On Sat, May 18, 2019 at 03:45:11PM -0400, Tom Lane wrote:
>Tomas Vondra <tomas.vondra@2ndquadrant.com> writes:
>> On Sat, May 18, 2019 at 11:49:06AM -0700, Andres Freund wrote:
>>>> On Sat, 18 May 2019 at 16:13, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>>>> It seems like what we need here is to have a separation between the
>>>>> *definition* of a stats object (which is what pg_dump needs access
>>>>> to) and the current actual *data* in it.
>
>>> Otoh, having a suboptimal catalog representation that we'll likely have
>>> to change in one of the next releases also isn't great. Seems likely
>>> that we'll need post beta1 catversion bumps anyway?
>
>> But that's not an issue intruduced by PG12, it works like that even for
>> the extended statistics introduced in PG10.
>
>Yeah, but no time like the present to fix it if it's wrong ...
>
Sorry, not sure I understand. Are you saying we should try to rework
this before the beta1 release, or that we don't have time to do that?
I think we have four options - rework it before beta1, rework it after
beta1, rework it in PG13 and leave it as it is now.
If the pg_dump thing si the only issue, maybe there's a simple solution
that reworking all the catalogs. Not sure. Are there any other reasons
why the current catalog representation would be suboptimal, or do we
have some precedent of a catalog split this way? I can't think of any.
regards
--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services