Problems with acumulated values

Hi all,
I'm trying to create a query that mix some information by day, and after that the value accumulated in period.
Is something like that:
Var: 0CALDAY (interval)
Colums: Structure
         - Real Amount (listed all days defined in Var)
         - Forecast Amount (last day of interval + 60 days)
         - Accumulated Real
         - Accmulated Forecast
        0CALDAY
Lines: 0LQPOSIT (hierarchy)
With that configuration, even the accumulated values are showed in days. Is there a way to don't show the days for the accumulated KF?
Thanks,
Marcel Will

I found out now, that when  i open the form again and change the value again and try to submit it to the document form, it works. it only doesn't work the first time!

Similar Messages

  • Problem with offset values

    Hi Experts,
    I have a problem with offset values. I have create a variable on OFISCPER, it's characteristic value, indivdual value, mandatory and ready for the input. I need to show the varibel period and the l previous period and so on till 12 periods. I used the offset values (-1, -2, etc.) but when I execute the query I can show just the variable period and the period previous I don`t see. The query is based on multiprovider and don´t has filters. Can anyone help me?

    Leonel,
    You may nt need to have 12 columns in the query designer.
    You could include the keyfigure and restrict it to the variable and create the offset for the 12 periods
    Check this http://help.sap.com/saphelp_nw04/helpdata/en/00/e8d13f7fb44c21e10000000a1550b0/frameset.htm
    -Doodle

  • LOV Problem with multiple values

    HI All,
    I have a problem with LOV .When ever i click LOV after search button all values are displaying fine.
    But when i get so many values i want to select only one vlaue that is not cmng to the main page ....Cursor is in running state always after that time out error is coming in my application .
    This problem is coming with with only single value selection in lOV only problem with Multiple values retrival that time only...
    (Iam using 11.1.1.3 Jdeveloper.)
    Thanx in advance...

    duplicate of {thread:id=2286814}

  • Facing Problem with passing Values from One report to another

    Hi,
    I am Hemanth, I have developed 2 reports. Firast Report High Level Summary, Secong is detailed. First report is developed using Union(4 union) , I am having 4 columns. The report is generating the data. I have used Navigation option on Client Column to move on to Second report. In Second report with in Filter i am using Prompted option. Due to some problem, the client value from first report is not passing to the second one. The second report is getting generated for all the clients.
    Normally i have used this Navigate option in other reports and that works fine. I have even tested with Union, it works. This time it is giving some trouble.
    Please, let me know whats going wrong.
    Thanks for the help.

    sorry for the late updation.
    My First Report, Summary level is a Pivot Table.
    I tried the same report with Table option, the value is passing correctly.
    Is there a way to get rid of this situation using Pivot table.
    Thanks for your help.
    below is the original request.
    Hi,
    I am Hemanth, I have developed 2 reports. Firast Report High Level Summary, Secong is detailed. First report is developed using Union(4 union) , I am having 4 columns. The report is generating the data. I have used Navigation option on Client Column to move on to Second report. In Second report with in Filter i am using Prompted option. Due to some problem, the client value from first report is not passing to the second one. The second report is getting generated for all the clients.
    Normally i have used this Navigate option in other reports and that works fine. I have even tested with Union, it works. This time it is giving some trouble.
    Please, let me know whats going wrong.
    Thanks for the help.

  • Problem with checkbox value and app.settings.getSetting()

    Hello everyone.
    I`ve got a problem with getting saved settings for checkbox.
    Saved settings doesn`t apply on checkbox value.
    It`s strange, but it always "true".
    So here piece of code:
    checkMe = 11;    //any value for the first time running
    if (app.settings.haveSetting("savea", "chck")) {                           
                          checkMe = app.settings.getSetting("savea", "chck");  // get saved value, else 11
                          alert("checkMe = "+checkMe)
                        if(checkMe == 11){
                            checkMe= false;   //if there is no saved settings, checkbox value will be false/ unchecked
                            alert("You run this script first time, so manual value will be 'false'")
                            }else{
                                checkMe = app.settings.getSetting("savea", "chck");     // if saved settings exist, get true or false
         alert("Yes, checkMe will = "+checkMe)       
    win.checkPanel.chkOne.value = checkMe;  /// checkbox value will equal saved checkMe
    var fff = win.checkPanel.chkOne.value;  // string for alert
    alert("aha! but checkbox value is  = " +fff+ "\n Why so?")     // here we`ve got always "true"
    win.checkPanel.chkOne.onClick  = function () {
        var path11 = win.checkPanel.chkOne.value;  //  checkbox value to variable
        app.settings.saveSetting("savea", "chck", path11);  //saving checkbox value
        alert("Saved this value - " + app.settings.getSetting("savea", "chck"))
    win.checkPanel.chkOne.value is always true.
    Can someone exmplain why and how to get rid of this static value?
    P.S. just in case, here is full code:
    function mainFun()
        this.windowRef = null;
    mainFun.prototype.run = function()
        /*----- UI -----*/
              var retval = true;
              var win = new Window("palette", "Check option", [150, 150, 460, 455]);
              this.windowRef = win;
              win.checkPanel = win.add("panel", [25, 150, 285, 265], "Checkbox");
              win.checkPanel.chkOne = win.checkPanel.add("checkbox", [10, 15, 125, 35], "Checkbox One");
              //win.checkPanel.chkTxtOne = win.checkPanel.add('edittext', [140, 15, 230, 35], '');
         win.quitBtn = win.add("button", [110,275,200,295], "Close");
    checkMe = 11;    //any value for the first time running
    if (app.settings.haveSetting("savea", "chck")) {                           
                          checkMe = app.settings.getSetting("savea", "chck");  // get saved value, else 11
                          alert("checkMe = "+checkMe)
                        if(checkMe == 11){
                            checkMe= false;   //if there is no saved settings, checkbox value will be false/ unchecked
                            alert("You run this script first time, so manual value will be 'false'")
                            }else{
                                checkMe = app.settings.getSetting("savea", "chck");     // if saved settings exist, get true or false
         alert("Yes, checkMe will = "+checkMe)       
    win.checkPanel.chkOne.value = checkMe;  /// checkbox value will equal saved checkMe
    var fff = win.checkPanel.chkOne.value;  // string for alert
    alert("aha! but checkbox value is  = " +fff+ "\n Why so?")     // here we`ve got always "true"
    win.checkPanel.chkOne.onClick  = function () {
        var path11 = win.checkPanel.chkOne.value;  //  checkbox value to variable
        app.settings.saveSetting("savea", "chck", path11);  //saving checkbox value
        alert("Saved this value - " + app.settings.getSetting("savea", "chck"))
              win.quitBtn.onClick = function() {
                  win.close();
        win.center();
              win.show();
              return retval;
    if(typeof(mainFun_unitTest) == "undefined") {
        new mainFun().run();

    Played around with different values, combinations...
    Doesn`t work too:
    checkMe = 111;    //any value for the first time running
    if (app.settings.haveSetting("333", "check")) {                           
                          checkMe = app.settings.getSetting("333", "check");
    alert(checkMe);
                    if(checkMe == 111){
                             win.checkPanel.chkOne.value= false;
                            }else{
                                if (checkMe != false){
                                win.checkPanel.chkOne.value= true;
                                }else{
                                win.checkPanel.chkOne.value= false;
    win.checkPanel.chkOne.onClick  = function () {
        var path11 = win.checkPanel.chkOne.value;
        var path12 = path11.toString();
        app.settings.saveSetting("333", "check", path12);
        alert("Saved value - " + app.settings.getSetting("333", "check"))

  • Problem with the Value property node (MacOS)

    As far as I have tested it, Value Property nodes (and Value (signaling)) don't work in MacOS : The value property appears as a boolean, the value(signaling) as a cluster (width, height).
    Even with boolean controls, the node is not working.
    No such problem in LV 8.6.
    Am-I missing something ? Is that some "intended use"  ? Or simply a bug ?
    Chilly Charly    (aka CC)
             E-List Master - Kudos glutton - Press the yellow button on the left...        
    Solved!
    Go to Solution.

    That sounds like some sort of indexing problem with the property nodes like somesort of corrupted installation.  What if you just delete the property node and recreate it?
    I dropped a numeric control and looked at the shortcut menu for the list of property nodes.  Right below Value and Value (signalling) are Visible (which would be a boolean) and Xcontrol, which when you step down through the submenus, the first choice is Container bounds, a cluster of Width and Height.  So these 2 datatypes that are 2 lower on the list match the ones you are getting.
    What happens if you pick another property node a few elements up or down on the shortcut menu?  Does is happen for some other properties, and if so, are thereany properties that are okay?
    Here is what my property nodes look like for a numeric and I draggged the box up and down to show the natural order of the property nodes.  Note that I have scripting installed, so there are a few more property nodes (and the blue box at the top) that you might not have.
    PS.  Check you signature.  The Kudos button is now moved to the left since the forum upgrade. 
    Attachments:
    Example_VI_BD.png ‏4 KB

  • Problems with F4 Value Help in NW2004s Portal

    Hello,
    we are having some troubles with the F4 ValueHelp Extension from the Advanced BSP Programming book (and the blogs of course).
    We've implemented the solution on a 4.7 backend and portal 6.0. No problems there, everything was working fine.
    We now moved to erp2005 (WebAS 700) and NW2004s Portal (7.0).
    The popup shows and I can select entries from the list. But when clicking the ok icon a javascript error is thrown stating that 'parent.document.getElementById(...)' is null or no object.
    I think the problem is, that by making a call to 'parent' we are searching in the portal frame and not the BSP. There (in the portal frame) the element is not known so the error is thrown.
    I allready changed from the dialog_ie6.js provided on the BookCD to the one delivered by SAP. Did nothing regarding the error.
    I also tried changing ZCL_BSP_M_INPUT_HELP->BUILD_ACCEPT_SELECTION_SCRIPT. I changed all the "parent.document.get..." "to document.get...". The error message was still the same, 'parent.document.getElementById(...)' is null or no object.
    So now I'm kind of stuck. If I try the example outside the portal everything works as expected. So this clearly seems to be a problem with the portal integration. All involved BSP-Apps are marked as "supports portal integration".
    Any ideas on this topic?
    Thanks and kind regards,
    Alexander Sperling

    I have a working solution to the problem.  It isn't the nicest solution because it will probably require some modifications to the application that hosts the value help, but we will get to that in a second.
    I tried many different ways to come up with a solution that would still work via JavaScript. The portal is now loading the popup javascript in the master frame.  It registers event handlers so that if any application triggers a popup internally it "hi-jacks" the event.  So regardless of what I change in the BSP javascript library, it gets overriden by the portal one. The portal also now even enforeces its theme on the inner page in the popup.  Truly the portal takes complete control over the inner html of the popup.
    To make things worse the popop has not path to reach the source that triggered it.  The source objects are tracked in the javascript, but because the master frame takes over; the popup window only has access and visibility to the master frame and the navigation frame.  In essence in the DOM it is now higher in the stack than the object that actually triggered it. 
    Web Dynpro gets around this because it doesn't use any client side mechanisms to communicate it results back to the source frame. Web Dynpro has the luxury of always being stateful and therefore can make certain assumptions that would never work in BSP.  The hi-jacking is an excellent solution for Web Dynpro.  The portal can then override the functionality of Web Dynpro - providing the new functionality of floating the dialog box over the entire portal - without having to make any changes to the Web Dynpro Framework.  However they never had our little BSP solution in bind obviously.
    It might be possible to completely rewrite the popup_ie.js and popup_ns.js to use all different event, method and variable names.  This separate version could then provide complete separation from the portal script. However this is just a theory. I'm afraid that I wouldn't have the bandwidth to rewrite the entire script right now so I focused on a quicker solution.
    The only approach that I could come up was to follow the Web Dynpro approach and return to the server with the value. To do that I used Portal Eventing.  Portal Eventing allows us to pass a value to the source frame regaurdless of our relative positions in the DOM. 
    In my inner BSP application for the value help (ZES_VALUEHELP), I made the following change in the model method BUILD_ACCEPT_SEL_SCRIPT_V2.  At line 141 I changed the logic.  Instead of using JavaScript to write the selected value back into the source element, we will just trigger a Portal Event:
    ****With an InputField we write the contents directly back into the element
          ELSE.
            CONCATENATE me->js_close1
            `portalFireEvent('com.sap','`
            me->element_id
            `',"`
            key
            `");`
             INTO me->js_close1.
            EXIT.
    The other change is within the Element handler class for inputHelpV2 - ZCL_ES_BSP_ELMNT_INPUT_HELP_V2.  We need to register the listener for the Portal Event. Therefore in DO_AT_BEGINNING we will add the follwing at line 23 (right after we place the Element Id into the URL Parameters Table.
      data: portal_event type ref to cl_portal_event.
      portal_event = cl_portal_event=>factory(
        name = <wa_params>-value
        namespace = 'com.sap' ).
      while m_page_context->element_process( element = portal_event ) = co_element_continue.
      endwhile. "End portalEvent Render
    Overally using the portal eventing actually is an intersting approach.  There is only one drawback, now we have a server side event.  That being the case, our application now needs some event logic to process this event and populate the value. Depending upon if you are using MVC, Data Binding, or Flow Logic this could vary - and would have to handled by each application. The nice thing about the old approach was that it all happened via JavaScript on the Frontend and didn't require the application programmer to do anything - that was obviously its weakness as well.

  • Problems with numeric values in the barcode

    Hi
    I have a problem with the data from a numeric field, when it is encoded in the barcode the validation pattern seems to be ignored.
    I have set the validation pattern of the numeric field to be
    num{9999}
    And when I enter a value like "0200" it is displayed in the form as "0200", which is correct.
    BUT in the barcode the value is captured as "200", which i NOT what I want. I need all four characters to be in the barcode. Even when the first one i 0 (zero).
    How can I do that?
    Kirstine

    Can you send the form to [email protected] and I will have a look. Please include a description of the issue in the email as the email and this forum are not tied together.
    paul

  • Problem with RGB Values for Canvas Type ‚Horizontal Toolbar' in Webforms 10

    Hello,
    I have a problem with the canvas type ‘Horizontal Toolbar’ in Webforms 10g.
    When I try to set or change the RGB value of a ‘Horizontal Toolbar’ with the forms function
    Set_Canvas_Property(‘my_horizontal_toolbar’, BACKGROUND_COLOR, rgb-value)
    (at runtime) it has no effect in Webforms 10g for the canvas type ‘Horizontal Toolbar’ (the background color of the horizontal toolbar doesn’t change).
    But it works fine for other canvas types like ‘Stacked’ or ‘Content’.
    In Oracle 6i (Client / Server) this function works quiet well (even for the ‘Horizontal Toolbar’ canvas type).
    The reason why we are using rgb values instead of visual attributes in this case is, that the color of the ‘Horizontal Toolbar’ should be customizeable (via a RGB Field) for our customers (at runtime).
    Further more we are using the same technique for Text-Field-Items, Content- and Stacked-Canvases (and all works fine).
    Maybe it has something to do with the pre-configured IAS color-schema (blaf).
    But changing the color schema (e.g.: from 'blaf' to 'red') has no effect also.
    Thanks for your help.

    This is probably bug 777312. Call Support for the latest info on this bug and for help getting it fixed.

  • Table editor problem with default value for column

    Hi there,
    I'm trying to use SQL Developer 1.2.1 on Ubuntu 7.10 (though I don't think the OS is relevant) and I'm running into a snag when trying to update table structures. If you go to edit the table to say modify a column name, and that column has a default value set, SQL Developer is encapsulating the value with brackets and quotes. When trying to save the changes, it complains about the default value being 4 chars too many for the column width. An example of this would be a currency code column of varchar2 (3) set with EUR as the default. SQL Developer changes the value to ('EUR') hence the error.
    Is there anyway to prevent this from happening?
    Cheers,
    Chris

    Could you please be more specific? In 1.2.1.32.40 I
    added the column COL1 VARCHAR2(3) with default value
    'EUR'. The data display field refreshed to show the
    default value, so I edited it in a couple of entries
    and comitted the change without problem. Then I
    renamed the column from the "Edit table" popup menu.
    Finally, I renamed the column from the "Rename
    column" popup menu, no problem either.Ah, you are correct. That does work. What I was trying is select tablename in the object browser, then right click on it and select "Edit" from the contextual menu. You are then presented with edit table dialog box. If you then select a column in that box you will see a column properties box. In that section you can change the name, default value, datatype et cetera. Note that your default column value which should just be EUR (or 'EUR' to denote that it's a string) is now ('EUR'). Once you try changing the name to of the column to something else and then click OK, you should then get an error complaining of "Invalid varchar2 default value" with a more specific message of "Default values size 7 is greater than the allowed 3"
    You have provided a work around, and thanks for that, but it would be nice to edit everything in the Edit table dialog box if you're modifying more than one column name. I'm tasked with cleaning up a bit of a messy database, and there are a good deal of columns to be renamed.
    Cheers,
    Chris

  • ADF Code Corner 069 : problem with returned value

    Hello,
    I use jdeveloper 11.1.2.2.0 and am trying to follow the sample 69 from ADF Code Croner : how-to create a custom LOV using bounded task flows. http://www.oracle.com/technetwork/developer-tools/adf/learnmore/69-custom-lov-with-btf-276178.pdf
    I have a problem with the returned value. The correct value is returned in the inputText but it seems that it is just at UI level and that binding layer is not impacted. If user click on next button, all others rows show the same value in the inputText (the value chosen before in the lov), and not the correct value which must be value from db.
    How can I solve this problem?
    note : opening the 69-sample, jdeveloper ask me to migrate some files (jpr/jws) from previous version of jdeveloper.
    Edited by: h0s on 8 oct. 2012 06:14

    Yes It happens with the downloaded sample. You can do this test :
    Launch EmployeesForm activity -> Open the LOV and select a department -> click next and see that the department id is the same for all rows.
    jdeveloper 11.1.2.2.0

  • Strange problem with Domain Values

    Dear All,
      I encounter one strange problem with Domain Fixed Values.
    Background:
       We have a custom domain ( ZXREF2) created for custom data element(ZXREF2) which has replaced standard dataelement in BSEG for field XREF2. This domain has 12 Fixed values. The idea was to have F4 help for field "Reference Key2" in FBL5N/FBL01 tcodes. It is working fine with these 12 Fixed value, means, on F4 we could see all these 12 values and when selecting a line from F4 the entry got thru the screen validation ( plz note hat this validation is like checking the mandatory value so this is done even before PAI).
    Problem:
      The requirement was to add a new line to F4 values( tecnically, 13th record in domain ZXREF2). We added this line to domain and moved to Quality system. Its working fine in both Dev and Qality, but the strange problem occured after moving and testing in Production.
      In production, we could see all 13 lines in F4 help. <b>BUT</b>, after selecting the new one( 13th one), and after the value appears on screen field, when Enter is pressed, it gives error message "Enter Valid value".
       This error message also comes when we enter any value which is not there in F4. That means, system is able to recognize the new value in F4, but falied to validate that.
      We thought that, this might be the problem with Buffer, and request Client to re-start the system. Even after system restart, it still gives the same error message.
    Please note that, the error message is triggered even before PAI of the screen called. So noway( I already tried system debugging) we can debug this error message to findout the rootcause.
    Any quick help would be highly appriciated.
    Thanks and Regards,
    Prasad.

    Hello Max,
      I guess DB Utility ( SE14 ) is to be run only when there is a change at structure level, to make Runtime and Database objects in sync. Here, we just added a new line to existing domain, and this added a new row to the Database table DD07T/DD07L.
      Could you please expalin it more clearly if your view is different that what I just described. 
    Prasad.

  • Problems with Audigy2 Value and CMS

    I have an Audigy2 value sound card running on windows xp with sp2. My speakers are 4. Creative 4400. I installed the drivers that came along with the card and they work fine. The trouble is that unless I turn on CMSS 3D, I get sound only from the front left and front right speakers. Only if CMSS 3D is on does the rear channel activate.
    This is very odd considering that I had a SBLi've! before this and I had no such problems with my speakers. I've tried tweaking the software but to no avail. Also tried reinstalling - but no change.
    Please help.Message Edited by shadowfall on 0-06-2005 05:59 PM

    You are a victim of the align to pixel grid option.
    Starting in CS5 if when you chose a new document profile of web, that feature gets turned on. Select All and remove uncheck "Align to pixel grid".
    Adiitonally uncheck Align New Object To Pixel Grid, so this does not come back when you draw new objects.
    Now if you are working on a web image, then that option is good to have on, but work in units of pixels, and all should be better as long as you work in whol pixel numbers.

  • XML Parsing - problem with a value of an element if starting with space

    Hi Experts,
    I need your valuable guidence to get out of a problem in parsing an XML file.
    An XML file is read into xstring variable and it is processed to be split into the corresponding itab.
    Please find the code below
    types: begin of ty-itab,
                field(4096),
              end of ty-itab.
    Data:
    g_ixml                       TYPE REF TO if_ixml,
    g_ixmldocument               TYPE REF TO if_ixml_document,
    g_ixmlstreamfactory          TYPE REF TO if_ixml_stream_factory,
    g_ixmlstream                 TYPE REF TO if_ixml_istream,
    g_ixmlparser              TYPE REF TO if_ixml_parser,
    g_ixmlnodemainlist        TYPE REF TO if_ixml_node_list,
    g_ixmlnodelistmainelement TYPE REF TO if_ixml_node_list,
    g_sxmldata                TYPE string,
    g_ixmlnode                TYPE REF TO if_ixml_node.
      DATA: wa_xmltab TYPE xstring,
                itab type table of ty-itab.
    OPEN DATASET l_file IN BINARY MODE FOR INPUT MESSAGE v_msg.
    READ DATASET l_file INTO wa_xmltab.
    PERFORM f_create_xmltable USING wa_xmltab
                                  CHANGING v_msg.
    IF NOT v_msg IS INITIAL.
          msg = v_msg.
          EXIT.
        ELSE.
          itab[] = it_xmldata[].
        ENDIF.
    CLOSE DATASET l_file.
    Subroutine to convert xstring to char type ITAB.
    FORM f_create_xmltable USING  value(pi_inputxmlstring) TYPE xstring
                           CHANGING v_msg.
    *-- create the main factory
      g_ixml = cl_ixml=>create( ).
    *-- create the initial document
      g_ixmldocument = g_ixml->create_document( ).
    *-- create the stream factory
      g_ixmlstreamfactory = g_ixml->create_stream_factory( ).
    *create input stream
      g_ixmlstream = g_ixmlstreamfactory->create_istream_xstring( string =
    pi_inputxmlstring ).
    *-- create the parser
      g_ixmlparser = g_ixml->create_parser( stream_factory =
    g_ixmlstreamfactory
                                      istream        = g_ixmlstream
                                      document       = g_ixmldocument ).
    *-- parse the stream
      IF g_ixmlparser->parse( ) NE 0.
    *if parser cannot be created then give error exit
        IF g_ixmlparser->num_errors( ) NE 0.
          EXIT.
        ENDIF.
      ENDIF.
    *-- we don't need the stream any more, so let's close it...
      CALL METHOD g_ixmlstream->close( ).
      CLEAR g_ixmlstream.
    *get the number of main nodes of the XML document
      g_ixmlnodemainlist = g_ixmldocument->get_children( ).
    *set number of elemtns
      g_inummainelements = 0.
      g_imainelementsctr = 0.
      g_inummainelements = g_ixmlnodemainlist->get_length( ).
      g_ifirstlevelctr = 0.
    *loop through the document till all have nodes have been covered.
      WHILE g_ifirstlevelctr LT g_inummainelements.
    *get the first node
        g_ixmlnode = g_ixmlnodemainlist->get_item( g_ifirstlevelctr ).
    *check the type of node
        g_isnodeelement = g_ixmlnode->get_type( ).
    *if node is not of type Element then continue
    *because we have got to read only text from element nodes.
        IF g_isnodeelement NE c_nodeelement.
          g_ifirstlevelctr = g_ifirstlevelctr + 1.
          CONTINUE.
        ENDIF.
    *get nodes of the element just found.
        g_ixmlnodelistmainelement = g_ixmlnode->get_children( ).
    *get number of children of main element
        g_inumchildelements = g_ixmlnodelistmainelement->get_length( ).
    *loop trhough the number of children
        WHILE g_imainelementsctr LT g_inumchildelements.
          g_ixmlnodemainelement = g_ixmlnodelistmainelement->get_item(
        g_imainelementsctr ).
    *get type of node
          g_isnodeelement = g_ixmlnodemainelement->get_type( ).
          IF g_isnodeelement NE c_nodeelement.
            g_imainelementsctr = g_imainelementsctr + 1.
            CONTINUE.
          ENDIF.
    *get name of the node.
          g_selementvalue = g_ixmlnodemainelement->get_name( ).
    *get children of node
          g_childnodelist = g_ixmlnodemainelement->get_children( ).
          g_inumchildren =  g_childnodelist->get_length( ).
          g_ichildelementcounter = 0.
    *while there are number of children of node.loop through
          WHILE g_ichildelementcounter LT g_inumchildren.
    *get the child node
            g_childnode = g_childnodelist->get_item(
    g_ichildelementcounter ).
    *check the type of node
            g_isnodeelement = g_childnode->get_type( ).
    *if node is not of element type continue
            IF g_isnodeelement NE c_nodeelement.
              g_ichildelementcounter = g_ichildelementcounter + 1.
              CONTINUE.
            ENDIF.
    *otherwise get element name
            g_selementname = g_childnode->get_name( ).
    *get value stored in this node.
            g_selementvalue = g_childnode->get_value( ).
            g_numelem = g_inumchildren - 1.
            IF g_ichildelementcounter EQ g_numelem.
              CONCATENATE: g_slinedata g_selementvalue
                      INTO g_slinedata.
            ELSE.
    *      store the value of noide in delimiter ~ line
    *      Check for Invalid characters in file
                IF g_selementvalue CA '&'.
                  REPLACE ALL OCCURRENCES OF '&amp;' IN g_selementvalue WITH
                endif.
    *if value contains delimiter then error
                IF g_selementvalue CA '~'.             
                  V_MSG = text-003.
                  EXIT.
                ELSE.
                  CONCATENATE: g_slinedata g_selementvalue '~'  .
                  INTO g_slinedata.
                ENDIF.
            ENDIF.
    *continue
            g_ichildelementcounter = g_ichildelementcounter + 1.
          ENDWHILE.
          g_ichildelementcounter = 0.
    *increment the main element counter by one to go to the next node
          g_imainelementsctr = g_imainelementsctr  + 1.
    *move the current delimiter line creted to internal table to be given
    *back to the calling program
          MOVE g_slinedata TO wa_xmldata-data.
          APPEND wa_xmldata TO it_xmldata.
          MOVE '' TO g_slinedata.
          MOVE '' TO wa_xmldata-data.
        ENDWHILE.
    *increment counter to move to hte next node.
        g_ifirstlevelctr = g_ifirstlevelctr + 1.
        g_imainelementsctr = 0.
      ENDWHILE.
    ENDFORM.                    "f_create_xmltable
    XML structure
    <?xml version="1.0" encoding="utf-8"?>
    <ABCInbound xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="C:\XYZSchema\ABCInbound.xsd">
    <HH>
    <RecordType>HH</RecordType>
    <Source>ABC</Source>
    <Destination>XYZ</Destination>
    <TimeStamp>20050909220546</TimeStamp>
    </HH>
    <BH>
    <RecordType>BH</RecordType>
    <DocType>AB</DocType>
    <Reference>2205516125</Reference>
    <DocumentDate>20080909</DocumentDate>
    <PostingDate></PostingDate>
    <CompanyCode>ABC</CompanyCode>
    <Currency>INR</Currency>
    <ExchangeRate>1.0000</ExchangeRate>
    <Park></Park>
    <ItemNumber>2</ItemNumber>
    </BH>
    <BL>
    <RecordType>BL</RecordType>
    <Reference>2205516125</Reference>
    <RefLineItem>1</RefLineItem>
    <AcctType>K</AcctType>
    <DrCrIndicator>H</DrCrIndicator>
    <Account>01000003</Account>
    <Amount>364.00</Amount>
    <VendorName-1>TOM &amp; JERRY IS MY</VendorName-1>
    <VendorName-2> NAME TO BE PAID</VendorName-2>
    <VendorName-3>1987566Z</VendorName-3>
    <VendorName-4>22</VendorName-4>
    <Street>UCX STREET</Street>
    <City>ROAD 4</City>
    <PostalCode>515004</PostalCode>
    <Country>IND</Country>
    <ContactPerson></ContactPerson>
    <AlternatePayeeCode></AlternatePayeeCode>
    <AlternatePayeeName-1></AlternatePayeeName-1>
    <AlternatePayeeName-2></AlternatePayeeName-2>
    <AlternatePayeeName-3></AlternatePayeeName-3>
    <PaymentTerms></PaymentTerms>
    <BaselineDate></BaselineDate>
    <PaymentMethods></PaymentMethods>
    <Allocation></Allocation>
    <LineItemText>item text</LineItemText>
    <TaxCode></TaxCode>
    <TaxAmount>0.00</TaxAmount>
    <WHTaxCode></WHTaxCode>
    <WHTaxbase>0.00</WHTaxbase>
    <Fund></Fund>
    <FundCenter></FundCenter>
    <CostCenter></CostCenter>
    <InternalOrder></InternalOrder>
    <TaxAutomatically></TaxAutomatically>
    <SpecialGLIndicator></SpecialGLIndicator>
    </BL>
    <TT>
    <RecordType>TT</RecordType>
    <TotalRecords>1</TotalRecords>
    <TotalValue>222</TotalValue>
    </TT>
    </ABCInbound>
    when the above xml file is read and populated into ITAB, for element vendorname-2 which has a space in first position , that space is ignored.
    This is being used for a FB01 posting and vendor is paid based on Name1+Name2 printed on cheque and due to the space ignoring problem, the vendor name is displayed wrongly thus causing problems.
    I appreciate if someone could guide me thru and help me in solving this problem.
    How to preserve the leading or trailing space.
    g_selementvalue = g_childnode->get_value( ).
    when i check g_selementvalue, space is ignored.
    i will be greateful if someone could guide me through.
    Regards,
    Simha
    Edited by: Simha on Dec 11, 2008 10:49 AM

    0.02: A C C E P T: Request="NEXT"
    0.06: Fetch session state from database
    0.08: ...Check session ... owner
    0.08: ...Metadata: Fetch Page, Computation, Process, and Branch
    0.08: Session: Fetch session header information
    0.08: ...Metadata: Fetch page attributes for application ..., page 330
    0.08: ...Validate item page affinity.
    0.08: ...Validate hidden_protected items.
    0.08: ...Check authorization security schemes
    0.08: Session State: Save form items and p_arg_values
    0.08: ...Session State: Save "P330_PROJECT" - saving same value: ""
    0.09: ...Session State: Saved Item "P330_SHUTTLE" New Value=""
    0.09: Processing point: ON_SUBMIT_BEFORE_COMPUTATION
    0.09: Branch point: BEFORE_COMPUTATION
    0.09: Computation point: AFTER_SUBMIT
    0.09: Tabs: Perform Branching for Tab Requests
    0.09: Branch point: BEFORE_VALIDATION
    0.09: ...Evaluating Branch: BEFORE_VALIDATION type: "REDIRECT_URL" button: 12904321314585385 branch: (Unconditional)
    0.09: Perform validations:
    0.09: ...Item Not Null Validation: P330_SHUTTLE
    0.09: ...Validation did NOT pass
    This is from debugging, what does the point "0.09: ...Session State: Saved Item "P330_SHUTTLE" New Value="" "
    mean ? I think it means, that the Shuttle somehow sets itself to null ?! but i dont understand why...
    There is no computation or anything that deletes the shuttle, the page is quite small, there is not much more than the shuttle and the validations and computations to give default value and transforming the shuttle items into a collection to continue working with the IDs.
    I dont check whats the problem
    Edited by: user12154443 on 21.07.2010 09:47

  • Problem with return value of stored function

    Hi,
    I've made a stored function that insert a new row in a table and return the primary key (see at the end of the message the function script).
    In VS 2005 with Visual basic, using the designer, I've created a dataset containing a tableAdapter to be able to use the Pl/SQL function.
    My problem is I can't get the proper return value. The VB code below works without error except that I get 0 as value.
    What's wrong?
    Cheers,
    Sebastien
    VB code
    Dim myAddSession As New ICISDataSetTableAdapters.AddSessionTableAdapter
    Dim intSessionId As Integer
    intSessionId = myAddSession.ADD_SESSION(tbxUsername.Text, _
    tempFolder.Substring(tempFolder.LastIndexOf("\") + 1), _
    "toto", Environment.GetEnvironmentVariable("COMPUTERNAME"), _
    myLevelAccess.icisUserId, myLevelAccess.levelId)
    The debugger tells me that the ADD_SESSION function return value is Object.
    add_session PL/SQL script:
    CREATE OR REPLACE FUNCTION ICISSEC.add_session (
    orausername IN icis_session.ora_user_name%TYPE,
    ctxsessionid IN icis_session.ctx_session_id%TYPE,
    ctxsessionname IN icis_session.ctx_session_name%TYPE,
    ctxservername IN icis_session.ctx_server_name%TYPE,
    icisuserid IN icis_session.icis_user_id%TYPE,
    startlevelid IN icis_session.start_lvl_id%TYPE
    RETURN icis_session.ICIS_SESSION_ID%TYPE
    IS
    tmpvar icis_session.ICIS_SESSION_ID%TYPE;
    BEGIN
    INSERT INTO icis_session
    (ora_user_name, ctx_session_id, ctx_server_name,
    icis_user_id, start_lvl_id, ctx_session_name
    VALUES (orausername, ctxsessionid, ctxservername,
    icisuserid, startlevelid, ctxsessionname
    RETURNING icis_session_id
    INTO tmpvar;
    COMMIT;
    RETURN tmpvar;
    EXCEPTION
    WHEN OTHERS
    THEN
    RAISE;
    END add_session;

    naama wrote:
    that if a value is null to convert it to 0 ? . i mean in the part of the declaration of parameter
    Nope, can't do that in the signature. You'll need to handle that by validating the passed parameters at the start of the function.
    It's simple enough. In your case you might just as well test for NULL and fail the function like this:
        FUNCTION date_post_message(
                user_lock_in IN NUMBER,
                form_type_in IN NUMBER DEFAULT 0 ,
                date_in      IN DATE)
            RETURN BOOLEAN
        IS
            v_num NUMBER(1);
        BEGIN
            dbms_output.put_line('Value of parameters : user_lock_in : '||user_lock_in || ' : form_type_in : '||form_type_in||' : date_in : '||date_in );
            IF user_lock_in = 1 THEN
                RETURN FALSE;
            END IF;
            IF form_type_in IS NULL THEN
                RETURN FALSE;
            ELSIF form_type_in NOT IN (1,2) THEN
                RETURN FALSE;
            END IF;
            RETURN TRUE;
        EXCEPTION
        WHEN NO_DATA_FOUND THEN
            RETURN FALSE;
        END;In other use cases I would declare a local variable and process it like this:
            IF form_type_in IS NULL THEN
                l_form_type := 0;
            ELSE
                 l_form_type := form_type_in;
            END IF;Obviously the code should use the local variable rather than the parameter.
    Validating the values passed in parameters at the start of a function is good practice. If you really want to go to town, read about Design By Contract.
    Cheers, APC
    Edited by: APC on Nov 9, 2011 1:36 PM
    Added example, as requested by OP

Maybe you are looking for

  • RAM issues when plugging in external devices

    When I plug external devices in such as a USB drive, iPhone or a Firewire external HDD the RAM usage on my Mac shoots up beyond normal levels. This has only started happening recently. Say for example I plug in my iPhone to sync it. With only iTunes

  • Problems in r/3

    what are the problems in r/3 system?

  • Send connector and recieve connector

    i recently figure out that send connector and reciece connector are not at the same level in EMC, why should it like this if it is different, then please point me out some known issues about configuration about them

  • Help! CS5 illustrator keeps quitting!

    Hi, My illustrator CS5 on my iMac keeps quitting out.  At first I was able to open files but I'd get an error message about "previewing in overprint preview" - Something about too many spot colors... followed by a "not enough memory" error.  I've alr

  • System Using Load Balancing

    Hi gurus, we have to do sso to ECC through portal we have ECC system with three instance ,i am going for  creating SAP System using Load Blancing, what is the value of server port should i enter. regards K Naveen Kishore.