3 * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
4 * 2000, 2001, 2002, 2003, 2004, 2005, 2006, 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 static const char ident_too_long[] = "Identifier too long";
29 static const char commaless_variable_list[] = "comma-less variable list";
31 static void restore_rsfp(pTHX_ void *f);
32 #ifndef PERL_NO_UTF16_FILTER
33 static I32 utf16_textfilter(pTHX_ int idx, SV *sv, int maxlen);
34 static I32 utf16rev_textfilter(pTHX_ int idx, SV *sv, int maxlen);
38 # define CURMAD(slot,sv) if (PL_madskills) { curmad(slot,sv); sv = 0; }
39 # define NEXTVAL_NEXTTOKE PL_nexttoke[PL_curforce].next_val
41 # define CURMAD(slot,sv)
42 # define NEXTVAL_NEXTTOKE PL_nextval[PL_nexttoke]
45 #define XFAKEBRACK 128
48 #ifdef USE_UTF8_SCRIPTS
49 # define UTF (!IN_BYTES)
51 # define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
54 /* In variables named $^X, these are the legal values for X.
55 * 1999-02-27 mjd-perl-patch@plover.com */
56 #define isCONTROLVAR(x) (isUPPER(x) || strchr("[\\]^_?", (x)))
58 /* On MacOS, respect nonbreaking spaces */
59 #ifdef MACOS_TRADITIONAL
60 #define SPACE_OR_TAB(c) ((c)==' '||(c)=='\312'||(c)=='\t')
62 #define SPACE_OR_TAB(c) ((c)==' '||(c)=='\t')
65 /* LEX_* are values for PL_lex_state, the state of the lexer.
66 * They are arranged oddly so that the guard on the switch statement
67 * can get by with a single comparison (if the compiler is smart enough).
70 /* #define LEX_NOTPARSING 11 is done in perl.h. */
72 #define LEX_NORMAL 10 /* normal code (ie not within "...") */
73 #define LEX_INTERPNORMAL 9 /* code within a string, eg "$foo[$x+1]" */
74 #define LEX_INTERPCASEMOD 8 /* expecting a \U, \Q or \E etc */
75 #define LEX_INTERPPUSH 7 /* starting a new sublex parse level */
76 #define LEX_INTERPSTART 6 /* expecting the start of a $var */
78 /* at end of code, eg "$x" followed by: */
79 #define LEX_INTERPEND 5 /* ... eg not one of [, { or -> */
80 #define LEX_INTERPENDMAYBE 4 /* ... eg one of [, { or -> */
82 #define LEX_INTERPCONCAT 3 /* expecting anything, eg at start of
83 string or after \E, $foo, etc */
84 #define LEX_INTERPCONST 2 /* NOT USED */
85 #define LEX_FORMLINE 1 /* expecting a format line */
86 #define LEX_KNOWNEXT 0 /* next token known; just return it */
90 static const char* const lex_state_names[] = {
109 #include "keywords.h"
111 /* CLINE is a macro that ensures PL_copline has a sane value */
116 #define CLINE (PL_copline = (CopLINE(PL_curcop) < PL_copline ? CopLINE(PL_curcop) : PL_copline))
119 # define SKIPSPACE0(s) skipspace0(s)
120 # define SKIPSPACE1(s) skipspace1(s)
121 # define SKIPSPACE2(s,tsv) skipspace2(s,&tsv)
122 # define PEEKSPACE(s) skipspace2(s,0)
124 # define SKIPSPACE0(s) skipspace(s)
125 # define SKIPSPACE1(s) skipspace(s)
126 # define SKIPSPACE2(s,tsv) skipspace(s)
127 # define PEEKSPACE(s) skipspace(s)
131 * Convenience functions to return different tokens and prime the
132 * lexer for the next token. They all take an argument.
134 * TOKEN : generic token (used for '(', DOLSHARP, etc)
135 * OPERATOR : generic operator
136 * AOPERATOR : assignment operator
137 * PREBLOCK : beginning the block after an if, while, foreach, ...
138 * PRETERMBLOCK : beginning a non-code-defining {} block (eg, hash ref)
139 * PREREF : *EXPR where EXPR is not a simple identifier
140 * TERM : expression term
141 * LOOPX : loop exiting command (goto, last, dump, etc)
142 * FTST : file test operator
143 * FUN0 : zero-argument function
144 * FUN1 : not used, except for not, which isn't a UNIOP
145 * BOop : bitwise or or xor
147 * SHop : shift operator
148 * PWop : power operator
149 * PMop : pattern-matching operator
150 * Aop : addition-level operator
151 * Mop : multiplication-level operator
152 * Eop : equality-testing operator
153 * Rop : relational operator <= != gt
155 * Also see LOP and lop() below.
158 #ifdef DEBUGGING /* Serve -DT. */
159 # define REPORT(retval) tokereport((I32)retval)
161 # define REPORT(retval) (retval)
164 #define TOKEN(retval) return ( PL_bufptr = s, REPORT(retval))
165 #define OPERATOR(retval) return (PL_expect = XTERM, PL_bufptr = s, REPORT(retval))
166 #define AOPERATOR(retval) return ao((PL_expect = XTERM, PL_bufptr = s, REPORT(retval)))
167 #define PREBLOCK(retval) return (PL_expect = XBLOCK,PL_bufptr = s, REPORT(retval))
168 #define PRETERMBLOCK(retval) return (PL_expect = XTERMBLOCK,PL_bufptr = s, REPORT(retval))
169 #define PREREF(retval) return (PL_expect = XREF,PL_bufptr = s, REPORT(retval))
170 #define TERM(retval) return (CLINE, PL_expect = XOPERATOR, PL_bufptr = s, REPORT(retval))
171 #define LOOPX(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)LOOPEX))
172 #define FTST(f) return (yylval.ival=f, PL_expect=XTERMORDORDOR, PL_bufptr=s, REPORT((int)UNIOP))
173 #define FUN0(f) return (yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC0))
174 #define FUN1(f) return (yylval.ival=f, PL_expect=XOPERATOR, PL_bufptr=s, REPORT((int)FUNC1))
175 #define BOop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITOROP)))
176 #define BAop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)BITANDOP)))
177 #define SHop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)SHIFTOP)))
178 #define PWop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)POWOP)))
179 #define PMop(f) return(yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MATCHOP))
180 #define Aop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)ADDOP)))
181 #define Mop(f) return ao((yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)MULOP)))
182 #define Eop(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)EQOP))
183 #define Rop(f) return (yylval.ival=f, PL_expect=XTERM, PL_bufptr=s, REPORT((int)RELOP))
185 /* This bit of chicanery makes a unary function followed by
186 * a parenthesis into a function with one argument, highest precedence.
187 * The UNIDOR macro is for unary functions that can be followed by the //
188 * operator (such as C<shift // 0>).
190 #define UNI2(f,x) { \
194 PL_last_uni = PL_oldbufptr; \
195 PL_last_lop_op = f; \
197 return REPORT( (int)FUNC1 ); \
199 return REPORT( *s=='(' ? (int)FUNC1 : (int)UNIOP ); \
201 #define UNI(f) UNI2(f,XTERM)
202 #define UNIDOR(f) UNI2(f,XTERMORDORDOR)
204 #define UNIBRACK(f) { \
207 PL_last_uni = PL_oldbufptr; \
209 return REPORT( (int)FUNC1 ); \
211 return REPORT( (*s == '(') ? (int)FUNC1 : (int)UNIOP ); \
214 /* grandfather return to old style */
215 #define OLDLOP(f) return(yylval.ival=f,PL_expect = XTERM,PL_bufptr = s,(int)LSTOP)
219 /* how to interpret the yylval associated with the token */
223 TOKENTYPE_OPNUM, /* yylval.ival contains an opcode number */
229 static struct debug_tokens {
231 enum token_type type;
233 } const debug_tokens[] =
235 { ADDOP, TOKENTYPE_OPNUM, "ADDOP" },
236 { ANDAND, TOKENTYPE_NONE, "ANDAND" },
237 { ANDOP, TOKENTYPE_NONE, "ANDOP" },
238 { ANONSUB, TOKENTYPE_IVAL, "ANONSUB" },
239 { ARROW, TOKENTYPE_NONE, "ARROW" },
240 { ASSIGNOP, TOKENTYPE_OPNUM, "ASSIGNOP" },
241 { BITANDOP, TOKENTYPE_OPNUM, "BITANDOP" },
242 { BITOROP, TOKENTYPE_OPNUM, "BITOROP" },
243 { COLONATTR, TOKENTYPE_NONE, "COLONATTR" },
244 { CONTINUE, TOKENTYPE_NONE, "CONTINUE" },
245 { DEFAULT, TOKENTYPE_NONE, "DEFAULT" },
246 { DO, TOKENTYPE_NONE, "DO" },
247 { DOLSHARP, TOKENTYPE_NONE, "DOLSHARP" },
248 { DORDOR, TOKENTYPE_NONE, "DORDOR" },
249 { DOROP, TOKENTYPE_OPNUM, "DOROP" },
250 { DOTDOT, TOKENTYPE_IVAL, "DOTDOT" },
251 { ELSE, TOKENTYPE_NONE, "ELSE" },
252 { ELSIF, TOKENTYPE_IVAL, "ELSIF" },
253 { EQOP, TOKENTYPE_OPNUM, "EQOP" },
254 { FOR, TOKENTYPE_IVAL, "FOR" },
255 { FORMAT, TOKENTYPE_NONE, "FORMAT" },
256 { FUNC, TOKENTYPE_OPNUM, "FUNC" },
257 { FUNC0, TOKENTYPE_OPNUM, "FUNC0" },
258 { FUNC0SUB, TOKENTYPE_OPVAL, "FUNC0SUB" },
259 { FUNC1, TOKENTYPE_OPNUM, "FUNC1" },
260 { FUNCMETH, TOKENTYPE_OPVAL, "FUNCMETH" },
261 { GIVEN, TOKENTYPE_IVAL, "GIVEN" },
262 { HASHBRACK, TOKENTYPE_NONE, "HASHBRACK" },
263 { IF, TOKENTYPE_IVAL, "IF" },
264 { LABEL, TOKENTYPE_PVAL, "LABEL" },
265 { LOCAL, TOKENTYPE_IVAL, "LOCAL" },
266 { LOOPEX, TOKENTYPE_OPNUM, "LOOPEX" },
267 { LSTOP, TOKENTYPE_OPNUM, "LSTOP" },
268 { LSTOPSUB, TOKENTYPE_OPVAL, "LSTOPSUB" },
269 { MATCHOP, TOKENTYPE_OPNUM, "MATCHOP" },
270 { METHOD, TOKENTYPE_OPVAL, "METHOD" },
271 { MULOP, TOKENTYPE_OPNUM, "MULOP" },
272 { MY, TOKENTYPE_IVAL, "MY" },
273 { MYSUB, TOKENTYPE_NONE, "MYSUB" },
274 { NOAMP, TOKENTYPE_NONE, "NOAMP" },
275 { NOTOP, TOKENTYPE_NONE, "NOTOP" },
276 { OROP, TOKENTYPE_IVAL, "OROP" },
277 { OROR, TOKENTYPE_NONE, "OROR" },
278 { PACKAGE, TOKENTYPE_NONE, "PACKAGE" },
279 { PMFUNC, TOKENTYPE_OPVAL, "PMFUNC" },
280 { POSTDEC, TOKENTYPE_NONE, "POSTDEC" },
281 { POSTINC, TOKENTYPE_NONE, "POSTINC" },
282 { POWOP, TOKENTYPE_OPNUM, "POWOP" },
283 { PREDEC, TOKENTYPE_NONE, "PREDEC" },
284 { PREINC, TOKENTYPE_NONE, "PREINC" },
285 { PRIVATEREF, TOKENTYPE_OPVAL, "PRIVATEREF" },
286 { REFGEN, TOKENTYPE_NONE, "REFGEN" },
287 { RELOP, TOKENTYPE_OPNUM, "RELOP" },
288 { SHIFTOP, TOKENTYPE_OPNUM, "SHIFTOP" },
289 { SUB, TOKENTYPE_NONE, "SUB" },
290 { THING, TOKENTYPE_OPVAL, "THING" },
291 { UMINUS, TOKENTYPE_NONE, "UMINUS" },
292 { UNIOP, TOKENTYPE_OPNUM, "UNIOP" },
293 { UNIOPSUB, TOKENTYPE_OPVAL, "UNIOPSUB" },
294 { UNLESS, TOKENTYPE_IVAL, "UNLESS" },
295 { UNTIL, TOKENTYPE_IVAL, "UNTIL" },
296 { USE, TOKENTYPE_IVAL, "USE" },
297 { WHEN, TOKENTYPE_IVAL, "WHEN" },
298 { WHILE, TOKENTYPE_IVAL, "WHILE" },
299 { WORD, TOKENTYPE_OPVAL, "WORD" },
300 { 0, TOKENTYPE_NONE, NULL }
303 /* dump the returned token in rv, plus any optional arg in yylval */
306 S_tokereport(pTHX_ I32 rv)
310 const char *name = NULL;
311 enum token_type type = TOKENTYPE_NONE;
312 const struct debug_tokens *p;
313 SV* const report = newSVpvs("<== ");
315 for (p = debug_tokens; p->token; p++) {
316 if (p->token == (int)rv) {
323 Perl_sv_catpv(aTHX_ report, name);
324 else if ((char)rv > ' ' && (char)rv < '~')
325 Perl_sv_catpvf(aTHX_ report, "'%c'", (char)rv);
327 sv_catpvs(report, "EOF");
329 Perl_sv_catpvf(aTHX_ report, "?? %"IVdf, (IV)rv);
332 case TOKENTYPE_GVVAL: /* doesn't appear to be used */
335 Perl_sv_catpvf(aTHX_ report, "(ival=%"IVdf")", (IV)yylval.ival);
337 case TOKENTYPE_OPNUM:
338 Perl_sv_catpvf(aTHX_ report, "(ival=op_%s)",
339 PL_op_name[yylval.ival]);
342 Perl_sv_catpvf(aTHX_ report, "(pval=\"%s\")", yylval.pval);
344 case TOKENTYPE_OPVAL:
346 Perl_sv_catpvf(aTHX_ report, "(opval=op_%s)",
347 PL_op_name[yylval.opval->op_type]);
348 if (yylval.opval->op_type == OP_CONST) {
349 Perl_sv_catpvf(aTHX_ report, " %s",
350 SvPEEK(cSVOPx_sv(yylval.opval)));
355 sv_catpvs(report, "(opval=null)");
358 PerlIO_printf(Perl_debug_log, "### %s\n\n", SvPV_nolen_const(report));
364 /* print the buffer with suitable escapes */
367 S_printbuf(pTHX_ const char* fmt, const char* s)
369 SV* const tmp = newSVpvs("");
370 PerlIO_printf(Perl_debug_log, fmt, pv_display(tmp, s, strlen(s), 0, 60));
379 * This subroutine detects &&=, ||=, and //= and turns an ANDAND, OROR or DORDOR
380 * into an OP_ANDASSIGN, OP_ORASSIGN, or OP_DORASSIGN
384 S_ao(pTHX_ int toketype)
387 if (*PL_bufptr == '=') {
389 if (toketype == ANDAND)
390 yylval.ival = OP_ANDASSIGN;
391 else if (toketype == OROR)
392 yylval.ival = OP_ORASSIGN;
393 else if (toketype == DORDOR)
394 yylval.ival = OP_DORASSIGN;
402 * When Perl expects an operator and finds something else, no_op
403 * prints the warning. It always prints "<something> found where
404 * operator expected. It prints "Missing semicolon on previous line?"
405 * if the surprise occurs at the start of the line. "do you need to
406 * predeclare ..." is printed out for code like "sub bar; foo bar $x"
407 * where the compiler doesn't know if foo is a method call or a function.
408 * It prints "Missing operator before end of line" if there's nothing
409 * after the missing operator, or "... before <...>" if there is something
410 * after the missing operator.
414 S_no_op(pTHX_ const char *what, char *s)
417 char * const oldbp = PL_bufptr;
418 const bool is_first = (PL_oldbufptr == PL_linestart);
424 yywarn(Perl_form(aTHX_ "%s found where operator expected", what));
425 if (ckWARN_d(WARN_SYNTAX)) {
427 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
428 "\t(Missing semicolon on previous line?)\n");
429 else if (PL_oldoldbufptr && isIDFIRST_lazy_if(PL_oldoldbufptr,UTF)) {
431 for (t = PL_oldoldbufptr; (isALNUM_lazy_if(t,UTF) || *t == ':'); t++)
433 if (t < PL_bufptr && isSPACE(*t))
434 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
435 "\t(Do you need to predeclare %.*s?)\n",
436 (int)(t - PL_oldoldbufptr), PL_oldoldbufptr);
440 Perl_warner(aTHX_ packWARN(WARN_SYNTAX),
441 "\t(Missing operator before %.*s?)\n", (int)(s - oldbp), oldbp);
449 * Complain about missing quote/regexp/heredoc terminator.
450 * If it's called with NULL then it cauterizes the line buffer.
451 * If we're in a delimited string and the delimiter is a control
452 * character, it's reformatted into a two-char sequence like ^C.
457 S_missingterm(pTHX_ char *s)
463 char * const nl = strrchr(s,'\n');
469 iscntrl(PL_multi_close)
471 PL_multi_close < 32 || PL_multi_close == 127
475 tmpbuf[1] = (char)toCTRL(PL_multi_close);
480 *tmpbuf = (char)PL_multi_close;
484 q = strchr(s,'"') ? '\'' : '"';
485 Perl_croak(aTHX_ "Can't find string terminator %c%s%c anywhere before EOF",q,s,q);
488 #define FEATURE_IS_ENABLED(name) \
489 ((0 != (PL_hints & HINT_LOCALIZE_HH)) \
490 && S_feature_is_enabled(aTHX_ STR_WITH_LEN(name)))
492 * S_feature_is_enabled
493 * Check whether the named feature is enabled.
496 S_feature_is_enabled(pTHX_ const char *name, STRLEN namelen)
499 HV * const hinthv = GvHV(PL_hintgv);
500 char he_name[32] = "feature_";
501 (void) my_strlcpy(&he_name[8], name, 24);
503 return (hinthv && hv_exists(hinthv, he_name, 8 + namelen));
511 Perl_deprecate(pTHX_ const char *s)
513 if (ckWARN(WARN_DEPRECATED))
514 Perl_warner(aTHX_ packWARN(WARN_DEPRECATED), "Use of %s is deprecated", s);
518 Perl_deprecate_old(pTHX_ const char *s)
520 /* This function should NOT be called for any new deprecated warnings */
521 /* Use Perl_deprecate instead */
523 /* It is here to maintain backward compatibility with the pre-5.8 */
524 /* warnings category hierarchy. The "deprecated" category used to */
525 /* live under the "syntax" category. It is now a top-level category */
526 /* in its own right. */
528 if (ckWARN2(WARN_DEPRECATED, WARN_SYNTAX))
529 Perl_warner(aTHX_ packWARN2(WARN_DEPRECATED, WARN_SYNTAX),
530 "Use of %s is deprecated", s);
534 * experimental text filters for win32 carriage-returns, utf16-to-utf8 and
535 * utf16-to-utf8-reversed.
538 #ifdef PERL_CR_FILTER
542 register const char *s = SvPVX_const(sv);
543 register const char * const e = s + SvCUR(sv);
544 /* outer loop optimized to do nothing if there are no CR-LFs */
546 if (*s++ == '\r' && *s == '\n') {
547 /* hit a CR-LF, need to copy the rest */
548 register char *d = s - 1;
551 if (*s == '\r' && s[1] == '\n')
562 S_cr_textfilter(pTHX_ int idx, SV *sv, int maxlen)
564 const I32 count = FILTER_READ(idx+1, sv, maxlen);
565 if (count > 0 && !maxlen)
573 * Initialize variables. Uses the Perl save_stack to save its state (for
574 * recursive calls to the parser).
578 Perl_lex_start(pTHX_ SV *line)
584 SAVEI32(PL_lex_dojoin);
585 SAVEI32(PL_lex_brackets);
586 SAVEI32(PL_lex_casemods);
587 SAVEI32(PL_lex_starts);
588 SAVEI32(PL_lex_state);
589 SAVEVPTR(PL_lex_inpat);
590 SAVEI32(PL_lex_inwhat);
592 if (PL_lex_state == LEX_KNOWNEXT) {
593 I32 toke = PL_lasttoke;
594 while (--toke >= 0) {
595 SAVEI32(PL_nexttoke[toke].next_type);
596 SAVEVPTR(PL_nexttoke[toke].next_val);
598 SAVEVPTR(PL_nexttoke[toke].next_mad);
600 SAVEI32(PL_lasttoke);
603 SAVESPTR(PL_thistoken);
604 SAVESPTR(PL_thiswhite);
605 SAVESPTR(PL_nextwhite);
606 SAVESPTR(PL_thisopen);
607 SAVESPTR(PL_thisclose);
608 SAVESPTR(PL_thisstuff);
609 SAVEVPTR(PL_thismad);
610 SAVEI32(PL_realtokenstart);
611 SAVEI32(PL_faketokens);
613 SAVEI32(PL_curforce);
615 if (PL_lex_state == LEX_KNOWNEXT) {
616 I32 toke = PL_nexttoke;
617 while (--toke >= 0) {
618 SAVEI32(PL_nexttype[toke]);
619 SAVEVPTR(PL_nextval[toke]);
621 SAVEI32(PL_nexttoke);
624 SAVECOPLINE(PL_curcop);
627 SAVEPPTR(PL_oldbufptr);
628 SAVEPPTR(PL_oldoldbufptr);
629 SAVEPPTR(PL_last_lop);
630 SAVEPPTR(PL_last_uni);
631 SAVEPPTR(PL_linestart);
632 SAVESPTR(PL_linestr);
633 SAVEGENERICPV(PL_lex_brackstack);
634 SAVEGENERICPV(PL_lex_casestack);
635 SAVEDESTRUCTOR_X(restore_rsfp, PL_rsfp);
636 SAVESPTR(PL_lex_stuff);
637 SAVEI32(PL_lex_defer);
638 SAVEI32(PL_sublex_info.sub_inwhat);
639 SAVESPTR(PL_lex_repl);
641 SAVEINT(PL_lex_expect);
643 PL_lex_state = LEX_NORMAL;
647 Newx(PL_lex_brackstack, 120, char);
648 Newx(PL_lex_casestack, 12, char);
650 *PL_lex_casestack = '\0';
662 PL_sublex_info.sub_inwhat = 0;
664 s = SvPV_const(line, len);
669 PL_linestr = newSVpvs("\n;");
670 } else if (SvREADONLY(line) || s[len-1] != ';') {
671 PL_linestr = newSVsv(line);
673 sv_catpvs(PL_linestr, "\n;");
676 SvREFCNT_inc_simple_void_NN(line);
679 /* PL_linestr needs to survive until end of scope, not just the next
680 FREETMPS. See changes 17505 and 17546 which fixed the symptoms only. */
681 SAVEFREESV(PL_linestr);
682 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = PL_linestart = SvPVX(PL_linestr);
683 PL_bufend = PL_bufptr + SvCUR(PL_linestr);
684 PL_last_lop = PL_last_uni = NULL;
690 * Finalizer for lexing operations. Must be called when the parser is
691 * done with the lexer.
698 PL_doextract = FALSE;
703 * This subroutine has nothing to do with tilting, whether at windmills
704 * or pinball tables. Its name is short for "increment line". It
705 * increments the current line number in CopLINE(PL_curcop) and checks
706 * to see whether the line starts with a comment of the form
707 * # line 500 "foo.pm"
708 * If so, it sets the current line number and file to the values in the comment.
712 S_incline(pTHX_ char *s)
720 CopLINE_inc(PL_curcop);
723 while (SPACE_OR_TAB(*s))
725 if (strnEQ(s, "line", 4))
729 if (SPACE_OR_TAB(*s))
733 while (SPACE_OR_TAB(*s))
741 while (SPACE_OR_TAB(*s))
743 if (*s == '"' && (t = strchr(s+1, '"'))) {
753 while (SPACE_OR_TAB(*e) || *e == '\r' || *e == '\f')
755 if (*e != '\n' && *e != '\0')
756 return; /* false alarm */
762 const char * const cf = CopFILE(PL_curcop);
763 STRLEN tmplen = cf ? strlen(cf) : 0;
764 if (tmplen > 7 && strnEQ(cf, "(eval ", 6)) {
765 /* must copy *{"::_<(eval N)[oldfilename:L]"}
766 * to *{"::_<newfilename"} */
767 char smallbuf[256], smallbuf2[256];
768 char *tmpbuf, *tmpbuf2;
770 STRLEN tmplen2 = strlen(s);
771 if (tmplen + 3 < sizeof smallbuf)
774 Newx(tmpbuf, tmplen + 3, char);
775 if (tmplen2 + 3 < sizeof smallbuf2)
778 Newx(tmpbuf2, tmplen2 + 3, char);
779 tmpbuf[0] = tmpbuf2[0] = '_';
780 tmpbuf[1] = tmpbuf2[1] = '<';
781 memcpy(tmpbuf + 2, cf, ++tmplen);
782 memcpy(tmpbuf2 + 2, s, ++tmplen2);
784 gvp = (GV**)hv_fetch(PL_defstash, tmpbuf, tmplen, FALSE);
786 gv2 = *(GV**)hv_fetch(PL_defstash, tmpbuf2, tmplen2, TRUE);
788 gv_init(gv2, PL_defstash, tmpbuf2, tmplen2, FALSE);
789 /* adjust ${"::_<newfilename"} to store the new file name */
790 GvSV(gv2) = newSVpvn(tmpbuf2 + 2, tmplen2 - 2);
791 GvHV(gv2) = (HV*)SvREFCNT_inc(GvHV(*gvp));
792 GvAV(gv2) = (AV*)SvREFCNT_inc(GvAV(*gvp));
795 if (tmpbuf != smallbuf) Safefree(tmpbuf);
796 if (tmpbuf2 != smallbuf2) Safefree(tmpbuf2);
799 CopFILE_free(PL_curcop);
800 CopFILE_set(PL_curcop, s);
803 CopLINE_set(PL_curcop, atoi(n)-1);
807 /* skip space before PL_thistoken */
810 S_skipspace0(pTHX_ register char *s)
817 PL_thiswhite = newSVpvs("");
818 sv_catsv(PL_thiswhite, PL_skipwhite);
819 sv_free(PL_skipwhite);
822 PL_realtokenstart = s - SvPVX(PL_linestr);
826 /* skip space after PL_thistoken */
829 S_skipspace1(pTHX_ register char *s)
831 const char *start = s;
832 I32 startoff = start - SvPVX(PL_linestr);
837 start = SvPVX(PL_linestr) + startoff;
838 if (!PL_thistoken && PL_realtokenstart >= 0) {
839 const char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
840 PL_thistoken = newSVpvn(tstart, start - tstart);
842 PL_realtokenstart = -1;
845 PL_nextwhite = newSVpvs("");
846 sv_catsv(PL_nextwhite, PL_skipwhite);
847 sv_free(PL_skipwhite);
854 S_skipspace2(pTHX_ register char *s, SV **svp)
857 const I32 bufptroff = PL_bufptr - SvPVX(PL_linestr);
858 const I32 startoff = s - SvPVX(PL_linestr);
861 PL_bufptr = SvPVX(PL_linestr) + bufptroff;
862 if (!PL_madskills || !svp)
864 start = SvPVX(PL_linestr) + startoff;
865 if (!PL_thistoken && PL_realtokenstart >= 0) {
866 char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
867 PL_thistoken = newSVpvn(tstart, start - tstart);
868 PL_realtokenstart = -1;
873 sv_setsv(*svp, PL_skipwhite);
874 sv_free(PL_skipwhite);
883 S_update_debugger_info_pv(pTHX_ const char *buf, STRLEN len)
885 AV *av = CopFILEAVx(PL_curcop);
887 SV * const sv = newSV(0);
888 sv_upgrade(sv, SVt_PVMG);
889 sv_setpvn(sv, buf, len);
892 av_store(av, (I32)CopLINE(PL_curcop), sv);
897 S_update_debugger_info_sv(pTHX_ SV *orig_sv)
899 AV *av = CopFILEAVx(PL_curcop);
901 SV * const sv = newSV(0);
902 sv_upgrade(sv, SVt_PVMG);
903 sv_setsv(sv, orig_sv);
906 av_store(av, (I32)CopLINE(PL_curcop), sv);
912 * Called to gobble the appropriate amount and type of whitespace.
913 * Skips comments as well.
917 S_skipspace(pTHX_ register char *s)
922 int startoff = s - SvPVX(PL_linestr);
925 sv_free(PL_skipwhite);
930 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
931 while (s < PL_bufend && SPACE_OR_TAB(*s))
941 SSize_t oldprevlen, oldoldprevlen;
942 SSize_t oldloplen = 0, oldunilen = 0;
943 while (s < PL_bufend && isSPACE(*s)) {
944 if (*s++ == '\n' && PL_in_eval && !PL_rsfp)
949 if (s < PL_bufend && *s == '#') {
950 while (s < PL_bufend && *s != '\n')
954 if (PL_in_eval && !PL_rsfp) {
961 /* only continue to recharge the buffer if we're at the end
962 * of the buffer, we're not reading from a source filter, and
963 * we're in normal lexing mode
965 if (s < PL_bufend || !PL_rsfp || PL_sublex_info.sub_inwhat ||
966 PL_lex_state == LEX_FORMLINE)
973 /* try to recharge the buffer */
975 curoff = s - SvPVX(PL_linestr);
978 if ((s = filter_gets(PL_linestr, PL_rsfp,
979 (prevlen = SvCUR(PL_linestr)))) == NULL)
982 if (PL_madskills && curoff != startoff) {
984 PL_skipwhite = newSVpvs("");
985 sv_catpvn(PL_skipwhite, SvPVX(PL_linestr) + startoff,
989 /* mustn't throw out old stuff yet if madpropping */
990 SvCUR(PL_linestr) = curoff;
991 s = SvPVX(PL_linestr) + curoff;
993 if (curoff && s[-1] == '\n')
997 /* end of file. Add on the -p or -n magic */
998 /* XXX these shouldn't really be added here, can't set PL_faketokens */
1001 sv_catpv(PL_linestr,
1002 ";}continue{print or die qq(-p destination: $!\\n);}");
1004 sv_setpv(PL_linestr,
1005 ";}continue{print or die qq(-p destination: $!\\n);}");
1007 PL_minus_n = PL_minus_p = 0;
1009 else if (PL_minus_n) {
1011 sv_catpvn(PL_linestr, ";}", 2);
1013 sv_setpvn(PL_linestr, ";}", 2);
1019 sv_catpvn(PL_linestr,";", 1);
1021 sv_setpvn(PL_linestr,";", 1);
1024 /* reset variables for next time we lex */
1025 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = s = PL_linestart
1031 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
1032 PL_last_lop = PL_last_uni = NULL;
1034 /* Close the filehandle. Could be from -P preprocessor,
1035 * STDIN, or a regular file. If we were reading code from
1036 * STDIN (because the commandline held no -e or filename)
1037 * then we don't close it, we reset it so the code can
1038 * read from STDIN too.
1041 if (PL_preprocess && !PL_in_eval)
1042 (void)PerlProc_pclose(PL_rsfp);
1043 else if ((PerlIO*)PL_rsfp == PerlIO_stdin())
1044 PerlIO_clearerr(PL_rsfp);
1046 (void)PerlIO_close(PL_rsfp);
1051 /* not at end of file, so we only read another line */
1052 /* make corresponding updates to old pointers, for yyerror() */
1053 oldprevlen = PL_oldbufptr - PL_bufend;
1054 oldoldprevlen = PL_oldoldbufptr - PL_bufend;
1056 oldunilen = PL_last_uni - PL_bufend;
1058 oldloplen = PL_last_lop - PL_bufend;
1059 PL_linestart = PL_bufptr = s + prevlen;
1060 PL_bufend = s + SvCUR(PL_linestr);
1062 PL_oldbufptr = s + oldprevlen;
1063 PL_oldoldbufptr = s + oldoldprevlen;
1065 PL_last_uni = s + oldunilen;
1067 PL_last_lop = s + oldloplen;
1070 /* debugger active and we're not compiling the debugger code,
1071 * so store the line into the debugger's array of lines
1073 if (PERLDB_LINE && PL_curstash != PL_debstash)
1074 update_debugger_info_pv(PL_bufptr, PL_bufend - PL_bufptr);
1081 PL_skipwhite = newSVpvs("");
1082 curoff = s - SvPVX(PL_linestr);
1083 if (curoff - startoff)
1084 sv_catpvn(PL_skipwhite, SvPVX(PL_linestr) + startoff,
1093 * Check the unary operators to ensure there's no ambiguity in how they're
1094 * used. An ambiguous piece of code would be:
1096 * This doesn't mean rand() + 5. Because rand() is a unary operator,
1097 * the +5 is its argument.
1107 if (PL_oldoldbufptr != PL_last_uni)
1109 while (isSPACE(*PL_last_uni))
1112 while (isALNUM_lazy_if(s,UTF) || *s == '-')
1114 if ((t = strchr(s, '(')) && t < PL_bufptr)
1117 if (ckWARN_d(WARN_AMBIGUOUS)){
1118 Perl_warner(aTHX_ packWARN(WARN_AMBIGUOUS),
1119 "Warning: Use of \"%.*s\" without parentheses is ambiguous",
1120 (int)(s - PL_last_uni), PL_last_uni);
1125 * LOP : macro to build a list operator. Its behaviour has been replaced
1126 * with a subroutine, S_lop() for which LOP is just another name.
1129 #define LOP(f,x) return lop(f,x,s)
1133 * Build a list operator (or something that might be one). The rules:
1134 * - if we have a next token, then it's a list operator [why?]
1135 * - if the next thing is an opening paren, then it's a function
1136 * - else it's a list operator
1140 S_lop(pTHX_ I32 f, int x, char *s)
1147 PL_last_lop = PL_oldbufptr;
1148 PL_last_lop_op = (OPCODE)f;
1151 return REPORT(LSTOP);
1154 return REPORT(LSTOP);
1157 return REPORT(FUNC);
1160 return REPORT(FUNC);
1162 return REPORT(LSTOP);
1168 * Sets up for an eventual force_next(). start_force(0) basically does
1169 * an unshift, while start_force(-1) does a push. yylex removes items
1174 S_start_force(pTHX_ int where)
1178 if (where < 0) /* so people can duplicate start_force(PL_curforce) */
1179 where = PL_lasttoke;
1180 assert(PL_curforce < 0 || PL_curforce == where);
1181 if (PL_curforce != where) {
1182 for (i = PL_lasttoke; i > where; --i) {
1183 PL_nexttoke[i] = PL_nexttoke[i-1];
1187 if (PL_curforce < 0) /* in case of duplicate start_force() */
1188 Zero(&PL_nexttoke[where], 1, NEXTTOKE);
1189 PL_curforce = where;
1192 curmad('^', newSVpvs(""));
1193 CURMAD('_', PL_nextwhite);
1198 S_curmad(pTHX_ char slot, SV *sv)
1204 if (PL_curforce < 0)
1205 where = &PL_thismad;
1207 where = &PL_nexttoke[PL_curforce].next_mad;
1210 sv_setpvn(sv, "", 0);
1213 if (UTF && is_utf8_string((U8*)SvPVX(sv), SvCUR(sv)))
1215 else if (PL_encoding) {
1216 sv_recode_to_utf8(sv, PL_encoding);
1221 /* keep a slot open for the head of the list? */
1222 if (slot != '_' && *where && (*where)->mad_key == '^') {
1223 (*where)->mad_key = slot;
1224 sv_free((*where)->mad_val);
1225 (*where)->mad_val = (void*)sv;
1228 addmad(newMADsv(slot, sv), where, 0);
1231 # define start_force(where) NOOP
1232 # define curmad(slot, sv) NOOP
1237 * When the lexer realizes it knows the next token (for instance,
1238 * it is reordering tokens for the parser) then it can call S_force_next
1239 * to know what token to return the next time the lexer is called. Caller
1240 * will need to set PL_nextval[] (or PL_nexttoke[].next_val with PERL_MAD),
1241 * and possibly PL_expect to ensure the lexer handles the token correctly.
1245 S_force_next(pTHX_ I32 type)
1249 if (PL_curforce < 0)
1250 start_force(PL_lasttoke);
1251 PL_nexttoke[PL_curforce].next_type = type;
1252 if (PL_lex_state != LEX_KNOWNEXT)
1253 PL_lex_defer = PL_lex_state;
1254 PL_lex_state = LEX_KNOWNEXT;
1255 PL_lex_expect = PL_expect;
1258 PL_nexttype[PL_nexttoke] = type;
1260 if (PL_lex_state != LEX_KNOWNEXT) {
1261 PL_lex_defer = PL_lex_state;
1262 PL_lex_expect = PL_expect;
1263 PL_lex_state = LEX_KNOWNEXT;
1269 S_newSV_maybe_utf8(pTHX_ const char *start, STRLEN len)
1272 SV * const sv = newSVpvn(start,len);
1273 if (UTF && !IN_BYTES && is_utf8_string((const U8*)start, len))
1280 * When the lexer knows the next thing is a word (for instance, it has
1281 * just seen -> and it knows that the next char is a word char, then
1282 * it calls S_force_word to stick the next word into the PL_next lookahead.
1285 * char *start : buffer position (must be within PL_linestr)
1286 * int token : PL_next will be this type of bare word (e.g., METHOD,WORD)
1287 * int check_keyword : if true, Perl checks to make sure the word isn't
1288 * a keyword (do this if the word is a label, e.g. goto FOO)
1289 * int allow_pack : if true, : characters will also be allowed (require,
1290 * use, etc. do this)
1291 * int allow_initial_tick : used by the "sub" lexer only.
1295 S_force_word(pTHX_ register char *start, int token, int check_keyword, int allow_pack, int allow_initial_tick)
1301 start = SKIPSPACE1(start);
1303 if (isIDFIRST_lazy_if(s,UTF) ||
1304 (allow_pack && *s == ':') ||
1305 (allow_initial_tick && *s == '\'') )
1307 s = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, allow_pack, &len);
1308 if (check_keyword && keyword(PL_tokenbuf, len, 0))
1310 start_force(PL_curforce);
1312 curmad('X', newSVpvn(start,s-start));
1313 if (token == METHOD) {
1318 PL_expect = XOPERATOR;
1321 NEXTVAL_NEXTTOKE.opval
1322 = (OP*)newSVOP(OP_CONST,0,
1323 S_newSV_maybe_utf8(aTHX_ PL_tokenbuf, len));
1324 NEXTVAL_NEXTTOKE.opval->op_private |= OPpCONST_BARE;
1332 * Called when the lexer wants $foo *foo &foo etc, but the program
1333 * text only contains the "foo" portion. The first argument is a pointer
1334 * to the "foo", and the second argument is the type symbol to prefix.
1335 * Forces the next token to be a "WORD".
1336 * Creates the symbol if it didn't already exist (via gv_fetchpv()).
1340 S_force_ident(pTHX_ register const char *s, int kind)
1344 const STRLEN len = strlen(s);
1345 OP* const o = (OP*)newSVOP(OP_CONST, 0, newSVpvn(s, len));
1346 start_force(PL_curforce);
1347 NEXTVAL_NEXTTOKE.opval = o;
1350 o->op_private = OPpCONST_ENTERED;
1351 /* XXX see note in pp_entereval() for why we forgo typo
1352 warnings if the symbol must be introduced in an eval.
1354 gv_fetchpvn_flags(s, len,
1355 PL_in_eval ? (GV_ADDMULTI | GV_ADDINEVAL)
1357 kind == '$' ? SVt_PV :
1358 kind == '@' ? SVt_PVAV :
1359 kind == '%' ? SVt_PVHV :
1367 Perl_str_to_version(pTHX_ SV *sv)
1372 const char *start = SvPV_const(sv,len);
1373 const char * const end = start + len;
1374 const bool utf = SvUTF8(sv) ? TRUE : FALSE;
1375 while (start < end) {
1379 n = utf8n_to_uvchr((U8*)start, len, &skip, 0);
1384 retval += ((NV)n)/nshift;
1393 * Forces the next token to be a version number.
1394 * If the next token appears to be an invalid version number, (e.g. "v2b"),
1395 * and if "guessing" is TRUE, then no new token is created (and the caller
1396 * must use an alternative parsing method).
1400 S_force_version(pTHX_ char *s, int guessing)
1406 I32 startoff = s - SvPVX(PL_linestr);
1415 while (isDIGIT(*d) || *d == '_' || *d == '.')
1419 start_force(PL_curforce);
1420 curmad('X', newSVpvn(s,d-s));
1423 if (*d == ';' || isSPACE(*d) || *d == '}' || !*d) {
1425 s = scan_num(s, &yylval);
1426 version = yylval.opval;
1427 ver = cSVOPx(version)->op_sv;
1428 if (SvPOK(ver) && !SvNIOK(ver)) {
1429 SvUPGRADE(ver, SVt_PVNV);
1430 SvNV_set(ver, str_to_version(ver));
1431 SvNOK_on(ver); /* hint that it is a version */
1434 else if (guessing) {
1437 sv_free(PL_nextwhite); /* let next token collect whitespace */
1439 s = SvPVX(PL_linestr) + startoff;
1447 if (PL_madskills && !version) {
1448 sv_free(PL_nextwhite); /* let next token collect whitespace */
1450 s = SvPVX(PL_linestr) + startoff;
1453 /* NOTE: The parser sees the package name and the VERSION swapped */
1454 start_force(PL_curforce);
1455 NEXTVAL_NEXTTOKE.opval = version;
1463 * Tokenize a quoted string passed in as an SV. It finds the next
1464 * chunk, up to end of string or a backslash. It may make a new
1465 * SV containing that chunk (if HINT_NEW_STRING is on). It also
1470 S_tokeq(pTHX_ SV *sv)
1474 register char *send;
1482 s = SvPV_force(sv, len);
1483 if (SvTYPE(sv) >= SVt_PVIV && SvIVX(sv) == -1)
1486 while (s < send && *s != '\\')
1491 if ( PL_hints & HINT_NEW_STRING ) {
1492 pv = sv_2mortal(newSVpvn(SvPVX_const(pv), len));
1498 if (s + 1 < send && (s[1] == '\\'))
1499 s++; /* all that, just for this */
1504 SvCUR_set(sv, d - SvPVX_const(sv));
1506 if ( PL_hints & HINT_NEW_STRING )
1507 return new_constant(NULL, 0, "q", sv, pv, "q");
1512 * Now come three functions related to double-quote context,
1513 * S_sublex_start, S_sublex_push, and S_sublex_done. They're used when
1514 * converting things like "\u\Lgnat" into ucfirst(lc("gnat")). They
1515 * interact with PL_lex_state, and create fake ( ... ) argument lists
1516 * to handle functions and concatenation.
1517 * They assume that whoever calls them will be setting up a fake
1518 * join call, because each subthing puts a ',' after it. This lets
1521 * join($, , 'lower ', lcfirst( 'uPpEr', ) ,)
1523 * (I'm not sure whether the spurious commas at the end of lcfirst's
1524 * arguments and join's arguments are created or not).
1529 * Assumes that yylval.ival is the op we're creating (e.g. OP_LCFIRST).
1531 * Pattern matching will set PL_lex_op to the pattern-matching op to
1532 * make (we return THING if yylval.ival is OP_NULL, PMFUNC otherwise).
1534 * OP_CONST and OP_READLINE are easy--just make the new op and return.
1536 * Everything else becomes a FUNC.
1538 * Sets PL_lex_state to LEX_INTERPPUSH unless (ival was OP_NULL or we
1539 * had an OP_CONST or OP_READLINE). This just sets us up for a
1540 * call to S_sublex_push().
1544 S_sublex_start(pTHX)
1547 register const I32 op_type = yylval.ival;
1549 if (op_type == OP_NULL) {
1550 yylval.opval = PL_lex_op;
1554 if (op_type == OP_CONST || op_type == OP_READLINE) {
1555 SV *sv = tokeq(PL_lex_stuff);
1557 if (SvTYPE(sv) == SVt_PVIV) {
1558 /* Overloaded constants, nothing fancy: Convert to SVt_PV: */
1560 const char * const p = SvPV_const(sv, len);
1561 SV * const nsv = newSVpvn(p, len);
1567 yylval.opval = (OP*)newSVOP(op_type, 0, sv);
1568 PL_lex_stuff = NULL;
1569 /* Allow <FH> // "foo" */
1570 if (op_type == OP_READLINE)
1571 PL_expect = XTERMORDORDOR;
1574 else if (op_type == OP_BACKTICK && PL_lex_op) {
1575 /* readpipe() vas overriden */
1576 cSVOPx(cLISTOPx(cUNOPx(PL_lex_op)->op_first)->op_first->op_sibling)->op_sv = tokeq(PL_lex_stuff);
1577 yylval.opval = PL_lex_op;
1579 PL_lex_stuff = NULL;
1583 PL_sublex_info.super_state = PL_lex_state;
1584 PL_sublex_info.sub_inwhat = op_type;
1585 PL_sublex_info.sub_op = PL_lex_op;
1586 PL_lex_state = LEX_INTERPPUSH;
1590 yylval.opval = PL_lex_op;
1600 * Create a new scope to save the lexing state. The scope will be
1601 * ended in S_sublex_done. Returns a '(', starting the function arguments
1602 * to the uc, lc, etc. found before.
1603 * Sets PL_lex_state to LEX_INTERPCONCAT.
1612 PL_lex_state = PL_sublex_info.super_state;
1613 SAVEI32(PL_lex_dojoin);
1614 SAVEI32(PL_lex_brackets);
1615 SAVEI32(PL_lex_casemods);
1616 SAVEI32(PL_lex_starts);
1617 SAVEI32(PL_lex_state);
1618 SAVEVPTR(PL_lex_inpat);
1619 SAVEI32(PL_lex_inwhat);
1620 SAVECOPLINE(PL_curcop);
1621 SAVEPPTR(PL_bufptr);
1622 SAVEPPTR(PL_bufend);
1623 SAVEPPTR(PL_oldbufptr);
1624 SAVEPPTR(PL_oldoldbufptr);
1625 SAVEPPTR(PL_last_lop);
1626 SAVEPPTR(PL_last_uni);
1627 SAVEPPTR(PL_linestart);
1628 SAVESPTR(PL_linestr);
1629 SAVEGENERICPV(PL_lex_brackstack);
1630 SAVEGENERICPV(PL_lex_casestack);
1632 PL_linestr = PL_lex_stuff;
1633 PL_lex_stuff = NULL;
1635 PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart
1636 = SvPVX(PL_linestr);
1637 PL_bufend += SvCUR(PL_linestr);
1638 PL_last_lop = PL_last_uni = NULL;
1639 SAVEFREESV(PL_linestr);
1641 PL_lex_dojoin = FALSE;
1642 PL_lex_brackets = 0;
1643 Newx(PL_lex_brackstack, 120, char);
1644 Newx(PL_lex_casestack, 12, char);
1645 PL_lex_casemods = 0;
1646 *PL_lex_casestack = '\0';
1648 PL_lex_state = LEX_INTERPCONCAT;
1649 CopLINE_set(PL_curcop, (line_t)PL_multi_start);
1651 PL_lex_inwhat = PL_sublex_info.sub_inwhat;
1652 if (PL_lex_inwhat == OP_MATCH || PL_lex_inwhat == OP_QR || PL_lex_inwhat == OP_SUBST)
1653 PL_lex_inpat = PL_sublex_info.sub_op;
1655 PL_lex_inpat = NULL;
1662 * Restores lexer state after a S_sublex_push.
1669 if (!PL_lex_starts++) {
1670 SV * const sv = newSVpvs("");
1671 if (SvUTF8(PL_linestr))
1673 PL_expect = XOPERATOR;
1674 yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
1678 if (PL_lex_casemods) { /* oops, we've got some unbalanced parens */
1679 PL_lex_state = LEX_INTERPCASEMOD;
1683 /* Is there a right-hand side to take care of? (s//RHS/ or tr//RHS/) */
1684 if (PL_lex_repl && (PL_lex_inwhat == OP_SUBST || PL_lex_inwhat == OP_TRANS)) {
1685 PL_linestr = PL_lex_repl;
1687 PL_bufend = PL_bufptr = PL_oldbufptr = PL_oldoldbufptr = PL_linestart = SvPVX(PL_linestr);
1688 PL_bufend += SvCUR(PL_linestr);
1689 PL_last_lop = PL_last_uni = NULL;
1690 SAVEFREESV(PL_linestr);
1691 PL_lex_dojoin = FALSE;
1692 PL_lex_brackets = 0;
1693 PL_lex_casemods = 0;
1694 *PL_lex_casestack = '\0';
1696 if (SvEVALED(PL_lex_repl)) {
1697 PL_lex_state = LEX_INTERPNORMAL;
1699 /* we don't clear PL_lex_repl here, so that we can check later
1700 whether this is an evalled subst; that means we rely on the
1701 logic to ensure sublex_done() is called again only via the
1702 branch (in yylex()) that clears PL_lex_repl, else we'll loop */
1705 PL_lex_state = LEX_INTERPCONCAT;
1715 PL_endwhite = newSVpvs("");
1716 sv_catsv(PL_endwhite, PL_thiswhite);
1720 sv_setpvn(PL_thistoken,"",0);
1722 PL_realtokenstart = -1;
1726 PL_bufend = SvPVX(PL_linestr);
1727 PL_bufend += SvCUR(PL_linestr);
1728 PL_expect = XOPERATOR;
1729 PL_sublex_info.sub_inwhat = 0;
1737 Extracts a pattern, double-quoted string, or transliteration. This
1740 It looks at PL_lex_inwhat and PL_lex_inpat to find out whether it's
1741 processing a pattern (PL_lex_inpat is true), a transliteration
1742 (PL_lex_inwhat == OP_TRANS is true), or a double-quoted string.
1744 Returns a pointer to the character scanned up to. If this is
1745 advanced from the start pointer supplied (i.e. if anything was
1746 successfully parsed), will leave an OP for the substring scanned
1747 in yylval. Caller must intuit reason for not parsing further
1748 by looking at the next characters herself.
1752 double-quoted style: \r and \n
1753 regexp special ones: \D \s
1756 case and quoting: \U \Q \E
1757 stops on @ and $, but not for $ as tail anchor
1759 In transliterations:
1760 characters are VERY literal, except for - not at the start or end
1761 of the string, which indicates a range. If the range is in bytes,
1762 scan_const expands the range to the full set of intermediate
1763 characters. If the range is in utf8, the hyphen is replaced with
1764 a certain range mark which will be handled by pmtrans() in op.c.
1766 In double-quoted strings:
1768 double-quoted style: \r and \n
1770 deprecated backrefs: \1 (in substitution replacements)
1771 case and quoting: \U \Q \E
1774 scan_const does *not* construct ops to handle interpolated strings.
1775 It stops processing as soon as it finds an embedded $ or @ variable
1776 and leaves it to the caller to work out what's going on.
1778 embedded arrays (whether in pattern or not) could be:
1779 @foo, @::foo, @'foo, @{foo}, @$foo, @+, @-.
1781 $ in double-quoted strings must be the symbol of an embedded scalar.
1783 $ in pattern could be $foo or could be tail anchor. Assumption:
1784 it's a tail anchor if $ is the last thing in the string, or if it's
1785 followed by one of "()| \r\n\t"
1787 \1 (backreferences) are turned into $1
1789 The structure of the code is
1790 while (there's a character to process) {
1791 handle transliteration ranges
1792 skip regexp comments /(?#comment)/ and codes /(?{code})/
1793 skip #-initiated comments in //x patterns
1794 check for embedded arrays
1795 check for embedded scalars
1797 leave intact backslashes from leaveit (below)
1798 deprecate \1 in substitution replacements
1799 handle string-changing backslashes \l \U \Q \E, etc.
1800 switch (what was escaped) {
1801 handle \- in a transliteration (becomes a literal -)
1802 handle \132 (octal characters)
1803 handle \x15 and \x{1234} (hex characters)
1804 handle \N{name} (named characters)
1805 handle \cV (control characters)
1806 handle printf-style backslashes (\f, \r, \n, etc)
1808 } (end if backslash)
1809 } (end while character to read)
1814 S_scan_const(pTHX_ char *start)
1817 register char *send = PL_bufend; /* end of the constant */
1818 SV *sv = newSV(send - start); /* sv for the constant */
1819 register char *s = start; /* start of the constant */
1820 register char *d = SvPVX(sv); /* destination for copies */
1821 bool dorange = FALSE; /* are we in a translit range? */
1822 bool didrange = FALSE; /* did we just finish a range? */
1823 I32 has_utf8 = FALSE; /* Output constant is UTF8 */
1824 I32 this_utf8 = UTF; /* The source string is assumed to be UTF8 */
1827 UV literal_endpoint = 0;
1828 bool native_range = TRUE; /* turned to FALSE if the first endpoint is Unicode. */
1831 if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
1832 /* If we are doing a trans and we know we want UTF8 set expectation */
1833 has_utf8 = PL_sublex_info.sub_op->op_private & (OPpTRANS_FROM_UTF|OPpTRANS_TO_UTF);
1834 this_utf8 = PL_sublex_info.sub_op->op_private & (PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
1838 while (s < send || dorange) {
1839 /* get transliterations out of the way (they're most literal) */
1840 if (PL_lex_inwhat == OP_TRANS) {
1841 /* expand a range A-Z to the full set of characters. AIE! */
1843 I32 i; /* current expanded character */
1844 I32 min; /* first character in range */
1845 I32 max; /* last character in range */
1856 char * const c = (char*)utf8_hop((U8*)d, -1);
1860 *c = (char)UTF_TO_NATIVE(0xff);
1861 /* mark the range as done, and continue */
1867 i = d - SvPVX_const(sv); /* remember current offset */
1870 SvLEN(sv) + (has_utf8 ?
1871 (512 - UTF_CONTINUATION_MARK +
1874 /* How many two-byte within 0..255: 128 in UTF-8,
1875 * 96 in UTF-8-mod. */
1877 SvGROW(sv, SvLEN(sv) + 256); /* never more than 256 chars in a range */
1879 d = SvPVX(sv) + i; /* refresh d after realloc */
1883 for (j = 0; j <= 1; j++) {
1884 char * const c = (char*)utf8_hop((U8*)d, -1);
1885 const UV uv = utf8n_to_uvchr((U8*)c, d - c, NULL, 0);
1891 max = (U8)0xff; /* only to \xff */
1892 uvmax = uv; /* \x{100} to uvmax */
1894 d = c; /* eat endpoint chars */
1899 d -= 2; /* eat the first char and the - */
1900 min = (U8)*d; /* first char in range */
1901 max = (U8)d[1]; /* last char in range */
1908 "Invalid range \"%c-%c\" in transliteration operator",
1909 (char)min, (char)max);
1913 if (literal_endpoint == 2 &&
1914 ((isLOWER(min) && isLOWER(max)) ||
1915 (isUPPER(min) && isUPPER(max)))) {
1917 for (i = min; i <= max; i++)
1919 *d++ = NATIVE_TO_NEED(has_utf8,i);
1921 for (i = min; i <= max; i++)
1923 *d++ = NATIVE_TO_NEED(has_utf8,i);
1928 for (i = min; i <= max; i++)
1931 const U8 ch = (U8)NATIVE_TO_UTF(i);
1932 if (UNI_IS_INVARIANT(ch))
1935 *d++ = (U8)UTF8_EIGHT_BIT_HI(ch);
1936 *d++ = (U8)UTF8_EIGHT_BIT_LO(ch);
1945 d = (char*)uvchr_to_utf8((U8*)d, 0x100);
1947 *d++ = (char)UTF_TO_NATIVE(0xff);
1949 d = (char*)uvchr_to_utf8((U8*)d, uvmax);
1953 /* mark the range as done, and continue */
1957 literal_endpoint = 0;
1962 /* range begins (ignore - as first or last char) */
1963 else if (*s == '-' && s+1 < send && s != start) {
1965 Perl_croak(aTHX_ "Ambiguous range in transliteration operator");
1972 *d++ = (char)UTF_TO_NATIVE(0xff); /* use illegal utf8 byte--see pmtrans */
1982 literal_endpoint = 0;
1983 native_range = TRUE;
1988 /* if we get here, we're not doing a transliteration */
1990 /* skip for regexp comments /(?#comment)/ and code /(?{code})/,
1991 except for the last char, which will be done separately. */
1992 else if (*s == '(' && PL_lex_inpat && s[1] == '?') {
1994 while (s+1 < send && *s != ')')
1995 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
1997 else if (s[2] == '{' /* This should match regcomp.c */
1998 || ((s[2] == 'p' || s[2] == '?') && s[3] == '{'))
2001 char *regparse = s + (s[2] == '{' ? 3 : 4);
2004 while (count && (c = *regparse)) {
2005 if (c == '\\' && regparse[1])
2013 if (*regparse != ')')
2014 regparse--; /* Leave one char for continuation. */
2015 while (s < regparse)
2016 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2020 /* likewise skip #-initiated comments in //x patterns */
2021 else if (*s == '#' && PL_lex_inpat &&
2022 ((PMOP*)PL_lex_inpat)->op_pmflags & PMf_EXTENDED) {
2023 while (s+1 < send && *s != '\n')
2024 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2027 /* check for embedded arrays
2028 (@foo, @::foo, @'foo, @{foo}, @$foo, @+, @-)
2030 else if (*s == '@' && s[1]) {
2031 if (isALNUM_lazy_if(s+1,UTF))
2033 if (strchr(":'{$", s[1]))
2035 if (!PL_lex_inpat && (s[1] == '+' || s[1] == '-'))
2036 break; /* in regexp, neither @+ nor @- are interpolated */
2039 /* check for embedded scalars. only stop if we're sure it's a
2042 else if (*s == '$') {
2043 if (!PL_lex_inpat) /* not a regexp, so $ must be var */
2045 if (s + 1 < send && !strchr("()| \r\n\t", s[1]))
2046 break; /* in regexp, $ might be tail anchor */
2049 /* End of else if chain - OP_TRANS rejoin rest */
2052 if (*s == '\\' && s+1 < send) {
2055 /* deprecate \1 in strings and substitution replacements */
2056 if (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat &&
2057 isDIGIT(*s) && *s != '0' && !isDIGIT(s[1]))
2059 if (ckWARN(WARN_SYNTAX))
2060 Perl_warner(aTHX_ packWARN(WARN_SYNTAX), "\\%c better written as $%c", *s, *s);
2065 /* string-change backslash escapes */
2066 if (PL_lex_inwhat != OP_TRANS && *s && strchr("lLuUEQ", *s)) {
2070 /* skip any other backslash escapes in a pattern */
2071 else if (PL_lex_inpat) {
2072 *d++ = NATIVE_TO_NEED(has_utf8,'\\');
2073 goto default_action;
2076 /* if we get here, it's either a quoted -, or a digit */
2079 /* quoted - in transliterations */
2081 if (PL_lex_inwhat == OP_TRANS) {
2088 if ((isALPHA(*s) || isDIGIT(*s)) &&
2090 Perl_warner(aTHX_ packWARN(WARN_MISC),
2091 "Unrecognized escape \\%c passed through",
2093 /* default action is to copy the quoted character */
2094 goto default_action;
2097 /* \132 indicates an octal constant */
2098 case '0': case '1': case '2': case '3':
2099 case '4': case '5': case '6': case '7':
2103 uv = grok_oct(s, &len, &flags, NULL);
2106 goto NUM_ESCAPE_INSERT;
2108 /* \x24 indicates a hex constant */
2112 char* const e = strchr(s, '}');
2113 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES |
2114 PERL_SCAN_DISALLOW_PREFIX;
2119 yyerror("Missing right brace on \\x{}");
2123 uv = grok_hex(s, &len, &flags, NULL);
2129 I32 flags = PERL_SCAN_DISALLOW_PREFIX;
2130 uv = grok_hex(s, &len, &flags, NULL);
2136 /* Insert oct or hex escaped character.
2137 * There will always enough room in sv since such
2138 * escapes will be longer than any UTF-8 sequence
2139 * they can end up as. */
2141 /* We need to map to chars to ASCII before doing the tests
2144 if (!UNI_IS_INVARIANT(NATIVE_TO_UNI(uv))) {
2145 if (!has_utf8 && uv > 255) {
2146 /* Might need to recode whatever we have
2147 * accumulated so far if it contains any
2150 * (Can't we keep track of that and avoid
2151 * this rescan? --jhi)
2155 for (c = (U8 *) SvPVX(sv); c < (U8 *)d; c++) {
2156 if (!NATIVE_IS_INVARIANT(*c)) {
2161 const STRLEN offset = d - SvPVX_const(sv);
2163 d = SvGROW(sv, SvLEN(sv) + hicount + 1) + offset;
2167 while (src >= (const U8 *)SvPVX_const(sv)) {
2168 if (!NATIVE_IS_INVARIANT(*src)) {
2169 const U8 ch = NATIVE_TO_ASCII(*src);
2170 *dst-- = (U8)UTF8_EIGHT_BIT_LO(ch);
2171 *dst-- = (U8)UTF8_EIGHT_BIT_HI(ch);
2181 if (has_utf8 || uv > 255) {
2182 d = (char*)uvchr_to_utf8((U8*)d, uv);
2184 if (PL_lex_inwhat == OP_TRANS &&
2185 PL_sublex_info.sub_op) {
2186 PL_sublex_info.sub_op->op_private |=
2187 (PL_lex_repl ? OPpTRANS_FROM_UTF
2191 if (uv > 255 && !dorange)
2192 native_range = FALSE;
2204 /* \N{LATIN SMALL LETTER A} is a named character */
2208 char* e = strchr(s, '}');
2215 yyerror("Missing right brace on \\N{}");
2219 if (e > s + 2 && s[1] == 'U' && s[2] == '+') {
2221 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES |
2222 PERL_SCAN_DISALLOW_PREFIX;
2225 uv = grok_hex(s, &len, &flags, NULL);
2226 if ( e > s && len != (STRLEN)(e - s) ) {
2230 goto NUM_ESCAPE_INSERT;
2232 res = newSVpvn(s + 1, e - s - 1);
2233 type = newSVpvn(s - 2,e - s + 3);
2234 res = new_constant( NULL, 0, "charnames",
2235 res, NULL, SvPVX(type) );
2238 sv_utf8_upgrade(res);
2239 str = SvPV_const(res,len);
2240 #ifdef EBCDIC_NEVER_MIND
2241 /* charnames uses pack U and that has been
2242 * recently changed to do the below uni->native
2243 * mapping, so this would be redundant (and wrong,
2244 * the code point would be doubly converted).
2245 * But leave this in just in case the pack U change
2246 * gets revoked, but the semantics is still
2247 * desireable for charnames. --jhi */
2249 UV uv = utf8_to_uvchr((const U8*)str, 0);
2252 U8 tmpbuf[UTF8_MAXBYTES+1], *d;
2254 d = uvchr_to_utf8(tmpbuf, UNI_TO_NATIVE(uv));
2255 sv_setpvn(res, (char *)tmpbuf, d - tmpbuf);
2256 str = SvPV_const(res, len);
2260 if (!has_utf8 && SvUTF8(res)) {
2261 const char * const ostart = SvPVX_const(sv);
2262 SvCUR_set(sv, d - ostart);
2265 sv_utf8_upgrade(sv);
2266 /* this just broke our allocation above... */
2267 SvGROW(sv, (STRLEN)(send - start));
2268 d = SvPVX(sv) + SvCUR(sv);
2271 if (len > (STRLEN)(e - s + 4)) { /* I _guess_ 4 is \N{} --jhi */
2272 const char * const odest = SvPVX_const(sv);
2274 SvGROW(sv, (SvLEN(sv) + len - (e - s + 4)));
2275 d = SvPVX(sv) + (d - odest);
2279 native_range = FALSE; /* \N{} is guessed to be Unicode */
2281 Copy(str, d, len, char);
2288 yyerror("Missing braces on \\N{}");
2291 /* \c is a control character */
2300 *d++ = NATIVE_TO_NEED(has_utf8,toCTRL(c));
2303 yyerror("Missing control char name in \\c");
2307 /* printf-style backslashes, formfeeds, newlines, etc */
2309 *d++ = NATIVE_TO_NEED(has_utf8,'\b');
2312 *d++ = NATIVE_TO_NEED(has_utf8,'\n');
2315 *d++ = NATIVE_TO_NEED(has_utf8,'\r');
2318 *d++ = NATIVE_TO_NEED(has_utf8,'\f');
2321 *d++ = NATIVE_TO_NEED(has_utf8,'\t');
2324 *d++ = ASCII_TO_NEED(has_utf8,'\033');
2327 *d++ = ASCII_TO_NEED(has_utf8,'\007');
2333 } /* end if (backslash) */
2340 /* If we started with encoded form, or already know we want it
2341 and then encode the next character */
2342 if ((has_utf8 || this_utf8) && !NATIVE_IS_INVARIANT((U8)(*s))) {
2344 const UV nextuv = (this_utf8) ? utf8n_to_uvchr((U8*)s, send - s, &len, 0) : (UV) ((U8) *s);
2345 const STRLEN need = UNISKIP(NATIVE_TO_UNI(nextuv));
2348 /* encoded value larger than old, need extra space (NOTE: SvCUR() not set here) */
2349 const STRLEN off = d - SvPVX_const(sv);
2350 d = SvGROW(sv, SvLEN(sv) + (need-len)) + off;
2352 d = (char*)uvchr_to_utf8((U8*)d, nextuv);
2355 if (uv > 255 && !dorange)
2356 native_range = FALSE;
2360 *d++ = NATIVE_TO_NEED(has_utf8,*s++);
2362 } /* while loop to process each character */
2364 /* terminate the string and set up the sv */
2366 SvCUR_set(sv, d - SvPVX_const(sv));
2367 if (SvCUR(sv) >= SvLEN(sv))
2368 Perl_croak(aTHX_ "panic: constant overflowed allocated space");
2371 if (PL_encoding && !has_utf8) {
2372 sv_recode_to_utf8(sv, PL_encoding);
2378 if (PL_lex_inwhat == OP_TRANS && PL_sublex_info.sub_op) {
2379 PL_sublex_info.sub_op->op_private |=
2380 (PL_lex_repl ? OPpTRANS_FROM_UTF : OPpTRANS_TO_UTF);
2384 /* shrink the sv if we allocated more than we used */
2385 if (SvCUR(sv) + 5 < SvLEN(sv)) {
2386 SvPV_shrink_to_cur(sv);
2389 /* return the substring (via yylval) only if we parsed anything */
2390 if (s > PL_bufptr) {
2391 if ( PL_hints & ( PL_lex_inpat ? HINT_NEW_RE : HINT_NEW_STRING ) )
2392 sv = new_constant(start, s - start,
2393 (const char *)(PL_lex_inpat ? "qr" : "q"),
2396 (( PL_lex_inwhat == OP_TRANS
2398 : ( (PL_lex_inwhat == OP_SUBST && !PL_lex_inpat)
2401 yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
2408 * Returns TRUE if there's more to the expression (e.g., a subscript),
2411 * It deals with "$foo[3]" and /$foo[3]/ and /$foo[0123456789$]+/
2413 * ->[ and ->{ return TRUE
2414 * { and [ outside a pattern are always subscripts, so return TRUE
2415 * if we're outside a pattern and it's not { or [, then return FALSE
2416 * if we're in a pattern and the first char is a {
2417 * {4,5} (any digits around the comma) returns FALSE
2418 * if we're in a pattern and the first char is a [
2420 * [SOMETHING] has a funky algorithm to decide whether it's a
2421 * character class or not. It has to deal with things like
2422 * /$foo[-3]/ and /$foo[$bar]/ as well as /$foo[$\d]+/
2423 * anything else returns TRUE
2426 /* This is the one truly awful dwimmer necessary to conflate C and sed. */
2429 S_intuit_more(pTHX_ register char *s)
2432 if (PL_lex_brackets)
2434 if (*s == '-' && s[1] == '>' && (s[2] == '[' || s[2] == '{'))
2436 if (*s != '{' && *s != '[')
2441 /* In a pattern, so maybe we have {n,m}. */
2458 /* On the other hand, maybe we have a character class */
2461 if (*s == ']' || *s == '^')
2464 /* this is terrifying, and it works */
2465 int weight = 2; /* let's weigh the evidence */
2467 unsigned char un_char = 255, last_un_char;
2468 const char * const send = strchr(s,']');
2469 char tmpbuf[sizeof PL_tokenbuf * 4];
2471 if (!send) /* has to be an expression */
2474 Zero(seen,256,char);
2477 else if (isDIGIT(*s)) {
2479 if (isDIGIT(s[1]) && s[2] == ']')
2485 for (; s < send; s++) {
2486 last_un_char = un_char;
2487 un_char = (unsigned char)*s;
2492 weight -= seen[un_char] * 10;
2493 if (isALNUM_lazy_if(s+1,UTF)) {
2495 scan_ident(s, send, tmpbuf, sizeof tmpbuf, FALSE);
2496 len = (int)strlen(tmpbuf);
2497 if (len > 1 && gv_fetchpvn_flags(tmpbuf, len, 0, SVt_PV))
2502 else if (*s == '$' && s[1] &&
2503 strchr("[#!%*<>()-=",s[1])) {
2504 if (/*{*/ strchr("])} =",s[2]))
2513 if (strchr("wds]",s[1]))
2515 else if (seen[(U8)'\''] || seen[(U8)'"'])
2517 else if (strchr("rnftbxcav",s[1]))
2519 else if (isDIGIT(s[1])) {
2521 while (s[1] && isDIGIT(s[1]))
2531 if (strchr("aA01! ",last_un_char))
2533 if (strchr("zZ79~",s[1]))
2535 if (last_un_char == 255 && (isDIGIT(s[1]) || s[1] == '$'))
2536 weight -= 5; /* cope with negative subscript */
2539 if (!isALNUM(last_un_char)
2540 && !(last_un_char == '$' || last_un_char == '@'
2541 || last_un_char == '&')
2542 && isALPHA(*s) && s[1] && isALPHA(s[1])) {
2547 if (keyword(tmpbuf, d - tmpbuf, 0))
2550 if (un_char == last_un_char + 1)
2552 weight -= seen[un_char];
2557 if (weight >= 0) /* probably a character class */
2567 * Does all the checking to disambiguate
2569 * between foo(bar) and bar->foo. Returns 0 if not a method, otherwise
2570 * FUNCMETH (bar->foo(args)) or METHOD (bar->foo args).
2572 * First argument is the stuff after the first token, e.g. "bar".
2574 * Not a method if bar is a filehandle.
2575 * Not a method if foo is a subroutine prototyped to take a filehandle.
2576 * Not a method if it's really "Foo $bar"
2577 * Method if it's "foo $bar"
2578 * Not a method if it's really "print foo $bar"
2579 * Method if it's really "foo package::" (interpreted as package->foo)
2580 * Not a method if bar is known to be a subroutine ("sub bar; foo bar")
2581 * Not a method if bar is a filehandle or package, but is quoted with
2586 S_intuit_method(pTHX_ char *start, GV *gv, CV *cv)
2589 char *s = start + (*start == '$');
2590 char tmpbuf[sizeof PL_tokenbuf];
2598 if (SvTYPE(gv) == SVt_PVGV && GvIO(gv))
2602 const char *proto = SvPVX_const(cv);
2613 s = scan_word(s, tmpbuf, sizeof tmpbuf, TRUE, &len);
2614 /* start is the beginning of the possible filehandle/object,
2615 * and s is the end of it
2616 * tmpbuf is a copy of it
2619 if (*start == '$') {
2620 if (gv || PL_last_lop_op == OP_PRINT || isUPPER(*PL_tokenbuf))
2623 len = start - SvPVX(PL_linestr);
2627 start = SvPVX(PL_linestr) + len;
2631 return *s == '(' ? FUNCMETH : METHOD;
2633 if (!keyword(tmpbuf, len, 0)) {
2634 if (len > 2 && tmpbuf[len - 2] == ':' && tmpbuf[len - 1] == ':') {
2638 soff = s - SvPVX(PL_linestr);
2642 indirgv = gv_fetchpvn_flags(tmpbuf, len, 0, SVt_PVCV);
2643 if (indirgv && GvCVu(indirgv))
2645 /* filehandle or package name makes it a method */
2646 if (!gv || GvIO(indirgv) || gv_stashpvn(tmpbuf, len, FALSE)) {
2648 soff = s - SvPVX(PL_linestr);
2651 if ((PL_bufend - s) >= 2 && *s == '=' && *(s+1) == '>')
2652 return 0; /* no assumptions -- "=>" quotes bearword */
2654 start_force(PL_curforce);
2655 NEXTVAL_NEXTTOKE.opval = (OP*)newSVOP(OP_CONST, 0,
2656 newSVpvn(tmpbuf,len));
2657 NEXTVAL_NEXTTOKE.opval->op_private = OPpCONST_BARE;
2659 curmad('X', newSVpvn(start,SvPVX(PL_linestr) + soff - start));
2664 PL_bufptr = SvPVX(PL_linestr) + soff; /* restart before space */
2666 return *s == '(' ? FUNCMETH : METHOD;
2674 * Return a string of Perl code to load the debugger. If PERL5DB
2675 * is set, it will return the contents of that, otherwise a
2676 * compile-time require of perl5db.pl.
2684 const char * const pdb = PerlEnv_getenv("PERL5DB");
2688 SETERRNO(0,SS_NORMAL);
2689 return "BEGIN { require 'perl5db.pl' }";
2695 /* Encoded script support. filter_add() effectively inserts a
2696 * 'pre-processing' function into the current source input stream.
2697 * Note that the filter function only applies to the current source file
2698 * (e.g., it will not affect files 'require'd or 'use'd by this one).
2700 * The datasv parameter (which may be NULL) can be used to pass
2701 * private data to this instance of the filter. The filter function
2702 * can recover the SV using the FILTER_DATA macro and use it to
2703 * store private buffers and state information.
2705 * The supplied datasv parameter is upgraded to a PVIO type
2706 * and the IoDIRP/IoANY field is used to store the function pointer,
2707 * and IOf_FAKE_DIRP is enabled on datasv to mark this as such.
2708 * Note that IoTOP_NAME, IoFMT_NAME, IoBOTTOM_NAME, if set for
2709 * private use must be set using malloc'd pointers.
2713 Perl_filter_add(pTHX_ filter_t funcp, SV *datasv)
2719 if (!PL_rsfp_filters)
2720 PL_rsfp_filters = newAV();
2723 SvUPGRADE(datasv, SVt_PVIO);
2724 IoANY(datasv) = FPTR2DPTR(void *, funcp); /* stash funcp into spare field */
2725 IoFLAGS(datasv) |= IOf_FAKE_DIRP;
2726 DEBUG_P(PerlIO_printf(Perl_debug_log, "filter_add func %p (%s)\n",
2727 FPTR2DPTR(void *, IoANY(datasv)),
2728 SvPV_nolen(datasv)));
2729 av_unshift(PL_rsfp_filters, 1);
2730 av_store(PL_rsfp_filters, 0, datasv) ;
2735 /* Delete most recently added instance of this filter function. */
2737 Perl_filter_del(pTHX_ filter_t funcp)
2743 DEBUG_P(PerlIO_printf(Perl_debug_log, "filter_del func %p",
2744 FPTR2DPTR(void*, funcp)));
2746 if (!PL_rsfp_filters || AvFILLp(PL_rsfp_filters)<0)
2748 /* if filter is on top of stack (usual case) just pop it off */
2749 datasv = FILTER_DATA(AvFILLp(PL_rsfp_filters));
2750 if (IoANY(datasv) == FPTR2DPTR(void *, funcp)) {
2751 IoFLAGS(datasv) &= ~IOf_FAKE_DIRP;
2752 IoANY(datasv) = (void *)NULL;
2753 sv_free(av_pop(PL_rsfp_filters));
2757 /* we need to search for the correct entry and clear it */
2758 Perl_die(aTHX_ "filter_del can only delete in reverse order (currently)");
2762 /* Invoke the idxth filter function for the current rsfp. */
2763 /* maxlen 0 = read one text line */
2765 Perl_filter_read(pTHX_ int idx, SV *buf_sv, int maxlen)
2770 /* This API is bad. It should have been using unsigned int for maxlen.
2771 Not sure if we want to change the API, but if not we should sanity
2772 check the value here. */
2773 const unsigned int correct_length
2782 if (!PL_rsfp_filters)
2784 if (idx > AvFILLp(PL_rsfp_filters)) { /* Any more filters? */
2785 /* Provide a default input filter to make life easy. */
2786 /* Note that we append to the line. This is handy. */
2787 DEBUG_P(PerlIO_printf(Perl_debug_log,
2788 "filter_read %d: from rsfp\n", idx));
2789 if (correct_length) {
2792 const int old_len = SvCUR(buf_sv);
2794 /* ensure buf_sv is large enough */
2795 SvGROW(buf_sv, (STRLEN)(old_len + correct_length)) ;
2796 if ((len = PerlIO_read(PL_rsfp, SvPVX(buf_sv) + old_len,
2797 correct_length)) <= 0) {
2798 if (PerlIO_error(PL_rsfp))
2799 return -1; /* error */
2801 return 0 ; /* end of file */
2803 SvCUR_set(buf_sv, old_len + len) ;
2806 if (sv_gets(buf_sv, PL_rsfp, SvCUR(buf_sv)) == NULL) {
2807 if (PerlIO_error(PL_rsfp))
2808 return -1; /* error */
2810 return 0 ; /* end of file */
2813 return SvCUR(buf_sv);
2815 /* Skip this filter slot if filter has been deleted */
2816 if ( (datasv = FILTER_DATA(idx)) == &PL_sv_undef) {
2817 DEBUG_P(PerlIO_printf(Perl_debug_log,
2818 "filter_read %d: skipped (filter deleted)\n",
2820 return FILTER_READ(idx+1, buf_sv, correct_length); /* recurse */
2822 /* Get function pointer hidden within datasv */
2823 funcp = DPTR2FPTR(filter_t, IoANY(datasv));
2824 DEBUG_P(PerlIO_printf(Perl_debug_log,
2825 "filter_read %d: via function %p (%s)\n",
2826 idx, (void*)datasv, SvPV_nolen_const(datasv)));
2827 /* Call function. The function is expected to */
2828 /* call "FILTER_READ(idx+1, buf_sv)" first. */
2829 /* Return: <0:error, =0:eof, >0:not eof */
2830 return (*funcp)(aTHX_ idx, buf_sv, correct_length);
2834 S_filter_gets(pTHX_ register SV *sv, register PerlIO *fp, STRLEN append)
2837 #ifdef PERL_CR_FILTER
2838 if (!PL_rsfp_filters) {
2839 filter_add(S_cr_textfilter,NULL);
2842 if (PL_rsfp_filters) {
2844 SvCUR_set(sv, 0); /* start with empty line */
2845 if (FILTER_READ(0, sv, 0) > 0)
2846 return ( SvPVX(sv) ) ;
2851 return (sv_gets(sv, fp, append));
2855 S_find_in_my_stash(pTHX_ const char *pkgname, I32 len)
2860 if (len == 11 && *pkgname == '_' && strEQ(pkgname, "__PACKAGE__"))
2864 (pkgname[len - 2] == ':' && pkgname[len - 1] == ':') &&
2865 (gv = gv_fetchpvn_flags(pkgname, len, 0, SVt_PVHV)))
2867 return GvHV(gv); /* Foo:: */
2870 /* use constant CLASS => 'MyClass' */
2871 gv = gv_fetchpvn_flags(pkgname, len, 0, SVt_PVCV);
2872 if (gv && GvCV(gv)) {
2873 SV * const sv = cv_const_sv(GvCV(gv));
2875 pkgname = SvPV_nolen_const(sv);
2878 return gv_stashpv(pkgname, FALSE);
2882 * S_readpipe_override
2883 * Check whether readpipe() is overriden, and generates the appropriate
2884 * optree, provided sublex_start() is called afterwards.
2887 S_readpipe_override(pTHX)
2890 GV *gv_readpipe = gv_fetchpvs("readpipe", GV_NOTQUAL, SVt_PVCV);
2891 yylval.ival = OP_BACKTICK;
2893 && GvCVu(gv_readpipe) && GvIMPORTED_CV(gv_readpipe))
2895 ((gvp = (GV**)hv_fetchs(PL_globalstash, "readpipe", FALSE))
2896 && (gv_readpipe = *gvp) != (GV*)&PL_sv_undef
2897 && GvCVu(gv_readpipe) && GvIMPORTED_CV(gv_readpipe)))
2899 PL_lex_op = (OP*)newUNOP(OP_ENTERSUB, OPf_STACKED,
2900 append_elem(OP_LIST,
2901 newSVOP(OP_CONST, 0, &PL_sv_undef), /* value will be read later */
2902 newCVREF(0, newGVOP(OP_GV, 0, gv_readpipe))));
2912 * The intent of this yylex wrapper is to minimize the changes to the
2913 * tokener when we aren't interested in collecting madprops. It remains
2914 * to be seen how successful this strategy will be...
2921 char *s = PL_bufptr;
2923 /* make sure PL_thiswhite is initialized */
2927 /* just do what yylex would do on pending identifier; leave PL_thiswhite alone */
2928 if (PL_pending_ident)
2929 return S_pending_ident(aTHX);
2931 /* previous token ate up our whitespace? */
2932 if (!PL_lasttoke && PL_nextwhite) {
2933 PL_thiswhite = PL_nextwhite;
2937 /* isolate the token, and figure out where it is without whitespace */
2938 PL_realtokenstart = -1;
2942 assert(PL_curforce < 0);
2944 if (!PL_thismad || PL_thismad->mad_key == '^') { /* not forced already? */
2945 if (!PL_thistoken) {
2946 if (PL_realtokenstart < 0 || !CopLINE(PL_curcop))
2947 PL_thistoken = newSVpvs("");
2949 char * const tstart = SvPVX(PL_linestr) + PL_realtokenstart;
2950 PL_thistoken = newSVpvn(tstart, s - tstart);
2953 if (PL_thismad) /* install head */
2954 CURMAD('X', PL_thistoken);
2957 /* last whitespace of a sublex? */
2958 if (optype == ')' && PL_endwhite) {
2959 CURMAD('X', PL_endwhite);
2964 /* if no whitespace and we're at EOF, bail. Otherwise fake EOF below. */
2965 if (!PL_thiswhite && !PL_endwhite && !optype) {
2966 sv_free(PL_thistoken);
2971 /* put off final whitespace till peg */
2972 if (optype == ';' && !PL_rsfp) {
2973 PL_nextwhite = PL_thiswhite;
2976 else if (PL_thisopen) {
2977 CURMAD('q', PL_thisopen);
2979 sv_free(PL_thistoken);
2983 /* Store actual token text as madprop X */
2984 CURMAD('X', PL_thistoken);
2988 /* add preceding whitespace as madprop _ */
2989 CURMAD('_', PL_thiswhite);
2993 /* add quoted material as madprop = */
2994 CURMAD('=', PL_thisstuff);
2998 /* add terminating quote as madprop Q */
2999 CURMAD('Q', PL_thisclose);
3003 /* special processing based on optype */
3007 /* opval doesn't need a TOKEN since it can already store mp */
3018 append_madprops(PL_thismad, yylval.opval, 0);
3026 addmad(newMADsv('p', PL_endwhite), &PL_thismad, 0);
3035 /* remember any fake bracket that lexer is about to discard */
3036 if (PL_lex_brackets == 1 &&
3037 ((expectation)PL_lex_brackstack[0] & XFAKEBRACK))
3040 while (s < PL_bufend && (*s == ' ' || *s == '\t'))
3043 PL_thiswhite = newSVpvn(PL_bufptr, ++s - PL_bufptr);
3044 addmad(newMADsv('#', PL_thiswhite), &PL_thismad, 0);
3047 break; /* don't bother looking for trailing comment */
3056 /* attach a trailing comment to its statement instead of next token */
3060 if (PL_bufptr > PL_oldbufptr && PL_bufptr[-1] == optype) {
3062 while (s < PL_bufend && (*s == ' ' || *s == '\t'))
3064 if (*s == '\n' || *s == '#') {
3065 while (s < PL_bufend && *s != '\n')
3069 PL_thiswhite = newSVpvn(PL_bufptr, s - PL_bufptr);
3070 addmad(newMADsv('#', PL_thiswhite), &PL_thismad, 0);
3087 /* Create new token struct. Note: opvals return early above. */
3088 yylval.tkval = newTOKEN(optype, yylval, PL_thismad);
3095 S_tokenize_use(pTHX_ int is_use, char *s) {
3097 if (PL_expect != XSTATE)
3098 yyerror(Perl_form(aTHX_ "\"%s\" not allowed in expression",
3099 is_use ? "use" : "no"));
3101 if (isDIGIT(*s) || (*s == 'v' && isDIGIT(s[1]))) {
3102 s = force_version(s, TRUE);
3103 if (*s == ';' || (s = SKIPSPACE1(s), *s == ';')) {
3104 start_force(PL_curforce);
3105 NEXTVAL_NEXTTOKE.opval = NULL;
3108 else if (*s == 'v') {
3109 s = force_word(s,WORD,FALSE,TRUE,FALSE);
3110 s = force_version(s, FALSE);
3114 s = force_word(s,WORD,FALSE,TRUE,FALSE);
3115 s = force_version(s, FALSE);
3117 yylval.ival = is_use;
3121 static const char* const exp_name[] =
3122 { "OPERATOR", "TERM", "REF", "STATE", "BLOCK", "ATTRBLOCK",
3123 "ATTRTERM", "TERMBLOCK", "TERMORDORDOR"
3130 Works out what to call the token just pulled out of the input
3131 stream. The yacc parser takes care of taking the ops we return and
3132 stitching them into a tree.
3138 if read an identifier
3139 if we're in a my declaration
3140 croak if they tried to say my($foo::bar)
3141 build the ops for a my() declaration
3142 if it's an access to a my() variable
3143 are we in a sort block?
3144 croak if my($a); $a <=> $b
3145 build ops for access to a my() variable
3146 if in a dq string, and they've said @foo and we can't find @foo
3148 build ops for a bareword
3149 if we already built the token before, use it.
3154 #pragma segment Perl_yylex
3160 register char *s = PL_bufptr;
3165 /* orig_keyword, gvp, and gv are initialized here because
3166 * jump to the label just_a_word_zero can bypass their
3167 * initialization later. */
3168 I32 orig_keyword = 0;
3173 SV* tmp = newSVpvs("");
3174 PerlIO_printf(Perl_debug_log, "### %"IVdf":LEX_%s/X%s %s\n",
3175 (IV)CopLINE(PL_curcop),
3176 lex_state_names[PL_lex_state],
3177 exp_name[PL_expect],
3178 pv_display(tmp, s, strlen(s), 0, 60));
3181 /* check if there's an identifier for us to look at */
3182 if (PL_pending_ident)
3183 return REPORT(S_pending_ident(aTHX));
3185 /* no identifier pending identification */
3187 switch (PL_lex_state) {
3189 case LEX_NORMAL: /* Some compilers will produce faster */
3190 case LEX_INTERPNORMAL: /* code if we comment these out. */
3194 /* when we've already built the next token, just pull it out of the queue */
3198 yylval = PL_nexttoke[PL_lasttoke].next_val;
3200 PL_thismad = PL_nexttoke[PL_lasttoke].next_mad;
3201 PL_nexttoke[PL_lasttoke].next_mad = 0;
3202 if (PL_thismad && PL_thismad->mad_key == '_') {
3203 PL_thiswhite = (SV*)PL_thismad->mad_val;
3204 PL_thismad->mad_val = 0;
3205 mad_free(PL_thismad);
3210 PL_lex_state = PL_lex_defer;
3211 PL_expect = PL_lex_expect;
3212 PL_lex_defer = LEX_NORMAL;
3213 if (!PL_nexttoke[PL_lasttoke].next_type)
3218 yylval = PL_nextval[PL_nexttoke];
3220 PL_lex_state = PL_lex_defer;
3221 PL_expect = PL_lex_expect;
3222 PL_lex_defer = LEX_NORMAL;
3226 /* FIXME - can these be merged? */
3227 return(PL_nexttoke[PL_lasttoke].next_type);
3229 return REPORT(PL_nexttype[PL_nexttoke]);
3232 /* interpolated case modifiers like \L \U, including \Q and \E.
3233 when we get here, PL_bufptr is at the \
3235 case LEX_INTERPCASEMOD:
3237 if (PL_bufptr != PL_bufend && *PL_bufptr != '\\')
3238 Perl_croak(aTHX_ "panic: INTERPCASEMOD");
3240 /* handle \E or end of string */
3241 if (PL_bufptr == PL_bufend || PL_bufptr[1] == 'E') {
3243 if (PL_lex_casemods) {
3244 const char oldmod = PL_lex_casestack[--PL_lex_casemods];
3245 PL_lex_casestack[PL_lex_casemods] = '\0';
3247 if (PL_bufptr != PL_bufend
3248 && (oldmod == 'L' || oldmod == 'U' || oldmod == 'Q')) {
3250 PL_lex_state = LEX_INTERPCONCAT;
3253 PL_thistoken = newSVpvs("\\E");
3259 while (PL_bufptr != PL_bufend &&
3260 PL_bufptr[0] == '\\' && PL_bufptr[1] == 'E') {
3262 PL_thiswhite = newSVpvs("");
3263 sv_catpvn(PL_thiswhite, PL_bufptr, 2);
3267 if (PL_bufptr != PL_bufend)
3270 PL_lex_state = LEX_INTERPCONCAT;
3274 DEBUG_T({ PerlIO_printf(Perl_debug_log,
3275 "### Saw case modifier\n"); });
3277 if (s[1] == '\\' && s[2] == 'E') {
3280 PL_thiswhite = newSVpvs("");
3281 sv_catpvn(PL_thiswhite, PL_bufptr, 4);
3284 PL_lex_state = LEX_INTERPCONCAT;
3289 if (!PL_madskills) /* when just compiling don't need correct */
3290 if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
3291 tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
3292 if ((*s == 'L' || *s == 'U') &&
3293 (strchr(PL_lex_casestack, 'L') || strchr(PL_lex_casestack, 'U'))) {
3294 PL_lex_casestack[--PL_lex_casemods] = '\0';
3297 if (PL_lex_casemods > 10)
3298 Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
3299 PL_lex_casestack[PL_lex_casemods++] = *s;
3300 PL_lex_casestack[PL_lex_casemods] = '\0';
3301 PL_lex_state = LEX_INTERPCONCAT;
3302 start_force(PL_curforce);
3303 NEXTVAL_NEXTTOKE.ival = 0;
3305 start_force(PL_curforce);
3307 NEXTVAL_NEXTTOKE.ival = OP_LCFIRST;
3309 NEXTVAL_NEXTTOKE.ival = OP_UCFIRST;
3311 NEXTVAL_NEXTTOKE.ival = OP_LC;
3313 NEXTVAL_NEXTTOKE.ival = OP_UC;
3315 NEXTVAL_NEXTTOKE.ival = OP_QUOTEMETA;
3317 Perl_croak(aTHX_ "panic: yylex");
3319 SV* const tmpsv = newSVpvs("");
3320 Perl_sv_catpvf(aTHX_ tmpsv, "\\%c", *s);
3326 if (PL_lex_starts) {
3332 sv_free(PL_thistoken);
3333 PL_thistoken = newSVpvs("");
3336 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
3337 if (PL_lex_casemods == 1 && PL_lex_inpat)
3346 case LEX_INTERPPUSH:
3347 return REPORT(sublex_push());
3349 case LEX_INTERPSTART:
3350 if (PL_bufptr == PL_bufend)
3351 return REPORT(sublex_done());
3352 DEBUG_T({ PerlIO_printf(Perl_debug_log,
3353 "### Interpolated variable\n"); });
3355 PL_lex_dojoin = (*PL_bufptr == '@');
3356 PL_lex_state = LEX_INTERPNORMAL;
3357 if (PL_lex_dojoin) {
3358 start_force(PL_curforce);
3359 NEXTVAL_NEXTTOKE.ival = 0;
3361 start_force(PL_curforce);
3362 force_ident("\"", '$');
3363 start_force(PL_curforce);
3364 NEXTVAL_NEXTTOKE.ival = 0;
3366 start_force(PL_curforce);
3367 NEXTVAL_NEXTTOKE.ival = 0;
3369 start_force(PL_curforce);
3370 NEXTVAL_NEXTTOKE.ival = OP_JOIN; /* emulate join($", ...) */
3373 if (PL_lex_starts++) {
3378 sv_free(PL_thistoken);
3379 PL_thistoken = newSVpvs("");
3382 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
3383 if (!PL_lex_casemods && PL_lex_inpat)
3390 case LEX_INTERPENDMAYBE:
3391 if (intuit_more(PL_bufptr)) {
3392 PL_lex_state = LEX_INTERPNORMAL; /* false alarm, more expr */
3398 if (PL_lex_dojoin) {
3399 PL_lex_dojoin = FALSE;
3400 PL_lex_state = LEX_INTERPCONCAT;
3404 sv_free(PL_thistoken);
3405 PL_thistoken = newSVpvs("");
3410 if (PL_lex_inwhat == OP_SUBST && PL_linestr == PL_lex_repl
3411 && SvEVALED(PL_lex_repl))
3413 if (PL_bufptr != PL_bufend)
3414 Perl_croak(aTHX_ "Bad evalled substitution pattern");
3418 case LEX_INTERPCONCAT:
3420 if (PL_lex_brackets)
3421 Perl_croak(aTHX_ "panic: INTERPCONCAT");
3423 if (PL_bufptr == PL_bufend)
3424 return REPORT(sublex_done());
3426 if (SvIVX(PL_linestr) == '\'') {
3427 SV *sv = newSVsv(PL_linestr);
3430 else if ( PL_hints & HINT_NEW_RE )
3431 sv = new_constant(NULL, 0, "qr", sv, sv, "q");
3432 yylval.opval = (OP*)newSVOP(OP_CONST, 0, sv);
3436 s = scan_const(PL_bufptr);
3438 PL_lex_state = LEX_INTERPCASEMOD;
3440 PL_lex_state = LEX_INTERPSTART;
3443 if (s != PL_bufptr) {
3444 start_force(PL_curforce);
3446 curmad('X', newSVpvn(PL_bufptr,s-PL_bufptr));
3448 NEXTVAL_NEXTTOKE = yylval;
3451 if (PL_lex_starts++) {
3455 sv_free(PL_thistoken);
3456 PL_thistoken = newSVpvs("");
3459 /* commas only at base level: /$a\Ub$c/ => ($a,uc(b.$c)) */
3460 if (!PL_lex_casemods && PL_lex_inpat)
3473 PL_lex_state = LEX_NORMAL;
3474 s = scan_formline(PL_bufptr);
3475 if (!PL_lex_formbrack)
3481 PL_oldoldbufptr = PL_oldbufptr;
3487 sv_free(PL_thistoken);
3490 PL_realtokenstart = s - SvPVX(PL_linestr); /* assume but undo on ws */
3494 if (isIDFIRST_lazy_if(s,UTF))
3496 Perl_croak(aTHX_ "Unrecognized character \\x%02X", *s & 255);
3499 goto fake_eof; /* emulate EOF on ^D or ^Z */
3508 if (PL_lex_brackets) {
3509 yyerror((const char *)
3511 ? "Format not terminated"
3512 : "Missing right curly or square bracket"));
3514 DEBUG_T( { PerlIO_printf(Perl_debug_log,
3515 "### Tokener got EOF\n");
3519 if (s++ < PL_bufend)
3520 goto retry; /* ignore stray nulls */
3523 if (!PL_in_eval && !PL_preambled) {
3524 PL_preambled = TRUE;
3529 sv_setpv(PL_linestr,incl_perldb());
3530 if (SvCUR(PL_linestr))
3531 sv_catpvs(PL_linestr,";");
3533 while(AvFILLp(PL_preambleav) >= 0) {
3534 SV *tmpsv = av_shift(PL_preambleav);
3535 sv_catsv(PL_linestr, tmpsv);
3536 sv_catpvs(PL_linestr, ";");
3539 sv_free((SV*)PL_preambleav);
3540 PL_preambleav = NULL;
3542 if (PL_minus_n || PL_minus_p) {
3543 sv_catpvs(PL_linestr, "LINE: while (<>) {");
3545 sv_catpvs(PL_linestr,"chomp;");
3548 if ((*PL_splitstr == '/' || *PL_splitstr == '\''
3549 || *PL_splitstr == '"')
3550 && strchr(PL_splitstr + 1, *PL_splitstr))
3551 Perl_sv_catpvf(aTHX_ PL_linestr, "our @F=split(%s);", PL_splitstr);
3553 /* "q\0${splitstr}\0" is legal perl. Yes, even NUL
3554 bytes can be used as quoting characters. :-) */
3555 const char *splits = PL_splitstr;
3556 sv_catpvs(PL_linestr, "our @F=split(q\0");
3559 if (*splits == '\\')
3560 sv_catpvn(PL_linestr, splits, 1);
3561 sv_catpvn(PL_linestr, splits, 1);
3562 } while (*splits++);
3563 /* This loop will embed the trailing NUL of
3564 PL_linestr as the last thing it does before
3566 sv_catpvs(PL_linestr, ");");
3570 sv_catpvs(PL_linestr,"our @F=split(' ');");
3574 sv_catpvs(PL_linestr,"use feature ':5.10';");
3575 sv_catpvs(PL_linestr, "\n");
3576 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3577 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3578 PL_last_lop = PL_last_uni = NULL;
3579 if (PERLDB_LINE && PL_curstash != PL_debstash)
3580 update_debugger_info_sv(PL_linestr);
3584 bof = PL_rsfp ? TRUE : FALSE;
3585 if ((s = filter_gets(PL_linestr, PL_rsfp, 0)) == NULL) {
3588 PL_realtokenstart = -1;
3591 if (PL_preprocess && !PL_in_eval)
3592 (void)PerlProc_pclose(PL_rsfp);
3593 else if ((PerlIO *)PL_rsfp == PerlIO_stdin())
3594 PerlIO_clearerr(PL_rsfp);
3596 (void)PerlIO_close(PL_rsfp);
3598 PL_doextract = FALSE;
3600 if (!PL_in_eval && (PL_minus_n || PL_minus_p)) {
3605 sv_setpv(PL_linestr,
3608 ? ";}continue{print;}" : ";}"));
3609 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3610 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3611 PL_last_lop = PL_last_uni = NULL;
3612 PL_minus_n = PL_minus_p = 0;
3615 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3616 PL_last_lop = PL_last_uni = NULL;
3617 sv_setpvn(PL_linestr,"",0);
3618 TOKEN(';'); /* not infinite loop because rsfp is NULL now */
3620 /* If it looks like the start of a BOM or raw UTF-16,
3621 * check if it in fact is. */
3627 #ifdef PERLIO_IS_STDIO
3628 # ifdef __GNU_LIBRARY__
3629 # if __GNU_LIBRARY__ == 1 /* Linux glibc5 */
3630 # define FTELL_FOR_PIPE_IS_BROKEN
3634 # if __GLIBC__ == 1 /* maybe some glibc5 release had it like this? */
3635 # define FTELL_FOR_PIPE_IS_BROKEN
3640 #ifdef FTELL_FOR_PIPE_IS_BROKEN
3641 /* This loses the possibility to detect the bof
3642 * situation on perl -P when the libc5 is being used.
3643 * Workaround? Maybe attach some extra state to PL_rsfp?
3646 bof = PerlIO_tell(PL_rsfp) == SvCUR(PL_linestr);
3648 bof = PerlIO_tell(PL_rsfp) == (Off_t)SvCUR(PL_linestr);
3651 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3652 s = swallow_bom((U8*)s);
3656 /* Incest with pod. */
3659 sv_catsv(PL_thiswhite, PL_linestr);
3661 if (*s == '=' && strnEQ(s, "=cut", 4)) {
3662 sv_setpvn(PL_linestr, "", 0);
3663 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3664 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3665 PL_last_lop = PL_last_uni = NULL;
3666 PL_doextract = FALSE;
3670 } while (PL_doextract);
3671 PL_oldoldbufptr = PL_oldbufptr = PL_bufptr = PL_linestart = s;
3672 if (PERLDB_LINE && PL_curstash != PL_debstash)
3673 update_debugger_info_sv(PL_linestr);
3674 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3675 PL_last_lop = PL_last_uni = NULL;
3676 if (CopLINE(PL_curcop) == 1) {
3677 while (s < PL_bufend && isSPACE(*s))
3679 if (*s == ':' && s[1] != ':') /* for csh execing sh scripts */
3683 PL_thiswhite = newSVpvn(PL_linestart, s - PL_linestart);
3687 if (*s == '#' && *(s+1) == '!')
3689 #ifdef ALTERNATE_SHEBANG
3691 static char const as[] = ALTERNATE_SHEBANG;
3692 if (*s == as[0] && strnEQ(s, as, sizeof(as) - 1))
3693 d = s + (sizeof(as) - 1);
3695 #endif /* ALTERNATE_SHEBANG */
3704 while (*d && !isSPACE(*d))
3708 #ifdef ARG_ZERO_IS_SCRIPT
3709 if (ipathend > ipath) {
3711 * HP-UX (at least) sets argv[0] to the script name,
3712 * which makes $^X incorrect. And Digital UNIX and Linux,
3713 * at least, set argv[0] to the basename of the Perl
3714 * interpreter. So, having found "#!", we'll set it right.
3716 SV * const x = GvSV(gv_fetchpvs("\030", GV_ADD|GV_NOTQUAL,
3718 assert(SvPOK(x) || SvGMAGICAL(x));
3719 if (sv_eq(x, CopFILESV(PL_curcop))) {
3720 sv_setpvn(x, ipath, ipathend - ipath);
3726 const char *bstart = SvPV_const(CopFILESV(PL_curcop),blen);
3727 const char * const lstart = SvPV_const(x,llen);
3729 bstart += blen - llen;
3730 if (strnEQ(bstart, lstart, llen) && bstart[-1] == '/') {
3731 sv_setpvn(x, ipath, ipathend - ipath);
3736 TAINT_NOT; /* $^X is always tainted, but that's OK */
3738 #endif /* ARG_ZERO_IS_SCRIPT */
3743 d = instr(s,"perl -");
3745 d = instr(s,"perl");
3747 /* avoid getting into infinite loops when shebang
3748 * line contains "Perl" rather than "perl" */
3750 for (d = ipathend-4; d >= ipath; --d) {
3751 if ((*d == 'p' || *d == 'P')
3752 && !ibcmp(d, "perl", 4))
3762 #ifdef ALTERNATE_SHEBANG
3764 * If the ALTERNATE_SHEBANG on this system starts with a
3765 * character that can be part of a Perl expression, then if
3766 * we see it but not "perl", we're probably looking at the
3767 * start of Perl code, not a request to hand off to some
3768 * other interpreter. Similarly, if "perl" is there, but
3769 * not in the first 'word' of the line, we assume the line
3770 * contains the start of the Perl program.
3772 if (d && *s != '#') {
3773 const char *c = ipath;
3774 while (*c && !strchr("; \t\r\n\f\v#", *c))
3777 d = NULL; /* "perl" not in first word; ignore */
3779 *s = '#'; /* Don't try to parse shebang line */
3781 #endif /* ALTERNATE_SHEBANG */
3782 #ifndef MACOS_TRADITIONAL
3787 !instr(s,"indir") &&
3788 instr(PL_origargv[0],"perl"))
3795 while (s < PL_bufend && isSPACE(*s))
3797 if (s < PL_bufend) {
3798 Newxz(newargv,PL_origargc+3,char*);
3800 while (s < PL_bufend && !isSPACE(*s))
3803 Copy(PL_origargv+1, newargv+2, PL_origargc+1, char*);
3806 newargv = PL_origargv;
3809 PerlProc_execv(ipath, EXEC_ARGV_CAST(newargv));
3811 Perl_croak(aTHX_ "Can't exec %s", ipath);
3815 while (*d && !isSPACE(*d))
3817 while (SPACE_OR_TAB(*d))
3821 const bool switches_done = PL_doswitches;
3822 const U32 oldpdb = PL_perldb;
3823 const bool oldn = PL_minus_n;
3824 const bool oldp = PL_minus_p;
3827 if (*d == 'M' || *d == 'm' || *d == 'C') {
3828 const char * const m = d;
3829 while (*d && !isSPACE(*d))
3831 Perl_croak(aTHX_ "Too late for \"-%.*s\" option",
3834 d = moreswitches(d);
3836 if (PL_doswitches && !switches_done) {
3837 int argc = PL_origargc;
3838 char **argv = PL_origargv;
3841 } while (argc && argv[0][0] == '-' && argv[0][1]);
3842 init_argv_symbols(argc,argv);
3844 if ((PERLDB_LINE && !oldpdb) ||
3845 ((PL_minus_n || PL_minus_p) && !(oldn || oldp)))
3846 /* if we have already added "LINE: while (<>) {",
3847 we must not do it again */
3849 sv_setpvn(PL_linestr, "", 0);
3850 PL_oldoldbufptr = PL_oldbufptr = s = PL_linestart = SvPVX(PL_linestr);
3851 PL_bufend = SvPVX(PL_linestr) + SvCUR(PL_linestr);
3852 PL_last_lop = PL_last_uni = NULL;
3853 PL_preambled = FALSE;
3855 (void)gv_fetchfile(PL_origfilename);
3862 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
3864 PL_lex_state = LEX_FORMLINE;
3869 #ifdef PERL_STRICT_CR
3870 Perl_warn(aTHX_ "Illegal character \\%03o (carriage return)", '\r');
3872 "\t(Maybe you didn't strip carriage returns after a network transfer?)\n");
3874 case ' ': case '\t': case '\f': case 013:
3875 #ifdef MACOS_TRADITIONAL
3879 PL_realtokenstart = -1;
3888 PL_realtokenstart = -1;
3892 if (PL_lex_state != LEX_NORMAL || (PL_in_eval && !PL_rsfp)) {
3893 if (*s == '#' && s == PL_linestart && PL_in_eval && !PL_rsfp) {
3894 /* handle eval qq[#line 1 "foo"\n ...] */
3895 CopLINE_dec(PL_curcop);
3898 if (PL_madskills && !PL_lex_formbrack && !PL_in_eval) {
3900 if (!PL_in_eval || PL_rsfp)
3905 while (d < PL_bufend && *d != '\n')
3909 else if (d > PL_bufend) /* Found by Ilya: feed random input to Perl. */
3910 Perl_croak(aTHX_ "panic: input overflow");
3913 PL_thiswhite = newSVpvn(s, d - s);
3918 if (PL_lex_formbrack && PL_lex_brackets <= PL_lex_formbrack) {
3920 PL_lex_state = LEX_FORMLINE;
3926 if (PL_madskills && CopLINE(PL_curcop) >= 1 && !PL_lex_formbrack) {
3927 if (CopLINE(PL_curcop) == 1 && s[0] == '#' && s[1] == '!') {
3930 TOKEN(PEG); /* make sure any #! line is accessible */
3935 /* if (PL_madskills && PL_lex_formbrack) { */
3937 while (d < PL_bufend && *d != '\n')
3941 else if (d > PL_bufend) /* Found by Ilya: feed random input to Perl. */
3942 Perl_croak(aTHX_ "panic: input overflow");
3943 if (PL_madskills && CopLINE(PL_curcop) >= 1) {
3945 PL_thiswhite = newSVpvs("");
3946 if (CopLINE(PL_curcop) == 1) {
3947 sv_setpvn(PL_thiswhite, "", 0);
3950 sv_catpvn(PL_thiswhite, s, d - s);
3964 if (s[1] && isALPHA(s[1]) && !isALNUM(s[2])) {
3972 while (s < PL_bufend && SPACE_OR_TAB(*s))
3975 if (strnEQ(s,"=>",2)) {
3976 s = force_word(PL_bufptr,WORD,FALSE,FALSE,FALSE);
3977 DEBUG_T( { printbuf("### Saw unary minus before =>, forcing word %s\n", s); } );
3978 OPERATOR('-'); /* unary minus */
3980 PL_last_uni = PL_oldbufptr;
3982 case 'r': ftst = OP_FTEREAD; break;
3983 case 'w': ftst = OP_FTEWRITE; break;
3984 case 'x': ftst = OP_FTEEXEC; break;
3985 case 'o': ftst = OP_FTEOWNED; break;
3986 case 'R': ftst = OP_FTRREAD; break;
3987 case 'W': ftst = OP_FTRWRITE; break;
3988 case 'X': ftst = OP_FTREXEC; break;
3989 case 'O': ftst = OP_FTROWNED; break;
3990 case 'e': ftst = OP_FTIS; break;
3991 case 'z': ftst = OP_FTZERO; break;
3992 case 's': ftst = OP_FTSIZE; break;
3993 case 'f': ftst = OP_FTFILE; break;
3994 case 'd': ftst = OP_FTDIR; break;
3995 case 'l': ftst = OP_FTLINK; break;
3996 case 'p': ftst = OP_FTPIPE; break;
3997 case 'S': ftst = OP_FTSOCK; break;
3998 case 'u': ftst = OP_FTSUID; break;
3999 case 'g': ftst = OP_FTSGID; break;
4000 case 'k': ftst = OP_FTSVTX; break;
4001 case 'b': ftst = OP_FTBLK; break;
4002 case 'c': ftst = OP_FTCHR; break;
4003 case 't': ftst = OP_FTTTY; break;
4004 case 'T': ftst = OP_FTTEXT; break;
4005 case 'B': ftst = OP_FTBINARY; break;
4006 case 'M': case 'A': case 'C':
4007 gv_fetchpvs("\024", GV_ADD|GV_NOTQUAL, SVt_PV);
4009 case 'M': ftst = OP_FTMTIME; break;
4010 case 'A': ftst = OP_FTATIME; break;
4011 case 'C': ftst = OP_FTCTIME; break;
4019 PL_last_lop_op = (OPCODE)ftst;
4020 DEBUG_T( { PerlIO_printf(Perl_debug_log,
4021 "### Saw file test %c\n", (int)tmp);
4026 /* Assume it was a minus followed by a one-letter named
4027 * subroutine call (or a -bareword), then. */
4028 DEBUG_T( { PerlIO_printf(Perl_debug_log,
4029 "### '-%c' looked like a file test but was not\n",
4036 const char tmp = *s++;
4039 if (PL_expect == XOPERATOR)
4044 else if (*s == '>') {
4047 if (isIDFIRST_lazy_if(s,UTF)) {
4048 s = force_word(s,METHOD,FALSE,TRUE,FALSE);
4056 if (PL_expect == XOPERATOR)
4059 if (isSPACE(*s) || !isSPACE(*PL_bufptr))
4061 OPERATOR('-'); /* unary minus */
4067 const char tmp = *s++;
4070 if (PL_expect == XOPERATOR)
4075 if (PL_expect == XOPERATOR)
4078 if (isSPACE(*s) || !isSPACE(*PL_bufptr))
4085 if (PL_expect != XOPERATOR) {
4086 s = scan_ident(s, PL_bufend, PL_tokenbuf, sizeof PL_tokenbuf, TRUE);
4087 PL_expect = XOPERATOR;
4088 force_ident(PL_tokenbuf, '*');
4101 if (PL_expect == XOPERATOR) {
4105 PL_tokenbuf[0] = '%';
4106 s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, TRUE);
4107 if (!PL_tokenbuf[1]) {
4110 PL_pending_ident = '%';
4121 && (PL_expect == XOPERATOR || PL_expect == XTERMORDORDOR)
4122 && FEATURE_IS_ENABLED("~~"))
4129 const char tmp = *s++;
4135 goto just_a_word_zero_gv;
4138 switch (PL_expect) {
4144 if (!PL_in_my || PL_lex_state != LEX_NORMAL)
4146 PL_bufptr = s; /* update in case we back off */
4152 PL_expect = XTERMBLOCK;
4155 stuffstart = s - SvPVX(PL_linestr) - 1;
4159 while (isIDFIRST_lazy_if(s,UTF)) {
4162 d = scan_word(s, PL_tokenbuf, sizeof PL_tokenbuf, FALSE, &len);
4163 if (isLOWER(*s) && (tmp = keyword(PL_tokenbuf, len, 0))) {
4164 if (tmp < 0) tmp = -tmp;
4179 sv = newSVpvn(s, len);
4181 d = scan_str(d,TRUE,TRUE);
4183 /* MUST advance bufptr here to avoid bogus
4184 "at end of line" context messages from yyerror().
4186 PL_bufptr = s + len;
4187 yyerror("Unterminated attribute parameter in attribute list");
4191 return REPORT(0); /* EOF indicator */
4195 sv_catsv(sv, PL_lex_stuff);
4196 attrs = append_elem(OP_LIST, attrs,
4197 newSVOP(OP_CONST, 0, sv));
4198 SvREFCNT_dec(PL_lex_stuff);
4199 PL_lex_stuff = NULL;
4202 if (len == 6 && strnEQ(SvPVX(sv), "unique", len)) {
4204 if (PL_in_my == KEY_our) {
4206 GvUNIQUE_on(cGVOPx_gv(yylval.opval));
4208 /* skip to avoid loading attributes.pm */
4210 deprecate(":unique");
4213 Perl_croak(aTHX_ "The 'unique' attribute may only be applied to 'our' variables");
4216 /* NOTE: any CV attrs applied here need to be part of
4217 the CVf_BUILTIN_ATTRS define in cv.h! */
4218 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "lvalue", len)) {
4220 CvLVALUE_on(PL_compcv);
4222 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "locked", len)) {
4224 CvLOCKED_on(PL_compcv);
4226 else if (!PL_in_my && len == 6 && strnEQ(SvPVX(sv), "method", len)) {
4228 CvMETHOD_on(PL_compcv);
4230 else if (!PL_in_my && len == 9 && strnEQ(SvPVX(sv), "assertion", len)) {
4232 CvASSERTION_on(PL_compcv);
4234 /* After we've set the flags, it could be argued that
4235 we don't need to do the attributes.pm-based setting
4236 process, and shouldn't bother appending recognized
4237 flags. To experiment with that, uncomment the
4238 following "else". (Note that's already been
4239 uncommented. That keeps the above-applied built-in
4240 attributes from being intercepted (and possibly
4241 rejected) by a package's attribute routines, but is
4242 justified by the performance win for the common case
4243 of applying only built-in attributes.) */
4245 attrs = append_elem(OP_LIST, attrs,
4246 newSVOP(OP_CONST, 0,
4250 if (*s == ':' && s[1] != ':')
4253 break; /* require real whitespace or :'s */
4254 /* XXX losing whitespace on sequential attributes here */
4258 = (PL_expect == XOPERATOR ? '=' : '{'); /*'}(' for vi */
4259 if (*s != ';' && *s != '}' && *s != tmp
4260 && (tmp != '=' || *s != ')')) {
4261 const char q = ((*s == '\'') ? '"' : '\'');
4262 /* If here for an expression, and parsed no attrs, back
4264 if (tmp == '=' && !attrs) {
4268 /* MUST advance bufptr here to avoid bogus "at end of line"
4269 context messages from yyerror().
4272 yyerror( (const char *)
4274 ? Perl_form(aTHX_ "Invalid separator character "
4275 "%c%c%c in attribute list", q, *s, q)
4276 : "Unterminated attribute list" ) );
4284 start_force(PL_curforce);
4285 NEXTVAL_NEXTTOKE.opval = attrs;
4286 CURMAD('_', PL_nextwhite);
4291 PL_thistoken = newSVpvn(SvPVX(PL_linestr) + stuffstart,
4292 (s - SvPVX(PL_linestr)) - stuffstart);
4300 if (PL_last_lop == PL_oldoldbufptr || PL_last_uni == PL_oldoldbufptr)
4301 PL_oldbufptr = PL_oldoldbufptr; /* allow print(STDOUT 123) */
4309 const char tmp = *s++;
4314 const char tmp = *s++;
4322 if (PL_lex_brackets <= 0)
4323 yyerror("Unmatched right square bracket");
4326 if (PL_lex_state == LEX_INTERPNORMAL) {
4327 if (PL_lex_brackets == 0) {
4328 if (*s != '[' && *s != '{' && (*s != '-' || s[1] != '>'))
4329 PL_lex_state = LEX_INTERPEND;
4336 if (PL_lex_brackets > 100) {
4337 Renew(PL_lex_brackstack, PL_lex_brackets + 10, char);
4339 switch (PL_expect) {
4341 if (PL_lex_formbrack) {
4345 if (PL_oldoldbufptr == PL_last_lop)
4346 PL_lex_brackstack[PL_lex_brackets++] = XTERM;
4348 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
4349 OPERATOR(HASHBRACK);
4351 while (s < PL_bufend && SPACE_OR_TAB(*s))
4354 PL_tokenbuf[0] = '\0';
4355 if (d < PL_bufend && *d == '-') {
4356 PL_tokenbuf[0] = '-';
4358 while (d < PL_bufend && SPACE_OR_TAB(*d))
4361 if (d < PL_bufend && isIDFIRST_lazy_if(d,UTF)) {
4362 d = scan_word(d, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1,
4364 while (d < PL_bufend && SPACE_OR_TAB(*d))
4367 const char minus = (PL_tokenbuf[0] == '-');
4368 s = force_word(s + minus, WORD, FALSE, TRUE, FALSE);
4376 PL_lex_brackstack[PL_lex_brackets++] = XSTATE;
4381 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
4386 if (PL_oldoldbufptr == PL_last_lop)
4387 PL_lex_brackstack[PL_lex_brackets++] = XTERM;
4389 PL_lex_brackstack[PL_lex_brackets++] = XOPERATOR;
4392 if (PL_expect == XREF && PL_lex_state == LEX_INTERPNORMAL) {
4394 /* This hack is to get the ${} in the message. */
4396 yyerror("syntax error");
4399 OPERATOR(HASHBRACK);
4401 /* This hack serves to disambiguate a pair of curlies
4402 * as being a block or an anon hash. Normally, expectation
4403 * determines that, but in cases where we're not in a
4404 * position to expect anything in particular (like inside
4405 * eval"") we have to resolve the ambiguity. This code
4406 * covers the case where the first term in the curlies is a
4407 * quoted string. Most other cases need to be explicitly
4408 * disambiguated by prepending a "+" before the opening
4409 * curly in order to force resolution as an anon hash.
4411 * XXX should probably propagate the outer expectation
4412 * into eval"" to rely less on this hack, but that could
4413 * potentially break current behavior of eval"".