Re: A thought on Index Organized Tables

Поиск
Список
Период
Сортировка
От Gokulakannan Somasundaram
Тема Re: A thought on Index Organized Tables
Дата
Msg-id 9362e74e1002240409q16ef5cf9m3160cbb0a6a703a1@mail.gmail.com
обсуждение исходный текст
Ответ на Re: A thought on Index Organized Tables  (Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>)
Список pgsql-hackers
Forgot to include the group..

On Wed, Feb 24, 2010 at 5:38 PM, Gokulakannan Somasundaram <gokul007@gmail.com> wrote:
I am not familiar with this term "broken data types", and I just looked for it in the source code and couldn't find it.

What exactly are you referring to?

cheers

andrew

Sorry i missed this. Actually if we create a function A which uses functions like time(), date() and random(), then this function A won't give the same output, even if we give the same input. So if a person has created a data type, which uses these functions, then it can't be made as a primary key in an Index organized table, because i need to reach the same tuple by applying the function on the supplied values. But since the function is mutable, we can't reach the same tuple.

If we decide to support only datatypes containing immutable functions, then there might be people who have created these kind of functions and marked it as immutable( while they are mutable functions). So those functions will result in index-corruption / failed operation. Only if we resolve this issue we can have data structures like IOT.

Hope, i was clear.

Thanks,
Gokul.


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Heikki Linnakangas
Дата:
Сообщение: Re: A thought on Index Organized Tables
Следующее
От: "William"
Дата:
Сообщение: FW: Unable to install PostgreSQL on Windows Server 2003 SP2