I've written this as the type detector. It works very well in all cases -- so far whatever I've tested, but fails only in one case.
First, here is the snippet: jsfiddle
var TypeOf = function ( thing ) {
var typeOfThing = typeof thing;
if ( typeOfThing === 'object' ) {
typeOfThing = Object.prototype.toString.call(thing);
if ( typeOfThing === '[object Object]') {
if ( thing.constructor.name )
typeOfThing = thing.constructor.name;
else if ( thing.constructor.toString().charAt(0) === '[' )
typeOfThing = typeOfThing.substring(8,typeOfThing.length - 1);
else
typeOfThing = thing.constructor.toString().match(/function\s*(\w+)/)[1];
} else {
typeOfThing = typeOfThing.substring(8,typeOfThing.length - 1);
}
return typeOfThing.toLowerCase();
} else {
return typeOfThing;
}
}
The problem is that if I define the function at parse-time then it will work perfectly fine:
function me () {};
var you = new me();
console.log('Type of you: ' + TypeOf(you)); // me
However if I just define it at run-time, then it won't work:
var me = function () {};
var you = new me();
console.log('Type of you: ' + TypeOf(you)); // nope
As far as I can see in the parse-time case the constructor is something like function me () {}, so I can get it with /function\s*(\w+)/, however in the run-time case the constructor is function () {}.
Is there any way to get this to work? And also, are there any other cases that this snippet might potentially fail to detect the type?
Update as @lonesomeday mentioned in the comments, it seems that I'm trying to get the name of an anonymous function. That looks a little bit scary to me. Is that even possible?
Update as @jackson said below, it seems that the correct output should be function not me. Is that really correct?
