Join

Description

Join two tables into one table, based on one or multiple shared columns.

Application

Data is often not stored in a single place in engineering applications. For instance, the design parameters, operating conditions and measured performances might be stored in different tables. The join function enables to combine these different tables to obtain one big table with a format more suitable for exploration or modelling purposes.

How to use

This function requires two different tables to work properly which have to share one or more common columns in order to join them (called index column below). The joining operation will join rows from the two tables when they share the same value for that selected column(s).

Available Join methods

Left		The resulting table will keep all information of the left table and additionally add all matching record between both tables. This can be used for example when we are mainly interested in the rows of the left table, but want to add information coming from another table.
Right		The method is very similar to the left join, but will keep all information of the right table instead.
Inner		The resulting table will only contain rows for which matching index columns is found in both tables. This can be used for example when we want a table that only show results for which we have all data.
Outer		The resulting table will contain all rows from the two tables, whether they have matching index columns or not. The resulting table will be bigger, but is more likely to have missing data. This can be used when we want a table containing all of the data from the two original tables. However, further processing might be needed.

The index column(s) selected in the fields Common column(s) from left/right table will all be kept in the final table. If the tables contain other columns with the same name on both tables (e.g. column_name), they will be both kept in the final table but renamed in order to differentiate them. In the left table the suffix _left will be added to the column name, _right will be added in the right table (for the example this would result in column_name_left and column_name_right).

Most of the time, the index column used to join tables is a unique identifier. However, sometimes more than one index column is required to uniquely identify a row. If multiple index columns are selected, the joining operation will join rows from the two tables when they share the same values on all the index columns. For this to work, the same number of index columns have to be selected for each table, and they have to be selected in the same order.

There are different join methods, depending on what information we want to have in the resulting table:

Examples

For these examples, let us consider the two following tables: The left table containing design parameters, the right table containing performance parameters.