Re: SQL/JSON features for v15 - Mailing list pgsql-hackers

From Tom Lane
Subject Re: SQL/JSON features for v15
Date
Msg-id 35146.1661971558@sss.pgh.pa.us
Whole thread Raw
In response to Re: SQL/JSON features for v15  (Bruce Momjian <bruce@momjian.us>)
Responses Re: SQL/JSON features for v15
List pgsql-hackers
Bruce Momjian <bruce@momjian.us> writes:
> I guess you are saying that setting a cut-off was a bad idea, or that
> the cut-off was too close to the final release date.  For me, I think
> there were three questions:

> 1.  Were subtransactions acceptable, consensus no
> 2.  Could trapping errors work for PG 15, consensus no
> 3.  Could the feature be trimmed back for PG 15 to avoid these, consensus ?

We could probably have accomplished #3 if there was more time,
but we're out of time.  (I'm not entirely convinced that spending
effort towards #3 was productive anyway, given that we're now thinking
about a much differently-scoped patch with API changes.)

> I don't think our community works well when there are three issues in
> play at once.

To the extent that there was a management failure here, it was that
we didn't press for a resolution sooner.  Given the scale of the
concerns raised in June, I kind of agree with Andres' opinion that
fixing them post-freeze was doomed to failure.  It was definitely
doomed once we reached August with no real work done towards it.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Peter Geoghegan
Date:
Subject: Re: effective_multixact_freeze_max_age issue
Next
From: Andres Freund
Date:
Subject: Re: Tracking last scan time