>Hi,
>I used to see couples of thread that asking about how to delete the duplicated record.
>I just would like to know, why duplicated record will occured? Does PK or candicate key take care of the uniqueness, doesn't it?
>Or, any other case that will cause duplicated record?
>
>Sorry for my innocent.
>
>Thank you.

A primary or candidate index is a very good way to avoid duplicates in the first place. You should usually not get duplicates if you have the indices. In rare occasions, you might still get duplicate records - perhaps due to index corruption.

You will still have to eliminate duplicate records created before you started using a candidate index. In fact, VFP won't even allow you to create a candidate index if there are duplicates, so you have to delete the duplicates (or change some values) before indexing.