tgsmith61591 / bear

Streamline python project creation, including setup scripts, CI/CD pipelines & documentation

🍇 bear

Create more. Innovate more easily. Bear fruit more quickly.

Bear helps to structurally automate creating a python package and its corresponding documentation and CI/CD pipelines so a developer can jump right into developing the next big thing.

What is it?

Often times a developer is struck with an idea and then ends up spending valuable time that could be spent prototyping just putting together the structure of a project (including doc, CI/CD pipelines, etc.). Bear streamlines that to allow you to move straight into the code-writing.

After bear is installed, you can start creating new project skeletons right away. There are two options for creating projects with bear:

• "create"
• "yaml"

"Create"

The fastest way to use bear is from the CLI with create mode:

$ python -m bear create --project_name my_project

With the create command, you can get as complicated as you like:

$ python -m bear create \
  --path . \
  --project_name new_project \
  --git_user newBearUser42 \
  --python ">=3.5" \
  --requirements "numpy,scipy,scikit-learn" \
  --version 1.0.0 \
  --verbose

"Yaml"

You can do the same thing with a YAML file. To create the exact same package as above with a YAML file:

config.txt
path: .
project_name: new_project
git_user: newBearUser42
python: ">=3.5"
requirements: numpy,scipy,scikit-learn
version: "1.0.0"
verbose: true

...and running the package:

$ python -m bear yaml --file config.yml

Note that any option on the command line is available via YAML and goes by the same key.

Including CI tools

You can make your life easier by adding auto-configured CI tools and configuration files to your package. As of v1.0.9, the following CI/CD tools are available and can be added with the respective options:

• Travis CI (--travis)

  Auto-includes build_tools/travis helper scripts for testing and deploying on Linux and Mac OS, as well as a .travis.yml file.

• Circle CI (--circle)

  Auto-includes a .circleci/config.yml file as well as a test runner in build_tools/circle/build_test_pypy.sh for testing PyPy distributions of Python.