Awesome

Iterator Chunking

A TC39 proposal to consume an iterator as either overlapping or non-overlapping subsequences of configurable size.

Stage: 1

Specification: https://tc39.es/proposal-iterator-chunking/

presentations to committee

• January 2024
• October 2024

motivation

It can be useful to consume a stream by more than one value at a time. For example, certain algorithms require looking at adjacent elements.

chunking

This is commonly solved for non-overlapping subsequences with a "chunking" method that works like the following:

const digits = () => [0, 1, 2, 3, 4, 5, 6, 7, 8, 9].values();

let chunksOf2 = Array.from(digits().chunks(2));
// [ [0, 1], [2, 3], [4, 5], [6, 7], [8, 9] ]

let chunksOf3 = Array.from(digits().chunks(3));
// [ [0, 1, 2], [3, 4, 5], [6, 7, 8], [9] ]

let chunksOf4 = Array.from(digits().chunks(4));
// [ [0, 1, 2, 3], [4, 5, 6, 7], [8, 9] ]

use cases for chunking

• pagination
• columnar/grid layouts, such as calendars
• batch/stream processing
• matrix operations
• formatting/encoding
• bucketing (using a computed chunk size, for iterators of known size)

sliding window

When overlapping sequences are needed, this is commonly called a "sliding window".

let windowsOf2 = Array.from(digits().windows(2));
// [ [0, 1], [1, 2], [2, 3], [3, 4], [4, 5], [5, 6], [6, 7], [7, 8], [8, 9] ]

let windowsOf3 = Array.from(digits().windows(3));
// [ [0, 1, 2], [1, 2, 3], [2, 3, 4], [3, 4, 5], [4, 5, 6], [5, 6, 7], [6, 7, 8], [7, 8, 9] ]

let windowsOf4 = Array.from(digits().windows(4));
// [ [0, 1, 2, 3], [1, 2, 3, 4], [2, 3, 4, 5], [3, 4, 5, 6], [4, 5, 6, 7], [5, 6, 7, 8], [6, 7, 8, 9] ]

use cases for sliding windows

• running/continuous computations, such as averages
• context-sensitive algorithms, such as pairwise comparisons
• carousels and their analogues (when applied to an infinite cycle)

prior art

other languages

JS libraries