X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/0453d815b8a74697ff1e5451c27aba2fe537b8e0..8850bf83ea1d396a0a5bb6dd7e4d4b9556496409:/utf8.c diff --git a/utf8.c b/utf8.c index 4bb2e9b..7a652b4 100644 --- a/utf8.c +++ b/utf8.c @@ -1,6 +1,6 @@ /* utf8.c * - * Copyright (c) 1998-1999, Larry Wall + * Copyright (c) 1998-2000, 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. @@ -27,21 +27,24 @@ /* Unicode support */ U8 * -Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) +Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) /* the d must be UTF8_MAXLEN+1 deep */ { if (uv < 0x80) { *d++ = uv; + *d = 0; return d; } if (uv < 0x800) { *d++ = (( uv >> 6) | 0xc0); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } if (uv < 0x10000) { *d++ = (( uv >> 12) | 0xe0); *d++ = (((uv >> 6) & 0x3f) | 0x80); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } if (uv < 0x200000) { @@ -49,6 +52,7 @@ Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) *d++ = (((uv >> 12) & 0x3f) | 0x80); *d++ = (((uv >> 6) & 0x3f) | 0x80); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } if (uv < 0x4000000) { @@ -57,6 +61,7 @@ Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) *d++ = (((uv >> 12) & 0x3f) | 0x80); *d++ = (((uv >> 6) & 0x3f) | 0x80); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } if (uv < 0x80000000) { @@ -66,10 +71,11 @@ Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) *d++ = (((uv >> 12) & 0x3f) | 0x80); *d++ = (((uv >> 6) & 0x3f) | 0x80); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } -#ifdef Quad_t - if (uv < 0x2000000000) +#ifdef HAS_QUAD + if (uv < UTF8_QUAD_MAX) #endif { *d++ = 0xfe; /* Can't match U+FEFF! */ @@ -79,11 +85,17 @@ Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) *d++ = (((uv >> 12) & 0x3f) | 0x80); *d++ = (((uv >> 6) & 0x3f) | 0x80); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } -#ifdef Quad_t +#ifdef HAS_QUAD { *d++ = 0xff; /* Can't match U+FFFE! */ + *d++ = 0x80; /* 6 Reserved bits */ + *d++ = (((uv >> 60) & 0x0f) | 0x80); /* 2 Reserved bits */ + *d++ = (((uv >> 54) & 0x3f) | 0x80); + *d++ = (((uv >> 48) & 0x3f) | 0x80); + *d++ = (((uv >> 42) & 0x3f) | 0x80); *d++ = (((uv >> 36) & 0x3f) | 0x80); *d++ = (((uv >> 30) & 0x3f) | 0x80); *d++ = (((uv >> 24) & 0x3f) | 0x80); @@ -91,83 +103,363 @@ Perl_uv_to_utf8(pTHX_ U8 *d, UV uv) *d++ = (((uv >> 12) & 0x3f) | 0x80); *d++ = (((uv >> 6) & 0x3f) | 0x80); *d++ = (( uv & 0x3f) | 0x80); + *d = 0; return d; } #endif } +/* Tests if some arbitrary number of bytes begins in a valid UTF-8 character. + * The actual number of bytes in the UTF-8 character will be returned if it + * is valid, otherwise 0. */ +STRLEN +Perl_is_utf8_char(pTHX_ U8 *s) +{ + U8 u = *s; + STRLEN slen, len; + UV uv, ouv; + + if (u <= 0x7f) + return 1; + + if (u >= 0x80 && u <= 0xbf) + return 0; + + len = UTF8SKIP(s); + + if (len < 2 || (u >= 0xc0 && u <= 0xfd && s[1] < 0x80)) + return 0; + + slen = len - 1; + s++; + uv = u; + ouv = uv; + while (slen--) { + if ((*s & 0xc0) != 0x80) + return 0; + uv = UTF8_ACCUMULATE(uv, *s); + if (uv < ouv) + return 0; + ouv = uv; + s++; + } + + if (UNISKIP(uv) < len) + return 0; + + return len; +} + +/* +=for apidoc Am|is_utf8_string|U8 *s|STRLEN len + +Returns true if first C bytes of the given string form valid a UTF8 +string, false otherwise. + +=cut +*/ + +bool +Perl_is_utf8_string(pTHX_ U8 *s, STRLEN len) +{ + U8* x = s; + U8* send = s + len; + STRLEN c; + + while (x < send) { + c = is_utf8_char(x); + if (!c) + return FALSE; + x += c; + if (x > send) + return FALSE; + } + + return TRUE; +} + +/* +=for apidoc Am|U8* s|utf8_to_uv|STRLEN curlen|STRLEN *retlen|U32 flags + +Returns the character value of the first character in the string C +which is assumed to be in UTF8 encoding and no longer than C; +C will be set to the length, in bytes, of that character, +and the pointer C will be advanced to the end of the character. + +If C does not point to a well-formed UTF8 character, the behaviour +is dependent on the value of C: if it contains UTF8_CHECK_ONLY, +it is assumed that the caller will raise a warning, and this function +will set C to C<-1> and return zero. If the C does not +contain UTF8_CHECK_ONLY, the UNICODE_REPLACEMENT (0xFFFD) will be +returned, and C will be set to the expected length of the +UTF-8 character in bytes. The C can also contain various flags +to allow deviations from the strict UTF-8 encoding (see F). + +=cut */ + UV -Perl_utf8_to_uv(pTHX_ U8* s, I32* retlen) +Perl_utf8_to_uv(pTHX_ U8* s, STRLEN curlen, STRLEN* retlen, U32 flags) { - UV uv = *s; - int len; - if (!(uv & 0x80)) { - if (retlen) - *retlen = 1; - return *s; + UV uv = *s, ouv; + STRLEN len = 1; +#ifdef EBCDIC + bool dowarn = 0; +#else + bool dowarn = ckWARN_d(WARN_UTF8); +#endif + STRLEN expectlen = 0; + + if (curlen == 0) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (an empty string)"); + goto malformed; } - if (!(uv & 0x40)) { - dTHR; - if (ckWARN_d(WARN_UTF8)) - Perl_warner(aTHX_ WARN_UTF8, "Malformed UTF-8 character"); + + if (UTF8_IS_ASCII(uv)) { if (retlen) *retlen = 1; return *s; } - if (!(uv & 0x20)) { len = 2; uv &= 0x1f; } - else if (!(uv & 0x10)) { len = 3; uv &= 0x0f; } - else if (!(uv & 0x08)) { len = 4; uv &= 0x07; } - else if (!(uv & 0x04)) { len = 5; uv &= 0x03; } - else if (!(uv & 0x02)) { len = 6; uv &= 0x01; } - else if (!(uv & 0x01)) { len = 7; uv &= 0x00; } - else len = 8; /* whoa! */ + if (UTF8_IS_CONTINUATION(uv) && + !(flags & UTF8_ALLOW_CONTINUATION)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (unexpected continuation byte 0x%02"UVxf")", + uv); + goto malformed; + } + if (UTF8_IS_START(uv) && curlen > 1 && !UTF8_IS_CONTINUATION(s[1]) && + !(flags & UTF8_ALLOW_NON_CONTINUATION)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (unexpected non-continuation byte 0x%02"UVxf" after start byte 0x%02"UVxf")", + (UV)s[1], uv); + goto malformed; + } + + if ((uv == 0xfe || uv == 0xff) && + !(flags & UTF8_ALLOW_FE_FF)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (byte 0x%02"UVxf")", + uv); + goto malformed; + } + + if (!(uv & 0x20)) { len = 2; uv &= 0x1f; } + else if (!(uv & 0x10)) { len = 3; uv &= 0x0f; } + else if (!(uv & 0x08)) { len = 4; uv &= 0x07; } + else if (!(uv & 0x04)) { len = 5; uv &= 0x03; } + else if (!(uv & 0x02)) { len = 6; uv &= 0x01; } + else if (!(uv & 0x01)) { len = 7; uv = 0; } + else { len = 13; uv = 0; } /* whoa! */ + if (retlen) *retlen = len; - --len; + + expectlen = len; + + if ((curlen < expectlen) && + !(flags & UTF8_ALLOW_SHORT)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (%d byte%s, need %d)", + curlen, curlen == 1 ? "" : "s", expectlen); + goto malformed; + } + + len--; s++; + ouv = uv; + while (len--) { - if ((*s & 0xc0) != 0x80) { - dTHR; - if (ckWARN_d(WARN_UTF8)) - Perl_warner(aTHX_ WARN_UTF8, "Malformed UTF-8 character"); - if (retlen) - *retlen -= len + 1; - return 0xfffd; + if (!UTF8_IS_CONTINUATION(*s) && + !(flags & UTF8_ALLOW_NON_CONTINUATION)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (unexpected non-continuation byte 0x%02x)", + *s); + goto malformed; } else - uv = (uv << 6) | (*s++ & 0x3f); + uv = UTF8_ACCUMULATE(uv, *s); + if (uv < ouv) { + /* This cannot be allowed. */ + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (overflow at 0x%"UVxf", byte 0x%02x)", + ouv, *s); + goto malformed; + } + s++; + ouv = uv; } + + if (UNICODE_IS_SURROGATE(uv) && + !(flags & UTF8_ALLOW_SURROGATE)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (UTF-16 surrogate 0x%04"UVxf")", + uv); + goto malformed; + } else if (UNICODE_IS_BYTE_ORDER_MARK(uv) && + !(flags & UTF8_ALLOW_BOM)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (byte order mark 0x%04"UVxf")", + uv); + goto malformed; + } else if ((expectlen > UNISKIP(uv)) && + !(flags & UTF8_ALLOW_LONG)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (%d byte%s, need %d)", + expectlen, expectlen == 1 ? "": "s", UNISKIP(uv)); + goto malformed; + } else if (UNICODE_IS_ILLEGAL(uv) && + !(flags & UTF8_ALLOW_FFFF)) { + if (dowarn) + Perl_warner(aTHX_ WARN_UTF8, + "Malformed UTF-8 character (character 0x%04"UVxf")", + uv); + goto malformed; + } + return uv; + +malformed: + + if (flags & UTF8_CHECK_ONLY) { + if (retlen) + *retlen = -1; + return 0; + } + + if (retlen) + *retlen = expectlen; + + return UNICODE_REPLACEMENT; } -/* utf8_distance(a,b) is intended to be a - b in pointer arithmetic */ +/* +=for apidoc Am|U8* s|utf8_to_uv_simple|STRLEN *retlen + +Returns the character value of the first character in the string C +which is assumed to be in UTF8 encoding; C will be set to the +length, in bytes, of that character, and the pointer C will be +advanced to the end of the character. -I32 +If C does not point to a well-formed UTF8 character, zero is +returned and retlen is set, if possible, to -1. + +=cut +*/ + +UV +Perl_utf8_to_uv_simple(pTHX_ U8* s, STRLEN* retlen) +{ + return Perl_utf8_to_uv(aTHX_ s, UTF8_MAXLEN, retlen, 0); +} + +/* +=for apidoc Am|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 +up past C, croaks. + +=cut +*/ + +STRLEN +Perl_utf8_length(pTHX_ U8* s, U8* e) +{ + STRLEN len = 0; + + /* Note: cannot use UTF8_IS_...() too eagerly here since e.g. + * the bitops (especially ~) can create illegal UTF-8. + * In other words: in Perl UTF-8 is not just for Unicode. */ + + if (e < s) + Perl_croak(aTHX_ "panic: utf8_length: unexpected end"); + while (s < e) { + U8 t = UTF8SKIP(s); + + if (e - s < t) + Perl_croak(aTHX_ "panic: utf8_length: unaligned end"); + s += t; + len++; + } + + return len; +} + +/* +=for apidoc Am|IV|utf8_distance|U8 *a|U8 *b + +Returns the number of UTF8 characters between the UTF-8 pointers C +and C. + +WARNING: use only if you *know* that the pointers point inside the +same UTF-8 buffer. + +=cut */ + +IV Perl_utf8_distance(pTHX_ U8 *a, U8 *b) { - I32 off = 0; + IV off = 0; + + /* Note: cannot use UTF8_IS_...() too eagerly here since e.g. + * the bitops (especially ~) can create illegal UTF-8. + * In other words: in Perl UTF-8 is not just for Unicode. */ + if (a < b) { while (a < b) { - a += UTF8SKIP(a); + U8 c = UTF8SKIP(a); + + if (b - a < c) + Perl_croak(aTHX_ "panic: utf8_distance: unaligned end"); + a += c; off--; } } else { while (b < a) { - b += UTF8SKIP(b); + U8 c = UTF8SKIP(b); + + if (a - b < c) + Perl_croak(aTHX_ "panic: utf8_distance: unaligned end"); + b += c; off++; } } + return off; } -/* WARNING: do not use the following unless you *know* off is within bounds */ +/* +=for apidoc Am|U8*|utf8_hop|U8 *s|I32 off + +Return the UTF-8 pointer C displaced by C characters, either +forward or backward. + +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 */ U8 * Perl_utf8_hop(pTHX_ U8 *s, I32 off) { + /* Note: cannot use UTF8_IS_...() too eagerly here since e.g + * the bitops (especially ~) can create illegal UTF-8. + * In other words: in Perl UTF-8 is not just for Unicode. */ + if (off >= 0) { while (off--) s += UTF8SKIP(s); @@ -175,28 +467,114 @@ Perl_utf8_hop(pTHX_ U8 *s, I32 off) else { while (off++) { s--; - if (*s & 0x80) { - while ((*s & 0xc0) == 0x80) - s--; - } + while (UTF8_IS_CONTINUATION(*s)) + s--; } } return s; } -/* XXX NOTHING CALLS THE FOLLOWING TWO ROUTINES YET!!! */ /* - * Convert native or reversed UTF-16 to UTF-8. +=for apidoc Am|U8 *|utf8_to_bytes|U8 *s|STRLEN *len + +Converts a string C of length C from UTF8 into byte encoding. +Unlike C, this over-writes the original string, and +updates len to contain the new length. +Returns zero on failure, setting C to -1. + +=cut +*/ + +U8 * +Perl_utf8_to_bytes(pTHX_ U8* s, STRLEN *len) +{ + U8 *send; + U8 *d; + U8 *save = s; + + /* ensure valid UTF8 and chars < 256 before updating string */ + for (send = s + *len; s < send; ) { + U8 c = *s++; + + if (c >= 0x80 && + ((s >= send) || + ((*s++ & 0xc0) != 0x80) || ((c & 0xfe) != 0xc2))) { + *len = -1; + return 0; + } + } + + d = s = save; + while (s < send) { + if (*s < 0x80) { + *d++ = *s++; + } + else { + STRLEN ulen; + *d++ = (U8)utf8_to_uv_simple(s, &ulen); + s += ulen; + } + } + *d = '\0'; + *len = d - save; + return save; +} + +/* +=for apidoc Am|U8 *|bytes_to_utf8|U8 *s|STRLEN *len + +Converts a string C of length C from ASCII into UTF8 encoding. +Returns a pointer to the newly-created string, and sets C to +reflect the new length. + +=cut +*/ + +U8* +Perl_bytes_to_utf8(pTHX_ U8* s, STRLEN *len) +{ + U8 *send; + U8 *d; + U8 *dst; + send = s + (*len); + + Newz(801, d, (*len) * 2 + 1, U8); + dst = d; + + while (s < send) { + if (*s < 0x80) + *d++ = *s++; + else { + UV uv = *s++; + *d++ = (( uv >> 6) | 0xc0); + *d++ = (( uv & 0x3f) | 0x80); + } + } + *d = '\0'; + *len = d-dst; + return dst; +} + +/* + * Convert native (big-endian) or reversed (little-endian) UTF-16 to UTF-8. * * Destination must be pre-extended to 3/2 source. Do not use in-place. * We optimize for native, for obvious reasons. */ U8* -Perl_utf16_to_utf8(pTHX_ U16* p, U8* d, I32 bytelen) +Perl_utf16_to_utf8(pTHX_ U8* p, U8* d, I32 bytelen, I32 *newlen) { - U16* pend = p + bytelen / 2; + U8* pend; + U8* dstart = d; + + if (bytelen & 1) + Perl_croak(aTHX_ "panic: utf16_to_utf8: odd bytelen"); + + pend = p + bytelen; + while (p < pend) { - UV uv = *p++; + UV uv = (p[0] << 8) + p[1]; /* UTF-16BE */ + p += 2; if (uv < 0x80) { *d++ = uv; continue; @@ -207,14 +585,9 @@ Perl_utf16_to_utf8(pTHX_ U16* p, U8* d, I32 bytelen) continue; } if (uv >= 0xd800 && uv < 0xdbff) { /* surrogates */ - dTHR; - int low = *p++; - if (low < 0xdc00 || low >= 0xdfff) { - if (ckWARN_d(WARN_UTF8)) - Perl_warner(aTHX_ WARN_UTF8, "Malformed UTF-16 surrogate"); - p--; - uv = 0xfffd; - } + UV low = *p++; + if (low < 0xdc00 || low >= 0xdfff) + Perl_croak(aTHX_ "Malformed UTF-16 surrogate"); uv = ((uv - 0xd800) << 10) + (low - 0xdc00) + 0x10000; } if (uv < 0x10000) { @@ -231,13 +604,14 @@ Perl_utf16_to_utf8(pTHX_ U16* p, U8* d, I32 bytelen) continue; } } + *newlen = d - dstart; return d; } /* Note: this one is slightly destructive of the source. */ U8* -Perl_utf16_to_utf8_reversed(pTHX_ U16* p, U8* d, I32 bytelen) +Perl_utf16_to_utf8_reversed(pTHX_ U8* p, U8* d, I32 bytelen, I32 *newlen) { U8* s = (U8*)p; U8* send = s + bytelen; @@ -247,7 +621,7 @@ Perl_utf16_to_utf8_reversed(pTHX_ U16* p, U8* d, I32 bytelen) s[1] = tmp; s += 2; } - return utf16_to_utf8(p, d, bytelen); + return utf16_to_utf8(p, d, bytelen, newlen); } /* for now these are all defined (inefficiently) in terms of the utf8 versions */ @@ -255,15 +629,23 @@ Perl_utf16_to_utf8_reversed(pTHX_ U16* p, U8* d, I32 bytelen) bool Perl_is_uni_alnum(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_alnum(tmpbuf); } bool +Perl_is_uni_alnumc(pTHX_ U32 c) +{ + U8 tmpbuf[UTF8_MAXLEN+1]; + uv_to_utf8(tmpbuf, (UV)c); + return is_utf8_alnumc(tmpbuf); +} + +bool Perl_is_uni_idfirst(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_idfirst(tmpbuf); } @@ -271,15 +653,23 @@ Perl_is_uni_idfirst(pTHX_ U32 c) bool Perl_is_uni_alpha(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_alpha(tmpbuf); } bool +Perl_is_uni_ascii(pTHX_ U32 c) +{ + U8 tmpbuf[UTF8_MAXLEN+1]; + uv_to_utf8(tmpbuf, (UV)c); + return is_utf8_ascii(tmpbuf); +} + +bool Perl_is_uni_space(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_space(tmpbuf); } @@ -287,7 +677,7 @@ Perl_is_uni_space(pTHX_ U32 c) bool Perl_is_uni_digit(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_digit(tmpbuf); } @@ -295,7 +685,7 @@ Perl_is_uni_digit(pTHX_ U32 c) bool Perl_is_uni_upper(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_upper(tmpbuf); } @@ -303,23 +693,55 @@ Perl_is_uni_upper(pTHX_ U32 c) bool Perl_is_uni_lower(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_lower(tmpbuf); } bool +Perl_is_uni_cntrl(pTHX_ U32 c) +{ + U8 tmpbuf[UTF8_MAXLEN+1]; + uv_to_utf8(tmpbuf, (UV)c); + return is_utf8_cntrl(tmpbuf); +} + +bool +Perl_is_uni_graph(pTHX_ U32 c) +{ + U8 tmpbuf[UTF8_MAXLEN+1]; + uv_to_utf8(tmpbuf, (UV)c); + return is_utf8_graph(tmpbuf); +} + +bool Perl_is_uni_print(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return is_utf8_print(tmpbuf); } +bool +Perl_is_uni_punct(pTHX_ U32 c) +{ + U8 tmpbuf[UTF8_MAXLEN+1]; + uv_to_utf8(tmpbuf, (UV)c); + return is_utf8_punct(tmpbuf); +} + +bool +Perl_is_uni_xdigit(pTHX_ U32 c) +{ + U8 tmpbuf[UTF8_MAXLEN+1]; + uv_to_utf8(tmpbuf, (UV)c); + return is_utf8_xdigit(tmpbuf); +} + U32 Perl_to_uni_upper(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return to_utf8_upper(tmpbuf); } @@ -327,7 +749,7 @@ Perl_to_uni_upper(pTHX_ U32 c) U32 Perl_to_uni_title(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return to_utf8_title(tmpbuf); } @@ -335,7 +757,7 @@ Perl_to_uni_title(pTHX_ U32 c) U32 Perl_to_uni_lower(pTHX_ U32 c) { - U8 tmpbuf[10]; + U8 tmpbuf[UTF8_MAXLEN+1]; uv_to_utf8(tmpbuf, (UV)c); return to_utf8_lower(tmpbuf); } @@ -349,6 +771,12 @@ Perl_is_uni_alnum_lc(pTHX_ U32 c) } bool +Perl_is_uni_alnumc_lc(pTHX_ U32 c) +{ + return is_uni_alnumc(c); /* XXX no locale support yet */ +} + +bool Perl_is_uni_idfirst_lc(pTHX_ U32 c) { return is_uni_idfirst(c); /* XXX no locale support yet */ @@ -361,6 +789,12 @@ Perl_is_uni_alpha_lc(pTHX_ U32 c) } bool +Perl_is_uni_ascii_lc(pTHX_ U32 c) +{ + return is_uni_ascii(c); /* XXX no locale support yet */ +} + +bool Perl_is_uni_space_lc(pTHX_ U32 c) { return is_uni_space(c); /* XXX no locale support yet */ @@ -385,11 +819,35 @@ Perl_is_uni_lower_lc(pTHX_ U32 c) } bool +Perl_is_uni_cntrl_lc(pTHX_ U32 c) +{ + return is_uni_cntrl(c); /* XXX no locale support yet */ +} + +bool +Perl_is_uni_graph_lc(pTHX_ U32 c) +{ + return is_uni_graph(c); /* XXX no locale support yet */ +} + +bool Perl_is_uni_print_lc(pTHX_ U32 c) { return is_uni_print(c); /* XXX no locale support yet */ } +bool +Perl_is_uni_punct_lc(pTHX_ U32 c) +{ + return is_uni_punct(c); /* XXX no locale support yet */ +} + +bool +Perl_is_uni_xdigit_lc(pTHX_ U32 c) +{ + return is_uni_xdigit(c); /* XXX no locale support yet */ +} + U32 Perl_to_uni_upper_lc(pTHX_ U32 c) { @@ -408,12 +866,16 @@ Perl_to_uni_lower_lc(pTHX_ U32 c) return to_uni_lower(c); /* XXX no locale support yet */ } - bool Perl_is_utf8_alnum(pTHX_ U8 *p) { + if (!is_utf8_char(p)) + return FALSE; if (!PL_utf8_alnum) - PL_utf8_alnum = swash_init("utf8", "IsAlnum", &PL_sv_undef, 0, 0); + /* NOTE: "IsWord", not "IsAlnum", since Alnum is a true + * 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 *p == '_' || is_utf8_alpha(p) || is_utf8_digit(p); */ #ifdef SURPRISINGLY_SLOWER /* probably because alpha is usually true */ @@ -425,6 +887,23 @@ Perl_is_utf8_alnum(pTHX_ U8 *p) } bool +Perl_is_utf8_alnumc(pTHX_ U8 *p) +{ + if (!is_utf8_char(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 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); +#endif +} + +bool Perl_is_utf8_idfirst(pTHX_ U8 *p) { return *p == '_' || is_utf8_alpha(p); @@ -433,22 +912,38 @@ Perl_is_utf8_idfirst(pTHX_ U8 *p) bool Perl_is_utf8_alpha(pTHX_ U8 *p) { + if (!is_utf8_char(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); } bool +Perl_is_utf8_ascii(pTHX_ U8 *p) +{ + if (!is_utf8_char(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); +} + +bool Perl_is_utf8_space(pTHX_ U8 *p) { + if (!is_utf8_char(p)) + return FALSE; if (!PL_utf8_space) - PL_utf8_space = swash_init("utf8", "IsSpace", &PL_sv_undef, 0, 0); + PL_utf8_space = swash_init("utf8", "IsSpacePerl", &PL_sv_undef, 0, 0); return swash_fetch(PL_utf8_space, p); } bool Perl_is_utf8_digit(pTHX_ U8 *p) { + if (!is_utf8_char(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); @@ -457,6 +952,8 @@ Perl_is_utf8_digit(pTHX_ U8 *p) bool Perl_is_utf8_upper(pTHX_ U8 *p) { + if (!is_utf8_char(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); @@ -465,22 +962,68 @@ Perl_is_utf8_upper(pTHX_ U8 *p) bool Perl_is_utf8_lower(pTHX_ U8 *p) { + if (!is_utf8_char(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); } bool +Perl_is_utf8_cntrl(pTHX_ U8 *p) +{ + if (!is_utf8_char(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); +} + +bool +Perl_is_utf8_graph(pTHX_ U8 *p) +{ + if (!is_utf8_char(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); +} + +bool Perl_is_utf8_print(pTHX_ U8 *p) { + if (!is_utf8_char(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); } bool +Perl_is_utf8_punct(pTHX_ U8 *p) +{ + if (!is_utf8_char(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); +} + +bool +Perl_is_utf8_xdigit(pTHX_ U8 *p) +{ + if (!is_utf8_char(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); +} + +bool Perl_is_utf8_mark(pTHX_ U8 *p) { + if (!is_utf8_char(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); @@ -494,7 +1037,7 @@ Perl_to_utf8_upper(pTHX_ U8 *p) 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 ? uv : utf8_to_uv(p,0); + return uv ? uv : utf8_to_uv(p,UTF8_MAXLEN,0,0); } UV @@ -505,7 +1048,7 @@ Perl_to_utf8_title(pTHX_ U8 *p) 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 ? uv : utf8_to_uv(p,0); + return uv ? uv : utf8_to_uv(p,UTF8_MAXLEN,0,0); } UV @@ -516,7 +1059,7 @@ Perl_to_utf8_lower(pTHX_ U8 *p) 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 ? uv : utf8_to_uv(p,0); + return uv ? uv : utf8_to_uv(p,UTF8_MAXLEN,0,0); } /* a "swash" is a swatch hash */ @@ -526,7 +1069,14 @@ Perl_swash_init(pTHX_ char* pkg, char* name, SV *listsv, I32 minbits, I32 none) { SV* retval; char tmpbuf[256]; - dSP; + dSP; + + if (!gv_stashpv(pkg, 0)) { /* demand load utf8 */ + ENTER; + Perl_load_module(aTHX_ PERL_LOADMOD_NOIMPORT, newSVpv(pkg,0), Nullsv); + LEAVE; + } + SPAGAIN; PUSHSTACKi(PERLSI_MAGIC); PUSHMARK(SP); EXTEND(SP,5); @@ -543,7 +1093,7 @@ 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 */ strncpy(tmpbuf, PL_tokenbuf, sizeof tmpbuf); if (call_method("SWASHNEW", G_SCALAR)) - retval = newSVsv(*PL_stack_sp--); + retval = newSVsv(*PL_stack_sp--); else retval = &PL_sv_undef; LEAVE; @@ -580,7 +1130,7 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) if (hv == PL_last_swash_hv && klen == PL_last_swash_klen && - (!klen || memEQ(ptr,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; @@ -599,11 +1149,11 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) PUSHMARK(SP); EXTEND(SP,3); PUSHs((SV*)sv); - PUSHs(sv_2mortal(newSViv(utf8_to_uv(ptr, 0) & ~(needents - 1)))); + PUSHs(sv_2mortal(newSViv(utf8_to_uv(ptr, UTF8_MAXLEN, 0, 0) & ~(needents - 1)))); PUSHs(sv_2mortal(newSViv(needents))); PUTBACK; if (call_method("SWASHGET", G_SCALAR)) - retval = newSVsv(*PL_stack_sp--); + retval = newSVsv(*PL_stack_sp--); else retval = &PL_sv_undef; POPSTACK; @@ -626,7 +1176,7 @@ Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr) Copy(ptr, PL_last_swash_key, klen, U8); } - switch ((slen << 3) / needents) { + switch ((int)((slen << 3) / needents)) { case 1: bit = 1 << (off & 7); off >>= 3;