Kyle Gray

Since this PR doesn't have any tests, I'm going to go ahead and close it. @WesleyMiller1998 if you're still running into an issue, please open up a bug report with...

@roccoblues I recently merged #1759 which is probably the source of your merge conflicts. I was surprised by the number of changes in the generated code. A quick search lead...

I got this to work by using regular `?` parameters inside of named parameters. The MySQL parser we're using is not happy with function calls after `LIMIT` and `OFFSET` https://play.sqlc.dev/p/ac63085049628c6b20fd85479e03ea071df9e42f5224a025863f7b6adba84ab4...

Possibly fixed here https://github.com/kyleconroy/sqlc/pull/1746

If the query structs are private, how are you calling the methods?

By using an `internal` package you can get the same level of encapsulation. Take a look at the attached example, but if you put an internal package inside your storage...

Please take a look at my comment here (https://github.com/kyleconroy/sqlc/issues/1364#issuecomment-1233613432). I'd like to hold off on merging this change until we've tried to solve it a few different ways.

Agreed with @akutschera about this being a duplicate Duplicate of https://github.com/kyleconroy/sqlc/issues/1676

> I guess the parser doesn't understand that the alias after UNNEST defines a column That is indeed the case.

The newest version of pgx/v5 has a generic [Array](https://pkg.go.dev/github.com/jackc/pgx/[email protected]/pgtype#Array) type that we'll start to use as part of https://github.com/kyleconroy/sqlc/issues/1823. We should only change the multi-dimension support for lib/pq.