/* toke.c
*
- * Copyright (c) 1991-2002, Larry Wall
+ * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
+ * 2000, 2001, 2002, 2003, 2004, 2005, by Larry Wall and others
*
* You may distribute under the terms of either the GNU General Public
* License or the Artistic License, as specified in the README file.
#define PERL_IN_TOKE_C
#include "perl.h"
-#define yychar PL_yychar
-#define yylval PL_yylval
+#define yychar (*PL_yycharp)
+#define yylval (*PL_yylvalp)
static char ident_too_long[] = "Identifier too long";
static char c_without_g[] = "Use of /c modifier is meaningless without /g";
#ifdef USE_UTF8_SCRIPTS
# define UTF (!IN_BYTES)
#else
-# ifdef EBCDIC /* For now 'use utf8' does not affect tokenizer on EBCDIC */
-# define UTF (PL_linestr && DO_UTF8(PL_linestr))
-# else
-# define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
-# endif
+# define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
#endif
/* In variables named $^X, these are the legal values for X.
#define LEX_FORMLINE 1
#define LEX_KNOWNEXT 0
-#ifdef ff_next
-#undef ff_next
+#ifdef DEBUGGING
+static char* lex_state_names[] = {
+ "KNOWNEXT",
+ "FORMLINE",
+ "INTERPCONST",
+ "INTERPCONCAT",
+ "INTERPENDMAYBE",
+ "INTERPEND",
+ "INTERPSTART",
+ "INTERPPUSH",
+ "INTERPCASEMOD",
+ "INTERPNORMAL",
+ "NORMAL"
+};
#endif
-#ifdef USE_PURE_BISON
-# ifndef YYMAXLEVEL
-# define YYMAXLEVEL 100
-# endif
-YYSTYPE* yylval_pointer[YYMAXLEVEL];
-int* yychar_pointer[YYMAXLEVEL];
-int yyactlevel = -1;
-# undef yylval
-# undef yychar
-# define yylval (*yylval_pointer[yyactlevel])
-# define yychar (*yychar_pointer[yyactlevel])
-# define PERL_YYLEX_PARAM yylval_pointer[yyactlevel],yychar_pointer[yyactlevel]
-# undef yylex
-# define yylex() Perl_yylex_r(aTHX_ yylval_pointer[yyactlevel],yychar_pointer[yyactlevel])
+#ifdef ff_next
+#undef ff_next
#endif
#include "keywords.h"
* Also see LOP and lop() below.
*/
-/* Note that REPORT() and REPORT2() will be expressions that supply
- * their own trailing comma, not suitable for statements as such. */
#ifdef DEBUGGING /* Serve -DT. */
-# define REPORT(x,retval) tokereport(x,s,(int)retval),
-# define REPORT2(x,retval) tokereport(x,s, yylval.ival),
+# define REPORT(retval) tokereport(s,(int)retval)
#else
-# define REPORT(x,retval)
-# define REPORT2(x,retval)
+# define REPORT(retval) (retval)
#endif
-#define TOKEN(retval) return (REPORT2("token",retval) PL_bufptr = s,(int)retval)
-#define OPERATOR(retval) return (REPORT2("operator",retval) PL_expect = XTERM, PL_bufptr = s,(int)retval)
-#define AOPERATOR(retval) return ao((REPORT2("aop",retval) PL_expect = XTERM, PL_bufptr = s,(int)retval))
-#define PREBLOCK(retval) return (REPORT2("preblock",retval) PL_expect = XBLOCK,PL_bufptr = s,(int)retval)
-#define PRETERMBLOCK(retval) return (REPORT2("pretermblock",retval) PL_expect = XTERMBLOCK,PL_bufptr = s,(int)retval)
-#define PREREF(retval) return (REPORT2("preref",retval) PL_expect = XREF,PL_bufptr = s,(int)retval)
-#define TERM(retval) return (CLINE, REPORT2("term",retval) PL_expect = XOPERATOR, PL_bufptr = s,(int)retval)
-#define LOOPX(f) return(yylval.ival=f, REPORT("loopx",f) PL_expect = XTERM,PL_bufptr = s,(int)LOOPEX)
-#define FTST(f) return(yylval.ival=f, REPORT("ftst",f) PL_expect = XTERM,PL_bufptr = s,(int)UNIOP)
-#define FUN0(f) return(yylval.ival = f, REPORT("fun0",f) PL_expect = XOPERATOR,PL_bufptr = s,(int)FUNC0)
-#define FUN1(f) return(yylval.ival = f, REPORT("fun1",f) PL_expect = XOPERATOR,PL_bufptr = s,(int)FUNC1)
-#define BOop(f) return ao((yylval.ival=f, REPORT("bitorop",f) PL_expect = XTERM,PL_bufptr = s,(int)BITOROP))
-#define BAop(f) return ao((yylval.ival=f, REPORT("bitandop",f) PL_expect = XTERM,PL_bufptr = s,(int)BITANDOP))
-#define SHop(f) return ao((yylval.ival=f, REPORT("shiftop",f) PL_expect = XTERM,PL_bufptr = s,(int)SHIFTOP))
-#define PWop(f) return ao((yylval.ival=f, REPORT("powop",f) PL_expect = XTERM,PL_bufptr = s,(int)POWOP))
-#define PMop(f) return(yylval.ival=f, REPORT("matchop",f) PL_expect = XTERM,PL_bufptr = s,(int)MATCHOP)
-#define Aop(f) return ao((yylval.ival=f, REPORT("add",f) PL_expect = XTERM,PL_bufptr = s,(int)ADDOP))
-#define Mop(f) return ao((yylval.ival=f, REPORT("mul",f) PL_expect = XTERM,PL_bufptr = s,(int)MULOP))
-#define Eop(f) return(yylval.ival=f, REPORT("eq",f) PL_expect = XTERM,PL_bufptr = s,(int)EQOP)
-#define Rop(f) return(yylval.ival=f, REPORT("rel",f) PL_expect = XTERM,PL_bufptr = s,(int)RELOP)
+#define TOKEN(retval) return ( PL_bufptr = s, REPORT(retval))
+#define OPERATOR(retval) return (PL_expect = XTERM, PL_bufptr = s, REPORT(retval))
+#define AOPERATOR(retval) return ao((PL_expect = XTERM, PL_bufptr = s, REPORT(retval)))
+#define PREBLOCK(retval) return (PL_expect = XBLOCK,PL_bufptr = s, REPORT(retval))
+#define PRETERMBLOCK(retval) return (PL_expect = XTERMBLOCK,PL_bufptr = s, REPORT(retval))
+#define PREREF(retval) return (PL_expect = XREF,PL_bufptr = s, REPORT(retval))
+#define TERM(retval) return (CLINE, PL_expect = XOPERATOR, PL_bufptr = s, REPORT(retval))
+#define LOOPX(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)LOOPEX))
+#define FTST(f) return (yylval.ival=f, PL_expect=XTERMORDORDOR, PL_bufptr=s, REPORT((int)UNIOP))
+#define FUN0(f) return (yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC0))
+#define FUN1(f) return (yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC1))
+#define BOop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITOROP)))
+#define BAop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITANDOP)))
+#define SHop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)SHIFTOP)))
+#define PWop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)POWOP)))
+#define PMop(f) return(yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MATCHOP))
+#define Aop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)ADDOP)))
+#define Mop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MULOP)))
+#define Eop(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)EQOP))
+#define Rop(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)RELOP))
/* This bit of chicanery makes a unary function followed by
* a parenthesis into a function with one argument, highest precedence.
+ * The UNIDOR macro is for unary functions that can be followed by the //
+ * operator (such as C<shift // 0>).
*/
-#define UNI(f) return(yylval.ival = f, \
- REPORT("uni",f) \
- PL_expect = XTERM, \
+#define UNI2(f,x) return ( \
+ yylval.ival = f, \
+ PL_expect = x, \
PL_bufptr = s, \
PL_last_uni = PL_oldbufptr, \
PL_last_lop_op = f, \
- (*s == '(' || (s = skipspace(s), *s == '(') ? (int)FUNC1 : (int)UNIOP) )
-
-#define UNIBRACK(f) return(yylval.ival = f, \
- REPORT("uni",f) \
+ REPORT( \
+ (*s == '(' || (s = skipspace(s), *s == '(') \
+ ? (int)FUNC1 : (int)UNIOP)))
+#define UNI(f) UNI2(f,XTERM)
+#define UNIDOR(f) UNI2(f,XTERMORDORDOR)
+
+#define UNIBRACK(f) return ( \
+ yylval.ival = f, \
PL_bufptr = s, \
PL_last_uni = PL_oldbufptr, \
- (*s == '(' || (s = skipspace(s), *s == '(') ? (int)FUNC1 : (int)UNIOP) )
+ REPORT( \
+ (*s == '(' || (s = skipspace(s), *s == '(') \
+ ? (int)FUNC1 : (int)UNIOP)))
/* grandfather return to old style */
#define OLDLOP(f) return(yylval.ival=f,PL_expect = XTERM,PL_bufptr = s,(int)LSTOP)
#ifdef DEBUGGING
-STATIC void
-S_tokereport(pTHX_ char *thing, char* s, I32 rv)
+/* how to interpret the yylval associated with the token */
+enum token_type {
+ TOKENTYPE_NONE,
+ TOKENTYPE_IVAL,
+ TOKENTYPE_OPNUM, /* yylval.ival contains an opcode number */
+ TOKENTYPE_PVAL,
+ TOKENTYPE_OPVAL,
+ TOKENTYPE_GVVAL
+};
+
+static struct debug_tokens { int token, type; char *name;} debug_tokens[] =
{
- DEBUG_T({
- SV* report = newSVpv(thing, 0);
- Perl_sv_catpvf(aTHX_ report, ":line %d:%"IVdf":", CopLINE(PL_curcop),
- (IV)rv);
+ { ADDOP, TOKENTYPE_OPNUM, "ADDOP" },
+ { ANDAND, TOKENTYPE_NONE, "ANDAND" },
+ { ANDOP, TOKENTYPE_NONE, "ANDOP" },
+ { ANONSUB, TOKENTYPE_IVAL, "ANONSUB" },
+ { ARROW, TOKENTYPE_NONE, "ARROW" },
+ { ASSIGNOP, TOKENTYPE_OPNUM, "ASSIGNOP" },
+ { BITANDOP, TOKENTYPE_OPNUM, "BITANDOP" },
+ { BITOROP, TOKENTYPE_OPNUM, "BITOROP" },
+ { COLONATTR, TOKENTYPE_NONE, "COLONATTR" },
+ { CONTINUE, TOKENTYPE_NONE, "CONTINUE" },
+ { DO, TOKENTYPE_NONE, "DO" },
+ { DOLSHARP, TOKENTYPE_NONE, "DOLSHARP" },
+ { DORDOR, TOKENTYPE_NONE, "DORDOR" },
+ { DOROP, TOKENTYPE_OPNUM, "DOROP" },
+ { DOTDOT, TOKENTYPE_IVAL, "DOTDOT" },
+ { ELSE, TOKENTYPE_NONE, "ELSE" },
+ { ELSIF, TOKENTYPE_IVAL, "ELSIF" },
+ { EQOP, TOKENTYPE_OPNUM, "EQOP" },
+ { FOR, TOKENTYPE_IVAL, "FOR" },
+ { FORMAT, TOKENTYPE_NONE, "FORMAT" },
+ { FUNC, TOKENTYPE_OPNUM, "FUNC" },
+ { FUNC0, TOKENTYPE_OPNUM, "FUNC0" },
+ { FUNC0SUB, TOKENTYPE_OPVAL, "FUNC0SUB" },
+ { FUNC1, TOKENTYPE_OPNUM, "FUNC1" },
+ { FUNCMETH, TOKENTYPE_OPVAL, "FUNCMETH" },
+ { HASHBRACK, TOKENTYPE_NONE, "HASHBRACK" },
+ { IF, TOKENTYPE_IVAL, "IF" },
+ { LABEL, TOKENTYPE_PVAL, "LABEL" },
+ { LOCAL, TOKENTYPE_IVAL, "LOCAL" },
+ { LOOPEX, TOKENTYPE_OPNUM, "LOOPEX" },
+ { LSTOP, TOKENTYPE_OPNUM, "LSTOP" },
+ { LSTOPSUB, TOKENTYPE_OPVAL, "LSTOPSUB" },
+ { MATCHOP, TOKENTYPE_OPNUM, "MATCHOP" },
+ { METHOD, TOKENTYPE_OPVAL, "METHOD" },
+ { MULOP, TOKENTYPE_OPNUM, "MULOP" },
+ { MY, TOKENTYPE_IVAL, "MY" },
+ { MYSUB, TOKENTYPE_NONE, "MYSUB" },
+ { NOAMP, TOKENTYPE_NONE, "NOAMP" },
+ { NOTOP, TOKENTYPE_NONE, "NOTOP" },
+ { OROP, TOKENTYPE_IVAL, "OROP" },
+ { OROR, TOKENTYPE_NONE, "OROR" },
+ { PACKAGE, TOKENTYPE_NONE, "PACKAGE" },
+ { PMFUNC, TOKENTYPE_OPVAL, "PMFUNC" },
+ { POSTDEC, TOKENTYPE_NONE, "POSTDEC" },
+ { POSTINC, TOKENTYPE_NONE, "POSTINC" },
+ { POWOP, TOKENTYPE_OPNUM, "POWOP" },
+ { PREDEC, TOKENTYPE_NONE, "PREDEC" },
+ { PREINC, TOKENTYPE_NONE, "PREINC" },
+ { PRIVATEREF, TOKENTYPE_OPVAL, "PRIVATEREF" },
+ { REFGEN, TOKENTYPE_NONE, "REFGEN" },
+ { RELOP, TOKENTYPE_OPNUM, "RELOP" },
+ { SHIFTOP, TOKENTYPE_OPNUM, "SHIFTOP" },
+ { SUB, TOKENTYPE_NONE, "SUB" },
+ { THING, TOKENTYPE_OPVAL, "THING" },
+ { UMINUS, TOKENTYPE_NONE, "UMINUS" },
+ { UNIOP, TOKENTYPE_OPNUM, "UNIOP" },
+ { UNIOPSUB, TOKENTYPE_OPVAL, "UNIOPSUB" },
+ { UNLESS, TOKENTYPE_IVAL, "UNLESS" },
+ { UNTIL, TOKENTYPE_IVAL, "UNTIL" },
+ { USE, TOKENTYPE_IVAL, "USE" },
+ { WHILE, TOKENTYPE_IVAL, "WHILE" },
+ { WORD, TOKENTYPE_OPVAL, "WORD" },
+ { 0, TOKENTYPE_NONE, 0 }
+};
+
+/* dump the returned token in rv, plus any optional arg in yylval */
+STATIC int
+S_tokereport(pTHX_ char* s, I32 rv)
+{
+ if (DEBUG_T_TEST) {
+ char *name = Nullch;
+ enum token_type type = TOKENTYPE_NONE;
+ struct debug_tokens *p;
+ SV* report = NEWSV(0, 60);
+
+ Perl_sv_catpvf(aTHX_ report, "<== ");
+
+ for (p = debug_tokens; p->token; p++) {
+ if (p->token == (int)rv) {
+ name = p->name;
+ type = p->type;
+ break;
+ }
+ }
+ if (name)
+ Perl_sv_catpvf(aTHX_ report, "%s", name);
+ else if ((char)rv > ' ' && (char)rv < '~')
+ Perl_sv_catpvf(aTHX_ report, "'%c'", (char)rv);
+ else if (!rv)
+ Perl_sv_catpvf(aTHX_ report, "EOF");
+ else
+ Perl_sv_catpvf(aTHX_ report, "?? %"IVdf, (IV)rv);
+ switch (type) {
+ case TOKENTYPE_NONE:
+ case TOKENTYPE_GVVAL: /* doesn't appear to be used */
+ break;
+ case TOKENTYPE_IVAL:
+ Perl_sv_catpvf(aTHX_ report, "(ival=%"IVdf")", yylval.ival);
+ break;
+ case TOKENTYPE_OPNUM:
+ Perl_sv_catpvf(aTHX_ report, "(ival=op_%s)",
+ PL_op_name[yylval.ival]);
+ break;
+ case TOKENTYPE_PVAL:
+ Perl_sv_catpvf(aTHX_ report, "(pval=\"%s\")", yylval.pval);
+ break;
+ case TOKENTYPE_OPVAL:
+ if (yylval.opval)
+ Perl_sv_catpvf(aTHX_ report, "(opval=op_%s)",
+ PL_op_name[yylval.opval->op_type]);
+ else
+ Perl_sv_catpv(aTHX_ report, "(opval=null)");
+ break;
+ }
+ Perl_sv_catpvf(aTHX_ report, " at line %d [", CopLINE(PL_curcop));
if (s - PL_bufptr > 0)
sv_catpvn(report, PL_bufptr, s - PL_bufptr);
else {
if (PL_oldbufptr && *PL_oldbufptr)
sv_catpv(report, PL_tokenbuf);
}
- PerlIO_printf(Perl_debug_log, "### %s\n", SvPV_nolen(report));
- });
+ PerlIO_printf(Perl_debug_log, "### %s]\n", SvPV_nolen(report));
+ };
+ return (int)rv;
}
#endif
/*
* S_ao
*
- * This subroutine detects &&= and ||= and turns an ANDAND or OROR
- * into an OP_ANDASSIGN or OP_ORASSIGN
+ * This subroutine detects &&=, ||=, and //= and turns an ANDAND, OROR or DORDOR
+ * into an OP_ANDASSIGN, OP_ORASSIGN, or OP_DORASSIGN
*/
STATIC int
yylval.ival = OP_ANDASSIGN;
else if (toketype == OROR)
yylval.ival = OP_ORASSIGN;
+ else if (toketype == DORDOR)
+ yylval.ival = OP_DORASSIGN;
toketype = ASSIGNOP;
}
return toketype;
else
PL_bufptr = s;
yywarn(Perl_form(aTHX_ "%s found where operator expected", what));
- if (is_first)
- Perl_warn(aTHX_ "\t(Missing semicolon on previous line?)\n");
- else if (PL_oldoldbufptr && isIDFIRST_lazy_if(PL_oldoldbufptr,UTF)) {
- char *t;
- for (t = PL_oldoldbufptr; *t && (isALNUM_lazy_if(t,UTF) || *t == ':'); t++) ;
- if (t < PL_bufptr && isSPACE(*t))
- Perl_warn(aTHX_ "\t(Do you need to predeclare %.*s?)\n",
- t - PL_oldoldbufptr, PL_oldoldbufptr);
- }
- else {
- assert(s >= oldbp);
- Perl_warn(aTHX_ "\t(Missing operator before %.*s?)\n", s - oldbp, oldbp);
+ 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)) {
+ char *t;
+ for (t = PL_oldoldbufptr; *t && (isALNUM_lazy_if(t,UTF) || *t == ':'); t++) ;
+ if (t < PL_bufptr && isSPACE(*t))
+ Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
+ "\t(Do you need to predeclare %.*s?)\n",
+ t - PL_oldoldbufptr, PL_oldoldbufptr);
+ }
+ else {
+ assert(s >= oldbp);
+ Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
+ "\t(Missing operator before %.*s?)\n", s - oldbp, oldbp);
+ }
}
PL_bufptr = oldbp;
}
SAVEPPTR(PL_last_uni);
SAVEPPTR(PL_linestart);
SAVESPTR(PL_linestr);
- SAVEPPTR(PL_lex_brackstack);
- SAVEPPTR(PL_lex_casestack);
+ SAVEGENERICPV(PL_lex_brackstack);
+ SAVEGENERICPV(PL_lex_casestack);
SAVEDESTRUCTOR_X(restore_rsfp, PL_rsfp);
SAVESPTR(PL_lex_stuff);
SAVEI32(PL_lex_defer);
PL_lex_brackets = 0;
New(899, PL_lex_brackstack, 120, char);
New(899, PL_lex_casestack, 12, char);
- SAVEFREEPV(PL_lex_brackstack);
- SAVEFREEPV(PL_lex_casestack);
PL_lex_casemods = 0;
*PL_lex_casestack = '\0';
PL_lex_dojoin = 0;
if (SvREADONLY(PL_linestr))
PL_linestr = sv_2mortal(newSVsv(PL_linestr));
s = SvPV(PL_linestr, len);
- if (len && s[len-1] != ';') {
+ if (!len || s[len-1] != ';') {
if (!(SvFLAGS(PL_linestr) & SVs_TEMP))
PL_linestr = sv_2mortal(newSVsv(PL_linestr));
sv_catpvn(PL_linestr, "\n;", 2);
PL_minus_n = PL_minus_p = 0;
}
else
- sv_setpv(PL_linestr,";");
+ sv_setpvn(PL_linestr,";", 1);
/* reset variables for next time we lex */
PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = s = PL_linestart
char ch = *s;
*s = '\0';
Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
- "Warning: Use of \"%s\" without parens is ambiguous",
+ "Warning: Use of \"%s\" without parentheses is ambiguous",
PL_last_uni);
*s = ch;
}
{
yylval.ival = f;
CLINE;
- REPORT("lop", f)
PL_expect = x;
PL_bufptr = s;
PL_last_lop = PL_oldbufptr;
PL_last_lop_op = (OPCODE)f;
if (PL_nexttoke)
- return LSTOP;
+ return REPORT(LSTOP);
if (*s == '(')
- return FUNC;
+ return REPORT(FUNC);
s = skipspace(s);
if (*s == '(')
- return FUNC;
+ return REPORT(FUNC);
else
- return LSTOP;
+ return REPORT(LSTOP);
}
/*
}
PL_nextval[PL_nexttoke].opval = (OP*)newSVOP(OP_CONST,0, newSVpv(PL_tokenbuf,0));
PL_nextval[PL_nexttoke].opval->op_private |= OPpCONST_BARE;
+ if (UTF && !IN_BYTES && is_utf8_string((U8*)PL_tokenbuf, len))
+ SvUTF8_on(((SVOP*)PL_nextval[PL_nexttoke].opval)->op_sv);
force_next(token);
}
return s;
}
yylval.opval = (OP*)newSVOP(op_type, 0, sv);
PL_lex_stuff = Nullsv;
+ /* Allow <FH> // "foo" */
+ if (op_type == OP_READLINE)
+ PL_expect = XTERMORDORDOR;
return THING;
}
SAVEPPTR(PL_last_uni);
SAVEPPTR(PL_linestart);
SAVESPTR(PL_linestr);
- SAVEPPTR(PL_lex_brackstack);
- SAVEPPTR(PL_lex_casestack);
+ SAVEGENERICPV(PL_lex_brackstack);
+ SAVEGENERICPV(PL_lex_casestack);
PL_linestr = PL_lex_stuff;
PL_lex_stuff = Nullsv;
PL_lex_brackets = 0;
New(899, PL_lex_brackstack, 120, char);
New(899, PL_lex_casestack, 12, char);
- SAVEFREEPV(PL_lex_brackstack);
- SAVEFREEPV(PL_lex_casestack);
PL_lex_casemods = 0;
*PL_lex_casestack = '\0';
PL_lex_starts = 0;
const char *leaveit = /* set of acceptably-backslashed characters */
PL_lex_inpat
- ? "\\.^$@AGZdDwWsSbBpPXC+*?|()-nrtfeaxcz0123456789[{]} \t\n\r\f\v#"
+ ? "\\.^$@AGZdDwWsSbBpPXC+*?|()-nrtfeaxz0123456789[{]} \t\n\r\f\v#"
: "";
if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
if (min > max) {
Perl_croak(aTHX_
- "Invalid [] range \"%c-%c\" in transliteration operator",
+ "Invalid range \"%c-%c\" in transliteration operator",
(char)min, (char)max);
}
except for the last char, which will be done separately. */
else if (*s == '(' && PL_lex_inpat && s[1] == '?') {
if (s[2] == '#') {
- while (s < send && *s != ')')
+ while (s+1 < send && *s != ')')
*d++ = NATIVE_TO_NEED(has_utf8,*s++);
}
else if (s[2] == '{' /* This should match regcomp.c */
count--;
regparse++;
}
- if (*regparse != ')') {
+ if (*regparse != ')')
regparse--; /* Leave one char for continuation. */
- yyerror("Sequence (?{...}) not terminated or not {}-balanced");
- }
while (s < regparse)
*d++ = NATIVE_TO_NEED(has_utf8,*s++);
}
e = s - 1;
goto cont_scan;
}
+ if (e > s + 2 && s[1] == 'U' && s[2] == '+') {
+ /* \N{U+...} */
+ I32 flags = PERL_SCAN_ALLOW_UNDERSCORES |
+ PERL_SCAN_DISALLOW_PREFIX;
+ s += 3;
+ len = e - s;
+ uv = grok_hex(s, &len, &flags, NULL);
+ s = e + 1;
+ goto NUM_ESCAPE_INSERT;
+ }
res = newSVpvn(s + 1, e - s - 1);
res = new_constant( Nullch, 0, "charnames",
res, Nullsv, "\\N{...}" );
/* \c is a control character */
case 'c':
s++;
- {
+ if (s < send) {
U8 c = *s++;
#ifdef EBCDIC
if (isLOWER(c))
#endif
*d++ = NATIVE_TO_NEED(has_utf8,toCTRL(c));
}
+ else {
+ yyerror("Missing control char name in \\c");
+ }
continue;
/* printf-style backslashes, formfeeds, newlines, etc */
*d = '\0';
SvCUR_set(sv, d - SvPVX(sv));
if (SvCUR(sv) >= SvLEN(sv))
- Perl_croak(aTHX_ "panic: constant overflowed allocated space");
+ Perl_croak(aTHX_ "panic: constant overflowed allocated space");
SvPOK_on(sv);
if (PL_encoding && !has_utf8) {
- sv_recode_to_utf8(sv, PL_encoding);
- has_utf8 = TRUE;
+ sv_recode_to_utf8(sv, PL_encoding);
+ if (SvUTF8(sv))
+ has_utf8 = TRUE;
}
if (has_utf8) {
SvUTF8_on(sv);
if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
- PL_sublex_info.sub_op->op_private |=
+ PL_sublex_info.sub_op->op_private |=
(PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
}
}
weight -= 5; /* cope with negative subscript */
break;
default:
- if (!isALNUM(last_un_char) && !strchr("$@&",last_un_char) &&
- isALPHA(*s) && s[1] && isALPHA(s[1])) {
+ if (!isALNUM(last_un_char)
+ && !(last_un_char == '$' || last_un_char == '@'
+ || last_un_char == '&')
+ && isALPHA(*s) && s[1] && isALPHA(s[1])) {
char *d = tmpbuf;
while (isALPHA(*s))
*d++ = *s++;
* Method if it's "foo $bar"
* Not a method if it's really "print foo $bar"
* Method if it's really "foo package::" (interpreted as package->foo)
- * Not a method if bar is known to be a subroutne ("sub bar; foo bar")
+ * Not a method if bar is known to be a subroutine ("sub bar; foo bar")
* Not a method if bar is a filehandle or package, but is quoted with
* =>
*/
if (pdb)
return pdb;
- SETERRNO(0,SS$_NORMAL);
+ SETERRNO(0,SS_NORMAL);
return "BEGIN { require 'perl5db.pl' }";
}
return "";
}
-/* Invoke the n'th filter function for the current rsfp. */
+/* Invoke the idxth filter function for the current rsfp. */
+/* maxlen 0 = read one text line */
I32
Perl_filter_read(pTHX_ int idx, SV *buf_sv, int maxlen)
-
-
- /* 0 = read one text line */
{
filter_t funcp;
SV *datasv = NULL;
if (!PL_rsfp_filters)
return -1;
- if (idx > AvFILLp(PL_rsfp_filters)){ /* Any more filters? */
+ if (idx > AvFILLp(PL_rsfp_filters)) { /* Any more filters? */
/* Provide a default input filter to make life easy. */
/* Note that we append to the line. This is handy. */
DEBUG_P(PerlIO_printf(Perl_debug_log,
return SvCUR(buf_sv);
}
/* Skip this filter slot if filter has been deleted */
- if ( (datasv = FILTER_DATA(idx)) == &PL_sv_undef){
+ if ( (datasv = FILTER_DATA(idx)) == &PL_sv_undef) {
DEBUG_P(PerlIO_printf(Perl_debug_log,
"filter_read %d: skipped (filter deleted)\n",
idx));
}
#endif
if (PL_rsfp_filters) {
-
if (!append)
SvCUR_set(sv, 0); /* start with empty line */
if (FILTER_READ(0, sv, 0) > 0)
#ifdef DEBUGGING
static char* exp_name[] =
{ "OPERATOR", "TERM", "REF", "STATE", "BLOCK", "ATTRBLOCK",
- "ATTRTERM", "TERMBLOCK"
+ "ATTRTERM", "TERMBLOCK", "TERMORDORDOR"
};
#endif
if we already built the token before, use it.
*/
-#ifdef USE_PURE_BISON
-int
-Perl_yylex_r(pTHX_ YYSTYPE *lvalp, int *lcharp)
-{
- int r;
-
- yyactlevel++;
- yylval_pointer[yyactlevel] = lvalp;
- yychar_pointer[yyactlevel] = lcharp;
- if (yyactlevel >= YYMAXLEVEL)
- Perl_croak(aTHX_ "panic: YYMAXLEVEL");
-
- r = Perl_yylex(aTHX);
-
- if (yyactlevel > 0)
- yyactlevel--;
-
- return r;
-}
-#endif
#ifdef __SC__
#pragma segment Perl_yylex
int
Perl_yylex(pTHX)
{
- register char *s;
+ register char *s = PL_bufptr;
register char *d;
register I32 tmp;
STRLEN len;
GV *gv = Nullgv;
GV **gvp = 0;
bool bof = FALSE;
+ I32 orig_keyword = 0;
+ DEBUG_T( {
+ PerlIO_printf(Perl_debug_log, "### LEX_%s\n",
+ lex_state_names[PL_lex_state]);
+ } );
/* check if there's an identifier for us to look at */
if (PL_pending_ident)
- return S_pending_ident(aTHX);
+ return REPORT(S_pending_ident(aTHX));
/* no identifier pending identification */
"### Next token after '%s' was known, type %"IVdf"\n", PL_bufptr,
(IV)PL_nexttype[PL_nexttoke]); });
- return(PL_nexttype[PL_nexttoke]);
+ return REPORT(PL_nexttype[PL_nexttoke]);
/* interpolated case modifiers like \L \U, including \Q and \E.
when we get here, PL_bufptr is at the \
oldmod = PL_lex_casestack[--PL_lex_casemods];
PL_lex_casestack[PL_lex_casemods] = '\0';
- if (PL_bufptr != PL_bufend && strchr("LUQ", oldmod)) {
+ if (PL_bufptr != PL_bufend
+ && (oldmod == 'L' || oldmod == 'U' || oldmod == 'Q')) {
PL_bufptr += 2;
PL_lex_state = LEX_INTERPCONCAT;
}
- return ')';
+ return REPORT(')');
}
if (PL_bufptr != PL_bufend)
PL_bufptr += 2;
DEBUG_T({ PerlIO_printf(Perl_debug_log,
"### Saw case modifier at '%s'\n", PL_bufptr); });
s = PL_bufptr + 1;
- if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
- tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
- if (strchr("LU", *s) &&
- (strchr(PL_lex_casestack, 'L') || strchr(PL_lex_casestack, 'U')))
- {
- PL_lex_casestack[--PL_lex_casemods] = '\0';
- return ')';
- }
- if (PL_lex_casemods > 10) {
- char* newlb = Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
- if (newlb != PL_lex_casestack) {
- SAVEFREEPV(newlb);
- PL_lex_casestack = newlb;
+ if (s[1] == '\\' && s[2] == 'E') {
+ PL_bufptr = s + 3;
+ PL_lex_state = LEX_INTERPCONCAT;
+ return yylex();
+ }
+ else {
+ if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
+ tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
+ if ((*s == 'L' || *s == 'U') &&
+ (strchr(PL_lex_casestack, 'L') || strchr(PL_lex_casestack, 'U'))) {
+ PL_lex_casestack[--PL_lex_casemods] = '\0';
+ return REPORT(')');
}
+ if (PL_lex_casemods > 10)
+ Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
+ PL_lex_casestack[PL_lex_casemods++] = *s;
+ PL_lex_casestack[PL_lex_casemods] = '\0';
+ PL_lex_state = LEX_INTERPCONCAT;
+ PL_nextval[PL_nexttoke].ival = 0;
+ force_next('(');
+ if (*s == 'l')
+ PL_nextval[PL_nexttoke].ival = OP_LCFIRST;
+ else if (*s == 'u')
+ PL_nextval[PL_nexttoke].ival = OP_UCFIRST;
+ else if (*s == 'L')
+ PL_nextval[PL_nexttoke].ival = OP_LC;
+ else if (*s == 'U')
+ PL_nextval[PL_nexttoke].ival = OP_UC;
+ else if (*s == 'Q')
+ PL_nextval[PL_nexttoke].ival = OP_QUOTEMETA;
+ else
+ Perl_croak(aTHX_ "panic: yylex");
+ PL_bufptr = s + 1;
}
- PL_lex_casestack[PL_lex_casemods++] = *s;
- PL_lex_casestack[PL_lex_casemods] = '\0';
- PL_lex_state = LEX_INTERPCONCAT;
- PL_nextval[PL_nexttoke].ival = 0;
- force_next('(');
- if (*s == 'l')
- PL_nextval[PL_nexttoke].ival = OP_LCFIRST;
- else if (*s == 'u')
- PL_nextval[PL_nexttoke].ival = OP_UCFIRST;
- else if (*s == 'L')
- PL_nextval[PL_nexttoke].ival = OP_LC;
- else if (*s == 'U')
- PL_nextval[PL_nexttoke].ival = OP_UC;
- else if (*s == 'Q')
- PL_nextval[PL_nexttoke].ival = OP_QUOTEMETA;
- else
- Perl_croak(aTHX_ "panic: yylex");
- PL_bufptr = s + 1;
force_next(FUNC);
if (PL_lex_starts) {
s = PL_bufptr;
PL_lex_starts = 0;
- Aop(OP_CONCAT);
+ /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
+ if (PL_lex_casemods == 1 && PL_lex_inpat)
+ OPERATOR(',');
+ else
+ Aop(OP_CONCAT);
}
else
return yylex();
}
case LEX_INTERPPUSH:
- return sublex_push();
+ return REPORT(sublex_push());
case LEX_INTERPSTART:
if (PL_bufptr == PL_bufend)
- return sublex_done();
+ return REPORT(sublex_done());
DEBUG_T({ PerlIO_printf(Perl_debug_log,
"### Interpolated variable at '%s'\n", PL_bufptr); });
PL_expect = XTERM;
if (PL_lex_dojoin) {
PL_nextval[PL_nexttoke].ival = 0;
force_next(',');
-#ifdef USE_5005THREADS
- PL_nextval[PL_nexttoke].opval = newOP(OP_THREADSV, 0);
- PL_nextval[PL_nexttoke].opval->op_targ = find_threadsv("\"");
- force_next(PRIVATEREF);
-#else
force_ident("\"", '$');
-#endif /* USE_5005THREADS */
PL_nextval[PL_nexttoke].ival = 0;
force_next('$');
PL_nextval[PL_nexttoke].ival = 0;
}
if (PL_lex_starts++) {
s = PL_bufptr;
- Aop(OP_CONCAT);
+ /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
+ if (!PL_lex_casemods && PL_lex_inpat)
+ OPERATOR(',');
+ else
+ Aop(OP_CONCAT);
}
return yylex();
if (PL_lex_dojoin) {
PL_lex_dojoin = FALSE;
PL_lex_state = LEX_INTERPCONCAT;
- return ')';
+ return REPORT(')');
}
if (PL_lex_inwhat == OP_SUBST && PL_linestr == PL_lex_repl
&& SvEVALED(PL_lex_repl))
Perl_croak(aTHX_ "panic: INTERPCONCAT");
#endif
if (PL_bufptr == PL_bufend)
- return sublex_done();
+ return REPORT(sublex_done());
if (SvIVX(PL_linestr) == '\'') {
SV *sv = newSVsv(PL_linestr);
PL_nextval[PL_nexttoke] = yylval;
PL_expect = XTERM;
force_next(THING);
- if (PL_lex_starts++)
- Aop(OP_CONCAT);
+ if (PL_lex_starts++) {
+ /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
+ if (!PL_lex_casemods && PL_lex_inpat)
+ OPERATOR(',');
+ else
+ Aop(OP_CONCAT);
+ }
else {
PL_bufptr = s;
return yylex();
PL_oldoldbufptr = PL_oldbufptr;
PL_oldbufptr = s;
DEBUG_T( {
- PerlIO_printf(Perl_debug_log, "### Tokener expecting %s at %s\n",
+ PerlIO_printf(Perl_debug_log, "### Tokener expecting %s at [%s]\n",
exp_name[PL_expect], s);
} );
if (!PL_rsfp) {
PL_last_uni = 0;
PL_last_lop = 0;
- if (PL_lex_brackets)
- yyerror("Missing right curly or square bracket");
+ if (PL_lex_brackets) {
+ if (PL_lex_formbrack)
+ yyerror("Format not terminated");
+ else
+ yyerror("Missing right curly or square bracket");
+ }
DEBUG_T( { PerlIO_printf(Perl_debug_log,
"### Tokener got EOF\n");
} );
PL_preambled = TRUE;
sv_setpv(PL_linestr,incl_perldb());
if (SvCUR(PL_linestr))
- sv_catpv(PL_linestr,";");
+ sv_catpvn(PL_linestr,";", 1);
if (PL_preambleav){
while(AvFILLp(PL_preambleav) >= 0) {
SV *tmpsv = av_shift(PL_preambleav);
sv_catsv(PL_linestr, tmpsv);
- sv_catpv(PL_linestr, ";");
+ sv_catpvn(PL_linestr, ";", 1);
sv_free(tmpsv);
}
sv_free((SV*)PL_preambleav);
sv_catpv(PL_linestr,"chomp;");
if (PL_minus_a) {
if (PL_minus_F) {
- if (strchr("/'\"", *PL_splitstr)
+ if ((*PL_splitstr == '/' || *PL_splitstr == '\''
+ || *PL_splitstr == '"')
&& strchr(PL_splitstr + 1, *PL_splitstr))
Perl_sv_catpvf(aTHX_ PL_linestr, "our @F=split(%s);", PL_splitstr);
else {
sv_catpv(PL_linestr,"our @F=split(' ');");
}
}
- sv_catpv(PL_linestr, "\n");
+ sv_catpvn(PL_linestr, "\n", 1);
PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
PL_last_lop = PL_last_uni = Nullch;
sv_setpv(PL_linestr,"");
TOKEN(';'); /* not infinite loop because rsfp is NULL now */
}
- /* if it looks like the start of a BOM, check if it in fact is */
- else if (bof && (!*s || *(U8*)s == 0xEF || *(U8*)s >= 0xFE)) {
+ /* If it looks like the start of a BOM or raw UTF-16,
+ * check if it in fact is. */
+ else if (bof &&
+ (*s == 0 ||
+ *(U8*)s == 0xEF ||
+ *(U8*)s >= 0xFE ||
+ s[1] == 0)) {
#ifdef PERLIO_IS_STDIO
# ifdef __GNU_LIBRARY__
# if __GNU_LIBRARY__ == 1 /* Linux glibc5 */
sv_setpvn(x, ipath, ipathend - ipath);
SvSETMAGIC(x);
}
+ else {
+ STRLEN blen;
+ STRLEN llen;
+ char *bstart = SvPV(CopFILESV(PL_curcop),blen);
+ char *lstart = SvPV(x,llen);
+ if (llen < blen) {
+ bstart += blen - llen;
+ if (strnEQ(bstart, lstart, llen) && bstart[-1] == '/') {
+ sv_setpvn(x, ipath, ipathend - ipath);
+ SvSETMAGIC(x);
+ }
+ }
+ }
TAINT_NOT; /* $^X is always tainted, but that's OK */
}
#endif /* ARG_ZERO_IS_SCRIPT */
else
newargv = PL_origargv;
newargv[0] = ipath;
+ PERL_FPU_PRE_EXEC
PerlProc_execv(ipath, EXEC_ARGV_CAST(newargv));
+ PERL_FPU_POST_EXEC
Perl_croak(aTHX_ "Can't exec %s", ipath);
}
#endif
}
d = moreswitches(d);
} while (d);
+ if (PL_doswitches && !switches_done) {
+ int argc = PL_origargc;
+ char **argv = PL_origargv;
+ do {
+ argc--,argv++;
+ } while (argc && argv[0][0] == '-' && argv[0][1]);
+ init_argv_symbols(argc,argv);
+ }
if ((PERLDB_LINE && !oldpdb) ||
((PL_minus_n || PL_minus_p) && !(oldn || oldp)))
/* if we have already added "LINE: while (<>) {",
/* Assume it was a minus followed by a one-letter named
* subroutine call (or a -bareword), then. */
DEBUG_T( { PerlIO_printf(Perl_debug_log,
- "### %c looked like a file test but was not\n",
- (int)ftst);
+ "### '-%c' looked like a file test but was not\n",
+ tmp);
} );
- s -= 2;
+ s = --PL_bufptr;
}
}
tmp = *s++;
PL_tokenbuf[0] = '%';
s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, TRUE);
if (!PL_tokenbuf[1]) {
- if (s == PL_bufend)
- yyerror("Final % should be \\% or %name");
PREREF('%');
}
PL_pending_ident = '%';
switch (tmp) {
case KEY_or:
case KEY_and:
+ case KEY_err:
case KEY_for:
case KEY_unless:
case KEY_if:
yyerror("Unterminated attribute parameter in attribute list");
if (attrs)
op_free(attrs);
- return 0; /* EOF indicator */
+ return REPORT(0); /* EOF indicator */
}
}
if (PL_lex_stuff) {
PL_lex_stuff = Nullsv;
}
else {
+ if (len == 6 && strnEQ(s, "unique", len)) {
+ if (PL_in_my == KEY_our)
+#ifdef USE_ITHREADS
+ GvUNIQUE_on(cGVOPx_gv(yylval.opval));
+#else
+ ; /* skip to avoid loading attributes.pm */
+#endif
+ else
+ Perl_croak(aTHX_ "The 'unique' attribute may only be applied to 'our' variables");
+ }
+
/* NOTE: any CV attrs applied here need to be part of
the CVf_BUILTIN_ATTRS define in cv.h! */
- if (!PL_in_my && len == 6 && strnEQ(s, "lvalue", len))
+ else if (!PL_in_my && len == 6 && strnEQ(s, "lvalue", len))
CvLVALUE_on(PL_compcv);
else if (!PL_in_my && len == 6 && strnEQ(s, "locked", len))
CvLOCKED_on(PL_compcv);
else if (!PL_in_my && len == 6 && strnEQ(s, "method", len))
CvMETHOD_on(PL_compcv);
-#ifdef USE_ITHREADS
- else if (PL_in_my == KEY_our && len == 6 &&
- strnEQ(s, "unique", len))
- GvUNIQUE_on(cGVOPx_gv(yylval.opval));
-#endif
+ else if (!PL_in_my && len == 9 && strnEQ(s, "assertion", len))
+ CvASSERTION_on(PL_compcv);
/* After we've set the flags, it could be argued that
we don't need to do the attributes.pm-based setting
process, and shouldn't bother appending recognized
break; /* require real whitespace or :'s */
}
tmp = (PL_expect == XOPERATOR ? '=' : '{'); /*'}(' for vi */
- if (*s != ';' && *s != tmp && (tmp != '=' || *s != ')')) {
+ if (*s != ';' && *s != '}' && *s != tmp && (tmp != '=' || *s != ')')) {
char q = ((*s == '\'') ? '"' : '\'');
/* If here for an expression, and parsed no attrs, back off. */
if (tmp == '=' && !attrs) {
PL_oldbufptr = PL_oldoldbufptr; /* allow print(STDOUT 123) */
else
PL_expect = XTERM;
+ s = skipspace(s);
TOKEN('(');
case ';':
CLINE;
leftbracket:
s++;
if (PL_lex_brackets > 100) {
- char* newlb = Renew(PL_lex_brackstack, PL_lex_brackets + 1, char);
- if (newlb != PL_lex_brackstack) {
- SAVEFREEPV(newlb);
- PL_lex_brackstack = newlb;
- }
+ Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
}
switch (PL_expect) {
case XTERM:
|| ((*t == 'q' || *t == 'x') && ++t < PL_bufend
&& !isALNUM(*t))))
{
+ /* skip q//-like construct */
char *tmps;
char open, close, term;
I32 brackets = 1;
while (t < PL_bufend && isSPACE(*t))
t++;
+ /* check for q => */
+ if (t+1 < PL_bufend && t[0] == '=' && t[1] == '>') {
+ OPERATOR(HASHBRACK);
+ }
term = *t;
open = term;
if (term && (tmps = strchr("([{< )]}> )]}>",term)))
else if (*t == open)
break;
}
- else
+ else {
for (t++; t < PL_bufend; t++) {
if (*t == '\\' && t+1 < PL_bufend)
t++;
else if (*t == open)
brackets++;
}
+ }
+ t++;
}
- t++;
+ else
+ /* skip plain q word */
+ while (t < PL_bufend && isALNUM_lazy_if(t,UTF))
+ t += UTF8SKIP(t);
}
else if (isALNUM_lazy_if(t,UTF)) {
t += UTF8SKIP(t);
case '!':
s++;
tmp = *s++;
- if (tmp == '=')
+ if (tmp == '=') {
+ /* was this !=~ where !~ was meant?
+ * warn on m:!=~\s+([/?]|[msy]\W|tr\W): */
+
+ if (*s == '~' && ckWARN(WARN_SYNTAX)) {
+ char *t = s+1;
+
+ while (t < PL_bufend && isSPACE(*t))
+ ++t;
+
+ if (*t == '/' || *t == '?' ||
+ ((*t == 'm' || *t == 's' || *t == 'y') && !isALNUM(t[1])) ||
+ (*t == 't' && t[1] == 'r' && !isALNUM(t[2])))
+ Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
+ "!=~ should be !~");
+ }
Eop(OP_NE);
+ }
if (tmp == '~')
PMop(OP_NOT);
s--;
if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) {
PL_expect = XTERM;
depcom();
- return ','; /* grandfather non-comma-format format */
+ return REPORT(','); /* grandfather non-comma-format format */
}
}
}
}
else {
- GV *gv = gv_fetchpv(tmpbuf, FALSE, SVt_PVCV);
- if (gv && GvCVu(gv))
- PL_expect = XTERM; /* e.g. print $fh subr() */
+ PL_expect = XTERM; /* e.g. print $fh subr() */
}
}
else if (isDIGIT(*s))
PL_expect = XTERM; /* e.g. print $fh 3 */
else if (*s == '.' && isDIGIT(s[1]))
PL_expect = XTERM; /* e.g. print $fh .3 */
- else if (strchr("/?-+", *s) && !isSPACE(s[1]) && s[1] != '=')
- PL_expect = XTERM; /* e.g. print $fh -1 */
+ else if ((*s == '?' || *s == '-' || *s == '+')
+ && !isSPACE(s[1]) && s[1] != '=')
+ PL_expect = XTERM; /* e.g. print $fh -1 */
+ else if (*s == '/' && !isSPACE(s[1]) && s[1] != '=' && s[1] != '/')
+ PL_expect = XTERM; /* e.g. print $fh /.../
+ XXX except DORDOR operator */
else if (*s == '<' && s[1] == '<' && !isSPACE(s[2]) && s[2] != '=')
PL_expect = XTERM; /* print $fh <<"EOF" */
}
PL_tokenbuf[0] = '@';
s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, FALSE);
if (!PL_tokenbuf[1]) {
- if (s == PL_bufend)
- yyerror("Final @ should be \\@ or @name");
PREREF('@');
}
if (PL_lex_state == LEX_NORMAL)
PL_pending_ident = '@';
TERM('@');
- case '/': /* may either be division or pattern */
- case '?': /* may either be conditional or pattern */
- if (PL_expect != XOPERATOR) {
- /* Disable warning on "study /blah/" */
- if (PL_oldoldbufptr == PL_last_uni
- && (*PL_last_uni != 's' || s - PL_last_uni < 5
- || memNE(PL_last_uni, "study", 5)
- || isALNUM_lazy_if(PL_last_uni+5,UTF)))
- check_uni();
- s = scan_pat(s,OP_MATCH);
- TERM(sublex_start());
- }
- tmp = *s++;
- if (tmp == '/')
- Mop(OP_DIVIDE);
- OPERATOR(tmp);
+ case '/': /* may be division, defined-or, or pattern */
+ if (PL_expect == XTERMORDORDOR && s[1] == '/') {
+ s += 2;
+ AOPERATOR(DORDOR);
+ }
+ case '?': /* may either be conditional or pattern */
+ if(PL_expect == XOPERATOR) {
+ tmp = *s++;
+ if(tmp == '?') {
+ OPERATOR('?');
+ }
+ else {
+ tmp = *s++;
+ if(tmp == '/') {
+ /* A // operator. */
+ AOPERATOR(DORDOR);
+ }
+ else {
+ s--;
+ Mop(OP_DIVIDE);
+ }
+ }
+ }
+ else {
+ /* Disable warning on "study /blah/" */
+ if (PL_oldoldbufptr == PL_last_uni
+ && (*PL_last_uni != 's' || s - PL_last_uni < 5
+ || memNE(PL_last_uni, "study", 5)
+ || isALNUM_lazy_if(PL_last_uni+5,UTF)
+ ))
+ check_uni();
+ s = scan_pat(s,OP_MATCH);
+ TERM(sublex_start());
+ }
case '.':
if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack
if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) {
PL_expect = XTERM;
depcom();
- return ','; /* grandfather non-comma-format format */
+ return REPORT(','); /* grandfather non-comma-format format */
}
else
no_op("String",s);
if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) {
PL_expect = XTERM;
depcom();
- return ','; /* grandfather non-comma-format format */
+ return REPORT(','); /* grandfather non-comma-format format */
}
else
no_op("String",s);
TERM(THING);
}
/* avoid v123abc() or $h{v1}, allow C<print v10;> */
- else if (!isALPHA(*start) && (PL_expect == XTERM || PL_expect == XREF || PL_expect == XSTATE)) {
+ else if (!isALPHA(*start) && (PL_expect == XTERM
+ || PL_expect == XREF || PL_expect == XSTATE
+ || PL_expect == XTERMORDORDOR)) {
char c = *start;
GV *gv;
*start = '\0';
case 'z': case 'Z':
keylookup: {
+ orig_keyword = 0;
gv = Nullgv;
gvp = 0;
}
}
if (ogv) {
+ orig_keyword = tmp;
tmp = 0; /* overridden by import or by GLOBAL */
}
else if (gv && !gvp
{
tmp = 0; /* any sub overrides "weak" keyword */
}
+ else if (gv && !gvp
+ && tmp == -KEY_err
+ && GvCVu(gv)
+ && PL_expect != XOPERATOR
+ && PL_expect != XTERMORDORDOR)
+ {
+ /* any sub overrides the "err" keyword, except when really an
+ * operator is expected */
+ tmp = 0;
+ }
else { /* no override */
tmp = -tmp;
if (tmp == KEY_dump && ckWARN(WARN_MISC)) {
CLINE;
yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
yylval.opval->op_private = OPpCONST_BARE;
+ /* UTF-8 package name? */
+ if (UTF && !IN_BYTES &&
+ is_utf8_string((U8*)SvPVX(sv), SvCUR(sv)))
+ SvUTF8_on(sv);
/* And if "Foo::", then that's what it certainly is. */
/* Two barewords in a row may indicate method call. */
if ((isIDFIRST_lazy_if(s,UTF) || *s == '$') && (tmp=intuit_method(s,gv)))
- return tmp;
+ return REPORT(tmp);
/* If not a declared subroutine, it's an indirect object. */
/* (But it's an indir obj regardless for sort.) */
/* If followed by a bareword, see if it looks like indir obj. */
- if ((isIDFIRST_lazy_if(s,UTF) || *s == '$') && (tmp = intuit_method(s,gv)))
- return tmp;
+ if (!orig_keyword
+ && (isIDFIRST_lazy_if(s,UTF) || *s == '$')
+ && (tmp = intuit_method(s,gv)))
+ return REPORT(tmp);
/* Not a method, so call it a subroutine (if defined) */
char *proto = SvPV((SV*)cv, len);
if (!len)
TERM(FUNC0SUB);
- if (strEQ(proto, "$"))
+ if (*proto == '$' && proto[1] == '\0')
OPERATOR(UNIOPSUB);
+ while (*proto == ';')
+ proto++;
if (*proto == '&' && *s == '{') {
sv_setpv(PL_subname, PL_curstash ?
"__ANON__" : "__ANON__::__ANON__");
if (ckWARN(WARN_RESERVED)) {
if (lastchar != '-') {
for (d = PL_tokenbuf; *d && isLOWER(*d); d++) ;
- if (!*d && strNE(PL_tokenbuf,"main"))
+ if (!*d && !gv_stashpv(PL_tokenbuf,FALSE))
Perl_warner(aTHX_ packWARN(WARN_RESERVED), PL_warn_reserved,
PL_tokenbuf);
}
}
safe_bareword:
- if (lastchar && strchr("*%&", lastchar) && ckWARN_d(WARN_AMBIGUOUS)) {
+ if ((lastchar == '*' || lastchar == '%' || lastchar == '&')
+ && ckWARN_d(WARN_AMBIGUOUS)) {
Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
"Operator or semicolon missing before %c%s",
lastchar, PL_tokenbuf);
case KEY___PACKAGE__:
yylval.opval = (OP*)newSVOP(OP_CONST, 0,
(PL_curstash
- ? newSVsv(PL_curstname)
+ ? newSVpv(HvNAME(PL_curstash), 0)
: &PL_sv_undef));
TERM(THING);
}
#endif
#ifdef PERLIO_LAYERS
- if (UTF && !IN_BYTES)
- PerlIO_apply_layers(aTHX_ PL_rsfp, NULL, ":utf8");
+ if (!IN_BYTES) {
+ if (UTF)
+ PerlIO_apply_layers(aTHX_ PL_rsfp, NULL, ":utf8");
+ else if (PL_encoding) {
+ SV *name;
+ dSP;
+ ENTER;
+ SAVETMPS;
+ PUSHMARK(sp);
+ EXTEND(SP, 1);
+ XPUSHs(PL_encoding);
+ PUTBACK;
+ call_method("name", G_SCALAR);
+ SPAGAIN;
+ name = POPs;
+ PUTBACK;
+ PerlIO_apply_layers(aTHX_ PL_rsfp, NULL,
+ Perl_form(aTHX_ ":encoding(%"SVf")",
+ name));
+ FREETMPS;
+ LEAVE;
+ }
+ }
#endif
PL_rsfp = Nullfp;
}
case KEY_eof:
UNI(OP_EOF);
+ case KEY_err:
+ OPERATOR(DOROP);
+
case KEY_exp:
UNI(OP_EXP);
UNI(OP_GMTIME);
case KEY_getc:
- UNI(OP_GETC);
+ UNIDOR(OP_GETC);
case KEY_getppid:
FUN0(OP_GETPPID);
if (isIDFIRST_lazy_if(s,UTF)) {
char *t;
for (d = s; isALNUM_lazy_if(d,UTF); d++) ;
- t = skipspace(d);
- if (strchr("|&*+-=!?:.", *t) && ckWARN_d(WARN_PRECEDENCE))
+ for (t=d; *t && isSPACE(*t); t++) ;
+ if ( *t && strchr("|&*+-=!?:.", *t) && ckWARN_d(WARN_PRECEDENCE)
+ /* [perl #16184] */
+ && !(t[0] == '=' && t[1] == '>')
+ ) {
Perl_warner(aTHX_ packWARN(WARN_PRECEDENCE),
"Precedence problem: open %.*s should be open(%.*s)",
- d-s,s, d-s,s);
+ d - s, s, d - s, s);
+ }
}
LOP(OP_OPEN,XTERM);
LOP(OP_PUSH,XTERM);
case KEY_pop:
- UNI(OP_POP);
+ UNIDOR(OP_POP);
case KEY_pos:
- UNI(OP_POS);
+ UNIDOR(OP_POS);
case KEY_pack:
LOP(OP_PACK,XTERM);
case KEY_readline:
set_csh();
- UNI(OP_READLINE);
+ UNIDOR(OP_READLINE);
case KEY_readpipe:
set_csh();
LOP(OP_REVERSE,XTERM);
case KEY_readlink:
- UNI(OP_READLINK);
+ UNIDOR(OP_READLINK);
case KEY_ref:
UNI(OP_REF);
LOP(OP_SSOCKOPT,XTERM);
case KEY_shift:
- UNI(OP_SHIFT);
+ UNIDOR(OP_SHIFT);
case KEY_shmctl:
LOP(OP_SHMCTL,XTERM);
d[tmp] = '\0';
if (bad_proto && ckWARN(WARN_SYNTAX))
Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
- "Illegal character in prototype for %s : %s",
- SvPVX(PL_subname), d);
+ "Illegal character in prototype for %"SVf" : %s",
+ PL_subname, d);
SvCUR(PL_lex_stuff) = tmp;
have_proto = TRUE;
if (*s == ':' && s[1] != ':')
PL_expect = attrful;
+ else if (*s != '{' && key == KEY_sub) {
+ if (!have_name)
+ Perl_croak(aTHX_ "Illegal declaration of anonymous subroutine");
+ else if (*s != ';')
+ Perl_croak(aTHX_ "Illegal declaration of subroutine %"SVf, PL_subname);
+ }
if (have_proto) {
PL_nextval[PL_nexttoke].opval =
LOP(OP_UNLINK,XTERM);
case KEY_undef:
- UNI(OP_UNDEF);
+ UNIDOR(OP_UNDEF);
case KEY_unpack:
LOP(OP_UNPACK,XTERM);
LOP(OP_UTIME,XTERM);
case KEY_umask:
- UNI(OP_UMASK);
+ UNIDOR(OP_UMASK);
case KEY_unshift:
LOP(OP_UNSHIFT,XTERM);
S_pending_ident(pTHX)
{
register char *d;
- register I32 tmp;
+ register I32 tmp = 0;
/* pit holds the identifier we read and pending_ident is reset */
char pit = PL_pending_ident;
PL_pending_ident = 0;
yyerror(Perl_form(aTHX_ "No package name allowed for "
"variable %s in \"our\"",
PL_tokenbuf));
- tmp = pad_allocmy(PL_tokenbuf);
+ tmp = allocmy(PL_tokenbuf);
}
else {
if (strchr(PL_tokenbuf,':'))
yyerror(Perl_form(aTHX_ PL_no_myglob,PL_tokenbuf));
yylval.opval = newOP(OP_PADANY, 0);
- yylval.opval->op_targ = pad_allocmy(PL_tokenbuf);
+ yylval.opval->op_targ = allocmy(PL_tokenbuf);
return PRIVATEREF;
}
}
*/
if (!strchr(PL_tokenbuf,':')) {
-#ifdef USE_5005THREADS
- /* Check for single character per-thread SVs */
- if (PL_tokenbuf[0] == '$' && PL_tokenbuf[2] == '\0'
- && !isALPHA(PL_tokenbuf[1]) /* Rule out obvious non-threadsvs */
- && (tmp = find_threadsv(&PL_tokenbuf[1])) != NOT_IN_PAD)
- {
- yylval.opval = newOP(OP_THREADSV, 0);
- yylval.opval->op_targ = tmp;
- return PRIVATEREF;
- }
-#endif /* USE_5005THREADS */
- if ((tmp = pad_findmy(PL_tokenbuf)) != NOT_IN_PAD) {
- SV *namesv = AvARRAY(PL_comppad_name)[tmp];
+ if (!PL_in_my)
+ tmp = pad_findmy(PL_tokenbuf);
+ if (tmp != NOT_IN_PAD) {
/* might be an "our" variable" */
- if (SvFLAGS(namesv) & SVpad_OUR) {
+ if (PAD_COMPNAME_FLAGS(tmp) & SVpad_OUR) {
/* build ops for a bareword */
- SV *sym = newSVpv(HvNAME(GvSTASH(namesv)),0);
+ SV *sym = newSVpv(HvNAME(PAD_COMPNAME_OURSTASH(tmp)), 0);
sv_catpvn(sym, "::", 2);
sv_catpv(sym, PL_tokenbuf+1);
yylval.opval = (OP*)newSVOP(OP_CONST, 0, sym);
break;
case 3:
if (strEQ(d,"eof")) return -KEY_eof;
+ if (strEQ(d,"err")) return -KEY_err;
if (strEQ(d,"exp")) return -KEY_exp;
break;
case 4:
break;
case 6:
if (strEQ(d,"exists")) return KEY_exists;
- if (strEQ(d,"elseif")) Perl_warn(aTHX_ "elseif should be elsif");
+ if (strEQ(d,"elseif") && ckWARN_d(WARN_SYNTAX))
+ Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
+ "elseif should be elsif");
break;
case 8:
if (strEQ(d,"endgrent")) return -KEY_endgrent;
else if (*d == 'l') {
if (strEQ(d,"login")) return -KEY_getlogin;
}
- else if (strEQ(d,"c")) return -KEY_getc;
+ else if (*d == 'c' && d[1] == '\0') return -KEY_getc;
break;
}
switch (len) {
}
break;
case 'q':
- if (len <= 2) {
- if (strEQ(d,"q")) return KEY_q;
- if (strEQ(d,"qr")) return KEY_qr;
- if (strEQ(d,"qq")) return KEY_qq;
- if (strEQ(d,"qw")) return KEY_qw;
- if (strEQ(d,"qx")) return KEY_qx;
+ if (len == 1) {
+ return KEY_q;
+ }
+ else if (len == 2) {
+ switch (d[1]) {
+ case 'r': return KEY_qr;
+ case 'q': return KEY_qq;
+ case 'w': return KEY_qw;
+ case 'x': return KEY_qx;
+ };
}
else if (strEQ(d,"quotemeta")) return -KEY_quotemeta;
break;
return s;
}
if (*s == '$' && s[1] &&
- (isALNUM_lazy_if(s+1,UTF) || strchr("${", s[1]) || strnEQ(s+1,"::",2)) )
+ (isALNUM_lazy_if(s+1,UTF) || s[1] == '$' || s[1] == '{' || strnEQ(s+1,"::",2)) )
{
return s;
}
}
if (*s == '}') {
s++;
- if (PL_lex_state == LEX_INTERPNORMAL && !PL_lex_brackets)
+ if (PL_lex_state == LEX_INTERPNORMAL && !PL_lex_brackets) {
PL_lex_state = LEX_INTERPEND;
+ PL_expect = XREF;
+ }
if (funny == '#')
funny = '@';
if (PL_lex_state == LEX_NORMAL) {
}
complement = del = squash = 0;
- while (strchr("cds", *s)) {
- if (*s == 'c')
+ while (1) {
+ switch (*s) {
+ case 'c':
complement = OPpTRANS_COMPLEMENT;
- else if (*s == 'd')
+ break;
+ case 'd':
del = OPpTRANS_DELETE;
- else if (*s == 's')
+ break;
+ case 's':
squash = OPpTRANS_SQUASH;
+ break;
+ default:
+ goto no_more;
+ }
s++;
}
+ no_more:
New(803, tbl, complement&&!del?258:256, short);
o = newPVOP(OP_TRANS, 0, (char*)tbl);
- o->op_private = del|squash|complement|
+ o->op_private &= ~OPpTRANS_ALL;
+ o->op_private |= del|squash|complement|
(DO_UTF8(PL_lex_stuff)? OPpTRANS_FROM_UTF : 0)|
(DO_UTF8(PL_lex_repl) ? OPpTRANS_TO_UTF : 0);
if (!outer)
*d++ = '\n';
for (peek = s; SPACE_OR_TAB(*peek); peek++) ;
- if (*peek && strchr("`'\"",*peek)) {
+ if (*peek == '`' || *peek == '\'' || *peek =='"') {
s = peek;
term = *s++;
s = delimcpy(d, e, s, PL_bufend, term, &len);
sv_setpvn(tmpstr,d+1,s-d);
s += len - 1;
sv_catpvn(herewas,s,bufend-s);
- (void)strcpy(bufptr,SvPVX(herewas));
+ Copy(SvPVX(herewas),bufptr,SvCUR(herewas) + 1,char);
s = olds;
goto retval;
av_store(CopFILEAV(PL_curcop), (I32)CopLINE(PL_curcop),sv);
}
if (*s == term && memEQ(s,PL_tokenbuf,len)) {
- s = PL_bufend - 1;
- *s = ' ';
+ STRLEN off = PL_bufend - 1 - SvPVX(PL_linestr);
+ *(SvPVX(PL_linestr) + off ) = ' ';
sv_catsv(PL_linestr,herewas);
PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
+ s = SvPVX(PL_linestr) + off; /* In case PV of PL_linestr moved. */
}
else {
s = PL_bufend;
Renew(SvPVX(tmpstr), SvLEN(tmpstr), char);
}
SvREFCNT_dec(herewas);
- if (UTF && !IN_BYTES && is_utf8_string((U8*)SvPVX(tmpstr), SvCUR(tmpstr)))
- SvUTF8_on(tmpstr);
+ if (!IN_BYTES) {
+ if (UTF && is_utf8_string((U8*)SvPVX(tmpstr), SvCUR(tmpstr)))
+ SvUTF8_on(tmpstr);
+ else if (PL_encoding)
+ sv_recode_to_utf8(tmpstr, PL_encoding);
+ }
PL_lex_stuff = tmpstr;
yylval.ival = op_type;
return s;
/* turn <> into <ARGV> */
if (!len)
- (void)strcpy(d,"ARGV");
+ Copy("ARGV",d,5,char);
/* Check whether readline() is overriden */
if (((gv_readline = gv_fetchpv("readline", FALSE, SVt_PVCV))
add symbol table ops
*/
if ((tmp = pad_findmy(d)) != NOT_IN_PAD) {
- SV *namesv = AvARRAY(PL_comppad_name)[tmp];
- if (SvFLAGS(namesv) & SVpad_OUR) {
- SV *sym = sv_2mortal(newSVpv(HvNAME(GvSTASH(namesv)),0));
+ if (PAD_COMPNAME_FLAGS(tmp) & SVpad_OUR) {
+ SV *sym = sv_2mortal(
+ newSVpv(HvNAME(PAD_COMPNAME_OURSTASH(tmp)),0));
sv_catpvn(sym, "::", 2);
sv_catpv(sym, d+1);
d = SvPVX(sym);
register char *to; /* current position in the sv's data */
I32 brackets = 1; /* bracket nesting level */
bool has_utf8 = FALSE; /* is there any utf8 content? */
+ I32 termcode; /* terminating char. code */
+ U8 termstr[UTF8_MAXLEN]; /* terminating string */
+ STRLEN termlen; /* length of terminating string */
+ char *last = NULL; /* last position for nesting bracket */
/* skip space before the delimiter */
if (isSPACE(*s))
/* after skipping whitespace, the next character is the terminator */
term = *s;
- if (!UTF8_IS_INVARIANT((U8)term) && UTF)
- has_utf8 = TRUE;
+ if (!UTF) {
+ termcode = termstr[0] = term;
+ termlen = 1;
+ }
+ else {
+ termcode = utf8_to_uvchr((U8*)s, &termlen);
+ Copy(s, termstr, termlen, U8);
+ if (!UTF8_IS_INVARIANT(term))
+ has_utf8 = TRUE;
+ }
/* mark where we are */
PL_multi_start = CopLINE(PL_curcop);
/* find corresponding closing delimiter */
if (term && (tmps = strchr("([{< )]}> )]}>",term)))
- term = tmps[5];
+ termcode = termstr[0] = term = tmps[5];
+
PL_multi_close = term;
/* create a new SV to hold the contents. 87 is leak category, I'm
assuming. 79 is the SV's initial length. What a random number. */
sv = NEWSV(87,79);
sv_upgrade(sv, SVt_PVIV);
- SvIVX(sv) = term;
+ SvIVX(sv) = termcode;
(void)SvPOK_only(sv); /* validate pointer */
/* move past delimiter and try to read a complete string */
if (keep_delims)
- sv_catpvn(sv, s, 1);
- s++;
+ sv_catpvn(sv, s, termlen);
+ s += termlen;
for (;;) {
+ if (PL_encoding && !UTF) {
+ bool cont = TRUE;
+
+ while (cont) {
+ int offset = s - SvPVX(PL_linestr);
+ bool found = sv_cat_decode(sv, PL_encoding, PL_linestr,
+ &offset, (char*)termstr, termlen);
+ char *ns = SvPVX(PL_linestr) + offset;
+ char *svlast = SvEND(sv) - 1;
+
+ for (; s < ns; s++) {
+ if (*s == '\n' && !PL_rsfp)
+ CopLINE_inc(PL_curcop);
+ }
+ if (!found)
+ goto read_more_line;
+ else {
+ /* handle quoted delimiters */
+ if (SvCUR(sv) > 1 && *(svlast-1) == '\\') {
+ char *t;
+ for (t = svlast-2; t >= SvPVX(sv) && *t == '\\';)
+ t--;
+ if ((svlast-1 - t) % 2) {
+ if (!keep_quoted) {
+ *(svlast-1) = term;
+ *svlast = '\0';
+ SvCUR_set(sv, SvCUR(sv) - 1);
+ }
+ continue;
+ }
+ }
+ if (PL_multi_open == PL_multi_close) {
+ cont = FALSE;
+ }
+ else {
+ char *t, *w;
+ if (!last)
+ last = SvPVX(sv);
+ for (w = t = last; t < svlast; w++, t++) {
+ /* At here, all closes are "was quoted" one,
+ so we don't check PL_multi_close. */
+ if (*t == '\\') {
+ if (!keep_quoted && *(t+1) == PL_multi_open)
+ t++;
+ else
+ *w++ = *t++;
+ }
+ else if (*t == PL_multi_open)
+ brackets++;
+
+ *w = *t;
+ }
+ if (w < t) {
+ *w++ = term;
+ *w = '\0';
+ SvCUR_set(sv, w - SvPVX(sv));
+ }
+ last = w;
+ if (--brackets <= 0)
+ cont = FALSE;
+ }
+ }
+ }
+ if (!keep_delims) {
+ SvCUR_set(sv, SvCUR(sv) - 1);
+ *SvEND(sv) = '\0';
+ }
+ break;
+ }
+
/* extend sv if need be */
SvGROW(sv, SvCUR(sv) + (PL_bufend - s) + 1);
/* set 'to' to the next character in the sv's string */
}
/* terminate when run out of buffer (the for() condition), or
have found the terminator */
- else if (*s == term)
- break;
+ else if (*s == term) {
+ if (termlen == 1)
+ break;
+ if (s+termlen <= PL_bufend && memEQ(s, (char*)termstr, termlen))
+ break;
+ }
else if (!has_utf8 && !UTF8_IS_INVARIANT((U8)*s) && UTF)
has_utf8 = TRUE;
*to = *s;
to[-1] = '\n';
#endif
+ read_more_line:
/* if we're out of file, or a read fails, bail and reset the current
line marker so we can report where the unterminated string began
*/
/* at this point, we have successfully read the delimited string */
- if (keep_delims)
- sv_catpvn(sv, s, 1);
- if (has_utf8)
+ if (!PL_encoding || UTF) {
+ if (keep_delims)
+ sv_catpvn(sv, s, termlen);
+ s += termlen;
+ }
+ if (has_utf8 || PL_encoding)
SvUTF8_on(sv);
+
PL_multi_end = CopLINE(PL_curcop);
- s++;
/* if we allocated too much space, give some back */
if (SvCUR(sv) + 5 < SvLEN(sv)) {
UV u = 0;
I32 shift;
bool overflowed = FALSE;
+ bool just_zero = TRUE; /* just plain 0 or binary number? */
static NV nvshift[5] = { 1.0, 2.0, 4.0, 8.0, 16.0 };
static char* bases[5] = { "", "binary", "", "octal",
"hexadecimal" };
if (s[1] == 'x') {
shift = 4;
s += 2;
+ just_zero = FALSE;
} else if (s[1] == 'b') {
shift = 1;
s += 2;
+ just_zero = FALSE;
}
/* check for a decimal in disguise */
else if (s[1] == '.' || s[1] == 'e' || s[1] == 'E')
*/
digit:
+ just_zero = FALSE;
if (!overflowed) {
x = u << shift; /* make room for the digit */
#endif
sv_setuv(sv, u);
}
- if (PL_hints & HINT_NEW_BINARY)
+ if (just_zero && (PL_hints & HINT_NEW_INTEGER))
+ sv = new_constant(start, s - start, "integer",
+ sv, Nullsv, NULL);
+ else if (PL_hints & HINT_NEW_BINARY)
sv = new_constant(start, s - start, "binary", sv, Nullsv, NULL);
}
break;
}
/* read exponent part, if present */
- if (*s && strchr("eE",*s) && strchr("+-0123456789_", s[1])) {
+ if ((*s == 'e' || *s == 'E') && strchr("+-0123456789_", s[1])) {
floatit = TRUE;
s++;
case 'v':
vstring:
sv = NEWSV(92,5); /* preallocate storage space */
- s = new_vstring(s,sv);
+ s = scan_vstring(s,sv);
break;
}
register char *t;
SV *stuff = newSVpvn("",0);
bool needargs = FALSE;
+ bool eofmt = FALSE;
while (!needargs) {
- if (*s == '.' || *s == /*{*/'}') {
+ if (*s == '.') {
/*SUPPRESS 530*/
#ifdef PERL_STRICT_CR
for (t = s+1;SPACE_OR_TAB(*t); t++) ;
#else
for (t = s+1;SPACE_OR_TAB(*t) || *t == '\r'; t++) ;
#endif
- if (*t == '\n' || t == PL_bufend)
+ if (*t == '\n' || t == PL_bufend) {
+ eofmt = TRUE;
break;
+ }
}
if (PL_in_eval && !PL_rsfp) {
- eol = strchr(s,'\n');
+ eol = memchr(s,'\n',PL_bufend-s);
if (!eol++)
eol = PL_bufend;
}
PL_last_lop = PL_last_uni = Nullch;
if (!s) {
s = PL_bufptr;
- yyerror("Format not terminated");
break;
}
}
}
else
PL_lex_state = LEX_FORMLINE;
+ if (!IN_BYTES) {
+ if (UTF && is_utf8_string((U8*)SvPVX(stuff), SvCUR(stuff)))
+ SvUTF8_on(stuff);
+ else if (PL_encoding)
+ sv_recode_to_utf8(stuff, PL_encoding);
+ }
PL_nextval[PL_nexttoke].opval = (OP*)newSVOP(OP_CONST, 0, stuff);
force_next(THING);
PL_nextval[PL_nexttoke].ival = OP_FORMLINE;
}
else {
SvREFCNT_dec(stuff);
- PL_lex_formbrack = 0;
+ if (eofmt)
+ PL_lex_formbrack = 0;
PL_bufptr = s;
}
return s;
{
I32 oldsavestack_ix = PL_savestack_ix;
CV* outsidecv = PL_compcv;
- AV* comppadlist;
if (PL_compcv) {
assert(SvTYPE(PL_compcv) == SVt_PVCV);
}
SAVEI32(PL_subline);
save_item(PL_subname);
- SAVEI32(PL_padix);
- SAVECOMPPAD();
- SAVESPTR(PL_comppad_name);
SAVESPTR(PL_compcv);
- SAVEI32(PL_comppad_name_fill);
- SAVEI32(PL_min_intro_pending);
- SAVEI32(PL_max_intro_pending);
- SAVEI32(PL_pad_reset_pending);
PL_compcv = (CV*)NEWSV(1104,0);
sv_upgrade((SV *)PL_compcv, is_format ? SVt_PVFM : SVt_PVCV);
CvFLAGS(PL_compcv) |= flags;
- PL_comppad = newAV();
- av_push(PL_comppad, Nullsv);
- PL_curpad = AvARRAY(PL_comppad);
- PL_comppad_name = newAV();
- PL_comppad_name_fill = 0;
- PL_min_intro_pending = 0;
- PL_padix = 0;
PL_subline = CopLINE(PL_curcop);
-#ifdef USE_5005THREADS
- av_store(PL_comppad_name, 0, newSVpvn("@_", 2));
- PL_curpad[0] = (SV*)newAV();
- SvPADMY_on(PL_curpad[0]); /* XXX Needed? */
-#endif /* USE_5005THREADS */
-
- comppadlist = newAV();
- AvREAL_off(comppadlist);
- av_store(comppadlist, 0, (SV*)PL_comppad_name);
- av_store(comppadlist, 1, (SV*)PL_comppad);
-
- CvPADLIST(PL_compcv) = comppadlist;
+ CvPADLIST(PL_compcv) = pad_new(padnew_SAVE|padnew_SAVESUB);
CvOUTSIDE(PL_compcv) = (CV*)SvREFCNT_inc(outsidecv);
-#ifdef USE_5005THREADS
- CvOWNER(PL_compcv) = 0;
- New(666, CvMUTEXP(PL_compcv), 1, perl_mutex);
- MUTEX_INIT(CvMUTEXP(PL_compcv));
-#endif /* USE_5005THREADS */
+ CvOUTSIDE_SEQ(PL_compcv) = PL_cop_seqmax;
return oldsavestack_ix;
}
where = "at EOF";
else if (PL_bufptr > PL_oldoldbufptr && PL_bufptr - PL_oldoldbufptr < 200 &&
PL_oldoldbufptr != PL_oldbufptr && PL_oldbufptr != PL_bufptr) {
+ /*
+ Only for NetWare:
+ The code below is removed for NetWare because it abends/crashes on NetWare
+ when the script has error such as not having the closing quotes like:
+ if ($var eq "value)
+ Checking of white spaces is anyway done in NetWare code.
+ */
+#ifndef NETWARE
while (isSPACE(*PL_oldoldbufptr))
PL_oldoldbufptr++;
+#endif
context = PL_oldoldbufptr;
contlen = PL_bufptr - PL_oldoldbufptr;
}
else if (PL_bufptr > PL_oldbufptr && PL_bufptr - PL_oldbufptr < 200 &&
PL_oldbufptr != PL_bufptr) {
+ /*
+ Only for NetWare:
+ The code below is removed for NetWare because it abends/crashes on NetWare
+ when the script has error such as not having the closing quotes like:
+ if ($var eq "value)
+ Checking of white spaces is anyway done in NetWare code.
+ */
+#ifndef NETWARE
while (isSPACE(*PL_oldbufptr))
PL_oldbufptr++;
+#endif
context = PL_oldbufptr;
contlen = PL_bufptr - PL_oldbufptr;
}
else if (yychar > 255)
where = "next token ???";
-#ifdef USE_PURE_BISON
-/* GNU Bison sets the value -2 */
- else if (yychar == -2) {
-#else
- else if ((yychar & 127) == 127) {
-#endif
+ else if (yychar == -2) { /* YYEMPTY */
if (PL_lex_state == LEX_NORMAL ||
(PL_lex_state == LEX_KNOWNEXT && PL_lex_defer == LEX_NORMAL))
where = "at end of line";
(int)PL_multi_open,(int)PL_multi_close,(IV)PL_multi_start);
PL_multi_end = 0;
}
- if (PL_in_eval & EVAL_WARNONLY)
- Perl_warn(aTHX_ "%"SVf, msg);
+ if (PL_in_eval & EVAL_WARNONLY && ckWARN_d(WARN_SYNTAX))
+ Perl_warner(aTHX_ packWARN(WARN_SYNTAX), "%"SVf, msg);
else
qerror(msg);
if (PL_error_count >= 10) {
{
STRLEN slen;
slen = SvCUR(PL_linestr);
- switch (*s) {
+ switch (s[0]) {
case 0xFF:
if (s[1] == 0xFE) {
- /* UTF-16 little-endian */
+ /* UTF-16 little-endian? (or UTF32-LE?) */
if (s[2] == 0 && s[3] == 0) /* UTF-32 little-endian */
- Perl_croak(aTHX_ "Unsupported script encoding");
+ Perl_croak(aTHX_ "Unsupported script encoding UTF32-LE");
#ifndef PERL_NO_UTF16_FILTER
- DEBUG_p(PerlIO_printf(Perl_debug_log, "UTF-LE script encoding\n"));
+ if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF16-LE script encoding (BOM)\n");
s += 2;
+ utf16le:
if (PL_bufend > (char*)s) {
U8 *news;
I32 newlen;
filter_add(utf16rev_textfilter, NULL);
New(898, news, (PL_bufend - (char*)s) * 3 / 2 + 1, U8);
- PL_bufend = (char*)utf16_to_utf8_reversed(s, news,
- PL_bufend - (char*)s - 1,
- &newlen);
- Copy(news, s, newlen, U8);
- SvCUR_set(PL_linestr, newlen);
- PL_bufend = SvPVX(PL_linestr) + newlen;
- news[newlen++] = '\0';
+ utf16_to_utf8_reversed(s, news,
+ PL_bufend - (char*)s - 1,
+ &newlen);
+ sv_setpvn(PL_linestr, (const char*)news, newlen);
Safefree(news);
+ SvUTF8_on(PL_linestr);
+ s = (U8*)SvPVX(PL_linestr);
+ PL_bufend = SvPVX(PL_linestr) + newlen;
}
#else
- Perl_croak(aTHX_ "Unsupported script encoding");
+ Perl_croak(aTHX_ "Unsupported script encoding UTF16-LE");
#endif
}
break;
case 0xFE:
- if (s[1] == 0xFF) { /* UTF-16 big-endian */
+ if (s[1] == 0xFF) { /* UTF-16 big-endian? */
#ifndef PERL_NO_UTF16_FILTER
- DEBUG_p(PerlIO_printf(Perl_debug_log, "UTF-16BE script encoding\n"));
+ if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16BE script encoding (BOM)\n");
s += 2;
+ utf16be:
if (PL_bufend > (char *)s) {
U8 *news;
I32 newlen;
filter_add(utf16_textfilter, NULL);
New(898, news, (PL_bufend - (char*)s) * 3 / 2 + 1, U8);
- PL_bufend = (char*)utf16_to_utf8(s, news,
- PL_bufend - (char*)s,
- &newlen);
- Copy(news, s, newlen, U8);
- SvCUR_set(PL_linestr, newlen);
- PL_bufend = SvPVX(PL_linestr) + newlen;
- news[newlen++] = '\0';
+ utf16_to_utf8(s, news,
+ PL_bufend - (char*)s,
+ &newlen);
+ sv_setpvn(PL_linestr, (const char*)news, newlen);
Safefree(news);
+ SvUTF8_on(PL_linestr);
+ s = (U8*)SvPVX(PL_linestr);
+ PL_bufend = SvPVX(PL_linestr) + newlen;
}
#else
- Perl_croak(aTHX_ "Unsupported script encoding");
+ Perl_croak(aTHX_ "Unsupported script encoding UTF16-BE");
#endif
}
break;
case 0xEF:
if (slen > 2 && s[1] == 0xBB && s[2] == 0xBF) {
- DEBUG_p(PerlIO_printf(Perl_debug_log, "UTF-8 script encoding\n"));
+ if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-8 script encoding (BOM)\n");
s += 3; /* UTF-8 */
}
break;
case 0:
- if (slen > 3 && s[1] == 0 && /* UTF-32 big-endian */
- s[2] == 0xFE && s[3] == 0xFF)
- {
- Perl_croak(aTHX_ "Unsupported script encoding");
+ if (slen > 3) {
+ if (s[1] == 0) {
+ if (s[2] == 0xFE && s[3] == 0xFF) {
+ /* UTF-32 big-endian */
+ Perl_croak(aTHX_ "Unsupported script encoding UTF32-BE");
+ }
+ }
+ else if (s[2] == 0 && s[3] != 0) {
+ /* Leading bytes
+ * 00 xx 00 xx
+ * are a good indicator of UTF-16BE. */
+ if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16BE script encoding (no BOM)\n");
+ goto utf16be;
+ }
}
+ default:
+ if (slen > 3 && s[1] == 0 && s[2] != 0 && s[3] == 0) {
+ /* Leading bytes
+ * xx 00 xx 00
+ * are a good indicator of UTF-16LE. */
+ if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16LE script encoding (no BOM)\n");
+ goto utf16le;
+ }
}
return (char*)s;
}
static I32
utf16_textfilter(pTHX_ int idx, SV *sv, int maxlen)
{
+ STRLEN old = SvCUR(sv);
I32 count = FILTER_READ(idx+1, sv, maxlen);
+ DEBUG_P(PerlIO_printf(Perl_debug_log,
+ "utf16_textfilter(%p): %d %d (%d)\n",
+ utf16_textfilter, idx, maxlen, count));
if (count) {
U8* tmps;
- U8* tend;
I32 newlen;
New(898, tmps, SvCUR(sv) * 3 / 2 + 1, U8);
- if (!*SvPV_nolen(sv))
- /* Game over, but don't feed an odd-length string to utf16_to_utf8 */
- return count;
-
- tend = utf16_to_utf8((U8*)SvPVX(sv), tmps, SvCUR(sv), &newlen);
- sv_usepvn(sv, (char*)tmps, tend - tmps);
+ Copy(SvPVX(sv), tmps, old, char);
+ utf16_to_utf8((U8*)SvPVX(sv) + old, tmps + old,
+ SvCUR(sv) - old, &newlen);
+ sv_usepvn(sv, (char*)tmps, (STRLEN)newlen + old);
}
- return count;
+ DEBUG_P({sv_dump(sv);});
+ return SvCUR(sv);
}
static I32
utf16rev_textfilter(pTHX_ int idx, SV *sv, int maxlen)
{
+ STRLEN old = SvCUR(sv);
I32 count = FILTER_READ(idx+1, sv, maxlen);
+ DEBUG_P(PerlIO_printf(Perl_debug_log,
+ "utf16rev_textfilter(%p): %d %d (%d)\n",
+ utf16rev_textfilter, idx, maxlen, count));
if (count) {
U8* tmps;
- U8* tend;
I32 newlen;
- if (!*SvPV_nolen(sv))
- /* Game over, but don't feed an odd-length string to utf16_to_utf8 */
- return count;
-
New(898, tmps, SvCUR(sv) * 3 / 2 + 1, U8);
- tend = utf16_to_utf8_reversed((U8*)SvPVX(sv), tmps, SvCUR(sv), &newlen);
- sv_usepvn(sv, (char*)tmps, tend - tmps);
+ Copy(SvPVX(sv), tmps, old, char);
+ utf16_to_utf8((U8*)SvPVX(sv) + old, tmps + old,
+ SvCUR(sv) - old, &newlen);
+ sv_usepvn(sv, (char*)tmps, (STRLEN)newlen + old);
}
+ DEBUG_P({ sv_dump(sv); });
return count;
}
#endif
+/*
+Returns a pointer to the next character after the parsed
+vstring, as well as updating the passed in sv.
+
+Function must be called like
+
+ sv = NEWSV(92,5);
+ s = scan_vstring(s,sv);
+
+The sv should already be large enough to store the vstring
+passed in, for performance reasons.
+
+*/
+
+char *
+Perl_scan_vstring(pTHX_ char *s, SV *sv)
+{
+ char *pos = s;
+ char *start = s;
+ if (*pos == 'v') pos++; /* get past 'v' */
+ while (pos < PL_bufend && (isDIGIT(*pos) || *pos == '_'))
+ pos++;
+ if ( *pos != '.') {
+ /* this may not be a v-string if followed by => */
+ char *next = pos;
+ while (next < PL_bufend && isSPACE(*next))
+ ++next;
+ if ((PL_bufend - next) >= 2 && *next == '=' && next[1] == '>' ) {
+ /* return string not v-string */
+ sv_setpvn(sv,(char *)s,pos-s);
+ return pos;
+ }
+ }
+
+ if (!isALPHA(*pos)) {
+ UV rev;
+ U8 tmpbuf[UTF8_MAXLEN+1];
+ U8 *tmpend;
+
+ if (*s == 'v') s++; /* get past 'v' */
+
+ sv_setpvn(sv, "", 0);
+
+ for (;;) {
+ rev = 0;
+ {
+ /* this is atoi() that tolerates underscores */
+ char *end = pos;
+ UV mult = 1;
+ while (--end >= s) {
+ UV orev;
+ if (*end == '_')
+ continue;
+ orev = rev;
+ rev += (*end - '0') * mult;
+ mult *= 10;
+ if (orev > rev && ckWARN_d(WARN_OVERFLOW))
+ Perl_warner(aTHX_ packWARN(WARN_OVERFLOW),
+ "Integer overflow in decimal number");
+ }
+ }
+#ifdef EBCDIC
+ if (rev > 0x7FFFFFFF)
+ Perl_croak(aTHX_ "In EBCDIC the v-string components cannot exceed 2147483647");
+#endif
+ /* Append native character for the rev point */
+ tmpend = uvchr_to_utf8(tmpbuf, rev);
+ sv_catpvn(sv, (const char*)tmpbuf, tmpend - tmpbuf);
+ if (!UNI_IS_INVARIANT(NATIVE_TO_UNI(rev)))
+ SvUTF8_on(sv);
+ if (pos + 1 < PL_bufend && *pos == '.' && isDIGIT(pos[1]))
+ s = ++pos;
+ else {
+ s = pos;
+ break;
+ }
+ while (pos < PL_bufend && (isDIGIT(*pos) || *pos == '_'))
+ pos++;
+ }
+ SvPOK_on(sv);
+ sv_magic(sv,NULL,PERL_MAGIC_vstring,(const char*)start, pos-start);
+ SvRMAGICAL_on(sv);
+ }
+ return s;
+}
+
https://perl5.git.perl.org / perl5.git / blobdiff