[Webkit-unassigned] [Bug 43258] Implement chromium WebDeviceOrientationClient wrapper and have WebViewImpl get it from WebViewClient.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=43258





--- Comment #10 from Jeremy Orlow <jorlow at chromium.org>  2010-08-03 08:52:19 PST ---
(In reply to comment #9)
> > WebKit/chromium/public/WebDeviceOrientationClient.h:50
> >  +      WebDeviceOrientationClient() { }
> > This works??  I'm pretty sure this will need to be public and virtual.
> AFAIK, constructors can't be virtual. Why would it need to be public? It's only called by subclass constructors (and create(), when that is implemented).

Oops, thought that was a destructor.

> 
> > 
> > WebKit/chromium/src/WebViewImpl.cpp:267
> >  +      , m_webDeviceOrientationClient(WebRuntimeFeatures::isDeviceOrientationEnabled() ? (client ? client->createWebDeviceOrientationClient() : WebDeviceOrientationClientDummy::create()) : 0)
> > What is the dummy for?  I don't think client can ever be non-null.  And, if it can be, then we should just assume the device orientation client is null too then.
> client will be null during some UI tests. This check is analogue to Satish's check in Bug 43240. Either we handle it, or we require it to be non-NULL.

Can we make that not be true easily?  Or can those tests not set the flag?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.