• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 671
  • Last Modified:

Parse single field into multiple columns

Greetings Experts,

Using SQL Server 2005, I need to parse a single field into 10 columns separated by the '-'. Please review the attached sample data below:

1-01-003-2843-000100-9495445-0003-0009-2434-082906
1-01-004-3518-001472-5650553-0007-0011-2434-071906
1-01-004-3518-001840-5650395-0016-0023-2434-071806
1-01-004-3518-006300-5650625-0070-0180-0696-090906
1-01-025-3520-001200-0263780-0013-0019-2434-072106
1-01-036-2318-006720-9453664-0214-0509-2434-082206
1-01-036-2318-010080-9492002-0433-0764-2434-070406


Your help is greatly appreciated,

0
malc77
Asked:
malc77
1 Solution
 
AaronAbendCommented:
create this function to pivot your rows... may take some tweaking.. you did not provide much for table structures...

create procedure pivoter   as
declare getrows cursor for
SELECT myfield
FROM mytable
WHERE .... - your constraints

declare @onecode varchar(20)
declare @alllcodes varchar(2000)
declare @dash varchar(1)
begin
set @dash='' --starts empty
open getrows;
fetch next from getrows into @onecode;
while (@@fetch_status=0)
begin
   set @allcodes = @allcodes + @dash+ @onecode    
   set @dash= '-'
   fetch next from getrows into @onecode;

end
close getrows;
deallocate getrows;

end
0
 
dduserCommented:
Create Procedure TempProc as

Declare @TempVal as Varchar(255)
Declare @InsertVal as Varchar(20)
Declare @SelectVal as Varchar(200)

Declare GetData Cursor For
Select myField from myTable

Fetch Next from GetData into @SelectVal

While (@@Fetch_Status = 0)
Begin

                      Set @TempVal = @SelectVal
                      While (PatIndex('%-%',@TempVal) <> 0)
                      Begin
                                        Set @InsertVal = Left(@TempVal,PatIndex('%-%',@TempVal) - 1)
                                        Set @TempVal = Right(@TempVal,Len(@TempVal) - PatIndex('%-%',@tempVal))
                                        Insert into TempTable (@InsertVal)
                      End
                      Insert into TempTable (@InsertVal)
End

Return




0
 
LandyJCommented:
If you want to place the data into individual columns, just a bit more manipulation will be needed.  This solution will work as long as the data has the same format for all rows and there will always be 10 columns of data output.

hth,
Landy

Here's the whole test script:

drop table #Data
drop table #Step
drop table #Pivot

CREATE TABLE #Data (
      Col01 varchar(100)
)

CREATE TABLE #Step (
      rowID int,
      colID int identity,
      val varchar(100)
)

CREATE TABLE #Pivot (
      rowID int,
      Col01 varchar(100),
      Col02 varchar(100),
      Col03 varchar(100),
      Col04 varchar(100),
      Col05 varchar(100),
      Col06 varchar(100),
      Col07 varchar(100),
      Col08 varchar(100),
      Col09 varchar(100),
      Col10 varchar(100)
)

-- Replace these with the INSERT INTO #Data SELECT FieldName from MyTable WHERE xFld = Cond1 ....
insert into #Data VALUES ('1-01-003-2843-000100-9495445-0003-0009-2434-082906')
insert into #Data VALUES ('1-01-004-3518-001472-5650553-0007-0011-2434-071906')
insert into #Data VALUES ('1-01-004-3518-001840-5650395-0016-0023-2434-071806')
insert into #Data VALUES ('1-01-004-3518-006300-5650625-0070-0180-0696-090906')
insert into #Data VALUES ('1-01-025-3520-001200-0263780-0013-0019-2434-072106')
insert into #Data VALUES ('1-01-036-2318-006720-9453664-0214-0509-2434-082206')
insert into #Data VALUES ('1-01-036-2318-010080-9492002-0433-0764-2434-070406')

declare cur_Pivot cursor for
      Select * from #Data
declare @row int, @col int,
      @Data varchar(100)
      
select @row= 0, @col = 0, @Data = ''

open cur_Pivot
FETCH NEXT from cur_Pivot INTO @Data

WHILE @@Fetch_Status = 0
  Begin
      INSERT INTO #Step
            (rowID, val)
      SELECT @row, *
      FROM udf_strSplit(@Data, '-')

      select @row = @row + 1
      FETCH NEXT from cur_Pivot INTO @Data
  End
 
close cur_Pivot
deallocate cur_Pivot

insert into #Pivot
SELECT rowID,
    MAX(CASE colID % 10 WHEN 1 THEN val ELSE '' END) AS Col01,
    MAX(CASE colID % 10 WHEN 2 THEN val ELSE '' END) AS Col02,
    MAX(CASE colID % 10 WHEN 3 THEN val ELSE '' END) AS Col03,
    MAX(CASE colID % 10 WHEN 4 THEN val ELSE '' END) AS Col04,
    MAX(CASE colID % 10 WHEN 5 THEN val ELSE '' END) AS Col05,
    MAX(CASE colID % 10 WHEN 6 THEN val ELSE '' END) AS Col06,
    MAX(CASE colID % 10 WHEN 7 THEN val ELSE '' END) AS Col07,
    MAX(CASE colID % 10 WHEN 8 THEN val ELSE '' END) AS Col08,
    MAX(CASE colID % 10 WHEN 9 THEN val ELSE '' END) AS Col09,
    MAX(CASE colID % 10 WHEN 0 THEN val ELSE '' END) AS Col10
from #Step
group by rowID

select * from #Pivot


AND, you'll need this udf:

CREATE FUNCTION  dbo.udf_strSplit (
                  @string varchar(8000),
                  @splitter char( 1)
                  )
RETURNS @res TABLE (val varchar( 8000))
AS
Begin
      IF SUBSTRING (@string, len ( @string), 1) <> @splitter
            SET @string= @string+@splitter

      Declare @start bigint,
            @word varchar( 8000),
            @charindex bigint,
            @i bigint
      SELECT @i=1, @start=1, @charindex= charindex( @splitter, @string, @start)

      WHILE (@charindex <> 0)
        Begin
            SET @word= substring( @string, @start, @charindex - @start)
            SET @start= @charindex +1
            SET @charindex= charindex( @splitter, @string, @start)
            INSERT INTO @res  VALUES (@word)
            SET @i=@i+1
      End

      RETURN
end

0
What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

 
malc77Author Commented:
Greetings,

Thanks for responding and forgive for not adding the table structure, (still a bit of a novice).  

My Table structure is as follows

AuxAcctAsmnt_1      nvarchar
Cost Ctr                      nvarchar
Cost Elem#      nvarchar
Cost element name      nvarchar
Nme                      nvarchar                                                   !!!!!!!! (parsed into the fields below)!!!!!!!!!     Nme = '1-01-003-2843-000100-9495445-0003-0009-2434-082906'
[Name of offsetting account]      nvarchar
RefDocNo                      nvarchar
Per                      nvarchar
User                      nvarchar
Value ObjCurr      float
Doc# Date                      datetime
Posting Date      datetime
Mde                      varchar                  Insert from parsed Nme field
SrvTyp                      varchar                  Insert from parsed Nme field
Supplier                      varchar                  Insert from parsed Nme field
Origin                      varchar                  Insert from parsed Nme field
Quantity                      varchar                  Insert from parsed Nme field
Material                      varchar                  Insert from parsed Nme field
ActlWght                      varchar                  Insert from parsed Nme field
DimWght                      varchar                  Insert from parsed Nme field
FnlDest                      varchar                  Insert from parsed Nme field  
ShipDte                      varchar                  Insert from parsed Nme field

thanks,
0
 
malc77Author Commented:
Greetings Experts,

Sorry for the delay in response...I will try the solutions provided above today.  thanks for your patience.
0
 
malc77Author Commented:
Greetings LandyJ,

Thanks for your response this is a great solution.

Each row has a unique key field 'RefDocNo' already assigned.  Is it possible to make this field the rowid?

Sorry for not mentioning it in the beginning.

Thanks,
0
 
LandyJCommented:
Sure, it doesn't matter what the rowID field is.  The important one is the colID identity field.  That's the one that will sort the split-out rows into the correct columns.  By using the RefDocNo as the rowID in #Pivot, it will let you join the two tables to get your final result:

SELECT
      AuxAcctAsmnt_1,
      CostCtr,
      CostElem,
      Cost,
      Col01,
      Col02,
      Col03,
      Col04,
      Col05,
      Col06,
      Col07,
      Col08,
      Col09,
      Col10,
      RefDocNo,
      Per,
      User,
      ValueObjCurr ,
      DocDate,
      PostingDate,
      Mde,
      SrvTyp,
      Supplier,
      Origin,
      Quantity,
      Material,
      ActlWght,
      DimWght ,
      FnlDest,
      ShipDte
FROM
      MyTable mt
      INNER JOIN #Pivot t ON mt.RefDocNo = t.RefDocNo


hth,
Landy
0
 
malc77Author Commented:
Greetings,

Thanks for your assistance....This has increase my abilities as a developing DBA.

Malcolm
0

Featured Post

Concerto's Cloud Advisory Services

Want to avoid the missteps to gaining all the benefits of the cloud? Learn more about the different assessment options from our Cloud Advisory team.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now