select-columns

Description

select-columns adaptor selects a list of columns from a datatable in a specified order.

Columns can be selected based on a list of columns, text string matching, or a regular expression. These methods can be combined, so a list of columns can be supplied in addition to a regular expression to select additional columns.

Inputs

data Type: datatable Required: Yes The datatable containing columns to be selected.

column names Type: list Required: No The list of columns to be included in the specified order.

pattern Type: text Required: No Specifies the pattern (a text or a regular expression) to select columns whose name matches that pattern. Matching columns will be added after any columns specified in column names, and in the order in which they appear in the original datatable. The pattern is treated as a regular expression if it begins and ends with / (e.g. /.*/).

Outputs

data Type: datatable A datatable with the selected columns.

Examples

Example 1: Default behaviour.

Inputs:

• data:

  Year	Month	Sales (USD)	Sales (GBP)	Sales (EUR)
  2020	Jan	1458	1156	1345
  2020	Feb	9874	7835	9110
  2020	Mar	4562	3601	4173
  2020	Apr	3654	2885	3342

• column names:

  1. Month

  2. Sales (GBP)

Outputs:

• data:

  Month	Sales (GBP)
  Jan	1156
  Feb	7835
  Mar	3601
  Apr	2885

Example 2: Selecting columns using a regular expression.

In this example, we use the regular expression /^Sales .*/ as a value for the pattern input to select columns which start with Sales.

Inputs:

• data:

  Year	Month	Sales (USD)	Sales (GBP)	Sales (EUR)
  2020	Jan	1458	1156	1345
  2020	Feb	9874	7835	9110
  2020	Mar	4562	3601	4173
  2020	Apr	3654	2885	3342

• column names:

  1. Month

• pattern: /Sales .*/

Outputs:

• data:

  Month	Sales (USD)	Sales (GBP)	Sales (EUR)
  Jan	1458	1156	1345
  Feb	9874	7835	9110
  Mar	4562	3601	4173
  Apr	3654	2885	3342

Use Cases

• Remove duplicate columns after joining multiple datatables.

• Reorder columns before exporting.