From: | Richard Huxton <dev(at)archonet(dot)com> |
---|---|
To: | Clodoaldo <clodoaldo(dot)pinto(dot)neto(at)gmail(dot)com> |
Cc: | Brent Friedman <bfriedman(at)scanonline(dot)com>, Martijn van Oosterhout <kleptog(at)svana(dot)org>, PostgreSQL - General ML <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Insert vs Update syntax |
Date: | 2008-02-29 19:04:17 |
Message-ID: | 47C85731.4090004@archonet.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Clodoaldo wrote:
> 2008/2/29, Brent Friedman <bfriedman(at)scanonline(dot)com>:
>> If you don't like the standard sql implementation, you could use plsql
>> or any language to make an abstraction layer/wrapper for this
>> functionality. Just pass everything as a key/value pair, in an array or
>> hashtable structure, to your abstraction layer/wrapper, and it can cycle
>> through the data structure to do the insert or update for you.
>
> Ok. I have been playing these tricks for the last 25 years. Please no
> more tricks.
Tricks?
> If someone knows something about the reason for the
> insert syntax in instead of the Update syntax then please elaborate on
> it.
I believe it's because of the way it reads in English: "INSERT INTO
<some columns> <these values>". One of the key benefits of SQL was
supposed to be its ability for non-programmers to use it. As a result,
they sacrificed consistency for the sake of ease-of-learning.
Of course, once you start writing queries of any complexity, you lose
the benefits of reading as English.
If you come across a time machine, pop back to the first standards
meeting and have a word with them, would you?
--
Richard Huxton
Archonet Ltd
From | Date | Subject | |
---|---|---|---|
Next Message | Chris Bowlby | 2008-02-29 19:07:11 | Re: issue with an assembled date field |
Previous Message | brian | 2008-02-29 18:53:18 | Re: issue with an assembled date field |