Question

If I have a table as follows:

ID | Name | Category | Parent | URL | LastModified

If two rows have the same Name and parent, then they are not unique. How do I get the unique set of rows in this case (but return more than the columns that make them unique)?

So, for more details:
This is a corporate keywords table, where keywords are organized by category. Each keyword can only have one category. Each keyword can have child keywords, so if parent=0 or NULL, it is a root keyword. If keywords have the same name and parent in a category, then they are not unique (regardless of the other columns). If two keywords have the same name and category, and parent=0 or NULL then they are not unique. If there are duplicates, then I only want the 1st one. The reason is I am putting these into a system that will not allow a keyword to have two children with the same name.

I would also like to see what rows ARE duplicates to see what is causing me the trouble!

Thanks a million so far for excellent responses. I am obviously not a SQL guy… 🙁

Answer 1

This query finds all rows where no other row has the same name and parent. If two rows have parent set to NULL, these rows are not considered to have the same parent.

SELECT T1.*
FROM Table1 T1
LEFT JOIN Table1 T2
ON T1.ID != T2.ID AND T1.Name = T2.Name AND T1.Parent = T2.Parent
WHERE T2.ID IS NULL