make intuit_start() handle mixed utf8-ness

Fix a bug in intuit_start() that makes it fail when the utf8-ness of the
string and pattern differ. This was mostly masked, since pp_match() skips
calling intuit in this case (and has done since 2000, presumably as a
workaround for this issue, and possibly for other issues since fixed).
But pp_subst() didn't skip, so code like this would fail:

    $c = "\x{c0}";
    utf8::upgrade($c);
    print "ok\n" if $c =~ s/\xC0{1,2}$/\xC0/i;

Now that intuit is (hopefully) fixed, also remove the guard in pp_match().

diff --git a/pp_hot.c b/pp_hot.c
index 07a7893..9994280 100644 (file)
--- a/pp_hot.c
+++ b/pp_hot.c
@@ -1432,8 +1432,7 @@ PP(pp_match)
        if (update_minmatch++)
            minmatch = had_zerolen;
     }
-    if (RX_EXTFLAGS(rx) & RXf_USE_INTUIT &&
-       DO_UTF8(TARG) == (RX_UTF8(rx) != 0)) {
+    if (RX_EXTFLAGS(rx) & RXf_USE_INTUIT) {
        s = CALLREG_INTUIT_START(rx, TARG, truebase,
                         (char *)s, (char *)strend, r_flags, NULL);
 

diff --git a/regexec.c b/regexec.c
index 0b27961..dc4d288 100644 (file)
--- a/regexec.c
+++ b/regexec.c
@@ -1112,8 +1112,11 @@ Perl_re_intuit_start(pTHX_
        /* If regstclass takes bytelength more than 1: If charlength==1, OK.
           This leaves EXACTF-ish only, which are dealt with in find_byclass().  */
         const U8* const str = (U8*)STRING(progi->regstclass);
+        /* XXX this value could be pre-computed */
         const int cl_l = (PL_regkind[OP(progi->regstclass)] == EXACT
-                   ? CHR_DIST(str+STR_LEN(progi->regstclass), str)
+                   ?  (reginfo->is_utf8_pat
+                        ? utf8_distance(str + STR_LEN(progi->regstclass), str)
+                        : STR_LEN(progi->regstclass))
                    : 1);
        char * endpos;
        if (prog->anchored_substr || prog->anchored_utf8 || ml_anch)