CSC 161	Grinnell College	Spring, 2009

Imperative Problem Solving and Data Structures

Scheme-like Lists in C

Goals

This lab applies ideas of box-and-pointer representations and provides practice implementing lists in C.

Background: Lists in Scheme

The reading for this lab describes box-and-pointer diagrams as a graphical model for lists.

Draw box-and-pointer diagrams for each of the following lists:
```
   ((x) y z)
   (x (y z))
   ((a) b (c ()))
```

Implementing Scheme List Operations in Java

The reading also describes the implementation of lists in C and presents program ~walker/c/lists/scheme-lists.c that presents relevant C functions and several test cases.

The car function returns the contents of the data field in the listNode designated by list. This information is obtainable as first.data.
For the cdr function, we must choose whether to alter the current list, return a pointer to the tail of the current one, or return a new one. For illustration, we return a pointer to the tail of the current list.
We implement cons by building a new node and copying relevant information into the data and next fields.
The listPrint function uses a temporary variable listPtr to move progressively down the list node by node, printing as it goes.
The listInit function requires that we pass the address of a first pointer. The formal header is
```
   void listInit (struct ** list)
```
or equivalently
```
   void listInit (listType * list)
```
Assignment of NULL to this address uses the syntax:
```
   *list = NULL
```
The listDelete procedure recursively deallocates space for the rest of the list before it uses C's free function to deallocate space for the head of a list.

Copy program ~walker/c/lists/scheme-lists.c to your account, compile it, and run it. This program will serve as the basis for the remaining steps of this lab. Be sure to ask about any sections you do not understand.

The first part of the program contains function prototypes — function signatures that specify the function's return type and parameters. These prototypes conclude with a semicolon, indicating that the body of the functions will be defined later.
1. Change the return type of listInit to int in the prototype, but not in definition that comes later. Recompile and note what happens. (Then change the return type back for the next steps.)
2. Change the return type of cons to struct Node * in the prototype, but not in the later definition. Recompile and note what happens. (The typedef makes listType a synonym for struct Node *, and this experiment clarifies how such a typedef statement works.)
Add a function second to this program to return returns the second element in a list (if present) or an empty stringif the list is null or has only one element. (In this and subsequent exercise, you will want to add lines to main to test your methods.)
Add a function count which counts how many times a specified item appears on a list. count should have two parameters: the list (of type listType) and the desired item (of type char *) for the search. In writing this code, you should use an iterative approach.
Add a function last which returns the last item on the list.
The program ~walker/c/lists/scheme-lists.c deallocates all nodes for d's list and also sets d to NULL. However, listDelete would not affect variables a, b, c, or e. For these variables, the nodes have been deallocated, but the variables still refer to the old memory locations. (Thus, the program sets each of these variables to NULL explicitly.)
1. What happens if try to print list c or d immediately the statement listDelete (&d); (before the NULL assignments?
2. Why do you think you get this result?

This document is available on the World Wide Web as

http://www.walker.cs.grinnell.edu/courses/161.sp09/labs/lab-lists-c-1.shtml

created 4 May 2000 last revised 17 April 2009
For more information, please contact Henry M. Walker at walker@cs.grinnell.edu.