top of page
  • Writer's pictureBill Bialas

Demystifying Google's Application Default Credentials (ADC) for Python

Updated: Oct 19, 2023


If you're developing applications that interact with Google Cloud services using Python, you've probably come across the term "Application Default Credentials" or ADC. But what exactly are ADC, and how do they work? In this blog post, we'll demystify ADC and explain how they make authentication a breeze for Python client libraries.


What are Application Default Credentials (ADC)?

Application Default Credentials (ADC) are a feature provided by Google Cloud to simplify the authentication process when interacting with Google Cloud services. They allow your Python code to automatically obtain the necessary credentials for authenticating with various Google services without the need for you to hardcode credentials in your code.


How ADC Works for Python Client Libraries

Let's break down how ADC works step by step:


1. Environment Detection

When you use a Google client library in Python, it's smart enough to detect the environment in which your code is running. This could be:

  • Google Cloud Environments: If your code is running on a Google Cloud environment like Compute Engine or App Engine, the client library will automatically check for credentials associated with the instance.

  • Local Development: If you're developing locally, the library will look for a JSON key file (typically named `google-application-credentials.json`) in a predefined location.


2. Metadata Server (if applicable)

In Google Cloud environments, the library will first check a metadata server to fetch the appropriate credentials. These credentials are often associated with the instance through IAM roles.


3. Authentication: Simplified by the Metadata Server

One of the significant advantages of using Application Default Credentials is the seamless authentication process. When your code is running on a Google Compute Engine (GCE) instance, it can utilize the metadata server to obtain the credentials it needs to authenticate with Google Cloud services. This is typically done using the service account associated with the instance.


Here's how it works in a nutshell: Your code running on the GCE instance queries the metadata server for authentication credentials. The metadata server provides the necessary credentials, which your code can then use to access Google Cloud services. This process eliminates the need for manual credential management and greatly simplifies the development workflow.


Here's an example of how to use ADC in Python on a GCE instance using python:


from google.cloud import storage


# Create a client without specifying credentials

client = storage.Client()


# Use the client to interact with Google Cloud Storage

buckets = list(client.list_buckets())

print("Available Buckets:")

for bucket in buckets:

print(bucket.name)


4. Local Development

Service Account Key

For local development, you might have set up your Google Cloud SDK with a service account key. Here's an example of how to use a service account key JSON file explicitly using Python:


from google.oauth2 import service_account

from google.cloud import storage


# Path to your service account key JSON file

key_path = '/path/to/your-service-account-key.json'


# Load the credentials from the JSON key file

credentials = service_account.Credentials.from_service_account_file(

key_path, scopes=['https://www.googleapis.com/auth/cloud-platform']

)


# Create a client using the obtained credentials

client = storage.Client(credentials=credentials)


# Use the client to interact with Google Cloud Storage

buckets = list(client.list_buckets())

print("Available Buckets:")

for bucket in buckets:

print(bucket.name)


gcloud auth application-default

The `gcloud auth application-default login` command is a powerful tool provided by the Google Cloud SDK (Software Development Kit) that simplifies the process of setting up application default credentials for local development. Here's a step-by-step explanation of how it works:


1. Initialization: When you run `gcloud auth application-default login`, it initializes a workflow to establish application default credentials on your local machine. These credentials are used by client libraries when your Python code makes requests to Google Cloud services, such as Cloud Storage or BigQuery.


2. Web Authentication Prompt: After running the command, your terminal will prompt you to authenticate with your Google account. It opens a web browser window with a secure authentication prompt. You'll need to log in to your Google account if you're not already logged in.


3. OAuth2 Authentication: The authentication process used is OAuth2, a secure and widely adopted standard for authentication and authorization. When you log in, you grant the Google Cloud SDK the necessary permissions to access your Google account information for the purpose of generating application default credentials.


4. Token Retrieval: Once you've successfully authenticated, the Google Cloud SDK retrieves an OAuth2 access token, which is a short-lived credential that allows your local applications to interact with Google Cloud services on your behalf.


5. Local Storage: The access token is stored securely on your local machine. It's important to note that only the access token is stored locally, not your Google account credentials. This token is used to authenticate your Python applications when they make requests to Google Cloud services.


6. Scoped to Your Google Account: The access token is scoped to your Google account, which means it can access resources and services associated with your account. This ensures that any actions performed by your local applications using these credentials are authorized and logged under your account.


7. Seamless Integration: With the application default credentials set up, your Python code can now seamlessly use these credentials when interacting with Google Cloud services. You don't need to manually specify the credentials in your code; the client libraries automatically use the application default credentials when available.


8. Secure and Convenient: The `gcloud auth application-default login` command provides a secure and convenient way to authenticate your local development environment without the need to manage service account keys or other credentials manually.


In essence, `gcloud auth application-default login` bridges the gap between your local development environment and Google Cloud services, allowing you to work on your Python applications knowing that they have the necessary credentials to interact securely with Google Cloud.


5. Access Control

Remember that access to Google Cloud services is determined by the permissions associated with the credentials. Ensure that the associated service account or user account has the necessary IAM roles to perform the actions your application requires.


Using ADC in Python

In Python, using Application Default Credentials is a breeze. You can typically initiate this process without explicitly specifying the credentials by using the `google-auth` library. Here's a simple example:


import google


# Get the credentials and project ID

credentials, project_id = google.auth.default()


# You can now use 'credentials' to authenticate with Google services

# For example, using the Google Cloud Storage client library:

from google.cloud import storage


# Create a client using the obtained credentials

client = storage.Client(credentials=credentials, project=project_id)



# Use the client to interact with Google Cloud Storage

buckets = list(client.list_buckets())

print("Available Buckets:")

for bucket in buckets:

print(bucket.name)


This code snippet automatically selects the appropriate credentials based on the runtime environment, whether it's local development or a Google Cloud environment.


Conclusion

Google's Application Default Credentials (ADC) for Python client libraries streamline the authentication process for applications interacting with Google Cloud services. By automatically detecting the runtime environment and fetching the appropriate credentials, ADC allows developers to focus on building their applications rather than managing authentication tokens.


92 views0 comments

Commentaires


bottom of page