Creating a Job to deduplicate data
You can generate a ready-to-use Job to deduplicate data in a specific file in the Talend Studio metadata. Using the component settings of this automatically-generated Job, you can choose to output the duplicates and the unique values in two separate files or databases.
- Selecting the file you want to deduplicate.
- Choosing the columns on which to run the deduplicating Job.
- If required, defining a blocking key to partition the data to be processed. A blocking key is usually needed when there is a lot of data in the file.
- Choosing where to write the unique and duplicated records.
- Running the generated Job.
Procedure
-
Click OK to close the dialog
box.
The Cheat Sheet panel is displayed in Talend Studio.
-
On the cheat sheet icon bar, click the drop-down
arrow, and from the menu select Launch
Other....
The Cheat Sheet Selection dialog box is displayed.
-
Expand , select Deduplicate Data and click
OK to close the dialog box.
The corresponding page is displayed in the Cheat Sheet panel. This page guides you through the steps on how to create a ready-to-use Job on certain columns in a specific file.
-
Read the introduction and then click Click to
Begin.
This will expand the first step in the procedure: Select File.
-
Select the file to cleanse from the metadata connection and click
OK.
The next step in the cheat sheet is expanded.
-
Follow the instructions and switch between the wizard and the steps in the cheat
sheet page till the last step: Review and Run the Generated
Job.
The wizard configures all the components and the metadata in the repository according to the settings you defined in the wizard different views and generates the Job. Talend Studio switches to the Integration perspective to display the generated Job that should look something like the following:
Results
The unique and duplicate values in the file are identified and stored in the defined output files or databases. The generated Job is stored under the Job Designs node in the Repository tree view.