PageLoadMetrics renderer and browser implementation (deprecated)

components/page_load_metrics/renderer/page_load_metrics_render_frame_observer.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 "base/time/time.h"
+#include "components/page_load_metrics/common/page_load_metrics_messages.h"
+#include "components/page_load_metrics/renderer/page_load_metrics_render_frame_observer.h"
+#include "content/public/renderer/render_frame.h"
+#include "third_party/WebKit/public/platform/WebURLResponse.h"
+#include "third_party/WebKit/public/web/WebDataSource.h"
+#include "third_party/WebKit/public/web/WebDocument.h"
+#include "third_party/WebKit/public/web/WebLocalFrame.h"
+#include "third_party/WebKit/public/web/WebPerformance.h"
+#include "url/gurl.h"
+
+using base::Time;
+
+namespace page_load_metrics {
+
+namespace {
+
+// PerformanceTiming is the concrete implementation of the Timing interface
+// that gets its timing data from a blink::WebPerformance instance.
+class PerformanceTiming : public PageLoadMetricsRenderFrameObserver::Timing {
+ public:
+  explicit PerformanceTiming(const blink::WebPerformance& performance)
+      : performance_(performance) {}
+
+  Time GetNavigationStart() const override {
+    return Time::FromDoubleT(performance_.navigationStart());
+  }
+
+  Time GetDOMContentLoadedEvent() const override {
+    return Time::FromDoubleT(performance_.domContentLoadedEventStart());
+  }
+
+  Time GetLoadEvent() const override {
+    return Time::FromDoubleT(performance_.loadEventStart());
+  }
+
+  Time GetFirstLayout() const override {
+    return Time::FromDoubleT(performance_.firstLayout());
+  }
+
+ private:
+  blink::WebPerformance performance_;
+
+  DISALLOW_COPY_AND_ASSIGN(PerformanceTiming);
+};
+
+}  // namespace
+
+// PageMetricsSender is a helper class responsible for sending page load
+// metrics over the lifetime of a webpage. A PageMetricsSender instance is
+// active from the time a provisional load is committed, until the time the
+// provisional load for the next page is started.
+class PageLoadMetricsRenderFrameObserver::PageMetricsSender {
+ public:
+  explicit PageMetricsSender(PageLoadMetricsRenderFrameObserver* parent);
+  void MaybeSendMetrics(const Timing& timing);
+
+ private:
+  void Send(IPC::Message* message) { parent_->Send(message); }
+  int routing_id() const { return parent_->routing_id(); }
+
+  void MaybeSendFirstLayout(const Timing& timing);
+  void MaybeSendDOMContentLoadedEvent(const Timing& timing);
+  void MaybeSendLoadEvent(const Timing& timing);
+
+  PageLoadMetricsRenderFrameObserver* const parent_;
+  bool did_send_first_layout_;
+  bool did_send_dom_content_loaded_event_;
+  bool did_send_load_event_;

[kinuko, 2015/08/18 10:37:39]

So this object's created for each page/document mainly for book-keeping which
fields of the Performance object are sent?

Do you have a plan about how we send custom metrics that aren't covered by
Performance API?

[Bryan McQuade, 2015/08/19 11:45:14]

The Blink WebPerformance object that exposes the timing information uses the nav
timing API (http://www.w3.org/TR/navigation-timing-2/) for many of its metrics,
but we also expose additional interesting metrics on WebPerformance that aren't
part of nav timing, such as time to first layout. My current plan is for us to
extend WebPerformance to expose other interesting metrics (e.g. time to first
text visible) as long as they are generally useful.

+};
+
+PageLoadMetricsRenderFrameObserver::PageLoadMetricsRenderFrameObserver(
+    content::RenderFrame* render_frame)
+    : content::RenderFrameObserver(render_frame) {}
+
+PageLoadMetricsRenderFrameObserver::~PageLoadMetricsRenderFrameObserver() {}
+
+// TODO(bmcquade): use DidChangePerformanceTiming instead, once available.
+void PageLoadMetricsRenderFrameObserver::DidFinishDocumentLoad() {
+  MaybeSendMetrics();
+}
+
+// TODO(bmcquade): use DidChangePerformanceTiming instead, once available.

[Randy Smith (Not in Mondays), 2015/08/17 19:31:07]

Is there a pointer to issue # or CL to include here so that someone coming on
the code later could figure out if this TODO should be executed?

[Bryan McQuade, 2015/08/19 11:45:14]

Sure - this should actually be addressed before I submit this change - his code
is in review now: https://codereview.chromium.org/1289053003/.

+void PageLoadMetricsRenderFrameObserver::DidFinishLoad() {
+  MaybeSendMetrics();
+}
+
+void PageLoadMetricsRenderFrameObserver::DidStartProvisionalLoad() {
+  // A provisional page load has started. Shut down metrics reporting until the
+  // next page load is committed. Note that a provisoinal load may fail. We
+  // simply stop reporting metrics in that case, until the next successful page
+  // load is committed.
+  page_metrics_sender_.reset();
+}
+
+void PageLoadMetricsRenderFrameObserver::DidCommitProvisionalLoad(
+    bool is_new_navigation, bool is_same_page_navigation) {
+  // Same-page navigations (e.g. an in-document navigation from a fragment
+  // link) aren't full page loads, since they don't go to network to load the
+  // main HTML resource.
+  if (is_same_page_navigation)
+    return;

[Randy Smith (Not in Mondays), 2015/08/17 19:31:07]

Does this mean that a same page navigation will nuke sending of metrics (because
DidStartProvisionalLoad() reset page_metrics_sender_) if the/some statistics
haven't yet been sent?  Is that ok?  Would it instead be useful/right to just
keep around the old page_metrics_sender_?  I think it's probably worth
commenting this issue, whatever the answer.

[Bryan McQuade, 2015/08/19 11:45:14]

DidStartProvisionalLoad doesn't get invoked for same-page navigations. I'll add
a comment for this.

+
+  if (ShouldSendMetrics())
+    page_metrics_sender_.reset(new PageMetricsSender(this));

[Randy Smith (Not in Mondays), 2015/08/17 19:31:07]

I'd like a comment somewhere (not sure where, but here might be fine) that this
function is when the go/no-go about sending metrics is done, but the actual
sending of the metrics is delayed until we actually have the data.

[Bryan McQuade, 2015/08/25 13:53:35]

Done.

+  else
+    page_metrics_sender_.reset();

[Randy Smith (Not in Mondays), 2015/08/17 19:31:07]

Under what circumstances will this have an effect?  I suspect there's something
about the life cycle I'm not understanding, since I would think we'd always have
a null page_metrics_sender_ at this point, meaning that this should/could be a
DCHECK?

[Bryan McQuade, 2015/08/19 11:45:14]

Good point. A DCHECK seems much better here. Will make that change.

+}
+
+void PageLoadMetricsRenderFrameObserver::MaybeSendMetrics() {
+  if (!page_metrics_sender_)
+    return;
+
+  scoped_ptr<Timing> timing(CreateTiming());
+  page_metrics_sender_->MaybeSendMetrics(*timing);
+}
+
+bool PageLoadMetricsRenderFrameObserver::ShouldSendMetrics() {
+  const blink::WebLocalFrame& frame = *render_frame()->GetWebFrame();
+  // We only generate historgrams for main frames.
+  if (frame.parent())
+    return false;
+
+  const blink::WebDocument& document = frame.document();
+  // Ignore non-HTTP schemes (e.g. chrome://).
+  const GURL& url = document.url();
+  if (!url.SchemeIs("http") && !url.SchemeIs("https"))

[kinuko, 2015/08/18 10:37:39]

nit: !url.SchemeIsHTTPOrHTTPS()

[Bryan McQuade, 2015/08/25 13:53:35]

Thanks, did not know about that. Fixed.

+    return false;
+
+  const blink::WebURLResponse& url_response = frame.dataSource()->response();
+  // Ignore multipart responses (e.g. MHTML).
+  if (url_response.isMultipartPayload())
+    return false;
+
+  // Ignore non-HTML documents (e.g. SVG). Note that images are treated by
+  // Blink as HTML documents, so to exclude images, we must perform
+  // additional mime type checking below.
+  if (!document.isHTMLDocument() && !document.isXHTMLDocument())
+    return false;
+
+  // Ignore non-HTML mime types (e.g. images).
+  std::string mime_type = url_response.mimeType().utf8();
+  if (mime_type != "text/html" && mime_type != "application/xhtml+xml")
+    return false;
+
+  return true;
+}
+
+scoped_ptr<PageLoadMetricsRenderFrameObserver::Timing>
+PageLoadMetricsRenderFrameObserver::CreateTiming() {
+  blink::WebLocalFrame* frame = render_frame()->GetWebFrame();
+  scoped_ptr<Timing> timing(new PerformanceTiming(
+      frame ? frame->performance() : blink::WebPerformance()));
+  return timing.Pass();
+}
+
+PageLoadMetricsRenderFrameObserver::PageMetricsSender::PageMetricsSender(
+    PageLoadMetricsRenderFrameObserver* parent)
+    : parent_(parent),
+      did_send_first_layout_(false),
+      did_send_dom_content_loaded_event_(false),
+      did_send_load_event_(false) {}
+
+void PageLoadMetricsRenderFrameObserver::PageMetricsSender::MaybeSendMetrics(
+    const Timing& timing) {
+  MaybeSendFirstLayout(timing);
+  MaybeSendDOMContentLoadedEvent(timing);
+  MaybeSendLoadEvent(timing);

[Randy Smith (Not in Mondays), 2015/08/17 19:31:08]

I'll happily accept this architecture if you tell me it's necessary, but I'll
just call out that if there's a) some linear dependency between these events,
and b) the last event doesn't generally happen "too late" for us to get
statistics, I'd rather hold off until the last event and send all three.  

Even if (a) isn't true (as I suspect it isn't), I'll call out that doing is this
way means that in interpreting the stats, we may need to deal with web pages
that get killed having sent only a completely arbitrary subset of the stats,
which is going to make statistics analysis ... "interesting".  We might want
less data that we can make stronger assumptions about, e.g. waiting until all
three of these things happen and then sending.  Or maybe we want exactly this,
but to track how often we don't get all the information for a web page.  

Basically, I'd like us to think through the analysis portion of these stats
before we finish landing this CL--if you were analyzing the UMA from this CL,
how would you do it and what assumptions would you be able to make and not make?

[Bryan McQuade, 2015/08/19 11:45:14]

It's generally hard to predict when in the page load life cycle these events
will happen. First layout can happen before or after dom content, and before or
after onload. As we add other metrics (e.g. time-to-first-visible-text) they'll
further complicate any possible orderings. An issue with holding off is that we
do want to log events eagerly as they happen - if a tab has a first layout but
gets closed before the onload event fires, we do want to record the first layout
time in UMA, since a user did experience a layout for that page. Future work
will want to record the number of page loads that get aborted before first
layout/first paint, and it wouldn't be possible to record that information if we
deferred sending until a later metric.

Note too that we considered having csharrison's new notification API indicate
which metric or metrics actually changed, which would alleviate the need for
testing to see if each metric is ready to send, but that ends up adding
additional complexity to the notification API (enum to list all possible event
types, plumbing that through various parts of blink) so we decided that it was
overall simpler to have a less complex notification API from blink that only
says 'the metrics changed' and then we'd add logic here to figure out which
metrics are actually available/changed.

+}
+
+void PageLoadMetricsRenderFrameObserver::PageMetricsSender::
+    MaybeSendFirstLayout(const Timing& timing) {
+  if (did_send_first_layout_)
+    return;
+
+  if (timing.GetNavigationStart().is_null() ||
+      timing.GetFirstLayout().is_null())
+    return;
+
+  Send(new PageLoadMetricsMsg_FirstLayout(
+      routing_id(),
+      timing.GetFirstLayout() - timing.GetNavigationStart()));
+  did_send_first_layout_ = true;
+}
+
+void PageLoadMetricsRenderFrameObserver::PageMetricsSender::
+    MaybeSendDOMContentLoadedEvent(const Timing& timing) {
+  if (did_send_dom_content_loaded_event_)
+    return;
+
+  if (timing.GetNavigationStart().is_null() ||
+      timing.GetDOMContentLoadedEvent().is_null())
+    return;
+
+  Send(new PageLoadMetricsMsg_DOMContentLoadedEvent(
+      routing_id(),
+      timing.GetDOMContentLoadedEvent() - timing.GetNavigationStart()));
+  did_send_dom_content_loaded_event_ = true;
+}
+
+void PageLoadMetricsRenderFrameObserver::PageMetricsSender::MaybeSendLoadEvent(
+    const Timing& timing) {
+  if (did_send_load_event_)
+    return;
+
+  if (timing.GetNavigationStart().is_null() || timing.GetLoadEvent().is_null())
+    return;
+
+  Send(new PageLoadMetricsMsg_LoadEvent(
+      routing_id(),
+      timing.GetLoadEvent() - timing.GetNavigationStart()));
+  did_send_load_event_ = true;
+}
+
+}  // namespace page_load_metrics