Review Request: Rewrite kinetic scrolling on ScrollWidget

igorto igor_trindade at yahoo.com.br
Wed Mar 17 13:20:50 CET 2010


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard.kde.org/r/3312/#review4534
-----------------------------------------------------------


Instead of change the ScrollWidget code you should have a look in plasma/private/kineticscroll.* because there is the old kinetic scroll implementation and it is used by others classes(like plasma/webview) and would be difficult maintain two different kinetic scrolling implementations.

- igorto


On 2010-03-17 04:59:11, Zack Rusin wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviewboard.kde.org/r/3312/
> -----------------------------------------------------------
> 
> (Updated 2010-03-17 04:59:11)
> 
> 
> Review request for Plasma.
> 
> 
> Summary
> -------
> 
> As previously discussed with this approach we can actually properly intercept events from child items. Furthermore now we properly "steal" events which cause a flick (it's important for child items to properly act on mouseReleaseEvents and not on mousePressEvents as some like to do, since it's the release events that cause a flick) so items don't get clicks when flicked. The physics and especially the overshoot behavior is a lot better in this code as well.
> I tried to preserve the old behavior and emit the scrollStateChanged when needed, but I'm not quite sure what that signal was meant to be good for.
> 
> 
> Diffs
> -----
> 
>   trunk/KDE/kdelibs/plasma/widgets/scrollwidget.h 1102878 
>   trunk/KDE/kdelibs/plasma/widgets/scrollwidget.cpp 1102878 
> 
> Diff: http://reviewboard.kde.org/r/3312/diff
> 
> 
> Testing
> -------
> 
> Done in a custom app. Would be nice if someone double checked it with other things that use ScrollWidget though (e.g. the notebook shell :) ).
> 
> 
> Thanks,
> 
> Zack
> 
>



More information about the Plasma-devel mailing list