Starter to get going with Remix and Cloudflare Workers.

This template is based on the starter created by Jacob(remix-cloudflare-worker-template).

This turborepo uses npm as a package manager. It includes the following packages/apps:

• counter-do: a Durable Object class for managing the count state
• remix-app: a Remix app (with Tailwind)
• worker: a Worker handler function
• cloudflare-env: types for Cloudflare environement variables
• eslint-config-custom: eslint configurations (includes @remix-run/eslint-config and eslint-config-prettier)
• tsconfig: tsconfig.json used throughout the monorepo

Let's setup wrangler CLI instalation (wrangler requires a minimum Node version of 16)

npm i @cloudflare/wrangler -g

After installation, log in to your Cloudflare account from the CLI.

wrangler login

Let's install the dependencies.

npm install

Now we can set up the project.

npm run setup

The setup script will ask for your worker's name. It can be any name. And then, we also need to add the Account ID, which you can find by visiting the Cloudflare Dashboard.

You can add the Worker name and the Account ID later to the `wrangler.toml.

That's all we're ready to push to prod! 🚀

Before pushing to the worker via GitHub action, we have to configure the CF_API_TOKEN secret in GitHub. We can generate an API Token from here. When presented with the list of templates to choose from, select the "Edit Cloudflare Workers" option. This template should have the necessary permissions to push a Worker from GitHub. Now we can commit the changes made to wrangler.toml and push the changes.

git commit -am "<message>"
git push

npm install @cloudflare/wrangler -g
wrangler login
npm install
npm run setup
# configure CF_API_TOKEN action secret
git add -A -m "<message>"
git push

Durable Objects are only available with a Workers paid subscription. However, you can upgrade to the paid plan from the Dashboard.

This starter template comes with a simple DO implementation to keep track of the number of times the root loader is invoked.

If you're starting with DO and not sure what it is, go through the official docs on Durable Objects will be a good start! And checkout using durable objects for more applications of DO.

This template comes with a script to create the boilerplate for a new Durable Object class.

npm run new:do

The script will have instructions to initialise the DO with the worker. Don't forget to follow them!

You can skip this section if you have used the script to generate the DO class. Continue for more information on DO :)

To define a DO class, check out the docs.

To include the DO class into the worker, we have to add the new DO package as a dependendency of the worker. Then we need to create a binding for the DO in the configuration file wrangler.toml.

[durable_objects]
bindings = [
  {name = "<DO_BINDING_NAME>", class_name = "<DO_CLASS_NAME>"},
]

For development add the following to wrangler.dev.toml

[env.dev.durable_objects]
bindings = [
  {name = "<DO_BINDING_NAME>", class_name = "<DO_CLASS_NAME>"},
]

We must create a migration to register the DO in the configuration file.

[[migrations]]
tag = "v<RUNNING_TAG_ID>"
new_classes = ["<DO_CLASS_NAME>"]

More info about uploading a DO here.

The DO binding will be available in the data functions (loader/action) through the context argument. Types for the context can be added at cloudflare-env. To add a type for a newly created DO, we have to add the following to the .d.ts file

<DO_BINDING_NAME>: DurableObjectNamespace

Now we can access the DO binding from the data function through the context.

export let loader: LoaderFunction = ({ context }) => {
  context.MY_DO
  //        ^ Will have proper type definitions
  return null
}

It requires a migration to delete a DO. More info here

• KV
• Wrangler KV CLI

This template does not come with a KV namespace attached to it. However, you can create one using the Wrangler CLI.

wrangler kv:namespace create "MY_KV"

The above command will create a KV namespace. Now we need to bind the created namespace with the worker. When we run the create command, the CLI will print the binding configuration we need to add to our wrangler.toml configuration file. It will look like

kv_namespaces = [
  { binding = "MY_KV", id = "xxxx" }
]

We must add this above the [site] block in the wrangler.toml file.

We have added the KV namespace binding for the production environment, but we also need a namespace for dev. We can do that by creating a new namespace for dev.

wrangler kv:namespace create "MY_KV" --preview

This will generate a namespace for the dev environment, and we must add this below the [env.dev] block in the dev configuration file wrangler.dev.toml.

[env.dev]
kv_namespaces = [
  { binding = "MY_KV", id = "xxxx", preview_id = "xxxx" }
]

Note: We need to add the preview_id key to the configuration file along with the id key with the same value (ref: stackoverflow).

The bounded KV will be available in the loader/action via the context argument passed to the functions. We define types for the context at cloudflare-env. To add types for a newly bounded KV, we have to add the following to the .d.ts file

MY_KV: KVNamespace

Now we can access the KV namespace from the data function through the context.

export let loader: LoaderFunction = ({ context }) => {
  context.MY_KV
  //        ^ Will have proper type definitions
  return null
}

You must run wrangler commands from a directory which contains the wrangler.toml file. Either we can cd into the worker directory present at packages/worker, or we can specify the path to the configuration file in the CLI.

To set a worker secret, we can

cd packages/worker
wrangler secret put <SECRET_NAME>

or,

wrangler secret -c packages/worker/wrangler.toml

Like DO/KV binding, the Env variables will be passed to the data functions via the context argument. But we have only configured the production worker with the secret. So let's configure the local environment with the secret.

When we ran npm run setup, the CLI would have created packages/worker/wrangler.dev.toml. And the configuration file should have a vars key under the [env.dev] table. So we can add the new secret there.

[env.dev]
vars = {SECRET_KEY = "<value>"}

One last thing to do is add the type definition for the Env var at config/cloudflare-env/index.d.ts.

SECRET_KEY: string

Now, we can access SESSION_SECRET via context.env.SESSION_SECRET in the data functions inside our Remix app.

This repository is used in the npx create-turbo@latest command and selected when choosing which package manager you wish to use with your monorepo (npm).

To build all apps and packages, run the following command:

npm run build

To develop all apps and packages, run the following command:

npm run dev

Turborepo can use a technique known as Remote Caching (Beta) to share cache artefacts across machines, enabling you to share build caches with your team and CI/CD pipelines.

By default, Turborepo will cache locally. To enable Remote Caching (Beta), you will need an account with Vercel. If you don't have an account, you can create one, then enter the following commands:

npx turbo login

This will authenticate the Turborepo CLI with your Vercel account.

Next, you can link your Turborepo to your Remote Cache by running the following command from the root of your turborepo:

npx turbo link

• Remix
• Cloudflare
• Workers
• KV
• Durable Objects
• Usage and Practical applications of DO
• Wrangler
• Wrangler Configuration file reference
• Wrangler KV

Learn more about the power of Turborepo:

• Pipelines
• Caching
• Remote Caching (Beta)
• Scoped Tasks
• Configuration Options
• CLI Usage