basic string scope in std namespace

I have a simple script but I am missing how vars are used. I get the compile error "error: ‘the_string’ was not declared in this scope" on the line that has " cout << the_string;"

Whats the correct way to access "the_string".

Code:

---

#include <string>
#include <iostream>

using namespace std;

int main(int argc, char* argv[])
{
    bool test1 = false;
    if(test1)
    {
        string the_string("its true");
    }else{
        string the_string("its false");
    }

    cout << the_string;
    return 0;
}

---

And again your program means its false. :eek: I am sad :sad:

Quote:

---

Originally Posted by Khiem

And again your program means its false. :eek: I am sad :sad:

---

umm ok?


If anyone has a real answer. I assume I am doing this the wrong way. Basically I am playing with cURL. I don't have a lot of experience with C++ as its my first project. I need to store the return data in a var. I don't know the best way to store the buffer since its going to be a variable length. My programing experience is all high level stuff like PHP. I have seen examples using the string lib but my use of it must not be right (or maybe g++ issue). Bottom line i'm confused why a var defined inside an if statement is outside the scope after it. I assume its the use of the string lib or the namespace.

Quote:

---

Originally Posted by ryanmills

Bottom line i'm confused why a var defined inside an if statement is outside the scope after it.

---

It shouldn't be. The compiler is telling you this. The variable the_string must exist at the highest level at which it will be used.

However, it may be *modified* at any level you like under that. I'd change it to:

Code:

---

#include <string>
#include <iostream>

using namespace std;

int main(int argc, char* argv[])
{
    bool test1 = false;
    string the_string;
    if(test1)
    {
        the_string = "its true";
    }else{
        the_string = "its false";
    }

    cout << the_string;
    return 0;
}

---

I guess i am not thinking about scope the right way but your example works. But I do have another question. Is there a risk of a buffer over run if you reset the "the_string" and whats the char limit for string?

example code:

Code:

---

#include <string>
#include <iostream>

using namespace std;

int main(int argc, char* argv[])
{
    string the_string;

    bool test1 = false;

    if(test1)
    {
        the_string = "its true";
    }else{
        the_string = "its false";
    }

    the_string = "this is a longer string to test for a buffer overrun";

    cout << the_string;
    return 0;
}

---

There is no limit* to the length of a string a string object can store. The string object will re-allocate the memory it stores internally if it needs to, to accomodate the longer string you assign to it.

*OK, there is a limit, but it's on the order of 2 billion, so for all practical purposes, there is no limit

No, the std::string class was created specifically to avoid buffer overruns being a possibility during normal operation (it's possible to screw them up, but they make it pretty difficult).

Its size is limited only by available memory and/or address space.

And now what maybe a complex question. Is this the best way to store data as a "buffer" in the context of content returned from a cURL call?

I'm not familiar with cURL. However, there are two questions you need to ask:

1) Will this data ever need to be passed to a function expecting a (non-const) char*?
2) Is it possible to have a 0 byte prior to the end of the data? (I don't mean the character '0', I mean the NULL byte.)

If the answer to either of these is yes, you'll be better off with a std::vector<char>. Otherwise a std::string should be just fine.

I think I need to get my head wrapped around strings better first. I thought I had a grasp on what * was used for. But I am seeing it used in contexts I don't understand. I can see now why people pick python over C, but I really want to get a handle on this.

Example that I get:

Code:

---


a = '6';
b = &a;  // points to the mem address
c = *b;
cout << c; // should return 6

---

But then I see this for example, does this mean its passing a reference?

Code:

---

int main(int argc, char* argv[])

---

Also when you say "expecting a (non-const) char*" do you mean the function is expecting a pointer and not the real content. Such as my example, its expecting the var c not a?

there is a difference - it depends on whether you are declaring a var, or using a var.

e.g.:

Code:

---

int* a = new int; // declares a pointer to an int, and sets it pointing to some uninitialized int memory
int& b = a;  // declares b as a reference, and sets reference to a;


int c;
c = *a; // here * is dereferencing a.  c = <value pointed at by a>

int* d;
d = &c; //  the pointer d is given the address of c to point to.

---

Hopefully you can now tell what the below means:

Code:

---

int main(int argc, char* argv[])

---

main expects an int, and a pointer to a char ( the [] means it actually expects there to be more than one char* in a contiguous block of memory )

char* argv[] is simply an array of C-style strings corresponding to command-line arguments. You'll sometimes see the parameter declared as char ** argv, which is equivalent.

At the lowest level---where C lives---any string is just an array of characters. Hence the type is char*. However, classes such as std::string were creating specifically so that you wouldn't have to worry about such details. Consider the progression:

Array of char arrays: char**
Array of std::strings: std::string*
std::vector of std::strings: std::vector<std::string>

At each step we wrap a C++ class around a pointer, allowing us not to worry about the specifics of dealing with pointers. That's not to say that the three types are 100&#37; compatible in all cases, but they're essentially similar.