Results 1 to 2 of 2
  1. #1
    New Lounger
    Join Date
    Aug 2008
    Thanked 0 Times in 0 Posts

    Access (Not sure)

    I am doing work for someone who is in the process of converting Excel data into an Access database. In the original database each contact had multiple entries with varying information because the account managers had all entered their info separately. In the first round there were 8 of us working around the clock to consolidate all the info in one record. The data in the last batch of duplicates was corrupt but it took me a while to realize it. The only thing I could figure out that a sort had gone wrong in the original Excel data. Here is an example:

    John McNamara - Email says: Jmcmanus@xxxxx Tim McPhail - Email says: Tmcnamara@xxxxx Steven McPhee - Email says: smcphail@xxxxxxx

    Some of these records were added to the master database before I realized there was a problem. The new master database contains 50,000 records. Is there any way to identify the corrupt data without having to comb through all these records. Is there a way in Access where the systems administrator could pull data for the dates I worked on it? I would very much appreciate any input. The IT person doing the conversion does not seem that familiar with Access and, being a free lancer for the firm, I do not want to comment. As far as Access goes, I think the old dbase was far better.


  2. #2
    Super Moderator
    Join Date
    Jun 2002
    Mt Macedon, Victoria, Australia
    Thanked 45 Times in 44 Posts

    Re: Access (Not sure)

    Access does not automatically timestamp additions/changes to records, so it is unlikely you can find the records that were added in a particular batch unless a timestamp field has been added to the table.
    If the table has an autonumber field then that number would be a clue to the order in which records were added.

    Is there any consistent pattern to the way email addresses should be constructed? If they are just a collection of addresses for customers, for example, there would not be a pattern, but if they were all addresses internal to an organisation, they may have a rule about how addresses are made from people's names.
    If that was the case you use a query to build a calculated field showing what a person's address should be, and compare that to what is listed. The problem records are those where the two are not equal.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts