github-matchmaker

A project to help you pick a github project to contribute to. Also, you can contribute to this one :)

Proposed as a VMware Borathon project by team who-has-free-time.

Public Demo Site

Description

Sometimes we have got a bit of free time, and would love to help fix issues or add features to an existing project. If we could identify a high visibility issue fit our preferences, then we would be more likely to submit patches that have a big impact for an existing project.

Those preferences may be something like:

• issue could be completed in our allotted time frame
• issue matches our skill level
• project is in a specific language or technology area
• other criteria....

A loose direction

• Create a stateless UI to

  • gather preferences,
  • query a set of query-generator services with those preferences
  • Assemble responses from those services into a github query
  • run the github query
  • display the results
  • possibly allow rating of results to be sent back to query-generators to influence their algorithms

• Create a set of query-generator services that

  • accept some seed input
  • map the input to a github query term the service finds relevent
  • return the github search query term

• ideas for query-generators:

  • language. seed=perl might yield query=lang:perl
  • repository visibility. seed=low might yeild query=comments>10
  • level of pain of the issue. seed=high might yield query=comments>30 +1 in:comments
  • available time commitment, or difficulty of the fix. seed=low might yield query="not hard" in:comments or easy in:comments

Outcomes

• Prevent project sprawl by helping to ID code repositories doing things you are interested in.
• give best bang for your contribution
• enable low skill coders to contribute to projects that may not be tagged in a way to facilitate contributions
• allow experimentation in a language you would like to learn
• allow scratching the code itch without starting a whole new project

Build and run:

docker-compose build
docker-compose up

curl http://localhost:4200

Testing

Tested on a mac, but real requirement is docker-compose.

# git clone this repo

git clone https://github.com/tompscanlan/github-matchmaker
cd github-matchmaker/integ-test/
./test.sh

Deploy

cd digital-ocean export DO_TOKEN=00000000000000000000 terraform apply
-var "do_token=${DO_TOKEN}"
-var "pub_key=$HOME/.ssh/id_rsa.pub"
-var "pvt_key=$HOME/.ssh/id_rsa"

License

Copyright VMware