merges r28728 from trunk into ruby_1_9_2.

--
* re.c (rb_reg_expr_str): fixed out-of-boundary access at invalid
  multibyte characters.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_2@28748 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

2 files changed, 14 insertions, 6 deletions

diff --git a/ChangeLog b/ChangeLog
index 14b7117b9e..6fa83202a1 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+Fri Jul 23 09:02:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_expr_str): fixed out-of-boundary access at invalid
+	  multibyte characters.
+
 Thu Jul 22 16:27:41 2010  NARUSE, Yui  <naruse@ruby-lang.org>
 
 	* re.c (rb_reg_expr_str): fix broken Regexp#inspect when it
diff --git a/re.c b/re.c
index f0213235c9..8484c68ee2 100644
--- a/re.c
+++ b/re.c
@@ -370,18 +370,20 @@ rb_reg_expr_str(VALUE str, const char *s, long len,
 		rb_str_buf_cat(str, p, clen);
 	    }
 	    else if (c == -1) {
-                int l;
+		clen = rb_enc_precise_mbclen(p, pend, enc);
+		if (!MBCLEN_CHARFOUND_P(clen)) {
+		    c = (unsigned char)*p;
+		    clen = 1;
+		    goto hex;
+		}
 		if (resenc) {
 		    unsigned int c = rb_enc_mbc_to_codepoint(p, pend, enc);
-		    l = rb_enc_codelen(c, enc);
 		    rb_str_buf_cat_escaped_char(str, c, unicode_p);
 		}
 		else {
-		    l = mbclen(p, pend, enc);
-		    rb_str_buf_cat(str, p, l);
+		    clen = MBCLEN_CHARFOUND_LEN(clen);
+		    rb_str_buf_cat(str, p, clen);
 		}
-		p += l;
-		continue;
 	    }
 	    else if (rb_enc_isprint(c, enc)) {
 		rb_str_buf_cat(str, p, clen);
@@ -389,6 +391,7 @@ rb_reg_expr_str(VALUE str, const char *s, long len,
 	    else if (!rb_enc_isspace(c, enc)) {
 		char b[8];
 
+	      hex:
 		snprintf(b, sizeof(b), "\\x%02X", c);
 		rb_str_buf_cat(str, b, 4);
 	    }