On Sun, Mar 25, 2012 at 6:11 AM, Marc Mamin <M(dot)Mamin(at)intershop(dot)de> wrote:
> Here is something we'd like to have:
> As we are quite busy and this issue hasn't a high priority, we haven't
> followed it until now :-(
> I'm only a Postgres user, not a hacker, so I don't have the knowledge to
> help on this nor to evaluate if this is might be a good Gssoc project.
> Just an idea for the case you are looking for another topic.
Good idea. If anyone want so pursue it, I'd strongly suggest building
it as a contrib module rather than dedicated syntax, because I'm not
sure there'd be any consensus on adding syntax for it to core.
Actually, though, I wonder how much faster it would be than CREATE
TABLE AS? Block-level copy should be faster than tuple-level copy,
but I'm not sure whether it would be a lot faster or only slightly
The Enterprise PostgreSQL Company
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2012-03-25 15:59:22|
|Subject: Re: [PATCH] Never convert n_distinct < 2 values to a ratio when computing stats |
|Previous:||From: Robert Haas||Date: 2012-03-25 15:27:33|
|Subject: Re: [PATCH] Never convert n_distinct < 2 values to a ratio
when computing stats|