My workspace is using ChangeDetector with two data sets original and revised. It appears that if the revised records contain two records with the same key and selected attribute and this key does not already exist in the original table then ChangeDetector will want to insert both records in the original table. This happens when the key is not already in the original table before the run. On subsequent runs, ChangeDetector finds the key and selected attribute in the original table, and it behaves as expected and does not insert a record. It treats it as an update or unchanged. How should I account for this? Attached is an image of my workspace. Thank you.