Posts by Clockowl

Shoot, now I have to look at the moon when I'm coding, thanks Narue.

You say it's a bad idea to overload pointers, which I get, but specifically pointers to Foo objects for the > operator, would that really be a bad idea you think?

Hey guys,

I know the title isn't possible, but I'd like to here how you'd do it then.

Suppose we have this code:

Class Foo {public: int x; };

//somewhere in the code..

int main()
    Foo one, two;
    one.x = 10;
    two.x = 20;

    Foo *a = &one;
    Foo *b = &two;

    //We want to compare Foo::x here, not the pointers.
    bool result = a > b;
    return 0;
}

How would you implement this? I've thought about a compare class, as the STL containers do, and ofcourse a compare functions, but I'd like to hear if there are other options, and if you pick compare classes over compare functions and why etc.

Here is the same problem in a small program, I still haven't figured it out why it thinks that set<ElementClass, CompareClass>::iterator is a value instead of a type.

#include <iostream>
#include <vector>
#include <set>
using namespace std;

class IntCompare {
	bool operator() (const int &a, const int &b){
		return (a == b);
	}
		
};

template <class ElementClass, class CompareClass>
class A {
public:
	virtual void foo(const int &x){
		pair< set<ElementClass, CompareClass>::iterator, bool> result;
	}
	
	set<ElementClass, CompareClass> elements;
};

/*class B : public A<int, IntCompare> {
public:
	virtual void foo(const int &x){
		cout << x*x << endl;
	}
};*/

int main (int argc, char * const argv[]) {
	A<int, IntCompare> mySet;
	for(int i = 0; i < 10; i++){
		mySet.foo(10);
	}
	
			
    return 0;
}

This code gives these compile errors:

main.cpp:17: error: type/value mismatch at argument 1 in template parameter list for 'template<class _T1, class _T2> struct std::pair'

main.cpp:17: error:   expected a type, got 'std::__debug::set<ElementClass,CompareClass,std::allocator<_CharT> >::iterator'

main.cpp:17: error: invalid type in declaration before ';' token

Hey guys,

I have this code snippet:

template <typename ElementType, typename CompareClass>
struct Delegate {
	virtual bool geef(const ElementType &element){
		pair <set<ElementType, CompareClass>::iterator, bool> ret;  //this is line 196
		ret = elements.insert(element); //197
		return (ret.second); //198
	}
	
	virtual void output() = 0;
	
	set<ElementType, CompareClass> elements;
};

struct ArtiestenPrinter : public Delegate<Lied, LiedCompare> {
	virtual bool geef(const Lied &lied){
		bool nieuwElement = Delegate<Lied, LiedCompare>::geef(lied); // 210
		if (nieuwElement) {
			cout << lied.artiest;
		}
		return nieuwElement;
	}
	
	virtual void output(){
		for(set<Lied, LiedCompare>::iterator it = elements.begin();
			it != elements.end();
			it++){
			
			cout << it->artiest << endl;
		}
	}
};

Which gives me these errors:

196: error: type/value mismatch at argument 1 in template parameter list for 'template<class _T1, class _T2> struct std::pair'
196: error:   expected a type, got 'std::set<ElementType,CompareClass,std::allocator<_CharT> >::iterator'
196: error: invalid type in declaration before ';' token
198: error: request for member 'second' in 'ret', which is of non-class type 'int'
210:   instantiated from here
197: error: cannot convert 'std::pair<std::_Rb_tree_const_iterator<Lied>, bool>' to 'int' in assignment

I can't get my head around why the compiler is complaining, so any help is greatly appreciated.

Thanks in advance,

Nevermind.

OH err, the above was a question. xD

"Can you read files larger than 2^32 bytes in a 32 bit environment with streams?"

Say that's a regular int (assuming 4 bytes), you can't open files more than ~5GB in a 32-bit environment (since the stream can't be bigger than that).

Thanks for that though! I used MAX_INT before, but this is definitely better.

What wildgoose gives as a solution is called a "forward declaration", so you can read up about it (where what why how etc.)

And yes, Pancake stacks are most of the time FILO ones. xD ;)

Hey guys,

Can I tell istream to ignore ALL the characters until the delimiter char? It can be done with a loop but it's kinda weird I can't tell istream "discard until this char" imho, so maybe there's an hidden option somewhere.

Thanks in advance!

Hey, sorry to go offtopic anilopo, but Narue reminded me,

If I have a class like this:

class X{
public:
  X(int y) : y(y) {}
 
private:
  int y;
}

Does the ctor invoke undefined behavior or is it valid C++? And does it do what I expect it to do: Set X::y to ctor param y? It works in MinGW GCC 3.4 afaik.

Again sorry for offtopic.

You can't call calc() in the product class. calc() isn't implemented in the Product class! ;)

What you're doing in the ctor is basically:

Product :: Product (...)
{
   this->price = this->calc ();
}

And since "this" in that context is a Product*, you can't call calc().

Hehe, I messed up the linking order.

Say that Program A depends on Lib B, and A and B contain references to Lib C, then A needs to supply B before C, not after.

Hey guys,

So I have compiled my first home-written static library (named PixGUI) that happens to call OpenGL functions. However, I noticed that when compiling (using MinGWs GCC with Code::Blocks project set to "static library") it doesn't matter whether I supply the linker with libopengl32.a or not, it simply compiles each time.

However, when I link PixGUI to another project, i.e. Pixlings, and try to compile that, it gives me these errors:

||=== Pixlings, Release ===|
..\PixGUI\lib\libPixGUI.a(button.o):button.cpp:(.text+0xc1)||undefined reference to `glColor3fv@4'|
..\PixGUI\lib\libPixGUI.a(button.o):button.cpp:(.text+0xea)||undefined reference to `glRectf@16'|
||=== Build finished: 2 errors, 0 warnings ===|

The build log of libPixGUI.a (without supplying the linker with opengl32, but doesn't matter for undefined references)

-------------- Build: Release in PixGUI ---------------

Compiling: element.cpp
Compiling: engine.cpp
Compiling: window.cpp
Compiling: button.cpp
Linking static library: lib\libPixGUI.a
ar.exe: creating lib\libPixGUI.a
Output size is 39.72 KB
Process terminated with status 0 (0 minutes, 2 seconds)
0 errors, 0 warnings

Build log of Pixlings

-------------- Build: Debug in Pixlings ---------------

Compiling: Program.cpp
Compiling: game.cpp
Compiling: main.cpp
Compiling: Menu.cpp
Linking console executable: bin\Debug\Pixlings.exe
..\PixGUI\lib\libPixGUI.a(button.o):button.cpp:(.text+0xc1): undefined reference to `glColor3fv@4'
..\PixGUI\lib\libPixGUI.a(button.o):button.cpp:(.text+0xea): undefined reference to `glRectf@16'
collect2: ld returned 1 exit status
Process terminated with status 1 (0 minutes, 1 seconds)
2 errors, 0 warnings

How would I solve these undefined references?

This might come in handy as well:

http://www.cplusplus.com/reference/std/typeinfo/type_info/

Thanks Narue!

Hey guys,

Below code isn't working as I expect it to do. I expect it to read all params when 6 are given, buuuuuuuuuuuut it only reads one, the rest remains zero. It does enter the case, but the stringstream buf seems empty. What am I doing wrong?

#include <iostream>
#include <fstream>
#include <sstream>
using namespace std;

#define SCREAM cout << "OMAGAWD" << endl

int main(int argc, char *argv[])
{
//    <src> <dest> [<offset> [<bytes, 0 = all> [<offset>]]
    //writes bytes bytes from src+offset to dest+offset

    size_t in_offset;
    size_t bytes;
    size_t out_offset;

    in_offset = bytes = out_offset = 0;

    stringstream buf;

    switch (argc)
    {
    case 6:
    SCREAM;
        buf.str(argv[5]);
        buf >> out_offset;

    case 5:
    SCREAM;
        buf.str(argv[4]);
        buf >> bytes;

    case 4:
    SCREAM;
        buf.str(argv[3]);
        buf >> in_offset;

    default:
        ifstream in(argv[1], ios::binary);
        ofstream out(argv[2], ios::binary);
        break;
    }

    cout << "Writing from " << argv[1] << " + " << in_offset <<  " to " << argv[2] << " + " << out_offset << " for " << bytes << " bytes" << endl;

    return 0;
}

Example params: program.exe aap noot 1 2 3 example output:

OMAGAWD
OMAGAWD
OMAGAWD
Writing from aap + 0 to noot + 3 for 0 bytes

Thanks in advance,

Ugh, fixed it, I forgot the const. With normal functions the error messages are a lot clearer in GCC.. hope they "fix" this soon.

Here's the working code:

//...
    private:
        template <typename T1, typename T2>
        T2 &get_indexed(multimap<T1, T2> &container, const T1 &key, size_t index = 0) {
            size_t count = container.count(key);
//...

Hehe, back again. Templates are difficult if you're not precise I guess. I'm stuck on this one:

sss.hpp

#include <string>
#include <map>
#include <stdexcept>

#ifndef SSSHPP
#define SSSHPP

namespace sss {
    using namespace std;

    class node {

    public:
        multimap<string, string>  values;
        multimap<string, node>    children;

    private:
        template <typename T1, typename T2>
        T2 &get_indexed(multimap<T1, T2> &container, T1 &key, size_t index = 0) {
            size_t count = container.count(key);

            if (count != 0 && count-1 >= index) {
                //it exists and index is within range
                //grab iterator, increment index times and return the node belonging to that.
                typename multimap<T1, T2>::iterator it = container.find(key);
                while (index--) it++;
                return it->second;
            } else {
                //Doesn't exist: throw OOR
                throw out_of_range((string)"get_indexed->"+"Element " + key + "doesn't exist!");
            }
        }
    };
}

#endif // SSSHPP

Call with:

sss.cpp

string &node::getValue(const string &name, size_t index) throw (out_of_range) {
    get_indexed<string, string>(values, name, index);
}

Doesn't compile with:

Code\SSSParser\sss.cpp In member function `std::string& sss::node::getValue(const std::string&, size_t)':
Code\SSSParser\sss.cpp error: no matching function for call to `sss::node::get_indexed(std::multimap<std::string, std::string, std::less<std::string>, std::allocator<std::pair<const std::string, std::string> > >&, const std::basic_string<char, std::char_traits<char>, std::allocator<char> >&, size_t&)

Ah, sorry. They are big functions. :) I finished the code and documented it, it's for review at my website, here's a quick link:

www.dotsimplicity.net/?p=191

Thanks for your input! That typeid() will come in handy one day.

Why are you using two separate functions read and write then ?

Because one reads and one writes? Hahahaha

I am not looking for a runtime solution, that's why I hadn't.

Here's a working example:

template <typename T>
        bool read_or_write(const char *filename) {
            T f(filename);
            if (!f.good()) {
                return false;
            }

            bool success;
            if (typeid(T) == typeid(ofstream)) {
                success = write(f);
            } else if (typeid(T) == typeid(ifstream)) {
                success = read(f);
            }
            f.close();

            return success;
        }

But this way, I also have to define write(ifstream ... ), while I know 100% sure that the compiler will never have to call such a function. It kinda defeats it's use (making the code smaller) for this small function, but for bigger functions this sort of template is very usable I think.

Any compiler-time suggestions?

PS: For anyone reading this and thinking, why the hell are you creating such a function for 20 lines of code? It's because I'm not going to use read_or_write(), but would like to know how I'd accomplish maximum code re-usage in C++.

Oh I'm sure that doesn't work, never seen it, just made it up. But I want to know how I'd do that in C++. :D

Well, a design wouldn't need it. I don't know if its needed but can't you say something like: T can only be ifstream, ofstream

Code could look a bit like this:

template <typename T = {ifstream, ofstream}>
bool node::read_or_write(const char *filename) {
    T f(filename);

    if (!f.good()) {
        return false;
    }

    bool success;
    if(T == typename ifstream){
        success = write(f);
    } else if (T == typename ofstream){
        success = read(f);
    }
    f.close();

    return success;
}

Compiler should throw an error when you try to instantiate any other type of it.. I'm trying to create such a class now, but I'm not very good at it.

This is the best I could come up with, but I didn't compile/test it yet due to diner being ready.

template <typename T>
class filestuffs {
    filestuffs(const char *filename) : filename(filename) {
        f(filename);
        success = f.good();
        success = read_or_write();
        f.close();
    }
    
    //can't do this unless we know T
    bool read_or_write(){ return false; }
            
    T f;
    bool success;
    const char *filename;
    const char [] errmsg = "Can't do this on anything but: ifstream, ofstream";
}
template <>
class filestuffs<ifstream> {
    //now we know T! :D
    bool read_or_write(){
        return read(f);
    }
}

template<>
class filestuffs<ofstream> {
    bool read_or_write(){
        return write(f);
    }
}

Hey guys,

How would I partially specilize a template in a manner as below? Is that even possible?

I have these two functions:

bool node::read(const char *filename) {
    ifstream f(filename);

    if (!f.good()) return false;

    bool success = read(f);
    f.close();

    return success;
}

bool node::write(const char *filename) {
    ofstream f(filename);

    if (!f.good()) {
        return false;
    }

    bool success = write(f);
    f.close();

    return success;
}

And I'd like to create a function like this:

template <typename T>
bool node::read_or_write(const char *filename) {
    T f(filename);

    if (!f.good()) {
        return false;
    }

    bool success;
    if(T == ifstream){
        success = write(f);
    } else if (T == ofstream){
        success = read(f);
    }
    f.close();

    return success;
}

How would I do that?

Thanks in advance,

Okay dat was stoopid.

node& node::child(const string name, size_t index) {
    return searchInsert<string, node>(values, name, index);
}

Should of course be..

node& node::child(const string name, size_t index) {
    return searchInsert<string, node>(children, name, index);
}

No wonder.

Hey guys,

What am I doing wrong here?

sss.hpp

#include <fstream>
#include <string>
#include <map>

namespace sss {
    using namespace std;
    class node {
    public:
        multimap<string, string>  values;
        multimap<string, node>    children;

        string &value(const string name, size_t index = 0);
        node &child(const string name, size_t index = 0);

    private:
    
        template <typename T1, typename T2>
        T2 &searchInsert(multimap<T1, T2> &container, const T1 &name, size_t index = 0) {
            typename multimap<T1, T2>::iterator it = container.find(name);

            //found none, insert...
            if (it == container.end()) {
                it = container.insert(pair<T1, T2>(name, T2()));
                return it->second;
            }

            //found one, multimap is sorted, iterate to index-th element or the end.
            for (size_t i = 0; i < index; i++) {
                it++;
                if (it == container.end()) {
                    it = container.insert(pair<T1, T2>(name, T2()));
                    return it->second;
                }
            }

            return it->second;
        }
    };
}

sss.cpp

#include <iostream>
#include <fstream>
#include <map>
#include <string>

#include "sss.hpp"

using namespace std;
using sss::node;

string& node::value(const string name, size_t index) {
    return searchInsert<string, string>(values, name, index);
}

node& node::child(const string name, size_t index) {
    return searchInsert<string, node>(values, name, index);
}

It won't compile with the following errors:

\Code\SSSParser\sss.cpp error: no matching function for call to `sss::node::searchInsert(std::multimap<std::string, std::string, std::less<std::string>, std::allocator<std::pair<const std::string, std::string> > >&, const std::string&, size_t&)'

I don't really get what it's trying to do.

Thanks in advance,

Thanks guys, worked like a charm and I understand it too. woopiedoo.

Hey guys,

I have a class with a friend function and a member function with the same name. Calling the friend function from the member function gives me an error: it looks for <classname>::<function> while the friend function is of course simply <function>.

In this case, MinGW GCC says:

\Code\RationalNumber\rational.cpp | In member function `long long unsigned int rational::gcd() const':
\Code\RationalNumber\rational.cpp | error: no matching function for call to `rational::gcd(const long long unsigned int&, const long long unsigned int&) const'
\Code\RationalNumber\rational.cpp | note: candidates are: long long unsigned int rational::gcd() const

How do I circumvent this naming collision without altering the function names? Can I tell the rational::gcd() to look for the other gcd without namespaces? Or with some sort of "not in any" namespace?

Here's the relevant code:

rational.hpp

class rational {
    friend unsigned long long gcd(unsigned long long x, unsigned long long y);

public:
    unsigned long long gcd() const;
};

rational.cpp

unsigned long long gcd(unsigned long long x, unsigned long long y) {
    if (x == y) return x;

    //Euclidean method
    unsigned long long diff;
    while (x != 0 && y != 0) {
        if (x > y) {
            diff = x - y;
            x = diff;
        } else {
            diff = y - x;
            y = diff;
        }
    }
    return (x == 0) ? y : x;
}

unsigned long long rational::gcd() const {
    return gcd(nominator, denominator);
}

Thanks in Advance,
Nick

Sorry for those semi-gotos. Here are equivalent function without lazy-programmerness.

string& node::value(const string name, int index) {
    using sss::value;

    valueiterator it = values.find(name);
    if (it == values.end()) {
        it = values.insert(value(name,string()));
        return it->second;
    }


    for (int i = 0; i < index; i++) {
        it++;
        if (it == values.end()) {
            it = values.insert(value(name,string()));
            return it->second;
        }
    }

    return it->second;
}


node& node::child(const string name, int index) {
    using sss::child;

    childiterator it = children.find(name);
    if (it == children.end()) {
        it = children.insert(child(name, node()));
        return it->second;
    }

    for (int i = 0; i < index; i++) {
        it++;
        if (it == children.end()) {
            it = children.insert(child(name, node()));
            return it->second;
        }
    }

    return it->second;
}

PS:
In C I'd do something like this:

#define nodeOrValue(returnType, type, plural, myIterator, insertType)\
returnType node::type(const string name, int index) {\
using sss::type;\
myIterator it = plural.find(name);\
    if (it == plural.end()) {\
        it = plural.insert(type(name, insertType()));\
        return it->second;\
    }\
    for (int i = 0; i < index; i++) {\
        it++;\
        if (it == plural.end()) {\
            it = plural.insert(type(name, insertType()));\
            return it->second;\
        }\
    }\
    return it->second;\
}

nodeOrValue(string&, value, values, valueiterator, string);
nodeOrValue(node&, child, children, childiterator, node);

So, considering this is (SPARTA!) C++, how should I do it? ;)

Hey guys,

With this defintion:

sss.hpp

#include <fstream>
#include <string>
#include <map>

namespace sss {

    using namespace std;

    class node {
    public:
        multimap<string, string>  values;
        multimap<string, node>    children;

        string &value(const string name, int index = 0);
        node &child(const string name, int index = 0);
        //...
    };

    typedef pair<string, string> value;
    typedef pair<string, node>   child;

    typedef multimap<string, string>::iterator valueiterator;
    typedef multimap<string, node>::iterator   childiterator;
}

and this implementation:

sss.cpp

#include <iostream>
#include <fstream>
#include <map>
#include <string>

#include "sss.hpp"

using namespace std;
using sss::node;
using sss::valueiterator;
using sss::childiterator;

string& node::value(const string name,int index) {
    valueiterator it = values.find(name);
    try {
        int i;
        if (it == values.end()) throw 0;
        for (i = 0; i<index; i++) {
            it++;
            if (it == values.end()) throw 0;
        }
    } catch (int e) {
        it = values.insert(sss::value(name,string()));
    }
    return it->second;
}

node& node::child(const string name,int index) {
    childiterator it = children.find(name);
    try {
        int i;
        if (it == children.end()) throw 0;
        for (i = 0; i<index; i++) {
            it++;
            if (it == children.end()) throw 0;
        }
    } catch (int e) {
        it = children.insert(sss::child(name, node()));
    }
    return it->second;
}

How would I merge these two functions? They have so much it common! It seemed redundant... Is there a way with templates and/or with a function that has the shared code?

Thanks in advance,

Second, has ClockOwl noticed that REGARDLESS of the template issue
he has written the algorithm wrong and you get absolutely no items in x after kill_dupes is called you might as well write

Yup, see my second post and the post with the "working code".

main.cpp|12|note: say `typename  std::vector<T,std::allocator<_CharT> >::iterator' if a type is meant|

So that's what it meant. Thanks. :)

Working code:

template <typename T>
void kill_dupes(vector<T> &x){
    vector<T> y;

    for(typename vector<T>::iterator it = x.begin(); it != x.end(); it++){
        if(find(y.begin(), y.end(), *it) == y.end()){
            y.push_back(*it);
        }
    }

    x = y;

    return;
}

Could someone explain why typename is needed though? I still don't quite get it...

Durr, code should be:

void kill_dupes(vector<int> &x){
    vector<int> y;

    for(vector<int>::iterator it = x.begin(); it != x.end(); it++){
        if(find(y.begin(), y.end(), *it) == y.end()){
            y.push_back(*it);
            cout << *it << " ";
        }
    }

    x = y;

    return;
}

Of course. But I want a template like that, as stated above.

That will not do the trick: Unique only removes when consecutive duplicates are found:

Copies the values of the elements in the range [first,last) to the range positions beginning at result, except for the duplicate consecutive elements, which are not copied

Hey guys,

I got this code, and I can't get it to compile, no clue what I'm doing wrong.

#include <iostream>
#include <vector>
#include <iterator>
using namespace std;

template <class T>
void kill_dupes(vector<T> &x){
    vector<T> y;

    for(vector<T>::iterator it = x.begin(); it != x.end(); it++){
        if(find(x.begin(), x.end(), *it) == x.end()){
            y.push_back(*it);
        }
    }

    x = y;

    return;
}

int main(){
    return 0;
}

GCC MinGW says:

main.cpp||In function `void kill_dupes(std::vector<T, std::allocator<_CharT> >&)':|
main.cpp|12|error: expected `;' before "it"|
main.cpp|12|error: `it' was not declared in this scope|
main.cpp||In function `void kill_dupes(std::vector<T, std::allocator<_CharT> >&) [with T = int]':|
main.cpp|31|instantiated from here|
main.cpp|12|error: dependent-name ` std::vector<T,std::allocator<_CharT> >::iterator' is parsed as a non-type, but instantiation yields a type|
main.cpp|12|note: say `typename  std::vector<T,std::allocator<_CharT> >::iterator' if a type is meant|
||=== Build finished: 3 errors, 0 warnings ===|

Of course it wasn't declared, I'm trying to declare it (no pun intended).

kill_dupes is called with a vector<int>.

Thanks in advance,

Edit:

This works as expected:

void kill_dupes(vector<int> &x){
    vector<int> y;

    for(vector<int>::iterator it = x.begin(); it != x.end(); it++){
        if(find(x.begin(), x.end(), *it) == x.end()){
            y.push_back(*it);
        }
    }

    x = y;

    return;
}

What's the difference? *-)

Oh that back_inserter is nice. I'm reading up on the STL as of lately, so much to learn. :D Thanks Narue.

Ugh. Nevermind. Forgot to allocate. That's sooooooooo C. xD

(Correct?) Working code:

int main() {
    int afrom[4] = {0,0,1,2};
    int ato[4] = {1,2,3,4};
    vector<int> from(&afrom[0], &afrom[4]);
    vector<int> to(&ato[0], &ato[4]);
    vector<int> length;

    vector<int> all_nodes(from.size() + to.size());
    sort(from.begin(), from.end());
    sort(to.begin(), to.end());

    merge(from.begin(), from.end(), to.begin(), to.end(), all_nodes.begin());
    

    return 0;
}

Hi guys,

What am I doing wrong here? Program dies on me @ merge(). :(

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

int main() {
    int afrom[4] = {0,0,1,2};
    int ato[4] = {1,2,3,4};
    vector<int> from(&afrom[0], &afrom[4]);
    vector<int> to(&ato[0], &ato[4]);
    vector<int> length;

    vector<int> all_nodes;
    sort(from.begin(), from.end());
    sort(to.begin(), to.end());
    merge(from.begin(), from.end(), to.begin(), to.end(), all_nodes.begin());

    return 0;
}

Thanks in advance,

Yes, that should be it. remember to use spaces tho. And put the input file in the same directory as tabu_sdvrp.exe of course.

Windows has a table of resources (Windows, buttons, icons, mouse pointers, menus, etc.)
Each resource has an unique ID.
With that ID, you can get that resource, and functions that create a new resource return such an ID.
The simplest example being

HANDLE WINAPI CreateFile(
  __in      LPCTSTR lpFileName,
  __in      DWORD dwDesiredAccess,
  __in      DWORD dwShareMode,
  __in_opt  LPSECURITY_ATTRIBUTES lpSecurityAttributes,
  __in      DWORD dwCreationDisposition,
  __in      DWORD dwFlagsAndAttributes,
  __in_opt  HANDLE hTemplateFile
);

Also on http://msdn.microsoft.com/en-us/library/aa363858(VS.85).aspx

This function creates a resource (a file) and gives you its ID, so you can pass that to ReadFile() for example:

BOOL WINAPI ReadFile(
  __in         HANDLE hFile,
  __out        LPVOID lpBuffer,
  __in         DWORD nNumberOfBytesToRead,
  __out_opt    LPDWORD lpNumberOfBytesRead,
  __inout_opt  LPOVERLAPPED lpOverlapped
);

Second param. :)

So, in this case, HANDLE is like C's FILE*: An unique ID for a file.

Tux4life is right, you can't delete one element of an array in C(++). Either use a vector or do as tux4life said.

Your overloaded >> is a bit complicated, here's a simpler version:

int main() {
    int nom, denom;
    string buf;

    getline(cin, buf);
    stringstream ss(buf);

    ss >> nom;
    //discard all the remaining chars, stop when / is found
    ss.ignore(ss.str().length(), '/');
    ss >> denom;

    cout << "Read: " << nom << "/" << denom << endl;

    return 0;
}

I know this doesn't overload anything, but the code is there. Understand what it does and build it in. It doesn't do any error checking as you can see, so you might want to add that yourself. ;) (Or, you know, screw the user. He should double check his entries. Saves a lot of code :D)

See here why: http://www.informit.com/guides/content.aspx?g=cplusplus&seqNum=134

Boils down to C-style casts being static_casts or reinterpret_casts. But I never got reinterpret_cast to work anyway.

Linux desktop environment KDE is also QT-powered ...

As addition: GNOME is GTK+ powered, however, that was due to the fact that back when QT was still non-free and used by KDE. Only a month ago QT was released with the LGPL license, allowing it to be used just as freely as GTK, so.. have your pick. ;)

I suggest to follow some tutorials for WxWidgets, QT and GTKmm (the OOP-brother of GTK) and decide for yourself what works the best. It'll look good on your resume too! :)

You can only do that when they are defined as constants, so MSVC++ can create the 2D arrays at compile time rather then at runtime.

I suggest reading how to make your own 2D arrays, or using vectors.

Is it a Windows handle, the one in windows.h? It's not an initializer, it's what int and char are: a type.

If it's a window handle, it's basically as specific as a pointer. A pointer used by windows to keep track of resources, varying from windows (window handles) to files (file handles).

Here's a bit more info on windows variables, including HANDLE: http://en.wikibooks.org/wiki/Windows_Programming/Handles_and_Data_Types

You can look all that info up in a STL List reference, as the one found here:

http://www.cplusplus.com/reference/stl/list/

See the "STL Algorithms" for sorting data inside STL containers.

You want to read a screen capture of it?

Yes, the idea was to screen capture the board.

How many different pieces in mahjong?

I have no clue, but I think it differs per puzzle.

you could probably figure out something simple.

Perhaps, hehe. But I don't think it depends on how many different pieces there are, my algorithm with search for exact matches, which is possible since the tiles are exactly the same.

what is this vector class you mentioned?

Short version: STL's resizable arrays.

Long version: I hope you realize I'm not superior to www.cplusplus.com or any reference on the STL when it comes to questions like this right? ;)

If you get stuck on vectors, create a new thread about vectors, but I don't think you'll get stuck on it. It's a great and easy to use class.

I know that's the problem, and MSVC++ is sometimes quite archaic when it comes to "C code". You can try and use the vector class? MSVC++ apparently can't create 2D "standard" arrays dynamically.

You can of course create the array yourself with new, see:
http://www.parashift.com/c++-faq-lite/freestore-mgmt.html#faq-16.16

And bookmark that website. Lots of useful info.