Why would anyone need JavaScript generator functions?

Generators are presented as an odd, low-level JavaScript feature that can be hard to read in isolation, but useful in specific cases. The article focuses on three uses: lazy processing of data, building infinite iterators, and passing messages between functions.

Reading notes

#

• Generators are described as a low-level construct that works like a tool for building other tools.
• Their syntax is presented as unusual because of starred function definitions and the yield keyword.
• Lazy iterators are introduced as the most immediate use case.
• The Tim Tam example shows how generator-based pipelines can process items one at a time instead of all at once.
• Utility functions such as map, take, and forEach are used to make generator pipelines work.
• Laziness is said to help with large data sets by loading one item at a time into memory.
• Generators are also used to create infinite iterators, such as repeating a value or generating natural numbers.
• The article uses scanl, filter, pop, drop, and sieve to build a prime-number generator.
• It says generators and iterators are mutable, so consuming one item changes the sequence that remains.
• The article notes that iterator helpers are not built in yet, but libraries such as Itertools and IxJS can fill the gap.
• Message passing is presented as another major use, especially for emulating async/await with generators.
• It shows a generator-based asyncDo wrapper that resolves yielded promises step by step.
• The article also shows how generator message passing can simplify Either-based error handling.
• It ends by saying generators are useful for efficient data processing, infinite sequences, and communication between functions.