Re: [MiNT] AES desktop extension

[Olivier Landemarre <olivier.landemarre@utbm.fr>]

Philipp Donzé wrote:

> Hi,
>
> I've followed your discussion about extending the AES for a better 
> desktop window management.
> Here some thoughts from me:
>
> We should use the normal wind_set(WF_NEWDESK) function to implement 
> your idea. I.e.
> -wind_set(0, WF_NEWDESK, &objc_addr) sets the appropriate desktop object.
> -wind_set(0, WF_NEWDESK, 0L) removes the desktop object and disables 
> desktop management
> -wind_set(0, WF_NEWDESK, -1L) enables the sending of WM_REDRAW messages
>
> Sure, an AES implementation should make sure that if applications are 
> switched it is perhaps necessary to send a redraw message to the 
> dektop window.
>
> If we go further in this direction, we could use the WF_NEWDESK 
> feature for normal windows too!
> It could be used for an easy implementation of window dialogs. I.e. an 
> application can add an object tree to a window just by using 
> wind_set(whandle, WF_NEWDESK, &my_dialog).
> Sure in this case the name WF_NEWDESK should be replaced by something 
> like WF_OBJECT or WF_OBJTREE.
>
> I think this would lead to a proper implementation of your desired 
> feature and further it could be a proper way of implementing window 
> dialogs (and get rid of WDIALOG ;-)).
> Sure for backwards compatibility we need to add this features to old 
> OS like TOS and MagiC.

I have propose to Arnaud this binding:
wind_set(0, WF_NEWDESK, w1+w2 -> &objc_addr,w3 -> -l,w4 ->-1) the two 
first are standard objc_addr we could put a simple object with for 
example only one object w3 and w4 with value -1 to enable sending 
WM_REDRAW messages, other values standard wind_set(), ok this will not 
do the work we need on old AES but it doesn't crash, I don't see how to 
add  this features to old AES, looks difficult.

Olivier