Might be worth their while to consider using an ORM of some sort that supports PGSQL then? I know some don't like ORM's but for projects that are hosted in a diverse amount of setups / environments they make sense to me.
Whichever approach is fine, the end result is that less effort is made to directly support PGSQL in the long run, and you'd get to support any database back-end as a result. I only mention ORM's cause some have decent tooling (EntityFramework (Core), Diesel.rs, etc) which can be useful for generating schemas out of object models. Also in some cases ORMs are using DB abstraction libraries like for example Entity Framework is built on top of ADO .NET.
