Question

I am using SQL Server Management Studio 2012. I have a table with thousands of rows of data in it. Many of the rows are duplicates, and I need to remove them. Each row has a unique identifier [OwnerID] which is set to Identity Specification with an Identity Increment of 1. For each row, the duplicates lie in these columns: [FirstName], [LastName], and [CompanyName].

So I need to delete rows with duplicate combinations of values in these 3 columns. After the deletion, is there t-sql I can write to reset the Identity Specification in
[OwnerID] to start at 1 for the first row and assign values to the rest of the rows in increments of 1?

Thanks for any assistance.

Answer 1

Here is my try using the temporary tables SQL Fiddle:

SELECT 
firstName, LastName, CompanyName, COUNT(*) thecount, MIN(ID) min_id
INTO #temp
FROM tab
GROUP BY firstName, LastName, CompanyName;

SELECT 
a.id ID, b.min_id
INTO #temp1
FROM tab a, #temp b
WHERE  a.firstName  = b.FirstName 
AND a.LastName    = b.LastName
AND a.CompanyName = b.CompanyName
AND b.thecount > 1;

-- run this query on all referenced tables:
UPDATE tab2 SET tab2.ID = t.min_id
FROM tab2, #temp1 t
WHERE tab2.ID = t.ID;

DELETE t
FROM tab t, #temp1 a
WHERE t.id = a.id and a.id <> a.min_id;