Paolo Insogna
Paolo Insogna
@joyeecheung The option was renamed when I turned it on so yes, we need a full revert unfortunately. Anyway, just for context, I'd like to emphasize that the option existed...
That might be an acceptable solution, but I'm still pretty skeptical about the approach. I think we might be able to solve all the (major) issues in few weeks. So...
Hello! First of all thanks for your PR. You haven't messed up anything, no worries. The only missing piece is a test to verify this change. About the subject of...
First of all, can you fix the lint problem outlined in the "Files changed tab"? Second, you'll need to edit `net.md` and change the doc for `maxConnections`. Before the option...
@juanarbol Nope, it's not. That test only included those addresses already in order, so there was no verification the attempt order was correct. In #48258 a test has been specifically...
I think it's bad luck. None of the failures seems to be related to this change. :(
Thanks for all your incredible effort Yagiz. I hope you recover soon! ^^
> This is missing a way for _other_ assertion frameworks to tap into the test planning. If I recall correctly, the need for this was what stopped @cjihrig working on...
I would avoid enforcing people to use `t.assert` and then move away from it (since, imho, is suboptimal DX). So if we find a way to do that from day...
Note that a very trivial initial implementation based a global symbol and AsynchronousLocalStorage would make the trick.