[webkit-qt] on QQWVP::initializeDesktop and ::initializeTouch and d->useTraditionalDesktopBehaviour

[Alexis Menard]

On Thu, Nov 10, 2011 at 6:45 PM, Kenneth Rohde Christiansen
<kenneth.christiansen at gmail.com> wrote:
> Hi there,

Hi,

>
> I cannot agree more. Currently we are also seeing a convergence of
> mobile/desktop. Safari was one of the first; there seems to be work
> going on in Chromium adding touch interaction, and Windows 8
> definitely will support more mobile behavior.

Granted.

>
> We shouldn't really be a follower but a leader here and we can be so.
>
> I also think that we should be able to use the tiling code always. It
> is performing extremely well on my laptop, and I only "see tiles"
> while scrolling in the situations there a normal browser would stall.

Since my patch both case use TiledBackingStore and AC. And yes
you're right it works nicely.

>
> The viewport meta handling can easily be fixed (I did the first patch
> already) by not layout out non-viewport-meta enabled page by 980, but
> by the maximum of the actual viewport and the 980. Then the 980 could
> be configurable and easily set to 0 in QML, enabling normal desktop
> behavior for such pages.

I'm not sure to follow the viewport meta handling jargon here :D, sorry.

>
> Cheers,
> Kenneth
>
> On Thu, Nov 10, 2011 at 10:35 PM,  <simon.hausmann at nokia.com> wrote:
>> Hi,
>>
>> I think we can eliminate this boolean altogether. Why don't we always have the interaction engine
>> around (might be nice to get it working on desktop computers with an attached touch screen)?
>> It should act only on touch events.
>>
>> One remaining point is the resizeToContents behaviour, which in turn will toggle whether to show
>> scrollbars or indicators (missing atm).
>>
>> We clearly need a way to forcibly enable/disable resizeToContents, at least for WTR. But ideally
>> we're smart about it and we just initialize it "correctly".
>>
>> We could play with heuristics, such as looking at the screen resolution and the dpi (QGuiApplication::primaryScreen()).
>> If the physical dimension of the "display" is below a certain shreshold (10"?), we simply figure that
>> scrollbars don't make sense and go to resizeToContents mode.
>>
>> Anyway, I think you get what I'm getting at: Let's see if there's a way to avoid a mode/boolean
>> that distinguishes between "desktop" and "mobile".
>>
>> Simon
>> _______________________________________________
>> webkit-qt mailing list
>> webkit-qt at lists.webkit.org
>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-qt
>>
>
>
>
> --
> Kenneth Rohde Christiansen
> Senior Engineer
> Application and Service Frameworks, Nokia Danmark A/S
> Phone  +45 4093 0598 / E-mail kenneth.christiansen at gmail.com
>
> http://codeposts.blogspot.com ﹆﹆﹆
> _______________________________________________
> webkit-qt mailing list
> webkit-qt at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-qt
>



-- 
Alexis Menard (darktears)
Software Engineer
INdT Recife Brazil