strings,pointers, and using namespace std;...................?

what does this mean?

Code:

---

using namespace std;

---

and I don't have a referance right now, but which one is declaring a string?

Code:

---


char* cString = "hello!";

//or

char *cString = "hello!";

---

I Know what pointers are, I just forgot which was which....

and when you use an array of chars, like this

Code:

---

chr cString[6] = "hello!";

//it can only hold a max of 7

chr cString[] = "hello!";

//I didnt specify what size the array was
//but it can only hold 7 now because thats what I
//assigned to it

//if i use the method above(char * .... etc)
//will I be able to change the length(make it longer)
//after the first time I change it?

//and finally

//how do I redim an array?

---

Thanks.......
:D

In C++, a namespace is simply a logical area where data and functions are stored:

Code:

---

namespace Mike {
    void function() {
    }
    int another(int a) {
        return a*2;
    }
};

---

To use these, you reference them with the Scope Operator (::):

Code:

---

Mike::function();
cout << Mike::another(50) << endl;

---

Namespaces are usually used to ensure that no function names are used which are already taken.

--------------------------------------------------------

Both of these declare a string. (hehe). Technically, though, you should use:

Code:

---

char *pcString = "Hello!";

---

...because the * operator is being applied to the variable.

--------------------------------------------------------

To redimension an array, you have to make a new buffer that is longer, copy the original contents into it, and set the pointer. BEWARE!!! You cannot do this with a statically-initialised variable:

Code:

---

char *pcString = "Hello!"; // cannot be redimensioned

---

To redimension, you must do something like this, which concatenates pcHello and pcWorld:

Code:

---

char *pcHello = "Hello ";
char *pcWorld = "World!";
char *pcString;
pcString = new char[strlen(pcHello) + strlen(pcWorld) + 1]; // The +1 is important
strncpy(pcString, pcHello, strlen(pcHello));
strncpy(pcString + strlen(pcHello), pcWorld, strlen(pcWorld));
pcString[strlen(strlen(pcHello) + strlen(pcWorld) + 1] = 0; // Terminator
cout << pcString << endl;
delete pcString;

---

pcString now contains "Hello World!". That last strncpy demonstrates one of the useful features of pointers - you can add to them. In this case, all it does is to reference the second half of pcString. The final line frees the memory allocated by new.

Ok, but I don't think you answered one of my questions...
when you do

Code:

---

char *mystring = "whatever";

---

can it be changed to

Code:

---

mystring = "longer text than before"

---

?

I did answer that one. You can't do it that way. Once you initialise it like that, you can't change it around. Anyway, you can use the STL string class:

Code:

---

#include <iostream>
#include <string>

using namespace std;

void main() {
    string sMyString = "This is a string";

    sMyString = "A different string"; // Allowed
    sMyString = sMyString + " and another one";

    cout << sMyString << endl;
}

---

This will print:

Code:

---

A different string and another one

---

Oh yeah, I forgot in my last post...namespaces. To avoid having to reference them explicitly, there is the using namespace Mike; code. Otherwise, the code about would be:

Code:

---

#include <iostream>
#include <string>

void main() {
    stl::string sMyString = "This is a string";

    sMyString = "A different string"; // Allowed
    sMyString = sMyString + " and another one";

    stl::cout << sMyString << stl::endl;
}

---

I think you can see the disadvantage here ;).

As far as I know, namespaces are useful when you have more than one person writing code for a project. If each person uses their own namespace then there can be no variable naming conflicts. It's a way of making your own scope - that's why you use the scope resolution operator to use identifiers in other namespaces.

Parksie, what's the function template for the strncpy function? I'm pretty sure I followed the code there, but I'd like to make sure.

Also,

Code:

---

pcString[strlen(strlen(pcHello) + strlen(pcWorld) + 1] = 0; // Terminator

---

Aren't you meant to assign the character '\0' as a string terminator for platform independance or something? (Been a while since I looked in my big book o' C)

I think it's

Code:

---

char* strncpy(char *dest, const char *source, size_t length);

---

It copies length characters from source to dest, without adding a terminator.

char is an integral type, so these are equivalent:

Code:

---

char c;

c = '\0';
c = 0;

---

It's not platform independence, just convention that strings end with NULL. Although in DOS, they used to end with a '$' symbol (don't ask why - this is still needed in asm).

Oh right, so the escape character '\0' is just for convenience when writing string literals in your code then? I thought that the string terminator for different platforms could be different.

'Integral type' - not a term I've come across before. I knew that char was just an integer data type really and you could assign ascii codes to it, but I didn't realise there was some terminology involved. I'll go look it up...

Cheers :)

"Integral" just means that it stores integers...;).
As for the string terminators, it's convention in C, because since a null character ('\0') cannot be displayed, it is used as the terminator. It also allows things like this:

Code:

---

void myprint(char *pcString) {
    char *p = pcString;
    while(*(p++))
        putc(*p);
}

---

In Basic, I think strings are stored as a count followed by the string. (BBC BASIC terminated strings with the RETURN character (Char 10?).

Back to C...all implementations of C use the null-terminator format.