Memory in the Shell

We love the shell. bash or zsh or fish or powershell or name your own. But the shell forgets, especially when we are across multiple systems. It would be incredible if the shell could safely remember things for us, across systems. So let us extend the shell to remember things for us.

Problem

The overall system will probably have multiple components. It may also need to handle some of the nuances of each shell. But we love to keep things simple. So, let us first build a simple API that can store the typed commands.

Requirements

1. The API should be able to store the commands typed by the user.
2. The API should be able to search the command history by keyword.

For example, the following command should store a command:

curl -X POST http://localhost:8080/api/v1/commands -d "command=ls -l"

Example command to search history:

curl -X GET http://localhost:8080/api/v1/commands?keyword=ls

Prerequisites

To run this application, you’ll need:

• Docker and Docker Compose installed.
• (Optional) Node.js and npm if you want to run the app without Docker.

Installation and Setup

1. Clone the repository:

git clone https://github.com/sabhisharma-ise/memory-in-the-shell.git
cd memory-in-the-shell

2. Set up environment variables: Create a .env file in the root directory with your PostgreSQL configuration:

DB_USER=your_db_username
DB_PASSWORD=your_db_password
DB_HOST=localhost
DB_PORT=5432
DB_NAME=your_database_name

3. Database Initialization: The application uses an init.sql file to initialize the database with a commands table containing id, command, and timestamp fields.

Running the Application

Using Docker Compose

1. Build and start the application:

   docker-compose up --build

   This command will:

   • Start a PostgreSQL database and initialize it with the required table structure.
   • Start the Express server.

2. Stopping the application: To stop the application, press Ctrl+C in the terminal where it’s running, then run:

   docker-compose down

Database Schema

The commands table is structured as follows:

Column	Type	Description
id	SERIAL	Unique ID for each command entry
command	TEXT	The shell command
timestamp	TIMESTAMP	The timestamp for when the command was added