Switch Speech Recognition to asynchronous callback-based AudioManager interactions.

media/audio/audio_system.h

+// Copyright 2017 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 MEDIA_AUDIO_AUDIO_SYSTEM_H_
+#define MEDIA_AUDIO_AUDIO_SYSTEM_H_
+
+#include "base/callback.h"
+#include "media/audio/audio_device_description.h"
+#include "media/base/audio_parameters.h"
+#include "media/base/media_export.h"
+
+namespace base {
+class SingleThreadTaskRunner;
+}
+
+namespace media {
+class AudioManager;
+
+// Work in progress: Provides asynchronous interface to AudioManager. All the
+// AudioManager clients will be switched to it, in preparation for moving
+// to Mojo audio service.
+class MEDIA_EXPORT AudioSystem {
+ public:
+  // Replies are asynchronously sent to the thread the call is issued on.
+  using OnAudioParamsCallback = base::Callback<void(const AudioParameters&)>;
+
+  virtual base::SingleThreadTaskRunner* GetTaskRunner() const = 0;
+
+  // Callback will receive invalid parameters if the device is not found.
+  virtual void GetInputStreamParameters(const std::string& device_id,

[tommi (sloooow) - chröme, 2017/02/05 20:14:54]

Since we're making a new interface, would it make sense for this to be const?

[o1ka, 2017/02/06 12:06:08]

Right! Done.

+                                        OnAudioParamsCallback on_params_cb) = 0;
+
+  // Must not be used for anything but stream creation.
+  virtual AudioManager* GetAudioManager() = 0;
+};
+
+}  // namespace media
+
+#endif  // MEDIA_AUDIO_AUDIO_SYSTEM_H_