Special Columns

In the section on multi-tables we introduced the special column table_name which lists the name of the source table for each row returned from a multi-table query. There are two more data.world special columns called row_index and row_address. These columns exist for every table in a data.world dataset. The first, row_index, just refers to the order of the row in the table, starting with 0. The row_address column contains a unique URL generated for the row. Under the covers, SQL at data.world is implemented using semantic web technology. One of the key ideas of the semantic web is that every entity should have its own unique URL.

Unlike normal columns, row_index and row_address will not be returned by default using SELECT * in a query. Instead they must be explicitly listed. To query all of the data for the products table in our dataset and it’s associated special columns, we would write a query like this:

SELECT products.*,
       row_index,
       row_address
  FROM products

The resulting table would look like this: