Cyou can
CIf it's a git project you can direct upload, if it's a direct upload project it can never git though
Cit just works via wrangler, I've done it before
Ryeah but i dont have wrangler setup on thos proejct locally
Rits just a directory that i push to git on vscode
Cif it's static all you really need to do is have wrangler global, and then npx wrangler pages deploy $directory, and it'll walk you through picking the right account & project, and push a direct upload deploy
Cor you can just wait out github, they may break often but it's usually not too long before they fix
NOk, stupid quesiton, please confirm, when we use CF page, there is not any more step we cloud config for CDN for best performance with CF, correct ?
UI dont really understand what you mean? Pages will handle its own cache to be as fast as it can if thats what you mean.
Nyes. that is what looking for. Thank you.
Nwhat is the best way to use Page access function, when we send a unique link to someone and keep track of any visit,
Nno pin/code/email needed, if possible, Just a unique URL something we can do with Page access /zero trust ?
Cif u're using framework like nextjs, sveltekit, ... they'll handle the cache thing for you. they make everything cache forever except api endpoints.
Nwhich means if we write simple html page, we don't need to do anything?
so if we use anything like static site generater, we need to dig deeper ?
so if we use anything like static site generater, we need to dig deeper ?
CI mean if you use framework, they have the part that handle cache-control.
Cif u do on your own, you gotta set that header by yourself.
Nthe problem is I use WordPress plugin to generate static site, which is not common framework. I guess need to swith to hugo or next framework for next project.
UStatic content is perfect for pages. The framework does not control the cache when it is static pages.
EPages is already distributed and already has its own cache, you don't need to change anything
Ncool. Thank you all for confirm and idea for the framework solutions.
GHi all. I've posted in pages-help but thought I'd quickly ask here as well : is there any way to debug when you get an error message referencing minified js output? [This is for a nuxt3 pages app]
GThis has been driving me insane, so any help would be greatly appreciated 
GI get a similar error when trying to debug locally using wrangler :
GI can never track down the location of this file [bh25zprwrab.js in this case] so not even sure where to start.
CHi Pages team! It's been a while since I've checked in, but is there now a way to cache data between builds? We need this for Gatsby.
FIt would not be a good idea for CF Pages to have by default hotlinking prevention? To prevent abuse of CF resources
EIs it not possible to use the pg package with Pages? This is for Workers https://developers.cloudflare.com/workers/tutorials/postgres/ but doesn't work in Pages (dev at least)
In this tutorial, you will learn how to create a Cloudflare Workers application and connect it to a PostgreSQL database using TCP Sockets. The Workers …
KWorkers & Pages Functions are the same thing, Functions is just a wrapper around it - what isn’t working specifically?
EI'm using Remix with Pages, and I have a few issues.
When I use the
When I use the
pgSo I added in some
Which comes from the compiled:
serverNodeBuiltinsPolyfill (https://remix.run/docs/en/1.19.3/file-conventions/remix-config#servernodebuiltinspolyfill), which got around those issues, but then I get:Which comes from the compiled:
EThis stuff confuses me - I guess it's some issue with how Remix is bundling the code and what the cf runtime looks for or something
CPages runs on the edge (cf's worker runtime), so a lot of things, packages, features that node provides won't be available.
If you're interested, some serverless hosts like: neon, vercel, supabase, planetscale, turso, cloudflare d1 all have libraries for interacting with their own databases on the edge. They kind of cheat, though, by using fetch requests to their api or some platform-specific tech rather than establishing a database connection. which makes the libraries proprietary (I think) to their provider.
If you're interested, some serverless hosts like: neon, vercel, supabase, planetscale, turso, cloudflare d1 all have libraries for interacting with their own databases on the edge. They kind of cheat, though, by using fetch requests to their api or some platform-specific tech rather than establishing a database connection. which makes the libraries proprietary (I think) to their provider.
CI'm not completely sure if there's a way to connect to pg directly from the edge, as I've just learned this stuff recently, but if you're content with using a serverless-compatable database, those are some good database driver options.
EThanks - my understanding is Cloudflare now suppports TCP connections. I was hoping to use Drizzle to connect directly to Supabase (https://orm.drizzle.team/docs/installation-and-db-connection/postgresql/supabase
- scroll to bottom) which suggests you can directly connect directly
EMy issue with using the Supabase client is that it's HTTP like you said, I can't use transactions which is pretty much a requirement
KThe
pgnode_compatUnfortunate that pages can't run node_compat. If you're needing transactions but still want to use the supabase database for sure, neon's serverless driver is able to run on any postgres database, as long as you run their websocket proxy for your edge clients.
E@kian Ah I have been running
wrangler pages dev ./public --compatibility-flags=nodejs_compat- although, in the dashboard for pages it does allow you to set that?
Knodejs_compatnode_compatnode_compatpgnodejs_compatnode:Well TIL I didn't even notice that
Kpages dev--node-compatpages deployso uh, I don't think it's usable in production
EI get
service core:user:worker: No such compatibility flag: node_compatI'm not requring supabase, the main thing is being able to run transactions (via code) from Pages on a database:
So any SQL DB would be ok, I'm just trying to connect at the moment
I've been using Prisma which is great, but their data proxy reguarly hangs, and their support is slow and not helpful
So any SQL DB would be ok, I'm just trying to connect at the moment
I've been using Prisma which is great, but their data proxy reguarly hangs, and their support is slow and not helpfulEI either get cold starts (on their service) or timeouts
