Skip to main content

Welcome to CS

Section 18.2 Program Design

One of the secrets of solving large, complex problems is to do our best to turn them into a group of smaller problems. Then, we can try to solve these small problems and test our solutions one at a time.
🔗
The process of breaking a large problem up into smaller sub-problems is known as top-down design. Start with the big problem and then use this process:
🔗
1. Decide if you can solve the problem with a small number of lines of code (ideally ~10 or less). If so, great, solve the problem. (Or put on a list of problems you will solve when you are done breaking everything down.)
🔗
2. If the problem is too complex to solve with a small amount of code, break it up into parts. Then repeat this process (starting from step 1) with each of the parts.
🔗
As you identify parts of a task, think about what information you will need to give to the procedure responsible for that task so it can do its job. These will be the parameters for your procedure. In the case of a turtle graphics program, every procedure will need to know the name of the turtle it is supposed to use. We also may need parameters to specify things like a size or color.
🔗
Along the way, look for places where you are writing the same code (or code that is essentially the same) in multiple places. Those are places you can maybe use a procedure to do the common part of the work. You may have to use parameters to make the procedure flexible enough to work in multiple situations.
🔗
Don’t worry about coming up with a perfect design on your first try. Sometimes you will think a problem is going to be easy and when you start working on it, realize it is much harder than you anticipated. In those cases, you can stop and revise your design by breaking the complex procedure into smaller pieces.
🔗

Subsection 18.2.1 Top-Down Design Example

Say I want to draw a house using turtle graphics that looks like this:
🔗
My design process might look like this:
🔗
Can I drawHouse in ~10 lines of code? Pretty sure I can not. So let me break it down. Well, I need to:
🔗
  • drawWall by making a rectangle
  • drawDoor by making a smaller rectangle
  • drawWindow make a square with a cross in the middle. I’ll need to do this twice.
  • drawRoof which is a triangle
🔗
OK, lets start with the drawWall… That seems pretty easy. I bet I can do that in about 10 lines of code.
🔗
Moving on to the drawDoor. That seems easy too. And it is a rectangle, like the wall. Maybe I should make a drawRectangle that both of these use. If I do that, it needs to allow me to specify a width and height so I can make rectangles of different sizes.
🔗
drawRectangle does not sound too bad - I can make that in under 10 lines. And if I do, the door and wall will be a snap.
🔗
OK, how about drawWindow. Well, it might be more than 10 lines of code. But I can use the drawRectangle to make this easier. If I have it to use, I think this won’t be too bad. I can just draw four small squares to make a window.
🔗
drawRoof doesn’t sound too bad - I can do it in a few lines of code.
🔗
Now that I’ve broken everything down into manageable chunks, we are ready to start implementing…
🔗

Checkpoint 18.2.1.

Let’s make sure you understand the design listed above. Arrange the blocks to implement the rough design.
For the auto-checker to work, you need to define your procedures in this order: rectangle, drawWall, drawDoor, drawWindow , drawRoof, drawHouse. The main part of the program should come after the procedures.
Don’t forget to indent the body of each procedure, even if it is not real code yet.
🔗
🔗
You have attempted 1 of 2 activities on this page.
🔗
PrevTopNext