Introduction to Programming in C

1
Introduction to Programming in C

• Class 18
• 03/31/2003
• Rohit Valsakumar

2
Operator Overloading Continued

• Rules on overloading operators
• Some special constructors
• Overloading ltlt and gtgt
• Some rules of class design

3
Rules on operator overloading

• At least one argument of the overloaded operator
  should be of a class type
• An overloaded operator can be friend of a class
  or its member function or an ordinary function
• Only existing operators such as ,-,/, etc. can
  be overloaded, new operators cannot be created
• We cannot change the number of arguments the an
  operator takes e.g., always two arguments
  whereas takes one argument only
• We cannot change the precedence of the operator
• The following operators cannot be downloaded
  viz., (.), (), operator ., ?
• Assignment operator () has to be overloaded in a
  special manner
• Operators and -gt also need to be overloaded in
  a special manner

4
Constructors for automatic type conversion

• class complex
• public
• complex ()
• complex (double re, double im)
• complex (int x)
• complex operator (complex c1, complex c2)
• private
• double re_part, im_part
• void main()
• complex c1, c2
• c2 c1 34

• In the main() function we notice that we are
  trying to add an integer with a complex object
• Ideally this will succeed only if the
  operator is overloaded to allow such an operation
• However, if the operator is not overloaded we
  still find that the operation succeeds
• This is because C compiler checks for and finds
  a constructor that takes a single integer value
  as argument to construct the object
• It uses that constructor to convert that integer
  value into a complex object and then adds the two
  complex objects together
• If such a constructor did not exist then the
  operation will not be carried out and the
  compiler will signal an error

5
Overloading Unary Operators

• Unary operators can be overloaded just like
  binary operators
• Unary operators like -, , -- are
  frequently overloaded
• Eg., with respect to complex class
• complex operator-(complex c1)
• complex c2
• c2.re_part -c1.re_part
• c2.im_part -c1.im_part
• return c2

6
Overloading Unary Operators

• The single argument in the function definition
  indicates to the compiler that it is an unary
  operator overload and not a binary operator
  overload
• Now following operations can be carried out
• complex obj1, obj2(4.5, 6.7)
• obj1 -obj2
• Similarly other unary operators can also be
  overloaded

7
Overloading ltlt and gtgt

• The insertion operator ltlt is used with cout
  or any other ostream object to send streams to
  the output
• It can also be overloaded like any other binary
  operator
• However there is a special technique to do it
• This is because in the case of the ltlt operator
  we dont know what to return

8
Overloading operator ltlt

• ostream operator ltlt (ostream outs, const
  complex c1)
• outs ltlt c1.re_part ltlt i ltlt c1.im_part
• return outs // remember this

9
Overloading operator ltlt

• Always remember to have as return type ostream
  when you overload the ltlt operator
• This means the function returns a reference to an
  ostream object
• This is the only way the operato ltlt overload
  function will work

10
Overloading operator gtgt

• istream operator ltlt (ostream ins, complex c1)
• ins gtgt c1.re_part gtgt c1.im_part
• return ins // remember this

11
Overloading operator gtgt

• The extraction operator (gtgt) is overloaded using
  most of the same rules as the insertion operator
  overload
• Only one thing changes the const modifier is
  removed form the parameter list
• Why do you think that is done?
• Please read chapter 8 thoroughly

12
Strings

• String is a sequence of characters
• There are two types of strings in C
• An array of characters terminated by a null
  character
• A class string is the second string type

13
C-string Values and C-String Variables

• In C strings are represented as an array of
  characters terminated by a null character
• A null character is represented as \0
• E.g., the string Program will be represented as
  eight indexed consecutive characters as shown
  below
• P r o g r a m \o
• The terminating null character is the compilers
  way of keeping track of the end of the string

14
C-string Values and C-String Variables

• A C-String variable is just an array of
  characters
• char str10
• The above variable can be used to store a string
  of maximum 9 characters and one null characters
• Let us try to store the previous string Program
  in str10
• Str0 P str1 r str2 o str3
  g str4 r str5 a str6 m
  str7 \0
• As seen above we need to occupy only eight of the
  ten available locations, the remaining two
  locations viz, str8 and str9 will be
  uundefined

15
C-string Values and C-String Variables

• A C string variable can be initialized at the
  point of declaration
• E.g.,
• char str20 hello world
• There will be a null character in the 12th
  location
• The last 8 locations of the above array will be
  undefined
• When defining at the point of declaration we need
  not specify the size inside the square braces
• E.g.,
• char str1 abc
• The size of the array will be automatically
  determined to be 4 characters
• However char str2 a, b, c will not
  have the effect of automatically introducing a
  null character at the end
• Thus size of str2 will be three and str2 is not
  same as str1