* re.c (rb_reg_preprocess): force fixed encoding when ASCII

  incompatible source string.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@15260 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

3 files changed, 18 insertions, 1 deletions

diff --git a/ChangeLog b/ChangeLog
index c74745c496..6687baac1e 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+Sun Jan 27 05:56:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_preprocess): force fixed encoding when ASCII
+	  incompatible source string.
+
 Sat Jan 26 23:46:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
 
 	* sprintf.c (rb_str_format): zero-precision zero should be empty.
diff --git a/re.c b/re.c
index a03bea5751..0237319076 100644
--- a/re.c
+++ b/re.c
@@ -1908,7 +1908,13 @@ rb_reg_preprocess(const char *p, const char *end, rb_encoding *enc,
 
     buf = rb_str_buf_new(0);
 
-    *fixed_enc = 0;
+    if (rb_enc_asciicompat(enc))
+        *fixed_enc = 0;
+    else {
+        *fixed_enc = enc;
+        rb_enc_associate(buf, enc);
+    }
+
     if (unescape_nonascii(p, end, enc, buf, fixed_enc, err) != 0)
         return Qnil;
 
diff --git a/test/ruby/test_utf16.rb b/test/ruby/test_utf16.rb
index 0a4bd4586c..aeb9cea133 100644
--- a/test/ruby/test_utf16.rb
+++ b/test/ruby/test_utf16.rb
@@ -115,4 +115,10 @@ class TestUTF16 < Test::Unit::TestCase
   def test_regexp_union
     enccall(Regexp, :union, "aa".force_encoding("utf-16be"), "bb".force_encoding("utf-16be"))
   end
+
+  def test_empty_regexp
+    s = "".force_encoding("utf-16be")
+    assert_equal(Encoding.find("utf-16be"), Regexp.new(s).encoding,
+                "Regexp.new(#{encdump s}).encoding")
+  end
 end