Table valued Parameter Equivalent in Postgresql

Question

In postgresql- what is the equivalent of stored procedure with table valued paramater(MSSQL)?

Answer 1

@HamidKhan -Is your development language Java or C #?

CREATE TABLE public.employee (
  emp_id INTEGER NOT NULL,
  emp_nm VARCHAR(40),
  first_in_time TIMESTAMP WITH TIME ZONE DEFAULT clock_timestamp(),
  last_chg_time TIMESTAMP WITH TIME ZONE DEFAULT clock_timestamp(),
  CONSTRAINT employee_pkey PRIMARY KEY(emp_id)
)
WITH (oids = false);

CREATE TYPE public.employee_udt AS (
  emp_id INTEGER,
  emp_nm VARCHAR(40)
);

--c# code 1

public class EmployeeUdt
{
    [PgName("emp_id")] 
    public int EmpId { get; set; } 

    [PgName("emp_nm")]
    public string EmpNm { get; set; }
}

--c# code 2

List<EmployeeUdt> lst_param = new List<EmployeeUdt>();

for (int i = 0; i < this.dataGridView1.Rows.Count; i++)
{
lst_param.Add(
new EmployeeUdt
{
    EmpId = Convert.ToInt32(this.dataGridView1[0, i].Value),
    EmpNm = this.dataGridView1[1, i].Value.ToString()
}
);
}

var _param = new[] {
new NpgsqlParameter
{
    ParameterName="p_employee",
    NpgsqlDbType = NpgsqlDbType.Composite,
    SpecificType = typeof(EmployeeUdt[]),
    NpgsqlValue = lst_param
}

};

SqlHelper.ExecuteNonQuery<EmployeeUdt>(this.connstring, "usp_set_emp", _param);

Answer 2

PostgreSQL has no table variables. The most similar type is a composite array:

create type foo_typ as (a int, b int);

do $$
declare
  f foo_typ[];
  r foo_typ;
begin
  f := ARRAY(SELECT row(10, 20)::foo_typ from generate_series(1,10));
  for r in select * from unnest(f) g(v)
  loop
    raise notice 'a:%, b:%', r.a, r.b;
  end loop;
end;
$$;
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
NOTICE:  a:10, b:20
DO

Writing stored procedures in PostgreSQL, you are practically starting from zero coming from T-SQL in terms of syntax familiarity. The PostgreSQL syntax is similar to Oracle and DB2 - but very different to SQL Server.