Thread: UPDATE... RETURNING AFTER/BEFORE

Hello,
the core of the feature is done. It works for simple statements like:
UPDATE foo2 SET bar1=bar1+1 RETURNING before.bar1, after.bar1,
foo2.bar1, foo2.bar1*before.bar1 AS sth;
or
UPDATE foo SET bar = bar || ' __ ' RETURNING md5(after.bar),
md5(before.bar);
However some of regression tests fail (71 of 136) and it is the next
step of my work. The other problem is the performance, it unnecessary
Seq Scan over whole table (values from these scans are not used).
Regards,
Karol Trzcionka

This week I've finished the feature patch. It is sent to -hackers and
posted on commitfest. The remaining work is to prepare regression tests
and fix any issues reported on -hackers (if any).
Regards,
Karol

----- Original Message -----
> This week I've finished the feature patch. It is sent to -hackers and
> posted on commitfest. The remaining work is to prepare regression tests
> and fix any issues reported on -hackers (if any).
> Regards,
> Karol

I wouldn't expect any real feedback on it until CF1 ends, except from your mentor.

--Josh Berkus

On Fri, Jul 05, 2013 at 10:56:24PM +0200, Karol Trzcionka wrote:
> This week I've finished the feature patch.

Not quite :)

> It is sent to -hackers and posted on commitfest. The remaining work
> is to prepare regression tests and fix any issues reported on
> -hackers (if any).

Regression tests and bug fixes are good.  Documentation may need
improving, too.  This is all part of the craft of actually delivering
working software that people will use, maintain, etc.

Cheers,
David.
--
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778  AIM: dfetter666  Yahoo!: dfetter
Skype: davidfetter      XMPP: david.fetter@gmail.com
iCal: webcal://www.tripit.com/feed/ical/people/david74/tripit.ics

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate