Update troubleshooting docs for Python (#488)

Closes https://github.com/google-github-actions/auth/pull/487

---------

Co-authored-by: Abhi Srivastava <bits.abhi@gmail.com>

README.md

@@ -322,7 +322,6 @@ regardless of the authentication mechanism.
     "token_format" is "id_token".
 
 
-
 <a id="setup"></a>
 ## Setup
 

docs/EXAMPLES.md

@@ -187,6 +187,69 @@ jobs:
       run: |-
         curl https://myapp-uvehjacqzq.a.run.app \
           --header "Authorization: Bearer ${{ steps.auth.outputs.id_token }}"
+
+    # Example of using ID token in Python code
+    - id: 'python-example'
+      run: |-
+        python -c "
+        import os
+        import requests
+
+        # ID token is available as environment variable
+        id_token = os.environ.get('GOOGLE_ID_TOKEN', '${{ steps.auth.outputs.id_token }}')
+
+        # Use the token to invoke a Cloud Run service
+        response = requests.get(
+            'https://myapp-uvehjacqzq.a.run.app',
+            headers={'Authorization': f'Bearer {id_token}'}
+        )
+        print(response.text)
+        "
+```
+
+### Using Default Credentials with Scopes in Python
+
+When using Workload Identity Federation with Python libraries, you may need to
+add scopes before refreshing credentials:
+
+```yaml
+jobs:
+  job_id:
+    permissions:
+      contents: 'read'
+      id-token: 'write'
+
+    steps:
+    - uses: 'actions/checkout@v4'
+
+    - id: 'auth'
+      uses: 'google-github-actions/auth@v2'
+      with:
+        workload_identity_provider: 'projects/123456789/locations/global/workloadIdentityPools/my-pool/providers/my-provider'
+        service_account: 'my-service-account@my-project.iam.gserviceaccount.com'
+
+    - id: 'python-auth'
+      run: |-
+        python -c "
+        from google.auth import default
+        from google.auth.transport.requests import Request
+
+        # Get default credentials
+        credentials, project = default()
+
+        # Add scopes before refreshing for impersonation
+        credentials = credentials.with_scopes(
+            ['https://www.googleapis.com/auth/cloud-platform']
+        )
+
+        # Refresh to get the token
+        credentials.refresh(request=Request())
+
+        # Now you can use the credentials
+        print(f'Access token: {credentials.token}')
+        if hasattr(credentials, 'id_token'):
+            print(f'ID token: {credentials.id_token}')
+        "
 ```
 
 [github-markdown-toc]: https://github.blog/changelog/2021-04-13-table-of-contents-support-in-markdown-files/

docs/TROUBLESHOOTING.md

@@ -230,6 +230,53 @@ tool like `jq`:
 cat credentials.json | jq -r tostring
 ```
 
+<a name="cannot-refresh"></a>
+
+## Cannot refresh credentials to retrieve an ID token
+
+If you get an error like:
+
+```text
+google.auth.exceptions.RefreshError: ('Unable to acquire impersonated credentials', '{"error": {"code": 400, "message": "Request contains an invalid argument.", "status": "INVALID_ARGUMENT"}}')
+```
+
+when trying to refresh credentials in Python code to get an ID token, this is
+usually because the credentials are missing required scopes. The Google Auth
+library requires scopes to be set when refreshing credentials for impersonation.
+
+To fix this issue, add the required scopes before refreshing:
+
+```python
+from google.auth import default
+from google.auth.transport.requests import Request
+
+credentials, project = default()
+
+# Add scopes before refreshing
+credentials = credentials.with_scopes(
+    ["https://www.googleapis.com/auth/cloud-platform"]
+)
+credentials.refresh(request=Request())
+
+# Now you can access the ID token
+print(credentials.id_token)
+```
+
+Alternatively, you can use the `token_format` parameter of this action to
+generate an ID token directly:
+
+```yaml
+- uses: 'google-github-actions/auth@v2'
+  with:
+    workload_identity_provider: ${{ secrets.WIF_PROVIDER }}
+    service_account: ${{ secrets.WIF_SERVICE_ACCOUNT }}
+    token_format: 'id_token'
+    id_token_audience: 'https://example.com'
+```
+
+This will export the ID token as an environment variable that you can use in
+your Python code.
+
 ## Organizational Policy Constraints
 
 > **ℹ️ NOTE!** Your Google Cloud organization administrator controls these