3 * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
4 * 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 by Larry Wall and others
6 * You may distribute under the terms of either the GNU General Public
7 * License or the Artistic License, as specified in the README file.
12 * 'It all comes from here, the stench and the peril.' --Frodo
14 * [p.719 of _The Lord of the Rings_, IV/ix: "Shelob's Lair"]
18 * This file is the lexer for Perl. It's closely linked to the
21 * The main routine is yylex(), which returns the next token.
25 =head1 Lexer interface
27 This is the lower layer of the Perl parser, managing characters and tokens.
29 =for apidoc AmU|yy_parser *|PL_parser
31 Pointer to a structure encapsulating the state of the parsing operation
32 currently in progress. The pointer can be locally changed to perform
33 a nested parse without interfering with the state of an outer parse.
34 Individual members of C<PL_parser> have their own documentation.
40 #define PERL_IN_TOKE_C
42 #include "dquote_static.c"
44 #define new_constant(a,b,c,d,e,f,g) \
45 S_new_constant(aTHX_ a,b,STR_WITH_LEN(c),d,e,f, g)
47 #define pl_yylval (PL_parser->yylval)
49 /* XXX temporary backwards compatibility */
50 #define PL_lex_brackets (PL_parser->lex_brackets)
51 #define PL_lex_allbrackets (PL_parser->lex_allbrackets)
52 #define PL_lex_fakeeof (PL_parser->lex_fakeeof)
53 #define PL_lex_brackstack (PL_parser->lex_brackstack)
54 #define PL_lex_casemods (PL_parser->lex_casemods)
55 #define PL_lex_casestack (PL_parser->lex_casestack)
56 #define PL_lex_defer (PL_parser->lex_defer)
57 #define PL_lex_dojoin (PL_parser->lex_dojoin)
58 #define PL_lex_expect (PL_parser->lex_expect)
59 #define PL_lex_formbrack (PL_parser->lex_formbrack)
60 #define PL_lex_inpat (PL_parser->lex_inpat)
61 #define PL_lex_inwhat (PL_parser->lex_inwhat)
62 #define PL_lex_op (PL_parser->lex_op)
63 #define PL_lex_repl (PL_parser->lex_repl)
64 #define PL_lex_starts (PL_parser->lex_starts)
65 #define PL_lex_stuff (PL_parser->lex_stuff)
66 #define PL_multi_start (PL_parser->multi_start)
67 #define PL_multi_open (PL_parser->multi_open)
68 #define PL_multi_close (PL_parser->multi_close)
69 #define PL_pending_ident (PL_parser->pending_ident)
70 #define PL_preambled (PL_parser->preambled)
71 #define PL_sublex_info (PL_parser->sublex_info)
72 #define PL_linestr (PL_parser->linestr)
73 #define PL_expect (PL_parser->expect)
74 #define PL_copline (PL_parser->copline)
75 #define PL_bufptr (PL_parser->bufptr)
76 #define PL_oldbufptr (PL_parser->oldbufptr)
77 #define PL_oldoldbufptr (PL_parser->oldoldbufptr)
78 #define PL_linestart (PL_parser->linestart)
79 #define PL_bufend (PL_parser->bufend)
80 #define PL_last_uni (PL_parser->last_uni)
81 #define PL_last_lop (PL_parser->last_lop)
82 #define PL_last_lop_op (PL_parser->last_lop_op)
83 #define PL_lex_state (PL_parser->lex_state)
84 #define PL_rsfp (PL_parser->rsfp)
85 #define PL_rsfp_filters (PL_parser->rsfp_filters)
86 #define PL_in_my (PL_parser->in_my)
87 #define PL_in_my_stash (PL_parser->in_my_stash)
88 #define PL_tokenbuf (PL_parser->tokenbuf)
89 #define PL_multi_end (PL_parser->multi_end)
90 #define PL_error_count (PL_parser->error_count)
93 # define PL_endwhite (PL_parser->endwhite)
94 # define PL_faketokens (PL_parser->faketokens)
95 # define PL_lasttoke (PL_parser->lasttoke)
96 # define PL_nextwhite (PL_parser->nextwhite)
97 # define PL_realtokenstart (PL_parser->realtokenstart)
98 # define PL_skipwhite (PL_parser->skipwhite)
99 # define PL_thisclose (PL_parser->thisclose)
100 # define PL_thismad (PL_parser->thismad)
101 # define PL_thisopen (PL_parser->thisopen)
102 # define PL_thisstuff (PL_parser->thisstuff)
103 # define PL_thistoken (PL_parser->thistoken)
104 # define PL_thiswhite (PL_parser->thiswhite)
105 # define PL_thiswhite (PL_parser->thiswhite)
106 # define PL_nexttoke (PL_parser->nexttoke)
107 # define PL_curforce (PL_parser->curforce)
109 # define PL_nexttoke (PL_parser->nexttoke)
110 # define PL_nexttype (PL_parser->nexttype)
111 # define PL_nextval (PL_parser->nextval)
114 /* This can't be done with embed.fnc, because struct yy_parser contains a
115 member named pending_ident, which clashes with the generated #define */
117 S_pending_ident(pTHX);
119 static const char ident_too_long[] = "Identifier too long";
122 # define CURMAD(slot,sv) if (PL_madskills) { curmad(slot,sv); sv = 0; }
123 # define NEXTVAL_NEXTTOKE PL_nexttoke[PL_curforce].next_val
125 # define CURMAD(slot,sv)
126 # define NEXTVAL_NEXTTOKE PL_nextval[PL_nexttoke]
129 #define XENUMMASK 0x3f
130 #define XFAKEEOF 0x40
131 #define XFAKEBRACK 0x80
133 #ifdef USE_UTF8_SCRIPTS
134 # define UTF (!IN_BYTES)
136 # define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || ( !(PL_parser->lex_flags & LEX_IGNORE_UTF8_HINTS) && (PL_hints & HINT_UTF8)))
139 /* The maximum number of characters preceding the unrecognized one to display */
140 #define UNRECOGNIZED_PRECEDE_COUNT 10
142 /* In variables named $^X, these are the legal values for X.
143 * 1999-02-27 mjd-perl-patch@plover.com */
144 #define isCONTROLVAR(x) (isUPPER(x) || strchr("[\\]^_?", (x)))
146 #define SPACE_OR_TAB(c) ((c)==' '||(c)=='\t')
148 /* LEX_* are values for PL_lex_state, the state of the lexer.
149 * They are arranged oddly so that the guard on the switch statement
150 * can get by with a single comparison (if the compiler is smart enough).
152 * These values refer to the various states within a sublex parse,
153 * i.e. within a double quotish string
156 /* #define LEX_NOTPARSING 11 is done in perl.h. */
158 #define LEX_NORMAL 10 /* normal code (ie not within "...") */
159 #define LEX_INTERPNORMAL 9 /* code within a string, eg "$foo[$x+1]" */
160 #define LEX_INTERPCASEMOD 8 /* expecting a \U, \Q or \E etc */
161 #define LEX_INTERPPUSH 7 /* starting a new sublex parse level */
162 #define LEX_INTERPSTART 6 /* expecting the start of a $var */
164 /* at end of code, eg "$x" followed by: */
165 #define LEX_INTERPEND 5 /* ... eg not one of [, { or -> */
166 #define LEX_INTERPENDMAYBE 4 /* ... eg one of [, { or -> */
168 #define LEX_INTERPCONCAT 3 /* expecting anything, eg at start of
169 string or after \E, $foo, etc */
170 #define LEX_INTERPCONST 2 /* NOT USED */
171 #define LEX_FORMLINE 1 /* expecting a format line */
172 #define LEX_KNOWNEXT 0 /* next token known; just return it */
176 static const char* const lex_state_names[] = {
195 #include "keywords.h"
197 /* CLINE is a macro that ensures PL_copline has a sane value */
202 #define CLINE (PL_copline = (CopLINE(PL_curcop) < PL_copline ? CopLINE(PL_curcop) : PL_copline))
205 # define SKIPSPACE0(s) skipspace0(s)
206 # define SKIPSPACE1(s) skipspace1(s)
207 # define SKIPSPACE2(s,tsv) skipspace2(s,&tsv)
208 # define PEEKSPACE(s) skipspace2(s,0)
210 # define SKIPSPACE0(s) skipspace(s)
211 # define SKIPSPACE1(s) skipspace(s)
212 # define SKIPSPACE2(s,tsv) skipspace(s)
213 # define PEEKSPACE(s) skipspace(s)
217 * Convenience functions to return different tokens and prime the
218 * lexer for the next token. They all take an argument.
220 * TOKEN : generic token (used for '(', DOLSHARP, etc)
221 * OPERATOR : generic operator
222 * AOPERATOR : assignment operator
223 * PREBLOCK : beginning the block after an if, while, foreach, ...
224 * PRETERMBLOCK : beginning a non-code-defining {} block (eg, hash ref)
225 * PREREF : *EXPR where EXPR is not a simple identifier
226 * TERM : expression term
227 * LOOPX : loop exiting command (goto, last, dump, etc)
228 * FTST : file test operator
229 * FUN0 : zero-argument function
230 * FUN0OP : zero-argument function, with its op created in this file
231 * FUN1 : not used, except for not, which isn't a UNIOP
232 * BOop : bitwise or or xor
234 * SHop : shift operator
235 * PWop : power operator
236 * PMop : pattern-matching operator
237 * Aop : addition-level operator
238 * Mop : multiplication-level operator
239 * Eop : equality-testing operator
240 * Rop : relational operator <= != gt
242 * Also see LOP and lop() below.
245 #ifdef DEBUGGING /* Serve -DT. */
246 # define REPORT(retval) tokereport((I32)retval, &pl_yylval)
248 # define REPORT(retval) (retval)
251 #define TOKEN(retval) return ( PL_bufptr = s, REPORT(retval))
252 #define OPERATOR(retval) return (PL_expect = XTERM, PL_bufptr = s, REPORT(retval))
253 #define AOPERATOR(retval) return ao((PL_expect = XTERM, PL_bufptr = s, REPORT(retval)))
254 #define PREBLOCK(retval) return (PL_expect = XBLOCK,PL_bufptr = s, REPORT(retval))
255 #define PRETERMBLOCK(retval) return (PL_expect = XTERMBLOCK,PL_bufptr = s, REPORT(retval))
256 #define PREREF(retval) return (PL_expect = XREF,PL_bufptr = s, REPORT(retval))
257 #define TERM(retval) return (CLINE, PL_expect = XOPERATOR, PL_bufptr = s, REPORT(retval))
258 #define LOOPX(f) return (pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)LOOPEX))
259 #define FTST(f) return (pl_yylval.ival=f, PL_expect=XTERMORDORDOR, PL_bufptr=s, REPORT((int)UNIOP))
260 #define FUN0(f) return (pl_yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC0))
261 #define FUN0OP(f) return (pl_yylval.opval=f, CLINE, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC0OP))
262 #define FUN1(f) return (pl_yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC1))
263 #define BOop(f) return ao((pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITOROP)))
264 #define BAop(f) return ao((pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITANDOP)))
265 #define SHop(f) return ao((pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)SHIFTOP)))
266 #define PWop(f) return ao((pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)POWOP)))
267 #define PMop(f) return(pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MATCHOP))
268 #define Aop(f) return ao((pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)ADDOP)))
269 #define Mop(f) return ao((pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MULOP)))
270 #define Eop(f) return (pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)EQOP))
271 #define Rop(f) return (pl_yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)RELOP))
273 /* This bit of chicanery makes a unary function followed by
274 * a parenthesis into a function with one argument, highest precedence.
275 * The UNIDOR macro is for unary functions that can be followed by the //
276 * operator (such as C<shift // 0>).
278 #define UNI2(f,x) { \
279 pl_yylval.ival = f; \
282 PL_last_uni = PL_oldbufptr; \
283 PL_last_lop_op = f; \
285 return REPORT( (int)FUNC1 ); \
287 return REPORT( *s=='(' ? (int)FUNC1 : (int)UNIOP ); \
289 #define UNI(f) UNI2(f,XTERM)
290 #define UNIDOR(f) UNI2(f,XTERMORDORDOR)
291 #define UNIPROTO(f,optional) { \
292 if (optional) PL_last_uni = PL_oldbufptr; \
296 #define UNIBRACK(f) { \
297 pl_yylval.ival = f; \
299 PL_last_uni = PL_oldbufptr; \
301 return REPORT( (int)FUNC1 ); \
303 return REPORT( (*s == '(') ? (int)FUNC1 : (int)UNIOP ); \
306 /* grandfather return to old style */
309 if (!PL_lex_allbrackets && PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC) \
310 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC; \
311 pl_yylval.ival = (f); \
319 /* how to interpret the pl_yylval associated with the token */
323 TOKENTYPE_OPNUM, /* pl_yylval.ival contains an opcode number */
329 static struct debug_tokens {
331 enum token_type type;
333 } const debug_tokens[] =
335 { ADDOP, TOKENTYPE_OPNUM, "ADDOP" },
336 { ANDAND, TOKENTYPE_NONE, "ANDAND" },
337 { ANDOP, TOKENTYPE_NONE, "ANDOP" },
338 { ANONSUB, TOKENTYPE_IVAL, "ANONSUB" },
339 { ARROW, TOKENTYPE_NONE, "ARROW" },
340 { ASSIGNOP, TOKENTYPE_OPNUM, "ASSIGNOP" },
341 { BITANDOP, TOKENTYPE_OPNUM, "BITANDOP" },
342 { BITOROP, TOKENTYPE_OPNUM, "BITOROP" },
343 { COLONATTR, TOKENTYPE_NONE, "COLONATTR" },
344 { CONTINUE, TOKENTYPE_NONE, "CONTINUE" },
345 { DEFAULT, TOKENTYPE_NONE, "DEFAULT" },
346 { DO, TOKENTYPE_NONE, "DO" },
347 { DOLSHARP, TOKENTYPE_NONE, "DOLSHARP" },
348 { DORDOR, TOKENTYPE_NONE, "DORDOR" },
349 { DOROP, TOKENTYPE_OPNUM, "DOROP" },
350 { DOTDOT, TOKENTYPE_IVAL, "DOTDOT" },
351 { ELSE, TOKENTYPE_NONE, "ELSE" },
352 { ELSIF, TOKENTYPE_IVAL, "ELSIF" },
353 { EQOP, TOKENTYPE_OPNUM, "EQOP" },
354 { FOR, TOKENTYPE_IVAL, "FOR" },
355 { FORMAT, TOKENTYPE_NONE, "FORMAT" },
356 { FUNC, TOKENTYPE_OPNUM, "FUNC" },
357 { FUNC0, TOKENTYPE_OPNUM, "FUNC0" },
358 { FUNC0OP, TOKENTYPE_OPVAL, "FUNC0OP" },
359 { FUNC0SUB, TOKENTYPE_OPVAL, "FUNC0SUB" },
360 { FUNC1, TOKENTYPE_OPNUM, "FUNC1" },
361 { FUNCMETH, TOKENTYPE_OPVAL, "FUNCMETH" },
362 { GIVEN, TOKENTYPE_IVAL, "GIVEN" },
363 { HASHBRACK, TOKENTYPE_NONE, "HASHBRACK" },
364 { IF, TOKENTYPE_IVAL, "IF" },
365 { LABEL, TOKENTYPE_OPVAL, "LABEL" },
366 { LOCAL, TOKENTYPE_IVAL, "LOCAL" },
367 { LOOPEX, TOKENTYPE_OPNUM, "LOOPEX" },
368 { LSTOP, TOKENTYPE_OPNUM, "LSTOP" },
369 { LSTOPSUB, TOKENTYPE_OPVAL, "LSTOPSUB" },
370 { MATCHOP, TOKENTYPE_OPNUM, "MATCHOP" },
371 { METHOD, TOKENTYPE_OPVAL, "METHOD" },
372 { MULOP, TOKENTYPE_OPNUM, "MULOP" },
373 { MY, TOKENTYPE_IVAL, "MY" },
374 { MYSUB, TOKENTYPE_NONE, "MYSUB" },
375 { NOAMP, TOKENTYPE_NONE, "NOAMP" },
376 { NOTOP, TOKENTYPE_NONE, "NOTOP" },
377 { OROP, TOKENTYPE_IVAL, "OROP" },
378 { OROR, TOKENTYPE_NONE, "OROR" },
379 { PACKAGE, TOKENTYPE_NONE, "PACKAGE" },
380 { PLUGEXPR, TOKENTYPE_OPVAL, "PLUGEXPR" },
381 { PLUGSTMT, TOKENTYPE_OPVAL, "PLUGSTMT" },
382 { PMFUNC, TOKENTYPE_OPVAL, "PMFUNC" },
383 { POSTDEC, TOKENTYPE_NONE, "POSTDEC" },
384 { POSTINC, TOKENTYPE_NONE, "POSTINC" },
385 { POWOP, TOKENTYPE_OPNUM, "POWOP" },
386 { PREDEC, TOKENTYPE_NONE, "PREDEC" },
387 { PREINC, TOKENTYPE_NONE, "PREINC" },
388 { PRIVATEREF, TOKENTYPE_OPVAL, "PRIVATEREF" },
389 { REFGEN, TOKENTYPE_NONE, "REFGEN" },
390 { RELOP, TOKENTYPE_OPNUM, "RELOP" },
391 { SHIFTOP, TOKENTYPE_OPNUM, "SHIFTOP" },
392 { SUB, TOKENTYPE_NONE, "SUB" },
393 { THING, TOKENTYPE_OPVAL, "THING" },
394 { UMINUS, TOKENTYPE_NONE, "UMINUS" },
395 { UNIOP, TOKENTYPE_OPNUM, "UNIOP" },
396 { UNIOPSUB, TOKENTYPE_OPVAL, "UNIOPSUB" },
397 { UNLESS, TOKENTYPE_IVAL, "UNLESS" },
398 { UNTIL, TOKENTYPE_IVAL, "UNTIL" },
399 { USE, TOKENTYPE_IVAL, "USE" },
400 { WHEN, TOKENTYPE_IVAL, "WHEN" },
401 { WHILE, TOKENTYPE_IVAL, "WHILE" },
402 { WORD, TOKENTYPE_OPVAL, "WORD" },
403 { YADAYADA, TOKENTYPE_IVAL, "YADAYADA" },
404 { 0, TOKENTYPE_NONE, NULL }
407 /* dump the returned token in rv, plus any optional arg in pl_yylval */
410 S_tokereport(pTHX_ I32 rv, const YYSTYPE* lvalp)
414 PERL_ARGS_ASSERT_TOKEREPORT;
417 const char *name = NULL;
418 enum token_type type = TOKENTYPE_NONE;
419 const struct debug_tokens *p;
420 SV* const report = newSVpvs("<== ");
422 for (p = debug_tokens; p->token; p++) {
423 if (p->token == (int)rv) {
430 Perl_sv_catpv(aTHX_ report, name);
431 else if ((char)rv > ' ' && (char)rv < '~')
432 Perl_sv_catpvf(aTHX_ report, "'%c'", (char)rv);
434 sv_catpvs(report, "EOF");
436 Perl_sv_catpvf(aTHX_ report, "?? %"IVdf, (IV)rv);
439 case TOKENTYPE_GVVAL: /* doesn't appear to be used */
442 Perl_sv_catpvf(aTHX_ report, "(ival=%"IVdf")", (IV)lvalp->ival);
444 case TOKENTYPE_OPNUM:
445 Perl_sv_catpvf(aTHX_ report, "(ival=op_%s)",
446 PL_op_name[lvalp->ival]);
449 Perl_sv_catpvf(aTHX_ report, "(pval=\"%s\")", lvalp->pval);
451 case TOKENTYPE_OPVAL:
453 Perl_sv_catpvf(aTHX_ report, "(opval=op_%s)",
454 PL_op_name[lvalp->opval->op_type]);
455 if (lvalp->opval->op_type == OP_CONST) {
456 Perl_sv_catpvf(aTHX_ report, " %s",
457 SvPEEK(cSVOPx_sv(lvalp->opval)));
462 sv_catpvs(report, "(opval=null)");
465 PerlIO_printf(Perl_debug_log, "### %s\n\n", SvPV_nolen_const(report));
471 /* print the buffer with suitable escapes */
474 S_printbuf(pTHX_ const char *const fmt, const char *const s)
476 SV* const tmp = newSVpvs("");
478 PERL_ARGS_ASSERT_PRINTBUF;
480 PerlIO_printf(Perl_debug_log, fmt, pv_display(tmp, s, strlen(s), 0, 60));
487 S_deprecate_commaless_var_list(pTHX) {
489 deprecate("comma-less variable list");
490 return REPORT(','); /* grandfather non-comma-format format */
496 * This subroutine detects &&=, ||=, and //= and turns an ANDAND, OROR or DORDOR
497 * into an OP_ANDASSIGN, OP_ORASSIGN, or OP_DORASSIGN
501 S_ao(pTHX_ int toketype)
504 if (*PL_bufptr == '=') {
506 if (toketype == ANDAND)
507 pl_yylval.ival = OP_ANDASSIGN;
508 else if (toketype == OROR)
509 pl_yylval.ival = OP_ORASSIGN;
510 else if (toketype == DORDOR)
511 pl_yylval.ival = OP_DORASSIGN;
519 * When Perl expects an operator and finds something else, no_op
520 * prints the warning. It always prints "<something> found where
521 * operator expected. It prints "Missing semicolon on previous line?"
522 * if the surprise occurs at the start of the line. "do you need to
523 * predeclare ..." is printed out for code like "sub bar; foo bar $x"
524 * where the compiler doesn't know if foo is a method call or a function.
525 * It prints "Missing operator before end of line" if there's nothing
526 * after the missing operator, or "... before <...>" if there is something
527 * after the missing operator.
531 S_no_op(pTHX_ const char *const what, char *s)
534 char * const oldbp = PL_bufptr;
535 const bool is_first = (PL_oldbufptr == PL_linestart);
537 PERL_ARGS_ASSERT_NO_OP;
543 yywarn(Perl_form(aTHX_ "%s found where operator expected", what), UTF ? SVf_UTF8 : 0);
544 if (ckWARN_d(WARN_SYNTAX)) {
546 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
547 "\t(Missing semicolon on previous line?)\n");
548 else if (PL_oldoldbufptr && isIDFIRST_lazy_if(PL_oldoldbufptr,UTF)) {
550 for (t = PL_oldoldbufptr; (isALNUM_lazy_if(t,UTF) || *t == ':');
551 t += UTF ? UTF8SKIP(t) : 1)
553 if (t < PL_bufptr && isSPACE(*t))
554 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
555 "\t(Do you need to predeclare %"SVf"?)\n",
556 SVfARG(newSVpvn_flags(PL_oldoldbufptr, (STRLEN)(t - PL_oldoldbufptr),
557 SVs_TEMP | (UTF ? SVf_UTF8 : 0))));
561 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
562 "\t(Missing operator before %"SVf"?)\n",
563 SVfARG(newSVpvn_flags(oldbp, (STRLEN)(s - oldbp),
564 SVs_TEMP | (UTF ? SVf_UTF8 : 0))));
572 * Complain about missing quote/regexp/heredoc terminator.
573 * If it's called with NULL then it cauterizes the line buffer.
574 * If we're in a delimited string and the delimiter is a control
575 * character, it's reformatted into a two-char sequence like ^C.
580 S_missingterm(pTHX_ char *s)
586 char * const nl = strrchr(s,'\n');
590 else if (isCNTRL(PL_multi_close)) {
592 tmpbuf[1] = (char)toCTRL(PL_multi_close);
597 *tmpbuf = (char)PL_multi_close;
601 q = strchr(s,'"') ? '\'' : '"';
602 Perl_croak(aTHX_ "Can't find string terminator %c%s%c anywhere before EOF",q,s,q);
608 * Check whether the named feature is enabled.
611 Perl_feature_is_enabled(pTHX_ const char *const name, STRLEN namelen)
614 char he_name[8 + MAX_FEATURE_LEN] = "feature_";
616 PERL_ARGS_ASSERT_FEATURE_IS_ENABLED;
618 assert(CURRENT_FEATURE_BUNDLE == FEATURE_BUNDLE_CUSTOM);
620 if (namelen > MAX_FEATURE_LEN)
622 memcpy(&he_name[8], name, namelen);
624 return cBOOL(cop_hints_fetch_pvn(PL_curcop, he_name, 8 + namelen, 0,
625 REFCOUNTED_HE_EXISTS));
629 * experimental text filters for win32 carriage-returns, utf16-to-utf8 and
630 * utf16-to-utf8-reversed.
633 #ifdef PERL_CR_FILTER
637 register const char *s = SvPVX_const(sv);
638 register const char * const e = s + SvCUR(sv);
640 PERL_ARGS_ASSERT_STRIP_RETURN;
642 /* outer loop optimized to do nothing if there are no CR-LFs */
644 if (*s++ == '\r' && *s == '\n') {
645 /* hit a CR-LF, need to copy the rest */
646 register char *d = s - 1;
649 if (*s == '\r' && s[1] == '\n')
660 S_cr_textfilter(pTHX_ int idx, SV *sv, int maxlen)
662 const I32 count = FILTER_READ(idx+1, sv, maxlen);
663 if (count > 0 && !maxlen)
670 =for apidoc Amx|void|lex_start|SV *line|PerlIO *rsfp|U32 flags
672 Creates and initialises a new lexer/parser state object, supplying
673 a context in which to lex and parse from a new source of Perl code.
674 A pointer to the new state object is placed in L</PL_parser>. An entry
675 is made on the save stack so that upon unwinding the new state object
676 will be destroyed and the former value of L</PL_parser> will be restored.
677 Nothing else need be done to clean up the parsing context.
679 The code to be parsed comes from I<line> and I<rsfp>. I<line>, if
680 non-null, provides a string (in SV form) containing code to be parsed.
681 A copy of the string is made, so subsequent modification of I<line>
682 does not affect parsing. I<rsfp>, if non-null, provides an input stream
683 from which code will be read to be parsed. If both are non-null, the
684 code in I<line> comes first and must consist of complete lines of input,
685 and I<rsfp> supplies the remainder of the source.
687 The I<flags> parameter is reserved for future use. Currently it is only
688 used by perl internally, so extensions should always pass zero.
693 /* LEX_START_SAME_FILTER indicates that this is not a new file, so it
694 can share filters with the current parser.
695 LEX_START_DONT_CLOSE indicates that the file handle wasn't opened by the
696 caller, hence isn't owned by the parser, so shouldn't be closed on parser
697 destruction. This is used to handle the case of defaulting to reading the
698 script from the standard input because no filename was given on the command
699 line (without getting confused by situation where STDIN has been closed, so
700 the script handle is opened on fd 0) */
703 Perl_lex_start(pTHX_ SV *line, PerlIO *rsfp, U32 flags)
706 const char *s = NULL;
707 yy_parser *parser, *oparser;
708 if (flags && flags & ~LEX_START_FLAGS)
709 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_start");
711 /* create and initialise a parser */
713 Newxz(parser, 1, yy_parser);
714 parser->old_parser = oparser = PL_parser;
717 parser->stack = NULL;
719 parser->stack_size = 0;
721 /* on scope exit, free this parser and restore any outer one */
723 parser->saved_curcop = PL_curcop;
725 /* initialise lexer state */
728 parser->curforce = -1;
730 parser->nexttoke = 0;
732 parser->error_count = oparser ? oparser->error_count : 0;
733 parser->copline = NOLINE;
734 parser->lex_state = LEX_NORMAL;
735 parser->expect = XSTATE;
737 parser->rsfp_filters =
738 !(flags & LEX_START_SAME_FILTER) || !oparser
740 : MUTABLE_AV(SvREFCNT_inc(
741 oparser->rsfp_filters
742 ? oparser->rsfp_filters
743 : (oparser->rsfp_filters = newAV())
746 Newx(parser->lex_brackstack, 120, char);
747 Newx(parser->lex_casestack, 12, char);
748 *parser->lex_casestack = '\0';
752 s = SvPV_const(line, len);
753 parser->linestr = flags & LEX_START_COPIED
754 ? SvREFCNT_inc_simple_NN(line)
755 : newSVpvn_flags(s, len, SvUTF8(line));
756 if (!len || s[len-1] != ';')
757 sv_catpvs(parser->linestr, "\n;");
759 parser->linestr = newSVpvs("\n;");
761 parser->oldoldbufptr =
764 parser->linestart = SvPVX(parser->linestr);
765 parser->bufend = parser->bufptr + SvCUR(parser->linestr);
766 parser->last_lop = parser->last_uni = NULL;
767 parser->lex_flags = flags & (LEX_IGNORE_UTF8_HINTS|LEX_EVALBYTES
768 |LEX_DONT_CLOSE_RSFP);
770 parser->in_pod = parser->filtered = 0;
774 /* delete a parser object */
777 Perl_parser_free(pTHX_ const yy_parser *parser)
779 PERL_ARGS_ASSERT_PARSER_FREE;
781 PL_curcop = parser->saved_curcop;
782 SvREFCNT_dec(parser->linestr);
784 if (PL_parser->lex_flags & LEX_DONT_CLOSE_RSFP)
785 PerlIO_clearerr(parser->rsfp);
786 else if (parser->rsfp && (!parser->old_parser ||
787 (parser->old_parser && parser->rsfp != parser->old_parser->rsfp)))
788 PerlIO_close(parser->rsfp);
789 SvREFCNT_dec(parser->rsfp_filters);
791 Safefree(parser->lex_brackstack);
792 Safefree(parser->lex_casestack);
793 PL_parser = parser->old_parser;
799 =for apidoc AmxU|SV *|PL_parser-E<gt>linestr
801 Buffer scalar containing the chunk currently under consideration of the
802 text currently being lexed. This is always a plain string scalar (for
803 which C<SvPOK> is true). It is not intended to be used as a scalar by
804 normal scalar means; instead refer to the buffer directly by the pointer
805 variables described below.
807 The lexer maintains various C<char*> pointers to things in the
808 C<PL_parser-E<gt>linestr> buffer. If C<PL_parser-E<gt>linestr> is ever
809 reallocated, all of these pointers must be updated. Don't attempt to
810 do this manually, but rather use L</lex_grow_linestr> if you need to
811 reallocate the buffer.
813 The content of the text chunk in the buffer is commonly exactly one
814 complete line of input, up to and including a newline terminator,
815 but there are situations where it is otherwise. The octets of the
816 buffer may be intended to be interpreted as either UTF-8 or Latin-1.
817 The function L</lex_bufutf8> tells you which. Do not use the C<SvUTF8>
818 flag on this scalar, which may disagree with it.
820 For direct examination of the buffer, the variable
821 L</PL_parser-E<gt>bufend> points to the end of the buffer. The current
822 lexing position is pointed to by L</PL_parser-E<gt>bufptr>. Direct use
823 of these pointers is usually preferable to examination of the scalar
824 through normal scalar means.
826 =for apidoc AmxU|char *|PL_parser-E<gt>bufend
828 Direct pointer to the end of the chunk of text currently being lexed, the
829 end of the lexer buffer. This is equal to C<SvPVX(PL_parser-E<gt>linestr)
830 + SvCUR(PL_parser-E<gt>linestr)>. A NUL character (zero octet) is
831 always located at the end of the buffer, and does not count as part of
832 the buffer's contents.
834 =for apidoc AmxU|char *|PL_parser-E<gt>bufptr
836 Points to the current position of lexing inside the lexer buffer.
837 Characters around this point may be freely examined, within
838 the range delimited by C<SvPVX(L</PL_parser-E<gt>linestr>)> and
839 L</PL_parser-E<gt>bufend>. The octets of the buffer may be intended to be
840 interpreted as either UTF-8 or Latin-1, as indicated by L</lex_bufutf8>.
842 Lexing code (whether in the Perl core or not) moves this pointer past
843 the characters that it consumes. It is also expected to perform some
844 bookkeeping whenever a newline character is consumed. This movement
845 can be more conveniently performed by the function L</lex_read_to>,
846 which handles newlines appropriately.
848 Interpretation of the buffer's octets can be abstracted out by
849 using the slightly higher-level functions L</lex_peek_unichar> and
850 L</lex_read_unichar>.
852 =for apidoc AmxU|char *|PL_parser-E<gt>linestart
854 Points to the start of the current line inside the lexer buffer.
855 This is useful for indicating at which column an error occurred, and
856 not much else. This must be updated by any lexing code that consumes
857 a newline; the function L</lex_read_to> handles this detail.
863 =for apidoc Amx|bool|lex_bufutf8
865 Indicates whether the octets in the lexer buffer
866 (L</PL_parser-E<gt>linestr>) should be interpreted as the UTF-8 encoding
867 of Unicode characters. If not, they should be interpreted as Latin-1
868 characters. This is analogous to the C<SvUTF8> flag for scalars.
870 In UTF-8 mode, it is not guaranteed that the lexer buffer actually
871 contains valid UTF-8. Lexing code must be robust in the face of invalid
874 The actual C<SvUTF8> flag of the L</PL_parser-E<gt>linestr> scalar
875 is significant, but not the whole story regarding the input character
876 encoding. Normally, when a file is being read, the scalar contains octets
877 and its C<SvUTF8> flag is off, but the octets should be interpreted as
878 UTF-8 if the C<use utf8> pragma is in effect. During a string eval,
879 however, the scalar may have the C<SvUTF8> flag on, and in this case its
880 octets should be interpreted as UTF-8 unless the C<use bytes> pragma
881 is in effect. This logic may change in the future; use this function
882 instead of implementing the logic yourself.
888 Perl_lex_bufutf8(pTHX)
894 =for apidoc Amx|char *|lex_grow_linestr|STRLEN len
896 Reallocates the lexer buffer (L</PL_parser-E<gt>linestr>) to accommodate
897 at least I<len> octets (including terminating NUL). Returns a
898 pointer to the reallocated buffer. This is necessary before making
899 any direct modification of the buffer that would increase its length.
900 L</lex_stuff_pvn> provides a more convenient way to insert text into
903 Do not use C<SvGROW> or C<sv_grow> directly on C<PL_parser-E<gt>linestr>;
904 this function updates all of the lexer's variables that point directly
911 Perl_lex_grow_linestr(pTHX_ STRLEN len)
915 STRLEN bufend_pos, bufptr_pos, oldbufptr_pos, oldoldbufptr_pos;
916 STRLEN linestart_pos, last_uni_pos, last_lop_pos;
917 linestr = PL_parser->linestr;
918 buf = SvPVX(linestr);
919 if (len <= SvLEN(linestr))
921 bufend_pos = PL_parser->bufend - buf;
922 bufptr_pos = PL_parser->bufptr - buf;
923 oldbufptr_pos = PL_parser->oldbufptr - buf;
924 oldoldbufptr_pos = PL_parser->oldoldbufptr - buf;
925 linestart_pos = PL_parser->linestart - buf;
926 last_uni_pos = PL_parser->last_uni ? PL_parser->last_uni - buf : 0;
927 last_lop_pos = PL_parser->last_lop ? PL_parser->last_lop - buf : 0;
928 buf = sv_grow(linestr, len);
929 PL_parser->bufend = buf + bufend_pos;
930 PL_parser->bufptr = buf + bufptr_pos;
931 PL_parser->oldbufptr = buf + oldbufptr_pos;
932 PL_parser->oldoldbufptr = buf + oldoldbufptr_pos;
933 PL_parser->linestart = buf + linestart_pos;
934 if (PL_parser->last_uni)
935 PL_parser->last_uni = buf + last_uni_pos;
936 if (PL_parser->last_lop)
937 PL_parser->last_lop = buf + last_lop_pos;
942 =for apidoc Amx|void|lex_stuff_pvn|const char *pv|STRLEN len|U32 flags
944 Insert characters into the lexer buffer (L</PL_parser-E<gt>linestr>),
945 immediately after the current lexing point (L</PL_parser-E<gt>bufptr>),
946 reallocating the buffer if necessary. This means that lexing code that
947 runs later will see the characters as if they had appeared in the input.
948 It is not recommended to do this as part of normal parsing, and most
949 uses of this facility run the risk of the inserted characters being
950 interpreted in an unintended manner.
952 The string to be inserted is represented by I<len> octets starting
953 at I<pv>. These octets are interpreted as either UTF-8 or Latin-1,
954 according to whether the C<LEX_STUFF_UTF8> flag is set in I<flags>.
955 The characters are recoded for the lexer buffer, according to how the
956 buffer is currently being interpreted (L</lex_bufutf8>). If a string
957 to be inserted is available as a Perl scalar, the L</lex_stuff_sv>
958 function is more convenient.
964 Perl_lex_stuff_pvn(pTHX_ const char *pv, STRLEN len, U32 flags)
968 PERL_ARGS_ASSERT_LEX_STUFF_PVN;
969 if (flags & ~(LEX_STUFF_UTF8))
970 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_stuff_pvn");
972 if (flags & LEX_STUFF_UTF8) {
976 const char *p, *e = pv+len;
977 for (p = pv; p != e; p++)
978 highhalf += !!(((U8)*p) & 0x80);
981 lex_grow_linestr(SvCUR(PL_parser->linestr)+1+len+highhalf);
982 bufptr = PL_parser->bufptr;
983 Move(bufptr, bufptr+len+highhalf, PL_parser->bufend+1-bufptr, char);
984 SvCUR_set(PL_parser->linestr,
985 SvCUR(PL_parser->linestr) + len+highhalf);
986 PL_parser->bufend += len+highhalf;
987 for (p = pv; p != e; p++) {
990 *bufptr++ = (char)(0xc0 | (c >> 6));
991 *bufptr++ = (char)(0x80 | (c & 0x3f));
998 if (flags & LEX_STUFF_UTF8) {
1000 const char *p, *e = pv+len;
1001 for (p = pv; p != e; p++) {
1004 Perl_croak(aTHX_ "Lexing code attempted to stuff "
1005 "non-Latin-1 character into Latin-1 input");
1006 } else if (c >= 0xc2 && p+1 != e &&
1007 (((U8)p[1]) & 0xc0) == 0x80) {
1010 } else if (c >= 0x80) {
1011 /* malformed UTF-8 */
1013 SAVESPTR(PL_warnhook);
1014 PL_warnhook = PERL_WARNHOOK_FATAL;
1015 utf8n_to_uvuni((U8*)p, e-p, NULL, 0);
1021 lex_grow_linestr(SvCUR(PL_parser->linestr)+1+len-highhalf);
1022 bufptr = PL_parser->bufptr;
1023 Move(bufptr, bufptr+len-highhalf, PL_parser->bufend+1-bufptr, char);
1024 SvCUR_set(PL_parser->linestr,
1025 SvCUR(PL_parser->linestr) + len-highhalf);
1026 PL_parser->bufend += len-highhalf;
1027 for (p = pv; p != e; p++) {
1030 *bufptr++ = (char)(((c & 0x3) << 6) | (p[1] & 0x3f));
1033 *bufptr++ = (char)c;
1038 lex_grow_linestr(SvCUR(PL_parser->linestr)+1+len);
1039 bufptr = PL_parser->bufptr;
1040 Move(bufptr, bufptr+len, PL_parser->bufend+1-bufptr, char);
1041 SvCUR_set(PL_parser->linestr, SvCUR(PL_parser->linestr) + len);
1042 PL_parser->bufend += len;
1043 Copy(pv, bufptr, len, char);
1049 =for apidoc Amx|void|lex_stuff_pv|const char *pv|U32 flags
1051 Insert characters into the lexer buffer (L</PL_parser-E<gt>linestr>),
1052 immediately after the current lexing point (L</PL_parser-E<gt>bufptr>),
1053 reallocating the buffer if necessary. This means that lexing code that
1054 runs later will see the characters as if they had appeared in the input.
1055 It is not recommended to do this as part of normal parsing, and most
1056 uses of this facility run the risk of the inserted characters being
1057 interpreted in an unintended manner.
1059 The string to be inserted is represented by octets starting at I<pv>
1060 and continuing to the first nul. These octets are interpreted as either
1061 UTF-8 or Latin-1, according to whether the C<LEX_STUFF_UTF8> flag is set
1062 in I<flags>. The characters are recoded for the lexer buffer, according
1063 to how the buffer is currently being interpreted (L</lex_bufutf8>).
1064 If it is not convenient to nul-terminate a string to be inserted, the
1065 L</lex_stuff_pvn> function is more appropriate.
1071 Perl_lex_stuff_pv(pTHX_ const char *pv, U32 flags)
1073 PERL_ARGS_ASSERT_LEX_STUFF_PV;
1074 lex_stuff_pvn(pv, strlen(pv), flags);
1078 =for apidoc Amx|void|lex_stuff_sv|SV *sv|U32 flags
1080 Insert characters into the lexer buffer (L</PL_parser-E<gt>linestr>),
1081 immediately after the current lexing point (L</PL_parser-E<gt>bufptr>),
1082 reallocating the buffer if necessary. This means that lexing code that
1083 runs later will see the characters as if they had appeared in the input.
1084 It is not recommended to do this as part of normal parsing, and most
1085 uses of this facility run the risk of the inserted characters being
1086 interpreted in an unintended manner.
1088 The string to be inserted is the string value of I<sv>. The characters
1089 are recoded for the lexer buffer, according to how the buffer is currently
1090 being interpreted (L</lex_bufutf8>). If a string to be inserted is
1091 not already a Perl scalar, the L</lex_stuff_pvn> function avoids the
1092 need to construct a scalar.
1098 Perl_lex_stuff_sv(pTHX_ SV *sv, U32 flags)
1102 PERL_ARGS_ASSERT_LEX_STUFF_SV;
1104 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_stuff_sv");
1106 lex_stuff_pvn(pv, len, flags | (SvUTF8(sv) ? LEX_STUFF_UTF8 : 0));
1110 =for apidoc Amx|void|lex_unstuff|char *ptr
1112 Discards text about to be lexed, from L</PL_parser-E<gt>bufptr> up to
1113 I<ptr>. Text following I<ptr> will be moved, and the buffer shortened.
1114 This hides the discarded text from any lexing code that runs later,
1115 as if the text had never appeared.
1117 This is not the normal way to consume lexed text. For that, use
1124 Perl_lex_unstuff(pTHX_ char *ptr)
1128 PERL_ARGS_ASSERT_LEX_UNSTUFF;
1129 buf = PL_parser->bufptr;
1131 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_unstuff");
1134 bufend = PL_parser->bufend;
1136 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_unstuff");
1137 unstuff_len = ptr - buf;
1138 Move(ptr, buf, bufend+1-ptr, char);
1139 SvCUR_set(PL_parser->linestr, SvCUR(PL_parser->linestr) - unstuff_len);
1140 PL_parser->bufend = bufend - unstuff_len;
1144 =for apidoc Amx|void|lex_read_to|char *ptr
1146 Consume text in the lexer buffer, from L</PL_parser-E<gt>bufptr> up
1147 to I<ptr>. This advances L</PL_parser-E<gt>bufptr> to match I<ptr>,
1148 performing the correct bookkeeping whenever a newline character is passed.
1149 This is the normal way to consume lexed text.
1151 Interpretation of the buffer's octets can be abstracted out by
1152 using the slightly higher-level functions L</lex_peek_unichar> and
1153 L</lex_read_unichar>.
1159 Perl_lex_read_to(pTHX_ char *ptr)
1162 PERL_ARGS_ASSERT_LEX_READ_TO;
1163 s = PL_parser->bufptr;
1164 if (ptr < s || ptr > PL_parser->bufend)
1165 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_read_to");
1166 for (; s != ptr; s++)
1168 CopLINE_inc(PL_curcop);
1169 PL_parser->linestart = s+1;
1171 PL_parser->bufptr = ptr;
1175 =for apidoc Amx|void|lex_discard_to|char *ptr
1177 Discards the first part of the L</PL_parser-E<gt>linestr> buffer,
1178 up to I<ptr>. The remaining content of the buffer will be moved, and
1179 all pointers into the buffer updated appropriately. I<ptr> must not
1180 be later in the buffer than the position of L</PL_parser-E<gt>bufptr>:
1181 it is not permitted to discard text that has yet to be lexed.
1183 Normally it is not necessarily to do this directly, because it suffices to
1184 use the implicit discarding behaviour of L</lex_next_chunk> and things
1185 based on it. However, if a token stretches across multiple lines,
1186 and the lexing code has kept multiple lines of text in the buffer for
1187 that purpose, then after completion of the token it would be wise to
1188 explicitly discard the now-unneeded earlier lines, to avoid future
1189 multi-line tokens growing the buffer without bound.
1195 Perl_lex_discard_to(pTHX_ char *ptr)
1199 PERL_ARGS_ASSERT_LEX_DISCARD_TO;
1200 buf = SvPVX(PL_parser->linestr);
1202 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_discard_to");
1205 if (ptr > PL_parser->bufptr)
1206 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_discard_to");
1207 discard_len = ptr - buf;
1208 if (PL_parser->oldbufptr < ptr)
1209 PL_parser->oldbufptr = ptr;
1210 if (PL_parser->oldoldbufptr < ptr)
1211 PL_parser->oldoldbufptr = ptr;
1212 if (PL_parser->last_uni && PL_parser->last_uni < ptr)
1213 PL_parser->last_uni = NULL;
1214 if (PL_parser->last_lop && PL_parser->last_lop < ptr)
1215 PL_parser->last_lop = NULL;
1216 Move(ptr, buf, PL_parser->bufend+1-ptr, char);
1217 SvCUR_set(PL_parser->linestr, SvCUR(PL_parser->linestr) - discard_len);
1218 PL_parser->bufend -= discard_len;
1219 PL_parser->bufptr -= discard_len;
1220 PL_parser->oldbufptr -= discard_len;
1221 PL_parser->oldoldbufptr -= discard_len;
1222 if (PL_parser->last_uni)
1223 PL_parser->last_uni -= discard_len;
1224 if (PL_parser->last_lop)
1225 PL_parser->last_lop -= discard_len;
1229 =for apidoc Amx|bool|lex_next_chunk|U32 flags
1231 Reads in the next chunk of text to be lexed, appending it to
1232 L</PL_parser-E<gt>linestr>. This should be called when lexing code has
1233 looked to the end of the current chunk and wants to know more. It is
1234 usual, but not necessary, for lexing to have consumed the entirety of
1235 the current chunk at this time.
1237 If L</PL_parser-E<gt>bufptr> is pointing to the very end of the current
1238 chunk (i.e., the current chunk has been entirely consumed), normally the
1239 current chunk will be discarded at the same time that the new chunk is
1240 read in. If I<flags> includes C<LEX_KEEP_PREVIOUS>, the current chunk
1241 will not be discarded. If the current chunk has not been entirely
1242 consumed, then it will not be discarded regardless of the flag.
1244 Returns true if some new text was added to the buffer, or false if the
1245 buffer has reached the end of the input text.
1250 #define LEX_FAKE_EOF 0x80000000
1253 Perl_lex_next_chunk(pTHX_ U32 flags)
1257 STRLEN old_bufend_pos, new_bufend_pos;
1258 STRLEN bufptr_pos, oldbufptr_pos, oldoldbufptr_pos;
1259 STRLEN linestart_pos, last_uni_pos, last_lop_pos;
1260 bool got_some_for_debugger = 0;
1262 if (flags & ~(LEX_KEEP_PREVIOUS|LEX_FAKE_EOF))
1263 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_next_chunk");
1264 linestr = PL_parser->linestr;
1265 buf = SvPVX(linestr);
1266 if (!(flags & LEX_KEEP_PREVIOUS) &&
1267 PL_parser->bufptr == PL_parser->bufend) {
1268 old_bufend_pos = bufptr_pos = oldbufptr_pos = oldoldbufptr_pos = 0;
1270 if (PL_parser->last_uni != PL_parser->bufend)
1271 PL_parser->last_uni = NULL;
1272 if (PL_parser->last_lop != PL_parser->bufend)
1273 PL_parser->last_lop = NULL;
1274 last_uni_pos = last_lop_pos = 0;
1278 old_bufend_pos = PL_parser->bufend - buf;
1279 bufptr_pos = PL_parser->bufptr - buf;
1280 oldbufptr_pos = PL_parser->oldbufptr - buf;
1281 oldoldbufptr_pos = PL_parser->oldoldbufptr - buf;
1282 linestart_pos = PL_parser->linestart - buf;
1283 last_uni_pos = PL_parser->last_uni ? PL_parser->last_uni - buf : 0;
1284 last_lop_pos = PL_parser->last_lop ? PL_parser->last_lop - buf : 0;
1286 if (flags & LEX_FAKE_EOF) {
1288 } else if (!PL_parser->rsfp && !PL_parser->filtered) {
1290 } else if (filter_gets(linestr, old_bufend_pos)) {
1292 got_some_for_debugger = 1;
1294 if (!SvPOK(linestr)) /* can get undefined by filter_gets */
1295 sv_setpvs(linestr, "");
1297 /* End of real input. Close filehandle (unless it was STDIN),
1298 * then add implicit termination.
1300 if (PL_parser->lex_flags & LEX_DONT_CLOSE_RSFP)
1301 PerlIO_clearerr(PL_parser->rsfp);
1302 else if (PL_parser->rsfp)
1303 (void)PerlIO_close(PL_parser->rsfp);
1304 PL_parser->rsfp = NULL;
1305 PL_parser->in_pod = PL_parser->filtered = 0;
1307 if (PL_madskills && !PL_in_eval && (PL_minus_p || PL_minus_n))
1310 if (!PL_in_eval && PL_minus_p) {
1312 /*{*/";}continue{print or die qq(-p destination: $!\\n);}");
1313 PL_minus_n = PL_minus_p = 0;
1314 } else if (!PL_in_eval && PL_minus_n) {
1315 sv_catpvs(linestr, /*{*/";}");
1318 sv_catpvs(linestr, ";");
1321 buf = SvPVX(linestr);
1322 new_bufend_pos = SvCUR(linestr);
1323 PL_parser->bufend = buf + new_bufend_pos;
1324 PL_parser->bufptr = buf + bufptr_pos;
1325 PL_parser->oldbufptr = buf + oldbufptr_pos;
1326 PL_parser->oldoldbufptr = buf + oldoldbufptr_pos;
1327 PL_parser->linestart = buf + linestart_pos;
1328 if (PL_parser->last_uni)
1329 PL_parser->last_uni = buf + last_uni_pos;
1330 if (PL_parser->last_lop)
1331 PL_parser->last_lop = buf + last_lop_pos;
1332 if (got_some_for_debugger && (PERLDB_LINE || PERLDB_SAVESRC) &&
1333 PL_curstash != PL_debstash) {
1334 /* debugger active and we're not compiling the debugger code,
1335 * so store the line into the debugger's array of lines
1337 update_debugger_info(NULL, buf+old_bufend_pos,
1338 new_bufend_pos-old_bufend_pos);
1344 =for apidoc Amx|I32|lex_peek_unichar|U32 flags
1346 Looks ahead one (Unicode) character in the text currently being lexed.
1347 Returns the codepoint (unsigned integer value) of the next character,
1348 or -1 if lexing has reached the end of the input text. To consume the
1349 peeked character, use L</lex_read_unichar>.
1351 If the next character is in (or extends into) the next chunk of input
1352 text, the next chunk will be read in. Normally the current chunk will be
1353 discarded at the same time, but if I<flags> includes C<LEX_KEEP_PREVIOUS>
1354 then the current chunk will not be discarded.
1356 If the input is being interpreted as UTF-8 and a UTF-8 encoding error
1357 is encountered, an exception is generated.
1363 Perl_lex_peek_unichar(pTHX_ U32 flags)
1367 if (flags & ~(LEX_KEEP_PREVIOUS))
1368 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_peek_unichar");
1369 s = PL_parser->bufptr;
1370 bufend = PL_parser->bufend;
1376 if (!lex_next_chunk(flags))
1378 s = PL_parser->bufptr;
1379 bufend = PL_parser->bufend;
1385 len = PL_utf8skip[head];
1386 while ((STRLEN)(bufend-s) < len) {
1387 if (!lex_next_chunk(flags | LEX_KEEP_PREVIOUS))
1389 s = PL_parser->bufptr;
1390 bufend = PL_parser->bufend;
1393 unichar = utf8n_to_uvuni((U8*)s, bufend-s, &retlen, UTF8_CHECK_ONLY);
1394 if (retlen == (STRLEN)-1) {
1395 /* malformed UTF-8 */
1397 SAVESPTR(PL_warnhook);
1398 PL_warnhook = PERL_WARNHOOK_FATAL;
1399 utf8n_to_uvuni((U8*)s, bufend-s, NULL, 0);
1405 if (!lex_next_chunk(flags))
1407 s = PL_parser->bufptr;
1414 =for apidoc Amx|I32|lex_read_unichar|U32 flags
1416 Reads the next (Unicode) character in the text currently being lexed.
1417 Returns the codepoint (unsigned integer value) of the character read,
1418 and moves L</PL_parser-E<gt>bufptr> past the character, or returns -1
1419 if lexing has reached the end of the input text. To non-destructively
1420 examine the next character, use L</lex_peek_unichar> instead.
1422 If the next character is in (or extends into) the next chunk of input
1423 text, the next chunk will be read in. Normally the current chunk will be
1424 discarded at the same time, but if I<flags> includes C<LEX_KEEP_PREVIOUS>
1425 then the current chunk will not be discarded.
1427 If the input is being interpreted as UTF-8 and a UTF-8 encoding error
1428 is encountered, an exception is generated.
1434 Perl_lex_read_unichar(pTHX_ U32 flags)
1437 if (flags & ~(LEX_KEEP_PREVIOUS))
1438 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_read_unichar");
1439 c = lex_peek_unichar(flags);
1442 CopLINE_inc(PL_curcop);
1444 PL_parser->bufptr += UTF8SKIP(PL_parser->bufptr);
1446 ++(PL_parser->bufptr);
1452 =for apidoc Amx|void|lex_read_space|U32 flags
1454 Reads optional spaces, in Perl style, in the text currently being
1455 lexed. The spaces may include ordinary whitespace characters and
1456 Perl-style comments. C<#line> directives are processed if encountered.
1457 L</PL_parser-E<gt>bufptr> is moved past the spaces, so that it points
1458 at a non-space character (or the end of the input text).
1460 If spaces extend into the next chunk of input text, the next chunk will
1461 be read in. Normally the current chunk will be discarded at the same
1462 time, but if I<flags> includes C<LEX_KEEP_PREVIOUS> then the current
1463 chunk will not be discarded.
1468 #define LEX_NO_NEXT_CHUNK 0x80000000
1471 Perl_lex_read_space(pTHX_ U32 flags)
1474 bool need_incline = 0;
1475 if (flags & ~(LEX_KEEP_PREVIOUS|LEX_NO_NEXT_CHUNK))
1476 Perl_croak(aTHX_ "Lexing code internal error (%s)", "lex_read_space");
1479 sv_free(PL_skipwhite);
1480 PL_skipwhite = NULL;
1483 PL_skipwhite = newSVpvs("");
1484 #endif /* PERL_MAD */
1485 s = PL_parser->bufptr;
1486 bufend = PL_parser->bufend;
1492 } while (!(c == '\n' || (c == 0 && s == bufend)));
1493 } else if (c == '\n') {
1495 PL_parser->linestart = s;
1500 } else if (isSPACE(c)) {
1502 } else if (c == 0 && s == bufend) {
1506 sv_catpvn(PL_skipwhite, PL_parser->bufptr, s-PL_parser->bufptr);
1507 #endif /* PERL_MAD */
1508 if (flags & LEX_NO_NEXT_CHUNK)
1510 PL_parser->bufptr = s;
1511 CopLINE_inc(PL_curcop);
1512 got_more = lex_next_chunk(flags);
1513 CopLINE_dec(PL_curcop);
1514 s = PL_parser->bufptr;
1515 bufend = PL_parser->bufend;
1518 if (need_incline && PL_parser->rsfp) {
1528 sv_catpvn(PL_skipwhite, PL_parser->bufptr, s-PL_parser->bufptr);
1529 #endif /* PERL_MAD */
1530 PL_parser->bufptr = s;
1535 * This subroutine has nothing to do with tilting, whether at windmills
1536 * or pinball tables. Its name is short for "increment line". It
1537 * increments the current line number in CopLINE(PL_curcop) and checks
1538 * to see whether the line starts with a comment of the form
1539 * # line 500 "foo.pm"
1540 * If so, it sets the current line number and file to the values in the comment.
1544 S_incline(pTHX_ const char *s)
1552 PERL_ARGS_ASSERT_INCLINE;
1554 CopLINE_inc(PL_curcop);
1557 while (SPACE_OR_TAB(*s))
1559 if (strnEQ(s, "line", 4))
1563 if (SPACE_OR_TAB(*s))
1567 while (SPACE_OR_TAB(*s))
1575 if (!SPACE_OR_TAB(*s) && *s != '\r' && *s != '\n' && *s != '\0')
1577 while (SPACE_OR_TAB(*s))
1579 if (*s == '"' && (t = strchr(s+1, '"'))) {
1585 while (!isSPACE(*t))
1589 while (SPACE_OR_TAB(*e) || *e == '\r' || *e == '\f')
1591 if (*e != '\n' && *e != '\0')
1592 return; /* false alarm */
1594 line_num = atoi(n)-1;
1597 const STRLEN len = t - s;
1598 SV *const temp_sv = CopFILESV(PL_curcop);
1603 cf = SvPVX(temp_sv);
1604 tmplen = SvCUR(temp_sv);
1610 if (!PL_rsfp && !PL_parser->filtered) {
1611 /* must copy *{"::_<(eval N)[oldfilename:L]"}
1612 * to *{"::_<newfilename"} */
1613 /* However, the long form of evals is only turned on by the
1614 debugger - usually they're "(eval %lu)" */
1618 STRLEN tmplen2 = len;
1619 if (tmplen + 2 <= sizeof smallbuf)
1622 Newx(tmpbuf, tmplen + 2, char);
1625 memcpy(tmpbuf + 2, cf, tmplen);
1627 gvp = (GV**)hv_fetch(PL_defstash, tmpbuf, tmplen, FALSE);
1632 if (tmplen2 + 2 <= sizeof smallbuf)
1635 Newx(tmpbuf2, tmplen2 + 2, char);
1637 if (tmpbuf2 != smallbuf || tmpbuf != smallbuf) {
1638 /* Either they malloc'd it, or we malloc'd it,
1639 so no prefix is present in ours. */
1644 memcpy(tmpbuf2 + 2, s, tmplen2);
1647 gv2 = *(GV**)hv_fetch(PL_defstash, tmpbuf2, tmplen2, TRUE);
1649 gv_init(gv2, PL_defstash, tmpbuf2, tmplen2, FALSE);
1650 /* adjust ${"::_<newfilename"} to store the new file name */
1651 GvSV(gv2) = newSVpvn(tmpbuf2 + 2, tmplen2 - 2);
1652 /* The line number may differ. If that is the case,
1653 alias the saved lines that are in the array.
1654 Otherwise alias the whole array. */
1655 if (CopLINE(PL_curcop) == line_num) {
1656 GvHV(gv2) = MUTABLE_HV(SvREFCNT_inc(GvHV(*gvp)));
1657 GvAV(gv2) = MUTABLE_AV(SvREFCNT_inc(GvAV(*gvp)));
1659 else if (GvAV(*gvp)) {
1660 AV * const av = GvAV(*gvp);
1661 const I32 start = CopLINE(PL_curcop)+1;
1662 I32 items = AvFILLp(av) - start;
1664 AV * const av2 = GvAVn(gv2);
1665 SV **svp = AvARRAY(av) + start;
1666 I32 l = (I32)line_num+1;
1668 av_store(av2, l++, SvREFCNT_inc(*svp++));
1673 if (tmpbuf2 != smallbuf) Safefree(tmpbuf2);
1675 if (tmpbuf != smallbuf) Safefree(tmpbuf);
1677 CopFILE_free(PL_curcop);
1678 CopFILE_setn(PL_curcop, s, len);
1680 CopLINE_set(PL_curcop, line_num);
1684 /* skip space before PL_thistoken */
1687 S_skipspace0(pTHX_ register char *s)
1689 PERL_ARGS_ASSERT_SKIPSPACE0;
1696 PL_thiswhite = newSVpvs("");
1697 sv_catsv(PL_thiswhite, PL_skipwhite);
1698 sv_free(PL_skipwhite);
1701 PL_realtokenstart = s - SvPVX(PL_linestr);
1705 /* skip space after PL_thistoken */
1708 S_skipspace1(pTHX_ register char *s)
1710 const char *start = s;
1711 I32 startoff = start - SvPVX(PL_linestr);
1713 PERL_ARGS_ASSERT_SKIPSPACE1;
1718 start = SvPVX(PL_linestr) + startoff;
1719 if (!PL_thistoken && PL_realtokenstart >= 0) {
1720 const char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
1721 PL_thistoken = newSVpvn(tstart, start - tstart);
1723 PL_realtokenstart = -1;
1726 PL_nextwhite = newSVpvs("");
1727 sv_catsv(PL_nextwhite, PL_skipwhite);
1728 sv_free(PL_skipwhite);
1735 S_skipspace2(pTHX_ register char *s, SV **svp)
1738 const I32 bufptroff = PL_bufptr - SvPVX(PL_linestr);
1739 const I32 startoff = s - SvPVX(PL_linestr);
1741 PERL_ARGS_ASSERT_SKIPSPACE2;
1744 PL_bufptr = SvPVX(PL_linestr) + bufptroff;
1745 if (!PL_madskills || !svp)
1747 start = SvPVX(PL_linestr) + startoff;
1748 if (!PL_thistoken && PL_realtokenstart >= 0) {
1749 char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
1750 PL_thistoken = newSVpvn(tstart, start - tstart);
1751 PL_realtokenstart = -1;
1755 *svp = newSVpvs("");
1756 sv_setsv(*svp, PL_skipwhite);
1757 sv_free(PL_skipwhite);
1766 S_update_debugger_info(pTHX_ SV *orig_sv, const char *const buf, STRLEN len)
1768 AV *av = CopFILEAVx(PL_curcop);
1770 SV * const sv = newSV_type(SVt_PVMG);
1772 sv_setsv(sv, orig_sv);
1774 sv_setpvn(sv, buf, len);
1777 av_store(av, (I32)CopLINE(PL_curcop), sv);
1783 * Called to gobble the appropriate amount and type of whitespace.
1784 * Skips comments as well.
1788 S_skipspace(pTHX_ register char *s)
1792 #endif /* PERL_MAD */
1793 PERL_ARGS_ASSERT_SKIPSPACE;
1796 sv_free(PL_skipwhite);
1797 PL_skipwhite = NULL;
1799 #endif /* PERL_MAD */
1800 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
1801 while (s < PL_bufend && SPACE_OR_TAB(*s))
1804 STRLEN bufptr_pos = PL_bufptr - SvPVX(PL_linestr);
1806 lex_read_space(LEX_KEEP_PREVIOUS |
1807 (PL_sublex_info.sub_inwhat || PL_lex_state == LEX_FORMLINE ?
1808 LEX_NO_NEXT_CHUNK : 0));
1810 PL_bufptr = SvPVX(PL_linestr) + bufptr_pos;
1811 if (PL_linestart > PL_bufptr)
1812 PL_bufptr = PL_linestart;
1817 PL_skipwhite = newSVpvn(start, s-start);
1818 #endif /* PERL_MAD */
1824 * Check the unary operators to ensure there's no ambiguity in how they're
1825 * used. An ambiguous piece of code would be:
1827 * This doesn't mean rand() + 5. Because rand() is a unary operator,
1828 * the +5 is its argument.
1838 if (PL_oldoldbufptr != PL_last_uni)
1840 while (isSPACE(*PL_last_uni))
1843 while (isALNUM_lazy_if(s,UTF) || *s == '-')
1845 if ((t = strchr(s, '(')) && t < PL_bufptr)
1848 Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS),
1849 "Warning: Use of \"%.*s\" without parentheses is ambiguous",
1850 (int)(s - PL_last_uni), PL_last_uni);
1854 * LOP : macro to build a list operator. Its behaviour has been replaced
1855 * with a subroutine, S_lop() for which LOP is just another name.
1858 #define LOP(f,x) return lop(f,x,s)
1862 * Build a list operator (or something that might be one). The rules:
1863 * - if we have a next token, then it's a list operator [why?]
1864 * - if the next thing is an opening paren, then it's a function
1865 * - else it's a list operator
1869 S_lop(pTHX_ I32 f, int x, char *s)
1873 PERL_ARGS_ASSERT_LOP;
1879 PL_last_lop = PL_oldbufptr;
1880 PL_last_lop_op = (OPCODE)f;
1889 return REPORT(FUNC);
1892 return REPORT(FUNC);
1895 if (!PL_lex_allbrackets && PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
1896 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
1897 return REPORT(LSTOP);
1904 * Sets up for an eventual force_next(). start_force(0) basically does
1905 * an unshift, while start_force(-1) does a push. yylex removes items
1910 S_start_force(pTHX_ int where)
1914 if (where < 0) /* so people can duplicate start_force(PL_curforce) */
1915 where = PL_lasttoke;
1916 assert(PL_curforce < 0 || PL_curforce == where);
1917 if (PL_curforce != where) {
1918 for (i = PL_lasttoke; i > where; --i) {
1919 PL_nexttoke[i] = PL_nexttoke[i-1];
1923 if (PL_curforce < 0) /* in case of duplicate start_force() */
1924 Zero(&PL_nexttoke[where], 1, NEXTTOKE);
1925 PL_curforce = where;
1928 curmad('^', newSVpvs(""));
1929 CURMAD('_', PL_nextwhite);
1934 S_curmad(pTHX_ char slot, SV *sv)
1940 if (PL_curforce < 0)
1941 where = &PL_thismad;
1943 where = &PL_nexttoke[PL_curforce].next_mad;
1949 if (UTF && is_utf8_string((U8*)SvPVX(sv), SvCUR(sv)))
1951 else if (PL_encoding) {
1952 sv_recode_to_utf8(sv, PL_encoding);
1957 /* keep a slot open for the head of the list? */
1958 if (slot != '_' && *where && (*where)->mad_key == '^') {
1959 (*where)->mad_key = slot;
1960 sv_free(MUTABLE_SV(((*where)->mad_val)));
1961 (*where)->mad_val = (void*)sv;
1964 addmad(newMADsv(slot, sv), where, 0);
1967 # define start_force(where) NOOP
1968 # define curmad(slot, sv) NOOP
1973 * When the lexer realizes it knows the next token (for instance,
1974 * it is reordering tokens for the parser) then it can call S_force_next
1975 * to know what token to return the next time the lexer is called. Caller
1976 * will need to set PL_nextval[] (or PL_nexttoke[].next_val with PERL_MAD),
1977 * and possibly PL_expect to ensure the lexer handles the token correctly.
1981 S_force_next(pTHX_ I32 type)
1986 PerlIO_printf(Perl_debug_log, "### forced token:\n");
1987 tokereport(type, &NEXTVAL_NEXTTOKE);
1991 if (PL_curforce < 0)
1992 start_force(PL_lasttoke);
1993 PL_nexttoke[PL_curforce].next_type = type;
1994 if (PL_lex_state != LEX_KNOWNEXT)
1995 PL_lex_defer = PL_lex_state;
1996 PL_lex_state = LEX_KNOWNEXT;
1997 PL_lex_expect = PL_expect;
2000 PL_nexttype[PL_nexttoke] = type;
2002 if (PL_lex_state != LEX_KNOWNEXT) {
2003 PL_lex_defer = PL_lex_state;
2004 PL_lex_expect = PL_expect;
2005 PL_lex_state = LEX_KNOWNEXT;
2013 int yyc = PL_parser->yychar;
2014 if (yyc != YYEMPTY) {
2017 NEXTVAL_NEXTTOKE = PL_parser->yylval;
2018 if (yyc == '{'/*}*/ || yyc == HASHBRACK || yyc == '['/*]*/) {
2019 PL_lex_allbrackets--;
2021 yyc |= (3<<24) | (PL_lex_brackstack[PL_lex_brackets] << 16);
2022 } else if (yyc == '('/*)*/) {
2023 PL_lex_allbrackets--;
2028 PL_parser->yychar = YYEMPTY;
2033 S_newSV_maybe_utf8(pTHX_ const char *const start, STRLEN len)
2036 SV * const sv = newSVpvn_utf8(start, len,
2039 && !is_ascii_string((const U8*)start, len)
2040 && is_utf8_string((const U8*)start, len));
2046 * When the lexer knows the next thing is a word (for instance, it has
2047 * just seen -> and it knows that the next char is a word char, then
2048 * it calls S_force_word to stick the next word into the PL_nexttoke/val
2052 * char *start : buffer position (must be within PL_linestr)
2053 * int token : PL_next* will be this type of bare word (e.g., METHOD,WORD)
2054 * int check_keyword : if true, Perl checks to make sure the word isn't
2055 * a keyword (do this if the word is a label, e.g. goto FOO)
2056 * int allow_pack : if true, : characters will also be allowed (require,
2057 * use, etc. do this)
2058 * int allow_initial_tick : used by the "sub" lexer only.
2062 S_force_word(pTHX_ register char *start, int token, int check_keyword, int allow_pack, int allow_initial_tick)
2068 PERL_ARGS_ASSERT_FORCE_WORD;
2070 start = SKIPSPACE1(start);
2072 if (isIDFIRST_lazy_if(s,UTF) ||
2073 (allow_pack && *s == ':') ||
2074 (allow_initial_tick && *s == '\'') )
2076 s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, allow_pack, &len);
2077 if (check_keyword && keyword(PL_tokenbuf, len, 0))
2079 start_force(PL_curforce);
2081 curmad('X', newSVpvn(start,s-start));
2082 if (token == METHOD) {
2087 PL_expect = XOPERATOR;
2091 curmad('g', newSVpvs( "forced" ));
2092 NEXTVAL_NEXTTOKE.opval
2093 = (OP*)newSVOP(OP_CONST,0,
2094 S_newSV_maybe_utf8(aTHX_ PL_tokenbuf, len));
2095 NEXTVAL_NEXTTOKE.opval->op_private |= OPpCONST_BARE;
2103 * Called when the lexer wants $foo *foo &foo etc, but the program
2104 * text only contains the "foo" portion. The first argument is a pointer
2105 * to the "foo", and the second argument is the type symbol to prefix.
2106 * Forces the next token to be a "WORD".
2107 * Creates the symbol if it didn't already exist (via gv_fetchpv()).
2111 S_force_ident(pTHX_ register const char *s, int kind)
2115 PERL_ARGS_ASSERT_FORCE_IDENT;
2118 const STRLEN len = strlen(s);
2119 OP* const o = (OP*)newSVOP(OP_CONST, 0, newSVpvn_flags(s, len,
2120 UTF ? SVf_UTF8 : 0));
2121 start_force(PL_curforce);
2122 NEXTVAL_NEXTTOKE.opval = o;
2125 o->op_private = OPpCONST_ENTERED;
2126 /* XXX see note in pp_entereval() for why we forgo typo
2127 warnings if the symbol must be introduced in an eval.
2129 gv_fetchpvn_flags(s, len,
2130 (PL_in_eval ? (GV_ADDMULTI | GV_ADDINEVAL)
2131 : GV_ADD) | ( UTF ? SVf_UTF8 : 0 ),
2132 kind == '$' ? SVt_PV :
2133 kind == '@' ? SVt_PVAV :
2134 kind == '%' ? SVt_PVHV :
2142 Perl_str_to_version(pTHX_ SV *sv)
2147 const char *start = SvPV_const(sv,len);
2148 const char * const end = start + len;
2149 const bool utf = SvUTF8(sv) ? TRUE : FALSE;
2151 PERL_ARGS_ASSERT_STR_TO_VERSION;
2153 while (start < end) {
2157 n = utf8n_to_uvchr((U8*)start, len, &skip, 0);
2162 retval += ((NV)n)/nshift;
2171 * Forces the next token to be a version number.
2172 * If the next token appears to be an invalid version number, (e.g. "v2b"),
2173 * and if "guessing" is TRUE, then no new token is created (and the caller
2174 * must use an alternative parsing method).
2178 S_force_version(pTHX_ char *s, int guessing)
2184 I32 startoff = s - SvPVX(PL_linestr);
2187 PERL_ARGS_ASSERT_FORCE_VERSION;
2195 while (isDIGIT(*d) || *d == '_' || *d == '.')
2199 start_force(PL_curforce);
2200 curmad('X', newSVpvn(s,d-s));
2203 if (*d == ';' || isSPACE(*d) || *d == '{' || *d == '}' || !*d) {
2205 #ifdef USE_LOCALE_NUMERIC
2206 char *loc = savepv(setlocale(LC_NUMERIC, NULL));
2207 setlocale(LC_NUMERIC, "C");
2209 s = scan_num(s, &pl_yylval);
2210 #ifdef USE_LOCALE_NUMERIC
2211 setlocale(LC_NUMERIC, loc);
2214 version = pl_yylval.opval;
2215 ver = cSVOPx(version)->op_sv;
2216 if (SvPOK(ver) && !SvNIOK(ver)) {
2217 SvUPGRADE(ver, SVt_PVNV);
2218 SvNV_set(ver, str_to_version(ver));
2219 SvNOK_on(ver); /* hint that it is a version */
2222 else if (guessing) {
2225 sv_free(PL_nextwhite); /* let next token collect whitespace */
2227 s = SvPVX(PL_linestr) + startoff;
2235 if (PL_madskills && !version) {
2236 sv_free(PL_nextwhite); /* let next token collect whitespace */
2238 s = SvPVX(PL_linestr) + startoff;
2241 /* NOTE: The parser sees the package name and the VERSION swapped */
2242 start_force(PL_curforce);
2243 NEXTVAL_NEXTTOKE.opval = version;
2250 * S_force_strict_version
2251 * Forces the next token to be a version number using strict syntax rules.
2255 S_force_strict_version(pTHX_ char *s)
2260 I32 startoff = s - SvPVX(PL_linestr);
2262 const char *errstr = NULL;
2264 PERL_ARGS_ASSERT_FORCE_STRICT_VERSION;
2266 while (isSPACE(*s)) /* leading whitespace */
2269 if (is_STRICT_VERSION(s,&errstr)) {
2271 s = (char *)scan_version(s, ver, 0);
2272 version = newSVOP(OP_CONST, 0, ver);
2274 else if ( (*s != ';' && *s != '{' && *s != '}' ) &&
2275 (s = SKIPSPACE1(s), (*s != ';' && *s != '{' && *s != '}' )))
2279 yyerror(errstr); /* version required */
2284 if (PL_madskills && !version) {
2285 sv_free(PL_nextwhite); /* let next token collect whitespace */
2287 s = SvPVX(PL_linestr) + startoff;
2290 /* NOTE: The parser sees the package name and the VERSION swapped */
2291 start_force(PL_curforce);
2292 NEXTVAL_NEXTTOKE.opval = version;
2300 * Tokenize a quoted string passed in as an SV. It finds the next
2301 * chunk, up to end of string or a backslash. It may make a new
2302 * SV containing that chunk (if HINT_NEW_STRING is on). It also
2307 S_tokeq(pTHX_ SV *sv)
2311 register char *send;
2316 PERL_ARGS_ASSERT_TOKEQ;
2321 s = SvPV_force(sv, len);
2322 if (SvTYPE(sv) >= SVt_PVIV && SvIVX(sv) == -1)
2325 /* This is relying on the SV being "well formed" with a trailing '\0' */
2326 while (s < send && !(*s == '\\' && s[1] == '\\'))
2331 if ( PL_hints & HINT_NEW_STRING ) {
2332 pv = newSVpvn_flags(SvPVX_const(pv), len, SVs_TEMP | SvUTF8(sv));
2336 if (s + 1 < send && (s[1] == '\\'))
2337 s++; /* all that, just for this */
2342 SvCUR_set(sv, d - SvPVX_const(sv));
2344 if ( PL_hints & HINT_NEW_STRING )
2345 return new_constant(NULL, 0, "q", sv, pv, "q", 1);
2350 * Now come three functions related to double-quote context,
2351 * S_sublex_start, S_sublex_push, and S_sublex_done. They're used when
2352 * converting things like "\u\Lgnat" into ucfirst(lc("gnat")). They
2353 * interact with PL_lex_state, and create fake ( ... ) argument lists
2354 * to handle functions and concatenation.
2358 * stringify ( const[foo] concat lcfirst ( const[bar] ) )
2363 * Assumes that pl_yylval.ival is the op we're creating (e.g. OP_LCFIRST).
2365 * Pattern matching will set PL_lex_op to the pattern-matching op to
2366 * make (we return THING if pl_yylval.ival is OP_NULL, PMFUNC otherwise).
2368 * OP_CONST and OP_READLINE are easy--just make the new op and return.
2370 * Everything else becomes a FUNC.
2372 * Sets PL_lex_state to LEX_INTERPPUSH unless (ival was OP_NULL or we
2373 * had an OP_CONST or OP_READLINE). This just sets us up for a
2374 * call to S_sublex_push().
2378 S_sublex_start(pTHX)
2381 register const I32 op_type = pl_yylval.ival;
2383 if (op_type == OP_NULL) {
2384 pl_yylval.opval = PL_lex_op;
2388 if (op_type == OP_CONST || op_type == OP_READLINE) {
2389 SV *sv = tokeq(PL_lex_stuff);
2391 if (SvTYPE(sv) == SVt_PVIV) {
2392 /* Overloaded constants, nothing fancy: Convert to SVt_PV: */
2394 const char * const p = SvPV_const(sv, len);
2395 SV * const nsv = newSVpvn_flags(p, len, SvUTF8(sv));
2399 pl_yylval.opval = (OP*)newSVOP(op_type, 0, sv);
2400 PL_lex_stuff = NULL;
2401 /* Allow <FH> // "foo" */
2402 if (op_type == OP_READLINE)
2403 PL_expect = XTERMORDORDOR;
2406 else if (op_type == OP_BACKTICK && PL_lex_op) {
2407 /* readpipe() vas overriden */
2408 cSVOPx(cLISTOPx(cUNOPx(PL_lex_op)->op_first)->op_first->op_sibling)->op_sv = tokeq(PL_lex_stuff);
2409 pl_yylval.opval = PL_lex_op;
2411 PL_lex_stuff = NULL;
2415 PL_sublex_info.super_state = PL_lex_state;
2416 PL_sublex_info.sub_inwhat = (U16)op_type;
2417 PL_sublex_info.sub_op = PL_lex_op;
2418 PL_lex_state = LEX_INTERPPUSH;
2422 pl_yylval.opval = PL_lex_op;
2432 * Create a new scope to save the lexing state. The scope will be
2433 * ended in S_sublex_done. Returns a '(', starting the function arguments
2434 * to the uc, lc, etc. found before.
2435 * Sets PL_lex_state to LEX_INTERPCONCAT.
2444 PL_lex_state = PL_sublex_info.super_state;
2445 SAVEBOOL(PL_lex_dojoin);
2446 SAVEI32(PL_lex_brackets);
2447 SAVEI32(PL_lex_allbrackets);
2448 SAVEI8(PL_lex_fakeeof);
2449 SAVEI32(PL_lex_casemods);
2450 SAVEI32(PL_lex_starts);
2451 SAVEI8(PL_lex_state);
2452 SAVEPPTR(PL_sublex_info.re_eval_start);
2453 SAVEVPTR(PL_lex_inpat);
2454 SAVEI16(PL_lex_inwhat);
2455 SAVECOPLINE(PL_curcop);
2456 SAVEPPTR(PL_bufptr);
2457 SAVEPPTR(PL_bufend);
2458 SAVEPPTR(PL_oldbufptr);
2459 SAVEPPTR(PL_oldoldbufptr);
2460 SAVEPPTR(PL_last_lop);
2461 SAVEPPTR(PL_last_uni);
2462 SAVEPPTR(PL_linestart);
2463 SAVESPTR(PL_linestr);
2464 SAVEGENERICPV(PL_lex_brackstack);
2465 SAVEGENERICPV(PL_lex_casestack);
2467 PL_linestr = PL_lex_stuff;
2468 PL_lex_stuff = NULL;
2469 PL_sublex_info.re_eval_start = NULL;
2471 PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart
2472 = SvPVX(PL_linestr);
2473 PL_bufend += SvCUR(PL_linestr);
2474 PL_last_lop = PL_last_uni = NULL;
2475 SAVEFREESV(PL_linestr);
2477 PL_lex_dojoin = FALSE;
2478 PL_lex_brackets = 0;
2479 PL_lex_allbrackets = 0;
2480 PL_lex_fakeeof = LEX_FAKEEOF_NEVER;
2481 Newx(PL_lex_brackstack, 120, char);
2482 Newx(PL_lex_casestack, 12, char);
2483 PL_lex_casemods = 0;
2484 *PL_lex_casestack = '\0';
2486 PL_lex_state = LEX_INTERPCONCAT;
2487 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
2489 PL_lex_inwhat = PL_sublex_info.sub_inwhat;
2490 if (PL_lex_inwhat == OP_TRANSR) PL_lex_inwhat = OP_TRANS;
2491 if (PL_lex_inwhat == OP_MATCH || PL_lex_inwhat == OP_QR || PL_lex_inwhat == OP_SUBST)
2492 PL_lex_inpat = PL_sublex_info.sub_op;
2494 PL_lex_inpat = NULL;
2501 * Restores lexer state after a S_sublex_push.
2508 if (!PL_lex_starts++) {
2509 SV * const sv = newSVpvs("");
2510 if (SvUTF8(PL_linestr))
2512 PL_expect = XOPERATOR;
2513 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
2517 if (PL_lex_casemods) { /* oops, we've got some unbalanced parens */
2518 PL_lex_state = LEX_INTERPCASEMOD;
2522 /* Is there a right-hand side to take care of? (s//RHS/ or tr//RHS/) */
2523 assert(PL_lex_inwhat != OP_TRANSR);
2524 if (PL_lex_repl && (PL_lex_inwhat == OP_SUBST || PL_lex_inwhat == OP_TRANS)) {
2525 PL_linestr = PL_lex_repl;
2527 PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart = SvPVX(PL_linestr);
2528 PL_bufend += SvCUR(PL_linestr);
2529 PL_last_lop = PL_last_uni = NULL;
2530 SAVEFREESV(PL_linestr);
2531 PL_lex_dojoin = FALSE;
2532 PL_lex_brackets = 0;
2533 PL_lex_allbrackets = 0;
2534 PL_lex_fakeeof = LEX_FAKEEOF_NEVER;
2535 PL_lex_casemods = 0;
2536 *PL_lex_casestack = '\0';
2538 if (SvEVALED(PL_lex_repl)) {
2539 PL_lex_state = LEX_INTERPNORMAL;
2541 /* we don't clear PL_lex_repl here, so that we can check later
2542 whether this is an evalled subst; that means we rely on the
2543 logic to ensure sublex_done() is called again only via the
2544 branch (in yylex()) that clears PL_lex_repl, else we'll loop */
2547 PL_lex_state = LEX_INTERPCONCAT;
2557 PL_endwhite = newSVpvs("");
2558 sv_catsv(PL_endwhite, PL_thiswhite);
2562 sv_setpvs(PL_thistoken,"");
2564 PL_realtokenstart = -1;
2568 PL_bufend = SvPVX(PL_linestr);
2569 PL_bufend += SvCUR(PL_linestr);
2570 PL_expect = XOPERATOR;
2571 PL_sublex_info.sub_inwhat = 0;
2579 Extracts the next constant part of a pattern, double-quoted string,
2580 or transliteration. This is terrifying code.
2582 For example, in parsing the double-quoted string "ab\x63$d", it would
2583 stop at the '$' and return an OP_CONST containing 'abc'.
2585 It looks at PL_lex_inwhat and PL_lex_inpat to find out whether it's
2586 processing a pattern (PL_lex_inpat is true), a transliteration
2587 (PL_lex_inwhat == OP_TRANS is true), or a double-quoted string.
2589 Returns a pointer to the character scanned up to. If this is
2590 advanced from the start pointer supplied (i.e. if anything was
2591 successfully parsed), will leave an OP_CONST for the substring scanned
2592 in pl_yylval. Caller must intuit reason for not parsing further
2593 by looking at the next characters herself.
2597 \N{ABC} => \N{U+41.42.43}
2600 all other \-char, including \N and \N{ apart from \N{ABC}
2603 @ and $ where it appears to be a var, but not for $ as tail anchor
2608 In transliterations:
2609 characters are VERY literal, except for - not at the start or end
2610 of the string, which indicates a range. If the range is in bytes,
2611 scan_const expands the range to the full set of intermediate
2612 characters. If the range is in utf8, the hyphen is replaced with
2613 a certain range mark which will be handled by pmtrans() in op.c.
2615 In double-quoted strings:
2617 double-quoted style: \r and \n
2618 constants: \x31, etc.
2619 deprecated backrefs: \1 (in substitution replacements)
2620 case and quoting: \U \Q \E
2623 scan_const does *not* construct ops to handle interpolated strings.
2624 It stops processing as soon as it finds an embedded $ or @ variable
2625 and leaves it to the caller to work out what's going on.
2627 embedded arrays (whether in pattern or not) could be:
2628 @foo, @::foo, @'foo, @{foo}, @$foo, @+, @-.
2630 $ in double-quoted strings must be the symbol of an embedded scalar.
2632 $ in pattern could be $foo or could be tail anchor. Assumption:
2633 it's a tail anchor if $ is the last thing in the string, or if it's
2634 followed by one of "()| \r\n\t"
2636 \1 (backreferences) are turned into $1 in substitutions
2638 The structure of the code is
2639 while (there's a character to process) {
2640 handle transliteration ranges
2641 skip regexp comments /(?#comment)/ and codes /(?{code})/
2642 skip #-initiated comments in //x patterns
2643 check for embedded arrays
2644 check for embedded scalars
2646 deprecate \1 in substitution replacements
2647 handle string-changing backslashes \l \U \Q \E, etc.
2648 switch (what was escaped) {
2649 handle \- in a transliteration (becomes a literal -)
2650 if a pattern and not \N{, go treat as regular character
2651 handle \132 (octal characters)
2652 handle \x15 and \x{1234} (hex characters)
2653 handle \N{name} (named characters, also \N{3,5} in a pattern)
2654 handle \cV (control characters)
2655 handle printf-style backslashes (\f, \r, \n, etc)
2658 } (end if backslash)
2659 handle regular character
2660 } (end while character to read)
2665 S_scan_const(pTHX_ char *start)
2668 register char *send = PL_bufend; /* end of the constant */
2669 SV *sv = newSV(send - start); /* sv for the constant. See
2670 note below on sizing. */
2671 register char *s = start; /* start of the constant */
2672 register char *d = SvPVX(sv); /* destination for copies */
2673 bool dorange = FALSE; /* are we in a translit range? */
2674 bool didrange = FALSE; /* did we just finish a range? */
2675 bool in_charclass = FALSE; /* within /[...]/ */
2676 bool has_utf8 = FALSE; /* Output constant is UTF8 */
2677 bool this_utf8 = cBOOL(UTF); /* Is the source string assumed
2678 to be UTF8? But, this can
2679 show as true when the source
2680 isn't utf8, as for example
2681 when it is entirely composed
2684 /* Note on sizing: The scanned constant is placed into sv, which is
2685 * initialized by newSV() assuming one byte of output for every byte of
2686 * input. This routine expects newSV() to allocate an extra byte for a
2687 * trailing NUL, which this routine will append if it gets to the end of
2688 * the input. There may be more bytes of input than output (eg., \N{LATIN
2689 * CAPITAL LETTER A}), or more output than input if the constant ends up
2690 * recoded to utf8, but each time a construct is found that might increase
2691 * the needed size, SvGROW() is called. Its size parameter each time is
2692 * based on the best guess estimate at the time, namely the length used so
2693 * far, plus the length the current construct will occupy, plus room for
2694 * the trailing NUL, plus one byte for every input byte still unscanned */
2698 UV literal_endpoint = 0;
2699 bool native_range = TRUE; /* turned to FALSE if the first endpoint is Unicode. */
2702 PERL_ARGS_ASSERT_SCAN_CONST;
2704 assert(PL_lex_inwhat != OP_TRANSR);
2705 if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
2706 /* If we are doing a trans and we know we want UTF8 set expectation */
2707 has_utf8 = PL_sublex_info.sub_op->op_private & (OPpTRANS_FROM_UTF|OPpTRANS_TO_UTF);
2708 this_utf8 = PL_sublex_info.sub_op->op_private & (PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
2712 while (s < send || dorange) {
2714 /* get transliterations out of the way (they're most literal) */
2715 if (PL_lex_inwhat == OP_TRANS) {
2716 /* expand a range A-Z to the full set of characters. AIE! */
2718 I32 i; /* current expanded character */
2719 I32 min; /* first character in range */
2720 I32 max; /* last character in range */
2731 char * const c = (char*)utf8_hop((U8*)d, -1);
2735 *c = (char)UTF_TO_NATIVE(0xff);
2736 /* mark the range as done, and continue */
2742 i = d - SvPVX_const(sv); /* remember current offset */
2745 SvLEN(sv) + (has_utf8 ?
2746 (512 - UTF_CONTINUATION_MARK +
2749 /* How many two-byte within 0..255: 128 in UTF-8,
2750 * 96 in UTF-8-mod. */
2752 SvGROW(sv, SvLEN(sv) + 256); /* never more than 256 chars in a range */
2754 d = SvPVX(sv) + i; /* refresh d after realloc */
2758 for (j = 0; j <= 1; j++) {
2759 char * const c = (char*)utf8_hop((U8*)d, -1);
2760 const UV uv = utf8n_to_uvchr((U8*)c, d - c, NULL, 0);
2766 max = (U8)0xff; /* only to \xff */
2767 uvmax = uv; /* \x{100} to uvmax */
2769 d = c; /* eat endpoint chars */
2774 d -= 2; /* eat the first char and the - */
2775 min = (U8)*d; /* first char in range */
2776 max = (U8)d[1]; /* last char in range */
2783 "Invalid range \"%c-%c\" in transliteration operator",
2784 (char)min, (char)max);
2788 if (literal_endpoint == 2 &&
2789 ((isLOWER(min) && isLOWER(max)) ||
2790 (isUPPER(min) && isUPPER(max)))) {
2792 for (i = min; i <= max; i++)
2794 *d++ = NATIVE_TO_NEED(has_utf8,i);
2796 for (i = min; i <= max; i++)
2798 *d++ = NATIVE_TO_NEED(has_utf8,i);
2803 for (i = min; i <= max; i++)
2806 const U8 ch = (U8)NATIVE_TO_UTF(i);
2807 if (UNI_IS_INVARIANT(ch))
2810 *d++ = (U8)UTF8_EIGHT_BIT_HI(ch);
2811 *d++ = (U8)UTF8_EIGHT_BIT_LO(ch);
2820 d = (char*)uvchr_to_utf8((U8*)d, 0x100);
2822 *d++ = (char)UTF_TO_NATIVE(0xff);
2824 d = (char*)uvchr_to_utf8((U8*)d, uvmax);
2828 /* mark the range as done, and continue */
2832 literal_endpoint = 0;
2837 /* range begins (ignore - as first or last char) */
2838 else if (*s == '-' && s+1 < send && s != start) {
2840 Perl_croak(aTHX_ "Ambiguous range in transliteration operator");
2847 *d++ = (char)UTF_TO_NATIVE(0xff); /* use illegal utf8 byte--see pmtrans */
2857 literal_endpoint = 0;
2858 native_range = TRUE;
2863 /* if we get here, we're not doing a transliteration */
2865 else if (*s == '[' && PL_lex_inpat && !in_charclass) {
2868 while (s1 >= start && *s1-- == '\\')
2871 in_charclass = TRUE;
2874 else if (*s == ']' && PL_lex_inpat && in_charclass) {
2877 while (s1 >= start && *s1-- == '\\')
2880 in_charclass = FALSE;
2883 /* skip for regexp comments /(?#comment)/, except for the last
2884 * char, which will be done separately.
2885 * Stop on (?{..}) and friends */
2887 else if (*s == '(' && PL_lex_inpat && s[1] == '?') {
2889 while (s+1 < send && *s != ')')
2890 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2892 else if (!PL_lex_casemods && !in_charclass &&
2893 ( s[2] == '{' /* This should match regcomp.c */
2894 || (s[2] == '?' && s[3] == '{')))
2900 /* likewise skip #-initiated comments in //x patterns */
2901 else if (*s == '#' && PL_lex_inpat &&
2902 ((PMOP*)PL_lex_inpat)->op_pmflags & RXf_PMf_EXTENDED) {
2903 while (s+1 < send && *s != '\n')
2904 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2907 /* no further processing of single-quoted regex */
2908 else if (PL_lex_inpat && SvIVX(PL_linestr) == '\'')
2909 goto default_action;
2911 /* check for embedded arrays
2912 (@foo, @::foo, @'foo, @{foo}, @$foo, @+, @-)
2914 else if (*s == '@' && s[1]) {
2915 if (isALNUM_lazy_if(s+1,UTF))
2917 if (strchr(":'{$", s[1]))
2919 if (!PL_lex_inpat && (s[1] == '+' || s[1] == '-'))
2920 break; /* in regexp, neither @+ nor @- are interpolated */
2923 /* check for embedded scalars. only stop if we're sure it's a
2926 else if (*s == '$') {
2927 if (!PL_lex_inpat) /* not a regexp, so $ must be var */
2929 if (s + 1 < send && !strchr("()| \r\n\t", s[1])) {
2931 Perl_ck_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
2932 "Possible unintended interpolation of $\\ in regex");
2934 break; /* in regexp, $ might be tail anchor */
2938 /* End of else if chain - OP_TRANS rejoin rest */
2941 if (*s == '\\' && s+1 < send) {
2942 char* e; /* Can be used for ending '}', etc. */
2946 /* warn on \1 - \9 in substitution replacements, but note that \11
2947 * is an octal; and \19 is \1 followed by '9' */
2948 if (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat &&
2949 isDIGIT(*s) && *s != '0' && !isDIGIT(s[1]))
2951 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX), "\\%c better written as $%c", *s, *s);
2956 /* string-change backslash escapes */
2957 if (PL_lex_inwhat != OP_TRANS && *s && strchr("lLuUEQF", *s)) {
2961 /* In a pattern, process \N, but skip any other backslash escapes.
2962 * This is because we don't want to translate an escape sequence
2963 * into a meta symbol and have the regex compiler use the meta
2964 * symbol meaning, e.g. \x{2E} would be confused with a dot. But
2965 * in spite of this, we do have to process \N here while the proper
2966 * charnames handler is in scope. See bugs #56444 and #62056.
2967 * There is a complication because \N in a pattern may also stand
2968 * for 'match a non-nl', and not mean a charname, in which case its
2969 * processing should be deferred to the regex compiler. To be a
2970 * charname it must be followed immediately by a '{', and not look
2971 * like \N followed by a curly quantifier, i.e., not something like
2972 * \N{3,}. regcurly returns a boolean indicating if it is a legal
2974 else if (PL_lex_inpat
2977 || regcurly(s + 1)))
2979 *d++ = NATIVE_TO_NEED(has_utf8,'\\');
2980 goto default_action;
2985 /* quoted - in transliterations */
2987 if (PL_lex_inwhat == OP_TRANS) {
2995 Perl_ck_warner(aTHX_ packWARN(WARN_MISC),
2996 "Unrecognized escape \\%c passed through",
2998 /* default action is to copy the quoted character */
2999 goto default_action;
3002 /* eg. \132 indicates the octal constant 0132 */
3003 case '0': case '1': case '2': case '3':
3004 case '4': case '5': case '6': case '7':
3008 uv = NATIVE_TO_UNI(grok_oct(s, &len, &flags, NULL));
3011 goto NUM_ESCAPE_INSERT;
3013 /* eg. \o{24} indicates the octal constant \024 */
3019 bool valid = grok_bslash_o(s, &uv, &len, &error, 1);
3025 goto NUM_ESCAPE_INSERT;
3028 /* eg. \x24 indicates the hex constant 0x24 */
3034 bool valid = grok_bslash_x(s, &uv, &len, &error, 1);
3043 /* Insert oct or hex escaped character. There will always be
3044 * enough room in sv since such escapes will be longer than any
3045 * UTF-8 sequence they can end up as, except if they force us
3046 * to recode the rest of the string into utf8 */
3048 /* Here uv is the ordinal of the next character being added in
3049 * unicode (converted from native). */
3050 if (!UNI_IS_INVARIANT(uv)) {
3051 if (!has_utf8 && uv > 255) {
3052 /* Might need to recode whatever we have accumulated so
3053 * far if it contains any chars variant in utf8 or
3056 SvCUR_set(sv, d - SvPVX_const(sv));
3059 /* See Note on sizing above. */
3060 sv_utf8_upgrade_flags_grow(sv,
3061 SV_GMAGIC|SV_FORCE_UTF8_UPGRADE,
3062 UNISKIP(uv) + (STRLEN)(send - s) + 1);
3063 d = SvPVX(sv) + SvCUR(sv);
3068 d = (char*)uvuni_to_utf8((U8*)d, uv);
3069 if (PL_lex_inwhat == OP_TRANS &&
3070 PL_sublex_info.sub_op) {
3071 PL_sublex_info.sub_op->op_private |=
3072 (PL_lex_repl ? OPpTRANS_FROM_UTF
3076 if (uv > 255 && !dorange)
3077 native_range = FALSE;
3090 /* In a non-pattern \N must be a named character, like \N{LATIN
3091 * SMALL LETTER A} or \N{U+0041}. For patterns, it also can
3092 * mean to match a non-newline. For non-patterns, named
3093 * characters are converted to their string equivalents. In
3094 * patterns, named characters are not converted to their
3095 * ultimate forms for the same reasons that other escapes
3096 * aren't. Instead, they are converted to the \N{U+...} form
3097 * to get the value from the charnames that is in effect right
3098 * now, while preserving the fact that it was a named character
3099 * so that the regex compiler knows this */
3101 /* This section of code doesn't generally use the
3102 * NATIVE_TO_NEED() macro to transform the input. I (khw) did
3103 * a close examination of this macro and determined it is a
3104 * no-op except on utfebcdic variant characters. Every
3105 * character generated by this that would normally need to be
3106 * enclosed by this macro is invariant, so the macro is not
3107 * needed, and would complicate use of copy(). XXX There are
3108 * other parts of this file where the macro is used
3109 * inconsistently, but are saved by it being a no-op */
3111 /* The structure of this section of code (besides checking for
3112 * errors and upgrading to utf8) is:
3113 * Further disambiguate between the two meanings of \N, and if
3114 * not a charname, go process it elsewhere
3115 * If of form \N{U+...}, pass it through if a pattern;
3116 * otherwise convert to utf8
3117 * Otherwise must be \N{NAME}: convert to \N{U+c1.c2...} if a
3118 * pattern; otherwise convert to utf8 */
3120 /* Here, s points to the 'N'; the test below is guaranteed to
3121 * succeed if we are being called on a pattern as we already
3122 * know from a test above that the next character is a '{'.
3123 * On a non-pattern \N must mean 'named sequence, which
3124 * requires braces */
3127 yyerror("Missing braces on \\N{}");
3132 /* If there is no matching '}', it is an error. */
3133 if (! (e = strchr(s, '}'))) {
3134 if (! PL_lex_inpat) {
3135 yyerror("Missing right brace on \\N{}");
3137 yyerror("Missing right brace on \\N{} or unescaped left brace after \\N.");
3142 /* Here it looks like a named character */
3146 /* XXX This block is temporary code. \N{} implies that the
3147 * pattern is to have Unicode semantics, and therefore
3148 * currently has to be encoded in utf8. By putting it in
3149 * utf8 now, we save a whole pass in the regular expression
3150 * compiler. Once that code is changed so Unicode
3151 * semantics doesn't necessarily have to be in utf8, this
3152 * block should be removed. However, the code that parses
3153 * the output of this would have to be changed to not
3154 * necessarily expect utf8 */
3156 SvCUR_set(sv, d - SvPVX_const(sv));
3159 /* See Note on sizing above. */
3160 sv_utf8_upgrade_flags_grow(sv,
3161 SV_GMAGIC|SV_FORCE_UTF8_UPGRADE,
3162 /* 5 = '\N{' + cur char + NUL */
3163 (STRLEN)(send - s) + 5);
3164 d = SvPVX(sv) + SvCUR(sv);
3169 if (*s == 'U' && s[1] == '+') { /* \N{U+...} */
3170 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
3171 | PERL_SCAN_DISALLOW_PREFIX;
3174 /* For \N{U+...}, the '...' is a unicode value even on
3175 * EBCDIC machines */
3176 s += 2; /* Skip to next char after the 'U+' */
3178 uv = grok_hex(s, &len, &flags, NULL);
3179 if (len == 0 || len != (STRLEN)(e - s)) {
3180 yyerror("Invalid hexadecimal number in \\N{U+...}");
3187 /* On non-EBCDIC platforms, pass through to the regex
3188 * compiler unchanged. The reason we evaluated the
3189 * number above is to make sure there wasn't a syntax
3190 * error. But on EBCDIC we convert to native so
3191 * downstream code can continue to assume it's native
3193 s -= 5; /* Include the '\N{U+' */
3195 d += my_snprintf(d, e - s + 1 + 1, /* includes the }
3198 (unsigned int) UNI_TO_NATIVE(uv));
3200 Copy(s, d, e - s + 1, char); /* 1 = include the } */
3204 else { /* Not a pattern: convert the hex to string */
3206 /* If destination is not in utf8, unconditionally
3207 * recode it to be so. This is because \N{} implies
3208 * Unicode semantics, and scalars have to be in utf8
3209 * to guarantee those semantics */
3211 SvCUR_set(sv, d - SvPVX_const(sv));
3214 /* See Note on sizing above. */
3215 sv_utf8_upgrade_flags_grow(
3217 SV_GMAGIC|SV_FORCE_UTF8_UPGRADE,
3218 UNISKIP(uv) + (STRLEN)(send - e) + 1);
3219 d = SvPVX(sv) + SvCUR(sv);
3223 /* Add the string to the output */
3224 if (UNI_IS_INVARIANT(uv)) {
3227 else d = (char*)uvuni_to_utf8((U8*)d, uv);
3230 else { /* Here is \N{NAME} but not \N{U+...}. */
3232 SV *res; /* result from charnames */
3233 const char *str; /* the string in 'res' */
3234 STRLEN len; /* its length */
3236 /* Get the value for NAME */
3237 res = newSVpvn(s, e - s);
3238 res = new_constant( NULL, 0, "charnames",
3239 /* includes all of: \N{...} */
3240 res, NULL, s - 3, e - s + 4 );
3242 /* Most likely res will be in utf8 already since the
3243 * standard charnames uses pack U, but a custom translator
3244 * can leave it otherwise, so make sure. XXX This can be
3245 * revisited to not have charnames use utf8 for characters
3246 * that don't need it when regexes don't have to be in utf8
3247 * for Unicode semantics. If doing so, remember EBCDIC */
3248 sv_utf8_upgrade(res);
3249 str = SvPV_const(res, len);
3251 /* Don't accept malformed input */
3252 if (! is_utf8_string((U8 *) str, len)) {
3253 yyerror("Malformed UTF-8 returned by \\N");
3255 else if (PL_lex_inpat) {
3257 if (! len) { /* The name resolved to an empty string */
3258 Copy("\\N{}", d, 4, char);
3262 /* In order to not lose information for the regex
3263 * compiler, pass the result in the specially made
3264 * syntax: \N{U+c1.c2.c3...}, where c1 etc. are
3265 * the code points in hex of each character
3266 * returned by charnames */
3268 const char *str_end = str + len;
3269 STRLEN char_length; /* cur char's byte length */
3270 STRLEN output_length; /* and the number of bytes
3271 after this is translated
3273 const STRLEN off = d - SvPVX_const(sv);
3275 /* 2 hex per byte; 2 chars for '\N'; 2 chars for
3276 * max('U+', '.'); and 1 for NUL */
3277 char hex_string[2 * UTF8_MAXBYTES + 5];
3279 /* Get the first character of the result. */
3280 U32 uv = utf8n_to_uvuni((U8 *) str,
3285 /* The call to is_utf8_string() above hopefully
3286 * guarantees that there won't be an error. But
3287 * it's easy here to make sure. The function just
3288 * above warns and returns 0 if invalid utf8, but
3289 * it can also return 0 if the input is validly a
3290 * NUL. Disambiguate */
3291 if (uv == 0 && NATIVE_TO_ASCII(*str) != '\0') {
3292 uv = UNICODE_REPLACEMENT;
3295 /* Convert first code point to hex, including the
3296 * boiler plate before it. For all these, we
3297 * convert to native format so that downstream code
3298 * can continue to assume the input is native */
3300 my_snprintf(hex_string, sizeof(hex_string),
3302 (unsigned int) UNI_TO_NATIVE(uv));
3304 /* Make sure there is enough space to hold it */
3305 d = off + SvGROW(sv, off
3307 + (STRLEN)(send - e)
3308 + 2); /* '}' + NUL */
3310 Copy(hex_string, d, output_length, char);
3313 /* For each subsequent character, append dot and
3314 * its ordinal in hex */
3315 while ((str += char_length) < str_end) {
3316 const STRLEN off = d - SvPVX_const(sv);
3317 U32 uv = utf8n_to_uvuni((U8 *) str,
3321 if (uv == 0 && NATIVE_TO_ASCII(*str) != '\0') {
3322 uv = UNICODE_REPLACEMENT;
3326 my_snprintf(hex_string, sizeof(hex_string),
3328 (unsigned int) UNI_TO_NATIVE(uv));
3330 d = off + SvGROW(sv, off
3332 + (STRLEN)(send - e)
3333 + 2); /* '}' + NUL */
3334 Copy(hex_string, d, output_length, char);
3338 *d++ = '}'; /* Done. Add the trailing brace */
3341 else { /* Here, not in a pattern. Convert the name to a
3344 /* If destination is not in utf8, unconditionally
3345 * recode it to be so. This is because \N{} implies
3346 * Unicode semantics, and scalars have to be in utf8
3347 * to guarantee those semantics */
3349 SvCUR_set(sv, d - SvPVX_const(sv));
3352 /* See Note on sizing above. */
3353 sv_utf8_upgrade_flags_grow(sv,
3354 SV_GMAGIC|SV_FORCE_UTF8_UPGRADE,
3355 len + (STRLEN)(send - s) + 1);
3356 d = SvPVX(sv) + SvCUR(sv);
3358 } else if (len > (STRLEN)(e - s + 4)) { /* I _guess_ 4 is \N{} --jhi */
3360 /* See Note on sizing above. (NOTE: SvCUR() is not
3361 * set correctly here). */
3362 const STRLEN off = d - SvPVX_const(sv);
3363 d = off + SvGROW(sv, off + len + (STRLEN)(send - s) + 1);
3365 Copy(str, d, len, char);
3370 /* Deprecate non-approved name syntax */
3371 if (ckWARN_d(WARN_DEPRECATED)) {
3372 bool problematic = FALSE;
3375 /* For non-ut8 input, look to see that the first
3376 * character is an alpha, then loop through the rest
3377 * checking that each is a continuation */
3379 if (! isALPHAU(*i)) problematic = TRUE;
3380 else for (i = s + 1; i < e; i++) {
3381 if (isCHARNAME_CONT(*i)) continue;
3387 /* Similarly for utf8. For invariants can check
3388 * directly. We accept anything above the latin1
3389 * range because it is immaterial to Perl if it is
3390 * correct or not, and is expensive to check. But
3391 * it is fairly easy in the latin1 range to convert
3392 * the variants into a single character and check
3394 if (UTF8_IS_INVARIANT(*i)) {
3395 if (! isALPHAU(*i)) problematic = TRUE;
3396 } else if (UTF8_IS_DOWNGRADEABLE_START(*i)) {
3397 if (! isALPHAU(UNI_TO_NATIVE(TWO_BYTE_UTF8_TO_UNI(*i,
3403 if (! problematic) for (i = s + UTF8SKIP(s);
3407 if (UTF8_IS_INVARIANT(*i)) {
3408 if (isCHARNAME_CONT(*i)) continue;
3409 } else if (! UTF8_IS_DOWNGRADEABLE_START(*i)) {
3411 } else if (isCHARNAME_CONT(
3413 TWO_BYTE_UTF8_TO_UNI(*i, *(i+1)))))
3422 /* The e-i passed to the final %.*s makes sure that
3423 * should the trailing NUL be missing that this
3424 * print won't run off the end of the string */
3425 Perl_warner(aTHX_ packWARN(WARN_DEPRECATED),
3426 "Deprecated character in \\N{...}; marked by <-- HERE in \\N{%.*s<-- HERE %.*s",
3427 (int)(i - s + 1), s, (int)(e - i), i + 1);
3430 } /* End \N{NAME} */
3433 native_range = FALSE; /* \N{} is defined to be Unicode */
3435 s = e + 1; /* Point to just after the '}' */
3438 /* \c is a control character */
3442 *d++ = grok_bslash_c(*s++, has_utf8, 1);
3445 yyerror("Missing control char name in \\c");
3449 /* printf-style backslashes, formfeeds, newlines, etc */
3451 *d++ = NATIVE_TO_NEED(has_utf8,'\b');
3454 *d++ = NATIVE_TO_NEED(has_utf8,'\n');
3457 *d++ = NATIVE_TO_NEED(has_utf8,'\r');
3460 *d++ = NATIVE_TO_NEED(has_utf8,'\f');
3463 *d++ = NATIVE_TO_NEED(has_utf8,'\t');
3466 *d++ = ASCII_TO_NEED(has_utf8,'\033');
3469 *d++ = ASCII_TO_NEED(has_utf8,'\007');
3475 } /* end if (backslash) */
3482 /* If we started with encoded form, or already know we want it,
3483 then encode the next character */
3484 if (! NATIVE_IS_INVARIANT((U8)(*s)) && (this_utf8 || has_utf8)) {
3488 /* One might think that it is wasted effort in the case of the
3489 * source being utf8 (this_utf8 == TRUE) to take the next character
3490 * in the source, convert it to an unsigned value, and then convert
3491 * it back again. But the source has not been validated here. The
3492 * routine that does the conversion checks for errors like
3495 const UV nextuv = (this_utf8) ? utf8n_to_uvchr((U8*)s, send - s, &len, 0) : (UV) ((U8) *s);
3496 const STRLEN need = UNISKIP(NATIVE_TO_UNI(nextuv));
3498 SvCUR_set(sv, d - SvPVX_const(sv));
3501 /* See Note on sizing above. */
3502 sv_utf8_upgrade_flags_grow(sv,
3503 SV_GMAGIC|SV_FORCE_UTF8_UPGRADE,
3504 need + (STRLEN)(send - s) + 1);
3505 d = SvPVX(sv) + SvCUR(sv);
3507 } else if (need > len) {
3508 /* encoded value larger than old, may need extra space (NOTE:
3509 * SvCUR() is not set correctly here). See Note on sizing
3511 const STRLEN off = d - SvPVX_const(sv);
3512 d = SvGROW(sv, off + need + (STRLEN)(send - s) + 1) + off;
3516 d = (char*)uvchr_to_utf8((U8*)d, nextuv);
3518 if (uv > 255 && !dorange)
3519 native_range = FALSE;
3523 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
3525 } /* while loop to process each character */
3527 /* terminate the string and set up the sv */
3529 SvCUR_set(sv, d - SvPVX_const(sv));
3530 if (SvCUR(sv) >= SvLEN(sv))
3531 Perl_croak(aTHX_ "panic: constant overflowed allocated space, %"UVuf
3532 " >= %"UVuf, (UV)SvCUR(sv), (UV)SvLEN(sv));
3535 if (PL_encoding && !has_utf8) {
3536 sv_recode_to_utf8(sv, PL_encoding);
3542 if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
3543 PL_sublex_info.sub_op->op_private |=
3544 (PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
3548 /* shrink the sv if we allocated more than we used */
3549 if (SvCUR(sv) + 5 < SvLEN(sv)) {
3550 SvPV_shrink_to_cur(sv);
3553 /* return the substring (via pl_yylval) only if we parsed anything */
3554 if (s > PL_bufptr) {
3555 if ( PL_hints & ( PL_lex_inpat ? HINT_NEW_RE : HINT_NEW_STRING ) ) {
3556 const char *const key = PL_lex_inpat ? "qr" : "q";
3557 const STRLEN keylen = PL_lex_inpat ? 2 : 1;
3561 if (PL_lex_inwhat == OP_TRANS) {
3564 } else if (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat) {
3567 } else if (PL_lex_inpat && SvIVX(PL_linestr) == '\'') {
3575 sv = S_new_constant(aTHX_ start, s - start, key, keylen, sv, NULL,
3578 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
3585 * Returns TRUE if there's more to the expression (e.g., a subscript),
3588 * It deals with "$foo[3]" and /$foo[3]/ and /$foo[0123456789$]+/
3590 * ->[ and ->{ return TRUE
3591 * { and [ outside a pattern are always subscripts, so return TRUE
3592 * if we're outside a pattern and it's not { or [, then return FALSE
3593 * if we're in a pattern and the first char is a {
3594 * {4,5} (any digits around the comma) returns FALSE
3595 * if we're in a pattern and the first char is a [
3597 * [SOMETHING] has a funky algorithm to decide whether it's a
3598 * character class or not. It has to deal with things like
3599 * /$foo[-3]/ and /$foo[$bar]/ as well as /$foo[$\d]+/
3600 * anything else returns TRUE
3603 /* This is the one truly awful dwimmer necessary to conflate C and sed. */
3606 S_intuit_more(pTHX_ register char *s)
3610 PERL_ARGS_ASSERT_INTUIT_MORE;
3612 if (PL_lex_brackets)
3614 if (*s == '-' && s[1] == '>' && (s[2] == '[' || s[2] == '{'))
3616 if (*s != '{' && *s != '[')
3621 /* In a pattern, so maybe we have {n,m}. */
3629 /* On the other hand, maybe we have a character class */
3632 if (*s == ']' || *s == '^')
3635 /* this is terrifying, and it works */
3636 int weight = 2; /* let's weigh the evidence */
3638 unsigned char un_char = 255, last_un_char;
3639 const char * const send = strchr(s,']');
3640 char tmpbuf[sizeof PL_tokenbuf * 4];
3642 if (!send) /* has to be an expression */
3645 Zero(seen,256,char);
3648 else if (isDIGIT(*s)) {
3650 if (isDIGIT(s[1]) && s[2] == ']')
3656 for (; s < send; s++) {
3657 last_un_char = un_char;
3658 un_char = (unsigned char)*s;
3663 weight -= seen[un_char] * 10;
3664 if (isALNUM_lazy_if(s+1,UTF)) {
3666 scan_ident(s, send, tmpbuf, sizeof tmpbuf, FALSE);
3667 len = (int)strlen(tmpbuf);
3668 if (len > 1 && gv_fetchpvn_flags(tmpbuf, len,
3669 UTF ? SVf_UTF8 : 0, SVt_PV))
3674 else if (*s == '$' && s[1] &&
3675 strchr("[#!%*<>()-=",s[1])) {
3676 if (/*{*/ strchr("])} =",s[2]))
3685 if (strchr("wds]",s[1]))
3687 else if (seen[(U8)'\''] || seen[(U8)'"'])
3689 else if (strchr("rnftbxcav",s[1]))
3691 else if (isDIGIT(s[1])) {
3693 while (s[1] && isDIGIT(s[1]))
3703 if (strchr("aA01! ",last_un_char))
3705 if (strchr("zZ79~",s[1]))
3707 if (last_un_char == 255 && (isDIGIT(s[1]) || s[1] == '$'))
3708 weight -= 5; /* cope with negative subscript */
3711 if (!isALNUM(last_un_char)
3712 && !(last_un_char == '$' || last_un_char == '@'
3713 || last_un_char == '&')
3714 && isALPHA(*s) && s[1] && isALPHA(s[1])) {
3719 if (keyword(tmpbuf, d - tmpbuf, 0))
3722 if (un_char == last_un_char + 1)
3724 weight -= seen[un_char];
3729 if (weight >= 0) /* probably a character class */
3739 * Does all the checking to disambiguate
3741 * between foo(bar) and bar->foo. Returns 0 if not a method, otherwise
3742 * FUNCMETH (bar->foo(args)) or METHOD (bar->foo args).
3744 * First argument is the stuff after the first token, e.g. "bar".
3746 * Not a method if bar is a filehandle.
3747 * Not a method if foo is a subroutine prototyped to take a filehandle.
3748 * Not a method if it's really "Foo $bar"
3749 * Method if it's "foo $bar"
3750 * Not a method if it's really "print foo $bar"
3751 * Method if it's really "foo package::" (interpreted as package->foo)
3752 * Not a method if bar is known to be a subroutine ("sub bar; foo bar")
3753 * Not a method if bar is a filehandle or package, but is quoted with
3758 S_intuit_method(pTHX_ char *start, GV *gv, CV *cv)
3761 char *s = start + (*start == '$');
3762 char tmpbuf[sizeof PL_tokenbuf];
3769 PERL_ARGS_ASSERT_INTUIT_METHOD;
3772 if (SvTYPE(gv) == SVt_PVGV && GvIO(gv))
3776 const char *proto = CvPROTO(cv);
3787 s = scan_word(s, tmpbuf, sizeof tmpbuf, TRUE, &len);
3788 /* start is the beginning of the possible filehandle/object,
3789 * and s is the end of it
3790 * tmpbuf is a copy of it
3793 if (*start == '$') {
3794 if (gv || PL_last_lop_op == OP_PRINT || PL_last_lop_op == OP_SAY ||
3795 isUPPER(*PL_tokenbuf))
3798 len = start - SvPVX(PL_linestr);
3802 start = SvPVX(PL_linestr) + len;
3806 return *s == '(' ? FUNCMETH : METHOD;
3808 if (!keyword(tmpbuf, len, 0)) {
3809 if (len > 2 && tmpbuf[len - 2] == ':' && tmpbuf[len - 1] == ':') {
3813 soff = s - SvPVX(PL_linestr);
3817 indirgv = gv_fetchpvn_flags(tmpbuf, len, ( UTF ? SVf_UTF8 : 0 ), SVt_PVCV);
3818 if (indirgv && GvCVu(indirgv))
3820 /* filehandle or package name makes it a method */
3821 if (!gv || GvIO(indirgv) || gv_stashpvn(tmpbuf, len, UTF ? SVf_UTF8 : 0)) {
3823 soff = s - SvPVX(PL_linestr);
3826 if ((PL_bufend - s) >= 2 && *s == '=' && *(s+1) == '>')
3827 return 0; /* no assumptions -- "=>" quotes bareword */
3829 start_force(PL_curforce);
3830 NEXTVAL_NEXTTOKE.opval = (OP*)newSVOP(OP_CONST, 0,
3831 S_newSV_maybe_utf8(aTHX_ tmpbuf, len));
3832 NEXTVAL_NEXTTOKE.opval->op_private = OPpCONST_BARE;
3834 curmad('X', newSVpvn_flags(start,SvPVX(PL_linestr) + soff - start,
3835 ( UTF ? SVf_UTF8 : 0 )));
3840 PL_bufptr = SvPVX(PL_linestr) + soff; /* restart before space */
3842 return *s == '(' ? FUNCMETH : METHOD;
3848 /* Encoded script support. filter_add() effectively inserts a
3849 * 'pre-processing' function into the current source input stream.
3850 * Note that the filter function only applies to the current source file
3851 * (e.g., it will not affect files 'require'd or 'use'd by this one).
3853 * The datasv parameter (which may be NULL) can be used to pass
3854 * private data to this instance of the filter. The filter function
3855 * can recover the SV using the FILTER_DATA macro and use it to
3856 * store private buffers and state information.
3858 * The supplied datasv parameter is upgraded to a PVIO type
3859 * and the IoDIRP/IoANY field is used to store the function pointer,
3860 * and IOf_FAKE_DIRP is enabled on datasv to mark this as such.
3861 * Note that IoTOP_NAME, IoFMT_NAME, IoBOTTOM_NAME, if set for
3862 * private use must be set using malloc'd pointers.
3866 Perl_filter_add(pTHX_ filter_t funcp, SV *datasv)
3875 if (PL_parser->lex_flags & LEX_IGNORE_UTF8_HINTS)
3876 Perl_croak(aTHX_ "Source filters apply only to byte streams");
3878 if (!PL_rsfp_filters)
3879 PL_rsfp_filters = newAV();
3882 SvUPGRADE(datasv, SVt_PVIO);
3883 IoANY(datasv) = FPTR2DPTR(void *, funcp); /* stash funcp into spare field */
3884 IoFLAGS(datasv) |= IOf_FAKE_DIRP;
3885 DEBUG_P(PerlIO_printf(Perl_debug_log, "filter_add func %p (%s)\n",
3886 FPTR2DPTR(void *, IoANY(datasv)),
3887 SvPV_nolen(datasv)));
3888 av_unshift(PL_rsfp_filters, 1);
3889 av_store(PL_rsfp_filters, 0, datasv) ;
3891 !PL_parser->filtered
3892 && PL_parser->lex_flags & LEX_EVALBYTES
3893 && PL_bufptr < PL_bufend
3895 const char *s = PL_bufptr;
3896 while (s < PL_bufend) {
3898 SV *linestr = PL_parser->linestr;
3899 char *buf = SvPVX(linestr);
3900 STRLEN const bufptr_pos = PL_parser->bufptr - buf;
3901 STRLEN const oldbufptr_pos = PL_parser->oldbufptr - buf;
3902 STRLEN const oldoldbufptr_pos=PL_parser->oldoldbufptr-buf;
3903 STRLEN const linestart_pos = PL_parser->linestart - buf;
3904 STRLEN const last_uni_pos =
3905 PL_parser->last_uni ? PL_parser->last_uni - buf : 0;
3906 STRLEN const last_lop_pos =
3907 PL_parser->last_lop ? PL_parser->last_lop - buf : 0;
3908 av_push(PL_rsfp_filters, linestr);
3909 PL_parser->linestr =
3910 newSVpvn(SvPVX(linestr), ++s-SvPVX(linestr));
3911 buf = SvPVX(PL_parser->linestr);
3912 PL_parser->bufend = buf + SvCUR(PL_parser->linestr);
3913 PL_parser->bufptr = buf + bufptr_pos;
3914 PL_parser->oldbufptr = buf + oldbufptr_pos;
3915 PL_parser->oldoldbufptr = buf + oldoldbufptr_pos;
3916 PL_parser->linestart = buf + linestart_pos;
3917 if (PL_parser->last_uni)
3918 PL_parser->last_uni = buf + last_uni_pos;
3919 if (PL_parser->last_lop)
3920 PL_parser->last_lop = buf + last_lop_pos;
3921 SvLEN(linestr) = SvCUR(linestr);
3922 SvCUR(linestr) = s-SvPVX(linestr);
3923 PL_parser->filtered = 1;
3933 /* Delete most recently added instance of this filter function. */
3935 Perl_filter_del(pTHX_ filter_t funcp)
3940 PERL_ARGS_ASSERT_FILTER_DEL;
3943 DEBUG_P(PerlIO_printf(Perl_debug_log, "filter_del func %p",
3944 FPTR2DPTR(void*, funcp)));
3946 if (!PL_parser || !PL_rsfp_filters || AvFILLp(PL_rsfp_filters)<0)
3948 /* if filter is on top of stack (usual case) just pop it off */
3949 datasv = FILTER_DATA(AvFILLp(PL_rsfp_filters));
3950 if (IoANY(datasv) == FPTR2DPTR(void *, funcp)) {
3951 sv_free(av_pop(PL_rsfp_filters));
3955 /* we need to search for the correct entry and clear it */
3956 Perl_die(aTHX_ "filter_del can only delete in reverse order (currently)");
3960 /* Invoke the idxth filter function for the current rsfp. */
3961 /* maxlen 0 = read one text line */
3963 Perl_filter_read(pTHX_ int idx, SV *buf_sv, int maxlen)
3968 /* This API is bad. It should have been using unsigned int for maxlen.
3969 Not sure if we want to change the API, but if not we should sanity
3970 check the value here. */
3971 unsigned int correct_length
3980 PERL_ARGS_ASSERT_FILTER_READ;
3982 if (!PL_parser || !PL_rsfp_filters)
3984 if (idx > AvFILLp(PL_rsfp_filters)) { /* Any more filters? */
3985 /* Provide a default input filter to make life easy. */
3986 /* Note that we append to the line. This is handy. */
3987 DEBUG_P(PerlIO_printf(Perl_debug_log,
3988 "filter_read %d: from rsfp\n", idx));
3989 if (correct_length) {
3992 const int old_len = SvCUR(buf_sv);
3994 /* ensure buf_sv is large enough */
3995 SvGROW(buf_sv, (STRLEN)(old_len + correct_length + 1)) ;
3996 if ((len = PerlIO_read(PL_rsfp, SvPVX(buf_sv) + old_len,
3997 correct_length)) <= 0) {
3998 if (PerlIO_error(PL_rsfp))
3999 return -1; /* error */
4001 return 0 ; /* end of file */
4003 SvCUR_set(buf_sv, old_len + len) ;
4004 SvPVX(buf_sv)[old_len + len] = '\0';
4007 if (sv_gets(buf_sv, PL_rsfp, SvCUR(buf_sv)) == NULL) {
4008 if (PerlIO_error(PL_rsfp))
4009 return -1; /* error */
4011 return 0 ; /* end of file */
4014 return SvCUR(buf_sv);
4016 /* Skip this filter slot if filter has been deleted */
4017 if ( (datasv = FILTER_DATA(idx)) == &PL_sv_undef) {
4018 DEBUG_P(PerlIO_printf(Perl_debug_log,
4019 "filter_read %d: skipped (filter deleted)\n",
4021 return FILTER_READ(idx+1, buf_sv, correct_length); /* recurse */
4023 if (SvTYPE(datasv) != SVt_PVIO) {
4024 if (correct_length) {
4026 const STRLEN remainder = SvLEN(datasv) - SvCUR(datasv);
4027 if (!remainder) return 0; /* eof */
4028 if (correct_length > remainder) correct_length = remainder;
4029 sv_catpvn(buf_sv, SvEND(datasv), correct_length);
4030 SvCUR_set(datasv, SvCUR(datasv) + correct_length);
4033 const char *s = SvEND(datasv);
4034 const char *send = SvPVX(datasv) + SvLEN(datasv);
4042 if (s == send) return 0; /* eof */
4043 sv_catpvn(buf_sv, SvEND(datasv), s-SvEND(datasv));
4044 SvCUR_set(datasv, s-SvPVX(datasv));
4046 return SvCUR(buf_sv);
4048 /* Get function pointer hidden within datasv */
4049 funcp = DPTR2FPTR(filter_t, IoANY(datasv));
4050 DEBUG_P(PerlIO_printf(Perl_debug_log,
4051 "filter_read %d: via function %p (%s)\n",
4052 idx, (void*)datasv, SvPV_nolen_const(datasv)));
4053 /* Call function. The function is expected to */
4054 /* call "FILTER_READ(idx+1, buf_sv)" first. */
4055 /* Return: <0:error, =0:eof, >0:not eof */
4056 return (*funcp)(aTHX_ idx, buf_sv, correct_length);
4060 S_filter_gets(pTHX_ register SV *sv, STRLEN append)
4064 PERL_ARGS_ASSERT_FILTER_GETS;
4066 #ifdef PERL_CR_FILTER
4067 if (!PL_rsfp_filters) {
4068 filter_add(S_cr_textfilter,NULL);
4071 if (PL_rsfp_filters) {
4073 SvCUR_set(sv, 0); /* start with empty line */
4074 if (FILTER_READ(0, sv, 0) > 0)
4075 return ( SvPVX(sv) ) ;
4080 return (sv_gets(sv, PL_rsfp, append));
4084 S_find_in_my_stash(pTHX_ const char *pkgname, STRLEN len)
4089 PERL_ARGS_ASSERT_FIND_IN_MY_STASH;
4091 if (len == 11 && *pkgname == '_' && strEQ(pkgname, "__PACKAGE__"))
4095 (pkgname[len - 2] == ':' && pkgname[len - 1] == ':') &&
4096 (gv = gv_fetchpvn_flags(pkgname, len, ( UTF ? SVf_UTF8 : 0 ), SVt_PVHV)))
4098 return GvHV(gv); /* Foo:: */
4101 /* use constant CLASS => 'MyClass' */
4102 gv = gv_fetchpvn_flags(pkgname, len, UTF ? SVf_UTF8 : 0, SVt_PVCV);
4103 if (gv && GvCV(gv)) {
4104 SV * const sv = cv_const_sv(GvCV(gv));
4106 pkgname = SvPV_const(sv, len);
4109 return gv_stashpvn(pkgname, len, UTF ? SVf_UTF8 : 0);
4113 * S_readpipe_override
4114 * Check whether readpipe() is overridden, and generates the appropriate
4115 * optree, provided sublex_start() is called afterwards.
4118 S_readpipe_override(pTHX)
4121 GV *gv_readpipe = gv_fetchpvs("readpipe", GV_NOTQUAL, SVt_PVCV);
4122 pl_yylval.ival = OP_BACKTICK;
4124 && GvCVu(gv_readpipe) && GvIMPORTED_CV(gv_readpipe))
4126 ((gvp = (GV**)hv_fetchs(PL_globalstash, "readpipe", FALSE))
4127 && (gv_readpipe = *gvp) && isGV_with_GP(gv_readpipe)
4128 && GvCVu(gv_readpipe) && GvIMPORTED_CV(gv_readpipe)))
4130 PL_lex_op = (OP*)newUNOP(OP_ENTERSUB, OPf_STACKED,
4131 op_append_elem(OP_LIST,
4132 newSVOP(OP_CONST, 0, &PL_sv_undef), /* value will be read later */
4133 newCVREF(0, newGVOP(OP_GV, 0, gv_readpipe))));
4140 * The intent of this yylex wrapper is to minimize the changes to the
4141 * tokener when we aren't interested in collecting madprops. It remains
4142 * to be seen how successful this strategy will be...
4149 char *s = PL_bufptr;
4151 /* make sure PL_thiswhite is initialized */
4155 /* just do what yylex would do on pending identifier; leave PL_thiswhite alone */
4156 if (PL_lex_state != LEX_KNOWNEXT && PL_pending_ident)
4157 return S_pending_ident(aTHX);
4159 /* previous token ate up our whitespace? */
4160 if (!PL_lasttoke && PL_nextwhite) {
4161 PL_thiswhite = PL_nextwhite;
4165 /* isolate the token, and figure out where it is without whitespace */
4166 PL_realtokenstart = -1;
4170 assert(PL_curforce < 0);
4172 if (!PL_thismad || PL_thismad->mad_key == '^') { /* not forced already? */
4173 if (!PL_thistoken) {
4174 if (PL_realtokenstart < 0 || !CopLINE(PL_curcop))
4175 PL_thistoken = newSVpvs("");
4177 char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
4178 PL_thistoken = newSVpvn(tstart, s - tstart);
4181 if (PL_thismad) /* install head */
4182 CURMAD('X', PL_thistoken);
4185 /* last whitespace of a sublex? */
4186 if (optype == ')' && PL_endwhite) {
4187 CURMAD('X', PL_endwhite);
4192 /* if no whitespace and we're at EOF, bail. Otherwise fake EOF below. */
4193 if (!PL_thiswhite && !PL_endwhite && !optype) {
4194 sv_free(PL_thistoken);
4199 /* put off final whitespace till peg */
4200 if (optype == ';' && !PL_rsfp && !PL_parser->filtered) {
4201 PL_nextwhite = PL_thiswhite;
4204 else if (PL_thisopen) {
4205 CURMAD('q', PL_thisopen);
4207 sv_free(PL_thistoken);
4211 /* Store actual token text as madprop X */
4212 CURMAD('X', PL_thistoken);
4216 /* add preceding whitespace as madprop _ */
4217 CURMAD('_', PL_thiswhite);
4221 /* add quoted material as madprop = */
4222 CURMAD('=', PL_thisstuff);
4226 /* add terminating quote as madprop Q */
4227 CURMAD('Q', PL_thisclose);
4231 /* special processing based on optype */
4235 /* opval doesn't need a TOKEN since it can already store mp */
4246 if (pl_yylval.opval)
4247 append_madprops(PL_thismad, pl_yylval.opval, 0);
4255 addmad(newMADsv('p', PL_endwhite), &PL_thismad, 0);
4264 /* remember any fake bracket that lexer is about to discard */
4265 if (PL_lex_brackets == 1 &&
4266 ((expectation)PL_lex_brackstack[0] & XFAKEBRACK))
4269 while (s < PL_bufend && (*s == ' ' || *s == '\t'))
4272 PL_thiswhite = newSVpvn(PL_bufptr, ++s - PL_bufptr);
4273 addmad(newMADsv('#', PL_thiswhite), &PL_thismad, 0);
4276 break; /* don't bother looking for trailing comment */
4285 /* attach a trailing comment to its statement instead of next token */
4289 if (PL_bufptr > PL_oldbufptr && PL_bufptr[-1] == optype) {
4291 while (s < PL_bufend && (*s == ' ' || *s == '\t'))
4293 if (*s == '\n' || *s == '#') {
4294 while (s < PL_bufend && *s != '\n')
4298 PL_thiswhite = newSVpvn(PL_bufptr, s - PL_bufptr);
4299 addmad(newMADsv('#', PL_thiswhite), &PL_thismad, 0);
4312 /* Create new token struct. Note: opvals return early above. */
4313 pl_yylval.tkval = newTOKEN(optype, pl_yylval, PL_thismad);
4320 S_tokenize_use(pTHX_ int is_use, char *s) {
4323 PERL_ARGS_ASSERT_TOKENIZE_USE;
4325 if (PL_expect != XSTATE)
4326 yyerror(Perl_form(aTHX_ "\"%s\" not allowed in expression",
4327 is_use ? "use" : "no"));
4329 if (isDIGIT(*s) || (*s == 'v' && isDIGIT(s[1]))) {
4330 s = force_version(s, TRUE);
4331 if (*s == ';' || *s == '}'
4332 || (s = SKIPSPACE1(s), (*s == ';' || *s == '}'))) {
4333 start_force(PL_curforce);
4334 NEXTVAL_NEXTTOKE.opval = NULL;
4337 else if (*s == 'v') {
4338 s = force_word(s,WORD,FALSE,TRUE,FALSE);
4339 s = force_version(s, FALSE);
4343 s = force_word(s,WORD,FALSE,TRUE,FALSE);
4344 s = force_version(s, FALSE);
4346 pl_yylval.ival = is_use;
4350 static const char* const exp_name[] =
4351 { "OPERATOR", "TERM", "REF", "STATE", "BLOCK", "ATTRBLOCK",
4352 "ATTRTERM", "TERMBLOCK", "TERMORDORDOR"
4356 #define word_takes_any_delimeter(p,l) S_word_takes_any_delimeter(p,l)
4358 S_word_takes_any_delimeter(char *p, STRLEN len)
4360 return (len == 1 && strchr("msyq", p[0])) ||
4362 (p[0] == 't' && p[1] == 'r') ||
4363 (p[0] == 'q' && strchr("qwxr", p[1]))));
4369 Works out what to call the token just pulled out of the input
4370 stream. The yacc parser takes care of taking the ops we return and
4371 stitching them into a tree.
4377 if read an identifier
4378 if we're in a my declaration
4379 croak if they tried to say my($foo::bar)
4380 build the ops for a my() declaration
4381 if it's an access to a my() variable
4382 are we in a sort block?
4383 croak if my($a); $a <=> $b
4384 build ops for access to a my() variable
4385 if in a dq string, and they've said @foo and we can't find @foo
4387 build ops for a bareword
4388 if we already built the token before, use it.
4393 #pragma segment Perl_yylex
4399 register char *s = PL_bufptr;
4405 /* orig_keyword, gvp, and gv are initialized here because
4406 * jump to the label just_a_word_zero can bypass their
4407 * initialization later. */
4408 I32 orig_keyword = 0;
4413 SV* tmp = newSVpvs("");
4414 PerlIO_printf(Perl_debug_log, "### %"IVdf":LEX_%s/X%s %s\n",
4415 (IV)CopLINE(PL_curcop),
4416 lex_state_names[PL_lex_state],
4417 exp_name[PL_expect],
4418 pv_display(tmp, s, strlen(s), 0, 60));
4421 /* check if there's an identifier for us to look at */
4422 if (PL_lex_state != LEX_KNOWNEXT && PL_pending_ident)
4423 return REPORT(S_pending_ident(aTHX));
4425 /* no identifier pending identification */
4427 switch (PL_lex_state) {
4429 case LEX_NORMAL: /* Some compilers will produce faster */
4430 case LEX_INTERPNORMAL: /* code if we comment these out. */
4434 /* when we've already built the next token, just pull it out of the queue */
4438 pl_yylval = PL_nexttoke[PL_lasttoke].next_val;
4440 PL_thismad = PL_nexttoke[PL_lasttoke].next_mad;
4441 PL_nexttoke[PL_lasttoke].next_mad = 0;
4442 if (PL_thismad && PL_thismad->mad_key == '_') {
4443 PL_thiswhite = MUTABLE_SV(PL_thismad->mad_val);
4444 PL_thismad->mad_val = 0;
4445 mad_free(PL_thismad);
4450 PL_lex_state = PL_lex_defer;
4451 PL_expect = PL_lex_expect;
4452 PL_lex_defer = LEX_NORMAL;
4453 if (!PL_nexttoke[PL_lasttoke].next_type)
4458 pl_yylval = PL_nextval[PL_nexttoke];
4460 PL_lex_state = PL_lex_defer;
4461 PL_expect = PL_lex_expect;
4462 PL_lex_defer = LEX_NORMAL;
4468 next_type = PL_nexttoke[PL_lasttoke].next_type;
4470 next_type = PL_nexttype[PL_nexttoke];
4472 if (next_type & (7<<24)) {
4473 if (next_type & (1<<24)) {
4474 if (PL_lex_brackets > 100)
4475 Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
4476 PL_lex_brackstack[PL_lex_brackets++] =
4477 (char) ((next_type >> 16) & 0xff);
4479 if (next_type & (2<<24))
4480 PL_lex_allbrackets++;
4481 if (next_type & (4<<24))
4482 PL_lex_allbrackets--;
4483 next_type &= 0xffff;
4486 /* FIXME - can these be merged? */
4489 return REPORT(next_type);
4493 /* interpolated case modifiers like \L \U, including \Q and \E.
4494 when we get here, PL_bufptr is at the \
4496 case LEX_INTERPCASEMOD:
4498 if (PL_bufptr != PL_bufend && *PL_bufptr != '\\')
4500 "panic: INTERPCASEMOD bufptr=%p, bufend=%p, *bufptr=%u",
4501 PL_bufptr, PL_bufend, *PL_bufptr);
4503 /* handle \E or end of string */
4504 if (PL_bufptr == PL_bufend || PL_bufptr[1] == 'E') {
4506 if (PL_lex_casemods) {
4507 const char oldmod = PL_lex_casestack[--PL_lex_casemods];
4508 PL_lex_casestack[PL_lex_casemods] = '\0';
4510 if (PL_bufptr != PL_bufend
4511 && (oldmod == 'L' || oldmod == 'U' || oldmod == 'Q'
4512 || oldmod == 'F')) {
4514 PL_lex_state = LEX_INTERPCONCAT;
4517 PL_thistoken = newSVpvs("\\E");
4520 PL_lex_allbrackets--;
4523 else if ( PL_bufptr != PL_bufend && PL_bufptr[1] == 'E' ) {
4524 /* Got an unpaired \E */
4525 Perl_ck_warner(aTHX_ packWARN(WARN_MISC),
4526 "Useless use of \\E");
4529 while (PL_bufptr != PL_bufend &&
4530 PL_bufptr[0] == '\\' && PL_bufptr[1] == 'E') {
4532 PL_thiswhite = newSVpvs("");
4533 sv_catpvn(PL_thiswhite, PL_bufptr, 2);
4537 if (PL_bufptr != PL_bufend)
4540 PL_lex_state = LEX_INTERPCONCAT;
4544 DEBUG_T({ PerlIO_printf(Perl_debug_log,
4545 "### Saw case modifier\n"); });
4547 if (s[1] == '\\' && s[2] == 'E') {
4550 PL_thiswhite = newSVpvs("");
4551 sv_catpvn(PL_thiswhite, PL_bufptr, 4);
4554 PL_lex_state = LEX_INTERPCONCAT;
4559 if (!PL_madskills) /* when just compiling don't need correct */
4560 if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
4561 tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
4562 if ((*s == 'L' || *s == 'U' || *s == 'F') &&
4563 (strchr(PL_lex_casestack, 'L')
4564 || strchr(PL_lex_casestack, 'U')
4565 || strchr(PL_lex_casestack, 'F'))) {
4566 PL_lex_casestack[--PL_lex_casemods] = '\0';
4567 PL_lex_allbrackets--;
4570 if (PL_lex_casemods > 10)
4571 Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
4572 PL_lex_casestack[PL_lex_casemods++] = *s;
4573 PL_lex_casestack[PL_lex_casemods] = '\0';
4574 PL_lex_state = LEX_INTERPCONCAT;
4575 start_force(PL_curforce);
4576 NEXTVAL_NEXTTOKE.ival = 0;
4577 force_next((2<<24)|'(');
4578 start_force(PL_curforce);
4580 NEXTVAL_NEXTTOKE.ival = OP_LCFIRST;
4582 NEXTVAL_NEXTTOKE.ival = OP_UCFIRST;
4584 NEXTVAL_NEXTTOKE.ival = OP_LC;
4586 NEXTVAL_NEXTTOKE.ival = OP_UC;
4588 NEXTVAL_NEXTTOKE.ival = OP_QUOTEMETA;
4590 NEXTVAL_NEXTTOKE.ival = OP_FC;
4592 Perl_croak(aTHX_ "panic: yylex, *s=%u", *s);
4594 SV* const tmpsv = newSVpvs("\\ ");
4595 /* replace the space with the character we want to escape
4597 SvPVX(tmpsv)[1] = *s;
4603 if (PL_lex_starts) {
4609 sv_free(PL_thistoken);
4610 PL_thistoken = newSVpvs("");
4613 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
4614 if (PL_lex_casemods == 1 && PL_lex_inpat)
4623 case LEX_INTERPPUSH:
4624 return REPORT(sublex_push());
4626 case LEX_INTERPSTART:
4627 if (PL_bufptr == PL_bufend)
4628 return REPORT(sublex_done());
4629 DEBUG_T({ if(*PL_bufptr != '(') PerlIO_printf(Perl_debug_log,
4630 "### Interpolated variable\n"); });
4632 PL_lex_dojoin = (*PL_bufptr == '@');
4633 PL_lex_state = LEX_INTERPNORMAL;
4634 if (PL_lex_dojoin) {
4635 start_force(PL_curforce);
4636 NEXTVAL_NEXTTOKE.ival = 0;
4638 start_force(PL_curforce);
4639 force_ident("\"", '$');
4640 start_force(PL_curforce);
4641 NEXTVAL_NEXTTOKE.ival = 0;
4643 start_force(PL_curforce);
4644 NEXTVAL_NEXTTOKE.ival = 0;
4645 force_next((2<<24)|'(');
4646 start_force(PL_curforce);
4647 NEXTVAL_NEXTTOKE.ival = OP_JOIN; /* emulate join($", ...) */
4650 /* Convert (?{...}) and friends to 'do {...}' */
4651 if (PL_lex_inpat && *PL_bufptr == '(') {
4652 PL_sublex_info.re_eval_start = PL_bufptr;
4654 if (*PL_bufptr != '{')
4656 start_force(PL_curforce);
4657 /* XXX probably need a CURMAD(something) here */
4658 PL_expect = XTERMBLOCK;
4662 if (PL_lex_starts++) {
4667 sv_free(PL_thistoken);
4668 PL_thistoken = newSVpvs("");
4671 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
4672 if (!PL_lex_casemods && PL_lex_inpat)
4679 case LEX_INTERPENDMAYBE:
4680 if (intuit_more(PL_bufptr)) {
4681 PL_lex_state = LEX_INTERPNORMAL; /* false alarm, more expr */
4687 if (PL_lex_dojoin) {
4688 PL_lex_dojoin = FALSE;
4689 PL_lex_state = LEX_INTERPCONCAT;
4693 sv_free(PL_thistoken);
4694 PL_thistoken = newSVpvs("");
4697 PL_lex_allbrackets--;
4700 if (PL_lex_inwhat == OP_SUBST && PL_linestr == PL_lex_repl
4701 && SvEVALED(PL_lex_repl))
4703 if (PL_bufptr != PL_bufend)
4704 Perl_croak(aTHX_ "Bad evalled substitution pattern");
4707 if (PL_sublex_info.re_eval_start) {
4708 if (*PL_bufptr != ')')
4709 Perl_croak(aTHX_ "Sequence (?{...}) not terminated with ')'");
4711 /* having compiled a (?{..}) expression, return the original
4712 * text too, as a const */
4713 start_force(PL_curforce);
4714 /* XXX probably need a CURMAD(something) here */
4715 NEXTVAL_NEXTTOKE.opval =
4716 (OP*)newSVOP(OP_CONST, 0,
4717 newSVpvn(PL_sublex_info.re_eval_start,
4718 PL_bufptr - PL_sublex_info.re_eval_start));
4720 PL_sublex_info.re_eval_start = NULL;
4726 case LEX_INTERPCONCAT:
4728 if (PL_lex_brackets)
4729 Perl_croak(aTHX_ "panic: INTERPCONCAT, lex_brackets=%ld",
4730 (long) PL_lex_brackets);
4732 if (PL_bufptr == PL_bufend)
4733 return REPORT(sublex_done());
4735 /* m'foo' still needs to be parsed for possible (?{...}) */
4736 if (SvIVX(PL_linestr) == '\'' && !PL_lex_inpat) {
4737 SV *sv = newSVsv(PL_linestr);
4739 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
4743 s = scan_const(PL_bufptr);
4745 PL_lex_state = LEX_INTERPCASEMOD;
4747 PL_lex_state = LEX_INTERPSTART;
4750 if (s != PL_bufptr) {
4751 start_force(PL_curforce);
4753 curmad('X', newSVpvn(PL_bufptr,s-PL_bufptr));
4755 NEXTVAL_NEXTTOKE = pl_yylval;
4758 if (PL_lex_starts++) {
4762 sv_free(PL_thistoken);
4763 PL_thistoken = newSVpvs("");
4766 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
4767 if (!PL_lex_casemods && PL_lex_inpat)
4780 PL_lex_state = LEX_NORMAL;
4781 s = scan_formline(PL_bufptr);
4782 if (!PL_lex_formbrack)
4788 PL_oldoldbufptr = PL_oldbufptr;
4794 sv_free(PL_thistoken);
4797 PL_realtokenstart = s - SvPVX(PL_linestr); /* assume but undo on ws */
4801 if (isIDFIRST_lazy_if(s,UTF))
4804 SV *dsv = newSVpvs_flags("", SVs_TEMP);
4805 const char *c = UTF ? savepv(sv_uni_display(dsv, newSVpvn_flags(s,
4807 SVs_TEMP | SVf_UTF8),
4808 10, UNI_DISPLAY_ISPRINT))
4809 : Perl_form(aTHX_ "\\x%02X", (unsigned char)*s);
4810 len = UTF ? Perl_utf8_length(aTHX_ (U8 *) PL_linestart, (U8 *) s) : (STRLEN) (s - PL_linestart);
4811 if (len > UNRECOGNIZED_PRECEDE_COUNT) {
4812 d = UTF ? (char *) Perl_utf8_hop(aTHX_ (U8 *) s, -UNRECOGNIZED_PRECEDE_COUNT) : s - UNRECOGNIZED_PRECEDE_COUNT;
4820 Perl_croak(aTHX_ "Unrecognized character %s; marked by <-- HERE after %"SVf"<-- HERE near column %d", c, SVfARG(dsv), (int) len + 1);
4824 goto fake_eof; /* emulate EOF on ^D or ^Z */
4830 if (!PL_rsfp && (!PL_parser->filtered || s+1 < PL_bufend)) {
4833 if (PL_lex_brackets &&
4834 PL_lex_brackstack[PL_lex_brackets-1] != XFAKEEOF) {
4835 yyerror((const char *)
4837 ? "Format not terminated"
4838 : "Missing right curly or square bracket"));
4840 DEBUG_T( { PerlIO_printf(Perl_debug_log,
4841 "### Tokener got EOF\n");
4845 if (s++ < PL_bufend)
4846 goto retry; /* ignore stray nulls */
4849 if (!PL_in_eval && !PL_preambled) {
4850 PL_preambled = TRUE;
4856 /* Generate a string of Perl code to load the debugger.
4857 * If PERL5DB is set, it will return the contents of that,
4858 * otherwise a compile-time require of perl5db.pl. */
4860 const char * const pdb = PerlEnv_getenv("PERL5DB");
4863 sv_setpv(PL_linestr, pdb);
4864 sv_catpvs(PL_linestr,";");
4866 SETERRNO(0,SS_NORMAL);
4867 sv_setpvs(PL_linestr, "BEGIN { require 'perl5db.pl' };");
4870 sv_setpvs(PL_linestr,"");
4871 if (PL_preambleav) {
4872 SV **svp = AvARRAY(PL_preambleav);
4873 SV **const end = svp + AvFILLp(PL_preambleav);
4875 sv_catsv(PL_linestr, *svp);
4877 sv_catpvs(PL_linestr, ";");
4879 sv_free(MUTABLE_SV(PL_preambleav));
4880 PL_preambleav = NULL;
4883 sv_catpvs(PL_linestr,
4884 "use feature ':5." STRINGIFY(PERL_VERSION) "';");
4885 if (PL_minus_n || PL_minus_p) {
4886 sv_catpvs(PL_linestr, "LINE: while (<>) {"/*}*/);
4888 sv_catpvs(PL_linestr,"chomp;");
4891 if ((*PL_splitstr == '/' || *PL_splitstr == '\''
4892 || *PL_splitstr == '"')
4893 && strchr(PL_splitstr + 1, *PL_splitstr))
4894 Perl_sv_catpvf(aTHX_ PL_linestr, "our @F=split(%s);", PL_splitstr);
4896 /* "q\0${splitstr}\0" is legal perl. Yes, even NUL
4897 bytes can be used as quoting characters. :-) */
4898 const char *splits = PL_splitstr;
4899 sv_catpvs(PL_linestr, "our @F=split(q\0");
4902 if (*splits == '\\')
4903 sv_catpvn(PL_linestr, splits, 1);
4904 sv_catpvn(PL_linestr, splits, 1);
4905 } while (*splits++);
4906 /* This loop will embed the trailing NUL of
4907 PL_linestr as the last thing it does before
4909 sv_catpvs(PL_linestr, ");");
4913 sv_catpvs(PL_linestr,"our @F=split(' ');");
4916 sv_catpvs(PL_linestr, "\n");
4917 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
4918 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
4919 PL_last_lop = PL_last_uni = NULL;
4920 if ((PERLDB_LINE || PERLDB_SAVESRC) && PL_curstash != PL_debstash)
4921 update_debugger_info(PL_linestr, NULL, 0);
4926 bof = PL_rsfp ? TRUE : FALSE;
4929 fake_eof = LEX_FAKE_EOF;
4931 PL_bufptr = PL_bufend;
4932 CopLINE_inc(PL_curcop);
4933 if (!lex_next_chunk(fake_eof)) {
4934 CopLINE_dec(PL_curcop);
4936 TOKEN(';'); /* not infinite loop because rsfp is NULL now */
4938 CopLINE_dec(PL_curcop);
4941 PL_realtokenstart = -1;
4944 /* If it looks like the start of a BOM or raw UTF-16,
4945 * check if it in fact is. */
4946 if (bof && PL_rsfp &&
4951 Off_t offset = (IV)PerlIO_tell(PL_rsfp);
4952 bof = (offset == (Off_t)SvCUR(PL_linestr));
4953 #if defined(PERLIO_USING_CRLF) && defined(PERL_TEXTMODE_SCRIPTS)
4954 /* offset may include swallowed CR */
4956 bof = (offset == (Off_t)SvCUR(PL_linestr)+1);
4959 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
4960 s = swallow_bom((U8*)s);
4963 if (PL_parser->in_pod) {
4964 /* Incest with pod. */
4967 sv_catsv(PL_thiswhite, PL_linestr);
4969 if (*s == '=' && strnEQ(s, "=cut", 4) && !isALPHA(s[4])) {
4970 sv_setpvs(PL_linestr, "");
4971 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
4972 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
4973 PL_last_lop = PL_last_uni = NULL;
4974 PL_parser->in_pod = 0;
4977 if (PL_rsfp || PL_parser->filtered)
4979 } while (PL_parser->in_pod);
4980 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = PL_linestart = s;
4981 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
4982 PL_last_lop = PL_last_uni = NULL;
4983 if (CopLINE(PL_curcop) == 1) {
4984 while (s < PL_bufend && isSPACE(*s))
4986 if (*s == ':' && s[1] != ':') /* for csh execing sh scripts */
4990 PL_thiswhite = newSVpvn(PL_linestart, s - PL_linestart);
4994 if (*s == '#' && *(s+1) == '!')
4996 #ifdef ALTERNATE_SHEBANG
4998 static char const as[] = ALTERNATE_SHEBANG;
4999 if (*s == as[0] && strnEQ(s, as, sizeof(as) - 1))
5000 d = s + (sizeof(as) - 1);
5002 #endif /* ALTERNATE_SHEBANG */
5011 while (*d && !isSPACE(*d))
5015 #ifdef ARG_ZERO_IS_SCRIPT
5016 if (ipathend > ipath) {
5018 * HP-UX (at least) sets argv[0] to the script name,
5019 * which makes $^X incorrect. And Digital UNIX and Linux,
5020 * at least, set argv[0] to the basename of the Perl
5021 * interpreter. So, having found "#!", we'll set it right.
5023 SV * const x = GvSV(gv_fetchpvs("\030", GV_ADD|GV_NOTQUAL,
5025 assert(SvPOK(x) || SvGMAGICAL(x));
5026 if (sv_eq(x, CopFILESV(PL_curcop))) {
5027 sv_setpvn(x, ipath, ipathend - ipath);
5033 const char *bstart = SvPV_const(CopFILESV(PL_curcop),blen);
5034 const char * const lstart = SvPV_const(x,llen);
5036 bstart += blen - llen;
5037 if (strnEQ(bstart, lstart, llen) && bstart[-1] == '/') {
5038 sv_setpvn(x, ipath, ipathend - ipath);
5043 TAINT_NOT; /* $^X is always tainted, but that's OK */
5045 #endif /* ARG_ZERO_IS_SCRIPT */
5050 d = instr(s,"perl -");
5052 d = instr(s,"perl");
5054 /* avoid getting into infinite loops when shebang
5055 * line contains "Perl" rather than "perl" */
5057 for (d = ipathend-4; d >= ipath; --d) {
5058 if ((*d == 'p' || *d == 'P')
5059 && !ibcmp(d, "perl", 4))
5069 #ifdef ALTERNATE_SHEBANG
5071 * If the ALTERNATE_SHEBANG on this system starts with a
5072 * character that can be part of a Perl expression, then if
5073 * we see it but not "perl", we're probably looking at the
5074 * start of Perl code, not a request to hand off to some
5075 * other interpreter. Similarly, if "perl" is there, but
5076 * not in the first 'word' of the line, we assume the line
5077 * contains the start of the Perl program.
5079 if (d && *s != '#') {
5080 const char *c = ipath;
5081 while (*c && !strchr("; \t\r\n\f\v#", *c))
5084 d = NULL; /* "perl" not in first word; ignore */
5086 *s = '#'; /* Don't try to parse shebang line */
5088 #endif /* ALTERNATE_SHEBANG */
5093 !instr(s,"indir") &&
5094 instr(PL_origargv[0],"perl"))
5101 while (s < PL_bufend && isSPACE(*s))
5103 if (s < PL_bufend) {
5104 Newx(newargv,PL_origargc+3,char*);
5106 while (s < PL_bufend && !isSPACE(*s))
5109 Copy(PL_origargv+1, newargv+2, PL_origargc+1, char*);
5112 newargv = PL_origargv;
5115 PerlProc_execv(ipath, EXEC_ARGV_CAST(newargv));
5117 Perl_croak(aTHX_ "Can't exec %s", ipath);
5120 while (*d && !isSPACE(*d))
5122 while (SPACE_OR_TAB(*d))
5126 const bool switches_done = PL_doswitches;
5127 const U32 oldpdb = PL_perldb;
5128 const bool oldn = PL_minus_n;
5129 const bool oldp = PL_minus_p;
5133 bool baduni = FALSE;
5135 const char *d2 = d1 + 1;
5136 if (parse_unicode_opts((const char **)&d2)
5140 if (baduni || *d1 == 'M' || *d1 == 'm') {
5141 const char * const m = d1;
5142 while (*d1 && !isSPACE(*d1))
5144 Perl_croak(aTHX_ "Too late for \"-%.*s\" option",
5147 d1 = moreswitches(d1);
5149 if (PL_doswitches && !switches_done) {
5150 int argc = PL_origargc;
5151 char **argv = PL_origargv;
5154 } while (argc && argv[0][0] == '-' && argv[0][1]);
5155 init_argv_symbols(argc,argv);
5157 if (((PERLDB_LINE || PERLDB_SAVESRC) && !oldpdb) ||
5158 ((PL_minus_n || PL_minus_p) && !(oldn || oldp)))
5159 /* if we have already added "LINE: while (<>) {",
5160 we must not do it again */
5162 sv_setpvs(PL_linestr, "");
5163 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
5164 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
5165 PL_last_lop = PL_last_uni = NULL;
5166 PL_preambled = FALSE;
5167 if (PERLDB_LINE || PERLDB_SAVESRC)
5168 (void)gv_fetchfile(PL_origfilename);
5175 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
5177 PL_lex_state = LEX_FORMLINE;
5182 #ifdef PERL_STRICT_CR
5183 Perl_warn(aTHX_ "Illegal character \\%03o (carriage return)", '\r');
5185 "\t(Maybe you didn't strip carriage returns after a network transfer?)\n");
5187 case ' ': case '\t': case '\f': case 013:
5189 PL_realtokenstart = -1;
5191 PL_thiswhite = newSVpvs("");
5192 sv_catpvn(PL_thiswhite, s, 1);
5199 PL_realtokenstart = -1;
5203 if (PL_lex_state != LEX_NORMAL ||
5204 (PL_in_eval && !PL_rsfp && !PL_parser->filtered)) {
5205 if (*s == '#' && s == PL_linestart && PL_in_eval
5206 && !PL_rsfp && !PL_parser->filtered) {
5207 /* handle eval qq[#line 1 "foo"\n ...] */
5208 CopLINE_dec(PL_curcop);
5211 if (PL_madskills && !PL_lex_formbrack && !PL_in_eval) {
5213 if (!PL_in_eval || PL_rsfp || PL_parser->filtered)
5218 while (d < PL_bufend && *d != '\n')
5222 else if (d > PL_bufend) /* Found by Ilya: feed random input to Perl. */
5223 Perl_croak(aTHX_ "panic: input overflow, %p > %p",
5227 PL_thiswhite = newSVpvn(s, d - s);
5232 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
5234 PL_lex_state = LEX_FORMLINE;
5240 if (PL_madskills && CopLINE(PL_curcop) >= 1 && !PL_lex_formbrack) {
5241 if (CopLINE(PL_curcop) == 1 && s[0] == '#' && s[1] == '!') {
5244 TOKEN(PEG); /* make sure any #! line is accessible */
5249 /* if (PL_madskills && PL_lex_formbrack) { */
5251 while (d < PL_bufend && *d != '\n')
5255 else if (d > PL_bufend) /* Found by Ilya: feed random input to Perl. */
5256 Perl_croak(aTHX_ "panic: input overflow");
5257 if (PL_madskills && CopLINE(PL_curcop) >= 1) {
5259 PL_thiswhite = newSVpvs("");
5260 if (CopLINE(PL_curcop) == 1) {
5261 sv_setpvs(PL_thiswhite, "");
5264 sv_catpvn(PL_thiswhite, s, d - s);
5278 if (s[1] && isALPHA(s[1]) && !isALNUM(s[2])) {
5286 while (s < PL_bufend && SPACE_OR_TAB(*s))
5289 if (strnEQ(s,"=>",2)) {
5290 s = force_word(PL_bufptr,WORD,FALSE,FALSE,FALSE);
5291 DEBUG_T( { printbuf("### Saw unary minus before =>, forcing word %s\n", s); } );
5292 OPERATOR('-'); /* unary minus */
5294 PL_last_uni = PL_oldbufptr;
5296 case 'r': ftst = OP_FTEREAD; break;
5297 case 'w': ftst = OP_FTEWRITE; break;
5298 case 'x': ftst = OP_FTEEXEC; break;
5299 case 'o': ftst = OP_FTEOWNED; break;
5300 case 'R': ftst = OP_FTRREAD; break;
5301 case 'W': ftst = OP_FTRWRITE; break;
5302 case 'X': ftst = OP_FTREXEC; break;
5303 case 'O': ftst = OP_FTROWNED; break;
5304 case 'e': ftst = OP_FTIS; break;
5305 case 'z': ftst = OP_FTZERO; break;
5306 case 's': ftst = OP_FTSIZE; break;
5307 case 'f': ftst = OP_FTFILE; break;
5308 case 'd': ftst = OP_FTDIR; break;
5309 case 'l': ftst = OP_FTLINK; break;
5310 case 'p': ftst = OP_FTPIPE; break;
5311 case 'S': ftst = OP_FTSOCK; break;
5312 case 'u': ftst = OP_FTSUID; break;
5313 case 'g': ftst = OP_FTSGID; break;
5314 case 'k': ftst = OP_FTSVTX; break;
5315 case 'b': ftst = OP_FTBLK; break;
5316 case 'c': ftst = OP_FTCHR; break;
5317 case 't': ftst = OP_FTTTY; break;
5318 case 'T': ftst = OP_FTTEXT; break;
5319 case 'B': ftst = OP_FTBINARY; break;
5320 case 'M': case 'A': case 'C':
5321 gv_fetchpvs("\024", GV_ADD|GV_NOTQUAL, SVt_PV);
5323 case 'M': ftst = OP_FTMTIME; break;
5324 case 'A': ftst = OP_FTATIME; break;
5325 case 'C': ftst = OP_FTCTIME; break;
5333 PL_last_lop_op = (OPCODE)ftst;
5334 DEBUG_T( { PerlIO_printf(Perl_debug_log,
5335 "### Saw file test %c\n", (int)tmp);
5340 /* Assume it was a minus followed by a one-letter named
5341 * subroutine call (or a -bareword), then. */
5342 DEBUG_T( { PerlIO_printf(Perl_debug_log,
5343 "### '-%c' looked like a file test but was not\n",
5350 const char tmp = *s++;
5353 if (PL_expect == XOPERATOR)
5358 else if (*s == '>') {
5361 if (isIDFIRST_lazy_if(s,UTF)) {
5362 s = force_word(s,METHOD,FALSE,TRUE,FALSE);
5370 if (PL_expect == XOPERATOR) {
5371 if (*s == '=' && !PL_lex_allbrackets &&
5372 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
5379 if (isSPACE(*s) || !isSPACE(*PL_bufptr))
5381 OPERATOR('-'); /* unary minus */
5387 const char tmp = *s++;
5390 if (PL_expect == XOPERATOR)
5395 if (PL_expect == XOPERATOR) {
5396 if (*s == '=' && !PL_lex_allbrackets &&
5397 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
5404 if (isSPACE(*s) || !isSPACE(*PL_bufptr))
5411 if (PL_expect != XOPERATOR) {
5412 s = scan_ident(s, PL_bufend, PL_tokenbuf, sizeof PL_tokenbuf, TRUE);
5413 PL_expect = XOPERATOR;
5414 force_ident(PL_tokenbuf, '*');
5422 if (*s == '=' && !PL_lex_allbrackets &&
5423 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
5429 if (*s == '=' && !PL_lex_allbrackets &&
5430 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
5437 if (PL_expect == XOPERATOR) {
5438 if (s[1] == '=' && !PL_lex_allbrackets &&
5439 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN)
5444 PL_tokenbuf[0] = '%';
5445 s = scan_ident(s, PL_bufend, PL_tokenbuf + 1,
5446 sizeof PL_tokenbuf - 1, FALSE);
5447 if (!PL_tokenbuf[1]) {
5450 PL_pending_ident = '%';
5454 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
5455 (s[1] == '=' ? LEX_FAKEEOF_ASSIGN : LEX_FAKEEOF_BITWISE))
5460 if (PL_lex_brackets > 100)
5461 Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
5462 PL_lex_brackstack[PL_lex_brackets++] = 0;
5463 PL_lex_allbrackets++;
5465 const char tmp = *s++;
5470 && (PL_expect == XOPERATOR || PL_expect == XTERMORDORDOR))
5472 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
5480 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMMA)
5487 goto just_a_word_zero_gv;
5490 switch (PL_expect) {
5496 if (!PL_in_my || PL_lex_state != LEX_NORMAL)
5498 PL_bufptr = s; /* update in case we back off */
5501 "Use of := for an empty attribute list is not allowed");
5508 PL_expect = XTERMBLOCK;
5511 stuffstart = s - SvPVX(PL_linestr) - 1;
5515 while (isIDFIRST_lazy_if(s,UTF)) {
5518 d = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &len);
5519 if (isLOWER(*s) && (tmp = keyword(PL_tokenbuf, len, 0))) {
5520 if (tmp < 0) tmp = -tmp;
5535 sv = newSVpvn_flags(s, len, UTF ? SVf_UTF8 : 0);
5537 d = scan_str(d,TRUE,TRUE,FALSE);
5539 /* MUST advance bufptr here to avoid bogus
5540 "at end of line" context messages from yyerror().
5542 PL_bufptr = s + len;
5543 yyerror("Unterminated attribute parameter in attribute list");
5547 return REPORT(0); /* EOF indicator */
5551 sv_catsv(sv, PL_lex_stuff);
5552 attrs = op_append_elem(OP_LIST, attrs,
5553 newSVOP(OP_CONST, 0, sv));
5554 SvREFCNT_dec(PL_lex_stuff);
5555 PL_lex_stuff = NULL;
5558 if (len == 6 && strnEQ(SvPVX(sv), "unique", len)) {
5560 if (PL_in_my == KEY_our) {
5561 deprecate(":unique");
5564 Perl_croak(aTHX_ "The 'unique' attribute may only be applied to 'our' variables");
5567 /* NOTE: any CV attrs applied here need to be part of
5568 the CVf_BUILTIN_ATTRS define in cv.h! */
5569 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "lvalue", len)) {
5571 CvLVALUE_on(PL_compcv);
5573 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "locked", len)) {
5575 deprecate(":locked");
5577 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "method", len)) {
5579 CvMETHOD_on(PL_compcv);
5581 /* After we've set the flags, it could be argued that
5582 we don't need to do the attributes.pm-based setting
5583 process, and shouldn't bother appending recognized
5584 flags. To experiment with that, uncomment the
5585 following "else". (Note that's already been
5586 uncommented. That keeps the above-applied built-in
5587 attributes from being intercepted (and possibly
5588 rejected) by a package's attribute routines, but is
5589 justified by the performance win for the common case
5590 of applying only built-in attributes.) */
5592 attrs = op_append_elem(OP_LIST, attrs,
5593 newSVOP(OP_CONST, 0,
5597 if (*s == ':' && s[1] != ':')
5600 break; /* require real whitespace or :'s */
5601 /* XXX losing whitespace on sequential attributes here */
5605 = (PL_expect == XOPERATOR ? '=' : '{'); /*'}(' for vi */
5606 if (*s != ';' && *s != '}' && *s != tmp
5607 && (tmp != '=' || *s != ')')) {
5608 const char q = ((*s == '\'') ? '"' : '\'');
5609 /* If here for an expression, and parsed no attrs, back
5611 if (tmp == '=' && !attrs) {
5615 /* MUST advance bufptr here to avoid bogus "at end of line"
5616 context messages from yyerror().
5619 yyerror( (const char *)
5621 ? Perl_form(aTHX_ "Invalid separator character "
5622 "%c%c%c in attribute list", q, *s, q)
5623 : "Unterminated attribute list" ) );
5631 start_force(PL_curforce);
5632 NEXTVAL_NEXTTOKE.opval = attrs;
5633 CURMAD('_', PL_nextwhite);
5638 PL_thistoken = newSVpvn(SvPVX(PL_linestr) + stuffstart,
5639 (s - SvPVX(PL_linestr)) - stuffstart);
5644 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_CLOSING) {
5648 PL_lex_allbrackets--;
5652 if (PL_last_lop == PL_oldoldbufptr || PL_last_uni == PL_oldoldbufptr)
5653 PL_oldbufptr = PL_oldoldbufptr; /* allow print(STDOUT 123) */
5657 PL_lex_allbrackets++;
5660 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
5666 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_CLOSING)
5669 PL_lex_allbrackets--;
5675 if (PL_lex_brackets && PL_lex_brackstack[PL_lex_brackets-1] == XFAKEEOF)
5678 if (PL_lex_brackets <= 0)
5679 yyerror("Unmatched right square bracket");
5682 PL_lex_allbrackets--;
5683 if (PL_lex_state == LEX_INTERPNORMAL) {
5684 if (PL_lex_brackets == 0) {
5685 if (*s == '-' && s[1] == '>')
5686 PL_lex_state = LEX_INTERPENDMAYBE;
5687 else if (*s != '[' && *s != '{')
5688 PL_lex_state = LEX_INTERPEND;
5695 if (PL_lex_brackets > 100) {
5696 Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
5698 switch (PL_expect) {
5700 if (PL_lex_formbrack) {
5704 if (PL_oldoldbufptr == PL_last_lop)
5705 PL_lex_brackstack[PL_lex_brackets++] = XTERM;
5707 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
5708 PL_lex_allbrackets++;
5709 OPERATOR(HASHBRACK);
5711 while (s < PL_bufend && SPACE_OR_TAB(*s))
5714 PL_tokenbuf[0] = '\0';
5715 if (d < PL_bufend && *d == '-') {
5716 PL_tokenbuf[0] = '-';
5718 while (d < PL_bufend && SPACE_OR_TAB(*d))
5721 if (d < PL_bufend && isIDFIRST_lazy_if(d,UTF)) {
5722 d = scan_word(d, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1,
5724 while (d < PL_bufend && SPACE_OR_TAB(*d))
5727 const char minus = (PL_tokenbuf[0] == '-');
5728 s = force_word(s + minus, WORD, FALSE, TRUE, FALSE);
5736 PL_lex_brackstack[PL_lex_brackets++] = XSTATE;
5737 PL_lex_allbrackets++;
5742 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
5743 PL_lex_allbrackets++;
5748 if (PL_oldoldbufptr == PL_last_lop)
5749 PL_lex_brackstack[PL_lex_brackets++] = XTERM;
5751 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
5752 PL_lex_allbrackets++;
5755 if (PL_expect == XREF && PL_lex_state == LEX_INTERPNORMAL) {
5757 /* This hack is to get the ${} in the message. */
5759 yyerror("syntax error");
5762 OPERATOR(HASHBRACK);
5764 /* This hack serves to disambiguate a pair of curlies
5765 * as being a block or an anon hash. Normally, expectation
5766 * determines that, but in cases where we're not in a
5767 * position to expect anything in particular (like inside
5768 * eval"") we have to resolve the ambiguity. This code
5769 * covers the case where the first term in the curlies is a
5770 * quoted string. Most other cases need to be explicitly
5771 * disambiguated by prepending a "+" before the opening
5772 * curly in order to force resolution as an anon hash.
5774 * XXX should probably propagate the outer expectation
5775 * into eval"" to rely less on this hack, but that could
5776 * potentially break current behavior of eval"".
5780 if (*s == '\'' || *s == '"' || *s == '`') {
5781 /* common case: get past first string, handling escapes */
5782 for (t++; t < PL_bufend && *t != *s;)
5783 if (*t++ == '\\' && (*t == '\\' || *t == *s))
5787 else if (*s == 'q') {
5790 || ((*t == 'q' || *t == 'x') && ++t < PL_bufend
5793 /* skip q//-like construct */
5795 char open, close, term;
5798 while (t < PL_bufend && isSPACE(*t))
5800 /* check for q => */
5801 if (t+1 < PL_bufend && t[0] == '=' && t[1] == '>') {
5802 OPERATOR(HASHBRACK);
5806 if (term && (tmps = strchr("([{< )]}> )]}>",term)))
5810 for (t++; t < PL_bufend; t++) {
5811 if (*t == '\\' && t+1 < PL_bufend && open != '\\')
5813 else if (*t == open)
5817 for (t++; t < PL_bufend; t++) {
5818 if (*t == '\\' && t+1 < PL_bufend)
5820 else if (*t == close && --brackets <= 0)
5822 else if (*t == open)
5829 /* skip plain q word */
5830 while (t < PL_bufend && isALNUM_lazy_if(t,UTF))
5833 else if (isALNUM_lazy_if(t,UTF)) {
5835 while (t < PL_bufend && isALNUM_lazy_if(t,UTF))
5838 while (t < PL_bufend && isSPACE(*t))
5840 /* if comma follows first term, call it an anon hash */
5841 /* XXX it could be a comma expression with loop modifiers */
5842 if (t < PL_bufend && ((*t == ',' && (*s == 'q' || !isLOWER(*s)))
5843 || (*t == '=' && t[1] == '>')))
5844 OPERATOR(HASHBRACK);
5845 if (PL_expect == XREF)
5848 PL_lex_brackstack[PL_lex_brackets-1] = XSTATE;
5854 pl_yylval.ival = CopLINE(PL_curcop);
5855 if (isSPACE(*s) || *s == '#')
5856 PL_copline = NOLINE; /* invalidate current command line number */
5859 if (PL_lex_brackets && PL_lex_brackstack[PL_lex_brackets-1] == XFAKEEOF)
5863 if (PL_lex_brackets <= 0)
5864 yyerror("Unmatched right curly bracket");
5866 PL_expect = (expectation)PL_lex_brackstack[--PL_lex_brackets];
5867 PL_lex_allbrackets--;
5868 if (PL_lex_brackets < PL_lex_formbrack && PL_lex_state != LEX_INTERPNORMAL)
5869 PL_lex_formbrack = 0;
5870 if (PL_lex_state == LEX_INTERPNORMAL) {
5871 if (PL_lex_brackets == 0) {
5872 if (PL_expect & XFAKEBRACK) {
5873 PL_expect &= XENUMMASK;
5874 PL_lex_state = LEX_INTERPEND;
5879 PL_thiswhite = newSVpvs("");
5880 sv_catpvs(PL_thiswhite,"}");
5883 return yylex(); /* ignore fake brackets */
5885 if (*s == '-' && s[1] == '>')
5886 PL_lex_state = LEX_INTERPENDMAYBE;
5887 else if (*s != '[' && *s != '{')
5888 PL_lex_state = LEX_INTERPEND;
5891 if (PL_expect & XFAKEBRACK) {
5892 PL_expect &= XENUMMASK;
5894 return yylex(); /* ignore fake brackets */
5896 start_force(PL_curforce);
5898 curmad('X', newSVpvn(s-1,1));
5899 CURMAD('_', PL_thiswhite);
5904 PL_thistoken = newSVpvs("");
5910 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
5911 (*s == '=' ? LEX_FAKEEOF_ASSIGN : LEX_FAKEEOF_LOGIC)) {
5918 if (PL_expect == XOPERATOR) {
5919 if (PL_bufptr == PL_linestart && ckWARN(WARN_SEMICOLON)
5920 && isIDFIRST_lazy_if(s,UTF))
5922 CopLINE_dec(PL_curcop);
5923 Perl_warner(aTHX_ packWARN(WARN_SEMICOLON), "%s", PL_warn_nosemi);
5924 CopLINE_inc(PL_curcop);
5926 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
5927 (*s == '=' ? LEX_FAKEEOF_ASSIGN : LEX_FAKEEOF_BITWISE)) {
5934 s = scan_ident(s - 1, PL_bufend, PL_tokenbuf, sizeof PL_tokenbuf, TRUE);
5936 PL_expect = XOPERATOR;
5937 force_ident(PL_tokenbuf, '&');
5941 pl_yylval.ival = (OPpENTERSUB_AMPER<<8);
5947 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
5948 (*s == '=' ? LEX_FAKEEOF_ASSIGN : LEX_FAKEEOF_LOGIC)) {
5955 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
5956 (*s == '=' ? LEX_FAKEEOF_ASSIGN : LEX_FAKEEOF_BITWISE)) {
5964 const char tmp = *s++;
5966 if (!PL_lex_allbrackets &&
5967 PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
5974 if (!PL_lex_allbrackets &&
5975 PL_lex_fakeeof >= LEX_FAKEEOF_COMMA) {
5983 if (tmp && isSPACE(*s) && ckWARN(WARN_SYNTAX)
5984 && strchr("+-*/%.^&|<",tmp))
5985 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
5986 "Reversed %c= operator",(int)tmp);
5988 if (PL_expect == XSTATE && isALPHA(tmp) &&
5989 (s == PL_linestart+1 || s[-2] == '\n') )
5991 if (PL_in_eval && !PL_rsfp && !PL_parser->filtered) {
5996 if (strnEQ(s,"=cut",4)) {
6012 PL_thiswhite = newSVpvs("");
6013 sv_catpvn(PL_thiswhite, PL_linestart,
6014 PL_bufend - PL_linestart);
6018 PL_parser->in_pod = 1;
6022 if (PL_lex_brackets < PL_lex_formbrack) {
6024 #ifdef PERL_STRICT_CR
6025 while (SPACE_OR_TAB(*t))
6027 while (SPACE_OR_TAB(*t) || *t == '\r')
6030 if (*t == '\n' || *t == '#') {
6036 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
6045 const char tmp = *s++;
6047 /* was this !=~ where !~ was meant?
6048 * warn on m:!=~\s+([/?]|[msy]\W|tr\W): */
6050 if (*s == '~' && ckWARN(WARN_SYNTAX)) {
6051 const char *t = s+1;
6053 while (t < PL_bufend && isSPACE(*t))
6056 if (*t == '/' || *t == '?' ||
6057 ((*t == 'm' || *t == 's' || *t == 'y')
6058 && !isALNUM(t[1])) ||
6059 (*t == 't' && t[1] == 'r' && !isALNUM(t[2])))
6060 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
6061 "!=~ should be !~");
6063 if (!PL_lex_allbrackets &&
6064 PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
6076 if (PL_expect != XOPERATOR) {
6077 if (s[1] != '<' && !strchr(s,'>'))
6080 s = scan_heredoc(s);
6082 s = scan_inputsymbol(s);
6083 TERM(sublex_start());
6089 if (*s == '=' && !PL_lex_allbrackets &&
6090 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
6094 SHop(OP_LEFT_SHIFT);
6099 if (!PL_lex_allbrackets &&
6100 PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
6107 if (!PL_lex_allbrackets &&
6108 PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
6116 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
6124 const char tmp = *s++;
6126 if (*s == '=' && !PL_lex_allbrackets &&
6127 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
6131 SHop(OP_RIGHT_SHIFT);
6133 else if (tmp == '=') {
6134 if (!PL_lex_allbrackets &&
6135 PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
6143 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE) {
6152 if (PL_expect == XOPERATOR) {
6153 if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) {
6154 return deprecate_commaless_var_list();
6158 if (s[1] == '#' && (isIDFIRST_lazy_if(s+2,UTF) || strchr("{$:+-@", s[2]))) {
6159 PL_tokenbuf[0] = '@';
6160 s = scan_ident(s + 1, PL_bufend, PL_tokenbuf + 1,
6161 sizeof PL_tokenbuf - 1, FALSE);
6162 if (PL_expect == XOPERATOR)
6163 no_op("Array length", s);
6164 if (!PL_tokenbuf[1])
6166 PL_expect = XOPERATOR;
6167 PL_pending_ident = '#';
6171 PL_tokenbuf[0] = '$';
6172 s = scan_ident(s, PL_bufend, PL_tokenbuf + 1,
6173 sizeof PL_tokenbuf - 1, FALSE);
6174 if (PL_expect == XOPERATOR)
6176 if (!PL_tokenbuf[1]) {
6178 yyerror("Final $ should be \\$ or $name");
6184 const char tmp = *s;
6185 if (PL_lex_state == LEX_NORMAL || PL_lex_brackets)
6188 if ((PL_expect != XREF || PL_oldoldbufptr == PL_last_lop)
6189 && intuit_more(s)) {
6191 PL_tokenbuf[0] = '@';
6192 if (ckWARN(WARN_SYNTAX)) {
6195 while (isSPACE(*t) || isALNUM_lazy_if(t,UTF) || *t == '$')
6198 PL_bufptr = PEEKSPACE(PL_bufptr); /* XXX can realloc */
6199 while (t < PL_bufend && *t != ']')
6201 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
6202 "Multidimensional syntax %.*s not supported",
6203 (int)((t - PL_bufptr) + 1), PL_bufptr);
6207 else if (*s == '{') {
6209 PL_tokenbuf[0] = '%';
6210 if (strEQ(PL_tokenbuf+1, "SIG") && ckWARN(WARN_SYNTAX)
6211 && (t = strchr(s, '}')) && (t = strchr(t, '=')))
6213 char tmpbuf[sizeof PL_tokenbuf];
6216 } while (isSPACE(*t));
6217 if (isIDFIRST_lazy_if(t,UTF)) {
6219 t = scan_word(t, tmpbuf, sizeof tmpbuf, TRUE,
6224 && get_cvn_flags(tmpbuf, len, UTF ? SVf_UTF8 : 0))
6225 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
6226 "You need to quote \"%"SVf"\"",
6227 SVfARG(newSVpvn_flags(tmpbuf, len,
6228 SVs_TEMP | (UTF ? SVf_UTF8 : 0))));
6234 PL_expect = XOPERATOR;
6235 if (PL_lex_state == LEX_NORMAL && isSPACE((char)tmp)) {
6236 const bool islop = (PL_last_lop == PL_oldoldbufptr);
6237 if (!islop || PL_last_lop_op == OP_GREPSTART)
6238 PL_expect = XOPERATOR;
6239 else if (strchr("$@\"'`q", *s))
6240 PL_expect = XTERM; /* e.g. print $fh "foo" */
6241 else if (strchr("&*<%", *s) && isIDFIRST_lazy_if(s+1,UTF))
6242 PL_expect = XTERM; /* e.g. print $fh &sub */
6243 else if (isIDFIRST_lazy_if(s,UTF)) {
6244 char tmpbuf[sizeof PL_tokenbuf];
6246 scan_word(s, tmpbuf, sizeof tmpbuf, TRUE, &len);
6247 if ((t2 = keyword(tmpbuf, len, 0))) {
6248 /* binary operators exclude handle interpretations */
6260 PL_expect = XTERM; /* e.g. print $fh length() */
6265 PL_expect = XTERM; /* e.g. print $fh subr() */
6268 else if (isDIGIT(*s))
6269 PL_expect = XTERM; /* e.g. print $fh 3 */
6270 else if (*s == '.' && isDIGIT(s[1]))
6271 PL_expect = XTERM; /* e.g. print $fh .3 */
6272 else if ((*s == '?' || *s == '-' || *s == '+')
6273 && !isSPACE(s[1]) && s[1] != '=')
6274 PL_expect = XTERM; /* e.g. print $fh -1 */
6275 else if (*s == '/' && !isSPACE(s[1]) && s[1] != '='
6277 PL_expect = XTERM; /* e.g. print $fh /.../
6278 XXX except DORDOR operator
6280 else if (*s == '<' && s[1] == '<' && !isSPACE(s[2])
6282 PL_expect = XTERM; /* print $fh <<"EOF" */
6285 PL_pending_ident = '$';
6289 if (PL_expect == XOPERATOR)
6291 PL_tokenbuf[0] = '@';
6292 s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, FALSE);
6293 if (!PL_tokenbuf[1]) {
6296 if (PL_lex_state == LEX_NORMAL)
6298 if ((PL_expect != XREF || PL_oldoldbufptr == PL_last_lop) && intuit_more(s)) {
6300 PL_tokenbuf[0] = '%';
6302 /* Warn about @ where they meant $. */
6303 if (*s == '[' || *s == '{') {
6304 if (ckWARN(WARN_SYNTAX)) {
6305 const char *t = s + 1;
6306 while (*t && (isALNUM_lazy_if(t,UTF) || strchr(" \t$#+-'\"", *t)))
6307 t += UTF ? UTF8SKIP(t) : 1;
6308 if (*t == '}' || *t == ']') {
6310 PL_bufptr = PEEKSPACE(PL_bufptr); /* XXX can realloc */
6311 /* diag_listed_as: Scalar value @%s[%s] better written as $%s[%s] */
6312 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
6313 "Scalar value %"SVf" better written as $%"SVf,
6314 SVfARG(newSVpvn_flags(PL_bufptr, (STRLEN)(t-PL_bufptr),
6315 SVs_TEMP | (UTF ? SVf_UTF8 : 0 ))),
6316 SVfARG(newSVpvn_flags(PL_bufptr+1, (STRLEN)(t-PL_bufptr-1),
6317 SVs_TEMP | (UTF ? SVf_UTF8 : 0 ))));
6322 PL_pending_ident = '@';
6325 case '/': /* may be division, defined-or, or pattern */
6326 if (PL_expect == XTERMORDORDOR && s[1] == '/') {
6327 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
6328 (s[2] == '=' ? LEX_FAKEEOF_ASSIGN : LEX_FAKEEOF_LOGIC))
6333 case '?': /* may either be conditional or pattern */
6334 if (PL_expect == XOPERATOR) {
6337 if (!PL_lex_allbrackets &&
6338 PL_lex_fakeeof >= LEX_FAKEEOF_IFELSE) {
6342 PL_lex_allbrackets++;
6348 /* A // operator. */
6349 if (!PL_lex_allbrackets && PL_lex_fakeeof >=
6350 (*s == '=' ? LEX_FAKEEOF_ASSIGN :
6351 LEX_FAKEEOF_LOGIC)) {
6359 if (*s == '=' && !PL_lex_allbrackets &&
6360 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
6369 /* Disable warning on "study /blah/" */
6370 if (PL_oldoldbufptr == PL_last_uni
6371 && (*PL_last_uni != 's' || s - PL_last_uni < 5
6372 || memNE(PL_last_uni, "study", 5)
6373 || isALNUM_lazy_if(PL_last_uni+5,UTF)
6377 deprecate("?PATTERN? without explicit operator");
6378 s = scan_pat(s,OP_MATCH);
6379 TERM(sublex_start());
6383 if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack
6384 #ifdef PERL_STRICT_CR
6387 && (s[1] == '\n' || (s[1] == '\r' && s[2] == '\n'))
6389 && (s == PL_linestart || s[-1] == '\n') )
6391 PL_lex_formbrack = 0;
6395 if (PL_expect == XSTATE && s[1] == '.' && s[2] == '.') {
6399 if (PL_expect == XOPERATOR || !isDIGIT(s[1])) {
6402 if (!PL_lex_allbrackets &&
6403 PL_lex_fakeeof >= LEX_FAKEEOF_RANGE) {
6410 pl_yylval.ival = OPf_SPECIAL;
6416 if (*s == '=' && !PL_lex_allbrackets &&
6417 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN) {
6424 case '0': case '1': case '2': case '3': case '4':
6425 case '5': case '6': case '7': case '8': case '9':
6426 s = scan_num(s, &pl_yylval);
6427 DEBUG_T( { printbuf("### Saw number in %s\n", s); } );
6428 if (PL_expect == XOPERATOR)
6433 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
6434 DEBUG_T( { printbuf("### Saw string before %s\n", s); } );
6435 if (PL_expect == XOPERATOR) {
6436 if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) {
6437 return deprecate_commaless_var_list();
6444 pl_yylval.ival = OP_CONST;
6445 TERM(sublex_start());
6448 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
6449 DEBUG_T( { printbuf("### Saw string before %s\n", s); } );
6450 if (PL_expect == XOPERATOR) {
6451 if (PL_lex_formbrack && PL_lex_brackets == PL_lex_formbrack) {
6452 return deprecate_commaless_var_list();
6459 pl_yylval.ival = OP_CONST;
6460 /* FIXME. I think that this can be const if char *d is replaced by
6461 more localised variables. */
6462 for (d = SvPV(PL_lex_stuff, len); len; len--, d++) {
6463 if (*d == '$' || *d == '@' || *d == '\\' || !UTF8_IS_INVARIANT((U8)*d)) {
6464 pl_yylval.ival = OP_STRINGIFY;
6468 TERM(sublex_start());
6471 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
6472 DEBUG_T( { printbuf("### Saw backtick string before %s\n", s); } );
6473 if (PL_expect == XOPERATOR)
6474 no_op("Backticks",s);
6477 readpipe_override();
6478 TERM(sublex_start());
6482 if (PL_lex_inwhat && isDIGIT(*s))
6483 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),"Can't use \\%c to mean $%c in expression",
6485 if (PL_expect == XOPERATOR)
6486 no_op("Backslash",s);
6490 if (isDIGIT(s[1]) && PL_expect != XOPERATOR) {
6491 char *start = s + 2;
6492 while (isDIGIT(*start) || *start == '_')
6494 if (*start == '.' && isDIGIT(start[1])) {
6495 s = scan_num(s, &pl_yylval);
6498 /* avoid v123abc() or $h{v1}, allow C<print v10;> */
6499 else if (!isALPHA(*start) && (PL_expect == XTERM
6500 || PL_expect == XREF || PL_expect == XSTATE
6501 || PL_expect == XTERMORDORDOR)) {
6502 GV *const gv = gv_fetchpvn_flags(s, start - s,
6503 UTF ? SVf_UTF8 : 0, SVt_PVCV);
6505 s = scan_num(s, &pl_yylval);
6512 if (isDIGIT(s[1]) && PL_expect == XOPERATOR) {
6555 s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &len);
6557 /* Some keywords can be followed by any delimiter, including ':' */
6558 anydelim = word_takes_any_delimeter(PL_tokenbuf, len);
6560 /* x::* is just a word, unless x is "CORE" */
6561 if (!anydelim && *s == ':' && s[1] == ':' && strNE(PL_tokenbuf, "CORE"))
6565 while (d < PL_bufend && isSPACE(*d))
6566 d++; /* no comments skipped here, or s### is misparsed */
6568 /* Is this a word before a => operator? */
6569 if (*d == '=' && d[1] == '>') {
6572 = (OP*)newSVOP(OP_CONST, 0,
6573 S_newSV_maybe_utf8(aTHX_ PL_tokenbuf, len));
6574 pl_yylval.opval->op_private = OPpCONST_BARE;
6578 /* Check for plugged-in keyword */
6582 char *saved_bufptr = PL_bufptr;
6584 result = PL_keyword_plugin(aTHX_ PL_tokenbuf, len, &o);
6586 if (result == KEYWORD_PLUGIN_DECLINE) {
6587 /* not a plugged-in keyword */
6588 PL_bufptr = saved_bufptr;
6589 } else if (result == KEYWORD_PLUGIN_STMT) {
6590 pl_yylval.opval = o;
6593 return REPORT(PLUGSTMT);
6594 } else if (result == KEYWORD_PLUGIN_EXPR) {
6595 pl_yylval.opval = o;
6597 PL_expect = XOPERATOR;
6598 return REPORT(PLUGEXPR);
6600 Perl_croak(aTHX_ "Bad plugin affecting keyword '%s'",
6605 /* Check for built-in keyword */
6606 tmp = keyword(PL_tokenbuf, len, 0);
6608 /* Is this a label? */
6609 if (!anydelim && PL_expect == XSTATE
6610 && d < PL_bufend && *d == ':' && *(d + 1) != ':') {
6612 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0,
6613 newSVpvn_flags(PL_tokenbuf,
6614 len, UTF ? SVf_UTF8 : 0));
6619 if (tmp < 0) { /* second-class keyword? */
6620 GV *ogv = NULL; /* override (winner) */
6621 GV *hgv = NULL; /* hidden (loser) */
6622 if (PL_expect != XOPERATOR && (*s != ':' || s[1] != ':')) {
6624 if ((gv = gv_fetchpvn_flags(PL_tokenbuf, len,
6625 UTF ? SVf_UTF8 : 0, SVt_PVCV)) &&
6628 if (GvIMPORTED_CV(gv))
6630 else if (! CvMETHOD(cv))
6634 (gvp = (GV**)hv_fetch(PL_globalstash, PL_tokenbuf,
6635 UTF ? -(I32)len : (I32)len, FALSE)) &&
6636 (gv = *gvp) && isGV_with_GP(gv) &&
6637 GvCVu(gv) && GvIMPORTED_CV(gv))
6644 tmp = 0; /* overridden by import or by GLOBAL */
6647 && -tmp==KEY_lock /* XXX generalizable kludge */
6650 tmp = 0; /* any sub overrides "weak" keyword */
6652 else { /* no override */
6654 if (tmp == KEY_dump) {
6655 Perl_ck_warner(aTHX_ packWARN(WARN_MISC),
6656 "dump() better written as CORE::dump()");
6660 if (hgv && tmp != KEY_x && tmp != KEY_CORE) /* never ambiguous */
6661 Perl_ck_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
6662 "Ambiguous call resolved as CORE::%s(), "
6663 "qualify as such or use &",
6671 default: /* not a keyword */
6672 /* Trade off - by using this evil construction we can pull the
6673 variable gv into the block labelled keylookup. If not, then
6674 we have to give it function scope so that the goto from the
6675 earlier ':' case doesn't bypass the initialisation. */
6677 just_a_word_zero_gv:
6685 const char lastchar = (PL_bufptr == PL_oldoldbufptr ? 0 : PL_bufptr[-1]);
6689 SV *nextPL_nextwhite = 0;
6693 /* Get the rest if it looks like a package qualifier */
6695 if (*s == '\'' || (*s == ':' && s[1] == ':')) {
6697 s = scan_word(s, PL_tokenbuf + len, sizeof PL_tokenbuf - len,
6700 Perl_croak(aTHX_ "Bad name after %"SVf"%s",
6701 SVfARG(newSVpvn_flags(PL_tokenbuf, len,
6702 (UTF ? SVf_UTF8 : 0) | SVs_TEMP )),
6703 *s == '\'' ? "'" : "::");
6708 if (PL_expect == XOPERATOR) {
6709 if (PL_bufptr == PL_linestart) {
6710 CopLINE_dec(PL_curcop);
6711 Perl_warner(aTHX_ packWARN(WARN_SEMICOLON), "%s", PL_warn_nosemi);
6712 CopLINE_inc(PL_curcop);
6715 no_op("Bareword",s);
6718 /* Look for a subroutine with this name in current package,
6719 unless name is "Foo::", in which case Foo is a bareword
6720 (and a package name). */
6722 if (len > 2 && !PL_madskills &&
6723 PL_tokenbuf[len - 2] == ':' && PL_tokenbuf[len - 1] == ':')
6725 if (ckWARN(WARN_BAREWORD)
6726 && ! gv_fetchpvn_flags(PL_tokenbuf, len, UTF ? SVf_UTF8 : 0, SVt_PVHV))
6727 Perl_warner(aTHX_ packWARN(WARN_BAREWORD),
6728 "Bareword \"%"SVf"\" refers to nonexistent package",
6729 SVfARG(newSVpvn_flags(PL_tokenbuf, len,
6730 (UTF ? SVf_UTF8 : 0) | SVs_TEMP)));
6732 PL_tokenbuf[len] = '\0';
6738 /* Mustn't actually add anything to a symbol table.
6739 But also don't want to "initialise" any placeholder
6740 constants that might already be there into full
6741 blown PVGVs with attached PVCV. */
6742 gv = gv_fetchpvn_flags(PL_tokenbuf, len,
6743 GV_NOADD_NOINIT | ( UTF ? SVf_UTF8 : 0 ),
6749 /* if we saw a global override before, get the right name */
6751 sv = S_newSV_maybe_utf8(aTHX_ PL_tokenbuf,
6752 len ? len : strlen(PL_tokenbuf));
6754 SV * const tmp_sv = sv;
6755 sv = newSVpvs("CORE::GLOBAL::");
6756 sv_catsv(sv, tmp_sv);
6757 SvREFCNT_dec(tmp_sv);
6761 if (PL_madskills && !PL_thistoken) {
6762 char *start = SvPVX(PL_linestr) + PL_realtokenstart;
6763 PL_thistoken = newSVpvn(start,s - start);
6764 PL_realtokenstart = s - SvPVX(PL_linestr);
6768 /* Presume this is going to be a bareword of some sort. */
6770 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
6771 pl_yylval.opval->op_private = OPpCONST_BARE;
6773 /* And if "Foo::", then that's what it certainly is. */
6778 OP *const_op = newSVOP(OP_CONST, 0, SvREFCNT_inc_NN(sv));
6779 const_op->op_private = OPpCONST_BARE;
6780 rv2cv_op = newCVREF(0, const_op);
6782 cv = rv2cv_op_cv(rv2cv_op, 0);
6784 /* See if it's the indirect object for a list operator. */
6786 if (PL_oldoldbufptr &&
6787 PL_oldoldbufptr < PL_bufptr &&
6788 (PL_oldoldbufptr == PL_last_lop
6789 || PL_oldoldbufptr == PL_last_uni) &&
6790 /* NO SKIPSPACE BEFORE HERE! */
6791 (PL_expect == XREF ||
6792 ((PL_opargs[PL_last_lop_op] >> OASHIFT)& 7) == OA_FILEREF))
6794 bool immediate_paren = *s == '(';
6796 /* (Now we can afford to cross potential line boundary.) */
6797 s = SKIPSPACE2(s,nextPL_nextwhite);
6799 PL_nextwhite = nextPL_nextwhite; /* assume no & deception */
6802 /* Two barewords in a row may indicate method call. */
6804 if ((isIDFIRST_lazy_if(s,UTF) || *s == '$') &&
6805 (tmp = intuit_method(s, gv, cv))) {
6807 if (tmp == METHOD && !PL_lex_allbrackets &&
6808 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
6809 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
6813 /* If not a declared subroutine, it's an indirect object. */
6814 /* (But it's an indir obj regardless for sort.) */
6815 /* Also, if "_" follows a filetest operator, it's a bareword */
6818 ( !immediate_paren && (PL_last_lop_op == OP_SORT ||
6820 (PL_last_lop_op != OP_MAPSTART &&
6821 PL_last_lop_op != OP_GREPSTART))))
6822 || (PL_tokenbuf[0] == '_' && PL_tokenbuf[1] == '\0'
6823 && ((PL_opargs[PL_last_lop_op] & OA_CLASS_MASK) == OA_FILESTATOP))
6826 PL_expect = (PL_last_lop == PL_oldoldbufptr) ? XTERM : XOPERATOR;
6831 PL_expect = XOPERATOR;
6834 s = SKIPSPACE2(s,nextPL_nextwhite);
6835 PL_nextwhite = nextPL_nextwhite;
6840 /* Is this a word before a => operator? */
6841 if (*s == '=' && s[1] == '>' && !pkgname) {
6844 sv_setpv(((SVOP*)pl_yylval.opval)->op_sv, PL_tokenbuf);
6845 if (UTF && !IN_BYTES && is_utf8_string((U8*)PL_tokenbuf, len))
6846 SvUTF8_on(((SVOP*)pl_yylval.opval)->op_sv);
6850 /* If followed by a paren, it's certainly a subroutine. */
6855 while (SPACE_OR_TAB(*d))
6857 if (*d == ')' && (sv = cv_const_sv(cv))) {
6864 PL_nextwhite = PL_thiswhite;
6867 start_force(PL_curforce);
6869 NEXTVAL_NEXTTOKE.opval = pl_yylval.opval;
6870 PL_expect = XOPERATOR;
6873 PL_nextwhite = nextPL_nextwhite;
6874 curmad('X', PL_thistoken);
6875 PL_thistoken = newSVpvs("");
6884 /* If followed by var or block, call it a method (unless sub) */
6886 if ((*s == '$' || *s == '{') && !cv) {
6888 PL_last_lop = PL_oldbufptr;
6889 PL_last_lop_op = OP_METHOD;
6890 if (!PL_lex_allbrackets &&
6891 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
6892 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
6896 /* If followed by a bareword, see if it looks like indir obj. */
6899 && (isIDFIRST_lazy_if(s,UTF) || *s == '$')
6900 && (tmp = intuit_method(s, gv, cv))) {
6902 if (tmp == METHOD && !PL_lex_allbrackets &&
6903 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
6904 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
6908 /* Not a method, so call it a subroutine (if defined) */
6911 if (lastchar == '-') {
6912 const SV *tmpsv = newSVpvn_flags( PL_tokenbuf, len ? len : strlen(PL_tokenbuf), (UTF ? SVf_UTF8 : 0) | SVs_TEMP );
6913 Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS),
6914 "Ambiguous use of -%"SVf" resolved as -&%"SVf"()",
6915 SVfARG(tmpsv), SVfARG(tmpsv));
6917 /* Check for a constant sub */
6918 if ((sv = cv_const_sv(cv))) {
6921 SvREFCNT_dec(((SVOP*)pl_yylval.opval)->op_sv);
6922 ((SVOP*)pl_yylval.opval)->op_sv = SvREFCNT_inc_simple(sv);
6923 pl_yylval.opval->op_private = 0;
6924 pl_yylval.opval->op_flags |= OPf_SPECIAL;
6928 op_free(pl_yylval.opval);
6929 pl_yylval.opval = rv2cv_op;
6930 pl_yylval.opval->op_private |= OPpENTERSUB_NOPAREN;
6931 PL_last_lop = PL_oldbufptr;
6932 PL_last_lop_op = OP_ENTERSUB;
6933 /* Is there a prototype? */
6940 STRLEN protolen = CvPROTOLEN(cv);
6941 const char *proto = CvPROTO(cv);
6945 if ((optional = *proto == ';'))
6948 while (*proto == ';');
6952 *proto == '$' || *proto == '_'
6953 || *proto == '*' || *proto == '+'
6958 *proto == '\\' && proto[1] && proto[2] == '\0'
6961 UNIPROTO(UNIOPSUB,optional);
6962 if (*proto == '\\' && proto[1] == '[') {
6963 const char *p = proto + 2;
6964 while(*p && *p != ']')
6966 if(*p == ']' && !p[1])
6967 UNIPROTO(UNIOPSUB,optional);
6969 if (*proto == '&' && *s == '{') {
6971 sv_setpvs(PL_subname, "__ANON__");
6973 sv_setpvs(PL_subname, "__ANON__::__ANON__");
6974 if (!PL_lex_allbrackets &&
6975 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
6976 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
6983 PL_nextwhite = PL_thiswhite;
6986 start_force(PL_curforce);
6987 NEXTVAL_NEXTTOKE.opval = pl_yylval.opval;
6990 PL_nextwhite = nextPL_nextwhite;
6991 curmad('X', PL_thistoken);
6992 PL_thistoken = newSVpvs("");
6995 if (!PL_lex_allbrackets &&
6996 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
6997 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
7002 /* Guess harder when madskills require "best effort". */
7003 if (PL_madskills && (!gv || !GvCVu(gv))) {
7004 int probable_sub = 0;
7005 if (strchr("\"'`$@%0123456789!*+{[<", *s))
7007 else if (isALPHA(*s)) {
7011 d = scan_word(d, tmpbuf, sizeof tmpbuf, TRUE, &tmplen);
7012 if (!keyword(tmpbuf, tmplen, 0))
7015 while (d < PL_bufend && isSPACE(*d))
7017 if (*d == '=' && d[1] == '>')
7022 gv = gv_fetchpv(PL_tokenbuf, GV_ADD | ( UTF ? SVf_UTF8 : 0 ),
7024 op_free(pl_yylval.opval);
7025 pl_yylval.opval = rv2cv_op;
7026 pl_yylval.opval->op_private |= OPpENTERSUB_NOPAREN;
7027 PL_last_lop = PL_oldbufptr;
7028 PL_last_lop_op = OP_ENTERSUB;
7029 PL_nextwhite = PL_thiswhite;
7031 start_force(PL_curforce);
7032 NEXTVAL_NEXTTOKE.opval = pl_yylval.opval;
7034 PL_nextwhite = nextPL_nextwhite;
7035 curmad('X', PL_thistoken);
7036 PL_thistoken = newSVpvs("");
7038 if (!PL_lex_allbrackets &&
7039 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
7040 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
7044 NEXTVAL_NEXTTOKE.opval = pl_yylval.opval;
7047 if (!PL_lex_allbrackets &&
7048 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
7049 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
7054 /* Call it a bare word */
7056 if (PL_hints & HINT_STRICT_SUBS)
7057 pl_yylval.opval->op_private |= OPpCONST_STRICT;
7060 /* after "print" and similar functions (corresponding to
7061 * "F? L" in opcode.pl), whatever wasn't already parsed as
7062 * a filehandle should be subject to "strict subs".
7063 * Likewise for the optional indirect-object argument to system
7064 * or exec, which can't be a bareword */
7065 if ((PL_last_lop_op == OP_PRINT
7066 || PL_last_lop_op == OP_PRTF
7067 || PL_last_lop_op == OP_SAY
7068 || PL_last_lop_op == OP_SYSTEM
7069 || PL_last_lop_op == OP_EXEC)
7070 && (PL_hints & HINT_STRICT_SUBS))
7071 pl_yylval.opval->op_private |= OPpCONST_STRICT;
7072 if (lastchar != '-') {
7073 if (ckWARN(WARN_RESERVED)) {
7077 if (!*d && !gv_stashpv(PL_tokenbuf, UTF ? SVf_UTF8 : 0))
7078 Perl_warner(aTHX_ packWARN(WARN_RESERVED), PL_warn_reserved,
7086 if ((lastchar == '*' || lastchar == '%' || lastchar == '&')) {
7087 Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS),
7088 "Operator or semicolon missing before %c%"SVf,
7089 lastchar, SVfARG(newSVpvn_flags(PL_tokenbuf,
7090 strlen(PL_tokenbuf),
7091 SVs_TEMP | (UTF ? SVf_UTF8 : 0))));
7092 Perl_ck_warner_d(aTHX_ packWARN(WARN_AMBIGUOUS),
7093 "Ambiguous use of %c resolved as operator %c",
7094 lastchar, lastchar);
7101 (OP*)newSVOP(OP_CONST, 0, newSVpv(CopFILE(PL_curcop),0))
7106 (OP*)newSVOP(OP_CONST, 0,
7107 Perl_newSVpvf(aTHX_ "%"IVdf, (IV)CopLINE(PL_curcop)))
7110 case KEY___PACKAGE__:
7112 (OP*)newSVOP(OP_CONST, 0,
7114 ? newSVhek(HvNAME_HEK(PL_curstash))
7121 if (PL_rsfp && (!PL_in_eval || PL_tokenbuf[2] == 'D')) {
7122 const char *pname = "main";
7125 if (PL_tokenbuf[2] == 'D')
7128 PL_curstash ? PL_curstash : PL_defstash;
7129 pname = HvNAME_get(stash);
7130 plen = HvNAMELEN (stash);
7131 if(HvNAMEUTF8(stash)) putf8 = SVf_UTF8;
7133 gv = gv_fetchpvn_flags(
7134 Perl_form(aTHX_ "%*s::DATA", (int)plen, pname),
7135 plen+6, GV_ADD|putf8, SVt_PVIO
7139 GvIOp(gv) = newIO();
7140 IoIFP(GvIOp(gv)) = PL_rsfp;
7141 #if defined(HAS_FCNTL) && defined(F_SETFD)
7143 const int fd = PerlIO_fileno(PL_rsfp);
7144 fcntl(fd,F_SETFD,fd >= 3);
7147 /* Mark this internal pseudo-handle as clean */
7148 IoFLAGS(GvIOp(gv)) |= IOf_UNTAINT;
7149 if ((PerlIO*)PL_rsfp == PerlIO_stdin())
7150 IoTYPE(GvIOp(gv)) = IoTYPE_STD;
7152 IoTYPE(GvIOp(gv)) = IoTYPE_RDONLY;
7153 #if defined(WIN32) && !defined(PERL_TEXTMODE_SCRIPTS)
7154 /* if the script was opened in binmode, we need to revert
7155 * it to text mode for compatibility; but only iff it has CRs
7156 * XXX this is a questionable hack at best. */
7157 if (PL_bufend-PL_bufptr > 2
7158 && PL_bufend[-1] == '\n' && PL_bufend[-2] == '\r')
7161 if (IoTYPE(GvIOp(gv)) == IoTYPE_RDONLY) {
7162 loc = PerlIO_tell(PL_rsfp);
7163 (void)PerlIO_seek(PL_rsfp, 0L, 0);
7166 if (PerlLIO_setmode(PL_rsfp, O_TEXT) != -1) {
7168 if (PerlLIO_setmode(PerlIO_fileno(PL_rsfp), O_TEXT) != -1) {
7169 #endif /* NETWARE */
7171 PerlIO_seek(PL_rsfp, loc, 0);
7175 #ifdef PERLIO_LAYERS
7178 PerlIO_apply_layers(aTHX_ PL_rsfp, NULL, ":utf8");
7179 else if (PL_encoding) {
7186 XPUSHs(PL_encoding);
7188 call_method("name", G_SCALAR);
7192 PerlIO_apply_layers(aTHX_ PL_rsfp, NULL,
7193 Perl_form(aTHX_ ":encoding(%"SVf")",
7202 if (PL_realtokenstart >= 0) {
7203 char *tstart = SvPVX(PL_linestr) + PL_realtokenstart;
7205 PL_endwhite = newSVpvs("");
7206 sv_catsv(PL_endwhite, PL_thiswhite);
7208 sv_catpvn(PL_endwhite, tstart, PL_bufend - tstart);
7209 PL_realtokenstart = -1;
7211 while ((s = filter_gets(PL_endwhite, SvCUR(PL_endwhite)))
7221 FUN0OP(newPVOP(OP_RUNCV,0,NULL));
7230 if (PL_expect == XSTATE) {
7237 if (*s == ':' && s[1] == ':') {
7240 s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &len);
7241 if (!(tmp = keyword(PL_tokenbuf, len, 1)))
7242 Perl_croak(aTHX_ "CORE::%"SVf" is not a keyword",
7243 SVfARG(newSVpvn_flags(PL_tokenbuf, len,
7244 (UTF ? SVf_UTF8 : 0) | SVs_TEMP)));
7247 else if (tmp == KEY_require || tmp == KEY_do
7249 /* that's a way to remember we saw "CORE::" */
7262 LOP(OP_ACCEPT,XTERM);
7265 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_LOWLOGIC)
7270 LOP(OP_ATAN2,XTERM);
7276 LOP(OP_BINMODE,XTERM);
7279 LOP(OP_BLESS,XTERM);
7288 /* We have to disambiguate the two senses of
7289 "continue". If the next token is a '{' then
7290 treat it as the start of a continue block;
7291 otherwise treat it as a control operator.
7301 (void)gv_fetchpvs("ENV", GV_ADD|GV_NOTQUAL, SVt_PVHV);
7311 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7320 if (!PL_cryptseen) {
7321 PL_cryptseen = TRUE;
7325 LOP(OP_CRYPT,XTERM);
7328 LOP(OP_CHMOD,XTERM);
7331 LOP(OP_CHOWN,XTERM);
7334 LOP(OP_CONNECT,XTERM);
7353 s = force_word(s,WORD,TRUE,TRUE,FALSE);
7354 if (orig_keyword == KEY_do) {
7363 PL_hints |= HINT_BLOCK_SCOPE;
7373 Perl_populate_isa(aTHX_ STR_WITH_LEN("AnyDBM_File::ISA"),
7374 STR_WITH_LEN("NDBM_File::"),
7375 STR_WITH_LEN("DB_File::"),
7376 STR_WITH_LEN("GDBM_File::"),
7377 STR_WITH_LEN("SDBM_File::"),
7378 STR_WITH_LEN("ODBM_File::"),
7380 LOP(OP_DBMOPEN,XTERM);
7386 s = force_word(s,WORD,TRUE,FALSE,FALSE);
7393 pl_yylval.ival = CopLINE(PL_curcop);
7397 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7411 if (*s == '{') { /* block eval */
7412 PL_expect = XTERMBLOCK;
7413 UNIBRACK(OP_ENTERTRY);
7415 else { /* string eval */
7417 UNIBRACK(OP_ENTEREVAL);
7422 UNIBRACK(-OP_ENTEREVAL);
7436 case KEY_endhostent:
7442 case KEY_endservent:
7445 case KEY_endprotoent:
7456 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
7458 pl_yylval.ival = CopLINE(PL_curcop);
7460 if (PL_expect == XSTATE && isIDFIRST_lazy_if(s,UTF)) {
7463 int soff = s - SvPVX(PL_linestr); /* for skipspace realloc */
7466 if ((PL_bufend - p) >= 3 &&
7467 strnEQ(p, "my", 2) && isSPACE(*(p + 2)))
7469 else if ((PL_bufend - p) >= 4 &&
7470 strnEQ(p, "our", 3) && isSPACE(*(p + 3)))
7473 if (isIDFIRST_lazy_if(p,UTF)) {
7474 p = scan_ident(p, PL_bufend,
7475 PL_tokenbuf, sizeof PL_tokenbuf, TRUE);
7479 Perl_croak(aTHX_ "Missing $ on loop variable");
7481 s = SvPVX(PL_linestr) + soff;
7487 LOP(OP_FORMLINE,XTERM);
7496 LOP(OP_FCNTL,XTERM);
7502 LOP(OP_FLOCK,XTERM);
7505 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7510 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7515 LOP(OP_GREPSTART, XREF);
7518 s = force_word(s,WORD,TRUE,FALSE,FALSE);
7533 case KEY_getpriority:
7534 LOP(OP_GETPRIORITY,XTERM);
7536 case KEY_getprotobyname:
7539 case KEY_getprotobynumber:
7540 LOP(OP_GPBYNUMBER,XTERM);
7542 case KEY_getprotoent:
7554 case KEY_getpeername:
7555 UNI(OP_GETPEERNAME);
7557 case KEY_gethostbyname:
7560 case KEY_gethostbyaddr:
7561 LOP(OP_GHBYADDR,XTERM);
7563 case KEY_gethostent:
7566 case KEY_getnetbyname:
7569 case KEY_getnetbyaddr:
7570 LOP(OP_GNBYADDR,XTERM);
7575 case KEY_getservbyname:
7576 LOP(OP_GSBYNAME,XTERM);
7578 case KEY_getservbyport:
7579 LOP(OP_GSBYPORT,XTERM);
7581 case KEY_getservent:
7584 case KEY_getsockname:
7585 UNI(OP_GETSOCKNAME);
7587 case KEY_getsockopt:
7588 LOP(OP_GSOCKOPT,XTERM);
7603 pl_yylval.ival = CopLINE(PL_curcop);
7608 orig_keyword==KEY_glob ? (orig_keyword=0, -OP_GLOB) : OP_GLOB,
7616 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
7618 pl_yylval.ival = CopLINE(PL_curcop);
7622 LOP(OP_INDEX,XTERM);
7628 LOP(OP_IOCTL,XTERM);
7640 s = force_word(s,WORD,TRUE,FALSE,FALSE);
7657 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7662 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7676 LOP(OP_LISTEN,XTERM);
7685 s = scan_pat(s,OP_MATCH);
7686 TERM(sublex_start());
7689 LOP(OP_MAPSTART, XREF);
7692 LOP(OP_MKDIR,XTERM);
7695 LOP(OP_MSGCTL,XTERM);
7698 LOP(OP_MSGGET,XTERM);
7701 LOP(OP_MSGRCV,XTERM);
7704 LOP(OP_MSGSND,XTERM);
7709 PL_in_my = (U16)tmp;
7711 if (isIDFIRST_lazy_if(s,UTF)) {
7715 s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, TRUE, &len);
7716 if (len == 3 && strnEQ(PL_tokenbuf, "sub", 3))
7718 PL_in_my_stash = find_in_my_stash(PL_tokenbuf, len);
7719 if (!PL_in_my_stash) {
7722 my_snprintf(tmpbuf, sizeof(tmpbuf), "No such class %.1000s", PL_tokenbuf);
7723 yyerror_pv(tmpbuf, UTF ? SVf_UTF8 : 0);
7726 if (PL_madskills) { /* just add type to declarator token */
7727 sv_catsv(PL_thistoken, PL_nextwhite);
7729 sv_catpvn(PL_thistoken, start, s - start);
7737 s = force_word(s,WORD,TRUE,FALSE,FALSE);
7741 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_COMPARE)
7746 s = tokenize_use(0, s);
7750 if (*s == '(' || (s = SKIPSPACE1(s), *s == '('))
7753 if (!PL_lex_allbrackets &&
7754 PL_lex_fakeeof > LEX_FAKEEOF_LOWLOGIC)
7755 PL_lex_fakeeof = LEX_FAKEEOF_LOWLOGIC;
7761 if (isIDFIRST_lazy_if(s,UTF)) {
7763 for (d = s; isALNUM_lazy_if(d,UTF);) {
7764 d += UTF ? UTF8SKIP(d) : 1;
7766 while (UTF8_IS_CONTINUED(*d) && is_utf8_mark((U8*)d)) {
7767 d += UTF ? UTF8SKIP(d) : 1;
7771 for (t=d; isSPACE(*t);)
7773 if ( *t && strchr("|&*+-=!?:.", *t) && ckWARN_d(WARN_PRECEDENCE)
7775 && !(t[0] == '=' && t[1] == '>')
7776 && !(t[0] == ':' && t[1] == ':')
7777 && !keyword(s, d-s, 0)
7779 SV *tmpsv = newSVpvn_flags(s, (STRLEN)(d-s),
7780 SVs_TEMP | (UTF ? SVf_UTF8 : 0));
7781 Perl_warner(aTHX_ packWARN(WARN_PRECEDENCE),
7782 "Precedence problem: open %"SVf" should be open(%"SVf")",
7783 SVfARG(tmpsv), SVfARG(tmpsv));
7789 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_LOWLOGIC)
7791 pl_yylval.ival = OP_OR;
7801 LOP(OP_OPEN_DIR,XTERM);
7804 checkcomma(s,PL_tokenbuf,"filehandle");
7808 checkcomma(s,PL_tokenbuf,"filehandle");
7827 s = force_word(s,WORD,FALSE,TRUE,FALSE);
7829 s = force_strict_version(s);
7830 PL_lex_expect = XBLOCK;
7834 LOP(OP_PIPE_OP,XTERM);
7837 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
7840 pl_yylval.ival = OP_CONST;
7841 TERM(sublex_start());
7848 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
7851 PL_expect = XOPERATOR;
7852 if (SvCUR(PL_lex_stuff)) {
7853 int warned_comma = !ckWARN(WARN_QW);
7854 int warned_comment = warned_comma;
7855 d = SvPV_force(PL_lex_stuff, len);
7857 for (; isSPACE(*d) && len; --len, ++d)
7862 if (!warned_comma || !warned_comment) {
7863 for (; !isSPACE(*d) && len; --len, ++d) {
7864 if (!warned_comma && *d == ',') {
7865 Perl_warner(aTHX_ packWARN(WARN_QW),
7866 "Possible attempt to separate words with commas");
7869 else if (!warned_comment && *d == '#') {
7870 Perl_warner(aTHX_ packWARN(WARN_QW),
7871 "Possible attempt to put comments in qw() list");
7877 for (; !isSPACE(*d) && len; --len, ++d)
7880 sv = newSVpvn_utf8(b, d-b, DO_UTF8(PL_lex_stuff));
7881 words = op_append_elem(OP_LIST, words,
7882 newSVOP(OP_CONST, 0, tokeq(sv)));
7887 words = newNULLLIST();
7889 SvREFCNT_dec(PL_lex_stuff);
7890 PL_lex_stuff = NULL;
7892 PL_expect = XOPERATOR;
7893 pl_yylval.opval = sawparens(words);
7898 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
7901 pl_yylval.ival = OP_STRINGIFY;
7902 if (SvIVX(PL_lex_stuff) == '\'')
7903 SvIV_set(PL_lex_stuff, 0); /* qq'$foo' should interpolate */
7904 TERM(sublex_start());
7907 s = scan_pat(s,OP_QR);
7908 TERM(sublex_start());
7911 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
7914 readpipe_override();
7915 TERM(sublex_start());
7923 s = force_version(s, FALSE);
7925 else if (*s != 'v' || !isDIGIT(s[1])
7926 || (s = force_version(s, TRUE), *s == 'v'))
7928 *PL_tokenbuf = '\0';
7929 s = force_word(s,WORD,TRUE,TRUE,FALSE);
7930 if (isIDFIRST_lazy_if(PL_tokenbuf,UTF))
7931 gv_stashpvn(PL_tokenbuf, strlen(PL_tokenbuf),
7932 GV_ADD | (UTF ? SVf_UTF8 : 0));
7934 yyerror("<> should be quotes");
7936 if (orig_keyword == KEY_require) {
7944 PL_last_uni = PL_oldbufptr;
7945 PL_last_lop_op = OP_REQUIRE;
7947 return REPORT( (int)REQUIRE );
7953 s = force_word(s,WORD,TRUE,FALSE,FALSE);
7957 LOP(OP_RENAME,XTERM);
7966 LOP(OP_RINDEX,XTERM);
7975 UNIDOR(OP_READLINE);
7978 UNIDOR(OP_BACKTICK);
7987 LOP(OP_REVERSE,XTERM);
7990 UNIDOR(OP_READLINK);
7997 if (pl_yylval.opval)
7998 TERM(sublex_start());
8000 TOKEN(1); /* force error */
8003 checkcomma(s,PL_tokenbuf,"filehandle");
8013 LOP(OP_SELECT,XTERM);
8019 LOP(OP_SEMCTL,XTERM);
8022 LOP(OP_SEMGET,XTERM);
8025 LOP(OP_SEMOP,XTERM);
8031 LOP(OP_SETPGRP,XTERM);
8033 case KEY_setpriority:
8034 LOP(OP_SETPRIORITY,XTERM);
8036 case KEY_sethostent:
8042 case KEY_setservent:
8045 case KEY_setprotoent:
8055 LOP(OP_SEEKDIR,XTERM);
8057 case KEY_setsockopt:
8058 LOP(OP_SSOCKOPT,XTERM);
8064 LOP(OP_SHMCTL,XTERM);
8067 LOP(OP_SHMGET,XTERM);
8070 LOP(OP_SHMREAD,XTERM);
8073 LOP(OP_SHMWRITE,XTERM);
8076 LOP(OP_SHUTDOWN,XTERM);
8085 LOP(OP_SOCKET,XTERM);
8087 case KEY_socketpair:
8088 LOP(OP_SOCKPAIR,XTERM);
8091 checkcomma(s,PL_tokenbuf,"subroutine name");
8094 s = force_word(s,WORD,TRUE,TRUE,FALSE);
8098 LOP(OP_SPLIT,XTERM);
8101 LOP(OP_SPRINTF,XTERM);
8104 LOP(OP_SPLICE,XTERM);
8119 LOP(OP_SUBSTR,XTERM);
8125 char tmpbuf[sizeof PL_tokenbuf];
8126 SSize_t tboffset = 0;
8127 expectation attrful;
8128 bool have_name, have_proto;
8129 const int key = tmp;
8134 char *tstart = SvPVX(PL_linestr) + PL_realtokenstart;
8135 SV *subtoken = newSVpvn_flags(tstart, s - tstart, SvUTF8(PL_linestr));
8139 s = SKIPSPACE2(s,tmpwhite);
8144 if (isIDFIRST_lazy_if(s,UTF) || *s == '\'' ||
8145 (*s == ':' && s[1] == ':'))
8148 SV *nametoke = NULL;
8152 attrful = XATTRBLOCK;
8153 /* remember buffer pos'n for later force_word */
8154 tboffset = s - PL_oldbufptr;
8155 d = scan_word(s, tmpbuf, sizeof tmpbuf, TRUE, &len);
8158 nametoke = newSVpvn_flags(s, d - s, SvUTF8(PL_linestr));
8160 if (memchr(tmpbuf, ':', len))
8161 sv_setpvn(PL_subname, tmpbuf, len);
8163 sv_setsv(PL_subname,PL_curstname);
8164 sv_catpvs(PL_subname,"::");
8165 sv_catpvn(PL_subname,tmpbuf,len);
8167 if (SvUTF8(PL_linestr))
8168 SvUTF8_on(PL_subname);
8174 CURMAD('X', nametoke);
8175 CURMAD('_', tmpwhite);
8176 (void) force_word(PL_oldbufptr + tboffset, WORD,
8179 s = SKIPSPACE2(d,tmpwhite);
8186 Perl_croak(aTHX_ "Missing name in \"my sub\"");
8187 PL_expect = XTERMBLOCK;
8188 attrful = XATTRTERM;
8189 sv_setpvs(PL_subname,"?");
8193 if (key == KEY_format) {
8195 PL_lex_formbrack = PL_lex_brackets + 1;
8197 PL_thistoken = subtoken;
8201 (void) force_word(PL_oldbufptr + tboffset, WORD,
8207 /* Look for a prototype */
8210 bool bad_proto = FALSE;
8211 bool in_brackets = FALSE;
8212 char greedy_proto = ' ';
8213 bool proto_after_greedy_proto = FALSE;
8214 bool must_be_last = FALSE;
8215 bool underscore = FALSE;
8216 bool seen_underscore = FALSE;
8217 const bool warnillegalproto = ckWARN(WARN_ILLEGALPROTO);
8220 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
8222 Perl_croak(aTHX_ "Prototype not terminated");
8223 /* strip spaces and check for bad characters */
8224 d = SvPV(PL_lex_stuff, tmplen);
8226 for (p = d; tmplen; tmplen--, ++p) {
8230 if (warnillegalproto) {
8232 proto_after_greedy_proto = TRUE;
8233 if (!strchr("$@%*;[]&\\_+", *p) || *p == '\0') {
8238 if ( !strchr(";@%", *p) )
8245 else if ( *p == ']' ) {
8246 in_brackets = FALSE;
8248 else if ( (*p == '@' || *p == '%') &&
8249 ( tmp < 2 || d[tmp-2] != '\\' ) &&
8251 must_be_last = TRUE;
8254 else if ( *p == '_' ) {
8255 underscore = seen_underscore = TRUE;
8262 if (proto_after_greedy_proto)
8263 Perl_warner(aTHX_ packWARN(WARN_ILLEGALPROTO),
8264 "Prototype after '%c' for %"SVf" : %s",
8265 greedy_proto, SVfARG(PL_subname), d);
8267 SV *dsv = newSVpvs_flags("", SVs_TEMP);
8268 Perl_warner(aTHX_ packWARN(WARN_ILLEGALPROTO),
8269 "Illegal character %sin prototype for %"SVf" : %s",
8270 seen_underscore ? "after '_' " : "",
8272 SvUTF8(PL_lex_stuff)
8273 ? sv_uni_display(dsv,
8274 newSVpvn_flags(d, tmp, SVs_TEMP | SVf_UTF8),
8276 UNI_DISPLAY_ISPRINT)
8277 : pv_pretty(dsv, d, tmp, 60, NULL, NULL,
8278 PERL_PV_ESCAPE_NONASCII));
8280 SvCUR_set(PL_lex_stuff, tmp);
8285 CURMAD('q', PL_thisopen);
8286 CURMAD('_', tmpwhite);
8287 CURMAD('=', PL_thisstuff);
8288 CURMAD('Q', PL_thisclose);
8289 NEXTVAL_NEXTTOKE.opval =
8290 (OP*)newSVOP(OP_CONST, 0, PL_lex_stuff);
8291 PL_lex_stuff = NULL;
8294 s = SKIPSPACE2(s,tmpwhite);
8302 if (*s == ':' && s[1] != ':')
8303 PL_expect = attrful;
8304 else if (*s != '{' && key == KEY_sub) {
8306 Perl_croak(aTHX_ "Illegal declaration of anonymous subroutine");
8307 else if (*s != ';' && *s != '}')
8308 Perl_croak(aTHX_ "Illegal declaration of subroutine %"SVf, SVfARG(PL_subname));
8315 curmad('^', newSVpvs(""));
8316 CURMAD('_', tmpwhite);
8320 PL_thistoken = subtoken;
8323 NEXTVAL_NEXTTOKE.opval =
8324 (OP*)newSVOP(OP_CONST, 0, PL_lex_stuff);
8325 PL_lex_stuff = NULL;
8331 sv_setpvs(PL_subname, "__ANON__");
8333 sv_setpvs(PL_subname, "__ANON__::__ANON__");
8337 (void) force_word(PL_oldbufptr + tboffset, WORD,
8346 LOP(OP_SYSTEM,XREF);
8349 LOP(OP_SYMLINK,XTERM);
8352 LOP(OP_SYSCALL,XTERM);
8355 LOP(OP_SYSOPEN,XTERM);
8358 LOP(OP_SYSSEEK,XTERM);
8361 LOP(OP_SYSREAD,XTERM);
8364 LOP(OP_SYSWRITE,XTERM);
8368 TERM(sublex_start());
8389 LOP(OP_TRUNCATE,XTERM);
8401 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
8403 pl_yylval.ival = CopLINE(PL_curcop);
8407 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
8409 pl_yylval.ival = CopLINE(PL_curcop);
8413 LOP(OP_UNLINK,XTERM);
8419 LOP(OP_UNPACK,XTERM);
8422 LOP(OP_UTIME,XTERM);
8428 LOP(OP_UNSHIFT,XTERM);
8431 s = tokenize_use(1, s);
8441 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
8443 pl_yylval.ival = CopLINE(PL_curcop);
8447 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_NONEXPR)
8449 pl_yylval.ival = CopLINE(PL_curcop);
8453 PL_hints |= HINT_BLOCK_SCOPE;
8460 LOP(OP_WAITPID,XTERM);
8469 ctl_l[0] = toCTRL('L');
8471 gv_fetchpvn_flags(ctl_l, 1, GV_ADD|GV_NOTQUAL, SVt_PV);
8474 /* Make sure $^L is defined */
8475 gv_fetchpvs("\f", GV_ADD|GV_NOTQUAL, SVt_PV);
8480 if (PL_expect == XOPERATOR) {
8481 if (*s == '=' && !PL_lex_allbrackets &&
8482 PL_lex_fakeeof >= LEX_FAKEEOF_ASSIGN)
8490 if (!PL_lex_allbrackets && PL_lex_fakeeof >= LEX_FAKEEOF_LOWLOGIC)
8492 pl_yylval.ival = OP_XOR;
8497 TERM(sublex_start());
8502 #pragma segment Main
8506 S_pending_ident(pTHX)
8511 /* pit holds the identifier we read and pending_ident is reset */
8512 char pit = PL_pending_ident;
8513 const STRLEN tokenbuf_len = strlen(PL_tokenbuf);
8514 /* All routes through this function want to know if there is a colon. */
8515 const char *const has_colon = (const char*) memchr (PL_tokenbuf, ':', tokenbuf_len);
8516 PL_pending_ident = 0;
8518 /* PL_realtokenstart = realtokenend = PL_bufptr - SvPVX(PL_linestr); */
8519 DEBUG_T({ PerlIO_printf(Perl_debug_log,
8520 "### Pending identifier '%s'\n", PL_tokenbuf); });
8522 /* if we're in a my(), we can't allow dynamics here.
8523 $foo'bar has already been turned into $foo::bar, so
8524 just check for colons.
8526 if it's a legal name, the OP is a PADANY.
8529 if (PL_in_my == KEY_our) { /* "our" is merely analogous to "my" */
8531 yyerror_pv(Perl_form(aTHX_ "No package name allowed for "
8532 "variable %s in \"our\"",
8533 PL_tokenbuf), UTF ? SVf_UTF8 : 0);
8534 tmp = allocmy(PL_tokenbuf, tokenbuf_len, UTF ? SVf_UTF8 : 0);
8538 yyerror_pv(Perl_form(aTHX_ PL_no_myglob,
8539 PL_in_my == KEY_my ? "my" : "state", PL_tokenbuf),
8540 UTF ? SVf_UTF8 : 0);
8542 pl_yylval.opval = newOP(OP_PADANY, 0);
8543 pl_yylval.opval->op_targ = allocmy(PL_tokenbuf, tokenbuf_len,
8544 UTF ? SVf_UTF8 : 0);
8550 build the ops for accesses to a my() variable.
8552 Deny my($a) or my($b) in a sort block, *if* $a or $b is
8553 then used in a comparison. This catches most, but not
8554 all cases. For instance, it catches
8555 sort { my($a); $a <=> $b }
8557 sort { my($a); $a < $b ? -1 : $a == $b ? 0 : 1; }
8558 (although why you'd do that is anyone's guess).
8563 tmp = pad_findmy_pvn(PL_tokenbuf, tokenbuf_len,
8564 UTF ? SVf_UTF8 : 0);
8565 if (tmp != NOT_IN_PAD) {
8566 /* might be an "our" variable" */
8567 if (PAD_COMPNAME_FLAGS_isOUR(tmp)) {
8568 /* build ops for a bareword */
8569 HV * const stash = PAD_COMPNAME_OURSTASH(tmp);
8570 HEK * const stashname = HvNAME_HEK(stash);
8571 SV * const sym = newSVhek(stashname);
8572 sv_catpvs(sym, "::");
8573 sv_catpvn_flags(sym, PL_tokenbuf+1, tokenbuf_len - 1, (UTF ? SV_CATUTF8 : SV_CATBYTES ));
8574 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, sym);
8575 pl_yylval.opval->op_private = OPpCONST_ENTERED;
8578 ? (GV_ADDMULTI | GV_ADDINEVAL)
8581 ((PL_tokenbuf[0] == '$') ? SVt_PV
8582 : (PL_tokenbuf[0] == '@') ? SVt_PVAV
8587 /* if it's a sort block and they're naming $a or $b */
8588 if (PL_last_lop_op == OP_SORT &&
8589 PL_tokenbuf[0] == '$' &&
8590 (PL_tokenbuf[1] == 'a' || PL_tokenbuf[1] == 'b')
8593 for (d = PL_in_eval ? PL_oldoldbufptr : PL_linestart;
8594 d < PL_bufend && *d != '\n';
8597 if (strnEQ(d,"<=>",3) || strnEQ(d,"cmp",3)) {
8598 Perl_croak(aTHX_ "Can't use \"my %s\" in sort comparison",
8604 pl_yylval.opval = newOP(OP_PADANY, 0);
8605 pl_yylval.opval->op_targ = tmp;
8611 Whine if they've said @foo in a doublequoted string,
8612 and @foo isn't a variable we can find in the symbol
8615 if (ckWARN(WARN_AMBIGUOUS) &&
8616 pit == '@' && PL_lex_state != LEX_NORMAL && !PL_lex_brackets) {
8617 GV *const gv = gv_fetchpvn_flags(PL_tokenbuf + 1, tokenbuf_len - 1,
8618 ( UTF ? SVf_UTF8 : 0 ), SVt_PVAV);
8619 if ((!gv || ((PL_tokenbuf[0] == '@') ? !GvAV(gv) : !GvHV(gv)))
8620 /* DO NOT warn for @- and @+ */
8621 && !( PL_tokenbuf[2] == '\0' &&
8622 ( PL_tokenbuf[1] == '-' || PL_tokenbuf[1] == '+' ))
8625 /* Downgraded from fatal to warning 20000522 mjd */
8626 Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
8627 "Possible unintended interpolation of %"SVf" in string",
8628 SVfARG(newSVpvn_flags(PL_tokenbuf, tokenbuf_len,
8629 SVs_TEMP | ( UTF ? SVf_UTF8 : 0 ))));
8633 /* build ops for a bareword */
8634 pl_yylval.opval = (OP*)newSVOP(OP_CONST, 0, newSVpvn_flags(PL_tokenbuf + 1,
8636 UTF ? SVf_UTF8 : 0 ));
8637 pl_yylval.opval->op_private = OPpCONST_ENTERED;
8638 gv_fetchpvn_flags(PL_tokenbuf+1, tokenbuf_len - 1,
8639 (PL_in_eval ? (GV_ADDMULTI | GV_ADDINEVAL) : GV_ADD)
8640 | ( UTF ? SVf_UTF8 : 0 ),
8641 ((PL_tokenbuf[0] == '$') ? SVt_PV
8642 : (PL_tokenbuf[0] == '@') ? SVt_PVAV
8648 S_checkcomma(pTHX_ const char *s, const char *name, const char *what)
8652 PERL_ARGS_ASSERT_CHECKCOMMA;
8654 if (*s == ' ' && s[1] == '(') { /* XXX gotta be a better way */
8655 if (ckWARN(WARN_SYNTAX)) {
8658 for (w = s+2; *w && level; w++) {
8666 /* the list of chars below is for end of statements or
8667 * block / parens, boolean operators (&&, ||, //) and branch
8668 * constructs (or, and, if, until, unless, while, err, for).
8669 * Not a very solid hack... */
8670 if (!*w || !strchr(";&/|})]oaiuwef!=", *w))
8671 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
8672 "%s (...) interpreted as function",name);
8675 while (s < PL_bufend && isSPACE(*s))
8679 while (s < PL_bufend && isSPACE(*s))
8681 if (isIDFIRST_lazy_if(s,UTF)) {
8682 const char * const w = s;
8683 s += UTF ? UTF8SKIP(s) : 1;
8684 while (isALNUM_lazy_if(s,UTF))
8685 s += UTF ? UTF8SKIP(s) : 1;
8686 while (s < PL_bufend && isSPACE(*s))
8690 if (keyword(w, s - w, 0))
8693 gv = gv_fetchpvn_flags(w, s - w, ( UTF ? SVf_UTF8 : 0 ), SVt_PVCV);
8694 if (gv && GvCVu(gv))
8696 Perl_croak(aTHX_ "No comma allowed after %s", what);
8701 /* Either returns sv, or mortalizes sv and returns a new SV*.
8702 Best used as sv=new_constant(..., sv, ...).
8703 If s, pv are NULL, calls subroutine with one argument,
8704 and type is used with error messages only. */
8707 S_new_constant(pTHX_ const char *s, STRLEN len, const char *key, STRLEN keylen,
8708 SV *sv, SV *pv, const char *type, STRLEN typelen)
8711 HV * table = GvHV(PL_hintgv); /* ^H */
8715 const char *why1 = "", *why2 = "", *why3 = "";
8717 PERL_ARGS_ASSERT_NEW_CONSTANT;
8719 /* charnames doesn't work well if there have been errors found */
8720 if (PL_error_count > 0 && strEQ(key,"charnames"))
8721 return &PL_sv_undef;
8724 || ! (PL_hints & HINT_LOCALIZE_HH)
8725 || ! (cvp = hv_fetch(table, key, keylen, FALSE))
8730 /* Here haven't found what we're looking for. If it is charnames,
8731 * perhaps it needs to be loaded. Try doing that before giving up */
8732 if (strEQ(key,"charnames")) {
8733 Perl_load_module(aTHX_
8735 newSVpvs("_charnames"),
8736 /* version parameter; no need to specify it, as if
8737 * we get too early a version, will fail anyway,
8738 * not being able to find '_charnames' */
8744 table = GvHV(PL_hintgv);
8746 && (PL_hints & HINT_LOCALIZE_HH)
8747 && (cvp = hv_fetch(table, key, keylen, FALSE))
8753 if (!table || !(PL_hints & HINT_LOCALIZE_HH)) {
8754 msg = Perl_newSVpvf(aTHX_
8755 "Constant(%s) unknown", (type ? type: "undef"));
8760 why3 = "} is not defined";
8762 msg = Perl_newSVpvf(aTHX_ "Constant(%s): %s%s%s",
8763 (type ? type: "undef"), why1, why2, why3);
8765 yyerror(SvPVX_const(msg));
8770 sv_2mortal(sv); /* Parent created it permanently */
8773 pv = newSVpvn_flags(s, len, SVs_TEMP);
8775 typesv = newSVpvn_flags(type, typelen, SVs_TEMP);
8777 typesv = &PL_sv_undef;
8779 PUSHSTACKi(PERLSI_OVERLOAD);
8791 call_sv(cv, G_SCALAR | ( PL_in_eval ? 0 : G_EVAL));
8795 /* Check the eval first */
8796 if (!PL_in_eval && SvTRUE(ERRSV)) {
8797 sv_catpvs(ERRSV, "Propagated");
8798 yyerror(SvPV_nolen_const(ERRSV)); /* Duplicates the message inside eval */
8800 res = SvREFCNT_inc_simple(sv);
8804 SvREFCNT_inc_simple_void(res);
8813 why1 = "Call to &{$^H{";
8815 why3 = "}} did not return a defined value";
8823 /* Returns a NUL terminated string, with the length of the string written to
8827 S_scan_word(pTHX_ register char *s, char *dest, STRLEN destlen, int allow_package, STRLEN *slp)
8830 register char *d = dest;
8831 register char * const e = d + destlen - 3; /* two-character token, ending NUL */
8833 PERL_ARGS_ASSERT_SCAN_WORD;
8837 Perl_croak(aTHX_ ident_too_long);
8838 if (isALNUM(*s) || (!UTF && isALNUMC_L1(*s))) /* UTF handled below */
8840 else if (allow_package && (*s == '\'') && isIDFIRST_lazy_if(s+1,UTF)) {
8845 else if (allow_package && (s[0] == ':') && (s[1] == ':') && (s[2] != '$')) {
8849 else if (UTF && UTF8_IS_START(*s) && isALNUM_utf8((U8*)s)) {
8850 char *t = s + UTF8SKIP(s);
8852 while (UTF8_IS_CONTINUED(*t) && is_utf8_mark((U8*)t))
8856 Perl_croak(aTHX_ ident_too_long);
8857 Copy(s, d, len, char);
8870 S_scan_ident(pTHX_ register char *s, register const char *send, char *dest, STRLEN destlen, I32 ck_uni)
8873 char *bracket = NULL;
8875 register char *d = dest;
8876 register char * const e = d + destlen - 3; /* two-character token, ending NUL */
8878 PERL_ARGS_ASSERT_SCAN_IDENT;
8883 while (isDIGIT(*s)) {
8885 Perl_croak(aTHX_ ident_too_long);
8892 Perl_croak(aTHX_ ident_too_long);
8893 if (isALNUM(*s)) /* UTF handled below */
8895 else if (*s == '\'' && isIDFIRST_lazy_if(s+1,UTF)) {
8900 else if (*s == ':' && s[1] == ':') {
8904 else if (UTF && UTF8_IS_START(*s) && isALNUM_utf8((U8*)s)) {
8905 char *t = s + UTF8SKIP(s);
8906 while (UTF8_IS_CONTINUED(*t) && is_utf8_mark((U8*)t))
8908 if (d + (t - s) > e)
8909 Perl_croak(aTHX_ ident_too_long);
8910 Copy(s, d, t - s, char);
8921 if (PL_lex_state != LEX_NORMAL)
8922 PL_lex_state = LEX_INTERPENDMAYBE;
8925 if (*s == '$' && s[1] &&
8926 (isALNUM_lazy_if(s+1,UTF) || s[1] == '$' || s[1] == '{' || strnEQ(s+1,"::",2)) )
8936 const STRLEN skip = UTF8SKIP(s);
8939 for ( i = 0; i < skip; i++ )
8947 if (*d == '^' && *s && isCONTROLVAR(*s)) {
8951 else if (ck_uni && !bracket)
8954 if (isSPACE(s[-1])) {
8956 const char ch = *s++;
8957 if (!SPACE_OR_TAB(ch)) {
8963 if (isIDFIRST_lazy_if(d,UTF)) {
8967 while ((end < send && isALNUM_lazy_if(end,UTF)) || *end == ':') {
8968 end += UTF8SKIP(end);
8969 while (end < send && UTF8_IS_CONTINUED(*end) && is_utf8_mark((U8*)end))
8970 end += UTF8SKIP(end);
8972 Copy(s, d, end - s, char);
8977 while ((isALNUM(*s) || *s == ':') && d < e)
8980 Perl_croak(aTHX_ ident_too_long);
8983 while (s < send && SPACE_OR_TAB(*s))
8985 if ((*s == '[' || (*s == '{' && strNE(dest, "sub")))) {
8986 if (ckWARN(WARN_AMBIGUOUS) && keyword(dest, d - dest, 0)) {
8987 const char * const brack =
8989 ((*s == '[') ? "[...]" : "{...}");
8990 /* diag_listed_as: Ambiguous use of %c{%s[...]} resolved to %c%s[...] */
8991 Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
8992 "Ambiguous use of %c{%s%s} resolved to %c%s%s",
8993 funny, dest, brack, funny, dest, brack);
8996 PL_lex_brackstack[PL_lex_brackets++] = (char)(XOPERATOR | XFAKEBRACK);
8997 PL_lex_allbrackets++;
9001 /* Handle extended ${^Foo} variables
9002 * 1999-02-27 mjd-perl-patch@plover.com */
9003 else if (!isALNUM(*d) && !isPRINT(*d) /* isCTRL(d) */
9007 while (isALNUM(*s) && d < e) {
9011 Perl_croak(aTHX_ ident_too_long);
9016 if (PL_lex_state == LEX_INTERPNORMAL && !PL_lex_brackets) {
9017 PL_lex_state = LEX_INTERPEND;
9020 if (PL_lex_state == LEX_NORMAL) {
9021 if (ckWARN(WARN_AMBIGUOUS) &&
9022 (keyword(dest, d - dest, 0)
9023 || get_cvn_flags(dest, d - dest, UTF ? SVf_UTF8 : 0)))
9025 SV *tmp = newSVpvn_flags( dest, d - dest,
9026 SVs_TEMP | (UTF ? SVf_UTF8 : 0) );
9029 Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
9030 "Ambiguous use of %c{%"SVf"} resolved to %c%"SVf,
9031 funny, tmp, funny, tmp);
9036 s = bracket; /* let the parser handle it */
9040 else if (PL_lex_state == LEX_INTERPNORMAL && !PL_lex_brackets && !intuit_more(s))
9041 PL_lex_state = LEX_INTERPEND;
9046 S_pmflag(pTHX_ const char* const valid_flags, U32 * pmfl, char** s, char* charset) {
9048 /* Adds, subtracts to/from 'pmfl' based on regex modifier flags found in
9049 * the parse starting at 's', based on the subset that are valid in this
9050 * context input to this routine in 'valid_flags'. Advances s. Returns
9051 * TRUE if the input should be treated as a valid flag, so the next char
9052 * may be as well; otherwise FALSE. 'charset' should point to a NUL upon
9053 * first call on the current regex. This routine will set it to any
9054 * charset modifier found. The caller shouldn't change it. This way,
9055 * another charset modifier encountered in the parse can be detected as an
9056 * error, as we have decided to allow only one */
9059 STRLEN charlen = UTF ? UTF8SKIP(*s) : 1;
9061 if ( charlen != 1 || ! strchr(valid_flags, c) ) {
9062 if (isALNUM_lazy_if(*s, UTF)) {
9063 yyerror_pv(Perl_form(aTHX_ "Unknown regexp modifier \"/%.*s\"", charlen, *s),
9064 UTF ? SVf_UTF8 : 0);
9066 /* Pretend that it worked, so will continue processing before
9075 CASE_STD_PMMOD_FLAGS_PARSE_SET(pmfl);
9076 case GLOBAL_PAT_MOD: *pmfl |= PMf_GLOBAL; break;
9077 case CONTINUE_PAT_MOD: *pmfl |= PMf_CONTINUE; break;
9078 case ONCE_PAT_MOD: *pmfl |= PMf_KEEP; break;
9079 case KEEPCOPY_PAT_MOD: *pmfl |= RXf_PMf_KEEPCOPY; break;
9080 case NONDESTRUCT_PAT_MOD: *pmfl |= PMf_NONDESTRUCT; break;
9081 case LOCALE_PAT_MOD:
9083 goto multiple_charsets;
9085 set_regex_charset(pmfl, REGEX_LOCALE_CHARSET);
9088 case UNICODE_PAT_MOD:
9090 goto multiple_charsets;
9092 set_regex_charset(pmfl, REGEX_UNICODE_CHARSET);
9095 case ASCII_RESTRICT_PAT_MOD:
9097 set_regex_charset(pmfl, REGEX_ASCII_RESTRICTED_CHARSET);
9101 /* Error if previous modifier wasn't an 'a', but if it was, see
9102 * if, and accept, a second occurrence (only) */
9104 || get_regex_charset(*pmfl)
9105 != REGEX_ASCII_RESTRICTED_CHARSET)
9107 goto multiple_charsets;
9109 set_regex_charset(pmfl, REGEX_ASCII_MORE_RESTRICTED_CHARSET);
9113 case DEPENDS_PAT_MOD:
9115 goto multiple_charsets;
9117 set_regex_charset(pmfl, REGEX_DEPENDS_CHARSET);
9126 if (*charset != c) {
9127 yyerror(Perl_form(aTHX_ "Regexp modifiers \"/%c\" and \"/%c\" are mutually exclusive", *charset, c));
9129 else if (c == 'a') {
9130 yyerror("Regexp modifier \"/a\" may appear a maximum of twice");
9133 yyerror(Perl_form(aTHX_ "Regexp modifier \"/%c\" may not appear twice", c));
9136 /* Pretend that it worked, so will continue processing before dieing */
9142 S_scan_pat(pTHX_ char *start, I32 type)
9146 char *s = scan_str(start,!!PL_madskills,FALSE, PL_reg_state.re_reparsing);
9147 const char * const valid_flags =
9148 (const char *)((type == OP_QR) ? QR_PAT_MODS : M_PAT_MODS);
9149 char charset = '\0'; /* character set modifier */
9154 PERL_ARGS_ASSERT_SCAN_PAT;
9156 /* this was only needed for the initial scan_str; set it to false
9157 * so that any (?{}) code blocks etc are parsed normally */
9158 PL_reg_state.re_reparsing = FALSE;
9160 const char * const delimiter = skipspace(start);
9164 ? "Search pattern not terminated or ternary operator parsed as search pattern"
9165 : "Search pattern not terminated" ));
9168 pm = (PMOP*)newPMOP(type, 0);
9169 if (PL_multi_open == '?') {
9170 /* This is the only point in the code that sets PMf_ONCE: */
9171 pm->op_pmflags |= PMf_ONCE;
9173 /* Hence it's safe to do this bit of PMOP book-keeping here, which
9174 allows us to restrict the list needed by reset to just the ??
9176 assert(type != OP_TRANS);
9178 MAGIC *mg = mg_find((const SV *)PL_curstash, PERL_MAGIC_symtab);
9181 mg = sv_magicext(MUTABLE_SV(PL_curstash), 0, PERL_MAGIC_symtab, 0, 0,
9184 elements = mg->mg_len / sizeof(PMOP**);
9185 Renewc(mg->mg_ptr, elements + 1, PMOP*, char);
9186 ((PMOP**)mg->mg_ptr) [elements++] = pm;
9187 mg->mg_len = elements * sizeof(PMOP**);
9188 PmopSTASH_set(pm,PL_curstash);
9195 /* if qr/...(?{..}).../, then need to parse the pattern within a new
9196 * anon CV. False positives like qr/[(?{]/ are harmless */
9198 if (type == OP_QR) {
9200 char *e, *p = SvPV(PL_lex_stuff, len);
9202 for (; p < e; p++) {
9203 if (p[0] == '(' && p[1] == '?'
9204 && (p[2] == '{' || (p[2] == '?' && p[3] == '{')))
9206 pm->op_pmflags |= PMf_HAS_CV;
9210 pm->op_pmflags |= PMf_IS_QR;
9213 while (*s && S_pmflag(aTHX_ valid_flags, &(pm->op_pmflags), &s, &charset)) {};
9215 if (PL_madskills && modstart != s) {
9216 SV* tmptoken = newSVpvn(modstart, s - modstart);
9217 append_madprops(newMADPROP('m', MAD_SV, tmptoken, 0), (OP*)pm, 0);
9220 /* issue a warning if /c is specified,but /g is not */
9221 if ((pm->op_pmflags & PMf_CONTINUE) && !(pm->op_pmflags & PMf_GLOBAL))
9223 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP),
9224 "Use of /c modifier is meaningless without /g" );
9227 PL_lex_op = (OP*)pm;
9228 pl_yylval.ival = OP_MATCH;
9233 S_scan_subst(pTHX_ char *start)
9240 char charset = '\0'; /* character set modifier */
9245 PERL_ARGS_ASSERT_SCAN_SUBST;
9247 pl_yylval.ival = OP_NULL;
9249 s = scan_str(start,!!PL_madskills,FALSE,FALSE);
9252 Perl_croak(aTHX_ "Substitution pattern not terminated");
9254 if (s[-1] == PL_multi_open)
9258 CURMAD('q', PL_thisopen);
9259 CURMAD('_', PL_thiswhite);
9260 CURMAD('E', PL_thisstuff);
9261 CURMAD('Q', PL_thisclose);
9262 PL_realtokenstart = s - SvPVX(PL_linestr);
9266 first_start = PL_multi_start;
9267 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
9270 SvREFCNT_dec(PL_lex_stuff);
9271 PL_lex_stuff = NULL;
9273 Perl_croak(aTHX_ "Substitution replacement not terminated");
9275 PL_multi_start = first_start; /* so whole substitution is taken together */
9277 pm = (PMOP*)newPMOP(OP_SUBST, 0);
9281 CURMAD('z', PL_thisopen);
9282 CURMAD('R', PL_thisstuff);
9283 CURMAD('Z', PL_thisclose);
9289 if (*s == EXEC_PAT_MOD) {
9293 else if (! S_pmflag(aTHX_ S_PAT_MODS, &(pm->op_pmflags), &s, &charset))
9302 curmad('m', newSVpvn(modstart, s - modstart));
9303 append_madprops(PL_thismad, (OP*)pm, 0);
9307 if ((pm->op_pmflags & PMf_CONTINUE)) {
9308 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), "Use of /c modifier is meaningless in s///" );
9312 SV * const repl = newSVpvs("");
9314 PL_sublex_info.super_bufptr = s;
9315 PL_sublex_info.super_bufend = PL_bufend;
9317 pm->op_pmflags |= PMf_EVAL;
9320 sv_catpvs(repl, "eval ");
9322 sv_catpvs(repl, "do ");
9324 sv_catpvs(repl, "{");
9325 sv_catsv(repl, PL_lex_repl);
9326 if (strchr(SvPVX(PL_lex_repl), '#'))
9327 sv_catpvs(repl, "\n");
9328 sv_catpvs(repl, "}");
9330 SvREFCNT_dec(PL_lex_repl);
9334 PL_lex_op = (OP*)pm;
9335 pl_yylval.ival = OP_SUBST;
9340 S_scan_trans(pTHX_ char *start)
9348 bool nondestruct = 0;
9353 PERL_ARGS_ASSERT_SCAN_TRANS;
9355 pl_yylval.ival = OP_NULL;
9357 s = scan_str(start,!!PL_madskills,FALSE,FALSE);
9359 Perl_croak(aTHX_ "Transliteration pattern not terminated");
9361 if (s[-1] == PL_multi_open)
9365 CURMAD('q', PL_thisopen);
9366 CURMAD('_', PL_thiswhite);
9367 CURMAD('E', PL_thisstuff);
9368 CURMAD('Q', PL_thisclose);
9369 PL_realtokenstart = s - SvPVX(PL_linestr);
9373 s = scan_str(s,!!PL_madskills,FALSE,FALSE);
9376 SvREFCNT_dec(PL_lex_stuff);
9377 PL_lex_stuff = NULL;
9379 Perl_croak(aTHX_ "Transliteration replacement not terminated");
9382 CURMAD('z', PL_thisopen);
9383 CURMAD('R', PL_thisstuff);
9384 CURMAD('Z', PL_thisclose);
9387 complement = del = squash = 0;
9394 complement = OPpTRANS_COMPLEMENT;
9397 del = OPpTRANS_DELETE;
9400 squash = OPpTRANS_SQUASH;
9412 o = newPVOP(nondestruct ? OP_TRANSR : OP_TRANS, 0, (char*)NULL);
9413 o->op_private &= ~OPpTRANS_ALL;
9414 o->op_private |= del|squash|complement|
9415 (DO_UTF8(PL_lex_stuff)? OPpTRANS_FROM_UTF : 0)|
9416 (DO_UTF8(PL_lex_repl) ? OPpTRANS_TO_UTF : 0);
9419 pl_yylval.ival = nondestruct ? OP_TRANSR : OP_TRANS;
9424 curmad('m', newSVpvn(modstart, s - modstart));
9425 append_madprops(PL_thismad, o, 0);
9434 S_scan_heredoc(pTHX_ register char *s)
9438 I32 op_type = OP_SCALAR;
9442 const char *found_newline;
9446 const int outer = (PL_rsfp || PL_parser->filtered)
9447 && !(PL_lex_inwhat == OP_SCALAR);
9449 I32 stuffstart = s - SvPVX(PL_linestr);
9452 PL_realtokenstart = -1;
9455 PERL_ARGS_ASSERT_SCAN_HEREDOC;
9459 e = PL_tokenbuf + sizeof PL_tokenbuf - 1;
9463 while (SPACE_OR_TAB(*peek))
9465 if (*peek == '`' || *peek == '\'' || *peek =='"') {
9468 s = delimcpy(d, e, s, PL_bufend, term, &len);
9478 if (!isALNUM_lazy_if(s,UTF))
9479 deprecate("bare << to mean <<\"\"");
9480 for (; isALNUM_lazy_if(s,UTF); s++) {
9485 if (d >= PL_tokenbuf + sizeof PL_tokenbuf - 1)
9486 Perl_croak(aTHX_ "Delimiter for here document is too long");
9489 len = d - PL_tokenbuf;
9493 tstart = PL_tokenbuf + !outer;
9494 PL_thisclose = newSVpvn(tstart, len - !outer);
9495 tstart = SvPVX(PL_linestr) + stuffstart;
9496 PL_thisopen = newSVpvn(tstart, s - tstart);
9497 stuffstart = s - SvPVX(PL_linestr);
9500 #ifndef PERL_STRICT_CR
9501 d = strchr(s, '\r');
9503 char * const olds = s;
9505 while (s < PL_bufend) {
9511 else if (*s == '\n' && s[1] == '\r') { /* \015\013 on a mac? */
9520 SvCUR_set(PL_linestr, PL_bufend - SvPVX_const(PL_linestr));
9527 if ( outer || !(found_newline = (char*)memchr((void*)s, '\n', PL_bufend - s)) ) {
9528 herewas = newSVpvn(s,PL_bufend-s);
9532 herewas = newSVpvn(s-1,found_newline-s+1);
9535 herewas = newSVpvn(s,found_newline-s);
9540 tstart = SvPVX(PL_linestr) + stuffstart;
9542 sv_catpvn(PL_thisstuff, tstart, s - tstart);
9544 PL_thisstuff = newSVpvn(tstart, s - tstart);
9547 s += SvCUR(herewas);
9550 stuffstart = s - SvPVX(PL_linestr);
9556 tmpstr = newSV_type(SVt_PVIV);
9560 SvIV_set(tmpstr, -1);
9562 else if (term == '`') {
9563 op_type = OP_BACKTICK;
9564 SvIV_set(tmpstr, '\\');
9568 PL_multi_start = CopLINE(PL_curcop);
9569 PL_multi_open = PL_multi_close = '<';
9570 term = *PL_tokenbuf;
9571 if (PL_lex_inwhat == OP_SUBST && PL_in_eval && !PL_rsfp
9572 && !PL_parser->filtered) {
9573 char * const bufptr = PL_sublex_info.super_bufptr;
9574 char * const bufend = PL_sublex_info.super_bufend;
9575 char * const olds = s - SvCUR(herewas);
9576 s = strchr(bufptr, '\n');
9580 while (s < bufend &&
9581 (*s != term || memNE(s,PL_tokenbuf,len)) ) {
9583 CopLINE_inc(PL_curcop);
9586 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
9587 missingterm(PL_tokenbuf);
9589 sv_setpvn(herewas,bufptr,d-bufptr+1);
9590 sv_setpvn(tmpstr,d+1,s-d);
9592 sv_catpvn(herewas,s,bufend-s);
9593 Copy(SvPVX_const(herewas),bufptr,SvCUR(herewas) + 1,char);
9600 while (s < PL_bufend &&
9601 (*s != term || memNE(s,PL_tokenbuf,len)) ) {
9603 CopLINE_inc(PL_curcop);
9605 if (s >= PL_bufend) {
9606 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
9607 missingterm(PL_tokenbuf);
9609 sv_setpvn(tmpstr,d+1,s-d);
9613 sv_catpvn(PL_thisstuff, d + 1, s - d);
9615 PL_thisstuff = newSVpvn(d + 1, s - d);
9616 stuffstart = s - SvPVX(PL_linestr);
9620 CopLINE_inc(PL_curcop); /* the preceding stmt passes a newline */
9622 sv_catpvn(herewas,s,PL_bufend-s);
9623 sv_setsv(PL_linestr,herewas);
9624 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = s = PL_linestart = SvPVX(PL_linestr);
9625 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
9626 PL_last_lop = PL_last_uni = NULL;
9629 sv_setpvs(tmpstr,""); /* avoid "uninitialized" warning */
9630 while (s >= PL_bufend) { /* multiple line string? */
9633 tstart = SvPVX(PL_linestr) + stuffstart;
9635 sv_catpvn(PL_thisstuff, tstart, PL_bufend - tstart);
9637 PL_thisstuff = newSVpvn(tstart, PL_bufend - tstart);
9641 CopLINE_inc(PL_curcop);
9642 if (!outer || !lex_next_chunk(0)) {
9643 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
9644 missingterm(PL_tokenbuf);
9646 CopLINE_dec(PL_curcop);
9649 stuffstart = s - SvPVX(PL_linestr);
9651 CopLINE_inc(PL_curcop);
9652 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
9653 PL_last_lop = PL_last_uni = NULL;
9654 #ifndef PERL_STRICT_CR
9655 if (PL_bufend - PL_linestart >= 2) {
9656 if ((PL_bufend[-2] == '\r' && PL_bufend[-1] == '\n') ||
9657 (PL_bufend[-2] == '\n' && PL_bufend[-1] == '\r'))
9659 PL_bufend[-2] = '\n';
9661 SvCUR_set(PL_linestr, PL_bufend - SvPVX_const(PL_linestr));
9663 else if (PL_bufend[-1] == '\r')
9664 PL_bufend[-1] = '\n';
9666 else if (PL_bufend - PL_linestart == 1 && PL_bufend[-1] == '\r')
9667 PL_bufend[-1] = '\n';
9669 if (*s == term && memEQ(s,PL_tokenbuf,len)) {
9670 STRLEN off = PL_bufend - 1 - SvPVX_const(PL_linestr);
9671 *(SvPVX(PL_linestr) + off ) = ' ';
9672 lex_grow_linestr(SvCUR(PL_linestr) + SvCUR(herewas) + 1);
9673 sv_catsv(PL_linestr,herewas);
9674 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
9675 s = SvPVX(PL_linestr) + off; /* In case PV of PL_linestr moved. */
9679 sv_catsv(tmpstr,PL_linestr);
9684 PL_multi_end = CopLINE(PL_curcop);
9685 if (SvCUR(tmpstr) + 5 < SvLEN(tmpstr)) {
9686 SvPV_shrink_to_cur(tmpstr);
9688 SvREFCNT_dec(herewas);
9690 if (UTF && is_utf8_string((U8*)SvPVX_const(tmpstr), SvCUR(tmpstr)))
9692 else if (PL_encoding)
9693 sv_recode_to_utf8(tmpstr, PL_encoding);
9695 PL_lex_stuff = tmpstr;
9696 pl_yylval.ival = op_type;
9701 takes: current position in input buffer
9702 returns: new position in input buffer
9703 side-effects: pl_yylval and lex_op are set.
9708 <FH> read from filehandle
9709 <pkg::FH> read from package qualified filehandle
9710 <pkg'FH> read from package qualified filehandle
9711 <$fh> read from filehandle in $fh
9717 S_scan_inputsymbol(pTHX_ char *start)
9720 register char *s = start; /* current position in buffer */
9723 char *d = PL_tokenbuf; /* start of temp holding space */
9724 const char * const e = PL_tokenbuf + sizeof PL_tokenbuf; /* end of temp holding space */
9726 PERL_ARGS_ASSERT_SCAN_INPUTSYMBOL;
9728 end = strchr(s, '\n');
9731 s = delimcpy(d, e, s + 1, end, '>', &len); /* extract until > */
9733 /* die if we didn't have space for the contents of the <>,
9734 or if it didn't end, or if we see a newline
9737 if (len >= (I32)sizeof PL_tokenbuf)
9738 Perl_croak(aTHX_ "Excessively long <> operator");
9740 Perl_croak(aTHX_ "Unterminated <> operator");
9745 Remember, only scalar variables are interpreted as filehandles by
9746 this code. Anything more complex (e.g., <$fh{$num}>) will be
9747 treated as a glob() call.
9748 This code makes use of the fact that except for the $ at the front,
9749 a scalar variable and a filehandle look the same.
9751 if (*d == '$' && d[1]) d++;
9753 /* allow <Pkg'VALUE> or <Pkg::VALUE> */
9754 while (*d && (isALNUM_lazy_if(d,UTF) || *d == '\'' || *d == ':'))
9755 d += UTF ? UTF8SKIP(d) : 1;
9757 /* If we've tried to read what we allow filehandles to look like, and
9758 there's still text left, then it must be a glob() and not a getline.
9759 Use scan_str to pull out the stuff between the <> and treat it
9760 as nothing more than a string.
9763 if (d - PL_tokenbuf != len) {
9764 pl_yylval.ival = OP_GLOB;
9765 s = scan_str(start,!!PL_madskills,FALSE,FALSE);
9767 Perl_croak(aTHX_ "Glob not terminated");
9771 bool readline_overriden = FALSE;
9774 /* we're in a filehandle read situation */
9777 /* turn <> into <ARGV> */
9779 Copy("ARGV",d,5,char);
9781 /* Check whether readline() is overriden */
9782 gv_readline = gv_fetchpvs("readline", GV_NOTQUAL, SVt_PVCV);
9784 && GvCVu(gv_readline) && GvIMPORTED_CV(gv_readline))
9786 ((gvp = (GV**)hv_fetchs(PL_globalstash, "readline", FALSE))
9787 && (gv_readline = *gvp) && isGV_with_GP(gv_readline)
9788 && GvCVu(gv_readline) && GvIMPORTED_CV(gv_readline)))
9789 readline_overriden = TRUE;
9791 /* if <$fh>, create the ops to turn the variable into a
9795 /* try to find it in the pad for this block, otherwise find
9796 add symbol table ops
9798 const PADOFFSET tmp = pad_findmy_pvn(d, len, UTF ? SVf_UTF8 : 0);
9799 if (tmp != NOT_IN_PAD) {
9800 if (PAD_COMPNAME_FLAGS_isOUR(tmp)) {
9801 HV * const stash = PAD_COMPNAME_OURSTASH(tmp);
9802 HEK * const stashname = HvNAME_HEK(stash);
9803 SV * const sym = sv_2mortal(newSVhek(stashname));
9804 sv_catpvs(sym, "::");
9810 OP * const o = newOP(OP_PADSV, 0);
9812 PL_lex_op = readline_overriden
9813 ? (OP*)newUNOP(OP_ENTERSUB, OPf_STACKED,
9814 op_append_elem(OP_LIST, o,
9815 newCVREF(0, newGVOP(OP_GV,0,gv_readline))))
9816 : (OP*)newUNOP(OP_READLINE, 0, o);
9825 ? (GV_ADDMULTI | GV_ADDINEVAL)
9826 : GV_ADDMULTI) | ( UTF ? SVf_UTF8 : 0 ),
9828 PL_lex_op = readline_overriden
9829 ? (OP*)newUNOP(OP_ENTERSUB, OPf_STACKED,
9830 op_append_elem(OP_LIST,
9831 newUNOP(OP_RV2SV, 0, newGVOP(OP_GV, 0, gv)),
9832 newCVREF(0, newGVOP(OP_GV, 0, gv_readline))))
9833 : (OP*)newUNOP(OP_READLINE, 0,
9834 newUNOP(OP_RV2SV, 0,
9835 newGVOP(OP_GV, 0, gv)));
9837 if (!readline_overriden)
9838 PL_lex_op->op_flags |= OPf_SPECIAL;
9839 /* we created the ops in PL_lex_op, so make pl_yylval.ival a null op */
9840 pl_yylval.ival = OP_NULL;
9843 /* If it's none of the above, it must be a literal filehandle
9844 (<Foo::BAR> or <FOO>) so build a simple readline OP */
9846 GV * const gv = gv_fetchpv(d, GV_ADD | ( UTF ? SVf_UTF8 : 0 ), SVt_PVIO);
9847 PL_lex_op = readline_overriden
9848 ? (OP*)newUNOP(OP_ENTERSUB, OPf_STACKED,
9849 op_append_elem(OP_LIST,
9850 newGVOP(OP_GV, 0, gv),
9851 newCVREF(0, newGVOP(OP_GV, 0, gv_readline))))
9852 : (OP*)newUNOP(OP_READLINE, 0, newGVOP(OP_GV, 0, gv));
9853 pl_yylval.ival = OP_NULL;
9862 takes: start position in buffer
9863 keep_quoted preserve \ on the embedded delimiter(s)
9864 keep_delims preserve the delimiters around the string
9865 re_reparse compiling a run-time /(?{})/:
9866 collapse // to /, and skip encoding src
9867 returns: position to continue reading from buffer
9868 side-effects: multi_start, multi_close, lex_repl or lex_stuff, and
9869 updates the read buffer.
9871 This subroutine pulls a string out of the input. It is called for:
9872 q single quotes q(literal text)
9873 ' single quotes 'literal text'
9874 qq double quotes qq(interpolate $here please)
9875 " double quotes "interpolate $here please"
9876 qx backticks qx(/bin/ls -l)
9877 ` backticks `/bin/ls -l`
9878 qw quote words @EXPORT_OK = qw( func() $spam )
9879 m// regexp match m/this/
9880 s/// regexp substitute s/this/that/
9881 tr/// string transliterate tr/this/that/
9882 y/// string transliterate y/this/that/
9883 ($*@) sub prototypes sub foo ($)
9884 (stuff) sub attr parameters sub foo : attr(stuff)
9885 <> readline or globs <FOO>, <>, <$fh>, or <*.c>
9887 In most of these cases (all but <>, patterns and transliterate)
9888 yylex() calls scan_str(). m// makes yylex() call scan_pat() which
9889 calls scan_str(). s/// makes yylex() call scan_subst() which calls
9890 scan_str(). tr/// and y/// make yylex() call scan_trans() which
9893 It skips whitespace before the string starts, and treats the first
9894 character as the delimiter. If the delimiter is one of ([{< then
9895 the corresponding "close" character )]}> is used as the closing
9896 delimiter. It allows quoting of delimiters, and if the string has
9897 balanced delimiters ([{<>}]) it allows nesting.
9899 On success, the SV with the resulting string is put into lex_stuff or,
9900 if that is already non-NULL, into lex_repl. The second case occurs only
9901 when parsing the RHS of the special constructs s/// and tr/// (y///).
9902 For convenience, the terminating delimiter character is stuffed into
9907 S_scan_str(pTHX_ char *start, int keep_quoted, int keep_delims, int re_reparse)
9910 SV *sv; /* scalar value: string */
9911 const char *tmps; /* temp string, used for delimiter matching */
9912 register char *s = start; /* current position in the buffer */
9913 register char term; /* terminating character */
9914 register char *to; /* current position in the sv's data */
9915 I32 brackets = 1; /* bracket nesting level */
9916 bool has_utf8 = FALSE; /* is there any utf8 content? */
9917 I32 termcode; /* terminating char. code */
9918 U8 termstr[UTF8_MAXBYTES]; /* terminating string */
9919 STRLEN termlen; /* length of terminating string */
9920 int last_off = 0; /* last position for nesting bracket */
9926 PERL_ARGS_ASSERT_SCAN_STR;
9928 /* skip space before the delimiter */
9934 if (PL_realtokenstart >= 0) {
9935 stuffstart = PL_realtokenstart;
9936 PL_realtokenstart = -1;
9939 stuffstart = start - SvPVX(PL_linestr);
9941 /* mark where we are, in case we need to report errors */
9944 /* after skipping whitespace, the next character is the terminator */
9947 termcode = termstr[0] = term;
9951 termcode = utf8_to_uvchr_buf((U8*)s, (U8*)PL_bufend, &termlen);
9952 Copy(s, termstr, termlen, U8);
9953 if (!UTF8_IS_INVARIANT(term))
9957 /* mark where we are */
9958 PL_multi_start = CopLINE(PL_curcop);
9959 PL_multi_open = term;
9961 /* find corresponding closing delimiter */
9962 if (term && (tmps = strchr("([{< )]}> )]}>",term)))
9963 termcode = termstr[0] = term = tmps[5];
9965 PL_multi_close = term;
9967 /* create a new SV to hold the contents. 79 is the SV's initial length.
9968 What a random number. */
9969 sv = newSV_type(SVt_PVIV);
9971 SvIV_set(sv, termcode);
9972 (void)SvPOK_only(sv); /* validate pointer */
9974 /* move past delimiter and try to read a complete string */
9976 sv_catpvn(sv, s, termlen);
9979 tstart = SvPVX(PL_linestr) + stuffstart;
9980 if (!PL_thisopen && !keep_delims) {
9981 PL_thisopen = newSVpvn(tstart, s - tstart);
9982 stuffstart = s - SvPVX(PL_linestr);
9986 if (PL_encoding && !UTF && !re_reparse) {
9990 int offset = s - SvPVX_const(PL_linestr);
9991 const bool found = sv_cat_decode(sv, PL_encoding, PL_linestr,
9992 &offset, (char*)termstr, termlen);
9993 const char * const ns = SvPVX_const(PL_linestr) + offset;
9994 char * const svlast = SvEND(sv) - 1;
9996 for (; s < ns; s++) {
9997 if (*s == '\n' && !PL_rsfp && !PL_parser->filtered)
9998 CopLINE_inc(PL_curcop);
10001 goto read_more_line;
10003 /* handle quoted delimiters */
10004 if (SvCUR(sv) > 1 && *(svlast-1) == '\\') {
10006 for (t = svlast-2; t >= SvPVX_const(sv) && *t == '\\';)
10008 if ((svlast-1 - t) % 2) {
10009 if (!keep_quoted) {
10010 *(svlast-1) = term;
10012 SvCUR_set(sv, SvCUR(sv) - 1);
10017 if (PL_multi_open == PL_multi_close) {
10023 for (t = w = SvPVX(sv)+last_off; t < svlast; w++, t++) {
10024 /* At here, all closes are "was quoted" one,
10025 so we don't check PL_multi_close. */
10027 if (!keep_quoted && *(t+1) == PL_multi_open)
10032 else if (*t == PL_multi_open)
10040 SvCUR_set(sv, w - SvPVX_const(sv));
10042 last_off = w - SvPVX(sv);
10043 if (--brackets <= 0)
10048 if (!keep_delims) {
10049 SvCUR_set(sv, SvCUR(sv) - 1);
10055 /* extend sv if need be */
10056 SvGROW(sv, SvCUR(sv) + (PL_bufend - s) + 1);
10057 /* set 'to' to the next character in the sv's string */
10058 to = SvPVX(sv)+SvCUR(sv);
10060 /* if open delimiter is the close delimiter read unbridle */
10061 if (PL_multi_open == PL_multi_close) {
10062 for (; s < PL_bufend; s++,to++) {
10063 /* embedded newlines increment the current line number */
10064 if (*s == '\n' && !PL_rsfp && !PL_parser->filtered)
10065 CopLINE_inc(PL_curcop);
10066 /* handle quoted delimiters */
10067 if (*s == '\\' && s+1 < PL_bufend && term != '\\') {
10070 || (re_reparse && s[1] == '\\'))
10073 /* any other quotes are simply copied straight through */
10077 /* terminate when run out of buffer (the for() condition), or
10078 have found the terminator */
10079 else if (*s == term) {
10082 if (s+termlen <= PL_bufend && memEQ(s, (char*)termstr, termlen))
10085 else if (!has_utf8 && !UTF8_IS_INVARIANT((U8)*s) && UTF)
10091 /* if the terminator isn't the same as the start character (e.g.,
10092 matched brackets), we have to allow more in the quoting, and
10093 be prepared for nested brackets.
10096 /* read until we run out of string, or we find the terminator */
10097 for (; s < PL_bufend; s++,to++) {
10098 /* embedded newlines increment the line count */
10099 if (*s == '\n' && !PL_rsfp && !PL_parser->filtered)
10100 CopLINE_inc(PL_curcop);
10101 /* backslashes can escape the open or closing characters */
10102 if (*s == '\\' && s+1 < PL_bufend) {
10103 if (!keep_quoted &&
10104 ((s[1] == PL_multi_open) || (s[1] == PL_multi_close)))
10109 /* allow nested opens and closes */
10110 else if (*s == PL_multi_close && --brackets <= 0)
10112 else if (*s == PL_multi_open)
10114 else if (!has_utf8 && !UTF8_IS_INVARIANT((U8)*s) && UTF)
10119 /* terminate the copied string and update the sv's end-of-string */
10121 SvCUR_set(sv, to - SvPVX_const(sv));
10124 * this next chunk reads more into the buffer if we're not done yet
10128 break; /* handle case where we are done yet :-) */
10130 #ifndef PERL_STRICT_CR
10131 if (to - SvPVX_const(sv) >= 2) {
10132 if ((to[-2] == '\r' && to[-1] == '\n') ||
10133 (to[-2] == '\n' && to[-1] == '\r'))
10137 SvCUR_set(sv, to - SvPVX_const(sv));
10139 else if (to[-1] == '\r')
10142 else if (to - SvPVX_const(sv) == 1 && to[-1] == '\r')
10147 /* if we're out of file, or a read fails, bail and reset the current
10148 line marker so we can report where the unterminated string began
10151 if (PL_madskills) {
10152 char * const tstart = SvPVX(PL_linestr) + stuffstart;
10154 sv_catpvn(PL_thisstuff, tstart, PL_bufend - tstart);
10156 PL_thisstuff = newSVpvn(tstart, PL_bufend - tstart);
10159 CopLINE_inc(PL_curcop);
10160 PL_bufptr = PL_bufend;
10161 if (!lex_next_chunk(0)) {
10163 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
10172 /* at this point, we have successfully read the delimited string */
10174 if (!PL_encoding || UTF || re_reparse) {
10176 if (PL_madskills) {
10177 char * const tstart = SvPVX(PL_linestr) + stuffstart;
10178 const int len = s - tstart;
10180 sv_catpvn(PL_thisstuff, tstart, len);
10182 PL_thisstuff = newSVpvn(tstart, len);
10183 if (!PL_thisclose && !keep_delims)
10184 PL_thisclose = newSVpvn(s,termlen);
10189 sv_catpvn(sv, s, termlen);
10194 if (PL_madskills) {
10195 char * const tstart = SvPVX(PL_linestr) + stuffstart;
10196 const int len = s - tstart - termlen;
10198 sv_catpvn(PL_thisstuff, tstart, len);
10200 PL_thisstuff = newSVpvn(tstart, len);
10201 if (!PL_thisclose && !keep_delims)
10202 PL_thisclose = newSVpvn(s - termlen,termlen);
10206 if (has_utf8 || (PL_encoding && !re_reparse))
10209 PL_multi_end = CopLINE(PL_curcop);
10211 /* if we allocated too much space, give some back */
10212 if (SvCUR(sv) + 5 < SvLEN(sv)) {
10213 SvLEN_set(sv, SvCUR(sv) + 1);
10214 SvPV_renew(sv, SvLEN(sv));
10217 /* decide whether this is the first or second quoted string we've read
10230 takes: pointer to position in buffer
10231 returns: pointer to new position in buffer
10232 side-effects: builds ops for the constant in pl_yylval.op
10234 Read a number in any of the formats that Perl accepts:
10236 \d(_?\d)*(\.(\d(_?\d)*)?)?[Ee][\+\-]?(\d(_?\d)*) 12 12.34 12.
10237 \.\d(_?\d)*[Ee][\+\-]?(\d(_?\d)*) .34
10240 0x[0-9A-Fa-f](_?[0-9A-Fa-f])*
10242 Like most scan_ routines, it uses the PL_tokenbuf buffer to hold the
10245 If it reads a number without a decimal point or an exponent, it will
10246 try converting the number to an integer and see if it can do so
10247 without loss of precision.
10251 Perl_scan_num(pTHX_ const char *start, YYSTYPE* lvalp)
10254 register const char *s = start; /* current position in buffer */
10255 register char *d; /* destination in temp buffer */
10256 register char *e; /* end of temp buffer */
10257 NV nv; /* number read, as a double */
10258 SV *sv = NULL; /* place to put the converted number */
10259 bool floatit; /* boolean: int or float? */
10260 const char *lastub = NULL; /* position of last underbar */
10261 static char const number_too_long[] = "Number too long";
10263 PERL_ARGS_ASSERT_SCAN_NUM;
10265 /* We use the first character to decide what type of number this is */
10269 Perl_croak(aTHX_ "panic: scan_num, *s=%d", *s);
10271 /* if it starts with a 0, it could be an octal number, a decimal in
10272 0.13 disguise, or a hexadecimal number, or a binary number. */
10276 u holds the "number so far"
10277 shift the power of 2 of the base
10278 (hex == 4, octal == 3, binary == 1)
10279 overflowed was the number more than we can hold?
10281 Shift is used when we add a digit. It also serves as an "are
10282 we in octal/hex/binary?" indicator to disallow hex characters
10283 when in octal mode.
10288 bool overflowed = FALSE;
10289 bool just_zero = TRUE; /* just plain 0 or binary number? */
10290 static const NV nvshift[5] = { 1.0, 2.0, 4.0, 8.0, 16.0 };
10291 static const char* const bases[5] =
10292 { "", "binary", "", "octal", "hexadecimal" };
10293 static const char* const Bases[5] =
10294 { "", "Binary", "", "Octal", "Hexadecimal" };
10295 static const char* const maxima[5] =
10297 "0b11111111111111111111111111111111",
10301 const char *base, *Base, *max;
10303 /* check for hex */
10304 if (s[1] == 'x' || s[1] == 'X') {
10308 } else if (s[1] == 'b' || s[1] == 'B') {
10313 /* check for a decimal in disguise */
10314 else if (s[1] == '.' || s[1] == 'e' || s[1] == 'E')
10316 /* so it must be octal */
10323 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10324 "Misplaced _ in number");
10328 base = bases[shift];
10329 Base = Bases[shift];
10330 max = maxima[shift];
10332 /* read the rest of the number */
10334 /* x is used in the overflow test,
10335 b is the digit we're adding on. */
10340 /* if we don't mention it, we're done */
10344 /* _ are ignored -- but warned about if consecutive */
10346 if (lastub && s == lastub + 1)
10347 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10348 "Misplaced _ in number");
10352 /* 8 and 9 are not octal */
10353 case '8': case '9':
10355 yyerror(Perl_form(aTHX_ "Illegal octal digit '%c'", *s));
10359 case '2': case '3': case '4':
10360 case '5': case '6': case '7':
10362 yyerror(Perl_form(aTHX_ "Illegal binary digit '%c'", *s));
10365 case '0': case '1':
10366 b = *s++ & 15; /* ASCII digit -> value of digit */
10370 case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
10371 case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
10372 /* make sure they said 0x */
10375 b = (*s++ & 7) + 9;
10377 /* Prepare to put the digit we have onto the end
10378 of the number so far. We check for overflows.
10384 x = u << shift; /* make room for the digit */
10386 if ((x >> shift) != u
10387 && !(PL_hints & HINT_NEW_BINARY)) {
10390 Perl_ck_warner_d(aTHX_ packWARN(WARN_OVERFLOW),
10391 "Integer overflow in %s number",
10394 u = x | b; /* add the digit to the end */
10397 n *= nvshift[shift];
10398 /* If an NV has not enough bits in its
10399 * mantissa to represent an UV this summing of
10400 * small low-order numbers is a waste of time
10401 * (because the NV cannot preserve the
10402 * low-order bits anyway): we could just
10403 * remember when did we overflow and in the
10404 * end just multiply n by the right
10412 /* if we get here, we had success: make a scalar value from
10417 /* final misplaced underbar check */
10418 if (s[-1] == '_') {
10419 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX), "Misplaced _ in number");
10423 if (n > 4294967295.0)
10424 Perl_ck_warner(aTHX_ packWARN(WARN_PORTABLE),
10425 "%s number > %s non-portable",
10431 if (u > 0xffffffff)
10432 Perl_ck_warner(aTHX_ packWARN(WARN_PORTABLE),
10433 "%s number > %s non-portable",
10438 if (just_zero && (PL_hints & HINT_NEW_INTEGER))
10439 sv = new_constant(start, s - start, "integer",
10440 sv, NULL, NULL, 0);
10441 else if (PL_hints & HINT_NEW_BINARY)
10442 sv = new_constant(start, s - start, "binary", sv, NULL, NULL, 0);
10447 handle decimal numbers.
10448 we're also sent here when we read a 0 as the first digit
10450 case '1': case '2': case '3': case '4': case '5':
10451 case '6': case '7': case '8': case '9': case '.':
10454 e = PL_tokenbuf + sizeof PL_tokenbuf - 6; /* room for various punctuation */
10457 /* read next group of digits and _ and copy into d */
10458 while (isDIGIT(*s) || *s == '_') {
10459 /* skip underscores, checking for misplaced ones
10463 if (lastub && s == lastub + 1)
10464 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10465 "Misplaced _ in number");
10469 /* check for end of fixed-length buffer */
10471 Perl_croak(aTHX_ number_too_long);
10472 /* if we're ok, copy the character */
10477 /* final misplaced underbar check */
10478 if (lastub && s == lastub + 1) {
10479 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX), "Misplaced _ in number");
10482 /* read a decimal portion if there is one. avoid
10483 3..5 being interpreted as the number 3. followed
10486 if (*s == '.' && s[1] != '.') {
10491 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10492 "Misplaced _ in number");
10496 /* copy, ignoring underbars, until we run out of digits.
10498 for (; isDIGIT(*s) || *s == '_'; s++) {
10499 /* fixed length buffer check */
10501 Perl_croak(aTHX_ number_too_long);
10503 if (lastub && s == lastub + 1)
10504 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10505 "Misplaced _ in number");
10511 /* fractional part ending in underbar? */
10512 if (s[-1] == '_') {
10513 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10514 "Misplaced _ in number");
10516 if (*s == '.' && isDIGIT(s[1])) {
10517 /* oops, it's really a v-string, but without the "v" */
10523 /* read exponent part, if present */
10524 if ((*s == 'e' || *s == 'E') && strchr("+-0123456789_", s[1])) {
10528 /* regardless of whether user said 3E5 or 3e5, use lower 'e' */
10529 *d++ = 'e'; /* At least some Mach atof()s don't grok 'E' */
10531 /* stray preinitial _ */
10533 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10534 "Misplaced _ in number");
10538 /* allow positive or negative exponent */
10539 if (*s == '+' || *s == '-')
10542 /* stray initial _ */
10544 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10545 "Misplaced _ in number");
10549 /* read digits of exponent */
10550 while (isDIGIT(*s) || *s == '_') {
10553 Perl_croak(aTHX_ number_too_long);
10557 if (((lastub && s == lastub + 1) ||
10558 (!isDIGIT(s[1]) && s[1] != '_')))
10559 Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
10560 "Misplaced _ in number");
10568 We try to do an integer conversion first if no characters
10569 indicating "float" have been found.
10574 const int flags = grok_number (PL_tokenbuf, d - PL_tokenbuf, &uv);
10576 if (flags == IS_NUMBER_IN_UV) {
10578 sv = newSViv(uv); /* Prefer IVs over UVs. */
10581 } else if (flags == (IS_NUMBER_IN_UV | IS_NUMBER_NEG)) {
10582 if (uv <= (UV) IV_MIN)
10583 sv = newSViv(-(IV)uv);
10590 /* terminate the string */
10592 nv = Atof(PL_tokenbuf);
10597 ? (PL_hints & HINT_NEW_FLOAT) : (PL_hints & HINT_NEW_INTEGER) ) {
10598 const char *const key = floatit ? "float" : "integer";
10599 const STRLEN keylen = floatit ? 5 : 7;
10600 sv = S_new_constant(aTHX_ PL_tokenbuf, d - PL_tokenbuf,
10601 key, keylen, sv, NULL, NULL, 0);
10605 /* if it starts with a v, it could be a v-string */
10608 sv = newSV(5); /* preallocate storage space */
10609 s = scan_vstring(s, PL_bufend, sv);
10613 /* make the op for the constant and return */
10616 lvalp->opval = newSVOP(OP_CONST, 0, sv);
10618 lvalp->opval = NULL;
10624 S_scan_formline(pTHX_ register char *s)
10627 register char *eol;
10629 SV * const stuff = newSVpvs("");
10630 bool needargs = FALSE;
10631 bool eofmt = FALSE;
10633 char *tokenstart = s;
10634 SV* savewhite = NULL;
10636 if (PL_madskills) {
10637 savewhite = PL_thiswhite;
10642 PERL_ARGS_ASSERT_SCAN_FORMLINE;
10644 while (!needargs) {
10647 #ifdef PERL_STRICT_CR
10648 while (SPACE_OR_TAB(*t))
10651 while (SPACE_OR_TAB(*t) || *t == '\r')
10654 if (*t == '\n' || t == PL_bufend) {
10659 if (PL_in_eval && !PL_rsfp && !PL_parser->filtered) {
10660 eol = (char *) memchr(s,'\n',PL_bufend-s);
10665 eol = PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
10667 for (t = s; t < eol; t++) {
10668 if (*t == '~' && t[1] == '~' && SvCUR(stuff)) {
10670 goto enough; /* ~~ must be first line in formline */
10672 if (*t == '@' || *t == '^')
10676 sv_catpvn(stuff, s, eol-s);
10677 #ifndef PERL_STRICT_CR
10678 if (eol-s > 1 && eol[-2] == '\r' && eol[-1] == '\n') {
10679 char *end = SvPVX(stuff) + SvCUR(stuff);
10682 SvCUR_set(stuff, SvCUR(stuff) - 1);
10690 if (PL_rsfp || PL_parser->filtered) {
10693 if (PL_madskills) {
10695 sv_catpvn(PL_thistoken, tokenstart, PL_bufend - tokenstart);
10697 PL_thistoken = newSVpvn(tokenstart, PL_bufend - tokenstart);
10700 PL_bufptr = PL_bufend;
10701 CopLINE_inc(PL_curcop);
10702 got_some = lex_next_chunk(0);
10703 CopLINE_dec(PL_curcop);
10706 tokenstart = PL_bufptr;
10714 if (SvCUR(stuff)) {
10717 PL_lex_state = LEX_NORMAL;
10718 start_force(PL_curforce);
10719 NEXTVAL_NEXTTOKE.ival = 0;
10723 PL_lex_state = LEX_FORMLINE;
10725 if (UTF && is_utf8_string((U8*)SvPVX_const(stuff), SvCUR(stuff)))
10727 else if (PL_encoding)
10728 sv_recode_to_utf8(stuff, PL_encoding);
10730 start_force(PL_curforce);
10731 NEXTVAL_NEXTTOKE.opval = (OP*)newSVOP(OP_CONST, 0, stuff);
10733 start_force(PL_curforce);
10734 NEXTVAL_NEXTTOKE.ival = OP_FORMLINE;
10738 SvREFCNT_dec(stuff);
10740 PL_lex_formbrack = 0;
10744 if (PL_madskills) {
10746 sv_catpvn(PL_thistoken, tokenstart, s - tokenstart);
10748 PL_thistoken = newSVpvn(tokenstart, s - tokenstart);
10749 PL_thiswhite = savewhite;
10756 Perl_start_subparse(pTHX_ I32 is_format, U32 flags)
10759 const I32 oldsavestack_ix = PL_savestack_ix;
10760 CV* const outsidecv = PL_compcv;
10763 assert(SvTYPE(PL_compcv) == SVt_PVCV);
10765 SAVEI32(PL_subline);
10766 save_item(PL_subname);
10767 SAVESPTR(PL_compcv);
10769 PL_compcv = MUTABLE_CV(newSV_type(is_format ? SVt_PVFM : SVt_PVCV));
10770 CvFLAGS(PL_compcv) |= flags;
10772 PL_subline = CopLINE(PL_curcop);
10773 CvPADLIST(PL_compcv) = pad_new(padnew_SAVE|padnew_SAVESUB);
10774 CvOUTSIDE(PL_compcv) = MUTABLE_CV(SvREFCNT_inc_simple(outsidecv));
10775 CvOUTSIDE_SEQ(PL_compcv) = PL_cop_seqmax;
10777 return oldsavestack_ix;
10781 #pragma segment Perl_yylex
10784 S_yywarn(pTHX_ const char *const s, U32 flags)
10788 PERL_ARGS_ASSERT_YYWARN;
10790 PL_in_eval |= EVAL_WARNONLY;
10791 yyerror_pv(s, flags);
10792 PL_in_eval &= ~EVAL_WARNONLY;
10797 Perl_yyerror(pTHX_ const char *const s)
10799 PERL_ARGS_ASSERT_YYERROR;
10800 return yyerror_pvn(s, strlen(s), 0);
10804 Perl_yyerror_pv(pTHX_ const char *const s, U32 flags)
10806 PERL_ARGS_ASSERT_YYERROR_PV;
10807 return yyerror_pvn(s, strlen(s), flags);
10811 Perl_yyerror_pvn(pTHX_ const char *const s, STRLEN len, U32 flags)
10814 const char *context = NULL;
10817 SV * const where_sv = newSVpvs_flags("", SVs_TEMP);
10818 int yychar = PL_parser->yychar;
10819 U32 is_utf8 = flags & SVf_UTF8;
10821 PERL_ARGS_ASSERT_YYERROR_PVN;
10823 if (!yychar || (yychar == ';' && !PL_rsfp))
10824 sv_catpvs(where_sv, "at EOF");
10825 else if (PL_oldoldbufptr && PL_bufptr > PL_oldoldbufptr &&
10826 PL_bufptr - PL_oldoldbufptr < 200 && PL_oldoldbufptr != PL_oldbufptr &&
10827 PL_oldbufptr != PL_bufptr) {
10830 The code below is removed for NetWare because it abends/crashes on NetWare
10831 when the script has error such as not having the closing quotes like:
10832 if ($var eq "value)
10833 Checking of white spaces is anyway done in NetWare code.
10836 while (isSPACE(*PL_oldoldbufptr))
10839 context = PL_oldoldbufptr;
10840 contlen = PL_bufptr - PL_oldoldbufptr;
10842 else if (PL_oldbufptr && PL_bufptr > PL_oldbufptr &&
10843 PL_bufptr - PL_oldbufptr < 200 && PL_oldbufptr != PL_bufptr) {
10846 The code below is removed for NetWare because it abends/crashes on NetWare
10847 when the script has error such as not having the closing quotes like:
10848 if ($var eq "value)
10849 Checking of white spaces is anyway done in NetWare code.
10852 while (isSPACE(*PL_oldbufptr))
10855 context = PL_oldbufptr;
10856 contlen = PL_bufptr - PL_oldbufptr;
10858 else if (yychar > 255)
10859 sv_catpvs(where_sv, "next token ???");
10860 else if (yychar == -2) { /* YYEMPTY */
10861 if (PL_lex_state == LEX_NORMAL ||
10862 (PL_lex_state == LEX_KNOWNEXT && PL_lex_defer == LEX_NORMAL))
10863 sv_catpvs(where_sv, "at end of line");
10864 else if (PL_lex_inpat)
10865 sv_catpvs(where_sv, "within pattern");
10867 sv_catpvs(where_sv, "within string");
10870 sv_catpvs(where_sv, "next char ");
10872 Perl_sv_catpvf(aTHX_ where_sv, "^%c", toCTRL(yychar));
10873 else if (isPRINT_LC(yychar)) {
10874 const char string = yychar;
10875 sv_catpvn(where_sv, &string, 1);
10878 Perl_sv_catpvf(aTHX_ where_sv, "\\%03o", yychar & 255);
10880 msg = sv_2mortal(newSVpvn_flags(s, len, is_utf8));
10881 Perl_sv_catpvf(aTHX_ msg, " at %s line %"IVdf", ",
10882 OutCopFILE(PL_curcop), (IV)CopLINE(PL_curcop));
10884 Perl_sv_catpvf(aTHX_ msg, "near \"%"SVf"\"\n",
10885 SVfARG(newSVpvn_flags(context, contlen,
10886 SVs_TEMP | (UTF ? SVf_UTF8 : 0))));
10888 Perl_sv_catpvf(aTHX_ msg, "%"SVf"\n", SVfARG(where_sv));
10889 if (PL_multi_start < PL_multi_end && (U32)(CopLINE(PL_curcop) - PL_multi_end) <= 1) {
10890 Perl_sv_catpvf(aTHX_ msg,
10891 " (Might be a runaway multi-line %c%c string starting on line %"IVdf")\n",
10892 (int)PL_multi_open,(int)PL_multi_close,(IV)PL_multi_start);
10895 if (PL_in_eval & EVAL_WARNONLY) {
10896 Perl_ck_warner_d(aTHX_ packWARN(WARN_SYNTAX), "%"SVf, SVfARG(msg));
10900 if (PL_error_count >= 10) {
10901 if (PL_in_eval && SvCUR(ERRSV))
10902 Perl_croak(aTHX_ "%"SVf"%s has too many errors.\n",
10903 SVfARG(ERRSV), OutCopFILE(PL_curcop));
10905 Perl_croak(aTHX_ "%s has too many errors.\n",
10906 OutCopFILE(PL_curcop));
10909 PL_in_my_stash = NULL;
10913 #pragma segment Main
10917 S_swallow_bom(pTHX_ U8 *s)
10920 const STRLEN slen = SvCUR(PL_linestr);
10922 PERL_ARGS_ASSERT_SWALLOW_BOM;
10926 if (s[1] == 0xFE) {
10927 /* UTF-16 little-endian? (or UTF-32LE?) */
10928 if (s[2] == 0 && s[3] == 0) /* UTF-32 little-endian */
10929 /* diag_listed_as: Unsupported script encoding %s */
10930 Perl_croak(aTHX_ "Unsupported script encoding UTF-32LE");
10931 #ifndef PERL_NO_UTF16_FILTER
10932 if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16LE script encoding (BOM)\n");
10934 if (PL_bufend > (char*)s) {
10935 s = add_utf16_textfilter(s, TRUE);
10938 /* diag_listed_as: Unsupported script encoding %s */
10939 Perl_croak(aTHX_ "Unsupported script encoding UTF-16LE");
10944 if (s[1] == 0xFF) { /* UTF-16 big-endian? */
10945 #ifndef PERL_NO_UTF16_FILTER
10946 if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16BE script encoding (BOM)\n");
10948 if (PL_bufend > (char *)s) {
10949 s = add_utf16_textfilter(s, FALSE);
10952 /* diag_listed_as: Unsupported script encoding %s */
10953 Perl_croak(aTHX_ "Unsupported script encoding UTF-16BE");
10958 if (slen > 2 && s[1] == 0xBB && s[2] == 0xBF) {
10959 if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-8 script encoding (BOM)\n");
10960 s += 3; /* UTF-8 */
10966 if (s[2] == 0xFE && s[3] == 0xFF) {
10967 /* UTF-32 big-endian */
10968 /* diag_listed_as: Unsupported script encoding %s */
10969 Perl_croak(aTHX_ "Unsupported script encoding UTF-32BE");
10972 else if (s[2] == 0 && s[3] != 0) {
10975 * are a good indicator of UTF-16BE. */
10976 #ifndef PERL_NO_UTF16_FILTER
10977 if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16BE script encoding (no BOM)\n");
10978 s = add_utf16_textfilter(s, FALSE);
10980 /* diag_listed_as: Unsupported script encoding %s */
10981 Perl_croak(aTHX_ "Unsupported script encoding UTF-16BE");
10987 if (slen > 3 && s[1] == 0x73 && s[2] == 0x66 && s[3] == 0x73) {
10988 if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-8 script encoding (BOM)\n");
10989 s += 4; /* UTF-8 */
10995 if (slen > 3 && s[1] == 0 && s[2] != 0 && s[3] == 0) {
10998 * are a good indicator of UTF-16LE. */
10999 #ifndef PERL_NO_UTF16_FILTER
11000 if (DEBUG_p_TEST || DEBUG_T_TEST) PerlIO_printf(Perl_debug_log, "UTF-16LE script encoding (no BOM)\n");
11001 s = add_utf16_textfilter(s, TRUE);
11003 /* diag_listed_as: Unsupported script encoding %s */
11004 Perl_croak(aTHX_ "Unsupported script encoding UTF-16LE");
11012 #ifndef PERL_NO_UTF16_FILTER
11014 S_utf16_textfilter(pTHX_ int idx, SV *sv, int maxlen)
11017 SV *const filter = FILTER_DATA(idx);
11018 /* We re-use this each time round, throwing the contents away before we
11020 SV *const utf16_buffer = MUTABLE_SV(IoTOP_GV(filter));
11021 SV *const utf8_buffer = filter;
11022 IV status = IoPAGE(filter);
11023 const bool reverse = cBOOL(IoLINES(filter));
11026 PERL_ARGS_ASSERT_UTF16_TEXTFILTER;
11028 /* As we're automatically added, at the lowest level, and hence only called
11029 from this file, we can be sure that we're not called in block mode. Hence
11030 don't bother writing code to deal with block mode. */
11032 Perl_croak(aTHX_ "panic: utf16_textfilter called in block mode (for %d characters)", maxlen);
11035 Perl_croak(aTHX_ "panic: utf16_textfilter called after error (status=%"IVdf")", status);
11037 DEBUG_P(PerlIO_printf(Perl_debug_log,
11038 "utf16_textfilter(%p,%ce): idx=%d maxlen=%d status=%"IVdf" utf16=%"UVuf" utf8=%"UVuf"\n",
11039 FPTR2DPTR(void *, S_utf16_textfilter),
11040 reverse ? 'l' : 'b', idx, maxlen, status,
11041 (UV)SvCUR(utf16_buffer), (UV)SvCUR(utf8_buffer)));
11048 /* First, look in our buffer of existing UTF-8 data: */
11049 char *nl = (char *)memchr(SvPVX(utf8_buffer), '\n', SvCUR(utf8_buffer));
11053 } else if (status == 0) {
11055 IoPAGE(filter) = 0;
11056 nl = SvEND(utf8_buffer);
11059 STRLEN got = nl - SvPVX(utf8_buffer);
11060 /* Did we have anything to append? */
11062 sv_catpvn(sv, SvPVX(utf8_buffer), got);
11063 /* Everything else in this code works just fine if SVp_POK isn't
11064 set. This, however, needs it, and we need it to work, else
11065 we loop infinitely because the buffer is never consumed. */
11066 sv_chop(utf8_buffer, nl);
11070 /* OK, not a complete line there, so need to read some more UTF-16.
11071 Read an extra octect if the buffer currently has an odd number. */
11075 if (SvCUR(utf16_buffer) >= 2) {
11076 /* Location of the high octet of the last complete code point.
11077 Gosh, UTF-16 is a pain. All the benefits of variable length,
11078 *coupled* with all the benefits of partial reads and
11080 const U8 *const last_hi = (U8*)SvPVX(utf16_buffer)
11081 + ((SvCUR(utf16_buffer) & ~1) - (reverse ? 1 : 2));
11083 if (*last_hi < 0xd8 || *last_hi > 0xdb) {
11087 /* We have the first half of a surrogate. Read more. */
11088 DEBUG_P(PerlIO_printf(Perl_debug_log, "utf16_textfilter partial surrogate detected at %p\n", last_hi));
11091 status = FILTER_READ(idx + 1, utf16_buffer,
11092 160 + (SvCUR(utf16_buffer) & 1));
11093 DEBUG_P(PerlIO_printf(Perl_debug_log, "utf16_textfilter status=%"IVdf" SvCUR(sv)=%"UVuf"\n", status, (UV)SvCUR(utf16_buffer)));
11094 DEBUG_P({ sv_dump(utf16_buffer); sv_dump(utf8_buffer);});
11097 IoPAGE(filter) = status;
11102 chars = SvCUR(utf16_buffer) >> 1;
11103 have = SvCUR(utf8_buffer);
11104 SvGROW(utf8_buffer, have + chars * 3 + 1);
11107 end = utf16_to_utf8_reversed((U8*)SvPVX(utf16_buffer),
11108 (U8*)SvPVX_const(utf8_buffer) + have,
11109 chars * 2, &newlen);
11111 end = utf16_to_utf8((U8*)SvPVX(utf16_buffer),
11112 (U8*)SvPVX_const(utf8_buffer) + have,
11113 chars * 2, &newlen);
11115 SvCUR_set(utf8_buffer, have + newlen);
11118 /* No need to keep this SV "well-formed" with a '\0' after the end, as
11119 it's private to us, and utf16_to_utf8{,reversed} take a
11120 (pointer,length) pair, rather than a NUL-terminated string. */
11121 if(SvCUR(utf16_buffer) & 1) {
11122 *SvPVX(utf16_buffer) = SvEND(utf16_buffer)[-1];
11123 SvCUR_set(utf16_buffer, 1);
11125 SvCUR_set(utf16_buffer, 0);
11128 DEBUG_P(PerlIO_printf(Perl_debug_log,
11129 "utf16_textfilter: returns, status=%"IVdf" utf16=%"UVuf" utf8=%"UVuf"\n",
11131 (UV)SvCUR(utf16_buffer), (UV)SvCUR(utf8_buffer)));
11132 DEBUG_P({ sv_dump(utf8_buffer); sv_dump(sv);});
11137 S_add_utf16_textfilter(pTHX_ U8 *const s, bool reversed)
11139 SV *filter = filter_add(S_utf16_textfilter, NULL);
11141 PERL_ARGS_ASSERT_ADD_UTF16_TEXTFILTER;
11143 IoTOP_GV(filter) = MUTABLE_GV(newSVpvn((char *)s, PL_bufend - (char*)s));
11144 sv_setpvs(filter, "");
11145 IoLINES(filter) = reversed;
11146 IoPAGE(filter) = 1; /* Not EOF */
11148 /* Sadly, we have to return a valid pointer, come what may, so we have to
11149 ignore any error return from this. */
11150 SvCUR_set(PL_linestr, 0);
11151 if (FILTER_READ(0, PL_linestr, 0)) {
11152 SvUTF8_on(PL_linestr);
11154 SvUTF8_on(PL_linestr);
11156 PL_bufend = SvEND(PL_linestr);
11157 return (U8*)SvPVX(PL_linestr);
11162 Returns a pointer to the next character after the parsed
11163 vstring, as well as updating the passed in sv.
11165 Function must be called like
11168 s = scan_vstring(s,e,sv);
11170 where s and e are the start and end of the string.
11171 The sv should already be large enough to store the vstring
11172 passed in, for performance reasons.
11177 Perl_scan_vstring(pTHX_ const char *s, const char *const e, SV *sv)
11180 const char *pos = s;
11181 const char *start = s;
11183 PERL_ARGS_ASSERT_SCAN_VSTRING;
11185 if (*pos == 'v') pos++; /* get past 'v' */
11186 while (pos < e && (isDIGIT(*pos) || *pos == '_'))
11188 if ( *pos != '.') {
11189 /* this may not be a v-string if followed by => */
11190 const char *next = pos;
11191 while (next < e && isSPACE(*next))
11193 if ((e - next) >= 2 && *next == '=' && next[1] == '>' ) {
11194 /* return string not v-string */
11195 sv_setpvn(sv,(char *)s,pos-s);
11196 return (char *)pos;
11200 if (!isALPHA(*pos)) {
11201 U8 tmpbuf[UTF8_MAXBYTES+1];
11204 s++; /* get past 'v' */
11209 /* this is atoi() that tolerates underscores */
11212 const char *end = pos;
11214 while (--end >= s) {
11216 const UV orev = rev;
11217 rev += (*end - '0') * mult;
11220 /* diag_listed_as: Integer overflow in %s number */
11221 Perl_ck_warner_d(aTHX_ packWARN(WARN_OVERFLOW),
11222 "Integer overflow in decimal number");
11226 if (rev > 0x7FFFFFFF)
11227 Perl_croak(aTHX_ "In EBCDIC the v-string components cannot exceed 2147483647");
11229 /* Append native character for the rev point */
11230 tmpend = uvchr_to_utf8(tmpbuf, rev);
11231 sv_catpvn(sv, (const char*)tmpbuf, tmpend - tmpbuf);
11232 if (!UNI_IS_INVARIANT(NATIVE_TO_UNI(rev)))
11234 if (pos + 1 < e && *pos == '.' && isDIGIT(pos[1]))
11240 while (pos < e && (isDIGIT(*pos) || *pos == '_'))
11244 sv_magic(sv,NULL,PERL_MAGIC_vstring,(const char*)start, pos-start);
11251 Perl_keyword_plugin_standard(pTHX_
11252 char *keyword_ptr, STRLEN keyword_len, OP **op_ptr)
11254 PERL_ARGS_ASSERT_KEYWORD_PLUGIN_STANDARD;
11255 PERL_UNUSED_CONTEXT;
11256 PERL_UNUSED_ARG(keyword_ptr);
11257 PERL_UNUSED_ARG(keyword_len);
11258 PERL_UNUSED_ARG(op_ptr);
11259 return KEYWORD_PLUGIN_DECLINE;
11262 #define parse_recdescent(g,p) S_parse_recdescent(aTHX_ g,p)
11264 S_parse_recdescent(pTHX_ int gramtype, I32 fakeeof)
11266 SAVEI32(PL_lex_brackets);
11267 if (PL_lex_brackets > 100)
11268 Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
11269 PL_lex_brackstack[PL_lex_brackets++] = XFAKEEOF;
11270 SAVEI32(PL_lex_allbrackets);
11271 PL_lex_allbrackets = 0;
11272 SAVEI8(PL_lex_fakeeof);
11273 PL_lex_fakeeof = (U8)fakeeof;
11274 if(yyparse(gramtype) && !PL_parser->error_count)
11275 qerror(Perl_mess(aTHX_ "Parse error"));
11278 #define parse_recdescent_for_op(g,p) S_parse_recdescent_for_op(aTHX_ g,p)
11280 S_parse_recdescent_for_op(pTHX_ int gramtype, I32 fakeeof)
11284 SAVEVPTR(PL_eval_root);
11285 PL_eval_root = NULL;
11286 parse_recdescent(gramtype, fakeeof);
11292 #define parse_expr(p,f) S_parse_expr(aTHX_ p,f)
11294 S_parse_expr(pTHX_ I32 fakeeof, U32 flags)
11297 if (flags & ~PARSE_OPTIONAL)
11298 Perl_croak(aTHX_ "Parsing code internal error (%s)", "parse_expr");
11299 exprop = parse_recdescent_for_op(GRAMEXPR, fakeeof);
11300 if (!exprop && !(flags & PARSE_OPTIONAL)) {
11301 if (!PL_parser->error_count)
11302 qerror(Perl_mess(aTHX_ "Parse error"));
11303 exprop = newOP(OP_NULL, 0);
11309 =for apidoc Amx|OP *|parse_arithexpr|U32 flags
11311 Parse a Perl arithmetic expression. This may contain operators of precedence
11312 down to the bit shift operators. The expression must be followed (and thus
11313 terminated) either by a comparison or lower-precedence operator or by
11314 something that would normally terminate an expression such as semicolon.
11315 If I<flags> includes C<PARSE_OPTIONAL> then the expression is optional,
11316 otherwise it is mandatory. It is up to the caller to ensure that the
11317 dynamic parser state (L</PL_parser> et al) is correctly set to reflect
11318 the source of the code to be parsed and the lexical context for the
11321 The op tree representing the expression is returned. If an optional
11322 expression is absent, a null pointer is returned, otherwise the pointer
11325 If an error occurs in parsing or compilation, in most cases a valid op
11326 tree is returned anyway. The error is reflected in the parser state,
11327 normally resulting in a single exception at the top level of parsing
11328 which covers all the compilation errors that occurred. Some compilation
11329 errors, however, will throw an exception immediately.
11335 Perl_parse_arithexpr(pTHX_ U32 flags)
11337 return parse_expr(LEX_FAKEEOF_COMPARE, flags);
11341 =for apidoc Amx|OP *|parse_termexpr|U32 flags
11343 Parse a Perl term expression. This may contain operators of precedence
11344 down to the assignment operators. The expression must be followed (and thus
11345 terminated) either by a comma or lower-precedence operator or by
11346 something that would normally terminate an expression such as semicolon.
11347 If I<flags> includes C<PARSE_OPTIONAL> then the expression is optional,
11348 otherwise it is mandatory. It is up to the caller to ensure that the
11349 dynamic parser state (L</PL_parser> et al) is correctly set to reflect
11350 the source of the code to be parsed and the lexical context for the
11353 The op tree representing the expression is returned. If an optional
11354 expression is absent, a null pointer is returned, otherwise the pointer
11357 If an error occurs in parsing or compilation, in most cases a valid op
11358 tree is returned anyway. The error is reflected in the parser state,
11359 normally resulting in a single exception at the top level of parsing
11360 which covers all the compilation errors that occurred. Some compilation
11361 errors, however, will throw an exception immediately.
11367 Perl_parse_termexpr(pTHX_ U32 flags)
11369 return parse_expr(LEX_FAKEEOF_COMMA, flags);
11373 =for apidoc Amx|OP *|parse_listexpr|U32 flags
11375 Parse a Perl list expression. This may contain operators of precedence
11376 down to the comma operator. The expression must be followed (and thus
11377 terminated) either by a low-precedence logic operator such as C<or> or by
11378 something that would normally terminate an expression such as semicolon.
11379 If I<flags> includes C<PARSE_OPTIONAL> then the expression is optional,
11380 otherwise it is mandatory. It is up to the caller to ensure that the
11381 dynamic parser state (L</PL_parser> et al) is correctly set to reflect
11382 the source of the code to be parsed and the lexical context for the
11385 The op tree representing the expression is returned. If an optional
11386 expression is absent, a null pointer is returned, otherwise the pointer
11389 If an error occurs in parsing or compilation, in most cases a valid op
11390 tree is returned anyway. The error is reflected in the parser state,
11391 normally resulting in a single exception at the top level of parsing
11392 which covers all the compilation errors that occurred. Some compilation
11393 errors, however, will throw an exception immediately.
11399 Perl_parse_listexpr(pTHX_ U32 flags)
11401 return parse_expr(LEX_FAKEEOF_LOWLOGIC, flags);
11405 =for apidoc Amx|OP *|parse_fullexpr|U32 flags
11407 Parse a single complete Perl expression. This allows the full
11408 expression grammar, including the lowest-precedence operators such
11409 as C<or>. The expression must be followed (and thus terminated) by a
11410 token that an expression would normally be terminated by: end-of-file,
11411 closing bracketing punctuation, semicolon, or one of the keywords that
11412 signals a postfix expression-statement modifier. If I<flags> includes
11413 C<PARSE_OPTIONAL> then the expression is optional, otherwise it is
11414 mandatory. It is up to the caller to ensure that the dynamic parser
11415 state (L</PL_parser> et al) is correctly set to reflect the source of
11416 the code to be parsed and the lexical context for the expression.
11418 The op tree representing the expression is returned. If an optional
11419 expression is absent, a null pointer is returned, otherwise the pointer
11422 If an error occurs in parsing or compilation, in most cases a valid op
11423 tree is returned anyway. The error is reflected in the parser state,
11424 normally resulting in a single exception at the top level of parsing
11425 which covers all the compilation errors that occurred. Some compilation
11426 errors, however, will throw an exception immediately.
11432 Perl_parse_fullexpr(pTHX_ U32 flags)
11434 return parse_expr(LEX_FAKEEOF_NONEXPR, flags);
11438 =for apidoc Amx|OP *|parse_block|U32 flags
11440 Parse a single complete Perl code block. This consists of an opening
11441 brace, a sequence of statements, and a closing brace. The block
11442 constitutes a lexical scope, so C<my> variables and various compile-time
11443 effects can be contained within it. It is up to the caller to ensure
11444 that the dynamic parser state (L</PL_parser> et al) is correctly set to
11445 reflect the source of the code to be parsed and the lexical context for
11448 The op tree representing the code block is returned. This is always a
11449 real op, never a null pointer. It will normally be a C<lineseq> list,
11450 including C<nextstate> or equivalent ops. No ops to construct any kind
11451 of runtime scope are included by virtue of it being a block.
11453 If an error occurs in parsing or compilation, in most cases a valid op
11454 tree (most likely null) is returned anyway. The error is reflected in
11455 the parser state, normally resulting in a single exception at the top
11456 level of parsing which covers all the compilation errors that occurred.
11457 Some compilation errors, however, will throw an exception immediately.
11459 The I<flags> parameter is reserved for future use, and must always
11466 Perl_parse_block(pTHX_ U32 flags)
11469 Perl_croak(aTHX_ "Parsing code internal error (%s)", "parse_block");
11470 return parse_recdescent_for_op(GRAMBLOCK, LEX_FAKEEOF_NEVER);
11474 =for apidoc Amx|OP *|parse_barestmt|U32 flags
11476 Parse a single unadorned Perl statement. This may be a normal imperative
11477 statement or a declaration that has compile-time effect. It does not
11478 include any label or other affixture. It is up to the caller to ensure
11479 that the dynamic parser state (L</PL_parser> et al) is correctly set to
11480 reflect the source of the code to be parsed and the lexical context for
11483 The op tree representing the statement is returned. This may be a
11484 null pointer if the statement is null, for example if it was actually
11485 a subroutine definition (which has compile-time side effects). If not
11486 null, it will be ops directly implementing the statement, suitable to
11487 pass to L</newSTATEOP>. It will not normally include a C<nextstate> or
11488 equivalent op (except for those embedded in a scope contained entirely
11489 within the statement).
11491 If an error occurs in parsing or compilation, in most cases a valid op
11492 tree (most likely null) is returned anyway. The error is reflected in
11493 the parser state, normally resulting in a single exception at the top
11494 level of parsing which covers all the compilation errors that occurred.
11495 Some compilation errors, however, will throw an exception immediately.
11497 The I<flags> parameter is reserved for future use, and must always
11504 Perl_parse_barestmt(pTHX_ U32 flags)
11507 Perl_croak(aTHX_ "Parsing code internal error (%s)", "parse_barestmt");
11508 return parse_recdescent_for_op(GRAMBARESTMT, LEX_FAKEEOF_NEVER);
11512 =for apidoc Amx|SV *|parse_label|U32 flags
11514 Parse a single label, possibly optional, of the type that may prefix a
11515 Perl statement. It is up to the caller to ensure that the dynamic parser
11516 state (L</PL_parser> et al) is correctly set to reflect the source of
11517 the code to be parsed. If I<flags> includes C<PARSE_OPTIONAL> then the
11518 label is optional, otherwise it is mandatory.
11520 The name of the label is returned in the form of a fresh scalar. If an
11521 optional label is absent, a null pointer is returned.
11523 If an error occurs in parsing, which can only occur if the label is
11524 mandatory, a valid label is returned anyway. The error is reflected in
11525 the parser state, normally resulting in a single exception at the top
11526 level of parsing which covers all the compilation errors that occurred.
11532 Perl_parse_label(pTHX_ U32 flags)
11534 if (flags & ~PARSE_OPTIONAL)
11535 Perl_croak(aTHX_ "Parsing code internal error (%s)", "parse_label");
11536 if (PL_lex_state == LEX_KNOWNEXT) {
11537 PL_parser->yychar = yylex();
11538 if (PL_parser->yychar == LABEL) {
11540 PL_parser->yychar = YYEMPTY;
11541 lsv = newSV_type(SVt_PV);
11542 sv_copypv(lsv, cSVOPx(pl_yylval.opval)->op_sv);
11550 STRLEN wlen, bufptr_pos;
11553 if (!isIDFIRST_lazy_if(s, UTF))
11555 t = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &wlen);
11556 if (word_takes_any_delimeter(s, wlen))
11558 bufptr_pos = s - SvPVX(PL_linestr);
11560 lex_read_space(LEX_KEEP_PREVIOUS);
11562 s = SvPVX(PL_linestr) + bufptr_pos;
11563 if (t[0] == ':' && t[1] != ':') {
11564 PL_oldoldbufptr = PL_oldbufptr;
11567 return newSVpvn_flags(s, wlen, UTF ? SVf_UTF8 : 0);
11571 if (flags & PARSE_OPTIONAL) {
11574 qerror(Perl_mess(aTHX_ "Parse error"));
11575 return newSVpvs("x");
11582 =for apidoc Amx|OP *|parse_fullstmt|U32 flags
11584 Parse a single complete Perl statement. This may be a normal imperative
11585 statement or a declaration that has compile-time effect, and may include
11586 optional labels. It is up to the caller to ensure that the dynamic
11587 parser state (L</PL_parser> et al) is correctly set to reflect the source
11588 of the code to be parsed and the lexical context for the statement.
11590 The op tree representing the statement is returned. This may be a
11591 null pointer if the statement is null, for example if it was actually
11592 a subroutine definition (which has compile-time side effects). If not
11593 null, it will be the result of a L</newSTATEOP> call, normally including
11594 a C<nextstate> or equivalent op.
11596 If an error occurs in parsing or compilation, in most cases a valid op
11597 tree (most likely null) is returned anyway. The error is reflected in
11598 the parser state, normally resulting in a single exception at the top
11599 level of parsing which covers all the compilation errors that occurred.
11600 Some compilation errors, however, will throw an exception immediately.
11602 The I<flags> parameter is reserved for future use, and must always
11609 Perl_parse_fullstmt(pTHX_ U32 flags)
11612 Perl_croak(aTHX_ "Parsing code internal error (%s)", "parse_fullstmt");
11613 return parse_recdescent_for_op(GRAMFULLSTMT, LEX_FAKEEOF_NEVER);
11617 =for apidoc Amx|OP *|parse_stmtseq|U32 flags
11619 Parse a sequence of zero or more Perl statements. These may be normal
11620 imperative statements, including optional labels, or declarations
11621 that have compile-time effect, or any mixture thereof. The statement
11622 sequence ends when a closing brace or end-of-file is encountered in a
11623 place where a new statement could have validly started. It is up to
11624 the caller to ensure that the dynamic parser state (L</PL_parser> et al)
11625 is correctly set to reflect the source of the code to be parsed and the
11626 lexical context for the statements.
11628 The op tree representing the statement sequence is returned. This may
11629 be a null pointer if the statements were all null, for example if there
11630 were no statements or if there were only subroutine definitions (which
11631 have compile-time side effects). If not null, it will be a C<lineseq>
11632 list, normally including C<nextstate> or equivalent ops.
11634 If an error occurs in parsing or compilation, in most cases a valid op
11635 tree is returned anyway. The error is reflected in the parser state,
11636 normally resulting in a single exception at the top level of parsing
11637 which covers all the compilation errors that occurred. Some compilation
11638 errors, however, will throw an exception immediately.
11640 The I<flags> parameter is reserved for future use, and must always
11647 Perl_parse_stmtseq(pTHX_ U32 flags)
11652 Perl_croak(aTHX_ "Parsing code internal error (%s)", "parse_stmtseq");
11653 stmtseqop = parse_recdescent_for_op(GRAMSTMTSEQ, LEX_FAKEEOF_CLOSING);
11654 c = lex_peek_unichar(0);
11655 if (c != -1 && c != /*{*/'}')
11656 qerror(Perl_mess(aTHX_ "Parse error"));
11662 * c-indentation-style: bsd
11663 * c-basic-offset: 4
11664 * indent-tabs-mode: nil
11667 * ex: set ts=8 sts=4 sw=4 et: