Re: self referencing tables/ nested sets etc...

From: Manfred Koizar <mkoi-pg(at)aon(dot)at>
To: Rob Hoopman <rob(at)tuna(dot)nl>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: self referencing tables/ nested sets etc...
Date: 2004-03-25 21:01:30
Message-ID: 5ih660tq1i1ovtsiacg40kus9gu77ud3rb@email.aon.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Thu, 25 Mar 2004 20:56:35 +0100, Rob Hoopman <rob(at)tuna(dot)nl> wrote:
>> > It appears that when adding more than 48 sub nodes
>> >to any node in the tree, craps out because of an INT8 column overflowing.
>>
>> AFAICS it doesn't depend on the number of siblings, but it fails when
>> the sum of the numbers in dotted path notation exceeds 62.
>>
>Maybe, but some of the intermediate steps are larger than the number that gets
>stored in the end. Actually that's where this implementation broke for me.

Rob, do you still have the functions and the data that led to the
overflow? If so, would you care to locate the parent of the node you
failed to insert and this parent's last child. Then please post the
output of

SELECT pk, numer, denom, path(numer, denom)
FROM yourtable
WHERE pk = 'parentpk' OR pk = 'childpk';

I'd like to find out whether OMPM is flawed or my theory about it.

Thanks.
Servus
Manfred

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Neil Conway 2004-03-25 21:09:07 Re: subversion vs cvs (Was: Re: linked list rewrite)
Previous Message Thomas Swan 2004-03-25 21:01:05 Re: subversion vs cvs (Was: Re: linked list rewrite)