Re: Materialized views don't show up in information_schema
| От | Stephen Frost |
|---|---|
| Тема | Re: Materialized views don't show up in information_schema |
| Дата | |
| Msg-id | 20141011004406.GQ28859@tamriel.snowman.net обсуждение исходный текст |
| Ответ на | Re: Materialized views don't show up in information_schema (Peter Eisentraut <peter_e@gmx.net>) |
| Ответы |
Re: Materialized views don't show up in information_schema
Re: Materialized views don't show up in information_schema |
| Список | pgsql-hackers |
* Peter Eisentraut (peter_e@gmx.net) wrote:
> On 10/10/14 6:53 PM, Stephen Frost wrote:
> > I'm not particularly thrilled with this answer. I'd aruge that the
> > 'materialized' part of mat views isn't relevant to the standard, which
> > does not concern itself with such performance-oriented considerations,
> > and therefore, to the standard's view (pun rather intended), they're
> > views.
>
> For example, you can't drop a materialized view with DROP VIEW. So any
> tool that offers a list of views to manipulate based on the information
> schema would be confused. This is different from temporary views, for
> example.
And users will be confused when using a tool which doesn't see mat
views, which is what started this thread. Overall, I'm inclined to view
materialized views as a subset of views, which would mean that we'd
perhaps add the ability to drop them with 'drop view'.
As a comparison, what about unlogged tables? They're not normal tables
and they aren't defined by the SQL standard either.
Thanks,
Stephen
В списке pgsql-hackers по дате отправления: