Re: Very long "WHERE" list.

From: Daniel Morgan <damorgan_at_x.washington.edu>
Date: Fri, 09 Jul 2004 17:04:58 -0700
Message-ID: <1089417913.117742@yasure>

Mikito Harakiri wrote:
> "Daniel Morgan" <damorgan_at_x.washington.edu> wrote in message
> news:1089413739.39227_at_yasure...
>

>>Mikito Harakiri wrote:
>>
>>
>>>wizofoz2k_at_yahoo.com.au (Noons) wrote in message

>
> news:<73e20c6c.0407082252.14e6fd5c_at_posting.google.com>...
>

>>>>>Unfortunately Oracle can not process this request directly.
>>>>
>>>>Correction: NO DATABASE whatsoever can process that query,
>>>>not just Oracle.
>>>
>>>
>>>You have to make a little research before dumping blant statements.
>>>What fundamental limitation on the implementation side would make
>>>processing such query impossible?
>>
>>Whether the statement is "impossible" it is ludicrous and a strong
>>indication of an incredbily bad design.

>
>
> No, it is not a job of a server to tell the client that his SQL query is
> stupid. Oracle never raizes an exception when a user submits a statement
> that has 20 tables with no join condition (Cartesian Product). It humbly
> tries to execute the statement (and fails at runtime).
>
> Handling long in-list is not something terribly difficult to implement. And
> it would save user some time when he naturally would invent gimmics like
> splitting long list into smaller pieces and concateneted predicates with OR
> condition.
>
> In general, ad-hock limitations like "the list can't be bigger than 1000
> elements" (which oracle generously spiced its implementation) look
> ridiculous.

When you get into a car do you expect the car to tell you that you are too drunk to drive?

If you walked into a hospital surgical theatre would you expect the scalpel to tell you that you are wholly unqualified to be a surgeon?

Why is it the databases responsibility to tell you that you don't understand how to design and implement a relational design?

Why is it Oracle's responsibility to tell you that you should be flipping burgers not bytes?

Daniel Morgan Received on Fri Jul 09 2004 - 19:04:58 CDT