Closed nthiery closed 6 years ago
By and large putting packages on PyPI for installation by pip is simply a matter of:
setup.py
(using plain distutils or, ideally, setuptools, depending on what features the package has)setup.py
to register the package on PyPI, and build/upload a source tarball.I will deploy these all on sagemathcloud too...
On Thursday, February 11, 2016, Erik Bray notifications@github.com wrote:
By and large putting packages on PyPI for installation by pip is simply a matter of:
- Writing a correct setup.py (using plain distutils or, ideally, setuptools, depending on what features the package has)
- Use the setup.py to register the package on PyPI, and build/upload a source tarball.
— Reply to this email directly or view it on GitHub https://github.com/sagemath/docker-images/issues/10#issuecomment-182795521 .
Sent from my massive iPhone 6 plus.
Now that those kernels are package for pip and Sage, this has become trivial. See: https://github.com/nthiery/test-binder-sage
Should I just move the three relevant lines to our base docker image?
Done by @embray in 0c536f732d9a5999b81e406e612a265280d923cd.
Step 1: get in touch with the authors of those kernels, and build pip packages Step 2: update the sagemath-jupyter container to include those packages by default
With this, the sagemath-jupyter docker container (and therefore Windows SageMath distribution) would give easy access to all components.