Extracts comments from a file into an array or mulit-dementional array when the "text" arg contains mulitple package or namepsace definitions.

Each entry in the returned array will be an object containing 3 properties

• start : The line number that the comment started on

• end : The line number that the comment ended on

• data : The contents of the comment

  var myComments = extract(str);
  yields : [
  {
  start : 12,
  end : 32,
  data : "the descript"
  },
  {
  start : 12,
  end : 32,
  data : "the descript"
  }
  ]

A few things of note:

• The data will NOT include the beginDoc, nor the endDoc strings.
• Comment prefixing is stripped
  - * [star space]
  - tabs
  - spaces
• Code blocks maintain indentation.
• Splitting on package or namespace. When a single file contains mulitple references to a "package" or "namespace" comments will split into multiple arrays -- treating the single source file as being mulitple files.

Split Example:

/**
* Class A
* @package foo <-- this designates a new "page"
*&#47;

&#47;**
* Something for A
* @method something
*&#47;

&#47;**
* Class B
* @package bar <-- this designates a new "page"
*&#47;

&#47;**
* Something for B
* @method something
*&#47;

var myComments = extract(str);
yields : [
            [ // the first "page"
                { 
                    start : 12,
                    end : 32,
                    data : "Class A ... "
                },
                { 
                    start : 12,
                    end : 32,
                    data : "Something for A ..."
                },

            ],
            [ // the second "page"
                { 
                    start : 64,
                    end : 96,
                    data : "Class B ... "
                },
                { 
                    start : 128,
                    end : 142,
                    data : "Something for B ..."
                },
            ,
        ]