DV360 Audience Manager Integration

This project provides a stateless FastAPI microservice and a web frontend to ingest audience list data into Display & Video 360 (DV360) using the Data Manager API.

Features

• Stateless Backend: Acts as a pass-through to Google's Data Manager API.
• Dynamic Payload Construction: Supports Partner and Advertiser level destinations.
• CSV Parsing: Handles client-side CSV parsing.
• Secure Handling: No sensitive data is logged; simple pass-through.

Prerequisites

• Python 3.9+
• pip (Python package installer)

Installation

1. Clone the Repository and navigate to this directory.
2. Install Dependencies:

   pip install fastapi uvicorn requests pydantic

Running Locally

1. Start the FastAPI Backend

Initialize your microservice to accept ingress queries:

python3 -m uvicorn main:app --reload --port 8000

(The server will run on http://localhost:8000/)

2. Launch the Web Frontend

Navigating to the served root address will load the interactive elements instead of file-caching! Open your web browser directly to:

http://localhost:8000/

How to Test

1. Acquire Access Tone:
   • Go to Google OAuth 2.0 Playground.
   • Enter scope: https://www.googleapis.com/auth/datamanager.
   • Click "Authorize APIs" and exchange the authentication authorization code.
2. Input Details on the Frontend: Complete the fields (Bearer Token, Partner ID, Advertiser ID, Audience ID (Destination), CSV upload).
3. Push Data: Click "Push to DV360".

Checking Request Status

You can trace specific upload events using this automated helper script:

python3 check_status.py <requestId>

(Optionally include the access token as a subsequent argument if yours expired).

deploy.sh

Usage

1. Defaults: The script comes with predefined default values for SERVICE_NAME, PROJECT_ID, BUCKET_NAME, and REGION.
2. Interactive Configuration: When run, it prompts you to confirm or change these deployment configuration values. You can hit Enter to accept the defaults for each prompt.
3. GCS Bucket Creation: It attempts to create the specified GCS bucket (gsutil mb) in your chosen project and region. It will silently succeed if the bucket already exists.
4. Cloud Run Deployment: It then deploys the service to Google Cloud Run using the gcloud run deploy command.
   • It deploys the source code from the current directory (.).
   • Uses a specified SERVICE_ACCOUNT for deployment.
   • Sets a timeout of 3600 seconds, memory to 4Gi, and cpu to 2.
   • Deploys without allowing unauthenticated access (--no-allow-unauthenticated).
   • Sets environment variables PROJECT_ID, GCS_BUCKET, and ENV for your application to use at runtime.

How to Run

Simply execute the script in your terminal:

bash deploy.sh