An incremental data definitions file is a data definitions file that contains the schema differences between two databases. When you load the incremental data definitions file into a target database, the target database will have the same schema as the source database. For example, you might use this to upgrade a production database to the same schema as a development database.

Note: If the data type of the extent of a field has changed, the Data tool writes a DROP statement to the .df file to delete the existing field, and then an ADD statement to re-create the field with its new definition. When the field is deleted from the database, the contents of that field are also deleted. To save the contents of the field, save the data before running this procedure, or rename the old field, create a new field separately, and write an ABL procedure that transfers the data from the old field to the new field. Then use the incremental dump procedure to update the rest of the schema, including deleting the obsolete old field.

To create an incremental data definitions file:

  1. Access the appropriate Data tool (the Data Administration tool if you are using a graphical interface or the Data Dictionary if you are using a character interface).
  2. Select the source database as the current working database. The source database is the database that has the schema that you want the definitions file to have.
  3. Choose Admin > Dump Data and Definitions > Create Incremental .df File. The Data tool lists all the connected databases.
  4. If you have more than two databases connected, choose the target database. The target database is the database to update. The Data tool prompts you for the file where you want to write the differences.
  5. Specify the file where you want to write the differences. The default filename is delta.df. The Data tool displays the file, field, sequence, and index names as it compares the databases.
  6. After comparing the database, the Data Administration tool or the Data Dictionary returns you to the main window.
    Note: If you use this option to create a .df file in conjunction with r-code files to update schema changes, you must load the .df file and recompile before you can run the new r-code. You must recompile because the Data tool reorders the indexes during the dump and load procedure.