provide dynamically loading client settings by object instead of using load_by_string

Following the example, we need to make a yaml file to set client settings and use GoogleAdsClient.load_from_storage to create a new client instance, but it does not make sense in production service that provides multiple customers to fetch their ads.

My current solution is like this,

   configuration_string = '\n'.join([
        'developer_token: %s' % developer_token,
        'client_id: %s' % client_id,
        'client_secret: %s' % client_secret,
        'refresh_token: %s' % refresh_token,
        'login_customer_id: %s' % login_customer_id
    ])
    client = google.ads.google_ads.client.GoogleAdsClient.load_from_string(configuration_string)

I use GoogleAdsClient.load_from_string to make a string as yaml format, but I think it had better provide a function to load settings by dictionary(object), and it will be more easy to use.

[BenRKarl]

@yo8568 Thanks for that info, it does make sense. I’ll keep this issue open as a feature request to add a method load_from_dict that allows the client to receive a dict of credentials.

In the meantime you can sort-of achieve this by initializing the client directly. You’ll need to import the OAuth client and create a set of Credentials through it, like so (we do this here in the client):

from google.oauth2.credentials import Credentials
from google.auth.transport.requests import Request
from google.ads.google_ads.client import GoogleAdsClient
 
credentials = Credentials(
    None,
    refresh_token=refresh_token,
    client_id=client_id,
    client_secret=client_secret,
    token_uri='https://accounts.google.com/o/oauth2/token')

credentials.refresh(Request())

client = GoogleAdsClient(credentials, developer_token, login_customer_id=login_customer_id)

You can reference the initialization method here to see what other keyword parameters are available.

[BenRKarl]

#139 was merged and release in library version 3.1.0.