I'm trying to write a Python 3.5 Flask application that redirects a user to an OAuth URL, for authentication / authorization. As part of that redirection, I have to include the Authorization header. The built-in redirect() method in Flask doesn't seem to support adding HTTP headers.
What's the proper way of handling this in such an application?
You will need to build your own response object to add headers. You can check out the docs here: http://docs.python-requests.org/en/master/api/#requests.Response
A simple example for your use case would be something like:
response = Response(headers={'Authorization': 'whatever'},
is_redirect=True,
url="https://your-redirected-url.com")
return response
Edit: Further info
Also, I would check out https://github.com/lepture/flask-oauthlib if you are interested in using a library. It has support for oAuth1 and oAuth2 and it is relatively easy to setup with a standard Flask app.
Edit: Another way of doing it
This morning I remembered a simpler way to do this. You can call the redirect function and it will return a flask Response object. Then you are able to set the headers on that newly created object.
response = redirect('https://url')
response.headers = {'authorization': 'whatever'}
return response
Related
I wonder if there is a way to authenticate each redirected URL when working with Python modules such as httpx or requests?
Problem Statement
I am trying to connect to an API endpoint under the company network. Due to the company's cyber security measures, the API endpoint will be randomly masked with a company proxy, which causes the 307 Redirect status code.
my current code snippet looks like the below:
import httpx
api_url = 'https://demo.vizionapi.com/carriers'
head = {
'X-API-Key':'API KEY'
}
response = httpx.get(url=api_url, verify='supporting_files/cacert.pem',
headers=head, auth=('my username', 'my password'),
follow_redirects=True)
With above code, I received the 401 authentication needed error (But auth has been passed). This error will only happen when redirection occurs due to the company proxy.
Question:
My assumption is the authentication is only being passed into the first URL not the redirected URL. Therefore, I wonder if anyone know how I can use the same auth parameter for all URLs (direct & redirect)?
Any suggestion will be deeply appracaited.
I don't know what requests behavior with regards to auth during redirect is, but the first solution to come to mind is to manually follow the redirects yourself. Put your request in a loop that checks for the 3xx response codes, and handle auth however you want to.
I'm trying to use Streamlabs API. Streamlabs API uses Oauth2 for creating apps. So first I send whoever's using my app to an authorization link containing my app's client id and the scopes I want to use.
(Something like this: streamlabs.com/api/v1.0/authorize?client_id=CLIENT-ID-HERE&redirect_uri=REDIRECT-URI&response_type=code&scope=SOME+SCOPES+HERE)
Once I've done that I receive a code at the redirect uri specified. I then use that code to get the access token for permanent access to the connected user's account. I then receive the access token from a POST request that works perfectly... Now I run into the problem. When getting the temporary code before the access token I specified the scopes: "donations.read +donations.create+alerts.write+alerts.create".
When authorizing, the app asks for permission to the different scopes. The scope in focus is "alerts.write" so that I can send test alerts using POST requests. But this doesn't work for some reason. To send a test alert I have to send a POST request to this url: "https://streamlabs.com/api/alerts/send_test_alert"
I've tried doing that in two different ways.
1:
import requests
url = "https://streamlabs.com/api/alerts/send_test_alert"
data = {
"access_token":"UserAccessTokenHere",
"type":"donation"
}
response = requests.post(url=url, data=data)
print(response.text)
2:
import requests
url = "https://streamlabs.com/api/alerts/send_test_alert?access_token=UserAccessTokenHere&type=donation"
response = requests.post(url=url)
print(response.text)
If I do print(response) it prints "Response [405]".
But if I do print(response.text) I get a long HTML document for this page: Error response page
Any ideas what's going wrong with my Python requests? send_test_alert documentation here: Link
I've contacted support and looks like you've made the same error as me.
You're not actually sending a request to the right URL.
You are a sending a request to: "https://streamlabs.com/api/alerts/send_test_alert"
You should be using the URL: "https://streamlabs.com/api/v1.0/alerts/send_test_alert"
I have been trying to use the Django-REST authentication to validate the user name /password given in a desktop app.
On the server side, I have installed the following DJANGO-REST-FRAMEWORK-JWT package found here:
https://github.com/GetBlimp/django-rest-framework-jwt
I have gone through the example and when I run the following on the command line get a token as a response:
curl -X POST -d "username=luca&password=letmein123" http://localhost:8000/api-token-auth/
And I get:
{"token":"eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VybmFtZSI6InBhbmthaiIsInVzZXJfaWQiOjIsImVtYWlsIjoiIiwiZXhwIjoxNDc5MTE5NzQ2fQ.RA085m-YSnGFheykCCxSVxI_9rW9AC9kEaOkUB5Gm0A"}
I tried something like:
import requests
resp = requests.post('http://127.0.0.1:8000/api-token-auth/', data={}, auth=('luca', 'letmein123'))
However, this always returns response code 400 with Bad request
My question is how can I do that from my desktop python app. I basically want to call the same API with the username and passord and be able to process the response and access protected APIs.
The auth parameter of requests.request is by default meant for Basic/Digest/Custom HTTP Auth, analogous to the -u, --user <user:password> parameter of curl.
You could define your own custom Authentication class to achieve the desired result, but the basic way to achieve the same result as your original curl request is:
resp = requests.post(
'http://localhost:8000/api-token-auth/',
data={"username": "luca", "password": "letmein123"})
The data dictionary can alternatively be supplied as json by using the json parameter if you prefer (the request would be different, but also supported by Django REST framework JWT).
You can then use the token (extracted with token = resp.json()['token']) to access the restricted urls as following:
requests.post(
'http://localhost:8000/some-personal-function/',
json={"foo": 42},
headers={'Authorization': 'JWT ' + token})
By the way looking at response.text might help finding the reason, which would, in the case of your 400 response, contain the following:
'{"username":["This field is required."],"password":["This field is required."]}'
I am trying to add a lead to a Zoho CRM module with Python. I keep getting:
< response>< error>< code>4600< /code>< message>Unable to process your request. Please verify if the name and value is appropriate for the "xmlData" parameter.< /message>< /error>< /response>
from the server. I have no idea if I am posting correctly or if it is a problem with our Xml Data. I am using urllib and urllib2 to format the post request.
The post request looks like this.
url = ("https://crm.zoho.com/crm/private/xml/Leads/insertRecords?authtoken="
""+str(self.authToken)+"&scope=crmapi")
params = {"xmlData":self.xml}
data = urllib.urlencode(params)
request = urllib2.Request(url = url, data =data)
request.add_header("Content-Type",'application/xml')
response = urllib2.urlopen(request)
You cannot combine HTTP GET query parameters (ones in URL) and HTTP POST parameters.
This is limitation on the HTTP protocol level, not in Python or Zoho.
Most likely you are doing it wrong. Revisit Zoho documentation how it should be done.
Here is another old library doing Zoho + CRM, written in Python. You might want to check it for inspiration: https://github.com/miohtama/mfabrik.zoho
I am using Flask (based on Werkzeug) which uses Python.
The user can download a file, I'm using the send_from_directory-function.
However when actually downloading the file, the HTTP header content-length is not set. So the user has no idea how big the file being downloaded is.
I can use os.path.getsize(FILE_LOCATION) in Python to get the file size (in bytes), but cannot find a way to set the content-length header in Flask.
Any ideas?
I needed this also, but for every requests, so here's what I did (based on the doc) :
#app.after_request
def after_request(response):
response.headers.add('Access-Control-Allow-Origin', '*')
return response
Since version 0.6 the canonical way to add headers to a response object is via the make_response method (see Flask docs).
def index():
response = make_response(render_template('index.html', foo=42))
response.headers['X-Parachutes'] = 'parachutes are cool'
return response
I believe you'd do something like this (untested):
from flask import Response
response = Response()
response.headers.add('content-length', str(os.path.getsize(FILE_LOCATION)))
See: Werkzug's Headers object and Flask's Response object.