Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This is nice, apart from a quick play when they were new I've never really looked at CF Workers and KV again. This made me go and have a look at the docs and find that it's actually looks quite useful. I didn't know they'd added Python and other languages, so that's cool in itself.

I like Jon's example here, a single and fairly short file that does just enough to demo this all, nice.



I'm building a blog directory/reader/search engine [1] on Cloudflare Workers with D1 as the main database. Very pleased with the ecosystem, including Queues and scheduled jobs. DB migrations are simple, local development is a breeze (scheduled jobs don't work though, but there's a simple workaround), and the DB is an SQLite implementation. I am worried about vendor lock in, but as long as I keep the app relatively straight-forward, it shouldn't be a huge deal to migrate to some other Node or Deno backend.

[1] https://minifeed.net/


I'd have thought the SQLite database (D1) would be a more obvious fit for a blog?


Good point, I'm making my way through exploring more of CF so I'll likely try a D1 variant of this. Ta


I have been very impressed with what you can build on it.

I built this little service to add UK specific political location data to Action Network CRM. 300 lines for the actual service!

code here: https://github.com/jms301/ANUpdaterUK/blob/main/src/index.js

Sure it's only taking a webhook, reading an API, looking up a db entry and then updating via API.

But just knowing it's running on someone else's infra and there's so little code to validate is very comforting.

Prior to this I had a version running on a VPS and yeah I can do it. But you end up with a LOT of code / config to run the web server, the queue, the database... And any of that could be introducing bugs and security problems.


I was stoked on D1 when it was announced.

But the maximum DB size is 10GB (https://developers.cloudflare.com/d1/platform/limits/).

Really?


I assume the intention is you store actual data in R2 objects and just hold references to them in D1.


Note that you can create up to 50,000 databases.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: