TSQL Compare the result of 2 queries - sql-server

Question

does anyone have ideas how to compare the result of 2 queries, that have the same columns names, but in different order?
I know that if I had both queries returning the same columns in the same order, I could use except, this isn't the case.
[EDIT]
To be more specific, I need to compare the value of each row, and each column (with the same name) from 2 different queries.
Example:
result query 1:
A|B|C|D
1|4|7|11
2|5|8|21
3|**6**|9|31
result query 2:
A|B |D
1|4 |11
2|5 |21
3|**99**|31
In this case, I would like to detect that Query2 on 3º row in column B, have a different value.
I don't care that Query2 don't have the column C, I just want that all common columns, between both queries, have the same values.
Thanks

Answer 1

Given these tables and data:
USE tempdb;
GO
CREATE TABLE dbo.TableA
(
A INT,
B INT,
C INT,
D INT
);
CREATE TABLE dbo.TableB
(
A INT,
D INT,
B INT
);
INSERT dbo.TableA SELECT 1,4,7,11
UNION ALL SELECT 2,5,8,21
UNION ALL SELECT 3,6,9,31;
INSERT dbo.TableB SELECT 1,11,4
UNION ALL SELECT 2,21,5
UNION ALL SELECT 3,31,99;
What you seem to be looking for is one of the following:
-- those where at least one column doesn't match:
SELECT A,B,D FROM dbo.TableA
EXCEPT
SELECT A,B,D FROM dbo.TableB;
Results (from the A side):
A B D
---- ---- ----
3 6 31
OR
-- those where all columns DO match:
SELECT A,B,D FROM dbo.TableA
INTERSECT
SELECT A,B,D FROM dbo.TableB;
Results:
A B D
---- ---- ----
1 4 11
2 5 21
If you don't know the columns or don't want to write them out manually, you can do this with dynamic SQL by just passing the two table names (with schema) into variables. Note that this doesn't trap for the errors that will occur if no columns are shared by the two tables, or if the same column names exist but are of incompatible data types. That error handling is easy to add if you want to make the solution more robust.
DECLARE
#sql NVARCHAR(MAX),
#cols NVARCHAR(MAX),
#t1 NVARCHAR(511),
#t2 NVARCHAR(511);
SELECT
#sql = N'',
#cols = N'',
#t1 = N'dbo.TableA',
#t2 = N'dbo.TableB';
SELECT #cols = #cols + ',' + a.name
FROM sys.columns AS a
INNER JOIN sys.columns AS b
ON a.name = b.name
WHERE a.[object_id] = OBJECT_ID(#t1)
AND b.[object_id] = OBJECT_ID(#t2);
SET #cols = STUFF(#cols, 1, 1, N'');
-- those where at least one column doesn't match:
SELECT #sql = N'SELECT ' + #cols + '
FROM ' + #t1 + ' EXCEPT
SELECT ' + #cols + ' FROM ' + #t2 + ';';
EXEC sp_executesql #sql;
-- those where all columns DO match:
SELECT #sql = N'SELECT ' + #cols + '
FROM ' + #t1 + ' INTERSECT
SELECT ' + #cols + ' FROM ' + #t2 + ';';
EXEC sp_executesql #sql;
Don't forget to clean up:
DROP TABLE dbo.TableA, dbo.TableB;

Answer 2

You can wrap your queries as subqueries and then reselect the columns in any order you want.

Answer 3

You can do it in only one step:
SELECT *
FROM (
--compare query a vs query b
SELECT ad.id_addetto,'not in b'y
FROM addetti ad
WHERE ad.id_addetto < 125 -- query a
EXCEPT
SELECT ad.id_addetto,'not in b'y
FROM addetti ad
WHERE ad.id_addetto < 166 -- query b
UNION
--compare query b vs query a
SELECT ad.id_addetto, 'not in a'y
FROM addetti ad
WHERE ad.id_addetto < 166 -- query b
EXCEPT
SELECT ad.id_addetto ,'not in a'y
FROM addetti ad
WHERE ad.id_addetto < 125 -- query a
) xx