Query to Identify Duplicate Families
I have been working on a project to audit my Parents fact for each person to be sure that it is in sync with the actual ChildTable and FamilyTable structure. In the process of working on the project, I identified one duplicate FamilyTable entry. I'm going to merge the duplicate FamilyTable entry out of existence by unlinking and relinking in the RM user interface. I don't think it merits trying to write an UPDATE script to do the merge. There is only one bad entry and I can fix it more easily and more reliably from RM than from a script.
The script does not try to identify all the different ways that families can be linked incorrectly. Rather, it identifies one particular way in which families can be linked incorrectly. As the comments in the script indicate, it's difficult and possibly impossible create this kind of duplicate FamilyTable entry directly from the RM user interface. But it apparently is possible for this to happen via GEDCOM import and subsequent merges.
Uploaded files:
Thanks Jerry.
In my tree, your script found 3 families where the parents were linked twice.
These families were not imported by gedcom but were transferred by either ancestry or FamilySearch.