I am importing a series of records into a table with Update matching records in the found set turned on and add remaining data as new records also turned on.

All of the records from the source have the same match field id, so in theory it will create the first record because it does not match anything existing, but every subsequent record will match this newly created record in the target and thus overwrite it.

I have tried tests and they all work yet in my working project they generate multiple records.

Any help would be appreciated.

Hi,

I have encountered a this problem before. Usually what I have found to be the problem is a rogue space(s) in the import set, once imported if you click into the key field you can see it. Look at the import set to see if you have this problem.

i.e. "1001" is not equal to "1001 "

Let me know what you find.

HTH

Tim

thanks for the reply.

no i thought the same, but there are no rogue characters. The records can be generated from a duplicate step and yet they still don't match.

I was wondering whether it had something to do with indexing and that it hadn't quite registered that the first record had generated a new record in the target file. I have fiddled with indexing but this hasn't worked either.

I think it imports all records because none of them has a matching record in the found set.

The found set being the found set before starting the import.

If there were matching records in the initial found set, they would get updated by the last matching record in the imported set, and no new records would be created.

I thought along those lines too, however I did a test with a new file and imported some rows of data with the same link field id. From no records in the database it became one record with the last rows data being displayed.

Thanks for your interest.

I suspect this is going to be smoethiong so obvious it has been slapping me in the face and i haven't noticed!!!!

I am afraid I am not following you. If there are no records in the file, you cannot select "Update matching records in the found set". If there are no records in the found set, and "Update matching records in the found set" is selected, no records will be imported - at least that's how it behaves here.

Oh, very interesting your import set has dupilcates in it. i.e. record 1 creates a new record and record 10 contains the same key field value. This should update record 1 but instead adds record 10 also. Just like record 1 was not added to the data set.

I have to agree with your idea about the index and suggest some type of duplicate record script to clean up after your import. But what has changed between your test file and your live file? Very perplexing.

My 2cents,

Tim

Sorry for wasting your time,

I must have been really stupid. You are correct, if there are no existing records in the found set it will not match them.