| From: | Karsten Hilbert <Karsten(dot)Hilbert(at)gmx(dot)net> |
|---|---|
| To: | pgsql-general(at)lists(dot)postgresql(dot)org |
| Subject: | Re: How should we design our tables and indexes |
| Date: | 2024-02-11 09:36:01 |
| Message-ID: | ZciVAQKU9-mCH3Ua@hermes.hilbert.loc |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Am Sun, Feb 11, 2024 at 12:53:10PM +0530 schrieb veem v:
> >> Pagination is already a hard problem, and does not even make sense when
> > combined with "a continuous stream of inserts". What should the user see
> > when they click on page 2?
> >
>
> When the user clicks to the second page , it will see the next set of rows
> i.e 100 to 200 and next will see 200 to 300 and so on till the result set
> finishes.
Given a continuous stream of inserts "second page" or "next
set of rows" is undefined -- if you aim for live data,
because interleaving data may have been inserting while the
user inspected the first batch of results.
A "second page" is only defined in terms of "what the original
query returned on the first run".
Karsten
--
GPG 40BE 5B0E C98E 1713 AFA6 5BC0 3BEA AC80 7D4F C89B
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Laurenz Albe | 2024-02-11 11:00:22 | Re: How to do faster DML |
| Previous Message | veem v | 2024-02-11 07:55:10 | Re: How to do faster DML |