[webkit-qt] QtWebKit2 and QML Items expected behavior - "There and back again"

[Andras Becsi]

Hi,

On 20 March 2012 01:10, Jesus Sanchez-Palencia <jesus at webkit.org> wrote:
> Hi folks,
>
> As we all know, the only public piece of the brand new QtWebKit2 (Qt5
> + QtQuick 2 + WebKit2) API is our WebView.
> This WebView is a QQuickItem, pretty much like everything else in QML 2.
>
> Therefore I would like to ask what would you guys expect from the
> following cases:
>
> 1- A WebView that has a child Rect.
> Example: https://gist.github.com/2127190

As for this case, the WebView is soon going to be a subclass of
Flickable which makes a child rect (a sibling rect of the page)
automatically reparented to the contentItem of the Flickable, so the
WebView should behave much like the ListView or GridView.
Since we do not hijack the default property of our internal Flickable,
for the flickable WebView currently this can be achieved with the
experimental API like this:

WebView {
    id: webView
    experimental.flickableData: [
       Rect {
           id: foo
       }
    ]
}

We decided to move away from the internal Flickable because for
components like the ScrollDecorator we needed to expose the Flickable,
which makes the forwarded API redundant.
Inheriting Flickable should also make fixing the Touch/Mouse issues we
experience easier.

> 2- A WebView having a shader applied to it.
> Example: https://gist.github.com/2051871
>
>
> I'm not talking about performance issues here and the matter is way
> simpler than even discussing if we should support this use-cases or
> not (or even why supporting them, let's say). When I first thought
> about these 2 examples with Lauro, we were trying something that is
> the very inner base of QML (case 1) and one of the biggest features of
> QML 2 (case 2).
>
> If we are NOT going to support this, then someone needs to document it
> when time comes (Qt5 beta?).
> If we are suppose to support this and these are critical bugs, then I
> can open them and block some sort of meta-bug of the Qt5 release, but
> I'm afraid there is nothing more I can do about it as it (qml
> rendering, webkit rendering, scenegraph, and the junction of all the
> aforementioned) is beyond my field of knowledge.
>
>
> Cheers! <3
> jesus
> _______________________________________________
> webkit-qt mailing list
> webkit-qt at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-qt

/András