Chapter6, Pause Reflect

• 6.1--6.8

• 6.9--6.14

• 6.15--6.18

• 6.19--6.25

• 6.26--6.31

• 6.1: You use a stereo by manipulating the buttons, e.g., to change radio stations, to play a given track on a CD, or to adjust the volume. The dials/displays provide a means to determine the values of "private data", again, the track or time of a track of a CD, the volume setting, the frequency of the radio station. Component stereos make it possible to replace one part without replacing all parts. However, some components aren't compatible with others. For example, some speakers require lots of power to drive them, so low-power amplifiers or tuners won't work.

  Just as a header file provides clues as to how a class can be used, but doesn't show how it's implemented, the buttons/dials on a stereo provide clues as to how to use the stereo, but not how it's put together. The header file does give away some information since the private section is visible to the user/client even though the variables cannot be accessed. A see-through stereo might be similar: users could see the circuit-boards and so on, but not access or touch them except via the dials/buttons of the stero.

• 6.2: You may know how a soda machine works, it's somewhat complicated. For example: what part of the machine is responsible for giving back change? How are cans stored? How is the price of an item changed --- is the machine re-programmed?

• 6.3: There are lots of comments in dice.h so that clients of the class can be sure how to use the class and how the class works. Since there is no access to the implementation of the class (dice.cpp), users of the class need guidance to be able to use the class properly.

• 6.4: The data field mySides is private and hence not accessible to client (user) programs. The member function NumSides() provides access to the value of the data. The line cout << "# of sides = " << tetra.mySides << endl;

  will not compile (accept in the implementation file dice.cpp). Instead the access functionDice::NumSides() is used.

  cout << "# of sides = " << tetra.NumSides() << endl;

• 6.5: The local (to the member function Dice::Roll()) varaible gen is of type RandGen. The class RandGen is declared in the header file randgen.h.

• 6.6: To allow the user to enter the number of sides of the dice the following could be used (note that die is defined after the user-entered value is obtained.) int sides; cout << "enter # of sides : "; cin >> sides; Dice die(sides); // ... continue with program The number of rolls can be entered as well, a loop is needed for this to work, however.

• 6.7: The statement myRollCount++ can be used in place of myRollCount = myRollCount+1.

• 6.8: To implement a function that returns the last value rolled a new data member/instance field (called myLastRollValue in the code below, but the name can be anything) must be added to the class---the code below would be in the class declaration in dice.h: private: int myRollCount; // # times die rolled int mySides; // # sides on die int myLastRollValue; // last value rolled

  The function Dice::Roll would change as follows:

  int Dice::Roll() // postcondition: number of rolls updated // random 'die' roll returned { RandGen gen; // random number generator myRollCount= myRollCount + 1; // update # of times die rolled myLastRollValue = gen.RandInt(1,mySides); // in range [1..mySides] return myLastRollValue; } Finally, a new member function that returns the last value can be added. This function should be const since it doesn't change the state of a Dice object, it acts as an accessor and not a mutator member function. int Dice::LastRoll() const // post: returns value of last time Roll() was called { return myLastRollValue; }

• 6.9: Two functions are shown below: void GetName(string & first, string & last) // postcondition: return first/last names entered by user { cout << "enter first name: "; cin >> first; cout << "enter last name: "; cin >> last; } It's possible to do all input on one line: cout << "enter first and last name (with space between): "; cin >> first >> last;

• 6.10: One version of a function is shown below: void Roots(double a, double b, double c, double & root1, double & root2) // precondition: a,b,c coefficients of ax^2 + bx + c // postcondition: sets root1 and root2 to roots of quadratic // exception: if roots are imaginary, set to infinity { double discriminant = b*b - 4*a*c; if (discriminant < 0) { root1 = root2 = infinity(); // from math.h } else // get both roots { double base = sqrt(discriminant); root1 = (-b + base)/(2*a); root2 = (-b - base)/(2*a); } }

• 6.11: The values printed are 3 and 4.5. Because a function cannot modify value parameters, it doesn't matter what the function Mystery does to its parameters. This assumes that as shown in the code fragment in the book num and top are NOT global variables.

• 6.12: Here's one possible function header: void DayCalculator(int month, int year, int & weekDays, int& weekEndDays) // pre: 1 <= month <= 12 (1 == january, ... 12 == december) // 1700 <= year // post: returns in weekDays the number of weekdays (MTWThF) in the month/year // and returns in weekEndDays the weekend days (Sat/Sun) in the month/year

• 6.13: The memory location associated with the variable num in main is referred to by the identifiers first AND second in the function Change (see diagram below). Thus the statement first += 2; references the memory diagrammed below (defined for num in main) and changes the value there to 10 (8 + 2).

  The statement

  second *= 2; references the same memory and multiplies 10*2 to get 20, storing the value in the only memory location shown, yielding 20. void Change(int & first, int & second) { | / first += 2; | / second *= 2; | / } | / | / main() | / { +-----+ int num = 8; | | +-----+ Change(num,num); cout << num << endl; }

  If the value of num in main is initialized to 3, the value printed will be 10 (first 2 is added to num making it 5, then this value is multiplied by 2).

• 6.14: To swap to integers: void Swap(int & a, in t& b) // postcondition: interchange values of a and b { int temp = a; // need temporary storage a = b; b = temp; }

• 6.15: The loop below tracks both negative and positive numbers (zero is considered neither positive nor negative). const int SENTINEL = 0; int count = 0; int num; int numNeg = 0; int numPos = 0; cin >> num; while(num != SENTINEL) { if (num > 0) { numPos++; } else { numNeg++; } count++; cin >> num; } cout << "number of positive #'s = " << numPos << endl; cout << "number of negative #'s = " << numNeg << endl;

• 6.16: On most systems input is buffered so that entering

    This is the start, this is the end --- nothing
    is in between
  

  will indicate 7 words have been read. The words are "This", "is", "the", "start," , "this", "is", "the".

• 6.17: One modification is shown below word = "dummy"; // force loop to execute once while (word != LAST_WORD) { cin >> word; if (word != LAST_WORD) { numWords++; } }

• 6.18: Two sentinel values: const string LAST_WORD = "end"; const string LAST_WORD_II = "finish"; int numWords = 0; string word = "dummy"; while (word != LAST_WORD && word != LAST_WORD_II) { cin >> word; if (word != LAST_WORD && word != LAST_WORD_II) { numWords++; } }

• 6.19: Here's a modified program: int main() { string word; int numWords = 0; // initially no words int sum = 0; // sum of all word lengths ifstream input; const SMALL = 3; const MEDIUM = 7; int smallCount = 0; // # small (1-3 letter) words int mediumCount = 0; // # 4-7 letter words int largeCount = 0; // # 8+ letter words string filename = PromptString("enter name of file: "); input.open(filename.c_str()); // bind input to named file while (input >> word) // read succeeded { numWords++; if (word.length() <= SMALL) { smallCount++; } else if (word.length() <= MEDIUM) { mediumCount++; } else { // must be large largeCount++; } sum += word.length(); } cout << "number of words read = " << numWords << endl; cout << "average word length = " << sum/numWords << endl; cout << "# of small words " << smallCount << endl; cout << "# of medium words " << mediumCount << endl; cout << "# of large words " << largeCount << endl; return 0; }

• 6.20: Here's one function header that fulfills the task void FileStatistics(const string& filename, int & small, int & medium, int & large) // pre: filename corresponds to an accessible file that can be open for reading // post: small = # small words, medium = # medium words, large = # large words

• 6.21: The value of 1/2 is zero since integer division truncates. In contrast, the value of 1/2.0 is 0.5 since double arithmetic is used when one of the arguments is a double (the int is promoted to a double).

  The value of 20/static_cast<double>(6) is 3.3333 since the cast converts the 6 to 6.0 and double arithmetic is used.

• 6.22: Here's a program that does the specified task. Arguably this should be broken into a function for copying and a function for getting names of files.
  #include <fstream> #include <iostream> #include <string> using namespace std; int main() { string filename, word; ifstream input; ofstream output; cout << "file for input: "; cin >> filename; input.open(filename.c_str()); cout << "file for output: "; cin >> filename; output.open(filename.c_str()); while (input >> word) { output << word << endl; } return 0; }

• 6.23: The statements in question are: string s; int m,n; cin >> s >> m >> n; If the user types "hello 12 3" (without quotes) the value of n will be 3, the last int entered. If ther user types "hello 1 2 3 4 5" the value of n will be 2, but if the statement executes again (the reading statement) the values left on the stream are 3 4 5, so the value of n is 5 (the string s has the value 3 since that's the first white-space delimited sequence of characters processed).

• 6.24: Here's one solution: #include <iostream> #include <fstream> #include <string> using namespace std; void getAverage(ifstream& input, string& first, string& last, double& average) // pre: input open for reading // post: returns first/last name and average for 5 scores // returns value of -1.0 for average if reading fails { int k, value; int total = 0; if (input >> first >> last) { for(k=0; k < 5; k++) { input >> value; total += value; } average = total/5.0; } else { average = -1.0; } } int main() { ifstream input; string filename; string first, last; double average = 1.0; // prime loop cout << "enter filename: "; cin >> filename; input.open(filename.c_str()); while (average != -1.0) { getAverage(input,first, last, average); if (average != -1) { cout << first << " " << last << " " << average << endl; } } return 0; }

• 6.25: The WordStreamIterator class returns strings. The loop/program above needs to read int values. It's possible to convert strings to ints using the function atoi in strutils.h.

• 6.26: To change mindata.cpp (Program 6.9) to read double values, change the type of the variables number and minimum from int to double. No other changes are necessary.

• 6.27: There is neither a largest nor a smallest value for BigInt objects, so the right approach for solving a problem with extreme BigInt values is to use the first value read as the initial value of either the max/min value being calculated (see 6.4.2).

• 6.28: If the statement below is used, the average will be correctly printed. cout << "average word length = " << (double) sum/numWords << endl; This is because the cast operator has higher precedence than the division operator.

  If the statement below is used, the average is not correctly printed (note: typo in book).

  cout << "average word length = " << double(sum/numWords) << endl; Here the division happens first, and if sum and numWords are int values, the result of the division is a truncated int. The cast needs to ensure that a floating-point/double division happens.

• 6.29: (run program on your system with right include files). cout << "smallest int = " << INT_MIN << endl; cout << "largest int = " << INT_MAX << endl;

• 6.30: Here's a modification of mindatat2.cpp, program 6.10. #include <iostream> using namespace std; // determine maximum of all numbers in input stream // illustrates fencepost problem: first item is maximum initialization int main() { int numNums = 0; // initially, no numbers int maximum; // largest number entered int number; // user entered number if (cin >> number) // read in first value { maximum = number; // to initialize maximum numNums++; } while (cin >> number) // read in any remaining values { numNums++; if (number > maximum) { maximum = number; } } if (numNums > 0) { cout << "number of numbers = " << numNums << endl; cout << "maximal number is " << maximum << endl; } else { cout << "no numbers entered, no maximum found" << endl; } return 0; }
• 6.31: Here's a function header (and body). #include <iostream> #include <string> using namespace std; void maxStrings(string& first, string& last) // post: first is alphabetically first string read from input // last is alphabetically last string read from input { string word; if (cin >> word) { first = word; last = word; } while (cin >> word) { if (word < first) { first = word; } if (word > last) { last = word; } } }