| From: | Jeff Davis <pgsql(at)j-davis(dot)com> |
|---|---|
| To: | Darren Duncan <darren(at)darrenduncan(dot)net> |
| Cc: | Guillaume Lelarge <guillaume(at)lelarge(dot)info>, mike beeper <mbeeper(at)hotmail(dot)com>, pgsql-general(at)postgresql(dot)org, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: [HACKERS] Creating temp tables inside read only transactions |
| Date: | 2011-07-09 06:19:03 |
| Message-ID: | 1310192343.3012.192.camel@jdavis |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general pgsql-hackers |
On Fri, 2011-07-08 at 12:34 -0700, Darren Duncan wrote:
> Yes, but that would just be in-memory or in temporary places external to every
> database. On disk internal to a database there would just be the oid. In fact,
> another aspect of the database model I defined is that each "database" is
> entirely self-contained; while you can do cross-database queries, you don't have
> cross-database constraints, in the general case.
Yes, you can have a "local oid" and a "fully-qualified oid". It sounds
like it might take some effort (which is an understatement) to go
through the system and figure out which ones should be local and which
ones should be fully-qualified.
Regards,
Jeff Davis
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Darren Duncan | 2011-07-09 06:39:48 | Re: [HACKERS] Creating temp tables inside read only transactions |
| Previous Message | Tom Lane | 2011-07-09 05:24:18 | Re: [HACKERS] Creating temp tables inside read only transactions |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Darren Duncan | 2011-07-09 06:39:48 | Re: [HACKERS] Creating temp tables inside read only transactions |
| Previous Message | Tom Lane | 2011-07-09 05:24:18 | Re: [HACKERS] Creating temp tables inside read only transactions |