X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=utf8.c;h=e5079cb709724d9bd6d703cb652b40d59421656a;hb=908fcb8bef8cbab8cfe098520d89599eb7d1a16c;hp=813a64fcddbdeaebc1577e8453a03bdaad44b044;hpb=979f29225180f8c09f4adec52f850ae45694fd81;p=p5sagit%2Fp5-mst-13.2.git diff --git a/utf8.c b/utf8.c index 813a64f..e5079cb 100644 --- a/utf8.c +++ b/utf8.c @@ -1,7 +1,7 @@ /* utf8.c * - * Copyright (C) 2000, 2001, 2002, 2003, 2004, 2005 by Larry Wall and - * others + * Copyright (C) 2000, 2001, 2002, 2003, 2004, 2005, 2006, + * by Larry Wall and others * * You may distribute under the terms of either the GNU General Public * License or the Artistic License, as specified in the README file. @@ -186,7 +186,7 @@ five bytes or more. =cut */ STATIC STRLEN -S_is_utf8_char_slow(pTHX_ const U8 *s, const STRLEN len) +S_is_utf8_char_slow(const U8 *s, const STRLEN len) { U8 u = *s; STRLEN slen; @@ -238,6 +238,7 @@ STRLEN Perl_is_utf8_char(pTHX_ const U8 *s) { const STRLEN len = UTF8SKIP(s); + PERL_UNUSED_CONTEXT; #ifdef IS_UTF8_CHAR if (IS_UTF8_CHAR_FAST(len)) return IS_UTF8_CHAR(s, len) ? len : 0; @@ -264,6 +265,7 @@ Perl_is_utf8_string(pTHX_ const U8 *s, STRLEN len) const U8* x = s; const U8* send; + PERL_UNUSED_CONTEXT; if (!len) len = strlen((const char *)s); send = s + len; @@ -329,6 +331,7 @@ Perl_is_utf8_string_loclen(pTHX_ const U8 *s, STRLEN len, const U8 **ep, STRLEN const U8* x = s; const U8* send; STRLEN c; + PERL_UNUSED_CONTEXT; if (!len) len = strlen((const char *)s); @@ -399,6 +402,7 @@ Most code should use utf8_to_uvchr() rather than call this directly. UV Perl_utf8n_to_uvuni(pTHX_ const U8 *s, STRLEN curlen, STRLEN *retlen, U32 flags) { + dVAR; const U8 *s0 = s; UV uv = *s, ouv = 0; STRLEN len = 1; @@ -533,12 +537,12 @@ malformed: } if (dowarn) { - SV* const sv = sv_2mortal(newSVpv("Malformed UTF-8 character ", 0)); + SV* const sv = sv_2mortal(newSVpvs("Malformed UTF-8 character ")); switch (warning) { case 0: /* Intentionally empty. */ break; case UTF8_WARN_EMPTY: - Perl_sv_catpv(aTHX_ sv, "(empty string)"); + sv_catpvs(sv, "(empty string)"); break; case UTF8_WARN_CONTINUATION: Perl_sv_catpvf(aTHX_ sv, "(unexpected continuation byte 0x%02"UVxf", with no preceding start byte)", uv); @@ -577,7 +581,7 @@ malformed: Perl_sv_catpvf(aTHX_ sv, "(character 0x%04"UVxf")", uv); break; default: - Perl_sv_catpv(aTHX_ sv, "(unknown reason)"); + sv_catpvs(sv, "(unknown reason)"); break; } @@ -655,6 +659,7 @@ up past C, croaks. STRLEN Perl_utf8_length(pTHX_ const U8 *s, const U8 *e) { + dVAR; STRLEN len = 0; /* Note: cannot use UTF8_IS_...() too eagerly here since e.g. @@ -698,6 +703,7 @@ same UTF-8 buffer. IV Perl_utf8_distance(pTHX_ const U8 *a, const U8 *b) { + dVAR; IV off = 0; /* Note: cannot use UTF8_IS_...() too eagerly here since e.g. @@ -752,6 +758,7 @@ on the first byte of character or just after the last byte of a character. U8 * Perl_utf8_hop(pTHX_ const U8 *s, I32 off) { + PERL_UNUSED_CONTEXT; /* 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. */ @@ -832,6 +839,7 @@ Perl_bytes_from_utf8(pTHX_ const U8 *s, STRLEN *len, bool *is_utf8) const U8 *send; I32 count = 0; + PERL_UNUSED_CONTEXT; if (!*is_utf8) return (U8 *)start; @@ -849,7 +857,7 @@ Perl_bytes_from_utf8(pTHX_ const U8 *s, STRLEN *len, bool *is_utf8) *is_utf8 = 0; - Newxz(d, (*len) - count + 1, U8); + Newx(d, (*len) - count + 1, U8); s = start; start = d; while (s < send) { U8 c = *s++; @@ -884,8 +892,9 @@ Perl_bytes_to_utf8(pTHX_ const U8 *s, STRLEN *len) const U8 * const send = s + (*len); U8 *d; U8 *dst; + PERL_UNUSED_CONTEXT; - Newxz(d, (*len) * 2 + 1, U8); + Newx(d, (*len) * 2 + 1, U8); dst = d; while (s < send) { @@ -1240,16 +1249,18 @@ static bool S_is_utf8_common(pTHX_ const U8 *const p, SV **swash, const char *const swashname) { + dVAR; if (!is_utf8_char(p)) return FALSE; if (!*swash) - *swash = swash_init("utf8", swashname, &PL_sv_undef, 0, 0); + *swash = swash_init("utf8", swashname, &PL_sv_undef, 1, 0); return swash_fetch(*swash, p, TRUE) != 0; } bool Perl_is_utf8_alnum(pTHX_ const U8 *p) { + dVAR; /* NOTE: "IsWord", not "IsAlnum", since Alnum is a true * descendant of isalnum(3), in other words, it doesn't * contain the '_'. --jhi */ @@ -1259,12 +1270,14 @@ Perl_is_utf8_alnum(pTHX_ const U8 *p) bool Perl_is_utf8_alnumc(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_alnumc, "IsAlnumC"); } bool Perl_is_utf8_idfirst(pTHX_ const U8 *p) /* The naming is historical. */ { + dVAR; if (*p == '_') return TRUE; /* is_utf8_idstart would be more logical. */ @@ -1274,6 +1287,7 @@ Perl_is_utf8_idfirst(pTHX_ const U8 *p) /* The naming is historical. */ bool Perl_is_utf8_idcont(pTHX_ const U8 *p) { + dVAR; if (*p == '_') return TRUE; return S_is_utf8_common(aTHX_ p, &PL_utf8_idcont, "IdContinue"); @@ -1282,72 +1296,84 @@ Perl_is_utf8_idcont(pTHX_ const U8 *p) bool Perl_is_utf8_alpha(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_alpha, "IsAlpha"); } bool Perl_is_utf8_ascii(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_ascii, "IsAscii"); } bool Perl_is_utf8_space(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_space, "IsSpacePerl"); } bool Perl_is_utf8_digit(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_digit, "IsDigit"); } bool Perl_is_utf8_upper(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_upper, "IsUppercase"); } bool Perl_is_utf8_lower(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_lower, "IsLowercase"); } bool Perl_is_utf8_cntrl(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_cntrl, "IsCntrl"); } bool Perl_is_utf8_graph(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_graph, "IsGraph"); } bool Perl_is_utf8_print(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_print, "IsPrint"); } bool Perl_is_utf8_punct(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_punct, "IsPunct"); } bool Perl_is_utf8_xdigit(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_xdigit, "Isxdigit"); } bool Perl_is_utf8_mark(pTHX_ const U8 *p) { + dVAR; return S_is_utf8_common(aTHX_ p, &PL_utf8_mark, "IsM"); } @@ -1380,6 +1406,7 @@ UV Perl_to_utf8_case(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, const char *normal, const char *special) { + dVAR; U8 tmpbuf[UTF8_MAXBYTES_CASE+1]; STRLEN len = 0; @@ -1480,6 +1507,7 @@ The first character of the uppercased version is returned UV Perl_to_utf8_upper(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp) { + dVAR; return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, &PL_utf8_toupper, "ToUpper", "utf8::ToSpecUpper"); } @@ -1500,6 +1528,7 @@ The first character of the titlecased version is returned UV Perl_to_utf8_title(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp) { + dVAR; return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, &PL_utf8_totitle, "ToTitle", "utf8::ToSpecTitle"); } @@ -1520,6 +1549,7 @@ The first character of the lowercased version is returned UV Perl_to_utf8_lower(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp) { + dVAR; return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, &PL_utf8_tolower, "ToLower", "utf8::ToSpecLower"); } @@ -1541,12 +1571,17 @@ The first character of the foldcased version is returned UV Perl_to_utf8_fold(pTHX_ const U8 *p, U8* ustrp, STRLEN *lenp) { + dVAR; return Perl_to_utf8_case(aTHX_ p, ustrp, lenp, &PL_utf8_tofold, "ToFold", "utf8::ToSpecFold"); } -/* a "swash" is a swatch hash */ - +/* Note: + * A "swash" is a swatch hash. + * A "swatch" is a bit vector generated by utf8.c:S_swash_get(). + * C is a pointer to a package name for SWASHNEW, should be "utf8". + * For other parameters, see utf8::SWASHNEW in lib/utf8_heavy.pl. + */ SV* Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits, I32 none) { @@ -1567,8 +1602,15 @@ Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits if (!gv_fetchmeth(stash, "SWASHNEW", 8, -1)) { /* demand load utf8 */ ENTER; errsv_save = newSVsv(ERRSV); + /* It is assumed that callers of this routine are not passing in any + user derived data. */ + /* Need to do this after save_re_context() as it will set PL_tainted to + 1 while saving $1 etc (see the code after getrx: in Perl_magic_get). + Even line to create errsv_save can turn on PL_tainted. */ + SAVEBOOL(PL_tainted); + PL_tainted = 0; Perl_load_module(aTHX_ PERL_LOADMOD_NOIMPORT, newSVpvn(pkg,pkg_len), - Nullsv); + NULL); if (!SvTRUE(ERRSV)) sv_setsv(ERRSV, errsv_save); SvREFCNT_dec(errsv_save); @@ -1622,11 +1664,18 @@ Perl_swash_init(pTHX_ const char* pkg, const char* name, SV *listsv, I32 minbits * the lower-level routine, and it is similarly broken for returning * multiple values. --jhi */ /* Now SWASHGET is recasted into S_swash_get in this file. */ + +/* Note: + * Returns the value of property/mapping C for the first character + * of the string C. If C is true, the string C is + * assumed to be in utf8. If C is false, the string C is + * assumed to be in native 8-bit encoding. Caches the swatch in C. + */ UV -Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8) +Perl_swash_fetch(pTHX_ SV *swash, const U8 *ptr, bool do_utf8) { dVAR; - HV* const hv = (HV*)SvRV(sv); + HV* const hv = (HV*)SvRV(swash); U32 klen; U32 off; STRLEN slen; @@ -1692,7 +1741,7 @@ Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8) const UV code_point = utf8n_to_uvuni(ptr, UTF8_MAXBYTES, 0, ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY); - swatch = swash_get(sv, + swatch = swash_get(swash, /* On EBCDIC & ~(0xA0-1) isn't a useful thing to do */ (klen) ? (code_point & ~(needents - 1)) : 0, needents); @@ -1704,7 +1753,7 @@ Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8) if (!svp || !(tmps = (U8*)SvPV(*svp, slen)) || (slen << 3) < needents) - Perl_croak(aTHX_ "The swatch does not have proper length"); + Perl_croak(aTHX_ "panic: swash_fetch got improper swatch"); } PL_last_swash_hv = hv; @@ -1730,7 +1779,7 @@ Perl_swash_fetch(pTHX_ SV *sv, const U8 *ptr, bool do_utf8) off <<= 2; return (tmps[off] << 24) + (tmps[off+1] << 16) + (tmps[off+2] << 8) + tmps[off + 3] ; } - Perl_croak(aTHX_ "panic: swash_fetch"); + Perl_croak(aTHX_ "panic: swash_fetch got swatch of unexpected bit width"); return 0; } @@ -1744,33 +1793,34 @@ STATIC SV* S_swash_get(pTHX_ SV* swash, UV start, UV span) { SV *swatch; - U8 *l, *lend, *x, *xend, *s, *nl; + U8 *l, *lend, *x, *xend, *s; STRLEN lcur, xcur, scur; HV* const hv = (HV*)SvRV(swash); - SV** listsvp = hv_fetch(hv, "LIST", 4, FALSE); - SV** typesvp = hv_fetch(hv, "TYPE", 4, FALSE); - SV** bitssvp = hv_fetch(hv, "BITS", 4, FALSE); - SV** nonesvp = hv_fetch(hv, "NONE", 4, FALSE); - SV** extssvp = hv_fetch(hv, "EXTRAS", 6, FALSE); - U8* typestr = (U8*)SvPV_nolen(*typesvp); - int typeto = typestr[0] == 'T' && typestr[1] == 'o'; - STRLEN bits = SvUV(*bitssvp); - STRLEN octets = bits >> 3; /* if bits == 1, then octets == 0 */ - UV none = SvUV(*nonesvp); - UV end = start + span; + SV** const listsvp = hv_fetchs(hv, "LIST", FALSE); + SV** const typesvp = hv_fetchs(hv, "TYPE", FALSE); + SV** const bitssvp = hv_fetchs(hv, "BITS", FALSE); + SV** const nonesvp = hv_fetchs(hv, "NONE", FALSE); + SV** const extssvp = hv_fetchs(hv, "EXTRAS", FALSE); + const U8* const typestr = (U8*)SvPV_nolen(*typesvp); + const int typeto = typestr[0] == 'T' && typestr[1] == 'o'; + const STRLEN bits = SvUV(*bitssvp); + const STRLEN octets = bits >> 3; /* if bits == 1, then octets == 0 */ + const UV none = SvUV(*nonesvp); + const UV end = start + span; if (bits != 1 && bits != 8 && bits != 16 && bits != 32) { - Perl_croak(aTHX_ "swash_get: unknown bits %"UVuf, (UV) bits); + Perl_croak(aTHX_ "panic: swash_get doesn't expect bits %"UVuf, + (UV)bits); } /* create and initialize $swatch */ - swatch = newSVpvn("",0); + swatch = newSVpvs(""); scur = octets ? (span * octets) : (span + 7) / 8; SvGROW(swatch, scur + 1); s = (U8*)SvPVX(swatch); if (octets && none) { - const U8* e = s + scur; + const U8* const e = s + scur; while (s < e) { if (bits == 8) *s++ = (U8)(none & 0xff); @@ -1801,7 +1851,7 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) STRLEN numlen; I32 flags = PERL_SCAN_SILENT_ILLDIGIT | PERL_SCAN_DISALLOW_PREFIX; - nl = (U8*)memchr(l, '\n', lend - l); + U8* const nl = (U8*)memchr(l, '\n', lend - l); numlen = lend - l; min = grok_hex((char *)l, &numlen, &flags, NULL); @@ -1846,6 +1896,8 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) } } } + else + val = 0; /* bits == 1, then val should be ignored */ } else { max = min; @@ -1855,6 +1907,8 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) Perl_croak(aTHX_ "%s: illegal mapping '%s'", typestr, l); } } + else + val = 0; /* bits == 1, then val should be ignored */ } if (nl) @@ -1895,11 +1949,11 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) ++val; } } - else { + else { /* bits == 1, then val should be ignored */ if (min < start) min = start; for (key = min; key <= max; key++) { - STRLEN offset = (STRLEN)(key - start); + const STRLEN offset = (STRLEN)(key - start); if (key >= end) goto go_out_list; s[offset >> 3] |= 1 << (offset & 7); @@ -1918,7 +1972,7 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) HV* otherhv; STRLEN otherbits; SV **otherbitssvp, *other; - U8 *s, *o; + U8 *s, *o, *nl; STRLEN slen, olen; U8 opc = *x++; @@ -1949,28 +2003,23 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) } othersvp = hv_fetch(hv, (char *)namestr, namelen, FALSE); - if (*othersvp && SvROK(*othersvp) && - SvTYPE(SvRV(*othersvp))==SVt_PVHV) - otherhv = (HV*)SvRV(*othersvp); - else - Perl_croak(aTHX_ "otherhv is not a hash reference"); - - otherbitssvp = hv_fetch(otherhv, "BITS", 4, FALSE); + otherhv = (HV*)SvRV(*othersvp); + otherbitssvp = hv_fetchs(otherhv, "BITS", FALSE); otherbits = (STRLEN)SvUV(*otherbitssvp); if (bits < otherbits) - Perl_croak(aTHX_ "swash_get: swatch size mismatch"); + Perl_croak(aTHX_ "panic: swash_get found swatch size mismatch"); /* The "other" swatch must be destroyed after. */ other = swash_get(*othersvp, start, span); o = (U8*)SvPV(other, olen); if (!olen) - Perl_croak(aTHX_ "swash_get didn't return valid swatch for other"); + Perl_croak(aTHX_ "panic: swash_get got improper swatch"); s = (U8*)SvPV(swatch, slen); if (bits == 1 && otherbits == 1) { if (slen != olen) - Perl_croak(aTHX_ "swash_get: swatch length mismatch"); + Perl_croak(aTHX_ "panic: swash_get found swatch length mismatch"); switch (opc) { case '+': @@ -1993,8 +2042,7 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) break; } } - else { /* bits >= 8 */ - /* XXX: but weirdly otherval is treated as boolean */ + else { STRLEN otheroctets = otherbits >> 3; STRLEN offset = 0; U8* send = s + slen; @@ -2015,8 +2063,8 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) } } - if (opc == '+' && otherval) - otherval = 1; + if (opc == '+' && otherval) + /*EMPTY*/; /* replace with otherval */ else if (opc == '!' && !otherval) otherval = 1; else if (opc == '-' && otherval) @@ -2024,7 +2072,7 @@ S_swash_get(pTHX_ SV* swash, UV start, UV span) else if (opc == '&' && !otherval) otherval = 0; else { - s += octets; /* not modify orig swatch */ + s += octets; /* no replacement */ continue; } @@ -2141,7 +2189,7 @@ Perl_pv_uni_display(pTHX_ SV *dsv, const U8 *spv, STRLEN len, STRLEN pvlim, UV f u = utf8_to_uvchr((U8*)s, 0); if (u < 256) { const unsigned char c = (unsigned char)u & 0xFF; - if (!ok && (flags & UNI_DISPLAY_BACKSLASH)) { + if (flags & UNI_DISPLAY_BACKSLASH) { switch (c) { case '\n': ok = 'n'; break; @@ -2171,7 +2219,7 @@ Perl_pv_uni_display(pTHX_ SV *dsv, const U8 *spv, STRLEN len, STRLEN pvlim, UV f Perl_sv_catpvf(aTHX_ dsv, "\\x{%"UVxf"}", u); } if (truncated) - sv_catpvn(dsv, "...", 3); + sv_catpvs(dsv, "..."); return SvPVX(dsv); } @@ -2223,11 +2271,15 @@ http://www.unicode.org/unicode/reports/tr21/ (Case Mappings). 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) { + dVAR; register const U8 *p1 = (const U8*)s1; register const U8 *p2 = (const U8*)s2; - register const U8 *f1 = 0, *f2 = 0; - register U8 *e1 = 0, *q1 = 0; - register U8 *e2 = 0, *q2 = 0; + register const U8 *f1 = NULL; + register const U8 *f2 = NULL; + register U8 *e1 = NULL; + register U8 *q1 = NULL; + register U8 *e2 = NULL; + register U8 *q2 = NULL; STRLEN n1 = 0, n2 = 0; U8 foldbuf1[UTF8_MAXBYTES_CASE+1]; U8 foldbuf2[UTF8_MAXBYTES_CASE+1];