(U8*)(off >= 0 ? reginfo->strend : reginfo->strbeg)) \
: (U8*)(pos + off))
-#define HOPBACKc(pos, off) \
- (char*)(reginfo->is_utf8_target \
- ? reghopmaybe3((U8*)pos, (SSize_t)0-off, (U8*)(reginfo->strbeg)) \
- : (pos - off >= reginfo->strbeg) \
- ? (U8*)pos - off \
+/* like HOPMAYBE3 but backwards. lim must be +ve. Returns NULL on overshoot */
+#define HOPBACK3(pos, off, lim) \
+ (reginfo->is_utf8_target \
+ ? reghopmaybe3((U8*)pos, (SSize_t)0-off, (U8*)(lim)) \
+ : (pos - off >= lim) \
+ ? (U8*)pos - off \
: NULL)
+#define HOPBACKc(pos, off) ((char*)HOPBACK3(pos, off, reginfo->strbeg))
+
#define HOP3(pos,off,lim) (reginfo->is_utf8_target ? reghop3((U8*)(pos), off, (U8*)(lim)) : (U8*)(pos + off))
#define HOP3c(pos,off,lim) ((char*)HOP3(pos,off,lim))
(IV)prog->check_end_shift);
});
- end_point = HOP3(strend, -end_shift, strbeg);
+ end_point = HOPBACK3(strend, end_shift, rx_origin);
+ if (!end_point)
+ goto fail_finish;
start_point = HOPMAYBE3(rx_origin, start_shift, end_point);
if (!start_point)
goto fail_finish;
/* Do not break within emoji flag sequences. That is, do not
* break between regional indicator (RI) symbols if there is an
* odd number of RI characters before the break point.
- * GB12 ^ (RI RI)* RI × RI
+ * GB12 sot (RI RI)* RI × RI
* GB13 [^RI] (RI RI)* RI × RI */
while (backup_one_GCB(strbeg,
* only if there are an even number of regional indicators
* preceding the position of the break.
*
- * sot (RI RI)* RI × RI
+ * sot (RI RI)* RI × RI
* [^RI] (RI RI)* RI × RI */
while (backup_one_LB(strbeg,
* odd number of RI characters before the potential break
* point.
*
- * WB15 ^ (RI RI)* RI × RI
+ * WB15 sot (RI RI)* RI × RI
* WB16 [^RI] (RI RI)* RI × RI */
while (backup_one_WB(&previous,
U8 gimme = G_SCALAR;
CV *caller_cv = NULL; /* who called us */
CV *last_pushed_cv = NULL; /* most recently called (?{}) CV */
- CHECKPOINT runops_cp; /* savestack position before executing EVAL */
U32 maxopenparen = 0; /* max '(' index seen so far */
int to_complement; /* Invert the result? */
_char_class_number classnum;
if (scan->flags == EXACTL || scan->flags == EXACTFLU8) {
_CHECK_AND_WARN_PROBLEMATIC_LOCALE;
if (utf8_target
+ && nextchr >= 0 /* guard against negative EOS value in nextchr */
&& UTF8_IS_ABOVE_LATIN1(nextchr)
&& scan->flags == EXACTL)
{
{
U8 *uc;
if ( ST.jump ) {
+ /* undo any captures done in the tail part of a branch,
+ * e.g.
+ * /(?:X(.)(.)|Y(.)).../
+ * where the trie just matches X then calls out to do the
+ * rest of the branch */
REGCP_UNWIND(ST.cp);
UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
}
goto eval_recurse_doit;
/* NOTREACHED */
- case EVAL: /* /(?{A})B/ /(??{A})B/ and /(?(?{A})X|Y)B/ */
+ case EVAL: /* /(?{...})B/ /(??{A})B/ and /(?(?{...})X|Y)B/ */
if (cur_eval && cur_eval->locinput==locinput) {
if ( ++nochange_depth > max_nochange_depth )
Perl_croak(aTHX_ "EVAL without pos change exceeded limit in regex");
/* save *all* paren positions */
regcppush(rex, 0, maxopenparen);
- REGCP_SET(runops_cp);
+ REGCP_SET(ST.lastcp);
if (!caller_cv)
caller_cv = find_runcv(NULL);
/* Some notes about MULTICALL and the context and save stacks.
*
* In something like
- * /...(?{ my $x)}...(?{ my $z)}...(?{ my $z)}.../
+ * /...(?{ my $x)}...(?{ my $y)}...(?{ my $z)}.../
* since codeblocks don't introduce a new scope (so that
* local() etc accumulate), at the end of a successful
* match there will be a SAVEt_CLEARSV on the savestack
* *everything* SAVEd by this function is undone and then
* the context popped, rather than e.g., popping the
* context (and restoring the original PL_comppad) then
- * popping more of the savestack and restoiring a bad
+ * popping more of the savestack and restoring a bad
* PL_comppad.
*/
* in the regexp code uses the pad ! */
PL_op = oop;
PL_curcop = ocurcop;
- regcp_restore(rex, runops_cp, &maxopenparen);
+ regcp_restore(rex, ST.lastcp, &maxopenparen);
PL_curpm_under = PL_curpm;
PL_curpm = PL_reg_curpm;
- if (logical != 2)
- break;
+ if (logical != 2) {
+ PUSH_STATE_GOTO(EVAL_B, next, locinput);
+ /* NOTREACHED */
+ }
}
/* only /(??{})/ from now on */
ST.prev_eval = cur_eval;
cur_eval = st;
/* now continue from first node in postoned RE */
- PUSH_YES_STATE_GOTO(EVAL_AB, startpoint, locinput);
+ PUSH_YES_STATE_GOTO(EVAL_postponed_AB, startpoint, locinput);
NOT_REACHED; /* NOTREACHED */
}
- case EVAL_AB: /* cleanup after a successful (??{A})B */
+ case EVAL_postponed_AB: /* cleanup after a successful (??{A})B */
/* note: this is called twice; first after popping B, then A */
DEBUG_STACK_r({
Perl_re_exec_indentf( aTHX_ "EVAL_AB cur_eval=%p prev_eval=%p\n",
sayYES;
- case EVAL_AB_fail: /* unsuccessfully ran A or B in (??{A})B */
+ case EVAL_B_fail: /* unsuccessful B in (?{...})B */
+ REGCP_UNWIND(ST.lastcp);
+ sayNO;
+
+ case EVAL_postponed_AB_fail: /* unsuccessfully ran A or B in (??{A})B */
/* note: this is called twice; first after popping B, then A */
DEBUG_STACK_r({
Perl_re_exec_indentf( aTHX_ "EVAL_AB_fail cur_eval=%p prev_eval=%p\n",
if (cur_curlyx->u.curlyx.minmod) {
ST.save_curlyx = cur_curlyx;
cur_curlyx = cur_curlyx->u.curlyx.prev_curlyx;
- ST.cp = regcppush(rex, ST.save_curlyx->u.curlyx.parenfloor,
- maxopenparen);
- REGCP_SET(ST.lastcp);
PUSH_YES_STATE_GOTO(WHILEM_B_min, ST.save_curlyx->u.curlyx.B,
locinput);
NOT_REACHED; /* NOTREACHED */
CACHEsayNO;
NOT_REACHED; /* NOTREACHED */
- case WHILEM_A_min_fail: /* just failed to match A in a minimal match */
- /* FALLTHROUGH */
case WHILEM_A_pre_fail: /* just failed to match even minimal A */
REGCP_UNWIND(ST.lastcp);
regcppop(rex, &maxopenparen);
+ /* FALLTHROUGH */
+ case WHILEM_A_min_fail: /* just failed to match A in a minimal match */
cur_curlyx->u.curlyx.lastloc = ST.save_lastloc;
cur_curlyx->u.curlyx.count--;
CACHEsayNO;
case WHILEM_B_min_fail: /* just failed to match B in a minimal match */
cur_curlyx = ST.save_curlyx;
- REGCP_UNWIND(ST.lastcp);
- regcppop(rex, &maxopenparen);
if (cur_curlyx->u.curlyx.count >= /*max*/ARG2(cur_curlyx->u.curlyx.me)) {
/* Maximum greed exceeded */
);
/* Try grabbing another A and see if it helps. */
cur_curlyx->u.curlyx.lastloc = locinput;
- ST.cp = regcppush(rex, cur_curlyx->u.curlyx.parenfloor,
- maxopenparen);
- REGCP_SET(ST.lastcp);
PUSH_STATE_GOTO(WHILEM_A_min,
/*A*/ NEXTOPER(ST.save_curlyx->u.curlyx.me) + EXTRA_STEP_2ARGS,
locinput);
SET_RECURSE_LOCINPUT("FAKE-END[after]", cur_eval->locinput);
- PUSH_YES_STATE_GOTO(EVAL_AB, st->u.eval.prev_eval->u.eval.B,
+ PUSH_YES_STATE_GOTO(EVAL_postponed_AB, st->u.eval.prev_eval->u.eval.B,
locinput); /* match B */
}
DEBUG_STACK_r({
regmatch_state *cur = st;
regmatch_state *curyes = yes_state;
- int curd = depth;
+ U32 i;
regmatch_slab *slab = PL_regmatch_slab;
- for (;curd > -1 && (depth-curd < 3);cur--,curd--) {
+ for (i = 0; i < 3 && i <= depth; cur--,i++) {
if (cur < SLAB_FIRST(slab)) {
slab = slab->prev;
cur = SLAB_LAST(slab);
}
- Perl_re_exec_indentf( aTHX_ "#%-3d %-10s %s\n",
+ Perl_re_exec_indentf( aTHX_ "%4s #%-3d %-10s %s\n",
depth,
- curd, PL_reg_name[cur->resume_state],
+ i ? " " : "push",
+ depth - i, PL_reg_name[cur->resume_state],
(curyes == cur) ? "yes" : ""
);
if (curyes == cur)
return TRUE;
}
+#ifndef PERL_IN_XSUB_RE
+
bool
Perl__is_grapheme(pTHX_ const U8 * strbeg, const U8 * s, const U8 * strend, const UV cp)
{
return isGCB(cp_gcb_val, next_cp_gcb_val, strbeg, s, TRUE);
}
+#endif
https://perl5.git.perl.org / perl5.git / blobdiff