There are several different approaches to programming, but the one that is easiest for me to grasp is imperative programming. The imperative approach allows you to tell the computer exactly what you want it to do and how you want it done. The best analogy I can think of is a cooking recipe. You have a list of ingredients and a specific set of instructions to follow. Veer from the recipe and you may not be happy with the results. Veer too far from the recipe and your house could burn down.

Control structures are the basic building blocks of imperative programming. They describe what kind of instructions the programming language understands. I will explore control structures, using examples from both the cooking world - hey, this is a good analogy and I want to keep using it - and Python, since Python’s syntax is clean enough that you can see the control structures easily.

I kind of played with the existing names for these structures to give them what I think is a more poetic quality.

Sequence

The fundamental control structure doesn’t seem like a control structure at all. Sequence refers to the fact that one thing usually happens after another.

• Beat the eggs and the oil together in a large bowl
• Mix the sugar and vanilla extract into the bowl.

Here’s a simple sequence example.

name = "Brian"
print "Hello", name

You don’t try to beat the eggs in the bowl after you’ve put the sugar in. I suppose you could, but you are just making your life more difficult. Similarly, you don’t try to print the name before you’ve defined it. Bad things will happen if you even try.

Selection

Selection structures allow you to mark some code as optional.

• (optional) Mix in walnuts

I like walnuts, but not in cookies. I’m not going to mix them into the dough. Maybe I’ll just eat them here.

name = raw_input("Who are you? ")
if name == "Brian":
    print "Hey!", name, "is here!"

There are a few refinements to the Selection structure. The main refinement is the if/else structure. This allows you to perform one block of code if the test is true, and another block if the test fails.

Instead of:

if name == "Brian":
    print "Hey! Brian is here!"
if name != "Brian":
    print "What are you doing on Brian's computer?"

We can say:

name = raw_input("Who are you? ")
if name == "Brian":
    print "Hey! Brian is here!"
else:
    print "What are you doing on Brian's computer?"

The other main refinement is the if/else-if/else structure, which lets you chain tests together. Instead of:

if name == "Brian":
    print "Hey! Brian is here!"
else:
    if name == "Brooke":
        print "This computer is really yours, isn't it?"
    else:
        print "What are you doing on this computer?"

… we can say:

if name == "Brian":
    print "Hey! Brian is here!"
elif name == "Brooke":
    print "This computer is really yours, isn't it?"
else:
    print "What are you doing on this computer?"

WARNING

Pay attention to how the else-if refinement is spelled in your favorite programming language! It’s elif in Python, and in other languages could be elsif, elseif, else if, or something else I haven’t come across yet.

Repetition

The next control structure is repetition, which will execute a block of code repeatedly until some condition is true.

• Cook in a 350 degree preheated oven for 10 minutes

Don’t you hate it when they tell you what temperature the oven is supposed to be at the last minute? I suppose that’s why we’re suppose to read the recipe before we start cooking. Anyways, the repetition here is that you keep on baking until those cookies are done!

Repetition in programming has a very wide range of refinements. I’ll just look at the major ones.

The first one is the while-loop.

number = 0
while number < 10:
    print "We have been baking for", number, "minutes"
    number += 1        # The same as saying -> number = number + 1

The basic idea of a while-loop is this:

1. We set up a test. We had to create a variable in this case, to make sure the test has some meaning the first time through.
2. We apply the test. Is number less than 10?
3. If the test passes, we execute the block.
4. Hopefully that block contains some code which affects the test. We added one to number in our block.
5. We go back to 2, run this again. In fact, we keep going back to 2 until number is equal to or greater than 10.

Generally, any test that your programming language can understand in a sense of true or false can be used. This applies to Selection structures too, but I forgot to mention it then.

Another kind of repetition structure is the list-loop (better name? Somebody? Please?). The list-loop steps through each of the values in a list and executes its block with the current item in the loop.

breakfast = [ "steak", "eggs", "potatoes" ]
for item in breakfast:
    print "I like", item, "in my breakfast."

Some languages won’t let you change the original value of item in a normal list-loop. For example, in Python I can uppercase item in my loop, but the original will still be lowercase. Try it yourself:

breakfast = [ "steak", "eggs", "potatoes" ]
for item in breakfast:
    item = item.capitalize()  # This is a *method* available to String *objects*
    print "I like", item, "in my breakfast."
for item in breakfast:
    print "Item:", item

Replacing the original value of item is more of a challenge, and often not worth the effort involved. You could step through the list using indexes, and change things that way:

breakfast = [ "steak", "eggs", "potatoes" ]
for i in range(0, len(breakfast):
    breakfast[i] = breakfast[i].capitalize()
for item in breakfast:
    print "Item:", item

I think that approach is unattractive, though. I’ll do it if I have to, but I will use a prettier and easier tactic if it’s available. A better way is to build a new list based on a modified version of the old list. Python makes this easy with a language feature called list comprehensions, which are funny-looking but very easy once you get the hang of them.

breakfast = ["steak", "eggs", "potatoes"]
capitalized = [food.capitalize() for food in breakfast]
for item in capitalized:
    print item, "is a great part of breakfast"

Many other languages have similar approaches, such as Perl’s map function.

Remote

The last major control structure I am going to look at is remote code. This isn’t technically code on another machine, but I suppose it could be. Remote code was written outside of the main sequence of the application, but you can use it (repeatedly) from within the main sequence. I guess the easiest cooking equivalent would be to hire a person who makes the cookies for you.

Remote code usually takes the form of functions, classes (the source of the capitalize method used in Repetition), and modules (which are usually collections of classes or functions in a separate file).

def get_name():
    return raw_prompt("What is your name?")

def greet(name):
    print "Hello", name, "it's good to see you!"

greet( get_name() )

That’s it for now.

Added to vault 2024-01-15. Updated on 2024-01-26