I am having trouble with using Between in comparison using varchar columns. When I run this:
SELECT DISTINCT diag_1, Diag_2, Diag_Admit
FROM Data_Exchange c
WHERE (Diag_1 BETWEEN '29600' and '29606' )
OR (Diag_2 BETWEEN '29600' and '29606')
OR (Diag_Admit between '29600' and '29606')
------
OR (Diag_1 between '29640' and '29680')
OR (Diag_2 between '29640' and '29680')
OR (Diag_Admit between '29640' and '29680')
I get the following shows up among the results:
2967
I don't want that. If it does not have 5 digits it should not show up in the results.
How do I fix this?
Thanks.
If all you want to do is limit the length of strings to check to 5 characters then add a where clause for each column:
--Untested(sorry)
SELECT DISTINCT diag_1, Diag_2, Diag_Admit
FROM Data_Exchange c
WHERE LEN(Diag_1)=5
AND LEN(Diag_2)=5
AND LEN(Diag_Admit)=5
AND ( Diag_1 BETWEEN '29600' and '29606'
OR Diag_2 BETWEEN '29600' and '29606'
OR Diag_Admit between '29600' and '29606'
OR Diag_1 between '29640' and '29680'
OR Diag_2 between '29640' and '29680'
OR Diag_Admit between '29640' and '29680')
However:
Looking at all those numbers in a varchar column makes me concerned that something else might be going on...
When the column is an alpha type then it is sorted alphabetically so '2967' should be between '29640' and '29680'.
Ideally if the column only contains numbers you should convert the datatype to an appropriate numeric type, but if this is not possible here is a workaround.
Sorted alphabetically the list 1,2,11,12,13 would be shown as 1,11,12,13,2 as can be seen in the following SQL.
--Test data.
DECLARE #alphaSort AS TABLE(col1 VARCHAR(2))
INSERT INTO #alphaSort VALUES('1')
INSERT INTO #alphaSort VALUES('2')
INSERT INTO #alphaSort VALUES('11')
INSERT INTO #alphaSort VALUES('12')
INSERT INTO #alphaSort VALUES('13')
SELECT col1 FROM #alphaSort ORDER BY col1
--Results:
col1
1
11
12
13
2
If you want the BETWEEN statement to work predictably for numbers you must convert the column to a numeric type:
SELECT col1 FROM #alphaSort ORDER BY CONVERT(INT,col1)
--Results
col1
1
2
11
12
13
If you have mixed data in the column then you will need to use the ISNUMERIC function to avoid the following error:
--New test data.
DECLARE #alphaSort AS TABLE(col1 VARCHAR(2))
INSERT INTO #alphaSort VALUES('1')
INSERT INTO #alphaSort VALUES('2')
INSERT INTO #alphaSort VALUES('11')
INSERT INTO #alphaSort VALUES('12')
INSERT INTO #alphaSort VALUES('13')
INSERT INTO #alphaSort VALUES('a')
--WITHOUT ISNUMERIC
SELECT col1 FROM #alphaSort ORDER BY CONVERT(INT,col1)
--Results
Msg 245, Level 16, State 1, Line xx
Conversion failed when converting the nvarchar value 'a' to data type int.
--WITH ISNUMERIC
SELECT col1 FROM #alphaSort WHERE ISNUMERIC(col1) = 1 ORDER BY CONVERT(INT,col1)
--Results
col1
1
2
11
12
13
If all you want to do is test that the the size of the column is grater then 4 you could so something like
select * from Data_Exchange
where len(Diag_1) >4
If not i think you need to add another set of () around your where clause. I think its getting confused
SELECT DISTINCT diag_1, Diag_2, Diag_Admit
FROM Data_Exchange c
WHERE ((Diag_1 BETWEEN '29600' and '29606' )
OR (Diag_2 BETWEEN '29600' and '29606')
OR (Diag_Admit between '29600' and '29606')
OR (Diag_1 between '29640' and '29680')
OR (Diag_2 between '29640' and '29680')
OR (Diag_Admit between '29640' and '29680'))
Related
I have a table with a column named "Calc" varchar(50). This column contains math calculations such as 1+1, 9*9, 10/2, 10-2 etc.
Is there way in an update query to apply this calculation from the column and output the result in the same table in the "Results" column varchar(50).
As you know by now, SQL Server does not have an EVAL() function, nor does it support macro substitution.
However, this can be done via dynamic SQL
Example
Create table #YourTable (id int,[Calc] varchar(150),Results varchar(150))
Insert Into #YourTable Values
(1,'1+1',null)
,(2,'9*9',null)
,(3,'10/2',null)
,(4,'10-2',null)
,(5,'datediff(DAY,''2018-01-01'',getdate())',null) -- Added Just for Fun
Declare #SQL varchar(max) = Stuff((Select ',' + concat('(',ID,',',[Calc],')')
From #YourTable A
For XML Path (''))
,1,1,'')
Exec('Update A set Results = B.Value
From #YourTable A
Join ( Select * from (values ' + #SQL + ')A([ID],[Value]) ) B
on A.ID = B.ID
')
Select *
From #YourTable
The Updated Table
id Calc Results
1 1+1 2
2 9*9 81
3 10/2 5
4 10-2 8
5 datediff(DAY,'2018-01-01',getdate()) 1012
I'm currently doing a data conversion project and need to strip all alphabetical characters from a string. Unfortunately I can't create or use a function as we don't own the source machine making the methods I've found from searching for previous posts unusable.
What would be the best way to do this in a select statement? Speed isn't too much of an issue as this will only be running over 30,000 records or so and is a once off statement.
You can do this in a single statement. You're not really creating a statement with 200+ REPLACEs are you?!
update tbl
set S = U.clean
from tbl
cross apply
(
select Substring(tbl.S,v.number,1)
-- this table will cater for strings up to length 2047
from master..spt_values v
where v.type='P' and v.number between 1 and len(tbl.S)
and Substring(tbl.S,v.number,1) like '[0-9]'
order by v.number
for xml path ('')
) U(clean)
Working SQL Fiddle showing this query with sample data
Replicated below for posterity:
create table tbl (ID int identity, S varchar(500))
insert tbl select 'asdlfj;390312hr9fasd9uhf012 3or h239ur ' + char(13) + 'asdfasf'
insert tbl select '123'
insert tbl select ''
insert tbl select null
insert tbl select '123 a 124'
Results
ID S
1 390312990123239
2 123
3 (null)
4 (null)
5 123124
CTE comes for HELP here.
;WITH CTE AS
(
SELECT
[ProductNumber] AS OrigProductNumber
,CAST([ProductNumber] AS VARCHAR(100)) AS [ProductNumber]
FROM [AdventureWorks].[Production].[Product]
UNION ALL
SELECT OrigProductNumber
,CAST(STUFF([ProductNumber], PATINDEX('%[^0-9]%', [ProductNumber]), 1, '') AS VARCHAR(100) ) AS [ProductNumber]
FROM CTE WHERE PATINDEX('%[^0-9]%', [ProductNumber]) > 0
)
SELECT * FROM CTE
WHERE PATINDEX('%[^0-9]%', [ProductNumber]) = 0
OPTION (MAXRECURSION 0)
output:
OrigProductNumber ProductNumber
WB-H098 098
VE-C304-S 304
VE-C304-M 304
VE-C304-L 304
TT-T092 092
RichardTheKiwi's script in a function for use in selects without cross apply,
also added dot because in my case I use it for double and money values within a varchar field
CREATE FUNCTION dbo.ReplaceNonNumericChars (#string VARCHAR(5000))
RETURNS VARCHAR(1000)
AS
BEGIN
SET #string = REPLACE(#string, ',', '.')
SET #string = (SELECT SUBSTRING(#string, v.number, 1)
FROM master..spt_values v
WHERE v.type = 'P'
AND v.number BETWEEN 1 AND LEN(#string)
AND (SUBSTRING(#string, v.number, 1) LIKE '[0-9]'
OR SUBSTRING(#string, v.number, 1) LIKE '[.]')
ORDER BY v.number
FOR
XML PATH('')
)
RETURN #string
END
GO
Thanks RichardTheKiwi +1
Well if you really can't use a function, I suppose you could do something like this:
SELECT REPLACE(REPLACE(REPLACE(LOWER(col),'a',''),'b',''),'c','')
FROM dbo.table...
Obviously it would be a lot uglier than that, since I only handled the first three letters, but it should give the idea.
Why does this return . when col1 contains a blank value?
CONCAT(NULLIF([COL1],''),'.')
I have 3 columns that i need to concatenate with a . in between, sometimes the column contains a blank value. In that case the trailing . should not be concatenated. What functions do I use?
col1 col2 col3
A 1 x
B 2
expected results:
A.1.X
B.2
test code:
DECLARE #tbl TABLE(a varchar(100),b varchar(100),c varchar(100))
INSERT INTO #tbl
SELECT 'A','1','X' UNION
SELECT 'B','2','' UNION
SELECT 'C','','' UNION
SELECT '','1','X' UNION
SELECT 'B','','' UNION
SELECT 'C','',''
SELECT CONCAT ( Nullif(a,''),'.' + nullif(b,''), '.' + nullif(c,'')) AS Contact_Result FROM #tbl;
You can use SQL CONCAT this way
SELECT CONCAT ( a,IIF((NULLIF(a,'')+NULLIF(b,'')) IS NULL,'','.'),b,IIF((NULLIF(b,'')+NULLIF(c,'')) IS NULL,'','.'), c) AS Contact_Result FROM #tbl;
Test code below
DECLARE #tbl TABLE(a varchar(100),b varchar(100),c varchar(100))
INSERT INTO #tbl
SELECT 'A','1','X' UNION
SELECT 'B','2',NULL UNION
SELECT 'C',NULL,NULL
SELECT CONCAT ( a,IIF((NULLIF(a,'')+NULLIF(b,'')) IS NULL,'','.'),b,IIF((NULLIF(b,'')+NULLIF(c,'')) IS NULL,'','.'), c) AS Contact_Result FROM #tbl;
Contact_Result
A.1.X
B.2
C
Another common use of this kind of concats is to Concat a Full Name in this case the . (dot) is replaced by a ' ' (space), it makes things easier because you can use trim
DECLARE #tbl TABLE(a varchar(100),b varchar(100),c varchar(100))
INSERT INTO #tbl
SELECT 'FirtName','MiddleName','LastName' UNION
SELECT 'FistName','','LastName' UNION
SELECT '','','FullName'
SELECT LTRIM(CONCAT ( a,' ' + b,' ' + c)) AS Contact_Result FROM #tbl;
Result
FullName
FirtName MiddleName LastName
FistName LastName
THIS IS AN ANSWER THAT COVERS ALL POSSIBILITIES
SELECT SUBSTRING(CONCAT ('.' + NULLIF(a,''),'.' + NULLIF(b,''),'.' + NULLIF(c,'')),2,10000) AS Contact_Result FROM #tbl;
Complete test cases
DECLARE #tbl TABLE(a varchar(100),b varchar(100),c varchar(100))
INSERT INTO #tbl
SELECT 'a','','' UNION
SELECT 'a','b','' UNION
SELECT 'a','b','c' UNION
SELECT 'a','','c' UNION
SELECT '','b','c' UNION
SELECT '','b','' UNION
SELECT '','','c' UNION
SELECT '','',''
SELECT SUBSTRING(CONCAT ('.' + NULLIF(a,''),'.' + NULLIF(b,''),'.' + NULLIF(c,'')),2,10000) AS Contact_Result FROM #tbl;
Results
c
b
b.c
a
a.c
a.b
a.b.c
You'll have to go old school. I'm on my phone and can't test this.
ISNULL(NULLIF([COL1],'') + '.', '') +
ISNULL(NULLIF([COL2],'') + '.', '') +
ISNULL(NULLIF([COL3],''), '');
---------------- EDIT ----------------
Okay, this was not as easy on my phone as I thought. Here's my updated solution that works with SQL Server 2005+
-- sample data
declare #table table
(
id int identity,
col1 varchar(10),
col2 varchar(10),
col3 varchar(10)
);
insert #table (col1, col2, col3)
values ('a','b','c'), ('aa','bb',''), ('x','','z'), ('','p','pp'),
('!!!','',''), ('','','fff'), ('','','');
-- My solution
select col1, col2, col3, concatinatedValue =
case when cv like '.%' then stuff(cv, 1, 1,'') else cv end
from #table
cross apply (values
(isnull(nullif([col1],''), '') +
isnull('.'+nullif([col2],''), '') +
isnull('.'+nullif([col3],''), ''))) v(cv);
RETURNS
cv col1 col2 col3 concatinatedValue
--------- ----- ----- ----- -------------------
a.b.c a b c a.b.c
aa.bb aa bb aa.bb
x.z x z x.z
.p.pp p pp p.pp
!!! !!! !!!
.fff fff fff
<----------- blank line ----------->
I am developing a T-SQL query in SSMS 2008 R2 which returns one line only. But the problem is that in this one line there are four fields which I instead want to be unique rows. For example, my output line looks like:
Col. 1 Col. 2 Col. 3 Col. 4
xxxx yyyy zzzz aaaa
Instead, I want this to look like:
Question Answer
Col. 1 xxxx
Col. 2 yyyy
Col. 3 zzzz
Col. 4 aaaa
I have tried using the UNPIVOT operator for this, but it is not doing the above. How can I achieve this?
You should be able to use UNPIVOT for this:
Here is a static pivot where you hard code in the values of the columns:
create table t1
(
col1 varchar(5),
col2 varchar(5),
col3 varchar(5),
col4 varchar(5)
)
insert into t1 values ('xxxx', 'yyyy', 'zzzz', 'aaaa')
select question, answer
FROM t1
unpivot
(
answer
for question in (col1, col2, col3, col4)
) u
drop table t1
Here is a SQL Fiddle with a demo.
but you can also use a Dynamic Unpivot:
DECLARE #cols AS NVARCHAR(MAX),
#query AS NVARCHAR(MAX);
select #cols = stuff((select ','+quotename(C.name)
from sys.columns as C
where C.object_id = object_id('t1') and
C.name like 'Col%'
for xml path('')), 1, 1, '')
set #query = 'SELECT question, answer
from t1
unpivot
(
answer
for question in (' + #cols + ')
) p '
execute(#query)
This is from my data names but it is tested
select 'sID', sID as 'val'
from [CSdemo01].[dbo].[docSVsys]
where sID = 247
union
select 'sParID', sParID as 'val'
from [CSdemo01].[dbo].[docSVsys]
where sID = 247 ;
But UNPIVOT should work
UNION-ing together your four questions would look like:
SELECT 'column1' AS Question, MAX(column1) AS Answer UNION
SELECT 'column2' , MAX(column2) UNION
SELECT 'column3' , MAX(column3) UNION
SELECT 'column4' , MAX(column4)
(obv just using the MAX as an example)
My SQL is rusty -- I have a simple requirement to calculate the sum of the greater of two column values:
CREATE TABLE [dbo].[Test]
(
column1 int NOT NULL,
column2 int NOT NULL
);
insert into Test (column1, column2) values (2,3)
insert into Test (column1, column2) values (6,3)
insert into Test (column1, column2) values (4,6)
insert into Test (column1, column2) values (9,1)
insert into Test (column1, column2) values (5,8)
In the absence of the GREATEST function in SQL Server, I can get the larger of the two columns with this:
select column1, column2, (select max(c)
from (select column1 as c
union all
select column2) as cs) Greatest
from test
And I was hoping that I could simply sum them thus:
select sum((select max(c)
from (select column1 as c
union all
select column2) as cs))
from test
But no dice:
Msg 130, Level 15, State 1, Line 7
Cannot perform an aggregate function on an expression containing an aggregate or a subquery.
Is this possible in T-SQL without resorting to a procedure/temp table?
UPDATE: Eran, thanks - I used this approach. My final expression is a little more complicated, however, and I'm wondering about performance in this case:
SUM(CASE WHEN ABS(column1 * column2) > ABS(column3 * column4)
THEN column5 * ABS(column1 * column2) * column6
ELSE column5 * ABS(column3 * column4) * column6 END)
Try this:
SELECT SUM(CASE WHEN column1 > column2
THEN column1
ELSE column2 END)
FROM test
Try this... Its not the best performing option, but should work.
SELECT
'LargerValue' = CASE
WHEN SUM(c1) >= SUM(c2) THEN SUM(c1)
ELSE SUM(c2)
END
FROM Test
SELECT
SUM(MaximumValue)
FROM (
SELECT
CASE WHEN column1 > column2
THEN
column1
ELSE
column2
END AS MaximumValue
FROM
Test
) A
FYI, the more complicated case should be fine, so long as all of those columns are part of the same table. It's still looking up the same number of rows, so performance should be very similar to the simpler case (as SQL Server performance is usually IO bound).
How to find max from single row data
-- eg (empid , data1,data2,data3 )
select emplid , max(tmp.a)
from
(select emplid,date1 from table
union
select emplid,date2 from table
union
select emplid,date3 from table
) tmp , table
where tmp.emplid = table.emplid
select sum(id) from (
select (select max(c)
from (select column1 as c
union all
select column2) as cs) id
from test
)
The best answer to this is simply put :
;With Greatest_CTE As
(
Select ( Select Max(ValueField) From ( Values (column1), (column2) ) ValueTable(ValueField) ) Greatest
From Test
)
Select Sum(Greatest)
From Greatest_CTE
It scales a lot better than the other answers with more than two value columns.