Specifying a sort operation (DataStage®)

You can specify that the stage sorts data before processing it. You specify the sort operation on the Partitioning tab of the Input page to sort the data as it enters the stage.

Before you begin

You cannot sort data if you are using the Auto partitioning method.

About this task

You specify one or more columns in the input data to act as sorting keys. The operation sorts the key columns, and associated rows, into the specified alphanumeric order. If you require a more complex sort operation, you can use the Sort stage.

Procedure

  1. Select Perform Sort.
  2. Select Stable to preserve previously sorted data sets.
  3. Select Unique to specify that, if multiple rows have identical sorting key values, only one row is retained. If Stable is also selected, the first row is retained.
  4. Select one or more columns in the Available list to move them to the Selected list. These columns form the sort key.
  5. Right-click on each key column in the Selected list and select from the following options:
    Option Description
    Sort direction Select Ascending or Descending as required.
    Case sensitivity Select Case sensitive or Case insensitive as required.
    Sort as EBCDIC Select this option to sort characters as EBCDIC rather than ASCII.
    Nulls position Select First or Last to specify whether Null values should be sorted to the first or the last position.
    Usage If you are using a keyed partitioning method, specify whether the current column is to be used for sorting, partitioning, or both.
  6. If National Language Support is enabled, click the properties button in the sort box to open the NLS Locale tab of the Sort Properties window. You can view the current default collate convention, and select another one if required.