X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/9cd990bfb433aa02bf17c9b9f23ddc7c56a9a016..0fbfbb97fd8bda2b2f51041f575b8e41691c21f0:/regexec.c diff --git a/regexec.c b/regexec.c index 3b4a889..e74ca18 100644 --- a/regexec.c +++ b/regexec.c @@ -536,12 +536,10 @@ S_isFOO_utf8_lc(pTHX_ const U8 classnum, const U8* character) case _CC_ENUM_BLANK: return is_HORIZWS_high(character); case _CC_ENUM_XDIGIT: return is_XDIGIT_high(character); case _CC_ENUM_VERTSPACE: return is_VERTWS_high(character); - default: return 0; /* Things like CNTRL are always - below 256 */ + default: break; } - assert(0); /* NOTREACHED */ - return FALSE; + return FALSE; /* Things like CNTRL are always below 256 */ } /* @@ -705,6 +703,7 @@ Perl_re_intuit_start(pTHX_ goto fail; } + RX_MATCH_UTF8_set(rx,utf8_target); reginfo->is_utf8_target = cBOOL(utf8_target); reginfo->info_aux = NULL; reginfo->strbeg = strbeg; @@ -774,7 +773,7 @@ Perl_re_intuit_start(pTHX_ * be too fiddly (e.g. REXEC_IGNOREPOS). */ if ( strpos != strbeg - && (prog->intflags & (PREGf_ANCH_BOL|PREGf_ANCH_SBOL))) + && (prog->intflags & PREGf_ANCH_SBOL)) { DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " Not at start...\n")); @@ -898,7 +897,7 @@ Perl_re_intuit_start(pTHX_ /* If the regex is absolutely anchored to either the start of the - * string (BOL,SBOL) or to pos() (ANCH_GPOS), then + * string (SBOL) or to pos() (ANCH_GPOS), then * check_offset_max represents an upper bound on the string where * the substr could start. For the ANCH_GPOS case, we assume that * the caller of intuit will have already set strpos to @@ -2619,6 +2618,7 @@ Perl_regexec_flags(pTHX_ REGEXP * const rx, char *stringarg, char *strend, } RX_MATCH_TAINTED_off(rx); + RX_MATCH_UTF8_set(rx, utf8_target); reginfo->prog = rx; /* Yes, sorry that this is confusing. */ reginfo->intuit = 0; @@ -2639,7 +2639,6 @@ Perl_regexec_flags(pTHX_ REGEXP * const rx, char *stringarg, char *strend, magic belonging to this SV. Not newSVsv, either, as it does not COW. */ - assert(!IS_PADGV(sv)); reginfo->sv = newSV(0); SvSetSV_nosteal(reginfo->sv, sv); SAVEFREESV(reginfo->sv); @@ -2717,7 +2716,7 @@ Perl_regexec_flags(pTHX_ REGEXP * const rx, char *stringarg, char *strend, } /* Simplest case: anchored match need be tried only once. */ - /* [unless only anchor is BOL and multiline is set] */ + /* [unless only anchor is MBOL - implying multiline is set] */ if (prog->intflags & (PREGf_ANCH & ~PREGf_ANCH_GPOS)) { if (s == startpos && regtry(reginfo, &s)) goto got_it; @@ -2971,7 +2970,7 @@ Perl_regexec_flags(pTHX_ REGEXP * const rx, char *stringarg, char *strend, } DEBUG_EXECUTE_r({ SV * const prop = sv_newmortal(); - regprop(prog, prop, c, reginfo); + regprop(prog, prop, c, reginfo, NULL); { RE_PV_QUOTED_DECL(quoted,utf8_target,PERL_DEBUG_PAD_ZERO(1), s,strend-s,60); @@ -3077,7 +3076,8 @@ Perl_regexec_flags(pTHX_ REGEXP * const rx, char *stringarg, char *strend, * and replaced it with this one. Yves */ DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log, - "String does not contain required substring, cannot match.\n" + "%sString does not contain required substring, cannot match.%s\n", + PL_colors[4], PL_colors[5] )); goto phooey; } @@ -3139,8 +3139,6 @@ got_it: if (RXp_PAREN_NAMES(prog)) (void)hv_iterinit(RXp_PAREN_NAMES(prog)); - RX_MATCH_UTF8_set(rx, utf8_target); - /* make sure $`, $&, $', and $digit will work later */ if ( !(flags & REXEC_NOT_FIRST) ) S_reg_set_capture_string(aTHX_ rx, @@ -3993,7 +3991,7 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) SV * const prop = sv_newmortal(); regnode *rnext=regnext(scan); DUMP_EXEC_POS( locinput, scan, utf8_target ); - regprop(rex, prop, scan, reginfo); + regprop(rex, prop, scan, reginfo, NULL); PerlIO_printf(Perl_debug_log, "%3"IVdf":%*s%s(%"IVdf")\n", @@ -4015,8 +4013,7 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) assert(nextchr < 256 && (nextchr >= 0 || nextchr == NEXTCHR_EOS)); switch (state_num) { - case BOL: /* /^../ */ - case SBOL: /* /^../s */ + case SBOL: /* /^../ and /\A../ */ if (locinput == reginfo->strbeg) break; sayNO; @@ -4039,21 +4036,22 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) st->u.keeper.val = rex->offs[0].start; rex->offs[0].start = locinput - reginfo->strbeg; PUSH_STATE_GOTO(KEEPS_next, next, locinput); - assert(0); /*NOTREACHED*/ + /* NOTREACHED */ + assert(0); + case KEEPS_next_fail: /* rollback the start point change */ rex->offs[0].start = st->u.keeper.val; sayNO_SILENT; - assert(0); /*NOTREACHED*/ + /* NOTREACHED */ + assert(0); case MEOL: /* /..$/m */ if (!NEXTCHR_IS_EOS && nextchr != '\n') sayNO; break; - case EOL: /* /..$/ */ - /* FALLTHROUGH */ - case SEOL: /* /..$/s */ + case SEOL: /* /..$/ */ if (!NEXTCHR_IS_EOS && nextchr != '\n') sayNO; if (reginfo->strend - locinput > 1) @@ -4095,7 +4093,8 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5]) ); sayNO_SILENT; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } /* FALLTHROUGH */ case TRIE: /* (ab|cd) */ @@ -4282,7 +4281,8 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) ); goto trie_first_try; /* jump into the fail handler */ }} - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case TRIE_next_fail: /* we failed - try next alternative */ { @@ -4396,7 +4396,8 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) if (ST.accepted > 1 || has_cutgroup) { PUSH_STATE_GOTO(TRIE_next, scan, (char*)uc); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } /* only one choice left - just continue */ DEBUG_EXECUTE_r({ @@ -4420,7 +4421,8 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) locinput = (char*)uc; continue; /* execute rest of RE */ - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } #undef ST @@ -5187,9 +5189,6 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) case TAIL: /* placeholder while compiling (A|B|C) */ break; - case BACK: /* ??? doesn't appear to be used ??? */ - break; - #undef ST #define ST st->u.eval { @@ -5229,7 +5228,8 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) /* and then jump to the code we share with EVAL */ goto eval_recurse_doit; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case EVAL: /* /(?{A})B/ /(??{A})B/ and /(?(?{A})X|Y)B/ */ if (cur_eval && cur_eval->locinput==locinput) { @@ -5445,7 +5445,7 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) re_sv = rex->engine->op_comp(aTHX_ &ret, 1, NULL, rex->engine, NULL, NULL, /* copy /msix etc to inner pattern */ - scan->flags, + ARG2L(scan), pm_flags); if (!(SvFLAGS(ret) @@ -5510,7 +5510,8 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) cur_eval = st; /* now continue from first node in postoned RE */ PUSH_YES_STATE_GOTO(EVAL_AB, startpoint, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } case EVAL_AB: /* cleanup after a successful (??{A})B */ @@ -5619,7 +5620,7 @@ S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog) } } goto fake_end; - /*NOTREACHED*/ + /* NOTREACHED */ case GROUPP: /* (?(1)) */ n = ARG(scan); /* which paren pair */ @@ -5769,19 +5770,22 @@ NULL ST.lastloc = NULL; /* this will be updated by WHILEM */ PUSH_YES_STATE_GOTO(CURLYX_end, PREVOPER(next), locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } case CURLYX_end: /* just finished matching all of A*B */ cur_curlyx = ST.prev_curlyx; sayYES; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case CURLYX_end_fail: /* just failed to match all of A*B */ regcpblow(ST.cp); cur_curlyx = ST.prev_curlyx; sayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); #undef ST @@ -5819,7 +5823,8 @@ NULL REGCP_SET(ST.lastcp); PUSH_STATE_GOTO(WHILEM_A_pre, A, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } /* If degenerate A matches "", assume A done. */ @@ -5931,7 +5936,8 @@ NULL REGCP_SET(ST.lastcp); PUSH_YES_STATE_GOTO(WHILEM_B_min, ST.save_curlyx->u.curlyx.B, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } /* Prefer A over B for maximal matching. */ @@ -5942,24 +5948,28 @@ NULL cur_curlyx->u.curlyx.lastloc = locinput; REGCP_SET(ST.lastcp); PUSH_STATE_GOTO(WHILEM_A_max, A, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } goto do_whilem_B_max; } - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case WHILEM_B_min: /* just matched B in a minimal match */ case WHILEM_B_max: /* just matched B in a maximal match */ cur_curlyx = ST.save_curlyx; sayYES; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case WHILEM_B_max_fail: /* just failed to match B in a maximal match */ cur_curlyx = ST.save_curlyx; cur_curlyx->u.curlyx.lastloc = ST.save_lastloc; cur_curlyx->u.curlyx.count--; CACHEsayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case WHILEM_A_min_fail: /* just failed to match A in a minimal match */ /* FALLTHROUGH */ @@ -5969,7 +5979,8 @@ NULL cur_curlyx->u.curlyx.lastloc = ST.save_lastloc; cur_curlyx->u.curlyx.count--; CACHEsayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case WHILEM_A_max_fail: /* just failed to match A in a maximal match */ REGCP_UNWIND(ST.lastcp); @@ -5995,7 +6006,8 @@ NULL cur_curlyx = cur_curlyx->u.curlyx.prev_curlyx; PUSH_YES_STATE_GOTO(WHILEM_B_max, ST.save_curlyx->u.curlyx.B, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case WHILEM_B_min_fail: /* just failed to match B in a minimal match */ cur_curlyx = ST.save_curlyx; @@ -6029,7 +6041,8 @@ NULL PUSH_STATE_GOTO(WHILEM_A_min, /*A*/ NEXTOPER(ST.save_curlyx->u.curlyx.me) + EXTRA_STEP_2ARGS, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); #undef ST #define ST st->u.branch @@ -6054,13 +6067,15 @@ NULL } else { PUSH_STATE_GOTO(BRANCH_next, scan, locinput); } - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case CUTGROUP: /* /(*THEN)/ */ sv_yes_mark = st->u.mark.mark_name = scan->flags ? NULL : MUTABLE_SV(rexi->data->data[ ARG( scan ) ]); PUSH_STATE_GOTO(CUTGROUP_next, next, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case CUTGROUP_next_fail: do_cutgroup = 1; @@ -6068,11 +6083,13 @@ NULL if (st->u.mark.mark_name) sv_commit = st->u.mark.mark_name; sayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case BRANCH_next: sayYES; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case BRANCH_next_fail: /* that branch failed; try the next, if any */ if (do_cutgroup) { @@ -6094,7 +6111,8 @@ NULL sayNO_SILENT; } continue; /* execute next BRANCH[J] op */ - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case MINMOD: /* next op will be non-greedy, e.g. A*? */ minmod = 1; @@ -6138,7 +6156,8 @@ NULL curlym_do_A: /* execute the A in /A{m,n}B/ */ PUSH_YES_STATE_GOTO(CURLYM_A, ST.A, locinput); /* match A */ - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case CURLYM_A: /* we've just matched an A */ ST.count++; @@ -6274,7 +6293,8 @@ NULL } PUSH_STATE_GOTO(CURLYM_B, ST.B, locinput); /* match B */ - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case CURLYM_B_fail: /* just failed to match a B */ REGCP_UNWIND(ST.cp); @@ -6452,8 +6472,8 @@ NULL REGCP_SET(ST.cp); goto curly_try_B_max; } - assert(0); /* NOTREACHED */ - + /* NOTREACHED */ + assert(0); case CURLY_B_min_known_fail: /* failed to find B in a non-greedy match where c1,c2 valid */ @@ -6528,8 +6548,8 @@ NULL } PUSH_STATE_GOTO(CURLY_B_min_known, ST.B, locinput); } - assert(0); /* NOTREACHED */ - + /* NOTREACHED */ + assert(0); case CURLY_B_min_fail: /* failed to find B in a non-greedy match where c1,c2 invalid */ @@ -6561,8 +6581,8 @@ NULL } } sayNO; - assert(0); /* NOTREACHED */ - + /* NOTREACHED */ + assert(0); curly_try_B_max: /* a successful greedy match: now try to match B */ @@ -6592,7 +6612,8 @@ NULL if (ST.c1 == CHRTEST_VOID || could_match) { CURLY_SETPAREN(ST.paren, ST.count); PUSH_STATE_GOTO(CURLY_B_max, ST.B, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } } /* FALLTHROUGH */ @@ -6711,7 +6732,8 @@ NULL /* execute body of (?...A) */ PUSH_YES_STATE_GOTO(IFMATCH_A, NEXTOPER(NEXTOPER(scan)), newstart); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } case IFMATCH_A_fail: /* body of (?...A) failed */ @@ -6751,7 +6773,8 @@ NULL if (!scan->flags) sv_yes_mark = sv_commit = MUTABLE_SV(rexi->data->data[ ARG( scan ) ]); PUSH_STATE_GOTO(COMMIT_next, next, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case COMMIT_next_fail: no_final = 1; @@ -6759,7 +6782,8 @@ NULL case OPFAIL: /* (*FAIL) */ sayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); #define ST st->u.mark case MARKPOINT: /* (*MARK:foo) */ @@ -6769,12 +6793,14 @@ NULL mark_state = st; ST.mark_loc = locinput; PUSH_YES_STATE_GOTO(MARKPOINT_next, next, locinput); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case MARKPOINT_next: mark_state = ST.prev_mark; sayYES; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case MARKPOINT_next_fail: if (popmark && sv_eq(ST.mark_name,popmark)) @@ -6795,7 +6821,8 @@ NULL sv_yes_mark = mark_state ? mark_state->u.mark.mark_name : NULL; sayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); case SKIP: /* (*SKIP) */ if (scan->flags) { @@ -6840,7 +6867,8 @@ NULL } no_final = 1; sayNO; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); #undef ST case LNBREAK: /* \R */ @@ -6874,7 +6902,8 @@ NULL /* switch break jumps here */ scan = next; /* prepare to execute the next op and ... */ continue; /* ... jump back to the top, reusing st */ - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); push_yes_state: /* push a state that backtracks on success */ @@ -6917,7 +6946,8 @@ NULL locinput = pushinput; st = newst; continue; - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } } @@ -6926,7 +6956,7 @@ NULL * the terminating point. */ Perl_croak(aTHX_ "corrupted regexp pointers"); - /*NOTREACHED*/ + /* NOTREACHED */ sayNO; yes: @@ -7553,7 +7583,8 @@ S_regrepeat(pTHX_ regexp *prog, char **startposp, const regnode *p, default: Perl_croak(aTHX_ "panic: regrepeat() called with unrecognized node type %d='%s'", OP(p), PL_reg_name[OP(p)]); - assert(0); /* NOTREACHED */ + /* NOTREACHED */ + assert(0); } @@ -7567,7 +7598,7 @@ S_regrepeat(pTHX_ regexp *prog, char **startposp, const regnode *p, GET_RE_DEBUG_FLAGS_DECL; DEBUG_EXECUTE_r({ SV * const prop = sv_newmortal(); - regprop(prog, prop, p, reginfo); + regprop(prog, prop, p, reginfo, NULL); PerlIO_printf(Perl_debug_log, "%*s %s can match %"IVdf" times out of %"IVdf"...\n", REPORT_CODE_OFF + depth*2, "", SvPVX_const(prop),(IV)c,(IV)max); @@ -7593,121 +7624,9 @@ Perl_regclass_swash(pTHX_ const regexp *prog, const regnode* node, bool doinit, *altsvp = NULL; } - return newSVsv(_get_regclass_nonbitmap_data(prog, node, doinit, listsvp, NULL)); + return newSVsv(_get_regclass_nonbitmap_data(prog, node, doinit, listsvp, NULL, NULL)); } -SV * -Perl__get_regclass_nonbitmap_data(pTHX_ const regexp *prog, - const regnode* node, - bool doinit, - SV** listsvp, - SV** only_utf8_locale_ptr) -{ - /* For internal core use only. - * Returns the swash for the input 'node' in the regex 'prog'. - * If is 'true', will attempt to create the swash if not already - * done. - * If is non-null, will return the printable contents of the - * swash. This can be used to get debugging information even before the - * swash exists, by calling this function with 'doinit' set to false, in - * which case the components that will be used to eventually create the - * swash are returned (in a printable form). - * Tied intimately to how regcomp.c sets up the data structure */ - - SV *sw = NULL; - SV *si = NULL; /* Input swash initialization string */ - SV* invlist = NULL; - - RXi_GET_DECL(prog,progi); - const struct reg_data * const data = prog ? progi->data : NULL; - - PERL_ARGS_ASSERT__GET_REGCLASS_NONBITMAP_DATA; - - assert(ANYOF_FLAGS(node) - & (ANYOF_UTF8|ANYOF_NONBITMAP_NON_UTF8|ANYOF_LOC_FOLD)); - - if (data && data->count) { - const U32 n = ARG(node); - - if (data->what[n] == 's') { - SV * const rv = MUTABLE_SV(data->data[n]); - AV * const av = MUTABLE_AV(SvRV(rv)); - SV **const ary = AvARRAY(av); - U8 swash_init_flags = _CORE_SWASH_INIT_ACCEPT_INVLIST; - - si = *ary; /* ary[0] = the string to initialize the swash with */ - - /* Elements 3 and 4 are either both present or both absent. [3] is - * any inversion list generated at compile time; [4] indicates if - * that inversion list has any user-defined properties in it. */ - if (av_tindex(av) >= 2) { - if (only_utf8_locale_ptr - && ary[2] - && ary[2] != &PL_sv_undef) - { - *only_utf8_locale_ptr = ary[2]; - } - else { - assert(only_utf8_locale_ptr); - *only_utf8_locale_ptr = NULL; - } - - if (av_tindex(av) >= 3) { - invlist = ary[3]; - if (SvUV(ary[4])) { - swash_init_flags |= _CORE_SWASH_INIT_USER_DEFINED_PROPERTY; - } - } - else { - invlist = NULL; - } - } - - /* Element [1] is reserved for the set-up swash. If already there, - * return it; if not, create it and store it there */ - if (ary[1] && SvROK(ary[1])) { - sw = ary[1]; - } - else if (doinit && ((si && si != &PL_sv_undef) - || (invlist && invlist != &PL_sv_undef))) { - assert(si); - sw = _core_swash_init("utf8", /* the utf8 package */ - "", /* nameless */ - si, - 1, /* binary */ - 0, /* not from tr/// */ - invlist, - &swash_init_flags); - (void)av_store(av, 1, sw); - } - } - } - - /* If requested, return a printable version of what this swash matches */ - if (listsvp) { - SV* matches_string = newSVpvs(""); - - /* The swash should be used, if possible, to get the data, as it - * contains the resolved data. But this function can be called at - * compile-time, before everything gets resolved, in which case we - * return the currently best available information, which is the string - * that will eventually be used to do that resolving, 'si' */ - if ((! sw || (invlist = _get_swash_invlist(sw)) == NULL) - && (si && si != &PL_sv_undef)) - { - sv_catsv(matches_string, si); - } - - /* Add the inversion list to whatever we have. This may have come from - * the swash, or from an input parameter */ - if (invlist) { - sv_catsv(matches_string, _invlist_contents(invlist)); - } - *listsvp = matches_string; - } - - return sw; -} #endif /* !defined(PERL_IN_XSUB_RE) || defined(PLUGGABLE_RE_EXTENSION) */ /* @@ -7750,22 +7669,25 @@ S_reginclass(pTHX_ regexp * const prog, const regnode * const n, const U8* const } /* If this character is potentially in the bitmap, check it */ - if (c < 256) { + if (c < NUM_ANYOF_CODE_POINTS) { if (ANYOF_BITMAP_TEST(n, c)) match = TRUE; - else if (flags & ANYOF_NON_UTF8_NON_ASCII_ALL - && ! utf8_target - && ! isASCII(c)) + else if ((flags & ANYOF_MATCHES_ALL_NON_UTF8_NON_ASCII) + && ! utf8_target + && ! isASCII(c)) { match = TRUE; } else if (flags & ANYOF_LOCALE_FLAGS) { - if (flags & ANYOF_LOC_FOLD) { - if (ANYOF_BITMAP_TEST(n, PL_fold_locale[c])) { - match = TRUE; - } + if ((flags & ANYOF_LOC_FOLD) + && c < 256 + && ANYOF_BITMAP_TEST(n, PL_fold_locale[c])) + { + match = TRUE; } - if (! match && ANYOF_POSIXL_TEST_ANY_SET(n)) { + else if (ANYOF_POSIXL_TEST_ANY_SET(n) + && c < 256 + ) { /* The data structure is arranged so bits 0, 2, 4, ... are set * if the class includes the Posix character class given by @@ -7818,18 +7740,20 @@ S_reginclass(pTHX_ regexp * const prog, const regnode * const n, const U8* const /* If the bitmap didn't (or couldn't) match, and something outside the * bitmap could match, try that. */ if (!match) { - if (c >= 256 && (flags & ANYOF_ABOVE_LATIN1_ALL)) { - match = TRUE; /* Everything above 255 matches */ + if (c >= NUM_ANYOF_CODE_POINTS + && (flags & ANYOF_MATCHES_ALL_ABOVE_BITMAP)) + { + match = TRUE; /* Everything above the bitmap matches */ } - else if ((flags & ANYOF_NONBITMAP_NON_UTF8) - || (utf8_target && (flags & ANYOF_UTF8)) + else if ((flags & ANYOF_HAS_NONBITMAP_NON_UTF8_MATCHES) + || (utf8_target && (flags & ANYOF_HAS_UTF8_NONBITMAP_MATCHES)) || ((flags & ANYOF_LOC_FOLD) && IN_UTF8_CTYPE_LOCALE - && ARG(n) != ANYOF_NONBITMAP_EMPTY)) + && ARG(n) != ANYOF_ONLY_HAS_BITMAP)) { SV* only_utf8_locale = NULL; SV * const sw = _get_regclass_nonbitmap_data(prog, n, TRUE, 0, - &only_utf8_locale); + &only_utf8_locale, NULL); if (sw) { U8 utf8_buffer[2]; U8 * utf8_p;