Re: Ordering dependency problem

From: Nis Jorgensen <nis_at_dkik.dk>
Date: Fri, 18 Jan 2002 10:38:12 +0100
Message-ID: <u1qf4usk8jg6vvip2o2ahv7q4c91ka84e3_at_4ax.com>


On 17 Jan 2002 20:16:12 -0800, zhouqingqing_at_excite.com (Qingqing) wrote:

>> I think this is the answer you wanted. Butif you wanted to scale donw
>> everyone's salary to be less than any of their superiors, you will
>> have to work harder.
>
>Hi, --CELKO--, thanks for your careful answer. Seems it is a good way
>to solve this problem. We all agree that this problem is because the
>table generated by DDL is not formalized. I try to find out a simplier
>way to handle this problem - then I give formalized tables, but the
>problem seems still exists:
[cut]
>So I wonder does this means only the tree-decomposed algorithms in
>your solution works? Or there is something wrong in my method?

With your examples for implementation of the tree (which are both isomorphic to adjacency lists), you are not even able to get a list of all the superiors of an employee in one query, without using the WITH construct (for which support is very poor).

I think it should actually be possible to write your query using a WITH in a subquery - but you have to be very careful in defining the criteria and the desired result.

-- 
Nis Jorgensen
Amsterdam

Please include only relevant quotes, and reply below the quoted text. Thanks
Received on Fri Jan 18 2002 - 10:38:12 CET

Original text of this message