I am passing one string to store procedure : 1:20,2:30,4:50
It contains id and appropriate value for it.
how can I add value as shown in below table in database.
ID Value
1 20
2 30
4 50
I have already "stringSplit" function which works perfectly and gives out put in row value some think like this :
1:20
2:30
4:50
can anyone please help me to insert data into table with any solution.
i already try this solution
insert <table> (colname)
select y.item
from dbo.SplitString(#testString, ':') x
cross apply
dbo.SplitString(x.item, ',') y
but this will return duplicate value as more as id value.
my store procedure is
CREATE PROCEDURE [dbo].[temp_result_insert]
#dataString varchar(max)
AS
insert into tempTable(id,marks)
select x.Item,y.Item
from dbo.SplitStringVarcahr(#dataString, ':') x
cross apply
dbo.SplitStringVarcahr(x.Item,',') y
RETURN 0
As you already splitted into rows and you want insert into some table by splliting into two columns may be this works
CREATE TABLE #Test(ID INT,Val INT)
declare #t table (val varchar(50))
insert into #t (val)values ('1:20,2:30,4:50')
declare #str varchar(max)
;with cte as (
SELECT
Split.a.value('.', 'VARCHAR(100)') AS String
FROM (SELECT
CAST ('<M>' + REPLACE([val], ',', '</M><M>') + '</M>' AS XML) AS String
FROM #t) AS A CROSS APPLY String.nodes ('/M') AS Split(a))
INSERT INTO #Test
select SUBSTRING(String,0,CHARINDEX(':',String)),REVERSE(SUBSTRING(reverse(String),0,CHARINDEX(':',reverse(String)))) from cte
select * from #test
You can also try XML.nodes() and string functions to spit the data. Something like this
DECLARE #var VARCHAR(100) = '1:20,2:30,4:50'
DECLARE #xml xml = CONVERT(xml, '<r>' + REPLACE(#var,',','</r><r>') + '</r>')
SELECT LEFT(val,cindex - 1) c1,RIGHT(val,LEN(val) - cindex) c2
FROM
(
SELECT CHARINDEX(':',c.value('text()[1]','VARCHAR(100)')) cindex,c.value('text()[1]','VARCHAR(100)') val
FROM #xml.nodes('r') as t(c))c
Use substring and Charindex:
SELECT Substring(col, 0, Charindex(col, ':') - 1) AS id,
Substring(col, Charindex(col, ':') + 1, Len(col)-Charindex(col, ':')) AS value
FROM splittedtable
Related
I wish to loop through two comma-separated values and perform an insert
As an example lets consider two variables
Declare #Qid= 1,4,6,7,8 #Answers = 4,4,3,2,3
set #pos = 0
set #len = 0
WHILE CHARINDEX(',', #Answers, #pos+1)>0
BEGIN
set #len = CHARINDEX(',', #Answers, #pos+1) - #pos
set #value = SUBSTRING(#Answers, #pos, #len)
insert into table values(#fdid,#Qid,#fusid, #value) -- i need Qid also
set #pos = CHARINDEX(',', #Answers, #pos+#len) +1
END
Using this loop I am able to extract #Answers and can perform insert. But I wish to extract #Qid and insert inside the loop.
edit
for more clarity it is a feedback module. my result table have Qid and Answer field. Answers are ratings (1 to 5). The values we get in variables #Qid and #Answers are sequential. which means 1st answer will be for 1st question and so on.
edit
as per Shnugo's Answer
Declare #Qid varchar(100)= '1,4,6,7,8', #Answers varchar(100)= '4,4,3,2,3'
DECLARE #tbl TABLE(ID INT IDENTITY, Questions VARCHAR(100),Answers VARCHAR(100));
INSERT INTO #tbl VALUES(#Qid,#Answers)
INSERT INTO table(FeedbackId,QuestionId,FeedbackUserId,Answer)
SELECT 1,
A.qXml.value('(/x[sql:column("B.QuestionCount")])[1]','int') AS QuestionNumber,3
,A.aXml.value('(/x[sql:column("B.QuestionCount")])[1]','int') AS AnwerNumber
FROM #tbl t
CROSS APPLY(SELECT CAST('<x>' + REPLACE(#Qid,',','</x><x>') + '</x>' AS XML)
,CAST('<x>' + REPLACE(#Answers,',','</x><x>') + '</x>' AS XML)) A(qXml,aXml)
CROSS APPLY(SELECT TOP(A.qXml.value('count(/x)','int')) ROW_NUMBER() OVER(ORDER BY (SELECT NULL)) FROM master..spt_values) B(QuestionCount)
I'd prefer Zhorov's JSON answer (needs v2016+).
If you use a SQL-Server below 2016 you might use this position-safe XML-based solution:
A mockup table to simulate your issue with two different rows.
DECLARE #tbl TABLE(ID INT IDENTITY, Questions VARCHAR(100),Answers VARCHAR(100));
INSERT INTO #tbl VALUES('1,4,6,7,8','4,4,3,2,3')
,('1,2,3','4,5,6');
--The query
SELECT t.*
,A.qXml.value('(/x[sql:column("B.QuestionCount")])[1]','int') AS QuestionNumber
,A.aXml.value('(/x[sql:column("B.QuestionCount")])[1]','int') AS AnwerNumber
FROM #tbl t
CROSS APPLY(SELECT CAST('<x>' + REPLACE(t.Questions,',','</x><x>') + '</x>' AS XML)
,CAST('<x>' + REPLACE(t.Answers,',','</x><x>') + '</x>' AS XML)) A(qXml,aXml)
CROSS APPLY(SELECT TOP(A.qXml.value('count(/x)','int')) ROW_NUMBER() OVER(ORDER BY (SELECT NULL)) FROM master..spt_values) B(QuestionCount);
The idea in short:
We need a CROSS APPLY and some string methods to transform something like 1,2,3 to an xml like <x>1</x><x>2</x><x>3</x>.
Now we can use value() with XQuery count() to find the actual count of questions.
We need one more CROSS APPLY with a computed TOP() clause to get a set of running number from 1 to n with n=countOfQuestions. I do this against master..spt_values. This is just a well-filled standard table... We do not need the values, just any set to create the counter...
Finally we can use .value() in connection with sql:column() in order to fetch the question and the corresponding answer by their positions.
UPDATE: Non-tabular data
If you do not get these CSV parameters as a table you can use this:
Declare #Qid varchar(100)= '1,4,6,7,8', #Answers varchar(100)= '4,4,3,2,3'
--INSERT INTO table(FeedbackId,QuestionId,FeedbackUserId,Answer)
SELECT 1
,A.qXml.value('(/x[sql:column("B.QuestionCount")])[1]','int') AS QuestionNumber
,3
,A.aXml.value('(/x[sql:column("B.QuestionCount")])[1]','int') AS AnwerNumber
FROM (SELECT CAST('<x>' + REPLACE(#Qid,',','</x><x>') + '</x>' AS XML)
,CAST('<x>' + REPLACE(#Answers,',','</x><x>') + '</x>' AS XML)) A(qXml,aXml)
CROSS APPLY(SELECT TOP(A.qXml.value('count(/x)','int')) ROW_NUMBER() OVER(ORDER BY (SELECT NULL)) FROM master..spt_values) B(QuestionCount);
If you use SQL Server 2016 or higher, you may try to use the next JSON-based approach to map questions and answers by their positions in the input strings. You need to transform the input strings into valid JSON arrays and then use OPENJSON() with default schema to parse the arrays. The result is a table, with columns key, value and type and the key column holds the index of the element in the specified array.
Note, that STRING_SPLIT() function does not guarantee the order of the rows and the output rows might be in any order.
Statement:
DECLARE #Qid nvarchar(max) = N'1,4,6,7,8'
DECLARE #Answers nvarchar(max) = N'4,4,3,2,3'
-- Build your INSERT statement as you expect
-- INSERT INTO Table ...
SELECT j1.[value] AS Qid, j2.[value] AS Answers
FROM OPENJSON(CONCAT(N'[', #Qid, N']')) j1
JOIN OPENJSON(CONCAT(N'[', #Answers, N']')) j2 ON j1.[key] = j2.[key]
Result from the SELECT statement:
Qid Answers
1 4
4 4
6 3
7 2
8 3
You have not described relationship of question and its answers. I feel its one to one relationship and for that, I have given the answer.
declare #Qid varchar(200)= '1,4,6,7,8' , #Answers varchar(200) = '4,4,3,2,3'
;with cte
as(
select id, data qid from dbo.Split (#qid, ',')
),
cte1 as
(
select id, data ansid from dbo.Split (#answers, ',')
)
--insert into tablename
select
qid, ansid from cte join cte1 on cte.id = cte1.id
Result will be:
qid ansid
1 4
4 4
6 3
7 2
8 3
See other option for later version of sqlserver : Split function equivalent in T-SQL?
There are some objects encoded as key:value strings and stored in a table, I'd like to increase sequence number of all objects, which is one field in the object.
For example:
ID Value
--------------------------
504 s:0;d:n;e:test;
506 s:1;d:y;e:branch;
507 s:2;d:y;e:;
I'd like to change them to:
ID Value
--------------------------
504 s:1;d:n;e:test;
506 s:2;d:y;e:branch;
507 s:3;d:y;e:;
Is there a simple way to do this?
Is there a simple way to do this?
No not really.
You can find the positions of s: and d: and then use that to extract the number inbetween, increase it by one and stuff it back into where it belongs.
declare #T table
(
ID int,
Value varchar(50)
);
insert into #T values
(504, 's:0;d:n;e:test;'),
(506, 's:1;d:y;e:branch;'),
(507, 's:2;d:y;e:;');
select T.ID,
stuff(T.Value, P.S, P.D - P.S - 1, S.Value) as NewValue
from #T as T
cross apply (values(charindex('s:', T.Value) + 2,
charindex('d:', T.Value))) as P(S, D)
cross apply (values(substring(T.Value, P.S, P.D - P.S - 1) + 1)) as S(Value)
A version where you find the ; after s: instead of d: as suggested by Eric in a comment.
select T.ID,
stuff(T.Value, S.Pos, SEnd.Pos - S.Pos, V.NewValue) as NewValue
from #T as T
cross apply (values(charindex('s:', T.Value) + 2)) as S(Pos)
cross apply (values(charindex(';', T.Value, S.Pos))) as SEnd(Pos)
cross apply (values(substring(T.Value, S.Pos, SEnd.Pos - S.Pos) + 1)) as V(NewValue)
DECLARE #val nvarchar(200)
SET #val = 's:1;d:y;e:branch;'
SELECT 's:' + CONVERT(nvarchar(100), CONVERT(INT, SUBSTRING(#val, charindex(':', #val) + 1, charindex(';', #val) - charindex(':', #val) -1)) + 1) + SUBSTRING(#val, charindex(':', #val),1000)
You can use what's in the SELECT's query in an UPDATE statement to change the table values
Using the split string functions from here:Split strings the right way – or the next best way
declare #string varchar(max)
set #string='504 s:0;d:n;e:test;'
;with cte as(select * from
[dbo].[SplitStrings_Numbers]
(#string,':'))
select b.item+1 from cte c
cross apply
(select * from [dbo].[SplitStrings_Numbers](c.item,';')) b
where isnumeric(b.item)=1
This accounts for empty or non-integer values; it will ignore them in the event they can't be incremented by one.
-- Build Test Data
IF OBJECT_ID('tempdb..#test') IS NOT NULL DROP TABLE #test
CREATE TABLE #test (ID INT, Value VARCHAR(100))
INSERT #test
VALUES
(504,'s:0;d:n;e:test;'),
(506,'s:1;d:y;e:branch;'),
(507,'s:2;d:y;e:;'),
(508,'s:;d:y;e:;'),
(509,'s:xyz;d:y;e:;');
-- Update S: values
WITH sVals AS
(
SELECT ID, Value, TRY_PARSE(SUBSTRING(Value,CHARINDEX('s:',Value)+2,CHARINDEX(';',Value,CHARINDEX('s:',Value))-(CHARINDEX('s:',Value)+2)) AS INT) AS sVal
FROM #test AS t
)
UPDATE s
SET Value = IIF(sVal IS NOT NULL, STUFF(Value,CHARINDEX('s:',Value)+2,CHARINDEX(';',Value,CHARINDEX('s:',Value))-(CHARINDEX('s:',Value)+2),sVal+1), Value)
FROM sVals AS s
-- Check the results
SELECT *
FROM #test
You can as the below:
DECLARE #val VARCHAR(100) = 's:12;d:n;e:test;'
SELECT REPLACE(#val, ':' + SUBSTRING(#val, 3, PATINDEX('%;d:%', #val) - 3) + ';', ':' + CAST(SUBSTRING(#val, 3, PATINDEX('%;d:%', #val) - 3)+ 1 AS VARCHAR(MAX)) + ';')
Result: s:13;d:n;e:test;
from table I retrieves values, for example,
7752652:1,7752653:2,7752654:3,7752655:4
or
7752941:1,7752942:2
i.e. string may contain any quantity of substrings.
What I need: remove all occurrences of characters from char ':' to a comma char.
For example,
7752652:1,7752653:2,7752654:3,7752655:4
should be
7752652,7752653,7752654,7752655
How do it?
Replace : with start tag <X>.
Replace , with end tag </X> and an extra comma.
Add an extra end tag to the end </X>.
That will give you a string that look like 7752941<X>1</X>,7752942<X>2</X>.
Cast to XML and use query(text()) to get the root text values.
Cast the result back to string.
SQL Fiddle
MS SQL Server 2012 Schema Setup:
create table T
(
C varchar(100)
)
insert into T values
('7752652:1,7752653:2,7752654:3,7752655:4'),
('7752941:1,7752942:2')
Query 1:
select cast(cast(replace(replace(T.C, ':', '<X>'), ',', '</X>,')+'</X>' as xml).query('text()') as varchar(100)) as C
from T
Results:
| C |
|---------------------------------|
| 7752652,7752653,7752654,7752655 |
| 7752941,7752942 |
declare #query varchar(8000)
select #query= 'select '+ replace (
replace('7752652:1,7752653:2,7752654:3,7752655:4',',',' t union all select ')
,':',' t1 , ')
exec(';with cte as ( '+#query+' ) select cast(t1 as varchar)+'','' from cte for xml path('''')')
Try this:
DECLARE #Data VARCHAR(100) = '7752652:1,7752653:2,7752654:3,7752655:4'
DECLARE #Output VARCHAR(100) = ''
WHILE CHARINDEX(':', #Data) > 0
BEGIN
IF LEN(#Output) > 0 SET #Output = #Output + ','
SET #Output = #Output + LEFT(#Data, CHARINDEX(':', #Data)-1)
SET #Data = STUFF(#Data,
1,
(CASE CHARINDEX(',', #Data)
WHEN 0 THEN LEN(#Data)
ELSE CHARINDEX(',', #Data)
END) - CHARINDEX(':', #Data),
'')
END
SELECT #Output AS Result -- 7752652,7752653,7752654,7752655
Hope this will help.
I borrowed the Splitter function from here. You could use any delimiter parser you may already be using.
Parse the string to table values
Used Substring function to remove values after ':'
Use For xml to re-generate CSV
Test Data:'
IF OBJECT_ID(N'tempdb..#temp')>0
DROP TABLE #temp
CREATE TABLE #temp (id int, StringCSV VARCHAR(500))
INSERT INTO #temp VALUES ('1','7752652:1,7752653:2,7752654:3,7752655:4')
INSERT INTO #temp VALUES ('2','7752656:1,7752657:3,7752658:4')
INSERT INTO #temp VALUES ('3','7752659:1,7752660:2')
SELECT * FROM #temp t
Main Query:
;WITH cte_Remove(ID, REMOVE) AS
(
SELECT y.id AS ID,
SUBSTRING(fn.string, 1, CHARINDEX(':', fn.string) -1) AS Removed
FROM #temp AS y
CROSS APPLY dbo.fnParseStringTSQL(y.StringCSV, ',') AS fn
)
SELECT DISTINCT ID,
STUFF(
(
SELECT ',' + REMOVE
FROM cte_Remove AS t2
WHERE t2.ID = t1.ID
FOR XML PATH('')
),1,1,'') AS col2
FROM cte_Remove AS t1
Cleanup Test Data:
IF OBJECT_ID(N'tempdb..#temp') > 0
DROP TABLE #temp
I solved this problem with CLR function. It is more quickly and function can be used in complex queries
public static SqlString fnRemoveSuffics(SqlString source)
{
string pattern = #":(\d+)";
string replacement = "";
string result = Regex.Replace(source.Value, pattern, replacement);
return new SqlString(result);
}
i have table with one column having comma seperated values and I want in row..
like
col1
3,4,5
5,6,6
return result should be
col1
3
4
5
5
6
6
declare #tbl table(data nvarchar(100))
insert into #tbl values('1,2,3')
insert into #tbl values('4,5,6')
insert into #tbl values('7,8,9')
SELECT
Split.a.value('.', 'VARCHAR(100)') AS String
FROM (SELECT data,
CAST ('<M>' + REPLACE(data, ',', '</M><M>') + '</M>' AS XML) AS String
FROM #tbl) AS A CROSS APPLY String.nodes ('/M') AS Split(a);
I believe the below explains how to loop through comma separated values. You could just insert them into another variable to get your required output.
Splitting of comma separated values
You can use a recursive query to perform this split:
;with cte (item, list) as
(
select
cast(left(col1, charindex(',',col1+',')-1) as varchar(50)) item,
stuff(col1, 1, charindex(',',col1+','), '') list
from yourtable
union all
select
cast(left(list, charindex(',',list+',')-1) as varchar(50)) item,
stuff(list, 1, charindex(',',list+','), '') list
from cte
where list > ''
)
select item
from cte
See SQL Fiddle with Demo
How can I get each value of a column that has a comma separator in her value ?
Example:
ID ColumnUnified
1 12,34,56,78
2 80,99,70,56
What I want is a query to get the number without comma. If possible, in collumns.
12 34 56 78
This will work for any number of values http://beyondrelational.com/modules/2/blogs/70/posts/10844/splitting-delimited-data-to-columns-set-based-approach.aspx
The solution Madhivanan's link refers to is very creative, but I had a slight problem with it on SQL Server 2012 related to the name of one of the columns (Start). I've modified the code in his answer to use StartPos instead of Start for the column name.
I was not familiar with the system procedure spt_values, but I found a very informative description of the procedure here on SO for those who are interested in exactly how this solution works.
Finally, here's the (slightly) revised code from Madhivana's answer:
CREATE TABLE #test(id int, data varchar(100))
INSERT INTO #test VALUES (1,'This,is,a,test,string')
INSERT INTO #test VALUES (2,'See,if,it,can,be,split,into,many,columns')
DECLARE #pivot varchar(8000)
DECLARE #select varchar(8000)
SELECT #pivot = COALESCE(#pivot + ',', '') + '[col'
+ CAST(number + 1 AS VARCHAR(10)) + ']'
FROM master..spt_values
WHERE type = 'p'
AND number <= ( SELECT MAX(LEN(data) - LEN(REPLACE(data, ',', '')))
FROM #test
)
SELECT #select = '
select p.*
from (
select
id,substring(data, StartPos+2, endPos-StartPos-2) as token,
''col''+cast(row_number() over(partition by id order by StartPos) as varchar(10)) as n
from (
select
id, data, n as StartPos, charindex('','',data,n+2) endPos
from (select number as n from master..spt_values where type=''p'') num
cross join
(
select
id, '','' + data +'','' as data
from
#test
) m
where n < len(data)-1
and substring(data,n+1,1) = '','') as data
) pvt
Pivot ( max(token)for n in (' + #pivot + '))p'
EXEC(#select)
DROP TABLE #test