HTTP is changing under our feet. HTTP/1.1, our old friend, is being supplemented by the brand new HTTP/2.0 standard. HTTP/2.0 provides many benefits: improved speed, lower bandwidth usage, better connection management, and more.
Unfortunately, most web servers do not support HTTP/2.0 at this time. What's needed is to abstract the difference between HTTP/1.1 and HTTP/2.0, so that your application can reap the benefits of HTTP/2.0 when possible whilst maintaining maximum compatibility.
Enter hyper:
from hyper import HTTPConnection
conn = HTTPConnection("www.python.org")
conn.request("GET", "/index.html")
r1 = conn.getresponse()
Did that code use HTTP/1.1, or HTTP/2.0? You don't have to worry. Whatever
www.python.org supports, hyper will use.
hyper is intended to be a drop-in replacement for
httplib/http.client, with an identical API. You can get all of the
HTTP/2.0 goodness by simply replacing your import httplib or
import http.client line with import hyper as httplib or import hyper
as http.client. You should then be good to go.
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.