Adds a CancellableTaskFactory and exposes postLoadingTasks to Blink

Source/platform/scheduler/CancellableTaskFactory.cpp

+// 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 "config.h"
+#include "platform/scheduler/CancellableTaskFactory.h"
+
+#include "platform/TraceLocation.h"
+#include "public/platform/Platform.h"
+#include "wtf/InstanceCounter.h"
+
+namespace blink {
+
+void CancellableTaskFactory::cancel()
+{
+    m_weakPtrFactory.revokeAll();
+}
+
+WebThread::Task* CancellableTaskFactory::task()
+{
+    m_weakPtrFactory.revokeAll();

[Sami, 2015/01/26 16:08:14]

Perhaps just call cancel() here instead of duplicating the code?

[alex clarke (OOO till 29th), 2015/01/26 17:00:45]

Done.

+    return new CancellableTask(m_weakPtrFactory.createWeakPtr());
+}
+
+void CancellableTaskFactory::CancellableTask::run()
+{
+    if (m_weakPtr.get())
+        m_weakPtr->runInternal();

[Sami, 2015/01/26 16:08:14]

nit: I think you could call m_closure here directly to save having another
function (assuming the revokeAll call isn't needed)?

[alex clarke (OOO till 29th), 2015/01/26 17:00:45]

Done.

+}
+
+void CancellableTaskFactory::runInternal()
+{
+    m_weakPtrFactory.revokeAll();

[Sami, 2015/01/26 16:08:14]

Do we need to do this? Task::run is guaranteed to be called at most once.

[alex clarke (OOO till 29th), 2015/01/26 17:00:45]

Yes we do because there's code in the HTML parser scheduler that relies on
isPending() returning false in side a function called from the callback. 
Reminds me I need to add a test for that.

+    (*m_closure)();
+}
+
+} // namespace blink