Computer Science 3510
Practice Exam 1 Solutions

This is a closed book test. You may use one 8.5x11 sheet of prepared notes, written on both sides. You have 90 minutes. Answer all of the questions.

  1. Write a clearly legible T to the left of each of the following statements that is true, and a clearly legible F to the left of each that is false.

    1. A null-terminated string of length 30 occupies 30 bytes. False (It occupies 31, one byte used for the null-terminator.)

    2. If Node is a type, then expression new Node[15] returns a pointer of type Node*. True

    3. The heap is an area of memory that can change size while a program runs. True

    4. Most dynamic storage allocation is done in the static area of memory. False (It is done in the heap.)

    5. Storing a value at a memory address that is a dangling pointer can cause some minor problems, but the program usually recovers from them quickly. False (It typically has catastrophic effects from which recovery is impossible.)

    6. Normally, function implementations are not placed in a header file. True (Only a few function implementations, such as constructors for classes, are typically put into header files.)

    7. If a C++ program does not return memory to the free space pool when it is done with the memory, then the system will periodically sweep through memory and return unused memory to the free space pool automatically. False (If a program does not delete unused memory then the memory leaks away and can not be recovered by the program.)

    8. A dangling pointer is pointer to memory that has been deleted. True

    9. A hash table is an excellent tool to help to sort an array. False (Hash tables tend to mix up data, which has the opposite effect from sorting it.)

    10. All data structures have a logical size that is determined when the program is compiled. False (A stack, for example, can grow as more data is added to it.)

    11. You are allowed to add an integer to a pointer in C++. True

    12. An abstract data type hides the way in which it is implemented. True

  2. Which of the following are part of an abstract data type specification? (Select all that apply.) (Selected ones are shown in blue.)
    1. A type.
    2. Prototypes for functions in the abstract data type.
    3. A description of how the type is represented.
    4. Implementations of the functions.
    5. Conceptual meanings of the functions.

  3. Suppose that the following structure definition is given.
       struct Widget
       {
         int side;
         char* front;
         char bottom[22];
         Widget() {}
         Widget(int s, char* f, char *b)
         {
           side = s;
           front = f;
           strcpy(bottom, b);
         }
       };
    

    1. Show how to create a new widget called frodo with side = 4, front = "red" and bottom = "blue". Use the constructor.

          Widget frodo(4, "red", "blue");
          

    2. Show how to create the same widget frodo without using the constructor that has parameters. Build the widget using the parameterless constructor, and install the values yourself.
             Widget frodo;
             frodo.side = 4;
             frodo.front = "red"
             strcpy(frodo.bottom, "blue");
          

  4. When doing object-oriented programming, you typically find that functions have one less parameter than do corresponding functions that are used in procedural (non-object-oriented) computing. For example, a classical function to test whether a value x is present in a tree t has two parameters, x and t, but an object-oriented function has only one parameter, x. Explain why there is one less parameter.

    Each function that belongs to an object can access the object to which it belongs implicitly. So there is always one implicit parameter, the object in which the function resides.

    The implicit parameter is the one that occurs in front of the dot when the function is used. For example, if you write t.member(x), the implicit parameter is t. This compares with member(t,x) for the non-object-oriented function.

  5. The following C++ program is run. It causes a segmentation fault. Why? (Note: there was a problem with the html source of this question. It is fixed here.)
        #include <iostream.h>
        int main()
        {
          char* word;
          cin >> word;
          cout << word << endl;
          return 0;
        }
    

    No memory is allocated for word. So the line cin >> word is using an uninitialized pointer.

  6. What does the following function print when it is run?
         void test()
         {
           int* s;
           int* p = new int;
           int* q = p;
           int* r = new int;
           *r = 17;
           s = r;
           *r = 41;
           *q = *s;
           p = s;
           *r = 8;
           r = q;
           cout << "*p = " << *p << endl;
           cout << "*q = " << *q << endl;
           cout << "*r = " << *r << endl;
           cout << "*s = " << *s << endl;
         }
    

       *p = 8
       *q = 41
       *r = 41
       *s = 8
    

  7. The following function is supposed to test whether two strings are the same, provided upper and lower case letters are treated as equal. That is, 'a' and 'A' should be considered the same letter. It is supposed to return 1 if the two strings are the same (when case is ignored) and 0 when they are not the same. For example, sameUpper("abc", "ABC") = 1, but sameUpper("abc", "AbD") = 0.

    This function works by copying each string, replacing lower case letters by upper case letters, and then using library function strcmp to tell whether the resulting strings are the same. (strcmp(s,t) returns 0 if strings s and t are equal, -1 if s comes before t in alphabetical order, and 1 if s comes after t in alphabetical order.) The function also uses function toupper, which converts a lower case letter to upper case, and returns all other characters unchanged. So toupper('a') = 'A' and toupper(':') = ':'.

    There is a serious mistake in this function. Explain what the mistake is, and rewrite the function to avoid the mistake. You might need to make more than one change, but try to keep the spirit of the method instead of choosing a completely different method, even if you think this method can be improved upon. Make sure the external behavior of the function is correct.

        bool sameUpper(char* x, char* y)
        {
          char *cpyx, *cpyy;
          int xlen = strlen(x);
          int ylen = strlen(y);
          int i;
          
          if(xlen != ylen) return 0;
          
          for(i = 0; i <= xlen; i++) {
            cpyx[i] = toupper(x[i]);
            cpyy[i] = toupper(y[i]);
          }
          
          if(strcmp(cpyx, cpyy) == 0) return 1;
          else return 0;      
        }
    

    No memory is allocated for the copies. Just before the for-loop, add

        cpyx = new char[xlen + 1];
        cpyy = new char[ylen + 1];
    
    But now that these arrays have been allocated, we want to deallocate them before returning. So replace the if-statement at the end by
          bool result;
          if(strcmp(cpyx, cpyy) == 0) result = 1;
          else result = 0;      
          delete [] cpyx;
          delete [] cpyy;
          return result;
      
    The final function definition is as follows.
        bool sameUpper(char* x, char* y)
        {
          char *cpyx, *cpyy;
          int xlen = strlen(x);
          int ylen = strlen(y);
          int i;
          
          if(xlen != ylen) return 0;
          
          cpyx = new char[xlen + 1];
          cpyy = new char[ylen + 1];
    
          for(i = 0; i <= xlen; i++) {
            cpyx[i] = toupper(x[i]);
            cpyy[i] = toupper(y[i]);
          }
          
          bool result;
          if(strcmp(cpyx, cpyy) == 0) result = 1;
          else result = 0;      
          delete [] cpyx;
          delete [] cpyy;
          return result;
        }
      

  8. Write a function called stutter that takes a null-terminated string s as a parameter and produces, in newly allocated memory, the string that results from repeating each character in s twice. For example, stutter("abaac") = "aabbaaaacc". It should return a pointer to the memory where the new string is stored.

    There are many correct solutions. Here is one.

        char* stutter(const char *s)
        {
          int   slen   = strlen(s);
          char* result = new char[2*slen + 1];
          
          for(int k = 0; k < slen; k++) {
            result[2*k]     = s[k];
            result[2*k + 1] = s[k];
          }
          result[2*slen] = '\0';
          
          return result;
        }