Using Min and MAX function on Bit Field

Question

Hi All,
I am trying to write a SQL Query where i want to aggregate a bit field along with some  other numerical fields
I  have a Field of type Bit and it can contain data like
Scenario 1:              MyBit Field
                   Row1      Null
                  Row2       Null 
                  Row3       Null
Scenario 2:              MyBit Field
                   Row1      1
                   Row2       1
                  Row3       Null
Scenario 3:              MyBit Field
                   Row1      0
                  Row2       0
                  Row3       Null
Scenario 4:              MyBit Field
                   Row1      1
                   Row2       0
                  Row3       Null
Scenario 5:              MyBit Field
                   Row1      1
    Row2       1
                  Row3       0
Scenario 6:              MyBit Field
                   Row1      1
                  Row2       1
                  Row3       1
Scenario 7:              MyBit Field
                   Row1      0
                   Row2       0
                  Row3       0
These are all scenario i can think of, i want to have sql statement which should return me
null if all values are null,
if there is mix of 1, 0 then return NA,
if there is mix of 1,0,null return NA
if there is mix of 1's with Null then return 1 ,
if there is mix of 0's with null return 0 ,
if there is all 0's return 0
if there is all 1's return 1
IS there an elegant way in SQL to handle this
Note : I have other fields in query which are simply getting SUM.

Answer 1

Try:
DECLARE @T table (
Scenario int NOT NULL,
col1 int NOT NULL,
col2 bit NULL,
UNIQUE CLUSTERED (Scenario, col2, col1)
INSERT INTO @T
(Scenario, col1, col2)
VALUES
(1, 1, NULL),
(1, 2, NULL),
(1, 3, NULL),
(2, 1, 1),
(2, 2, 1),
(2, 3, NULL),
(3, 1, 0),
(3, 2, 0),
(3, 3, NULL),
(4, 1, 0),
(4, 2, 1),
(4, 3, NULL),
(5, 1, 1),
(5, 2, 1),
(5, 3, 0),
(6, 1, 1),
(6, 2, 1),
(6, 3, 1),
(7, 1, 0),
(7, 2, 0),
(7, 3, 0);
WITH T AS (
SELECT
Scenario,
MIN(col2 * 1) AS min_col2,
MAX(col2 * 1) AS max_col2,
COUNT(*) AS cnt
FROM
@T
GROUP BY
Scenario,
col2
, R AS (
SELECT
Scenario,
MIN(min_col2) AS min_col2,
MAX(max_col2) AS max_col2,
MAX(cnt) AS max_cnt,
SUM(cnt) AS total
FROM
T
GROUP BY
Scenario
SELECT
Scenario,
CASE
WHEN max_cnt = total THEN CAST(min_col2 AS char(1))
WHEN min_col2 = 1 AND max_cnt < total THEN CAST(min_col2 AS char(1))
WHEN max_col2 = 0 AND max_cnt < total THEN CAST(max_col2 AS char(1))
WHEN min_col2 <> max_col2 AND max_cnt < total THEN 'NA'
END AS col3
FROM
R;
GO
AMB
Some guidelines for posting questions...
AYÚDANOS A AYUDARTE, guía básica de consejos para formular preguntas