Seth Michael Larson
Seth Michael Larson
Also this seems similar to https://github.com/urllib3/urllib3/issues/3100
I'm also imagining a test case that checks the bytes sent over the socket for the absence of the "h2" ALPN in the TLS ClientHello, I believe we do something...
See this test which verifies that the hostname is in the initial TLS handshake via SNI: https://github.com/urllib3/urllib3/blob/main/test/with_dummyserver/test_socketlevel.py#L103 Can assert the `http/1.1` and `h2` ALPN strings in a similar way.
This PR depends on the global socket manager in: https://github.com/urllib3/urllib3/issues/3301
@pquentin That's correct!
@elminson Yes! It would be all the error conditions documented here: http://python-hyper.org/projects/h2/en/stable/api.html
@zawan-ila I was imagining something like that, but maybe we can save the original LifoQueue instance and do an "is" check before attempting to reimport from the queue module? Does...
I could see a feature like this being useful, especially how you describe it with checking a "last known good" time against the `idle_timeout` on the connection pool. Would cut...
Adding my two cents, if it's encoding as UTF-8 and the world isn't broken we may want to call this a "feature" and encode every body as UTF-8. That's what...
@pquentin Oh gotcha, yeah let's UTF-8 all the things!