On Thu, 7 Oct 2021 19:30:41 GMT, Phil Race <[email protected]> wrote:

> I don't think "we" have a problem. I think there is a macOS bug.

In both cases, we should execute the same code and request the same number of 
pixels(the one pixel) from the macOS. The robot native code should not be 
affected by the uiScale parameter, since the native macOS API gets the 
coordinates/size in the device space.

>  So let's leave the tests running using uiScale=1 here they are doing 
> something useful and withdraw this PR

That tests found a bug, and it will be better to fix it other than workaround 
it.

-------------

PR: https://git.openjdk.java.net/jdk/pull/5766

Reply via email to