yes use wrangler

actually im not sure if the code run in quick edit runs on GCP, or its just outbound requests being proxied through GCP

but regardless, cloudflare ips arent being used for quick edit

ok, but when i push the code with wrangler the worker is running at cloudflare and uses there ips?

I believe it should, but one from the team would be able to confirm. @User, maybe.

You could also deploy it to the .workers.dev subdomain just for testing or a different URL and put Access in front to limit access to the feature.

Access is free for a single user (and a bit more) or bypass by IP.

Ok i will try some things

If you need help we are here (or better, others are for the next few hours, I'll be sleeping...)

FWIW,

wranger dev

runs on our real edge, not GCP. But the online editor and its preview still run in GCP. Eventually we want to switch those over to the

wrangler dev

architecture. So much to do...

Hey guys , I am using nextjs to build our product page with authentication , I am having hard time implementing auth using AWS SDK , can't use Amplify because not compatible with cloudflare workers. Are there any FOSS authentication libraries you would suggest using which are compatible with Cloudflare workers ?

if you're interested in using auth0, i wrote a tutorial for our docs last year that covers using it in your workers app https://developers.cloudflare.com/workers/tutorials/authorize-users-with-auth0

I'm trying to work on some caching of assets fetched from a worker. In my worker, I'm see that the request headers are an empty object

{}

. However, I'm absolutely certain my browser is sending headers because I'm using Proxyman to watch the requests/responses. Is this expected behavior? Does Workers strip headers before they are processed?

addEventListener("fetch", (event) => {
  const { request } = event;
  console.log("addEventListener -> request.method, request.url");
  console.log(request.method, request.url);
  console.log("addEventListener -> request.headers");
  console.log(request.headers);

  event.respondWith(handleRequest(event));
});

This image shows the logging from the code above: This image shows the logging of the actual event that CF received using

wrangler tail

. It clearly shows that there were headers in the original request.

If you switch it out for

console.log(Object.fromEntries(request.headers.entries()));

, does that work?

it does!

Fab! The Headers object continually catches me out. It never behaves quite like I'd expect, but that's just the spec, not how it's been implemented in Workers. https://developer.mozilla.org/en-US/docs/Web/API/Headers

Thanks!

It's not just a plain object like I always expect it to be, so you need to do stuff like ^ to get all the data out of it.

Yep, I've been all over the MDN docs but failed to click the

.headers

link here: https://developer.mozilla.org/en-US/docs/Web/API/Request It makes it look like you can simply access them like :

Request.headers

. Thanks so much!

No worries! 🙂

Did cloudflare worker support hybrid apps like nextjs

no, but also yes at the same time

Next.JS is normally meant to be ran in node.js environments, which Workers are not.

However, there is https://fab.dev/ which does make it possible to create a bundle with next.js that works on Workers

thanks @User

I’m a bit confused about the recommended DNS configuration when you want to link a domain to Cloudflare Workers.

This it what it says in the documentation:

“If your route is configured to a hostname, you will need to add a DNS record to Cloudflare to ensure that the hostname can be resolved externally. If your Worker acts as your origin (the response comes directly from a Worker), you should enter a placeholder (dummy) AAAA record pointing to 100::, which is the reserved IPv6 discard prefixOpen external link.”

Wouldn’t you also need an A record?

Just a AAAA is enough 😊 (or you can use a similar discard A record instead if you'd prefer)