The following directives, applied at column level, enable the generation of synthetic data according to their defined rules.
If no directive is specified for a given column, the column is populated with the default value for column type; for example,
zero for numeric columns:
- Generation by dictionary
- Column values are obtained from a dictionary.
Data Express provides some pre-loaded dictionaries (for masking purposes), and allows you to add dictionaries as well. To generate using
a dictionary, you specify the dictionary name in the directive properties.
- Generation by domain value list
- Column values are obtained from a user-defined list. For example,if your column is a CHAR(1) and you want to assign it the
values M and F (as male and female) with probability distribution values to represent 60% M and 40% F, you can specify these
values in the directive properties.
- Generation by random value list
- Columns are randomly assigned values in the column's domain of definition.
- Generation by referential integrity (other file dependency)
- Column values are obtained from a list of values present in a column from another table. To ensure this, the parent and child
column must be of the same class.
Note: This feature is similar to the "filter by filtered list" feature in Data Subset Extraction.