Skip to content

Latest commit

 

History

History
118 lines (102 loc) · 4.3 KB

README.md

File metadata and controls

118 lines (102 loc) · 4.3 KB

Getting started with AWS Crypto Tools Python projects

This file introduces the various tools that we use in our Python projects and how to get started developing on these projects.

Tools

  • All of our Python projects use tox to orchestrate all activity, both in CI and locally. Some common tox environments that should be in every project include:

    • tox -e autoformat : Applies all autoformatting rules.
    • tox -e lint : On most projects, this also runs autoformat.
    • tox -e py{VERSION}-local : Runs all non-integ tests using Python VERSION. (ex: py38-local)
    • tox -e py{VERSION}-integ : Runs all integ tests using Python VERSION. (ex: py38-integ) NOTE: Integ tests usually require available AWS credentials.
    • tox -e py{VERSION}-all : Runs all tests using Python VERSION. (ex: py38-all)
    • tox -r -p {THREADS} : Runs all default tox environments concurrently, THREADS at a time. Adjust THREADS as appropriate for your environment. (ex: tox -r -p 7)

  • We use pytest to run our tests.

  • We use a variety of autoformatting and static analysis tools to ensure consistency among our projects.

    • black : Applies general formatting requirements.

    • isort : Sorts import statements.

    • flake8 : Looks for code issues and inconsistencies (often partially overlaps with pylint). We also commonly use several flake8 plugins:

    • pylint : Looks for code issues and inconsistencies (often partially overlaps with flake8).

    • doc8 : Looks for issues in reStructuredText docs.

    • check-manifest : Looks for issues with MANIFEST.in file used for building source distributions.

    • bandit : Looks for common security issues in Python code.

    • vulture : Looks for unused code. NOTE: Vulture has a high false-positive rate for libraries but offers a useful perspective.

Local Development Setup

Linux/MacOS

For local development, we recommend using pyenv. If you want to quickly bootstrap a local development environment, run:

# "install" pyenv https://github.com/pyenv/pyenv#installation
git clone https://github.com/pyenv/pyenv.git ~/.pyenv
# bash
echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.zprofile
echo 'export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.zprofile
echo 'eval "$(pyenv init -)"' >> ~/.zprofile
# Install Python build dependencies
# (varies by OS)
# https://github.com/pyenv/pyenv/wiki#suggested-build-environment

# Install latest Python versions
for MINOR_VERSION in 3.6 3.7 3.8 3.9 3.10;do
  PYTHON_VERSION=$(pyenv install -l | grep "^  ${MINOR_VERSION}" | tail -1)
  pyenv install ${PYTHON_VERSION}
done

# set "local" pyenv version
#  just sets .python-version file in current directory
#  you can delete this afterwards
#
# This is the version of Python that all pipx-managed tools will use.
pyenv local $(pyenv versions --bare | grep 3.8 | tail -1)
# install pipx-in-pipx
# https://pipx-in-pipx.readthedocs.io/
pyenv exec pip install pipx-in-pipx
# remove pyenv local setting (optional)
rm .python-version
# restart shell to get all current and future pipx-managed binaries into path
exec "$SHELL"

# install tox
pipx install tox
pipx inject tox tox-pyenv