| From: | Peter Eisentraut <peter(dot)eisentraut(at)2ndquadrant(dot)com> |
|---|---|
| To: | Stephen Frost <sfrost(at)snowman(dot)net>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Remove superuser() checks from pgstattuple |
| Date: | 2016-09-08 16:28:14 |
| Message-ID: | 682a41e2-075a-86d7-769a-c046c2a34011@2ndquadrant.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 9/4/16 7:36 PM, Stephen Frost wrote:
> Perhaps if the versioned install script was actually a non-versioned
> install script in the source tree, and the Makefile simply installed it
> into the correct place, then we could eliminate that part. (All very
> hand-wavy, I've not looked at what it'd take to do.)
A number of external extensions actually do it that way, but it also has
its problems. For example, if you don't preserve the old base install
scripts, you can't actually test whether your upgrade scripts work in
the sense that they upgrade you to the same place.
This is now being obsoleted by the later idea of allowing base installs
from a chain of upgrade scripts. But if your upgrade scripts contain
ALTER TABLE commands, you will probably still want to write base install
scripts that do a fresh CREATE TABLE instead.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2016-09-08 16:36:02 | Default make target in test modules |
| Previous Message | Daniel Verite | 2016-09-08 16:25:09 | Re: Surprising behaviour of \set AUTOCOMMIT ON |