I have the following SQL query:
SELECT DISTINCT ProductNumber, PageNumber FROM table
I am trying to modify the query so that PageNumber will be formatted. You see, PageNumber is in any of the following formats, where 'x' is a digit:
xxx, xxx
xxx
xxx-xxx
xx, xxx-xxx
xx-xx, xxx
xx-xx, xxx-xxx
I want to format PageNumber so that it is only in the format: xxx. To do so, I have parse out the following bolded numbers from the above formats:
xxx, xxx
xxx
xxx-xxx
xx, xxx-xxx
xx-xx, xxx
xx-xx, xxx-xxx
I want to do this all without writing any functions, but I don't know if that is possible. I am having trouble "detecting" all of the different formats, though:
Here is what I have so far:
SELECT ProductNumber,
CASE WHEN CHARINDEX(',', PageNumber) > 0
THEN SUBSTRING(PageNumber, 0, CHARINDEX('-', PageNumber))
WHEN CHARINDEX('-', PageNumber) > 0
THEN SUBSTRING(PageNumber, 0, CHARINDEX('-', PageNumber))
ELSE PageNumber
END AS PageNumber
FROM table
WHERE PageNumber IS NOT NULL
AND PageNumber <> ''
Can anyone offer me some help? Thanks!
Use pattern matching rather than CHARINDEX
CASE also forces ordering of evaluation which helps here for the 3rd case which overlaps with the first 2 cases.
Not tested, something like
CASE
WHEN PageNumber LIKE '[0-9][0-9][0-9]%' THEN LEFT(PageNumber, 3)
WHEN PageNumber LIKE '[0-9][0-9]-[0-9][0-9], [0-9][0-9][0-9]') THEN RIGHT(PageNumber , 3)
WHEN PageNumber LIKE '[0-9][0-9]%') THEN LEFT(PageNumber, 2)
END
try this:
DECLARE #YourTable table (ProductNumber int, PageNumber varchar(20))
INSERT #YourTable VALUES (1,'123, 456')
INSERT #YourTable VALUES (2,'123')
INSERT #YourTable VALUES (3,'123-456')
INSERT #YourTable VALUES (4,'12, 345-678')
INSERT #YourTable VALUES (5,'12-34, 567')
INSERT #YourTable VALUES (6,'12-34, 567-789')
;WITH AllNumbers AS ---builds a Numbers table 1-100
( SELECT 1 AS Number
UNION ALL
SELECT Number+1
FROM AllNumbers
WHERE Number<101
)
, RowChars AS --one row for each non-numeric single character value per #YourTable row
( SELECT DISTINCT
ProductNumber,Number, SUBSTRING(PageNumber,Number,1) AS CharacterOF
FROM #YourTable
INNER JOIN AllNumbers ON 1=1
WHERE SUBSTRING(PageNumber,Number,1) IS NOT NULL AND SUBSTRING(PageNumber,Number,1) NOT LIKE '[0-9]' AND SUBSTRING(PageNumber,Number,1)!=''
)
,FirstSplit AS --get first non-numeric single character value per #YourTable row
( SELECT
ProductNumber,MIN(Number) AS SplitOf
FROM RowChars
GROUP BY ProductNumber
)
SELECT
t.ProductNumber, LEFT(t.PageNumber,COALESCE(s.SplitOf-1,LEN(t.PageNumber))) AS NewPage,t.PageNumber AS OldPage
FROM #YourTable t
LEFT OUTER JOIN FirstSplit s ON t.ProductNumber=s.ProductNumber
OUTPUT:
ProductNumber NewPage OldPage
------------- -------------------- --------------------
1 123 123, 456
2 123 123
3 123 123-456
4 12 12, 345-678
5 12 12-34, 567
6 12 12-34, 567-789
(6 row(s) affected)
Related
I have a table that looks like this:
ProductId, Color
"1", "red, blue, green"
"2", null
"3", "purple, green"
And I want to expand it to this:
ProductId, Color
1, red
1, blue
1, green
2, null
3, purple
3, green
Whats the easiest way to accomplish this? Is it possible without a loop in a proc?
Take a look at this function. I've done similar tricks to split and transpose data in Oracle. Loop over the data inserting the decoded values into a temp table. The convent thing is that MS will let you do this on the fly, while Oracle requires an explicit temp table.
MS SQL Split Function
Better Split Function
Edit by author:
This worked great. Final code looked like this (after creating the split function):
select pv.productid, colortable.items as color
from product p
cross apply split(p.color, ',') as colortable
based on your tables:
create table test_table
(
ProductId int
,Color varchar(100)
)
insert into test_table values (1, 'red, blue, green')
insert into test_table values (2, null)
insert into test_table values (3, 'purple, green')
create a new table like this:
CREATE TABLE Numbers
(
Number int not null primary key
)
that has rows containing values 1 to 8000 or so.
this will return what you want:
EDIT
here is a much better query, slightly modified from the great answer from #Christopher Klein:
I added the "LTRIM()" so the spaces in the color list, would be handled properly: "red, blue, green". His solution requires no spaces "red,blue,green". Also, I prefer to use my own Number table and not use master.dbo.spt_values, this allows the removal of one derived table too.
SELECT
ProductId, LEFT(PartialColor, CHARINDEX(',', PartialColor + ',')-1) as SplitColor
FROM (SELECT
t.ProductId, LTRIM(SUBSTRING(t.Color, n.Number, 200)) AS PartialColor
FROM test_table t
LEFT OUTER JOIN Numbers n ON n.Number<=LEN(t.Color) AND SUBSTRING(',' + t.Color, n.Number, 1) = ','
) t
EDIT END
SELECT
ProductId, Color --,number
FROM (SELECT
ProductId
,CASE
WHEN LEN(List2)>0 THEN LTRIM(RTRIM(SUBSTRING(List2, number+1, CHARINDEX(',', List2, number+1)-number - 1)))
ELSE NULL
END AS Color
,Number
FROM (
SELECT ProductId,',' + Color + ',' AS List2
FROM test_table
) AS dt
LEFT OUTER JOIN Numbers n ON (n.Number < LEN(dt.List2)) OR (n.Number=1 AND dt.List2 IS NULL)
WHERE SUBSTRING(List2, number, 1) = ',' OR List2 IS NULL
) dt2
ORDER BY ProductId, Number, Color
here is my result set:
ProductId Color
----------- --------------
1 red
1 blue
1 green
2 NULL
3 purple
3 green
(6 row(s) affected)
which is the same order you want...
You can try this out, doesnt require any additional functions:
declare #t table (col1 varchar(10), col2 varchar(200))
insert #t
select '1', 'red,blue,green'
union all select '2', NULL
union all select '3', 'green,purple'
select col1, left(d, charindex(',', d + ',')-1) as e from (
select *, substring(col2, number, 200) as d from #t col1 left join
(select distinct number from master.dbo.spt_values where number between 1 and 200) col2
on substring(',' + col2, number, 1) = ',') t
I arrived this question 10 years after the post.
SQL server 2016 added STRING_SPLIT function.
By using that, this can be written as below.
declare #product table
(
ProductId int,
Color varchar(max)
);
insert into #product values (1, 'red, blue, green');
insert into #product values (2, null);
insert into #product values (3, 'purple, green');
select
p.ProductId as ProductId,
ltrim(split_table.value) as Color
from #product p
outer apply string_split(p.Color, ',') as split_table;
Fix your database if at all possible. Comma delimited lists in database cells indicate a flawed schema 99% of the time or more.
I would create a CLR table-defined function for this:
http://msdn.microsoft.com/en-us/library/ms254508(VS.80).aspx
The reason for this is that CLR code is going to be much better at parsing apart the strings (computational work) and can pass that information back as a set, which is what SQL Server is really good at (set management).
The CLR function would return a series of records based on the parsed values (and the input id value).
You would then use a CROSS APPLY on each element in your table.
Just convert your columns into xml and query it. Here's an example.
select
a.value('.', 'varchar(42)') c
from (select cast('<r><a>' + replace(#CSV, ',', '</a><a>') + '</a></r>' as xml) x) t1
cross apply x.nodes('//r/a') t2(a)
Why not use dynamic SQL for this purpose, something like this(adapt to your needs):
DECLARE #dynSQL VARCHAR(max)
SET #dynSQL = 'insert into DestinationTable(field) values'
select #dynSQL = #dynSQL + '('+ REPLACE(Color,',',''',''') + '),' from Table
SET #dynSql = LEFT(#dynSql,LEN(#dynSql) -1) -- delete the last comma
exec #dynSql
One advantage is that you can use it on any SQL Server version
I have two tables, in one I have data like this:
id description
2 12.07.13y 1000eur to bank account KZ21321o0002134
4 To bank account KZasd9093636 12 of May 2016y 200dusd
And I have a second table where I need to put filtered information from table first like:
id
data
bank_account
tranfered_money
First i need to split description,then i need to recognize ban_account which always started with "KZ",data and transfered_money
This is just awful but seems to be able to extract the ban_acount:
CREATE TABLE exp
(
column1 varchar(400)
);
Insert into exp (column1) values ('12.07.13y 1000eur to bank account KZ21321o0002134');
Insert into exp (column1) values ('To bank account KZasd9093636 12 of May 2016y 200dusd');
Select
CASE
WHEN CHARINDEX ( SPACE(1), SUBSTRING ( column1, CHARINDEX('KZ' , column1),LEN(column1))) = 0
THEN SUBSTRING ( column1, CHARINDEX('KZ' , column1),LEN(column1))
ELSE SUBSTRING ( SUBSTRING (column1, CHARINDEX('KZ' , column1),LEN(column1)), 0, CHARINDEX (SPACE(1), SUBSTRING(column1, CHARINDEX('KZ' , column1),LEN(column1))))
END result
From exp
At first convert your table to XML.
Then create table with month/weekdays names and digits from 1 to 3000 (or you can take 2016 as current year)
You will need a table with currency. I made one based on data from here.
DECLARE #x xml
;WITH YourTable AS ( --I use this CTE, you should use your table in scripts below
SELECT *
FROM (VALUES
(2, '12.07.13y 1000eur to bank account KZ21321o0002134'),
(4, 'To bank account KZasd9093636 12 of May 2016y 200dusd')
) as t(id, [description])
)
SELECT #x = ( --XML sample that we get you can see below after output
SELECT CAST(N'<row id="'+CAST(id as nvarchar(max))+'"><b>'+REPLACE([description],' ','</b><b>')+'</b></row>' as xml)
FROM YourTable
FOR XML PATH('')
)
;WITH CurrencyList AS ( --Currency table
SELECT *
FROM (VALUES
('AED', 'United Arab Emirates Dirham'),
('AFN', 'Afghanistan Afghani'),
('ALL', 'Albania Lek'),
('AMD', 'Armenia Dram'),
...
('ZAR', 'South Africa Rand'),
('ZMW', 'Zambia Kwacha'),
('ZWD', 'Zimbabwe Dollar')
) as t(code, countryname)
),cte AS ( --generate numbers 1 to 3000
SELECT 0 as d
UNION ALL
SELECT d+1
FROM cte
WHERE d < 3000
), datenames AS ( --generate datenames
SELECT d,
CASE WHEN d < 7 THEN DATENAME(weekday,DATEADD(day,d,'1970-01-01 00:00:00.000')) ELSE NULL END as weekday_name,
CASE WHEN d < 12 THEN DATENAME(month,DATEADD(month,d,'1970-01-01 00:00:00.000')) ELSE NULL END as mon_name
FROM cte
)
--Final query
SELECT t.c.value('../#id','int') as id,
t.c.value('.','nvarchar(max)') as str_part,
CASE WHEN t.c.value('.','nvarchar(max)') LIKE 'KZ%' THEN 'bank_account'
WHEN countryname IS NOT NULL THEN 'tranfered_money'
WHEN dn.d IS NOT NULL OR RIGHT(t.c.value('.','nvarchar(max)'),1) ='y' THEN 'datepart'
ELSE NULL END as what_is
FROM #x.nodes('/row/b') as t(c)
LEFT JOIN CurrencyList cl
ON RIGHT(t.c.value('.','nvarchar(max)'),3) = cl.code --check 3 last symbols of string with currency codes
LEFT JOIN datenames dn
ON dn.d = t.c.value('. cast as xs:int?','int') -- if it is a day/month/year number
OR t.c.value('.','nvarchar(max)') = dn.weekday_name -- or it is a week day name
OR t.c.value('.','nvarchar(max)') = dn.mon_name --or month name
OPTION (MAXRECURSION 0)
Will bring you:
id str_part what_is
2 12.07.13y datepart
2 1000eur tranfered_money
2 to NULL
2 bank NULL
2 account NULL
2 KZ21321o0002134 bank_account
4 To NULL
4 bank NULL
4 account NULL
4 KZasd9093636 bank_account
4 12 datepart
4 of NULL
4 May datepart
4 2016y datepart
4 200dusd tranfered_money
After that you need to bring dates in normal date form and that is all.
XML Sample:
<row id="2">
<b>12.07.13y</b>
<b>1000eur</b>
<b>to</b>
<b>bank</b>
<b>account</b>
<b>KZ21321o0002134</b>
</row>
<row id="4">
<b>To</b>
<b>bank</b>
<b>account</b>
<b>KZasd9093636</b>
<b>12</b>
<b>of</b>
<b>May</b>
<b>2016y</b>
<b>200dusd</b>
</row>
I have a table like this:
Device
DeviceId Parts
1 Part1, Part2, Part3
2 Part2, Part3, Part4
3 Part1
I would like to create a table 'Parts', export data from Parts column to the new table. I will drop the Parts column after that
Expected result
Parts
PartId PartName
1 Part1
2 Part2
3 Part3
4 Part4
DevicePart
DeviceId PartId
1 1
1 2
1 3
2 2
2 3
2 4
3 1
Can I do this in SQL Server 2008 without using cursors?
-- Setup:
declare #Device table(DeviceId int primary key, Parts varchar(1000))
declare #Part table(PartId int identity(1,1) primary key, PartName varchar(100))
declare #DevicePart table(DeviceId int, PartId int)
insert #Device
values
(1, 'Part1, Part2, Part3'),
(2, 'Part2, Part3, Part4'),
(3, 'Part1')
--Script:
declare #DevicePartTemp table(DeviceId int, PartName varchar(100))
insert #DevicePartTemp
select DeviceId, ltrim(x.value('.', 'varchar(100)'))
from
(
select DeviceId, cast('<x>' + replace(Parts, ',', '</x><x>') + '</x>' as xml) XmlColumn
from #Device
)tt
cross apply
XmlColumn.nodes('x') as Nodes(x)
insert #Part
select distinct PartName
from #DevicePartTemp
insert #DevicePart
select tmp.DeviceId, prt.PartId
from #DevicePartTemp tmp
join #Part prt on
prt.PartName = tmp.PartName
-- Result:
select *
from #Part
PartId PartName
----------- ---------
1 Part1
2 Part2
3 Part3
4 Part4
select *
from #DevicePart
DeviceId PartId
----------- -----------
1 1
1 2
1 3
2 2
2 3
2 4
3 1
You will need a Tally table to accomplish this without a cursor.
Follow the instructions to create a tally table here: Tally Tables by Jeff Moden
This script will put the table into your Temp database, so you probably want to change the "Use DB" statement
Then you can run the script below to insert a breakdown of Devices and Parts into a temp table. You should then be able to join on your part table by the part name (to get the ID) and insert into your new DevicePart table.
select *,
--substring(d.parts, 1, t.n)
substring(d.parts, t.n, charindex(', ', d.parts + ', ',t.n) - t.n) 'Part'
into #devicesparts
from device d
cross join tally t
where t.n < (select max(len(parts))+ 1 from device)
and substring(', ' + d.parts, t.n, 1) = ', '
Have a look at using fn_Split to create a table variable from the comma separated values.
You can then use this to drive your insert.
EDIT: Actually, I think you may still need a cursor. Leaving this answer incase fn_Split helps.
If there is a maximum number of parts per device then, yes, it can be done without a cursor, but this is quite complex.
Essentially, create a table (or view or subquery) that has a DeviceID and one PartID column for each possible index in the PartID string. This can be accomplished by making the PartID columns calculated columns using fn_split or another method of your choice. From there you do a multiple self-UNION of this table, with one table in the self-UNION for each PartID column. Each table in the self-UNION has only one of the PartID columns included in the select list of the query for the table.
I have a varchar column in one of my tables with data like:
1234abc
1234abcde456757
1234abc Supervisor
1234abc456 Administrator
I want to "clean it" by removing any letters and numbers immediately following them so for the above examples I want to have:
1234
1234
1234 Supervisor
1234 Administrator
In another word, I want to keep the initial number and the last word. I'm using the SUBSTRING and CHARINDEX but those functions remove everything till the end of the string and I don't know the length of the part I need to remove.
Any suggestions?
Thanks
You could search for the first non-digit and the first space in a subquery. That also works if the number of digits isn't exactly four:
declare #t table (col1 varchar(50))
insert into #t select '12abc'
union all select '1234abcde456757'
union all select '1234abc Supervisor'
union all select '1234abc456 Administrator'
union all select '123456abc456 Administrator'
select case when FirstNonDigit = 0 then col1
when FirstSpace = 0 then substring(col1, 1, FirstNonDigit-1)
else substring(col1, 1, FirstNonDigit-1) +
substring(col1, FirstSpace, len(col1) - FirstSpace + 1)
end
from (
select patindex('%[^0-9]%', col1) FirstNonDigit
, patindex('% %', col1) FirstSpace
, col1
from #t
) subqueryalias
-->
12
1234
1234 Supervisor
1234 Administrator
123456 Administrator
try this:
DECLARE #YourTable table (RowValue varchar(50))
INSERT #YourTable VALUES ('1234abc')
INSERT #YourTable VALUES ('1234abcde456757')
INSERT #YourTable VALUES ('1234abc Supervisor')
INSERT #YourTable VALUES ('1234abc456 Administrator')
UPDATE #YourTable
SET RowValue=LEFT(RowValue,4)+RIGHT(RowValue,CHARINDEX(' ',REVERSE(RowValue)))
FROM #YourTable
SELECT * FROM #YourTable
OUTPUT:
RowValue
--------------------------------------------------
1234
1234
1234 Supervisor
1234 Administrator
(4 row(s) affected)
EDIT: set based any number of digits and handles no digits or no words
DECLARE #YourTable table (RowValue varchar(50))
set nocount on
INSERT #YourTable VALUES ('13')
INSERT #YourTable VALUES ('1234abc')
INSERT #YourTable VALUES ('1234abc')
INSERT #YourTable VALUES ('1234abcde456757')
INSERT #YourTable VALUES ('1234abc Supervisor')
INSERT #YourTable VALUES ('1234abc456 Administrator')
INSERT #YourTable VALUES ('1234567abc456 Administrator')
INSERT #YourTable VALUES ('Administrator')
INSERT #YourTable VALUES ('abcde Administrator')
set nocount off
;WITH Digits AS
(SELECT 0 AS Digit UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9
)
,Numbers AS
(SELECT 1 AS Number
UNION ALL
SELECT Number+1 FROM Numbers where Number<1000
)
,FindDigits AS
(
SELECT
y.RowValue,n.Number,SUBSTRING(y.RowValue,n.Number,1) AS CharOf,CASE WHEN SUBSTRING(y.RowValue,n.Number,1) LIKE '[0-9]' THEN 'N' ELSE 'A' END AS TypeOf
FROM #YourTable y
INNER JOIN Numbers n ON 1=1
WHERE n.Number<=LEN(y.RowValue)
)
,LenOf AS
(
SELECT
RowValue,MIN(Number)-1 AS Digits
FROM FindDigits
WHERE TypeOf='A'
GROUP BY RowValue
HAVING MIN(Number)-1>0
UNION
SELECT
f.RowValue,LEN(f.RowValue)
FROM FindDigits f
WHERE NOT EXISTS (SELECT 1 FROM FindDigits f2 WHERE f.RowValue=f2.RowValue AND TypeOf='A')
)
UPDATE y
SET RowValue=CASE WHEN l.Digits IS NOT NULL THEN LEFT(y.RowValue,l.Digits)+RIGHT(y.RowValue,CHARINDEX(' ',REVERSE(y.RowValue)))
WHEN CHARINDEX(' ',REVERSE(y.RowValue))=0 THEN y.RowValue
ELSE RIGHT(y.RowValue,CHARINDEX(' ',REVERSE(y.RowValue))-1) END
FROM #YourTable y
LEFT JOIN LenOf l ON y.RowValue=l.RowValue
OPTION (MAXRECURSION 1000)
SELECT * FROM #YourTable
OUTPUT:
RowValue
--------------------------------------------------
13
1234
1234
1234
1234 Supervisor
1234 Administrator
1234567 Administrator
Administrator
Administrator
(9 row(s) affected)
You actually want two strings, the characters at indices 0-3 and those from the position after the space till the end of the string. I (think) this will work (have not tried it):
UPDATE TableName SET ColumnName = SUBSTRING(ColumnName,1,4) +
SUBSTRING(ColumnName,CHARINDEX(' ',ColumnName)+1,LEN(ColumnName))
The code below uses a "tally table" of values to find the first non-numeric character and the last space. KM's solution using PATINDEX is probably more elegant!
DECLARE #t TABLE
(
c VARCHAR(MAX)
);
INSERT INTO #t VALUES('1234abc');
INSERT INTO #t VALUES('1234abcde456757');
INSERT INTO #t VALUES('1234abc Supervisor');
INSERT INTO #t VALUES('1234abc456 Administrator');
WITH Tally AS
(
SELECT ROW_NUMBER() OVER (ORDER BY s1.[id]) AS i
FROM sys.sysobjects s1 CROSS JOIN sys.sysobjects s2 CROSS JOIN sys.sysobjects s3
),
NumPart AS
(
SELECT c, MIN(i) AS firstNonNumber
FROM #t CROSS JOIN Tally
WHERE i <= LEN(c)
AND SUBSTRING(c, i, 1) < '0' OR SUBSTRING(c, i, 1) > '9'
GROUP BY c
),
SpacePart AS
(
SELECT c, MAX(i) AS spacePos
FROM #t t CROSS JOIN Tally
WHERE i <= LEN(c)
AND SUBSTRING(c, i, 1) = ' '
GROUP BY c
)
UPDATE t
SET t.c = LEFT(n.c, n.firstNonNumber - 1) +
CASE WHEN ISNULL(s.SpacePos, 0) > 0 THEN
RIGHT(n.c, LEN(n.c) - s.SpacePos + 1)
ELSE
''
END
FROM #t t
INNER JOIN NumPart n ON t.c = n.c
LEFT JOIN SpacePart s ON n.c = s.c;
SELECT * FROM #t;
I have a table that looks like this:
ProductId, Color
"1", "red, blue, green"
"2", null
"3", "purple, green"
And I want to expand it to this:
ProductId, Color
1, red
1, blue
1, green
2, null
3, purple
3, green
Whats the easiest way to accomplish this? Is it possible without a loop in a proc?
Take a look at this function. I've done similar tricks to split and transpose data in Oracle. Loop over the data inserting the decoded values into a temp table. The convent thing is that MS will let you do this on the fly, while Oracle requires an explicit temp table.
MS SQL Split Function
Better Split Function
Edit by author:
This worked great. Final code looked like this (after creating the split function):
select pv.productid, colortable.items as color
from product p
cross apply split(p.color, ',') as colortable
based on your tables:
create table test_table
(
ProductId int
,Color varchar(100)
)
insert into test_table values (1, 'red, blue, green')
insert into test_table values (2, null)
insert into test_table values (3, 'purple, green')
create a new table like this:
CREATE TABLE Numbers
(
Number int not null primary key
)
that has rows containing values 1 to 8000 or so.
this will return what you want:
EDIT
here is a much better query, slightly modified from the great answer from #Christopher Klein:
I added the "LTRIM()" so the spaces in the color list, would be handled properly: "red, blue, green". His solution requires no spaces "red,blue,green". Also, I prefer to use my own Number table and not use master.dbo.spt_values, this allows the removal of one derived table too.
SELECT
ProductId, LEFT(PartialColor, CHARINDEX(',', PartialColor + ',')-1) as SplitColor
FROM (SELECT
t.ProductId, LTRIM(SUBSTRING(t.Color, n.Number, 200)) AS PartialColor
FROM test_table t
LEFT OUTER JOIN Numbers n ON n.Number<=LEN(t.Color) AND SUBSTRING(',' + t.Color, n.Number, 1) = ','
) t
EDIT END
SELECT
ProductId, Color --,number
FROM (SELECT
ProductId
,CASE
WHEN LEN(List2)>0 THEN LTRIM(RTRIM(SUBSTRING(List2, number+1, CHARINDEX(',', List2, number+1)-number - 1)))
ELSE NULL
END AS Color
,Number
FROM (
SELECT ProductId,',' + Color + ',' AS List2
FROM test_table
) AS dt
LEFT OUTER JOIN Numbers n ON (n.Number < LEN(dt.List2)) OR (n.Number=1 AND dt.List2 IS NULL)
WHERE SUBSTRING(List2, number, 1) = ',' OR List2 IS NULL
) dt2
ORDER BY ProductId, Number, Color
here is my result set:
ProductId Color
----------- --------------
1 red
1 blue
1 green
2 NULL
3 purple
3 green
(6 row(s) affected)
which is the same order you want...
You can try this out, doesnt require any additional functions:
declare #t table (col1 varchar(10), col2 varchar(200))
insert #t
select '1', 'red,blue,green'
union all select '2', NULL
union all select '3', 'green,purple'
select col1, left(d, charindex(',', d + ',')-1) as e from (
select *, substring(col2, number, 200) as d from #t col1 left join
(select distinct number from master.dbo.spt_values where number between 1 and 200) col2
on substring(',' + col2, number, 1) = ',') t
I arrived this question 10 years after the post.
SQL server 2016 added STRING_SPLIT function.
By using that, this can be written as below.
declare #product table
(
ProductId int,
Color varchar(max)
);
insert into #product values (1, 'red, blue, green');
insert into #product values (2, null);
insert into #product values (3, 'purple, green');
select
p.ProductId as ProductId,
ltrim(split_table.value) as Color
from #product p
outer apply string_split(p.Color, ',') as split_table;
Fix your database if at all possible. Comma delimited lists in database cells indicate a flawed schema 99% of the time or more.
I would create a CLR table-defined function for this:
http://msdn.microsoft.com/en-us/library/ms254508(VS.80).aspx
The reason for this is that CLR code is going to be much better at parsing apart the strings (computational work) and can pass that information back as a set, which is what SQL Server is really good at (set management).
The CLR function would return a series of records based on the parsed values (and the input id value).
You would then use a CROSS APPLY on each element in your table.
Just convert your columns into xml and query it. Here's an example.
select
a.value('.', 'varchar(42)') c
from (select cast('<r><a>' + replace(#CSV, ',', '</a><a>') + '</a></r>' as xml) x) t1
cross apply x.nodes('//r/a') t2(a)
Why not use dynamic SQL for this purpose, something like this(adapt to your needs):
DECLARE #dynSQL VARCHAR(max)
SET #dynSQL = 'insert into DestinationTable(field) values'
select #dynSQL = #dynSQL + '('+ REPLACE(Color,',',''',''') + '),' from Table
SET #dynSql = LEFT(#dynSql,LEN(#dynSql) -1) -- delete the last comma
exec #dynSql
One advantage is that you can use it on any SQL Server version