Re: Cross-ref IDs and de-dupe
- From: Tim Ferguson <FergusonTG@xxxxxxxxxxxx>
- Date: Mon, 15 Jan 2007 14:08:53 -0800
=?Utf-8?B?UGVuZHJhZ29u?= <Pendragon@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in
news:FEF14E7B-5ACF-49F0-9B77-5570EAD99390@xxxxxxxxxxxxx:
Label #1:
John Smith & Jane Smith
123 First Street
etc.
Label #2:
Jane Smith & John Smith
123 First Street
etc.
The classic solution would be to have a new table of Labels; each Member
record (actually, shouldn't that be Registrations? Still I digress...)
would have a FK field pointing at the Labels. Most Label records would
have only one Member record pointing at each of them, but some would have
two or more. The UI design is left as a problem for the reader..!
The "quick and dirty" solution would be to have a FK field in the Member
record called UsesLabelFor pointing at another Member record. In this
case, you would have to decide which of John or Jane is first on the
label, and his or her record would have UsesLabelFor set to NULL, while
the other record would contain the other's PersonID value.
The SQL for the first one is not easy: you do need some procedural logic
(i.e. VBA) to make the ampersand-separated list but at least it's easy to
get the database to watch out for illegal values. The problem with the
second version is that if you fill in UsesLabelFor fields for both
partners (or create a chain) then you will likely lose labels altogether
and the database engine cannot come to your rescue.
Best of luck
Tim F
.
- Prev by Date: Decimal precision change - MySQL & MS-Access
- Next by Date: Re: Scheduled Tasks
- Previous by thread: Decimal precision change - MySQL & MS-Access
- Next by thread: Re: Cross-ref IDs and de-dupe
- Index(es):
Relevant Pages
|
