Bug 74522 – [Qt] DeviceOrientationClientQt should initialize m_controller to zero.

• Component: WebKit Qt
• Status: RESOLVED FIXED

[Alexander Færøy]

We currently forget to set m_controller to zero upon construction. We should also forward any calls to setController() to our provider.

Patch coming.

[Alexander Færøy]

Created attachment 119253 [details]
Patch

[Tor Arne Vestbø]

Comment on attachment 119253 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=119253&action=review

> Source/WebKit/qt/WebCoreSupport/DeviceOrientationProviderQt.cpp:33
> +    m_controller = 0;

initializer

[Alexander Færøy]

Created attachment 119257 [details]
Patch

[Alexander Færøy]

Adding Tor Arne.

[Antonio Gomes]

Comment on attachment 119257 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=119257&action=review

> Source/WebKit/qt/WebCoreSupport/DeviceOrientationClientQt.cpp:43
> +    ASSERT(controller);
>      m_controller = controller;
> +    m_provider->setController(controller);

you seem to be doing more than what to changelog'ed.

> Source/WebKit/qt/WebCoreSupport/DeviceOrientationProviderQt.cpp:32
> +    : m_lastOrientation(DeviceOrientation::create())
> +    , m_controller(0)
> +    , m_sensor()

ditto

[Antonio Gomes]

(In reply to comment #5)
> (From update of attachment 119257 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=119257&action=review
> 
> > Source/WebKit/qt/WebCoreSupport/DeviceOrientationClientQt.cpp:43
> > +    ASSERT(controller);
> >      m_controller = controller;
> > +    m_provider->setController(controller);
> 
> you seem to be doing more than what to changelog'ed.

s/what to/what you/g

[Kenneth Rohde Christiansen]

Comment on attachment 119257 [details]
Patch

Write proper changelogs Alex! :-)

[Alexander Færøy]

(In reply to comment #7)
> (From update of attachment 119257 [details])
> Write proper changelogs Alex! :-)

It is because you are not around to pet me!

Jokes aside, new patch coming up in one moment. It is being split up such that the setController() changes are being moved to a separate patch.

[Kenneth Rohde Christiansen]

Comment on attachment 119257 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=119257&action=review

> Source/WebKit/qt/WebCoreSupport/DeviceOrientationProviderQt.cpp:31
> +    , m_controller(0)

why isn't it a smart controller? you seem to use it by two here in this patch? should it be reffed?

[Kenneth Rohde Christiansen]

(In reply to comment #9)
> (From update of attachment 119257 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=119257&action=review
> 
> > Source/WebKit/qt/WebCoreSupport/DeviceOrientationProviderQt.cpp:31
> > +    , m_controller(0)
> 
> why isn't it a smart controller? you seem to use it by two here in this patch? should it be reffed?

smart pointer :-)

[Alexander Færøy]

Created attachment 119278 [details]
Patch

[WebKit Review Bot]

Comment on attachment 119278 [details]
Patch

Clearing flags on attachment: 119278

Committed r102822: <http://trac.webkit.org/changeset/102822>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.