This is similar in concept to a SQL SELECT statement. It operates on simple text database files - each record is one line in the file, and data items are separated by the designated field separator character, generally tab. You can specify which fields to extract, and you can also use conditional expressions to pick out only certain records.
Input files can be in a few different formats - "merge" format with the field names in the first line of the file, "schema" format with the field names in a separate file, or name-value pairs.
The available conditionals are string equality, regular-expression matching, and numeric comparisons.