I missed this in my original client implementation. Using spawn_blocking means that tests tend to not terminate if something goes wrong, but the server is async, so that is just completely unnecessary. This neatly handles panics in tasks and properly finishes the test if something goes wrong.
I missed this in my original client implementation. Using
spawn_blockingmeans that tests tend to not terminate if something goes wrong, but the server is async, so that is just completely unnecessary. This neatly handles panics in tasks and properly finishes the test if something goes wrong.