PGSQL "macro" or "inplace subfunction"?

Поиск
Список
Период
Сортировка
От Durumdara
Тема PGSQL "macro" or "inplace subfunction"?
Дата
Msg-id CAEcMXh=Bp0w4o-4q5BpNTKJwEYfnYJvBNZSKsp+Yk+oL63810A@mail.gmail.com
обсуждение исходный текст
Список pgsql-general
Dear Members!

Many times I need to type same expressions in SQL queries.

select [exprs] from anytable
where [exprs] > 0
order by [exprs]

Sometimes I solve this with subquery, or "WITH" query, because it is easier to define value once, and use it more times...

select * from (
select [exprs] as e1 from anytable
) t where e1 > 0
order by e1

(But I can't do this with group by)

Sometimes the expression is a repeated code which I don't want to create a persistent function, but I need to use more times.

Pseudo code:

macro _Sgn(aCol1, aCol2):
    case 
        when(aCol1 = aCol2) then 0
        when (aCol1 < aCol2) then -1
        when (aCol1 > aCol2) then 1
        else NUL
   end

select
    _Sgn(X1, X2) as XDiff,
    _Sgn(Y1, Y2) as YDiff,
    _Sgn(Z1, Z2) as ZDiff
    ...

This changed by the compiler to...

select
    case 
        when(X1 = X2) then 0
        when (X1 < X2) then -1
        when (X1 > X2) then 1
        else NUL
   end
...
    case 
        when(Y1 = Y2) then 0
...

Do you know same thing in PGSQL queries?

Thank you for any help!

Best regards

dd

В списке pgsql-general по дате отправления:

Предыдущее
От: Luca Ferrari
Дата:
Сообщение: Re: Altering multiple column types
Следующее
От: Vladimir Soldatov
Дата:
Сообщение: SCRAM-SHA-256, is it possible to retrieve enough information from PGserver (pg_authid etc) to perform authentication as a client