A caching / streaming HTTP Proxy using Flask and Requests.

I've seen some patterns out there for doing request proxying through Flask, but my use-case required something a little bit more complicated, and this is what I've come up with.

The APIRequestProxy object takes a simple dictionary which describes a proxy requext, instantiates the upstream object to create the upstream request, and provides the results as a Flask Response object.

Most HTTP verbs are supported, and an optional JSON payload can be included -- stick it inside a key named 'data' inside the proxy_request dictionary.

A primitive caching implementation is included, but disabled by default. The fields 'enable_cache' and 'cache_age' control the response caching behavior.

The status code is set to the result from the upstream request by default, but this behavior can be disabled by including 'disable_status_passthrough=True' in the proxy_request dictionary.

The response can be stored as an attribute or streamed directly back to the client (more efficient when fetching large payloads.)

I recommend putting careful thought into possible attack vectors and input validation before implementing this code.

Usage Example 1:

proxy_request = {
    'url': 'http://some_valid_url',
    'method': 'POST',
}
proxy = APIRequestProxy()
proxy.proxy_request = proxy_request
print(proxy.stream_response())

Usage Example 2:

proxy_request = {
    'url': 'http://www.google.com',
    'method': 'GET',
    'enable_cache': True,
    'cache_age': 5,

}
payload = {
    'some_resource': {...}
}
proxy = APIRequestProxy()
proxy.proxy_request = proxy_request
proxy.payload = payload
proxy.make_response()
return proxy.response