Specific duplicate selection in MySQL

Question

I did take a look in and outside of SO and still don't know if this can be done. I have a table that looks like this:

User ID | Role | First Name | Last Name | Email |<br>
0001    | K    | John       | Smith     | e@e.co|<br>
0002    | Q    | Jane       | Dickens   | q@q.co|<br>
0003    | K    | John       | Smith     | e@e.co|<br>
0004    | J    | Jack       | Paper     | j@j.co|<br>

As you can see, the table contains a duplicate due to a user entering their information two separate times. I want to display the rows that

1. have the same first name

2. have the same last name

3. have the same email

4. do NOT have the same User ID

I can get the first three conditions to work with an inner join subquery, but I get 0 returned results when ever I try to to add in the fourth condition.

Thanks in advance for your help!

Answer 1

SELECT GROUP_CONCAT(`User ID`) as IDs, Role, `First Name`, `Last Name`, Email
FROM users_table
GROUP BY Role,`First Name`,`Last Name`,Email

Will give a table like

IDs       | Role | First Name | Last Name | Email |
0001,0003 |   K  |    John    |   Smith   | e@e.co|
0002      |   Q  |    Jane    |  Dickens  | q@q.co|
0004      |   J  |    Jack    |   Paper   | j@j.co|

The trick is to GROUP BY everything except ID.

You could also do:

SELECT COUNT(`User ID`) as numIDs, GROUP_CONCAT(`User ID`) as IDs, 
       Role, `First Name`, `Last Name`, Email
FROM users_table
GROUP BY Role,`First Name`,`Last Name`,Email
HAVING numIDs > 1

to get

numIDs |IDs       | Role | First Name | Last Name | Email |
    2  |0001,0003 |   K  |    John    |   Smith   | e@e.co|

Anyhow, you get the point of how to vary it to your purposes.

Answer 2

Try something like:

select *
  from tb_users
 where (first_name, last_name, email) in
   (select first_name, last_name, email
      from tb_users
     group by first_name, last_name, email
    having count(*) > 1)

Answer 3

I am assuming that your table does not contain true duplicate rows (where the User ID also matches). I think it should be as simple as this to get the relevant rows back (with my adjusted column and table naming scheme):

SELECT ui.user_id, ui.role, ui.first_name, ui.last_name, ui.email
FROM user_info AS ui
  INNER JOIN user_info AS udi
    ON ui.first_name = udi.first_name
      AND ui.last_name = udi.last_name
      AND ui.email = udi.email
      AND ui.user_id != udi.user_id;

Answer 4

I would use count(*) and group by clause.

Like this SELECT count(*) as count FROM table group by concat(firstname,'\n',last_name) having count=1;

Answer 5

I think you want to delete duplicate rows. Only take count by group by and delete duplicate rows.

select * from tb_users
     group by first_name, last_name, email
    having count(*) > 1