On Fri Nov 25 10:53:39 2022 +0000, Jacek Caban wrote:
This looks suspicious, the image could have changed on the server and the test is not really convincing. One way to explain what you see in tests would be that we should start loading immediately and finish immediately if the image from http protocol is cached. One hint that it's the case is in onerror tests: `pump_msgs(&called_img_onerror); /* FIXME: should not be needed */`.
Yes, I think that's the case, but I'm not sure how to check if it's cached.