- Make sure you have twine installed:
pip install twine
- Make sure you have added your PyPI credentials to
~/.pypirc - Make sure you have anaconda-client installed:
conda install anaconda-client
- Log into Anaconda Cloud
anaconda login
- Make sure the version in package.json updated.
- Navigate to the correct directory:
cd igv-jupyter
- Upload the files by running:
python setup.py sdist bdist_wheel; twine upload -r pypitest dist/*.tar.gz; twine upload -r pypitest dist/*.whl
- If the upload fails go to https://testpypi.python.org/pypi and manually upload dist/nbtools-*.tar.gz.
- Test the deploy by uninstalling and reinstalling the package:
sudo pip uninstall igv-jupyter; sudo pip install -i https://testpypi.python.org/pypi igv-jupyter
- First deploy to test and ensure everything is working correctly (see above).
- Navigate to the correct directory:
cd igv-jupyter
- Upload the files by running:
python setup.py sdist bdist_wheel; twine upload dist/*.tar.gz; twine upload dist/*.whl
- If the upload fails go to https://pypi.python.org/pypi and manually upload dist/nbtools-*.tar.gz.
- Test the deploy by uninstalling and reinstalling the package:
sudo pip uninstall igv-jupyter; sudo pip install igv-jupyter
- Deploy to Production PyPi
- Navigate to Anaconda directory
cd /anaconda3
- Activate a clean environment.
conda activate clean
- Run the following, removing the existing directory if necessary:
conda skeleton pypi igv-jupyter --version XXX
- Build the package:
conda build igv-jupyter
- Converting this package to builds for other operating systems can be done as shown below. You will need to upload each built version using a separate upload command.
conda convert --platform all /anaconda3/conda-bld/osx-64/igv-jupyter-XXX-py37_0.tar.bz2 -o conda-bld/
- Upload the newly built package:
anaconda upload /anaconda3/conda-bld/*/igv-jupyter-XXX-py37_0.tar.bz2 -u igvteam
- Log into the Anaconda website to make sure everything is good.