Skip to main content

List Functions

List functions allow you to create and interact with lists.

For information on performing scatter and gather using list:tokenize(), and list:collect(), refer to the examples in Data Pipeline Examples.

Input

Below event is sent to ProductComboStream,

['Ice Cream', 'Chocolate', 'Cake']

Output

After processing, the following events will be arriving at each stream:

  • NewListStream: [[Ice Cream, Chocolate, Cake]]
  • ListAnalysisStream: [true, true, false, Chocolate, 3]
  • UpdatedListStream: [[Ice Cream, Chocolate, Toffee]]

Example

This example shows how to use basic list functions.

CREATE STREAM ProductComboStream (product1 string, product2 string, product3 string);

@info(name = 'Create-list')
INSERT INTO NewListStream
SELECT list:create(product1, product2, product3) AS productList
FROM ProductComboStream;

@info(name = 'Check-list')
INSERT INTO ListAnalysisStream
SELECT list:isList(productList) AS isList,
list:contains(productList, 'Cake') AS isCakePresent,
list:isEmpty(productList) AS isEmpty,
list:get(productList, 1) AS valueAt1,
list:size(productList) AS size
FROM NewListStream;

@info(name = 'Clone-and-update')
INSERT INTO UpdatedListStream
SELECT list:remove(
list:add(list:clone(productList), "Toffee"),
"Cake") AS productList
FROM NewListStream;

This query defines a stream ProductComboStream with attributes product1, product2, and product3. It then creates a list using these attributes and sends the list to the NewListStream.

The Check-list query checks if the created object is a list, verifies if it contains a specific value ('Cake'), and provides information about the list's size, value at index 1, and whether it is empty or not.

Lastly, the Clone-and-update query clones the list, adds 'Toffee' to the end of the list, and removes 'Cake' from the list. The resulting list is then sent to the UpdatedListStream.