Best way to split one table into 2 tables thru a form

[Steve in MN]

This is like the reverse of a form/subform set up, where i am importing data
from a mainframe that is actually a text file. I import it and manipulate it
into one table thru my import process. The user would work thru a form to
determine which lines actually go together to make one load. Then a new
record would be created in the "One" table (as in one to many) and the
existing table would be the "many" table in the relationship.
Is there a resource out there that might have an example of this so that i
could look at to get some ideas?
This might seem archaic, but that is the way the older system runs and the
routes are put together manually after the fact and i need to design a
reverse form to be able to select rows and then add a line to the main table
for each pair in the many. Mainly because they dont pair them in the
mainframe, they route individually and them pair them manually after the fact.

any help or ideas would be appreciated.

 

[Mike Painter]

This is like the reverse of a form/subform set up, where i am
importing data from a mainframe that is actually a text file. I
import it and manipulate it into one table thru my import process.
The user would work thru a form to determine which lines actually go
together to make one load. Then a new record would be created in the
"One" table (as in one to many) and the existing table would be the
"many" table in the relationship.
Is there a resource out there that might have an example of this so
that i could look at to get some ideas?
This might seem archaic, but that is the way the older system runs
and the routes are put together manually after the fact and i need to
design a reverse form to be able to select rows and then add a line
to the main table for each pair in the many. Mainly because they
dont pair them in the mainframe, they route individually and them
pair them manually after the fact.

any help or ideas would be appreciated.

If it's always "each pair in the many" then you could open the dataset and
either concatenate them or mark them with duplicate keys.

If it has to be done manually the same method might be used. The user would
have to mark the records in some way, then push a button to concatenate and
remove the originals.

If you added the duplicates keys to the new list as a non duplicate you
could guard against the user using the same values on different sets.

 

[Steve in MN]

If it's always "each pair in the many" then you could open the dataset and
either concatenate them or mark them with duplicate keys.

If it has to be done manually the same method might be used. The user would
have to mark the records in some way, then push a button to concatenate and
remove the originals.

If you added the duplicates keys to the new list as a non duplicate you
could guard against the user using the same values on different sets.


So If i do this from the temp table (where the data first gets imported to) I could set up 2 extra fields per row. 1 used to designate which records get moved to the 1 table and then use matching numbers to designate which records get concatenated together for the "many" table. Then when the "one" table is sent an auto number could be created for each record and then when i move the many over, I can use the auto number to match in the "many".

Sounds like it will work, now i need to build the form around it.