The Column Selector Customizer contains two basic parts. The left side of the customizer allows you to configure how many datasets can be brought in for filtering. Each dataset added will add two additional custom properties to the Column Selector; an In dataset property and an Out filtered dataset property. Datasets can also be removed here, or moved up or down in the list. If there are multiple datasets, the columns from the first dataset in the list will be displayed at the top of the Column Selector, while the columns from the last will be at the bottom.
The right side of the customizer allows you to configure the settings for each dataset. When a dataset is highlighted on the left, we can see some basic information about it on the right, such as the Grouping Title and a list of all of the columns in that dataset. The Grouping Title is only used if there is more than one dataset in the Column Selector. In the component, each dataset's columns will be contained in a border and will display the Grouping Title. This can be configured to be anything, so that it is easier for a user to distinguish what each set of columns is for. In the Column Settings table, we see each one of the columns in that dataset listed out. Here, the Display column allows us to alter what name that column will display on the component, again allowing you to create names that are more meaningful to the user. Finally, the Excluded from Selection column allows you to exclude certain columns from being filtered. Columns that have this enabled will not show up in the list of columns on the component. This will not filter them out in the output dataset, but instead forces them to be in the output dataset.