The Relations command performs a virtual join between two files at a time by creating a relationship between data in one file with data in another file.
You can analyze the related data as though it existed in one file, and you can combine fields from related files in the View. Once you establish a relationship, you can use the Extract command to create a new file containing the related data fields.The Relations command produces results similar to the Join command with the All Primary option selected.
Data relationships are hierarchical. A parent file can have one or more child files, and each child file can have children of its own. A parent can be related to any number of child files. Any two files to be related must have one key field in common.
Because relations are generally less processor-intensive and require less disk space than joins, they are particularly useful when working with several large data files. Processing related files can be slow, especially if there are numerous indexes.
For more information on the Relations command see Relations.
Relations uses less disk space than Join for three reasons:
| • | Join requires one or both files to be physically sorted |
| • | Join creates a new file that can be larger than both the primary and secondary files together |
| • | Joins with multiple files require multiple sorts (or indexes) and joins |
For more information on comparing the Join and Relations commands see Join Versus Relations.