This document describes how to create packages for various platforms from your Python code.
# Build the desired image
make docker-cli
make docker-rest
# Or
ORG=my-docker-registry.something.org make docker-cli
# Upload to your registry
# (requires that you first tag the image appropriately for the Docker registry)
docker push $DOCKER_TAG
# Run locally, the Dockerfile is designed to take arguments as a CLI,
# change the Dockerfile as needed for your use-case
docker run someorg/pytemplate-cli version
From a Windows computer:
# First read windows/README.md for instructions on how to install dependencies
# Create a single file .exe
# Or use any other spec
pyinstaller windows/onefile-qt.spec
# Create an installer
python windows/release.py
From a MacOS computer:
# Install depdencies, first time only
./macos/homebrew_deps.sh
# Create an app bundle packaged in a DMG for MacOS
python3 macos/release.py
# Upload your package to PyPi so that anybody can install using pip.
#
# NOTE: This means the entire public internet. Do not enable the pypi
# option in tools/fork.py for private/proprietary Python packages.
#
# Requires `twine` to be installed, and a local twine config with your
# pypi username and password
make pypi
See appimage/README.md for instructions. Also see appimage/release.py -h.
# Remember to edit rpm.spec to include the correct information and dependencies
make rpm
# Remember to edit debian/control to include the correct information
# and dependencies
make deb