URL Formatter: Add StripSubdomain method that preserves eTLD + 1.

components/url_formatter/url_formatter.cc

 // Copyright 2015 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 "components/url_formatter/url_formatter.h"
 
 #include <algorithm>
 #include <utility>
 #include <vector>
 
 #include "base/lazy_instance.h"
 #include "base/macros.h"
 #include "base/numerics/safe_conversions.h"
 #include "base/strings/string_piece.h"
 #include "base/strings/string_util.h"
 #include "base/strings/utf_offset_string_conversions.h"
 #include "base/strings/utf_string_conversions.h"
 #include "base/threading/thread_local_storage.h"
 #include "components/url_formatter/idn_spoof_checker.h"
+#include "net/base/registry_controlled_domains/registry_controlled_domain.h"
 #include "third_party/icu/source/common/unicode/uidna.h"
 #include "third_party/icu/source/common/unicode/utypes.h"
 #include "url/gurl.h"
 #include "url/third_party/mozilla/url_parse.h"
 
 namespace url_formatter {
 
 namespace {
 
 base::string16 IDNToUnicodeWithAdjustments(
@@ skipping 541 matching lines @@
 base::string16 IDNToUnicode(base::StringPiece host) {
   return IDNToUnicodeWithAdjustments(host, nullptr);
 }
 
 base::string16 StripWWW(const base::string16& text) {
   const base::string16 www(base::ASCIIToUTF16("www."));
   return base::StartsWith(text, www, base::CompareCase::SENSITIVE)
       ? text.substr(www.length()) : text;
 }
 
-base::string16 StripWWWFromHost(const GURL& url) {
+std::string StripSubdomains(const GURL& url, StripSubdomainTypes subdomains) {
   DCHECK(url.is_valid());
-  return StripWWW(base::ASCIIToUTF16(url.host_piece()));
+
+  base::StringPiece host_piece = url.host_piece();
+  std::string domain_and_registry =
+      net::registry_controlled_domains::GetDomainAndRegistry(
+          host_piece,
+          net::registry_controlled_domains::INCLUDE_PRIVATE_REGISTRIES);
+
+  if (host_piece == domain_and_registry)
+    return host_piece.as_string();
+
+  std::string subdomain_piece =
+      host_piece.substr(0, host_piece.length() - domain_and_registry.length())
+          .as_string();
+
+  if (subdomains & kStripWWW)
+    base::ReplaceSubstringsAfterOffset(&subdomain_piece, 0, "www.", "");
+
+  if (subdomains & kStripM)
+    base::ReplaceSubstringsAfterOffset(&subdomain_piece, 0, "m.", "");
+
+  return subdomain_piece + domain_and_registry;
 }
 
 }  // namespace url_formatter