You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
That way, we could have a first protocol implementing the "proxy protocol", then upgrading to the pipe one, and have only one implementation of this (currently the TCP proxy has its own implementation of piping data back and forth)
The text was updated successfully, but these errors were encountered:
Hello,
I understand that first thing to do, it's to add Pipe to the Tcp Client. And that has to work like the Pipe in Http client.
But what I don't understand is how can we add this, without creating the upgrade method itself.
The Pipe is create in the Http::Client during a back_readable when a ProtocolResult::Upgrade is read. But the back_readable of Tcp::Client doesn't seem to follow the same behavior.
Thank in advance for your help 😄
for http, it is done in back_readable because we might upgrade from HTTP when the server answered a Connection: Upgrade. For TCP, no need to implement upgrade, you can create the Pipe right from the beginning
That way, we could have a first protocol implementing the "proxy protocol", then upgrading to the pipe one, and have only one implementation of this (currently the TCP proxy has its own implementation of piping data back and forth)
The text was updated successfully, but these errors were encountered: