从存储过程检索结果的惯用方式是什么？

Question

DROP function mytest();
CREATE OR REPLACE FUNCTION mytest() 
RETURNS TABLE(name text, age int)
 AS 
$$ 
SELECT name, age FROM names
$$ 
LANGUAGE sql;

Most of the examples I've seen with a stored procedure return a single row with a single column and can be used with QueryRow. Here I am using a table as the output. This above returns 4 rows:

    mytest   
------------
 (bob,12)
 (fred,18)
 (james,22)
 (bill,27)
(4 rows)

In Go, what is the idiomatic way to deal with the tuples:

rows, err := db.Query("SELECT mytest()")
if err != nil {
    panic(err)
}

defer rows.Close()
for rows.Next() {
    var items string

    if err = rows.Scan(&items); err != nil {
        panic(err)
    }

    log.Println(items) // items is a string...now what?

}
if err = rows.Err(); err != nil {
    panic(err)
}

// prints some tuples:

(bob,12)
(fred,18)
(james,22)
(bill,27)

Maybe there's a package for this but I haven't found it ;(

Answer 1

Instead of a Query, you could try a PostgreSQL typesafe1 ORM , like src-d/go-kallax, which deals with tuples.

You can see kalax presented in this blog post

• the first priority of kallax is to provide type safety to the data access layer.
• Another goal of kallax is to make sure all models are, first and foremost, Go structs without having to use database-specific types such as, for example, sql.NullInt64. Support for arrays and slices of all basic Go types and all JSON and arrays operators is provided as well.