On 9 Aug 2001, at 18:28, Stanley Loo wrote:
"Because that's what the customer wants" is probably the best
reason, even if the logic escapes mere mortals <g>.
I didn't have a data set at hand with dups so didn't test it. Hope it
works for you (Albert generally doesn't miss, so it's a good bet)
Ben Petersen
> Our client needs strong evidence of problems in a system that needs
> revamping. They want a list of present problems, and duplicate records are
> just one. Regarding duplicates, they have a choice between knowing which
> records have duplicates and how many, and listing all duplicates. Senior
> management wants all duplicates listed. I suspect they are looking for a
> cannon. What we plan to do is to give them both, as a pleasant surprise.
>
> By the way, it is a mainframe Adabas system from Software AG. For reasons
> of efficiency, I am using Rbase to look for problems in the key data
> downloaded from the mainframe.
>
> Ben, hope your curiosity is met. Also thank you for your idea. BTW, did
> you test
> it out with real data?
>
> Stan Loo
>
>
> ----- Original Message -----
> From: "Ben Petersen" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Wednesday, August 08, 2001 6:17 PM
> Subject: RE: Capturing Duplicates
>
>
> > I've been watching this thread and wondering what someone would
> > do with a bunch of duplicate records, so I'll add my 2 cents...
> >
> > Del dup fro tblName
> >
> > ---- or ----
> >
> > (using some of that very nice code from Albert earlier)
> >
> > Create view (col1,Col2,col3) as +
> > SELECT Col1,Col2,Col3 +
> > FROM OriginalTable t1, +
> > (SELECT Col1,Col2,Col3 FROM Original table +
> > group by Col1,col2,col3 having count(*) > 1) as t2 +
> > WHERE t2.Col1 = t1.Col1 AND +
> > t2.Col2 = t1.Col2 AND t2.Col3 = t1.Col3
> >
> > With a view the dups are always there to check... or if the data is
> > clean you'll know it because the view won't return anything. Or, just
> > use the select statement from Albert's code and put it on the
> > screen w/o creating tables/views.
> >
> > Of course, it's getting late and I may well be missing something
> > <g>.
> >
> > Ben Petersen
> >
> >
> > >
> > > >===== Original Message From [EMAIL PROTECTED] =====
> > > >Now that you have the #'s of each duplicates, they can be expanded to
> another
> > > >table to exactly represent the original duplicates using declare cursor
> and
> > > an
> > > >IF...ENDIF loop.
> > > >RRR
> > > >
> > > >> suredata wrote:
> > > >>
> > > >> Thank you Bill, Ron, Dennis, Phil, and Albert for responding.
> Apparently I
> > > >> did not make myself clear enough as to what I want.
> > > >>
> > > >> I need to save all duplicate rows to a new table or file, not just
> knowing
> > > >> which rows in the original table have one or more duplicates. In
> other
> > > >> words, if a row in the original table has 5 duplicates (or 6
> identical rows
> > > >> in the table), I want to save these five duplicates as separate rows
> in a
> > > new
> > > >> table. This new table would contain nothing but all duplicates in
> the
> > > >> original table. Is there an easy way to achieve this?
> > > >>
> > > >> I hope this is clearer. Thanks again
> > > >>
> > > >> Stan Loo
> > >
> >
> >
>
>
