Add URL filter to trigger punycode for Japanese homographic sequences....

net/base/net_util_unittest.cc

 // Copyright (c) 2010 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "net/base/net_util.h"
 
 #include "base/file_path.h"
 #include "base/format_macros.h"
 #include "base/string_util.h"
 #include "base/sys_string_conversions.h"
@@ skipping 110 matching lines @@
     false, false, false, false, false,
     false, false, false, true,  false,
     false}},
   // Katakana including U+30FC
   {"xn--tckm4i2e.jp", L"\x30b3\x30de\x30fc\x30b9.jp",
    {true, false, false, true,  false,
     false, false, false, false, false,
     false, false, false, false, false,
     false, false, false, true, false,
     }},
+  {"xn--3ck7a7g.jp", L"\u30ce\u30f3\u30bd.jp",
+   {true, false, false, true,  false,
+    false, false, false, false, false,
+    false, false, false, false, false,
+    false, false, false, true, false,
+    }},
   // Katakana + Latin (Japanese)
   // TODO(jungshik): Change 'false' in the first element to 'true'
   // after upgrading to ICU 4.2.1 to use new uspoof_* APIs instead
   // of our IsIDNComponentInSingleScript().
   {"xn--e-efusa1mzf.jp", L"e\x30b3\x30de\x30fc\x30b9.jp",
    {false, false, false, true,  false,
     false, false, false, false, false,
     false, false, false, false, false,
     false, false, false, true, false,
     }},
+  {"xn--3bkxe.jp", L"\x30c8\x309a.jp",
+   {false, false, false, true,  false,
+    false, false, false, false, false,
+    false, false, false, false, false,
+    false, false, false, true, false,
+    }},
   // Hangul (Korean)
   {"www.xn--or3b17p6jjc.kr", L"www.\xc804\xc790\xc815\xbd80.kr",
    {true,  false, false, false, true,
     false, false, false, false, false,
     false, false, false, false, false,
     false, false, false, true,  false,
     false}},
   // b<u-umlaut>cher (German)
   {"xn--bcher-kva.de", L"b\x00fc" L"cher.de",
    {true,  false, false, false, false,
@@ skipping 164 matching lines @@
     {false, false, false, false, false,
      false, false, false, false, false,
      false, false, false, false, false,
      false, false, false, false, false,
   }},
   {"google.xn--comabc-k8d", L"google.com\x0338" L"abc",
     {false, false, false, false, false,
      false, false, false, false, false,
      false, false, false, false, false,
      false, false, false, false, false,
-     }},
+  }},
+  {"google.xn--com-oh4ba.evil.jp", L"google.com\x309a\x309a.evil.jp",
+    {false, false, false, false, false,
+     false, false, false, false, false,
+     false, false, false, false, false,
+     false, false, false, false, false,
+  }},
+  {"google.xn--comevil-v04f.jp", L"google.com\x30ce" L"evil.jp",
+    {false, false, false, false, false,
+     false, false, false, false, false,
+     false, false, false, false, false,
+     false, false, false, false, false,
+  }},
 #if 0
   // These two cases are special. We need a separate test.
   // U+3000 and U+3002 are normalized to ASCII space and dot.
   {"xn-- -kq6ay5z.cn", L"\x4e2d\x56fd\x3000.cn",
     {false, false, true,  false, false,
      false, false, false, false, false,
      false, false, false, false, false,
      false, false, true,  false, false,
      true}},
   {"xn--fiqs8s.cn", L"\x4e2d\x56fd\x3002" L"cn",
@@ skipping 1661 matching lines @@
     EXPECT_TRUE(net::ParseCIDRBlock(tests[i].cidr_literal,
                                     &ip_prefix,
                                     &prefix_length_in_bits));
 
     EXPECT_EQ(tests[i].expected_to_match,
               net::IPNumberMatchesPrefix(ip_number,
                                          ip_prefix,
                                          prefix_length_in_bits));
   }
 }