X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=utf8.c;h=60933cde8028e2b7e315bd9d09e12eff18271f01;hb=3e6618b7a3a573729d121d2ba204ea2164d47107;hp=b95c7ad164c8b9500195eb1b330e7782de14f627;hpb=2b9d42f0ba1bb562fe21327dc7948ab1a5397a19;p=p5sagit%2Fp5-mst-13.2.git diff --git a/utf8.c b/utf8.c index b95c7ad..60933cd 100644 --- a/utf8.c +++ b/utf8.c @@ -1,6 +1,6 @@ /* utf8.c * - * Copyright (c) 1998-2001, Larry Wall + * Copyright (c) 1998-2002, Larry Wall * * You may distribute under the terms of either the GNU General Public * License or the Artistic License, as specified in the README file. @@ -24,18 +24,26 @@ #define PERL_IN_UTF8_C #include "perl.h" -/* Unicode support */ +/* +=head1 Unicode Support -/* -=for apidoc A|U8*|uvuni_to_utf8|U8 *d|UV uv +=for apidoc A|U8 *|uvuni_to_utf8_flags|U8 *d|UV uv|UV flags Adds the UTF8 representation of the Unicode codepoint C to the end of the string C; C should be have at least C free bytes available. The return value is the pointer to the byte after the end of the new character. In other words, + d = uvuni_to_utf8_flags(d, uv, flags); + +or, in most cases, + d = uvuni_to_utf8(d, uv); +(which is equivalent to) + + d = uvuni_to_utf8_flags(d, uv, 0); + is the recommended Unicode-aware way of saying *(d++) = uv; @@ -44,13 +52,34 @@ is the recommended Unicode-aware way of saying */ U8 * -Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv) -{ +Perl_uvuni_to_utf8_flags(pTHX_ U8 *d, UV uv, UV flags) +{ + if (ckWARN(WARN_UTF8)) { + if (UNICODE_IS_SURROGATE(uv) && + !(flags & UNICODE_ALLOW_SURROGATE)) + Perl_warner(aTHX_ WARN_UTF8, "UTF-16 surrogate 0x%04"UVxf, uv); + else if ( + ((uv >= 0xFDD0 && uv <= 0xFDEF && + !(flags & UNICODE_ALLOW_FDD0)) + || + ((uv & 0xFFFF) == 0xFFFE && + !(flags & UNICODE_ALLOW_FFFE)) + || + ((uv & 0xFFFF) == 0xFFFF && + !(flags & UNICODE_ALLOW_FFFF))) && + /* UNICODE_ALLOW_SUPER includes + * FFFEs and FFFFs beyond 0x10FFFF. */ + ((uv <= PERL_UNICODE_MAX) || + !(flags & UNICODE_ALLOW_SUPER)) + ) + Perl_warner(aTHX_ WARN_UTF8, + "Unicode character 0x%04"UVxf" is illegal", uv); + } if (UNI_IS_INVARIANT(uv)) { *d++ = UTF_TO_NATIVE(uv); return d; } -#if defined(EBCDIC) || 1 /* always for testing */ +#if defined(EBCDIC) else { STRLEN len = UNISKIP(uv); U8 *p = d+len-1; @@ -130,7 +159,12 @@ Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv) #endif #endif /* Loop style */ } - + +U8 * +Perl_uvuni_to_utf8(pTHX_ U8 *d, UV uv) +{ + return Perl_uvuni_to_utf8_flags(aTHX_ d, uv, 0); +} /* @@ -141,7 +175,8 @@ character. Note that an INVARIANT (i.e. ASCII) character is a valid UTF-8 chara The actual number of bytes in the UTF-8 character will be returned if it is valid, otherwise 0. -=cut */ +=cut +*/ STRLEN Perl_is_utf8_char(pTHX_ U8 *s) { @@ -162,14 +197,14 @@ Perl_is_utf8_char(pTHX_ U8 *s) slen = len - 1; s++; - /* The initial value is dubious */ + u &= UTF_START_MASK(len); uv = u; ouv = uv; while (slen--) { if (!UTF8_IS_CONTINUATION(*s)) return 0; uv = UTF8_ACCUMULATE(uv, *s); - if (uv < ouv) + if (uv < ouv) return 0; ouv = uv; s++; @@ -236,14 +271,17 @@ the strict UTF-8 encoding (see F). Most code should use utf8_to_uvchr() rather than call this directly. -=cut */ +=cut +*/ UV -Perl_utf8n_to_uvuni(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, U32 flags) +Perl_utf8n_to_uvuni(pTHX_ U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags) { - UV uv = *s, ouv; + U8 *s0 = s; + UV uv = *s, ouv = 0; STRLEN len = 1; bool dowarn = ckWARN_d(WARN_UTF8); + UV startbyte = *s; STRLEN expectlen = 0; U32 warning = 0; @@ -386,22 +424,28 @@ malformed: Perl_sv_catpvf(aTHX_ sv, "(empty string)"); break; case UTF8_WARN_CONTINUATION: - Perl_sv_catpvf(aTHX_ sv, "(unexpected continuation byte 0x%02"UVxf")", uv); + Perl_sv_catpvf(aTHX_ sv, "(unexpected continuation byte 0x%02"UVxf", with no preceding start byte)", uv); break; case UTF8_WARN_NON_CONTINUATION: - Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf" after start byte 0x%02"UVxf")", - (UV)s[1], uv); + if (s == s0) + Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", immediately after start byte 0x%02"UVxf")", + (UV)s[1], startbyte); + else + Perl_sv_catpvf(aTHX_ sv, "(unexpected non-continuation byte 0x%02"UVxf", %d byte%s after start byte 0x%02"UVxf", expected %d bytes)", + (UV)s[1], s - s0, s - s0 > 1 ? "s" : "", startbyte, expectlen); + break; case UTF8_WARN_FE_FF: Perl_sv_catpvf(aTHX_ sv, "(byte 0x%02"UVxf")", uv); break; case UTF8_WARN_SHORT: - Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d)", - curlen, curlen == 1 ? "" : "s", expectlen); + Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d, after start byte 0x%02"UVxf")", + curlen, curlen == 1 ? "" : "s", expectlen, startbyte); + expectlen = curlen; /* distance for caller to skip */ break; case UTF8_WARN_OVERFLOW: - Perl_sv_catpvf(aTHX_ sv, "(overflow at 0x%"UVxf", byte 0x%02x)", - ouv, *s); + Perl_sv_catpvf(aTHX_ sv, "(overflow at 0x%"UVxf", byte 0x%02x, after start byte 0x%02"UVxf")", + ouv, *s, startbyte); break; case UTF8_WARN_SURROGATE: Perl_sv_catpvf(aTHX_ sv, "(UTF-16 surrogate 0x%04"UVxf")", uv); @@ -410,8 +454,8 @@ malformed: Perl_sv_catpvf(aTHX_ sv, "(byte order mark 0x%04"UVxf")", uv); break; case UTF8_WARN_LONG: - Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d)", - expectlen, expectlen == 1 ? "": "s", UNISKIP(uv)); + Perl_sv_catpvf(aTHX_ sv, "(%d byte%s, need %d, after start byte 0x%02"UVxf")", + expectlen, expectlen == 1 ? "": "s", UNISKIP(uv), startbyte); break; case UTF8_WARN_FFFF: Perl_sv_catpvf(aTHX_ sv, "(character 0x%04"UVxf")", uv); @@ -426,7 +470,7 @@ malformed: if (PL_op) Perl_warner(aTHX_ WARN_UTF8, - "%s in %s", s, PL_op_desc[PL_op->op_type]); + "%s in %s", s, OP_DESC(PL_op)); else Perl_warner(aTHX_ WARN_UTF8, "%s", s); } @@ -439,7 +483,7 @@ malformed: } /* -=for apidoc A|U8* s|utf8_to_uvchr|STRLEN *retlen +=for apidoc A|UV|utf8_to_uvchr|U8 *s|STRLEN *retlen Returns the native character value of the first character in the string C which is assumed to be in UTF8 encoding; C will be set to the @@ -452,13 +496,13 @@ returned and retlen is set, if possible, to -1. */ UV -Perl_utf8_to_uvchr(pTHX_ U8* s, STRLEN* retlen) +Perl_utf8_to_uvchr(pTHX_ U8 *s, STRLEN *retlen) { return Perl_utf8n_to_uvchr(aTHX_ s, UTF8_MAXLEN, retlen, 0); } /* -=for apidoc A|U8* s|utf8_to_uvuni|STRLEN *retlen +=for apidoc A|UV|utf8_to_uvuni|U8 *s|STRLEN *retlen Returns the Unicode code point of the first character in the string C which is assumed to be in UTF8 encoding; C will be set to the @@ -474,14 +518,14 @@ returned and retlen is set, if possible, to -1. */ UV -Perl_utf8_to_uvuni(pTHX_ U8* s, STRLEN* retlen) +Perl_utf8_to_uvuni(pTHX_ U8 *s, STRLEN *retlen) { /* Call the low level routine asking for checks */ return Perl_utf8n_to_uvuni(aTHX_ s, UTF8_MAXLEN, retlen, 0); } /* -=for apidoc A|STRLEN|utf8_length|U8* s|U8 *e +=for apidoc A|STRLEN|utf8_length|U8 *s|U8 *e Return the length of the UTF-8 char encoded string C in characters. Stops at C (inclusive). If C s> or if the scan would end @@ -491,7 +535,7 @@ up past C, croaks. */ STRLEN -Perl_utf8_length(pTHX_ U8* s, U8* e) +Perl_utf8_length(pTHX_ U8 *s, U8 *e) { STRLEN len = 0; @@ -505,7 +549,7 @@ Perl_utf8_length(pTHX_ U8* s, U8* e) U8 t = UTF8SKIP(s); if (e - s < t) - Perl_croak(aTHX_ "panic: utf8_length: s=%p (%02X) e=%p l=%d - unaligned end",s,*s,e,t); + Perl_croak(aTHX_ "panic: utf8_length: unaligned end"); s += t; len++; } @@ -522,7 +566,8 @@ and C. WARNING: use only if you *know* that the pointers point inside the same UTF-8 buffer. -=cut */ +=cut +*/ IV Perl_utf8_distance(pTHX_ U8 *a, U8 *b) @@ -558,7 +603,7 @@ Perl_utf8_distance(pTHX_ U8 *a, U8 *b) } /* -=for apidoc A|U8*|utf8_hop|U8 *s|I32 off +=for apidoc A|U8 *|utf8_hop|U8 *s|I32 off Return the UTF-8 pointer C displaced by C characters, either forward or backward. @@ -567,7 +612,8 @@ WARNING: do not use the following unless you *know* C is within the UTF-8 data pointed to by C *and* that on entry C is aligned on the first byte of character or just after the last byte of a character. -=cut */ +=cut +*/ U8 * Perl_utf8_hop(pTHX_ U8 *s, I32 off) @@ -602,7 +648,7 @@ Returns zero on failure, setting C to -1. */ U8 * -Perl_utf8_to_bytes(pTHX_ U8* s, STRLEN *len) +Perl_utf8_to_bytes(pTHX_ U8 *s, STRLEN *len) { U8 *send; U8 *d; @@ -641,10 +687,11 @@ length. Returns the original string if no conversion occurs, C is unchanged. Do nothing if C points to 0. Sets C to 0 if C is converted or contains all 7bit characters. -=cut */ +=cut +*/ U8 * -Perl_bytes_from_utf8(pTHX_ U8* s, STRLEN *len, bool *is_utf8) +Perl_bytes_from_utf8(pTHX_ U8 *s, STRLEN *len, bool *is_utf8) { U8 *d; U8 *start = s; @@ -695,7 +742,7 @@ reflect the new length. */ U8* -Perl_bytes_to_utf8(pTHX_ U8* s, STRLEN *len) +Perl_bytes_to_utf8(pTHX_ U8 *s, STRLEN *len) { U8 *send; U8 *d; @@ -791,7 +838,7 @@ Perl_utf16_to_utf8_reversed(pTHX_ U8* p, U8* d, I32 bytelen, I32 *newlen) /* for now these are all defined (inefficiently) in terms of the utf8 versions */ bool -Perl_is_uni_alnum(pTHX_ U32 c) +Perl_is_uni_alnum(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -799,7 +846,7 @@ Perl_is_uni_alnum(pTHX_ U32 c) } bool -Perl_is_uni_alnumc(pTHX_ U32 c) +Perl_is_uni_alnumc(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -807,7 +854,7 @@ Perl_is_uni_alnumc(pTHX_ U32 c) } bool -Perl_is_uni_idfirst(pTHX_ U32 c) +Perl_is_uni_idfirst(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -815,7 +862,7 @@ Perl_is_uni_idfirst(pTHX_ U32 c) } bool -Perl_is_uni_alpha(pTHX_ U32 c) +Perl_is_uni_alpha(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -823,7 +870,7 @@ Perl_is_uni_alpha(pTHX_ U32 c) } bool -Perl_is_uni_ascii(pTHX_ U32 c) +Perl_is_uni_ascii(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -831,7 +878,7 @@ Perl_is_uni_ascii(pTHX_ U32 c) } bool -Perl_is_uni_space(pTHX_ U32 c) +Perl_is_uni_space(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -839,7 +886,7 @@ Perl_is_uni_space(pTHX_ U32 c) } bool -Perl_is_uni_digit(pTHX_ U32 c) +Perl_is_uni_digit(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -847,7 +894,7 @@ Perl_is_uni_digit(pTHX_ U32 c) } bool -Perl_is_uni_upper(pTHX_ U32 c) +Perl_is_uni_upper(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -855,7 +902,7 @@ Perl_is_uni_upper(pTHX_ U32 c) } bool -Perl_is_uni_lower(pTHX_ U32 c) +Perl_is_uni_lower(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -863,7 +910,7 @@ Perl_is_uni_lower(pTHX_ U32 c) } bool -Perl_is_uni_cntrl(pTHX_ U32 c) +Perl_is_uni_cntrl(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -871,7 +918,7 @@ Perl_is_uni_cntrl(pTHX_ U32 c) } bool -Perl_is_uni_graph(pTHX_ U32 c) +Perl_is_uni_graph(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -879,7 +926,7 @@ Perl_is_uni_graph(pTHX_ U32 c) } bool -Perl_is_uni_print(pTHX_ U32 c) +Perl_is_uni_print(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -887,7 +934,7 @@ Perl_is_uni_print(pTHX_ U32 c) } bool -Perl_is_uni_punct(pTHX_ U32 c) +Perl_is_uni_punct(pTHX_ UV c) { U8 tmpbuf[UTF8_MAXLEN+1]; uvchr_to_utf8(tmpbuf, (UV)c); @@ -895,119 +942,127 @@ Perl_is_uni_punct(pTHX_ U32 c) } bool -Perl_is_uni_xdigit(pTHX_ U32 c) +Perl_is_uni_xdigit(pTHX_ UV c) { - U8 tmpbuf[UTF8_MAXLEN+1]; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; uvchr_to_utf8(tmpbuf, (UV)c); return is_utf8_xdigit(tmpbuf); } -U32 -Perl_to_uni_upper(pTHX_ U32 c) +UV +Perl_to_uni_upper(pTHX_ UV c, U8* p, STRLEN *lenp) { - U8 tmpbuf[UTF8_MAXLEN+1]; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; uvchr_to_utf8(tmpbuf, (UV)c); - return to_utf8_upper(tmpbuf); + return to_utf8_upper(tmpbuf, p, lenp); } -U32 -Perl_to_uni_title(pTHX_ U32 c) +UV +Perl_to_uni_title(pTHX_ UV c, U8* p, STRLEN *lenp) { - U8 tmpbuf[UTF8_MAXLEN+1]; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; uvchr_to_utf8(tmpbuf, (UV)c); - return to_utf8_title(tmpbuf); + return to_utf8_title(tmpbuf, p, lenp); } -U32 -Perl_to_uni_lower(pTHX_ U32 c) +UV +Perl_to_uni_lower(pTHX_ UV c, U8* p, STRLEN *lenp) { - U8 tmpbuf[UTF8_MAXLEN+1]; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; uvchr_to_utf8(tmpbuf, (UV)c); - return to_utf8_lower(tmpbuf); + return to_utf8_lower(tmpbuf, p, lenp); +} + +UV +Perl_to_uni_fold(pTHX_ UV c, U8* p, STRLEN *lenp) +{ + U8 tmpbuf[UTF8_MAXLEN_FOLD+1]; + uvchr_to_utf8(tmpbuf, (UV)c); + return to_utf8_fold(tmpbuf, p, lenp); } /* for now these all assume no locale info available for Unicode > 255 */ bool -Perl_is_uni_alnum_lc(pTHX_ U32 c) +Perl_is_uni_alnum_lc(pTHX_ UV c) { return is_uni_alnum(c); /* XXX no locale support yet */ } bool -Perl_is_uni_alnumc_lc(pTHX_ U32 c) +Perl_is_uni_alnumc_lc(pTHX_ UV c) { return is_uni_alnumc(c); /* XXX no locale support yet */ } bool -Perl_is_uni_idfirst_lc(pTHX_ U32 c) +Perl_is_uni_idfirst_lc(pTHX_ UV c) { return is_uni_idfirst(c); /* XXX no locale support yet */ } bool -Perl_is_uni_alpha_lc(pTHX_ U32 c) +Perl_is_uni_alpha_lc(pTHX_ UV c) { return is_uni_alpha(c); /* XXX no locale support yet */ } bool -Perl_is_uni_ascii_lc(pTHX_ U32 c) +Perl_is_uni_ascii_lc(pTHX_ UV c) { return is_uni_ascii(c); /* XXX no locale support yet */ } bool -Perl_is_uni_space_lc(pTHX_ U32 c) +Perl_is_uni_space_lc(pTHX_ UV c) { return is_uni_space(c); /* XXX no locale support yet */ } bool -Perl_is_uni_digit_lc(pTHX_ U32 c) +Perl_is_uni_digit_lc(pTHX_ UV c) { return is_uni_digit(c); /* XXX no locale support yet */ } bool -Perl_is_uni_upper_lc(pTHX_ U32 c) +Perl_is_uni_upper_lc(pTHX_ UV c) { return is_uni_upper(c); /* XXX no locale support yet */ } bool -Perl_is_uni_lower_lc(pTHX_ U32 c) +Perl_is_uni_lower_lc(pTHX_ UV c) { return is_uni_lower(c); /* XXX no locale support yet */ } bool -Perl_is_uni_cntrl_lc(pTHX_ U32 c) +Perl_is_uni_cntrl_lc(pTHX_ UV c) { return is_uni_cntrl(c); /* XXX no locale support yet */ } bool -Perl_is_uni_graph_lc(pTHX_ U32 c) +Perl_is_uni_graph_lc(pTHX_ UV c) { return is_uni_graph(c); /* XXX no locale support yet */ } bool -Perl_is_uni_print_lc(pTHX_ U32 c) +Perl_is_uni_print_lc(pTHX_ UV c) { return is_uni_print(c); /* XXX no locale support yet */ } bool -Perl_is_uni_punct_lc(pTHX_ U32 c) +Perl_is_uni_punct_lc(pTHX_ UV c) { return is_uni_punct(c); /* XXX no locale support yet */ } bool -Perl_is_uni_xdigit_lc(pTHX_ U32 c) +Perl_is_uni_xdigit_lc(pTHX_ UV c) { return is_uni_xdigit(c); /* XXX no locale support yet */ } @@ -1015,19 +1070,31 @@ Perl_is_uni_xdigit_lc(pTHX_ U32 c) U32 Perl_to_uni_upper_lc(pTHX_ U32 c) { - return to_uni_upper(c); /* XXX no locale support yet */ + /* XXX returns only the first character -- do not use XXX */ + /* XXX no locale support yet */ + STRLEN len; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; + return (U32)to_uni_upper(c, tmpbuf, &len); } U32 Perl_to_uni_title_lc(pTHX_ U32 c) { - return to_uni_title(c); /* XXX no locale support yet */ + /* XXX returns only the first character XXX -- do not use XXX */ + /* XXX no locale support yet */ + STRLEN len; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; + return (U32)to_uni_title(c, tmpbuf, &len); } U32 Perl_to_uni_lower_lc(pTHX_ U32 c) { - return to_uni_lower(c); /* XXX no locale support yet */ + /* XXX returns only the first character -- do not use XXX */ + /* XXX no locale support yet */ + STRLEN len; + U8 tmpbuf[UTF8_MAXLEN_UCLC+1]; + return (U32)to_uni_lower(c, tmpbuf, &len); } bool @@ -1040,13 +1107,13 @@ Perl_is_utf8_alnum(pTHX_ U8 *p) * descendant of isalnum(3), in other words, it doesn't * contain the '_'. --jhi */ PL_utf8_alnum = swash_init("utf8", "IsWord", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_alnum, p); + return swash_fetch(PL_utf8_alnum, p, TRUE); /* return *p == '_' || is_utf8_alpha(p) || is_utf8_digit(p); */ #ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */ if (!PL_utf8_alnum) PL_utf8_alnum = swash_init("utf8", "", sv_2mortal(newSVpv("+utf8::IsAlpha\n+utf8::IsDigit\n005F\n",0)), 0, 0); - return swash_fetch(PL_utf8_alnum, p); + return swash_fetch(PL_utf8_alnum, p, TRUE); #endif } @@ -1057,13 +1124,13 @@ Perl_is_utf8_alnumc(pTHX_ U8 *p) return FALSE; if (!PL_utf8_alnum) PL_utf8_alnum = swash_init("utf8", "IsAlnumC", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_alnum, p); + return swash_fetch(PL_utf8_alnum, p, TRUE); /* return is_utf8_alpha(p) || is_utf8_digit(p); */ #ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */ if (!PL_utf8_alnum) PL_utf8_alnum = swash_init("utf8", "", sv_2mortal(newSVpv("+utf8::IsAlpha\n+utf8::IsDigit\n005F\n",0)), 0, 0); - return swash_fetch(PL_utf8_alnum, p); + return swash_fetch(PL_utf8_alnum, p, TRUE); #endif } @@ -1080,7 +1147,7 @@ Perl_is_utf8_alpha(pTHX_ U8 *p) return FALSE; if (!PL_utf8_alpha) PL_utf8_alpha = swash_init("utf8", "IsAlpha", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_alpha, p); + return swash_fetch(PL_utf8_alpha, p, TRUE); } bool @@ -1090,7 +1157,7 @@ Perl_is_utf8_ascii(pTHX_ U8 *p) return FALSE; if (!PL_utf8_ascii) PL_utf8_ascii = swash_init("utf8", "IsAscii", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_ascii, p); + return swash_fetch(PL_utf8_ascii, p, TRUE); } bool @@ -1100,7 +1167,7 @@ Perl_is_utf8_space(pTHX_ U8 *p) return FALSE; if (!PL_utf8_space) PL_utf8_space = swash_init("utf8", "IsSpacePerl", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_space, p); + return swash_fetch(PL_utf8_space, p, TRUE); } bool @@ -1110,7 +1177,7 @@ Perl_is_utf8_digit(pTHX_ U8 *p) return FALSE; if (!PL_utf8_digit) PL_utf8_digit = swash_init("utf8", "IsDigit", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_digit, p); + return swash_fetch(PL_utf8_digit, p, TRUE); } bool @@ -1120,7 +1187,7 @@ Perl_is_utf8_upper(pTHX_ U8 *p) return FALSE; if (!PL_utf8_upper) PL_utf8_upper = swash_init("utf8", "IsUpper", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_upper, p); + return swash_fetch(PL_utf8_upper, p, TRUE); } bool @@ -1130,7 +1197,7 @@ Perl_is_utf8_lower(pTHX_ U8 *p) return FALSE; if (!PL_utf8_lower) PL_utf8_lower = swash_init("utf8", "IsLower", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_lower, p); + return swash_fetch(PL_utf8_lower, p, TRUE); } bool @@ -1140,7 +1207,7 @@ Perl_is_utf8_cntrl(pTHX_ U8 *p) return FALSE; if (!PL_utf8_cntrl) PL_utf8_cntrl = swash_init("utf8", "IsCntrl", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_cntrl, p); + return swash_fetch(PL_utf8_cntrl, p, TRUE); } bool @@ -1150,7 +1217,7 @@ Perl_is_utf8_graph(pTHX_ U8 *p) return FALSE; if (!PL_utf8_graph) PL_utf8_graph = swash_init("utf8", "IsGraph", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_graph, p); + return swash_fetch(PL_utf8_graph, p, TRUE); } bool @@ -1160,7 +1227,7 @@ Perl_is_utf8_print(pTHX_ U8 *p) return FALSE; if (!PL_utf8_print) PL_utf8_print = swash_init("utf8", "IsPrint", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_print, p); + return swash_fetch(PL_utf8_print, p, TRUE); } bool @@ -1170,7 +1237,7 @@ Perl_is_utf8_punct(pTHX_ U8 *p) return FALSE; if (!PL_utf8_punct) PL_utf8_punct = swash_init("utf8", "IsPunct", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_punct, p); + return swash_fetch(PL_utf8_punct, p, TRUE); } bool @@ -1180,7 +1247,7 @@ Perl_is_utf8_xdigit(pTHX_ U8 *p) return FALSE; if (!PL_utf8_xdigit) PL_utf8_xdigit = swash_init("utf8", "IsXDigit", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_xdigit, p); + return swash_fetch(PL_utf8_xdigit, p, TRUE); } bool @@ -1190,40 +1257,170 @@ Perl_is_utf8_mark(pTHX_ U8 *p) return FALSE; if (!PL_utf8_mark) PL_utf8_mark = swash_init("utf8", "IsM", &PL_sv_undef, 0, 0); - return swash_fetch(PL_utf8_mark, p); + return swash_fetch(PL_utf8_mark, p, TRUE); } +/* +=for apidoc A|UV|to_utf8_case|U8 *p|U8* ustrp|STRLEN *lenp|SV **swash|char *normal|char *special + +The "p" contains the pointer to the UTF-8 string encoding +the character that is being converted. + +The "ustrp" is a pointer to the character buffer to put the +conversion result to. The "lenp" is a pointer to the length +of the result. + +The "swash" is a pointer to the swash to use. + +The "normal" is a string like "ToLower" which means the swash +$utf8::ToLower, which is stored in lib/unicore/To/Lower.pl, +and loaded by SWASHGET, using lib/utf8_heavy.pl. + +The "special" is a string like "utf8::ToSpecLower", which means +the hash %utf8::ToSpecLower, which is stored in the same file, +lib/unicore/To/Lower.pl, and also loaded by SWASHGET. The access +to the hash is by Perl_to_utf8_case(). + +=cut + */ + UV -Perl_to_utf8_upper(pTHX_ U8 *p) +Perl_to_utf8_case(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, char *normal, char *special) { UV uv; - if (!PL_utf8_toupper) - PL_utf8_toupper = swash_init("utf8", "ToUpper", &PL_sv_undef, 4, 0); - uv = swash_fetch(PL_utf8_toupper, p); - return uv ? UNI_TO_NATIVE(uv) : utf8_to_uvchr(p,0); + if (!*swashp) + *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0); + uv = swash_fetch(*swashp, p, TRUE); + if (!uv) { + HV *hv; + SV *keysv; + HE *he; + + uv = utf8_to_uvchr(p, 0); + + if ((hv = get_hv(special, FALSE)) && + (keysv = sv_2mortal(Perl_newSVpvf(aTHX_ "%04"UVXf, uv))) && + (he = hv_fetch_ent(hv, keysv, FALSE, 0))) { + SV *val = HeVAL(he); + char *s = SvPV(val, *lenp); + U8 c = *(U8*)s; + + if (*lenp > 1 || UNI_IS_INVARIANT(c)) + Copy(s, ustrp, *lenp, U8); + else { + /* something in the 0x80..0xFF range */ + ustrp[0] = UTF8_EIGHT_BIT_HI(c); + ustrp[1] = UTF8_EIGHT_BIT_LO(c); + *lenp = 2; + } +#ifdef EBCDIC + { + U8 tmpbuf[UTF8_MAXLEN_FOLD+1]; + U8 *d = tmpbuf; + U8 *t, *tend; + STRLEN tlen; + + for (t = ustrp, tend = t + *lenp; t < tend; t += tlen) { + UV c = utf8_to_uvchr(t, &tlen); + d = uvchr_to_utf8(d, UNI_TO_NATIVE(c)); + } + *lenp = d - tmpbuf; + Copy(tmpbuf, ustrp, *lenp, U8); + } +#endif + return utf8_to_uvchr(ustrp, 0); + } + uv = NATIVE_TO_UNI(uv); + } + if (lenp) + *lenp = UNISKIP(uv); + uvuni_to_utf8(ustrp, uv); + return uv; } +/* +=for apidoc A|UV|to_utf8_upper|U8 *p|U8 *ustrp|STRLEN *lenp + +Convert the UTF-8 encoded character at p to its uppercase version and +store that in UTF-8 in ustrp and its length in bytes in lenp. Note +that the ustrp needs to be at least UTF8_MAXLEN_UCLC+1 bytes since the +uppercase version may be longer than the original character (up to two +characters). + +The first character of the uppercased version is returned +(but note, as explained above, that there may be more.) + +=cut */ + UV -Perl_to_utf8_title(pTHX_ U8 *p) +Perl_to_utf8_upper(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp) { - UV uv; + return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, + &PL_utf8_toupper, "ToUpper", "utf8::ToSpecUpper"); +} + +/* +=for apidoc A|UV|to_utf8_title|U8 *p|U8 *ustrp|STRLEN *lenp + +Convert the UTF-8 encoded character at p to its titlecase version and +store that in UTF-8 in ustrp and its length in bytes in lenp. Note +that the ustrp needs to be at least UTF8_MAXLEN_UCLC+1 bytes since the +titlecase version may be longer than the original character (up to two +characters). - if (!PL_utf8_totitle) - PL_utf8_totitle = swash_init("utf8", "ToTitle", &PL_sv_undef, 4, 0); - uv = swash_fetch(PL_utf8_totitle, p); - return uv ? UNI_TO_NATIVE(uv) : utf8_to_uvchr(p,0); +The first character of the titlecased version is returned +(but note, as explained above, that there may be more.) + +=cut */ + +UV +Perl_to_utf8_title(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp) +{ + return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, + &PL_utf8_totitle, "ToTitle", "utf8::ToSpecTitle"); } +/* +=for apidoc A|UV|to_utf8_lower|U8 *p|U8 *ustrp|STRLEN *lenp + +Convert the UTF-8 encoded character at p to its lowercase version and +store that in UTF-8 in ustrp and its length in bytes in lenp. Note +that the ustrp needs to be at least UTF8_MAXLEN_UCLC+1 bytes since the +lowercase version may be longer than the original character (up to two +characters). + +The first character of the lowercased version is returned +(but note, as explained above, that there may be more.) + +=cut */ + UV -Perl_to_utf8_lower(pTHX_ U8 *p) +Perl_to_utf8_lower(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp) { - UV uv; + return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, + &PL_utf8_tolower, "ToLower", "utf8::ToSpecLower"); +} + +/* +=for apidoc A|UV|to_utf8_fold|U8 *p|U8 *ustrp|STRLEN *lenp + +Convert the UTF-8 encoded character at p to its foldcase version and +store that in UTF-8 in ustrp and its length in bytes in lenp. Note +that the ustrp needs to be at least UTF8_MAXLEN_FOLD+1 bytes since the +foldcase version may be longer than the original character (up to +three characters). - if (!PL_utf8_tolower) - PL_utf8_tolower = swash_init("utf8", "ToLower", &PL_sv_undef, 4, 0); - uv = swash_fetch(PL_utf8_tolower, p); - return uv ? UNI_TO_NATIVE(uv) : utf8_to_uvchr(p,0); +The first character of the foldcased version is returned +(but note, as explained above, that there may be more.) + +=cut */ + +UV +Perl_to_utf8_fold(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp) +{ + return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, + &PL_utf8_tofold, "ToFold", "utf8::ToSpecFold"); } /* a "swash" is a swatch hash */ @@ -1235,10 +1432,15 @@ Perl_swash_init(pTHX_ char* pkg, char* name, SV *listsv, I32 minbits, I32 none) SV* tokenbufsv = sv_2mortal(NEWSV(0,0)); dSP; HV *stash = gv_stashpvn(pkg, strlen(pkg), FALSE); + SV* errsv_save; if (!gv_fetchmeth(stash, "SWASHNEW", 8, -1)) { /* demand load utf8 */ ENTER; + errsv_save = newSVsv(ERRSV); Perl_load_module(aTHX_ PERL_LOADMOD_NOIMPORT, newSVpv(pkg,0), Nullsv); + if (!SvTRUE(ERRSV)) + sv_setsv(ERRSV, errsv_save); + SvREFCNT_dec(errsv_save); LEAVE; } SPAGAIN; @@ -1258,10 +1460,14 @@ Perl_swash_init(pTHX_ char* pkg, char* name, SV *listsv, I32 minbits, I32 none) if (PL_curcop == &PL_compiling) /* XXX ought to be handled by lex_start */ sv_setpv(tokenbufsv, PL_tokenbuf); + errsv_save = newSVsv(ERRSV); if (call_method("SWASHNEW", G_SCALAR)) retval = newSVsv(*PL_stack_sp--); else retval = &PL_sv_undef; + if (!SvTRUE(ERRSV)) + sv_setsv(ERRSV, errsv_save); + SvREFCNT_dec(errsv_save); LEAVE; POPSTACK; if (PL_curcop == &PL_compiling) { @@ -1276,17 +1482,53 @@ Perl_swash_init(pTHX_ char* pkg, char* name, SV *listsv, I32 minbits, I32 none) return retval; } + +/* This API is wrong for special case conversions since we may need to + * return several Unicode characters for a single Unicode character + * (see lib/unicore/SpecCase.txt) The SWASHGET in lib/utf8_heavy.pl is + * the lower-level routine, and it is similarly broken for returning + * multiple values. --jhi */ UV -Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) +Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr, bool do_utf8) { HV* hv = (HV*)SvRV(sv); - U32 klen = UTF8SKIP(ptr) - 1; - U32 off = ptr[klen] & 127; /* NB: 64 bit always 0 when len > 1 */ + U32 klen; + U32 off; STRLEN slen; - STRLEN needents = (klen ? 64 : 128); - U8 *tmps; + STRLEN needents; + U8 *tmps = NULL; U32 bit; SV *retval; + U8 tmputf8[2]; + UV c = NATIVE_TO_ASCII(*ptr); + + if (!do_utf8 && !UNI_IS_INVARIANT(c)) { + tmputf8[0] = UTF8_EIGHT_BIT_HI(c); + tmputf8[1] = UTF8_EIGHT_BIT_LO(c); + ptr = tmputf8; + } + /* Given a UTF-X encoded char 0xAA..0xYY,0xZZ + * then the "swatch" is a vec() for al the chars which start + * with 0xAA..0xYY + * So the key in the hash (klen) is length of encoded char -1 + */ + klen = UTF8SKIP(ptr) - 1; + off = ptr[klen]; + + if (klen == 0) + { + /* If char in invariant then swatch is for all the invariant chars + * In both UTF-8 and UTF8-MOD that happens to be UTF_CONTINUATION_MARK + */ + needents = UTF_CONTINUATION_MARK; + off = NATIVE_TO_UTF(ptr[klen]); + } + else + { + /* If char is encoded then swatch is for the prefix */ + needents = (1 << UTF_ACCUMULATION_SHIFT); + off = NATIVE_TO_UTF(ptr[klen]) & UTF_CONTINUATION_MASK; + } /* * This single-entry cache saves about 1/3 of the utf8 overhead in test @@ -1297,9 +1539,9 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) * NB: this code assumes that swatches are never modified, once generated! */ - if (hv == PL_last_swash_hv && + if (hv == PL_last_swash_hv && klen == PL_last_swash_klen && - (!klen || memEQ((char *)ptr,(char *)PL_last_swash_key,klen)) ) + (!klen || memEQ((char *)ptr, (char *)PL_last_swash_key, klen)) ) { tmps = PL_last_swash_tmps; slen = PL_last_swash_slen; @@ -1315,6 +1557,7 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) Unicode tables, not a native character number. */ UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXLEN, NULL, 0); + SV *errsv_save; ENTER; SAVETMPS; save_re_context(); @@ -1322,13 +1565,19 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) PUSHMARK(SP); EXTEND(SP,3); PUSHs((SV*)sv); - PUSHs(sv_2mortal(newSViv(code_point & ~(needents - 1)))); + /* On EBCDIC & ~(0xA0-1) isn't a useful thing to do */ + PUSHs(sv_2mortal(newSViv((klen) ? + (code_point & ~(needents - 1)) : 0))); PUSHs(sv_2mortal(newSViv(needents))); PUTBACK; + errsv_save = newSVsv(ERRSV); if (call_method("SWASHGET", G_SCALAR)) retval = newSVsv(*PL_stack_sp--); else retval = &PL_sv_undef; + if (!SvTRUE(ERRSV)) + sv_setsv(ERRSV, errsv_save); + SvREFCNT_dec(errsv_save); POPSTACK; FREETMPS; LEAVE; @@ -1337,7 +1586,7 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) svp = hv_store(hv, (char*)ptr, klen, retval, 0); - if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || slen < 8) + if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || (slen << 3) < needents) Perl_croak(aTHX_ "SWASHGET didn't return result of proper length"); } @@ -1369,7 +1618,7 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) /* -=for apidoc A|U8*|uvchr_to_utf8|U8 *d|UV uv +=for apidoc A|U8 *|uvchr_to_utf8|U8 *d|UV uv Adds the UTF8 representation of the Native codepoint C to the end of the string C; C should be have at least C free @@ -1392,12 +1641,17 @@ is the recommended wide native character-aware way of saying U8 * Perl_uvchr_to_utf8(pTHX_ U8 *d, UV uv) { - return Perl_uvuni_to_utf8(aTHX_ d, NATIVE_TO_UNI(uv)); + return Perl_uvuni_to_utf8_flags(aTHX_ d, NATIVE_TO_UNI(uv), 0); } +U8 * +Perl_uvchr_to_utf8_flags(pTHX_ U8 *d, UV uv, UV flags) +{ + return Perl_uvuni_to_utf8_flags(aTHX_ d, NATIVE_TO_UNI(uv), flags); +} /* -=for apidoc A|U8* s|utf8n_to_uvchr|STRLEN curlen, STRLEN *retlen, U32 flags +=for apidoc A|UV|utf8n_to_uvchr|U8 *s|STRLEN curlen|STRLEN *retlen|U32 flags Returns the native character value of the first character in the string C which is assumed to be in UTF8 encoding; C will be set to the @@ -1407,15 +1661,208 @@ Allows length and flags to be passed to low level routine. =cut */ -/* On ASCII machines this is normally a macro but we want a - real function in case XS code wants it +/* On ASCII machines this is normally a macro but we want + a real function in case XS code wants it */ #undef Perl_utf8n_to_uvchr UV -Perl_utf8n_to_uvchr(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, U32 flags) +Perl_utf8n_to_uvchr(pTHX_ U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags) { UV uv = Perl_utf8n_to_uvuni(aTHX_ s, curlen, retlen, flags); return UNI_TO_NATIVE(uv); } +/* +=for apidoc A|char *|pv_uni_display|SV *dsv|U8 *spv|STRLEN len|STRLEN pvlim|UV flags + +Build to the scalar dsv a displayable version of the string spv, +length len, the displayable version being at most pvlim bytes long +(if longer, the rest is truncated and "..." will be appended). + +The flags argument can have UNI_DISPLAY_ISPRINT set to display +isPRINT()able characters as themselves, UNI_DISPLAY_BACKSLASH +to display the \\[nrfta\\] as the backslashed versions (like '\n') +(UNI_DISPLAY_BACKSLASH is preferred over UNI_DISPLAY_ISPRINT for \\). +UNI_DISPLAY_QQ (and its alias UNI_DISPLAY_REGEX) have both +UNI_DISPLAY_BACKSLASH and UNI_DISPLAY_ISPRINT turned on. + +The pointer to the PV of the dsv is returned. + +=cut */ +char * +Perl_pv_uni_display(pTHX_ SV *dsv, U8 *spv, STRLEN len, STRLEN pvlim, UV flags) +{ + int truncated = 0; + char *s, *e; + + sv_setpvn(dsv, "", 0); + for (s = (char *)spv, e = s + len; s < e; s += UTF8SKIP(s)) { + UV u; + bool ok = FALSE; + + if (pvlim && SvCUR(dsv) >= pvlim) { + truncated++; + break; + } + u = utf8_to_uvchr((U8*)s, 0); + if (u < 256) { + if (!ok && (flags & UNI_DISPLAY_BACKSLASH)) { + switch (u & 0xFF) { + case '\n': + Perl_sv_catpvf(aTHX_ dsv, "\\n"); ok = TRUE; break; + case '\r': + Perl_sv_catpvf(aTHX_ dsv, "\\r"); ok = TRUE; break; + case '\t': + Perl_sv_catpvf(aTHX_ dsv, "\\t"); ok = TRUE; break; + case '\f': + Perl_sv_catpvf(aTHX_ dsv, "\\f"); ok = TRUE; break; + case '\a': + Perl_sv_catpvf(aTHX_ dsv, "\\a"); ok = TRUE; break; + case '\\': + Perl_sv_catpvf(aTHX_ dsv, "\\" ); ok = TRUE; break; + default: break; + } + } + /* isPRINT() is the locale-blind version. */ + if (!ok && (flags & UNI_DISPLAY_ISPRINT) && isPRINT(u & 0xFF)) { + Perl_sv_catpvf(aTHX_ dsv, "%c", u); + ok = TRUE; + } + } + if (!ok) + Perl_sv_catpvf(aTHX_ dsv, "\\x{%"UVxf"}", u); + } + if (truncated) + sv_catpvn(dsv, "...", 3); + + return SvPVX(dsv); +} + +/* +=for apidoc A|char *|sv_uni_display|SV *dsv|SV *ssv|STRLEN pvlim|UV flags + +Build to the scalar dsv a displayable version of the scalar sv, +the displayable version being at most pvlim bytes long +(if longer, the rest is truncated and "..." will be appended). + +The flags argument is as in pv_uni_display(). + +The pointer to the PV of the dsv is returned. + +=cut */ +char * +Perl_sv_uni_display(pTHX_ SV *dsv, SV *ssv, STRLEN pvlim, UV flags) +{ + return Perl_pv_uni_display(aTHX_ dsv, (U8*)SvPVX(ssv), SvCUR(ssv), + pvlim, flags); +} + +/* +=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 + +Return true if the strings s1 and s2 differ case-insensitively, false +if not (if they are equal case-insensitively). If u1 is true, the +string s1 is assumed to be in UTF-8-encoded Unicode. If u2 is true, +the string s2 is assumed to be in UTF-8-encoded Unicode. If u1 or u2 +are false, the respective string is assumed to be in native 8-bit +encoding. + +If the pe1 and pe2 are non-NULL, the scanning pointers will be copied +in there (they will point at the beginning of the I character). +If the pointers behind pe1 or pe2 are non-NULL, they are the end +pointers beyond which scanning will not continue under any +circustances. If the byte lengths l1 and l2 are non-zero, s1+l1 and +s2+l2 will be used as goal end pointers that will also stop the scan, +and which qualify towards defining a successful match: all the scans +that define an explicit length must reach their goal pointers for +a match to succeed). + +For case-insensitiveness, the "casefolding" of Unicode is used +instead of upper/lowercasing both the characters, see +http://www.unicode.org/unicode/reports/tr21/ (Case Mappings). + +=cut */ +I32 +Perl_ibcmp_utf8(pTHX_ const char *s1, char **pe1, register UV l1, bool u1, const char *s2, char **pe2, register UV l2, bool u2) +{ + register U8 *p1 = (U8*)s1; + register U8 *p2 = (U8*)s2; + register U8 *e1 = 0, *f1 = 0, *q1 = 0; + register U8 *e2 = 0, *f2 = 0, *q2 = 0; + STRLEN n1 = 0, n2 = 0; + U8 foldbuf1[UTF8_MAXLEN_FOLD+1]; + U8 foldbuf2[UTF8_MAXLEN_FOLD+1]; + U8 natbuf[1+1]; + STRLEN foldlen1, foldlen2; + bool match; + + if (pe1) + e1 = *(U8**)pe1; + if (e1 == 0 || (l1 && l1 < e1 - (U8*)s1)) + f1 = (U8*)s1 + l1; + if (pe2) + e2 = *(U8**)pe2; + if (e2 == 0 || (l2 && l2 < e2 - (U8*)s2)) + f2 = (U8*)s2 + l2; + + if ((e1 == 0 && f1 == 0) || (e2 == 0 && f2 == 0) || (f1 == 0 && f2 == 0)) + return 1; /* mismatch; possible infinite loop or false positive */ + + if (!u1 || !u2) + natbuf[1] = 0; /* Need to terminate the buffer. */ + + while ((e1 == 0 || p1 < e1) && + (f1 == 0 || p1 < f1) && + (e2 == 0 || p2 < e2) && + (f2 == 0 || p2 < f2)) { + if (n1 == 0) { + if (u1) + to_utf8_fold(p1, foldbuf1, &foldlen1); + else { + natbuf[0] = NATIVE_TO_UNI(*p1); + to_utf8_fold(natbuf, foldbuf1, &foldlen1); + } + q1 = foldbuf1; + n1 = foldlen1; + } + if (n2 == 0) { + if (u2) + to_utf8_fold(p2, foldbuf2, &foldlen2); + else { + natbuf[0] = NATIVE_TO_UNI(*p2); + to_utf8_fold(natbuf, foldbuf2, &foldlen2); + } + q2 = foldbuf2; + n2 = foldlen2; + } + while (n1 && n2) { + if ( UTF8SKIP(q1) != UTF8SKIP(q2) || + (UTF8SKIP(q1) == 1 && *q1 != *q2) || + memNE((char*)q1, (char*)q2, UTF8SKIP(q1)) ) + return 1; /* mismatch */ + n1 -= UTF8SKIP(q1); + q1 += UTF8SKIP(q1); + n2 -= UTF8SKIP(q2); + q2 += UTF8SKIP(q2); + } + if (n1 == 0) + p1 += u1 ? UTF8SKIP(p1) : 1; + if (n2 == 0) + p2 += u2 ? UTF8SKIP(p2) : 1; + + } + + /* A match is defined by all the scans that specified + * an explicit length reaching their final goals. */ + match = (f1 == 0 || p1 == f1) && (f2 == 0 || p2 == f2); + + if (match) { + if (pe1) + *pe1 = (char*)p1; + if (pe2) + *pe2 = (char*)p2; + } + + return match ? 0 : 1; /* 0 match, 1 mismatch */ +}