Getting Started

These three links summarize the setup process. For full instructions refer to these.

1. https://help.github.com/articles/using-jekyll-as-a-static-site-generator-with-github-pages/
2. https://help.github.com/articles/about-github-pages-and-jekyll/
3. https://help.github.com/articles/setting-up-your-github-pages-site-locally-with-jekyll/

For quick instructions to get setup, do the following.

1. Clone repository with git clone [email protected]:CoDataScience/codatascience.github.io.git --recursive
2. Run git submodule init
3. Run git submodule update
4. Ensure you have Ruby 2.0 or higher installed with ruby --version
5. Run gem install bundler
6. Run bundle install (from the repository directory)
7. To run a local development version of the site run bundle exec jekyll serve -w

Any changes committed to master will trigger a rebuild and deploy of the website. Confirmation of the commit and build status are posted in the #github channel on slack.

Updating Notebooks

All the notebooks are obtained from git submodules that are in _notebooks/repositories. By convention any repository in the data science team puts notebooks intended to be published on the website in <ROOT>/notebooks and no other files.

Steps

1. Update the git submodules of interest by running git pull in the submodule directory then use git add and git commit as you would add any other change to git.
2. cd _notebooks then run ./generate.py
3. This results in new html markdown files that need to be committed as with any other file. This is necessary to make github pages work.

Additional Info

Periodically we should run bundle update github-pages to update to the same versions that Github Pages uses