Open a Git Bash terminal.
A folder with a pyproject.toml is recognised as a project
uv init my_project --python 3.10
cd my_project
Take a look at the project
ls -a
cat pyproject.toml
Install the Python version in .python-version
uv venv
If a different Python version is required
uv venv --python 3.13
Activate the virtual environment
source .venv/scripts/activate
Check that main.py can be executed
uv run main.py
Make source code directory and move main.py
mkdir -p src/my_project
mv main.py src/my_project/main.py
Add the build system to pyproject.toml
echo '' >> pyproject.toml
echo '[build-system]' >> pyproject.toml
echo 'requires = ["hatchling"]' >> pyproject.toml
echo 'build-backend = "hatchling.build"' >> pyproject.toml
echo '' >> pyproject.toml
echo '[tool.hatch.build.targets.wheel]' >> pyproject.toml
echo 'packages = ["src/my_project"]' >> pyproject.toml
UV will automatically install dependencies and add them to pyproject.toml
uv add plotly scikit-learn "pandas[computation,excel,output-formatting,performance,plot]"
Remove dependencies like so
uv remove matplotlib
Add optional developer dependencies
uv add --dev pytest
```bash
Check that pyproject.toml has been updated
```bash
cat pyproject.toml
Check if the dependencies were installed in venv
uv pip list
UV already created a .git directory; to re-initialise
git init
Check that you're on the master branch with no commits yet and all files untracked
git status
Stage all files
git add .
Make initial commit
uv sync
git commit -m "initial commit"
Go to https://github.com/stabilefrisur?tab=repositories to add a new GitHub repository my_project
Establish the remote connection
git remote add origin https://github.com/stabilefrisur/my_project.git
Push to GitHub
git branch -M master
git push -u origin master
Remove private files from staging
git rm -r --cached private_file
Add private folder and file names to .gitignore, such as docs/ *.log todo.md
Create a new branch
git checkout -b new_feature
Stage all code changes
git add .
Commit code changes
git commit -m "Add new feature"
Push the new branch to GitHub
git push -u origin new_feature
Start a pull request:
Merge the pull request into master:
Delete the new_feature branch locally and remotely:
git checkout master
git pull origin master
git branch -d new_feature
git push origin --delete new_feature
Register an entry point to the package
echo '[project.scripts]' >> pyproject.toml
echo 'my_project = "my_project.main:main"' >> pyproject.toml
Install the package in editable mode
uv pip install -e .
The local package will now be listed in pip
uv pip list
Run the main program which is now registered under my_project
uv run my_project
Make tests directory for pytest
mkdir tests
Add the tests path to pyproject.toml
echo '' >> pyproject.toml
echo '[tool.pytest.ini_options]' >> pyproject.toml
echo 'pythonpath = "src"' >> pyproject.toml
echo 'testpaths = "tests"' >> pyproject.toml
Run pytest
pytest
Exclude private folders from PyPi package
echo '' >> pyproject.toml
echo '[tool.hatch.build]' >> pyproject.toml
echo 'exclude = ["docs", "output"]' >> pyproject.toml
Update the project environment
uv sync
Delete the dist directory
Build Python packages into source distributions and wheels
uv build
Commit changes and push to GitHub before publishing
git add .
git commit -m "MESSAGE"
git push -u origin master
Upload distributions to an index; input PyPi credentials
uv publish
Verify that the package is published on PyPi: https://pypi.org/project/my_project/
Create a test environment
python -m venv test_env
source test_env/scripts/activate
Install the package and check that it's listed
pip install my_project
pip list
pip show my_project
Import the package and run its entry point in a Python shell
python
>>> import my_project
>>> my_project.main()
Alternatively, run the main program as a module
python -m my_project.main
If the main module is registered in pyproject.toml, just run
my_project