About vectors.

Suppose I want the user himself to enter the data into a vector. How should I go about doing it? cin wouldn't work, neither does cin.get.

Example:

Code:

---

#include <iostream>
#include <vector>

using namespace std;

void main (void)
{
    vector<char>alphabet;
    //user inputs data...but how? fill in please =)
    for(int i = 0; i < alphabet.size(); i++)
    {
        cout << alphabet[i];
    }
    cout << endl;
    system("pause");
}

---

Also, there was a warning on using alphabet.size() over there while the program is as incomplete as it is.
Warning C4018: '<' : signed/unsigned mismatch

You should probably use push_back to add things to a vector.

And in this case your signed/unsigned warning is harmless, but the easy way to fix it is to declare i as unsigned.

What if I want the user to input data in like a character array, where the user can just type in a string of characters immediately and it goes into the character array? I could use a string vector but I need to be able to edit each character so I probably can't use a string vector.

If a vector<char> is good enough, why don't you just use a string to store the input?

Also, what makes you think that you cannot edit each character in a string vector?

Well umm...I'm trying to learn to use vectors so I have to use a vector xD and how do I change the individual characters in a string? Maybe you could somehow add to my code here.

Code:

---

#include <iostream>
#include <algorithm>
#include <vector>

using namespace std;

void main (void)
{
vector<char>alphabet;
while(1)
{
        if(alphabet.size() > 10)
        {
                cin.clear();
                cin.ignore(100,'\n');
                cout << "Too many letters! Max 10 letters are allowed : ";
        }
        else
        {
                break;
        }
}
for(unsigned int a = 0; a < alphabet.size(); a++)
{
        cout << alphabet[a];
}
cout << ", ";
for (unsigned int b = 0; b < alphabet.size(); b++)
{
        if(static_cast<int>(alphabet[b]>= 65 && static_cast<int>(alphabet[b] <= 90)))
        {
                alphabet[b] = static_cast<char>(static_cast<int>(alphabet[b]) + 32);
        }
        else if(static_cast<int>(alphabet[b]>= 91 && static_cast<int>(alphabet[b] <= 122)))
        {
                alphabet[b] = static_cast<char>(static_cast<int>(alphabet[b]) - 32);
        }
        cout << alphabet[b];
}
cout << endl;
system("pause");
}

---

Quote:

---

void main (void)

---

int main (void) - read FAQ

Code:

---

if(alphabet.size() > 10)

---

and where is the code that make this test possibly true?

Code:

---

cout << alphabet[a];

---

Do you plan to initialize alphabet with something before outputting?

If you manage to store anything in your vector, then all the static_casts are unnecessary too, as they would happen implicitly anyway.

You access individual characters in a string just as you do with the vector, using the [] operator.

Your use of magic numbers (65, 90, 91, 122) is a bad thing. Use the appropriate character constants instead. (This would also make it more explicit that you are trying to uppercase things like '['.)

You're bringing in the big guns with <algorithm> but I don't see why.

As mentioned before, you can add a char at a time to a vector<char> so you'll have to read things in a char at a time. But, checking whether alphabet.size() > 10 doesn't actually remove the extra bits from alphabet. You should check during input that only 10 letters are input.

I guess the code was pretty messy huh. This is the reworked code and it works fine.

Code:

---

#include <iostream>
#include <string>
#include <vector>

using namespace std;

void main (void)
{
        char word[100];
        vector<char>alphabet;
        while(1)
        {
                cin >> word;
                for(unsigned int i = 0; word[i] != '\0'; i++)
                {
                        alphabet.push_back(word[i]);
                }
                if(alphabet.size() > 10)
                {
                        cin.clear();
                        cin.ignore(100,'\n');
                        alphabet.clear();
                        cout << "Too many letters! Max 10 letters are allowed : ";
                }
                else
                {
                        break;
                }
        }
        for(unsigned int a = 0; a < alphabet.size(); a++)
        {
                        cout << alphabet[a];
        }
        cout << ", ";
        for (unsigned int b = 0; b < alphabet.size(); b++)
        {
                if(static_cast<int>(alphabet[b]>= 65 && static_cast<int>(alphabet[b] <= 90)))
                {
                        alphabet[b] = static_cast<char>(static_cast<int>(alphabet[b]) + 32);
                }
                else if(static_cast<int>(alphabet[b]>= 91 && static_cast<int>(alphabet[b] <= 122)))
                {
                        alphabet[b] = static_cast<char>(static_cast<int>(alphabet[b]) - 32);
                }
                cout << alphabet[b];
        }
        cout << endl;
        system("pause");
}

---

However, could you further explain how else I can check for upper and lower case letters?

As for not using int main() well, my course is currently teaching me to use void main() so I suppose its better to do so for now.

I also do not understand why exactly would I not need to use static_cast<int>. Does it automatically check for the int value of the character if I'm checking with an int value?

There are such things as isupper() and islower(). If you feel you must use ASCII code comparisons, you should use a literal 'a' and 'z' and a literal 'A' and 'Z' in place of 65, etc. (Just for fun, type in "bob]george" and still see if you claim it's working.)

Your belief that characters are not integers is charming, but incorrect. If you remove all the static_casts in your program (both to int and to char) it will work in exactly the same way. (I suppose there's nothing wrong with being explicit as to when you mean to interpret things as characters and when you mean to interpret things as integers, but it's irrelevant to C++.)

Why to work so hard for filling vector of chars? Just 2 days ago was a sample for this simple task

Code:

---

string s;
vector<char> v;
if(cin >> s)
    v.insert(v.end(), s.begin(), s.end());

---

http://cboard.cprogramming.com/showthread.php?t=106629

and still you have main as void

Quote:

---

Originally Posted by vart

and still you have main as void

---

Takes time to change and I'm trying to. xD

Ok thanks for the info on isupper() and islower() it is REALLY very helpful.

By the way, though a little off-topic, I must ask how should I continue to learn C++? I have learned that just waiting to be taught in lesson is not enough, but the thing is I don't know how to go about doing it. Just like people usually start with cin and not GetAsyncKeyState, is there possibly a way to learn C++ in order? I'm currently at Classes and have jumped to <vector>. I'm also trying to understand <map>, though this pace feels kinda wrong. =(

I'd recommend that you buy the book "Accelerated C++". It is one of the better of the MANY books written to teach C++. Follow the order of that book.

Of course, you don't need to know everything about everything in C++ to be able to write useful, meaningful and efficient programs - because some types of programs only use some parts of the C++ language and related libraries.

On the other hand of "you only need to know some parts" is "if you don't know it exists, you can't use it when you need it". So having a brief understanding that there is something called "map", and it works sort of like an array where the index can be "anything" is useful to know even if you have no need at all for maps in the code you are writing/planning to write at present.

--
Mats

How much does that book cover actually? and are there some good websites with at least a good list that tells me what I should learn in order? xD

I haven't actually read that book - it is what most people recommend however [there is a book topic in the C++ forum].

Once you have the basics, such as simple variables, input/output, flow-control, basic functions covered, and basic user defined data types (struct, enum, union), it is pretty much arbitrary how you continue from there, because you have the fundamentals. Of course there are still some things that depend on understanding lower level stuff, e.g. templates to make classes requires that you know classes.

I'm sure there are a whole heap of different web-sites with varying levels of "this is the order to learn things".

Another approach to learning is to set yourself a project [that you have reasonable confidence that you can solve], and start programming that project. A calculator is always a good example, because you can start simple, expand and extend it until you have a graphical user interface and a small programming language that allows you to program user specified functions.

--
Mats

matsp you NEVER fail me! Thank you! That was really helpful!

Although after a "beginners" book, there are lots of advanced books out there.
Then there are design books and such. You can never learn enough.

O0O
But you see actually the reason I'm trying to learn so much is because theres a small group(really small group) of people in my class who know ALOT more than what we are currently learning and I want to be able to do stuff like they do(adding pictures, etc, etc) but I'm not sure what exactly to learn. Currently all I know are...

-Cin(.ignore,.get,etc),

-Cout,

-Loops,

-If, Else if,

-Arrays(2D and onward),

mostly the really basic stuff

-A little on File I/O(indata, outdata, etc),

-Structs,

-Classes(though I'm not really good at it) no inheritance though,

Pointers(but I NEVER use them except during practicals ON vectors because I don't see why I would need them)

AND THEN I am currently trying to learn...

-Vectors(because they are cool)

-Iterators(because they came along with vectors and just like pointers, I don't really see the need for them, am I like pointer-blind? O_O)

-Algorithm

-Maps(but I only know how to declare them)

-How to make simple games!(I'm trying to do English Chess xD, good idea?)

Yeah, I think that's just about it.

Quote:

---

How to make simple games!(I'm trying to do English Chess xD, good idea?)

---

Chess AI is a science of its own, but a simple interface for two human player games (that can detect correctness of moves, draw and win conditions, do game timing, perhaps replays after the game is over, taking moves back etc) should be quite doable at some point.

Somewhere quite early you should be able to write a Guess a number game (could be both ways, human guessing a computer-picked number and the other way round).

Then perhaps a Hangman game.

Tic-Tac-Toe is popular too, but complete beginners tend to implement it in awful ways (there are examples of the game written without any arrays or loops at all :)), so that might be an intermediate level challenge.

Of simple graphical games, Pong, Tetris, Breakout and similar are popular choices.

Quote:

---

Pointers(but I NEVER use them except during practicals ON vectors because I don't see why I would need them)

---

I don't quite understand this sentence, but storing pointers in a vector, or storing pointers to vector items are not very good ideas.

Storing pointers in a vector is usually a good solution, though, when you don't want data copied needlessly.

Quote:

---

Originally Posted by Elysia

Storing pointers in a vector is usually a good solution, though, when you don't want data copied needlessly.

---

Then use a shared pointer implementation like the one boost has. Raw pointers in a vector is never a good idea because you have to manage memory explicitly.

Or use a special-purpose container, such as boost::ptr_vector.

When you store pointer to dynamically allocated objects in a regular vector, not only is it your responsibility to free the memory yourself, but also certain functions become unusable (such as std::remove).

Basically you'd be potentially required to write much more explicit memory handling code than just releasing the memory at the end.

In addition, it won't be exception safe.

Naturally. I never leave home without smart pointers these days!
Of course, a smart pointer is technically still a pointer, so the argument still holds ;)

Quote:

---

Originally Posted by anon

Chess AI is a science of its own, but a simple interface for two human player games (that can detect correctness of moves, draw and win conditions, do game timing, perhaps replays after the game is over, taking moves back etc) should be quite doable at some point.

---

Yep, my teacher showed us this video where this supercomputer beat this professional at chess.

Quote:

---

Somewhere quite early you should be able to write a Guess a number game (could be both ways, human guessing a computer-picked number and the other way round).

---

Yep, had to do it in one of my practicals, though still quite unclear about the effects of srand(time(NULL)).

Quote:

---

Then perhaps a Hangman game.

Tic-Tac-Toe is popular too, but complete beginners tend to implement it in awful ways (there are examples of the game written without any arrays or loops at all :)), so that might be an intermediate level challenge.

---

Cool! I have a book which uses these two games as an example.

Quote:

---

Of simple graphical games, Pong, Tetris, Breakout and similar are popular choices.

I don't quite understand this sentence, but storing pointers in a vector, or storing pointers to vector items are not very good ideas.

---

Thanks for the great ideas, same for everyone else. xD But I guess it'll be some time before I start using pointers instinctively like while loops.