Welcome to CS1315. Click on the python to add comments.
	This page removed for FERPA compliance

Final Exam Review Spring 2003: Questions on Objects

Questions, answers, comments?

1. A class defines the data and behavior of an object. An instance of a class is one particular object belonging to a class. For example, we created a class named slide. Class defines a type of object. Objects know methods, which are called using dot notation. Our class ?slide? knew two methods: show picture and and blockingPlay sound. The particular object (our slide) that was created when we passed a picture and a sound into the function as inputs was the instance. These ideas were first discussed in our introduction to methods and dot notation where methods such as capitalize only applied to certain types of data (in this example, strings). Remember that:

• All data in Python are actually objects
• Objects not only store data, but they respond to special functions that only objects of the same type understand.
• We call these special functions methods
• To execute a method, you use dot notation
• Object.method()

2. A function takes as input a certain type of data and then does something to it. The data could be anything from a directory to a picture to a string to a list. The type of data that the function needs is defined by the programmer when writing the program. Also in functional programming, a function is data that can be used as an input so that we can write functions that call other functions. A method is a special type of function that only applies to a certain class of objects. The methods that we have dealt with have mostly been predefined: for example, string.capitalize(). However, by defining a class (a certain type of object), we can create our own methods. While a function must be called in the form function(input), a method is called by object.method(). Methods are a somewhat stronger and better way to program since defining objects and classes is a REMOVED accurate way to model the real world.

3. Procedural programming focuses on process: to get our desired result, first we do this, then this, then this, etc. We write our function according to the actions that must be performed.

Functional programming is based on the idea that functions are data that can be used as inputs. Therefore we can write functions that call and/or manipulate other functions. This creates a process of layering functions to get the desired outcome. You don?t write complicated functions with large loops to process all of your data (as in procedural programming). Instead you just write small functions that process one piece of your data (using concepts of granularity and modularity). Since this allows you to do a lot in a few lines of code, it is useful when the problem is hard and/or ill-defined so that you can get as far is possible in the least number of lines. (One area where this is true is Artificial Intelligence research.) Functional programming also lets you use recursion, the most flexible and powerful form of looping.

Object oriented programming comes at the problem from a different direction. It starts with the real world and then attempts to program by modeling things (objects) that exist in the world. It is noun oriented, as opposed to the verb orientation of function and procedural programming. It focuses on defining the domain of the program (like people, cells, etc.). The three phases to this type of programming are:

1. Object-oriented analysis: Understand the domain. Define an object-based model of it.
2. Object-oriented design: Define an implementation. Design the solution.
3. Object-oriented programming: Build it.

4. When the same method name can be applied to REMOVED than one object, we call that method polymorphic. From the Greek ?many shaped?. A polymorphic method is very powerful for the programmer: you don?t need to know exactly what method is being executed, you don?t even need to know exactly what object it is that you?re telling to show(), you just know your goal: Show this object!

5. If we want to change how slides work, we change them in the definition of slides. We call that encapsulation: Combining data and behavior related to that data.

6. Being able to use other objects with our objects is powerful. Being able to make lists of objects, to be able to use objects (like picture and sound) in our objects. We call that aggregation: Combining objects, so that there are objects in other objects.

7. All software models the real world. The real world is made up of small component pieces: quarks make subatomic particles make atoms make compounds make cells make organs make people (or dogs, plants, sponges, etc.). Object oriented programming sees this phenomena and attempts to use it to help make an accurate version of the real world. This is very useful to programming. The closer the perspective on something, the REMOVED fundamental the activity is. This allows us to manage complexity. For example: A human being is a very complex organism that can carry out very complicated activities. Just looking at the whole human from the outside would not let us create a very effective model. But if we break each part of the human down into form and then function all of the way to the level of the cells, we will reach something simple enough for us to be able to understand and then model. Specifically, object oriented programming looks at cells to model the world since cells are independent, indivisible, and interact in standard ways (just like a good function). Cells also scale well ? they can be added together to create REMOVED complex structures. Other characteristics of cells that are emulated in object oriented programming:

• Complexity: Distributed responsibility
• Robustness: Independent
• REMOVEDpporting growth: Same mechanism everywhere
• Reuse: Provide services, just like in real world

Link to this Page

• Sp2003 Final Exam Review last edited on 30 April 2003 at 10:25 am by lect02.physics.gatech.edu