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
I can see that you've added Finch.async_request and doc even states this as a way to go to use it with Stream.resource, but
The biggest problem
It doesn't lazily read the data from socket on demand, it just sends whole response body into message queue which can just overflow the server. Any huge (let's say 1TB) response will just shut down any program using this approach. And people usually use Stream for these kind of problems (requesting huge files). So this is just misleading and has nothing to do with stream, please remove this from your documentation
The text was updated successfully, but these errors were encountered:
We should mention that async_request works as a firehose, it will immediately send messages as they come to the current process. I think discouraging its usage with Stream is fine. And we should mention Finch.stream as an option to lazily stream.
I can see that you've added
Finch.async_request
and doc even states this as a way to go to use it withStream.resource
, butThe biggest problem
It doesn't lazily read the data from socket on demand, it just sends whole response body into message queue which can just overflow the server. Any huge (let's say 1TB) response will just shut down any program using this approach. And people usually use
Stream
for these kind of problems (requesting huge files). So this is just misleading and has nothing to do with stream, please remove this from your documentationThe text was updated successfully, but these errors were encountered: