I create a basic IBasic interface with a static field
class IBasic
{
public:
IBasic();
virtual ~IBasic();
static std::vector< std::vector<char> > Field;
};
from which the Inherit class is inherited:
class Inherit : public IBasic
{
public:
Inherit(int);
~Inherit();
void Foo();
};
The Inherit class makes some manipulations with Field static member in constructor/or member function.
In order to create an instance of the Inherit class, we need to explicitly declare a static field in the main.cpp before the main function:
#include "Basic.h"
#include "Inherit.h"
std::vector< std::vector<char> > IBasic::Field;
int main()
{
Inherit(10);
return 0;
}
The questions are:
In what namespace does the static method actually exists (global?)? Because I know that static field/function is not a class member in fact.
Is there another way to declare this static method, for example, in a
class file, inside a main function, or through creation unnamed namespace? Is it only one right variant?
How is right? What should be considered first of all?
A static member of a class is a member of its class (that's a tautology) and its class namespace (a class is a namespace). It is not a nember of any other namespace.
A non-const static data member of a class must be defined exactly once in a program, outside of any class, in the same namespace its class is defined in (a global namespace in your case). A header file is inappropriate place for such declaration. It is normally placed in an implementation .cpp file that goes together with the header file.
Having said that, an interface should not have any static data members, much less public ones. It is most likely a grave design error.
In what namespace does the static method actually exists (global?)? Because I know that static field/function is not a class member in fact.
It is declared in scope of the class. In fact the static variable is a class member, your assumption is wrong.
Is there another way to declare this static method, for example, in a class file, inside a main function, or through creation unnamed namespace? Is it only one right variant?
The usual way is to define it in the translation unit that contains the function definitions for the class.
How is right? What should be considered first of all?
There's no right or wrong way, but as mentioned definition in the same translation unit as the class function definitions is the usual way.
Here's an example usage of a static member without any inheritance.
SomeClass.h
#ifndef SOME_CLASS_H
#define SOME_CLASS_H
class SomeClass {
private:
int x;
public:
static SomeClass* const get(); // Needed For Using class to get this pointer
SomeClass();
int getX() const { return x; }
void setX( int val ) { x = val; }
};
#endif // SOME_CLASS_H
SomeClass.cpp
#include "SomeClass.h"
static SomeClass* s_pSomeClass = nullptr;
SomeClass::SomeClass() {
s_pSomeClass = this;
}
SomeClass* const SomeClass::get() {
if ( nullptr == s_pSomeClass ) {
// throw exception
}
return s_pSomeClass;
}
Another class using above class as a static member
OtherClass.h
#ifndef OTHER_CLASS_H
#define OTHER_CLASS_H
class SomeClass; // Forward Declaration
class OtherClass {
private:
static SomeClass* pSomeClass; // The Static Member to this class
int y;
public:
OtherClass();
int getY() const { return y; }
void setY( int val ) { y = val; }
void useSomeClassToSetY();
};
#endif // OTHER_CLASS_H
OtherClass.cpp
#include "OtherClass.h"
#include "SomeClass.h"
SomeClass* OtherClass::pSomeClass = nullptr;
OtherClass::OtherClass() {
if ( nullptr == pSomeClass ) {
pSomeClass = SomeClass::get();
}
}
void OtherClass::useSomeClassToSetY() {
// First Set X To Some Value:
pSomeClass->setX( 10 ); // Use of Static Member
y = pSomeClass->getX(); // Use of Static Member
}
Static members still belong to the class, but they have static storage.
Related
So consider I have a class with a private member variable and a private function which I do not want to define in the header file, because I want to "hide" it from the user.
How can I make this? I can not access the private variable without the declaration of the function in the header.
So what works is something like this:
// header file
class Testclass {
public:
// ...
private:
const int m_i;
void func() const;
}
// cpp file
#include "TestClass.h"
Testclass::func() const {
int j = m_i; //access to a private member variable
}
// ...
But I want something like this:
// header file
class Testclass{
public:
//...
private:
const int m_i;
}
// cpp file
#include "TestClass.h"
Testclass::func() const {
int j = m_i; //access to a private member variable
}
// ...
Which possibilities do I have? I read something about the PIMPL Idiom, but I am not sure if this is what I want, since it looks a bit cumbersome coding-wise.
You can have non-member helper functions in the cpp file, that the class members can use. However, they would have to pass the private variable as a parameter.
// header file
class Testclass{
public:
//...
private:
const int m_i;
}
// cpp file
#include "TestClass.h"
void func(int m_i) {
int j = m_i; //private member variable supplied by caller
}
// ...
Normally one achieves this through the PIMPL (Pointer to IMPLementation) idiom. In your header file you have:
class MainClass
{
public:
void public_function();
private:
class Impl;
Impl* impl;
};
Note that the header file does not contain the definition of the Impl class, only its declaration.
You then define that class in your cpp file and forward the calls from your public interface to the functions of the impl class:
class MainClass::Impl
{
void actualfunc()
{
//do useful stuff here
}
};
void MainClass::public_function()
{
return impl->actualfunc();
}
Apart from your indended hiding of unwanted members from your class users, the PIMPL idiom provides the additional benefit that if no changes are made to the interface of the class, the users of the class need not be recompiled.
Consider these two classes that employ the Pimpl idiom:
ClassA: Pimpl class forward declaration and variable declaration on separate lines
ClassA.h:
#include <memory>
class ClassA {
public:
ClassA();
~ClassA();
void SetValue( int value );
int GetValue() const;
private:
class ClassA_Impl;
// ^^^^^^^^^^^^^^ class forward declaration on its own line
std::unique_ptr<ClassA_Impl> m_pImpl;
// ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ variable declaration on its own line
//EDIT:
//Even if we use a raw pointer instead of a smart pointer,
//i.e. instead of declaring the smart pointer above, if we declare:
ClassA_Impl *m_pImpl;
//the situation in the *.cpp files and my questions (below) are the same.
};
ClassA.cpp:
#include "ClassA.h"
class ClassA::ClassA_Impl {
public:
void SetValue( int value );
int GetValue() const;
private:
int value_;
};
// Private class implementation
void
ClassA::ClassA_Impl::SetValue( int value ) {
value_ = value;
}
int
ClassA::ClassA_Impl::GetValue() const {
return value_;
}
// ClassA implementation
ClassA::ClassA() : m_pImpl( new ClassA_Impl() ) {}
ClassA::~ClassA() {}
void
ClassA::SetValue( int value ) {
m_pImpl->SetValue( value );
}
int
ClassA::GetValue() const {
return m_pImpl->GetValue();
}
ClassB: Pimpl class forward declaration and variable declaration on one line
ClassB.h:
#include <memory>
class ClassB {
public:
ClassB();
~ClassB();
void SetValue( int value );
int GetValue() const;
private:
std::unique_ptr<class ClassB_Impl> m_pImpl;
// ^^^^^^^^^^^^^^^^^^ class forward declaration
// combined with variable declaration on one line,
// in one shot.
//EDIT:
//Even if we use a raw pointer instead of a smart pointer,
//i.e. instead of declaring the smart pointer above, if we declare:
class ClassB_Impl *m_pImpl;
//the situation in the *.cpp files and my questions (below) are the same.
};
ClassB.cpp:
#include "ClassB.h"
class ClassB_Impl {
public:
void SetValue( int value );
int GetValue() const;
private:
int value_;
};
// Private class implementation
void
ClassB_Impl::SetValue( int value ) {
value_ = value;
}
int
ClassB_Impl::GetValue() const {
return value_;
}
// ClassB implementation
ClassB::ClassB() : m_pImpl( new ClassB_Impl() ) {}
ClassB::~ClassB() {}
void
ClassB::SetValue( int nValue ) {
m_pImpl->SetValue( nValue );
}
int
ClassB::GetValue() const {
return m_pImpl->GetValue();
}
Questions:
Why does combining the forward declaration and variable declaration on one line in ClassB.h require ClassB_Impl to be "unscoped" in the implementation of the private class in ClassB.cpp?
i.e. in ClassA.cpp, private class method definitions begin with
void ClassA::ClassA_Impl::foo() {...
but in ClassB.cpp, private class method definitions begin with
void ClassB_Impl::foo() {...
What are the implications of each method? Which one is better?
(Follow-up question in response to Galik's answer)
When you combine forward declaration of a class and declaration of a variable of that class in one statement...
//one-liner
class ClassB_Impl *m_pImpl;
...what is this called? Is there a name for this kind of combined statement? And why exactly doesn't ClassB_Impl become an inner class of ClassB as a result of such a statement?
Compare this to...
//two-liner
class ClassA_Impl;
ClassA_Impl *m_pImpl;
...in which case ClassA_Impl does become an inner class of ClassA.
Why does the one-liner put ClassB_Impl into the global namepace, while the two-liner puts ClassA_Impl into ClassA's namespace? Why are they different?
Why does combining the forward declaration and variable declaration on
one line in ClassB.h require ClassB_Impl to be "unscoped" in the
implementation of the private class in ClassB.cpp?
Because in the first example you declare ClassA_Impl as an inner class of ClassA.
When you declare ClassB_Impl it in the template parameter that is not part of ClassB.
What are the implications of each method? Which one is better?
This is a matter of opinion. Personally I think inner classes are messy and harder to work with for little reward.
My preferred method uses a separate interface class which helps to reduce the number of times you have to redeclare the interface.
See: Is it possible to write an agile Pimpl in c++?
/*
* CDummy.h
*/
#ifndef CDUMMY_H_
#define CDUMMY_H_
class CDummy {
public:
CDummy();
virtual ~CDummy();
};
#endif /* CDUMMY_H_ */
I've read that one should not declare class variables in header files. ist this right?
So I declare it in the cpp file below:
/*
* CDummy.cpp
*/
#include "CDummy.h"
static int counter = 0; //so here is my static counter. is this now private or public? how can i make it public, i cannot introduce a public block here.
CDummy::CDummy() {
counter++;
}
CDummy::~CDummy() {
counter--;
}
using this code i cannot access the classvariable from my main program....
thx
A "class variable" needs to belong to a class. So it has to be declared in the class definition. If the class definition is in a header file, then it follows that the class variable declaration must also be in a header file.
The class variable's definition should go in an implementation file, typically the one in which the class' members are defined. Here's a simplified example:
Foo.h
struct Foo
{
void foo() const;
static int FOO; // declaration
};
Foo.cpp
void Foo::foo() {}
int Foo::FOO = 42; // definition
What you have here:
static int counter = 0;
is a static variable that is not a static member of any class. It is only non-member static variable, static to the compilation unit of CDummy.cpp.
static int counter = 0; //so here is my static counter. is this now private or public? how can i make it public, i cannot introduce a public block here.
From the code I see counter is just a global static variables as it is not declated anywhere in your CDummy
Static variables should be public so that you could initialize them outside the class declaration. Your code should look like this to have it public:
class CDummy {
public:
static int count;
CDummy();
virtual ~CDummy();
};
// inside CDummy.cpp
int CDummy::count = 0;
Here you can read more about how to use static variables in class declaration.
I am currently doing a checker for the top level domains of email addresses. In order to check, I am comparing it to a list that is a text file. I want to import the list into a static map container. However, when I try to instantiate it, it says that it cannot be defined in the current scope. Why is that?
This is some my header file:
class TldPart {
public:
static void LoadTlds();
private:
static map<string,bool> Tld;
}
Here is the implementation in the cpp:
void TldPart::LoadTlds()
{
map<string,bool> Tld;
...
}
It is telling me that ValidTld cannot be defined in the LoadTlds function.
Static members of a class exist outside of an object. You should define and initialize static member outside of the class.
Here we define and initialize a static class member:
header-file:
#pragma once
#include <map>
#include <string>
class TldPart {
public:
static void LoadTlds();
private:
static std::map<std::string, bool> tld;
};
your cpp-file:
#include "external.h"
std::map<std::string,bool> TldPart::tld;
void TldPart::LoadTlds()
{
tld.insert(std::make_pair("XX", true));
}
And don't forget semicolon at the end of the class.
EDIT: You can provide in-class initializers for static members of const integral type or static members that are constexprs and has literal type.
I am trying to implement the following class. However, when I try to instantiate an object of the class within its definition and pass "0" as value to initialize the object, i get an error:
"a type specifier is expected".
Can anyone explain how can i remove this error?
class MessageType
{
public:
static const MessageType msgEvent(0);
private:
MessageType();
virtual ~MessageType();
MessageType(int);
};
You need to initialize(define) it outside the class definition in a cpp file.
MessageType const MessageType::msgEvent;
However, Your intent in doing so is not very clear. Are you trying to implement a Singleton Pattern, probably this sample implementation might help, I leave it to you to decide, whether you really need a singleton, inspite of its disadvantages:
//MessageType.h
#include <boost/noncopyable.hpp>
class MessageType: private boost::noncopyable
{
public:
static MessageType* instance();
private:
MessageType();
~MessageType();
static bool g_initialised;
// static initialisation
static MessageType g_instance;
// dynamic initialisation
};
// MessageType.cpp
#include "MessageType.hpp"
#include <ostream>
#include <iostream>
#include <cstring>
bool MessageType::g_initialised;
// static initialisation
MessageType MessageType::g_instance;
// dynamic initialisation
MessageType::MessageType()
{
g_initialised = true;
}
MessageType::~MessageType()
{
g_initialised = false;
}
MessageType* MessageType::instance()
{
return g_initialised ? &g_instance : 0;
}
You can only initialize static member variables in the definition if they are of int type.
class MessageType
{
public:
static int sCount = 0; // That is fine.
static std::string sLogName; // That is fine.
static std::string sLogName("log.txt"); // Fail!
};
There's no way around this rule. If you want to initialize a static member variable, then you have to do it in the cpp:
std::string MessageType::sLogName("log.txt"); // fine, in the cpp.
This same rule applies directly to your MessageType instance, and has nothing to do with the fact that the class is of it's own type.