This is the python generator for RAML. It creates:
- requests - probably the most common
- httplib/urllib2 - built-in libraries
- httplib2/urllib2 - enhanced versions of built-in libraries
- fetchurl - not to be confused with the google app engine fetch
- Google Fetch API - Google app engine specific
- Tornado HTTPClient - Hi-speed Async requests
- pulsar - newer async framework
possible future implementations:
- pulsar - newer async framework
not supported
- urllib2 - only GET/POST supported. use httplib directly.
- grequests - gevent version of requests. Doesn't work at ALL with SSL (how useless is that?)
Mock Object - generates a mock object that returns minimal information.
- Can be used as a base object for overloading
- Can be used as a stub object for a guideline of what signamtures are expected
WSGI Middleware - the "standard" way to do it.
Gevent WSGI Server - gevent's high-performance async implementation of WSGI
Tornado Server - one of the most popular non-gevent async frameworks (from Facebook)
- Is there even a middleware concept in Tornado?
- If not, maybe just supply a routing table. Probably a list of (url,handler) tuples.
Pretty much most servers will server up WSGI middleware just fine