Cerebro

Search through github forks of a repo for search terms! Great for maintainers that want to know what downstream forks are doing.

Install it with:

$ gem install cerebro

Cerebro currently works by using the Github API to identify all the forks of the repo if interest. It then will clone all the repos down into a specifically named repo under the $HOME/.cerebro directory. It will then use grep searching to determine if your search term is present in each/which of the forks.

• Cloning down of the repos will be shallow by default. If you want deep clones, use cerebro search --deep.

• To clean up all repos that Cerebro has cloned onto your local file system:

$ GITHUB_TOKEN=<github_api_token> cerebro search <repo_owner> <repo_name> <search_term>

For example, if I'm looking to determine how many forks of the Cloud Foundry python buildpack have been forked to add or use graphviz, I might run:

GITHUB_TOKEN=<github_token> cerebro search cloudfoundry python-buildpack graphviz
Found 85 forks of cloudfoundry/python-buildpack
All forks will be stored in /Users/user/.cerebro/python-buildpack-forks
Cloning or updating all local fork repos...
<...>
Searching through these forks now...

----------------Search Results---------------------
Found graphviz in kdunn-pivotal-python-buildpack
Found graphviz in ssssam-python-buildpack

Found "graphviz" in 2 forks out of total 85 forks of cloudfoundry/python-buildpack

Clones of forked repos are located in /Users/user/.cerebro/python-buildpack-forks

• To clean up all repos that Cerebro has cloned onto your local file system:

$ cerebro clean --all

• To clean up only the repos that are forks for a specific repo:

$ cerebro clean <repo-name>

• This gem relies on git for cloning down forks of repos.
• This gem currently relies on $HOME being set on your machine for figuring out where to store cloned repos.
• This gem's searching for forks relies on the user providing a valid Github API token via GITHUB_TOKEN env vars. This token is not persisted or stored at all. This API token does not need any scopes specified, it is only required to prevent hitting the unauthenticated Github API request limit.

Bug reports and pull requests are welcome on GitHub at https://github.com/RochesterinNYC/cerebro or in https://cloudfoundry.slack.com/messages/buildpacks/.

To install the source of this repo as a local gem, use:

gem build cerebro.gemspec
gem install --local ./cerebro-*

However, bear in mind that while in this repo, running cerebro will run exe/cerebro and the source code by default.

• Make dirnames specific to owner and repo name instead of just repo name
• Command for cleaning up stored repos
• Specify where to store cloned repos
• Option for auto-clean up afterwards
• Better UX for specifying github auth
• Remove dependency on octokit
• Threading for parallel cloning
• Search caching?

The gem is available as open source under the terms of the MIT License.