Ce cours initie aux bases de la programmation en utilisant le langage C++ : variables, boucles, fonctions, ... Il ne présuppose pas de connaissance préalable. Les aspects plus avancés (programmation orientée objet) sont donnés dans un cours suivant, «Introduction à la programmation orientée objet (en C++)». Il s'appuie sur de nombreux éléments pédagogiques : vidéos sous-titrées, quizz dans et hors vidéos, exercices, devoirs notés automatiquement, notes de cours.
Fonctions : arguments par défaut et surcharge
Loading...
Del curso dictado por École Polytechnique Fédérale de Lausanne
Initiation à la programmation (en C++)
192 calificaciones
De la lección
Fonctions
Cette semaine aborde un sujet fondamental en programmation : les « fonctions » qui permettent de beaucoup mieux structurer les programmes et d'éviter d'avoir à récrire plusieurs fois la même portion de code.
Conoce a los instructores
Jean-Cédric ChappelierDr.
School of Computer and Communication Sciences
Jamila SamDr
School of Computer and Communication Sciences
Let's end with two fairly easy-to-understand notions.
First, default arguments.
In function prototypes, we can
give default values
to some parameters of the function.
And in that case, it becomes unnecessary
to supply values, that is, arguments, to these parameters
when we call the function.
To define such a parameter
with a default value,
all that is needed is to add, in the function's prototype, an equals sign ('=')
after the parameter, followed by the default value
that we wish to give
to this parameter.
Kind of like an assignment.
And be careful,
these parameters with default values
must absolutely appear last
in the list of parameters of the function.
Let's take as an example a function which will display a line
of several characters.
The prototype of this function is here,
the name of the function is <i>affiche_ligne</i> (TN: means "<i>display_line</i>").
This function takes two parameters:
the first parameter will define the character
that we wish to repeat on the line.
The type, <i>char</i>, simply means that
the parameter <i>elt</i> will contain a character.
The second parameter of the function is
the number of times that we wish to repeat the character.
And you can see that
this parameter has a default value of 5.
Which means that when I call
the <i>affiche_ligne</i> function in this way,
that is, with a single argument,
where the first argument is the '*' character (star),
The parameter <i>elt</i> will contain the star character.
Since I don't have a second argument
in this call to my <i>affiche_ligne</i> function,
the second parameter <i>nb</i> will take
the default value of 5.
And so, this call to the <i>affiche_ligne</i> function
will display a line with 5 stars.
Now,
the second call
to my <i>affiche_ligne</i> function
has two arguments,
where the first argument is the '+' (plus) character
and the second argument has a value of 8.
In this case,
the parameter <i>elt</i> will contain the '+' character
and the second parameter <i>nb</i>
will not take the default value of 5,
but the value which is passed as an argument here,
which is 8.
And so this call to my <i>affiche_ligne</i> function
will display the '+' character eight times.
Two remarks on this concept of default arguments:
arguments with default values
must be specified in the function prototype
and not in the definition of the function.
What's more, upon calling a function with several parameters
with default values,
omitted arguments must be the last ones
and omitted in the order of the parameter list
For example, look at this function
which has a first parameter called "i"
which has no default value,
a second parameter "c" which has the character "a"
as default value;
and a third parameter "x"
with a value of 0 by default.
If I now call my function "f" in this way,
that is, with one single argument,
then the parameter "i" will take on the value of 1,
the parameter "c" will take
the default character, "a",
and the parameter "x" will take its default value
which is 0.
I can also call my function "f"
in this way, with two arguments.
"i" will take
the value of the first argument, that is, 1,
"c" will contain the second argument,
that is, the character "b",
and the third parameter "x"
will still take its default value
of 0.
Now, if I try to call my "f" function
in this way,
the compiler will refuse the call
because the second argument here is of type <i>double</i>.
The compiler will try to assign this <i>double</i>-type value
to the second parameter "c" which is of type <i>char</i>,
so this call will be refused by the compiler.
Now, we might be tempted
to call our "f" function like this,
that is, by omitting the second parameter.
Unfortunately, the compiler will refuse such a call.
And finally,
if I try to call my "f" function like this,
with these three arguments,
what will happen?
The parameter "i" will take a value of 1,
the parameter "c" will contain the character "b",
and the parameter "x" will take the value of the third argument,
a value of 3.
We will finish by talking about overloading of functions.
In C++, it is possible to define
several functions with the same name
if these functions have different parameters.
That is, if the number or types of these parameters are different.
This will allow the compiler to distinguish between these functions.
This is called function overloading
and it will come in very handy when we use functions
which have similar tasks,
but operate on different kinds of data.
So, in C++, what determines the difference between functions
is not only the names of these functions,
but also the types of their parameters,
which is technically called the "signature" of the function.
Note that the return type
is not part of these distinctions.
You cannot have two functions
with the same name and same parameters
and simply with different return types.
So I can have two functions "f",
where the first function has a parameter of type <i>int</i>,
and where the second function has a parameter of type <i>double</i>,
and where the return type for these two functions
is the same, for example <i>int</i>.
However, I cannot have two functions "f"
which both have a parameter of type <i>int</i> for example,
and which vary only in their return type.
For example <i>int</i> for the first
and <i>double</i> for the second.
To sum up, in C++, one can have several functions with the same name,
as long as they have different types of parameters.
For example, I can have three functions
which are all named <i>affiche</i> (<i>display</i>).
The first function has a parameter of type <i>int</i>,
the second, a parameter of type <i>double</i>,
and the third, two parameters of type <i>int</i>.
These are indeed 3 different functions.
Now, if I call <i>affiche(1)</i>,
since the argument 1 is of type <i>int</i>,
it is the first function,
which has one parameter of type <i>int</i>, that will be called.
And in this case, I will get the message
"entier : 1". (TN : means "integer: 1")
If I now call <i>affiche(1.0)</i>,
since the argument (1.0) is of type <i>double</i>,
it is the second function,
which has one parameter of type <i>double</i>, that will be called,
and it will display the message
"reel : 1" (TN : means "real : 1").
If I call <i>d'affiche(1,1)</i>,
since both arguments have the same <i>int</i> type,
it is the third function
that will be called
and I will get, on the display, "couple : 11".
Careful: I cannot define these 2 functions
in the same program.
The first function has a parameter of type <i>int</i>;
the second has two parameters,
both of type <i>int</i>,
but the second parameter has a default value.
If I now call <i>affiche(2)</i>,
it is impossible to know whether I want to call
the first function "affiche"
by passing 2 as an argument to the parameter "x",
or the second function,
by passing 2 to the parameter "x1"
and using the default value of 1 for parameter "x2".
There is an ambiguity
between these two functions in this call.
This is why
defining these two functions together
in the same program is forbidden.
Explora nuestro catálogo
Inscríbete de manera gratuita y obtén recomendaciones personalizadas, actualizaciones y ofertas.
Coursera brinda acceso universal a la mejor educación del mundo, al asociarse con las mejores universidades y organizaciones, para ofrecer cursos en línea.
© 2018 Coursera Inc. Todos los derechos reservados.
