Re: outer join question

"matt lewis" <matt_at_lewis.co.uk> wrote in message news:fa4cq0$g2c$1_at_registered.motzarella.org...
> Hi, I'm studying a book which is full of errors and the errata seems
> sketchy.
>
> Is this statement correct?
>
> "In outer join statements on tables A and B, you specify the left outer
> join when you want all of table B's rows, even when no corresponding
> record exists in table A"
>
> I had this down as correct but the answer in the book has it as wrong.
>
> I'm pulling my hair out here!
>
> Is it right or wrong please?

As an informal description of outer joins generally then it's correct. I guess what the author meant was that A is written before B and that therefore the statement is supposed to describe a RIGHT outer join rather than LEFT. In other words table B is on the "preserved" side of the join ("all of table B's rows").

I wouldn't worry too much about it. The distinction between left and right outer joins is an almost trivial matter of syntax. i.e. with all other things being equal:

A LEFT OUTER JOIN B means the same as

B RIGHT OUTER JOIN A If it helps, just imagine that all examples of "non-symetric" joins are LEFT outer joins.

-- 
David Portas