About this code

This repository contains the source code for the Visual Studio Code extension "codehelper". This extension serves as a starting point and proof of concept for how one can recreate a GitHub Copilot-like extension using the open-source model deepseek-coder:1.5b.

Purpose

The primary goal of this extension is to demonstrate the capabilities of the deepseek-coder:1.5b model in providing intelligent code suggestions and assistance within the Visual Studio Code environment. It aims to replicate some of the functionalities of GitHub Copilot using an open-source approach.

Back to top

What a user can do and test using this repo

1. Install and Set Up: Follow the installation steps to set up the development environment.
2. Run and Debug: Use Visual Studio Code to run and debug the extension in a new VS Code window.
3. Test Functionality: Test the various features provided by the extension.
4. Contribute: Make improvements or add new features to the extension and contribute back to the repository.

Back to top

Installation steps

Step 1: Install Docker
Step 2: Install Ollama
Step 3: Pull deepseek-coder:1.3b
Step 4: Clone this repo and set up
Step 5: Make sure your Ollama is running and deepseek-coder is operational. Run your VSC in debug mode to test
Step 6: Package and deploy as needed

Back to top

Detailed installation: Installing Docker and Ollama on Windows, Linux, and Mac

Windows Installation

1. Download Docker Desktop for Windows from the official site:
   • Docker Desktop for Windows
2. Run the installer and follow the setup instructions.
3. After installation, start Docker Desktop.
4. Verify installation by running:

   docker --version

Back to top

Linux Installation

1. Update package lists:

   sudo apt update

2. Install required packages:

   sudo apt install -y ca-certificates curl gnupg

3. Add Docker’s official GPG key:

   sudo install -m 0755 -d /etc/apt/keyrings
   curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo tee /etc/apt/keyrings/docker.asc > /dev/null
   sudo chmod a+r /etc/apt/keyrings/docker.asc

4. Set up the repository:

   echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

5. Install Docker Engine:

   sudo apt update
   sudo apt install -y docker-ce docker-ce-cli containerd.io

6. Verify installation:

   docker --version

7. Docker Installation for Linux - Official Docs

Back to top

Mac Installation

1. Download Docker Desktop for Mac:
   • Docker Desktop for Mac
2. Open the downloaded .dmg file and follow the installation instructions.
3. Start Docker Desktop.
4. Verify installation by running:

   docker --version

Back to top

Installing Ollama on Docker

1. Pull the Ollama Docker image:

   docker pull ollama/ollama

2. Run the Ollama container:

   docker run -d --name ollama -p 11434:11434 ollama/ollama

3. Verify Ollama is running:

   curl http://localhost:11434

Back to top

Pulling and Testing deepseek-code:1.5b

1. Pull the deepseek-code:1.5b model:

   ollama pull deepseek-code:1.5b

2. Run a test command to check if the model is working:

   ollama run deepseek-code:1.5b "def hello_world(): print('Hello, world!')"

Back to top

Cloning and Setting Up This Repository

1. Clone this repository:

   git clone https://github.com/yourusername/codehelper.git
   cd codehelper

2. Install the necessary dependencies:

   npm install

3. Open the project in Visual Studio Code:

   code .

4. Make sure your Ollama is running and deepseek-coder is operational.
5. Run Visual Studio Code in debug mode to test the extension:
   • Press F5 to start debugging.

Back to top

Testing the Extension

1. Open the project in Visual Studio Code:

   code .

2. Press F5 to start debugging. This will open a new VS Code window with the extension loaded.

3. In the new window, open the command palette (Ctrl+Shift+P or Cmd+Shift+P on Mac) and run the command codehelper.codehelper-ext to open the CodeHelper Chat panel.

4. In the CodeHelper Chat panel, enter a prompt in the text area and click the "Send" button.

5. The extension will use the deepseek-coder:1.3b model to generate a response, which will be displayed in the panel. Screenshots below on the current experience.

   Invoking the extension using the command prompt -

   Extension chat page for your testing -

6. Check the output and debug console for any errors or logs to ensure the extension is working as expected.

7. Extend and contribute back your changes and enhancements.

Back to top

Additional Resources

• Docker Official Installation Guide
• Ollama Documentation

Back to top