I'm trying to parse only numbers from a string. My code must be pretty close, but something is off here, because several numbers in the last string are being truncated, although the first two strings seem fine.
Here is my code.
Drop Table SampleData
Create table SampleData
(id int, factor varchar(100))
insert into #source_Policy values (1 ,'AAA 1.058 (Protection Class)')
insert into #source_Policy values (2, 'BBB0.565 (Construction) ')
insert into #source_Policy values ( 3, 'CCCCC 1.04890616 (Building Limit Rel')
Select *
From SampleData
;with processTable as (
select id, factor, num
from SampleData
cross apply (
select (select C + ''
from (select N, substring(factor, N, 1) C from (values(1),(2),(3),(4),(5),(6),(7),(8),(9),(10),(11),(12)) Num(N) where N<=datalength(factor)) t
where PATINDEX('%[0-9.]%',C)> 0
order by N
for xml path(''))
) p0 (num)
)
SELECT id, factor, num
FROM processTable
This is the result that I get.
In the num column, instead of 1.04, I would like to see the full precision, so: 1.04890616
I would think something like this:
select s.*, v2.numstr
from sampledata s cross apply
(values (stuff(factor, 1, patindex('%[0-9]%', factor) - 1, ''))) v(str) cross apply
(values (left(v.str, patindex('%[^0-9.]%', v.str + 'x') - 1))) v2(numstr);
Here is a SQL Fiddle.
Related
In SQL I want to add 0 in front and , at the end of each character.
Example: A30F1 -> 0A,03,00,0F,01
I don't want to use cursor if possible.
Thanks!
EIDT:
I apologize for not asking the most appropriate question at the beginning.
In short, I have a table and for each value in the column name I have to convert it to the desired format. For example, we have a #Temp table:
CREATE TABLE #Temp (id INT, name VARCHAR(25))
INSERT INTO #Temp VALUES (1, 'A30F1'), (2, 'B51R9'), (3, 'L1721')
SELECT * FROM #Temp
One method would be to use a Tally to split the string into it's individual characters, and then use concatenation to add the 0 to the start, and STRING_AGG to comma delimit the results:
DECLARE #YourValue varchar(5) = 'A30F1';
WITH N AS(
SELECT N
FROM (VALUES(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL))N(N)),
Tally AS(
SELECT TOP (LEN(#YourValue))
ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) AS I
FROM N N1, N N2) --Up to 100 characters, add more cross joins for more characters
SELECT STRING_AGG(CONCAT('0',SS.C),',') WITHIN GROUP (ORDER BY T.I) AS NewString
FROM (VALUES(#YourValue))V(YourValue)
CROSS JOIN Tally T
CROSS APPLY (VALUES(SUBSTRING(V.YourValue,T.I,1)))SS(C);
It appears this is meant to be against a table, not a single value. This needs, however, very few changes to work against a table:
WITH N AS(
SELECT N
FROM (VALUES(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL),(NULL))N(N)),
Tally AS(
SELECT TOP (SELECT MAX(LEN(YourColumn)) FROM dbo.YourTable)
ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) AS I
FROM N N1, N N2) --Up to 100 characters, add more cross joins for more characters
SELECT STRING_AGG(CONCAT('0',SS.C),',') WITHIN GROUP (ORDER BY T.I) AS NewString
FROM dbo.YourTable YT
JOIN Tally T ON LEN(YT.YourColumn) >= T.I
CROSS APPLY (VALUES(SUBSTRING(YT.YourColumn,T.I,1)))SS(C)
GROUP BY YT.YourColumn;
db<>fiddle
I solved the simplest possible with a few variables, WHILE and SUBSTRING
DECLARE #var VARCHAR(20) = 'A30F1', #i INT = 1, #res NVARCHAR(20)
WHILE (#i <= LEN(#var))
BEGIN
SET #res = #res + '0' + SUBSTRING(#var, #i, 1) + ','
SET #i = #i + 1
END
SELECT LEFT(#res, LEN(#res) - 1) output
Check demo on DB<>FIDDLE.
Original answer:
A recursive CTE and a STRING_AGG() call is also an option (SQL Server 2017+ is needed):
DECLARE #text varchar(max) = 'A30F1';
WITH rCTE AS
(
SELECT 1 AS CharacterPosition, SUBSTRING(#text, 1, 1) AS Character
UNION ALL
SELECT CharacterPosition + 1, SUBSTRING(#text, CharacterPosition + 1, 1)
FROM rCTE
WHERE CharacterPosition < LEN(#text)
)
SELECT STRING_AGG('0' + Character, ',') WITHIN GROUP (ORDER BY CharacterPosition)
FROM rCTE
OPTION (MAXRECURSION 0);
Update:
You need a different statement, if the names are stored in a table, again using recursion and STRING_AGG():
Table:
CREATE TABLE #Temp (id INT, name VARCHAR(25))
INSERT INTO #Temp VALUES (1, 'A30F1'), (2, 'B51R9'), (3, 'L1721')
Statement:
; WITH rCTE AS (
SELECT
t.id AS id,
LEFT(t.name, 1) AS Character,
STUFF(t.name, 1, 1, '') AS CharactersRemaining,
1 AS CharacterPosition
FROM #Temp t
UNION ALL
SELECT
r.id,
LEFT(r.CharactersRemaining, 1),
STUFF(r.CharactersRemaining, 1, 1, ''),
CharacterPosition + 1
FROM rCTE r
WHERE LEN(r.CharactersRemaining) > 0
)
SELECT
id,
STRING_AGG('0' + Character, ',') WITHIN GROUP (ORDER BY CharacterPosition) AS name
FROM rCTE
GROUP BY id
OPTION (MAXRECURSION 0);
Result:
id name
1 0A,03,00,0F,01
2 0B,05,01,0R,09
3 0L,01,07,02,01
If you are only applying this to English alphabet characters and digits as in your example you could do this.
CREATE TABLE #Temp (id INT, name VARCHAR(25))
INSERT INTO #Temp VALUES (1, 'A30F1'), (2, 'B51R9'), (3, 'L1721'), (4, 'A')
SELECT SUBSTRING(REPLACE(
0x00 + CAST(CAST(name AS NVARCHAR(25)) AS BINARY(50)), CHAR(0), '0,')
, 3
, LEN(name) * 3 - 1)
FROM #Temp
returns
0A,03,00,0F,01
0B,05,01,0R,09
0L,01,07,02,01
0A
This takes advantage of the fact that the binary representation of the nvarchar and varchar is the same for this limited character set except for padding out with 0x00
'A30F1' -> 0x4133304631
N'A30F1' -> 0x41003300300046003100
Azure MSSQL Database
I have a column that contains values stored per transaction. The string can contain up to 7 values, separated by a '-'.
I need to be able to extract the value that is stored after the 3rd '-'. The issue is that the length of this column (and the characters that come before the 3rd '-') can vary.
For example:
DIM VALUE
1. NHL--WA-S-MOSG-SER-
2. VDS----HAST-SER-
3. ---D---SER
Row 1 needs to return 'S'
Row 2 needs to return '-'
Row 3 needs to return 'D'
This is by no means an optimal solution, but it works in SQL Server. 😊
TempTable added for testing purposes. Maybe it gives you a hint as of where to start.
Edit: added reference for string_split function (works from SQL Server 2016 up).
CREATE TABLE #tempStrings (
VAL VARCHAR(30)
);
INSERT INTO #tempStrings VALUES ('NHL--WA-S-MOSG-SER-');
INSERT INTO #tempStrings VALUES ('VDS----HAST-SER-');
INSERT INTO #tempStrings VALUES ('---D---SER');
INSERT INTO #tempStrings VALUES ('A-V-D-C--SER');
SELECT
t.VAL,
CASE t.PART WHEN '' THEN '-' ELSE t.PART END AS PART
FROM
(SELECT
t.VAL,
ROW_NUMBER() OVER (PARTITION BY VAL ORDER BY (SELECT NULL)) AS IX,
value AS PART
FROM #tempStrings t
CROSS APPLY string_split(VAL, '-')) t
WHERE t.IX = 4; --DASH COUNT + 1
DROP TABLE #tempStrings;
Output is...
VAL PART
---D---SER D
A-V-D-C--SER C
NHL--WA-S-MOSG-SER- S
VDS----HAST-SER- -
If you always want the fourth element then using CHARINDEX is relatively straightforward:
DROP TABLE IF EXISTS #tmp;
CREATE TABLE #tmp (
rowId INT IDENTITY PRIMARY KEY,
xval VARCHAR(30) NOT NULL
);
INSERT INTO #tmp
VALUES
( 'NHL--WA-S-MOSG-SER-' ),
( 'VDS----HAST-SER-' ),
( '---D---SER' ),
( 'A-V-D-C--SER' );
;WITH cte AS
( -- Work out the position of the 3rd dash
SELECT
rowId,
xval,
CHARINDEX( '-', xval, CHARINDEX( '-', xval, CHARINDEX( '-', xval ) + 1 ) + 1 ) + 1 xstart
FROM #tmp t
), cte2 AS
( -- Work out the length for the substring function
SELECT rowId, xval, xstart, CHARINDEX( '-', xval, xstart) - (xstart) AS xlen
FROM cte
)
SELECT rowId, ISNULL( NULLIF( SUBSTRING( xval, xstart, xlen ), '' ), '-' ) xpart
FROM cte2
I also did a volume test at 1 million rows and this was by far the fastest method compared with STRING_SPLIT, OPENJSON, recursive CTE (the worst at high volume). As a downside this method is less extensible, say you want the second or fifth items for example.
I have a record in one of the column and it looks like below:
132007700013213860001321264000
I want to split without using any delimiter based on char length of 10. After every 10 characters i need the records in a new line. Like below
1320077000
1321386000
1321264000
If the string is fixed width you could use 3 SUBSTRING functions and CROSS APPLY to 'unpivot' the 3 columns into 1 column, called 'new_col'.
declare #txtTable table(txt char(30) not null)
insert #txtTable(txt) values
('132007700013213860001321264000'),
('999999999999999999999999999999');
select u.*
from #txtTable t
cross apply (values (substring(t.txt, 1, 10),
substring(t.txt, 11, 10),
substring(t.txt, 21, 10))) v(c1, c2, c3)
cross apply (values (v.c1),(v.c2),(v.c3)) u(new_col);
new_col
1320077000
1321386000
1321264000
9999999999
9999999999
9999999999
You can use a recursive CTE:
with cte as (
select left(col, 10) as x10, convert(varchar(max), stuff(col, 1, 10, '')) as rest
from t
union all
select left(rest, 10), stuff(rest, 1, 10, '')
from cte
where rest <> ''
)
select x10
from cte;
Here is a db<>fiddle.
All credit goes to Gordon Linoff. I just added casting to make it work.
Somehow the DB Fiddle sample worked, but in SSMS it was an error.
SQL
DECLARE #tbl TABLE (col VARCHAR(30));
INSERT #tbl (col) VALUES
('132007700013213860001321264000'),
('999999999999999999999999999999');
WITH cte as
(
SELECT CAST(LEFT(col, 10) AS VARCHAR(10)) as x10, CAST(stuff(col, 1, 10, '') AS VARCHAR(MAX)) as rest
FROM #tbl
UNION ALL
SELECT CAST(LEFT(rest, 10) AS VARCHAR(10)), STUFF(rest, 1, 10, '')
FROM cte
WHERE rest <> ''
)
SELECT x10
FROM cte;
I need sql query WITHOUT FUNCTION with SELECT only numeric characters.
For example, I have in sql table 0f-gh 14-2t-4 /// and I want get this -> 01424. How I can do it with sql query SELECT, without anything, only with SELECT
This is the logic from digitsonlyEE which is the fastest T-SQL based "digits only" function available today.
declare #table table (somestring varchar(50));
insert #table VALUES('abc123xxx555!!!999'),('##123ttt999'),('555222!');
SELECT *
FROM #table t
CROSS APPLY
(
SELECT DigitsOnly =
(
SELECT SUBSTRING(t.somestring,n,1)
FROM
(
SELECT TOP (LEN(ISNULL(t.somestring,CHAR(32))))
(CHECKSUM(ROW_NUMBER() OVER (ORDER BY (SELECT NULL))))
FROM
(VALUES ($),($),($),($),($),($),($),($),($),($)) a(x),
(VALUES ($),($),($),($),($),($),($),($),($),($)) b(x),
(VALUES ($),($),($),($),($),($),($),($),($),($)) c(x),
(VALUES ($),($),($),($),($),($),($),($),($),($)) d(x)
) iTally(n)
WHERE ((ASCII(SUBSTRING(t.somestring,N,1)) - 48) & 0x7FFF) < 10
FOR XML PATH('')
)
) digitsOnlyEE(digitsOnly);
Results:
somestring digitsOnly
--------------------- ----------
abc123xxx555!!!999 123555999
##123ttt999 123999
555222! 555222
Here is an inline approach
Declare #YourTable table (ID int,SomeCol varchar(max))
Insert Into #YourTable values
(1,'0f-gh 14-2t-4 ///')
Select A.ID
,B.*
From #YourTable A
Cross Apply (
Select NewValue = (Select substring(A.SomeCol,N,1)
From (Select Top (len(A.SomeCol)) N=Row_Number() Over (Order By (Select NULL)) From master..spt_values n1) S
Where substring(A.SomeCol,N,1) like '[0-9]%'
Order By N
For XML Path (''))
) B
Returns
ID NewValue
1 01424
Note: Use Outer Apply if you want to see null values in the event where the string has NO numerics.
It's my data and every ThroughRouteSid record has the same pattern.
six number and five comma. then I just want to get three and five
number into two record to template Table and get the same Count()
value to these two record.
For example: First record in the picture.
ThroughRouteSid(3730,2428,2428,3935,3935,3938,) Count(32).
I want a result like this:
2428 32 3935 32
I get What number I want.become two record and both have same Count value into template table
you can use XML to get your result, please refer below sample code -
create table #t1( ThroughRouteSid varchar(500) , Cnt int)
insert into #t1
select '3730,2428,2428,3935,3935,3938,' , len('3730,2428,2428,3935,3935,3938,')
union all select '1111,2222,3333,4444,5555,6666,' , len('1111,2222,3333,4444,5555,6666,')
select cast( '<xml><td>' + REPLACE( SUBSTRING(ThroughRouteSid ,1 , len(ThroughRouteSid)-1),',','</td><td>') + '</td></xml>' as xml) XmlData , Cnt
into #t2 from #t1
select XmlData.value('(xml/td)[3]' ,'int' ), Cnt ,XmlData.value('(xml/td)[5]' ,'int' ), Cnt
from #t2
First create the function referring How to Split a string by delimited char in SQL Server. Then try Querying the following
select (SELECT CONVERT(varchar,splitdata) + ' '+ Convert(varchar, [Count])+' ' FROM (select splitdata, ROW_NUMBER() over (ORDER BY (SELECT 100)) row_no
from [dbo].[fnSplitString](ThroughRouteSid,',')
where splitdata != '') as temp where row_no in (2,5)
for xml path('')) as col1 from [yourtable]
If you are using SQL Server 2016 you can do something like this:
create table #temp (ThroughRouteSid varchar(1024),[Count] int)
insert into #temp values
('3730,2428,2428,3935,3935,3938,',32),
('730,428,428,335,935,938,',28)
select
spt.value,
t.[Count]
from #temp t
cross apply (
select value from STRING_SPLIT(t.ThroughRouteSid,',') where LEN(value) > 0
)spt