|
|
Posted by Ed Murphy on 10/05/06 17:18
MVM wrote:
> Could I get a reply without the inaccurate irrational ramblings? I didn't
> think this would be very difficult for a SQL expert.
Intuiting a positive correlation between (1) giving a sketchy problem
description and (2) making common mistakes is hardly irrational - though
of course it will be sometimes inaccurate, and sometimes overkill.
In your particular case, the parts that were (and still are) sketchy are
as follows. #1 is fairly simple, #2 is somewhat more complex.
1) When a row in Table 1 has no matching rows in Table 2, what do you
want your query to give for Acres?
a) 0
b) NULL
2) When a row in Table 1 has two or more matching rows in Table 2, what
do you want your query to give for Acres?
a) Sum of the matching rows' values
b) Exactly one of the matching rows' values:
i) Maximum
ii) Minimum
iii) One is right and all the others are wrong; I want it to
give the right one
iv) The one with the most recent value in a date column, which
I forgot to mention
c) Other, please specify
Also, it seems odd that Table 2 would have /no/ columns other than GID
(Government ID?) and Acres. Wouldn't you want to store a survey date,
or a description field (e.g. "northeast section"), or something?
Navigation:
[Reply to this message]
|