IRCTC Mini System (Backend Intern Assignment)

Simplified IRCTC-style backend built with Django + DRF, using:

• MySQL for transactional data (users, trains, bookings)
• MongoDB for API request logs + analytics
• JWT authentication (SimpleJWT)

This repository implements the exact APIs required in the assignment.

Requirements Checklist

• POST /api/register/ → create user, return JWT tokens
• POST /api/login/ → login by email+password, return JWT tokens
• JWT required for all other endpoints (Authorization: Bearer <token>)
• GET /api/trains/search/ → search trains (+ filters date, limit, offset)
• Train search logs to MongoDB with endpoint, params, user_id, execution_time_ms
• POST /api/trains/ → admin-only train create/update (upsert by train_number)
• POST /api/bookings/ → atomic seat booking, validates availability, deducts seats
• GET /api/bookings/my/ → returns user bookings with train details
• GET /api/analytics/top-routes/ → top 5 searched routes from Mongo logs

Quick Start (Windows)

1) Create venv + install dependencies

python -m venv .venv
\.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt

2) Configure environment

Copy .env.example → .env and fill in values:

• MYSQL_* (database credentials)
• MONGO_URI (MongoDB connection string)

If MYSQL_NAME is empty, the project falls back to SQLite for quick smoke tests.

3) Start MongoDB

If MongoDB is installed as a Windows service (PowerShell as Admin):

net start MongoDB

Confirm it’s listening:

netstat -ano | findstr 27017

4) Create MySQL DB + user (recommended)

If you get MySQL error 1045 Access denied, create a dedicated user:

CREATE DATABASE IF NOT EXISTS irctc_db CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;

CREATE USER IF NOT EXISTS 'irctc_user'@'localhost' IDENTIFIED BY 'your_strong_password';
GRANT ALL PRIVILEGES ON irctc_db.* TO 'irctc_user'@'localhost';
FLUSH PRIVILEGES;

Then set in .env:

MYSQL_NAME=irctc_db
MYSQL_USER=irctc_user
MYSQL_PASSWORD=your_strong_password

5) Run migrations

python manage.py makemigrations
python manage.py migrate

6) Create admin user (needed for /api/trains/)

python manage.py createsuperuser

7) Run the server

python manage.py runserver

Note: / intentionally returns 404. Use /admin/ or /api/*.

Auth

All endpoints (except register/login) require:

Authorization: Bearer <access_token>

API Usage Examples

Register

curl -X POST http://127.0.0.1:8000/api/register/ \
  -H "Content-Type: application/json" \
  -d "{\"name\":\"Aman\",\"email\":\"aman@example.com\",\"password\":\"Secret123!\"}"

Login

curl -X POST http://127.0.0.1:8000/api/login/ \
  -H "Content-Type: application/json" \
  -d "{\"email\":\"aman@example.com\",\"password\":\"Secret123!\"}"

Admin: Create/Update Train

curl -X POST http://127.0.0.1:8000/api/trains/ \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <ADMIN_ACCESS>" \
  -d "{\"train_number\":\"12951\",\"name\":\"Rajdhani\",\"source\":\"DEL\",\"destination\":\"BOM\",\"travel_date\":\"2026-02-25\",\"departure_time\":\"09:00:00\",\"arrival_time\":\"21:00:00\",\"total_seats\":200,\"available_seats\":200,\"is_active\":true}"

Search Trains (logs to MongoDB)

curl "http://127.0.0.1:8000/api/trains/search/?source=DEL&destination=BOM&date=2026-02-25&limit=10&offset=0" \
  -H "Authorization: Bearer <ACCESS>"

Book Seats

curl -X POST http://127.0.0.1:8000/api/bookings/ \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <ACCESS>" \
  -d "{\"train_id\":1,\"seats\":2}"

My Bookings

curl http://127.0.0.1:8000/api/bookings/my/ \
  -H "Authorization: Bearer <ACCESS>"

Analytics: Top Routes

curl http://127.0.0.1:8000/api/analytics/top-routes/ \
  -H "Authorization: Bearer <ACCESS>"

MongoDB Logging Details

Each train-search request inserts a document shaped like:

{
  "ts": "2026-02-24T12:00:00Z",
  "endpoint": "/api/trains/search/",
  "method": "GET",
  "user_id": 1,
  "params": {"source": ["DEL"], "destination": ["BOM"], "limit": ["10"], "offset": ["0"]},
  "execution_time_ms": 12,
  "source": "DEL",
  "destination": "BOM"
}

Optional sample logs are included in mongo_log_samples.json.

Implementation Notes

• Booking is protected against overselling using a DB transaction + row lock (select_for_update()).
• MongoDB logging is best-effort (never breaks train search). Analytics requires MongoDB running.