A location-based recommendation platform that helps users discover nearby activities and experiences.
- Log in to AWS console using registered account: https://d-9267bea535.awsapps.com/start
- Type in the command line aws configure sso
- Provide necessary info (client region, output format, profile name)
- Based on your given profile name (step 3), type in the command line aws s3 ls --profile eugenekim3107 (assuming profile name is "eugenekim3107")
- Copy and paste the AWS environment variables
To develop and test locally while connecting to MongoDB hosted on the EC2 instance, follow these steps:
Set up SSH port forwarding to connect to MongoDB on the EC2 instance:
- Secure the SSH key:
chmod 400 whatnext/backend/WhatNextAdminKey.pem
- Establish an SSH tunnel for MongoDB access:
ssh -i whatnext/backend/WhatNextAdminKey.pem -L 8000:localhost:27017 ubuntu@<Elastic-IP>
8000: Local port on your machine for MongoDB.localhost:27017: Remote destination (MongoDB port on EC2 instance).Elastic-IP: Replace with the actual Elastic IP address of your EC2 instance.
- Login to MongoDB (within EC2):
mongosh -u eugenekim -p whatnext -authenticationDatabase admin
Set up SSH port forwarding to connect to Redis-server on the EC2 instance:
- Secure the SSH key:
chmod 400 whatnext/backend/WhatNextAdminKey.pem
- Establish an SSH tunnel for Redis-server access:
ssh -i whatnext/backend/WhatNextAdminKey.pem -L 8001:localhost:6379 ubuntu@<Elastic-IP>
8001: Local port on your machine for Redis-server.localhost:6379: Remote destination (Redis-server port on EC2 instance).Elastic-IP: Replace with the actual Elastic IP address of your EC2 instance.
To start the FastAPI server locally:
uvicorn main:app --host 0.0.0.0 --port 8080 --reload
0.0.0.0: Binds the server to all IP addresses.8080: Specifies the port for the FastAPI server.--reload: Enables auto-reload on file changes for development.- Ensure the packages in requirements.txt are correctly installed in your environment.
To test an endpoint, use the following command:
curl -X 'GET' 'http://localhost:8080/nearby_locations'
curl -X POST "http://localhost:8080/chatgpt_response" -H "Content-Type: application/json" -d '{"user_id": "wiVOrMOJ8COqs7d6OgCBNVTV9lt2", "message": "I would like to drink some coffee", "latitude": 32.8723812680163, "longitude": -117.21242234341588}'
To continue a conversation, input the session_id into the POST:
curl -X POST "http://localhost:8080/chatgpt_response" -H "Content-Type: application/json" -d '{"user_id": "wiVOrMOJ8COqs7d6OgCBNVTV9lt2", "session_id": <session-id>, "message": "Thank you for the information!", "latitude": 32.8723812680163, "longitude": -117.21242234341588}'
For setting up your environment for production, follow these steps:
Repeat the steps for SSH port forwarding as in the local development and testing section to connect to MongoDB on the EC2 instance.
Ensure MongoDB is running on the EC2 instance. It should run automatically upon initialization. To check the status of MongoDB:
systemctl status mongod
If MongoDB is not running, initialize the database using:
sudo systemctl start mongod
sudo systemctl enable mongod
Ensure Redis-server is running on the EC2 instance. It should run automatically upon initalization. To check the status of the Redis-server:
systemctl status redis-server.service
If Redis-server is not running, initalize the database using:
sudo systemctl start redis-server.service
sudo systemctl enable redis-server.service
Ensure the Docker container is running on the EC2 instance. It should run automatically upon initialization. To check the status of the Docker container:
docker ps
If the Docker conatiner is not running, initialize the database using:
docker run -d --restart=always --name whatnext-container -p 8000:8000 --env OPENAI_API_KEY=$OPENAI_API_KEY
whatnext-container: The name of the container.whatnext-image: The name of the image.
To test an endpoint in production, use the following command:
curl -k "https://api.whatnext.live/nearby_locations"
curl -k -X POST "https://api.whatnext.live/chatgpt_response" -H "Content-Type: application/json" -d '{"user_id": "wiVOrMOJ8COqs7d6OgCBNVTV9lt2", "message": "I want to go hiking. Can you give me some suggestions?", "latitude": 32.8723812680163, "longitude": -117.21242234341588}'
Official Website: whatnext.live
To ensure clarity and maintain a structured approach to API interactions, all API endpoints adhere to a specific path convention:
Endpoint Prefix: All API endpoints begin with api.whatnext.live.
Examples of API Endpoint Structures:
- GET Requests: To retrieve data from our platform, you'll use the GET method with endpoints structured as follows:
curl -k "https://api.whatnext.live/[endpoint-name]"
- POST Requests: For submitting data to our platform, POST requests follow a similar structure:
curl -k -X POST "https://api.whatnext.live/[endpoint-name]" -H "Content-Type: application/json" -d '{[field-content]}'
By adhering to these guidelines, we ensure that API requests are efficiently processed to the back-end services, facilitating a reliable platform for all users.