Add SafeBrowsingNavigationObserver to listen to navigation events

chrome/browser/safe_browsing/safe_browsing_navigation_observer.h

+// Copyright 2016 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.
+
+#ifndef CHROME_BROWSER_SAFE_BROWSING_SAFE_BROWSING_NAVIGATION_OBSERVER_H_
+#define CHROME_BROWSER_SAFE_BROWSING_SAFE_BROWSING_NAVIGATION_OBSERVER_H_
+
+#include "base/supports_user_data.h"
+#include "content/public/browser/web_contents_observer.h"
+#include "url/gurl.h"
+
+namespace content {
+class NavigationHandle;
+struct ResourceRedirectDetails;
+}
+
+namespace safe_browsing {
+class SafeBrowsingNavigationObserverManager;
+
+// Struct to record the details of a navigation event for any frames.

[Charlie Reis, 2016/10/17 23:54:03]

nit: s/frames/frame/

[Jialiu Lin, 2016/10/18 19:06:05]

Done.

+// This information will be used to fill |url_chain| field in safe browsing
+// download pings.
+struct NavigationEvent {
+  NavigationEvent();
+  NavigationEvent(NavigationEvent&& nav_event);
+  NavigationEvent& operator=(NavigationEvent&& nav_event);
+  ~NavigationEvent();
+
+  GURL source_url;    // URL that caused this navigation to occur.
+  int source_tab_id;  // Which tab the above source_url is in. Tab ID is

[Charlie Reis, 2016/10/17 23:54:04]

Assuming that source_url might be a subframe, I think it might help to clarify
the comment here to say something like "Which tab contains the frame with
source_url."

[Jialiu Lin, 2016/10/18 19:06:05]

Ah, that's much clearer. Thanks!

+                      // returned by SessionTabHelper::IdForTab. This ID is
+                      // immutable for a given tab and unique across Chrome
+                      // within the current session.
+
+  GURL target_url;    // The original request URL of this navigation.
+  int target_tab_id;  // Which tab this target url is in.
+
+  int frame_id;  // Frame tree node ID of the frame where this navigation takes
+                 // place.
+  GURL main_frame_url;  // Main frame url of the source_url. Can be the same

[Charlie Reis, 2016/10/17 23:54:04]

This should probably be called source_main_frame_url, since it sounds more like
the destination URL.

[Jialiu Lin, 2016/10/18 19:06:05]

make sense. 

+                        // as source url, if source_url was loaded in main
+                        // frame.
+
+  base::Time timestamp;  // Timestamp of last time this object is updated.

[Charlie Reis, 2016/10/17 23:54:04]

last_updated, perhaps?

[Jialiu Lin, 2016/10/18 19:06:05]

Done.

+
+  bool is_user_initiated;  // browser_initiated || has_user_gesture.
+  bool has_committed;
+  bool has_server_redirect;
+  GURL server_redirect_url;  // Last server redirect url in server redirect

[Charlie Reis, 2016/10/17 23:54:04]

This sounds a bit awkward to use.  In most places (e.g., NavigationEntry's URL),
we store the URL that ended up committing after all the redirects, and you can
call GetOriginalRequestURL if you care about what it was before redirects.

With the current struct, there's no simple way to find what URL committed. 
You'd have to check server_redirect_url, and if that's empty, fall back to
target_url.

Would it be simpler to use if we change server_redirect_url to committed_url and
have it present even there's no redirects?

[Jialiu Lin, 2016/10/18 19:06:05]

I agree. though for some cases (e.g. download urls) this url may not be
technically committed.  how about we rename target_url to original_request_url,
and server_redirect_url to actual_target_url?

[Charlie Reis, 2016/10/19 17:10:45]

Good point about downloads not committing.  Maybe destination_url would be
closer to terminology we use elsewhere?

[Jialiu Lin, 2016/10/21 02:38:38]

SG. renamed to destination_url.

+                             // chain. In the absence of server redirect, this
+                             // field will be empty.
+};
+
+// Structure to keep tracks of resolved IP address of a host.

[Charlie Reis, 2016/10/17 23:54:04]

nit: track

[Jialiu Lin, 2016/10/18 19:06:05]

Done.

+struct ResolvedIPAddress {
+  ResolvedIPAddress() : timestamp(base::Time::Now()), ip() {}
+  ResolvedIPAddress(base::Time timestamp, const std::string& ip)
+      : timestamp(timestamp), ip(ip) {}
+  base::Time timestamp;  // Timestamp of when we get the resolved IP.
+  std::string ip;        // Resolved IP address
+};
+
+// Observes the navigation events for a single WebContents (both main-frame
+// and sub-frame navigations)
+class SafeBrowsingNavigationObserver : public base::SupportsUserData::Data,
+                                       public content::WebContentsObserver {
+ public:
+  static void MaybeCreateForWebContents(content::WebContents* web_contents);
+  static SafeBrowsingNavigationObserver* FromWebContents(
+      content::WebContents* web_contents);
+
+  SafeBrowsingNavigationObserver(
+      content::WebContents* contents,
+      const scoped_refptr<SafeBrowsingNavigationObserverManager>& manager);
+
+  ~SafeBrowsingNavigationObserver() override;
+
+ private:
+  typedef std::unordered_map<content::NavigationHandle*, NavigationEvent>
+      NavigationHandleMap;
+
+  // content::WebContentsObserver:
+  void DidStartNavigation(
+      content::NavigationHandle* navigation_handle) override;
+  void DidRedirectNavigation(
+      content::NavigationHandle* navigation_handle) override;
+  void DidFinishNavigation(
+      content::NavigationHandle* navigation_handle) override;
+  void DidGetResourceResponseStart(
+      const content::ResourceRequestDetails& details) override;
+  void DidGetUserInteraction(const blink::WebInputEvent::Type type) override;
+  void WebContentsDestroyed() override;
+
+  // Map keyed on NavigationHandle* to keep track of all the ongoing navigation
+  // events. NavigationHandle pointers are owned by RenderFrameHost. Since a
+  // NavigationHandle object will be destructed after navigation is done,
+  // corresponding entry in this map will be removed from navigation_handle_map_
+  // and added to navigation_map_ at the end of DidFinishNavigation(...).

[Charlie Reis, 2016/10/17 23:54:04]

Please clarify where navigation_map_ lives.

[Jialiu Lin, 2016/10/18 19:06:05]

Done.

+  NavigationHandleMap navigation_handle_map_;
+  scoped_refptr<SafeBrowsingNavigationObserverManager> manager_;

[Charlie Reis, 2016/10/17 23:54:04]

nit: I'd find this section more readable with blank lines between members, given
the comments.  Just my 2 cents, though.

[Jialiu Lin, 2016/10/18 19:06:05]

Blank lines added.

+  // If the observed WebContents just got an user gesture.
+  bool has_user_gesture_;
+  base::Time last_user_gesture_timestamp_;
+  DISALLOW_COPY_AND_ASSIGN(SafeBrowsingNavigationObserver);
+};
+
+}  // namespace safe_browsing
+
+#endif  // CHROME_BROWSER_SAFE_BROWSING_SAFE_BROWSING_NAVIGATION_OBSERVER_H_