X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/19c6248174a2dd57782271308b166ab6bc3e63c8..39c012bc2fc2f1cf310f6d4ba499ec58a7bad900:/toke.c diff --git a/toke.c b/toke.c index 08607ef..439ce15 100644 --- a/toke.c +++ b/toke.c @@ -148,6 +148,9 @@ static const char ident_too_long[] = "Identifier too long"; /* LEX_* are values for PL_lex_state, the state of the lexer. * They are arranged oddly so that the guard on the switch statement * can get by with a single comparison (if the compiler is smart enough). + * + * These values refer to the various states within a sublex parse, + * i.e. within a double quotish string */ /* #define LEX_NOTPARSING 11 is done in perl.h. */ @@ -272,9 +275,9 @@ static const char* const lex_state_names[] = { * The UNIDOR macro is for unary functions that can be followed by the // * operator (such as C). */ -#define UNI2(f,x) { \ +#define UNI3(f,x,have_x) { \ pl_yylval.ival = f; \ - PL_expect = x; \ + if (have_x) PL_expect = x; \ PL_bufptr = s; \ PL_last_uni = PL_oldbufptr; \ PL_last_lop_op = f; \ @@ -283,22 +286,14 @@ static const char* const lex_state_names[] = { s = PEEKSPACE(s); \ return REPORT( *s=='(' ? (int)FUNC1 : (int)UNIOP ); \ } -#define UNI(f) UNI2(f,XTERM) -#define UNIDOR(f) UNI2(f,XTERMORDORDOR) +#define UNI(f) UNI3(f,XTERM,1) +#define UNIDOR(f) UNI3(f,XTERMORDORDOR,1) #define UNIPROTO(f,optional) { \ if (optional) PL_last_uni = PL_oldbufptr; \ OPERATOR(f); \ } -#define UNIBRACK(f) { \ - pl_yylval.ival = f; \ - PL_bufptr = s; \ - PL_last_uni = PL_oldbufptr; \ - if (*s == '(') \ - return REPORT( (int)FUNC1 ); \ - s = PEEKSPACE(s); \ - return REPORT( (*s == '(') ? (int)FUNC1 : (int)UNIOP ); \ - } +#define UNIBRACK(f) UNI3(f,0,0) /* grandfather return to old style */ #define OLDLOP(f) \ @@ -359,7 +354,7 @@ static struct debug_tokens { { GIVEN, TOKENTYPE_IVAL, "GIVEN" }, { HASHBRACK, TOKENTYPE_NONE, "HASHBRACK" }, { IF, TOKENTYPE_IVAL, "IF" }, - { LABEL, TOKENTYPE_PVAL, "LABEL" }, + { LABEL, TOKENTYPE_OPVAL, "LABEL" }, { LOCAL, TOKENTYPE_IVAL, "LOCAL" }, { LOOPEX, TOKENTYPE_OPNUM, "LOOPEX" }, { LSTOP, TOKENTYPE_OPNUM, "LSTOP" }, @@ -537,24 +532,28 @@ S_no_op(pTHX_ const char *const what, char *s) s = oldbp; else PL_bufptr = s; - yywarn(Perl_form(aTHX_ "%s found where operator expected", what), 0); + yywarn(Perl_form(aTHX_ "%s found where operator expected", what), UTF ? SVf_UTF8 : 0); if (ckWARN_d(WARN_SYNTAX)) { if (is_first) Perl_warner(aTHX_ packWARN(WARN_SYNTAX), "\t(Missing semicolon on previous line?)\n"); else if (PL_oldoldbufptr && isIDFIRST_lazy_if(PL_oldoldbufptr,UTF)) { const char *t; - for (t = PL_oldoldbufptr; (isALNUM_lazy_if(t,UTF) || *t == ':'); t++) + for (t = PL_oldoldbufptr; (isALNUM_lazy_if(t,UTF) || *t == ':'); + t += UTF ? UTF8SKIP(t) : 1) NOOP; if (t < PL_bufptr && isSPACE(*t)) Perl_warner(aTHX_ packWARN(WARN_SYNTAX), - "\t(Do you need to predeclare %.*s?)\n", - (int)(t - PL_oldoldbufptr), PL_oldoldbufptr); + "\t(Do you need to predeclare %"SVf"?)\n", + SVfARG(newSVpvn_flags(PL_oldoldbufptr, (STRLEN)(t - PL_oldoldbufptr), + SVs_TEMP | (UTF ? SVf_UTF8 : 0)))); } else { assert(s >= oldbp); Perl_warner(aTHX_ packWARN(WARN_SYNTAX), - "\t(Missing operator before %.*s?)\n", (int)(s - oldbp), oldbp); + "\t(Missing operator before %"SVf"?)\n", + SVfARG(newSVpvn_flags(oldbp, (STRLEN)(s - oldbp), + SVs_TEMP | (UTF ? SVf_UTF8 : 0)))); } } PL_bufptr = oldbp; @@ -2345,14 +2344,10 @@ S_tokeq(pTHX_ SV *sv) * converting things like "\u\Lgnat" into ucfirst(lc("gnat")). They * interact with PL_lex_state, and create fake ( ... ) argument lists * to handle functions and concatenation. - * They assume that whoever calls them will be setting up a fake - * join call, because each subthing puts a ',' after it. This lets - * "lower \luPpEr" - * become - * join($, , 'lower ', lcfirst( 'uPpEr', ) ,) - * - * (I'm not sure whether the spurious commas at the end of lcfirst's - * arguments and join's arguments are created or not). + * For example, + * "foo\lbar" + * is tokenised as + * stringify ( const[foo] concat lcfirst ( const[bar] ) ) */ /* @@ -2446,6 +2441,7 @@ S_sublex_push(pTHX) SAVEI32(PL_lex_casemods); SAVEI32(PL_lex_starts); SAVEI8(PL_lex_state); + SAVEPPTR(PL_sublex_info.re_eval_start); SAVEVPTR(PL_lex_inpat); SAVEI16(PL_lex_inwhat); SAVECOPLINE(PL_curcop); @@ -2462,6 +2458,7 @@ S_sublex_push(pTHX) PL_linestr = PL_lex_stuff; PL_lex_stuff = NULL; + PL_sublex_info.re_eval_start = NULL; PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart = SvPVX(PL_linestr); @@ -2571,8 +2568,11 @@ S_sublex_done(pTHX) /* scan_const - Extracts a pattern, double-quoted string, or transliteration. This - is terrifying code. + Extracts the next constant part of a pattern, double-quoted string, + or transliteration. This is terrifying code. + + For example, in parsing the double-quoted string "ab\x63$d", it would + stop at the '$' and return an OP_CONST containing 'abc'. It looks at PL_lex_inwhat and PL_lex_inpat to find out whether it's processing a pattern (PL_lex_inpat is true), a transliteration @@ -2580,15 +2580,22 @@ S_sublex_done(pTHX) Returns a pointer to the character scanned up to. If this is advanced from the start pointer supplied (i.e. if anything was - successfully parsed), will leave an OP for the substring scanned + successfully parsed), will leave an OP_CONST for the substring scanned in pl_yylval. Caller must intuit reason for not parsing further by looking at the next characters herself. In patterns: - backslashes: - constants: \N{NAME} only - case and quoting: \U \Q \E - stops on @ and $, but not for $ as tail anchor + expand: + \N{ABC} => \N{U+41.42.43} + + pass through: + all other \-char, including \N and \N{ apart from \N{ABC} + + stops on: + @ and $ where it appears to be a var, but not for $ as tail anchor + \l \L \u \U \Q \E + (?{ or (??{ + In transliterations: characters are VERY literal, except for - not at the start or end @@ -2618,7 +2625,7 @@ S_sublex_done(pTHX) it's a tail anchor if $ is the last thing in the string, or if it's followed by one of "()| \r\n\t" - \1 (backreferences) are turned into $1 + \1 (backreferences) are turned into $1 in substitutions The structure of the code is while (there's a character to process) { @@ -2657,6 +2664,7 @@ S_scan_const(pTHX_ char *start) register char *d = SvPVX(sv); /* destination for copies */ bool dorange = FALSE; /* are we in a translit range? */ bool didrange = FALSE; /* did we just finish a range? */ + bool in_charclass = FALSE; /* within /[...]/ */ bool has_utf8 = FALSE; /* Output constant is UTF8 */ bool this_utf8 = cBOOL(UTF); /* Is the source string assumed to be UTF8? But, this can @@ -2846,33 +2854,38 @@ S_scan_const(pTHX_ char *start) /* if we get here, we're not doing a transliteration */ - /* skip for regexp comments /(?#comment)/ and code /(?{code})/, - except for the last char, which will be done separately. */ + else if (*s == '[' && PL_lex_inpat && !in_charclass) { + char *s1 = s-1; + int esc = 0; + while (s1 >= start && *s1-- == '\\') + esc = !esc; + if (!esc) + in_charclass = TRUE; + } + + else if (*s == ']' && PL_lex_inpat && in_charclass) { + char *s1 = s-1; + int esc = 0; + while (s1 >= start && *s1-- == '\\') + esc = !esc; + if (!esc) + in_charclass = FALSE; + } + + /* skip for regexp comments /(?#comment)/, except for the last + * char, which will be done separately. + * Stop on (?{..}) and friends */ + else if (*s == '(' && PL_lex_inpat && s[1] == '?') { if (s[2] == '#') { while (s+1 < send && *s != ')') *d++ = NATIVE_TO_NEED(has_utf8,*s++); } - else if (s[2] == '{' /* This should match regcomp.c */ - || (s[2] == '?' && s[3] == '{')) + else if (!PL_lex_casemods && !in_charclass && + ( s[2] == '{' /* This should match regcomp.c */ + || (s[2] == '?' && s[3] == '{'))) { - I32 count = 1; - char *regparse = s + (s[2] == '{' ? 3 : 4); - char c; - - while (count && (c = *regparse)) { - if (c == '\\' && regparse[1]) - regparse++; - else if (c == '{') - count++; - else if (c == '}') - count--; - regparse++; - } - if (*regparse != ')') - regparse--; /* Leave one char for continuation. */ - while (s < regparse) - *d++ = NATIVE_TO_NEED(has_utf8,*s++); + break; } } @@ -2883,6 +2896,10 @@ S_scan_const(pTHX_ char *start) *d++ = NATIVE_TO_NEED(has_utf8,*s++); } + /* no further processing of single-quoted regex */ + else if (PL_lex_inpat && SvIVX(PL_linestr) == '\'') + goto default_action; + /* check for embedded arrays (@foo, @::foo, @'foo, @{foo}, @$foo, @+, @-) */ @@ -2966,7 +2983,7 @@ S_scan_const(pTHX_ char *start) /* FALL THROUGH */ default: { - if ((isALPHA(*s) || isDIGIT(*s))) + if ((isALNUMC(*s))) Perl_ck_warner(aTHX_ packWARN(WARN_MISC), "Unrecognized escape \\%c passed through", *s); @@ -3002,29 +3019,16 @@ S_scan_const(pTHX_ char *start) /* eg. \x24 indicates the hex constant 0x24 */ case 'x': - ++s; - if (*s == '{') { - char* const e = strchr(s, '}'); - I32 flags = PERL_SCAN_ALLOW_UNDERSCORES | - PERL_SCAN_DISALLOW_PREFIX; + { STRLEN len; + const char* error; - ++s; - if (!e) { - yyerror("Missing right brace on \\x{}"); + bool valid = grok_bslash_x(s, &uv, &len, &error, 1); + s += len; + if (! valid) { + yyerror(error); continue; } - len = e - s; - uv = NATIVE_TO_UNI(grok_hex(s, &len, &flags, NULL)); - s = e + 1; - } - else { - { - STRLEN len = 2; - I32 flags = PERL_SCAN_DISALLOW_PREFIX; - uv = NATIVE_TO_UNI(grok_hex(s, &len, &flags, NULL)); - s += len; - } } NUM_ESCAPE_INSERT: @@ -3552,6 +3556,9 @@ S_scan_const(pTHX_ char *start) } else if (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat) { type = "s"; typelen = 1; + } else if (PL_lex_inpat && SvIVX(PL_linestr) == '\'') { + type = "q"; + typelen = 1; } else { type = "qq"; typelen = 2; @@ -3728,7 +3735,7 @@ S_intuit_more(pTHX_ register char *s) * * First argument is the stuff after the first token, e.g. "bar". * - * Not a method if bar is a filehandle. + * Not a method if foo is a filehandle. * Not a method if foo is a subroutine prototyped to take a filehandle. * Not a method if it's really "Foo $bar" * Method if it's "foo $bar" @@ -3766,8 +3773,7 @@ S_intuit_method(pTHX_ char *start, GV *gv, CV *cv) return 0; } } - } else - gv = NULL; + } } s = scan_word(s, tmpbuf, sizeof tmpbuf, TRUE, &len); /* start is the beginning of the possible filehandle/object, @@ -3776,7 +3782,7 @@ S_intuit_method(pTHX_ char *start, GV *gv, CV *cv) */ if (*start == '$') { - if (gv || PL_last_lop_op == OP_PRINT || PL_last_lop_op == OP_SAY || + if (cv || PL_last_lop_op == OP_PRINT || PL_last_lop_op == OP_SAY || isUPPER(*PL_tokenbuf)) return 0; #ifdef PERL_MAD @@ -3803,7 +3809,7 @@ S_intuit_method(pTHX_ char *start, GV *gv, CV *cv) if (indirgv && GvCVu(indirgv)) return 0; /* filehandle or package name makes it a method */ - if (!gv || GvIO(indirgv) || gv_stashpvn(tmpbuf, len, UTF ? SVf_UTF8 : 0)) { + if (!cv || GvIO(indirgv) || gv_stashpvn(tmpbuf, len, UTF ? SVf_UTF8 : 0)) { #ifdef PERL_MAD soff = s - SvPVX(PL_linestr); #endif @@ -4227,6 +4233,7 @@ Perl_madlex(pTHX) case FUNC0SUB: case UNIOPSUB: case LSTOPSUB: + case LABEL: if (pl_yylval.opval) append_madprops(PL_thismad, pl_yylval.opval, 0); PL_thismad = 0; @@ -4287,10 +4294,6 @@ Perl_madlex(pTHX) } break; - /* pval */ - case LABEL: - break; - /* ival */ default: break; @@ -4614,7 +4617,7 @@ Perl_yylex(pTHX) case LEX_INTERPSTART: if (PL_bufptr == PL_bufend) return REPORT(sublex_done()); - DEBUG_T({ PerlIO_printf(Perl_debug_log, + DEBUG_T({ if(*PL_bufptr != '(') PerlIO_printf(Perl_debug_log, "### Interpolated variable\n"); }); PL_expect = XTERM; PL_lex_dojoin = (*PL_bufptr == '@'); @@ -4635,6 +4638,18 @@ Perl_yylex(pTHX) NEXTVAL_NEXTTOKE.ival = OP_JOIN; /* emulate join($", ...) */ force_next(FUNC); } + /* Convert (?{...}) and friends to 'do {...}' */ + if (PL_lex_inpat && *PL_bufptr == '(') { + PL_sublex_info.re_eval_start = PL_bufptr; + PL_bufptr += 2; + if (*PL_bufptr != '{') + PL_bufptr++; + start_force(PL_curforce); + /* XXX probably need a CURMAD(something) here */ + PL_expect = XTERMBLOCK; + force_next(DO); + } + if (PL_lex_starts++) { s = PL_bufptr; #ifdef PERL_MAD @@ -4680,6 +4695,24 @@ Perl_yylex(pTHX) Perl_croak(aTHX_ "Bad evalled substitution pattern"); PL_lex_repl = NULL; } + if (PL_sublex_info.re_eval_start) { + if (*PL_bufptr != ')') + Perl_croak(aTHX_ "Sequence (?{...}) not terminated with ')'"); + PL_bufptr++; + /* having compiled a (?{..}) expression, return the original + * text too, as a const */ + start_force(PL_curforce); + /* XXX probably need a CURMAD(something) here */ + NEXTVAL_NEXTTOKE.opval = + (OP*)newSVOP(OP_CONST, 0, + newSVpvn(PL_sublex_info.re_eval_start, + PL_bufptr - PL_sublex_info.re_eval_start)); + force_next(THING); + PL_sublex_info.re_eval_start = NULL; + PL_expect = XTERM; + return REPORT(','); + } + /* FALLTHROUGH */ case LEX_INTERPCONCAT: #ifdef DEBUGGING @@ -4690,12 +4723,10 @@ Perl_yylex(pTHX) if (PL_bufptr == PL_bufend) return REPORT(sublex_done()); - if (SvIVX(PL_linestr) == '\'') { + /* m'foo' still needs to be parsed for possible (?{...}) */ + if (SvIVX(PL_linestr) == '\'' && !PL_lex_inpat) { SV *sv = newSVsv(PL_linestr); - if (!PL_lex_inpat) - sv = tokeq(sv); - else if ( PL_hints & HINT_NEW_RE ) - sv = new_constant(NULL, 0, "qr", sv, sv, "q", 1); + sv = tokeq(sv); pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv); s = PL_bufend; } @@ -4761,7 +4792,12 @@ Perl_yylex(pTHX) if (isIDFIRST_lazy_if(s,UTF)) goto keylookup; { - unsigned char c = *s; + SV *dsv = newSVpvs_flags("", SVs_TEMP); + const char *c = UTF ? savepv(sv_uni_display(dsv, newSVpvn_flags(s, + UTF8SKIP(s), + SVs_TEMP | SVf_UTF8), + 10, UNI_DISPLAY_ISPRINT)) + : Perl_form(aTHX_ "\\x%02X", (unsigned char)*s); len = UTF ? Perl_utf8_length(aTHX_ (U8 *) PL_linestart, (U8 *) s) : (STRLEN) (s - PL_linestart); if (len > UNRECOGNIZED_PRECEDE_COUNT) { d = UTF ? (char *) Perl_utf8_hop(aTHX_ (U8 *) s, -UNRECOGNIZED_PRECEDE_COUNT) : s - UNRECOGNIZED_PRECEDE_COUNT; @@ -4769,7 +4805,10 @@ Perl_yylex(pTHX) d = PL_linestart; } *s = '\0'; - Perl_croak(aTHX_ "Unrecognized character \\x%02X; marked by <-- HERE after %s<-- HERE near column %d", c, d, (int) len + 1); + sv_setpv(dsv, d); + if (UTF) + SvUTF8_on(dsv); + Perl_croak(aTHX_ "Unrecognized character %s; marked by <-- HERE after %"SVf"<-- HERE near column %d", c, SVfARG(dsv), (int) len + 1); } case 4: case 26: @@ -5486,7 +5525,7 @@ Perl_yylex(pTHX) } sv = newSVpvn_flags(s, len, UTF ? SVf_UTF8 : 0); if (*d == '(') { - d = scan_str(d,TRUE,TRUE); + d = scan_str(d,TRUE,TRUE,FALSE); if (!d) { /* MUST advance bufptr here to avoid bogus "at end of line" context messages from yyerror(). @@ -6172,10 +6211,12 @@ Perl_yylex(pTHX) &len); while (isSPACE(*t)) t++; - if (*t == ';' && get_cvn_flags(tmpbuf, len, 0)) + if (*t == ';' + && get_cvn_flags(tmpbuf, len, UTF ? SVf_UTF8 : 0)) Perl_warner(aTHX_ packWARN(WARN_SYNTAX), - "You need to quote \"%s\"", - tmpbuf); + "You need to quote \"%"SVf"\"", + SVfARG(newSVpvn_flags(tmpbuf, len, + SVs_TEMP | (UTF ? SVf_UTF8 : 0)))); } } } @@ -6254,15 +6295,17 @@ Perl_yylex(pTHX) if (ckWARN(WARN_SYNTAX)) { const char *t = s + 1; while (*t && (isALNUM_lazy_if(t,UTF) || strchr(" \t$#+-'\"", *t))) - t++; + t += UTF ? UTF8SKIP(t) : 1; if (*t == '}' || *t == ']') { t++; PL_bufptr = PEEKSPACE(PL_bufptr); /* XXX can realloc */ /* diag_listed_as: Scalar value @%s[%s] better written as $%s[%s] */ Perl_warner(aTHX_ packWARN(WARN_SYNTAX), - "Scalar value %.*s better written as $%.*s", - (int)(t-PL_bufptr), PL_bufptr, - (int)(t-PL_bufptr-1), PL_bufptr+1); + "Scalar value %"SVf" better written as $%"SVf, + SVfARG(newSVpvn_flags(PL_bufptr, (STRLEN)(t-PL_bufptr), + SVs_TEMP | (UTF ? SVf_UTF8 : 0 ))), + SVfARG(newSVpvn_flags(PL_bufptr+1, (STRLEN)(t-PL_bufptr-1), + SVs_TEMP | (UTF ? SVf_UTF8 : 0 )))); } } } @@ -6378,7 +6421,7 @@ Perl_yylex(pTHX) TERM(THING); case '\'': - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); DEBUG_T( { printbuf("### Saw string before %s\n", s); } ); if (PL_expect == XOPERATOR) { if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) { @@ -6393,7 +6436,7 @@ Perl_yylex(pTHX) TERM(sublex_start()); case '"': - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); DEBUG_T( { printbuf("### Saw string before %s\n", s); } ); if (PL_expect == XOPERATOR) { if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) { @@ -6416,7 +6459,7 @@ Perl_yylex(pTHX) TERM(sublex_start()); case '`': - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); DEBUG_T( { printbuf("### Saw backtick string before %s\n", s); } ); if (PL_expect == XOPERATOR) no_op("Backticks",s); @@ -6557,7 +6600,9 @@ Perl_yylex(pTHX) if (!anydelim && PL_expect == XSTATE && d < PL_bufend && *d == ':' && *(d + 1) != ':') { s = d + 1; - pl_yylval.pval = CopLABEL_alloc(PL_tokenbuf); + pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, + newSVpvn_flags(PL_tokenbuf, + len, UTF ? SVf_UTF8 : 0)); CLINE; TOKEN(LABEL); } @@ -6643,7 +6688,9 @@ Perl_yylex(pTHX) s = scan_word(s, PL_tokenbuf + len, sizeof PL_tokenbuf - len, TRUE, &morelen); if (!morelen) - Perl_croak(aTHX_ "Bad name after %s%s", PL_tokenbuf, + Perl_croak(aTHX_ "Bad name after %"SVf"%s", + SVfARG(newSVpvn_flags(PL_tokenbuf, len, + (UTF ? SVf_UTF8 : 0) | SVs_TEMP )), *s == '\'' ? "'" : "::"); len += morelen; pkgname = 1; @@ -6669,8 +6716,9 @@ Perl_yylex(pTHX) if (ckWARN(WARN_BAREWORD) && ! gv_fetchpvn_flags(PL_tokenbuf, len, UTF ? SVf_UTF8 : 0, SVt_PVHV)) Perl_warner(aTHX_ packWARN(WARN_BAREWORD), - "Bareword \"%s\" refers to nonexistent package", - PL_tokenbuf); + "Bareword \"%"SVf"\" refers to nonexistent package", + SVfARG(newSVpvn_flags(PL_tokenbuf, len, + (UTF ? SVf_UTF8 : 0) | SVs_TEMP))); len -= 2; PL_tokenbuf[len] = '\0'; gv = NULL; @@ -6851,17 +6899,19 @@ Perl_yylex(pTHX) /* Not a method, so call it a subroutine (if defined) */ if (cv) { - if (lastchar == '-') - Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS), - "Ambiguous use of -%s resolved as -&%s()", - PL_tokenbuf, PL_tokenbuf); + if (lastchar == '-') { + const SV *tmpsv = newSVpvn_flags( PL_tokenbuf, len ? len : strlen(PL_tokenbuf), (UTF ? SVf_UTF8 : 0) | SVs_TEMP ); + Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS), + "Ambiguous use of -%"SVf" resolved as -&%"SVf"()", + SVfARG(tmpsv), SVfARG(tmpsv)); + } /* Check for a constant sub */ if ((sv = cv_const_sv(cv))) { its_constant: op_free(rv2cv_op); SvREFCNT_dec(((SVOP*)pl_yylval.opval)->op_sv); ((SVOP*)pl_yylval.opval)->op_sv = SvREFCNT_inc_simple(sv); - pl_yylval.opval->op_private = 0; + pl_yylval.opval->op_private = OPpCONST_FOLDED; pl_yylval.opval->op_flags |= OPf_SPECIAL; TOKEN(WORD); } @@ -7026,8 +7076,10 @@ Perl_yylex(pTHX) safe_bareword: if ((lastchar == '*' || lastchar == '%' || lastchar == '&')) { Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS), - "Operator or semicolon missing before %c%s", - lastchar, PL_tokenbuf); + "Operator or semicolon missing before %c%"SVf, + lastchar, SVfARG(newSVpvn_flags(PL_tokenbuf, + strlen(PL_tokenbuf), + SVs_TEMP | (UTF ? SVf_UTF8 : 0)))); Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS), "Ambiguous use of %c resolved as operator %c", lastchar, lastchar); @@ -7178,7 +7230,9 @@ Perl_yylex(pTHX) d = s; s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &len); if (!(tmp = keyword(PL_tokenbuf, len, 1))) - Perl_croak(aTHX_ "CORE::%s is not a keyword", PL_tokenbuf); + Perl_croak(aTHX_ "CORE::%"SVf" is not a keyword", + SVfARG(newSVpvn_flags(PL_tokenbuf, len, + (UTF ? SVf_UTF8 : 0) | SVs_TEMP))); if (tmp < 0) tmp = -tmp; else if (tmp == KEY_require || tmp == KEY_do @@ -7286,8 +7340,13 @@ Perl_yylex(pTHX) s = SKIPSPACE1(s); if (*s == '{') PRETERMBLOCK(DO); - if (*s != '\'') - s = force_word(s,WORD,TRUE,TRUE,FALSE); + if (*s != '\'') { + d = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, 1, &len); + if (len) { + d = SKIPSPACE1(d); + if (*d == '(') s = force_word(s,WORD,TRUE,TRUE,FALSE); + } + } if (orig_keyword == KEY_do) { orig_keyword = 0; pl_yylval.ival = 1; @@ -7657,7 +7716,7 @@ Perl_yylex(pTHX) char tmpbuf[1024]; PL_bufptr = s; my_snprintf(tmpbuf, sizeof(tmpbuf), "No such class %.1000s", PL_tokenbuf); - yyerror(tmpbuf); + yyerror_pv(tmpbuf, UTF ? SVf_UTF8 : 0); } #ifdef PERL_MAD if (PL_madskills) { /* just add type to declarator token */ @@ -7697,8 +7756,14 @@ Perl_yylex(pTHX) s = SKIPSPACE1(s); if (isIDFIRST_lazy_if(s,UTF)) { const char *t; - for (d = s; isALNUM_lazy_if(d,UTF);) - d++; + for (d = s; isALNUM_lazy_if(d,UTF);) { + d += UTF ? UTF8SKIP(d) : 1; + if (UTF) { + while (UTF8_IS_CONTINUED(*d) && is_utf8_mark((U8*)d)) { + d += UTF ? UTF8SKIP(d) : 1; + } + } + } for (t=d; isSPACE(*t);) t++; if ( *t && strchr("|&*+-=!?:.", *t) && ckWARN_d(WARN_PRECEDENCE) @@ -7707,10 +7772,11 @@ Perl_yylex(pTHX) && !(t[0] == ':' && t[1] == ':') && !keyword(s, d-s, 0) ) { - int parms_len = (int)(d-s); + SV *tmpsv = newSVpvn_flags(s, (STRLEN)(d-s), + SVs_TEMP | (UTF ? SVf_UTF8 : 0)); Perl_warner(aTHX_ packWARN(WARN_PRECEDENCE), - "Precedence problem: open %.*s should be open(%.*s)", - parms_len, s, parms_len, s); + "Precedence problem: open %"SVf" should be open(%"SVf")", + SVfARG(tmpsv), SVfARG(tmpsv)); } } LOP(OP_OPEN,XTERM); @@ -7764,7 +7830,7 @@ Perl_yylex(pTHX) LOP(OP_PIPE_OP,XTERM); case KEY_q: - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) missingterm(NULL); pl_yylval.ival = OP_CONST; @@ -7775,7 +7841,7 @@ Perl_yylex(pTHX) case KEY_qw: { OP *words = NULL; - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) missingterm(NULL); PL_expect = XOPERATOR; @@ -7825,7 +7891,7 @@ Perl_yylex(pTHX) } case KEY_qq: - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) missingterm(NULL); pl_yylval.ival = OP_STRINGIFY; @@ -7838,7 +7904,7 @@ Perl_yylex(pTHX) TERM(sublex_start()); case KEY_qx: - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) missingterm(NULL); readpipe_override(); @@ -8147,7 +8213,7 @@ Perl_yylex(pTHX) const bool warnillegalproto = ckWARN(WARN_ILLEGALPROTO); STRLEN tmplen; - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) Perl_croak(aTHX_ "Prototype not terminated"); /* strip spaces and check for bad characters */ @@ -8199,9 +8265,13 @@ Perl_yylex(pTHX) "Illegal character %sin prototype for %"SVf" : %s", seen_underscore ? "after '_' " : "", SVfARG(PL_subname), - sv_uni_display(dsv, - newSVpvn_flags(d, tmp, SVs_TEMP | SvUTF8(PL_lex_stuff)), - tmp, UNI_DISPLAY_ISPRINT)); + SvUTF8(PL_lex_stuff) + ? sv_uni_display(dsv, + newSVpvn_flags(d, tmp, SVs_TEMP | SVf_UTF8), + tmp, + UNI_DISPLAY_ISPRINT) + : pv_pretty(dsv, d, tmp, 60, NULL, NULL, + PERL_PV_ESCAPE_NONASCII)); } SvCUR_set(PL_lex_stuff, tmp); have_proto = TRUE; @@ -8432,7 +8502,6 @@ static int S_pending_ident(pTHX) { dVAR; - register char *d; PADOFFSET tmp = 0; /* pit holds the identifier we read and pending_ident is reset */ char pit = PL_pending_ident; @@ -8454,15 +8523,16 @@ S_pending_ident(pTHX) if (PL_in_my) { if (PL_in_my == KEY_our) { /* "our" is merely analogous to "my" */ if (has_colon) - yyerror(Perl_form(aTHX_ "No package name allowed for " + yyerror_pv(Perl_form(aTHX_ "No package name allowed for " "variable %s in \"our\"", - PL_tokenbuf)); + PL_tokenbuf), UTF ? SVf_UTF8 : 0); tmp = allocmy(PL_tokenbuf, tokenbuf_len, UTF ? SVf_UTF8 : 0); } else { if (has_colon) - yyerror(Perl_form(aTHX_ PL_no_myglob, - PL_in_my == KEY_my ? "my" : "state", PL_tokenbuf)); + yyerror_pv(Perl_form(aTHX_ PL_no_myglob, + PL_in_my == KEY_my ? "my" : "state", PL_tokenbuf), + UTF ? SVf_UTF8 : 0); pl_yylval.opval = newOP(OP_PADANY, 0); pl_yylval.opval->op_targ = allocmy(PL_tokenbuf, tokenbuf_len, @@ -8473,14 +8543,6 @@ S_pending_ident(pTHX) /* build the ops for accesses to a my() variable. - - Deny my($a) or my($b) in a sort block, *if* $a or $b is - then used in a comparison. This catches most, but not - all cases. For instance, it catches - sort { my($a); $a <=> $b } - but not - sort { my($a); $a < $b ? -1 : $a == $b ? 0 : 1; } - (although why you'd do that is anyone's guess). */ if (!has_colon) { @@ -8509,23 +8571,6 @@ S_pending_ident(pTHX) return WORD; } - /* if it's a sort block and they're naming $a or $b */ - if (PL_last_lop_op == OP_SORT && - PL_tokenbuf[0] == '$' && - (PL_tokenbuf[1] == 'a' || PL_tokenbuf[1] == 'b') - && !PL_tokenbuf[2]) - { - for (d = PL_in_eval ? PL_oldoldbufptr : PL_linestart; - d < PL_bufend && *d != '\n'; - d++) - { - if (strnEQ(d,"<=>",3) || strnEQ(d,"cmp",3)) { - Perl_croak(aTHX_ "Can't use \"my %s\" in sort comparison", - PL_tokenbuf); - } - } - } - pl_yylval.opval = newOP(OP_PADANY, 0); pl_yylval.opval->op_targ = tmp; return PRIVATEREF; @@ -8549,8 +8594,9 @@ S_pending_ident(pTHX) { /* Downgraded from fatal to warning 20000522 mjd */ Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS), - "Possible unintended interpolation of %s in string", - PL_tokenbuf); + "Possible unintended interpolation of %"SVf" in string", + SVfARG(newSVpvn_flags(PL_tokenbuf, tokenbuf_len, + SVs_TEMP | ( UTF ? SVf_UTF8 : 0 )))); } } @@ -8603,9 +8649,10 @@ S_checkcomma(pTHX_ const char *s, const char *name, const char *what) while (s < PL_bufend && isSPACE(*s)) s++; if (isIDFIRST_lazy_if(s,UTF)) { - const char * const w = s++; + const char * const w = s; + s += UTF ? UTF8SKIP(s) : 1; while (isALNUM_lazy_if(s,UTF)) - s++; + s += UTF ? UTF8SKIP(s) : 1; while (s < PL_bufend && isSPACE(*s)) s++; if (*s == ',') { @@ -8758,7 +8805,7 @@ S_scan_word(pTHX_ register char *s, char *dest, STRLEN destlen, int allow_packag for (;;) { if (d >= e) Perl_croak(aTHX_ ident_too_long); - if (isALNUM(*s)) /* UTF handled below */ + if (isALNUM(*s) || (!UTF && isALNUMC_L1(*s))) /* UTF handled below */ *d++ = *s++; else if (allow_package && (*s == '\'') && isIDFIRST_lazy_if(s+1,UTF)) { *d++ = ':'; @@ -8854,8 +8901,6 @@ S_scan_ident(pTHX_ register char *s, register const char *send, char *dest, STRL bracket = s; s++; } - else if (ck_uni) - check_uni(); if (s < send) { if (UTF) { const STRLEN skip = UTF8SKIP(s); @@ -8873,6 +8918,8 @@ S_scan_ident(pTHX_ register char *s, register const char *send, char *dest, STRL *d = toCTRL(*s); s++; } + else if (ck_uni && !bracket) + check_uni(); if (bracket) { if (isSPACE(s[-1])) { while (s < send) { @@ -8943,13 +8990,15 @@ S_scan_ident(pTHX_ register char *s, register const char *send, char *dest, STRL if (PL_lex_state == LEX_NORMAL) { if (ckWARN(WARN_AMBIGUOUS) && (keyword(dest, d - dest, 0) - || get_cvn_flags(dest, d - dest, 0))) + || get_cvn_flags(dest, d - dest, UTF ? SVf_UTF8 : 0))) { + SV *tmp = newSVpvn_flags( dest, d - dest, + SVs_TEMP | (UTF ? SVf_UTF8 : 0) ); if (funny == '#') funny = '@'; Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS), - "Ambiguous use of %c{%s} resolved to %c%s", - funny, dest, funny, dest); + "Ambiguous use of %c{%"SVf"} resolved to %c%"SVf, + funny, tmp, funny, tmp); } } } @@ -8969,18 +9018,24 @@ S_pmflag(pTHX_ const char* const valid_flags, U32 * pmfl, char** s, char* charse /* Adds, subtracts to/from 'pmfl' based on regex modifier flags found in * the parse starting at 's', based on the subset that are valid in this * context input to this routine in 'valid_flags'. Advances s. Returns - * TRUE if the input was a valid flag, so the next char may be as well; - * otherwise FALSE. 'charset' should point to a NUL upon first call on the - * current regex. This routine will set it to any charset modifier found. - * The caller shouldn't change it. This way, another charset modifier - * encountered in the parse can be detected as an error, as we have decided - * allow only one */ + * TRUE if the input should be treated as a valid flag, so the next char + * may be as well; otherwise FALSE. 'charset' should point to a NUL upon + * first call on the current regex. This routine will set it to any + * charset modifier found. The caller shouldn't change it. This way, + * another charset modifier encountered in the parse can be detected as an + * error, as we have decided to allow only one */ const char c = **s; - - if (! strchr(valid_flags, c)) { - if (isALNUM(c)) { - goto deprecate; + STRLEN charlen = UTF ? UTF8SKIP(*s) : 1; + + if ( charlen != 1 || ! strchr(valid_flags, c) ) { + if (isALNUM_lazy_if(*s, UTF)) { + yyerror_pv(Perl_form(aTHX_ "Unknown regexp modifier \"/%.*s\"", (int)charlen, *s), + UTF ? SVf_UTF8 : 0); + (*s) += charlen; + /* Pretend that it worked, so will continue processing before + * dieing */ + return TRUE; } return FALSE; } @@ -8994,34 +9049,6 @@ S_pmflag(pTHX_ const char* const valid_flags, U32 * pmfl, char** s, char* charse case KEEPCOPY_PAT_MOD: *pmfl |= RXf_PMf_KEEPCOPY; break; case NONDESTRUCT_PAT_MOD: *pmfl |= PMf_NONDESTRUCT; break; case LOCALE_PAT_MOD: - - /* In 5.14, qr//lt is legal but deprecated; the 't' means they - * can't be regex modifiers. - * In 5.14, s///le is legal and ambiguous. Try to disambiguate as - * much as easily done. s///lei, for example, has to mean regex - * modifiers if it's not an error (as does any word character - * following the 'e'). Otherwise, we resolve to the backwards- - * compatible, but less likely 's/// le ...', i.e. as meaning - * less-than-or-equal. The reason it's not likely is that s// - * returns a number for code in the field (/r returns a string, but - * that wasn't added until the 5.13 series), and so '<=' should be - * used for comparing, not 'le'. */ - if (*((*s) + 1) == 't') { - goto deprecate; - } - else if (*((*s) + 1) == 'e' && ! isALNUM(*((*s) + 2))) { - - /* 'e' is valid only for substitutes, s///e. If it is not - * valid in the current context, then 'm//le' must mean the - * comparison operator, so use the regular deprecation message. - */ - if (! strchr(valid_flags, 'e')) { - goto deprecate; - } - Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS), - "Ambiguous use of 's//le...' resolved as 's// le...'; Rewrite as 's//el' if you meant 'use locale rules and evaluate rhs as an expression'. In Perl 5.18, it will be resolved the other way"); - return FALSE; - } if (*charset) { goto multiple_charsets; } @@ -9029,11 +9056,6 @@ S_pmflag(pTHX_ const char* const valid_flags, U32 * pmfl, char** s, char* charse *charset = c; break; case UNICODE_PAT_MOD: - /* In 5.14, qr//unless and qr//until are legal but deprecated; the - * 'n' means they can't be regex modifiers */ - if (*((*s) + 1) == 'n') { - goto deprecate; - } if (*charset) { goto multiple_charsets; } @@ -9041,12 +9063,6 @@ S_pmflag(pTHX_ const char* const valid_flags, U32 * pmfl, char** s, char* charse *charset = c; break; case ASCII_RESTRICT_PAT_MOD: - /* In 5.14, qr//and is legal but deprecated; the 'n' means they - * can't be regex modifiers */ - if (*((*s) + 1) == 'n') { - goto deprecate; - } - if (! *charset) { set_regex_charset(pmfl, REGEX_ASCII_RESTRICTED_CHARSET); } @@ -9076,11 +9092,6 @@ S_pmflag(pTHX_ const char* const valid_flags, U32 * pmfl, char** s, char* charse (*s)++; return TRUE; - deprecate: - Perl_ck_warner_d(aTHX_ packWARN(WARN_SYNTAX), - "Having no space between pattern and following word is deprecated"); - return FALSE; - multiple_charsets: if (*charset != c) { yyerror(Perl_form(aTHX_ "Regexp modifiers \"/%c\" and \"/%c\" are mutually exclusive", *charset, c)); @@ -9102,7 +9113,7 @@ S_scan_pat(pTHX_ char *start, I32 type) { dVAR; PMOP *pm; - char *s = scan_str(start,!!PL_madskills,FALSE); + char *s = scan_str(start,!!PL_madskills,FALSE, PL_reg_state.re_reparsing); const char * const valid_flags = (const char *)((type == OP_QR) ? QR_PAT_MODS : M_PAT_MODS); char charset = '\0'; /* character set modifier */ @@ -9112,6 +9123,9 @@ S_scan_pat(pTHX_ char *start, I32 type) PERL_ARGS_ASSERT_SCAN_PAT; + /* this was only needed for the initial scan_str; set it to false + * so that any (?{}) code blocks etc are parsed normally */ + PL_reg_state.re_reparsing = FALSE; if (!s) { const char * const delimiter = skipspace(start); Perl_croak(aTHX_ @@ -9147,6 +9161,25 @@ S_scan_pat(pTHX_ char *start, I32 type) #ifdef PERL_MAD modstart = s; #endif + + /* if qr/...(?{..}).../, then need to parse the pattern within a new + * anon CV. False positives like qr/[(?{]/ are harmless */ + + if (type == OP_QR) { + STRLEN len; + char *e, *p = SvPV(PL_lex_stuff, len); + e = p + len; + for (; p < e; p++) { + if (p[0] == '(' && p[1] == '?' + && (p[2] == '{' || (p[2] == '?' && p[3] == '{'))) + { + pm->op_pmflags |= PMf_HAS_CV; + break; + } + } + pm->op_pmflags |= PMf_IS_QR; + } + while (*s && S_pmflag(aTHX_ valid_flags, &(pm->op_pmflags), &s, &charset)) {}; #ifdef PERL_MAD if (PL_madskills && modstart != s) { @@ -9183,7 +9216,7 @@ S_scan_subst(pTHX_ char *start) pl_yylval.ival = OP_NULL; - s = scan_str(start,!!PL_madskills,FALSE); + s = scan_str(start,!!PL_madskills,FALSE,FALSE); if (!s) Perl_croak(aTHX_ "Substitution pattern not terminated"); @@ -9201,7 +9234,7 @@ S_scan_subst(pTHX_ char *start) #endif first_start = PL_multi_start; - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) { if (PL_lex_stuff) { SvREFCNT_dec(PL_lex_stuff); @@ -9291,7 +9324,7 @@ S_scan_trans(pTHX_ char *start) pl_yylval.ival = OP_NULL; - s = scan_str(start,!!PL_madskills,FALSE); + s = scan_str(start,!!PL_madskills,FALSE,FALSE); if (!s) Perl_croak(aTHX_ "Transliteration pattern not terminated"); @@ -9307,7 +9340,7 @@ S_scan_trans(pTHX_ char *start) } #endif - s = scan_str(s,!!PL_madskills,FALSE); + s = scan_str(s,!!PL_madskills,FALSE,FALSE); if (!s) { if (PL_lex_stuff) { SvREFCNT_dec(PL_lex_stuff); @@ -9699,7 +9732,7 @@ S_scan_inputsymbol(pTHX_ char *start) if (d - PL_tokenbuf != len) { pl_yylval.ival = OP_GLOB; - s = scan_str(start,!!PL_madskills,FALSE); + s = scan_str(start,!!PL_madskills,FALSE,FALSE); if (!s) Perl_croak(aTHX_ "Glob not terminated"); return s; @@ -9799,6 +9832,8 @@ intro_sym: takes: start position in buffer keep_quoted preserve \ on the embedded delimiter(s) keep_delims preserve the delimiters around the string + re_reparse compiling a run-time /(?{})/: + collapse // to /, and skip encoding src returns: position to continue reading from buffer side-effects: multi_start, multi_close, lex_repl or lex_stuff, and updates the read buffer. @@ -9839,7 +9874,7 @@ intro_sym: */ STATIC char * -S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims) +S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims, int re_reparse) { dVAR; SV *sv; /* scalar value: string */ @@ -9918,7 +9953,7 @@ S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims) } #endif for (;;) { - if (PL_encoding && !UTF) { + if (PL_encoding && !UTF && !re_reparse) { bool cont = TRUE; while (cont) { @@ -10000,9 +10035,12 @@ S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims) CopLINE_inc(PL_curcop); /* handle quoted delimiters */ if (*s == '\\' && s+1 < PL_bufend && term != '\\') { - if (!keep_quoted && s[1] == term) + if (!keep_quoted + && (s[1] == term + || (re_reparse && s[1] == '\\')) + ) s++; - /* any other quotes are simply copied straight through */ + /* any other quotes are simply copied straight through */ else *to++ = *s++; } @@ -10103,7 +10141,7 @@ S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims) /* at this point, we have successfully read the delimited string */ - if (!PL_encoding || UTF) { + if (!PL_encoding || UTF || re_reparse) { #ifdef PERL_MAD if (PL_madskills) { char * const tstart = SvPVX(PL_linestr) + stuffstart; @@ -10135,7 +10173,7 @@ S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims) } } #endif - if (has_utf8 || PL_encoding) + if (has_utf8 || (PL_encoding && !re_reparse)) SvUTF8_on(sv); PL_multi_end = CopLINE(PL_curcop); @@ -10740,18 +10778,6 @@ Perl_yyerror_pv(pTHX_ const char *const s, U32 flags) } int -Perl_yyerror_sv(pTHX_ SV * sv, U32 flags) -{ - char *s; - STRLEN len; - PERL_ARGS_ASSERT_YYERROR_SV; - s = SvPV(sv, len); - if (SvUTF8(sv)) - flags |= SVf_UTF8; - return yyerror_pvn(s, len, flags); -} - -int Perl_yyerror_pvn(pTHX_ const char *const s, STRLEN len, U32 flags) { dVAR; @@ -11480,15 +11506,10 @@ Perl_parse_label(pTHX_ U32 flags) if (PL_lex_state == LEX_KNOWNEXT) { PL_parser->yychar = yylex(); if (PL_parser->yychar == LABEL) { - char *lpv = pl_yylval.pval; - STRLEN llen = strlen(lpv); SV *lsv; PL_parser->yychar = YYEMPTY; lsv = newSV_type(SVt_PV); - SvPV_set(lsv, lpv); - SvCUR_set(lsv, llen); - SvLEN_set(lsv, llen+1); - SvPOK_on(lsv); + sv_copypv(lsv, cSVOPx(pl_yylval.opval)->op_sv); return lsv; } else { yyunlex(); @@ -11496,17 +11517,12 @@ Perl_parse_label(pTHX_ U32 flags) } } else { char *s, *t; - U8 c; STRLEN wlen, bufptr_pos; lex_read_space(0); t = s = PL_bufptr; - c = (U8)*s; - if (!isIDFIRST_A(c)) + if (!isIDFIRST_lazy_if(s, UTF)) goto no_label; - do { - c = (U8)*++t; - } while(isWORDCHAR_A(c)); - wlen = t - s; + t = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &wlen); if (word_takes_any_delimeter(s, wlen)) goto no_label; bufptr_pos = s - SvPVX(PL_linestr); @@ -11518,7 +11534,7 @@ Perl_parse_label(pTHX_ U32 flags) PL_oldoldbufptr = PL_oldbufptr; PL_oldbufptr = s; PL_bufptr = t+1; - return newSVpvn(s, wlen); + return newSVpvn_flags(s, wlen, UTF ? SVf_UTF8 : 0); } else { PL_bufptr = s; no_label: @@ -11611,29 +11627,12 @@ Perl_parse_stmtseq(pTHX_ U32 flags) return stmtseqop; } -void -Perl_munge_qwlist_to_paren_list(pTHX_ OP *qwlist) -{ - PERL_ARGS_ASSERT_MUNGE_QWLIST_TO_PAREN_LIST; - deprecate("qw(...) as parentheses"); - force_next((4<<24)|')'); - if (qwlist->op_type == OP_STUB) { - op_free(qwlist); - } - else { - start_force(PL_curforce); - NEXTVAL_NEXTTOKE.opval = qwlist; - force_next(THING); - } - force_next((2<<24)|'('); -} - /* * Local variables: * c-indentation-style: bsd * c-basic-offset: 4 - * indent-tabs-mode: t + * indent-tabs-mode: nil * End: * - * ex: set ts=8 sts=4 sw=4 noet: + * ex: set ts=8 sts=4 sw=4 et: */