webhook-devourer

Webhook Devourer

This is a simple and customisable HTTP(S) server ready to take action upon receiving any kind of webhook call it understands.

How does it work?

When a weebhook call is made, it's body is parsed by a specific handler mapped to the URL the request was sent to. Currently there are two kinds of handlers supported:

• Bitbucket
• DockerHub

Each handler can have assigned one or more so called runners. Runner is a piece of code doing the job you want as a reaction to webhook calls. Just before being executed it gets a context, provided by handler and containing some of the significant values extracted from the webhook's body. Currently there is only one runner supported:

• Bash

All this logic can be secured by using SSL (encryption) and/or API keys (to prevent unauthorized access).

How to use it?

Start by looking at included example of a CONFIG file. Define handlers and map them to unique urls. Apart from type, you can specify the auth flag, which turns on the api key verification, as well as log flag, which prints the webhook's body on the console.

Next, add as many runners to each handler as you want. Their parameters can contain placeholders like $REPO$, which are replaced with values from context passed by the handler on each webhook call.

In order to use encryption, provide your key and certificate files in the tls section.

Add api keys to the apikeys section. Only request with specified apikey request parameter will be allowed for handlers with auth flag set to true.

How to build it?

Webhook Devourer was developed within a dockerized environment. Hava a look at the WERCKER file for dev and build pipelines.

License

See the LICENSE file for license rights and limitations (MIT).