Limitations for EIS imports and exports

The limitations or restrictions when using the enterprise service discovery or enterprise data discovery wizards or the imports and exports they produce are listed here.

The following sections have limitations:

The Workarounds for C and COBOL limitations and Workaround for missing arrow on Add to import list button sections have steps that can circumvent some of these limitations.

General limitations

Connections

For CICS® and IMS, you can statically set values on the connectionSpec and interactionSpec properties. You cannot dynamically (programmatically) set or get their values.

Languages

C
  • The UNION clause is not supported.

COBOL

JMS bindings

User interface

WebSphere Business Integration adapters

Workarounds for C and COBOL limitations

The limitations listed above for C and COBOL can be addressed indirectly. Accessing a generated business object can be achieved by switching to the J2EE perspective, creating a JavaServer Page (JSP), and then using the JSP to manually access the business object created by the enterprise service discovery wizard. Several examples of such JSPs are shown in WebSphere test environment setup in which JSPs manually access business objects generated by the enterprise service wizard.

In the COBOL section, an error is identified if the COMMAREA contains invalid data. The following steps show how to circumvent the problem.

  1. Start the J2C Java™ Bean wizard. Right-click your module and from the menu select New > Other. Expand J2C and select J2C Java Bean. The New J2C Java Bean window opens.
  2. Select the ECI or IMS resource adapter and click Next.
  3. Specify a JNDI lookup name for the ManagedConnectionFactory class or specify the connection properties. Click Next.
  4. If you specify a JNDI lookup name, currently there is a problem where the J2C code cannot browse the JNDI namespace on the WebSphere Process Server. However, you can type the name JNDI lookup name in and click Yes when you see the message: Specified JNDI name could not be found.
  5. Specify how you want to save the service in the J2C Java Bean Output Properties page. Click Next.
  6. In the Java Methods page, add your methods. Click Next. In the following page, add the method name. Click Next.
  7. In the following pages, add your input and output, beginning with input by clicking New beside the Input type field. In the Data Import page, specify the mapping, COBOL to Java, and the COBOL file. Click Next and on the Importer page click Query to specify the data structure you wish to import. For example, DFHCOMMAREA.
  8. In the Saving Properties page, add or revise any names and folders for your service.
  9. Repeat the steps just shown for your output or select Use the input type for output type.
  10. In the following page, specify the InteractionSpec properties such as the function name of the program at the server. Click Finish. The J2C Java Bean is generated. If you specified a JNDI lookup name, then you may receive an error stating that the RAR deployment failed, which you can ignore. You can add the JNDI lookup name using the Admin console.
  11. Return to the Business Integration perspective. Open the assembly editor for your module. Change the view to one where you can see the generated J2C Java Bean (for example, the Java view or the Physical Resources view). Drag and drop the implementation onto the assembly editor canvas.
  12. Make sure the J2C Connection Factory is specified on the server if set on the J2C Java Bean.

Workaround for missing arrow on Add to import list button

The missing arrow is a code point. The reason the code point is not displaying on Windows is that a font mapping is required for the font in use to display the widget text in the IDE. Font linking on Windows is fully described in the Fonts section of Globalization Step-by-Step.

To add a font mapping to Windows that will add the arrow, follow these steps:

  1. From the menu in the workbench, select Window > Preferences.
  2. In the Preferences window, expand Workbench and select Colors and Fonts.
  3. Expand the Basic folder in the tree view in the right pane. Select Dialog Font.
  4. Click Change. In the Font dialog box, note the font name that is displayed.
  5. Click Cancel in the Font dialog box. Click Cancel in the Preferences window.
  6. Using Notepad or another text editor, create a file called fontmap.reg with the following content excluding the Start of File Content and End of File Content lines.
    -----------Start of File Content---------------
    REGEDIT4
    
    [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\FontLink\SystemLink]
    "Tahoma"=hex(7):54,49,4d,45,53,2e,54,54,46,2c,54,69,6d,65,73,20,4e,65,\
      77,20,52,6f,6d,61,6e,00,00
    
    -----------End of File Content---------------

    In this sample content, the font displayed in the dialog in the preference page was Tahoma. If your font had a different name, replace Tahoma with the font name that was displayed in the dialog. Save the fontmap.reg file.

  7. Open Windows Explorer. Find the fontmap.reg file you just created. Right-click fontmap.reg. Select Merge from the context menu. Click Yes when prompted to add the information to the registry.
  8. A registry entry will be created under the following path:
    HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\FontLink\SystemLink

    This registry entry contains the font mapping from your dialog font used in the IDE to the Times New Roman font.

  9. Reboot your system and the arrow will display correctly.

Feedback
(C) Copyright IBM Corporation 2005, 2006. All Rights Reserved.