[webkit-reviews] review denied: [Bug 89197] [Chromium] Implements DeviceMotion : [Attachment 159450] Patch
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Mon Aug 20 12:06:13 PDT 2012
Adam Barth <abarth at webkit.org> has denied Hans Wennborg <hans at chromium.org>'s
request for review:
Bug 89197: [Chromium] Implements DeviceMotion
https://bugs.webkit.org/show_bug.cgi?id=89197
Attachment 159450: Patch
https://bugs.webkit.org/attachment.cgi?id=159450&action=review
------- Additional Comments from Adam Barth <abarth at webkit.org>
Going "down" to the platform rather than "up" to the client is fine. However,
if you're going to adopt that approach, you should adopt it entirely. The
current patch is a confused mix of the two approaches. For example, the "down
to the platform" approach, we should get rid of DeviceMotionClient and the
WebKit layer shouldn't have any knowledge of DeviceMotionDetectorInternal.
The reason classes like DeviceMotionClient exist is to call "up" from WebCore
to the WebKit layer. In the platform-based approach, there isn't any need to
call up to the WebKit layer as you can just talk directly down to the platform
layer. Consequently, you don't need DeviceMotionClient.
Similarly, WebViewImpl shouldn't be involved in interactions with the platform
layer. WebViewImpl exists to mediate interactions with the client.
More information about the webkit-reviews
mailing list