GitHub - CageEye/opencv-python-with-ffmpeg: Automated CI toolchain to produce precompiled opencv-python, opencv-python-headless, opencv-contrib-python and opencv-contrib-python-headless packages.

go to find_version.py and edit line 65

    f.write('opencv_version = "{}"\n'.format("4.6.0"))

(bump version you want to publish)

then

docker build --build-arg REPO_PWD=***** . -t opencv-build

docker run -it opencv-build  bash

Inside the container run

pip wheel . --verbose

This will produce whl file that you need to copy to your host machine via docker cp and publish with twine

Name	Name	Last commit message	Last commit date
Latest commit History 848 Commits
.github	.github
cv2	cv2
docker	docker
multibuild @ b89bb90	multibuild @ b89bb90
opencv @ dad2633	opencv @ dad2633
opencv_contrib @ 49e8f12	opencv_contrib @ 49e8f12
opencv_extra @ 37294e3	opencv_extra @ 37294e3
patches	patches
scripts	scripts
tests	tests
.gitignore	.gitignore
.gitmodules	.gitmodules
CONTRIBUTING.md	CONTRIBUTING.md
Dockerfile	Dockerfile
LICENSE-3RD-PARTY.txt	LICENSE-3RD-PARTY.txt
LICENSE.txt	LICENSE.txt
MANIFEST.in	MANIFEST.in
README.md	README.md
find_version.py	find_version.py
patch_auditwheel_whitelist.py	patch_auditwheel_whitelist.py
pyproject.toml	pyproject.toml
setup.py	setup.py
travis_config.sh	travis_config.sh
travis_multibuild_customize.sh	travis_multibuild_customize.sh
travis_osx_brew_cache.sh	travis_osx_brew_cache.sh

Provide feedback