From: Peter Brawley
Date: July 19 2010 5:41pm
Subject: Re: [MySQL] Re: MySQL select matching
List-Archive: http://lists.mysql.com/mysql/222265
Message-Id: <4C448E4E.3000203@earthlink.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

  Ashley,

 > I can't go back to the client and tell them their schema
 > is really messed up and to store their data differently.

You can hope that's not not necessary in order to deliver the requested 
query, but it's a bad mistake to rule it out altogether, since it often 
happens that accomplishing the requested query efficiently requires DDL 
changes.

I suggest too that you're more likely to get good help if you describe 
the problem with a set of Create Table and Insert statements than with 
truncated results of Describe.

PB

-----

On 7/19/2010 12:29 PM, Ashley M. Kirchner wrote:
> On 7/19/2010 11:08 AM, Shawn Green (MySQL) wrote:
>> Just because someone hands you a set of data to *start* with, does 
>> not mean that you must only use that data to *work* with.
>>
>> You should be able create additional tables derived from the original 
>> data and work with those as part of your analysis project. No 
>> modification of the original *tables* will be required.
>>
>> But this would indeed be much easier to talk about if your columns 
>> had names.
>>
>
>     And that's okay, having to create temporary tables to work with.  
> What I meant was, I can't go back to the client and tell them their 
> schema is really messed up and to store their data differently.  I 
> know I can create additional tables to work with, and ultimately that 
> may indeed happen.
>
>     A
>
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.5.441 / Virus Database: 271.1.1/3015 - Release Date: 07/19/10 06:36:00
>