从一个表或另一个表中选择数据的最佳方法

Question

I have two temporary tables, say #t1 and #t2 in Sql Server 2008. I need to create #t3 such as:

• when #t1 has rows, indifferently of #t2 's contents, #t3 = select * from #t1
• when #t1 has no rows, #t3 = select * from #t2

we can assume #t1 and #t2 have the same columns but I don't think I would like to rely on that fact.

I was thinking of something that draws some logic out of ' if exists (select * ...) ' statements, but wouldn't there be better like some sort of bool operators ?

Answer 1

The simplest way is to implement the logic as:

if (exists (select * from #t1))
begin
    select *
    into #t3
    from #t1;
end;
else
begin
    select *
    into #t3
    from #t2;
end;

You can do this in one statement as:

select t.*
into #t3
 from ((select *
        from #t1
       )
       union all
       (select *
        from #t2
        where not exists(select * from #t1)
       )
      ) t

However, I think the explicit if is a clearer way to express your intent.

Answer 2

This query would do just that, would see if there are rows in t1 and in that case populate table t3 with the contents of t1 , provided they have the same columns, otherwise it populates t3 with the contents of t2 .

IF(select count(*) from t1) > 0 
    BEGIN 
        select * into t3 from t1
    END
ELSE
    BEGIN
        select * into t3 from t2
    END

Here is a SQLFiddle so you can see how it works. To test the case when t1 does not have any rows, just remove the line that inserts into t1 , rebuild schema and re-run the query.