表值作为SQL Server存储过程的参数

Question

I have table1 :col1, col2, col3 and table2: col1, col2, col3
My goal is to get all records

where     
t2.col1 like t1.col1 and    
t2.col2 like t1.col2 and    
t2.col3 like t1.col3   

........................................

One variant is the inner join method

select * from t2 inner join t1 on    
t2.col1 like t1.col1  and  
t2.col2 like t1.col2  and  
t2.col3 like t1.col3     

........................................

Another variant is a stored procedure based on the 'where' clause:

select *  from t2     
where t2.col1 like parameter1  and      
t2.col2 like parameter2  and     
t2.col3 like parameter3    

Then I call the procedure in VBA and I use a for next loop to go through all values/parameters from an excel table1
........................................
Execution time for the join method is slower(~20, 30%) than vba+sp method, but unfortunately, for a big set of parameters, excel freeze.
........................................
Is possible to apply loop method and go thru table1 values, as parameters for the stored procedure, inside sql server, in a sql script, no vba or c++ or perl etc. ?

I am a user with no access to db/tables design.

Thank you

Answer 1

First of all, your two queries in the question are not equivalent:

select * from t2 inner join t1 on    
t1.col1 like t2.col1  and  
t1.col2 like t2.col2  and  
t1.col3 like t2.col3     

Here you have t1 like t2

select *  from t2     
where t2.col1 like parameter1  and      
t2.col2 like parameter2  and     
t2.col3 like parameter3    

Here it is other way around t2 like t1 .

End result would be different.

Based on the sample data it looks like it should be t2 like t1 .

---

You can try to re-write the query using CROSS APPLY instead of JOIN , but it is unlikely to make any difference in performance.

SELECT *
FROM
    t1
    CROSS APPLY
    (
        SELECT
        FROM t2
        WHERE
            t2.col1 like t1.col1
            and t2.col2 like t1.col2
            and t2.col3 like t1.col3
    ) AS A
;

This query structure mimics your stored procedure approach where for each row from t1 you select a set of rows from t2 .