erikd (4/15/2013)
dhananjay.nagarkar (4/14/2013)
Sure I would love to share-Here's what I did-
created 3 temp tables with the filelds I need for each with same order, data type.
then UNION the 3 selects.
then in the UNION i calculated the field as below-
SELECT A.*,
CASE
WHEN [Age (Working Days)] IS NULL
THEN NULL -- DO NOT DISLAY THIS FIELD IF AGE WORKING DAYS IS NULL
ELSE -- DISPLAY AGE GROUP WHEN AGE IS PRESENT
CASE
WHEN CAST([Age (Working Days)] AS INT) < 30
THEN '1 - 29'
WHEN CAST([Age (Working Days)] AS INT) < 60
THEN '30 - 59'
WHEN CAST([Age (Working Days)] AS INT) < 90
THEN '60 - 89'
WHEN CAST([Age (Working Days)] AS INT) < 120
THEN '90 - 119'
ELSE'120 +'
END
END AS [Age Group (Working Days)]
FROM (
SELECT * FROM #tmp_JP
UNION
SELECT * FROM #tmp_WO
UNION
SELECT * FROM #tmp_WOR
)A
Thanks
Dhananjay
Doesn't everything get cast as either <120 or Over 120, or NULL? You didn't set any lower limits on your higher numbers?
WHEN CAST([Age (Working Days)] AS INT) < 30
THEN '1 - 29'
WHEN CAST([Age (Working Days)] AS INT) between 30 and 59
THEN '30 - 59'
WHEN CAST([Age (Working Days)] AS INT) between 60 and 89
THEN '60 - 89'
WHEN CAST([Age (Working Days)] AS INT) between 90 and 119
THEN '90 - 119'
ELSE '120 +'
Actually, no. If working days is less than 30 then the first WHEN is satisfied. If working days is less than 60, then the second WHEN is satisfied. If you flipped the CASE and started with the largest value first, then yes you would need the lower bounds defined as well. The CASE stops evaluating when it hits the first successful condition.