Re: VARCHAR length

Hi David,

That's because you didn't have Howard or myself teaching you ;)

P11-12 of the Oracle9i Database Administration Fundamentals I course discusses the structure of a row.

And Howard of course (only a little pun intended) is spot on.

Cheers

Richard
"David Sharples" <david.sharples3_at_ntlworld.com> wrote in message news:KxJo9.2188$QY.17594_at_newsfep3-gui.server.ntli.net...
>
> "Karsten Farell" <kfarrell_at_medimpact.com> wrote in message
> news:eaHo9.1596$i17.195196079_at_newssvr13.news.prodigy.com...
> > Howard J. Rogers wrote:
> > > I hope you're not claiming that CHAR datatypes don't use length bytes?
> > >
> > > They do.
> > >
> > > Regards
> > > HJR
> >
> > Correct me if I'm wrong, but it's been quite a while since I attended an
> > oracle architecture class. As you probably know, training is one of the
> > first things to go in a shrinking budget ... and new DBAs get trained
> > before old-timers - as it should be, I suppose, since they have more to
> > learn. So I haven't heard the lecture since Oracle7 first came out.
> > Something tells me Oracle just might have done a little work on their db
> > since then, eh?
> >
> > I thought the length of a CHAR field was stored in the data dictionary
> > (since it never changes) and the length of a VARCHAR2 field was stored
> > in the database row (since it potentially changes). Therefore, the
> > starting byte position of each CHAR field can be determined at SQL parse
> > time (if there are no variable-length fields in front of it); but the
> > starting position of VARCHAR2 fields cannot be determined until the row
> > is read into the buffer.
> >
>
> Having just attended an Oracle architecture class, that was definitely not
> covered.
>
>
Received on Wed Oct 09 2002 - 07:28:49 CDT