Assistance with PIVOT function - sql

I have a reference table that's currently delimited with '-' that I'm needing to split the values out into multiple columns. Our version of SQL DB doesn't support the string_split function.
The first part of the script (multiple CTE) is returning the results into multiple rows, which I'm then wanting to pivot into columns.
Is someone able to please assist with the PIVOT portion (or even a new statement if it achieves the same result). Am looking to have the results returned per final table format?
Thanks
Original Data:
ID
Value
Description
1
MV-RUC-DEBT-ASSESS
MV Debt Assessment
declare #T table (ID int, Col varchar(100), description varchar(50))
insert into #T values (1, 'MV-RUC-DEBT-ASSESS', 'MV Debt Assessment')
;
with cte as (
select a.ID
,replace(a.Col,'-', ' ') as "Col"
, a.description
from #T a
),
cte2 as (
select a.ID
, n.r.value('.', 'varchar(50)') "Value"
, a.description
from cte a
cross apply (select cast('<r>'+replace(replace(Col,'&','&'), ' ', '</r><r>')+'</r>' as xml)) as S(XMLCol)
cross apply S.XMLCol.nodes('r') as n(r)
)
select *
from cte2 a
pivot
(max(value) for a.ID in ([1], [2], [3], [4])) as "Pivot"
I'm expecting the results to look like
ID
Description
1
2
3
4
1
MV Debt Assessment
MV
RUC
DEBT
ASSESS

When you use a.ID as part of the pivot, it has only ones so that you only would get the max value of value.
adding a row_number would give you the wanted result
declare #T table (ID int, Col varchar(100), description varchar(50))
insert into #T values (1, 'MV-RUC-DEBT-ASSESS', 'MV Debt Assessment')
;
with cte as (
select a.ID
,replace(a.Col,'-', ' ') as "Col"
, a.description
from #T a
),
cte2 as (
select a.ID
, n.r.value('.', 'varchar(50)') "Value"
, row_number() OVER(PARTITION BY a.ID ORDER BY a.ID) rn
, a.description
from cte a
cross apply (select cast('<r>'+replace(replace(Col,'&','&'), ' ', '</r><r>')+'</r>' as xml)) as S(XMLCol)
cross apply S.XMLCol.nodes('r') as n(r)
)
select *
from cte2 a
pivot
(max(value) for a.rn in ([1], [2], [3], [4])) as "Pivot"
ID
description
1
2
3
4
1
MV Debt Assessment
MV
RUC
DEBT
ASSESS
fiddle

Related

Split SQL 2017 column apart by Carriage Return and Line feed

I'm working with an Address column in SSMS 2017 that has Carriage Returns and Line Feeds. One line in SQL is broken into three lines when pasting into Notepad.
Can someone help me break the single SQL column into multiple columns based on the Carriage Return and Line Feeds? I'm expecting to see the Address broken up into three different columns separated by CR and LF delimiter in SQL.
I've tried all the examples on the web that use CHAR(13) and Char(10) but they don't seem to be working.
Any help would be greatly appreciated.
You can use string_split and pivot to get the desired result.
Input text:
declare #t table (val nvarchar(1000))
insert into #t
select 'Smith, John
North Bend, 100 Bay Park, Field 2
Trever, NJ 11993
'
select * from #t
;WITH C AS
(
SELECT value ,ROW_NUMBER() OVER(PARTITION BY val ORDER BY (SELECT NULL)) as rn
FROM #t t
CROSS APPLY STRING_SPLIT(REPLACE(val, CHAR(13)+CHAR(10),'~'), '~') AS SPL
)
SELECT [1] AS [Name]
,[2] AS Addr_1
,[3] AS Addr_2
FROM C
PIVOT
(
MAX(VALUE)
FOR RN IN([1],[2],[3])
) as PVT
Result:
declare #table table(
input nvarchar(255))
insert into #table
values ('Smith, John
North Bend, 100 Bay Park, Field 2
Trever, NJ 11993')
select
pvt.[1] as [name] ,
pvt.[2] as [address],
pvt.[3] as [city state & zip]
from
(
select *
from #table cross apply
(select rowN = row_number() over (order by (select null)), value from string_split(input, char(10)))d)src
pivot (
max(value)
for src.rowN in ([1], [2], [3])
) pvt
link to fiddle uk

SQL Server Loop thru rows to form Groups

I using SQL Server 2008 R2 / 2014. I wish to find a SQL query that can do the following:
Rules:
Each [Group] must have [Number] 1 to 6 to be complete group.
[Name] in each [Group] must be unique.
Each row only can use 1 time.
Table before sorting is...
Name Number Group
---- ------ -----
A 1
B 6
A 123
C 3
B 4
C 23
D 45
D 4
C 56
A 12
D 56
After sorting, result I want is below or similar....
Name Number Group
---- ------ -----
A 1 1
C 23 1
D 45 1
B 6 1
A 123 2
D 4 2
C 56 2
A 12 3
C 3 3
B 4 3
D 56 3
What I tried before is to find a subgroup that have [Number] consist of 1-6 with below concatenate method...
SELECT *
FROM [Table1] ST2
WHERE
SUBSTRING((SELECT ST1.[Number] AS [text()]
FROM [Table1] ST1
-- WHERE ST1.[Group] = ST2.[Group]
ORDER BY LEFT(ST1.[Number],1)
FOR XML PATH ('')), 1, 1000) = '123456'
Maybe you should check ROW_NUMBER function.
select Name
, Number
, ROW_NUMBER () OVER(PARTITION BY Name ORDER BY Number) as Group
from [Table1]
If you have more than 6 rows with same NAME value then it will return more groups. You can filter additional groups out since you are interested in only 6 groups with unique values of NAME column.
I'm not sure if this can be done more simply or not, but here's my go at it...
Advanced warning, this requires some means of splitting strings. Since you're not on 2016, I've included a function at the beginning of the script.
The bulk of the work is a recursive CTE that builds the Name and Number columns into comma delimited groups. We then reduce our working set to only the groups where the numbers would create 123456, split the groups and use ROW_NUMBER() OVER... to identify them, and then select based on the new data.
Demo: http://rextester.com/NEXG53500
CREATE FUNCTION [dbo].[SplitStrings]
(
#List NVARCHAR(MAX),
#Delimiter NVARCHAR(255)
)
RETURNS TABLE
WITH SCHEMABINDING
AS
RETURN
(
SELECT Item = y.i.value('(./text())[1]', 'nvarchar(4000)')
FROM
(
SELECT x = CONVERT(XML, '<i>'
+ REPLACE(#List, #Delimiter, '</i><i>')
+ '</i>').query('.')
) AS a CROSS APPLY x.nodes('i') AS y(i)
);
GO
CREATE TABLE #temp
(
name VARCHAR(MAX),
number INT
)
INSERT INTO #temp
VALUES
('a',1),
('b',6),
('a',123),
('c',3),
('b',4),
('c',23),
('d',45),
('d',4),
('c',56),
('a',12),
('d',56);
/*** Recursively build groups based on information from #temp ***/
WITH groupFinder AS
(
SELECT CAST(name AS VARCHAR(MAX)) AS [groupNames], CAST(number AS VARCHAR(max)) AS [groupNumbers] FROM #temp
UNION ALL
SELECT
cast(CONCAT(t.[Name],',',g.[groupNames]) as VARCHAR(MAX)),
CAST(CONCAT(CAST(t.[Number] AS VARCHAR(max)),',',CAST(g.[groupNumbers] AS VARCHAR(max))) AS VARCHAR(max))
FROM #temp t
JOIN groupFinder g
ON
g.groupNames NOT LIKE '%' + t.name+'%'
AND g.[groupNumbers] NOT LIKE '%' + CAST(t.number/100 AS VARCHAR(10)) +'%'
AND g.[groupNumbers] NOT LIKE '%' + CAST(t.number/10 AS VARCHAR(10)) +'%'
AND g.[groupNumbers] NOT LIKE '%' + CAST(t.number%10 AS VARCHAR(10)) +'%'
)
/*** only get groups where the numbers form 123456 ***/
, groupPruner AS
(
SELECT *, ROW_NUMBER() OVER (ORDER BY [groupNames]) AS [rn] FROM groupFinder WHERE REPLACE([groupNumbers],',','') = '123456'
)
/*** split the name group and give it identifiers ***/
, nameIdentifier AS
(
SELECT g.*, c1.[item] AS [Name], ROW_NUMBER() OVER (PARTITION BY [rn] ORDER BY (SELECT NULL)) AS [rn1]
FROM groupPruner g
CROSS APPLY splitstrings(g.groupnames,',') c1
)
/*** split the number group and give it identifiers ***/
, numberIdentifier AS
(
SELECT g.*, c1.[item] AS [Number], ROW_NUMBER() OVER (PARTITION BY [rn], [rn1] ORDER BY (SELECT NULL)) AS [rn2]
FROM nameIdentifier g
CROSS APPLY splitstrings(g.groupNumbers,',') c1
)
SELECT [Name], [Number], [rn] AS [Group]
--,groupnames, groupNumbers /*uncomment this line to see the groups that were built*/
FROM numberIdentifier
WHERE rn1 = rn2
ORDER BY rn, rn1
DROP TABLE #temp

SQL Server - Dynamic Pivot with 2 Group Variables and 2 Aggregate Calculations

I have a dataset that is shaped like this:
I am trying to convert the data to this format:
As you can see, I'd like to sum the accounts and revenue (for each month) by State and Account Type. It is important to note that I seek a dynamic solution as these ARE NOT the only values (hard-coding is not an option!).
What SQL query can I write to accomplish this task, dynamically? (as these values are not the only ones present in the complete dataset).
Thanks!
I'm assuming you want to keep the columns in order by date, thus the top 100 percent ... order by in the section where we generate the columns
Example
Declare #SQL varchar(max) = '
Select *
From (
Select [State]
,[AccountType]
,B.*
From YourTable A
Cross Apply ( values (concat(''Accounts_'',format([Date],''MM/dd/yyyy'')),Accounts)
,(concat(''Revenue_'' ,format([Date],''MM/dd/yyyy'')),Revenue)
) B (Item,Value)
) A
Pivot (sum([Value]) For [Item] in (' + Stuff((Select ','+QuoteName('Accounts_'+format([Date],'MM/dd/yyyy'))
+','+QuoteName('Revenue_' +format([Date],'MM/dd/yyyy'))
From (Select top 100 percent [Date] from YourTable Group By [Date] Order by [Date] ) A
For XML Path('')),1,1,'') + ') ) p'
--Print #SQL
Exec(#SQL)
Returns
If it helps, the generated SQL looks like this:
Select *
From (
Select [State]
,[AccountType]
,B.*
From YourTable A
Cross Apply ( values (concat('Accounts_',format([Date],'MM/dd/yyyy')),Accounts)
,(concat('Revenue_' ,format([Date],'MM/dd/yyyy')),Revenue)
) B (Item,Value)
) A
Pivot (sum([Value]) For [Item] in ([Accounts_12/31/2017],[Revenue_12/31/2017],[Accounts_01/31/2018],[Revenue_01/31/2018]) ) p

Split SQL server string that has Decimal points into multiple Columns

I have the following table
Col
=========================
1270.8/847.2/254.16/106.9
And I would like to be split into columns like so:
Col1 Col2 Col3 Col4
============================================
1270.8 847.2 254.16 106.9
I have the code below, but it doesn't take the decimal into consideration.
Declare #Sample Table
(MachineName varchar(max))
Insert into #Sample
values ('1270.8/847.2/254.16');
SELECT
Reverse(ParseName(Replace(Reverse(MachineName), '/', ''), 1)) As [M1]
, Reverse(ParseName(Replace(Reverse(MachineName), '/', ''), 2)) As [M2]
, Reverse(ParseName(Replace(Reverse(MachineName), '/', ''), 3)) As [M3]
FROM #Sample
In SQL Server 2016+ you can use string_split().
In SQL Server pre-2016, using a CSV Splitter table valued function by Jeff Moden and conditional aggregation:
declare #Sample Table (id int not null identity(1,1), MachineName varchar(max));
insert into #Sample values ('1270.8/847.2/254.16'),('1270.8/847.2/254.16/106.9');
select
t.id
, m1 = max(case when s.ItemNumber = 1 then s.Item end)
, m2 = max(case when s.ItemNumber = 2 then s.Item end)
, m3 = max(case when s.ItemNumber = 3 then s.Item end)
, m4 = max(case when s.ItemNumber = 4 then s.Item end)
from #Sample t
cross apply dbo.delimitedsplit8K(MachineName,'/') s
group by id
rextester demo: http://rextester.com/WJVLB77682
returns:
+----+--------+-------+--------+-------+
| id | m1 | m2 | m3 | m4 |
+----+--------+-------+--------+-------+
| 1 | 1270.8 | 847.2 | 254.16 | NULL |
| 2 | 1270.8 | 847.2 | 254.16 | 106.9 |
+----+--------+-------+--------+-------+
splitting strings reference:
Tally OH! An Improved SQL 8K “CSV Splitter” Function - Jeff Moden
Splitting Strings : A Follow-Up - Aaron Bertrand
Split strings the right way – or the next best way - Aaron Bertrand
string_split() in SQL Server 2016 : Follow-Up #1 - Aaron Bertrand
Everyone should have a good split/parse function as illustrated by SQLZim (+1), but another option could be as follow:
Declare #YourTable table (ID int,Col varchar(max))
Insert Into #YourTable values
(1,'1270.8/847.2/254.16/106.9')
Select A.ID
,B.*
From #YourTable A
Cross Apply (
Select Col1 = xDim.value('/x[1]','float')
,Col2 = xDim.value('/x[2]','float')
,Col3 = xDim.value('/x[3]','float')
,Col4 = xDim.value('/x[4]','float')
From (Select Cast('<x>' + replace((Select replace(A.Col,'/','§§Split§§') as [*] For XML Path('')),'§§Split§§','</x><x>')+'</x>' as xml) as xDim) as A
) B
Returns
ID Col1 Col2 Col3 Col4
1 1270.8 847.2 254.16 106.9
EDIT - If 2012+, and just to be super-duper safe
Select A.ID
,B.*
From #YourTable A
Cross Apply (
Select Col1 = try_convert(float,xDim.value('/x[1]','varchar(100)'))
,Col2 = try_convert(float,xDim.value('/x[2]','varchar(100)'))
,Col3 = try_convert(float,xDim.value('/x[3]','varchar(100)'))
,Col4 = try_convert(float,xDim.value('/x[4]','varchar(100)'))
From (Select Cast('<x>' + replace((Select replace(A.Col,'/','§§Split§§') as [*] For XML Path('')),'§§Split§§','</x><x>')+'</x>' as xml) as xDim) as A
) B
If you are using SQL Server 2016, you can use String_Split()
;with cte as (
select RowN = row_number() over(order by (SELECT NULL)), * from string_split('1270.8/847.2/254.16/106.9','/')
) select * from cte
pivot (max(value) for RowN in ([1],[2],[3],[4])) p
If you are using less than SQL Server 2016 version then you might require to use custom split functions... Many ways to write custom split function one easier way is to write using xml
CREATE Function dbo.udf_split( #str varchar(max), #delimiter as varchar(5) )
RETURNS #retTable Table
( RowN int,
value varchar(max)
)
AS
BEGIN
DECLARE #xml as xml
SET #xml = cast(('<X>'+replace(#str,#delimiter ,'</X><X>')+'</X>') as xml)
INSERT INTO #retTable
SELECT RowN = Row_Number() over (order by (SELECT NULL)), N.value('.', 'varchar(MAX)') as value FROM #xml.nodes('X') as T(N)
RETURN
END
--Your query
;with cte as (
select * from udf_split('1270.8/847.2/254.16/106.9','/')
) select * from cte
pivot (max(value) for RowN in ([1],[2],[3],[4])) p
But mine is similar to John's solution... Just now only looking at that
If you are using in value in a table then you can use cross apply as below
create table #t (v varchar(50), i int)
insert into #t (v, i) values ('1270.8/847.2/254.16/106.9',1)
,('847.222/254.33/106.44',2)
select * from #t t cross apply string_split(t.v, '/')
create table #t (v varchar(50), i int)
insert into #t (v, i) values ('1270.8/847.2/254.16/106.9',1)
,('847.222/254.33/106.44',2)
--Just to get all the values
select * from #t t cross apply string_split(t.v, '/')
--Inorder to get into same row -pivoting the data
select * from (
select * from #t t cross apply (select RowN=Row_Number() over (Order by (SELECT NULL)), value from string_split(t.v, '/') ) d) src
pivot (max(value) for src.RowN in([1],[2],[3],[4])) p

SQL to Make XML Path list Columns

I have the following SQL Server query which cranks out a comma delimited list into one field.
Result looks like this 2003, 9083, 4567, 3214
Question: What would be the best way (SQL syntax) to put this into columns?
Meaning, I need these to show up as 1 column for "2003", 1 column for "9083" 1 column, for "4567" ..etc.
Obviously the number of columns would be dynamic based on the policy ID I give it . Any idea would be most appreciated.
My query is below .
SELECT DISTINCT x.ClassCode + ', '
FROM PremByClass x
WHERE x.PolicyId = 1673885
FOR XML PATH('')
If you take out the XML and the comma you are left with
SELECT DISTINCT x.ClassCode
FROM PremByClass x
WHERE x.PolicyId = 1673885
Which gives you a single column of the values, to turn this into columns you need to PIVOT it. However, you need to specify the names of the columns.
There is some more information in this answer https://stackoverflow.com/a/15931734/350188
You need PIVOT and if number of values could be different - dynamic SQL:
SELECT *
FROM (
SELECT DISTINCT ClassCode,
ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) RN
FROM PremByClass
WHERE PolicyId = 1673885
) as t
PIVOT (
MAX(ClassCode) FOR RN IN ([1],[2],[3],[4])
) as pvt
Will give you:
1 2 3 4
-----------------------------
2003 9083 4567 3214
Dynamic SQL will be something like:
DECLARE #sql nvarchar(max),
#columns nvarchar(max)
SELECT #columns = STUFF((
SELECT DISTINCT ','+QUOTENAME(ROW_NUMBER() OVER (ORDER BY (SELECT NULL)))
FROM PremByClass
WHERE PolicyId = 1673885
FOR XML PATH('')
),1,1,'')
SELECT #sql = N'
SELECT *
FROM (
SELECT DISTINCT ClassCode,
ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) RN
FROM PremByClass
WHERE PolicyId = 1673885
) as t
PIVOT (
MAX(ClassCode) FOR RN IN ('+#columns+')
) as pvt'
EXEC sp_executesql #sql
Assuming that there's a limit to the amount of numbers in that csv string.
You could cast or convert it to an xml type, and then put the values in as many columns you expect.
In this example it's assumed that there's no more than 6 values in the text:
declare #PolicyId INT = 1673885;
select PolicyId
,x.value('/x[1]','int') as n1
,x.value('/x[2]','int') as n2
,x.value('/x[3]','int') as n3
,x.value('/x[4]','int') as n4
,x.value('/x[5]','int') as n5
,x.value('/x[6]','int') as n6
from (
select
PolicyId,
cast('<x>'+replace(ClassCode,',','</x><x>')+'</x>' as xml) as x
from PremByClass
where PolicyId = #PolicyId
) q;