X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=regexec.c;h=d19b82ea5a613ea6f0086e8485392b7278aa12ef;hb=b42969c054c571b3b721d68818184e59ca761058;hp=fae084e061317aa8e758ef2c4113aceb4c91d1c8;hpb=3f7c398ef4badd9c6ec5b40ea29141484c160f63;p=p5sagit%2Fp5-mst-13.2.git diff --git a/regexec.c b/regexec.c index fae084e..d19b82e 100644 --- a/regexec.c +++ b/regexec.c @@ -55,7 +55,6 @@ # define PERL_NO_GET_CONTEXT #endif -/*SUPPRESS 112*/ /* * pregcomp and pregexec -- regsub and regerror are not used in perl * @@ -140,7 +139,12 @@ #define HOP3c(pos,off,lim) ((char*)HOP3(pos,off,lim)) #define HOPMAYBE3c(pos,off,lim) ((char*)HOPMAYBE3(pos,off,lim)) -#define LOAD_UTF8_CHARCLASS(a,b) STMT_START { if (!CAT2(PL_utf8_,a)) { ENTER; save_re_context(); (void)CAT2(is_utf8_, a)((const U8*)b); LEAVE; } } STMT_END +#define LOAD_UTF8_CHARCLASS(class,str) STMT_START { \ + if (!CAT2(PL_utf8_,class)) { bool ok; ENTER; save_re_context(); ok=CAT2(is_utf8_,class)((const U8*)str); assert(ok); LEAVE; } } STMT_END +#define LOAD_UTF8_CHARCLASS_ALNUM() LOAD_UTF8_CHARCLASS(alnum,"a") +#define LOAD_UTF8_CHARCLASS_DIGIT() LOAD_UTF8_CHARCLASS(digit,"0") +#define LOAD_UTF8_CHARCLASS_SPACE() LOAD_UTF8_CHARCLASS(space," ") +#define LOAD_UTF8_CHARCLASS_MARK() LOAD_UTF8_CHARCLASS(mark, "\xcd\x86") /* for use after a quantifier and before an EXACT-like node -- japhy */ #define JUMPABLE(rn) ( \ @@ -412,8 +416,8 @@ Perl_re_intuit_start(pTHX_ regexp *prog, SV *sv, char *strpos, char *check_at = Nullch; /* check substr found at this pos */ const I32 multiline = prog->reganch & PMf_MULTILINE; #ifdef DEBUGGING - char *i_strpos = strpos; - SV *dsv = PERL_DEBUG_PAD_ZERO(0); + const char * const i_strpos = strpos; + SV * const dsv = PERL_DEBUG_PAD_ZERO(0); #endif GET_RE_DEBUG_FLAGS_DECL; @@ -611,7 +615,8 @@ Perl_re_intuit_start(pTHX_ regexp *prog, SV *sv, char *strpos, if (check == (do_utf8 ? prog->float_utf8 : prog->float_substr)) { do_other_anchored: { - char *last = HOP3c(s, -start_shift, strbeg), *last1, *last2; + char * const last = HOP3c(s, -start_shift, strbeg); + char *last1, *last2; char *s1 = s; SV* must; @@ -857,7 +862,7 @@ Perl_re_intuit_start(pTHX_ regexp *prog, SV *sv, char *strpos, regstclass does not come from lookahead... */ /* If regstclass takes bytelength more than 1: If charlength==1, OK. This leaves EXACTF only, which is dealt with in find_byclass(). */ - const U8* str = (U8*)STRING(prog->regstclass); + const U8* const str = (U8*)STRING(prog->regstclass); const int cl_l = (PL_regkind[(U8)OP(prog->regstclass)] == EXACT ? CHR_DIST(str+STR_LEN(prog->regstclass), str) : 1); @@ -1034,14 +1039,15 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 U8 *sm = (U8 *) m; U8 tmpbuf1[UTF8_MAXBYTES_CASE+1]; U8 tmpbuf2[UTF8_MAXBYTES_CASE+1]; + const U32 uniflags = ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY; to_utf8_lower((U8*)m, tmpbuf1, &ulen1); to_utf8_upper((U8*)m, tmpbuf2, &ulen2); c1 = utf8n_to_uvchr(tmpbuf1, UTF8_MAXBYTES_CASE, - 0, ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY); + 0, uniflags); c2 = utf8n_to_uvchr(tmpbuf2, UTF8_MAXBYTES_CASE, - 0, ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY); + 0, uniflags); lnc = 0; while (sm < ((U8 *) m + ln)) { lnc++; @@ -1080,14 +1086,13 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 UV c, f; U8 tmpbuf [UTF8_MAXBYTES+1]; STRLEN len, foldlen; - + const U32 uniflags = ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY; if (c1 == c2) { /* Upper and lower of 1st char are equal - * probably not a "letter". */ while (s <= e) { c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); if ( c == c1 && (ln == len || ibcmp_utf8(s, (char **)0, 0, do_utf8, @@ -1114,8 +1119,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 else { while (s <= e) { c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); /* Handle some of the three Greek sigmas cases. * Note that not all the possible combinations @@ -1189,7 +1193,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 } tmp = ((OP(c) == BOUND ? isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0); - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (tmp == !(OP(c) == BOUND ? swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) : @@ -1232,7 +1236,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 } tmp = ((OP(c) == NBOUND ? isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0); - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (tmp == !(OP(c) == NBOUND ? swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) : @@ -1261,7 +1265,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 break; case ALNUM: if (do_utf8) { - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8)) { if (tmp && (norun || regtry(prog, s))) @@ -1319,7 +1323,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 break; case NALNUM: if (do_utf8) { - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (!swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8)) { if (tmp && (norun || regtry(prog, s))) @@ -1377,7 +1381,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 break; case SPACE: if (do_utf8) { - LOAD_UTF8_CHARCLASS(space," "); + LOAD_UTF8_CHARCLASS_SPACE(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8)) { if (tmp && (norun || regtry(prog, s))) @@ -1435,7 +1439,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 break; case NSPACE: if (do_utf8) { - LOAD_UTF8_CHARCLASS(space," "); + LOAD_UTF8_CHARCLASS_SPACE(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (!(*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8))) { if (tmp && (norun || regtry(prog, s))) @@ -1493,7 +1497,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 break; case DIGIT: if (do_utf8) { - LOAD_UTF8_CHARCLASS(digit,"0"); + LOAD_UTF8_CHARCLASS_DIGIT(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (swash_fetch(PL_utf8_digit,(U8*)s, do_utf8)) { if (tmp && (norun || regtry(prog, s))) @@ -1551,7 +1555,7 @@ S_find_byclass(pTHX_ regexp * prog, regnode *c, char *s, const char *strend, I32 break; case NDIGIT: if (do_utf8) { - LOAD_UTF8_CHARCLASS(digit,"0"); + LOAD_UTF8_CHARCLASS_DIGIT(); while (s + (uskip = UTF8SKIP(s)) <= strend) { if (!swash_fetch(PL_utf8_digit,(U8*)s, do_utf8)) { if (tmp && (norun || regtry(prog, s))) @@ -1646,7 +1650,7 @@ Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char * GET_RE_DEBUG_FLAGS_DECL; - (void)data; /* Currently unused */ + PERL_UNUSED_ARG(data); RX_MATCH_UTF8_set(prog,do_utf8); PL_regcc = 0; @@ -1844,7 +1848,6 @@ Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char * "Did not find anchored character...\n") ); } - /*SUPPRESS 560*/ else if (prog->anchored_substr != Nullsv || prog->anchored_utf8 != Nullsv || ((prog->float_substr != Nullsv || prog->float_utf8 != Nullsv) @@ -1944,8 +1947,8 @@ Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char * } DEBUG_EXECUTE_r({ SV *prop = sv_newmortal(); - char *s0; - char *s1; + const char *s0; + const char *s1; int len0; int len1; @@ -1953,7 +1956,7 @@ Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char * s0 = UTF ? pv_uni_display(dsv0, (U8*)SvPVX_const(prop), SvCUR(prop), 60, UNI_DISPLAY_REGEX) : - SvPVX(prop); + SvPVX_const(prop); len0 = UTF ? SvCUR(dsv0) : SvCUR(prop); s1 = UTF ? sv_uni_display(dsv1, sv, 60, UNI_DISPLAY_REGEX) : s; @@ -1989,7 +1992,7 @@ Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char * } else { STRLEN len; - const char * const little = SvPV(float_real, len); + const char * const little = SvPV_const(float_real, len); if (SvTAIL(float_real)) { if (memEQ(strend - len + 1, little, len - 1)) @@ -2056,7 +2059,7 @@ got_it: RX_MATCH_COPY_FREE(prog); if (flags & REXEC_COPY_STR) { I32 i = PL_regeol - startpos + (stringarg - strbeg); -#ifdef PERL_COPY_ON_WRITE +#ifdef PERL_OLD_COPY_ON_WRITE if ((SvIsCOW(sv) || (SvFLAGS(sv) & CAN_COW_MASK) == CAN_COW_FLAGS)) { if (DEBUG_C_TEST) { @@ -2065,7 +2068,7 @@ got_it: (int) SvTYPE(sv)); } prog->saved_copy = sv_setsv_cow(prog->saved_copy, sv); - prog->subbeg = SvPVX(prog->saved_copy); + prog->subbeg = (char *)SvPVX_const(prog->saved_copy); assert (SvPOKp(prog->saved_copy)); } else #endif @@ -2143,7 +2146,7 @@ S_regtry(pTHX_ regexp *prog, char *startpos) SAVEDESTRUCTOR_X(restore_pos, 0); } if (!PL_reg_curpm) { - Newz(22, PL_reg_curpm, 1, PMOP); + Newxz(PL_reg_curpm, 1, PMOP); #ifdef USE_ITHREADS { SV* repointer = newSViv(0); @@ -2164,7 +2167,7 @@ S_regtry(pTHX_ regexp *prog, char *startpos) $` inside (?{}) could fail... */ PL_reg_oldsaved = prog->subbeg; PL_reg_oldsavedlen = prog->sublen; -#ifdef PERL_COPY_ON_WRITE +#ifdef PERL_OLD_COPY_ON_WRITE PL_nrs = prog->saved_copy; #endif RX_MATCH_COPIED_off(prog); @@ -2189,7 +2192,7 @@ S_regtry(pTHX_ regexp *prog, char *startpos) if(PL_reg_start_tmp) Renew(PL_reg_start_tmp, PL_reg_start_tmpl, char*); else - New(22, PL_reg_start_tmp, PL_reg_start_tmpl, char*); + Newx(PL_reg_start_tmp, PL_reg_start_tmpl, char*); } /* XXXX What this code is doing here?!!! There should be no need @@ -2412,6 +2415,7 @@ S_regmatch(pTHX_ regnode *prog) SV *re_debug_flags = NULL; #endif + U32 uniflags = ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY; GET_RE_DEBUG_FLAGS; @@ -2579,8 +2583,6 @@ S_regmatch(pTHX_ regnode *prog) case TRIEF: case TRIEFL: { - - const U32 uniflags = ckWARN( WARN_UTF8 ) ? 0 : UTF8_ALLOW_ANY; U8 *uc = ( U8* )locinput; U32 state = 1; U16 charid = 0; @@ -2609,7 +2611,7 @@ S_regmatch(pTHX_ regnode *prog) if ( base ) { - if ( do_utf8 || UTF ) { + if ( do_utf8 ) { if ( foldlen>0 ) { uvc = utf8n_to_uvuni( uscan, UTF8_MAXLEN, &len, uniflags ); foldlen -= len; @@ -2648,7 +2650,6 @@ S_regmatch(pTHX_ regnode *prog) from previous if blocks */ case TRIE: { - const U32 uniflags = ckWARN( WARN_UTF8 ) ? 0 : UTF8_ALLOW_ANY; U8 *uc = (U8*)locinput; U32 state = 1; U16 charid = 0; @@ -2675,7 +2676,7 @@ S_regmatch(pTHX_ regnode *prog) if ( base ) { - if ( do_utf8 || UTF ) { + if ( do_utf8 ) { uvc = utf8n_to_uvuni( (U8*)uc, UTF8_MAXLEN, &len, uniflags ); } else { uvc = (U32)*uc; @@ -2722,7 +2723,7 @@ S_regmatch(pTHX_ regnode *prog) "%*s %sonly one match : #%d <%s>%s\n", REPORT_CODE_OFF+PL_regindent*2, "", PL_colors[4], accept_buff[ 0 ].wordnum, - tmp ? SvPV_nolen( *tmp ) : "not compiled under -Dr", + tmp ? SvPV_nolen_const( *tmp ) : "not compiled under -Dr", PL_colors[5] ); }); PL_reginput = (char *)accept_buff[ 0 ].endpos; @@ -2757,7 +2758,7 @@ S_regmatch(pTHX_ regnode *prog) PerlIO_printf( Perl_debug_log, "%*s %strying alternation #%d <%s> at 0x%p%s\n", REPORT_CODE_OFF+PL_regindent*2, "", PL_colors[4], accept_buff[best].wordnum, - tmp ? SvPV_nolen( *tmp ) : "not compiled under -Dr",scan, + tmp ? SvPV_nolen_const( *tmp ) : "not compiled under -Dr",scan, PL_colors[5] ); }); if ( best= PL_regeol) sayNO; if (do_utf8) { - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); if (OP(scan) == NALNUM ? swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8) : isALNUM_LC_utf8((U8*)locinput)) @@ -2985,7 +2984,7 @@ S_regmatch(pTHX_ regnode *prog) } if (OP(scan) == BOUND || OP(scan) == NBOUND) { ln = isALNUM_uni(ln); - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); n = swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8); } else { @@ -3017,7 +3016,7 @@ S_regmatch(pTHX_ regnode *prog) sayNO; if (do_utf8) { if (UTF8_IS_CONTINUED(nextchr)) { - LOAD_UTF8_CHARCLASS(space," "); + LOAD_UTF8_CHARCLASS_SPACE(); if (!(OP(scan) == SPACE ? swash_fetch(PL_utf8_space, (U8*)locinput, do_utf8) : isSPACE_LC_utf8((U8*)locinput))) @@ -3047,7 +3046,7 @@ S_regmatch(pTHX_ regnode *prog) if (!nextchr && locinput >= PL_regeol) sayNO; if (do_utf8) { - LOAD_UTF8_CHARCLASS(space," "); + LOAD_UTF8_CHARCLASS_SPACE(); if (OP(scan) == NSPACE ? swash_fetch(PL_utf8_space, (U8*)locinput, do_utf8) : isSPACE_LC_utf8((U8*)locinput)) @@ -3070,7 +3069,7 @@ S_regmatch(pTHX_ regnode *prog) if (!nextchr) sayNO; if (do_utf8) { - LOAD_UTF8_CHARCLASS(digit,"0"); + LOAD_UTF8_CHARCLASS_DIGIT(); if (!(OP(scan) == DIGIT ? swash_fetch(PL_utf8_digit, (U8*)locinput, do_utf8) : isDIGIT_LC_utf8((U8*)locinput))) @@ -3093,7 +3092,7 @@ S_regmatch(pTHX_ regnode *prog) if (!nextchr && locinput >= PL_regeol) sayNO; if (do_utf8) { - LOAD_UTF8_CHARCLASS(digit,"0"); + LOAD_UTF8_CHARCLASS_DIGIT(); if (OP(scan) == NDIGIT ? swash_fetch(PL_utf8_digit, (U8*)locinput, do_utf8) : isDIGIT_LC_utf8((U8*)locinput)) @@ -3113,7 +3112,7 @@ S_regmatch(pTHX_ regnode *prog) if (locinput >= PL_regeol) sayNO; if (do_utf8) { - LOAD_UTF8_CHARCLASS(mark,"~"); + LOAD_UTF8_CHARCLASS_MARK(); if (swash_fetch(PL_utf8_mark,(U8*)locinput, do_utf8)) sayNO; locinput += PL_utf8skip[nextchr]; @@ -3248,7 +3247,7 @@ S_regmatch(pTHX_ regnode *prog) } else { STRLEN len; - char *t = SvPV(ret, len); + const char *t = SvPV_const(ret, len); PMOP pm; char * const oprecomp = PL_regprecomp; const I32 osize = PL_regsize; @@ -3256,7 +3255,7 @@ S_regmatch(pTHX_ regnode *prog) Zero(&pm, 1, PMOP); if (DO_UTF8(ret)) pm.op_pmdynflags |= PMdf_DYN_UTF8; - re = CALLREGCOMP(aTHX_ t, t + len, &pm); + re = CALLREGCOMP(aTHX_ (char*)t, (char*)t + len, &pm); if (!(SvFLAGS(ret) & (SVs_TEMP | SVs_PADTMP | SVf_READONLY | SVs_GMG))) @@ -3551,7 +3550,7 @@ S_regmatch(pTHX_ regnode *prog) } else { PL_reg_poscache_size = size; - Newz(29, PL_reg_poscache, size, char); + Newxz(PL_reg_poscache, size, char); } DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log, @@ -3937,16 +3936,13 @@ S_regmatch(pTHX_ regnode *prog) to_utf8_upper((U8*)s, tmpbuf2, &ulen2); c1 = utf8n_to_uvuni(tmpbuf1, UTF8_MAXBYTES, 0, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); c2 = utf8n_to_uvuni(tmpbuf2, UTF8_MAXBYTES, 0, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); } else { c2 = c1 = utf8n_to_uvchr(s, UTF8_MAXBYTES, 0, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); } } } @@ -4007,8 +4003,7 @@ S_regmatch(pTHX_ regnode *prog) while (locinput <= e && utf8n_to_uvchr((U8*)locinput, UTF8_MAXBYTES, &len, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY) != (UV)c1) { + uniflags) != (UV)c1) { locinput += len; count++; } @@ -4019,8 +4014,7 @@ S_regmatch(pTHX_ regnode *prog) while (locinput <= e) { UV c = utf8n_to_uvchr((U8*)locinput, UTF8_MAXBYTES, &len, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); if (c == (UV)c1 || c == (UV)c2) break; locinput += len; @@ -4056,8 +4050,7 @@ S_regmatch(pTHX_ regnode *prog) if (do_utf8) c = utf8n_to_uvchr((U8*)PL_reginput, UTF8_MAXBYTES, 0, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); else c = UCHARAT(PL_reginput); /* If it could work, try it. */ @@ -4106,8 +4099,7 @@ S_regmatch(pTHX_ regnode *prog) if (do_utf8) c = utf8n_to_uvchr((U8*)PL_reginput, UTF8_MAXBYTES, 0, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); else c = UCHARAT(PL_reginput); } @@ -4129,8 +4121,7 @@ S_regmatch(pTHX_ regnode *prog) if (do_utf8) c = utf8n_to_uvchr((U8*)PL_reginput, UTF8_MAXBYTES, 0, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY); + uniflags); else c = UCHARAT(PL_reginput); } @@ -4323,7 +4314,6 @@ do_no: goto do_no; } /* Have more choice yet. Reuse the same uwb. */ - /*SUPPRESS 560*/ if ((n = (uwb->type == RE_UNWIND_BRANCH ? NEXT_OFF(next) : ARG(next)))) next += n; @@ -4437,7 +4427,7 @@ S_regrepeat(pTHX_ const regnode *p, I32 max) case ALNUM: if (do_utf8) { loceol = PL_regeol; - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); while (hardcount < max && scan < loceol && swash_fetch(PL_utf8_alnum, (U8*)scan, do_utf8)) { scan += UTF8SKIP(scan); @@ -4465,7 +4455,7 @@ S_regrepeat(pTHX_ const regnode *p, I32 max) case NALNUM: if (do_utf8) { loceol = PL_regeol; - LOAD_UTF8_CHARCLASS(alnum,"a"); + LOAD_UTF8_CHARCLASS_ALNUM(); while (hardcount < max && scan < loceol && !swash_fetch(PL_utf8_alnum, (U8*)scan, do_utf8)) { scan += UTF8SKIP(scan); @@ -4493,7 +4483,7 @@ S_regrepeat(pTHX_ const regnode *p, I32 max) case SPACE: if (do_utf8) { loceol = PL_regeol; - LOAD_UTF8_CHARCLASS(space," "); + LOAD_UTF8_CHARCLASS_SPACE(); while (hardcount < max && scan < loceol && (*scan == ' ' || swash_fetch(PL_utf8_space,(U8*)scan, do_utf8))) { @@ -4522,7 +4512,7 @@ S_regrepeat(pTHX_ const regnode *p, I32 max) case NSPACE: if (do_utf8) { loceol = PL_regeol; - LOAD_UTF8_CHARCLASS(space," "); + LOAD_UTF8_CHARCLASS_SPACE(); while (hardcount < max && scan < loceol && !(*scan == ' ' || swash_fetch(PL_utf8_space,(U8*)scan, do_utf8))) { @@ -4551,7 +4541,7 @@ S_regrepeat(pTHX_ const regnode *p, I32 max) case DIGIT: if (do_utf8) { loceol = PL_regeol; - LOAD_UTF8_CHARCLASS(digit,"0"); + LOAD_UTF8_CHARCLASS_DIGIT(); while (hardcount < max && scan < loceol && swash_fetch(PL_utf8_digit, (U8*)scan, do_utf8)) { scan += UTF8SKIP(scan); @@ -4565,7 +4555,7 @@ S_regrepeat(pTHX_ const regnode *p, I32 max) case NDIGIT: if (do_utf8) { loceol = PL_regeol; - LOAD_UTF8_CHARCLASS(digit,"0"); + LOAD_UTF8_CHARCLASS_DIGIT(); while (hardcount < max && scan < loceol && !swash_fetch(PL_utf8_digit, (U8*)scan, do_utf8)) { scan += UTF8SKIP(scan); @@ -4668,9 +4658,9 @@ Perl_regclass_swash(pTHX_ register const regnode* node, bool doinit, SV** listsv const U32 n = ARG(node); if (PL_regdata->what[n] == 's') { - SV *rv = (SV*)PL_regdata->data[n]; - AV *av = (AV*)SvRV((SV*)rv); - SV **ary = AvARRAY(av); + SV * const rv = (SV*)PL_regdata->data[n]; + AV * const av = (AV*)SvRV((SV*)rv); + SV **const ary = AvARRAY(av); SV **a, **b; /* See the end of regcomp.c:S_reglass() for @@ -4735,7 +4725,7 @@ S_reginclass(pTHX_ register const regnode *n, register const U8* p, STRLEN* lenp match = TRUE; if (!match) { AV *av; - SV *sw = regclass_swash(n, TRUE, 0, (SV**)&av); + SV * const sw = regclass_swash(n, TRUE, 0, (SV**)&av); if (sw) { if (swash_fetch(sw, p, do_utf8)) @@ -4743,11 +4733,10 @@ S_reginclass(pTHX_ register const regnode *n, register const U8* p, STRLEN* lenp else if (flags & ANYOF_FOLD) { if (!match && lenp && av) { I32 i; - for (i = 0; i <= av_len(av); i++) { - SV* sv = *av_fetch(av, i, FALSE); + SV* const sv = *av_fetch(av, i, FALSE); STRLEN len; - const char *s = SvPV(sv, len); + const char * const s = SvPV_const(sv, len); if (len <= plen && memEQ(s, (char*)p, len)) { *lenp = len; @@ -4898,12 +4887,12 @@ S_reghopmaybe3(pTHX_ U8* s, I32 off, U8* lim) static void restore_pos(pTHX_ void *arg) { - (void)arg; /* unused */ + PERL_UNUSED_ARG(arg); if (PL_reg_eval_set) { if (PL_reg_oldsaved) { PL_reg_re->subbeg = PL_reg_oldsaved; PL_reg_re->sublen = PL_reg_oldsavedlen; -#ifdef PERL_COPY_ON_WRITE +#ifdef PERL_OLD_COPY_ON_WRITE PL_reg_re->saved_copy = PL_nrs; #endif RX_MATCH_COPIED_on(PL_reg_re); @@ -4917,8 +4906,8 @@ restore_pos(pTHX_ void *arg) STATIC void S_to_utf8_substr(pTHX_ register regexp *prog) { - SV* sv; if (prog->float_substr && !prog->float_utf8) { + SV* sv; prog->float_utf8 = sv = newSVsv(prog->float_substr); sv_utf8_upgrade(sv); if (SvTAIL(prog->float_substr)) @@ -4927,6 +4916,7 @@ S_to_utf8_substr(pTHX_ register regexp *prog) prog->check_utf8 = sv; } if (prog->anchored_substr && !prog->anchored_utf8) { + SV* sv; prog->anchored_utf8 = sv = newSVsv(prog->anchored_substr); sv_utf8_upgrade(sv); if (SvTAIL(prog->anchored_substr)) @@ -4939,8 +4929,8 @@ S_to_utf8_substr(pTHX_ register regexp *prog) STATIC void S_to_byte_substr(pTHX_ register regexp *prog) { - SV* sv; if (prog->float_utf8 && !prog->float_substr) { + SV* sv; prog->float_substr = sv = newSVsv(prog->float_utf8); if (sv_utf8_downgrade(sv, TRUE)) { if (SvTAIL(prog->float_utf8)) @@ -4953,6 +4943,7 @@ S_to_byte_substr(pTHX_ register regexp *prog) prog->check_substr = sv; } if (prog->anchored_utf8 && !prog->anchored_substr) { + SV* sv; prog->anchored_substr = sv = newSVsv(prog->anchored_utf8); if (sv_utf8_downgrade(sv, TRUE)) { if (SvTAIL(prog->anchored_utf8))