Sharing an array between Convex and UI?
In the app I'm building for work, I have some data that (ideally) should be shared between Convex and my front-end UI, and I'm not sure how best to accomplish this.
As an example, I have a table of contacts. One of the fields in this table is
currentState. In the current iteration of our data (Airtable), this field has 23 different options. Rather than keep multiple copies of that list, I'd prefer to just maintain one.
One idea that came to mind was to do the following:
* Create the array of options in one of my Convex files
* Transform that into a validator for use in the schema and appropriate query/mutation functions (using the literals() helper function)
* Make a Convex query function that returns the array, to be used in my UI code via useQuery()
The first two are the no-brainers. The last item, though, I'm not sure about. Is it bad form to make a query that doesn't actually query the database? Is there a better way to go about this?8 Replies
Thanks for posting in <#1088161997662724167>.
Reminder: If you have a Convex Pro account, use the Convex Dashboard to file support tickets.
- Provide context: What are you trying to achieve, what is the end-user interaction, what are you seeing? (full error message, command output, etc.)
- Use search.convex.dev to search Docs, Stack, and Discord all at once.
- Additionally, you can post your questions in the Convex Community's <#1228095053885476985> channel to receive a response from AI.
- Avoid tagging staff unless specifically instructed.
Thank you!
I tried searching the docs/Discord for anything about this but came up empty
Nothing wrong with this at all
Convex functions are just serverless functions. The one you're describing would have a 500% cache hit rate too lol
I'm not sure I understand your comment about the 500% cache hit rate. Is that good or bad?
Oh it’s good, just joking about the fact that you’ll pretty much always hit cache if your function doesn’t access the database and takes no args.
It's also fine to define it in code and import it directly from both your frontend and convex functions.
Putting it in a query has benefits of having one source-of-truth, and when you deploy your convex functions any old frontend code will get the new values right away.
On the other hand, putting it in code means you can use it without calling useQuery (which may take a few milliseconds to load) and save on billed function calls.
Cannot believe this didn't occur to me lol, yes do this ☝️
I probably should have mentioned this, but this project is a monorepo (Turborepo specifically), with the Convex functions in a package that's separate from the app. Accessing the Convex stuff from within the app is all done via the package prefix; e.g.
import { Doc } from "@loaportal/backend/convex/_generated/dataModel" .
If I put this array in a file named something like shared.ts inside the convex folder and provide access to it by defining a path in the package exports, would that work?