Hi Pavel, it needs a redesign of original implementation, we should to change API to > use default values with named parameters > > but it doesn't help too much (although it can be readable little bit more) > > instead row_to_json(x, false, true) > > be > > row_ro_json(x, ignore_null := true) > > it is not too much work, but I need a names for parameters >
I have tried adding dummy names (a, b, c) in pg_proc entry you have added. But that is not sufficient. We need to have default values provided to these arguments to work row_ro_json(x, ignore_null := true) call. It was not trivial. So I have not put much thought on that. For name, I choose (row, pretty, ignore_nulls) or similar. However it was my thought. If it is too complex of not so useful then we can ignore it. Thanks -- Jeevan B Chalke Principal Software Engineer, Product Development EnterpriseDB Corporation The Enterprise PostgreSQL Company
