Re: Functional index performance question
| От | Arguile |
|---|---|
| Тема | Re: Functional index performance question |
| Дата | |
| Msg-id | 1064927204.26645.19.camel@broadswd обсуждение исходный текст |
| Ответ на | Functional index performance question (Mike Mascari <mascarm@mascari.com>) |
| Ответы |
Re: Functional index performance question
Re: Functional index performance question Re: Functional index performance question |
| Список | pgsql-general |
On Tue, 2003-09-30 at 07:06, Mike Mascari wrote:
[snip]
> CREATE INDEX i_employees ON employees(lower(name));
>
> Let's also assume that the lower() function is computationally
> expensive. Now if I have a query like:
>
> SELECT lower(name)
> FROM employees
> WHERE lower(name) = 'mike'
>
> will PostgreSQL re-evaluate lower(name)? Is it necessary?
No, it won't re-evaluate. Which is why functional indexes work and why
you can only declare a functional index on a referentially transparent
function (see IMMUTABLE flag in CREATE FUNCTION).
See also:
http://developer.postgresql.org/docs/postgres/indexes-expressional.html
http://developer.postgresql.org/docs/postgres/sql-createfunction.html
В списке pgsql-general по дате отправления: