Closed lfse-slafleur closed 1 month ago
Sure, if it's a requirement it's ok for me to apply the rerename.
I'm wondering how a different implementation of S2 would look like in practice. Let's say you want to communicate XML, we could transform the XML into a dict using xmltodict and apply the current classes. So in that case, It would be adding two new methods to the mixin (from_xml, to_xml).
Regarding the communication protocol, let's say HTTP, I guess that one approach would be to use polling to handle bidirectional communication. In that case, the schema should be the same as in websockets, right?
Thanks for raising a very valid point. The schema remains the same regardless of communication protocols which is why we renamed it in the first place. I did not raise this during our internal discussions. I am going to restart the discussion internally as I would really like to not change the name for many reasons, including the one you are raising.
Closing this issue as current time seems to be accepted.
@victorgarcia98
I notified some of my colleagues of our chosen rename and got some push-back. Their view is that having different implementations of S2 in other protocols than websockets+json is realistic for the future. So we do have to add that to the name before we release.
The first name was s2-ws-json-python. I agree that having a simpler name is more preferable but having both websockets & json in the identifier is given to me as a requirement. Would the previous name s2-ws-json-python (with
import s2wsjson
) suit your needs?