MochiKit.Iter - itertools for JavaScript; iteration made HARD, and then easy


theSum = sum(takewhile(
        partial(, 10),
            partial(operator.mul, 2),

assert( theSum == (0 + 0 + 2 + 4 + 6 + 8) );


All of the functional programming missing from MochiKit.Base lives here. The functionality in this module is largely inspired by Python's iteration protocol [1], and the itertools module [2].

MochiKit.Iter defines a standard way to iterate over anything, that you can extend with registerIterator, or by implementing the .iter() protocol. Iterators are lazy, so it can potentially be cheaper to built a filter chain of iterators than to build lots of intermediate arrays. Especially when the data set is very large, but the result is not.



Iteration for JavaScript

The best overview right now is in my Iteration for JavaScript [3] blog entry. This information will migrate here eventually.

API Reference



The singleton MochiKit.Base.NamedError that signifies the end of an iterator


applymap(fun, seq[, self]):

applymap(fun, seq) -->
fun.apply(self, seq0), fun.apply(self, seq1), ...

chain(p, q[, ...]):

chain(p, q, ...) --> p0, p1, ... plast, q0, q1, ...


count(n=0) --> n, n + 1, n + 2, ...


cycle(p) --> p0, p1, ... plast, p0, p1, ...

dropwhile(pred, seq):

dropwhile(pred, seq) --> seq[n], seq[n + 1], starting when
pred(seq[n]) fails

every(iterable, func):

Return true if func(item) is true for every item in iterable.


Exhausts an iterable without saving the results anywhere, like list(iterable) when you don't care what the output is.

forEach(iterable, func[, self]):

Call func for each item in iterable, and don't save the results.

groupby(iterable[, keyfunc]):

Make an iterator that returns consecutive keys and groups from the iterable. The key is a function computing a key value for each element. If not specified or is None, key defaults to an identity function and returns the element unchanged. Generally, the iterable needs to already be sorted on the same key function.

The returned group is itself an iterator that shares the underlying iterable with groupby(). Because the source is shared, when the groupby object is advanced, the previous group is no longer visible. So, if that data is needed later, it should be stored as an array:

var groups = [];
var uniquekeys = [];
forEach(groupby(data, keyfunc), function (key_group) {

As a convenience, groupby_as_array() is provided to suit the above use case.

groupby_as_array(iterable[, keyfunc]):

Perform the same task as groupby(), except return an array of arrays instead of an iterator of iterators.

iextend(lst, iterable):

Just like list(iterable), except it pushes results on lst rather than creating a new one.

ifilter(pred, seq):

ifilter(pred, seq) --> elements of seq where pred(elem) is true

ifilterfalse(pred, seq):

ifilterfalse(pred, seq) --> elements of seq where pred(elem) is

imap(fun, p, q[, ...]):

imap(fun, p, q, ...) --> fun(p0, q0, ...), fun(p1, q1, ...), ...

islice(seq, [start,] stop[, step]):

islice(seq, [start,] stop[, step]) --> elements from
seq[start:stop:step] (in Python slice syntax)

iter(iterable[, sentinel]):

Convert the given argument to an iterator (object implementing .next()).

  1. If iterable is an iterator (implements .next()), then it will be returned as-is.
  2. If iterable is an iterator factory (implements .iter()), then the result of iterable.iter() will be returned.
  3. Otherwise, the iterator factory MochiKit.Base.AdapterRegistry is used to find a match.
  4. If no factory is found, it will throw TypeError

Built-in iterator factories are present for Array-like objects, and objects that implement the iterateNext protocol (e.g. the result of Mozilla's document.evaluate).

When used directly, using an iterator should look like this:

var it = iter(iterable);
try {
    while (var o = {
        // use o
} catch (e) {
    if (e != StopIteration) {
        throw e;
    // pass

This is ugly, so you should use the higher order functions to work with iterators whenever possible.

izip(p, q[, ...]):

izip(p, q, ...) --> [p0, q0, ...], [p1, q1, ...], ...


Convert iterable to a new Array



range([start,] stop[, step]):

Return an iterator containing an arithmetic progression of integers.

range(i, j) returns iter([i, i + 1, i + 2, ..., j - 1])

start (!) defaults to 0. When step is given, it specifies the increment (or decrement). The end point is omitted!

For example, range(4) returns iter([0, 1, 2, 3]). This iterates over exactly the valid indexes for an array of 4 elements.

reduce(fn, iterable[, initial]):

Apply fn(a, b) cumulatively to the items of an iterable from left to right, so as to reduce the iterable to a single value.

For example:

reduce(function (a, b) { return x + y; }, [1, 2, 3, 4, 5])


((((1 + 2) + 3) + 4) + 5).

If initial is given, it is placed before the items of the sequence in the calculation, and serves as a default when the sequence is empty.

Note that the above example could be written more clearly as:

reduce(operator.add, [1, 2, 3, 4, 5])

Or even simpler:

sum([1, 2, 3, 4, 5])

registerIteratorFactory(name, check, iterfactory[, override]):

Register an iterator factory for use with the iter function.

check is a function(a) that returns true if a can be converted into an iterator with iterfactory.

iterfactory is a function(a) that returns an object with a .next() method that returns the next value in the sequence.

iterfactory is guaranteed to only be called if check(a) returns a true value.

If override is true, then it will be made the highest precedence iterator factory. Otherwise, the lowest.

repeat(elem[, n]):

repeat(elem, [,n]) --> elem, elem, elem, ... endlessly or up to n times


Return a reversed array from iterable.

some(iterable, func):

Return true if func(item) is true for at least one item in iterable.

sorted(iterable[, cmp]):

Return a sorted array from iterable.

sum(iterable, start=0):

Returns the sum of a sequence of numbers plus the value of parameter start (with a default of 0). When the sequence is empty, returns start.

Equivalent to:

reduce(operator.add, iterable, start);

takewhile(pred, seq):

takewhile(pred, seq) --> seq[0], seq[1], ... until pred(seq[n]) fails

tee(iterable, n=2):

tee(it, n=2) --> [it1, it2, it3, ... itn] splits one iterator into n

See Also

[1]The iteration protocol is described in PEP 234 - Iterators:
[2]Python's itertools module:
[3]Iteration in JavaScript:



Copyright 2005 Bob Ippolito <>. This program is dual-licensed free software; you can redistribute it and/or modify it under the terms of the MIT License or the Academic Free License v2.1.