1. On the platform where the source code exists, compile the source code with the STRING-XREF option.

2. Transfer the STRING-XREF output (the .xrf file) to the machine where the Translation Manager is running. You must also transfer the source files.

3. On the Translation Manager, re-create the same directory structure for your source files as you had on UNIX. Your working directory must be the same as the one you were using when you generated the STRING-XREF file.

4. In Translation Manger, create a project, but do not choose Extract on the Data tab folder.

5. When you load the files into the platform, select the .xrf file that you created in Step 1.

6. Continue with the translation process as you would for any other application.

7. Verify that the translation is complete by building and compiling the translated application in the Translation Manager.

8. Using the Data Administration tool, dump the data definitions (.df) and data (.d) files from the project database that the Translation Manager created.

9. Transfer the .df and .d files to the platform where the translated application will run.

10. Create an empty OpenEdge database on the target platform with the same name as the project database that the Translation Manager created.

11. Load the .df and .d files into the new database.

12. Using the Procedure Editor or Application Compiler, connect to the new database (you must use xlatedb as the Logical Database Name) and to the application database and compile the original source code with the LANGUAGES option.

13. When running the compiled code, you must use the Language (-lng) startup parameter and specify the appropriate language.