Question

I have a table we’ll call Table1 with a bunch of junk data in it and no unique identifier column.

I want to select some columns from Table1 and transfer the data to Table2. However, I want to prevent duplicate entries from 3 of the columns from being inserted.

Let’s say I have a row with desired columns [FirstName], [LastName], [CompanyName], [City], and [State] to be transferred. I want only the rows with unique combinations of [FirstName], [LastName], and [CompanyName] to be copied to Table2 along with their respective [City] and [State].

Answer 1

I am assuming you have data like the following which you want to scrub:

FirstName  | LastName  | CompanyName  | City       | State
----------------------------------------------------------
Bob        |  Smith    |  ABC         | New York   |  NY
Bob        |  Smith    |  ABC         | Newark     |  NJ
Jane       |  Adams    |  ACME        | Seattle    |  WA 

Therefore I am assuming you mean “their respective city and state” to mean “one randomly-selected city and state from the corresponding rows”, which is what actually makes the problem difficult:

WITH [Names] AS (
    SELECT DISTINCT [FirstName], [LastName], [CompanyName]
    FROM Table1
)
INSERT INTO Table2 ([FirstName], [LastName], [CompanyName], [City], [State])
SELECT [FirstName], [LastName], [CompanyName], [City], [State]
FROM [Names]
CROSS APPLY
(
    SELECT TOP 1 [City], [State]
    FROM Table1
    WHERE Table1.[FirstName] = [Names].[FirstName]
    AND Table1.[LastName] = [Names].[LastName]
    AND Table1.[CompanyName] = [Names].[CompanyName]
) [FirstRespectiveAddress]

If “some random row” does not work for you, you will have to explicitly define which city and state to take for each person.