I'm not understanding Parameterized Constructors in C++

[shivajikobardan]

Homework Statement

parameterized constructor in c++

Relevant Equations

none

I am not understanding this code meaning

#include<iostream>
#include<cstring>
using namespace std;

class String
{
private:
    char *s;
    int size;
public:
    String(const char *str)
    {

        size=strlen(str);
        s=new char[size+1];
        strcpy(s,str);
    }
    ~String()
    {
        delete[] s;
    }
    String(const String& old_str)
    {
        size=old_str.size;
        s=new char[size+1];
        strcpy(s,old_str.s);

    }
    void print()
    {
        cout<<s<<endl;
    }
    void change(const char *str)
    {
        delete [] s;
        size=strlen(str);
        s=new char[size+1];
        strcpy(s,str);
    }
};
int main()
{
    String str1("Hello World");
    String str2=str1;

    str1.print();
    str2.print();

    str2.change("Namaste");
    str1.print();
    str2.print();
    return 0;
}

Q1)

Let's look at the main function.

String str1("Hello world");

This invokes this constructor.

    String(const char *str)
    {

        size=strlen(str);
        s=new char[size+1];
        strcpy(s,str);
    }

I imagine a scenario like this:

How can we says strlen(str) finds the length of "Hello World"?
Because str=2000H. We need to have said strlen(*str) instead.

What's the effect of s= new char[size+1]? Is this the effect?

Again I don't understand how can we just do strcpy(s,str), should not we do strcpy(*s,*str) instead?

Now, we go to str1.print() in main() function.
Here we do

cout<<s

Again as I said above, s is just a pointer and stores the address of first memory location of string. For eg: it is 2000 in third picture.Q2) I have other questions in str2 as well which uses a copy constructor, but I'll ask them in a new post.

A request I'd like to make to mentors would be if possible share some pictures, it doesn't have to look wonderful, any random sketch would help a lot to me. A picture speaks 1000 words and an example speaks 1000 paragraphs. So, figures and examples will be much appreciated.

I'd be glad to see article links to read if there's already someone who has answered these questions.

 

[Mark44]

I imagine a scenario like this:

Your first picture is a reasonable representation of how things relate.

Your second and third pictures show some misunderstanding.
In the 2nd picture, you have *str = "Hello world"; and "Content of 2000H = "Hello world"
Both of these are incorrect.
str contains the address of the first character in that string; namely the letter 'H'.
That address is a byte address that can contain only a single character.

How can we says strlen(str) finds the length of "Hello World"?
Because str=2000H. We need to have said strlen(*str) instead.

strlen() is used on C-style character strings. strlen(*str) would generate a compile-time error, as strlen() requires the address of a character, not a character. The way strlen() works is that it starts at the address represented by its parameter, and increments an internal count variable as it iterates through the string. It stops when it finds the null character that marks the end of the string.

What's the effect of s= new char[size+1]? Is this the effect? Now, we go to str1.print() in main() function.
Here we do

cout<<s

Again as I said above, s is just a pointer and stores the address of first memory location of string. For eg: it is 2000 in third picture

In your first question above, the picture is reasonably accurate. The new operator allocates an array of length size + 1. The print() method is nothing more than a wrapper around the C++ way of displaying a string. What you probably don't understand is that the stream insertion operator, <<, has many different overloads, one for each type that it works with. If a C-style string is the argument, the insertion operator starts at the first character at the given address, and continues inserting characters into the stream until it hits a null character.

Again I don't understand how can we just do strcpy(s,str), should not we do strcpy(*s,*str) instead?

Because the C standard library function strcpy() expects its parameters to be of type "pointer to char"; i.e., char *. In your example, *s and *str would be of type char, not char *, so this wouldn't compile.

 

[shivajikobardan]

Your first picture is a reasonable representation of how things relate.

Your second and third pictures show some misunderstanding.
In the 2nd picture, you have *str = "Hello world"; and "Content of 2000H = "Hello world"
Both of these are incorrect.
str contains the address of the first character in that string; namely the letter 'H'.
That address is a byte address that can contain only a single character.strlen() is used on C-style character strings. strlen(*str) would generate a compile-time error, as strlen() requires the address of a character, not a character. The way strlen() works is that it starts at the address represented by its parameter, and increments an internal count variable as it iterates through the string. It stops when it finds the null character that marks the end of the string.In your first question above, the picture is reasonably accurate. The new operator allocates an array of length size + 1. The print() method is nothing more than a wrapper around the C++ way of displaying a string. What you probably don't understand is that the stream insertion operator, <<, has many different overloads, one for each type that it works with. If a C-style string is the argument, the insertion operator starts at the first character at the given address, and continues inserting characters into the stream until it hits a null character.

Because the C standard library function strcpy() expects its parameters to be of type "pointer to char"; i.e., char *. In your example, *s and *str would be of type char, not char *, so this wouldn't compile.

thanks for the detailed explanation.