Positronic Vibrations

I'm in love with prototype-based languages, but there only a few to play with ECMAScript (Javascript) and Self. Since I'm learning Python, I thought it would be a good experiment to implement prototypes in Python. For one thing, I would get to learn some of the meta capabilities and other advanced features. A pure thought experiment just for learning and seeing how far I could push another paradigm onto Python. Here is what I came up with:

class Prototype(object):
    def __init__(self, *args):
        self._parents = []
        for each in args:
            self._parents.append(each)
        
    def clone(self, *args):
        return self.__class__(self, *args)
    
    def __getattr__(self, name):
        for each in self._parents:
            try:
                return getattr(each, name)
            except AttributeError:
                pass
        raise AttributeError(name)

The implementation is suprisingly simple: three methods!

The first is the initialize method for new instances. It takes a list of objects that we will use as the parents of our new object. Of course, by simply calling Prototype's constructor with no arguments, we get a clean object. I made the ability to have multiple inheritance simple because of Self.

The second method is the clone method and it simply calls the constructor for another object makes itself the first parent with the rest of the arguments becoming the other parents. Nothing to it!

The last method is the real meat. It only get called by the Python engine when an attribute fails to be looked up. All I do is call each parent and if one succeeds, it returns the answer. This is a depth first search of the parent hierarchy. Simple.

And that's all we need to implement prototypes in Python. Now, I just need to explain some practical uses of this in some later blog entries.

How do I know all of this works? Here's my tests:

import unittest
class Test(unittest.TestCase):

    def testSimple(self):
        parent = Prototype()
        child = parent.clone()
        
        parent.value = 0
        self.assertEquals(0, child.value)
        child.value = 1
        
        self.assertEquals(1, child.value)
        self.assertEquals(0, parent.value)
        
        self.failUnlessRaises(AttributeError, lambda:parent.unknown)
    
    def testMultipleParents(self):
        parent1 = Prototype()
        parent2 = Prototype()
        child = parent1.clone(parent2)
        
        parent2.value = 2
        self.failUnlessRaises(AttributeError, lambda:parent1.value)
        self.assertEquals(2, child.value)

If anyone sees anything wrong in my implementation or un-Pythonic let me know. I'm still learning, but I thought it would be cool to show my progress.

Labels: prototypes, Python

posted by Blaine at 6:54 PM |  4 comments |  Post a Comment

I recently came to the realization that I've been fighting Javascript too much. It's because I've been going against the grain with it. It's like trying to force the OO paradigm in a functional language or the opposite. You might say that I've been treating Javascript like an OO language and you would be right. But, I've been forcing my class-based thinking on it and it's been not so nice.

I've been reacquainting myself with Self and Io again. And it hit me like a rhino slamming into me. I've been doing Javascript all wrong. I should have treating it like the prototype-based language that it is. Sure, it doesn't have the ability to inherit from multiple prototypes like Self, but the way to succeed is with prototypes. Forcing class-based OO is like walking up the rainbow and finding no pot of gold at the end.

First, let's write the following:

Object.prototype.clone=function() {
    var creator=function() { this.constructor=arguments.callee };
    creator.prototype=this;
    return new creator();
}

Object.prototype.mixIn=function(definitions) {
    for (var each in definitions) if (definitions.hasOwnProperty(each)) {
        this[each]=definitions[each];
    }
}

It's just two methods up on the Object prototype. Why clone is not part of the Javascript standard is beyond me. Clone is required in all the other prototype-bases languages I have seen. It is the only way to create new objects. Enough complaining, it was only a few lines of code. All my clone method does is make the object I call it on the prototype of the result. I get a clean object with all of the properties of the receiver inherited. Nice. This is the behavior we want. Next, I added a convenience method to Object to mix-in in other objects. This is helpful to keep my code organized. I can just create a hash object with my functions and properties like before. So far, this is standard stuff. But, here is how I use it:

Pounds = new Object();
Pounds.mixIn({
    of: function(value) {
        var result=this.clone();
        result.value=value;
        return result;
    },
    toString: function() {
        return this.value.toString() + " lbs";
    },
    value: 0
});

print(Pounds.of(5)); // => 5 lbs
print(Pounds); // => 0 lbs
print(Pounds.isPrototypeOf(Pounds.of(5))); // => true

Notice, normally, you define the constructor with the first letter of the name upper cased. I no longer need the constructor, so I upper case on the prototype that will be the example for all the instances that I clone off it. The example above is a measurement class. "Pounds" is my example so I default it with values that the objects that I clone from will have. This is how prototype-based languages work. It feels slightly unusual coming from the class-based background, but I think it makes my Javascript look a lot less alien. I like it.

Let's look at a more complicated example shall we?

Animal = new Object();
Animal.mixIn({
    name: "Unknown",
    sound: function() {
        return "?";
    },
    toString: function() {
        return this.name 
                + " goes " + this.sound() 
                + " and weighs " 
                + this.weight.toString();
    },
    weight: Pounds.of(0)
});

Cat = Animal.clone();
Cat.mixIn({
    name: "Unknown Cat",
    sound: function() {
        return "Meow";
    },
    weight: Pounds.of(10)
});

grendel = Cat.clone();
grendel.mixIn({
    name: "Grendel",
    sound: function() {
        return this.constructor.prototype.sound.call(this) + " and Purr";
    },
    weight: Pounds.of(20)
});

gracie = Cat.clone();
gracie.mixIn({
    name: "Gracie",
    sound: function() {
        return "Purr";
    }
});

print(Animal); // => Unknown goes ? and weighs 0 lbs
print(Cat); // => Unknown Cat goes Meow and weighs 10 lbs
print(Animal.isPrototypeOf(grendel)); // => true
print(grendel); // => Grendel goes Meow and Purr and weighs 20 lbs
print(gracie); // => Gracie goes Purr and weighs 10 lbs

"super" is implemented with "this.constructor.prototype". It seems wordy, but prototype on the object is not built-in to Javascript. Besides, you should use "super" sparingly. To me, this reads better and works better for inheritance. I have dropped all references to classes. This makes meta-programming easier as well (I simply walk up the prototype chain with "this.constructor.prototype" and iterate through the properties.

The constructor function used as a class always seemed awkward to me. It was inelegant, but the above fits Javascript coding better. The reason is because I'm now using it as a prototype-based language and not forcing class-based OO on it. The above is meant to get started playing around. Prototype-based programming is so cool and I'm still exploring all of the possibilities. There's little documentation, but the little that is well worth the effort to track down. Have fun!

The example above was tested in SpiderMonkey. A great little REPL for playing with Javascript.

Labels: javascript, programming, prototypes

posted by Blaine at 11:59 AM |  3 comments |  Post a Comment