Jack Christensen

Sorry, I don't understand what you are asking.

I think this is an unintended side-effect of the new `Codec` system as well as moving to generics for array support. In v4, each array type was made through code...

@elvizlai I think you might have a different issue. `[]string` insert with stdlib should work already. ```go package main import ( "database/sql" "fmt" "log" "os" _ "github.com/jackc/pgx/v5/stdlib" ) func main()...

Candidate solution: https://github.com/jackc/pgx/pull/2020

I agree this would be useful. This might be part of #1803. Otherwise, a new method could be added. I see 3 options for that method. 1. Return the parameter...

Yeah, an iterator would be possible. Though maybe overkill. Hmm... actually if that was done it would be best to use the new https://github.com/golang/go/issues/61405 Go iteration construct (though that would...

It's been a while so I don't remember the reason for sure. It could be a mistake. It also might be that it was not considered necessary since should always...

> If you agree that it's a problem, I'd like to try fixing it this weekend Please do!

> Identifier.Sanitize() should probably check for parts[0] == "" > > Maybe it should also ensure len(parts)

I don't think there is any way around this. The Go `time.Time` type can't represent the PostgreSQL `infinity` value. However, `pgtype.Timestamp` and `pgtype.Timestamptz` can.