Collect separate V8 histograms for Top10 non-Google sites

content/renderer/render_thread_impl.cc

Index: content/renderer/render_thread_impl.cc
diff --git a/content/renderer/render_thread_impl.cc b/content/renderer/render_thread_impl.cc
index 82feaa80f293e6038cd63a9ab82991e415c855b2..269ebc67e29df74ae98a319485f4d3329a1cd738 100644
--- a/content/renderer/render_thread_impl.cc
+++ b/content/renderer/render_thread_impl.cc
@@ -22,6 +22,7 @@
 #include "base/single_thread_task_runner.h"
 #include "base/strings/string16.h"
 #include "base/strings/string_number_conversions.h"
+#include "base/strings/string_split.h"
 #include "base/strings/string_tokenizer.h"
 #include "base/strings/sys_string_conversions.h"
 #include "base/strings/utf_string_conversions.h"
@@ -131,6 +132,7 @@
 #include "mojo/common/common_type_converters.h"
 #include "net/base/net_errors.h"
 #include "net/base/port_util.h"
+#include "net/base/registry_controlled_domains/registry_controlled_domain.h"
 #include "skia/ext/event_tracer_impl.h"
 #include "skia/ext/skia_memory_dump_provider.h"
 #include "third_party/WebKit/public/platform/WebImageGenerator.h"
@@ -277,6 +279,45 @@ class RenderViewZoomer : public RenderViewVisitor {
   DISALLOW_COPY_AND_ASSIGN(RenderViewZoomer);
 };
 
+bool IsAlexaTop10NonGoogleSite(const std::string& host) {

[jochen (gone - plz use gerrit), 2015/10/26 15:29:33]

just make this a method of HistogramCustomizer, then you can include your unit
tests

+  // The Top10 sites have different TLD and/or subdomains
+  // depending on the localization
+  if (host == "sina.com.cn")
+    return true;
+
+  std::string sanitized_host =
+        net::registry_controlled_domains::GetDomainAndRegistry(

[jochen (gone - plz use gerrit), 2015/10/26 15:29:33]

please use "git cl format" on this cl

+            host,
+          net::registry_controlled_domains::INCLUDE_PRIVATE_REGISTRIES);
+
+  if (sanitized_host == "facebook.com")
+    return true;
+  if (sanitized_host == "baidu.com")
+    return true;
+  if (sanitized_host == "qq.com")
+    return true;
+  if (sanitized_host == "twitter.com")
+    return true;
+  if (sanitized_host == "taobao.com")
+    return true;
+  if (sanitized_host == "live.com")
+    return true;
+
+  if (!sanitized_host.empty()) {
+     std::vector<base::StringPiece> host_tokens = base::SplitStringPiece(
+       sanitized_host, ".", base::TRIM_WHITESPACE, base::SPLIT_WANT_NONEMPTY);
+
+    if (host_tokens.size() >= 2) {
+      if ((host_tokens[0] == "yahoo") ||
+          (host_tokens[0] == "amazon") ||
+          (host_tokens[0] == "wikipedia")) {
+        return true;
+      }
+    }
+  }
+  return false;
+}
+
 std::string HostToCustomHistogramSuffix(const std::string& host) {
   if (host == "mail.google.com")
     return ".gmail";
@@ -288,6 +329,9 @@ std::string HostToCustomHistogramSuffix(const std::string& host) {
     return ".inbox";
   if (host == "www.youtube.com")
     return ".youtube";
+  if (IsAlexaTop10NonGoogleSite(host))
+    return ".top10";
+
   return std::string();
 }