Welcome to swrv... discussions!

[darrenjennings]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with vue/swrv, or if you are using swrv!

[davestaab]

Hello!

Thanks for your work on this Darren. With that said I don't actually use this library! 😞

I found this library after reading this blog post: https://epicreact.dev/my-state-management-mistake/
It got me thinking about server state vs app state. This project stood out as a potential solution for our vue app and I hope to try it soon. I know in my project we do alot of work to get the server state loaded/setup/transformed/updated/messaged/etc. It's no easy task and I think something like this might improve things.

If I understand things, it seems to be very chatty. Do you normally get new data every time a key is used? How are apps being setup? I see the default dedupingInterval is 2 seconds, is that typical or just a random default?

I'm interested in hearing more about how people are setting their defaults and how many requests deviate from the defaults.

Thanks again!

-Dave

[darrenjennings]

If I understand things, it seems to be very chatty

Chatty by design, see: https://github.com/Kong/swrv#why-does-swrv-make-so-many-requests One addendum to that FAQ I might say is that swrv is fetcher agnostic so, if you want, your fetcher can also utilize cache control headers so that your frontend client can be chatty/overfetch, while your server can instruct the browser to store responses to in-memory/disk cache with server determined cache expiry.

How are apps being setup?

Hard to say but at kong we use a wrapper hook on useSwrv kinda like https://github.com/Kong/swrv/blob/master/examples/axios-typescript-nuxt/pages/index.vue#L34-L43 so that we can mess with global dedupingInterval or give it a longer life for requests that don't need to revalidate much. For example, some users will fetch once and then never revalidate for something like a session/app config.

I see the default dedupingInterval is 2 seconds, is that typical or just a random default?

Random. It was mainly chosen to have feature parity with vercel/swr. Not sure there is any data to back up the number itself. If you need it to be more real-time, you can lower the number and use refreshInterval, or if you need it to prefer stale, you can raise the number.

Thanks for the feedback! I'm biased, but you should definitely give it a try ;)