import requests
import psycopg2
from google.auth.transport.requests import Request
from google_auth_oauthlib.flow import Flow

Fetch fitness data

Insert into PostgreSQL

This approach gives you complete control but requires more development time and ongoing maintenance.

## Setting Up Your PostgreSQL Schema

Your database schema should reflect the structure of Google Fitness API responses. Create tables that efficiently store different types of health data:

```sql
-- User activity tracking
CREATE TABLE user_activities (
    id SERIAL PRIMARY KEY,
    user_id VARCHAR(255) NOT NULL,
    activity_type VARCHAR(100),
    start_time BIGINT,
    end_time BIGINT,
    calories INTEGER,
    distance_meters INTEGER,
    duration_ms BIGINT,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

-- Heart rate data
CREATE TABLE heart_rate_readings (
    id SERIAL PRIMARY KEY,
    user_id VARCHAR(255) NOT NULL,
    heart_rate INTEGER,
    timestamp_ms BIGINT,
    accuracy INTEGER,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

-- Sleep data
CREATE TABLE sleep_sessions (
    id SERIAL PRIMARY KEY,
    user_id VARCHAR(255) NOT NULL,
    start_time BIGINT,
    end_time BIGINT,
    sleep_type VARCHAR(50), -- light, deep, rem
    duration_minutes INTEGER,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

Pro tip: Use appropriate data types for your use case. Google Fitness API often returns timestamps in milliseconds (BIGINT), distances in meters (INTEGER), and durations in milliseconds (BIGINT). Plan your schema accordingly to avoid data conversion issues later.

Handling Authentication and Security

Google Fitness API uses OAuth 2.0, which means you'll need to handle user authentication properly. Here's what you need to set up:

1. Create a Google Cloud Project and enable the Fitness API
2. Generate OAuth 2.0 credentials for your application
3. Implement the OAuth flow to get user consent
4. Store refresh tokens securely in your PostgreSQL database

Most no-code platforms handle this authentication flow automatically. They'll prompt users to authorize access to their Google Fitness data and manage token refreshing behind the scenes.

Optimizing Data Loading Performance

When working with health data, you'll often deal with large datasets. Here are strategies to keep your data loading fast and efficient:

Use batch insertions instead of individual row insertions:

INSERT INTO user_activities (user_id, activity_type, calories, timestamp_ms)
VALUES 
    ('user123', 'running', 450, 1638360000000),
    ('user123', 'walking', 120, 1638366000000),
    ('user123', 'cycling', 300, 1638372000000);

Create proper indexes on frequently queried columns:

CREATE INDEX idx_user_activities_user_id ON user_activities(user_id);
CREATE INDEX idx_user_activities_timestamp ON user_activities(timestamp_ms);
CREATE INDEX idx_heart_rate_user_timestamp ON heart_rate_readings(user_id, timestamp_ms);