akr	2008-01-06 21:15:48 +0900 (Sun, 06 Jan 2008)

  New Revision: 14911

  Modified files:
    trunk/ChangeLog
    trunk/re.c
    trunk/test/ruby/test_m17n.rb

  Log:
    * re.c (rb_reg_initialize_str): forbid raw non ASCII character
      for ASCII-8BIT regexp in non ASCII-8BIT script.


  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/ChangeLog?r1=14911&r2=14910&diff_format=u
  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/re.c?r1=14911&r2=14910&diff_format=u
  http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/test/ruby/test_m17n.rb?r1=14911&r2=14910&diff_format=u

Index: re.c
===================================================================
--- re.c	(revision 14910)
+++ re.c	(revision 14911)
@@ -2037,8 +2037,16 @@
 {
     int ret;
     rb_encoding *enc = rb_enc_get(str);
-    if (options & ARG_ENCODING_NONE)
-        enc = rb_ascii8bit_encoding();
+    if (options & ARG_ENCODING_NONE) {
+        rb_encoding *ascii8bit = rb_ascii8bit_encoding();
+        if (enc != ascii8bit) {
+            if (rb_enc_str_coderange(str) != ENC_CODERANGE_7BIT) {
+                strcpy(err, "/.../n has a non escaped non ASCII character in non ASCII-8BIT script");
+                return -1;
+            }
+            enc = ascii8bit;
+        }
+    }
     ret = rb_reg_initialize(obj, RSTRING_PTR(str), RSTRING_LEN(str), enc,
 			    options, err);
     RB_GC_GUARD(str);
Index: ChangeLog
===================================================================
--- ChangeLog	(revision 14910)
+++ ChangeLog	(revision 14911)
@@ -1,3 +1,8 @@
+Sun Jan  6 21:14:12 2008  Tanaka Akira  <akr@f...>
+
+	* re.c (rb_reg_initialize_str): forbid raw non ASCII character
+	  for ASCII-8BIT regexp in non ASCII-8BIT script.
+
 Sun Jan  6 18:19:12 2008  Tanaka Akira  <akr@f...>
 
 	* include/ruby/encoding.h (rb_enc_str_buf_cat): declared.
Index: test/ruby/test_m17n.rb
===================================================================
--- test/ruby/test_m17n.rb	(revision 14910)
+++ test/ruby/test_m17n.rb	(revision 14911)
@@ -320,6 +320,8 @@
     assert_regexp_fixed_ascii8bit(eval(a(%{/\xc2\xa1/n})))
     assert_regexp_fixed_ascii8bit(eval(a(%q{/\xc2\xa1/})))
 
+    assert_raise(SyntaxError) { eval("/\xa1\xa1/n".force_encoding("euc-jp")) }
+
     [/\xc2\xa1/n, eval(a(%{/\xc2\xa1/})), eval(a(%{/\xc2\xa1/n}))].each {|r|
       assert_equal(nil, r =~ a("a"))
       assert_equal(nil, r =~ e("a"))

--
ML: ruby-changes@q...
Info: http://www.atdot.net/~ko1/quickml