Count NULL columns

Question

Post reply

Count NULL columns

karthik M

One Orange Chip

Points: 29139
More actions
February 7, 2014 at 6:43 am

#303797

Hi All,
I am working on the below requirement.
use css
go
create table test
(
eno int,
t1 int null,
t2 int null,
t3 int null,
t4 int null
)
insert into test values (7,null,null,null,null)
insert into test values (1,null,null,null,null)
insert into test values (2,2,null,null,null)
insert into test values (3,1,7,null,null)
insert into test values (4,1,7,4,2)
insert into test values (5,null,null,4,2)
insert into test values (6,null,null,null,2)
insert into test values (7,null,null,null,null)
select eno, cnt from
(
select eno, case when t1 is null then 1 else 0 end + case when t2 is null then 1 else 0 end + case when t3 is null then 1 else 0 end + case when t4 is null then 1 else 0 end as cnt
from test
)x
where cnt > 2
output:
14
23
63
74
But my actual requirement is to test 119 columns. do I need to hard code all 119 column in the sql query?
We can use Dynamic SQL to achieve this, again the problem is the table contains 250 columns.
is there any trick available to achieve this? Inputs are welcome!
karthik

Viewing 5 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic. Login to reply

ChrisM@Work SSC Guru Points: 186127 More actions · Answer 1

SELECT t.eno, d.NullCount

FROM test t

CROSS APPLY (

SELECT NullCount = 4 - COUNT(col) FROM (VALUES (t1), (t2), (t3), (t4)) d (col)

) d

WHERE d.NullCount > 2

^{“Write the query the simplest way. If through testing it becomes clear that the performance is inadequate, consider alternative query forms.” - Gail Shaw}

For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden

karthik M One Orange Chip Points: 29139 More actions · Answer 2

VALUES (t1), (t2), (t3), (t4))

I have 119 columns in the actual table. Do I need to use all those names in VALUES clause?

karthik

ChrisM@Work SSC Guru Points: 186127 More actions · Answer 3

karthik M (2/7/2014)
VALUES (t1), (t2), (t3), (t4))
I have 119 columns in the actual table. Do I need to use all those names in VALUES clause?

You could use dynamic SQL, but tbh it won't take more than a few moments to copy'n'paste them in.

^{“Write the query the simplest way. If through testing it becomes clear that the performance is inadequate, consider alternative query forms.” - Gail Shaw}

For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden

ScottPletcher SSC Guru Points: 100942 More actions · Answer 4

Many of us prefer to generate such code from the table itself, something like below. Uncomment the EXEC(@sql) when ready to actually run the code:

USE css

DECLARE @nonrepeated_cols nvarchar(max)

DECLARE @repeated_cols nvarchar(max)

DECLARE @sql nvarchar(max)

SELECT @nonrepeated_cols = (

SELECT ', ' + c.name

FROM sys.columns c

WHERE

c.object_id = OBJECT_ID('test') AND

c.name NOT LIKE 't%'

FOR XML PATH('')

)

SELECT @repeated_cols = (

SELECT ' + CASE WHEN [' + c.name + '] IS NULL THEN 1 ELSE 0 END'

FROM sys.columns c

WHERE

c.object_id = OBJECT_ID('test') AND

c.name LIKE 't%'

FOR XML PATH('')

)

SELECT @nonrepeated_cols, @repeated_cols

SELECT @sql = '

Select ' + SUBSTRING(@nonrepeated_cols, 3, 2000000000) + ',

' + SUBSTRING(@repeated_cols, 4, 2000000000) + ' AS null_count

from test

where ' + SUBSTRING(@repeated_cols, 4, 2000000000) + ' >= 2'

SELECT @sql

--EXEC(@sql)

SQL DBA,SQL Server MVP(07, 08, 09) A socialist is someone who will give you the shirt off *someone else's* back.