From: | "Joe Conway" <joseph(dot)conway(at)home(dot)com> |
---|---|
To: | "Command Prompt, Inc(dot)" <pgsql-general(at)commandprompt(dot)com>, <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Practical Cursors |
Date: | 2001-09-17 21:56:42 |
Message-ID: | 021b01c13fc3$a3a29b80$6cdc10ac@jecw2k1 |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
> However, using something like PHP will not allow this because HTTP is
> stateless and PostgreSQL will not know from one transaction to the next
> that the results of the connection are related.
>
> Is this truly the case, or is there a way for PostgreSQL to remember the
> connection identifier so that the next time a PHP connection is made with
> the same identifier a transaction can be completed?
>
> Sincerely,
>
> Joshua Drake
>
I have actually been thinking recently about just this question. I looked
through the source for the PHP pgsql extension and it is clear that no
cursor is used. There is, however, the ability to open a persistent
connection to PostgreSQL from PHP. I think it would be possible to create a
persistent resource identifier to a cursor in a similar manner, and "reuse"
the cursor across multiple HTTP requests.
The downside I can see is that PHP would have no way to know when it could
garbage-collect the allocated resource. So while it could be done, what
would happen when the user closes their browser with your cursor still open?
-- Joe
From | Date | Subject | |
---|---|---|---|
Next Message | Doug McNaught | 2001-09-17 21:57:16 | Re: Left Joins |
Previous Message | Brett Schwarz | 2001-09-17 21:55:07 | RRules using existing data |