Horton
Menu

Re: pg_stats and range statistics

Alexander Korotkov <aekorotkov@gmail.com>

View thread
From: Alexander Korotkov <aekorotkov@gmail.com>
To: jian he <jian.universality@gmail.com>
Cc: "Gregory Stark (as CFM)" <stark.cfm@gmail.com>, Egor Rogov <e.rogov@postgrespro.ru>, Tomas Vondra <tomas.vondra@enterprisedb.com>, Justin Pryzby <pryzby@telsasoft.com>, Soumyadeep Chakraborty <soumyadeep2007@gmail.com>, pgsql-hackers@lists.postgresql.org
Date: 2023-11-25T16:55:11Z
Lists: pgsql-hackers

Attachments

On Sat, Nov 25, 2023 at 10:58 AM jian he <jian.universality@gmail.com> wrote:
> On Sat, Nov 25, 2023 at 7:06 AM Alexander Korotkov <aekorotkov@gmail.com> wrote:
> >
> > Hi!
> > Additionally, I found that the current patch can't handle infinite
> > range bounds and discards information about inclusiveness of range
> > bounds.  The infinite bounds could be represented as NULL (while I'm
> > not sure how good this representation is).  Regarding inclusiveness, I
> > don't see the possibility to represent them in a reasonable way within
> > an array of base types.  I also don't feel good about discarding the
> > accuracy in the pg_stats view.
> >
>
> in range_length_histogram, maybe we can document that when calculating
> the length of a range, inclusiveness will be true.

I've revised the patchset.  Edited comment in pg_statistic.h as you
proposed.  And I've added to the documentation a short note on how the
range length histogram is calculated.

------
Regards,
Alexander Korotkov

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Collect and use histograms of lower and upper bounds for range types.