1.3 Four Java Interfaces
to the Driver
The Driver for Delimited Text includes four Java* interfaces:
- InputSorter
- InputSource
- PreProcessor
- PostProcessor
These interfaces enable you to add extensions, which are optional.
The driver continues to function as before without extensions. However,
if you want to directly modify the behavior of the driver, but have
been unable to make these modifications from a style sheet or DirXML
Script, extending the Driver for Delimited Text can be useful.
By using Java classes that you write, you can use the interfaces
to customize the publish and subscribe processes in the following
ways:
Table 1-2 Customizing the Publish and Subscribe
Processes
Publish |
InputSorter |
Defines the processing order of multiple
input files.
The system where your driver is installed determines the default processing
order. For example, files on an NT system are processed in alphabetical
order. You can use the InputSorter to impose the processing order
that you require. |
Publish |
InputSource |
Provides data other than the files in
the default location for the driver to process.
For example, you could check an FTP server for input files
and then transfer the files to the local file system for processing. |
Publish |
PreProcessor |
Ties data manipulation required to prepare
input files for driver processing directly to the driver.
Before this interface was available, preprocessing was independent
of the driver. You could create a separate application that would
monitor another directory for input files, modify the files in some
way, and then copy the files to the input directory of the driver.
By creating a class that implements the PreProcessor, you can do
this type of preprocessing more directly. |
Subscribe |
PostProcessor |
Ties data manipulation required by the
application consuming Identity Vault output directly to the driver. |
These enhancements to the driver require Java programming.
For more information, see Using Java Interfaces
to Customize File Processing.