Solution for leveraging the performance benefits of getElementById, getElementsByTagName, and getElementsByClassName for simple CSS selectors when querying the DOM. Please refer to http://ryanmorr.com/abstract-away-the-performance-faults-of-queryselectorall for more detailed information regarding this project.

Usage mimics that of your standard selector engine, accepting any CSS selector to query the DOM. The context of the query can be limited by providing a DOM element as an optional second argument:

query('#foo');
query('.foo');
query('.foo.bar');
query('div', element);

An array rather than a node list is returned, making available all the methods of the Array prototype.

This project is dedicated to the public domain as described by the Unlicense.