Order by in Inner Join

Question

I am putting inner join in my query.I have got the result but didn\'t know that how the data is coming in output.Can anyone tell me that how the Inner join matching the data.Bel

Answer 1

Add an ORDER BY ONE.ID ASC at the end of your first query.

By default there is no ordering.

Answer 2

SQL doesn't return any ordering by default because it's faster this way. It doesn't have to go through your data first and then decide what to do.

You need to add an order by clause, and probably order by which ever ID you expect. (There's a duplicate of names, thus I'd assume you want One.ID)

select * From one
inner join two
ON one.one_name = two.one_name
ORDER BY one.ID

Answer 3

You have to sort it if you want the data to come back a certain way. When you say you are expecting "Mohit" to be the first row, I am assuming you say that because "Mohit" is the first row in the [One] table. However, when SQL Server joins tables, it doesn't necessarily join in the order you think.

If you want the first row from [One] to be returned, then try sorting by [One].[ID]. Alternatively, you can order by any other column.

Answer 4

I found this to be an issue when joining but you might find this blog useful in understanding how Joins work in the back. How Joins Work..

[Edited] @Shree Thank you for pointing that out. On the paragraph of Merge Join. It mentions on how joins work...

Like hash join, merge join consists of two steps. First, both tables of the join are sorted on the join attribute. This can be done with just two passes through each table via an external merge sort. Finally, the result tuples are generated as the next ordered element is pulled from each table and the join attributes are compared

.

Answer 5

In SQL, the order of the output is not defined unless you specify it in the ORDER BY clause.

Try this:

SELECT  *
FROM    one
JOIN    two
ON      one.one_name = two.one_name
ORDER BY
        one.id

Answer 6

Avoid SELECT * in your main query.

Avoid duplicate columns: the JOIN condition ensures One.One_Name and two.One_Name will be equal therefore you don't need to return both in the SELECT clause.

Avoid duplicate column names: rename One.ID and Two.ID using 'aliases'.

Add an ORDER BY clause using the column names ('alises' where applicable) from the SELECT clause.

Suggested re-write:

SELECT T1.ID AS One_ID, T1.One_Name, 
       T2.ID AS Two_ID, T2.Two_name
  FROM One AS T1
       INNER JOIN two AS T2
          ON T1.One_Name = T2.One_Name
 ORDER 
    BY One_ID;