Very Computer

I am wondering about creating  calculated fields that contain each
repetition of a row separated by returns, double return at the end.
Then exporting to a Word file, find & replace the returns with tabs,
the double returns with singles and then using this revised text file
as an import into the new relational file.

Am I on the path here? Better ideas, or just forget it altogether (hand
enter the data into the relational file)?

Cross platform FM4.1 soon to upgrade to FM5 and Server5.

Looking forward to hearing anyone's thoughts on bringing old repeating
field files into the light of day.

Peter.

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!

You are in luck.  FMPro does this in one step.  Make sure the child DB has
fields for all the values you want to capture, especially that unique
identifier.  Then make sure you don't have a found set in the parent DB at the
moment, go to the child DB and perform an import from the parent DB.  Choose and
match up all the fields you want to populate in the child DB.  Note that the
repeating fields in the parent DB appear only once.  When you have it set up and
choose Import, a dialogue box pops up asking whether you would like to keep the
repeating fields' values in one record or split them into separate records.  I
think your answer here would be "Split them, and thanks very much!"  You are
finished.  If you run into any problems or have other questions, just get back
to the group, but this is a simple process and you will get it right away.

Best of Luck -
James

Thank you very much, I was not aware that FM had this feature. I
suppose Claris anticipated the problem as they developed FM into a
fully relational DB.

One question though, why no found set in the parent before doing the
import?

  Then make sure you don't have a found set in the

Quote:> parent DB at the
> moment, go to the child DB and perform an import from the parent
> DB.

Peter

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!

In FMPro, an import is limited to the found set in the source DB.  If you forget
and only import some records, it can be really difficult to go back and get the
rest later if other records have been added.

Best of Luck -
James

1. I sometimes need to import a record from the same file. Duplicate
Record would copy too much, so I just do an Export of certain fields,
then an import.

2. I export a record consisting of summary data so I can re-import it.
This proved difficult/impossible for me to do without exporting then importing.

PB

Peter, can't you split repeating fields into separate records upon import?

Scott

Quote:> Condsider the challenge of creating a relational DB from a legacy file
> using repeating fielder (ERRR!). What I'm looking at is several
> repeating fields aligned together in a record in a flat file, in a
> sense mimicing a portal. Imagine, say for example three fields (Date,
> Name, Place) each a repeating field of say 10 repetitions each and
> displayed in alignment so that a row appears like a portal record.
> Picture Row_1:DateRep_1, NameRep_1, Place_Rep_1; Row_2: Date_Rep_2,
> NameRep_2, PlaceRep_2, and so on. Now, populate a relational database,
> one row of the repeating fields, including a primary key field unique
> to the record in the flat file in which they reside, to one record in a
> relational file.

> I am wondering about creating  calculated fields that contain each
> repetition of a row separated by returns, double return at the end.
> Then exporting to a Word file, find & replace the returns with tabs,
> the double returns with singles and then using this revised text file
> as an import into the new relational file.

> Am I on the path here? Better ideas, or just forget it altogether (hand
> enter the data into the relational file)?

> Cross platform FM4.1 soon to upgrade to FM5 and Server5.

> Looking forward to hearing anyone's thoughts on bringing old repeating
> field files into the light of day.

> Peter.

> * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network
*
> The fastest and easiest way to search and participate in Usenet - Free!

Actually I intend to import a limited found set then archive all the
others.

I ran a quick and dirty beta test and it works fine, each row of
repeated records become a separate record in the new file. This will
certainly save a lot of work compared to the convoluted approach I was
considering!!

Peter