X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=regexec.c;h=d19b82ea5a613ea6f0086e8485392b7278aa12ef;hb=b42969c054c571b3b721d68818184e59ca761058;hp=fa3eb1ba5770a216846cc7c85c5b323723ea671a;hpb=a6e20a404b42cad25eb97280fa2bcacffda64cd6;p=p5sagit%2Fp5-mst-13.2.git diff --git a/regexec.c b/regexec.c index fa3eb1b..d19b82e 100644 --- a/regexec.c +++ b/regexec.c @@ -416,8 +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; @@ -615,7 +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; @@ -861,7 +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); @@ -1038,14 +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++; @@ -1084,14 +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, @@ -1118,8 +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 @@ -1650,7 +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; @@ -2146,7 +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); @@ -2192,7 +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 @@ -2415,6 +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; @@ -2582,8 +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; @@ -2612,7 +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; @@ -2651,7 +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; @@ -2678,7 +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; @@ -2807,8 +2805,7 @@ S_regmatch(pTHX_ regnode *prog) sayNO; if (NATIVE_TO_UNI(*(U8*)s) != utf8n_to_uvuni((U8*)l, UTF8_MAXBYTES, &ulen, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY)) + uniflags)) sayNO; l += ulen; s ++; @@ -2822,8 +2819,7 @@ S_regmatch(pTHX_ regnode *prog) sayNO; if (NATIVE_TO_UNI(*((U8*)l)) != utf8n_to_uvuni((U8*)s, UTF8_MAXBYTES, &ulen, - ckWARN(WARN_UTF8) ? - 0 : UTF8_ALLOW_ANY)) + uniflags)) sayNO; s += ulen; l ++; @@ -3554,7 +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, @@ -3940,16 +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); } } } @@ -4010,8 +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++; } @@ -4022,8 +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; @@ -4059,8 +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. */ @@ -4109,8 +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); } @@ -4132,8 +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); } @@ -4670,9 +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 @@ -4737,7 +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)) @@ -4745,11 +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_const(sv, len); + const char * const s = SvPV_const(sv, len); if (len <= plen && memEQ(s, (char*)p, len)) { *lenp = len; @@ -4900,7 +4887,7 @@ 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; @@ -4919,8 +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)) @@ -4929,6 +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)) @@ -4941,8 +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)) @@ -4955,6 +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))