Working with Extract Packages

Extract packages enable you to run multiple extracts from an application with a single operation. When you run the extract package, each extract in the package is run and the results from each of them are added to a single zip file. Optionally, you can combine some or all of the extracts into a single output text file.

Example Use Cases

  • The ability to combine multiple extracts into a single file enables you to create sectional files such as those used for the Oracle Hyperion Financial Management .APP files or Oracle Data Relationship Management import files.
  • Additionally, because not all extracts in a package have to be combined you can create an extract package that creates a single zip file with the relationships from all of your hierarchy files combined into a single output text file and your list file with all of your node level properties as a separate output text file.

Considerations

  • While individual extracts are at the dimension level, extract packages are at the application level. This means that you can create an extract package that contains extracts from multiple dimensions in an application.
  • You cannot include extract packages in other extract packages.
  • You can run an extract package to a global connection or a file. The output is always zipped, even if there is only a single extract in the package.

Best Practices for Combined Extract Files

When combining multiple extracts into a single output text file, consider the following best practices:

  • Each individual extract that will be combined into a single output file in the extract package should use the same file and data format options (such as the delimiter character, line feed style, and date and time formats, see Specifying Extract Options).
  • If you are creating a single section from multiple extracts in a combined file, you should select the same columns to extract (see Selecting Extract Columns) in each extract.
  • Use the extract package header and footer options (see Creating, Editing, Copying, and Deleting Extract Packages) that are appropriate for the intent of your output file. For example, if you are creating a sectional file with one extract with relationship information and another extract with node information, you may retain the individual extract header information on each extract. But if you are including three extracts into a combined relationship file, you may want to retain the header information only on the first extract (excluding the header information on the other two extracts) and the footer on the last extract file.