Let's go through the process of taking a legacy (outdated) Python application and bring it up to the latest version of Python with the assitance of GitHub Copilot.

In this workshop, you will be working with a legacy Python application that used Python version 2.5. Here are some features:

1. All dependencies are pre-installed for you, but not using the legacy application which uses a dated version of Python.
2. The application uses Sqlite3 and dated Python syntax that you can update
3. The legacy application has documentation that can be used for functional testing
4. Both unit and functional tests are provided as part of the legacy code.

Use the @workspace agent to explain what is going on with this project.

• Open GitHub Copilot Chat and prefix your prompt with @workspace
• Ask questions like how is this project installing dependencies

Try to run the tests in the Python application. Use pytest which is pre-installed to get an output. What happens?

• Ask @workspace what are some problems that this code may have as I port it to modern Python?
• Try installing the dependencies with a virtual environment

In order to port the code, you will need to copy all of the code over to the upgraded directory. This is where you will be working on the code.

• Use the functional tests to determine a slice of functionality that you will focus on

Create a virtual environment and install the project. Do python setup.py develop and try to understand what the error is.

• Use GitHub Copilot chat to paste the error and get guidance on what is going on
• Ensure to add the file as context #file:setup.py
• Fix the errors with the suggestions and complete the installation

The project has the pytest framework and command-line tool installed and available. Try running the unit tests available and explore the errors.

• Use the output of the errors to ask Copilot for help. Include it in the chat with #terminalSelection. Make sure you have selected the output.
• Review the response from Copilot and ask for clarifications if needed
• Do not edit the code yet.

Use the functional tests to determine a slice of functionality that you will focus on. In this case you will fix the try/except errors.

• Use GitHub Copilot to help you with the actual problem to fix. Mention the old Python version.
• Fix all try/except errors and run the tests continously until these are fixed.

The next slice of the problem is dealing with ConfigParser not being available. Use GitHub Copilot to provide a strategy to choose for implementing a fix.

• Ask "This project was built using Python 2.5 and with Python 3.9 I get the following:" . Include a paste of the test output
• The fix might be simple. Choose an option to implement and verify immediately.
• Run the tests to ensure that the changes are correct and that the tests pass

[!NOTE]

All tests should now pass. If they don't, you can use the test output to ask Copilot for help and revise all the previous steps in this workshop

Pytest is the most common and powerful testing framework and testing command-line tool in Python. The legacy code uses unittest which is not as powerful and flexible as pytest. Port all the tests to pytest and run them.

• On any test file, use the @workspace agent to ask how to port the tests to pytest
• Write a single test, and try using test functions instead of classes when possible
• Once you write a test, use GitHub Copilot to suggest the next test to generate

The legacy code uses distribute which you had to fix earlier. But it is still using setuptools which is now deprecated with setup.py. Use pyproject.toml to package the project.

• Ask Copilot to help you with the pyproject.toml file. Use the @workspace agent to ask how to create a pyproject.toml file. Ensure you have the setup.py file open so that it can be used as context.
• Create a new pyproject.toml file and use Copilot suggestions to fill it in.
• Use the pyproject.toml file to install the project. You can use pip install . to install the project in editable mode.

The legacy project didn't have any automation. Use GitHub Actions to automate testing and validation of the project. The goal is to have a workdlow that will run on every pull request and will run the tests and try to install the project.

• Ask GitHub Copilot with the @workspace agent how to create a GitHub Actions workflow.
• Create the necessary files and directory to create a GitHub Actions workflow.
• Push your changes when done to verify GitHub Actions is working.