[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kwin
Subject:    Re: [PATCH] bug #243693  -> glide effect... ouch!
From:       Martin =?iso-8859-15?q?Gr=E4=DFlin?= <kde () martin-graesslin ! com>
Date:       2010-10-07 14:25:05
Message-ID: 201010071625.16369.kde () martin-graesslin ! com
[Download RAW message or body]

[Attachment #2 (multipart/signed)]


On Thursday 07 October 2010 16:00:21 Thomas Lübking wrote:
> a) QHash is not an array (and actually a QMap should be faster in this
> context, since the item number is usually pretty small)
Qt should remove the [] operator ;-)
> b) Pointers don't magically delete themself when set to NULL... (and
> there's no moc either)
> c) the effect kept _all_ mapped (and handled) windows in its internal list
> and if a window was in this list ( what's more or less "if (true)" ) set
> it to TRANSFORMED (this probably blocked the blur effect)
probably
> d) i recall that the TRANSFORMED mask has to be set in prePaintWindow and
> *not* in paintWindow, yesno?!
yes, if set in paintWindow it should crash in debug builds.
> -> set a property for IsGlideWindow (since all deleted aren't managed) and
> stripped down the entire thing
looks good, I think the Sheet effect needs the same adjustment (and the sheet 
effect was based on another effect). I especially like the idea of the 
property - that's something we could make as a pattern in more effects.

Concerning the notice: the repaints are required in case e.g. a fullscreen 
effect gets activated while a window is animated and the window is not shown 
(e.g. present windows). By adding the repaint the effect continues to add time 
to the progress so that the window does not continue to be animated when the 
window gets visible again.

["signature.asc" (application/pgp-signature)]

_______________________________________________
kwin mailing list
kwin@kde.org
https://mail.kde.org/mailman/listinfo/kwin


[prev in list] [next in list] [prev in thread] [next in thread]