Releasing nb2plots

• Review the open list of nb2plots issues. Check whether there are outstanding issues that can be closed, and whether there are any issues that should delay the release. Label them.

• Review and update the release notes. Review and update the Changelog file. Get a partial list of contributors with something like:

  git log 0.2.0.. | grep '^Author' | cut -d' ' -f 2- | sort | uniq
  

  where 0.2.0 was the last release tag name.

  Then manually go over git shortlog 0.2.0.. to make sure the release notes are as complete as possible and that every contributor was recognized.

• Use the opportunity to update the .mailmap file if there are any duplicate authors listed from git shortlog -ns.

• Check the copyright years in doc/conf.py and LICENSE;

• Check the output of:

  rst2html.py README.rst > ~/tmp/readme.html
  

  because this will be the output used by PyPi

• Check nb2plots travis-ci.

• Once everything looks good, you are ready to upload the source release to PyPi. See setuptools intro. Make sure you have a file \$HOME/.pypirc, of form:

  [distutils]
  index-servers =
      pypi
  
  [pypi]
  repository: https://upload.pypi.io/legacy/
  username:your.pypi.username
  password:your-password
  

• Tag the release. This will also set the version (we are using versioneer to manage versions via git tags):

  git tag -s 0.3
  

• Clean:

  make distclean
  # Check no files outside version control that you want to keep
  git status
  # Nuke
  git clean -fxd
  

• When ready:

  python setup.py sdist --formats=zip
  # -s flag to sign the release
  twine upload -r warehouse -s dist/nb2plots*zip
  

• Upload the release commit and tag to github:

  git push
  git push --tags
  

• Push the docs to github pages with:

  cd doc
  make github