The idea is

- to obtain a file over HTTP

- split that file up in some way (that way should be provided as a function by the user)

- process a function (again provided) on each chunk in parallel to, as they say in Star Wars, use the cores

- bring the pieces back together using, you guessed it, a function provided by the user


That's the use case.  The solution based around node.js

- use formidable to load file

- use async to schedule functions (it has all these function entry points)

In theory it should be done in < 50 lines
This file was modified by JetBrains WebStorm 3.0.1 for binding GitHub repository