Hi,
I have this code in a dw's other() event. I've stripped out most of the code, just to illustrate.
// v1 and v2, mjl, 13/04/20: Since win8 and w10 totally ignore the mouse settings when working remotely, I've re-introduced the code to scroll:
CONSTANT Long WM_MOUSEWHEEL = 522
CONSTANT Longptr WM_OUTLOOK_DROP = 111
CHOOSE CASE wparam
CASE WM_MOUSEWHEEL
RETURN 1
CASE WM_OUTLOOK_DROP // email dropped from Outlook?
RETURN 0
END CHOOSE
RETURN 0
There's no problem at all when scrolling with the mousewheel
Now, since the wparam really is a longptr, I decided to define my WM_MOUSEWHEEL also as such.
Result is first I don't see anything on the dw untill I scroll a bit and then I can scroll down, but cannot scroll up.
I've reported this as a private bug, since my video which accompanies it holds private data, but the only answer I've gotten (which I do appreciate), is that on 32bit, longptr's are interpreted as long at runtime. So far, It's not being considered as a possible bug. Fine, the explanation is good, but it doesn't make sense, since I'm comparing a longptr with a longptr in the choose case (after having changed the WM_MOUSEWHEEL constant to longptr).
In our full application based on pfc classes, everything just freezes / hangs.
Has anyone else have the idea that this is a bug or should I just have some more rum on my birthday?
Please watch this video which shows how to reproduce my case:
I'll now also publish this as a public possible bug.