Question

I have a table with a certain number of columns and a primary key column (suppose OriginalKey). I perform a GROUP BY on a certain sub-set of those columns and store them in a temporary table with primary key (suppose GroupKey). At a later stage, I may need to get more details about one or more of those groupings (which can be found in the temporary table) i.e. I need to know which were the rows from the original table that formed that group. Simply put, I need to know the mappings between GroupKey and OriginalKey. What’s the best way to do this? Thanks in advance.

Example:

Table Student(
StudentID INT PRIMARY KEY,
Level INT, --Grade/Class/Level depending on which country you are from)
HomeTown TEXT,
Gender CHAR)

INSERT INTO TempTable SELECT HomeTown, Gender, COUNT(*) AS NumStudents FROM Student GROUP BY HomeTown, Gender

On a later date, I would like to find out details about all towns that have more than 50 male students and know details of every one of them.

Answer 1

You’d need to store the fields you grouped on in your temporary table, so you can join back to the original table. e.g. if you grouped on fieldA, fieldB, and fieldC, you’d need something like:

select original.id
from original
inner join temptable on
   temptable.fieldA = original.fieldA and
   temptable.fieldB = original.fieldB and
   temptable.fieldC = original.fieldC