Re: user32: Only call SetWindowPos() in UpdateLayeredWindowIndirect() if needed. (try 4)
Adam Martinson adam.r.martinson@gmail.com wrote:
(We should not be calling SetWindowPos() in UpdateLayeredWindowIndirect() at all.) Fixes bug 26924.
That's not what I was asking about (a bunch of message tests with various parameters). Your current version essentially tests nothing.
You should call UpdateLayeredWindowIndirect with a buch of differnt parameters, and check what message sequence they produce and how they change window state.
See dlls/user32/tests/msg.c how to create a message test.
On 02/05/2012 09:46 PM, Dmitry Timoshkov wrote:
Adam Martinsonadam.r.martinson@gmail.com wrote:
(We should not be calling SetWindowPos() in UpdateLayeredWindowIndirect() at all.) Fixes bug 26924.
That's not what I was asking about (a bunch of message tests with various parameters). Your current version essentially tests nothing.
You should call UpdateLayeredWindowIndirect with a buch of differnt parameters, and check what message sequence they produce and how they change window state.
See dlls/user32/tests/msg.c how to create a message test.
The whole point of the tests I wrote was to show that UpdateLayeredWindowIndirect() *doesn't* generate window messages, at least not as far as I've seen. Is there a better way to test that? Wouldn't these tests fail if it did?
Adam Martinson adam.r.martinson@gmail.com wrote:
The whole point of the tests I wrote was to show that UpdateLayeredWindowIndirect() *doesn't* generate window messages, at least not as far as I've seen. Is there a better way to test that? Wouldn't these tests fail if it did?
A message test could help to prove that UpdateLayeredWindow really doesn't generate the messages. Also my point is not only about messages, but also about *a buncn* of tests which check window attributes before and after each call.
-
Adam Martinson -
Dmitry Timoshkov