ok, I see. No I do not think so. What is the alternative when people rely on package with node built
ok, I see. No I do not think so. What is the alternative when people rely on package with node built-in module ? Any cloudflare product
G(going to read the code of ytdl-core
- but the patch path seems a bit long to me)
IThere is no Cloudflare product that allows build-in modules unfortunately - it's either patching it to use
node:buffer or finding an alternative package.IActually, you might be able to.
Inode_compat = true to your wrangler.tomlIt's the older form of Node compat so it won't work as well for some things but it might work.
IOh this is Pages. Hmmm.
INot sure if you can use node_compat in Pages
GReading the code, as https://github.com/fent/node-ytdl-core/blob/master/package.json#L39 depends on m3u8stream,miniget and sax I guess I should also patch this packages ?
IYeah, probably. Assuming they use node built-ins
Ghttps://developers.cloudflare.com/workers/runtime-apis/nodejs/#enable-nodejs-with-pages-functions apparently yes it's possible
GI did that but no effect
IThat would be the newer nodejs_compat (similar names, it's a bit confusing) which requires you to change your imports to
node:buffer unfortunately.GI locally tested with
npx wrangler pages dev --compatibility-flags="nodejs_compat" -- npm run devII don't think the
node_compat (the original one) works with Pages.Goh ! I see 
node_compat vs nodejs_compatjs heheGok lets dive a bit into built-in stuff
Gand see what's possible
Gany recommandation to check if a package rely on node built-in module ?
IOther than deploying the package to test, not really. I suppose you could
npm installbufferfs, etc in node_modules but that might not catch everything.GThanks !
R@auth/sveltekit! I have the same issue on cf pages functions.
Rhttps://developers.cloudflare.com/workers/runtime-apis/nodejs/#enable-nodejs-from-the-cloudflare-dashboard suggests that those nodejs packages should work, yet they aren't. any help?
Implemented Node.js runtime APIs and enablement instructions for your Worker project.
HThis is a build-time error, not a runtime error. SvelteKit doesn’t know that you are enabling node compatibility, so it throws an error
HYou need to mark those node dependencies as external
ROh thanks, do you have a resource how to configure that?
HNot sure. I think I remember you have to do it via the Vite Options in SvelteKit. Might be able to dig it up later if you can't find it.
Rwould be great, can't find any resource 
GOh !
GSo, it would work for my package also ?
GIt's a bit hard to understand your choice on the
node_compat vs nodejs_compatOn netlify, we
just npm i -S and import, on Google we edit the package.json and it just work.GOn cloudflare, everything is super smooth and better except this part.
GDisclamer, I ended up using Google Cloud Functions but I've preferred not to
KWorkers and Functions aren’t Node, Netlify and GCF are.
Knode_compat are build time polyfills and nodejs_compat is a subset of modules built into the runtime
THello, I am using React JS and page function, how can I binding KV and use it in my react application locally, this is the code to access my local KV but it is not working, when I deploy then it works normally. Thanks a lot for your answers!
KHello, I am using page function that bind with R2 bucket. Is there a way to upload and delete image directly from url
SHello, I have set the maximum CPU timelimit to 1ms in standard pricing of cloudflare workers the usage shows that CPU time of "60,237 ms" for 40 requests, seems like a lot, is the comma misplaced instead of a dot?
JHello, anyone else getting this error randomly?? I have no call to a clone() in my function but it keeps getting thrown on each call..
Your worker created multiple branches of a single stream (for instance, by calling response.clone() or request.clone()) but did not read the body of both branches. This is wasteful, as it forces the system to buffer the entire stream of data in memory, rather than streaming it through. This may cause your worker to be unexpectedly terminated for going over the memory limit. If you only meant to copy the request or response headers and metadata (e.g. in order to be able to modify them), use the appropriate constructors instead (for instance, new Response(response.body, response), , etc).
KPages Functions are working fine when developing locally with
wrangler pages dev public, but when deployed to CF its throwing 404.html
K
Kany solution ??
WOu. I have the same issue. is it Next.js project?
If yes, I created a ticket here: https://github.com/cloudflare/next-on-pages/issues/644
If yes, I created a ticket here: https://github.com/cloudflare/next-on-pages/issues/644
GitHub
The main question: Is it possible to use a combination of Static export and POST Request function? If Yes, then it's a bug, if not, then should be documented better. next-on-pages environment r...
WBut I received the reply here, https://github.com/cloudflare/next-on-pages/issues/644#issuecomment-1896470172 that it's possible
GitHub
The main question: Is it possible to use a combination of Static export and POST Request function? If Yes, then it's a bug, if not, then should be documented better. next-on-pages environment r...
Pdoes
Looks like yes? https://github.com/cloudflare/workers-sdk/blob/89d78c0ac444885298ac052b0b3de23b69fb029b/packages/wrangler/src/pages/dev.ts#L340
wrangler pages dev only watch the /functions directory for file changes? trying to reason about how to structure a project that has types in /common and queries and mutations in the /database folder. Currently needing to restart the dev server when I change those shared files.Looks like yes? https://github.com/cloudflare/workers-sdk/blob/89d78c0ac444885298ac052b0b3de23b69fb029b/packages/wrangler/src/pages/dev.ts#L340
PSounds like the ideal path is to put all code that functions may need into
/functions and build out folder structure in there, and then when the UI needs shared stuff it can be a bit more configurable to watch code in the functions folderKno, just a simple HTML, CSS, JS webapp.
PIf you look at the deployment in the dashboard there's a functions tab that should provide some more info on what wrangler inferred as far as routing goes
PYep. It does not seem like a wise approach as, personally, I have code I do not want to be an endpoint 
PIf you find a better way to handle it I'd be super grateful if you let me know