Re: Multiple monitors, VirtualScreens and/or Xrandr

Hi Stefan,

Le 29/03/2018 à 18:07, Stefan Monnier a écrit :
> [snip]
>>>> * f.occupy window now uses layout (to stay visible);
>>> Not sure what this means.
>> f.occupy displays a window. In case the screen is not rectangular, this
>> makes the window always visible even when it could appear (even
>> partially) in a hidden zone...
>> +-------+-----+
>> | mon1  | mon |
>> +-------+  2  |
>>  hidden |     |
>>   zone  +-----+
> 
> Ah, I see.  I thought that it used to be displayed at the location of
> the affected window (tho a quick test seems to show it's actually
> placed wherever the mouse was when the action was triggered).

Yes, centered at this location. But when close to an edge, a part of the
window can be displayed outside. This case was handled in a
big-rectangle-screen fashion, but not when all monitors do not have the
same size. Now it is OK. Note that same policy applies when displaying a
menu and for f.identify window.

> [snip]
> 
>>>> Does anyone really use VirtualScreens?
>>> I'm not using it right now, but I've used it in the past and I liked it.
>> You're lucky, I have never succeeded to use it without any problem...
> 
> It's been a while since I managed to use it.  I liked the ability to
> switch workspaces separately.
> 
> I tried it with a manual
> 
>     VirtualScreens {
>      "1200x1600+0+0"
>      "1200x1600+1200+0"
>     }
> 
> and it seemed to be somewhat working.  For some reason it seems like it
> doesn't "bring in" the previously existing windows (at least I couldn't
> see them, I had to remove the VirtualScreens declaration and restart
> ctwm to recover my windows).  And clicking in the background of the
> second monitor brought up the menu in the first monitor :-(
> I'm sure there are many more quirks.> [snip]

That describes pretty well the problems I encountered :)

++

Max.