On Sunday 15 May 2011 15:03:30 Martin Gräßlin wrote: > On Sunday 15 May 2011 14:43:45 Michael Jansen wrote: > > On Sunday 15 May 2011 14:04:34 Martin Gräßlin wrote: > > > On Sunday 15 May 2011 13:36:14 Luciano Montanaro wrote: > > > > A feature like this should not be implemented in a style, IMHO, > > > > styles should be limited to changing the look, not the > > > > behaviour of widgets, and for sure it should not be the default > > > > behaviour. > > > > > > you are fighting windmills here. Reality is that the styles can > > > change the behavior and they do so. Claiming they should not, does > > > not fix anything. The only difference is that you are not aware of > > > a style "breaking" your app if it is not default. > > > > So you say the default kde style is entitled to break kde application at > > will because there are other styles out there which do it already? Is > > that really a accepted kde policy? > > > > Sometimes i wonder. > > So you think we should better disable the feature and close the eyes > although we know that there are bugs with other styles (including for > example the quite popular Ubuntu default style)? I prefer getting the bugs > fixed.
I prefer to keep perfectly valid code working as expected and advertised in the documentation (qt that is). We do not only talk about kde applications here. We talk about qt applications to and probably gtk ones (oxygen-gtk has the same bug/feature?) too. You are saying that an application developer has to check his application not only on every os and distribution but on each and every window manager / style combination and whatever problem he finds is his to fix. Even if the style is breaking perfectly valid code from the outside. You expect each and everyone with broken code too: 1. Find out what suddenly happens (It is not obvious to find. What do you think a pure qt developer will think about it?) 2. Trying to find out about some obscure way to unbreak his code with oxygen (and then most likely with all other styles you mentioned too) that is most likely only documented in some mailing list posts and somewhere deep in the kde api documentation). 3. Fix his code and continue to care for kde. I won't tell you what i think of such an attitude. A bug introduced by oxygen is a bug to be fixed in oxygen. Mike >> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
