Lightweight and efficient iterators

npm install nanoiterator

var nanoiterator = require('nanoiterator')

var values = [1, 2, 3, 4, null]
var ite = nanoiterator({
  next: cb => process.nextTick(cb, null, values.shift())
})

ite.next(console.log) // 1
ite.next(console.log) // 2
ite.next(console.log) // 3
ite.next(console.log) // 4
ite.next(console.log) // null

Create a new iterator.

Options include:

{
  open: cb => cb(null), // sets ._open
  next: cb => cb(null, nextValue), // sets ._next
  destroy: cb => cb(null) // sets ._destroy
}

Call this function to get the next value from the iterator. It is same to call this method as many times as you want without waiting for previous calls to finish.

Overwrite this function to your own iteration logic.

Call callback(null, nextValue) when you have a new value to return, or call callback(null, null) if you want to signal that the iterator has ended.

No matter how many times a user calls .next(cb) only one _next call will run at the same time.

Optionally overwrite this method with your own open logic.

Called the first time ._next is called and is run before the _next call runs.

Optionally overwrite this method with your own destruction logic.

Called once when a user calls .destroy(cb) and all subsequent .next() calls will result in an error.

Signals if the iterator has been ended (_next has returned (null, null)).

Signals if the iterator has been fully opened.

Signals if the iterator has been destroyed.

If you want to convert the iterator to a readable Node.js stream you can use the require('nanoiterator/to-stream') helper.

var toStream = require('nanoiterator/to-stream')
var stream = toStream(iterator)

stream.on('data', function (data) {
  // calls .next() behind the scene and pushes it to the stream.
})

MIT