.codegen - Automatic test generation for your projects

Never write a test from scratch again. Automatically generate tests for any file you open in your codebase.

Keep your team up to date with the latest best practices and conventions you adopt. Customize the templates to fit your team's needs.

We've built this tool internally to speed up writing tests for our monolith at June. The main idea is that across your codebase, you can have a set of instructions that are used to generate tests. These templates are configurable so no matter what framework or language your current file is in, you can generate tests that fit your team's needs.

Now we're open-sourcing it so you can use it too.

1. Install our CLI via Homebrew:

brew tap ferrucc-io/dotcodegen-tap
brew install dotcodegen

Or via RubyGems, this requires Ruby 3.3.0:

gem install dotcodegen

2. Initialise the .codegen directory in your codebase:

codegen --init

3. Configure the templates to fit your team's needs. See the configuration section for more details.

4. Run the codegen command in your terminal:

codegen path/to/the/file/you/want/to/test --openai_key <your_openai_key>

5. That's it! You're ready to start generating tests for your codebase.

Extra:

This code becomes very powerful when you integrate it with your editor, so you can open a file and generate tests with a single command.

In order to do that you can add a task to your tasks.json file in the .vscode directory of your project:

{
  "version": "2.0.0",
  "tasks": [
    {
      "label": "Generate test for the current file",
      "type": "shell",
      // Alternatively you can specify the OPENAI_KEY environment variable in your .env file
      "command": "codegen ${relativeFile} --openai_key your_openai_key",
      "group": "test",
      "problemMatcher": [],
      "options": {
          "cwd": "${workspaceFolder}"
      },
    }
  ]
}

We're currently building a VSCode extension to be able to generate tests directly from your editor. Stay tuned!

CLI takes an optional openai_org_id parameter. Used to specify the organisation to which the user making the API request belongs, especially in cases where the user is part of multiple organisations. This can be important for billing and access control, ensuring that the API usage is attributed to the correct organisation.

codegen path/to/the/file/you/want/to/test --openai_key <your_openai_key>  --openai_org_id <your_openai_organisation_id>

When testing .rb code, then Rubocop or StandardRB will automatically be applied to the generated test code, if these gems are in your GemFile. For Rubocop, a .rubocop.yml needs to be on the top-level also. If both Rubocop and StandardRB are correctly set up, only StandardRB runs against the generated test code.

The extension uses AI to generate tests for any file you open in your codebase. It uses a set of customizable templates to generate the tests. You can customize the templates to fit your team's needs.

• Easy to learn: Get started in minutes.
• AI powered scaffolding: Generate smart tests for any file you open in your codebase.
• Fully customisable: Customize the templates to fit your team's needs.
• Bring your own API key: Use your own OpenAI API key to generate the tests.
• 🚧 Integrates with VSCode: Use the extension to generate tests directly from your editor.

If you want to add some default templates for your language, feel free to open a PR. See the config/default directory for the ones we have already.

Bug reports and pull requests are welcome on GitHub at https://github.com/ferrucc-io/dotcodegen.

After checking out the repo, run bin/setup to install dependencies. Then, run rake spec to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and the created tag, and push the .gem file to rubygems.org.

The gem is available as open source under the terms of the MIT License.