/* toke.c
*
- * Copyright (c) 1991-2002, Larry Wall
+ * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
+ * 2000, 2001, 2002, 2003, by Larry Wall and others
*
* You may distribute under the terms of either the GNU General Public
* License or the Artistic License, as specified in the README file.
#ifdef DEBUGGING
static char* exp_name[] =
{ "OPERATOR", "TERM", "REF", "STATE", "BLOCK", "ATTRBLOCK",
- "ATTRTERM", "TERMBLOCK"
+ "ATTRTERM", "TERMBLOCK", "TERMORDORDOR"
};
#endif
DEBUG_T({ PerlIO_printf(Perl_debug_log,
"### Saw case modifier at '%s'\n", PL_bufptr); });
s = PL_bufptr + 1;
- if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
- tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
- if (strchr("LU", *s) &&
- (strchr(PL_lex_casestack, 'L') || strchr(PL_lex_casestack, 'U')))
- {
- PL_lex_casestack[--PL_lex_casemods] = '\0';
- return ')';
+ if (s[1] == '\\' && s[2] == 'E') {
+ PL_bufptr = s + 3;
+ PL_lex_state = LEX_INTERPCONCAT;
+ return yylex();
}
- if (PL_lex_casemods > 10) {
- Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
+ else {
+ if (strnEQ(s, "L\\u", 3) || strnEQ(s, "U\\l", 3))
+ tmp = *s, *s = s[2], s[2] = (char)tmp; /* misordered... */
+ if (strchr("LU", *s) &&
+ (strchr(PL_lex_casestack, 'L') || strchr(PL_lex_casestack, 'U'))) {
+ PL_lex_casestack[--PL_lex_casemods] = '\0';
+ return ')';
+ }
+ if (PL_lex_casemods > 10)
+ Renew(PL_lex_casestack, PL_lex_casemods + 2, char);
+ PL_lex_casestack[PL_lex_casemods++] = *s;
+ PL_lex_casestack[PL_lex_casemods] = '\0';
+ PL_lex_state = LEX_INTERPCONCAT;
+ PL_nextval[PL_nexttoke].ival = 0;
+ force_next('(');
+ if (*s == 'l')
+ PL_nextval[PL_nexttoke].ival = OP_LCFIRST;
+ else if (*s == 'u')
+ PL_nextval[PL_nexttoke].ival = OP_UCFIRST;
+ else if (*s == 'L')
+ PL_nextval[PL_nexttoke].ival = OP_LC;
+ else if (*s == 'U')
+ PL_nextval[PL_nexttoke].ival = OP_UC;
+ else if (*s == 'Q')
+ PL_nextval[PL_nexttoke].ival = OP_QUOTEMETA;
+ else
+ Perl_croak(aTHX_ "panic: yylex");
+ PL_bufptr = s + 1;
}
- PL_lex_casestack[PL_lex_casemods++] = *s;
- PL_lex_casestack[PL_lex_casemods] = '\0';
- PL_lex_state = LEX_INTERPCONCAT;
- PL_nextval[PL_nexttoke].ival = 0;
- force_next('(');
- if (*s == 'l')
- PL_nextval[PL_nexttoke].ival = OP_LCFIRST;
- else if (*s == 'u')
- PL_nextval[PL_nexttoke].ival = OP_UCFIRST;
- else if (*s == 'L')
- PL_nextval[PL_nexttoke].ival = OP_LC;
- else if (*s == 'U')
- PL_nextval[PL_nexttoke].ival = OP_UC;
- else if (*s == 'Q')
- PL_nextval[PL_nexttoke].ival = OP_QUOTEMETA;
- else
- Perl_croak(aTHX_ "panic: yylex");
- PL_bufptr = s + 1;
force_next(FUNC);
if (PL_lex_starts) {
s = PL_bufptr;
}
d = moreswitches(d);
} while (d);
+ if (PL_doswitches && !switches_done) {
+ int argc = PL_origargc;
+ char **argv = PL_origargv;
+ do {
+ argc--,argv++;
+ } while (argc && argv[0][0] == '-' && argv[0][1]);
+ init_argv_symbols(argc,argv);
+ }
if ((PERLDB_LINE && !oldpdb) ||
((PL_minus_n || PL_minus_p) && !(oldn || oldp)))
/* if we have already added "LINE: while (<>) {",
PL_tokenbuf[0] = '%';
s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, TRUE);
if (!PL_tokenbuf[1]) {
- if (s == PL_bufend)
- yyerror("Final % should be \\% or %name");
PREREF('%');
}
PL_pending_ident = '%';
CvLOCKED_on(PL_compcv);
else if (!PL_in_my && len == 6 && strnEQ(s, "method", len))
CvMETHOD_on(PL_compcv);
+ else if (!PL_in_my && len == 9 && strnEQ(s, "assertion", len))
+ CvASSERTION_on(PL_compcv);
#ifdef USE_ITHREADS
else if (PL_in_my == KEY_our && len == 6 &&
strnEQ(s, "unique", len))
PL_oldbufptr = PL_oldoldbufptr; /* allow print(STDOUT 123) */
else
PL_expect = XTERM;
+ s = skipspace(s);
TOKEN('(');
case ';':
CLINE;
|| ((*t == 'q' || *t == 'x') && ++t < PL_bufend
&& !isALNUM(*t))))
{
+ /* skip q//-like construct */
char *tmps;
char open, close, term;
I32 brackets = 1;
while (t < PL_bufend && isSPACE(*t))
t++;
+ /* check for q => */
+ if (t+1 < PL_bufend && t[0] == '=' && t[1] == '>') {
+ OPERATOR(HASHBRACK);
+ }
term = *t;
open = term;
if (term && (tmps = strchr("([{< )]}> )]}>",term)))
else if (*t == open)
break;
}
- else
+ else {
for (t++; t < PL_bufend; t++) {
if (*t == '\\' && t+1 < PL_bufend)
t++;
else if (*t == open)
brackets++;
}
+ }
+ t++;
}
- t++;
+ else
+ /* skip plain q word */
+ while (t < PL_bufend && isALNUM_lazy_if(t,UTF))
+ t += UTF8SKIP(t);
}
else if (isALNUM_lazy_if(t,UTF)) {
t += UTF8SKIP(t);
PL_tokenbuf[0] = '@';
s = scan_ident(s, PL_bufend, PL_tokenbuf + 1, sizeof PL_tokenbuf - 1, FALSE);
if (!PL_tokenbuf[1]) {
- if (s == PL_bufend)
- yyerror("Final @ should be \\@ or @name");
PREREF('@');
}
if (PL_lex_state == LEX_NORMAL)
TERM(FUNC0SUB);
if (strEQ(proto, "$"))
OPERATOR(UNIOPSUB);
+ while (*proto == ';')
+ proto++;
if (*proto == '&' && *s == '{') {
sv_setpv(PL_subname, PL_curstash ?
"__ANON__" : "__ANON__::__ANON__");
}
#endif
#ifdef PERLIO_LAYERS
- if (UTF && !IN_BYTES)
- PerlIO_apply_layers(aTHX_ PL_rsfp, NULL, ":utf8");
+ if (!IN_BYTES) {
+ if (UTF)
+ PerlIO_apply_layers(aTHX_ PL_rsfp, NULL, ":utf8");
+ else if (PL_encoding) {
+ SV *name;
+ dSP;
+ ENTER;
+ SAVETMPS;
+ PUSHMARK(sp);
+ EXTEND(SP, 1);
+ XPUSHs(PL_encoding);
+ PUTBACK;
+ call_method("name", G_SCALAR);
+ SPAGAIN;
+ name = POPs;
+ PUTBACK;
+ PerlIO_apply_layers(aTHX_ PL_rsfp, NULL,
+ Perl_form(aTHX_ ":encoding(%"SVf")",
+ name));
+ FREETMPS;
+ LEAVE;
+ }
+ }
#endif
PL_rsfp = Nullfp;
}
S_pending_ident(pTHX)
{
register char *d;
- register I32 tmp;
+ register I32 tmp = 0;
/* pit holds the identifier we read and pending_ident is reset */
char pit = PL_pending_ident;
PL_pending_ident = 0;
*/
if (!strchr(PL_tokenbuf,':')) {
- if ((tmp = pad_findmy(PL_tokenbuf)) != NOT_IN_PAD) {
+ if (!PL_in_my)
+ tmp = pad_findmy(PL_tokenbuf);
+ if (tmp != NOT_IN_PAD) {
/* might be an "our" variable" */
if (PAD_COMPNAME_FLAGS(tmp) & SVpad_OUR) {
/* build ops for a bareword */
}
if (*s == '}') {
s++;
- if (PL_lex_state == LEX_INTERPNORMAL && !PL_lex_brackets)
+ if (PL_lex_state == LEX_INTERPNORMAL && !PL_lex_brackets) {
PL_lex_state = LEX_INTERPEND;
+ PL_expect = XREF;
+ }
if (funny == '#')
funny = '@';
if (PL_lex_state == LEX_NORMAL) {
Renew(SvPVX(tmpstr), SvLEN(tmpstr), char);
}
SvREFCNT_dec(herewas);
- if (UTF && !IN_BYTES && is_utf8_string((U8*)SvPVX(tmpstr), SvCUR(tmpstr)))
- SvUTF8_on(tmpstr);
+ if (!IN_BYTES) {
+ if (UTF && is_utf8_string((U8*)SvPVX(tmpstr), SvCUR(tmpstr)))
+ SvUTF8_on(tmpstr);
+ else if (PL_encoding)
+ sv_recode_to_utf8(tmpstr, PL_encoding);
+ }
PL_lex_stuff = tmpstr;
yylval.ival = op_type;
return s;
register char *to; /* current position in the sv's data */
I32 brackets = 1; /* bracket nesting level */
bool has_utf8 = FALSE; /* is there any utf8 content? */
+ I32 termcode; /* terminating char. code */
+ U8 termstr[UTF8_MAXLEN]; /* terminating string */
+ STRLEN termlen; /* length of terminating string */
+ char *last = NULL; /* last position for nesting bracket */
/* skip space before the delimiter */
if (isSPACE(*s))
/* after skipping whitespace, the next character is the terminator */
term = *s;
- if (!UTF8_IS_INVARIANT((U8)term) && UTF)
- has_utf8 = TRUE;
+ if (!UTF) {
+ termcode = termstr[0] = term;
+ termlen = 1;
+ }
+ else {
+ termcode = utf8_to_uvchr((U8*)s, &termlen);
+ Copy(s, termstr, termlen, U8);
+ if (!UTF8_IS_INVARIANT(term))
+ has_utf8 = TRUE;
+ }
/* mark where we are */
PL_multi_start = CopLINE(PL_curcop);
/* find corresponding closing delimiter */
if (term && (tmps = strchr("([{< )]}> )]}>",term)))
- term = tmps[5];
+ termcode = termstr[0] = term = tmps[5];
+
PL_multi_close = term;
/* create a new SV to hold the contents. 87 is leak category, I'm
assuming. 79 is the SV's initial length. What a random number. */
sv = NEWSV(87,79);
sv_upgrade(sv, SVt_PVIV);
- SvIVX(sv) = term;
+ SvIVX(sv) = termcode;
(void)SvPOK_only(sv); /* validate pointer */
/* move past delimiter and try to read a complete string */
if (keep_delims)
- sv_catpvn(sv, s, 1);
- s++;
+ sv_catpvn(sv, s, termlen);
+ s += termlen;
for (;;) {
+ if (PL_encoding && !UTF) {
+ bool cont = TRUE;
+
+ while (cont) {
+ int offset = s - SvPVX(PL_linestr);
+ bool found = sv_cat_decode(sv, PL_encoding, PL_linestr,
+ &offset, (char*)termstr, termlen);
+ char *ns = SvPVX(PL_linestr) + offset;
+ char *svlast = SvEND(sv) - 1;
+
+ for (; s < ns; s++) {
+ if (*s == '\n' && !PL_rsfp)
+ CopLINE_inc(PL_curcop);
+ }
+ if (!found)
+ goto read_more_line;
+ else {
+ /* handle quoted delimiters */
+ if (*(svlast-1) == '\\') {
+ char *t;
+ for (t = svlast-2; t >= SvPVX(sv) && *t == '\\';)
+ t--;
+ if ((svlast-1 - t) % 2) {
+ if (!keep_quoted) {
+ *(svlast-1) = term;
+ *svlast = '\0';
+ SvCUR_set(sv, SvCUR(sv) - 1);
+ }
+ continue;
+ }
+ }
+ if (PL_multi_open == PL_multi_close) {
+ cont = FALSE;
+ }
+ else {
+ char *t, *w;
+ if (!last)
+ last = SvPVX(sv);
+ for (w = t = last; t < svlast; w++, t++) {
+ /* At here, all closes are "was quoted" one,
+ so we don't check PL_multi_close. */
+ if (*t == '\\') {
+ if (!keep_quoted && *(t+1) == PL_multi_open)
+ t++;
+ else
+ *w++ = *t++;
+ }
+ else if (*t == PL_multi_open)
+ brackets++;
+
+ *w = *t;
+ }
+ if (w < t) {
+ *w++ = term;
+ *w = '\0';
+ SvCUR_set(sv, w - SvPVX(sv));
+ }
+ last = w;
+ if (--brackets <= 0)
+ cont = FALSE;
+ }
+ }
+ }
+ if (!keep_delims) {
+ SvCUR_set(sv, SvCUR(sv) - 1);
+ *SvEND(sv) = '\0';
+ }
+ break;
+ }
+
/* extend sv if need be */
SvGROW(sv, SvCUR(sv) + (PL_bufend - s) + 1);
/* set 'to' to the next character in the sv's string */
}
/* terminate when run out of buffer (the for() condition), or
have found the terminator */
- else if (*s == term)
- break;
+ else if (*s == term) {
+ if (termlen == 1)
+ break;
+ if (s+termlen <= PL_bufend && memEQ(s, (char*)termstr, termlen))
+ break;
+ }
else if (!has_utf8 && !UTF8_IS_INVARIANT((U8)*s) && UTF)
has_utf8 = TRUE;
*to = *s;
to[-1] = '\n';
#endif
+ read_more_line:
/* if we're out of file, or a read fails, bail and reset the current
line marker so we can report where the unterminated string began
*/
/* at this point, we have successfully read the delimited string */
- if (keep_delims)
- sv_catpvn(sv, s, 1);
- if (has_utf8)
+ if (!PL_encoding || UTF) {
+ if (keep_delims)
+ sv_catpvn(sv, s, termlen);
+ s += termlen;
+ }
+ if (has_utf8 || PL_encoding)
SvUTF8_on(sv);
- else if (PL_encoding)
- sv_recode_to_utf8(sv, PL_encoding);
PL_multi_end = CopLINE(PL_curcop);
- s++;
/* if we allocated too much space, give some back */
if (SvCUR(sv) + 5 < SvLEN(sv)) {
}
else
PL_lex_state = LEX_FORMLINE;
+ if (!IN_BYTES) {
+ if (UTF && is_utf8_string((U8*)SvPVX(stuff), SvCUR(stuff)))
+ SvUTF8_on(stuff);
+ else if (PL_encoding)
+ sv_recode_to_utf8(stuff, PL_encoding);
+ }
PL_nextval[PL_nexttoke].opval = (OP*)newSVOP(OP_CONST, 0, stuff);
force_next(THING);
PL_nextval[PL_nexttoke].ival = OP_FORMLINE;