Azkaban

When you are aggregating data, for example:



Will give you a count of every row with the same name. But let's say you want all usernames that use the same name.

For example, if 5 users are named John Smith, and you have usernames like johnsmith,johnsmith2,js,etc. and you want a count of all names with multiple users, the above works. The record will produce

name | count
============
John Smith | 5
Jeff Green | 3

But if you want the rows transposed, is it doable? The output should look like

name | count
=============
John Smith | 5 | johnsmith,johnsmith1,js1,js2,jsmith
Jeff Green | 3 | jeffg,jeffgreen,jeffgreen2


Of course you can iterate by getting one table with the count, then going back into the table and pulling out all the usernames that matches each name, but it seems wasteful. Mysql processes the rows the first time, why not pull out all the data while there?