Laurence Isla

Thanks for the report. > Besides fixing the URL, an additional HTTP server redirect from https://postgrest.org/en/v12.0/api to https://postgrest.org/en/v12.0/references/api would probably make sense. Just added the redirect, so it should be...

Yes, this is a current limitation in the OpenAPI output. We're currently working on the [`postgrest-openapi` repository](https://github.com/PostgREST/postgrest-openapi), so this will be added there.

The pros are that [the library](https://hackage.haskell.org/package/wai-cors-0.2.7/docs/src/Network.Wai.Middleware.Cors.html#corsFailure) returns the errors nicely, with a message like "Origin header is missing" or "Unsupported origin..." using: ```haskell corsFailure msg = WAI.responseLBS HTTP.status400 [("Content-Type", "text/html;...

Found a problem with the library: it expects that any `OPTIONS` request that has an `Origin` header should be considered a `pre-flight` request and it complains when it doesn't have...

> Not sure yet of the shape of the queries. I'm giving this a shot in #3640. My approach is to wrap all the selected columns in `json_agg()`. This would...

Wrapping the key in quotations seems to work: ```bash curl "http://localhost:3000/jsonvals?a->>\"@key\"=eq.1" ``` ```json [{"@key":"1"}] ``` Doing this should work as another workaround: ```js .eq('media3Id->data->>schemaOrg2->>"@type"', schemaPrefix) ```

As a disclaimer, this was an issue reported by a user in a Discord channel for Supabase. They mentioned that they can use workarounds, but was just curious to know...

> Wait, but we are using RETURNING, right? Ah yes, but we're only `RETURNING` the base table columns, not the modified columns for the table inside the `TRIGGER`. Something like...

> What was the problem for this? The libraries `hasql-dynamic-statements` and `hasql-transaction` didn't have versions supporting `hasql` >= 1.7, which `hasql-notification` used. But I think that's fixed now upstream, I'll...

> Hm, the newly added env vars and db options look hard to maintain. > > I'm thinking we should somehow auto-generate them. That's a good idea, we could use...