Question

I have made a statement that used group by. Now, I need to know the details for every record but I need to perform this at one statement.

For example: I made the following query to query about the files that ends with _1 and share the same number:

SELECT number, COUNT(*) AS sum domains
FROM table
WHERE file LIKE ‘%_1’
GROUP BY number HAVING sum domains > 1

So, if I have the following table:

domain      |   file      | Number 
------------------------------------     
aaa.com     | aaa.com_1   | 111
bbb.com     | bbb.com_1   | 222
ccc.com     | ccc.com_2   | 111
ddd.com     | ddd.com_1   | 222
eee.com     | eee.com_1   | 333
qqq.com     | qqq.com_1   | 333

The result of the query is (the number that is shared by more than file and the count of the file(s) that ends with _1 and shared this number):

number | sum domains
------------------------
222    | 2
333    | 2

What I need to do is to print out the file names. I need:

number | file
------------------------
222    | bbb.com_1 
222    | ddd.com_1
333    | eee.com_1
333    | qqq.com_1

How can I do this since group by clause does not allow me to print the file(s) ?

Answer 1

You can JOIN your query back against the main table as a subquery, to get the original rows and filenames:

SELECT 
  main.number, 
  main.file
FROM 
  table AS main
  /* Joined against your query as a derived table */
  INNER JOIN (
    SELECT number, COUNT(*) AS sum domains 
    FROM table
    WHERE RIGHT(file, 2) = '_1' 
    GROUP BY number 
    HAVING sum domains > 1
    /* Matching `number` against the main table, and limiting to rows with _1 */
  ) as subq ON main.number = subq.number AND RIGHT(main.file, 2) = '_1'

http://sqlfiddle.com/#!2/cb05b/6

Note that I have replaced your LIKE '%_1' with RIGHT(file, 2) = '_1'. Hard to tell which will be faster without a benchmark though.