Agree

Totally agree!

OK! I promise! (vulcan salute)

Well, I had to hack in the words to get this result. ;) Thanks!

Hello,

The test code is indeed sensitive (tricky to test memory consumption and run time).

But there is a good reason it fails here. Let me precise the error message: you reach a memory limit.
Indeed, your code is "taking the chunks one by one", so that's good.
But you transform the chunk into an array with .split('').
Therefore, you make a copy of the chunk. So it takes more memory than necessary (and also slower). Especially at the end of the loop with buffStr hanging around.
(memory threshold is pretty hard i know :p)

Good news: you are going to learn a new thing today : String.prototype.charAt().
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/charAt
Basically, you don't need to use arrays. Just do:

var buff = this.reader.getChunk();
buff.charAt(5);

Note that buff[5] also works. See http://stackoverflow.com/questions/5943726/string-charatx-or-stringx for the nuances.

And then your code pass all tests :)

Tests cannot get changed after more than 500 warriors have solved them. See #123 and #163.