What is the difference between these two statements?

[najum98]

I was working on a homework assignment when I came across two methods to get a certain job done.
I am not going to write whole class code here. Actually I am doing aggregation practice.

so can anybody explain me the difference between these two approaches?

Approach # 1:
Function in class =

addCourse(Course* c)
{
    cptr[1]=c
}

In Main call=

addCourse(&c1);

Approach # 2:
Function in class=

addCourse(Course c)
{
    cptr[1]=&c
}

In Main call=

addCourse(c1);

----------------------------------
Considering that getID is a getter function and id is an int value
cptr[1].getID==id
above statement works fine when approach # 1 is used and it gets skipped when using the other approach..
Whats the matter?

[2kaud]

In approach #1, you are passing as a param for addCouse() a pointer to type Course as c. In the call statement you are obtaining the address of c1 and pasing this as the param. Thus in addCouse(), c is the address of c1 in Main(). Everything is ok and as you're found, works as expected.

However in approach #2, things are slightly different. Here the param to addCouse is of type Course - not a pointer (or a reference). Hence the data is 'passed by value' which means that in addCouse(), c is a copy of the variable c1 in Main() - not the same variable. As it's a different variable, it has a different address. Hence in addCouse(), &c is the address of the c variable used in addCouse() - which is a different address to that of c1 in main(). The variable c only exists until the end of the function addCouse() - so &c is only valid during addCouse() and not outside it. So in main(), cptr[] now has an invalid pointer.

To 'correct' approach #2, c needs to be passed 'by ref' so that address of c in addCourse() is the same as that of the argument c1 in main().

addCourse(Course& c)
{
    cptr[1] = &c;
}

[najum98]

I tried to correct the approach 2 the same way you did but the addCourse got red underline showing the following error
1 IntelliSense: declaration is incompatible with "bool Student::addCourse(Course *c)" (declared at line 14 of "c:\users\najam ul saqib\documents\visual studio 2013\projects\course student\course student\StudentHeader.h") c:\Users\Najam Ul Saqib\Documents\Visual Studio 2013\Projects\Course Student\Course Student\StudentEnc.cpp 28 15 Course Student

In approach #1, you are passing as a param for addCouse() a pointer to type Course as c. In the call statement you are obtaining the address of c1 and pasing this as the param. Thus in addCouse(), c is the address of c1 in Main(). Everything is ok and as you're found, works as expected.

However in approach #2, things are slightly different. Here the param to addCouse is of type Course - not a pointer (or a reference). Hence the data is 'passed by value' which means that in addCouse(), c is a copy of the variable c1 in Main() - not the same variable. As it's a different variable, it has a different address. Hence in addCouse(), &c is the address of the c variable used in addCouse() - which is a different address to that of c1 in main(). The variable c only exists until the end of the function addCouse() - so &c is only valid during addCouse() and not outside it. So in main(), cptr[] now has an invalid pointer.

To 'correct' approach #2, c needs to be passed 'by ref' so that address of c in addCourse() is the same as that of the argument c1 in main().

addCourse(Course& c)
{
    cptr[1] = &c;
}

[2kaud]

Yes, you will need to change the declaration of addCourse() in the class Student to match the revised function definition. The function declaration and definition needs to have the same interface specification.

[najum98]

Yes, you will need to change the declaration of addCourse() in the class Student to match the revised function definition. The function declaration and definition needs to have the same interface specification.

Thanks a lot mate!!