Mitigate spoofing attempt using Latin letters.

components/url_formatter/top_domains/make_top_domain_gperf.cc

Index: components/url_formatter/top_domains/make_top_domain_gperf.cc
diff --git a/components/url_formatter/top_domains/make_top_domain_gperf.cc b/components/url_formatter/top_domains/make_top_domain_gperf.cc
new file mode 100644
index 0000000000000000000000000000000000000000..9a6d2049b872fc92d5db8265be7ee5d4a62690e0
--- /dev/null
+++ b/components/url_formatter/top_domains/make_top_domain_gperf.cc
@@ -0,0 +1,117 @@
+// Copyright (c) 2017 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 <algorithm>
+#include <iostream>
+#include <sstream>
+#include <string>
+#include <vector>
+
+#include "base/base_paths.h"
+#include "base/files/file_path.h"
+#include "base/files/file_util.h"
+#include "base/i18n/icu_util.h"
+#include "base/path_service.h"
+#include "base/strings/string_split.h"
+#include "base/strings/string_util.h"
+#include "third_party/icu/source/common/unicode/unistr.h"
+#include "third_party/icu/source/common/unicode/utypes.h"
+#include "third_party/icu/source/i18n/unicode/uspoof.h"
+
+std::string GetSkeleton(const std::string& domain,
+                        const USpoofChecker* spoof_checker) {
+  UErrorCode status = U_ZERO_ERROR;
+  icu::UnicodeString ustr_skeleton;
+  uspoof_getSkeletonUnicodeString(spoof_checker, 0, /* not used */

[Peter Kasting, 2017/05/10 22:38:47]

Nit: Suggest ',' after */ instead of before

[jungshik at Google, 2017/05/14 09:36:23]

Done.

+                                  icu::UnicodeString::fromUTF8(domain),
+                                  ustr_skeleton, &status);
+  std::string skeleton;
+  return U_SUCCESS(status) ? ustr_skeleton.toUTF8String(skeleton) : skeleton;
+}
+
+base::FilePath GetPath(base::StringPiece basename) {
+  base::FilePath path;
+  base::PathService::Get(base::DIR_SOURCE_ROOT, &path);
+  return path.Append(FILE_PATH_LITERAL("components"))
+      .Append(FILE_PATH_LITERAL("url_formatter"))

[Peter Kasting, 2017/05/10 22:38:47]

Nit: Is this how git cl format indented this?  I'd have expected dots aligned.

[jungshik at Google, 2017/05/14 09:36:23]

Yes, even if I get dots aligned manually, 'git cl format' put them back where
they used to be.

+      .Append(FILE_PATH_LITERAL("top_domains"))
+      .AppendASCII(basename);
+}
+
+void WriteToFile(const std::string& content, base::StringPiece basename) {
+  base::FilePath path = GetPath(basename);
+  bool succeeded = base::WriteFile(path, content.data(), content.size()) != -1;

[Peter Kasting, 2017/05/10 22:38:47]

Again, shouldn't this check "== content.size()"?

[jungshik at Google, 2017/05/14 09:36:23]

Done.

+  CHECK(succeeded) << "Failed to write to " << path.AsUTF8Unsafe() << '\n';

[Peter Kasting, 2017/05/10 22:38:47]

Nit: Maybe rather than CHECK, return |succeeded|, so the caller can cerr/return
nonzero.

[jungshik at Google, 2017/05/14 09:36:23]

Done.

+}
+
+int main(int argc, const char** argv) {
+  if (argc != 1) {
+    std::cerr << "Generates the list of top domain skeletons to use as input to"
+                 "\nbase/dafsa/make_dafsa.py.\nUsage: "
+              << argv[0] << '\n';
+    return 1;
+  }
+
+  base::i18n::InitializeICU();
+  base::FilePath input_file = GetPath("alexa_domains.list");
+  std::string input_content;
+  if (!base::ReadFileToString(input_file, &input_content)) {
+    std::cerr << "Failed to read the input file " << input_file.AsUTF8Unsafe()
+              << '\n';
+    return 1;
+  }
+
+  UErrorCode status = U_ZERO_ERROR;
+  USpoofChecker* spoof_checker = uspoof_open(&status);
+  if (U_FAILURE(status)) {
+    std::cerr << "Failed to create an ICU uspoof_checker due to "
+              << u_errorName(status) << ".\n";
+    return 1;
+  }
+
+  std::stringstream input(input_content);
+  std::string output =
+      "// Copyright 2017 The Chromium Authors. All rights reserved.\n"

[Peter Kasting, 2017/05/10 22:38:47]

Nit: Maybe consider a raw string literal

[jungshik at Google, 2017/05/14 09:36:23]

Done.

+      "// Use of this source code is governed by a BSD-style license that can "
+      "be\n"
+      "// found in the LICENSE file.\n\n"
+      "// This file is generated by components/url_formatter/"
+      "make_top_domain_gperf.cc\n"
+      "// DO NOT MANUALLY EDIT!\n\n"
+      "// Each entry is the skeleton of a top domain for the confusability "
+      "check\n"
+      "// in components/url_formatter/url_formatter.cc.\n"
+      "%%\n";
+
+  std::string domain;
+  size_t max_labels = 0;
+  std::string domain_with_max_labels;
+  while (std::getline(input, domain)) {
+    if (domain[0] == '#')
+      continue;
+    std::string skeleton = GetSkeleton(domain, spoof_checker);
+    if (skeleton.empty()) {
+      std::cerr << "Failed to generate the skeleton of " << domain << '\n';
+      output += "// " + domain + '\n';
+    } else {
+      output += skeleton + ", 1\n";
+    }
+    std::vector<base::StringPiece> labels = base::SplitStringPiece(
+        domain, ".", base::TRIM_WHITESPACE, base::SPLIT_WANT_ALL);
+    if (labels.size() > max_labels) {
+      domain_with_max_labels = domain;
+      max_labels = labels.size();
+    }
+  }
+
+  output += "%%\n";
+
+  WriteToFile(output, "alexa_skeletons.gperf");
+
+  std::cout << "The first domain with the largest number of labels is "
+            << domain_with_max_labels << " and has " << max_labels
+            << " labels.\n";
+
+  return 0;
+}