RELEASING.md

Releasing

1. Check out a clean copy. Note that cargo publish packages up untracked files. Use --allow-dirty at your peril.
2. Fetch from the main repo
3. Ensure that the TensorFlow version is a real release, not a release candidate
4. Update changelog.
5. Bump version number of tensorflow-sys if necessary
   1. Run git log v${PREVIOUS_VERSION}..HEAD tensorflow-sys and see if there were any changes. If not, skip.
   2. Bump the version in tensorflow-sys/Cargo.toml
   3. Bump the version for tensorflow-sys in the root Cargo.toml
6. Bump the version number in Cargo.toml
7. Bump the version number in README.md
8. Run ./test-all
9. Run python examples/addition/addition.py using the version of TensorFlow being linked against. (Use pip in a virtualenv, see https://www.tensorflow.org/install/pip#2-create-a-virtual-environment-recommended)
   1. Run virtualenv --system-site-packages -p python3 ~/tensorflow-${TENSORFLOW_VERSION}
   2. Run source ~/tensorflow-${TENSORFLOW_VERSION}/bin/activate to activate the virtualenv
   3. Run pip install --upgrade pip
   4. Run pip install --upgrade tensorflow==${TENSORFLOW_VERSION}
   5. Run python examples/addition/addition.py
   6. Run deactivate to exit the virtualenv
10. Run ./run-valgrind
11. Commit and push the changes. (Push before publishing to ensure that the changes being published are up to date.)
12. If the version of tensorflow-sys was bumped, run cargo publish for tensorflow-sys. (Publish before tagging in case there are problems publishing and we need to add commits to fix them.)
13. Run cargo publish. (Publish before tagging in case there are problems publishing and we need to add commits to fix them.)
14. Add a v${VERSION} tag and push it
15. Run ./update-docs
16. Announce the release to the mailing list

Post-release

1. Update version numbers of dependencies
2. Remove any deprecated items scheduled to be removed