3 * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
4 * 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 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
16 * This file is the lexer for Perl. It's closely linked to the
19 * The main routine is yylex(), which returns the next token.
23 #define PERL_IN_TOKE_C
26 #define yylval (PL_parser->yylval)
28 /* YYINITDEPTH -- initial size of the parser's stacks. */
29 #define YYINITDEPTH 200
31 /* XXX temporary backwards compatibility */
32 #define PL_lex_brackets (PL_parser->lex_brackets)
33 #define PL_lex_brackstack (PL_parser->lex_brackstack)
34 #define PL_lex_casemods (PL_parser->lex_casemods)
35 #define PL_lex_casestack (PL_parser->lex_casestack)
36 #define PL_lex_defer (PL_parser->lex_defer)
37 #define PL_lex_dojoin (PL_parser->lex_dojoin)
38 #define PL_lex_expect (PL_parser->lex_expect)
39 #define PL_lex_formbrack (PL_parser->lex_formbrack)
40 #define PL_lex_inpat (PL_parser->lex_inpat)
41 #define PL_lex_inwhat (PL_parser->lex_inwhat)
42 #define PL_lex_op (PL_parser->lex_op)
43 #define PL_lex_repl (PL_parser->lex_repl)
44 #define PL_lex_starts (PL_parser->lex_starts)
45 #define PL_lex_stuff (PL_parser->lex_stuff)
46 #define PL_multi_start (PL_parser->multi_start)
47 #define PL_multi_open (PL_parser->multi_open)
48 #define PL_multi_close (PL_parser->multi_close)
49 #define PL_pending_ident (PL_parser->pending_ident)
50 #define PL_preambled (PL_parser->preambled)
51 #define PL_sublex_info (PL_parser->sublex_info)
54 # define PL_endwhite (PL_parser->endwhite)
55 # define PL_faketokens (PL_parser->faketokens)
56 # define PL_lasttoke (PL_parser->lasttoke)
57 # define PL_nextwhite (PL_parser->nextwhite)
58 # define PL_realtokenstart (PL_parser->realtokenstart)
59 # define PL_skipwhite (PL_parser->skipwhite)
60 # define PL_thisclose (PL_parser->thisclose)
61 # define PL_thismad (PL_parser->thismad)
62 # define PL_thisopen (PL_parser->thisopen)
63 # define PL_thisstuff (PL_parser->thisstuff)
64 # define PL_thistoken (PL_parser->thistoken)
65 # define PL_thiswhite (PL_parser->thiswhite)
69 S_pending_ident(pTHX);
71 static const char ident_too_long[] = "Identifier too long";
72 static const char commaless_variable_list[] = "comma-less variable list";
74 static void restore_rsfp(pTHX_ void *f);
75 #ifndef PERL_NO_UTF16_FILTER
76 static I32 utf16_textfilter(pTHX_ int idx, SV *sv, int maxlen);
77 static I32 utf16rev_textfilter(pTHX_ int idx, SV *sv, int maxlen);
81 # define CURMAD(slot,sv) if (PL_madskills) { curmad(slot,sv); sv = 0; }
82 # define NEXTVAL_NEXTTOKE PL_nexttoke[PL_curforce].next_val
84 # define CURMAD(slot,sv)
85 # define NEXTVAL_NEXTTOKE PL_nextval[PL_nexttoke]
88 #define XFAKEBRACK 128
91 #ifdef USE_UTF8_SCRIPTS
92 # define UTF (!IN_BYTES)
94 # define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
97 /* In variables named $^X, these are the legal values for X.
98 * 1999-02-27 mjd-perl-patch@plover.com */
99 #define isCONTROLVAR(x) (isUPPER(x) || strchr("[\\]^_?", (x)))
101 /* On MacOS, respect nonbreaking spaces */
102 #ifdef MACOS_TRADITIONAL
103 #define SPACE_OR_TAB(c) ((c)==' '||(c)=='\312'||(c)=='\t')
105 #define SPACE_OR_TAB(c) ((c)==' '||(c)=='\t')
108 /* LEX_* are values for PL_lex_state, the state of the lexer.
109 * They are arranged oddly so that the guard on the switch statement
110 * can get by with a single comparison (if the compiler is smart enough).
113 /* #define LEX_NOTPARSING 11 is done in perl.h. */
115 #define LEX_NORMAL 10 /* normal code (ie not within "...") */
116 #define LEX_INTERPNORMAL 9 /* code within a string, eg "$foo[$x+1]" */
117 #define LEX_INTERPCASEMOD 8 /* expecting a \U, \Q or \E etc */
118 #define LEX_INTERPPUSH 7 /* starting a new sublex parse level */
119 #define LEX_INTERPSTART 6 /* expecting the start of a $var */
121 /* at end of code, eg "$x" followed by: */
122 #define LEX_INTERPEND 5 /* ... eg not one of [, { or -> */
123 #define LEX_INTERPENDMAYBE 4 /* ... eg one of [, { or -> */
125 #define LEX_INTERPCONCAT 3 /* expecting anything, eg at start of
126 string or after \E, $foo, etc */
127 #define LEX_INTERPCONST 2 /* NOT USED */
128 #define LEX_FORMLINE 1 /* expecting a format line */
129 #define LEX_KNOWNEXT 0 /* next token known; just return it */
133 static const char* const lex_state_names[] = {
152 #include "keywords.h"
154 /* CLINE is a macro that ensures PL_copline has a sane value */
159 #define CLINE (PL_copline = (CopLINE(PL_curcop) < PL_copline ? CopLINE(PL_curcop) : PL_copline))
162 # define SKIPSPACE0(s) skipspace0(s)
163 # define SKIPSPACE1(s) skipspace1(s)
164 # define SKIPSPACE2(s,tsv) skipspace2(s,&tsv)
165 # define PEEKSPACE(s) skipspace2(s,0)
167 # define SKIPSPACE0(s) skipspace(s)
168 # define SKIPSPACE1(s) skipspace(s)
169 # define SKIPSPACE2(s,tsv) skipspace(s)
170 # define PEEKSPACE(s) skipspace(s)
174 * Convenience functions to return different tokens and prime the
175 * lexer for the next token. They all take an argument.
177 * TOKEN : generic token (used for '(', DOLSHARP, etc)
178 * OPERATOR : generic operator
179 * AOPERATOR : assignment operator
180 * PREBLOCK : beginning the block after an if, while, foreach, ...
181 * PRETERMBLOCK : beginning a non-code-defining {} block (eg, hash ref)
182 * PREREF : *EXPR where EXPR is not a simple identifier
183 * TERM : expression term
184 * LOOPX : loop exiting command (goto, last, dump, etc)
185 * FTST : file test operator
186 * FUN0 : zero-argument function
187 * FUN1 : not used, except for not, which isn't a UNIOP
188 * BOop : bitwise or or xor
190 * SHop : shift operator
191 * PWop : power operator
192 * PMop : pattern-matching operator
193 * Aop : addition-level operator
194 * Mop : multiplication-level operator
195 * Eop : equality-testing operator
196 * Rop : relational operator <= != gt
198 * Also see LOP and lop() below.
201 #ifdef DEBUGGING /* Serve -DT. */
202 # define REPORT(retval) tokereport((I32)retval)
204 # define REPORT(retval) (retval)
207 #define TOKEN(retval) return ( PL_bufptr = s, REPORT(retval))
208 #define OPERATOR(retval) return (PL_expect = XTERM, PL_bufptr = s, REPORT(retval))
209 #define AOPERATOR(retval) return ao((PL_expect = XTERM, PL_bufptr = s, REPORT(retval)))
210 #define PREBLOCK(retval) return (PL_expect = XBLOCK,PL_bufptr = s, REPORT(retval))
211 #define PRETERMBLOCK(retval) return (PL_expect = XTERMBLOCK,PL_bufptr = s, REPORT(retval))
212 #define PREREF(retval) return (PL_expect = XREF,PL_bufptr = s, REPORT(retval))
213 #define TERM(retval) return (CLINE, PL_expect = XOPERATOR, PL_bufptr = s, REPORT(retval))
214 #define LOOPX(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)LOOPEX))
215 #define FTST(f) return (yylval.ival=f, PL_expect=XTERMORDORDOR, PL_bufptr=s, REPORT((int)UNIOP))
216 #define FUN0(f) return (yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC0))
217 #define FUN1(f) return (yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC1))
218 #define BOop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITOROP)))
219 #define BAop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITANDOP)))
220 #define SHop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)SHIFTOP)))
221 #define PWop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)POWOP)))
222 #define PMop(f) return(yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MATCHOP))
223 #define Aop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)ADDOP)))
224 #define Mop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MULOP)))
225 #define Eop(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)EQOP))
226 #define Rop(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)RELOP))
228 /* This bit of chicanery makes a unary function followed by
229 * a parenthesis into a function with one argument, highest precedence.
230 * The UNIDOR macro is for unary functions that can be followed by the //
231 * operator (such as C<shift // 0>).
233 #define UNI2(f,x) { \
237 PL_last_uni = PL_oldbufptr; \
238 PL_last_lop_op = f; \
240 return REPORT( (int)FUNC1 ); \
242 return REPORT( *s=='(' ? (int)FUNC1 : (int)UNIOP ); \
244 #define UNI(f) UNI2(f,XTERM)
245 #define UNIDOR(f) UNI2(f,XTERMORDORDOR)
247 #define UNIBRACK(f) { \
250 PL_last_uni = PL_oldbufptr; \
252 return REPORT( (int)FUNC1 ); \
254 return REPORT( (*s == '(') ? (int)FUNC1 : (int)UNIOP ); \
257 /* grandfather return to old style */
258 #define OLDLOP(f) return(yylval.ival=f,PL_expect = XTERM,PL_bufptr = s,(int)LSTOP)
262 /* how to interpret the yylval associated with the token */
266 TOKENTYPE_OPNUM, /* yylval.ival contains an opcode number */
272 static struct debug_tokens {
274 enum token_type type;
276 } const debug_tokens[] =
278 { ADDOP, TOKENTYPE_OPNUM, "ADDOP" },
279 { ANDAND, TOKENTYPE_NONE, "ANDAND" },
280 { ANDOP, TOKENTYPE_NONE, "ANDOP" },
281 { ANONSUB, TOKENTYPE_IVAL, "ANONSUB" },
282 { ARROW, TOKENTYPE_NONE, "ARROW" },
283 { ASSIGNOP, TOKENTYPE_OPNUM, "ASSIGNOP" },
284 { BITANDOP, TOKENTYPE_OPNUM, "BITANDOP" },
285 { BITOROP, TOKENTYPE_OPNUM, "BITOROP" },
286 { COLONATTR, TOKENTYPE_NONE, "COLONATTR" },
287 { CONTINUE, TOKENTYPE_NONE, "CONTINUE" },
288 { DEFAULT, TOKENTYPE_NONE, "DEFAULT" },
289 { DO, TOKENTYPE_NONE, "DO" },
290 { DOLSHARP, TOKENTYPE_NONE, "DOLSHARP" },
291 { DORDOR, TOKENTYPE_NONE, "DORDOR" },
292 { DOROP, TOKENTYPE_OPNUM, "DOROP" },
293 { DOTDOT, TOKENTYPE_IVAL, "DOTDOT" },
294 { ELSE, TOKENTYPE_NONE, "ELSE" },
295 { ELSIF, TOKENTYPE_IVAL, "ELSIF" },
296 { EQOP, TOKENTYPE_OPNUM, "EQOP" },
297 { FOR, TOKENTYPE_IVAL, "FOR" },
298 { FORMAT, TOKENTYPE_NONE, "FORMAT" },
299 { FUNC, TOKENTYPE_OPNUM, "FUNC" },
300 { FUNC0, TOKENTYPE_OPNUM, "FUNC0" },
301 { FUNC0SUB, TOKENTYPE_OPVAL, "FUNC0SUB" },
302 { FUNC1, TOKENTYPE_OPNUM, "FUNC1" },
303 { FUNCMETH, TOKENTYPE_OPVAL, "FUNCMETH" },
304 { GIVEN, TOKENTYPE_IVAL, "GIVEN" },
305 { HASHBRACK, TOKENTYPE_NONE, "HASHBRACK" },
306 { IF, TOKENTYPE_IVAL, "IF" },
307 { LABEL, TOKENTYPE_PVAL, "LABEL" },
308 { LOCAL, TOKENTYPE_IVAL, "LOCAL" },
309 { LOOPEX, TOKENTYPE_OPNUM, "LOOPEX" },
310 { LSTOP, TOKENTYPE_OPNUM, "LSTOP" },
311 { LSTOPSUB, TOKENTYPE_OPVAL, "LSTOPSUB" },
312 { MATCHOP, TOKENTYPE_OPNUM, "MATCHOP" },
313 { METHOD, TOKENTYPE_OPVAL, "METHOD" },
314 { MULOP, TOKENTYPE_OPNUM, "MULOP" },
315 { MY, TOKENTYPE_IVAL, "MY" },
316 { MYSUB, TOKENTYPE_NONE, "MYSUB" },
317 { NOAMP, TOKENTYPE_NONE, "NOAMP" },
318 { NOTOP, TOKENTYPE_NONE, "NOTOP" },
319 { OROP, TOKENTYPE_IVAL, "OROP" },
320 { OROR, TOKENTYPE_NONE, "OROR" },
321 { PACKAGE, TOKENTYPE_NONE, "PACKAGE" },
322 { PMFUNC, TOKENTYPE_OPVAL, "PMFUNC" },
323 { POSTDEC, TOKENTYPE_NONE, "POSTDEC" },
324 { POSTINC, TOKENTYPE_NONE, "POSTINC" },
325 { POWOP, TOKENTYPE_OPNUM, "POWOP" },
326 { PREDEC, TOKENTYPE_NONE, "PREDEC" },
327 { PREINC, TOKENTYPE_NONE, "PREINC" },
328 { PRIVATEREF, TOKENTYPE_OPVAL, "PRIVATEREF" },
329 { REFGEN, TOKENTYPE_NONE, "REFGEN" },
330 { RELOP, TOKENTYPE_OPNUM, "RELOP" },
331 { SHIFTOP, TOKENTYPE_OPNUM, "SHIFTOP" },
332 { SUB, TOKENTYPE_NONE, "SUB" },
333 { THING, TOKENTYPE_OPVAL, "THING" },
334 { UMINUS, TOKENTYPE_NONE, "UMINUS" },
335 { UNIOP, TOKENTYPE_OPNUM, "UNIOP" },
336 { UNIOPSUB, TOKENTYPE_OPVAL, "UNIOPSUB" },
337 { UNLESS, TOKENTYPE_IVAL, "UNLESS" },
338 { UNTIL, TOKENTYPE_IVAL, "UNTIL" },
339 { USE, TOKENTYPE_IVAL, "USE" },
340 { WHEN, TOKENTYPE_IVAL, "WHEN" },
341 { WHILE, TOKENTYPE_IVAL, "WHILE" },
342 { WORD, TOKENTYPE_OPVAL, "WORD" },
343 { 0, TOKENTYPE_NONE, NULL }
346 /* dump the returned token in rv, plus any optional arg in yylval */
349 S_tokereport(pTHX_ I32 rv)
353 const char *name = NULL;
354 enum token_type type = TOKENTYPE_NONE;
355 const struct debug_tokens *p;
356 SV* const report = newSVpvs("<== ");
358 for (p = debug_tokens; p->token; p++) {
359 if (p->token == (int)rv) {
366 Perl_sv_catpv(aTHX_ report, name);
367 else if ((char)rv > ' ' && (char)rv < '~')
368 Perl_sv_catpvf(aTHX_ report, "'%c'", (char)rv);
370 sv_catpvs(report, "EOF");
372 Perl_sv_catpvf(aTHX_ report, "?? %"IVdf, (IV)rv);
375 case TOKENTYPE_GVVAL: /* doesn't appear to be used */
378 Perl_sv_catpvf(aTHX_ report, "(ival=%"IVdf")", (IV)yylval.ival);
380 case TOKENTYPE_OPNUM:
381 Perl_sv_catpvf(aTHX_ report, "(ival=op_%s)",
382 PL_op_name[yylval.ival]);
385 Perl_sv_catpvf(aTHX_ report, "(pval=\"%s\")", yylval.pval);
387 case TOKENTYPE_OPVAL:
389 Perl_sv_catpvf(aTHX_ report, "(opval=op_%s)",
390 PL_op_name[yylval.opval->op_type]);
391 if (yylval.opval->op_type == OP_CONST) {
392 Perl_sv_catpvf(aTHX_ report, " %s",
393 SvPEEK(cSVOPx_sv(yylval.opval)));
398 sv_catpvs(report, "(opval=null)");
401 PerlIO_printf(Perl_debug_log, "### %s\n\n", SvPV_nolen_const(report));
407 /* print the buffer with suitable escapes */
410 S_printbuf(pTHX_ const char* fmt, const char* s)
412 SV* const tmp = newSVpvs("");
413 PerlIO_printf(Perl_debug_log, fmt, pv_display(tmp, s, strlen(s), 0, 60));
422 * This subroutine detects &&=, ||=, and //= and turns an ANDAND, OROR or DORDOR
423 * into an OP_ANDASSIGN, OP_ORASSIGN, or OP_DORASSIGN
427 S_ao(pTHX_ int toketype)
430 if (*PL_bufptr == '=') {
432 if (toketype == ANDAND)
433 yylval.ival = OP_ANDASSIGN;
434 else if (toketype == OROR)
435 yylval.ival = OP_ORASSIGN;
436 else if (toketype == DORDOR)
437 yylval.ival = OP_DORASSIGN;
445 * When Perl expects an operator and finds something else, no_op
446 * prints the warning. It always prints "<something> found where
447 * operator expected. It prints "Missing semicolon on previous line?"
448 * if the surprise occurs at the start of the line. "do you need to
449 * predeclare ..." is printed out for code like "sub bar; foo bar $x"
450 * where the compiler doesn't know if foo is a method call or a function.
451 * It prints "Missing operator before end of line" if there's nothing
452 * after the missing operator, or "... before <...>" if there is something
453 * after the missing operator.
457 S_no_op(pTHX_ const char *what, char *s)
460 char * const oldbp = PL_bufptr;
461 const bool is_first = (PL_oldbufptr == PL_linestart);
467 yywarn(Perl_form(aTHX_ "%s found where operator expected", what));
468 if (ckWARN_d(WARN_SYNTAX)) {
470 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
471 "\t(Missing semicolon on previous line?)\n");
472 else if (PL_oldoldbufptr && isIDFIRST_lazy_if(PL_oldoldbufptr,UTF)) {
474 for (t = PL_oldoldbufptr; (isALNUM_lazy_if(t,UTF) || *t == ':'); t++)
476 if (t < PL_bufptr && isSPACE(*t))
477 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
478 "\t(Do you need to predeclare %.*s?)\n",
479 (int)(t - PL_oldoldbufptr), PL_oldoldbufptr);
483 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
484 "\t(Missing operator before %.*s?)\n", (int)(s - oldbp), oldbp);
492 * Complain about missing quote/regexp/heredoc terminator.
493 * If it's called with NULL then it cauterizes the line buffer.
494 * If we're in a delimited string and the delimiter is a control
495 * character, it's reformatted into a two-char sequence like ^C.
500 S_missingterm(pTHX_ char *s)
506 char * const nl = strrchr(s,'\n');
512 iscntrl(PL_multi_close)
514 PL_multi_close < 32 || PL_multi_close == 127
518 tmpbuf[1] = (char)toCTRL(PL_multi_close);
523 *tmpbuf = (char)PL_multi_close;
527 q = strchr(s,'"') ? '\'' : '"';
528 Perl_croak(aTHX_ "Can't find string terminator %c%s%c anywhere before EOF",q,s,q);
531 #define FEATURE_IS_ENABLED(name) \
532 ((0 != (PL_hints & HINT_LOCALIZE_HH)) \
533 && S_feature_is_enabled(aTHX_ STR_WITH_LEN(name)))
535 * S_feature_is_enabled
536 * Check whether the named feature is enabled.
539 S_feature_is_enabled(pTHX_ const char *name, STRLEN namelen)
542 HV * const hinthv = GvHV(PL_hintgv);
543 char he_name[32] = "feature_";
544 (void) my_strlcpy(&he_name[8], name, 24);
546 return (hinthv && hv_exists(hinthv, he_name, 8 + namelen));
554 Perl_deprecate(pTHX_ const char *s)
556 if (ckWARN(WARN_DEPRECATED))
557 Perl_warner(aTHX_ packWARN(WARN_DEPRECATED), "Use of %s is deprecated", s);
561 Perl_deprecate_old(pTHX_ const char *s)
563 /* This function should NOT be called for any new deprecated warnings */
564 /* Use Perl_deprecate instead */
566 /* It is here to maintain backward compatibility with the pre-5.8 */
567 /* warnings category hierarchy. The "deprecated" category used to */
568 /* live under the "syntax" category. It is now a top-level category */
569 /* in its own right. */
571 if (ckWARN2(WARN_DEPRECATED, WARN_SYNTAX))
572 Perl_warner(aTHX_ packWARN2(WARN_DEPRECATED, WARN_SYNTAX),
573 "Use of %s is deprecated", s);
577 * experimental text filters for win32 carriage-returns, utf16-to-utf8 and
578 * utf16-to-utf8-reversed.
581 #ifdef PERL_CR_FILTER
585 register const char *s = SvPVX_const(sv);
586 register const char * const e = s + SvCUR(sv);
587 /* outer loop optimized to do nothing if there are no CR-LFs */
589 if (*s++ == '\r' && *s == '\n') {
590 /* hit a CR-LF, need to copy the rest */
591 register char *d = s - 1;
594 if (*s == '\r' && s[1] == '\n')
605 S_cr_textfilter(pTHX_ int idx, SV *sv, int maxlen)
607 const I32 count = FILTER_READ(idx+1, sv, maxlen);
608 if (count > 0 && !maxlen)
618 * Initialize variables. Uses the Perl save_stack to save its state (for
619 * recursive calls to the parser).
623 Perl_lex_start(pTHX_ SV *line)
626 const char *s = NULL;
630 /* create and initialise a parser */
632 Newxz(parser, 1, yy_parser);
633 parser->old_parser = PL_parser;
636 Newx(parser->stack, YYINITDEPTH, yy_stack_frame);
637 parser->ps = parser->stack;
638 parser->stack_size = YYINITDEPTH;
640 parser->stack->state = 0;
641 parser->yyerrstatus = 0;
642 parser->yychar = YYEMPTY; /* Cause a token to be read. */
644 /* initialise lexer state */
646 SAVEI32(PL_lex_state);
648 if (PL_lex_state == LEX_KNOWNEXT) {
649 I32 toke = parser->old_parser->lasttoke;
650 while (--toke >= 0) {
651 SAVEI32(PL_nexttoke[toke].next_type);
652 SAVEVPTR(PL_nexttoke[toke].next_val);
654 SAVEVPTR(PL_nexttoke[toke].next_mad);
657 SAVEI32(PL_curforce);
659 if (PL_lex_state == LEX_KNOWNEXT) {
660 I32 toke = PL_nexttoke;
661 while (--toke >= 0) {
662 SAVEI32(PL_nexttype[toke]);
663 SAVEVPTR(PL_nextval[toke]);
665 SAVEI32(PL_nexttoke);
668 SAVECOPLINE(PL_curcop);
671 SAVEPPTR(PL_oldbufptr);
672 SAVEPPTR(PL_oldoldbufptr);
673 SAVEPPTR(PL_last_lop);
674 SAVEPPTR(PL_last_uni);
675 SAVEPPTR(PL_linestart);
676 SAVESPTR(PL_linestr);
677 SAVEDESTRUCTOR_X(restore_rsfp, PL_rsfp);
681 PL_lex_state = LEX_NORMAL;
683 Newx(parser->lex_brackstack, 120, char);
684 Newx(parser->lex_casestack, 12, char);
685 *parser->lex_casestack = '\0';
691 s = SvPV_const(line, len);
696 PL_linestr = newSVpvs("\n;");
697 } else if (SvREADONLY(line) || s[len-1] != ';') {
698 PL_linestr = newSVsv(line);
700 sv_catpvs(PL_linestr, "\n;");
703 SvREFCNT_inc_simple_void_NN(line);
706 /* PL_linestr needs to survive until end of scope, not just the next
707 FREETMPS. See changes 17505 and 17546 which fixed the symptoms only. */
708 SAVEFREESV(PL_linestr);
709 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = PL_linestart = SvPVX(PL_linestr);
710 PL_bufend = PL_bufptr + SvCUR(PL_linestr);
711 PL_last_lop = PL_last_uni = NULL;
717 * Finalizer for lexing operations. Must be called when the parser is
718 * done with the lexer.
725 PL_doextract = FALSE;
730 * This subroutine has nothing to do with tilting, whether at windmills
731 * or pinball tables. Its name is short for "increment line". It
732 * increments the current line number in CopLINE(PL_curcop) and checks
733 * to see whether the line starts with a comment of the form
734 * # line 500 "foo.pm"
735 * If so, it sets the current line number and file to the values in the comment.
739 S_incline(pTHX_ const char *s)
746 CopLINE_inc(PL_curcop);
749 while (SPACE_OR_TAB(*s))
751 if (strnEQ(s, "line", 4))
755 if (SPACE_OR_TAB(*s))
759 while (SPACE_OR_TAB(*s))
767 while (SPACE_OR_TAB(*s))
769 if (*s == '"' && (t = strchr(s+1, '"'))) {
779 while (SPACE_OR_TAB(*e) || *e == '\r' || *e == '\f')
781 if (*e != '\n' && *e != '\0')
782 return; /* false alarm */
785 const STRLEN len = t - s;
787 const char * const cf = CopFILE(PL_curcop);
788 STRLEN tmplen = cf ? strlen(cf) : 0;
789 if (tmplen > 7 && strnEQ(cf, "(eval ", 6)) {
790 /* must copy *{"::_<(eval N)[oldfilename:L]"}
791 * to *{"::_<newfilename"} */
792 /* However, the long form of evals is only turned on by the
793 debugger - usually they're "(eval %lu)" */
797 STRLEN tmplen2 = len;
798 if (tmplen + 2 <= sizeof smallbuf)
801 Newx(tmpbuf, tmplen + 2, char);
804 memcpy(tmpbuf + 2, cf, tmplen);
806 gvp = (GV**)hv_fetch(PL_defstash, tmpbuf, tmplen, FALSE);
811 if (tmplen2 + 2 <= sizeof smallbuf)
814 Newx(tmpbuf2, tmplen2 + 2, char);
816 if (tmpbuf2 != smallbuf || tmpbuf != smallbuf) {
817 /* Either they malloc'd it, or we malloc'd it,
818 so no prefix is present in ours. */
823 memcpy(tmpbuf2 + 2, s, tmplen2);
826 gv2 = *(GV**)hv_fetch(PL_defstash, tmpbuf2, tmplen2, TRUE);
828 gv_init(gv2, PL_defstash, tmpbuf2, tmplen2, FALSE);
829 /* adjust ${"::_<newfilename"} to store the new file name */
830 GvSV(gv2) = newSVpvn(tmpbuf2 + 2, tmplen2 - 2);
831 GvHV(gv2) = (HV*)SvREFCNT_inc(GvHV(*gvp));
832 GvAV(gv2) = (AV*)SvREFCNT_inc(GvAV(*gvp));
835 if (tmpbuf2 != smallbuf) Safefree(tmpbuf2);
837 if (tmpbuf != smallbuf) Safefree(tmpbuf);
840 CopFILE_free(PL_curcop);
841 CopFILE_setn(PL_curcop, s, len);
843 CopLINE_set(PL_curcop, atoi(n)-1);
847 /* skip space before PL_thistoken */
850 S_skipspace0(pTHX_ register char *s)
857 PL_thiswhite = newSVpvs("");
858 sv_catsv(PL_thiswhite, PL_skipwhite);
859 sv_free(PL_skipwhite);
862 PL_realtokenstart = s - SvPVX(PL_linestr);
866 /* skip space after PL_thistoken */
869 S_skipspace1(pTHX_ register char *s)
871 const char *start = s;
872 I32 startoff = start - SvPVX(PL_linestr);
877 start = SvPVX(PL_linestr) + startoff;
878 if (!PL_thistoken && PL_realtokenstart >= 0) {
879 const char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
880 PL_thistoken = newSVpvn(tstart, start - tstart);
882 PL_realtokenstart = -1;
885 PL_nextwhite = newSVpvs("");
886 sv_catsv(PL_nextwhite, PL_skipwhite);
887 sv_free(PL_skipwhite);
894 S_skipspace2(pTHX_ register char *s, SV **svp)
897 const I32 bufptroff = PL_bufptr - SvPVX(PL_linestr);
898 const I32 startoff = s - SvPVX(PL_linestr);
901 PL_bufptr = SvPVX(PL_linestr) + bufptroff;
902 if (!PL_madskills || !svp)
904 start = SvPVX(PL_linestr) + startoff;
905 if (!PL_thistoken && PL_realtokenstart >= 0) {
906 char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
907 PL_thistoken = newSVpvn(tstart, start - tstart);
908 PL_realtokenstart = -1;
913 sv_setsv(*svp, PL_skipwhite);
914 sv_free(PL_skipwhite);
923 S_update_debugger_info(pTHX_ SV *orig_sv, const char *buf, STRLEN len)
925 AV *av = CopFILEAVx(PL_curcop);
927 SV * const sv = newSV(0);
928 sv_upgrade(sv, SVt_PVMG);
930 sv_setsv(sv, orig_sv);
932 sv_setpvn(sv, buf, len);
935 av_store(av, (I32)CopLINE(PL_curcop), sv);
941 * Called to gobble the appropriate amount and type of whitespace.
942 * Skips comments as well.
946 S_skipspace(pTHX_ register char *s)
951 int startoff = s - SvPVX(PL_linestr);
954 sv_free(PL_skipwhite);
959 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
960 while (s < PL_bufend && SPACE_OR_TAB(*s))
970 SSize_t oldprevlen, oldoldprevlen;
971 SSize_t oldloplen = 0, oldunilen = 0;
972 while (s < PL_bufend && isSPACE(*s)) {
973 if (*s++ == '\n' && PL_in_eval && !PL_rsfp)
978 if (s < PL_bufend && *s == '#') {
979 while (s < PL_bufend && *s != '\n')
983 if (PL_in_eval && !PL_rsfp) {
990 /* only continue to recharge the buffer if we're at the end
991 * of the buffer, we're not reading from a source filter, and
992 * we're in normal lexing mode
994 if (s < PL_bufend || !PL_rsfp || PL_sublex_info.sub_inwhat ||
995 PL_lex_state == LEX_FORMLINE)
1002 /* try to recharge the buffer */
1004 curoff = s - SvPVX(PL_linestr);
1007 if ((s = filter_gets(PL_linestr, PL_rsfp,
1008 (prevlen = SvCUR(PL_linestr)))) == NULL)
1011 if (PL_madskills && curoff != startoff) {
1013 PL_skipwhite = newSVpvs("");
1014 sv_catpvn(PL_skipwhite, SvPVX(PL_linestr) + startoff,
1018 /* mustn't throw out old stuff yet if madpropping */
1019 SvCUR(PL_linestr) = curoff;
1020 s = SvPVX(PL_linestr) + curoff;
1022 if (curoff && s[-1] == '\n')
1026 /* end of file. Add on the -p or -n magic */
1027 /* XXX these shouldn't really be added here, can't set PL_faketokens */
1030 sv_catpvs(PL_linestr,
1031 ";}continue{print or die qq(-p destination: $!\\n);}");
1033 sv_setpvs(PL_linestr,
1034 ";}continue{print or die qq(-p destination: $!\\n);}");
1036 PL_minus_n = PL_minus_p = 0;
1038 else if (PL_minus_n) {
1040 sv_catpvn(PL_linestr, ";}", 2);
1042 sv_setpvn(PL_linestr, ";}", 2);
1048 sv_catpvn(PL_linestr,";", 1);
1050 sv_setpvn(PL_linestr,";", 1);
1053 /* reset variables for next time we lex */
1054 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = s = PL_linestart
1060 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
1061 PL_last_lop = PL_last_uni = NULL;
1063 /* Close the filehandle. Could be from -P preprocessor,
1064 * STDIN, or a regular file. If we were reading code from
1065 * STDIN (because the commandline held no -e or filename)
1066 * then we don't close it, we reset it so the code can
1067 * read from STDIN too.
1070 if (PL_preprocess && !PL_in_eval)
1071 (void)PerlProc_pclose(PL_rsfp);
1072 else if ((PerlIO*)PL_rsfp == PerlIO_stdin())
1073 PerlIO_clearerr(PL_rsfp);
1075 (void)PerlIO_close(PL_rsfp);
1080 /* not at end of file, so we only read another line */
1081 /* make corresponding updates to old pointers, for yyerror() */
1082 oldprevlen = PL_oldbufptr - PL_bufend;
1083 oldoldprevlen = PL_oldoldbufptr - PL_bufend;
1085 oldunilen = PL_last_uni - PL_bufend;
1087 oldloplen = PL_last_lop - PL_bufend;
1088 PL_linestart = PL_bufptr = s + prevlen;
1089 PL_bufend = s + SvCUR(PL_linestr);
1091 PL_oldbufptr = s + oldprevlen;
1092 PL_oldoldbufptr = s + oldoldprevlen;
1094 PL_last_uni = s + oldunilen;
1096 PL_last_lop = s + oldloplen;
1099 /* debugger active and we're not compiling the debugger code,
1100 * so store the line into the debugger's array of lines
1102 if (PERLDB_LINE && PL_curstash != PL_debstash)
1103 update_debugger_info(NULL, PL_bufptr, PL_bufend - PL_bufptr);
1110 PL_skipwhite = newSVpvs("");
1111 curoff = s - SvPVX(PL_linestr);
1112 if (curoff - startoff)
1113 sv_catpvn(PL_skipwhite, SvPVX(PL_linestr) + startoff,
1122 * Check the unary operators to ensure there's no ambiguity in how they're
1123 * used. An ambiguous piece of code would be:
1125 * This doesn't mean rand() + 5. Because rand() is a unary operator,
1126 * the +5 is its argument.
1136 if (PL_oldoldbufptr != PL_last_uni)
1138 while (isSPACE(*PL_last_uni))
1141 while (isALNUM_lazy_if(s,UTF) || *s == '-')
1143 if ((t = strchr(s, '(')) && t < PL_bufptr)
1146 if (ckWARN_d(WARN_AMBIGUOUS)){
1147 Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
1148 "Warning: Use of \"%.*s\" without parentheses is ambiguous",
1149 (int)(s - PL_last_uni), PL_last_uni);
1154 * LOP : macro to build a list operator. Its behaviour has been replaced
1155 * with a subroutine, S_lop() for which LOP is just another name.
1158 #define LOP(f,x) return lop(f,x,s)
1162 * Build a list operator (or something that might be one). The rules:
1163 * - if we have a next token, then it's a list operator [why?]
1164 * - if the next thing is an opening paren, then it's a function
1165 * - else it's a list operator
1169 S_lop(pTHX_ I32 f, int x, char *s)
1176 PL_last_lop = PL_oldbufptr;
1177 PL_last_lop_op = (OPCODE)f;
1180 return REPORT(LSTOP);
1183 return REPORT(LSTOP);
1186 return REPORT(FUNC);
1189 return REPORT(FUNC);
1191 return REPORT(LSTOP);
1197 * Sets up for an eventual force_next(). start_force(0) basically does
1198 * an unshift, while start_force(-1) does a push. yylex removes items
1203 S_start_force(pTHX_ int where)
1207 if (where < 0) /* so people can duplicate start_force(PL_curforce) */
1208 where = PL_lasttoke;
1209 assert(PL_curforce < 0 || PL_curforce == where);
1210 if (PL_curforce != where) {
1211 for (i = PL_lasttoke; i > where; --i) {
1212 PL_nexttoke[i] = PL_nexttoke[i-1];
1216 if (PL_curforce < 0) /* in case of duplicate start_force() */
1217 Zero(&PL_nexttoke[where], 1, NEXTTOKE);
1218 PL_curforce = where;
1221 curmad('^', newSVpvs(""));
1222 CURMAD('_', PL_nextwhite);
1227 S_curmad(pTHX_ char slot, SV *sv)
1233 if (PL_curforce < 0)
1234 where = &PL_thismad;
1236 where = &PL_nexttoke[PL_curforce].next_mad;
1239 sv_setpvn(sv, "", 0);
1242 if (UTF && is_utf8_string((U8*)SvPVX(sv), SvCUR(sv)))
1244 else if (PL_encoding) {
1245 sv_recode_to_utf8(sv, PL_encoding);
1250 /* keep a slot open for the head of the list? */
1251 if (slot != '_' && *where && (*where)->mad_key == '^') {
1252 (*where)->mad_key = slot;
1253 sv_free((*where)->mad_val);
1254 (*where)->mad_val = (void*)sv;
1257 addmad(newMADsv(slot, sv), where, 0);
1260 # define start_force(where) NOOP
1261 # define curmad(slot, sv) NOOP
1266 * When the lexer realizes it knows the next token (for instance,
1267 * it is reordering tokens for the parser) then it can call S_force_next
1268 * to know what token to return the next time the lexer is called. Caller
1269 * will need to set PL_nextval[] (or PL_nexttoke[].next_val with PERL_MAD),
1270 * and possibly PL_expect to ensure the lexer handles the token correctly.
1274 S_force_next(pTHX_ I32 type)
1278 if (PL_curforce < 0)
1279 start_force(PL_lasttoke);
1280 PL_nexttoke[PL_curforce].next_type = type;
1281 if (PL_lex_state != LEX_KNOWNEXT)
1282 PL_lex_defer = PL_lex_state;
1283 PL_lex_state = LEX_KNOWNEXT;
1284 PL_lex_expect = PL_expect;
1287 PL_nexttype[PL_nexttoke] = type;
1289 if (PL_lex_state != LEX_KNOWNEXT) {
1290 PL_lex_defer = PL_lex_state;
1291 PL_lex_expect = PL_expect;
1292 PL_lex_state = LEX_KNOWNEXT;
1298 S_newSV_maybe_utf8(pTHX_ const char *start, STRLEN len)
1301 SV * const sv = newSVpvn(start,len);
1302 if (UTF && !IN_BYTES && is_utf8_string((const U8*)start, len))
1309 * When the lexer knows the next thing is a word (for instance, it has
1310 * just seen -> and it knows that the next char is a word char, then
1311 * it calls S_force_word to stick the next word into the PL_nexttoke/val
1315 * char *start : buffer position (must be within PL_linestr)
1316 * int token : PL_next* will be this type of bare word (e.g., METHOD,WORD)
1317 * int check_keyword : if true, Perl checks to make sure the word isn't
1318 * a keyword (do this if the word is a label, e.g. goto FOO)
1319 * int allow_pack : if true, : characters will also be allowed (require,
1320 * use, etc. do this)
1321 * int allow_initial_tick : used by the "sub" lexer only.
1325 S_force_word(pTHX_ register char *start, int token, int check_keyword, int allow_pack, int allow_initial_tick)
1331 start = SKIPSPACE1(start);
1333 if (isIDFIRST_lazy_if(s,UTF) ||
1334 (allow_pack && *s == ':') ||
1335 (allow_initial_tick && *s == '\'') )
1337 s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, allow_pack, &len);
1338 if (check_keyword && keyword(PL_tokenbuf, len, 0))
1340 start_force(PL_curforce);
1342 curmad('X', newSVpvn(start,s-start));
1343 if (token == METHOD) {
1348 PL_expect = XOPERATOR;
1351 NEXTVAL_NEXTTOKE.opval
1352 = (OP*)newSVOP(OP_CONST,0,
1353 S_newSV_maybe_utf8(aTHX_ PL_tokenbuf, len));
1354 NEXTVAL_NEXTTOKE.opval->op_private |= OPpCONST_BARE;
1362 * Called when the lexer wants $foo *foo &foo etc, but the program
1363 * text only contains the "foo" portion. The first argument is a pointer
1364 * to the "foo", and the second argument is the type symbol to prefix.
1365 * Forces the next token to be a "WORD".
1366 * Creates the symbol if it didn't already exist (via gv_fetchpv()).
1370 S_force_ident(pTHX_ register const char *s, int kind)
1374 const STRLEN len = strlen(s);
1375 OP* const o = (OP*)newSVOP(OP_CONST, 0, newSVpvn(s, len));
1376 start_force(PL_curforce);
1377 NEXTVAL_NEXTTOKE.opval = o;
1380 o->op_private = OPpCONST_ENTERED;
1381 /* XXX see note in pp_entereval() for why we forgo typo
1382 warnings if the symbol must be introduced in an eval.
1384 gv_fetchpvn_flags(s, len,
1385 PL_in_eval ? (GV_ADDMULTI | GV_ADDINEVAL)
1387 kind == '$' ? SVt_PV :
1388 kind == '@' ? SVt_PVAV :
1389 kind == '%' ? SVt_PVHV :
1397 Perl_str_to_version(pTHX_ SV *sv)
1402 const char *start = SvPV_const(sv,len);
1403 const char * const end = start + len;
1404 const bool utf = SvUTF8(sv) ? TRUE : FALSE;
1405 while (start < end) {
1409 n = utf8n_to_uvchr((U8*)start, len, &skip, 0);
1414 retval += ((NV)n)/nshift;
1423 * Forces the next token to be a version number.
1424 * If the next token appears to be an invalid version number, (e.g. "v2b"),
1425 * and if "guessing" is TRUE, then no new token is created (and the caller
1426 * must use an alternative parsing method).
1430 S_force_version(pTHX_ char *s, int guessing)
1436 I32 startoff = s - SvPVX(PL_linestr);
1445 while (isDIGIT(*d) || *d == '_' || *d == '.')
1449 start_force(PL_curforce);
1450 curmad('X', newSVpvn(s,d-s));
1453 if (*d == ';' || isSPACE(*d) || *d == '}' || !*d) {
1455 s = scan_num(s, &yylval);
1456 version = yylval.opval;
1457 ver = cSVOPx(version)->op_sv;
1458 if (SvPOK(ver) && !SvNIOK(ver)) {
1459 SvUPGRADE(ver, SVt_PVNV);
1460 SvNV_set(ver, str_to_version(ver));
1461 SvNOK_on(ver); /* hint that it is a version */
1464 else if (guessing) {
1467 sv_free(PL_nextwhite); /* let next token collect whitespace */
1469 s = SvPVX(PL_linestr) + startoff;
1477 if (PL_madskills && !version) {
1478 sv_free(PL_nextwhite); /* let next token collect whitespace */
1480 s = SvPVX(PL_linestr) + startoff;
1483 /* NOTE: The parser sees the package name and the VERSION swapped */
1484 start_force(PL_curforce);
1485 NEXTVAL_NEXTTOKE.opval = version;
1493 * Tokenize a quoted string passed in as an SV. It finds the next
1494 * chunk, up to end of string or a backslash. It may make a new
1495 * SV containing that chunk (if HINT_NEW_STRING is on). It also
1500 S_tokeq(pTHX_ SV *sv)
1504 register char *send;
1512 s = SvPV_force(sv, len);
1513 if (SvTYPE(sv) >= SVt_PVIV && SvIVX(sv) == -1)
1516 while (s < send && *s != '\\')
1521 if ( PL_hints & HINT_NEW_STRING ) {
1522 pv = sv_2mortal(newSVpvn(SvPVX_const(pv), len));
1528 if (s + 1 < send && (s[1] == '\\'))
1529 s++; /* all that, just for this */
1534 SvCUR_set(sv, d - SvPVX_const(sv));
1536 if ( PL_hints & HINT_NEW_STRING )
1537 return new_constant(NULL, 0, "q", sv, pv, "q");
1542 * Now come three functions related to double-quote context,
1543 * S_sublex_start, S_sublex_push, and S_sublex_done. They're used when
1544 * converting things like "\u\Lgnat" into ucfirst(lc("gnat")). They
1545 * interact with PL_lex_state, and create fake ( ... ) argument lists
1546 * to handle functions and concatenation.
1547 * They assume that whoever calls them will be setting up a fake
1548 * join call, because each subthing puts a ',' after it. This lets
1551 * join($, , 'lower ', lcfirst( 'uPpEr', ) ,)
1553 * (I'm not sure whether the spurious commas at the end of lcfirst's
1554 * arguments and join's arguments are created or not).
1559 * Assumes that yylval.ival is the op we're creating (e.g. OP_LCFIRST).
1561 * Pattern matching will set PL_lex_op to the pattern-matching op to
1562 * make (we return THING if yylval.ival is OP_NULL, PMFUNC otherwise).
1564 * OP_CONST and OP_READLINE are easy--just make the new op and return.
1566 * Everything else becomes a FUNC.
1568 * Sets PL_lex_state to LEX_INTERPPUSH unless (ival was OP_NULL or we
1569 * had an OP_CONST or OP_READLINE). This just sets us up for a
1570 * call to S_sublex_push().
1574 S_sublex_start(pTHX)
1577 register const I32 op_type = yylval.ival;
1579 if (op_type == OP_NULL) {
1580 yylval.opval = PL_lex_op;
1584 if (op_type == OP_CONST || op_type == OP_READLINE) {
1585 SV *sv = tokeq(PL_lex_stuff);
1587 if (SvTYPE(sv) == SVt_PVIV) {
1588 /* Overloaded constants, nothing fancy: Convert to SVt_PV: */
1590 const char * const p = SvPV_const(sv, len);
1591 SV * const nsv = newSVpvn(p, len);
1597 yylval.opval = (OP*)newSVOP(op_type, 0, sv);
1598 PL_lex_stuff = NULL;
1599 /* Allow <FH> // "foo" */
1600 if (op_type == OP_READLINE)
1601 PL_expect = XTERMORDORDOR;
1604 else if (op_type == OP_BACKTICK && PL_lex_op) {
1605 /* readpipe() vas overriden */
1606 cSVOPx(cLISTOPx(cUNOPx(PL_lex_op)->op_first)->op_first->op_sibling)->op_sv = tokeq(PL_lex_stuff);
1607 yylval.opval = PL_lex_op;
1609 PL_lex_stuff = NULL;
1613 PL_sublex_info.super_state = PL_lex_state;
1614 PL_sublex_info.sub_inwhat = op_type;
1615 PL_sublex_info.sub_op = PL_lex_op;
1616 PL_lex_state = LEX_INTERPPUSH;
1620 yylval.opval = PL_lex_op;
1630 * Create a new scope to save the lexing state. The scope will be
1631 * ended in S_sublex_done. Returns a '(', starting the function arguments
1632 * to the uc, lc, etc. found before.
1633 * Sets PL_lex_state to LEX_INTERPCONCAT.
1642 PL_lex_state = PL_sublex_info.super_state;
1643 SAVEI32(PL_lex_dojoin);
1644 SAVEI32(PL_lex_brackets);
1645 SAVEI32(PL_lex_casemods);
1646 SAVEI32(PL_lex_starts);
1647 SAVEI32(PL_lex_state);
1648 SAVEVPTR(PL_lex_inpat);
1649 SAVEI32(PL_lex_inwhat);
1650 SAVECOPLINE(PL_curcop);
1651 SAVEPPTR(PL_bufptr);
1652 SAVEPPTR(PL_bufend);
1653 SAVEPPTR(PL_oldbufptr);
1654 SAVEPPTR(PL_oldoldbufptr);
1655 SAVEPPTR(PL_last_lop);
1656 SAVEPPTR(PL_last_uni);
1657 SAVEPPTR(PL_linestart);
1658 SAVESPTR(PL_linestr);
1659 SAVEGENERICPV(PL_lex_brackstack);
1660 SAVEGENERICPV(PL_lex_casestack);
1662 PL_linestr = PL_lex_stuff;
1663 PL_lex_stuff = NULL;
1665 PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart
1666 = SvPVX(PL_linestr);
1667 PL_bufend += SvCUR(PL_linestr);
1668 PL_last_lop = PL_last_uni = NULL;
1669 SAVEFREESV(PL_linestr);
1671 PL_lex_dojoin = FALSE;
1672 PL_lex_brackets = 0;
1673 Newx(PL_lex_brackstack, 120, char);
1674 Newx(PL_lex_casestack, 12, char);
1675 PL_lex_casemods = 0;
1676 *PL_lex_casestack = '\0';
1678 PL_lex_state = LEX_INTERPCONCAT;
1679 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
1681 PL_lex_inwhat = PL_sublex_info.sub_inwhat;
1682 if (PL_lex_inwhat == OP_MATCH || PL_lex_inwhat == OP_QR || PL_lex_inwhat == OP_SUBST)
1683 PL_lex_inpat = PL_sublex_info.sub_op;
1685 PL_lex_inpat = NULL;
1692 * Restores lexer state after a S_sublex_push.
1699 if (!PL_lex_starts++) {
1700 SV * const sv = newSVpvs("");
1701 if (SvUTF8(PL_linestr))
1703 PL_expect = XOPERATOR;
1704 yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
1708 if (PL_lex_casemods) { /* oops, we've got some unbalanced parens */
1709 PL_lex_state = LEX_INTERPCASEMOD;
1713 /* Is there a right-hand side to take care of? (s//RHS/ or tr//RHS/) */
1714 if (PL_lex_repl && (PL_lex_inwhat == OP_SUBST || PL_lex_inwhat == OP_TRANS)) {
1715 PL_linestr = PL_lex_repl;
1717 PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart = SvPVX(PL_linestr);
1718 PL_bufend += SvCUR(PL_linestr);
1719 PL_last_lop = PL_last_uni = NULL;
1720 SAVEFREESV(PL_linestr);
1721 PL_lex_dojoin = FALSE;
1722 PL_lex_brackets = 0;
1723 PL_lex_casemods = 0;
1724 *PL_lex_casestack = '\0';
1726 if (SvEVALED(PL_lex_repl)) {
1727 PL_lex_state = LEX_INTERPNORMAL;
1729 /* we don't clear PL_lex_repl here, so that we can check later
1730 whether this is an evalled subst; that means we rely on the
1731 logic to ensure sublex_done() is called again only via the
1732 branch (in yylex()) that clears PL_lex_repl, else we'll loop */
1735 PL_lex_state = LEX_INTERPCONCAT;
1745 PL_endwhite = newSVpvs("");
1746 sv_catsv(PL_endwhite, PL_thiswhite);
1750 sv_setpvn(PL_thistoken,"",0);
1752 PL_realtokenstart = -1;
1756 PL_bufend = SvPVX(PL_linestr);
1757 PL_bufend += SvCUR(PL_linestr);
1758 PL_expect = XOPERATOR;
1759 PL_sublex_info.sub_inwhat = 0;
1767 Extracts a pattern, double-quoted string, or transliteration. This
1770 It looks at PL_lex_inwhat and PL_lex_inpat to find out whether it's
1771 processing a pattern (PL_lex_inpat is true), a transliteration
1772 (PL_lex_inwhat == OP_TRANS is true), or a double-quoted string.
1774 Returns a pointer to the character scanned up to. If this is
1775 advanced from the start pointer supplied (i.e. if anything was
1776 successfully parsed), will leave an OP for the substring scanned
1777 in yylval. Caller must intuit reason for not parsing further
1778 by looking at the next characters herself.
1782 double-quoted style: \r and \n
1783 regexp special ones: \D \s
1786 case and quoting: \U \Q \E
1787 stops on @ and $, but not for $ as tail anchor
1789 In transliterations:
1790 characters are VERY literal, except for - not at the start or end
1791 of the string, which indicates a range. If the range is in bytes,
1792 scan_const expands the range to the full set of intermediate
1793 characters. If the range is in utf8, the hyphen is replaced with
1794 a certain range mark which will be handled by pmtrans() in op.c.
1796 In double-quoted strings:
1798 double-quoted style: \r and \n
1800 deprecated backrefs: \1 (in substitution replacements)
1801 case and quoting: \U \Q \E
1804 scan_const does *not* construct ops to handle interpolated strings.
1805 It stops processing as soon as it finds an embedded $ or @ variable
1806 and leaves it to the caller to work out what's going on.
1808 embedded arrays (whether in pattern or not) could be:
1809 @foo, @::foo, @'foo, @{foo}, @$foo, @+, @-.
1811 $ in double-quoted strings must be the symbol of an embedded scalar.
1813 $ in pattern could be $foo or could be tail anchor. Assumption:
1814 it's a tail anchor if $ is the last thing in the string, or if it's
1815 followed by one of "()| \r\n\t"
1817 \1 (backreferences) are turned into $1
1819 The structure of the code is
1820 while (there's a character to process) {
1821 handle transliteration ranges
1822 skip regexp comments /(?#comment)/ and codes /(?{code})/
1823 skip #-initiated comments in //x patterns
1824 check for embedded arrays
1825 check for embedded scalars
1827 leave intact backslashes from leaveit (below)
1828 deprecate \1 in substitution replacements
1829 handle string-changing backslashes \l \U \Q \E, etc.
1830 switch (what was escaped) {
1831 handle \- in a transliteration (becomes a literal -)
1832 handle \132 (octal characters)
1833 handle \x15 and \x{1234} (hex characters)
1834 handle \N{name} (named characters)
1835 handle \cV (control characters)
1836 handle printf-style backslashes (\f, \r, \n, etc)
1838 } (end if backslash)
1839 } (end while character to read)
1844 S_scan_const(pTHX_ char *start)
1847 register char *send = PL_bufend; /* end of the constant */
1848 SV *sv = newSV(send - start); /* sv for the constant */
1849 register char *s = start; /* start of the constant */
1850 register char *d = SvPVX(sv); /* destination for copies */
1851 bool dorange = FALSE; /* are we in a translit range? */
1852 bool didrange = FALSE; /* did we just finish a range? */
1853 I32 has_utf8 = FALSE; /* Output constant is UTF8 */
1854 I32 this_utf8 = UTF; /* The source string is assumed to be UTF8 */
1857 UV literal_endpoint = 0;
1858 bool native_range = TRUE; /* turned to FALSE if the first endpoint is Unicode. */
1861 if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
1862 /* If we are doing a trans and we know we want UTF8 set expectation */
1863 has_utf8 = PL_sublex_info.sub_op->op_private & (OPpTRANS_FROM_UTF|OPpTRANS_TO_UTF);
1864 this_utf8 = PL_sublex_info.sub_op->op_private & (PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
1868 while (s < send || dorange) {
1869 /* get transliterations out of the way (they're most literal) */
1870 if (PL_lex_inwhat == OP_TRANS) {
1871 /* expand a range A-Z to the full set of characters. AIE! */
1873 I32 i; /* current expanded character */
1874 I32 min; /* first character in range */
1875 I32 max; /* last character in range */
1886 char * const c = (char*)utf8_hop((U8*)d, -1);
1890 *c = (char)UTF_TO_NATIVE(0xff);
1891 /* mark the range as done, and continue */
1897 i = d - SvPVX_const(sv); /* remember current offset */
1900 SvLEN(sv) + (has_utf8 ?
1901 (512 - UTF_CONTINUATION_MARK +
1904 /* How many two-byte within 0..255: 128 in UTF-8,
1905 * 96 in UTF-8-mod. */
1907 SvGROW(sv, SvLEN(sv) + 256); /* never more than 256 chars in a range */
1909 d = SvPVX(sv) + i; /* refresh d after realloc */
1913 for (j = 0; j <= 1; j++) {
1914 char * const c = (char*)utf8_hop((U8*)d, -1);
1915 const UV uv = utf8n_to_uvchr((U8*)c, d - c, NULL, 0);
1921 max = (U8)0xff; /* only to \xff */
1922 uvmax = uv; /* \x{100} to uvmax */
1924 d = c; /* eat endpoint chars */
1929 d -= 2; /* eat the first char and the - */
1930 min = (U8)*d; /* first char in range */
1931 max = (U8)d[1]; /* last char in range */
1938 "Invalid range \"%c-%c\" in transliteration operator",
1939 (char)min, (char)max);
1943 if (literal_endpoint == 2 &&
1944 ((isLOWER(min) && isLOWER(max)) ||
1945 (isUPPER(min) && isUPPER(max)))) {
1947 for (i = min; i <= max; i++)
1949 *d++ = NATIVE_TO_NEED(has_utf8,i);
1951 for (i = min; i <= max; i++)
1953 *d++ = NATIVE_TO_NEED(has_utf8,i);
1958 for (i = min; i <= max; i++)
1961 const U8 ch = (U8)NATIVE_TO_UTF(i);
1962 if (UNI_IS_INVARIANT(ch))
1965 *d++ = (U8)UTF8_EIGHT_BIT_HI(ch);
1966 *d++ = (U8)UTF8_EIGHT_BIT_LO(ch);
1975 d = (char*)uvchr_to_utf8((U8*)d, 0x100);
1977 *d++ = (char)UTF_TO_NATIVE(0xff);
1979 d = (char*)uvchr_to_utf8((U8*)d, uvmax);
1983 /* mark the range as done, and continue */
1987 literal_endpoint = 0;
1992 /* range begins (ignore - as first or last char) */
1993 else if (*s == '-' && s+1 < send && s != start) {
1995 Perl_croak(aTHX_ "Ambiguous range in transliteration operator");
2002 *d++ = (char)UTF_TO_NATIVE(0xff); /* use illegal utf8 byte--see pmtrans */
2012 literal_endpoint = 0;
2013 native_range = TRUE;
2018 /* if we get here, we're not doing a transliteration */
2020 /* skip for regexp comments /(?#comment)/ and code /(?{code})/,
2021 except for the last char, which will be done separately. */
2022 else if (*s == '(' && PL_lex_inpat && s[1] == '?') {
2024 while (s+1 < send && *s != ')')
2025 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2027 else if (s[2] == '{' /* This should match regcomp.c */
2028 || ((s[2] == 'p' || s[2] == '?') && s[3] == '{'))
2031 char *regparse = s + (s[2] == '{' ? 3 : 4);
2034 while (count && (c = *regparse)) {
2035 if (c == '\\' && regparse[1])
2043 if (*regparse != ')')
2044 regparse--; /* Leave one char for continuation. */
2045 while (s < regparse)
2046 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2050 /* likewise skip #-initiated comments in //x patterns */
2051 else if (*s == '#' && PL_lex_inpat &&
2052 ((PMOP*)PL_lex_inpat)->op_pmflags & PMf_EXTENDED) {
2053 while (s+1 < send && *s != '\n')
2054 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2057 /* check for embedded arrays
2058 (@foo, @::foo, @'foo, @{foo}, @$foo, @+, @-)
2060 else if (*s == '@' && s[1]) {
2061 if (isALNUM_lazy_if(s+1,UTF))
2063 if (strchr(":'{$", s[1]))
2065 if (!PL_lex_inpat && (s[1] == '+' || s[1] == '-'))
2066 break; /* in regexp, neither @+ nor @- are interpolated */
2069 /* check for embedded scalars. only stop if we're sure it's a
2072 else if (*s == '$') {
2073 if (!PL_lex_inpat) /* not a regexp, so $ must be var */
2075 if (s + 1 < send && !strchr("()| \r\n\t", s[1]))
2076 break; /* in regexp, $ might be tail anchor */
2079 /* End of else if chain - OP_TRANS rejoin rest */
2082 if (*s == '\\' && s+1 < send) {
2085 /* deprecate \1 in strings and substitution replacements */
2086 if (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat &&
2087 isDIGIT(*s) && *s != '0' && !isDIGIT(s[1]))
2089 if (ckWARN(WARN_SYNTAX))
2090 Perl_warner(aTHX_ packWARN(WARN_SYNTAX), "\\%c better written as $%c", *s, *s);
2095 /* string-change backslash escapes */
2096 if (PL_lex_inwhat != OP_TRANS && *s && strchr("lLuUEQ", *s)) {
2100 /* skip any other backslash escapes in a pattern */
2101 else if (PL_lex_inpat) {
2102 *d++ = NATIVE_TO_NEED(has_utf8,'\\');
2103 goto default_action;
2106 /* if we get here, it's either a quoted -, or a digit */
2109 /* quoted - in transliterations */
2111 if (PL_lex_inwhat == OP_TRANS) {
2118 if ((isALPHA(*s) || isDIGIT(*s)) &&
2120 Perl_warner(aTHX_ packWARN(WARN_MISC),
2121 "Unrecognized escape \\%c passed through",
2123 /* default action is to copy the quoted character */
2124 goto default_action;
2127 /* \132 indicates an octal constant */
2128 case '0': case '1': case '2': case '3':
2129 case '4': case '5': case '6': case '7':
2133 uv = grok_oct(s, &len, &flags, NULL);
2136 goto NUM_ESCAPE_INSERT;
2138 /* \x24 indicates a hex constant */
2142 char* const e = strchr(s, '}');
2143 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES |
2144 PERL_SCAN_DISALLOW_PREFIX;
2149 yyerror("Missing right brace on \\x{}");
2153 uv = grok_hex(s, &len, &flags, NULL);
2159 I32 flags = PERL_SCAN_DISALLOW_PREFIX;
2160 uv = grok_hex(s, &len, &flags, NULL);
2166 /* Insert oct or hex escaped character.
2167 * There will always enough room in sv since such
2168 * escapes will be longer than any UTF-8 sequence
2169 * they can end up as. */
2171 /* We need to map to chars to ASCII before doing the tests
2174 if (!UNI_IS_INVARIANT(NATIVE_TO_UNI(uv))) {
2175 if (!has_utf8 && uv > 255) {
2176 /* Might need to recode whatever we have
2177 * accumulated so far if it contains any
2180 * (Can't we keep track of that and avoid
2181 * this rescan? --jhi)
2185 for (c = (U8 *) SvPVX(sv); c < (U8 *)d; c++) {
2186 if (!NATIVE_IS_INVARIANT(*c)) {
2191 const STRLEN offset = d - SvPVX_const(sv);
2193 d = SvGROW(sv, SvLEN(sv) + hicount + 1) + offset;
2197 while (src >= (const U8 *)SvPVX_const(sv)) {
2198 if (!NATIVE_IS_INVARIANT(*src)) {
2199 const U8 ch = NATIVE_TO_ASCII(*src);
2200 *dst-- = (U8)UTF8_EIGHT_BIT_LO(ch);
2201 *dst-- = (U8)UTF8_EIGHT_BIT_HI(ch);
2211 if (has_utf8 || uv > 255) {
2212 d = (char*)uvchr_to_utf8((U8*)d, uv);
2214 if (PL_lex_inwhat == OP_TRANS &&
2215 PL_sublex_info.sub_op) {
2216 PL_sublex_info.sub_op->op_private |=
2217 (PL_lex_repl ? OPpTRANS_FROM_UTF
2221 if (uv > 255 && !dorange)
2222 native_range = FALSE;
2234 /* \N{LATIN SMALL LETTER A} is a named character */
2238 char* e = strchr(s, '}');
2245 yyerror("Missing right brace on \\N{}");
2249 if (e > s + 2 && s[1] == 'U' && s[2] == '+') {
2251 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES |
2252 PERL_SCAN_DISALLOW_PREFIX;
2255 uv = grok_hex(s, &len, &flags, NULL);
2256 if ( e > s && len != (STRLEN)(e - s) ) {
2260 goto NUM_ESCAPE_INSERT;
2262 res = newSVpvn(s + 1, e - s - 1);
2263 type = newSVpvn(s - 2,e - s + 3);
2264 res = new_constant( NULL, 0, "charnames",
2265 res, NULL, SvPVX(type) );
2268 sv_utf8_upgrade(res);
2269 str = SvPV_const(res,len);
2270 #ifdef EBCDIC_NEVER_MIND
2271 /* charnames uses pack U and that has been
2272 * recently changed to do the below uni->native
2273 * mapping, so this would be redundant (and wrong,
2274 * the code point would be doubly converted).
2275 * But leave this in just in case the pack U change
2276 * gets revoked, but the semantics is still
2277 * desireable for charnames. --jhi */
2279 UV uv = utf8_to_uvchr((const U8*)str, 0);
2282 U8 tmpbuf[UTF8_MAXBYTES+1], *d;
2284 d = uvchr_to_utf8(tmpbuf, UNI_TO_NATIVE(uv));
2285 sv_setpvn(res, (char *)tmpbuf, d - tmpbuf);
2286 str = SvPV_const(res, len);
2290 if (!has_utf8 && SvUTF8(res)) {
2291 const char * const ostart = SvPVX_const(sv);
2292 SvCUR_set(sv, d - ostart);
2295 sv_utf8_upgrade(sv);
2296 /* this just broke our allocation above... */
2297 SvGROW(sv, (STRLEN)(send - start));
2298 d = SvPVX(sv) + SvCUR(sv);
2301 if (len > (STRLEN)(e - s + 4)) { /* I _guess_ 4 is \N{} --jhi */
2302 const char * const odest = SvPVX_const(sv);
2304 SvGROW(sv, (SvLEN(sv) + len - (e - s + 4)));
2305 d = SvPVX(sv) + (d - odest);
2309 native_range = FALSE; /* \N{} is guessed to be Unicode */
2311 Copy(str, d, len, char);
2318 yyerror("Missing braces on \\N{}");
2321 /* \c is a control character */
2330 *d++ = NATIVE_TO_NEED(has_utf8,toCTRL(c));
2333 yyerror("Missing control char name in \\c");
2337 /* printf-style backslashes, formfeeds, newlines, etc */
2339 *d++ = NATIVE_TO_NEED(has_utf8,'\b');
2342 *d++ = NATIVE_TO_NEED(has_utf8,'\n');
2345 *d++ = NATIVE_TO_NEED(has_utf8,'\r');
2348 *d++ = NATIVE_TO_NEED(has_utf8,'\f');
2351 *d++ = NATIVE_TO_NEED(has_utf8,'\t');
2354 *d++ = ASCII_TO_NEED(has_utf8,'\033');
2357 *d++ = ASCII_TO_NEED(has_utf8,'\007');
2363 } /* end if (backslash) */
2370 /* If we started with encoded form, or already know we want it
2371 and then encode the next character */
2372 if ((has_utf8 || this_utf8) && !NATIVE_IS_INVARIANT((U8)(*s))) {
2374 const UV nextuv = (this_utf8) ? utf8n_to_uvchr((U8*)s, send - s, &len, 0) : (UV) ((U8) *s);
2375 const STRLEN need = UNISKIP(NATIVE_TO_UNI(nextuv));
2378 /* encoded value larger than old, need extra space (NOTE: SvCUR() not set here) */
2379 const STRLEN off = d - SvPVX_const(sv);
2380 d = SvGROW(sv, SvLEN(sv) + (need-len)) + off;
2382 d = (char*)uvchr_to_utf8((U8*)d, nextuv);
2385 if (uv > 255 && !dorange)
2386 native_range = FALSE;
2390 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2392 } /* while loop to process each character */
2394 /* terminate the string and set up the sv */
2396 SvCUR_set(sv, d - SvPVX_const(sv));
2397 if (SvCUR(sv) >= SvLEN(sv))
2398 Perl_croak(aTHX_ "panic: constant overflowed allocated space");
2401 if (PL_encoding && !has_utf8) {
2402 sv_recode_to_utf8(sv, PL_encoding);
2408 if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
2409 PL_sublex_info.sub_op->op_private |=
2410 (PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
2414 /* shrink the sv if we allocated more than we used */
2415 if (SvCUR(sv) + 5 < SvLEN(sv)) {
2416 SvPV_shrink_to_cur(sv);
2419 /* return the substring (via yylval) only if we parsed anything */
2420 if (s > PL_bufptr) {
2421 if ( PL_hints & ( PL_lex_inpat ? HINT_NEW_RE : HINT_NEW_STRING ) )
2422 sv = new_constant(start, s - start,
2423 (const char *)(PL_lex_inpat ? "qr" : "q"),
2426 (( PL_lex_inwhat == OP_TRANS
2428 : ( (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat)
2431 yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
2438 * Returns TRUE if there's more to the expression (e.g., a subscript),
2441 * It deals with "$foo[3]" and /$foo[3]/ and /$foo[0123456789$]+/
2443 * ->[ and ->{ return TRUE
2444 * { and [ outside a pattern are always subscripts, so return TRUE
2445 * if we're outside a pattern and it's not { or [, then return FALSE
2446 * if we're in a pattern and the first char is a {
2447 * {4,5} (any digits around the comma) returns FALSE
2448 * if we're in a pattern and the first char is a [
2450 * [SOMETHING] has a funky algorithm to decide whether it's a
2451 * character class or not. It has to deal with things like
2452 * /$foo[-3]/ and /$foo[$bar]/ as well as /$foo[$\d]+/
2453 * anything else returns TRUE
2456 /* This is the one truly awful dwimmer necessary to conflate C and sed. */
2459 S_intuit_more(pTHX_ register char *s)
2462 if (PL_lex_brackets)
2464 if (*s == '-' && s[1] == '>' && (s[2] == '[' || s[2] == '{'))
2466 if (*s != '{' && *s != '[')
2471 /* In a pattern, so maybe we have {n,m}. */
2488 /* On the other hand, maybe we have a character class */
2491 if (*s == ']' || *s == '^')
2494 /* this is terrifying, and it works */
2495 int weight = 2; /* let's weigh the evidence */
2497 unsigned char un_char = 255, last_un_char;
2498 const char * const send = strchr(s,']');
2499 char tmpbuf[sizeof PL_tokenbuf * 4];
2501 if (!send) /* has to be an expression */
2504 Zero(seen,256,char);
2507 else if (isDIGIT(*s)) {
2509 if (isDIGIT(s[1]) && s[2] == ']')
2515 for (; s < send; s++) {
2516 last_un_char = un_char;
2517 un_char = (unsigned char)*s;
2522 weight -= seen[un_char] * 10;
2523 if (isALNUM_lazy_if(s+1,UTF)) {
2525 scan_ident(s, send, tmpbuf, sizeof tmpbuf, FALSE);
2526 len = (int)strlen(tmpbuf);
2527 if (len > 1 && gv_fetchpvn_flags(tmpbuf, len, 0, SVt_PV))
2532 else if (*s == '$' && s[1] &&
2533 strchr("[#!%*<>()-=",s[1])) {
2534 if (/*{*/ strchr("])} =",s[2]))
2543 if (strchr("wds]",s[1]))
2545 else if (seen[(U8)'\''] || seen[(U8)'"'])
2547 else if (strchr("rnftbxcav",s[1]))
2549 else if (isDIGIT(s[1])) {
2551 while (s[1] && isDIGIT(s[1]))
2561 if (strchr("aA01! ",last_un_char))
2563 if (strchr("zZ79~",s[1]))
2565 if (last_un_char == 255 && (isDIGIT(s[1]) || s[1] == '$'))
2566 weight -= 5; /* cope with negative subscript */
2569 if (!isALNUM(last_un_char)
2570 && !(last_un_char == '$' || last_un_char == '@'
2571 || last_un_char == '&')
2572 && isALPHA(*s) && s[1] && isALPHA(s[1])) {
2577 if (keyword(tmpbuf, d - tmpbuf, 0))
2580 if (un_char == last_un_char + 1)
2582 weight -= seen[un_char];
2587 if (weight >= 0) /* probably a character class */
2597 * Does all the checking to disambiguate
2599 * between foo(bar) and bar->foo. Returns 0 if not a method, otherwise
2600 * FUNCMETH (bar->foo(args)) or METHOD (bar->foo args).
2602 * First argument is the stuff after the first token, e.g. "bar".
2604 * Not a method if bar is a filehandle.
2605 * Not a method if foo is a subroutine prototyped to take a filehandle.
2606 * Not a method if it's really "Foo $bar"
2607 * Method if it's "foo $bar"
2608 * Not a method if it's really "print foo $bar"
2609 * Method if it's really "foo package::" (interpreted as package->foo)
2610 * Not a method if bar is known to be a subroutine ("sub bar; foo bar")
2611 * Not a method if bar is a filehandle or package, but is quoted with
2616 S_intuit_method(pTHX_ char *start, GV *gv, CV *cv)
2619 char *s = start + (*start == '$');
2620 char tmpbuf[sizeof PL_tokenbuf];
2628 if (SvTYPE(gv) == SVt_PVGV && GvIO(gv))
2632 const char *proto = SvPVX_const(cv);
2643 s = scan_word(s, tmpbuf, sizeof tmpbuf, TRUE, &len);
2644 /* start is the beginning of the possible filehandle/object,
2645 * and s is the end of it
2646 * tmpbuf is a copy of it
2649 if (*start == '$') {
2650 if (gv || PL_last_lop_op == OP_PRINT || isUPPER(*PL_tokenbuf))
2653 len = start - SvPVX(PL_linestr);
2657 start = SvPVX(PL_linestr) + len;
2661 return *s == '(' ? FUNCMETH : METHOD;
2663 if (!keyword(tmpbuf, len, 0)) {
2664 if (len > 2 && tmpbuf[len - 2] == ':' && tmpbuf[len - 1] == ':') {
2668 soff = s - SvPVX(PL_linestr);
2672 indirgv = gv_fetchpvn_flags(tmpbuf, len, 0, SVt_PVCV);
2673 if (indirgv && GvCVu(indirgv))
2675 /* filehandle or package name makes it a method */
2676 if (!gv || GvIO(indirgv) || gv_stashpvn(tmpbuf, len, 0)) {
2678 soff = s - SvPVX(PL_linestr);
2681 if ((PL_bufend - s) >= 2 && *s == '=' && *(s+1) == '>')
2682 return 0; /* no assumptions -- "=>" quotes bearword */
2684 start_force(PL_curforce);
2685 NEXTVAL_NEXTTOKE.opval = (OP*)newSVOP(OP_CONST, 0,
2686 newSVpvn(tmpbuf,len));
2687 NEXTVAL_NEXTTOKE.opval->op_private = OPpCONST_BARE;
2689 curmad('X', newSVpvn(start,SvPVX(PL_linestr) + soff - start));
2694 PL_bufptr = SvPVX(PL_linestr) + soff; /* restart before space */
2696 return *s == '(' ? FUNCMETH : METHOD;
2704 * Return a string of Perl code to load the debugger. If PERL5DB
2705 * is set, it will return the contents of that, otherwise a
2706 * compile-time require of perl5db.pl.
2714 const char * const pdb = PerlEnv_getenv("PERL5DB");
2718 SETERRNO(0,SS_NORMAL);
2719 return "BEGIN { require 'perl5db.pl' }";
2725 /* Encoded script support. filter_add() effectively inserts a
2726 * 'pre-processing' function into the current source input stream.
2727 * Note that the filter function only applies to the current source file
2728 * (e.g., it will not affect files 'require'd or 'use'd by this one).
2730 * The datasv parameter (which may be NULL) can be used to pass
2731 * private data to this instance of the filter. The filter function
2732 * can recover the SV using the FILTER_DATA macro and use it to
2733 * store private buffers and state information.
2735 * The supplied datasv parameter is upgraded to a PVIO type
2736 * and the IoDIRP/IoANY field is used to store the function pointer,
2737 * and IOf_FAKE_DIRP is enabled on datasv to mark this as such.
2738 * Note that IoTOP_NAME, IoFMT_NAME, IoBOTTOM_NAME, if set for
2739 * private use must be set using malloc'd pointers.
2743 Perl_filter_add(pTHX_ filter_t funcp, SV *datasv)
2749 if (!PL_rsfp_filters)
2750 PL_rsfp_filters = newAV();
2753 SvUPGRADE(datasv, SVt_PVIO);
2754 IoANY(datasv) = FPTR2DPTR(void *, funcp); /* stash funcp into spare field */
2755 IoFLAGS(datasv) |= IOf_FAKE_DIRP;
2756 DEBUG_P(PerlIO_printf(Perl_debug_log, "filter_add func %p (%s)\n",
2757 FPTR2DPTR(void *, IoANY(datasv)),
2758 SvPV_nolen(datasv)));
2759 av_unshift(PL_rsfp_filters, 1);
2760 av_store(PL_rsfp_filters, 0, datasv) ;
2765 /* Delete most recently added instance of this filter function. */
2767 Perl_filter_del(pTHX_ filter_t funcp)
2773 DEBUG_P(PerlIO_printf(Perl_debug_log, "filter_del func %p",
2774 FPTR2DPTR(void*, funcp)));
2776 if (!PL_rsfp_filters || AvFILLp(PL_rsfp_filters)<0)
2778 /* if filter is on top of stack (usual case) just pop it off */
2779 datasv = FILTER_DATA(AvFILLp(PL_rsfp_filters));
2780 if (IoANY(datasv) == FPTR2DPTR(void *, funcp)) {
2781 IoFLAGS(datasv) &= ~IOf_FAKE_DIRP;
2782 IoANY(datasv) = (void *)NULL;
2783 sv_free(av_pop(PL_rsfp_filters));
2787 /* we need to search for the correct entry and clear it */
2788 Perl_die(aTHX_ "filter_del can only delete in reverse order (currently)");
2792 /* Invoke the idxth filter function for the current rsfp. */
2793 /* maxlen 0 = read one text line */
2795 Perl_filter_read(pTHX_ int idx, SV *buf_sv, int maxlen)
2800 /* This API is bad. It should have been using unsigned int for maxlen.
2801 Not sure if we want to change the API, but if not we should sanity
2802 check the value here. */
2803 const unsigned int correct_length
2812 if (!PL_rsfp_filters)
2814 if (idx > AvFILLp(PL_rsfp_filters)) { /* Any more filters? */
2815 /* Provide a default input filter to make life easy. */
2816 /* Note that we append to the line. This is handy. */
2817 DEBUG_P(PerlIO_printf(Perl_debug_log,
2818 "filter_read %d: from rsfp\n", idx));
2819 if (correct_length) {
2822 const int old_len = SvCUR(buf_sv);
2824 /* ensure buf_sv is large enough */
2825 SvGROW(buf_sv, (STRLEN)(old_len + correct_length)) ;
2826 if ((len = PerlIO_read(PL_rsfp, SvPVX(buf_sv) + old_len,
2827 correct_length)) <= 0) {
2828 if (PerlIO_error(PL_rsfp))
2829 return -1; /* error */
2831 return 0 ; /* end of file */
2833 SvCUR_set(buf_sv, old_len + len) ;
2836 if (sv_gets(buf_sv, PL_rsfp, SvCUR(buf_sv)) == NULL) {
2837 if (PerlIO_error(PL_rsfp))
2838 return -1; /* error */
2840 return 0 ; /* end of file */
2843 return SvCUR(buf_sv);
2845 /* Skip this filter slot if filter has been deleted */
2846 if ( (datasv = FILTER_DATA(idx)) == &PL_sv_undef) {
2847 DEBUG_P(PerlIO_printf(Perl_debug_log,
2848 "filter_read %d: skipped (filter deleted)\n",
2850 return FILTER_READ(idx+1, buf_sv, correct_length); /* recurse */
2852 /* Get function pointer hidden within datasv */
2853 funcp = DPTR2FPTR(filter_t, IoANY(datasv));
2854 DEBUG_P(PerlIO_printf(Perl_debug_log,
2855 "filter_read %d: via function %p (%s)\n",
2856 idx, (void*)datasv, SvPV_nolen_const(datasv)));
2857 /* Call function. The function is expected to */
2858 /* call "FILTER_READ(idx+1, buf_sv)" first. */
2859 /* Return: <0:error, =0:eof, >0:not eof */
2860 return (*funcp)(aTHX_ idx, buf_sv, correct_length);
2864 S_filter_gets(pTHX_ register SV *sv, register PerlIO *fp, STRLEN append)
2867 #ifdef PERL_CR_FILTER
2868 if (!PL_rsfp_filters) {
2869 filter_add(S_cr_textfilter,NULL);
2872 if (PL_rsfp_filters) {
2874 SvCUR_set(sv, 0); /* start with empty line */
2875 if (FILTER_READ(0, sv, 0) > 0)
2876 return ( SvPVX(sv) ) ;
2881 return (sv_gets(sv, fp, append));
2885 S_find_in_my_stash(pTHX_ const char *pkgname, I32 len)
2890 if (len == 11 && *pkgname == '_' && strEQ(pkgname, "__PACKAGE__"))
2894 (pkgname[len - 2] == ':' && pkgname[len - 1] == ':') &&
2895 (gv = gv_fetchpvn_flags(pkgname, len, 0, SVt_PVHV)))
2897 return GvHV(gv); /* Foo:: */
2900 /* use constant CLASS => 'MyClass' */
2901 gv = gv_fetchpvn_flags(pkgname, len, 0, SVt_PVCV);
2902 if (gv && GvCV(gv)) {
2903 SV * const sv = cv_const_sv(GvCV(gv));
2905 pkgname = SvPV_nolen_const(sv);
2908 return gv_stashpv(pkgname, 0);
2912 * S_readpipe_override
2913 * Check whether readpipe() is overriden, and generates the appropriate
2914 * optree, provided sublex_start() is called afterwards.
2917 S_readpipe_override(pTHX)
2920 GV *gv_readpipe = gv_fetchpvs("readpipe", GV_NOTQUAL, SVt_PVCV);
2921 yylval.ival = OP_BACKTICK;
2923 && GvCVu(gv_readpipe) && GvIMPORTED_CV(gv_readpipe))
2925 ((gvp = (GV**)hv_fetchs(PL_globalstash, "readpipe", FALSE))
2926 && (gv_readpipe = *gvp) != (GV*)&PL_sv_undef
2927 && GvCVu(gv_readpipe) && GvIMPORTED_CV(gv_readpipe)))
2929 PL_lex_op = (OP*)newUNOP(OP_ENTERSUB, OPf_STACKED,
2930 append_elem(OP_LIST,
2931 newSVOP(OP_CONST, 0, &PL_sv_undef), /* value will be read later */
2932 newCVREF(0, newGVOP(OP_GV, 0, gv_readpipe))));
2942 * The intent of this yylex wrapper is to minimize the changes to the
2943 * tokener when we aren't interested in collecting madprops. It remains
2944 * to be seen how successful this strategy will be...
2951 char *s = PL_bufptr;
2953 /* make sure PL_thiswhite is initialized */
2957 /* just do what yylex would do on pending identifier; leave PL_thiswhite alone */
2958 if (PL_pending_ident)
2959 return S_pending_ident(aTHX);
2961 /* previous token ate up our whitespace? */
2962 if (!PL_lasttoke && PL_nextwhite) {
2963 PL_thiswhite = PL_nextwhite;
2967 /* isolate the token, and figure out where it is without whitespace */
2968 PL_realtokenstart = -1;
2972 assert(PL_curforce < 0);
2974 if (!PL_thismad || PL_thismad->mad_key == '^') { /* not forced already? */
2975 if (!PL_thistoken) {
2976 if (PL_realtokenstart < 0 || !CopLINE(PL_curcop))
2977 PL_thistoken = newSVpvs("");
2979 char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
2980 PL_thistoken = newSVpvn(tstart, s - tstart);
2983 if (PL_thismad) /* install head */
2984 CURMAD('X', PL_thistoken);
2987 /* last whitespace of a sublex? */
2988 if (optype == ')' && PL_endwhite) {
2989 CURMAD('X', PL_endwhite);
2994 /* if no whitespace and we're at EOF, bail. Otherwise fake EOF below. */
2995 if (!PL_thiswhite && !PL_endwhite && !optype) {
2996 sv_free(PL_thistoken);
3001 /* put off final whitespace till peg */
3002 if (optype == ';' && !PL_rsfp) {
3003 PL_nextwhite = PL_thiswhite;
3006 else if (PL_thisopen) {
3007 CURMAD('q', PL_thisopen);
3009 sv_free(PL_thistoken);
3013 /* Store actual token text as madprop X */
3014 CURMAD('X', PL_thistoken);
3018 /* add preceding whitespace as madprop _ */
3019 CURMAD('_', PL_thiswhite);
3023 /* add quoted material as madprop = */
3024 CURMAD('=', PL_thisstuff);
3028 /* add terminating quote as madprop Q */
3029 CURMAD('Q', PL_thisclose);
3033 /* special processing based on optype */
3037 /* opval doesn't need a TOKEN since it can already store mp */
3048 append_madprops(PL_thismad, yylval.opval, 0);
3056 addmad(newMADsv('p', PL_endwhite), &PL_thismad, 0);
3065 /* remember any fake bracket that lexer is about to discard */
3066 if (PL_lex_brackets == 1 &&
3067 ((expectation)PL_lex_brackstack[0] & XFAKEBRACK))
3070 while (s < PL_bufend && (*s == ' ' || *s == '\t'))
3073 PL_thiswhite = newSVpvn(PL_bufptr, ++s - PL_bufptr);
3074 addmad(newMADsv('#', PL_thiswhite), &PL_thismad, 0);
3077 break; /* don't bother looking for trailing comment */
3086 /* attach a trailing comment to its statement instead of next token */
3090 if (PL_bufptr > PL_oldbufptr && PL_bufptr[-1] == optype) {
3092 while (s < PL_bufend && (*s == ' ' || *s == '\t'))
3094 if (*s == '\n' || *s == '#') {
3095 while (s < PL_bufend && *s != '\n')
3099 PL_thiswhite = newSVpvn(PL_bufptr, s - PL_bufptr);
3100 addmad(newMADsv('#', PL_thiswhite), &PL_thismad, 0);
3117 /* Create new token struct. Note: opvals return early above. */
3118 yylval.tkval = newTOKEN(optype, yylval, PL_thismad);
3125 S_tokenize_use(pTHX_ int is_use, char *s) {
3127 if (PL_expect != XSTATE)
3128 yyerror(Perl_form(aTHX_ "\"%s\" not allowed in expression",
3129 is_use ? "use" : "no"));
3131 if (isDIGIT(*s) || (*s == 'v' && isDIGIT(s[1]))) {
3132 s = force_version(s, TRUE);
3133 if (*s == ';' || (s = SKIPSPACE1(s), *s == ';')) {
3134 start_force(PL_curforce);
3135 NEXTVAL_NEXTTOKE.opval = NULL;
3138 else if (*s == 'v') {
3139 s = force_word(s,WORD,FALSE,TRUE,FALSE);
3140 s = force_version(s, FALSE);
3144 s = force_word(s,WORD,FALSE,TRUE,FALSE);
3145 s = force_version(s, FALSE);
3147 yylval.ival = is_use;
3151 static const char* const exp_name[] =
3152 { "OPERATOR", "TERM", "REF", "STATE", "BLOCK", "ATTRBLOCK",
3153 "ATTRTERM", "TERMBLOCK", "TERMORDORDOR"
3160 Works out what to call the token just pulled out of the input
3161 stream. The yacc parser takes care of taking the ops we return and
3162 stitching them into a tree.
3168 if read an identifier
3169 if we're in a my declaration
3170 croak if they tried to say my($foo::bar)
3171 build the ops for a my() declaration
3172 if it's an access to a my() variable
3173 are we in a sort block?
3174 croak if my($a); $a <=> $b
3175 build ops for access to a my() variable
3176 if in a dq string, and they've said @foo and we can't find @foo
3178 build ops for a bareword
3179 if we already built the token before, use it.
3184 #pragma segment Perl_yylex
3190 register char *s = PL_bufptr;
3195 /* orig_keyword, gvp, and gv are initialized here because
3196 * jump to the label just_a_word_zero can bypass their
3197 * initialization later. */
3198 I32 orig_keyword = 0;
3203 SV* tmp = newSVpvs("");
3204 PerlIO_printf(Perl_debug_log, "### %"IVdf":LEX_%s/X%s %s\n",
3205 (IV)CopLINE(PL_curcop),
3206 lex_state_names[PL_lex_state],
3207 exp_name[PL_expect],
3208 pv_display(tmp, s, strlen(s), 0, 60));
3211 /* check if there's an identifier for us to look at */
3212 if (PL_pending_ident)
3213 return REPORT(S_pending_ident(aTHX));
3215 /* no identifier pending identification */
3217 switch (PL_lex_state) {
3219 case LEX_NORMAL: /* Some compilers will produce faster */
3220 case LEX_INTERPNORMAL: /* code if we comment these out. */
3224 /* when we've already built the next token, just pull it out of the queue */
3228 yylval = PL_nexttoke[PL_lasttoke].next_val;
3230 PL_thismad = PL_nexttoke[PL_lasttoke].next_mad;
3231 PL_nexttoke[PL_lasttoke].next_mad = 0;
3232 if (PL_thismad && PL_thismad->mad_key == '_') {
3233 PL_thiswhite = (SV*)PL_thismad->mad_val;
3234 PL_thismad->mad_val = 0;
3235 mad_free(PL_thismad);
3240 PL_lex_state = PL_lex_defer;
3241 PL_expect = PL_lex_expect;
3242 PL_lex_defer = LEX_NORMAL;
3243 if (!PL_nexttoke[PL_lasttoke].next_type)
3248 yylval = PL_nextval[PL_nexttoke];
3250 PL_lex_state = PL_lex_defer;
3251 PL_expect = PL_lex_expect;
3252 PL_lex_defer = LEX_NORMAL;
3256 /* FIXME - can these be merged? */
3257 return(PL_nexttoke[PL_lasttoke].next_type);
3259 return REPORT(PL_nexttype[PL_nexttoke]);
3262 /* interpolated case modifiers like \L \U, including \Q and \E.
3263 when we get here, PL_bufptr is at the \
3265 case LEX_INTERPCASEMOD:
3267 if (PL_bufptr != PL_bufend && *PL_bufptr != '\\')
3268 Perl_croak(aTHX_ "panic: INTERPCASEMOD");
3270 /* handle \E or end of string */
3271 if (PL_bufptr == PL_bufend || PL_bufptr[1] == 'E') {
3273 if (PL_lex_casemods) {
3274 const char oldmod = PL_lex_casestack[--PL_lex_casemods];
3275 PL_lex_casestack[PL_lex_casemods] = '\0';
3277 if (PL_bufptr != PL_bufend
3278 && (oldmod == 'L' || oldmod == 'U' || oldmod == 'Q')) {
3280 PL_lex_state = LEX_INTERPCONCAT;
3283 PL_thistoken = newSVpvs("\\E");
3289 while (PL_bufptr != PL_bufend &&
3290 PL_bufptr[0] == '\\' && PL_bufptr[1] == 'E') {
3292 PL_thiswhite = newSVpvs("");
3293 sv_catpvn(PL_thiswhite, PL_bufptr, 2);
3297 if (PL_bufptr != PL_bufend)
3300 PL_lex_state = LEX_INTERPCONCAT;
3304 DEBUG_T({ PerlIO_printf(Perl_debug_log,
3305 "### Saw case modifier\n"); });
3307 if (s[1] == '\\' && s[2] == 'E') {
3310 PL_thiswhite = newSVpvs("");
3311 sv_catpvn(PL_thiswhite, PL_bufptr, 4);
3314 PL_lex_state = LEX_INTERPCONCAT;
3319 if (!PL_madskills) /* when just compiling don't need correct */
3320 if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
3321 tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
3322 if ((*s == 'L' || *s == 'U') &&
3323 (strchr(PL_lex_casestack, 'L') || strchr(PL_lex_casestack, 'U'))) {
3324 PL_lex_casestack[--PL_lex_casemods] = '\0';
3327 if (PL_lex_casemods > 10)
3328 Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
3329 PL_lex_casestack[PL_lex_casemods++] = *s;
3330 PL_lex_casestack[PL_lex_casemods] = '\0';
3331 PL_lex_state = LEX_INTERPCONCAT;
3332 start_force(PL_curforce);
3333 NEXTVAL_NEXTTOKE.ival = 0;
3335 start_force(PL_curforce);
3337 NEXTVAL_NEXTTOKE.ival = OP_LCFIRST;
3339 NEXTVAL_NEXTTOKE.ival = OP_UCFIRST;
3341 NEXTVAL_NEXTTOKE.ival = OP_LC;
3343 NEXTVAL_NEXTTOKE.ival = OP_UC;
3345 NEXTVAL_NEXTTOKE.ival = OP_QUOTEMETA;
3347 Perl_croak(aTHX_ "panic: yylex");
3349 SV* const tmpsv = newSVpvs("");
3350 Perl_sv_catpvf(aTHX_ tmpsv, "\\%c", *s);
3356 if (PL_lex_starts) {
3362 sv_free(PL_thistoken);
3363 PL_thistoken = newSVpvs("");
3366 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
3367 if (PL_lex_casemods == 1 && PL_lex_inpat)
3376 case LEX_INTERPPUSH:
3377 return REPORT(sublex_push());
3379 case LEX_INTERPSTART:
3380 if (PL_bufptr == PL_bufend)
3381 return REPORT(sublex_done());
3382 DEBUG_T({ PerlIO_printf(Perl_debug_log,
3383 "### Interpolated variable\n"); });
3385 PL_lex_dojoin = (*PL_bufptr == '@');
3386 PL_lex_state = LEX_INTERPNORMAL;
3387 if (PL_lex_dojoin) {
3388 start_force(PL_curforce);
3389 NEXTVAL_NEXTTOKE.ival = 0;
3391 start_force(PL_curforce);
3392 force_ident("\"", '$');
3393 start_force(PL_curforce);
3394 NEXTVAL_NEXTTOKE.ival = 0;
3396 start_force(PL_curforce);
3397 NEXTVAL_NEXTTOKE.ival = 0;
3399 start_force(PL_curforce);
3400 NEXTVAL_NEXTTOKE.ival = OP_JOIN; /* emulate join($", ...) */
3403 if (PL_lex_starts++) {
3408 sv_free(PL_thistoken);
3409 PL_thistoken = newSVpvs("");
3412 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
3413 if (!PL_lex_casemods && PL_lex_inpat)
3420 case LEX_INTERPENDMAYBE:
3421 if (intuit_more(PL_bufptr)) {
3422 PL_lex_state = LEX_INTERPNORMAL; /* false alarm, more expr */
3428 if (PL_lex_dojoin) {
3429 PL_lex_dojoin = FALSE;
3430 PL_lex_state = LEX_INTERPCONCAT;
3434 sv_free(PL_thistoken);
3435 PL_thistoken = newSVpvs("");
3440 if (PL_lex_inwhat == OP_SUBST && PL_linestr == PL_lex_repl
3441 && SvEVALED(PL_lex_repl))
3443 if (PL_bufptr != PL_bufend)
3444 Perl_croak(aTHX_ "Bad evalled substitution pattern");
3448 case LEX_INTERPCONCAT:
3450 if (PL_lex_brackets)
3451 Perl_croak(aTHX_ "panic: INTERPCONCAT");
3453 if (PL_bufptr == PL_bufend)
3454 return REPORT(sublex_done());
3456 if (SvIVX(PL_linestr) == '\'') {
3457 SV *sv = newSVsv(PL_linestr);
3460 else if ( PL_hints & HINT_NEW_RE )
3461 sv = new_constant(NULL, 0, "qr", sv, sv, "q");
3462 yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
3466 s = scan_const(PL_bufptr);
3468 PL_lex_state = LEX_INTERPCASEMOD;
3470 PL_lex_state = LEX_INTERPSTART;
3473 if (s != PL_bufptr) {
3474 start_force(PL_curforce);
3476 curmad('X', newSVpvn(PL_bufptr,s-PL_bufptr));
3478 NEXTVAL_NEXTTOKE = yylval;
3481 if (PL_lex_starts++) {
3485 sv_free(PL_thistoken);
3486 PL_thistoken = newSVpvs("");
3489 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
3490 if (!PL_lex_casemods && PL_lex_inpat)
3503 PL_lex_state = LEX_NORMAL;
3504 s = scan_formline(PL_bufptr);
3505 if (!PL_lex_formbrack)
3511 PL_oldoldbufptr = PL_oldbufptr;
3517 sv_free(PL_thistoken);
3520 PL_realtokenstart = s - SvPVX(PL_linestr); /* assume but undo on ws */
3524 if (isIDFIRST_lazy_if(s,UTF))
3526 Perl_croak(aTHX_ "Unrecognized character \\x%02X", *s & 255);
3529 goto fake_eof; /* emulate EOF on ^D or ^Z */
3538 if (PL_lex_brackets) {
3539 yyerror((const char *)
3541 ? "Format not terminated"
3542 : "Missing right curly or square bracket"));
3544 DEBUG_T( { PerlIO_printf(Perl_debug_log,
3545 "### Tokener got EOF\n");
3549 if (s++ < PL_bufend)
3550 goto retry; /* ignore stray nulls */
3553 if (!PL_in_eval && !PL_preambled) {
3554 PL_preambled = TRUE;
3559 sv_setpv(PL_linestr,incl_perldb());
3560 if (SvCUR(PL_linestr))
3561 sv_catpvs(PL_linestr,";");
3563 while(AvFILLp(PL_preambleav) >= 0) {
3564 SV *tmpsv = av_shift(PL_preambleav);
3565 sv_catsv(PL_linestr, tmpsv);
3566 sv_catpvs(PL_linestr, ";");
3569 sv_free((SV*)PL_preambleav);
3570 PL_preambleav = NULL;
3572 if (PL_minus_n || PL_minus_p) {
3573 sv_catpvs(PL_linestr, "LINE: while (<>) {");
3575 sv_catpvs(PL_linestr,"chomp;");
3578 if ((*PL_splitstr == '/' || *PL_splitstr == '\''
3579 || *PL_splitstr == '"')
3580 && strchr(PL_splitstr + 1, *PL_splitstr))
3581 Perl_sv_catpvf(aTHX_ PL_linestr, "our @F=split(%s);", PL_splitstr);
3583 /* "q\0${splitstr}\0" is legal perl. Yes, even NUL
3584 bytes can be used as quoting characters. :-) */
3585 const char *splits = PL_splitstr;
3586 sv_catpvs(PL_linestr, "our @F=split(q\0");
3589 if (*splits == '\\')
3590 sv_catpvn(PL_linestr, splits, 1);
3591 sv_catpvn(PL_linestr, splits, 1);
3592 } while (*splits++);
3593 /* This loop will embed the trailing NUL of
3594 PL_linestr as the last thing it does before
3596 sv_catpvs(PL_linestr, ");");
3600 sv_catpvs(PL_linestr,"our @F=split(' ');");
3604 sv_catpvs(PL_linestr,"use feature ':5.10';");
3605 sv_catpvs(PL_linestr, "\n");
3606 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3607 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3608 PL_last_lop = PL_last_uni = NULL;
3609 if (PERLDB_LINE && PL_curstash != PL_debstash)
3610 update_debugger_info(PL_linestr, NULL, 0);
3614 bof = PL_rsfp ? TRUE : FALSE;
3615 if ((s = filter_gets(PL_linestr, PL_rsfp, 0)) == NULL) {
3618 PL_realtokenstart = -1;
3621 if (PL_preprocess && !PL_in_eval)
3622 (void)PerlProc_pclose(PL_rsfp);
3623 else if ((PerlIO *)PL_rsfp == PerlIO_stdin())
3624 PerlIO_clearerr(PL_rsfp);
3626 (void)PerlIO_close(PL_rsfp);
3628 PL_doextract = FALSE;
3630 if (!PL_in_eval && (PL_minus_n || PL_minus_p)) {
3635 sv_setpv(PL_linestr,
3638 ? ";}continue{print;}" : ";}"));
3639 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3640 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3641 PL_last_lop = PL_last_uni = NULL;
3642 PL_minus_n = PL_minus_p = 0;
3645 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3646 PL_last_lop = PL_last_uni = NULL;
3647 sv_setpvn(PL_linestr,"",0);
3648 TOKEN(';'); /* not infinite loop because rsfp is NULL now */
3650 /* If it looks like the start of a BOM or raw UTF-16,
3651 * check if it in fact is. */
3657 #ifdef PERLIO_IS_STDIO
3658 # ifdef __GNU_LIBRARY__
3659 # if __GNU_LIBRARY__ == 1 /* Linux glibc5 */
3660 # define FTELL_FOR_PIPE_IS_BROKEN
3664 # if __GLIBC__ == 1 /* maybe some glibc5 release had it like this? */
3665 # define FTELL_FOR_PIPE_IS_BROKEN
3670 #ifdef FTELL_FOR_PIPE_IS_BROKEN
3671 /* This loses the possibility to detect the bof
3672 * situation on perl -P when the libc5 is being used.
3673 * Workaround? Maybe attach some extra state to PL_rsfp?
3676 bof = PerlIO_tell(PL_rsfp) == SvCUR(PL_linestr);
3678 bof = PerlIO_tell(PL_rsfp) == (Off_t)SvCUR(PL_linestr);
3681 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3682 s = swallow_bom((U8*)s);
3686 /* Incest with pod. */
3689 sv_catsv(PL_thiswhite, PL_linestr);
3691 if (*s == '=' && strnEQ(s, "=cut", 4) && !isALPHA(s[4])) {
3692 sv_setpvn(PL_linestr, "", 0);
3693 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3694 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3695 PL_last_lop = PL_last_uni = NULL;
3696 PL_doextract = FALSE;
3700 } while (PL_doextract);
3701 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = PL_linestart = s;
3702 if (PERLDB_LINE && PL_curstash != PL_debstash)
3703 update_debugger_info(PL_linestr, NULL, 0);
3704 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3705 PL_last_lop = PL_last_uni = NULL;
3706 if (CopLINE(PL_curcop) == 1) {
3707 while (s < PL_bufend && isSPACE(*s))
3709 if (*s == ':' && s[1] != ':') /* for csh execing sh scripts */
3713 PL_thiswhite = newSVpvn(PL_linestart, s - PL_linestart);
3717 if (*s == '#' && *(s+1) == '!')
3719 #ifdef ALTERNATE_SHEBANG
3721 static char const as[] = ALTERNATE_SHEBANG;
3722 if (*s == as[0] && strnEQ(s, as, sizeof(as) - 1))
3723 d = s + (sizeof(as) - 1);
3725 #endif /* ALTERNATE_SHEBANG */
3734 while (*d && !isSPACE(*d))
3738 #ifdef ARG_ZERO_IS_SCRIPT
3739 if (ipathend > ipath) {
3741 * HP-UX (at least) sets argv[0] to the script name,
3742 * which makes $^X incorrect. And Digital UNIX and Linux,
3743 * at least, set argv[0] to the basename of the Perl
3744 * interpreter. So, having found "#!", we'll set it right.
3746 SV * const x = GvSV(gv_fetchpvs("\030", GV_ADD|GV_NOTQUAL,
3748 assert(SvPOK(x) || SvGMAGICAL(x));
3749 if (sv_eq(x, CopFILESV(PL_curcop))) {
3750 sv_setpvn(x, ipath, ipathend - ipath);
3756 const char *bstart = SvPV_const(CopFILESV(PL_curcop),blen);
3757 const char * const lstart = SvPV_const(x,llen);
3759 bstart += blen - llen;
3760 if (strnEQ(bstart, lstart, llen) && bstart[-1] == '/') {
3761 sv_setpvn(x, ipath, ipathend - ipath);
3766 TAINT_NOT; /* $^X is always tainted, but that's OK */
3768 #endif /* ARG_ZERO_IS_SCRIPT */
3773 d = instr(s,"perl -");
3775 d = instr(s,"perl");
3777 /* avoid getting into infinite loops when shebang
3778 * line contains "Perl" rather than "perl" */
3780 for (d = ipathend-4; d >= ipath; --d) {
3781 if ((*d == 'p' || *d == 'P')
3782 && !ibcmp(d, "perl", 4))
3792 #ifdef ALTERNATE_SHEBANG
3794 * If the ALTERNATE_SHEBANG on this system starts with a
3795 * character that can be part of a Perl expression, then if
3796 * we see it but not "perl", we're probably looking at the
3797 * start of Perl code, not a request to hand off to some
3798 * other interpreter. Similarly, if "perl" is there, but
3799 * not in the first 'word' of the line, we assume the line
3800 * contains the start of the Perl program.
3802 if (d && *s != '#') {
3803 const char *c = ipath;
3804 while (*c && !strchr("; \t\r\n\f\v#", *c))
3807 d = NULL; /* "perl" not in first word; ignore */
3809 *s = '#'; /* Don't try to parse shebang line */
3811 #endif /* ALTERNATE_SHEBANG */
3812 #ifndef MACOS_TRADITIONAL
3817 !instr(s,"indir") &&
3818 instr(PL_origargv[0],"perl"))
3825 while (s < PL_bufend && isSPACE(*s))
3827 if (s < PL_bufend) {
3828 Newxz(newargv,PL_origargc+3,char*);
3830 while (s < PL_bufend && !isSPACE(*s))
3833 Copy(PL_origargv+1, newargv+2, PL_origargc+1, char*);
3836 newargv = PL_origargv;
3839 PerlProc_execv(ipath, EXEC_ARGV_CAST(newargv));
3841 Perl_croak(aTHX_ "Can't exec %s", ipath);
3845 while (*d && !isSPACE(*d))
3847 while (SPACE_OR_TAB(*d))
3851 const bool switches_done = PL_doswitches;
3852 const U32 oldpdb = PL_perldb;
3853 const bool oldn = PL_minus_n;
3854 const bool oldp = PL_minus_p;
3857 if (*d == 'M' || *d == 'm' || *d == 'C') {
3858 const char * const m = d;
3859 while (*d && !isSPACE(*d))
3861 Perl_croak(aTHX_ "Too late for \"-%.*s\" option",
3864 d = moreswitches(d);
3866 if (PL_doswitches && !switches_done) {
3867 int argc = PL_origargc;
3868 char **argv = PL_origargv;
3871 } while (argc && argv[0][0] == '-' && argv[0][1]);
3872 init_argv_symbols(argc,argv);
3874 if ((PERLDB_LINE && !oldpdb) ||
3875 ((PL_minus_n || PL_minus_p) && !(oldn || oldp)))
3876 /* if we have already added "LINE: while (<>) {",
3877 we must not do it again */
3879 sv_setpvn(PL_linestr, "", 0);
3880 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3881 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3882 PL_last_lop = PL_last_uni = NULL;
3883 PL_preambled = FALSE;
3885 (void)gv_fetchfile(PL_origfilename);
3892 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
3894 PL_lex_state = LEX_FORMLINE;
3899 #ifdef PERL_STRICT_CR
3900 Perl_warn(aTHX_ "Illegal character \\%03o (carriage return)", '\r');
3902 "\t(Maybe you didn't strip carriage returns after a network transfer?)\n");
3904 case ' ': case '\t': case '\f': case 013:
3905 #ifdef MACOS_TRADITIONAL
3909 PL_realtokenstart = -1;
3918 PL_realtokenstart = -1;
3922 if (PL_lex_state != LEX_NORMAL || (PL_in_eval && !PL_rsfp)) {
3923 if (*s == '#' && s == PL_linestart && PL_in_eval && !PL_rsfp) {
3924 /* handle eval qq[#line 1 "foo"\n ...] */
3925 CopLINE_dec(PL_curcop);
3928 if (PL_madskills && !PL_lex_formbrack && !PL_in_eval) {
3930 if (!PL_in_eval || PL_rsfp)
3935 while (d < PL_bufend && *d != '\n')
3939 else if (d > PL_bufend) /* Found by Ilya: feed random input to Perl. */
3940 Perl_croak(aTHX_ "panic: input overflow");
3943 PL_thiswhite = newSVpvn(s, d - s);
3948 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
3950 PL_lex_state = LEX_FORMLINE;
3956 if (PL_madskills && CopLINE(PL_curcop) >= 1 && !PL_lex_formbrack) {
3957 if (CopLINE(PL_curcop) == 1 && s[0] == '#' && s[1] == '!') {
3960 TOKEN(PEG); /* make sure any #! line is accessible */
3965 /* if (PL_madskills && PL_lex_formbrack) { */
3967 while (d < PL_bufend && *d != '\n')
3971 else if (d > PL_bufend) /* Found by Ilya: feed random input to Perl. */
3972 Perl_croak(aTHX_ "panic: input overflow");
3973 if (PL_madskills && CopLINE(PL_curcop) >= 1) {
3975 PL_thiswhite = newSVpvs("");
3976 if (CopLINE(PL_curcop) == 1) {
3977 sv_setpvn(PL_thiswhite, "", 0);
3980 sv_catpvn(PL_thiswhite, s, d - s);
3994 if (s[1] && isALPHA(s[1]) && !isALNUM(s[2])) {
4002 while (s < PL_bufend && SPACE_OR_TAB(*s))
4005 if (strnEQ(s,"=>",2)) {
4006 s = force_word(PL_bufptr,WORD,FALSE,FALSE,FALSE);
4007 DEBUG_T( { printbuf("### Saw unary minus before =>, forcing word %s\n", s); } );
4008 OPERATOR('-'); /* unary minus */
4010 PL_last_uni = PL_oldbufptr;
4012 case 'r': ftst = OP_FTEREAD; break;
4013 case 'w': ftst = OP_FTEWRITE; break;
4014 case 'x': ftst = OP_FTEEXEC; break;
4015 case 'o': ftst = OP_FTEOWNED; break;
4016 case 'R': ftst = OP_FTRREAD; break;
4017 case 'W': ftst = OP_FTRWRITE; break;
4018 case 'X': ftst = OP_FTREXEC; break;
4019 case 'O': ftst = OP_FTROWNED; break;
4020 case 'e': ftst = OP_FTIS; break;
4021 case 'z': ftst = OP_FTZERO; break;
4022 case 's': ftst = OP_FTSIZE; break;
4023 case 'f': ftst = OP_FTFILE; break;
4024 case 'd': ftst = OP_FTDIR; break;
4025 case 'l': ftst = OP_FTLINK; break;
4026 case 'p': ftst = OP_FTPIPE; break;
4027 case 'S': ftst = OP_FTSOCK; break;
4028 case 'u': ftst = OP_FTSUID; break;
4029 case 'g': ftst = OP_FTSGID; break;
4030 case 'k': ftst = OP_FTSVTX; break;
4031 case 'b': ftst = OP_FTBLK; break;
4032 case 'c': ftst = OP_FTCHR; break;
4033 case 't': ftst = OP_FTTTY; break;
4034 case 'T': ftst = OP_FTTEXT; break;
4035 case 'B': ftst = OP_FTBINARY; break;
4036 case 'M': case 'A': case 'C':
4037 gv_fetchpvs("\024", GV_ADD|GV_NOTQUAL, SVt_PV);
4039 case 'M': ftst = OP_FTMTIME; break;
4040 case 'A': ftst = OP_FTATIME; break;
4041 case 'C': ftst = OP_FTCTIME; break;
4049 PL_last_lop_op = (OPCODE)ftst;
4050 DEBUG_T( { PerlIO_printf(Perl_debug_log,
4051 "### Saw file test %c\n", (int)tmp);
4056 /* Assume it was a minus followed by a one-letter named
4057 * subroutine call (or a -bareword), then. */
4058 DEBUG_T( { PerlIO_printf(Perl_debug_log,
4059 "### '-%c' looked like a file test but was not\n",
4066 const char tmp = *s++;
4069 if (PL_expect == XOPERATOR)
4074 else if (*s == '>') {
4077 if (isIDFIRST_lazy_if(s,UTF)) {
4078 s = force_word(s,METHOD,FALSE,TRUE,FALSE);
4086 if (PL_expect == XOPERATOR)
4089 if (isSPACE(*s) || !isSPACE(*PL_bufptr))
4091 OPERATOR('-'); /* unary minus */
4097 const char tmp = *s++;
4100 if (PL_expect == XOPERATOR)
4105 if (PL_expect == XOPERATOR)
4108 if (isSPACE(*s) || !isSPACE(*PL_bufptr))
4115 if (PL_expect != XOPERATOR) {
4116 s = scan_ident(s, PL_bufend, PL_tokenbuf, sizeof PL_tokenbuf, TRUE);
4117 PL_expect = XOPERATOR;
4118 force_ident(PL_tokenbuf, '*');
4131 if (PL_expect == XOPERATOR) {
4135 PL_tokenbuf[0] = '%';
4136 s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, TRUE);
4137 if (!PL_tokenbuf[1]) {
4140 PL_pending_ident = '%';
4151 && (PL_expect == XOPERATOR || PL_expect == XTERMORDORDOR))
4158 const char tmp = *s++;
4164 goto just_a_word_zero_gv;
4167 switch (PL_expect) {
4173 if (!PL_in_my || PL_lex_state != LEX_NORMAL)
4175 PL_bufptr = s; /* update in case we back off */
4181 PL_expect = XTERMBLOCK;
4184 stuffstart = s - SvPVX(PL_linestr) - 1;
4188 while (isIDFIRST_lazy_if(s,UTF)) {
4191 d = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &len);
4192 if (isLOWER(*s) && (tmp = keyword(PL_tokenbuf, len, 0))) {
4193 if (tmp < 0) tmp = -tmp;
4208 sv = newSVpvn(s, len);
4210 d = scan_str(d,TRUE,TRUE);
4212 /* MUST advance bufptr here to avoid bogus
4213 "at end of line" context messages from yyerror().
4215 PL_bufptr = s + len;
4216 yyerror("Unterminated attribute parameter in attribute list");
4220 return REPORT(0); /* EOF indicator */
4224 sv_catsv(sv, PL_lex_stuff);
4225 attrs = append_elem(OP_LIST, attrs,
4226 newSVOP(OP_CONST, 0, sv));
4227 SvREFCNT_dec(PL_lex_stuff);
4228 PL_lex_stuff = NULL;
4231 if (len == 6 && strnEQ(SvPVX(sv), "unique", len)) {
4233 if (PL_in_my == KEY_our) {
4235 GvUNIQUE_on(cGVOPx_gv(yylval.opval));
4237 /* skip to avoid loading attributes.pm */
4239 deprecate(":unique");
4242 Perl_croak(aTHX_ "The 'unique' attribute may only be applied to 'our' variables");
4245 /* NOTE: any CV attrs applied here need to be part of
4246 the CVf_BUILTIN_ATTRS define in cv.h! */
4247 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "lvalue", len)) {
4249 CvLVALUE_on(PL_compcv);
4251 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "locked", len)) {
4253 CvLOCKED_on(PL_compcv);
4255 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "method", len)) {
4257 CvMETHOD_on(PL_compcv);
4259 else if (!PL_in_my && len == 9 && strnEQ(SvPVX(sv), "assertion", len)) {
4261 CvASSERTION_on(PL_compcv);
4263 /* After we've set the flags, it could be argued that
4264 we don't need to do the attributes.pm-based setting
4265 process, and shouldn't bother appending recognized
4266 flags. To experiment with that, uncomment the
4267 following "else". (Note that's already been
4268 uncommented. That keeps the above-applied built-in
4269 attributes from being intercepted (and possibly
4270 rejected) by a package's attribute routines, but is
4271 justified by the performance win for the common case
4272 of applying only built-in attributes.) */
4274 attrs = append_elem(OP_LIST, attrs,
4275 newSVOP(OP_CONST, 0,
4279 if (*s == ':' && s[1] != ':')
4282 break; /* require real whitespace or :'s */
4283 /* XXX losing whitespace on sequential attributes here */
4287 = (PL_expect == XOPERATOR ? '=' : '{'); /*'}(' for vi */
4288 if (*s != ';' && *s != '}' && *s != tmp
4289 && (tmp != '=' || *s != ')')) {
4290 const char q = ((*s == '\'') ? '"' : '\'');
4291 /* If here for an expression, and parsed no attrs, back
4293 if (tmp == '=' && !attrs) {
4297 /* MUST advance bufptr here to avoid bogus "at end of line"
4298 context messages from yyerror().
4301 yyerror( (const char *)
4303 ? Perl_form(aTHX_ "Invalid separator character "
4304 "%c%c%c in attribute list", q, *s, q)
4305 : "Unterminated attribute list" ) );
4313 start_force(PL_curforce);
4314 NEXTVAL_NEXTTOKE.opval = attrs;
4315 CURMAD('_', PL_nextwhite);
4320 PL_thistoken = newSVpvn(SvPVX(PL_linestr) + stuffstart,
4321 (s - SvPVX(PL_linestr)) - stuffstart);
4329 if (PL_last_lop == PL_oldoldbufptr || PL_last_uni == PL_oldoldbufptr)
4330 PL_oldbufptr = PL_oldoldbufptr; /* allow print(STDOUT 123) */
4338 const char tmp = *s++;
4343 const char tmp = *s++;
4351 if (PL_lex_brackets <= 0)
4352 yyerror("Unmatched right square bracket");
4355 if (PL_lex_state == LEX_INTERPNORMAL) {
4356 if (PL_lex_brackets == 0) {
4357 if (*s != '[' && *s != '{' && (*s != '-' || s[1] != '>'))
4358 PL_lex_state = LEX_INTERPEND;
4365 if (PL_lex_brackets > 100) {
4366 Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
4368 switch (PL_expect) {
4370 if (PL_lex_formbrack) {
4374 if (PL_oldoldbufptr == PL_last_lop)
4375 PL_lex_brackstack[PL_lex_brackets++] = XTERM;
4377 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
4378 OPERATOR(HASHBRACK);
4380 while (s < PL_bufend && SPACE_OR_TAB(*s))
4383 PL_tokenbuf[0] = '\0';
4384 if (d < PL_bufend && *d == '-') {
4385 PL_tokenbuf[0] = '-';
4387 while (d < PL_bufend && SPACE_OR_TAB(*d))
4390 if (d < PL_bufend && isIDFIRST_lazy_if(d,UTF)) {
4391 d = scan_word(d, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1,
4393 while (d < PL_bufend && SPACE_OR_TAB(*d))
4396 const char minus = (PL_tokenbuf[0] == '-');
4397 s = force_word(s + minus, WORD, FALSE, TRUE, FALSE);
4405 PL_lex_brackstack[PL_lex_brackets++] = XSTATE;
4410 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
4415 if (PL_oldoldbufptr == PL_last_lop)
4416 PL_lex_brackstack[PL_lex_brackets++] = XTERM;