Initial fixes for window position at or beyond (0,0) under win32.
This is not the whole story, but is better.
We were messing up the initialisation of num_screens on Win32 in that
it was being set to a non-zero value by a call to
GetSystemMetrics(SM_CMONITORS) but was then subsequently incremented
even higher by each iteration call to the screen_cb(...) function,
so what we were storing for the screen rectangles was a bit messed up.
Under fltk-1.1, the code used to explicitly reset num_screens to zero
before starting to iterate the screen_cb(...) function, so that worked OK.
I have sort-of restored that behaviour, and removed the (now redundant)
call to GetSystemMetrics(SM_CMONITORS) since we *always* enumerate the
monitors now to get the per-screen DPI values out.
I've also put in a commented out stub where we could store the per-screen
work-area, which it seems we do also need now, though I have not yet
contrived to actually implement that.
Volunteers welcomed...
git-svn-id: file:///fltk/svn/fltk/branches/branch-1.3@8981 ea41ed52-d2ee-0310-a9c1-e6b18d33e121
@ -211,10 +208,10 @@ void Fl::screen_xywh(int &X, int &Y, int &W, int &H, int mx, int my) {
@@ -211,10 +208,10 @@ void Fl::screen_xywh(int &X, int &Y, int &W, int &H, int mx, int my) {
@ -300,20 +297,20 @@ void Fl::screen_xywh(int &X, int &Y, int &W, int &H, int mx, int my, int mw, int
@@ -300,20 +297,20 @@ void Fl::screen_xywh(int &X, int &Y, int &W, int &H, int mx, int my, int mw, int