Re: Bug(?) with cursors using aggregate functions. - Mailing list pgsql-general

From Dennis Gearon
Subject Re: Bug(?) with cursors using aggregate functions.
Date
Msg-id 3EAFEBF2.1020802@cvc.net
Whole thread Raw
In response to Re: Bug(?) with cursors using aggregate functions.  (Benjamin Scherrey <scherrey@proteus-tech.com>)
List pgsql-general
They cache all the results in the file system. I assume that there's a different command for moving the curosor, (a
ADODBobject method), and that it goes to the filesystem to fetch the results. Not really that much different than what
adatabase does with curors except maybe the db keeps it in memory. ADODB transparently uses native cursors if they are
available.So it's meant to take you to Oracle, Sybase, DB2, and others as your project grous. 

Benjamin Scherrey wrote:
> Actually I'm using pyPgSQL. How does ASODB do this? Does it know some lower-level calls to
> determine absolute position or just it just cache ALL results from the query up front? If the former,
> then I'd like to know how they do it and perhaps I can implement this for pyPgSQL.
>
>     thanx & later,
>
>         Ben Scherrey
>
> 4/29/2003 11:04:47 AM, Dennis Gearon <gearond@cvc.net> wrote:
>
>
>>If you are using PHP, the ADODB library will do caching of results on the server file system and
>
> allow you to 'cursor' through it. Of course, if you need requerying, you'd have to trigger that and pay
> the time penalty.
>
>
>
>


pgsql-general by date:

Previous
From: "Robert Fitzpatrick"
Date:
Subject: Function for field arrays
Next
From: Dennis Gearon
Date:
Subject: Re: ERROR: ExecutePlan: (junk) `ctid' is NULL!