Split single column to multiple

Hello dears.
please i have the following along column as below and i want to make it multiple column.
07700000295
07700000519
07700000937
07700000987
07700001113
07700001846
07700002397
07700002427
Output>> 07700000295 07700000937 07700001113 07700002397
07700000519 07700000987 07700001846 07700002427
appreciate your efforts..
Ali//

Hi, Ali,
user13721099 wrote:
... please i have the following along column as below and i want to make it multiple column...The links above show how to do String Aggregation , where you combine one column from several rows into one column on one row. That one big column can be formatted to look like multiple columns.
If you really need multiple columns, then you need to Pivot , which is covered in the links on this page:
SQL and PL/SQL FAQ
String Aggregation can be much simpler, especially if the number of columns you want depends on the data in the table.
If you'd like help, post your requirements. Why do you need the data pivoted (or aggregated)? How will the re-arranged data be used? Post your best attempt at a query, and CREATE TABLE and INSERT statements for the sample data, the results you want from that data, and an explanation of how you get those results from that data. If order is important, explain.
Always say which version of Oracle you're using.

Similar Messages

  • Split single column into multiple column using sql /plsql

    create table test (name varchar2(255);
    insert into test values('DH  RED 20 12/10 10 2 ');
    insert into test values('PM  STUD 20 15/10 20 29.55' );
    insert into test values('LS  MENTHOl FILTER ASC 18/70 60 240.66');
    insert into test values('WINSTON WHITE CLASSIC 13    18/70 60 240.66');
    My Output should be like as below in other table :
    create table test_result (brand varchar2(255),packet   varchar2(50),amount varchar2(25),total varchar2(25));
    BRAND                                   PACKET       AMOUNT           TOTAL
    DH  RED 20                            12/10            10              2
    PM  STUD 20                           15/10            20              29.55
    LS  MENTHOl FILTER ASC               18/70            60              240.66
    WINSTON WHITE CLASSIC 13             18/70            60              240.66can you please help me to solve this issue
    thanks in advance
    Edited by: A on Apr 21, 2012 11:33 PM
    Edited by: A on Apr 21, 2012 11:34 PM
    Edited by: A on Apr 21, 2012 11:34 PM

    h4. # Database should be 10g. If version is below 10g query need to be modified as per string format.
    h4. # Split string into two parts by '/'. First part contain brand + packet(1), Second part contain packet(2) + amount + total
    h4. # Your brand name can be of any length.
    String Format* : <Brand Name><space><packet(1)>/<packet(2)><space><amount><space><total>
    SELECT name,
           REGEXP_SUBSTR(first_part,'.+[[:space:]]',1,1) brand,
           REGEXP_SUBSTR(first_part,'[^[:space:]]+/',1,1) || REGEXP_SUBSTR(second_part,'[^[:space:]]+[[:space:]]',1,1) packet,
           REGEXP_SUBSTR(second_part,'[^[:space:]]+[[:space:]]',1,2) amount,
           REGEXP_SUBSTR(second_part,'[^[:space:]]+$',1) total
    FROM
      (SELECT trim(name) name,
              trim(substr(name,1, instr(name,'/'))) first_part,
              trim(substr(name,instr(name,'/')+1)) second_part
       FROM test )
    {code}                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               

  • Split single row in multiple rows based on date range

    I need sql script that can split single row in multiple rows based on start date and end date column in table.
    Thank you

    I agree to your suggestion of having a dates table permanently in the database. Thats how we also do for most of our projects as well
    But in most projects the ownership  of table creation etc lies with the client as they will be the DBAs and will be design approval authorities. What I've seen is the fact that though
    many of them are in favour of having calendar table they dont generally prefer having a permanent table for numbers in the db. The best that they would agree is for creating a UDF which will have
    tally table functionality built into it based on a number range and can be used in cases where we need to multiply records as above.
    Wherever we have the freedom of doing design then I would also prefer creating it as a permanent table with required indexes as you suggested.
    >> many of them are in favour of having calendar table they dont generally prefer having a permanent table
    Those people do not understand about database and are not DBAs :-)
    It is our job to tell them what is right or wrong.
    ** This is a real story! I had a client several years back, who was the CEO of a software company.
    He use the query:
    select * from table_name
    In order to get the last ID!
    The table_name was actually a view that took data from several tables, and the main table that he wanted to get the ID included several string columns, with lot of data!
    he actually pulled all this data to the application, just to get the lat ID in a specific table!
    It is our job as Consultants or DBAs to fix's his misunderstanding :-)
      Ronen Ariely
     [Personal Site]    [Blog]    [Facebook]

  • Performance of update query for single column vs multiple column

    Hi All,
    I could not find any answer for this, does it ever matter in terms of performance updating single column versus multiple column in a single update query.
    For eg. table consisting of 15 columns, what would be the difference in performance when one column is update in the single query against another update query where 15 columns are updated.
    Please do keep in mind my table columns in actually could be around 150+.
    Thanks for any information provided.

    If updated columns aren´t on where clause , then the only impact of 15 columns will be a increase on redo generation, and a possible chainning row.
    So since the redo is one of the things that have a large impact, the answer is yes.
    The performance will be slower.
    Regards
    Helio Dias.
    http://heliodias.com
    OCE SQL, OCP 9i

  • Split Single IDOC into Multiple IDOC's Based on Segment Type

    Hi Experts,
    I have a scenario IDOC to FILE ,  Split Single IDOC into Multiple IDOC's based on Segment Type
    Outbound:
    ZIdocName
    Control Record
    Data Record
    Segment 1
    Segment 2
    Segment 3
    Status Record
    I should get output like below
    Inbound:
    ZIdocName
    Control Record
    Data Record
    Segment 1
    Status Record
    ZIdocName
    Control Record
    Data Record
    Segment 2
    Status Record
    ZIdocName
    Control Record
    Data Record
    Segment 3
    Status Record
    Please suggest me step by step process to achieve this task.
    Thanks.

    Thanks a lot Harish for reply.
    I have small doubt. According to your reply , If we have N number of segments in single IDOC with same fields in all segments then for splitting Single IDOC into Multiple IDOC's based on Segment Type we need to duplicate N number of target IDOC tree structure.
    Is that possible to Split single IDOC into Multiple IDOC's based on Segment Type using only one Target IDOC structure without duplicating the Target IDOC structure tree.

  • Select single column but multiple rows in JTable

    Hi
    I have a jTable and want to be able to select multiple rows but only in a single column.
    I've set these properties which makes selection almost the way I would like it.
    table1.setCellSelectionEnabled(true);
    table1.setSelectionMode(ListSelectionModel.SINGLE_INTERVAL_SELECTION);the only problem now is that the user can select multiple columns.
    Is there a simple way to restrict selection to single column?
    regards
    abq

    table.setCellSelectionEnabled(true);
    table.setSelectionMode(ListSelectionModel.SINGLE_INTERVAL_SELECTION);
    DefaultListSelectionModel model =
         (DefaultListSelectionModel)table.getColumnModel().getSelectionModel();
    model.setSelectionMode(ListSelectionModel.SINGLE_SELECTION);

  • Splitting single video into multiple files - workflow improvements?

    I regularly need to split a single video into multiple separate files that are used by other programs.  I'm using Premiere Pro CS4 for a number of reasons including the ability to easily and accurately select a scene by individual frames, native audio and video effects/filters, and the ability to import and export various video types.  I'm open to using other tools, but I haven't yet found one that is even close to Premiere Pro when it comes to accurately selecting split points.
    I'm not a Premiere Pro expert.  I have been doing this for about four months and have read many resources (books, forum posts, help guides) to refine the workflow I'm currently using.  My question:  does anyone know any improvements that might save me time/keystrokes.
    Current workflow:
    I've bound F8 to File > Export > Media
    Import video file
    Open in source monitor
    Set In and Out points for scene in source monitor
    Ctrl-drag from source monitor to Project Panel (creates a subclip)
    Repeat 3 and 4 for all scenes/subclips
    Select all subclips from Project Panel and drag to timeline
    Set work area bar to first subclip in timeline by the following four keystrokes:  Home, Alt-[, PageDown, Alt-]
    Type F8 (File > Export > Media shortcut).  Type output file name, click OK and the first subclip is put in the AME queue
    For the remaining subclips, type: Alt-[, PageDown, Alt-], F8, output file name, OK and repeat until done
    Any thoughts/ideas on workflow improvements?
    Thank you!

    Thank you Hunt.
    For the sake of brevity, I didn't include all of the details about the various aspects of the projects.
    Early on I was doing what you suggested and yes, it is a little quicker.  I learned (painfully) a couple of months later that I sometimes have to go back and adjust the clips and this method (trim to timeline with no subclips) resulted in quite a bit of duplicate effort (to reobtain start/stop points).  I've since learned that managing the "scenes" as subclips in topical bins allows me to go back and make adjustments, long after I've totally forgotten about the project content, without a lot of unnecessary reworking.

  • Split single row into multiple rows containing time periods

    Hi,
    I have a table with rows like this:
    id, intime, outtime
    1, 2010-01-01 00:10, 2010-01-3 20:00
    I would like to split this row into multiple rows, 1 for each 24hr period in the record.
    i.e. The above should translate into:
    id, starttime, endtime, period
    1, 2010-01-01 00:10, 2010-01-02 00:10, 1
    1, 2010-01-02 00:10, 2010-01-03 00:10, 2
    1, 2010-01-03 00:10, 2010-01-03 20:00, 3
    The first starttime should be the intime and the last endtime should be the outtime.
    Is there a way to do this without hard-coding the 24hr periods?
    Thanks,
    Dan Scott
    http://danieljamesscott.org

    Thanks for all the feedback, Dan.
    It appears that the respective solutions provided will give you: a) different resultsets and b) different performance.
    Regarding your 'truly desired resultset' you haven't answered all questions from my previous post (there are differences in the provided examples), but anyway:
    I found that using CEIL or ROUND makes quite a difference, using my 'simple 3 record testset' (30 records vs. 66 records got initially returned, that's less than half of the original). That's quite a difference. However, I must call it a day (since it's almost midnight) for now, so there's room for more optimizement and I haven't thoroughly tested.
    But this might hopefully make a difference performancewise when compared to my previous 'dreaded example':
    SQL> drop table t;
    Table dropped.
    SQL> create table  t as
      2  select 1 id, to_date('2010-01-01 00:10', 'yyyy-mm-dd hh24:mi') intime, to_date('2010-01-03 20:00', 'yyyy-mm-dd hh24:mi') outtime from dual union all
      3  select 2 id, to_date('2010-02-01 00:10', 'yyyy-mm-dd hh24:mi') intime, to_date('2010-02-05 20:00', 'yyyy-mm-dd hh24:mi') outtime from dual union all
      4  select 3 id, to_date('2010-03-01 00:10', 'yyyy-mm-dd hh24:mi') intime, to_date('2010-03-03 00:10', 'yyyy-mm-dd hh24:mi') outtime from dual;
    Table created.
    SQL> select id
      2  ,      max(intime)+level-1 starttime
      3  ,      case
      4           when level = to_char(max(t.outtime), 'dd')
      5           then max(t.outtime)
      6           else max(t.intime)+level
      7         end outtime
      8  ,      level period      
      9  from   t
    10  connect by level <= round(outtime-intime)
    11  group by id, level
    12  order by 1,2;
            ID STARTTIME           OUTTIME                 PERIOD
             1 01-01-2010 00:10:00 02-01-2010 00:10:00          1
             1 02-01-2010 00:10:00 03-01-2010 00:10:00          2
             1 03-01-2010 00:10:00 03-01-2010 20:00:00          3
             2 01-02-2010 00:10:00 02-02-2010 00:10:00          1
             2 02-02-2010 00:10:00 03-02-2010 00:10:00          2
             2 03-02-2010 00:10:00 04-02-2010 00:10:00          3
             2 04-02-2010 00:10:00 05-02-2010 00:10:00          4
             2 05-02-2010 00:10:00 05-02-2010 20:00:00          5
             3 01-03-2010 00:10:00 02-03-2010 00:10:00          1
             3 02-03-2010 00:10:00 03-03-2010 00:10:00          2
    10 rows selected.
    SQL> By the way: I'm assuming you're on 10g, is that correct?
    Can you give us some information regarding the indexes present on your table?

  • Splitting  single row to multiple query result

    Hi All,
    I have a column is a table with data seperated by new like . I need to write a select query which can split this to multiple rows .. can any one help me out .
    please see the xample
    select data_comm from customer where rownum=1
    data_comm
    serial=0
    parallel=1
    comm=1
    a single row contain a field called data_comm which has got data stored seperated by new_line
    I need to write a query which can split data on data_comm and give as multiple rows.
    Thanks in advance

    I think something like this would do the trick.
    Part 1:
    select substr(data_comm, 1, instr(data_comm,chr(10),1)) from customer ;
    Part 2:
    select substr(data_comm,instr(data_comm,chr(10),1)+1,instr(data_comm,chr(10),2)) from customer ;
    Part 3:
    select substr(data_comm,instr(data_comm,chr(10),2)+1,instr(data_comm,chr(10),3)) from customer ;
    Maybe the chr(10) is not the used newline char, but you can figure that one out yourself

  • Split single transaction into multiple PDFs

    I have a single XML file that contains a form that overflows multiple times. I want each form to be treated as a separate transaction (i.e. I want each form to generate a separate pdf).
    From the FSISYS:
    < ExtractKeyField >
    SearchMask = 1,<Wrapper
    I want each of these Address Changes to result in a separate PDF.
    <Wrapper>
    <ProducerConfirmations beginTime="0001-01-01T00:00:00" endTime="0001-01-01T00:00:00" emailChange="true" nyFlag="NA">
    <AddressChange previousAddressId="ResidentAddr_Old" addressId="ResidentAddr" mode="email" responsys="e0740ace1f534391b464b03707874636" addressType="Residential" />
    <AddressChange previousAddressId="BusAddr1_Old" addressId="BusAddr1" mode="email" responsys="e0740ace1f534391b464b03707874636" addressType="BusinessAddress1" />
    <AddressChange previousAddressId="BusAddr2_Old" addressId="BusAddr2" mode="email" responsys="e0740ace1f534391b464b03707874636" addressType="BusinessAddress2" />
    <AddressChange previousAddressId="BusAddr3_Old" addressId="BusAddr3" mode="email" responsys="e0740ace1f534391b464b03707874636" addressType="BusinessAddress3" />
    <AddressChange previousAddressId="BusAddr4_Old" addressId="BusAddr4" mode="email" responsys="e0740ace1f534391b464b03707874636" addressType="BusinessAddress4" />
    </ProducerConfirmations>
    </Wraper>
    Any way I can force the PDF print driver to produce each page as a separate file?
    Muchas gracias,
    Dave

    You'd need to use class recipients and make this node the address field. There is not an other way to split at the transaction level.

  • Creating chart from single column of multiple tables in Numbers '09

    I have multiple tables (spread across multiple sheets) in my document. In column B of each table is a name. The other columns contain other categorized data. Each table has a variable number of rows, but each name is unique within a table. A particular name in any one table may or may not occur in another.
    What I want to do is create a chart that shows the total number of times each unique name occurs across all the tables. Not really sure how to do this...
    I would also like to be able to sort the chart from highest to lowest.
    Message was edited by: Sir Reginald

    That won't quite work, as I have hundreds of names to process. It was a good starting point though. Here's what I had to do:
    1) create a new table (mynames) containing all the names in mynames::A
    2) in mynames::B1 put the formula =SUM(COUNTIF(Table 1::A, A1), COUNTIF(Table 2::A, A1), ...)
    3) Select column mynames::B and then choose *INSERT -> Fill Down* from the menu
    It becomes a bit tedious to have to update the formula and refill the column every time I add another data table. Imagine if I had hundreds of tables to enter. Hmmm....
    I currently have one table per sheet, but if I were to move all the tables onto one sheet, categorize a column, and then collapse all the categories, that wouldn't take up too much space on one sheet. Say I had 100 tables, I could then reduce the formula to:
    =COUNTIF(Table 1:Table 100::A, A1)
    I would also have to make sure to place the mynames table before Table 1 in the table list. All in all, much less of a hassle to add new table data.
    Hopefully, Apple will add multi-sheet spanning collections, in the same way you can do multi-table collections now.
    Message was edited by: Sir Reginald

  • To split a column into multiple columns

    Hi,
    I have column and i want to split it into three columns as shown below given inputs and required outputs. I have written this query to get P_1, and p_2 but dont know how to get p_3.
    SELECT ADDRESS,SUBSTR(ADDRESS,1, INSTR(ADDRESS,' ')-1) P_1
         ,SUBSTR(ADDRESS, instr(address,' ',-1)+1) P_2
         FROM TT
    Input:
    ADDRESS     
    1 AVENUE OF THE AMERICAS     
    2 NORTH CAROLINE STREET EAST     
    236 TURPENTINE ROAD COOPER CREEK     
    REQUIRED OUTPUT
    P_1     P_2     P_3     
    1     AMERICAS     AVENUE OF THE
    2     EAST     NORTH CAROLINE STREET
    236     CREEK     TURPENTINE ROAD COOPER
    Thanks in advance

    If you're on 10g or 11g you can do something like this ...
    SQL*Plus: Release 10.2.0.1.0 - Production on Tue Mar 4 21:18:07 2008
    Connected to:
    Oracle Database 10g Express Edition Release 10.2.0.1.0 - Production
    column p_1 format a3
    column p_2 format a10
    column p_3 format a30
    with data as
    select '1 AVENUE OF THE AMERICAS' as ADDRESS from dual union all
    select '2 NORTH CAROLINE STREET EAST' as ADDRESS from dual union all
    select '236 TURPENTINE ROAD COOPER CREEK' as ADDRESS from dual
    select
      regexp_replace( address, '^([^ ]+) (.+?) ([^ ]+)$', '\1' ) as p_1 ,
      regexp_replace( address, '^([^ ]+) (.+?) ([^ ]+)$', '\3' ) as p_2 ,
      regexp_replace( address, '^([^ ]+) (.+?) ([^ ]+)$', '\2' ) as p_3
    from
      data
    P_1 P_2        P_3
    1   AMERICAS   AVENUE OF THE
    2   EAST       NORTH CAROLINE STREET
    236 CREEK      TURPENTINE ROAD COOPER
    3 rows selected.
    -- or try this shorter version --
    with data as
    select '1 AVENUE OF THE AMERICAS' as ADDRESS from dual union all
    select '2 NORTH CAROLINE STREET EAST' as ADDRESS from dual union all
    select '236 TURPENTINE ROAD COOPER CREEK' as ADDRESS from dual
    select
      regexp_replace( address, '^([^ ]+).*', '\1' ) as p_1 ,
      regexp_replace( address, '.*?([^ ]+)$', '\1' ) as p_2 ,
      regexp_replace( address, '^([^ ]+) (.+?) ([^ ]+)$', '\2' ) as p_3
    from
      data
    P_1 P_2        P_3
    1   AMERICAS   AVENUE OF THE
    2   EAST       NORTH CAROLINE STREET
    236 CREEK      TURPENTINE ROAD COOPER
    3 rows selected.See SQL Snippets: SQL Features Tutorials - Regular Expressions if you're unfamiliar with regular expressions.
    If you can't / don't want to use regular expressions then you'll need to settle for a more boring approach like this one.
    with data as
    select '1 AVENUE OF THE AMERICAS' as ADDRESS from dual union all
    select '2 NORTH CAROLINE STREET EAST' as ADDRESS from dual union all
    select '236 TURPENTINE ROAD COOPER CREEK' as ADDRESS from dual
    select
      substr( address, 1, instr( address, ' ' ) - 1 ) as p_1 ,
      substr( address, instr( address, ' ', - 1 ) + 1 ) as p_2 ,
      substr
      ( address,
        instr( address, ' ', + 1 ) + 1,
        instr( address, ' ', - 1 ) - instr( address, ' ' ) - 1
      ) as p_3
    from
      data
    P_1 P_2        P_3
    1   AMERICAS   AVENUE OF THE
    2   EAST       NORTH CAROLINE STREET
    236 CREEK      TURPENTINE ROAD COOPER--
    Joe Fuda
    SQL Snippets

  • Split 1 column into multiple columns

    Hi
    How can we acheive this in Sql:
    Col 1
    abc def ghi jkl
    convert to:
    Col 1
    Col 2
    Col 3
    Col 4
    abc
    def
    ghi
    jkl
    Royal Thomas

    The best approach is to NOT split them out into individual columns... It violates the 1st normal form...
    Under first normal form, all occurrences of a record[sic] type must contain the same number of fields[sic].
    First normal form excludes variable repeating fields[sic] and groups.
    The better approach is to split them into rows, in their own table, with a foreign key reference back to the table you are pulling them away from.
    If you check my 1st post you'll see that I split the string with a function. That same a function can be used to populate the new table.
    Something along these lines...
    IF OBJECT_ID('tempdb..#BadOldDesign') IS NOT NULL
    DROP TABLE #BadOldDesign;
    CREATE TABLE #BadOldDesign (
    ID INT NOT NULL IDENTITY(1,1) PRIMARY KEY,
    ListOfCodes VARCHAR(100) NOT NULL
    INSERT #BadOldDesign (ListOfCodes) VALUES
    ('abc htr tbj yfd hjk mth'),
    ('bbb hyt tbj vyk hjk htd'),
    ('vvv nud'),
    ('yyy hyn tbj bdr '),
    ('htr htf tbj yfd hjk mko'),
    ('kio bhn tbj bjd hjk'),
    ('byd loj tbj yfd hjk cds');
    SELECT * FROM #BadOldDesign bod
    IF OBJECT_ID('tempdb..#ShinyNewTable') IS NOT NULL
    DROP TABLE #ShinyNewTable;
    SELECT
    bod.ID,
    sc8.ItemNumber,
    sc8.Item
    INTO #ShinyNewTable
    FROM
    #BadOldDesign bod
    CROSS APPLY dbo.SplitCSVToTable8K(bod.ListOfCodes, ' ') sc8;
    ALTER TABLE #ShinyNewTable ALTER COLUMN id INT NOT NULL;
    GO
    ALTER TABLE #ShinyNewTable ALTER COLUMN ItemNumber INT NOT NULL;
    GO
    ALTER TABLE #ShinyNewTable ADD CONSTRAINT pk_ShinyNewTable PRIMARY KEY CLUSTERED (ID, ItemNumber);
    SELECT * FROM #ShinyNewTable snt
    The dbo.SplitCSVToTable8K function that I'm using can be found here... http://www.sqlservercentral.com/articles/Tally+Table/72993/
    HTH,
    Jason
    Jason Long

  • Splitting Single Column into 3 columns.

    I have a column data like this
    85052/123755 LAURITSEN/H HENRIK S MR
    87432/119975 MORTENSEN/P PIA BUBANDT MS
    73784/127530 PEDERSEN/M MORTEN H MR
    88579/128347 JENSEN/M MORTEN MR
    ./AIRTECH STEFANSEN/P PER MR
    10024/AIRTECH HOFFMANN/M MICHAEL KJUL MR
    75922/128415 MOELLER/F FRANK MR
    17639/128440 THOMSEN/P PETER MR
    53198/127655 THYGESEN/E EBBE MR
    85960/128473 ORAEINAMZADI/M MAZYAR MR
    52663/128270 MAINZ/J JAN MR
    17639/128440 THOMSEN/P PETER MR
    27510/. HANSEN/N NILS HARLADSTED
    43885/126300 LARSEN/J JOHNNI S MR
    87580/125410 STEFANSEN/R RASMUS MR
    63215/128445 NIELSSON/J JESPER MR
    80594/123334 OLSEN/H HENRIK W MR
    I need to split this into three columns
    85052/123755 LAURITSEN/H HENRIK S MR
    should be split as
    Column1: 85052
    Column2: 123755
    Column3: LAURITSEN/H HENRIK S MR
    but in some cases it may also be like this
    ./AIRTECH STEFANSEN/P PER MR
    this should be split into
    Column1: null
    Column2: null
    Column3: AIRTECH STEFANSEN/P PER MR
    one more scenario is like this
    27510/. HANSEN/N NILS HARLADSTED
    Column1: 27510
    Column2: null
    Column3: HANSEN/N NILS HARLADSTED
    Pls Help
    Thanks
    Vinoth.

    Based on the sample data you provided,
    SQL> WITH T
      2       AS (SELECT '85052/123755 LAURITSEN/H HENRIK S MR' str FROM DUAL
      3           UNION ALL
      4           SELECT './AIRTECH STEFANSEN/P PER MR' str FROM DUAL
      5           UNION ALL
      6           SELECT '10024/AIRTECH HOFFMANN/M MICHAEL KJUL MR' str FROM DUAL
      7           UNION ALL
      8           SELECT '27510/. HANSEN/N NILS HARLADSTED' str FROM DUAL)
      9  SELECT REGEXP_REPLACE (REGEXP_SUBSTR (str, '[^/]+', 1, 1), '[^[:digit:]]') col1
    10        ,REGEXP_REPLACE (REGEXP_SUBSTR (str, '[^/]+', 1, 2), '[^[:digit:]]') col2
    11        ,REGEXP_SUBSTR (str, '[[:alpha:]].*', 1, 1) col3
    12    FROM T;
    COL1                                     COL2                                     COL3
    85052                                    123755                                   LAURITSEN/H HENRIK S MR
                                                                                      AIRTECH STEFANSEN/P PER MR
    10024                                                                             AIRTECH HOFFMANN/M MICHAEL KJUL
    27510                                                                             HANSEN/N NILS HARLADSTED
    SQL> G.

  • Splitting single idoc into multiple xml's

    Hello,
    My requirement is such that, i need to split an custom IDoc into three xml which will be picked up by MDM, as i am new to PI 7.1 kindly guide me through the process.
    and the second requirement is, i need to map the same custom IDoc into three different MDM repositories pls guide me through the process

    Hi Abhishek,
    [quote]My requirement is such that, i need to split an custom IDoc into three xml which will be picked up by MDM, as i am new to PI 7.1 kindly guide me through the process
    For this, you have to go for multimapping.
    The links for it are already provided in the above posts.
    and the second requirement is, i need to map the same custom IDoc into three different MDM repositories pls guide me through the process
    Can you please explain this requirement further?
    -Supriya.

Maybe you are looking for