Rauth with google provider and python3 - python-3.x

I've got some problem using the rauth library with flask, python3 and google oauth, python2 works without problems.
It says TypeError: the JSON object must be str, not 'bytes'
Here is the log error:
I've found that Issue here and tried to convert the byte to string
decoder=lambda b: json.loads(str(b))
but without success.
Here is my Implementation
class GoogleSignIn(OAuthSignIn):
def __init__(self):
super(GoogleSignIn, self).__init__('google')
self.service = OAuth2Service(
name='google',
client_id=self.consumer_id,
client_secret=self.consumer_secret,
authorize_url='https://accounts.google.com/o/oauth2/auth',
access_token_url='https://accounts.google.com/o/oauth2/token',
base_url='https://www.googleapis.com/plus/v1/people/'
)
def authorize(self):
return redirect(self.service.get_authorize_url(
scope='email',
response_type='code',
redirect_uri=self.get_callback_url())
)
def callback(self):
if 'code' not in request.args:
return None, None, None
oauth_session = self.service.get_auth_session(
data={'code': request.args['code'],
'grant_type': 'authorization_code',
'redirect_uri': self.get_callback_url()},
decoder=json.loads
)
me = oauth_session.get('me').json()
me_email = None
for e in me['emails']:
if e['type'] == 'account':
me_email = e['value']
return (
me.get('id'),
me.get('displayName'),
me_email)
There is anothere way mentioned to use own decoder, but dont't know how to do that, please help me.

I figured out some dirty workaround using simplejson module (also got a loads method so nothing to change here)
importing that as
import simplejson as json
before Google class lets the authentification work as expected

Related

Using hydra.main on main method

Summary of the problem:
I'm running a requests call on an API endpoint, whose request params are hidden in a config file and I decided to try out hydra to retrieve those params [Reason being the request params do change as I'm working on collecting custom dataset using RapidAPI]
I have created a class called QueryParamsLocations which implements the getter methods to fetch the parameters to be later used by run_query method.
class QueryParamsLocations(QueryParams):
#hydra.main(config_path='configs', config_name='location_query')
def get_params_query_string(self, cfg: DictConfig) -> dict():
return {
'query': cfg.location_params.query,
'locale': cfg.location_params.locale,
'currency': cfg.location_params.currency
}
#hydra.main(config_path='configs', config_name='location_query')
def get_url(self, cfg: DictConfig) -> str():
return cfg.urls.location_url
#hydra.main(config_path='configs', config_name='location_query')
def get_headers(self, cfg: DictConfig) -> dict():
return {
'X-RapidAPI-Host': cfg.headers.x_rapidapi_host,
'X-RapidAPI-Key': cfg.headers.x_rapidapi_key
}
class QueryParams is an abstract class which has these 3 getter templates. run_query method is an external call to run the request.
#hydra.main(config_path='configs', config_name='location_query')
def run_query(cfg: DictConfig) -> None:
try:
LoggerFactory.get_logger('logs/logger.log', 'INFO').info('Running query for location')
qpl = QueryParamsLocations()
response = requests.request("GET", qpl.get_url(cfg), headers=qpl.get_headers(cfg), params=qpl.get_params_query_string(cfg))
print(response.json())
except Exception as e:
LoggerFactory.get_logger('logs/logger.log',
'ERROR').error(f'Error in running query: {e}')
run_query()
While running run_query without if name == 'main': and with it as well , the following error is encountered :
[2022-05-16 13:43:32,614][logs/logger.log][ERROR] - Error in running query: **decorated_main()** takes from 0 to 1 positional arguments but 2 were given
Although newer version of hydra (I'm using hydra-core==1.1.2) uses two arguments while creating cfg object however , I'm not sure as to whether there's other way of handling this as such.
Also, by searching through other threads, following was also tried - Compose API
however, from the docs, it requires an override parameter , which is not needed atm.
Would like to know if any other approach can be tried out. Happy to provide more details if needed.
Definitely use the Compose API and not hydra.main() for this use case.
You can just pass an empty array for your override list if you have nothing to override.

Pytest Patch: Different Mock is used

I am trying to patch an imported class called BotoAWSRequestsAuth, as I want to assert a call that uses the auth.
When I patch the imported class and then try to use the patch in my assertion is is failing because a different magic mock is used, I cant seem to figure out why this is happening.
There error I get is:
AssertionError: expected call not found.
Expected: get('https://x/x/x/x/x', auth=<MagicMock name='BotoAWSRequestsAuth' id='140061268160816'>)
Actual: get('x/x/x/x/x', auth=<MagicMock name='BotoAWSRequestsAuth()' id='140061270695792'>)
.....
Differing items:
{'auth': <MagicMock name='BotoAWSRequestsAuth()' id='140061268160816'>} != {'auth': <MagicMock name='BotoAWSRequestsAuth' id='140061270695792'>}
I have tried to create an extracted and minimal version of my code:
In the module that uses the auth:
from aws_requests_auth.boto_utils import BotoAWSRequestsAuth
def call_api(url):
auth = BotoAWSRequestsAuth(
aws_host = "host",
aws_region = 'eu-west-1',
aws_service = 'execute-api'
)
r = requests.get(url, auth=auth)
data = r.content
And in the test:
with patch(
"my_module.BotoAWSRequestsAuth"
) as mock_requests_auth:
my_module.call_api(url)
mocked_requests.get.assert_called_once_with(
url,
auth=mock_requests_auth
)
Is it obvious what I am doing wrong? Maybe I am not actually mocking the constructor call at all because the actual object looks like BotoAWSRequestsAuth() in the assertion error.

python django Mock SAML Response from onelogin.saml.auth library using python3-saml

I have implemented for our django back-end application (SP) possibility to login via SAML, as IDP im using Keycloak. It works fine, but I want to write tests to be sure that all logic is being executed correctly. For this I want to generate a post request with SAML as body and mock (unittest.mock.patch) the real request. But i stuck.
Here is my django view, which accepts get and post requests when I try to login via SAML:
class SamlLoginView(View):
#staticmethod
def prepare_django_request(request):
if 'HTTP_X_FORWARDED_FOR' in request.META:
server_port = 443
else:
server_port = request.META.get('SERVER_PORT')
result = {
'https': 'on' if request.is_secure() else 'off',
'http_host': request.META['HTTP_HOST'],
'script_name': request.META['PATH_INFO'],
'server_port': server_port,
'get_data': request.GET.copy(),
'post_data': request.POST.copy(),
}
return result
#never_cache
def get(self, *args, **kwargs):
req = SamlLoginView.prepare_django_request(self.request)
auth = OneLogin_Saml2_Auth(req, settings.SAML_IDP_SETTINGS)
return_url = self.request.GET.get('next') or settings.LOGIN_REDIRECT_URL
return HttpResponseRedirect(auth.login(return_to=return_url))
#never_cache
def post(self, *args, **kwargs):
req = SamlLoginView.prepare_django_request(self.request)
print(req['post_data']['SAMLResponse'])
auth = OneLogin_Saml2_Auth(req, settings.SAML_IDP_SETTINGS)
auth.process_response()
errors = auth.get_errors()
if not errors:
if auth.is_authenticated():
logger.info("Login", extra={'action': 'login',
'userid': auth.get_nameid()})
user = authenticate(request=self.request,
saml_authentication=auth)
login(self.request, user)
return HttpResponseRedirect("/")
else:
raise PermissionDenied()
else:
return HttpResponseBadRequest("Error when processing SAML Response: %s" % (', '.join(errors)))
In my tests, I wanted to directly call the post method, in which there will be a saml inside:
class TestSamlLogin(TestCase):
def test_saml_auth(self, prepare):
client = APIClient()
url = reverse_lazy("miri_auth:samllogin")
saml_resp='<xml with saml response>'
resp = client.post(url, data=saml_resp)
but obviously it shows that request.POST is empty.
I then decided to make a mock for the prepare_django_request function, and manually insert the saml:
def mocked_prepare_request(request):
post_query_dict = QueryDict(mutable=True)
post_data = {
'SAMLResponse': saml_xml,
'RelayState': '/accounts/profile/'
}
post_query_dict.update(post_data)
result = {
'https': 'on',
'http_host': '<http-host>',
'script_name': '/api/auth/samllogin/',
'server_port': '443',
'get_data': {},
'post_data': post_query_dict,
}
return result
class TestSamlLogin(TestCase):
#patch('miri_auth.views.SamlLoginView.prepare_django_request', side_effect=mocked_prepare_request)
def test_saml_auth(self, prepare):
client = APIClient()
url = reverse_lazy("miri_auth:samllogin")
saml_resp='<xml with saml response>'
resp = client.post(url, data=saml_resp)
and depending on how I pass the saml_xml it throws different errors, if i define it as string:
with open(os.path.join(TEST_FILES_PATH, 'saml.xml')) as f:
saml_xml = " ".join([x.strip() for x in f])
it returns: lxml.etree.XMLSyntaxError: Start tag expected, '<' not found, line 1, column 1, although I checked the output from saml_xml in the xml validator and it says that the xml is valid.
When i try to parse the file into xml in advance, i get another error later,
libraries with which I tried to parse:
import xml.etree.ElementTree as ET
from xml.dom import minidom
from lxml import etree
tree = etree.parse(os.path.join(TEST_FILES_PATH, 'saml.xml'))
it returns:
TypeError: argument should be a bytes-like object or ASCII string, not '_ElementTree'
Debugging these errors didn't lead me to any solution.
If anyone has any thoughts on how this can be implemented (Mocking response with SAML), or where I made a mistake, I would be glad to hear.
Thank in advance
I realized that the SAML Response must be encoded:
with open(os.path.join(TEST_FILES_PATH, 'saml.xml')) as f:
saml_xml = " ".join([x.strip() for x in f])
base64_saml = base64.b64encode(saml_xml.encode('ascii')).decode('ascii')
post_data = {'SAMLResponse': base64_saml, 'RelayState': '/accounts/profile/'}
url = reverse_lazy("miri_auth:samllogin")
request = self.client.post(url, post_data)
but now i am getting the following errors:
func=xmlSecOpenSSLEvpDigestVerify:file=digests.c:line=280:obj=sha256:subj=unknown:error=12:invalid data:data and digest do not match

Twisted - Request did not return bytes

I have basic twister app and i keep getting errors like that:
Request did not return bytes
Request:
Resource:
<main.MainPageDispatcher object at 0x7f049fa62be0>
Value:
'hello'
Everywhere, even in official docs' examples I see that string is returned and yet it not works for me. If I comment out first return and send bytes instead of string it is working.
Can anyone help me understand how it works? If it has to be in bytes then why official guides are returning strings?
My code:
from twisted.web.server import Site
from twisted.web.static import File
from twisted.web.resource import Resource
from twisted.internet import reactor
class MainPageDispatcher(Resource):
isLeaf = True
def __init__(self):
super().__init__()
def render_GET(self, request):
request.setHeader(b"content-type", b"text/html")
return "hello"
return bytes("hello", "utf-8")
root = MainPageDispatcher()
factory = Site(root)
reactor.listenTCP(8888, factory)
reactor.run()
In python3 I'm using:
def render_GET(self, request):
request.setHeader("Content-Type", "text/html; charset=utf-8")
return "<html>Hello, world!</html>".encode('utf-8')
str.encode('utf-8') returns a bytes representation of the Unicode string

Not found view doesn't work on Pyramid using traversal

I'm using Pyramid (1.5.7) + traversal and following the documentation I've tried all possible ways to get the "Not found exception view" working.
from pyramid.view import notfound_view_config,forbidden_view_config, view_config
#notfound_view_config(renderer="error/not_found.jinja2")
def not_found_view(request):
request.response.status = 404
return {}
#forbidden_view_config(renderer="error/forbidden.jinja2")
def forbidden_view(request):
return {}
Using contexts:
from pyramid.view import view_config
from pyramid.httpexceptions import HTTPForbidden, HTTPUnauthorized
#view_config(context=HTTPNotFound, renderer="error/not_found.jinja2")
def not_found_view(request):
request.response.status = 404
return {}
#view_config(context=HTTPForbidden, renderer="error/forbidden.jinja2")
def forbidden_view(request):
return {}
I'm using the Scan mode, but I've tried also adding a custom function to the configuration:
def main(globals, **settings):
config = Configurator()
config.add_notfound_view(notfound)
Not luck either, all time getting the following unhandled exception:
raise HTTPNotFound(msg)
pyramid.httpexceptions.HTTPNotFound: /example-url
Ouch... My bad! I was using a tween which was preventing Pyramid to load the Exceptions:
def predispatch_factory(handler, registry):
# one-time configuration code goes here
def predispatch(request):
# code to be executed for each request before
# the actual application code goes here
response = handler(request)
# code to be executed for each request after
# the actual application code goes here
return response
return predispatch
Still I don't know why but after removing this tween all seems to work as expected.

Resources