I am trying to use a pointer to a member function in my code so that I can easily replace the function to use without changing everywhere in the code. I get an error while compiling that I don't understand how to solve. Here is a minimum working example:
OrderBook.h
#include <list>
#include <string>
class Order
{
};
class OrderBook
{
typedef void(OrderBook::* MatchingAlgorithm)(Order&, std::list<Order>&);
public:
OrderBook(const std::string name);
void ExecuteFunction(Order&, std::list<Order>);
private:
void FunctionToUse(Order&, std::list<Order>&);
const std::string m_OrderBookName;
MatchingAlgorithm m_matchingAlgorithm;
};
OrderBook.cpp
#include "OrderBook.h"
OrderBook::OrderBook(
const std::string name
)
: m_OrderBookName(name)
{
m_matchingAlgorithm = &OrderBook::FunctionToUse;
}
void OrderBook::ExecuteFunction(Order & order, std::list<Order> listOfOrders)
{
(*m_matchingAlgorithm)(order, listOfOrders);
}
void OrderBook::FunctionToUse(Order &, std::list<Order>&)
{
// do nothing
}
Source.cpp
#include "OrderBook.h"
int main()
{
std::list<Order> mylist;
Order o1, o2;
mylist.push_back(o1);
mylist.push_back(o2);
OrderBook ob("my book");
ob.ExecuteFunction(o1, mylist);
return 0;
}
Compilation Errors
error C2171: '*': illegal on operands of type 'OrderBook::MatchingAlgorithm'
error C2064: term does not evaluate to a function taking 2 arguments
If I replace (*m_matchingAlgorithm) with FunctionToUse inside ExecuteFunction the code compiles without errors.
Change the function call to:
(this->*m_matchingAlgorithm)(order, listOfOrders);
Short answer
You can call your function using the pointer to member operator (.* or ->*):
void OrderBook::ExecuteFunction(Order & order, std::list<Order> listOfOrders)
{
(this->*m_matchingAlgorithm)(order, listOfOrders);
}
More info here.
Long explanations
We are all used to omit this-> for accessing members of the current object. The mecanism that allows this is name resolution principles. But these do not apply for pointer dereferencing operators.
The simple case of a pointer to a data member
Let's star with a simpler case : a pointer to an integer member.
class OrederBook {
...
int x,y;
int OrderBook::* px;
};
In the constructor you couldn't initialize px with &x because &x is an absolute address of an integer, and px is the (relative) address of an integer in an OrderBook. We would then initialize it:
OrderBook::OrderBook(..)
: m_OrderBookName(name), px(&OrderBook::x) {...}
But in this case, the ordinary dereferencing operator for px would not work either:
*px =2; // error because of unary operator *
In fact, for dereferencing such a pointer, you need to know not only the pointer, but also the object in which this pointer should be used:
this->*px =2; // yes ! take relative px address and apply it to this object.
Unary vs. Binary derefencing
The standard defines two different dereferencing operators:
5.3.1/1: The unary * operator performs indirection: the expression to which it
is applied shall be a pointer to an object type, or a pointer to a
function type and the result is an lvalue referring to the object or
function to which the expression points. If the type of the expression
is “pointer to T,” the type of the result is “T.”
5.5/3: The binary operator ->* binds its second operand, which shall be of type “pointer to member of T” to its first operand, which
shall be of type “pointer to T” or “pointer to a class of which T is
an unambiguous and accessible base class.” The expression E1->E2 is
converted into the equivalent form ((E1)).*E2.
Dereferencing a pointer to a member function
For the data example above, you have the choice of the weapons. You could build very easily code achieving the same results, but defining px as int *px instead of int OrderBook::*px and using the unary * as usual.
Unfortunately for poitners to non static member function, you can't use such shortcuts. You really need a pointer to a member function and not a pointer to a function: When you call a member function, in addition to the arguments that you have to pass, you always have to know for which object you call it. And the only way to tell this is to use (->*) or (.*).
Conclusion
Ommitting this->* and assuming the compiler would be intelligent enough to deduct that it's for the current object could seem natural for us. But this is not defined by the standard. It's unfortunate and more here.
Related
On this page I found a good example of function pointers in C++ (as well as of functors, but this question isn't about functors). Below is some copypasta from that page.
#include <iostream>
double add(double left, double right) {
return left + right;
}
double multiply(double left, double right) {
return left * right;
}
double binary_op(double left, double right, double (*f)(double, double)) {
return (*f)(left, right);
}
int main( ) {
double a = 5.0;
double b = 10.0;
std::cout << "Add: " << binary_op(a, b, add) << std::endl;
std::cout << "Multiply: " << binary_op(a, b, multiply) << std::endl;
return 0;
}
I understand the code in general terms, but there are a couple of things that I've always found confusing. Function binary_op() takes a function pointer *f, but when it's used, for example on line 19 binary_op(a, b, add), the function symbol add is passed in, not what one would think of as its pointer, &add. Now you may say that this is because the symbol add is a pointer; it's the address of the bit of code corresponding to the function add(). Very well, but then there still seems to be a type discrepancy here. The function binary_op() takes *f, which means f is a pointer to something. I pass in add, which itself is a pointer to code. (Right?) So then f is assigned the value of add, which makes f a pointer to code, which means that f is a function just like add, which means that f should be called like f(left, right), exactly how add should be called, but on line 12, it's called like (*f)(left, right), which doesn't seem right to me because it would be like writing (*add)(left, right), and *add isn't the function, it's the first character of the code that add points to. (Right?)
I know that replacing the original definition of binary_op() with the following also works.
double binary_op(double left, double right, double f(double, double)) {
return f(left, right);
}
And in fact, this makes much more sense to me, but the original syntax doesn't make sense as I explained above.
So, why is it syntactically correct to use (*f) instead of just f? If the symbol func is itself a pointer, then what precisely does the phrase "function pointer" or "pointer to a function" mean? As the original code currently stands, when we write double (*f)(double, double), what kind of thing is f then? A pointer to a pointer (because (*f) is itself a pointer to a bit of code)? Is the symbol add the same sort of thing as (*f), or the same sort of thing as f?
Now, if the answer to all of this is "Yeah C++ syntax is weird, just memorise function pointer syntax and don't question it.", then I'll reluctantly accept it, but I would really like a proper explanation of what I'm thinking wrong here.
I've read this question and I think I understand that, but haven't found it helpful in addressing my confusion. I've also read this question, which also didn't help because it doesn't directly address my type discrepancy problem. I could keep reading the sea of information on the internet to find my answer but hey, that's what Stack Overflow is for right?
This is because C function pointer are special.
First of, the expression add will decay into a pointer. Just like reference to array will decay into a pointer, reference to function will decay into a pointer to function.
Then, the weird stuff it there:
return (*f)(left, right);
So, why is it syntactically correct to use (*f) instead of just f?
Both are valid, you can rewrite the code like this:
return f(left, right);
This is because the dereference operator will return the reference to the function, and both a reference to a function or a function pointer are considered callable.
The funny thing is that a function reference decay so easily that it will decay back into a pointer when calling the dereference operator, allowing to dereference the function as many time as you want:
return (*******f)(left, right); // ah! still works
As the original code currently stands, when we write double (*f)(double, double), what kind of thing is f then?
The type of f is double (*)(double, double) i.e. it is a pointer to a function of type double(double,double).
because (*f) is itself a pointer
It is not.
Q: What do you get when you indirect through a pointer (such as in *f)? A: You get an lvalue reference. For example, given an object pointer int* ptr, the type of the expression *ptr is int& i.e. lvalue reference to int.
The same is true for function pointers: When you indirect through a function pointer, you get an lvalue reference to the pointed function. In the case of *f, the type is double (&)(double, double) i.e. reference to function of type double(double,double).
Is the symbol add the same sort of thing as (*f), or the same sort of thing as f?
The unqualified id expression add is the same sort of thing as *f i.e. it is an lvalue:
Standard draft [expr.prim.id.unqual]
... The expression is an lvalue if the entity is a function ...
the function symbol add is passed in, not what one would think of as its pointer, &add. Now you may say that this is because the symbol add is a pointer;
No. That's not the reason.
add is not a pointer. It is an lvalue. But lvalues of function type implicitly convert to a pointer (this is called decaying):
Standard draft [conv.func]
An lvalue of function type T can be converted to a prvalue of type “pointer to T”. The result is a pointer to the function.
As such, the following are semantically equivalent:
binary_op(a, b, add); // implicit function-to-pointer conversion
binary_op(a, b, &add); // explicit use of addressof operator
So, why is it syntactically correct to use (*f) instead of just f?
Turns out that calling a function lvalue has the same syntax as calling a function pointer:
Standard draft [expr.call]
A function call is a postfix expression followed by parentheses containing a possibly empty, comma-separated list of initializer-clauses which constitute the arguments to the function.
The postfix expression shall have function type or function pointer type.
For a call to a non-member function or to a static member function, the postfix expression shall either be an lvalue that refers to a function (in which case the function-to-pointer standard conversion ([conv.func]) is suppressed on the postfix expression), or have function pointer type.
These are all the same function call:
add(parameter_list); // lvalue
(*f)(parameter_list); // lvalue
(&add)(parameter_list); // pointer
f(parameter_list); // pointer
P.S. These two declarations are equivalent:
double binary_op(double, double, double (*)(double, double))
double binary_op(double, double, double (double, double))
This is because of the following rule, which is complementary to the implicit decay into function pointer:
Standard draft [dcl.fct]
The type of a function is determined using the following rules.
The type of each parameter (including function parameter packs) is determined from its own decl-specifier-seq and declarator.
After determining the type of each parameter, any parameter of type “array of T” or of function type T is adjusted to be “pointer to T” ...
First of all a function parameter specified as a function declaration is adjusted to pointer to the function when the compiler determinates the type of the parameter. So for example following function declarations
void f( void h() );
void f( void ( *h )() );
are equivalent and declare the same one function.
Consider the following demonstrative program
#include <iostream>
void f( void h() );
void f( void ( *h )() );
void h() { std::cout << "Hello Ray\n"; }
void f( void h() ) { h(); }
int main()
{
f( h );
}
From the c++ 17 Standard (11.3.5 Functions):
5 The type of a function is determined using the following rules. The
type of each parameter (including function parameter packs) is
determined from its own decl-specifier-seq and declarator. After
determining the type of each parameter, any parameter of type “array
of T” or of function type T is adjusted to be “pointer to T”.
On the other hand, according to the C++ 17 Standard
9 When there is no parameter for a given argument, the argument is
passed in such a way that the receiving function can obtain the value
of the argument by invoking va_arg (21.11). [ Note: This paragraph
does not apply to arguments passed to a function parameter pack.
Function parameter packs are expanded during template instantiation
(17.6.3), thus each such argument has a corresponding parameter when a
function template specialization is actually called. — end note ] The
lvalue-to-rvalue (7.1), array-to-pointer (7.2), and function-to-pointer (7.3)
standard conversions are performed on the argument expression
So what is the difference between these two declarations
void f( void h() );
void f( void ( *h )() );
For the first declaration you may consider the parameter h within the function body like a typedef for a function pointer.
typedef void ( *H )();
For example
#include <iostream>
void f( void h() );
void f( void ( *h )() );
void h() { std::cout << "Hello Ray\n"; }
typedef void ( *H )();
void f( H h ) { h(); }
int main()
{
f( h );
}
According to the C++ 17 Standard (8.5.1.2 Function call)
1 A function call is a postfix expression followed by parentheses
containing a possibly empty, comma-separated list of
initializer-clauses which constitute the arguments to the function.
The postfix expression shall have function type or function pointer
type.
So you may also define the function like
void f( void h() ) { ( *h )(); }
Or even like
void f( void h() ) { ( ******h )(); }
because when the operator * is applied to a function name then the function name is implicitly convereted to pijnter to the function.
When I'm passing function as parameter to other functions in c++ , do I have to specify it as
void callOtherFunctions(void f());
or
void callOtherFunctions(void (*f)());
I have no idea what happens under the hood , so I tried running both versions with a simple program as below , replacing the necessary part for 2nd run.
#include <iostream>
using namespace std;
void printHello();
void callOtherFunctions(void f());
int main() {
callOtherFunctions(printHello);
return 0;
}
void printHello(){
std::cout<<"\n Hello World";
}
void callOtherFunctions(void f()){
for (int i=0;i<5;++i){
f();
}
}
and to my surprise , both execute with same output and no warnings. So which is the preferred way , or correct way ( in case I'm doing something wrong ). And what actually happens in each case , when I pass pointer - does it executes address function there and when I pass function - does it copies down whole function there? or something else?
Here is Ideone Link
void callOtherFunctions(void f());
and
void callOtherFunctions(void (*f)());
are identical. Quoting N1570,
§6.7.6.3/8 A declaration of a parameter as "function returning type"
shall be adjusted to "pointer to function returning type", as in
6.3.2.1.
I would prefer the function pointer syntax because more people would be familiar with it and it's explicit what you mean. To answer your second question, a conversion also happens in certain cases (informally known as "decay"):
§6.3.2.1/4 A function designator is an expression that has function
type. Except when it is the operand of the sizeof operator, the
_Alignofoperator,65) or the unary & operator, a
function designator with type "function returning type" is converted to an expression that has type "pointer to function returning type".
Function parameter declarations are somewhat unusual; the compiler will adjust some of the declared types. This is one of them: function parameters of function type are adjusted to the corresponding pointer type.
Other common adjustments to function parameters are array to pointer type, and removing top-level const:
int foo(int a[5]); // a is a pointer
int foo(const int a); // foo() can be called with a non-const int argument.
Do these lines mean the same? Both works without any warning messages!
int (*pFunc)() = func1; // I learned this is right.
int (*pFunc)() = &func1; // Works well with an ampersand too.
Why do I have to put an ampersand in this case? Without it, doesn't work!
int (CMyClass::*pMemberFunc)() = &CMyClass::memberFunc1;
Why do I have to specify namespace for functions in Classes even if the type of the function pointer matches exactly?
int (*pMemberFunc)() = CMyClass::memberFunc1; // Compiler error
int (*pMemberFunc)() = &CMyClass::memberFunc1; // Compiler error
Why can't I specify namespace in this case?
namespace myNamespace {
int func1() {}
}
int (myNamespace::*pFunc)() = myNamespace::func1; // Compiler error
int (myNamespace::*pFunc)() = &myNamespace::func1; // Compiler error
int (*pFunc)() = &myNamespace::func1; // Works!
Your first question, legalese of the Standard:
A pointer to member is only formed when an explicit & is used and its operand is a qualified-id not enclosed
in parentheses. [Note: that is, the expression &(qualified-id), where the qualified-id is enclosed in
parentheses, does not form an expression of type “pointer to member.” Neither does qualified-id, because
there is no implicit conversion from a qualified-id for a non-static member function to the type “pointer to
member function” as there is from an lvalue of function type to the type “pointer to function” (4.3). Nor is
&unqualified-id a pointer to member, even within the scope of the unqualified-id’s class. —end note ]
While it seems inconsistent at first, I do like the fact that it makes the semantics for pointers to members (be them functions or not) equivalent. This certainly has benefits when dealing with templates and decltype.
For the second question, you scope the pointer variable with CMyClass because yours is not a simple "() -> int" function: memberFunc1() is implicitly passed a CmyClass* argument you normally refer to as "this".
If you could truly pass nothing, you'd be missing information (and possibly crash) for the method to do its job correctly. If you're accustomed to "delegates" in other languages, do remember these can optionally store a "Target" pointer to the "this" object if the method is not static/global.
As for the third, it's a free standing function, so it's truly () -> int, but you're attempting to scope your pointer to the namespace, when in fact you're not declaring your variable inside the namespace block.
While the namespace certainly alters how symbols are searched for, it doesn't affect the convention call of the function at all.
1) about Q1, looking at the following code, as the func1 is a right-value of the function, so with or without "&" doesn't change the actual function address.
Look at the following code:
#include <stdio.h>
void foo(){
printf("foo called\n");
}
int main(){
printf("%p\n", foo);
printf("%p\n", &foo);
void (*FUNC) ();
FUNC = foo;
FUNC();
printf("address %p\n", FUNC);
printf("address %p\n", &FUNC);
return 0;
}
output is
0x101406eb0
0x101406eb0
foo called
address 0x101406eb0
address 0x7fff5e7f9a80
Q2 & Q3 )
Pointers to Member Functions Are Not Pointers
You can refer it here
https://www.safaribooksonline.com/library/view/c-common-knowledge/0321321928/ch16.html
I am using the gtkmm library on linux to draw a simple menu for my GUI.
In the below code the compiler complained about unable to resolve address
sigc::mem_fun(*this, AppWindow::hide)));
^
appwindow.cpp:15:41: note: could not resolve address from overloaded function
But when I insert the & it compiles fine
m_menu_app.items().push_back(MenuElem("Quit",
sigc::mem_fun(*this, &AppWindow::hide)));
What difference does it make here? Isn't the hide function just an address in the first place?
This is the exact definition of the function-to-pointer conversion, [conv.func]:
An lvalue of function type T can be converted to a prvalue of type
“pointer to T.” The result is a pointer to the function.55
55) This conversion never applies to non-static member functions because an lvalue that refers to a non-static member function
cannot be obtained.
Thus the decay that we see with normal, non-member functions1 doesn't apply and you need to explicitly take the address.
I.e.
void f();
struct A {
void f();
static void g();
};
auto a = f; // Ok: auto deduced as void(*)()
auto b = A::f; // Error: A::f not an lvalue, auto cannot be deduced
auto c = A::g; // Ok: auto deduced as void(*)()
1 Or static member functions.
For global (non-member) functions, the name of the function evaluates to the address of that function except when passed to the & operator, so you can (for example) assign to a pointer to a function either with or without the & equivalently:
int f() {}
int (*pf1)() = f;
int (*pf2)() = &f;
So, in this case there's really no difference between the two.
For member functions1, however, the rules are a bit different. In this case, the & is required; if you attempt to omit the &, the code simply won't compile (assuming a properly functioning compiler, anyway).
There's no particular reason this would have to be the case--it's just how Bjarne decided things should be. If he'd decided he wanted the name of a member function to evaluate to a pointer to a member (equivalent to how things work for non-member functions) he could have done that.
1. Other than static member functions, which mostly act like non-member functions.
When a function is a non-static member function of a class, then it is necessary to use the form &ClassName::functionName when a pointer to the member function is expected in an expression.
When a function is a static member function of a class, both ClassName::functionName and &ClassName;:functionName can be used when a pointer to a function is expected in an expression.
When a function is a global, i.e. non-member, function, both functionName and &functionName can be used when a pointer to a function is expected in an expression.
I was looking at the emulated version of nullptr and saw this converting operator (a member of nullptr_t):
template<class C, class T> // or any type of null
operator T C::*() const // member pointer...
{ return 0; }
This syntax for pointer to member function confuses me. I usually expect to see such a type as something like
R (C::*)(I1, I2, ...)
With the template above, there's no input arguments. I can't figure out how the type deduction works in this case. I'm having trouble forming a specific question, other than, how does this work? If I have code like this:
typedef int (MyClass::*MyTypedef)(float);
MyTypedef m = nullptr;
I'm guessing T deduces to int, and C deduces to MyClass. What "happens" to float?
That is a pointer to member, not necessarily a pointer to member function. The difference is that it can generate a pointer to member function or a pointer to non-function member.
Now in the particular use case, the destination is a pointer to member, the compiler is seeing an expression in which it needs a int (MyClass::*)(float), and on the other hand it has a nullptr. It tries to find a conversion and it finds the operator T C::*(), which is a valid conversion if C is deduced to be MyClass and T is deduced to be int (float) [function taking a float and returning an int].
I also find this particular corner of the language a bit confusing (having typedefs, or deduced types for functions), for example this is legal if weird:
typedef void int_f(int);
struct X {
int_f m;
};
void X::m(int x) { std::cout << x << '\n'; }
The same thing is going on in the conversion operator that you are concerned with.