Added characters that look like padlocks to net IDN character blacklist.

net/base/net_util_icu_unittest.cc

 // Copyright 2014 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 <string.h>
 
 #include <vector>
 
@@ skipping 313 matching lines @@
      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,
   }},
+  // Padlock icon spoof.
+  {"xn--google-hj64e", L"\U0001f512google.com",
+    {false, false, false, false, false,
+     false, false, false, false, false,
+     false, false, false, false, false,
+     false, false, false, false, false,
+  }},
+  // Ensure that by blacklisting "\xd83d\xdd12", we have not inadvertently
+  // blacklisted all strings with the surrogate '\xdd12'.

[mmenke, 2015/06/05 15:08:06]

nit:  Don't use "we" in comments

[Matt Giuca, 2015/06/09 04:10:35]

Hmm, not a fan of this rule (it means rewriting active sentences as passive
sentences, and this one is a pretty bad one). But OK.

Also, the word "we" already appears immediately above, and 17 times in
net_util_icu.cc.

+  {"xn--fk9c.com", L"\U00010912.com",
+    {true,  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 71 matching lines @@
   for (size_t i = 0; i < arraysize(idn_cases); i++) {
     for (size_t j = 0; j < arraysize(kLanguages); j++) {
       // ja || zh-TW,en || ko,ja -> IDNToUnicodeSlow
       if (j == 3 || j == 17 || j == 18)
         continue;
       base::string16 output(IDNToUnicode(idn_cases[i].input, kLanguages[j]));
       base::string16 expected(idn_cases[i].unicode_allowed[j] ?
           WideToUTF16(idn_cases[i].unicode_output) :
           ASCIIToUTF16(idn_cases[i].input));
       AppendLanguagesToOutputs(kLanguages[j], &expected, &output);
-      EXPECT_EQ(expected, output);
+      EXPECT_EQ(expected, output) << "input: \"" << idn_cases[i].input
+                                  << "\", languages: \"" << kLanguages[j]
+                                  << "\"";
     }
   }
 }
 
 TEST(NetUtilTest, IDNToUnicodeSlow) {
   for (size_t i = 0; i < arraysize(idn_cases); i++) {
     for (size_t j = 0; j < arraysize(kLanguages); j++) {
       // !(ja || zh-TW,en || ko,ja) -> IDNToUnicodeFast
       if (!(j == 3 || j == 17 || j == 18))
         continue;
       base::string16 output(IDNToUnicode(idn_cases[i].input, kLanguages[j]));
       base::string16 expected(idn_cases[i].unicode_allowed[j] ?
           WideToUTF16(idn_cases[i].unicode_output) :
           ASCIIToUTF16(idn_cases[i].input));
       AppendLanguagesToOutputs(kLanguages[j], &expected, &output);
-      EXPECT_EQ(expected, output);
+      EXPECT_EQ(expected, output) << "input: \"" << idn_cases[i].input
+                                  << "\", languages: \"" << kLanguages[j]
+                                  << "\"";
     }
   }
 }
 
 // ulocdata_getExemplarSet may fail with some locales (currently bn, gu, and
 // te), which was causing a crash (See http://crbug.com/510551).  This may be an
 // icu bug, but regardless, that should not cause a crash.
 TEST(NetUtilTest, IDNToUnicodeNeverCrashes) {
   for (char c1 = 'a'; c1 <= 'z'; c1++) {
     for (char c2 = 'a'; c2 <= 'z'; c2++) {
@@ skipping 623 matching lines @@
 
   const size_t omit_all_offsets[] = {
     0, kNpos, kNpos, kNpos, kNpos, kNpos, kNpos, 0, kNpos, kNpos, kNpos, kNpos,
     0, 1, 2, 3, 4, 5, 6, 7
   };
   CheckAdjustedOffsets("http://user@foo.com/", "en", kFormatUrlOmitAll,
                        UnescapeRule::NORMAL, omit_all_offsets);
 }
 
 }  // namespace net