|From:||Fabien COELHO <coelho(at)cri(dot)ensmp(dot)fr>|
|To:||Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>|
|Cc:||PostgreSQL Developers <pgsql-hackers(at)postgresql(dot)org>|
|Subject:||Re: PATCH: numeric timestamp in log_line_prefix|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
>>> 1) fix the docs (explicitly say that it's a Unix epoch)
>> I would add the word "numeric" in front of timestamp both in the doc and
>> in the postgresql.conf.sample, as it justifies the chosen %n.
> I think we're already using 'unix epoch' in the docs without explicitly
> stating that it's a numeric value, so I don't think we should use it here as
> it'd be inconsistent.
The point was to justify the choice of 'n' somehow.
>>> 2) handle 'padding' properly
> Hmmm, I'm not entirely sure how exactly the padding is supposed to work (IIRC
> I've never used it), and I thought it behaved correctly. But maybe not - I
> think the safest thing is copy what 't' does, so I've done that in attached
> v3 of the patch.
Ok. Version 3 applies and compiles, and padding now works as expected.
Here is a v4 that I also tested, and where I just removed a spurious '.'
in the millisecond format.
|Next Message||Fabien COELHO||2015-08-23 11:25:37||pgbench progress with timestamp|
|Previous Message||Fabien COELHO||2015-08-23 07:03:47||Re: checkpointer continuous flushing|