Provide a simple HTTP server that tries to echo the request back in the response in the most sensible way possible. This can be useful for, testing, debugging, stubbing out a local server in systems that have a hard-coded assumption of making HTTP requests, etc.::
$ python -m httpdecho Echoing HTTP at http://localhost:8000 ...
Without specifying a port, the server will try to find the next available port starting at 8000 to try and be as predictable as possible::
import sys import time import subprocess from six.moves import SimpleHTTPServer startup_delay = 0.5 simple_popen = subprocess.Popen( ... [sys.executable, '-m', SimpleHTTPServer.name] ... ); time.sleep(1) echo_popen = subprocess.Popen( ... [sys.executable, '-m', 'httpdecho'] ... ); time.sleep(1) echo_popen.poll() simple_popen.kill()
Once running, HTTP requests are echoed in the responses. The default response
body format is basically HTTP header format, from
http.client.HTTPMessage::
import io import requests import email get_response = requests.delete('http://localhost:8001') get_body = email.message_from_string(get_response.text) print(get_body['Method']) DELETE print(get_body['Path']) / print(get_body.get_payload())
Query parameters are also included::
query_response = requests.get( ... 'http://localhost:8001', params=dict(Foo='foo')) query_body = email.message_from_string(query_response.text) print(query_body['Foo']) foo
If the request is a POST or another method that accepts a body on the
request, the body or the responses body will contain the POST body::
post_response = requests.patch( ... 'http://localhost:8001', data=dict(Bar='bar')) post_body = email.message_from_string(post_response.text) print(post_body.get_payload()) Bar=bar
Shutdown the server::
echo_popen.kill()
Features for future releases
Content-Type and Accept support for content negotiation:
Return the response body in the format specified in the Accept header if
given, otherwise in the same Content-Type as the request.