*
**** Alterations to Henry's code are...
****
- **** 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.
char *i_strpos = strpos;
SV *dsv = PERL_DEBUG_PAD_ZERO(0);
#endif
+ RX_MATCH_UTF8_set(prog,do_utf8);
if (prog->reganch & ROPT_UTF8) {
DEBUG_r(PerlIO_printf(Perl_debug_log,
start_shift + (s - strbeg), end_shift, pp, 0);
else
goto fail_finish;
+ /* we may be pointing at the wrong string */
+ if (s && RX_MATCH_COPIED(prog))
+ s = strbeg + (s - SvPVX(sv));
if (data)
*data->scream_olds = s;
}
t = s - prog->check_offset_max;
if (s - strpos > prog->check_offset_max /* signed-corrected t > strpos */
- && (!(prog->reganch & ROPT_UTF8)
+ && (!do_utf8
|| ((t = reghopmaybe3_c(s, -(prog->check_offset_max), strpos))
&& t > strpos)))
/* EMPTY */;
t = s - prog->check_offset_max;
if (s - strpos > prog->check_offset_max /* signed-corrected t > strpos */
- && (!(prog->reganch & ROPT_UTF8)
+ && (!do_utf8
|| ((t = reghopmaybe3_c(s, -prog->check_offset_max, strpos))
&& t > strpos))) {
/* Fixed substring is found far enough so that the match
char *startpos = strbeg;
t = s;
- if (prog->reganch & ROPT_UTF8) {
- PL_regdata = prog->data;
- PL_bostr = startpos;
- }
+ cache_re(prog);
s = find_byclass(prog, prog->regstclass, s, endpos, startpos, 1);
if (!s) {
#ifdef DEBUGGING
/* We know what class it must start with. */
switch (OP(c)) {
case ANYOF:
- while (s < strend) {
- STRLEN skip = do_utf8 ? UTF8SKIP(s) : 1;
-
- if (do_utf8 ?
- reginclass(c, (U8*)s, 0, do_utf8) :
- REGINCLASS(c, (U8*)s) ||
- (ANYOF_FOLD_SHARP_S(c, s, strend) &&
- /* The assignment of 2 is intentional:
- * for the sharp s, the skip is 2. */
- (skip = SHARP_S_SKIP)
- )) {
- if (tmp && (norun || regtry(prog, s)))
- goto got_it;
- else
- tmp = doevery;
- }
- else
- tmp = 1;
- s += skip;
+ if (do_utf8) {
+ while (s < strend) {
+ if ((ANYOF_FLAGS(c) & ANYOF_UNICODE) ||
+ !UTF8_IS_INVARIANT((U8)s[0]) ?
+ reginclass(c, (U8*)s, 0, do_utf8) :
+ REGINCLASS(c, (U8*)s)) {
+ if (tmp && (norun || regtry(prog, s)))
+ goto got_it;
+ else
+ tmp = doevery;
+ }
+ else
+ tmp = 1;
+ s += UTF8SKIP(s);
+ }
+ }
+ else {
+ while (s < strend) {
+ STRLEN skip = 1;
+
+ if (REGINCLASS(c, (U8*)s) ||
+ (ANYOF_FOLD_SHARP_S(c, s, strend) &&
+ /* The assignment of 2 is intentional:
+ * for the folded sharp s, the skip is 2. */
+ (skip = SHARP_S_SKIP))) {
+ if (tmp && (norun || regtry(prog, s)))
+ goto got_it;
+ else
+ tmp = doevery;
+ }
+ else
+ tmp = 1;
+ s += skip;
+ }
}
break;
case CANY:
SV *dsv0 = PERL_DEBUG_PAD_ZERO(0);
SV *dsv1 = PERL_DEBUG_PAD_ZERO(1);
#endif
+ RX_MATCH_UTF8_set(prog,do_utf8);
PL_regcc = 0;
: (s = fbm_instr((unsigned char*)HOP3(s, back_min, strend),
(unsigned char*)strend, must,
PL_multiline ? FBMrf_MULTILINE : 0))) ) {
+ /* we may be pointing at the wrong string */
+ if ((flags & REXEC_SCREAM) && RX_MATCH_COPIED(prog))
+ s = strbeg + (s - SvPVX(sv));
DEBUG_r( did_match = 1 );
if (HOPc(s, -back_max) > last1) {
last1 = HOPc(s, -back_min);
goto phooey;
}
else if ((c = prog->regstclass)) {
- if (minlen && PL_regkind[(U8)OP(prog->regstclass)] != EXACT)
+ if (minlen) {
+ I32 op = (U8)OP(prog->regstclass);
/* don't bother with what can't match */
- strend = HOPc(strend, -(minlen - 1));
+ if (PL_regkind[op] != EXACT && op != CANY)
+ strend = HOPc(strend, -(minlen - 1));
+ }
DEBUG_r({
SV *prop = sv_newmortal();
char *s0;
end_shift, &scream_pos, 1); /* last one */
if (!last)
last = scream_olds; /* Only one occurrence. */
+ /* we may be pointing at the wrong string */
+ else if (RX_MATCH_COPIED(prog))
+ s = strbeg + (s - SvPVX(sv));
}
else {
STRLEN len;
/* make sure $`, $&, $', and $digit will work later */
if ( !(flags & REXEC_NOT_FIRST) ) {
- if (RX_MATCH_COPIED(prog)) {
- Safefree(prog->subbeg);
- RX_MATCH_COPIED_off(prog);
- }
+ RX_MATCH_COPY_FREE(prog);
if (flags & REXEC_COPY_STR) {
I32 i = PL_regeol - startpos + (stringarg - strbeg);
-
- s = savepvn(strbeg, i);
- prog->subbeg = s;
+#ifdef PERL_COPY_ON_WRITE
+ if ((SvIsCOW(sv)
+ || (SvFLAGS(sv) & CAN_COW_MASK) == CAN_COW_FLAGS)) {
+ if (DEBUG_C_TEST) {
+ PerlIO_printf(Perl_debug_log,
+ "Copy on write: regexp capture, type %d\n",
+ (int) SvTYPE(sv));
+ }
+ prog->saved_copy = sv_setsv_cow(prog->saved_copy, sv);
+ prog->subbeg = SvPVX(prog->saved_copy);
+ assert (SvPOKp(prog->saved_copy));
+ } else
+#endif
+ {
+ RX_MATCH_COPIED_on(prog);
+ s = savepvn(strbeg, i);
+ prog->subbeg = s;
+ }
prog->sublen = i;
- RX_MATCH_COPIED_on(prog);
}
else {
prog->subbeg = strbeg;
$` inside (?{}) could fail... */
PL_reg_oldsaved = prog->subbeg;
PL_reg_oldsavedlen = prog->sublen;
+#ifdef PERL_COPY_ON_WRITE
+ PL_nrs = prog->saved_copy;
+#endif
RX_MATCH_COPIED_off(prog);
}
else
PL_reglastparen = &prog->lastparen;
PL_reglastcloseparen = &prog->lastcloseparen;
prog->lastparen = 0;
+ prog->lastcloseparen = 0;
PL_regsize = 0;
DEBUG_r(PL_reg_starttry = startpos);
if (PL_reg_start_tmpl <= prog->nparens) {
regprop(prop, scan);
{
char *s0 =
- do_utf8 ?
+ do_utf8 && OP(scan) != CANY ?
pv_uni_display(dsv0, (U8*)(locinput - pref_len),
pref0_len, 60, UNI_DISPLAY_REGEX) :
locinput - pref_len;
int len0 = do_utf8 ? strlen(s0) : pref0_len;
- char *s1 = do_utf8 ?
+ char *s1 = do_utf8 && OP(scan) != CANY ?
pv_uni_display(dsv1, (U8*)(locinput - pref_len + pref0_len),
pref_len - pref0_len, 60, UNI_DISPLAY_REGEX) :
locinput - pref_len + pref0_len;
int len1 = do_utf8 ? strlen(s1) : pref_len - pref0_len;
- char *s2 = do_utf8 ?
+ char *s2 = do_utf8 && OP(scan) != CANY ?
pv_uni_display(dsv2, (U8*)locinput,
PL_regeol - locinput, 60, UNI_DISPLAY_REGEX) :
locinput;
re_cc_state state;
CHECKPOINT cp, lastcp;
int toggleutf;
+ register SV *sv;
- if(SvROK(ret) || SvRMAGICAL(ret)) {
- SV *sv = SvROK(ret) ? SvRV(ret) : ret;
-
- if(SvMAGICAL(sv))
- mg = mg_find(sv, PERL_MAGIC_qr);
+ if(SvROK(ret) && SvSMAGICAL(sv = SvRV(ret)))
+ mg = mg_find(sv, PERL_MAGIC_qr);
+ else if (SvSMAGICAL(ret)) {
+ if (SvGMAGICAL(ret))
+ sv_unmagic(ret, PERL_MAGIC_qr);
+ else
+ mg = mg_find(ret, PERL_MAGIC_qr);
}
+
if (mg) {
re = (regexp *)mg->mg_obj;
(void)ReREFCNT_inc(re);
if (DO_UTF8(ret)) pm.op_pmdynflags |= PMdf_DYN_UTF8;
re = CALLREGCOMP(aTHX_ t, t + len, &pm);
if (!(SvFLAGS(ret)
- & (SVs_TEMP | SVs_PADTMP | SVf_READONLY)))
+ & (SVs_TEMP | SVs_PADTMP | SVf_READONLY
+ | SVs_GMG)))
sv_magic(ret,(SV*)ReREFCNT_inc(re),
PERL_MAGIC_qr,0,0);
PL_regprecomp = oprecomp;
if (PL_reg_oldsaved) {
PL_reg_re->subbeg = PL_reg_oldsaved;
PL_reg_re->sublen = PL_reg_oldsavedlen;
+#ifdef PERL_COPY_ON_WRITE
+ PL_reg_re->saved_copy = PL_nrs;
+#endif
RX_MATCH_COPIED_on(PL_reg_re);
}
PL_reg_magic->mg_len = PL_reg_oldpos;