HTTP is changing under our feet. HTTP/1.1, our old friend, is being supplemented by the brand new HTTP/2 standard. HTTP/2 provides many benefits: improved speed, lower bandwidth usage, better connection management, and more.

hyper provides these benefits to your Python code. How? Like this:

from hyper import HTTPConnection

conn = HTTPConnection('nghttp2.org:443')
conn.request('GET', '/httpbin/get')
resp = conn.get_response()

print(resp.read())

Simple.

Please be warned: hyper is in a very early alpha. You will encounter bugs when using it. In addition, there are very many rough edges. With that said, please try it out in your applications: I need your feedback to fix the bugs and file down the rough edges.

hyper supports the final draft of the HTTP/2 specification: additionally, it provides support for drafts 14, 15, and 16 of the HTTP/2 specification. It also supports the final draft of the HPACK specification.

hyper is intended to be a drop-in replacement for http.client, with a similar API. However, hyper intentionally does not name its classes the same way http.client does. This is because most servers do not support HTTP/2 at this time: I don't want you accidentally using hyper when you wanted http.client.

Looking to learn more? Documentation for hyper can be found on Read the Docs.

hyper welcomes contributions from anyone! Unlike many other projects we are happy to accept cosmetic contributions and small contributions, in addition to large feature requests and changes.

Before you contribute (either by opening an issue or filing a pull request), please read the contribution guidelines.

hyper is made available under the MIT License. For more details, see the LICENSE file in the repository.

hyper is maintained by Cory Benfield, with contributions from others. For more details about the contributors, please see CONTRIBUTORS.rst.