For enhanced flexibility and performance, tableBASE tables are structured differently from the typical row and column tables found in spreadsheets. While a tableBASE table can easily contain the data from a spreadsheet, it can also contain data of far greater organizational complexity.
tableBASE tables are a collection of fixed-length data rows. Each row contains a key and unstructured data.
Applications that access tableBASE tables—for example, tablesONLINE—apply column definitions to the table data. The column definitions themselves (metadata) are contained in tables in tablesONLINE.
Besides data rows, a tableBASE table also includes a table definition that contains attributes such as row length, key location, key size, organization, and search method. The table definition is used to generate the primary Index.
Although tableBASE tables are usually loaded from, and ultimately stored on, a hard disk, tableBASE tables are entirely resident in memory during processing, providing a greater performance advantage over disk-based tables for many types of data.
Each table that is stored on the tableBASE library must have a unique name. Tables created for temporary use by an application can have almost any name.