Re: A PGsendQuery API design for your review

From: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>
To: "Karl O(dot) Pinc" <kop(at)meme(dot)com>, Federico Di Gregorii <fog(at)dndg(dot)it>
Cc: psycopg(at)postgresql(dot)org
Subject: Re: A PGsendQuery API design for your review
Date: 2016-02-02 21:24:24
Message-ID: 56B11E88.1070404@aklaver.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: psycopg

On 02/02/2016 12:10 PM, Karl O. Pinc wrote:
> On Tue, 2 Feb 2016 14:04:08 -0600
> "Karl O. Pinc" <kop(at)meme(dot)com> wrote:
>
>> On Tue, 02 Feb 2016 19:58:16 +0100
>> Federico Di Gregorii <fog(at)dndg(dot)it> wrote:
>>
>>> I can say with some confidence that we'll never implement such an
>>> API. Even the
>>> ux, i.e., accepting multi-queries from the user is wrong, imho.
>
>> Seriously, the real world is complex.
>
> (Reminds me of an interaction I had with the Mozilla people.
> They'd changed Firefox so that a response had to show up
> within 5 minutes. My folks occasionally write queries
> that run hours. Mozilla changed their mind. Hope you will
> too.)

Not sure that was a good idea, holding a resource open for hours to do
nothing seems counter-productive to me. That is what LISTEN/NOTIFY,
Websockets, name your async method, are for.

>
> Regards,
>
> Karl <kop(at)meme(dot)com>
> Free Software: "You don't pay back, you pay forward."
> -- Robert A. Heinlein
>
>

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com

In response to

Responses

Browse psycopg by date

  From Date Subject
Next Message Karl O. Pinc 2016-02-02 21:30:03 Re: A PGsendQuery API design for your review
Previous Message Adrian Klaver 2016-02-02 21:20:34 Re: A PGsendQuery API design for your review