Rust Cloudflare Axum Worker

A Rust worker example for Cloudflare Workers using Axum.

---

This repo is a very basic example of running an Axum-based Rust worker on Cloudflare along with Vite integration tests and working live reloading during development. I put this together to help others get started quickly.

Github and Gitlab CI configs are also included. These run linters, tests and allow manual deployments to Cloudflare as a release or preview.

I will update with more features when I get time. This will include real unit tests & implementations of Cloudflare features like D1, KV etc.

Features

• Rust WASM worker
• Working live reload make dev-watch
• Wrangler config/vars in /wrangler
• Makefile support. Requires make, run to view available commands.
• Biome linting/formatting for TS integration tests.
• Working Vitest integration testing
• Github & Gitlab CI configs
• Local Github actions testing via Act
• Lint-staged checks (linting/formatting) via Husky
• Template rendering via Askama

Get Started

• Ensure Rust, Make & Node >= 18 is installed.
• Clone.
• Create wrangler/.dev.vars (cp wrangler/.dev.vars.sample wrangler/.dev.vars) and add your Cloudflare API Token & account ID.
• Run make deps to install node dependencies.
• Run make help to view available commands.

Running make dev should start a local version of the worker on port 8787

http://localhost:8787 - Shows an example message produced via Axum HTML. http://localhost:8787/api - Shows an example JSON response.

CI/CD

For both Github and Gitlab you will need to add the environment variables CLOUDFLARE_API_TOKEN & CLOUDFLARE_ACCOUNT_ID with their respective values.

For Gitlab you will also need to set the path to the .gitlab-ci.yaml file. This is because I like to keep my file structure as clean as possible and avoid config hell.

You can set the path to the path to this file under your repo's 'Settings' -> 'CI/CD' -> 'General Pipelines' -> 'CI/CD configuration file'. Set this to .gitlab/ci-pipeline.yml

In order to actually deploy to Cloudflare via Github or Gitlab, you will need to manually run either of the two deployment example actions/jobs.

Contribute

PRs welcome!