On 20.07.24 01:46 , Gary Gregory wrote: > I don't see a matching unit test in [2] so we are asking for a future > regression IMO...
This _is_ tested. Our tests connect to old OpenSSH, not using this key exchange method, and to new OpenSSH method, where it is used. The problem is that to test this, you'd have to test potentially many key exchanges. Against an external implementation, i.e., an OpenSSH container. And with the fix, they all would succeed. So when do you stop? After 1000? After 2000? You _could_ stop once you get a secret that does start with zero. But to the detect that, you'd have to change existing code just to make this observable for the test, and you still wouldn't know after how many key exchanges it would happen. So that's why there is no dedicated test for this. It _is_ tested implicitly in other tests. Cheers, Thomas --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@mina.apache.org For additional commands, e-mail: dev-h...@mina.apache.org
