Effective rect of stacks on Windows 10

[Paul Dupuis]

On 10/21/2019 2:11 PM, Trevor DeVore via use-livecode wrote:
> I am looking for input from people who are using `the effective rect of
> stack` and the values it reports on Windows. I have some code that uses the
> effective rect of a group to determine if the stack is still on the screen
> after the user moves the stack or when restoring window positions when an
> app relaunches.
>
> On Windows 10 I just noticed that `the effective rect of stack` includes
> the drop shadow that appears along the left, bottom, and right of a window
> on Windows 10. For my use case including the shadows in the `effective rect
> of stack` serves no value. I'm wondering if someone has a use case where it
> is helpful to have the drop shadow included in `the effective rect`?
>
> Here is how you can check behavior if you are interested:
>
> EXPECTED RESULT: The effective rect of a stack on Windows would return the
> rect of the stack that includes stack borders and title bar.
>
> OBSERVED RESULT: The effective rect includes the drop shadow of the window
> in the rect.
>
> RECIPE:
> 1. Create a new stack
> 2. Set the topleft of the stack to 0,0
> 3. `put the effective rect of this stack`. The left of the rect will be a
> negative value.
> 4. `set the effective topleft of this stack to 0,0`. The left of the stack
> will be offset by the negative value from (3).

Add your name to this bug: 
https://quality.livecode.com/show_bug.cgi?id=16305

Mark Waddigham chimed in at one point that the Windows API returns the 
8px border width under Windows 10 - the 1px visible border and a 7px 
"touch area" is apparently how Microsoft views it, so LC's position is 
that this is sort of an OS vendor issue rather than their.

I disagree of course.