Check time budget before starting new requests, and unittest.

components/offline_pages/background/request_coordinator.cc

Index: components/offline_pages/background/request_coordinator.cc
diff --git a/components/offline_pages/background/request_coordinator.cc b/components/offline_pages/background/request_coordinator.cc
index a78d58cbdcfe24663f5e444c1777d3ab4155c104..f120ffb1fb37f914ed857fadce06200b9bb9e9f6 100644
--- a/components/offline_pages/background/request_coordinator.cc
+++ b/components/offline_pages/background/request_coordinator.cc
@@ -35,14 +35,6 @@ const Scheduler::TriggerConditions kUserRequestTriggerConditions(
     false /* require_power_connected */,
     50 /* minimum_battery_percentage */,
     false /* require_unmetered_network */);
-
-// Timeout is 2.5 minutes based on the size of Marshmallow doze mode
-// maintenance window (3 minutes)
-// TODO(petewil): Find the optimal timeout based on data for 2G connections and
-// common EM websites. crbug.com/625204
-// TODO(petewil): Move this value into the policy object.
-long OFFLINER_TIMEOUT_SECONDS = 150;
-
 }  // namespace
 
 RequestCoordinator::RequestCoordinator(std::unique_ptr<OfflinerPolicy> policy,
@@ -51,13 +43,15 @@ RequestCoordinator::RequestCoordinator(std::unique_ptr<OfflinerPolicy> policy,
                                        std::unique_ptr<Scheduler> scheduler)
     : is_busy_(false),
       is_canceled_(false),
-      offliner_timeout_(base::TimeDelta::FromSeconds(OFFLINER_TIMEOUT_SECONDS)),
       offliner_(nullptr),
+      operation_start_time_(base::Time::Now()),

[jianli, 2016/07/25 21:50:38]

It seems to me that this initialization was not needed.

[Pete Williamson, 2016/07/25 22:01:33]

Done.

       policy_(std::move(policy)),
       factory_(std::move(factory)),
       queue_(std::move(queue)),
       scheduler_(std::move(scheduler)),
       last_offlining_status_(Offliner::RequestStatus::UNKNOWN),
+      offliner_timeout_(base::TimeDelta::FromSeconds(
+          policy_->GetSinglePageTimeBudgetInSeconds())),
       weak_ptr_factory_(this) {
   DCHECK(policy_ != nullptr);
   picker_.reset(new RequestPicker(queue_.get(), policy_.get()));
@@ -128,6 +122,10 @@ bool RequestCoordinator::StartProcessing(
   current_conditions_.reset(new DeviceConditions(device_conditions));
   if (is_busy_) return false;
 
+  // Mark the time at which we started processing so we can check our time
+  // budget.
+  operation_start_time_ = base::Time::Now();
+
   is_canceled_ = false;
   scheduler_callback_ = callback;
   // TODO(petewil): Check existing conditions (should be passed down from
@@ -139,6 +137,15 @@ bool RequestCoordinator::StartProcessing(
 }
 
 void RequestCoordinator::TryNextRequest() {
+  // If there is no time left in the budget, return to the scheduler, but leave
+  // the "safety" task, so we get called again next time.

[jianli, 2016/07/25 21:50:38]

I found it hard to understand what the "safety" task was meant for. Could you
please rephrase it?

[Pete Williamson, 2016/07/25 22:01:33]

Done.  I was referring to something in the design docs, but it is better if the
code is standalone.

+  if (base::Time::Now() - operation_start_time_ >
+      base::TimeDelta::FromSeconds(
+          policy_->GetBackgroundProcessingTimeBudgetSeconds())) {
+    // Let the scheduler know we are done processing.
+    scheduler_callback_.Run(true);

[jianli, 2016/07/25 21:50:38]

Do we need to return here?

[Pete Williamson, 2016/07/25 22:01:33]

Oh my goodness, you're right!  Added.

+  }
+
   // Choose a request to process that meets the available conditions.
   // This is an async call, and returns right away.
   picker_->ChooseNextRequest(
@@ -229,8 +236,6 @@ void RequestCoordinator::OfflinerDoneCallback(const SavePageRequest& request,
                    weak_ptr_factory_.GetWeakPtr()));
   }
 
-  // TODO(petewil): Check time budget. Return to the scheduler if we are out.
-  // Start another request if we have time.
   TryNextRequest();
 }