[Webkit-unassigned] [Bug 73544] [EFL] Support for Web Notification.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=73544





--- Comment #46 from Kihong Kwon <kihong.kwon at samsung.com>  2012-06-04 19:19:26 PST ---
(From update of attachment 145673)
View in context: https://bugs.webkit.org/attachment.cgi?id=145673&action=review

>> LayoutTests/platform/efl/TestExpectations:726
>> +// Efl doesn't support to simulateDesktopNotificationClick.
> 
> Efl -> EFL.

OK.

>> Source/WebCore/ChangeLog:8
>> +        Change requestPermission can be processed when argument is null on the JSC.
> 
> Change requestPermission -> To change requestPermission ?

I don't think so, I meant "Fix to run requestPermission even if arguemnt is empty on the JSC".
I will change this description~;-)

>> Source/WebKit/efl/WebCoreSupport/NotificationPresenterClientEfl.cpp:53
>> +                    printf("REPLACING NOTIFICATION %s\n", iter->first->isHTML() ? iter->first->url().string().utf8().data() : iter->first->title().utf8().data());
> 
> Remove printf(). If you wanna print debug log, use LOG_XXX. Or, if you wanna pass layout test, you should move this printf to DRT side.

QT port is implemented like this.
In the chromium port, there are additional classes only for the layout tests.
But, I think if I make additional classes for the layout tests, It's not a tests for real implementation. That's why I use printf in the middle of implementation like QT.
How do you think about this?

>> Source/WebKit/efl/WebCoreSupport/NotificationPresenterClientEfl.cpp:63
>> +    evas_object_smart_callback_call(m_view, "notification,show", static_cast<void*>(ewkNotification));
> 
> If you add new signal, you should add signal description to ewk_view.h.

OK.

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