question

How to identify duplicate rows has been addressed several times. Perhaps, if you review one of the previous answers, you'll be able to adapt it to your particular variation of the problem: [ http://ask.sqlservercentral.com/questions/48287/identify-and-remove-duplicates-help.html][1] [ http://ask.sqlservercentral.com/questions/41782/update-duplicate-records.html][2] [ http://ask.sqlservercentral.com/questions/89469/finding-and-returning-a-result-set-of-supplicate-r.html][3] [ http://ask.sqlservercentral.com/questions/3716/select-first-row-when-duplicate-rows-are-available.html][4] It is also helpful to define what constitutes a duplicate record - does the table have no primary key, so every single column is identical between the duplicate rows, or is there a primary key, but all of the non-primary key columns are identical between duplicate rows. Perhaps there are multiple portions of the same logical row which has had different columns duplicated across different rows due to an application error. "Single query" is another thing that can be difficult to pin down. Do you count using a CTE count as a single query? How about correlated subqueries within the context of a "single query" Union all? Can you only use the key word SELECT (or INSERT or UPDATE or DELETE) once? How about not at all: Executing a stored procedure is a "single query", though it can contain thousands of lines of SQL and tens or hundreds of queries. [1]: http://ask.sqlservercentral.com/questions/48287/identify-and-remove-duplicates-help.html [2]: http://ask.sqlservercentral.com/questions/41782/update-duplicate-records.html [3]: http://ask.sqlservercentral.com/questions/89469/finding-and-returning-a-result-set-of-supplicate-r.html [4]: http://ask.sqlservercentral.com/questions/3716/select-first-row-when-duplicate-rows-are-available.html