It is possible to include data in a table visualization (or in labels and tooltips of some other visualizations) that is not part of the original data table. A virtual column connects to a remote data source and collects data from there. This can be useful if, for example, a large quantity of data is stored in a database and the Spotfire analysis file would be too large if the data was included. The virtual columns can contain plain text but a more common use might be if the collected data were to consist of images or other large files.
Note: A virtual column type needs to be registered programmatically. Virtual columns are not available by default.
Note: Virtual columns are not supported for tables using in-db data sources.
Differences between virtual columns and regular columns
In the context of the table visualization, there are no major differences except that you cannot sort by the virtual column. However, the virtual column is only part of the table visualization and not part of the data table. This means that:
No filter will be created for the virtual column.
No export of the underlying data table will include the virtual column. To export the data, the visualization, and not the data table, must be chosen. Copy will wait until the value has been collected in the virtual column and then copy the value.
Settings, such as changing the column's name, are not managed from Data > Column properties, but from the Virtual Columns tab in the properties page for the table visualization.
The virtual column will not be available for selection on axes in the other visualization types.
The virtual column may take some time to load, especially if the column contains images. However, the column will not be loaded all at once. It is possible to start working with the column and the contents will be loaded as you scroll down in the table. The remote data source from where the virtual column is retrieved may require you to enter a username and password.