Class 51: Stacks

Held: Wednesday, 29 November 2006

Summary: Today we consider abstract data types in general and a particular abstract data type, the stack.

Related Pages:

• EBoard.
• Sample Code (stacks.scm).
• Reading on Stacks.

Notes:

• Only a few more days for questions on Exam 3.

Overview:

• Abstract Data Types (ADTs).
• Stacks.
• Implementing Stacks in Scheme.
• Some Applications.

Abstract Data Types

• As you've noted, when working with data we find many different ways to organize the data.
  • If we want to access information quickly by number, we use a vector.
  • If we want to add and remove information, we use a list.
  • If we want to access information by name, we use an association list.
• Good programmers decide on the key operations they wish to perform and then find the best possible implementation of those operations.
• A natural set of related operations is often called an abstract data type.
• The corresponding implementation of those operations is often called a data structure.

Stacks

• Stacks are an abstract data type with a few simple operations.
• Think about a stack of books or papers on a desk.
• You can add something to the top of the stack. Computer scientists call this operation push.
• You can remove something from the top of the stack. Computer scientists call this operation pop.
• You can look at the top of the stack. Computer scientists call this operation top or peek.

Implementing Stacks

• It is fairly easy to implement stacks in Scheme.
• We represent the stack as a list, with the newest item at the front.
• To push something, we cons it onto the front.
• To pop something, we cdr the list.
• How do we change a list in place? We don't, really. Instead, we store it in a vector.
• To make things even more fun, let's use object-oriented programming techniques.

Some Activities

• RPN calculator.
• Eliminating deep recursion.
• Matching tags in HTML.