I have read this thoroughly: https://developers.google.com/google-apps/documents-list/#using_google_apps_administrative_access_to_impersonate_other_domain_users
I have googled this to death.
So far I have been able to:
Authorise with:
clientLogin
OAuth tokens (using my domain key)
retrieve document feeds for all users in the domain (authorised either way in #1)
I am using the "entry" from the feed to Export/Download documents and always get forbidden for other users for documents not shared with admin. The feed query I am using is like:
https://docs.google.com/feeds/userid#mydomain.com/private/full/?v=3
(I have tried with and without the ?v=3)
I have also tried adding the xoauth_requestor_id (which I have also seen in posts as xoauth_requestor), both on the uri, and as a client property: client.xoauth_requestor_id = ...
Code fragments:
Client Login (using administrator credentials):
client.http_client.debug = cfg.get('HTTPDEBUG')
client.ClientLogin( cfg.get('ADMINUSER'), cfg.get('ADMINPASS'), 'HOSTED' )
OAuth:
client.http_client.debug = cfg.get('HTTPDEBUG')
client.SetOAuthInputParameters( gdata.auth.OAuthSignatureMethod.HMAC_SHA1, cfg.get('DOMAIN'), cfg.get('APPS.SECRET') )
oatip = gdata.auth.OAuthInputParams( gdata.auth.OAuthSignatureMethod.HMAC_SHA1, cfg.get('DOMAIN'), cfg.get('APPS.SECRET') )
oat = gdata.auth.OAuthToken( scopes = cfg.get('APPS.%s.SCOPES' % section), oauth_input_params = oatip )
oat.set_token_string( cfg.get('APPS.%s.TOKEN' % section) )
client.current_token = oat
Once the feed is retrieved:
# pathname eg whatever.doc
client.Export(entry, pathname)
# have also tried
client.Export(entry, pathname, extra_params = { 'v': 3 } )
# and tried
client.Export(entry, pathname, extra_params = { 'v': 3, 'xoauth_requestor_id': 'admin#mydomain.com' } )
Any suggestions, or pointers as to what I am missing here?
Thanks
You were very close to having a correct implementation. In your example above, you had:
client.Export(entry, pathname, extra_params = { 'v': 3, 'xoauth_requestor_id': 'admin#mydomain.com' } )
xoauth_requestor_id must be set to the user you're impersonating. Also what you need is to use 2-Legged OAuth 1.0a with the xoauth_requestor_id set either in the token or in the client.
import gdata.docs.client
import gdata.gauth
import tempfile
# Replace with values from your Google Apps domain admin console
CONSUMER_KEY = ''
CONSUMER_SECRET = ''
# Set this to the user you're impersonating, NOT the admin user
username = 'userid#mydomain.com'
destination = tempfile.mkstemp()
token = gdata.gauth.TwoLeggedOAuthHmacToken(
consumer_key, consumer_secret, username)
# Setting xoauth_requestor_id in the DocsClient constructor is not required
# because we set it in the token above, but I'm showing it here in case your
# token is constructed via some other mechanism and you need another way to
# set xoauth_requestor_id.
client = gdata.docs.client.DocsClient(
auth_token=token, xoauth_requestor_id=username)
# Replace this with the resource your application needs
resource = client.GetAllResources()[0]
client.DownloadResource(resource, path)
print 'Downloaded %s to %s' % (resource.title.text, destination)
Here is the reference in the source code to the TwoLeggedOAuthHmacToken class:
http://code.google.com/p/gdata-python-client/source/browse/src/gdata/gauth.py#1062
And here are the references in the source code that provide the xoauth_requestor_id constructor parameter (read these in order):
http://code.google.com/p/gdata-python-client/source/browse/src/atom/client.py#42
http://code.google.com/p/gdata-python-client/source/browse/src/atom/client.py#179
http://code.google.com/p/gdata-python-client/source/browse/src/gdata/client.py#136
Related
Need to obtain the list of all tokens in organization.
Used the token to make a call to https://vssps.dev.azure.com/{organization}/_apis/tokens/pats?api-version=6.1-preview.1
My permission in DevOps are set as the Collection Administrator.
Received response was:
{“$id”:“1”,“innerException”:null,“message”:“The requested operation is not allowed.”,“typeName”:“Microsoft.TeamFoundation.Framework.Server.InvalidAccessException, Microsoft.TeamFoundation.Framework.Server”,“typeKey”:“InvalidAccessException”,“errorCode”:0,“eventId”:3000}
Is there some lack of permissions or do I need to set up something else to get list of tokens in organization?
You don't mention how you get your token, and criteria for authentication flow but I will share my adventure that started similarly yours.
I got your exact error while following this guide: https://learn.microsoft.com/en-gb/azure/devops/organizations/accounts/manage-personal-access-tokens-via-api?view=azure-devops
The token I got from that python code just didn't work.
Then I found this code instead: https://learn.microsoft.com/en-us/azure/databricks/dev-tools/api/latest/aad/app-aad-token#--username-password-flow-programmatic
While using the same app registration from the link above, I copied my scope and tenantID from the dysfunctional code into this new code, and then go to your app registration --> authentication --> Allow public client flows to yes, see screenshot.
I ran the script after giving the credentials and now the token worked.
Dumping the code for future reference:
# Given the client ID and tenant ID for an app registered in Azure,
# along with an Azure username and password,
# provide an Azure AD access token and a refresh token.
# If the caller is not already signed in to Azure, the caller's
# web browser will prompt the caller to sign in first.
# pip install msal
from msal import PublicClientApplication
import sys
# You can hard-code the registered app's client ID and tenant ID here,
# along with the Azure username and password,
# or you can provide them as command-line arguments to this script.
client_id = '<client-id>'
tenant_id = '<tenant-id>'
username = '<username>'
password = '<password>'
# Do not modify this variable. It represents the programmatic ID for
# Azure Databricks along with the default scope of '/.default'.
scope = [ '2ff814a6-3304-4ab8-85cb-cd0e6f879c1d/.default' ]
# Check for too few or too many command-line arguments.
if (len(sys.argv) > 1) and (len(sys.argv) != 5):
print("Usage: get-tokens-for-user.py <client ID> <tenant ID> <username> <password>")
exit(1)
# If the registered app's client ID and tenant ID along with the
# Azure username and password are provided as command-line variables,
# set them here.
if len(sys.argv) > 1:
client_id = sys.argv[1]
tenant_id = sys.argv[2]
username = sys.argv[3]
password = sys.argv[4]
app = PublicClientApplication(
client_id = client_id,
authority = "https://login.microsoftonline.com/" + tenant_id
)
acquire_tokens_result = app.acquire_token_by_username_password(
username = username,
password = password,
scopes = scope
)
if 'error' in acquire_tokens_result:
print("Error: " + acquire_tokens_result['error'])
print("Description: " + acquire_tokens_result['error_description'])
else:
print("Access token:\n")
print(acquire_tokens_result['access_token'])
print("\nRefresh token:\n")
print(acquire_tokens_result['refresh_token'])
I want to implement Google's One tap sign-up and automatic sign-in in my website with the help of documentation https://developers.google.com/identity/one-tap/web/ but I am getting confused on how to implement in python.
def smartlock(request):
try:
CLIENT_ID='*******'
csrf_token_cookie = self.request.cookies.get('g_csrf_token')
if not csrf_token_cookie:
webapp2.abort(400, 'No CSRF token in Cookie.')
csrf_token_body = self.request.get('g_csrf_token')
if not csrf_token_body:
webapp2.abort(400, 'No CSRF token in post body.')
if csrf_token_cookie != csrf_token_body:
webapp2.abort(400, 'Failed to verify double submit cookie.')
# Specify the CLIENT_ID of the app that accesses the backend:
idinfo = id_token.verify_oauth2_token(csrf_token_cookie, requests.Request(), CLIENT_ID)
# Or, if multiple clients access the backend server:
# idinfo = id_token.verify_oauth2_token(token, requests.Request())
# if idinfo['aud'] not in [CLIENT_ID_1, CLIENT_ID_2, CLIENT_ID_3]:
# raise ValueError('Could not verify audience.')
if idinfo['iss'] not in ['accounts.google.com', 'https://accounts.google.com']:
raise ValueError('Wrong issuer.')
# If auth request is from a G Suite domain:
# if idinfo['hd'] != GSUITE_DOMAIN_NAME:
# raise ValueError('Wrong hosted domain.')
# ID token is valid. Get the user's Google Account ID from the decoded token.
userid = idinfo['sub']
except ValueError:
# Invalid token
pass
'''
As mentioned in the 'Key Point' section of this page: The ID token is returned in the credential field, instead of the g_csrf_token field.
So, you need to get the idinfo with the code as below:
credential = self.request.get('credential')
idinfo = id_token.verify_oauth2_token(credential, requests.Request(), CLIENT_ID)
The g_csrf_token parameter is for different purpose. It makes sure the request was submitted from a page in your own domain, so as to prevent the cross-site-request-forge attacks.
I'm building an app using Dailymotion API
https://developer.dailymotion.com/
with official python sdk
https://github.com/dailymotion/dailymotion-sdk-python
and writing video CRUD(Create, Read, Update, Delete) simply.
Create, Read, Delete were succeccfully completed, but encountering a weird response of the API about "Update".
Here is a simplified snippet of my code in my Django project,
def update(request, video_id):
user = request.user
video = get_object_or_404(Video, pk=video_id)
file_path = # define file_path from uploaded file object
input_title = # define input_title from post request
input_description = # define input_description from post request
d = get_dailymotion_d(user)
if d == 'revoked':
# do actions of logout and delete the user
try:
# get url for upload with the file_path on my server
url = d.upload(file_path)
# update
response = d.post('/video/' + video.dailymotion_video_id, {'url': url, 'title': input_title, 'description': input_description, 'published': 'true', 'channel': 'creation'})
# delete the video from my sever
video.file_field.delete()
return redirect('/videos')
except Exception as e:
print(e.args)
print('update failed..!')
# delete the video from my server
video.file_field.delete()
return redirect('/videos')
def get_dailymotion_d(user):
d = dailymotion.Dailymotion()
d.set_grant_type('token', api_key=settings.DAILYMOTION_API_KEY, api_secret=settings.DAILYMOTION_API_SECRET, scope=['email', 'userinfo', 'manage_videos'], info={'redirect_uri': settings.DAILYMOTION_REDIRECT_URI})
# get credentiaols from database
access_token = user.dailymotionuser.access_token
expires = user.dailymotionuser.expires
refresh_token = user.dailymotionuser.refresh_token
session_params = {'access_token': access_token, 'expires': expires, 'refresh_token': refresh_token}
# set the credentials
d._session_store.set(session_params)
# check if the user revoked or not
try:
force_refreshed_access_token = d.get_access_token(force_refresh=True)
except dailymotion.DailymotionAuthError as e:
print(e.args[0])
return 'revoked'
# get valid access token
valid_access_token = d.get_access_token()
# update database with the valid access token
DailymotionUser.objects.filter(user=user).update(access_token=valid_access_token, expires=expires, refresh_token=refresh_token)
# prepare dic of the valid access token
valid_access_token_dic = {'access_token': valid_access_token}
# set the valid access token
d._session_store.set(valid_access_token_dic)
return d
but the update fails except title field with the following message,
'access_forbidden: You are not allowed to change existing video source.'
from doc,
access_forbidden: Thrown when the user doesn't have the permission to access the data (e.g. missing a required scope to access certain fields).
but I'm sure that the permission has manage_videos scope which is an enough scope to update the existing video source because the doc says,
manage_videos: Allows to modify or delete the user's uploaded videos and to publish new ones.
and described above,
Only the title field of the video is updated with the input_title properly.
Thanks for reading, I researched their docs carefully, but still don't understand this response.
Only partner users can update the video source urls.
Best,
Google is deprecating the OpenID endpoint I was using (v1.0 I think, via the django_openid_auth module) and I need to update my app and migrate my users' accounts to use Google OAuth2.
I've changed the app to use python-social-auth and have it authenticating with social.backends.google.GoogleOAuth2 successfully.
I've written a pipeline function to find associated OpenID urls from the old table and this is working for the other backends I care about but Google:
def associate_legacy_user(backend, response, uid=None, user=None,
*args, **kwargs):
if uid and not user:
# Try to associate accounts registered in the old openid table
identity_url = None
if backend.name == 'google-oauth2':
# TODO: this isn't working
identity_url = response.get('open_id')
else:
# for all other backends, see if there is a claimed_id url
# matching the identity_url use identity_url instead of uid
# as uid may be the user's email or username
try:
identity_url = response.identity_url
except AttributeError:
identity_url = uid
if identity_url:
# raw sql as this is no longer an installed app
user_ids = sql_query.dbquery('SELECT user_id '
'FROM django_openid_auth_useropenid '
'WHERE claimed_id = %s',
(identity_url,))
if len(user_ids) == 1:
return {'user': User.objects.get(id=user_ids[0]['user_id'])}
As best I can tell from reading Google's migration guide, I need to add an openid.realm to the request, which I've done as follows in settings.py:
SOCIAL_AUTH_GOOGLE_OAUTH2_AUTH_EXTRA_ARGUMENTS \
= {'openid.realm': 'http://example.com/'}
But this doesn't seem to be returning the open_id value in the response passed into my pipeline function.
I seem to be stuck on Step 3:
I tried sub-classing the backend to change the RESPONSE_TYPE to add id_token but that returned an empty response:
import social.backends.google
class CustomGoogleOAuth2(social.backends.google.GoogleOAuth2):
RESPONSE_TYPE = 'code id_token'
I tried building an additional request to https://www.googleapis.com/oauth2/v3/token similar to this example, but I don't really know how to go about putting that together and debugging it.
Some more details:
My old claimed_ids for Google OpenID users look like: https://www.google.com/accounts/o8/id?id=AItOawmAW18QuHDdn6PZzaiI5BWUb84mZzNB9eo
I'm happy to use social.backends.google.GoogleOpenIdConnect or a similar alternative backend if that's an easier solution. And while it seems to be closer to what the Google docs are talking about, I wasn't able to get it to work when I tried:
I get a 400 Error: invalid_request Parameter not allowed for this message type: nonce
I can get past the nonce error using social.backends.google.GoogleOpenIdConnect by adding id_token to the RESPONSE_TYPE but then I get an AuthMissingParameter error in my /complete/google-openidconnect/ endpoint as the request's GET and POST are empty. (Tried 'code id_token', 'token id_token', 'id_token', ...)
I don't want to use social.backends.google.GooglePlusAuth as that doesn't integrate as nicely with my current login form.
Worst case, I should be able to use social.pipeline.social_auth.associate_by_email, but I only have email addresses for maybe 80% of the users so that leaves quite a few who will have a new account and need support to associate it manually.
Try as I might, I can't find any examples of people doing a similar migration with python-social-auth, but it must be happening to lots of people.
Any ideas?
Solution works for python social auth 0.1.26
In new versions (0.2.*) of python social auth, there is GoogleOpenIdConnect, but it does not work fine (at least I did not succeed). And my project has some legacy, so I can't use new version of social.
I wrote custom GoogleOpenIdConnect backend:
import datetime
from calendar import timegm
from jwt import InvalidTokenError, decode as jwt_decode
from social.backends.google import GoogleOAuth2
from social.exceptions import AuthTokenError
class GoogleOpenIdConnect(GoogleOAuth2):
name = 'google-openidconnect'
ACCESS_TOKEN_URL = 'https://www.googleapis.com/oauth2/v3/token'
DEFAULT_SCOPE = ['openid']
EXTRA_DATA = ['id_token', 'refresh_token', ('sub', 'id')]
ID_TOKEN_ISSUER = "accounts.google.com"
def user_data(self, access_token, *args, **kwargs):
return self.get_json(
'https://www.googleapis.com/plus/v1/people/me/openIdConnect',
params={'access_token': access_token, 'alt': 'json'}
)
def get_user_id(self, details, response):
return response['sub']
def request_access_token(self, *args, **kwargs):
"""
Retrieve the access token. Also, validate the id_token and
store it (temporarily).
"""
response = self.get_json(*args, **kwargs)
response['id_token_parsed'] = self.validate_and_return_id_token(response['id_token'])
return response
def validate_and_return_id_token(self, id_token):
"""
Validates the id_token according to the steps at
http://openid.net/specs/openid-connect-core-1_0.html#IDTokenValidation.
"""
try:
id_token = jwt_decode(id_token, verify=False)
except InvalidTokenError as err:
raise AuthTokenError(self, err)
# Verify the token was issued in the last 10 minutes
utc_timestamp = timegm(datetime.datetime.utcnow().utctimetuple())
if id_token['iat'] < (utc_timestamp - 600):
raise AuthTokenError(self, 'Incorrect id_token: iat')
return id_token
Notes:
get_user_id – An identifier for the user, unique among all Google accounts and never reused.
request_access_token – there is I add id_token_parsed to response, and it will be used in pipeline.
validate_and_return_id_token – validate of jwt is disabled, because in google developers console I have registered Client ID as web application so, I have no certificates for validate this data.
Then I created pipelines:
def social_user_google_backwards(strategy, uid, *args, **kwargs):
"""
Provide find user that was connect with google openID, but is logging with google oauth2
"""
result = social_user(strategy, uid, *args, **kwargs)
provider = strategy.backend.name
user = result.get('user')
if provider != 'google-openidconnect' or user is not None:
return result
openid_id = kwargs.get('response', {}).get('id_token_parsed', {}).get('openid_id')
if openid_id is None:
return result
social = _get_google_openid(strategy, openid_id)
if social is not None:
result.update({
'user': social.user,
'is_new': social.user is None,
'google_openid_social': social
})
return result
def _get_google_openid(strategy, openid_id):
social = strategy.storage.user.get_social_auth('openid', openid_id)
if social:
return social
return None
def associate_user(strategy, uid, user=None, social=None, *args, **kwargs):
result = social_associate_user(strategy, uid, user, social, *args, **kwargs)
google_openid_social = kwargs.pop('google_openid_social', None)
if google_openid_social is not None:
google_openid_social.delete()
return result
And changed my SOCIAL_AUTH_PIPELINE and AUTHENTICATION_BACKENDS settings:
AUTHENTICATION_BACKENDS = (
...
#'social.backends.open_id.OpenIdAuth' remove it
'social_extension.backends.google.GoogleOpenIdConnect', # add it
...
)
and
SOCIAL_AUTH_PIPELINE = (
'social.pipeline.social_auth.social_details',
'social.pipeline.social_auth.social_uid',
'social.pipeline.social_auth.auth_allowed',
# 'social.pipeline.social_auth.social_user', remove it
'social_extension.pipeline.social_user_google_backwards', # add it
'social.pipeline.user.get_username',
...
# 'social.pipeline.social_auth.associate_user', remove it
'social_extension.pipeline.associate_user', # add it
'social.pipeline.social_auth.load_extra_data',
...
)
Is there a way to get content owner information (attribution) for particular video via YouTube Partner API?
For example, this video: http://www.youtube.com/watch?v=I67cgXr6L6o&feature=c4-overview&list=UUGnjeahCJW1AF34HBmQTJ-Q is attributed to VEVO.
Is there a way to get that info somehow via API?
It's possbible now:
https://developers.google.com/youtube/partner/docs/v1/ownership/
[EDIT]
I've used this sample to create the snipped below and succesfully got the assets content owner.
You will need setup your python enviromment and the clients_secrets.file correctly
#!/usr/bin/python
import httplib2
import os
import sys
import logging
import requests
import json
import time
from apiclient.discovery import build
from oauth2client.file import Storage
from oauth2client.client import flow_from_clientsecrets
from oauth2client.tools import argparser, run_flow
from symbol import for_stmt
# The CLIENT_SECRETS_FILE variable specifies the name of a file that contains
# the OAuth 2.0 information for this application, including its client_id and
# client_secret. You can acquire an OAuth 2.0 client ID and client secret from
# the Google Developers Console at
# https://console.developers.google.com/.
# Please ensure that you have enabled the YouTube Data API for your project.
# For more information about using OAuth2 to access the YouTube Data API, see:
# https://developers.google.com/youtube/v3/guides/authentication
# For more information about the client_secrets.json file format, see:
# https://developers.google.com/api-client-library/python/guide/aaa_client_secrets
CLIENT_SECRETS_FILE = "client_secrets.json"
# This variable defines a message to display if the CLIENT_SECRETS_FILE is
# missing.
MISSING_CLIENT_SECRETS_MESSAGE = """
WARNING: Please configure OAuth 2.0
To make this sample run you will need to populate the client_secrets.json file
found at:
%s
with information from the Developers Console
https://console.developers.google.com/
For more information about the client_secrets.json file format, please visit:
https://developers.google.com/api-client-library/python/guide/aaa_client_secrets
""" % os.path.abspath(os.path.join(os.path.dirname(__file__), CLIENT_SECRETS_FILE))
YOUTUBE_SCOPES = (
# This OAuth 2.0 access scope allows for read-only access to the authenticated
# user's account, but not other types of account access.
"https://www.googleapis.com/auth/youtube.readonly",
# This OAuth 2.0 scope grants access to YouTube Content ID API functionality.
"https://www.googleapis.com/auth/youtubepartner")
YOUTUBE_API_SERVICE_NAME = "youtube"
YOUTUBE_API_VERSION = "v3"
YOUTUBE_PARTNER_API_SERVICE_NAME = "youtubePartner"
YOUTUBE_PARTNER_API_VERSION = "v1"
KEY = "USE YOUR KEY"
# Authorize the request and store authorization credentials.
def get_authenticated_services(args):
flow = flow_from_clientsecrets(CLIENT_SECRETS_FILE, scope=" ".join(YOUTUBE_SCOPES), message=MISSING_CLIENT_SECRETS_MESSAGE)
storage = Storage("%s-oauth2.json" % sys.argv[0])
credentials = storage.get()
if credentials is None or credentials.invalid:
credentials = run_flow(flow, storage, args)
youtube = build(YOUTUBE_API_SERVICE_NAME, YOUTUBE_API_VERSION, http=credentials.authorize(httplib2.Http()))
youtube_partner = build(YOUTUBE_PARTNER_API_SERVICE_NAME, YOUTUBE_PARTNER_API_VERSION, http=credentials.authorize(httplib2.Http()))
return (youtube, youtube_partner)
def get_content_owner_id(youtube_partner):
content_owners_list_response = youtube_partner.contentOwners().list(fetchMine=True).execute()
return content_owners_list_response["items"][0]["id"]
#def
def claim_search(youtube_partner, content_owner_id, _videoId):
video_info = get_video_info(_videoId)
print '\n---------- CLAIM SEARCH LIST - VIDEO (',video_info['title'],') ID: (',_videoId,')'
claims = youtube_partner.claimSearch().list( videoId=_videoId, onBehalfOfContentOwner=content_owner_id).execute()
print ' --- CLAIMS: ',claims
if claims['pageInfo']['totalResults'] < 1:
print " --- DOESN'T HAVE CLAIMS"
return
assetId = claims['items'][0]['assetId']
print ' --- ASSET ID: ', assetId
ownershipHistory = youtube_partner.ownershipHistory().list(assetId=assetId, onBehalfOfContentOwner=content_owner_id).execute()
print ' --- OWNERSHIP HISTORY: ', ownershipHistory
contentOwnerId = ownershipHistory['items'][0]['origination']['owner']
print ' --- CONTENT OWNER ID: ', contentOwnerId
contentOwners = youtube_partner.contentOwners().get(contentOwnerId=contentOwnerId, onBehalfOfContentOwner=content_owner_id).execute()
print ' --- CONTENT OWNERS: ', contentOwners
def get_video_info(videoId):
r = requests.get("https://www.googleapis.com/youtube/v3/videos?id="+videoId+"&part=id,snippet,contentDetails,status,statistics,topicDetails,recordingDetails&key="+KEY)
content = r.content
content = json.loads(content)
return content['items'][0]['snippet']
if __name__ == "__main__":
args = argparser.parse_args()
(youtube, youtube_partner) = get_authenticated_services(args)
#video id's. Ex: "https://www.youtube.com/watch?v=lgSLz5FeXUg"
videos = ['I67cgXr6L6o', 'lgSLz5FeXUg']
content_owner_id = get_content_owner_id(youtube_partner)
for vid in videos :
claim_search(youtube_partner, content_owner_id, vid)
time.sleep(0.5)
The (censored) proof :
There is no way to check other people's ownership or management rights on a video or channel. YouTube deliberately avoids that.
While ownershipHistory works, I saw GET https://www.googleapis.com/youtube/partner/v1/assets?id=[multiple asset ids] also worked when it's called with fetchOwnership=effective option
https://developers.google.com/youtube/partner/docs/v1/assets/list