Linking problem in 7.10G

Deskin, Bob Bob.Deskin@Cognos.COM
Mon, 7 Sep 1998 07:48:01 -0400


The SELECT file IF will work for a file in any position in the ACCESS
statement as long the linkage is not OPTIONAL, because then you want the
records even if there's no match.

As for expanding the short to a long, which is the best long-term approach,
ensure that you've caught all uses of the short before you change. Is the
short a pure number or are there embedded meanings (such as codes) in it. Is
it ever embedded in another item? Is it ever converted to character. And so
on. With one item, it should not be hard to track this sort of thing down.
There are many editors that can do a search for one pattern in many files.

Bob Deskin              
Senior Product Advisor  bob.deskin@cognos.com
Cognos Inc.             (613) 738-1338 ext 4205 FAX: (613) 228-3149
3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA


> ----------
> From: 	alex ling[SMTP:linga88@yahoo.com]
> Sent: 	September 7, 1998 3:13 AM
> To: 	Deskin, Bob; powerh-l@lists.swau.edu
> Subject: 	RE: Linking problem in 7.10G
> 
> Thanks for the two-pass/subfile workaround suggestion for an OPTIONAL
> link. It would be nice and easy if the "LONG-ITEM to SHORT-ITEM"
> linkage is the first linkage in the ACCESS statement like in your
> example. What if the linkage is the some 10th linkage in the ACCESS
> statement? What if there are more than one such linkage in one ACCESS
> statement? This will give us nightmare in handing it with a subfile in
> the two-pass/subfile approach. Then there is performance issue to
> consider. With the number of QTP/QUIZ programs we have here it is
> going to be hell for us.
> 
> Bob, we prefer to change the SHORT-ITEM to LONG-ITEM so that the
> linkage will become "LONG-ITEM to LONG-ITEM". What do you think? Any
> advice on the implication as a result of the change from a 2-byte to a
> 4-byte?   
> 
> Thanks,
> A Ling.
> 
> 
> ---"Deskin, Bob" <Bob.Deskin@Cognos.COM> wrote:
> >
> > I don't know if this has been suggested as a workaround, but Allison
> > suggested using a SELECT file IF so select out values that are too
> high. In
> > other words:
> > 
> > access A link LONG-ITEM to SHORT-ITEM of B
> > select A if long-item < 32768
> > 
> > If the test fails, the link won't even be attempted.
> > 
> > If LONG-ITEM is an index/segment, you could also use a CHOOSE. If
> you needed
> > an OPTIONAL link such that you want A even if the link won't work,
> then you
> > need two passes. Do one like the above and write the results to a
> subfile.
> > Then do another pass reversing the select and without the LINK TO
> like this:
> > 
> > access A
> > select A if long-item > 32767
> > 
> > and append the results to the subfile with appropriate defined default
> > values for the missing items. Then access the subfile and report or
> > whatever.
> > 
> > Bob Deskin              
> > Senior Product Advisor  bob.deskin@cognos.com
> > Cognos Inc.             (613) 738-1338 ext 4205 FAX: (613) 228-3149
> > 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA
> > 
> _________________________________________________________
> DO YOU YAHOO!?
> Get your free @yahoo.com address at http://mail.yahoo.com
> 
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
This list is closed, thus to post to the list, you must be a subscriber.