Re: "no unpinned buffers available" ? why? (hstore and

From: Richard Huxton <dev(at)archonet(dot)com>
To: hubert depesz lubaczewski <depesz(at)gmail(dot)com>
Cc: Postgres General <pgsql-general(at)postgresql(dot)org>
Subject: Re: "no unpinned buffers available" ? why? (hstore and
Date: 2007-01-04 09:18:25
Message-ID: 459CC661.6040605@archonet.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

hubert depesz lubaczewski wrote:
> On 1/3/07, Richard Huxton <dev(at)archonet(dot)com> wrote:
>>
>> hubert depesz lubaczewski wrote:
>> > On 1/3/07, Richard Huxton <dev(at)archonet(dot)com> wrote:
>> >>
>> >> If you do that separately at the start of the process, (one query per
>> >> custom column in the old table) then it becomes straightforward.
>> >>
>> >
>> > no, because meaning of "col1" in advert_custom_fields is different for
>> each
>> > record.
>> > for one record it's codename might be "email" for another record it
>> > might be
>> > "engine size".
>> And is that not what's stored in "v_category_custom_fields"? So you can
>> do the transformation and get (advert_id=1, codename='col1',
>> value='vvv') then use v_category_custom_fields to update the 'col1' part.
>>
>
> this information is stored there, yet i have no clue on how you would like
> to make it with standard sql statements? for every "advert" there are about
> 20-30 "custom fields" (in one record in advert_custom_fields). to do it
> your
> way i would need to make approximatelly 30 (numer of custom field) times
> 300000 (number of adverts) queries. that would be way slower and
> definitelly
> not automatic.

Show me the table definitions and some sample data and I'll see if the
SQL is do-able.

--
Richard Huxton
Archonet Ltd

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Richard Huxton 2007-01-04 09:19:39 Re: Backup Restore
Previous Message hubert depesz lubaczewski 2007-01-04 08:21:04 Re: "no unpinned buffers available" ? why? (hstore and plperl involved)