Jan Piotrowski

If you "set max_prepared_statements to 0" you disable the new named prepared statement support of PgBouncer and will need to keep using `pgbouncer=true` @MorenoMdz. The problem of the new PgBouncer...

So to recap: 1. PgBouncer Needs `pgbouncer=true` to work correctly 2. PgBouncer 1.21.0, max_prepared_statements = 0 => Needs `pgbouncer=true` to work correctly (same behavior as 0 => Will fail with...

Related issue: https://github.com/prisma/prisma/issues/10386

Support for the PostgreSQL index creation via Prisma Migrate is tracked in https://github.com/prisma/prisma/issues/10386

Here is a PR fixing the index definition in the docs: https://github.com/prisma/docs/pull/2683 The reason why index creation needs to use `to_tsvector('english', ...)` or similar: > Notice that the 2-argument version...

We are aware and trying to find a good solution. This one case here would be ok to fix - but there are many more combinations of queries and indexes...

Then you actually have a feature request, to support precomputed tsvector columns. That is of course a good request and you should create a new feature request for that. (That...

Interesting. You would suggest to add an attribute `@fulltextsearch` to a field in a model to influence the query that is created when one tries to run a `search` query,...

Why don't you rename the model or model field in the Prisma schema, and then generate a migration with `prisma migrate dev`? (Via the VSCode extension you can also do...

Then these are related: https://github.com/prisma/prisma/issues/7790 + https://github.com/prisma/prisma/issues/12387