Question

I am trying to find duplicates by comparing the first name and surname columns in a table. The first name can be a name or an initial.

Reading other posts I have managed to figure out how to get the duplicate surnames and list the first letter for first name. But I am unsure how to only show rows where there is a match of surname and the first letter of the first name.

SELECT * 
FROM table AS a
INNER JOIN (

SELECT LEFT( firstname, 1 ) , surname
FROM table
GROUP BY surname
HAVING COUNT( * ) > 1
) AS b ON a.surname = b.surname

id | firstname | surname
**************************
1  | joe       | bloggs
2  | j         | bloggs
3  | s         | bloggs
4  | f         | doe
5  | frank     | spencer

Currently this query would return

1  | joe       | bloggs
2  | j         | bloggs
3  | s         | bloggs

Result I would like would just contain the possible duplicates.

1  | joe       | bloggs
2  | j         | bloggs

Answer 1

I don’t quite get what you want. Yor provided a query, your current table and the expected result.

I’ve just created your table, run your query and got the expected result. What is wrong with this?

SELECT  FROM table1 AS a
INNER JOIN (
  SELECT surname FROM table1
  GROUP BY surname
  HAVING COUNT(*) > 1
) AS b ON a.surname = b.surname

This effectively result in your expected result:

joe | bloggs
j   | bloggs

Or am I missing something?

After re-reading… are you expecting to get only this?

j | bloggs

If that is the case, use this:

SELECT * FROM table1 AS a
INNER JOIN (
  SELECT surname FROM table1
  GROUP BY surname
  HAVING COUNT(*) > 1
) AS b ON a.surname = b.surname
WHERE CHAR_LENGTH(firstname) = 1

Edit:

After the expected result was properly explained I conclude the query should be:

SELECT a.firstname, a.surname FROM t1 AS a
INNER JOIN (
  SELECT LEFT(firstname, 1) AS firstChar, surname FROM t1
  GROUP BY surname, firstChar
  HAVING COUNT(surname) > 1
) AS b ON a.surname = b.surname AND b.firstChar = LEFT(a.firstname, 1)

Working example