连接列别名的SQL语法

Question

I have a situation where I need to join tables on a common column name. I have some example SQL that shows what I am trying to do.

declare @t table (assetid int)
declare @x table (id int)

insert into @t(assetid) values (1)
insert into @t(assetid) values (2)
insert into @t(assetid) values (3)

insert into @x(id) values (1)
insert into @x(id) values (2)
insert into @x(id) values (3)

-- SUCCESS
select assetid as assetid
from @t t 
inner join @x x on x.id = t.assetid

-- FAIL
select assetid as EntityId
from @t t 
inner join @x x on x.id = t.EntityId  <-- syntax error

How can I rewrite the FAIL section above where I can join not on the column name from the table, but instead on an alias of that column name?

Answer 1

You'll need to use a CTE or inline view (subquery). ON is evaluated well before the SELECT clause. (Scroll to "Logical Processing Order of the SELECT statement.")

It's the same reason you can't use a column alias in the WHERE clause, or, well, anywhere else except the ORDER BY clause.

Answer 2

This way:

select EntityId
    from (select assetid as EntityId from @t) t 
    inner join @x x on x.id = t.EntityId