s)[0] <= 0xF7 ) && ( ( ((U8*)s)[1] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[2] & 0xC0 ) == 0x80 ) ) && ( ( ((U8*)s)[3] & 0xC0 ) == 0x80 ) ) ? 4 : 0 )
714 =for apidoc Am|STRLEN|isUTF8_CHAR|const U8 *s|const U8 *e
716 Returns the number of bytes beginning at C<s> which form a legal UTF-8 (or
717 UTF-EBCDIC) encoded character, looking no further than S<C<e - s>> bytes into
718 C<s>. Returns 0 if the sequence starting at C<s> through S<C<e - 1>> is not
721 Note that an INVARIANT character (i.e. ASCII on non-EBCDIC
722 machines) is a valid UTF-8 character.
727 #define isUTF8_CHAR(s, e) (UNLIKELY((e) <= (s)) \
729 : (UTF8_IS_INVARIANT(*s)) \
731 : UNLIKELY(((e) - (s)) < UTF8SKIP(s)) \
733 : LIKELY(IS_UTF8_CHAR_FAST(UTF8SKIP(s))) \
734 ? is_UTF8_CHAR_utf8_no_length_checks(s) \
735 : _is_utf8_char_slow(s, e))
737 #define is_utf8_char_buf(buf, buf_end) isUTF8_CHAR(buf, buf_end)
739 /* Do not use; should be deprecated. Use isUTF8_CHAR() instead; this is
740 * retained solely for backwards compatibility */
741 #define IS_UTF8_CHAR(p, n) (isUTF8_CHAR(p, (p) + (n)) == n)
746 * ex: set ts=8 sts=4 sw=4 et: