I think the major changes to RadWindow could and should have been better documented. From the release notes not much appears to have changed:
RadWindow
What’s New:
Support for the Windows 7 gestures for maximize, minimize, etc.
What’s Fixed:
Greatly improved the performance in Windows XP and Windows Vista/7 without Aero.
If there is a slider placed in the RadWindow, the slider causes some layout issues.
From a forum post a very different story is told:
"We remade big part of RadWindow and decided that this is the correct way to do things. Previously when RadWindow was declared in XAML we were removing it from its Parent and put it in a popup then show it. Clients tend to bind elements in the window to another XAML element and this binding getting broken and there were also other problems like panels layout is refreshed because of removing the window etc...
We believe that it is better practice to either create RadWindow in code behind or use it as a root element of a UserControl especially in WPF as there it is supposed that new windows are always root elements. So that is the reason that we changed the way RadWindow works when declared in XAML."
(http://www.telerik.com/community/forums/silverlight/window/can-t-move-window-in-q2-2011-release.aspx#1721022)
I realise you may have disapproved of declaring windows in xaml, but the fact is this worked and was an efficient solution for simple windows with functionality that was bound up with the main form. After wasting many hours figuring out what was going wrong we've now gone back to the previous version of Telerik controls, as we don't have time to make the numerous changes necessary to run our application with the new binaries. My question is whether future service packs will re-enable xaml declarations of windows, or whether we do indeed need to embark on a major rewriting and testing exercise with no tangible benefits in terms of functionality. I understand that breaking changes are sometimes necessary, but do think you could have taken more care in letting us know about this one.
Thanks, James.
RadWindow
What’s New:
Support for the Windows 7 gestures for maximize, minimize, etc.
What’s Fixed:
Greatly improved the performance in Windows XP and Windows Vista/7 without Aero.
If there is a slider placed in the RadWindow, the slider causes some layout issues.
From a forum post a very different story is told:
"We remade big part of RadWindow and decided that this is the correct way to do things. Previously when RadWindow was declared in XAML we were removing it from its Parent and put it in a popup then show it. Clients tend to bind elements in the window to another XAML element and this binding getting broken and there were also other problems like panels layout is refreshed because of removing the window etc...
We believe that it is better practice to either create RadWindow in code behind or use it as a root element of a UserControl especially in WPF as there it is supposed that new windows are always root elements. So that is the reason that we changed the way RadWindow works when declared in XAML."
(http://www.telerik.com/community/forums/silverlight/window/can-t-move-window-in-q2-2011-release.aspx#1721022)
I realise you may have disapproved of declaring windows in xaml, but the fact is this worked and was an efficient solution for simple windows with functionality that was bound up with the main form. After wasting many hours figuring out what was going wrong we've now gone back to the previous version of Telerik controls, as we don't have time to make the numerous changes necessary to run our application with the new binaries. My question is whether future service packs will re-enable xaml declarations of windows, or whether we do indeed need to embark on a major rewriting and testing exercise with no tangible benefits in terms of functionality. I understand that breaking changes are sometimes necessary, but do think you could have taken more care in letting us know about this one.
Thanks, James.