Oracle Reports has comma as decimal separator in XML output

In XML, numbers should be canonical, otherwise BI Publisher errors with java.lang.NumberFormatException.
(This regards the e-business suite, by the way.)
We have trouble with our invoice report, a modified RAXINV, that suddenly makes XML files with comma. Which brought our invoicing process to a screeching halt. "Suddenly" means that it used to work, but now it does not work. I think someone has changed a setting somewhere, but nobody admits to doing that.
Can anybody point out where to look? What is actually controlling the decimal separator used by Oracle Reports in the XML output?
Thanksalotinadvance,
Tore.

Thanks for replying, Tim. (and thanks for a most enlightening "meet the experts" session at OOW)
The language settings for RAXINV are controlled by an MLS language function. I have logged an SR, but the problem seemed to solve itself, so the SR was closed, and then the problem reappeared. Darn.
However, I did some more research on Metalink and found this bug that seem to match our problem:
Bug 5920262 - NLS_NUMERIC_CHARACTERS KEEP CHANGING DYNAMICALLY FROM SESSION TO SESSION
I've ordered patching, so i think we'll have this fixed soon.

Similar Messages

  • Report Parameter - Comma as Decimal Separator

    Post Author: Jóhan E. M. J. Sivertsen
    CA Forum: General
    I want to use comma as decimal separator in Crystal Reports Server XI.
    When designing the report in CR Professional (version 11.0.0.1994), my report both displays and understand comma as decimal separator when filtering with a parameter.
    When running the same report in CR Server XI InfoView, the report correctly displays numbers with comma (,) as decimal separator, but the parameter uses dot/point (.) as decimal separator.
    The report runs against Oracle 10g database.
    Regional and Language Options is set to Danish (comma as decimal separator) on both my Windows XP PC (running CR Professional) and the Windows Server 2003 SP2 (running CR Server XI).
    NLS_LANG is set to DANISH_DENMARK.WE8MSWIN1252 (comma as decimal separator) in the Windows registry. [HKEY_LOCAL_MACHINESOFTWAREORACLEKEY_OraClient10g_home1]"NLS_LANG"="DANISH_DENMARK.WE8MSWIN1252"
    The report parameters are written directly in the query "command" in the CR database expert.
    Why does the CR Server uses dot/point (.) as decimal separator in the parameter when CR Professional on my PC uses comma (,) as decimal separator in the parameter?
    Is it possible to change this, so CR Server (CR InfoView) uses comma as decimal separator? If so, how? Please remember that the numbers displayed in the report itselves are written with comma as decimal separator. It is only the parameter that wants dot/point (.) as decimal separator.
    ThanksJohan

    Hi Sabcat,
    You can temporarily rename the first channel to create header rows above the channels and temporarily rename the last channel to create channel header rows, such as unit information, then rename the channels back to their original values after the CSV export.  It's low-tech, but it works, and uses the full efficiency of the CSV data export, which runs in C++ code.  I have attached an example VBScript below which does this programmatically.  You will need to edit the DataFileSave() line to include the extra XML from the other posts to get your custom delimiter and comma options.  I don't know of any way to affect the NoValue text.
    Brad Turpin
    DIAdem Product Support Engineer
    National Instruments
    Attachments:
    ASCII Export with CSV Headers.zip ‏60 KB

  • Illustrator Graphs; Add totals to a column design; Need , (comma) as decimal separator

    Hi all
    I am editing a German Annual Report.
    I live in Australia.
    My system runs OS X 10.10.2 and my Adobe Suite CC is English (or Australian, if that does make a difference in some library)
    I need to create a bunch of columns and bar graphs.
    The label next to a single column or bar is taken automatically from the data spread.
    The decimal separator in the data spread is a . (full stop, or dot)
    I need it do display a comma on the final graph. This is standard in many European languages.
    If I enter a comma in the data spread Illustrator refuses to draw a graph for this value.
    How can I do this?
    Regards
    Romano

    Thank you Jacob
    I have to play with this and see what happens if I open data with , as a separator at a later stage.
    Also I got lots of clients on the go at all times. I can not change my OS settings for this and then unwittingly create a problem on other clients files.
    I am looking at solving this for a specific client on a specific project. Also the files may be edited at the clients site (in Switzerland [Swiss German System]) at a later stage.
    I just sent them a test file and they tell me the separator is still a . when they open and edit the data.
    If it is in deed a OS issue... ...that should change the moment the edit the graph me thinks...
    Regards
    Romano

  • Calling XML Publisher Report from OAF using the conc programs xml output

    Hi all
    I am trying to call a report from oa Framework. I have developed the Template in the Ms-Word and for Generating the xml data i used the view object defined in OAF, this approach is working for me...i used the following code to do this in the AM as mentioned in many places...
    public XMLNode getEmpDataXML()
    OAViewObject vo = (OAViewObject)findViewObject("EmpVO1");
    XMLNode xmlNode = (XMLNode) vo.writeXML(4, XMLInterface.XML_OPT_ALL_ROWS);
    return xmlNode;
    now i want that instead of the view object i should use the xml generated by the report which is registered as concurrnet program with output as xml.
    I am submitting the concurrent program and able to get the request id ...
    Now from here what should i do.. please suggest...

    Hi,
    I am thinking of doing something very similar. What i was planning on doing is submitting the request, and then redirecting the user to the concurrent request page. They can then use the standard functionality to view output and get the xml to use the appropriate template.
    i would be interested in how you get the concurrent request id from the submitted job.
    Many thanks
    Rupesh

  • XML to Oracle interface has same insert count regardless of XML input file

    Not sure how to get around this one... I have an interface that simply transfers the contents of an XML file to an Oracle table. I keep getting the same insert count and records regardless of the XML file i'm using.
    I was using a properties file with this, but then just decided to use memory to process the file. So, in the physical architecture of the XML topology object, I just specify the file name, xsd and schema.
    Any thoughts on what's happening/what to check?

    Hi,
    I'm having a similar problem.
    I've settted up an XML topology using the following jdbc driver and jdbc url:
    com.sunopsis.jdbc.driver.xml.SnpsXmlDriver
    jdbc:snps:xml?f=#XML_Import_Test.CurrFileName&ro=true&dod=true&lf=c:/odilog.log&ll=255
    Then in the designer I've created a Package with the following steps:
    1) Declare Step for the CurrFileName variable
    2) A Set Variable Step for the variable TempFileName. Here I place the value of the file that I would like to process
    3) A refresh Step for the CurrFileName variable with the following refresh code: select '#TempFileName' from dual
    4) an Interace that maps the xml file that I would like to import to an Oracle Db Table
    Now, when I run the package from the designer I've to hit two times the execute button to have the data in the db. The first time that I execute the interface the process is using the previous processed file.
    Any ideas on what could be the problem? what I'm doing wrong?
    Thanks a lot for all the help
    Ben

  • Report will not generate PDF only the XML output.  View XML icon disabled

    I created a PDF format report.
    When the report completes, View Output shows XML code not the PDF output.
    View XML button is disabled from Diagnostic on view request screen.
    Any thoughts. All settings are the same as when other XML to PDF reports are created. Not sure why this one is not working.
    Thanks

    I'm afraid this is not an installation issue, it rather sounds and application server one, at the operative level.
    I suggest you to post your thread at the Fusion Middleware or Developer Tools forums.
    You didn't provide any OS nor Oracle / AS version, I suggest you to specify this information when posting in future threads.
    ~ Madrid.

  • How to generate oracle report output in a comma separated format in reports

    Pls any one can help in generation of oracle report in comma separated format in reports 10g.....
    thanks,
    prasad.
    1,Chennai,Nokiasiemen,Convent junction,5535

    Use desformat=delimited.
    http://download.oracle.com/docs/cd/E15523_01/bi.1111/b32122/orbr_howto007.htm#i1040102

  • Why is it such a pain to use java in a country that uses commas as decimal?

    Why is it such a pain to use java in a country that uses commas as decimal separator?
    A few weeks back I've asked here about the keypad decimal key. For some reason, java doesn't map the decimal key to a comma on the Portuguese (Portugal) keyboard layout. I've got no answer and I ended up using a custom plainDocument on the JTextFields to replace all points with commas.
    Now, I've just spent the whole morning trying to store and use decimal numbers properly. For some reason, a Double/Float .valueOf method (or the corresponding parse method) simply ignores the locale in use and uses US defaults when parsing the string. I can't parse anything with commas in those methods and I should, as it is the decimal separator for the system and default locale being used by java.
    First of all, I shouldn't be expected to perform replacements on every single operation that comes with a comma and I obviously can't be expected to program my own locale checking to decide what decimal separator to use in each final system. Second, is there any way to work with numbers seamlessly, without having to know the locale of the end user?
    I'm sorry if this is all my fault for doing something completly wrong, I'm new to java and I did search around to no avail. I'm really frustrated with what seems to be a complete lack of support in java for locales other than the US one.

    Good old Cobol has the "DECIMAL-POINT IS COMMA" clause... And isn't it great? :)
    Second, is there any way to work with numbers seamlessly, without having to know the locale of the end user?Consider "123.456". In some locales, this number is one hundred twenty-three thousand, four hundred fifty-six. In other locales it is one hundred twenty-three and four hundred fifty-size thousandths. How will you be able to determine which, without a locale?That's not what I've meant. Java should know the locale and behave accordingly. I don't have to know the locale of the end user since it might vary greatly. My point is that if strings are flying around with commas and if comma is the decimal separator on the end user's machine, any method aimed at parsing a numeric value out of a string should regard commas as such. I'm constantly replacing dots with commas and vice versa which could cause trouble if a different locale is used.
    And I mean that as a rant. Given my inexperience with Java, there might be good reasons for such a behaviour as baftos argued. What I'm really interested is in finding the proper way to deal with this issue.
    Have you tried the NumberFormat.parse? I will now.
    Edited by: Smigh on Apr 9, 2008 9:21 AM

  • How to get an effective EXCEL Spreadsheet as output for Oracle report 10g

    Hi All,
    I am a newbie in Oracle reporting. I am using 10g. I have an oracle report whose output format is PDF from the user interface. User asked for Spreadsheet output. I changed the JAVA CODE "desformat" = Spreadsheet and got the report in Excel. But this report is not as expected as it has many blank cells, misplacements etc and looks really ugly. so Is there anyway to get an excellent spreadsheet as output.
    If anybody needs any more information on this, please let me know.
    Thanks in advance.

    Thank you so much Marwim. Your suggestion is very helpful. But I have some other Oracle reports which have only one .rdf file giving output in both pdf and Excel. This doubt might be silly. Almost same features were existing for the rdf what I am working now. For the remaining reports, it was given that DESFORMAT=SPREADSHEET , what this report lacks. I googled and found these links http://docs.oracle.com/html/B13895_01/orbr_paptoexcel.htm &
    http://docs.oracle.com/html/B13895_01/orbr_paptoexcel.htm#CIHFDJAJ . But they are not intended for excel output. they generate only web layout in spreadsheet form. Could you please suggest any other ways to achieve this.
    Thank you in advacne.
    Edited by: user07960 on Jan 10, 2013 10:47 AM

  • Oracle Report or JarperReport

    hello
    I need a recommendation.
    that serious better for the reports because he/she also listens something on Oracle Report.
    which gives me but advantages in the handling of reports JasperReport or Oracle Report????
    which those advantages are????

    buks, et-al,
    first let me give a brief "when to use reports" statement. Oracle Reports has been our (Oracle's) enterprise reporting tool for a very long time (more than 15 years) and is used by a very large customer base.
    it is true, that oracle reports does not necessarily solve all problems in the reporting space, as it's primary audience is high-volume, high-fidelity enterprise reporting environments. as such, it is a highly specialized deveopment tool for exactly this problem space.
    over the years features have been added to allow those enterprise reporting customers to branch out into other areas and technologies (e.g. java importer to use java code from inside reports, extensibility using java plugins, etc.) but our primary target audience was and is the reporting crowd.
    i understand that from a java developer perspective, oracle reports might not provide enough capabilities to tie in with the different kinds of java technologies, but our java plugins and other capabilities solve the requirements of our core audience.
    specialized tools that focus on reporting in the java space focus on making reporting accessible for java developers, following java desing patterns and ideas. however in most cases (and i have to admit, i don't really know JasperReport in any detail) do not provide a fully fledged enterprise reporting environment or require significant java knowledge to work productivly.
    but now to go into more detail regarding your particular questions :
    1. Jdev/OC4J integration/Developer suite always a release behind.
    Oracle Reports is shipped as part of the Oracle Developer Suite, Oracle Application Server EE, Oracle BI and Oracle BI Tools and as such is synchronized with the respective versions of JDeveloper and OC4J that come with those respective releases. Oracle JDeveloper is being released as a stand alone product inbetween Developer Suite Releases and the same is true for the Container for Java, in some instances.
    2. The use of existing methods implemented in BC4J application modules.
    you can use the Java Importer to access java classes using PL/SQL wrappers or you can create your own data source plugin that uses BC4Js.
    3. Difficulty in setting up the layout.
    The reports layout model is designed to allow highly complex reports which are common in enterprise reporting environments. The concepts of hierarchical frames, etc. can make it complex at times, especially for end-users or occasional users of the product. But since reports is really targeted at a professional development crowd, it is our goal to provide most possible freedom in creating reports.
    4. How do you create a consistent layout so that reports look alike?
    not quite sure what exactly you are looking for here, you can use templates to define look and feel similarities, or you can create one master report as a basis that your developers can use to add their own parts.
    5. Can you use more than one template for a report?\
    not at this point. the templating model of reports is restricted to one at a time. using multiple tempates brings the problem of merging definitions and dealing with conflicts inside the templates.
    6. How do you put a footer for every page of the report?
    you can put objects in the margin area of your report and by that create page headers and/or footers. you can also define objects to be printerd on first/last/all-but-first/all-but-last/all pages and that way create things like repeating table headers for a table spanning mutliple pages.
    7. How do you get the footer to print something different for the last page?
    see above, all-but-last print condition on objects in the layout area will print them on accordingly. you can also use format triggers to determine whether an object should be printed or not.
    for further technical questions, please post them on the Oracle Reports forum here on OTN.
    thanks
    ph.

  • Oracle report installation and system flow

    Please help me to give some inputs for the oracle development. The 12 reports wont integrate with the existing system because it was not built by us and also I want to know how the oracle report installation will be done. Please update my diagram scenario of the proposed solution as below
    Problems Faced in the existing system:-
    ================================
    •     12 Critical Finance Reports not generated according to user requirements
    •     Most of basic feature for reporting purposes are not implemented in current system which is in Oracle Apps 11i - E-Business Suite
    •     Takes a lot of time to prepare reports which are needed to be presented for management meetings.
    Proposed System
    ===============
    Expectation
    •     The finance dept plans to redevelop 12 new critical reports for their system
    •     The new implementation must be compatible with the future migration of current Oracle Apps 11i - E-Business Suite to Oracle Apps version 12.
    Solution
    The new development will be built using the oracle reports and won’t be integrated with the existing application. Oracle reports do not require any specific installation to be done at client PC.
    Advantages
    •     If it's from a desktop users may find it easier to fire the reports as they may not require to login to Apps.
    •     When the Report is displayed, the user can simply generate the report to several formats like PDF, XLS, HTML, RTF, XML etc.
    •     Oracle Reports is reliable: You can serve thousands of concurrent users in a secure and reliable manner.
    •     Oracle Reports has the ability to apply changes to a report based on a customization file in XML either at runtime or in batch.
    Proposed system flow for Oracle reports development
    =========================================
    12 reports to be installed onto the Red Hat AS3 system <--------------Red Hat AS3(Oracle E-Business Suite Ver 11i (Apps) and Oracle Database Ver 9.2.0.5) -----> Oracle Database

    There are groups specific to e-Business Suite. Please delete this post and post your inquiry there.

  • Generating an Oracle Report in Excel Format

    Do any one of you worked on Oracle Reports? I need some information as to whether a oracle report can be sent to the Excel as output?
    What technologies can I use here ?
    Thanks,
    Srini

    thx for replay,
    actually i need to custom one report ok.
    which report generate more than one sheets n excel n in every sheet i need to display data ok
    but i want to do all these thing through one concurrent program
    if u have any querie then ask me.i will explain more if u not got it.
    even its happen by XML publisher , but i want exact answer , how to do that.
    Edited by: user13642081 on Jan 6, 2011 4:54 AM

  • How to use Oracle Reports in an enterprise app, wrt file i/o?

    How are people architecting their enterprise applications to make use of Oracle Reports, with respect to transmission of the output?
    Do you have session beans which call the Oracle Reports JSP pages, and have them write the output to file names, the names of which you then transmit to the web layer? Alternatively, is there any practical way to "stream" the output directly from the EJB layer to the web layer and to the client?

    Hello David,
    When you call the report server via an URL, with DESTYPE=CACHE, the output is directly sent to the web browser.
    I am not sure to well understand you question, but if you want to control the report from a session bean, you can write a bean that call report server with an URL and get the output. Today, we do not have any EJB to control Oracle Reports.
    I am interested by the type of beans/architecture you would like to see in Reports, could you please send me a detailled example of a type of bean you would like to have ? ([email protected])
    Regards
    Tugdual Grall

  • How do I read a report's XML Output from a procedure?

    I am using E-Business suite 11.5.10.2 and XMLP 5.5. I have a report that will output XML. When this report is run, I can click the 'View Output' button and view the XML. This appears to be a temp file somewhere, because the temp id in the URL changes each time a new report is run. Here's my problem: After running this report, I want to execute a procedure that will loop through the XML output and transform it into HTML. I will then output this HTML to the screen. But I'm not sure how to pass this file as a parameter to my procedure.
    Is there anyway to allow a procedure to consume the XML output?
    The reason I'm trying to do this, is because we need to transform certain reports into Excel format. This is becoming a critical issue. After testing, I found that if I can extract data into an HTML table and display it on the screen, it dumps perfectly into Excel. I tested it using a cursor, but that won't work for production reports. There is a ton of logic built into these Oracle reports, so I need to go the XML route after running the report.
    Looking at previous posts, the current Excel output process seems to be a bit buggy. And, we can not upgrade to Excel 2003 at this time.
    Any ideas would be greatly appreciated!!!
    Thanks.
    Mark K

    Hello Klaus. I appreciate your response. While I'm no expert with XMLP, I have created about 8 production reports using XMLP. I am using XMLP 5.5, and not 5.6.1 I am using Excel 2002 and not Excel 2003. From other posts (shown below), it is my understanding that the Excel output may have bugs unless you're using XMLP 5.6.1 and Excel 2003.
    Please see the following posts for more information:
    Re: EXCEL output not displaying properly
    Excel output in 5.6 version
    Also, when I tried to output to Excel using my desktop version, I received an error message. But, when I switch to PDF format, it opens just fine.
    We are not prepared to upgrade Excel and/or install another patch for XMLP at this time. If I can do what I need to do programmatically, I will have much more leverage when it comes to troubleshooting. On the other hand, when I run into a bug with Oracle software, I usually have to wait for a patch.
    Thanks again.
    Mark K.

  • Project in dutch language, but use a point as decimal separator (not comma)

    I am working with Haley v9.4.1.71.
    I am preparing a demo for coming wednesday in which I want Haleys to work together with Siebel. I have now ran in to a problem for which I can’t find the solution. I have made the rules in the Dutch language wich all works fine. As it seems it also uses the dutch format for numbers, this means that the decimal separator is a comma instead of a point. Siebel however, requires the decimal separator to be a (point).
    Data delivered from Haleys to Siebel is in the current situation therefore converted incorrectly.
    My question is:
    How can I keep the dutch language, but use the point as a decimal separator (instead of the comma).
    Thing I have tried already:
    - Configure the file hi-servlet.properties with the Haley Interactive Configuration Tool
    The ‘Application Output Numeric Format’ is #,###,###,##0.###############
    This looks correct, but still in the User Interface the comma is used as the decimal separator.
    - Set the Windows Regional and Language Options (Tab Regional Options) to English(United States) and customized the Decimal symbol to a point.
    - I have also looked in the Folder C:\Program Files\Haley\Office Rules 2008\Language\Dutch but couldn’t find specific settings for the comma separator
    I hope you can help me solve this issue.
    Sake Kampen

    Thank you for the reply.
    The problem to upgrading to v10 is that we make use of a OPA <-> Siebel integration developed by the local Dutch Oracle pre-sales team. And I don't know if we can replicate what Oracle has done for us if we upgrade.
    But the problem I'm having doesnt lie with the connection but with OPA itself, in OPA the decimal separator is a comma where I want it to be a point. This problem only exists when I create a Dutch project, when I create an English project (on the same windows environment with the same regional settings) the decimal separator is a point.
    I have advanced a little bit, but I’m not quite there yet.
    Haleys seems to use the Regional settings of Windows.
    The regional settings of my Windows-installation was English, so I temporarily went to Dutch to see what the settings were there. Ofcourse the numberformatting was in the dutch way with a comma as the decimal separator. So I changed this to a point and clicked ‘Apply’. Then I changed the language back to English and clicked ‘Apply’.
    In Haleys the same error persisted.
    So I went back to the regional settings of windows, and there I went back to dutch (English was still selected) and saw to my surprise that the ‘corrected’ decimal separator for dutch was set back to its default.
    So I changed it again to a point and now kept the dutch regional setting, and tested Haleys again. The behaviour of Haleys then changed. But surprisingly the decimal signs are now deleted.
    e.g.:
    a value derived from a table in excel is transformed as follows
    in Excel: 28.5 ==> in Haleys: 285
    Provided by user in Haleys: 543.1234 ==> shown in Haleys summary screen 5.431.234
    I can provide decimal numbers and they seem to be accepted, but if I show them in the summary screen (using the%%) I see that every number is rounded to a whole one in the way described above.
    Are you certain this is a known issue that has been adressed in Haleys 10?

Maybe you are looking for

  • Can I create a dashboard layout on the fly?

    Hi all, I am trying to create a dashboard layout "on-the-fly" (dynamically). for example: I can create the following layout using MXML very easily: [CODE] <mx:VBox width="100%" height="100%" horizontalAlign="center"> <mx:Canvas id="dashboardCanvas" w

  • Wish I didn't upgrade to iTunes 10... Having issues.

    Ok so I have over 1200 songs on my iTunes and I saw it upgraded so I figure why not... that was a mistake as it is flawed at the moment. Some of my music is in MIDI format which was working fine on the previous version. Now ALL MIDI files read as Qui

  • Is there a way to track emails?

    Can anyone please enlighten me as to whether there's a way to tell when the email you've sent to someone has been read? I'm using the Mac Mail program, and I've heard that other email apps will let you do this, but I don't seem to find a way to do th

  • Validation and Read Only Problem

    Hello there, I have this problem. I have an item that I need to hide when the page runs, and this item needs to have a validation. So my question will be. How can I bypass the validation if the item is on read only? Any ideas? Regards

  • Kde4-config: command not found (after upgrade of kde4 to version 4.6)

    When I upgraded my KDE desktop to latest version, it seems that it has removed the kde4-config command. Unfortunately I don't know which package it belongs to? I get the error when I execute startkde script from ~/.xinitrc I have tested to re-install