A program to handle web hooks: It listens on a http port for incoming web requests (launched by these so called "web hooks") and performs actions depending on what is specificed in its configuration file.

Clients can connect to the NATS server and register jobs; by publishing a job.register event with the following information:

{
  "name": "my job",                     // The job's name
  "path": "/some/path/{with}/{param}",  // The webhook's path (URL)
  "methods": ["GET", "POST", "PUT"],    // The allowed methods
  "reponse_headers": {                  // Optional response headers
    "x-my-super-header": "some value"
  },
  "statusCode": 203                     // Optional response status code
}

juk will register this job and when a request is made to it's path, it will be processed and any client listening will be notified.

Example use case:

Let's say that each time something is "pushed" to your GitHub repository you want to do something, for example:

Store information about the change
Run some tests
Push changes to another branch
Etc...

You just need to "register" a job, and implement the behaviour.

An example program is provided (using the Go programming language).

Usage

juk works with a (json) configuration file, named juk.json.

You can set the JUK_CONFIG environment variable to override this behaviour.

This configuration file sets:

The HTTP server address
The HTTP server port
Path of the certificate and key (Optional, only if you want to use HTTPS)
The NATS server host and port.

You can start juk by invoking

juk [optional config filepath]

Credits

Thanks to @dciccale for the logo (that I managed to loose) and ideas.

julien/juk

Usage

Credits