Wake Lock API implementation (Blink part)

public/platform/modules/wake_lock/WebWakeLockClient.h

Index: public/platform/modules/wake_lock/WebWakeLockClient.h
diff --git a/public/platform/modules/wake_lock/WebWakeLockClient.h b/public/platform/modules/wake_lock/WebWakeLockClient.h
new file mode 100644
index 0000000000000000000000000000000000000000..3c0a7c409022d113fdf3220f7854f6d22214dec6
--- /dev/null
+++ b/public/platform/modules/wake_lock/WebWakeLockClient.h
@@ -0,0 +1,19 @@
+// 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.
+
+#ifndef WebWakeLockClient_h
+#define WebWakeLockClient_h
+
+namespace blink {
+
+class WebWakeLockClient {
+public:
+    virtual ~WebWakeLockClient() { }

[mlamouri (slow - plz ping), 2015/07/05 14:50:14]

nit: use "= default;"

+
+    virtual void requestKeepScreenAwake(bool) { };

[mlamouri (slow - plz ping), 2015/07/05 14:50:15]

nit: why not "= 0;" ?

[alogvinov, 2015/07/06 09:26:29]

You suggested not using pure virtual functions in this very place in this patch:
https://codereview.chromium.org/1084923002/patch/1/10020

"Usually, Blink interfaces are not using pure virtual methods because if the
Blink interface changes, it would break Chromium."

Has the situation changed since then?

[mlamouri (slow - plz ping), 2015/07/06 10:13:01]

Eh :) I would say that is obviously a problem for classes that already exist
because it would break Chromium's build. I'm not sure if that's a problem for
new classes actually. I guess I'm confused :)

Jochen, do you know what's the expected practice?

+};
+
+} // namespace blink
+
+#endif