X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=sv.c;h=65a3279d6b25e813d1de9f1e1b41f8391ea2ccc5;hb=66670188cea1de9e5a28e46eaf556eccea649d2c;hp=b96cc454580d74b6a40a914f01edeb93bf7eec7f;hpb=5e1c5d9c059365c8c0b3de19e24481eaae21f7e5;p=p5sagit%2Fp5-mst-13.2.git diff --git a/sv.c b/sv.c index b96cc45..65a3279 100644 --- a/sv.c +++ b/sv.c @@ -137,6 +137,7 @@ S_more_sv(pTHX) if (PL_nice_chunk) { sv_add_arena(PL_nice_chunk, PL_nice_chunk_size, 0); PL_nice_chunk = Nullch; + PL_nice_chunk_size = 0; } else { char *chunk; /* must use New here to match call to */ @@ -147,20 +148,24 @@ S_more_sv(pTHX) return sv; } -STATIC void +STATIC I32 S_visit(pTHX_ SVFUNC_t f) { SV* sva; SV* sv; register SV* svend; + I32 visited = 0; for (sva = PL_sv_arenaroot; sva; sva = (SV*)SvANY(sva)) { svend = &sva[SvREFCNT(sva)]; for (sv = sva + 1; sv < svend; ++sv) { - if (SvTYPE(sv) != SVTYPEMASK) + if (SvTYPE(sv) != SVTYPEMASK && SvREFCNT(sv)) { (FCALL)(aTHXo_ sv); + ++visited; + } } } + return visited; } void @@ -181,12 +186,14 @@ Perl_sv_clean_objs(pTHX) PL_in_clean_objs = FALSE; } -void +I32 Perl_sv_clean_all(pTHX) { + I32 cleaned; PL_in_clean_all = TRUE; - visit(do_clean_all); + cleaned = visit(do_clean_all); PL_in_clean_all = FALSE; + return cleaned; } void @@ -1424,12 +1431,12 @@ S_not_a_number(pTHX_ SV *sv) { char tmpbuf[64]; char *d = tmpbuf; - char *s; char *limit = tmpbuf + sizeof(tmpbuf) - 8; /* each *s can expand to 4 chars + "...\0", i.e. need room for 8 chars */ - for (s = SvPVX(sv); *s && d < limit; s++) { + char *s, *end; + for (s = SvPVX(sv), end = s + SvCUR(sv); s < end && d < limit; s++) { int ch = *s & 0xFF; if (ch & 128 && !isPRINT_LC(ch)) { *d++ = 'M'; @@ -1452,6 +1459,10 @@ S_not_a_number(pTHX_ SV *sv) *d++ = '\\'; *d++ = '\\'; } + else if (ch == '\0') { + *d++ = '\\'; + *d++ = '0'; + } else if (isPRINT_LC(ch)) *d++ = ch; else { @@ -1459,7 +1470,7 @@ S_not_a_number(pTHX_ SV *sv) *d++ = toCTRL(ch); } } - if (*s) { + if (s < end) { *d++ = '.'; *d++ = '.'; *d++ = '.'; @@ -1720,7 +1731,7 @@ Perl_sv_2iv(pTHX_ register SV *sv) if (SvROK(sv)) { SV* tmpstr; if (SvAMAGIC(sv) && (tmpstr=AMG_CALLun(sv,numer)) && - (SvRV(tmpstr) != SvRV(sv))) + (SvTYPE(tmpstr) != SVt_RV || (SvRV(tmpstr) != SvRV(sv)))) return SvIV(tmpstr); return PTR2IV(SvRV(sv)); } @@ -1974,7 +1985,7 @@ Perl_sv_2uv(pTHX_ register SV *sv) if (SvROK(sv)) { SV* tmpstr; if (SvAMAGIC(sv) && (tmpstr=AMG_CALLun(sv,numer)) && - (SvRV(tmpstr) != SvRV(sv))) + (SvTYPE(tmpstr) != SVt_RV || (SvRV(tmpstr) != SvRV(sv)))) return SvUV(tmpstr); return PTR2UV(SvRV(sv)); } @@ -2258,7 +2269,7 @@ Perl_sv_2nv(pTHX_ register SV *sv) if (SvROK(sv)) { SV* tmpstr; if (SvAMAGIC(sv) && (tmpstr=AMG_CALLun(sv,numer)) && - (SvRV(tmpstr) != SvRV(sv))) + (SvTYPE(tmpstr) != SVt_RV || (SvRV(tmpstr) != SvRV(sv)))) return SvNV(tmpstr); return PTR2NV(SvRV(sv)); } @@ -2522,7 +2533,7 @@ Perl_looks_like_number(pTHX_ SV *sv) ) { #ifdef USE_LOCALE_NUMERIC if (specialradix) - s += SvCUR(PL_numeric_radix); + s += SvCUR(PL_numeric_radix_sv); else #endif s++; @@ -2538,7 +2549,7 @@ Perl_looks_like_number(pTHX_ SV *sv) ) { #ifdef USE_LOCALE_NUMERIC if (specialradix) - s += SvCUR(PL_numeric_radix); + s += SvCUR(PL_numeric_radix_sv); else #endif s++; @@ -2674,7 +2685,7 @@ Perl_sv_2pv(pTHX_ register SV *sv, STRLEN *lp) if (SvROK(sv)) { SV* tmpstr; if (SvAMAGIC(sv) && (tmpstr=AMG_CALLun(sv,string)) && - (SvRV(tmpstr) != SvRV(sv))) + (SvTYPE(tmpstr) != SVt_RV || (SvRV(tmpstr) != SvRV(sv)))) return SvPV(tmpstr,*lp); sv = (SV*)SvRV(sv); if (!sv) @@ -2914,7 +2925,7 @@ Perl_sv_2bool(pTHX_ register SV *sv) if (SvROK(sv)) { SV* tmpsv; if (SvAMAGIC(sv) && (tmpsv=AMG_CALLun(sv,bool_)) && - (SvRV(tmpsv) != SvRV(sv))) + (SvTYPE(tmpsv) != SVt_RV || (SvRV(tmpsv) != SvRV(sv)))) return SvTRUE(tmpsv); return SvRV(sv) != 0; } @@ -2954,37 +2965,41 @@ if all the bytes have hibit clear. STRLEN Perl_sv_utf8_upgrade(pTHX_ register SV *sv) { - char *s, *t, *e; + U8 *s, *t, *e; int hibit = 0; if (!sv) return 0; - if (!SvPOK(sv)) - (void) SvPV_nolen(sv); + if (!SvPOK(sv)) { + STRLEN len = 0; + (void) sv_2pv(sv,&len); + if (!SvPOK(sv)) + return len; + } if (SvUTF8(sv)) return SvCUR(sv); + if (SvREADONLY(sv) && SvFAKE(sv)) { + sv_force_normal(sv); + } + /* This function could be much more efficient if we had a FLAG in SVs * to signal if there are any hibit chars in the PV. * Given that there isn't make loop fast as possible */ - s = SvPVX(sv); - e = SvEND(sv); + s = (U8 *) SvPVX(sv); + e = (U8 *) SvEND(sv); t = s; while (t < e) { - if ((hibit = UTF8_IS_CONTINUED(*t++))) + U8 ch = *t++; + if ((hibit = !NATIVE_IS_INVARIANT(ch))) break; } - if (hibit) { STRLEN len; - if (SvREADONLY(sv) && SvFAKE(sv)) { - sv_force_normal(sv); - s = SvPVX(sv); - } len = SvCUR(sv) + 1; /* Plus the \0 */ SvPVX(sv) = (char*)bytes_to_utf8((U8*)s, &len); SvCUR(sv) = len - 1; @@ -3092,8 +3107,8 @@ bool Perl_sv_utf8_decode(pTHX_ register SV *sv) { if (SvPOK(sv)) { - char *c; - char *e; + U8 *c; + U8 *e; /* The octets may have got themselves encoded - get them back as bytes */ if (!sv_utf8_downgrade(sv, TRUE)) @@ -3102,12 +3117,13 @@ Perl_sv_utf8_decode(pTHX_ register SV *sv) /* it is actually just a matter of turning the utf8 flag on, but * we want to make sure everything inside is valid utf8 first. */ - c = SvPVX(sv); - if (!is_utf8_string((U8*)c, SvCUR(sv)+1)) + c = (U8 *) SvPVX(sv); + if (!is_utf8_string(c, SvCUR(sv)+1)) return FALSE; - e = SvEND(sv); + e = (U8 *) SvEND(sv); while (c < e) { - if (UTF8_IS_CONTINUED(*c++)) { + U8 ch = *c++; + if (!UTF8_IS_INVARIANT(ch)) { SvUTF8_on(sv); break; } @@ -3810,8 +3826,9 @@ Perl_sv_chop(pTHX_ register SV *sv, register char *ptr) /* like set but assuming =for apidoc sv_catpvn Concatenates the string onto the end of the string which is in the SV. The -C indicates number of bytes to copy. Handles 'get' magic, but not -'set' magic. See C. +C indicates number of bytes to copy. If the SV has the UTF8 +status set, then the bytes appended should be valid UTF8. +Handles 'get' magic, but not 'set' magic. See C. =cut */ @@ -3909,10 +3926,10 @@ Perl_sv_catsv_mg(pTHX_ SV *dsv, register SV *ssv) =for apidoc sv_catpv Concatenates the string onto the end of the string which is in the SV. -Handles 'get' magic, but not 'set' magic. See C. +If the SV has the UTF8 status set, then the bytes appended should be +valid UTF8. Handles 'get' magic, but not 'set' magic. See C. -=cut -*/ +=cut */ void Perl_sv_catpv(pTHX_ register SV *sv, register const char *ptr) @@ -4726,8 +4743,9 @@ Perl_sv_pos_b2u(pTHX_ register SV *sv, I32* offsetp) len = 0; while (s < send) { STRLEN n; - /* We can use low level directly here as we are not looking at the values */ - if (utf8n_to_uvuni(s, UTF8SKIP(s), &n, 0)) { + /* Call utf8n_to_uvchr() to validate the sequence */ + utf8n_to_uvchr(s, UTF8SKIP(s), &n, 0); + if (n > 0) { s += n; len++; } @@ -4755,8 +4773,7 @@ Perl_sv_eq(pTHX_ register SV *sv1, register SV *sv2) char *pv2; STRLEN cur2; I32 eq = 0; - bool pv1tmp = FALSE; - bool pv2tmp = FALSE; + char *tpv = Nullch; if (!sv1) { pv1 = ""; @@ -4775,31 +4792,33 @@ Perl_sv_eq(pTHX_ register SV *sv1, register SV *sv2) /* do not utf8ize the comparands as a side-effect */ if (cur1 && cur2 && SvUTF8(sv1) != SvUTF8(sv2) && !IN_BYTE) { bool is_utf8 = TRUE; - + /* UTF-8ness differs */ if (PL_hints & HINT_UTF8_DISTINCT) return FALSE; if (SvUTF8(sv1)) { + /* sv1 is the UTF-8 one , If is equal it must be downgrade-able */ char *pv = (char*)bytes_from_utf8((U8*)pv1, &cur1, &is_utf8); - - if ((pv1tmp = (pv != pv1))) - pv1 = pv; + if (pv != pv1) + pv1 = tpv = pv; } else { + /* sv2 is the UTF-8 one , If is equal it must be downgrade-able */ char *pv = (char *)bytes_from_utf8((U8*)pv2, &cur2, &is_utf8); - - if ((pv2tmp = (pv != pv2))) - pv2 = pv; + if (pv != pv2) + pv2 = tpv = pv; + } + if (is_utf8) { + /* Downgrade not possible - cannot be eq */ + return FALSE; } } if (cur1 == cur2) eq = memEQ(pv1, pv2, cur1); - if (pv1tmp) - Safefree(pv1); - if (pv2tmp) - Safefree(pv2); + if (tpv != Nullch) + Safefree(tpv); return eq; } @@ -6751,12 +6770,15 @@ Perl_sv_catpvf_mg_nocontext(SV *sv, const char* pat, ...) /* =for apidoc sv_catpvf -Processes its arguments like C and appends the formatted output -to an SV. Handles 'get' magic, but not 'set' magic. C must -typically be called after calling this function to handle 'set' magic. +Processes its arguments like C and appends the formatted +output to an SV. If the appended data contains "wide" characters +(including, but not limited to, SVs with a UTF-8 PV formatted with %s, +and characters >255 formatted with %c), the original SV might get +upgraded to UTF-8. Handles 'get' magic, but not 'set' magic. +C must typically be called after calling this function +to handle 'set' magic. -=cut -*/ +=cut */ void Perl_sv_catpvf(pTHX_ SV *sv, const char* pat, ...) @@ -7121,7 +7143,9 @@ Perl_sv_vcatpvfn(pTHX_ SV *sv, const char *pat, STRLEN patlen, va_list *args, SV case 'c': uv = args ? va_arg(*args, int) : SvIVx(argsv); - if ((uv > 255 || (uv > 127 && SvUTF8(sv))) && !IN_BYTE) { + if ((uv > 255 || + (!UNI_IS_INVARIANT(uv) && SvUTF8(sv))) + && !IN_BYTE) { eptr = (char*)utf8buf; elen = uvchr_to_utf8((U8*)eptr, uv) - utf8buf; is_utf = TRUE; @@ -7661,8 +7685,8 @@ Perl_gp_dup(pTHX_ GP *gp) MAGIC * Perl_mg_dup(pTHX_ MAGIC *mg) { - MAGIC *mgret = (MAGIC*)NULL; - MAGIC *mgprev; + MAGIC *mgprev = (MAGIC*)NULL; + MAGIC *mgret; if (!mg) return (MAGIC*)NULL; /* look for it in the table first */ @@ -7673,10 +7697,10 @@ Perl_mg_dup(pTHX_ MAGIC *mg) for (; mg; mg = mg->mg_moremagic) { MAGIC *nmg; Newz(0, nmg, 1, MAGIC); - if (!mgret) - mgret = nmg; - else + if (mgprev) mgprev->mg_moremagic = nmg; + else + mgret = nmg; nmg->mg_virtual = mg->mg_virtual; /* XXX copy dynamic vtable? */ nmg->mg_private = mg->mg_private; nmg->mg_type = mg->mg_type; @@ -8190,7 +8214,7 @@ dup_pvcv: CvROOT(dstr) = OpREFCNT_inc(CvROOT(sstr)); CvXSUB(dstr) = CvXSUB(sstr); CvXSUBANY(dstr) = CvXSUBANY(sstr); - CvGV(dstr) = gv_dup_inc(CvGV(sstr)); + CvGV(dstr) = gv_dup(CvGV(sstr)); CvDEPTH(dstr) = CvDEPTH(sstr); if (CvPADLIST(sstr) && !AvREAL(CvPADLIST(sstr))) { /* XXX padlists are real, but pretend to be not */ @@ -8258,7 +8282,7 @@ Perl_cx_dup(pTHX_ PERL_CONTEXT *cxs, I32 ix, I32 max) ncx->blk_sub.argarray = (cx->blk_sub.hasargs ? av_dup_inc(cx->blk_sub.argarray) : Nullav); - ncx->blk_sub.savearray = av_dup(cx->blk_sub.savearray); + ncx->blk_sub.savearray = av_dup_inc(cx->blk_sub.savearray); ncx->blk_sub.olddepth = cx->blk_sub.olddepth; ncx->blk_sub.hasargs = cx->blk_sub.hasargs; ncx->blk_sub.lval = cx->blk_sub.lval; @@ -8512,6 +8536,7 @@ Perl_ss_dup(pTHX_ PerlInterpreter *proto_perl) TOPIV(nss,ix) = iv; break; case SAVEt_FREESV: + case SAVEt_MORTALIZESV: sv = (SV*)POPPTR(ss,ix); TOPPTR(nss,ix) = sv_dup_inc(sv); break; @@ -8845,7 +8870,7 @@ perl_clone_using(PerlInterpreter *proto_perl, UV flags, PL_defgv = gv_dup(proto_perl->Idefgv); PL_argvgv = gv_dup(proto_perl->Iargvgv); PL_argvoutgv = gv_dup(proto_perl->Iargvoutgv); - PL_argvout_stack = av_dup(proto_perl->Iargvout_stack); + PL_argvout_stack = av_dup_inc(proto_perl->Iargvout_stack); /* shortcuts to regexp stuff */ PL_replgv = gv_dup(proto_perl->Ireplgv); @@ -9056,7 +9081,7 @@ perl_clone_using(PerlInterpreter *proto_perl, UV flags, PL_numeric_name = SAVEPV(proto_perl->Inumeric_name); PL_numeric_standard = proto_perl->Inumeric_standard; PL_numeric_local = proto_perl->Inumeric_local; - PL_numeric_radix = sv_dup_inc(proto_perl->Inumeric_radix); + PL_numeric_radix_sv = sv_dup_inc(proto_perl->Inumeric_radix_sv); #endif /* !USE_LOCALE_NUMERIC */ /* utf8 character classes */ @@ -9270,7 +9295,6 @@ perl_clone_using(PerlInterpreter *proto_perl, UV flags, PL_regendp = (I32*)NULL; PL_reglastparen = (U32*)NULL; PL_regtill = Nullch; - PL_regprev = '\n'; PL_reg_start_tmp = (char**)NULL; PL_reg_start_tmpl = 0; PL_regdata = (struct reg_data*)NULL;