This serves as a skeleton for setting up new projects.
To get started, follow these steps:
- Fork this repository to your project on github.
- Invite your team members and us as collaborators in the settings tab
- clone the repository to your machine:
git clone [email protected]:<USERNAME>/swp-teamproject - Look through the repository and try to understand its contents
- remove what you don't need, adapt what needs to be changed in your project (e.g. author, project and package names)
We recommend to setup a clean development environment for this project. This
has several benefits such as preventing version conflicts with packages
required by other projects and to make it easier to keep track of
dependencies (type pip list to get a list of installed packages).
A good choice for managing environments is to use conda (specifically, we recommend miniconda). After installation, you can create and activate python environments as follows:
conda create -n myproject python=3.8 conda activate myproject
Other noteworthy tools for managing environments are
- virtualenvwrapper (more low-level, can only create environments for installed python versions)
- and Pipenv a more functional approach to package dependency management, specifically for applications, but with its own set of drawbacks
Once you've activated the environment, install the current package and its dependencies with:
pip install -e .[dev]
After the package and all dependencies are installed, you can execute the code
that's contained in the teamproject/__main__.py by typing:
python -m teamproject
If you have executed the pip install line above, you can also type for
short:
teamproject
The following is a short overview of how a python project could be structured:
. ├── README.rst project front page ├── setup.py packaging script ├── setup.cfg package metadata and tool config ├── MANIFEST.in lists data files to be included ├── .gitignore lists files to be ignored by git │ ├── teamproject │ ├── __init__.py toplevel package variables │ ├── __main__.py invoked on `python -m teamproject` │ ├── crawler.py web-crawler / queries │ ├── gui.py defines the gui code │ └── models.py ML code for predictions │ └── tests ├── test_crawler.py tests grouped by functionality ├── .... └── test_models.py
We recommend to use at least flake8 for basic syntax and style checks, and twine for checking the created distributions.
Unit and integration tests are in the test/ subdirectory and can be
executed via pytest.
It is possible to automatically run certain actions such as tests or publish releases when pushing to github. This is called continuous integration or continuous deployment (CI/CD). Popular CI services are for example:
- GitHub Actions for linux/mac/windows testing
- travis-ci.com for linux/mac testing
- appveyor.com for windows testing
- coveralls.io for coverage reports
- readthedocs.org for documentation