Below is my SQL Server query which works great, but as you can see I am fetching the same row multiple times. Is there a way to optimise this by doing a SELECT only once? Thanks.
SELECT TOP 1 1
FROM some_table
WHERE col1 = @col1
AND (col2 IN ('N', 'C', 'U') OR
col3 != (SELECT t.col3 FROM some_table t
WHERE t.id = id AND t.revision_id = revision_id - 1) OR
col4 != (SELECT t.col4 FROM some_table t
WHERE t.id = id AND t.revision_id = revision_id - 1) OR
col5 != (SELECT t.col5 FROM some_table t
WHERE t.id = id AND t.revision_id = revision_id - 1) OR
col6 != (SELECT t.col6 FROM some_table t
WHERE t.id = id AND t.revision_id = revision_id - 1)
)
Note: I only need to see only if one such column exists (see SELECT 1 1
), which is more than likely to give be 1
most times. So I do NOT want to do a join of the table which has millions of rows.
CodePudding user response:
First, you have an error with the column names in the subqueries. See Qualifying column names in subqueries
column names in a statement are implicitly qualified by the table referenced in the FROM clause at the same level.
So, all of these columns belong to t
: t.id=id AND t.revision_id=revision_id-1
And the query:
SELECT TOP 1 1
FROM some_table s
WHERE col1=@col1
AND
(
col2 IN ('N','C','U') OR
EXISTS(SELECT * FROM some_table where id = s.id
AND revision_id = s.revision_id - 1 AND
(col3 != s.col3 OR col4 != s.col4 OR col5 != s.col5 OR col6 != s.col6)
)
)
CodePudding user response:
You can use apply
here to select multiple columns form a related row.
This is just an approximation of what you need as you haven't supplied any example data and expected results. You may need to use a top (1)
with an appropriate order by
clause or aggregate if the apply can return more than one row:
select *
from some_table t
outer apply (
select col3, col4, col5, col6
from some_table t2
where t2.id = t.id AND t2.revision_id = t.revision_id-1
)t2
where col1 = @col1
and (
col2 IN ('N','C','U') or
t.col3 != t2.col3 or
t.col4 != t2.col4 or
t.col5 != t2.col5 or
t.col6 != t2.col6
);