Open topic with navigation

A Writer is a special type of processor used to write out records to a Staged Data table, Data Interface or Real time consumer (for real time responses). You can select the attributes you choose to write, and map them to attributes with a different name in your Staged Data table, Data Interface, or real time response.

Once data is in a Staged Data table, it may exported to a Data Store, used in another process, or used in a lookup.

Use

Use a Writer or Writers at the end of a process to report on its results, or to return a response to an external system.

For example:

• From a batch auditing process, you might write out all valid records to one table, and all invalid records to another table.
• From a batch cleansing process, you might write out the records that you are intending to migrate to a new system into a Staged Data table, and then check, or recheck, that data in other processes before export.
• From a real time cleansing process, you might use a Writer to issues responses to each incoming record, with various corrections made.

Configuration

Inputs

Any attributes that you wish to write to a Staged Data table, to a Data Interface, or as a real time response.

Options

Note that unlike most other processors, the Writer does not launch its configuration dialog when it is connected on the Canvas. This is because Writers are often connected to multiple streams. To launch the Writer configuration dialog, double-click on the processor on the Canvas.

Select the Type of target that you want to write to:

• Staged data, where you wish to write to a table in the OEDQ repository, or directly to an external Data Store
• Data interface, where you wish to choose the target of the data in job configuration rather than in the process, using data interface mappings
• Real time consumer, where you have a configured Real time consumer that will pick up the response messages.

Then, choose which attributes to write, from the selection on the left. Selected attributes for writing appear in the middle pane for mapping to the target. There are two options above the attribute list to show more information - Show Actual Attributes, which will display the actual attribute name where you are using the latest version of an attribute, and Show Data Types, which will display the data types of the attributes.

The Staged data table to write to can either be:

• Created using the New button (this creates a Staged Data table corresponding to the selected set of attributes - the attributes may be renamed if required)
• Selected from the drop down of existing Staged Data tables.

Use the Auto button at the bottom of the screen to map all selected attributes directly to a selected Staged Data table.

Use the Clear button to clear all mappings between selected attributes and attributes in the Staged Data table. (Note that the attributes and columns remain until deleted).

Click and drag attributes in the middle column to reorder them.

It is also possible to edit the definition of the Staged Data table after it has been created, either from the Project Browser, or by launching the Staged Data editor from the Writer configuration dialog using the Edit button:

Execution

Results Browsing

The data written by the Writer can be viewed in the Results Browser.

Example

In this example, a number of invalid records are written to a Staged Data table called Quarantine output: