On Sat Jan 13 18:12:01 2024 +0000, Rémi Bernon wrote:
Only had a quick look so far and it looks okay, so feel free to add these too. Fwiw in general we prefer to add tests first, with todo_wine as needed, and then remove the todo_wine as things are implemented. This better shows the desired behavior which is being implemented.
That makes sense.
I had quite a hard time following the test code. I got it working, but I am still not clear on the `wait_hid_expect` calls. Comments indicate they are required for asynchronous items and they seem to be used in combination with * gain report due to acquire, * gain report due to set property, * gain report due to send reset command, * set parameters * Are these asynchronous due to the code writing writing the request out to the `hid_joystick` device handle (which uses the asynchronous `FILE_FLAG_OVERLAPPED` system)? What is it waiting for, and, even if they are asynchronous, why do you need to wait?
Also, it seems the device handle used for various other routines that there is no `wait_hid_expect` calls after, such as downloading, unacquiring, etc.?