Skip to content

Latest commit

 

History

History
71 lines (51 loc) · 2.7 KB

README.md

File metadata and controls

71 lines (51 loc) · 2.7 KB
Raw

pulumi-cljs

This is a small wrapper library making it easier to use ClojureScript to write Infrastructure As Code via Pulumi's NodeJS API.

It also includes a variety of utilities and patterns for working with AWS resources in Pulumi.

Environment

This library assumes the use of shadow-cljs to compile ClojureScript to JavaScript, prior to running Pulumi. In this model, Pulumi remains entirely unaware of ClojureScript and without any explicit integration - it operates only on the complied JavaScript output.

It is reccomended to read the documentation for both Pulumi and shadow-cljs to understand how to install them, and they work independently before using this tool.

A project directory using this library to deploy a Pulumi stack will typically have the following elements:

  1. A shadow-cljs configuration with shadow-cljs.edn and deps.edn.
  2. A Pulumi NodeJS configuration with Pulumi.yaml, Pulumi.<stack>.yaml, package.json, package-lock.json, etc.
  3. A shim JavaScript file. This script serves as the entry point for the Pulumi application, and invokes the compiled ClojureScript via JS->CLJS interop.

An example of such a project configuration is included in the examples/hello-world subdirectory.

Running Tests

To run tests in this project, run shadow-cljs compile test && node generated/tests.js.

Library Features

This library consists primarily of helper functions to use Pulumi's Node.js APIs fluently from ClojureScript.

Please see the docstrings in the code (or eventually, generated API) docs for descriptions and documentation on the available functions.

This code extends the ILookup protocol to Pulumi Resource objects, meaning that keyword lookups work as expected on Pulumi objects.

Other functions of note include:

  • resource - to create a Pulumi resource
  • cfg and cfg-obj - to efficiently retrieve values from the Pulumi config without needing to pass it around.
  • id - to generate unique Pulumi resource IDs that include the project name.
  • all - a macro with bindings as per let, which takes the inputs as Pulumi Outputs, and immediately returns a Pulumi Output. The value of the output is returned by the body of the macro.
  • json - Given a ClojureScript datastructure, recursively convert it to a Pulumi Output containing a JSON string, resolving any nested Outputs along the way.
  • prepare-output - Walks a Clojure data structure, converting it to a JavaScript object so it can be returned as stack outputs.

Usage

See the examples/hello-world for a simple project, or the example Wordpress Deployment for a more complex stack.