Magnolia Mock Server

This CLI tool allows you to quickly provide a copy of your CMS website content via a Magnolia-like API.

npx magnolia-mock-server ./content
# Server running at http://localhost:5000

• About
• Installation
• Usage
• Troubleshooting

About

Running Magnolia can be painful for front-end developers. It requires setting up infrastructure and tools we are not always familiar with so when they go wrong - which they do - that can leave people stuck and frustrated. And things are not always easy even when the CMS is up and running because Magnolia consumes a very large amount of resources which are vastly disproportionate to the requirements that front-end developers have; fetching a few hundred kilobytes of data.

For this reason many devs will configure their apps to fetch data from a remote instance of Magnolia instead of a local one but this can also lead to problems because - as a CMS - the content is always changing!

This small tool allows you to quickly serve a static copy of your JCR website content via a Magnolia-like API so you don't need to run Magnolia or depend upon a moving target.

Installation

This is a Node.js module available through the npm registry. Node 14 or above is required.

Installation is done using the npm install command:

$ npm install -D magnolia-mock-server

Usage

Command line

You can use the tool to serve a directory of YAML files containing your content using npx (which is installed with npm):

npx magnolia-mock-server --port 5000 ./content

Once running you can access the API via one of its endpoints.

Node.js

You can also use the tool programmatically by importing it into your JavaScript code:

const bootstrap = require("magnolia-mock-server");

try {
  await bootstrap(options);
} catch (err) {
  console.error(err);
}

Options

When using this tool via the CLI or in your Node.js modules it can be configured with a number of options. Most of the avilable options are analogous to Magnolia's delivery API configuration.

| Option | Type | Description | |----------------|----------|------------------------------------------------------------------| | port | number | Port number to run the mock server on. (default: 5000) | | sourceDir | string | Path to the directory containing your YAML files. | | rootPath | string | The root path of this endpoint. (default: "pages") | | depth | number | Specifies how deep the node tree will be traversed. (default: 0) | | nodeTypes | string[] | A comma-separated list of allowed node types for depth-0 nodes. | | childNodeTypes | string[] | A comma-separated list of allowed node types for child nodes. |

Endpoints

• Query nodes: http://localhost:5000/.rest/delivery/pages/ (not implemented)
• Read node: http://localhost:5000/.rest/delivery/pages/{path}
• Get children: http://localhost:5000/.rest/delivery/pages/{path}/@nodes

Project setup

1. Clone the Git repository and change to the new directory that has been created:

   git clone [email protected]:i-like-robots/magnolia-mock-server
   cd magnolia-mock-server

2. Install all of the project dependencies using npm:

   npm install

3. Copy your website data files (usually named website.{namespace}.yaml) into the content/ folder:

   cp ../magnolia-cms/website.example-site.yaml content/

4. You can now start and visit the running application:

   npm start

   Then open http://localhost:5000/.rest/delivery/pages/ in your browser.

Troubleshooting

Why can't I access the query endpoint?

This is not currently implemented, you can only fetch nodes and children of nodes.

Does this app support JCR data saved as XML?

No, the app only supports JCR data saved in YAML format.

Can I configure the node depth and node types to return?

Yes, like the Magnolia delivery endpoint configuration you can filter my node type and tree depth.