> On 5 Jan 2019, at 01:04, Michael Paquier wrote:
>
> On Fri, Jan 04, 2019 at 02:45:55PM +0100, Daniel Gustafsson wrote:
>> (Michael: sorry for not having responded to your comments on the patch, $life
>> has had little time over for hacking lately)
>
> No worries, I understand.
>
>> There is t
On Fri, Jan 04, 2019 at 02:45:55PM +0100, Daniel Gustafsson wrote:
> (Michael: sorry for not having responded to your comments on the patch, $life
> has had little time over for hacking lately)
No worries, I understand.
> There is that. We might not be excited about writing tests for this
> cont
(Michael: sorry for not having responded to your comments on the patch, $life
has had little time over for hacking lately)
> On 4 Jan 2019, at 13:49, Michael Paquier wrote:
>
> On Fri, Jan 04, 2019 at 01:06:24PM +0100, Peter Eisentraut wrote:
>> Do we really want to add user-facing options just
On Fri, Jan 04, 2019 at 01:06:24PM +0100, Peter Eisentraut wrote:
> Do we really want to add user-facing options just to be able to run
> tests? Should we write the tests differently instead?
The take is that the output of the plans generated includes data which
is run-dependent because the durat
On 26/12/2018 06:42, Michael Paquier wrote:
> On Tue, Oct 30, 2018 at 11:44:42AM +0100, Daniel Gustafsson wrote:
>> Circling back to this, I updated the patch with providing another
>> option as I couldn’t think of another way to do it cleanly. I’ll
>> add the patch to the next CF but as it’s just
On Tue, Oct 30, 2018 at 11:44:42AM +0100, Daniel Gustafsson wrote:
> Circling back to this, I updated the patch with providing another
> option as I couldn’t think of another way to do it cleanly. I’ll
> add the patch to the next CF but as it’s just about to start it
> should be moved to the next
> On 31 Jul 2018, at 14:23, Andrew Dunstan
> wrote:
>> I’m not sure it’s worth adding this much to the code just to be able to test
>> it, but it seemed like a good excercise to write to have something to reason
>> about.
>
> I think it probably is, buit I'm not very happy about the hack, so I
On 07/17/2018 02:03 PM, Daniel Gustafsson wrote:
On 17 Jul 2018, at 19:11, Andrew Dunstan wrote:
On 07/17/2018 12:04 PM, Daniel Gustafsson wrote:
Since DEBUG is not a defined loglevel, it seems superfluous to include it here.
It’s also omitted from the documentation so it should probably be
> On 17 Jul 2018, at 19:11, Andrew Dunstan
> wrote:
>
> On 07/17/2018 12:04 PM, Daniel Gustafsson wrote:
>> Since DEBUG is not a defined loglevel, it seems superfluous to include it
>> here.
>> It’s also omitted from the documentation so it should probably be omitted
>> from
>> here.
>>
>> +
On 07/17/2018 12:04 PM, Daniel Gustafsson wrote:
On 4 Jul 2018, at 15:34, Andrew Dunstan wrote:
On Wed, Jun 20, 2018 at 2:06 PM, Daniel Gustafsson wrote:
On 27 Apr 2018, at 04:24, Andres Freund wrote:
On 2018-04-27 11:52:18 +0930, Tom Dunstan wrote:
I'd argue this should contain the non
> On 4 Jul 2018, at 15:34, Andrew Dunstan
> wrote:
>
> On Wed, Jun 20, 2018 at 2:06 PM, Daniel Gustafsson wrote:
>>> On 27 Apr 2018, at 04:24, Andres Freund wrote:
>>>
>>> On 2018-04-27 11:52:18 +0930, Tom Dunstan wrote:
> I'd argue this should contain the non-error cases. It's just as
>>
On Wed, Jun 20, 2018 at 2:06 PM, Daniel Gustafsson wrote:
>> On 27 Apr 2018, at 04:24, Andres Freund wrote:
>>
>> On 2018-04-27 11:52:18 +0930, Tom Dunstan wrote:
I'd argue this should contain the non-error cases. It's just as
reasonable to want to add this as a debug level or such.
>>>
> On 27 Apr 2018, at 04:24, Andres Freund wrote:
>
> On 2018-04-27 11:52:18 +0930, Tom Dunstan wrote:
>>> I'd argue this should contain the non-error cases. It's just as
>>> reasonable to want to add this as a debug level or such.
>>
>> So all of warning, info, debug and debug1-5?
>
> Yea. Like
> "Tom" == Tom Dunstan writes:
Tom> Hi all
Tom> This patch allows a user to configure auto_explain to log to
Tom> NOTICE instead of the server log. This allows automated testing of
Tom> application-generated queries to ensure correct plans etc from
Tom> code that can inspect returned n
On 2018-04-27 11:52:18 +0930, Tom Dunstan wrote:
> > I'd argue this should contain the non-error cases. It's just as
> > reasonable to want to add this as a debug level or such.
> >
>
> So all of warning, info, debug and debug1-5?
Yea. Likely nobody will ever use debug5, but I don't see a point m
(Resent with subscribed email address, thanks gmail)
Hi Andres, thanks for the extremely fast review!
On 27 April 2018 at 11:46, Andres Freund wrote:
>
> > I don't see any tests for auto_explain so haven't added any test cases.
>
>> > Happy to do so if that's deemed necessary.
>
>>
> I'd be in
Hi,
On 2018-04-27 11:43:58 +0930, Tom Dunstan wrote:
> This patch allows a user to configure auto_explain to log to NOTICE instead
> of the server log. This allows automated testing of application-generated
> queries to ensure correct plans etc from code that can inspect returned
> notices but not
Hi all
This patch allows a user to configure auto_explain to log to NOTICE instead
of the server log. This allows automated testing of application-generated
queries to ensure correct plans etc from code that can inspect returned
notices but not the server log.
I don't see any tests for auto_expla
18 matches
Mail list logo
