On Tue May 6 11:34:26 2025 +0000, Jinoh Kang wrote:
[MSDN][wu] says WM_USER is reserved for use in class-specific messages, so I'd shy away from using it for predefined controls like ListBox. I cannot find a precedent for using WM_USER as a marker in msg.c either. Of course we could find guaranteed-unused ID or split seq, but I have a better idea: how about we test the invalidation / update region directly? We could use GetUpdateRect() or GetUpdateRgn() for that. Should be less flaky too. [wu]: https://learn.microsoft.com/en-us/windows/win32/winmsg/wm-user
Isn't this mostly a speculation? WM_USER is a normal window message regardless what MSDN states about it. It could potentially cause a problem if the control would use it, and it would appear in the message sequence. Personally I don't see any problem with using WM_USER for any purpose in the tests.