Re: Generate pg_stat_get_xact*() functions with Macros

Hi,

On 3/27/23 3:20 AM, Michael Paquier wrote:
> On Sat, Mar 25, 2023 at 11:50:50AM +0900, Michael Paquier wrote:
>> On Fri, Mar 24, 2023 at 06:58:30AM +0100, Drouvot, Bertrand wrote:
>>> - Does not include the refactoring for
>>> pg_stat_get_xact_function_total_time(),
>>> pg_stat_get_xact_function_self_time(),
>>> pg_stat_get_function_total_time() and
>>> pg_stat_get_function_self_time(). I think they can be done in a
>>> dedicated commit once we agree on the renaming for
>>> PG_STAT_GET_DBENTRY_FLOAT8 (see Andres's comment up-thread) so that
>>> the new macros can match the future agreement.
>
> Thanks for the reminder. I have completely missed that this is
> mentioned in [1], and that it is all about 8018ffb. The suggestion to
> prefix the macro names with a "_MS" to outline the conversion sounds
> like a good one seen from here. So please find attached a patch to do
> this adjustment, completed with a similar switch for the two counters
> of the function entries.
>

Thanks! LGTM, but what about also taking care of pg_stat_get_xact_function_total_time()
and pg_stat_get_xact_function_self_time() while at it?

>>> - Does include the refactoring of the new
>>> - pg_stat_get_xact_tuples_newpage_updated() function (added in
>>> - ae4fdde135)
>>
>> Fine by me. One step is better than no steps, and this helps:
>> 1 file changed, 29 insertions(+), 97 deletions(-)
>>
>> I'll go apply that if there are no objections.
>
> Just did this part to shave a bit more code.

Thanks!

Regards,

--
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com