3 * Copyright (C) 2000, 2001, 2002, 2003, 2004, 2005 by Larry Wall and
6 * You may distribute under the terms of either the GNU General Public
7 * License or the Artistic License, as specified in the README file.
12 * 'What a fix!' said Sam. 'That's the one place in all the lands we've ever
13 * heard of that we don't want to see any closer; and that's the one place
14 * we're trying to get to! And that's just where we can't get, nohow.'
16 * 'Well do I understand your speech,' he answered in the same language;
17 * 'yet few strangers do so. Why then do you not speak in the Common Tongue,
18 * as is the custom in the West, if you wish to be answered?'
20 * ...the travellers perceived that the floor was paved with stones of many
21 * hues; branching runes and strange devices intertwined beneath their feet.
25 #define PERL_IN_UTF8_C
28 static const char unees[] =
29 "Malformed UTF-8 character (unexpected end of string)";
32 =head1 Unicode Support
34 This file contains various utility functions for manipulating UTF8-encoded
35 strings. For the uninitiated, this is a method of representing arbitrary
36 Unicode characters as a variable number of bytes, in such a way that
37 characters in the ASCII range are unmodified, and a zero byte never appears
38 within non-zero characters.
40 =for apidoc A|U8 *|uvuni_to_utf8_flags|U8 *d|UV uv|UV flags
42 Adds the UTF-8 representation of the Unicode codepoint C<uv> to the end
43 of the string C<d>; C<d> should be have at least C<UTF8_MAXBYTES+1> free
44 bytes available. The return value is the pointer to the byte after the
45 end of the new character. In other words,
47 d = uvuni_to_utf8_flags(d, uv, flags);
51 d = uvuni_to_utf8(d, uv);
53 (which is equivalent to)
55 d = uvuni_to_utf8_flags(d, uv, 0);
57 is the recommended Unicode-aware way of saying
65 Perl_uvuni_to_utf8_flags(pTHX_ U8 *d, UV uv, UV flags)
67 if (ckWARN(WARN_UTF8)) {
68 if (UNICODE_IS_SURROGATE(uv) &&
69 !(flags & UNICODE_ALLOW_SURROGATE))
70 Perl_warner(aTHX_ packWARN(WARN_UTF8), "UTF-16 surrogate 0x%04"UVxf, uv);
72 ((uv >= 0xFDD0 && uv <= 0xFDEF &&
73 !(flags & UNICODE_ALLOW_FDD0))
75 ((uv & 0xFFFE) == 0xFFFE && /* Either FFFE or FFFF. */
76 !(flags & UNICODE_ALLOW_FFFF))) &&
77 /* UNICODE_ALLOW_SUPER includes
78 * FFFEs and FFFFs beyond 0x10FFFF. */
79 ((uv <= PERL_UNICODE_MAX) ||
80 !(flags & UNICODE_ALLOW_SUPER))
82 Perl_warner(aTHX_ packWARN(WARN_UTF8),
83 "Unicode character 0x%04"UVxf" is illegal", uv);
85 if (UNI_IS_INVARIANT(uv)) {
86 *d++ = (U8)UTF_TO_NATIVE(uv);
91 STRLEN len = UNISKIP(uv);
94 *p-- = (U8)UTF_TO_NATIVE((uv & UTF_CONTINUATION_MASK) | UTF_CONTINUATION_MARK);
95 uv >>= UTF_ACCUMULATION_SHIFT;
97 *p = (U8)UTF_TO_NATIVE((uv & UTF_START_MASK(len)) | UTF_START_MARK(len));
100 #else /* Non loop style */
102 *d++ = (U8)(( uv >> 6) | 0xc0);
103 *d++ = (U8)(( uv & 0x3f) | 0x80);
107 *d++ = (U8)(( uv >> 12) | 0xe0);
108 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
109 *d++ = (U8)(( uv & 0x3f) | 0x80);
113 *d++ = (U8)(( uv >> 18) | 0xf0);
114 *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
115 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
116 *d++ = (U8)(( uv & 0x3f) | 0x80);
119 if (uv < 0x4000000) {
120 *d++ = (U8)(( uv >> 24) | 0xf8);
121 *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
122 *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
123 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
124 *d++ = (U8)(( uv & 0x3f) | 0x80);
127 if (uv < 0x80000000) {
128 *d++ = (U8)(( uv >> 30) | 0xfc);
129 *d++ = (U8)(((uv >> 24) & 0x3f) | 0x80);
130 *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
131 *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
132 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
133 *d++ = (U8)(( uv & 0x3f) | 0x80);
137 if (uv < UTF8_QUAD_MAX)
140 *d++ = 0xfe; /* Can't match U+FEFF! */
141 *d++ = (U8)(((uv >> 30) & 0x3f) | 0x80);
142 *d++ = (U8)(((uv >> 24) & 0x3f) | 0x80);
143 *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
144 *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
145 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
146 *d++ = (U8)(( uv & 0x3f) | 0x80);
151 *d++ = 0xff; /* Can't match U+FFFE! */
152 *d++ = 0x80; /* 6 Reserved bits */
153 *d++ = (U8)(((uv >> 60) & 0x0f) | 0x80); /* 2 Reserved bits */
154 *d++ = (U8)(((uv >> 54) & 0x3f) | 0x80);
155 *d++ = (U8)(((uv >> 48) & 0x3f) | 0x80);
156 *d++ = (U8)(((uv >> 42) & 0x3f) | 0x80);
157 *d++ = (U8)(((uv >> 36) & 0x3f) | 0x80);
158 *d++ = (U8)(((uv >> 30) & 0x3f) | 0x80);
159 *d++ = (U8)(((uv >> 24) & 0x3f) | 0x80);
160 *d++ = (U8)(((uv >> 18) & 0x3f) | 0x80);
161 *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
162 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
163 *d++ = (U8)(( uv & 0x3f) | 0x80);
167 #endif /* Loop style */
171 Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv)
173 return Perl_uvuni_to_utf8_flags(aTHX_ d, uv, 0);
178 Tests if some arbitrary number of bytes begins in a valid UTF-8
179 character. Note that an INVARIANT (i.e. ASCII) character is a valid
180 UTF-8 character. The actual number of bytes in the UTF-8 character
181 will be returned if it is valid, otherwise 0.
183 This is the "slow" version as opposed to the "fast" version which is
184 the "unrolled" IS_UTF8_CHAR(). E.g. for t/uni/class.t the speed
185 difference is a factor of 2 to 3. For lengths (UTF8SKIP(s)) of four
186 or less you should use the IS_UTF8_CHAR(), for lengths of five or more
187 you should use the _slow(). In practice this means that the _slow()
188 will be used very rarely, since the maximum Unicode code point (as of
189 Unicode 4.1) is U+10FFFF, which encodes in UTF-8 to four bytes. Only
190 the "Perl extended UTF-8" (the infamous 'v-strings') will encode into
195 S_is_utf8_char_slow(pTHX_ const U8 *s, const STRLEN len)
201 if (UTF8_IS_INVARIANT(u))
204 if (!UTF8_IS_START(u))
207 if (len < 2 || !UTF8_IS_CONTINUATION(s[1]))
213 u = NATIVE_TO_UTF(u);
215 u &= UTF_START_MASK(len);
219 if (!UTF8_IS_CONTINUATION(*s))
221 uv = UTF8_ACCUMULATE(uv, *s);
228 if ((STRLEN)UNISKIP(uv) < len)
235 =for apidoc A|STRLEN|is_utf8_char|const U8 *s
237 Tests if some arbitrary number of bytes begins in a valid UTF-8
238 character. Note that an INVARIANT (i.e. ASCII) character is a valid
239 UTF-8 character. The actual number of bytes in the UTF-8 character
240 will be returned if it is valid, otherwise 0.
244 Perl_is_utf8_char(pTHX_ const U8 *s)
246 const STRLEN len = UTF8SKIP(s);
248 if (IS_UTF8_CHAR_FAST(len))
249 return IS_UTF8_CHAR(s, len) ? len : 0;
250 #endif /* #ifdef IS_UTF8_CHAR */
251 return is_utf8_char_slow(s, len);
255 =for apidoc A|bool|is_utf8_string|const U8 *s|STRLEN len
257 Returns true if first C<len> bytes of the given string form a valid
258 UTF-8 string, false otherwise. Note that 'a valid UTF-8 string' does
259 not mean 'a string that contains code points above 0x7F encoded in UTF-8'
260 because a valid ASCII string is a valid UTF-8 string.
262 See also is_utf8_string_loclen() and is_utf8_string_loc().
268 Perl_is_utf8_string(pTHX_ const U8 *s, STRLEN len)
274 len = strlen((const char *)s);
279 /* Inline the easy bits of is_utf8_char() here for speed... */
280 if (UTF8_IS_INVARIANT(*x))
282 else if (!UTF8_IS_START(*x))
285 /* ... and call is_utf8_char() only if really needed. */
288 if (IS_UTF8_CHAR_FAST(c)) {
289 if (!IS_UTF8_CHAR(x, c))
291 } else if (!is_utf8_char_slow(x, c))
295 #endif /* #ifdef IS_UTF8_CHAR */
310 =for apidoc A|bool|is_utf8_string_loclen|const U8 *s|STRLEN len|const U8 **ep|const STRLEN *el
312 Like is_ut8_string() but stores the location of the failure (in the
313 case of "utf8ness failure") or the location s+len (in the case of
314 "utf8ness success") in the C<ep>, and the number of UTF-8
315 encoded characters in the C<el>.
317 See also is_utf8_string_loc() and is_utf8_string().
323 Perl_is_utf8_string_loclen(pTHX_ const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
330 len = strlen((const char *)s);
336 /* Inline the easy bits of is_utf8_char() here for speed... */
337 if (UTF8_IS_INVARIANT(*x))
339 else if (!UTF8_IS_START(*x))
342 /* ... and call is_utf8_char() only if really needed. */
345 if (IS_UTF8_CHAR_FAST(c)) {
346 if (!IS_UTF8_CHAR(x, c))
349 c = is_utf8_char_slow(x, c);
352 #endif /* #ifdef IS_UTF8_CHAR */
371 =for apidoc A|bool|is_utf8_string_loc|const U8 *s|STRLEN len|const U8 **ep|const STRLEN *el
373 Like is_ut8_string() but stores the location of the failure (in the
374 case of "utf8ness failure") or the location s+len (in the case of
375 "utf8ness success") in the C<ep>.
377 See also is_utf8_string_loclen() and is_utf8_string().
383 Perl_is_utf8_string_loc(pTHX_ const U8 *s, STRLEN len, const U8 **ep)
385 return is_utf8_string_loclen(s, len, ep, 0);
389 =for apidoc A|UV|utf8n_to_uvuni|const U8 *s|STRLEN curlen|STRLEN *retlen|U32 flags
391 Bottom level UTF-8 decode routine.
392 Returns the unicode code point value of the first character in the string C<s>
393 which is assumed to be in UTF-8 encoding and no longer than C<curlen>;
394 C<retlen> will be set to the length, in bytes, of that character.
396 If C<s> does not point to a well-formed UTF-8 character, the behaviour
397 is dependent on the value of C<flags>: if it contains UTF8_CHECK_ONLY,
398 it is assumed that the caller will raise a warning, and this function
399 will silently just set C<retlen> to C<-1> and return zero. If the
400 C<flags> does not contain UTF8_CHECK_ONLY, warnings about
401 malformations will be given, C<retlen> will be set to the expected
402 length of the UTF-8 character in bytes, and zero will be returned.
404 The C<flags> can also contain various flags to allow deviations from
405 the strict UTF-8 encoding (see F<utf8.h>).
407 Most code should use utf8_to_uvchr() rather than call this directly.
413 Perl_utf8n_to_uvuni(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
418 const bool dowarn = ckWARN_d(WARN_UTF8);
419 const UV startbyte = *s;
420 STRLEN expectlen = 0;
423 /* This list is a superset of the UTF8_ALLOW_XXX. */
425 #define UTF8_WARN_EMPTY 1
426 #define UTF8_WARN_CONTINUATION 2
427 #define UTF8_WARN_NON_CONTINUATION 3
428 #define UTF8_WARN_FE_FF 4
429 #define UTF8_WARN_SHORT 5
430 #define UTF8_WARN_OVERFLOW 6
431 #define UTF8_WARN_SURROGATE 7
432 #define UTF8_WARN_LONG 8
433 #define UTF8_WARN_FFFF 9 /* Also FFFE. */
436 !(flags & UTF8_ALLOW_EMPTY)) {
437 warning = UTF8_WARN_EMPTY;
441 if (UTF8_IS_INVARIANT(uv)) {
444 return (UV) (NATIVE_TO_UTF(*s));
447 if (UTF8_IS_CONTINUATION(uv) &&
448 !(flags & UTF8_ALLOW_CONTINUATION)) {
449 warning = UTF8_WARN_CONTINUATION;
453 if (UTF8_IS_START(uv) && curlen > 1 && !UTF8_IS_CONTINUATION(s[1]) &&
454 !(flags & UTF8_ALLOW_NON_CONTINUATION)) {
455 warning = UTF8_WARN_NON_CONTINUATION;
460 uv = NATIVE_TO_UTF(uv);
462 if ((uv == 0xfe || uv == 0xff) &&
463 !(flags & UTF8_ALLOW_FE_FF)) {
464 warning = UTF8_WARN_FE_FF;
469 if (!(uv & 0x20)) { len = 2; uv &= 0x1f; }
470 else if (!(uv & 0x10)) { len = 3; uv &= 0x0f; }
471 else if (!(uv & 0x08)) { len = 4; uv &= 0x07; }
472 else if (!(uv & 0x04)) { len = 5; uv &= 0x03; }
474 else if (!(uv & 0x02)) { len = 6; uv &= 0x01; }
475 else { len = 7; uv &= 0x01; }
477 else if (!(uv & 0x02)) { len = 6; uv &= 0x01; }
478 else if (!(uv & 0x01)) { len = 7; uv = 0; }
479 else { len = 13; uv = 0; } /* whoa! */
487 if ((curlen < expectlen) &&
488 !(flags & UTF8_ALLOW_SHORT)) {
489 warning = UTF8_WARN_SHORT;
498 if (!UTF8_IS_CONTINUATION(*s) &&
499 !(flags & UTF8_ALLOW_NON_CONTINUATION)) {
501 warning = UTF8_WARN_NON_CONTINUATION;
505 uv = UTF8_ACCUMULATE(uv, *s);
507 /* These cannot be allowed. */
509 if (expectlen != 13 && !(flags & UTF8_ALLOW_LONG)) {
510 warning = UTF8_WARN_LONG;
514 else { /* uv < ouv */
515 /* This cannot be allowed. */
516 warning = UTF8_WARN_OVERFLOW;
524 if (UNICODE_IS_SURROGATE(uv) &&
525 !(flags & UTF8_ALLOW_SURROGATE)) {
526 warning = UTF8_WARN_SURROGATE;
528 } else if ((expectlen > (STRLEN)UNISKIP(uv)) &&
529 !(flags & UTF8_ALLOW_LONG)) {
530 warning = UTF8_WARN_LONG;
532 } else if (UNICODE_IS_ILLEGAL(uv) &&
533 !(flags & UTF8_ALLOW_FFFF)) {
534 warning = UTF8_WARN_FFFF;
542 if (flags & UTF8_CHECK_ONLY) {
549 SV* const sv = sv_2mortal(newSVpv("Malformed UTF-8 character ", 0));
552 case 0: /* Intentionally empty. */ break;
553 case UTF8_WARN_EMPTY:
554 Perl_sv_catpv(aTHX_ sv, "(empty string)");
556 case UTF8_WARN_CONTINUATION:
557 Perl_sv_catpvf(aTHX_ sv, "(unexpected continuation byte 0x%02"UVxf", with no preceding start byte)", uv);
559 case UTF8_WARN_NON_CONTINUATION:
561 Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", immediately after start byte 0x%02"UVxf")",
562 (UV)s[1], startbyte);
564 Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", %d byte%s after start byte 0x%02"UVxf", expected %d bytes)",
565 (UV)s[1], s - s0, s - s0 > 1 ? "s" : "", startbyte, (int)expectlen);
568 case UTF8_WARN_FE_FF:
569 Perl_sv_catpvf(aTHX_ sv, "(byte 0x%02"UVxf")", uv);
571 case UTF8_WARN_SHORT:
572 Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d, after start byte 0x%02"UVxf")",
573 (int)curlen, curlen == 1 ? "" : "s", (int)expectlen, startbyte);
574 expectlen = curlen; /* distance for caller to skip */
576 case UTF8_WARN_OVERFLOW:
577 Perl_sv_catpvf(aTHX_ sv, "(overflow at 0x%"UVxf", byte 0x%02x, after start byte 0x%02"UVxf")",
580 case UTF8_WARN_SURROGATE:
581 Perl_sv_catpvf(aTHX_ sv, "(UTF-16 surrogate 0x%04"UVxf")", uv);
584 Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d, after start byte 0x%02"UVxf")",
585 (int)expectlen, expectlen == 1 ? "": "s", UNISKIP(uv), startbyte);
588 Perl_sv_catpvf(aTHX_ sv, "(character 0x%04"UVxf")", uv);
591 Perl_sv_catpv(aTHX_ sv, "(unknown reason)");
596 const char * const s = SvPVX_const(sv);
599 Perl_warner(aTHX_ packWARN(WARN_UTF8),
600 "%s in %s", s, OP_DESC(PL_op));
602 Perl_warner(aTHX_ packWARN(WARN_UTF8), "%s", s);
607 *retlen = expectlen ? expectlen : len;
613 =for apidoc A|UV|utf8_to_uvchr|const U8 *s|STRLEN *retlen
615 Returns the native character value of the first character in the string C<s>
616 which is assumed to be in UTF-8 encoding; C<retlen> will be set to the
617 length, in bytes, of that character.
619 If C<s> does not point to a well-formed UTF-8 character, zero is
620 returned and retlen is set, if possible, to -1.
626 Perl_utf8_to_uvchr(pTHX_ const U8 *s, STRLEN *retlen)
628 return Perl_utf8n_to_uvchr(aTHX_ s, UTF8_MAXBYTES, retlen,
629 ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
633 =for apidoc A|UV|utf8_to_uvuni|const U8 *s|STRLEN *retlen
635 Returns the Unicode code point of the first character in the string C<s>
636 which is assumed to be in UTF-8 encoding; C<retlen> will be set to the
637 length, in bytes, of that character.
639 This function should only be used when returned UV is considered
640 an index into the Unicode semantic tables (e.g. swashes).
642 If C<s> does not point to a well-formed UTF-8 character, zero is
643 returned and retlen is set, if possible, to -1.
649 Perl_utf8_to_uvuni(pTHX_ const U8 *s, STRLEN *retlen)
651 /* Call the low level routine asking for checks */
652 return Perl_utf8n_to_uvuni(aTHX_ s, UTF8_MAXBYTES, retlen,
653 ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
657 =for apidoc A|STRLEN|utf8_length|const U8 *s|const U8 *e
659 Return the length of the UTF-8 char encoded string C<s> in characters.
660 Stops at C<e> (inclusive). If C<e E<lt> s> or if the scan would end
661 up past C<e>, croaks.
667 Perl_utf8_length(pTHX_ const U8 *s, const U8 *e)
671 /* Note: cannot use UTF8_IS_...() too eagerly here since e.g.
672 * the bitops (especially ~) can create illegal UTF-8.
673 * In other words: in Perl UTF-8 is not just for Unicode. */
676 goto warn_and_return;
678 const U8 t = UTF8SKIP(s);
681 if (ckWARN_d(WARN_UTF8)) {
683 Perl_warner(aTHX_ packWARN(WARN_UTF8),
684 "%s in %s", unees, OP_DESC(PL_op));
686 Perl_warner(aTHX_ packWARN(WARN_UTF8), unees);
698 =for apidoc A|IV|utf8_distance|const U8 *a|const U8 *b
700 Returns the number of UTF-8 characters between the UTF-8 pointers C<a>
703 WARNING: use only if you *know* that the pointers point inside the
710 Perl_utf8_distance(pTHX_ const U8 *a, const U8 *b)
714 /* Note: cannot use UTF8_IS_...() too eagerly here since e.g.
715 * the bitops (especially ~) can create illegal UTF-8.
716 * In other words: in Perl UTF-8 is not just for Unicode. */
720 const U8 c = UTF8SKIP(a);
722 goto warn_and_return;
729 const U8 c = UTF8SKIP(b);
733 if (ckWARN_d(WARN_UTF8)) {
735 Perl_warner(aTHX_ packWARN(WARN_UTF8),
736 "%s in %s", unees, OP_DESC(PL_op));
738 Perl_warner(aTHX_ packWARN(WARN_UTF8), unees);
751 =for apidoc A|U8 *|utf8_hop|U8 *s|I32 off
753 Return the UTF-8 pointer C<s> displaced by C<off> characters, either
756 WARNING: do not use the following unless you *know* C<off> is within
757 the UTF-8 data pointed to by C<s> *and* that on entry C<s> is aligned
758 on the first byte of character or just after the last byte of a character.
764 Perl_utf8_hop(pTHX_ const U8 *s, I32 off)
766 /* Note: cannot use UTF8_IS_...() too eagerly here since e.g
767 * the bitops (especially ~) can create illegal UTF-8.
768 * In other words: in Perl UTF-8 is not just for Unicode. */
777 while (UTF8_IS_CONTINUATION(*s))
785 =for apidoc A|U8 *|utf8_to_bytes|U8 *s|STRLEN *len
787 Converts a string C<s> of length C<len> from UTF-8 into byte encoding.
788 Unlike C<bytes_to_utf8>, this over-writes the original string, and
789 updates len to contain the new length.
790 Returns zero on failure, setting C<len> to -1.
796 Perl_utf8_to_bytes(pTHX_ U8 *s, STRLEN *len)
802 /* ensure valid UTF-8 and chars < 256 before updating string */
803 for (send = s + *len; s < send; ) {
806 if (!UTF8_IS_INVARIANT(c) &&
807 (!UTF8_IS_DOWNGRADEABLE_START(c) || (s >= send)
808 || !(c = *s++) || !UTF8_IS_CONTINUATION(c))) {
817 *d++ = (U8)utf8_to_uvchr(s, &ulen);
826 =for apidoc A|U8 *|bytes_from_utf8|const U8 *s|STRLEN *len|bool *is_utf8
828 Converts a string C<s> of length C<len> from UTF-8 into byte encoding.
829 Unlike C<utf8_to_bytes> but like C<bytes_to_utf8>, returns a pointer to
830 the newly-created string, and updates C<len> to contain the new
831 length. Returns the original string if no conversion occurs, C<len>
832 is unchanged. Do nothing if C<is_utf8> points to 0. Sets C<is_utf8> to
833 0 if C<s> is converted or contains all 7bit characters.
839 Perl_bytes_from_utf8(pTHX_ const U8 *s, STRLEN *len, bool *is_utf8)
849 /* ensure valid UTF-8 and chars < 256 before converting string */
850 for (send = s + *len; s < send;) {
852 if (!UTF8_IS_INVARIANT(c)) {
853 if (UTF8_IS_DOWNGRADEABLE_START(c) && s < send &&
854 (c = *s++) && UTF8_IS_CONTINUATION(c))
863 Newxz(d, (*len) - count + 1, U8);
864 s = start; start = d;
867 if (!UTF8_IS_INVARIANT(c)) {
868 /* Then it is two-byte encoded */
869 c = UTF8_ACCUMULATE(NATIVE_TO_UTF(c), *s++);
870 c = ASCII_TO_NATIVE(c);
880 =for apidoc A|U8 *|bytes_to_utf8|const U8 *s|STRLEN *len
882 Converts a string C<s> of length C<len> from ASCII into UTF-8 encoding.
883 Returns a pointer to the newly-created string, and sets C<len> to
884 reflect the new length.
886 If you want to convert to UTF-8 from other encodings than ASCII,
887 see sv_recode_to_utf8().
893 Perl_bytes_to_utf8(pTHX_ const U8 *s, STRLEN *len)
895 const U8 * const send = s + (*len);
899 Newxz(d, (*len) * 2 + 1, U8);
903 const UV uv = NATIVE_TO_ASCII(*s++);
904 if (UNI_IS_INVARIANT(uv))
905 *d++ = (U8)UTF_TO_NATIVE(uv);
907 *d++ = (U8)UTF8_EIGHT_BIT_HI(uv);
908 *d++ = (U8)UTF8_EIGHT_BIT_LO(uv);
917 * Convert native (big-endian) or reversed (little-endian) UTF-16 to UTF-8.
919 * Destination must be pre-extended to 3/2 source. Do not use in-place.
920 * We optimize for native, for obvious reasons. */
923 Perl_utf16_to_utf8(pTHX_ U8* p, U8* d, I32 bytelen, I32 *newlen)
928 if (bytelen == 1 && p[0] == 0) { /* Be understanding. */
935 Perl_croak(aTHX_ "panic: utf16_to_utf8: odd bytelen %"UVf, (UV)bytelen);
940 UV uv = (p[0] << 8) + p[1]; /* UTF-16BE */
947 *d++ = (U8)(( uv >> 6) | 0xc0);
948 *d++ = (U8)(( uv & 0x3f) | 0x80);
951 if (uv >= 0xd800 && uv < 0xdbff) { /* surrogates */
952 UV low = (p[0] << 8) + p[1];
954 if (low < 0xdc00 || low >= 0xdfff)
955 Perl_croak(aTHX_ "Malformed UTF-16 surrogate");
956 uv = ((uv - 0xd800) << 10) + (low - 0xdc00) + 0x10000;
959 *d++ = (U8)(( uv >> 12) | 0xe0);
960 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
961 *d++ = (U8)(( uv & 0x3f) | 0x80);
965 *d++ = (U8)(( uv >> 18) | 0xf0);
966 *d++ = (U8)(((uv >> 12) & 0x3f) | 0x80);
967 *d++ = (U8)(((uv >> 6) & 0x3f) | 0x80);
968 *d++ = (U8)(( uv & 0x3f) | 0x80);
972 *newlen = d - dstart;
976 /* Note: this one is slightly destructive of the source. */
979 Perl_utf16_to_utf8_reversed(pTHX_ U8* p, U8* d, I32 bytelen, I32 *newlen)
982 U8* send = s + bytelen;
989 return utf16_to_utf8(p, d, bytelen, newlen);
992 /* for now these are all defined (inefficiently) in terms of the utf8 versions */
995 Perl_is_uni_alnum(pTHX_ UV c)
997 U8 tmpbuf[UTF8_MAXBYTES+1];
998 uvchr_to_utf8(tmpbuf, c);
999 return is_utf8_alnum(tmpbuf);
1003 Perl_is_uni_alnumc(pTHX_ UV c)
1005 U8 tmpbuf[UTF8_MAXBYTES+1];
1006 uvchr_to_utf8(tmpbuf, c);
1007 return is_utf8_alnumc(tmpbuf);
1011 Perl_is_uni_idfirst(pTHX_ UV c)
1013 U8 tmpbuf[UTF8_MAXBYTES+1];
1014 uvchr_to_utf8(tmpbuf, c);
1015 return is_utf8_idfirst(tmpbuf);
1019 Perl_is_uni_alpha(pTHX_ UV c)
1021 U8 tmpbuf[UTF8_MAXBYTES+1];
1022 uvchr_to_utf8(tmpbuf, c);
1023 return is_utf8_alpha(tmpbuf);
1027 Perl_is_uni_ascii(pTHX_ UV c)
1029 U8 tmpbuf[UTF8_MAXBYTES+1];
1030 uvchr_to_utf8(tmpbuf, c);
1031 return is_utf8_ascii(tmpbuf);
1035 Perl_is_uni_space(pTHX_ UV c)
1037 U8 tmpbuf[UTF8_MAXBYTES+1];
1038 uvchr_to_utf8(tmpbuf, c);
1039 return is_utf8_space(tmpbuf);
1043 Perl_is_uni_digit(pTHX_ UV c)
1045 U8 tmpbuf[UTF8_MAXBYTES+1];
1046 uvchr_to_utf8(tmpbuf, c);
1047 return is_utf8_digit(tmpbuf);
1051 Perl_is_uni_upper(pTHX_ UV c)
1053 U8 tmpbuf[UTF8_MAXBYTES+1];
1054 uvchr_to_utf8(tmpbuf, c);
1055 return is_utf8_upper(tmpbuf);
1059 Perl_is_uni_lower(pTHX_ UV c)
1061 U8 tmpbuf[UTF8_MAXBYTES+1];
1062 uvchr_to_utf8(tmpbuf, c);
1063 return is_utf8_lower(tmpbuf);
1067 Perl_is_uni_cntrl(pTHX_ UV c)
1069 U8 tmpbuf[UTF8_MAXBYTES+1];
1070 uvchr_to_utf8(tmpbuf, c);
1071 return is_utf8_cntrl(tmpbuf);
1075 Perl_is_uni_graph(pTHX_ UV c)
1077 U8 tmpbuf[UTF8_MAXBYTES+1];
1078 uvchr_to_utf8(tmpbuf, c);
1079 return is_utf8_graph(tmpbuf);
1083 Perl_is_uni_print(pTHX_ UV c)
1085 U8 tmpbuf[UTF8_MAXBYTES+1];
1086 uvchr_to_utf8(tmpbuf, c);
1087 return is_utf8_print(tmpbuf);
1091 Perl_is_uni_punct(pTHX_ UV c)
1093 U8 tmpbuf[UTF8_MAXBYTES+1];
1094 uvchr_to_utf8(tmpbuf, c);
1095 return is_utf8_punct(tmpbuf);
1099 Perl_is_uni_xdigit(pTHX_ UV c)
1101 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
1102 uvchr_to_utf8(tmpbuf, c);
1103 return is_utf8_xdigit(tmpbuf);
1107 Perl_to_uni_upper(pTHX_ UV c, U8* p, STRLEN *lenp)
1109 uvchr_to_utf8(p, c);
1110 return to_utf8_upper(p, p, lenp);
1114 Perl_to_uni_title(pTHX_ UV c, U8* p, STRLEN *lenp)
1116 uvchr_to_utf8(p, c);
1117 return to_utf8_title(p, p, lenp);
1121 Perl_to_uni_lower(pTHX_ UV c, U8* p, STRLEN *lenp)
1123 uvchr_to_utf8(p, c);
1124 return to_utf8_lower(p, p, lenp);
1128 Perl_to_uni_fold(pTHX_ UV c, U8* p, STRLEN *lenp)
1130 uvchr_to_utf8(p, c);
1131 return to_utf8_fold(p, p, lenp);
1134 /* for now these all assume no locale info available for Unicode > 255 */
1137 Perl_is_uni_alnum_lc(pTHX_ UV c)
1139 return is_uni_alnum(c); /* XXX no locale support yet */
1143 Perl_is_uni_alnumc_lc(pTHX_ UV c)
1145 return is_uni_alnumc(c); /* XXX no locale support yet */
1149 Perl_is_uni_idfirst_lc(pTHX_ UV c)
1151 return is_uni_idfirst(c); /* XXX no locale support yet */
1155 Perl_is_uni_alpha_lc(pTHX_ UV c)
1157 return is_uni_alpha(c); /* XXX no locale support yet */
1161 Perl_is_uni_ascii_lc(pTHX_ UV c)
1163 return is_uni_ascii(c); /* XXX no locale support yet */
1167 Perl_is_uni_space_lc(pTHX_ UV c)
1169 return is_uni_space(c); /* XXX no locale support yet */
1173 Perl_is_uni_digit_lc(pTHX_ UV c)
1175 return is_uni_digit(c); /* XXX no locale support yet */
1179 Perl_is_uni_upper_lc(pTHX_ UV c)
1181 return is_uni_upper(c); /* XXX no locale support yet */
1185 Perl_is_uni_lower_lc(pTHX_ UV c)
1187 return is_uni_lower(c); /* XXX no locale support yet */
1191 Perl_is_uni_cntrl_lc(pTHX_ UV c)
1193 return is_uni_cntrl(c); /* XXX no locale support yet */
1197 Perl_is_uni_graph_lc(pTHX_ UV c)
1199 return is_uni_graph(c); /* XXX no locale support yet */
1203 Perl_is_uni_print_lc(pTHX_ UV c)
1205 return is_uni_print(c); /* XXX no locale support yet */
1209 Perl_is_uni_punct_lc(pTHX_ UV c)
1211 return is_uni_punct(c); /* XXX no locale support yet */
1215 Perl_is_uni_xdigit_lc(pTHX_ UV c)
1217 return is_uni_xdigit(c); /* XXX no locale support yet */
1221 Perl_to_uni_upper_lc(pTHX_ U32 c)
1223 /* XXX returns only the first character -- do not use XXX */
1224 /* XXX no locale support yet */
1226 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
1227 return (U32)to_uni_upper(c, tmpbuf, &len);
1231 Perl_to_uni_title_lc(pTHX_ U32 c)
1233 /* XXX returns only the first character XXX -- do not use XXX */
1234 /* XXX no locale support yet */
1236 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
1237 return (U32)to_uni_title(c, tmpbuf, &len);
1241 Perl_to_uni_lower_lc(pTHX_ U32 c)
1243 /* XXX returns only the first character -- do not use XXX */
1244 /* XXX no locale support yet */
1246 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
1247 return (U32)to_uni_lower(c, tmpbuf, &len);
1251 Perl_is_utf8_alnum(pTHX_ const U8 *p)
1253 if (!is_utf8_char(p))
1256 /* NOTE: "IsWord", not "IsAlnum", since Alnum is a true
1257 * descendant of isalnum(3), in other words, it doesn't
1258 * contain the '_'. --jhi */
1259 PL_utf8_alnum = swash_init("utf8", "IsWord", &PL_sv_undef, 0, 0);
1260 return swash_fetch(PL_utf8_alnum, p, TRUE) != 0;
1261 /* return *p == '_' || is_utf8_alpha(p) || is_utf8_digit(p); */
1262 #ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */
1264 PL_utf8_alnum = swash_init("utf8", "",
1265 sv_2mortal(newSVpv("+utf8::IsAlpha\n+utf8::IsDigit\n005F\n",0)), 0, 0);
1266 return swash_fetch(PL_utf8_alnum, p, TRUE) != 0;
1271 Perl_is_utf8_alnumc(pTHX_ const U8 *p)
1273 if (!is_utf8_char(p))
1276 PL_utf8_alnum = swash_init("utf8", "IsAlnumC", &PL_sv_undef, 0, 0);
1277 return swash_fetch(PL_utf8_alnum, p, TRUE) != 0;
1278 /* return is_utf8_alpha(p) || is_utf8_digit(p); */
1279 #ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */
1281 PL_utf8_alnum = swash_init("utf8", "",
1282 sv_2mortal(newSVpv("+utf8::IsAlpha\n+utf8::IsDigit\n005F\n",0)), 0, 0);
1283 return swash_fetch(PL_utf8_alnum, p, TRUE) != 0;
1288 Perl_is_utf8_idfirst(pTHX_ const U8 *p) /* The naming is historical. */
1292 if (!is_utf8_char(p))
1294 if (!PL_utf8_idstart) /* is_utf8_idstart would be more logical. */
1295 PL_utf8_idstart = swash_init("utf8", "IdStart", &PL_sv_undef, 0, 0);
1296 return swash_fetch(PL_utf8_idstart, p, TRUE) != 0;
1300 Perl_is_utf8_idcont(pTHX_ const U8 *p)
1304 if (!is_utf8_char(p))
1306 if (!PL_utf8_idcont)
1307 PL_utf8_idcont = swash_init("utf8", "IdContinue", &PL_sv_undef, 0, 0);
1308 return swash_fetch(PL_utf8_idcont, p, TRUE) != 0;
1312 Perl_is_utf8_alpha(pTHX_ const U8 *p)
1314 if (!is_utf8_char(p))
1317 PL_utf8_alpha = swash_init("utf8", "IsAlpha", &PL_sv_undef, 0, 0);
1318 return swash_fetch(PL_utf8_alpha, p, TRUE) != 0;
1322 Perl_is_utf8_ascii(pTHX_ const U8 *p)
1324 if (!is_utf8_char(p))
1327 PL_utf8_ascii = swash_init("utf8", "IsAscii", &PL_sv_undef, 0, 0);
1328 return swash_fetch(PL_utf8_ascii, p, TRUE) != 0;
1332 Perl_is_utf8_space(pTHX_ const U8 *p)
1334 if (!is_utf8_char(p))
1337 PL_utf8_space = swash_init("utf8", "IsSpacePerl", &PL_sv_undef, 0, 0);
1338 return swash_fetch(PL_utf8_space, p, TRUE) != 0;
1342 Perl_is_utf8_digit(pTHX_ const U8 *p)
1344 if (!is_utf8_char(p))
1347 PL_utf8_digit = swash_init("utf8", "IsDigit", &PL_sv_undef, 0, 0);
1348 return swash_fetch(PL_utf8_digit, p, TRUE) != 0;
1352 Perl_is_utf8_upper(pTHX_ const U8 *p)
1354 if (!is_utf8_char(p))
1357 PL_utf8_upper = swash_init("utf8", "IsUppercase", &PL_sv_undef, 0, 0);
1358 return swash_fetch(PL_utf8_upper, p, TRUE) != 0;
1362 Perl_is_utf8_lower(pTHX_ const U8 *p)
1364 if (!is_utf8_char(p))
1367 PL_utf8_lower = swash_init("utf8", "IsLowercase", &PL_sv_undef, 0, 0);
1368 return swash_fetch(PL_utf8_lower, p, TRUE) != 0;
1372 Perl_is_utf8_cntrl(pTHX_ const U8 *p)
1374 if (!is_utf8_char(p))
1377 PL_utf8_cntrl = swash_init("utf8", "IsCntrl", &PL_sv_undef, 0, 0);
1378 return swash_fetch(PL_utf8_cntrl, p, TRUE) != 0;
1382 Perl_is_utf8_graph(pTHX_ const U8 *p)
1384 if (!is_utf8_char(p))
1387 PL_utf8_graph = swash_init("utf8", "IsGraph", &PL_sv_undef, 0, 0);
1388 return swash_fetch(PL_utf8_graph, p, TRUE) != 0;
1392 Perl_is_utf8_print(pTHX_ const U8 *p)
1394 if (!is_utf8_char(p))
1397 PL_utf8_print = swash_init("utf8", "IsPrint", &PL_sv_undef, 0, 0);
1398 return swash_fetch(PL_utf8_print, p, TRUE) != 0;
1402 Perl_is_utf8_punct(pTHX_ const U8 *p)
1404 if (!is_utf8_char(p))
1407 PL_utf8_punct = swash_init("utf8", "IsPunct", &PL_sv_undef, 0, 0);
1408 return swash_fetch(PL_utf8_punct, p, TRUE) != 0;
1412 Perl_is_utf8_xdigit(pTHX_ const U8 *p)
1414 if (!is_utf8_char(p))
1416 if (!PL_utf8_xdigit)
1417 PL_utf8_xdigit = swash_init("utf8", "IsXDigit", &PL_sv_undef, 0, 0);
1418 return swash_fetch(PL_utf8_xdigit, p, TRUE) != 0;
1422 Perl_is_utf8_mark(pTHX_ const U8 *p)
1424 if (!is_utf8_char(p))
1427 PL_utf8_mark = swash_init("utf8", "IsM", &PL_sv_undef, 0, 0);
1428 return swash_fetch(PL_utf8_mark, p, TRUE) != 0;
1432 =for apidoc A|UV|to_utf8_case|U8 *p|U8* ustrp|STRLEN *lenp|SV **swash|char *normal|char *special
1434 The "p" contains the pointer to the UTF-8 string encoding
1435 the character that is being converted.
1437 The "ustrp" is a pointer to the character buffer to put the
1438 conversion result to. The "lenp" is a pointer to the length
1441 The "swashp" is a pointer to the swash to use.
1443 Both the special and normal mappings are stored lib/unicore/To/Foo.pl,
1444 and loaded by SWASHGET, using lib/utf8_heavy.pl. The special (usually,
1445 but not always, a multicharacter mapping), is tried first.
1447 The "special" is a string like "utf8::ToSpecLower", which means the
1448 hash %utf8::ToSpecLower. The access to the hash is through
1449 Perl_to_utf8_case().
1451 The "normal" is a string like "ToLower" which means the swash
1457 Perl_to_utf8_case(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp,
1458 SV **swashp, const char *normal, const char *special)
1460 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
1463 const UV uv0 = utf8_to_uvchr(p, NULL);
1464 /* The NATIVE_TO_UNI() and UNI_TO_NATIVE() mappings
1465 * are necessary in EBCDIC, they are redundant no-ops
1466 * in ASCII-ish platforms, and hopefully optimized away. */
1467 const UV uv1 = NATIVE_TO_UNI(uv0);
1468 uvuni_to_utf8(tmpbuf, uv1);
1470 if (!*swashp) /* load on-demand */
1471 *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0);
1473 /* The 0xDF is the only special casing Unicode code point below 0x100. */
1474 if (special && (uv1 == 0xDF || uv1 > 0xFF)) {
1475 /* It might be "special" (sometimes, but not always,
1476 * a multicharacter mapping) */
1480 if ((hv = get_hv(special, FALSE)) &&
1481 (svp = hv_fetch(hv, (const char*)tmpbuf, UNISKIP(uv1), FALSE)) &&
1485 s = SvPV_const(*svp, len);
1487 len = uvuni_to_utf8(ustrp, NATIVE_TO_UNI(*(U8*)s)) - ustrp;
1490 /* If we have EBCDIC we need to remap the characters
1491 * since any characters in the low 256 are Unicode
1492 * code points, not EBCDIC. */
1493 U8 *t = (U8*)s, *tend = t + len, *d;
1500 UV c = utf8_to_uvchr(t, &tlen);
1502 d = uvchr_to_utf8(d, UNI_TO_NATIVE(c));
1511 d = uvchr_to_utf8(d, UNI_TO_NATIVE(*t));
1516 Copy(tmpbuf, ustrp, len, U8);
1518 Copy(s, ustrp, len, U8);
1524 if (!len && *swashp) {
1525 UV uv2 = swash_fetch(*swashp, tmpbuf, TRUE);
1528 /* It was "normal" (a single character mapping). */
1529 UV uv3 = UNI_TO_NATIVE(uv2);
1531 len = uvchr_to_utf8(ustrp, uv3) - ustrp;
1535 if (!len) /* Neither: just copy. */
1536 len = uvchr_to_utf8(ustrp, uv0) - ustrp;
1541 return len ? utf8_to_uvchr(ustrp, 0) : 0;
1545 =for apidoc A|UV|to_utf8_upper|const U8 *p|U8 *ustrp|STRLEN *lenp
1547 Convert the UTF-8 encoded character at p to its uppercase version and
1548 store that in UTF-8 in ustrp and its length in bytes in lenp. Note
1549 that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since
1550 the uppercase version may be longer than the original character.
1552 The first character of the uppercased version is returned
1553 (but note, as explained above, that there may be more.)
1558 Perl_to_utf8_upper(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
1560 return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
1561 &PL_utf8_toupper, "ToUpper", "utf8::ToSpecUpper");
1565 =for apidoc A|UV|to_utf8_title|const U8 *p|U8 *ustrp|STRLEN *lenp
1567 Convert the UTF-8 encoded character at p to its titlecase version and
1568 store that in UTF-8 in ustrp and its length in bytes in lenp. Note
1569 that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since the
1570 titlecase version may be longer than the original character.
1572 The first character of the titlecased version is returned
1573 (but note, as explained above, that there may be more.)
1578 Perl_to_utf8_title(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
1580 return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
1581 &PL_utf8_totitle, "ToTitle", "utf8::ToSpecTitle");
1585 =for apidoc A|UV|to_utf8_lower|const U8 *p|U8 *ustrp|STRLEN *lenp
1587 Convert the UTF-8 encoded character at p to its lowercase version and
1588 store that in UTF-8 in ustrp and its length in bytes in lenp. Note
1589 that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since the
1590 lowercase version may be longer than the original character.
1592 The first character of the lowercased version is returned
1593 (but note, as explained above, that there may be more.)
1598 Perl_to_utf8_lower(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
1600 return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
1601 &PL_utf8_tolower, "ToLower", "utf8::ToSpecLower");
1605 =for apidoc A|UV|to_utf8_fold|const U8 *p|U8 *ustrp|STRLEN *lenp
1607 Convert the UTF-8 encoded character at p to its foldcase version and
1608 store that in UTF-8 in ustrp and its length in bytes in lenp. Note
1609 that the ustrp needs to be at least UTF8_MAXBYTES_CASE+1 bytes since the
1610 foldcase version may be longer than the original character (up to
1613 The first character of the foldcased version is returned
1614 (but note, as explained above, that there may be more.)
1619 Perl_to_utf8_fold(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp)
1621 return Perl_to_utf8_case(aTHX_ p, ustrp, lenp,
1622 &PL_utf8_tofold, "ToFold", "utf8::ToSpecFold");
1625 /* a "swash" is a swatch hash */
1628 Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits, I32 none)
1632 SV* const tokenbufsv = sv_newmortal();
1634 const size_t pkg_len = strlen(pkg);
1635 const size_t name_len = strlen(name);
1636 HV * const stash = gv_stashpvn(pkg, pkg_len, FALSE);
1639 PUSHSTACKi(PERLSI_MAGIC);
1644 if (!gv_fetchmeth(stash, "SWASHNEW", 8, -1)) { /* demand load utf8 */
1646 errsv_save = newSVsv(ERRSV);
1647 Perl_load_module(aTHX_ PERL_LOADMOD_NOIMPORT, newSVpvn(pkg,pkg_len),
1650 sv_setsv(ERRSV, errsv_save);
1651 SvREFCNT_dec(errsv_save);
1657 PUSHs(sv_2mortal(newSVpvn(pkg, pkg_len)));
1658 PUSHs(sv_2mortal(newSVpvn(name, name_len)));
1660 PUSHs(sv_2mortal(newSViv(minbits)));
1661 PUSHs(sv_2mortal(newSViv(none)));
1663 if (IN_PERL_COMPILETIME) {
1664 /* XXX ought to be handled by lex_start */
1667 sv_setpv(tokenbufsv, PL_tokenbuf);
1669 errsv_save = newSVsv(ERRSV);
1670 if (call_method("SWASHNEW", G_SCALAR))
1671 retval = newSVsv(*PL_stack_sp--);
1673 retval = &PL_sv_undef;
1675 sv_setsv(ERRSV, errsv_save);
1676 SvREFCNT_dec(errsv_save);
1679 if (IN_PERL_COMPILETIME) {
1681 const char* const pv = SvPV_const(tokenbufsv, len);
1683 Copy(pv, PL_tokenbuf, len+1, char);
1684 PL_curcop->op_private = (U8)(PL_hints & HINT_PRIVATE_MASK);
1686 if (!SvROK(retval) || SvTYPE(SvRV(retval)) != SVt_PVHV) {
1688 Perl_croak(aTHX_ "Can't find Unicode property definition \"%"SVf"\"",
1690 Perl_croak(aTHX_ "SWASHNEW didn't return an HV ref");
1696 /* This API is wrong for special case conversions since we may need to
1697 * return several Unicode characters for a single Unicode character
1698 * (see lib/unicore/SpecCase.txt) The SWASHGET in lib/utf8_heavy.pl is
1699 * the lower-level routine, and it is similarly broken for returning
1700 * multiple values. --jhi */
1702 Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8)
1705 HV* const hv = (HV*)SvRV(sv);
1710 const U8 *tmps = NULL;
1714 UV c = NATIVE_TO_ASCII(*ptr);
1716 if (!do_utf8 && !UNI_IS_INVARIANT(c)) {
1717 tmputf8[0] = (U8)UTF8_EIGHT_BIT_HI(c);
1718 tmputf8[1] = (U8)UTF8_EIGHT_BIT_LO(c);
1721 /* Given a UTF-X encoded char 0xAA..0xYY,0xZZ
1722 * then the "swatch" is a vec() for al the chars which start
1724 * So the key in the hash (klen) is length of encoded char -1
1726 klen = UTF8SKIP(ptr) - 1;
1731 /* If char in invariant then swatch is for all the invariant chars
1732 * In both UTF-8 and UTF-8-MOD that happens to be UTF_CONTINUATION_MARK
1734 needents = UTF_CONTINUATION_MARK;
1735 off = NATIVE_TO_UTF(ptr[klen]);
1739 /* If char is encoded then swatch is for the prefix */
1740 needents = (1 << UTF_ACCUMULATION_SHIFT);
1741 off = NATIVE_TO_UTF(ptr[klen]) & UTF_CONTINUATION_MASK;
1745 * This single-entry cache saves about 1/3 of the utf8 overhead in test
1746 * suite. (That is, only 7-8% overall over just a hash cache. Still,
1747 * it's nothing to sniff at.) Pity we usually come through at least
1748 * two function calls to get here...
1750 * NB: this code assumes that swatches are never modified, once generated!
1753 if (hv == PL_last_swash_hv &&
1754 klen == PL_last_swash_klen &&
1755 (!klen || memEQ((char *)ptr, (char *)PL_last_swash_key, klen)) )
1757 tmps = PL_last_swash_tmps;
1758 slen = PL_last_swash_slen;
1761 /* Try our second-level swatch cache, kept in a hash. */
1762 SV** svp = hv_fetch(hv, (const char*)ptr, klen, FALSE);
1764 /* If not cached, generate it via utf8::SWASHGET */
1765 if (!svp || !SvPOK(*svp) || !(tmps = (const U8*)SvPV_const(*svp, slen))) {
1767 /* We use utf8n_to_uvuni() as we want an index into
1768 Unicode tables, not a native character number.
1770 const UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXBYTES, 0,
1772 0 : UTF8_ALLOW_ANY);
1777 PUSHSTACKi(PERLSI_MAGIC);
1781 /* On EBCDIC & ~(0xA0-1) isn't a useful thing to do */
1782 PUSHs(sv_2mortal(newSViv((klen) ?
1783 (code_point & ~(needents - 1)) : 0)));
1784 PUSHs(sv_2mortal(newSViv(needents)));
1786 errsv_save = newSVsv(ERRSV);
1787 if (call_method("SWASHGET", G_SCALAR))
1788 retval = newSVsv(*PL_stack_sp--);
1790 retval = &PL_sv_undef;
1792 sv_setsv(ERRSV, errsv_save);
1793 SvREFCNT_dec(errsv_save);
1797 if (IN_PERL_COMPILETIME)
1798 PL_curcop->op_private = (U8)(PL_hints & HINT_PRIVATE_MASK);
1800 svp = hv_store(hv, (const char *)ptr, klen, retval, 0);
1802 if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || (slen << 3) < needents)
1803 Perl_croak(aTHX_ "SWASHGET didn't return result of proper length");
1806 PL_last_swash_hv = hv;
1807 PL_last_swash_klen = klen;
1808 /* FIXME change interpvar.h? */
1809 PL_last_swash_tmps = (U8 *) tmps;
1810 PL_last_swash_slen = slen;
1812 Copy(ptr, PL_last_swash_key, klen, U8);
1815 switch ((int)((slen << 3) / needents)) {
1817 bit = 1 << (off & 7);
1819 return (tmps[off] & bit) != 0;
1824 return (tmps[off] << 8) + tmps[off + 1] ;
1827 return (tmps[off] << 24) + (tmps[off+1] << 16) + (tmps[off+2] << 8) + tmps[off + 3] ;
1829 Perl_croak(aTHX_ "panic: swash_fetch");
1835 =for apidoc A|U8 *|uvchr_to_utf8|U8 *d|UV uv
1837 Adds the UTF-8 representation of the Native codepoint C<uv> to the end
1838 of the string C<d>; C<d> should be have at least C<UTF8_MAXBYTES+1> free
1839 bytes available. The return value is the pointer to the byte after the
1840 end of the new character. In other words,
1842 d = uvchr_to_utf8(d, uv);
1844 is the recommended wide native character-aware way of saying
1851 /* On ASCII machines this is normally a macro but we want a
1852 real function in case XS code wants it
1854 #undef Perl_uvchr_to_utf8
1856 Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv)
1858 return Perl_uvuni_to_utf8_flags(aTHX_ d, NATIVE_TO_UNI(uv), 0);
1862 Perl_uvchr_to_utf8_flags(pTHX_ U8 *d, UV uv, UV flags)
1864 return Perl_uvuni_to_utf8_flags(aTHX_ d, NATIVE_TO_UNI(uv), flags);
1868 =for apidoc A|UV|utf8n_to_uvchr|U8 *s|STRLEN curlen|STRLEN *retlen|U32 flags
1870 Returns the native character value of the first character in the string C<s>
1871 which is assumed to be in UTF-8 encoding; C<retlen> will be set to the
1872 length, in bytes, of that character.
1874 Allows length and flags to be passed to low level routine.
1878 /* On ASCII machines this is normally a macro but we want
1879 a real function in case XS code wants it
1881 #undef Perl_utf8n_to_uvchr
1883 Perl_utf8n_to_uvchr(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags)
1885 const UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags);
1886 return UNI_TO_NATIVE(uv);
1890 =for apidoc A|char *|pv_uni_display|SV *dsv|U8 *spv|STRLEN len|STRLEN pvlim|UV flags
1892 Build to the scalar dsv a displayable version of the string spv,
1893 length len, the displayable version being at most pvlim bytes long
1894 (if longer, the rest is truncated and "..." will be appended).
1896 The flags argument can have UNI_DISPLAY_ISPRINT set to display
1897 isPRINT()able characters as themselves, UNI_DISPLAY_BACKSLASH
1898 to display the \\[nrfta\\] as the backslashed versions (like '\n')
1899 (UNI_DISPLAY_BACKSLASH is preferred over UNI_DISPLAY_ISPRINT for \\).
1900 UNI_DISPLAY_QQ (and its alias UNI_DISPLAY_REGEX) have both
1901 UNI_DISPLAY_BACKSLASH and UNI_DISPLAY_ISPRINT turned on.
1903 The pointer to the PV of the dsv is returned.
1907 Perl_pv_uni_display(pTHX_ SV *dsv, const U8 *spv, STRLEN len, STRLEN pvlim, UV flags)
1912 sv_setpvn(dsv, "", 0);
1913 for (s = (const char *)spv, e = s + len; s < e; s += UTF8SKIP(s)) {
1915 /* This serves double duty as a flag and a character to print after
1916 a \ when flags & UNI_DISPLAY_BACKSLASH is true.
1920 if (pvlim && SvCUR(dsv) >= pvlim) {
1924 u = utf8_to_uvchr((U8*)s, 0);
1926 const unsigned char c = (unsigned char)u & 0xFF;
1927 if (!ok && (flags & UNI_DISPLAY_BACKSLASH)) {
1944 Perl_sv_catpvf(aTHX_ dsv, "\\%c", ok);
1947 /* isPRINT() is the locale-blind version. */
1948 if (!ok && (flags & UNI_DISPLAY_ISPRINT) && isPRINT(c)) {
1949 Perl_sv_catpvf(aTHX_ dsv, "%c", c);
1954 Perl_sv_catpvf(aTHX_ dsv, "\\x{%"UVxf"}", u);
1957 sv_catpvn(dsv, "...", 3);
1963 =for apidoc A|char *|sv_uni_display|SV *dsv|SV *ssv|STRLEN pvlim|UV flags
1965 Build to the scalar dsv a displayable version of the scalar sv,
1966 the displayable version being at most pvlim bytes long
1967 (if longer, the rest is truncated and "..." will be appended).
1969 The flags argument is as in pv_uni_display().
1971 The pointer to the PV of the dsv is returned.
1975 Perl_sv_uni_display(pTHX_ SV *dsv, SV *ssv, STRLEN pvlim, UV flags)
1977 return Perl_pv_uni_display(aTHX_ dsv, (const U8*)SvPVX_const(ssv),
1978 SvCUR(ssv), pvlim, flags);
1982 =for apidoc A|I32|ibcmp_utf8|const char *s1|char **pe1|register UV l1|bool u1|const char *s2|char **pe2|register UV l2|bool u2
1984 Return true if the strings s1 and s2 differ case-insensitively, false
1985 if not (if they are equal case-insensitively). If u1 is true, the
1986 string s1 is assumed to be in UTF-8-encoded Unicode. If u2 is true,
1987 the string s2 is assumed to be in UTF-8-encoded Unicode. If u1 or u2
1988 are false, the respective string is assumed to be in native 8-bit
1991 If the pe1 and pe2 are non-NULL, the scanning pointers will be copied
1992 in there (they will point at the beginning of the I<next> character).
1993 If the pointers behind pe1 or pe2 are non-NULL, they are the end
1994 pointers beyond which scanning will not continue under any
1995 circumstances. If the byte lengths l1 and l2 are non-zero, s1+l1 and
1996 s2+l2 will be used as goal end pointers that will also stop the scan,
1997 and which qualify towards defining a successful match: all the scans
1998 that define an explicit length must reach their goal pointers for
1999 a match to succeed).
2001 For case-insensitiveness, the "casefolding" of Unicode is used
2002 instead of upper/lowercasing both the characters, see
2003 http://www.unicode.org/unicode/reports/tr21/ (Case Mappings).
2007 Perl_ibcmp_utf8(pTHX_ const char *s1, char **pe1, register UV l1, bool u1, const char *s2, char **pe2, register UV l2, bool u2)
2009 register const U8 *p1 = (const U8*)s1;
2010 register const U8 *p2 = (const U8*)s2;
2011 register const U8 *f1 = 0, *f2 = 0;
2012 register U8 *e1 = 0, *q1 = 0;
2013 register U8 *e2 = 0, *q2 = 0;
2014 STRLEN n1 = 0, n2 = 0;
2015 U8 foldbuf1[UTF8_MAXBYTES_CASE+1];
2016 U8 foldbuf2[UTF8_MAXBYTES_CASE+1];
2018 STRLEN foldlen1, foldlen2;
2023 if (e1 == 0 || (l1 && l1 < (UV)(e1 - (const U8*)s1)))
2024 f1 = (const U8*)s1 + l1;
2027 if (e2 == 0 || (l2 && l2 < (UV)(e2 - (const U8*)s2)))
2028 f2 = (const U8*)s2 + l2;
2030 if ((e1 == 0 && f1 == 0) || (e2 == 0 && f2 == 0) || (f1 == 0 && f2 == 0))
2031 return 1; /* mismatch; possible infinite loop or false positive */
2034 natbuf[1] = 0; /* Need to terminate the buffer. */
2036 while ((e1 == 0 || p1 < e1) &&
2037 (f1 == 0 || p1 < f1) &&
2038 (e2 == 0 || p2 < e2) &&
2039 (f2 == 0 || p2 < f2)) {
2042 to_utf8_fold(p1, foldbuf1, &foldlen1);
2044 uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p1)));
2045 to_utf8_fold(natbuf, foldbuf1, &foldlen1);
2052 to_utf8_fold(p2, foldbuf2, &foldlen2);
2054 uvuni_to_utf8(natbuf, (UV) NATIVE_TO_UNI(((UV)*p2)));
2055 to_utf8_fold(natbuf, foldbuf2, &foldlen2);
2061 if ( UTF8SKIP(q1) != UTF8SKIP(q2) ||
2062 (UTF8SKIP(q1) == 1 && *q1 != *q2) ||
2063 memNE((char*)q1, (char*)q2, UTF8SKIP(q1)) )
2064 return 1; /* mismatch */
2071 p1 += u1 ? UTF8SKIP(p1) : 1;
2073 p2 += u2 ? UTF8SKIP(p2) : 1;
2077 /* A match is defined by all the scans that specified
2078 * an explicit length reaching their final goals. */
2079 match = (f1 == 0 || p1 == f1) && (f2 == 0 || p2 == f2);
2088 return match ? 0 : 1; /* 0 match, 1 mismatch */
2093 * c-indentation-style: bsd
2095 * indent-tabs-mode: t
2098 * ex: set ts=8 sts=4 sw=4 noet: