| From: | Andrew Dunstan <andrew(at)dunslane(dot)net> |
|---|---|
| To: | Sam Mason <sam(at)samason(dot)me(dot)uk> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Question about SPI_prepare |
| Date: | 2008-11-11 18:26:38 |
| Message-ID: | 4919CE5E.3050705@dunslane.net |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Sam Mason wrote:
> On Tue, Nov 11, 2008 at 11:33:41AM -0600, Tim Keitt wrote:
>
>> I have an application where I am building a plan with SPI_plan and
>> then this plan is called multiple times. There is one free parameter
>> ($1) to the plan. The issue is with the order of the values returned.
>> If $1 is identical during successive calls to SPI_execute_plan, is
>> there any guarantee that the rows will be returned in the same order?
>>
>
> No, AFAIK SeqScans will start from different places if there's someone
> else doing a scan on the table (or has done recently).
>
>
>
>
Not only that, but the physical order of rows in the table is not
constant. It can be changed by any insert/update/delete. If you need the
rows in some order, use ORDER BY.
cheers
andrew
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ron Mayer | 2008-11-11 18:32:35 | Re: Re: Updated interval patches (SQL std output, ISO8601 intervals, and interval rounding) |
| Previous Message | Joshua Tolley | 2008-11-11 18:22:29 | Re: Question about SPI_prepare |