RAMLocker

Construct a mock Express HTTP server based on a spec set out in RAML v0.8 files.

Quickstart

How to use on your project

To install ramlocker on your project, we recommend using NPM

npm install --save-dev ramlocker

You can then require the file into your file using your package manager of choice and start the server using the function this returns. On large projects, it can take a while to parse the RAML and construct the endpoints so this is an asynchronous process, that exposes a promise you can use to capture the server if, say, you want to shut it down.

let app;
const ramLocker = require('ramlocker');

ramLocker($PORT, $PATH_TO_RAML).then(server => {
    app = server;
});

How to develop on this project

Check out the code. The first time you do this you will need to install the dependencies

npm install

We currently have some smoke tests to ensure we don't break the fundamental HTTP requests. More tests are definitely on the todo list! You can run these using

gulp test

We'll use a fairly standard pull request model if you want to submit back to the project

API

ramlocker(port, pathToRaml) : Promise(Server)

The global export is a function that takes a port to run the server on, and a path to your raml directory. It will search at this level for RAML documentation.

It returns a promise that will resovle to an http.Server object.

TODOs

Functionality

• HTTP methods that are not GET or POST
• Custom HTTP status codes
• Mock response based on schema (using fakerJS) if no example is presented
• Wildcard content-types (application/*)
• Validation of request bodies against schema (for non-GET requests)
• Validate URL parameters based on data types
• Allow any RAML that fits v0.8 schema, not just those which are defined according to the motivating project schema

Technically

• More granular e2e tests + unit tests
• Linter
• Run tests on check-in
• Move RAML parsing to a stream-based API to prevent passing all routes across in one promise