Re: VIEWS compared to Nodes as Windows into data
Date: Mon, 3 May 2004 19:52:26 +0300
Message-ID: <40967812$1_at_post.usenet.com>
"Dawn M. Wolthuis" <dwolt_at_tincat-group.com> wrote in message
news:c75rev$tdq$1_at_news.netins.net...
> Yes -- good --that might be the question I should ask -- if one creates a
> view that joins in table G and then Pat queries against this view and
> includes no columns from G, does the join get executed? I would think
that
> under certain conditions the join would be required, such as in the case
of
> an inner join where there would be more rows if the join were not
performed.
What do you mean by "performing the join" or "the join get executed" ? :-) When Pat query against G, she gets what she was asking for .
What atomic operations would be "executed" by the DBMS in the two scenarios/DBMSs you have in mind ?
Why there would be more operations for a RDBMS than for any other DBMS ? As I said "Relational" is just a language.
> But do current RDBMS's figure out when they don't need to include the join
> bz they know (how?) that the join will not result in a different row set
and
> there are no attributes from table G in the query against the view?
There is a current RDBMS ? :-)
If Pat asked for the join, an RDBMS would need to consider it.
If Pat haven't asked for a field from table G, then an important
part of the join may be skipped.
Sometime the other part of the join can be skipped too. ;-)
It is called query optimization.
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
- Usenet.com - The #1 Usenet Newsgroup Service on The Planet! *** http://www.usenet.com Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=