Recover from HTTP errors like read timeout

[nemobis]

  File "/home/users/federico/.local/lib/python2.7/site-packages/mwclient/client.py", line 286, in api
    info = self.raw_api(action, http_method, **kwargs)
  File "/home/users/federico/.local/lib/python2.7/site-packages/mwclient/client.py", line 434, in raw_api
    http_method=http_method)
  File "/home/users/federico/.local/lib/python2.7/site-packages/mwclient/client.py", line 395, in raw_call
    stream = self.connection.request(http_method, url, **args)
  File "/home/users/federico/.local/lib/python2.7/site-packages/requests/sessions.py", line 533, in request
    resp = self.send(prep, **send_kwargs)
  File "/home/users/federico/.local/lib/python2.7/site-packages/requests/sessions.py", line 646, in send
    r = adapter.send(request, **kwargs)
  File "/home/users/federico/.local/lib/python2.7/site-packages/requests/adapters.py", line 529, in send
    raise ReadTimeout(e, request=request)
requests.exceptions.ReadTimeout: HTTPSConnectionPool(host='biologija.fandom.com', port=443): Read timed out. (read timeout=30)

I think it's reasonable to expect the library to try a little harder when it receives an unexpected response. (Here it's Wikia which every now and then terminates all connections for whatever reason.) Downstream code can catch such exceptions but it would be rather tedious for instance to tweak waiting times and timeouts outside of mwclient.

An approach I've used with reasponable success in dumpgenerator is to override the default HTTPAdapter as suggested in https://stackoverflow.com/a/35504626/1333493 or https://2.python-requests.org/en/master/api/#requests.adapters.HTTPAdapter