Implement GenericURLRequestJob::GetLoadTimingInfo

headless/public/util/generic_url_request_job.cc

Index: headless/public/util/generic_url_request_job.cc
diff --git a/headless/public/util/generic_url_request_job.cc b/headless/public/util/generic_url_request_job.cc
index 9c7245bc43093991dbe604ac089ce41ac9fae205..3aab15f5388622cd93212baa0c9563bcd333de72 100644
--- a/headless/public/util/generic_url_request_job.cc
+++ b/headless/public/util/generic_url_request_job.cc
@@ -140,6 +140,7 @@ void GenericURLRequestJob::OnFetchComplete(
     scoped_refptr<net::HttpResponseHeaders> response_headers,
     const char* body,
     size_t body_size) {
+  response_time_ = base::TimeTicks::Now();
   http_response_code_ = http_response_code;
   response_headers_ = response_headers;
   body_ = body;
@@ -186,4 +187,9 @@ bool GenericURLRequestJob::GetCharset(std::string* charset) {
   return response_headers_->GetCharset(charset);
 }
 
+void GenericURLRequestJob::GetLoadTimingInfo(
+    net::LoadTimingInfo* load_timing_info) const {
+  load_timing_info->receive_headers_end = response_time_;

[altimin, 2016/11/17 15:45:24]

Should we also fill other fields in net::LoadTimingInfo?

[alex clarke (OOO till 29th), 2016/11/17 16:05:59]

Possibly :)

I'm not really sure how they are used (and how important they are).  Currently
some of them we simply can't get with our current fetch interface.  I'd like to
understand this better before I rush off and implement something.

[altimin, 2016/11/17 16:08:39]

Let's add a TODO! :)

+}
+
 }  // namespace headless