Type erasure techniques - c++

(With type erasure, I mean hiding some or all of the type information regarding a class, somewhat like Boost.Any.)
I want to get a hold of type erasure techniques, while also sharing those, which I know of. My hope is kinda to find some crazy technique that somebody thought of in his/her darkest hour. :)
The first and most obvious, and commonly taken approach, that I know, are virtual functions. Just hide the implementation of your class inside an interface based class hierarchy. Many Boost libraries do this, for example Boost.Any does this to hide your type and Boost.Shared_ptr does this to hide the (de)allocation mechanic.
Then there is the option with function pointers to templated functions, while holding the actual object in a void* pointer, like Boost.Function does to hide the real type of the functor. Example implementations can be found at the end of the question.
So, for my actual question:
What other type erasure techniques do you know of? Please provide them, if possible, with an example code, use cases, your experience with them and maybe links for further reading.
Edit
(Since I wasn't sure wether to add this as an answer, or just edit the question, I'll just do the safer one.)
Another nice technique to hide the actual type of something without virtual functions or void* fiddling, is the one GMan employs here, with relevance to my question on how exactly this works.
Example code:
#include <iostream>
#include <string>
// NOTE: The class name indicates the underlying type erasure technique
// this behaves like the Boost.Any type w.r.t. implementation details
class Any_Virtual{
struct holder_base{
virtual ~holder_base(){}
virtual holder_base* clone() const = 0;
};
template<class T>
struct holder : holder_base{
holder()
: held_()
{}
holder(T const& t)
: held_(t)
{}
virtual ~holder(){
}
virtual holder_base* clone() const {
return new holder<T>(*this);
}
T held_;
};
public:
Any_Virtual()
: storage_(0)
{}
Any_Virtual(Any_Virtual const& other)
: storage_(other.storage_->clone())
{}
template<class T>
Any_Virtual(T const& t)
: storage_(new holder<T>(t))
{}
~Any_Virtual(){
Clear();
}
Any_Virtual& operator=(Any_Virtual const& other){
Clear();
storage_ = other.storage_->clone();
return *this;
}
template<class T>
Any_Virtual& operator=(T const& t){
Clear();
storage_ = new holder<T>(t);
return *this;
}
void Clear(){
if(storage_)
delete storage_;
}
template<class T>
T& As(){
return static_cast<holder<T>*>(storage_)->held_;
}
private:
holder_base* storage_;
};
// the following demonstrates the use of void pointers
// and function pointers to templated operate functions
// to safely hide the type
enum Operation{
CopyTag,
DeleteTag
};
template<class T>
void Operate(void*const& in, void*& out, Operation op){
switch(op){
case CopyTag:
out = new T(*static_cast<T*>(in));
return;
case DeleteTag:
delete static_cast<T*>(out);
}
}
class Any_VoidPtr{
public:
Any_VoidPtr()
: object_(0)
, operate_(0)
{}
Any_VoidPtr(Any_VoidPtr const& other)
: object_(0)
, operate_(other.operate_)
{
if(other.object_)
operate_(other.object_, object_, CopyTag);
}
template<class T>
Any_VoidPtr(T const& t)
: object_(new T(t))
, operate_(&Operate<T>)
{}
~Any_VoidPtr(){
Clear();
}
Any_VoidPtr& operator=(Any_VoidPtr const& other){
Clear();
operate_ = other.operate_;
operate_(other.object_, object_, CopyTag);
return *this;
}
template<class T>
Any_VoidPtr& operator=(T const& t){
Clear();
object_ = new T(t);
operate_ = &Operate<T>;
return *this;
}
void Clear(){
if(object_)
operate_(0,object_,DeleteTag);
object_ = 0;
}
template<class T>
T& As(){
return *static_cast<T*>(object_);
}
private:
typedef void (*OperateFunc)(void*const&,void*&,Operation);
void* object_;
OperateFunc operate_;
};
int main(){
Any_Virtual a = 6;
std::cout << a.As<int>() << std::endl;
a = std::string("oh hi!");
std::cout << a.As<std::string>() << std::endl;
Any_Virtual av2 = a;
Any_VoidPtr a2 = 42;
std::cout << a2.As<int>() << std::endl;
Any_VoidPtr a3 = a.As<std::string>();
a2 = a3;
a2.As<std::string>() += " - again!";
std::cout << "a2: " << a2.As<std::string>() << std::endl;
std::cout << "a3: " << a3.As<std::string>() << std::endl;
a3 = a;
a3.As<Any_Virtual>().As<std::string>() += " - and yet again!!";
std::cout << "a: " << a.As<std::string>() << std::endl;
std::cout << "a3->a: " << a3.As<Any_Virtual>().As<std::string>() << std::endl;
std::cin.get();
}

All type erasure techniques in C++ are done with function pointers (for behaviour) and void* (for data). The "different" methods simply differ in the way they add semantic sugar. Virtual functions, e.g., are just semantic sugar for
struct Class {
struct vtable {
void (*dtor)(Class*);
void (*func)(Class*,double);
} * vtbl
};
iow: function pointers.
That said, there's one technique I particularly like, though: It's shared_ptr<void>, simply because it blows the minds off of people who don't know you can do this: You can store any data in a shared_ptr<void>, and still have the correct destructor called at the end, because the shared_ptr constructor is a function template, and will use the type of the actual object passed for creating the deleter by default:
{
const shared_ptr<void> sp( new A );
} // calls A::~A() here
Of course, this is just the usual void*/function-pointer type erasure, but very conveniently packaged.

Fundamentally, those are your options: virtual functions or function pointers.
How you store the data and associate it with the functions can vary. For example, you could store a pointer-to-base, and have the derived class contain the data and the virtual function implementations, or you could store the data elsewhere (e.g. in a separately allocated buffer), and just have the derived class provide the virtual function implementations, which take a void* that points to the data. If you store the data in a separate buffer, then you could use function pointers rather than virtual functions.
Storing a pointer-to-base works well in this context, even if the data is stored separately, if there are multiple operations that you wish to apply to your type-erased data. Otherwise you end up with multiple function pointers (one for each of the type-erased functions), or functions with a parameter that specifies the operation to perform.

I would also consider (similar to void*) the use of "raw storage": char buffer[N].
In C++0x you have std::aligned_storage<Size,Align>::type for this.
You can store anything you want in there, as long as it's small enough and you deal with the alignment properly.

Stroustrup, in The C++ programming language (4th edition) ยง25.3, states:
Variants of the technique of using a single runt-time representation for values of a number of types and relying on the (static) type system to ensure that they are used only according to their declared type has been called type erasure.
In particular, no use of virtual functions or function pointers is needed to perform type erasure if we use templates. The case, already mentioned in other answers, of the correct destructor call according to the type stored in a std::shared_ptr<void> is an example of that.
The example provided in Stroustrup's book is just as enjoyable.
Think about implementing template<class T> class Vector, a container along the lines of std::vector. When you will use your Vector with a lot of different pointers types, as it often happens, the compiler will supposedly generate different code for every pointer type.
This code bloat can be prevented by defining a specialization of Vector for void* pointers and then using this specialization as a common base implementation of Vector<T*> for all others types T:
template<typename T>
class Vector<T*> : private Vector<void*>{
// all the dirty work is done once in the base class only
public:
// ...
// static type system ensures that a reference of right type is returned
T*& operator[](size_t i) { return reinterpret_cast<T*&>(Vector<void*>::operator[](i)); }
};
As you can see, we have a strongly typed container but Vector<Animal*>, Vector<Dog*>, Vector<Cat*>, ..., will share the same (C++ and binary) code for the implementation, having their pointer type erased behind void*.

See this series of posts for a (fairly short) list of type erasure techniques and the discussion about the trade-offs:
Part I,
Part II,
Part III,
Part IV.
The one I haven't seen mentioned yet is Adobe.Poly, and Boost.Variant, which can be considered a type erasure to some extent.

As stated by Marc, one can use cast std::shared_ptr<void>.
For example store the type in a function pointer, cast it and store in a functor of only one type:
#include <iostream>
#include <memory>
#include <functional>
using voidFun = void(*)(std::shared_ptr<void>);
template<typename T>
void fun(std::shared_ptr<T> t)
{
std::cout << *t << std::endl;
}
int main()
{
std::function<void(std::shared_ptr<void>)> call;
call = reinterpret_cast<voidFun>(fun<std::string>);
call(std::make_shared<std::string>("Hi there!"));
call = reinterpret_cast<voidFun>(fun<int>);
call(std::make_shared<int>(33));
call = reinterpret_cast<voidFun>(fun<char>);
call(std::make_shared<int>(33));
// Output:,
// Hi there!
// 33
// !
}

Related

Alternative to overloading functions with derived types

Sorry for the uninformative title, I don't really know what to call what I am asking.
I want to achieve the following: Having a container of a base class type with instances of derived types, accessing the containers and calling a function overload dependent on the type of the derived object accessed. In a question earlier I asked here I learned that the static design I had in mind so far, does not work. The way I tried is this:
struct Int2TypeBase{
};
template <int v>
struct Int2Type : public Int2TypeBase
{
enum
{
value = v
};
};
void f(const Int2Type<0>&){
std::cout << "f(const Int2Type<0>&)" << "\n";
}
void f(const Int2Type<1>&){
std::cout << "f(const Int2Type<1>&)" << "\n";
}
int main(){
using namespace std;
std::vector<std::reference_wrapper<Int2TypeBase>> v;
Int2Type<0> i2t_1;
v.emplace_back(i2t_1);
Int2Type<1> i2t_2;
v.emplace_back(i2t_2);
auto x0 = v[0];
auto x1 = v[1];
f(x0.get()); // After my imagination this would have called void f(const Int2Type<0>&)
f(x1.get()); // After my imagination this would have called void f(const Int2Type<1>&)
}
Ok, so I want the correct overload of f to be selected, this however does not compile as at compile time it is unknown which type x0 and x1 actually have. But is there some alternate design that can realize this behavior?
Overloading is a static mechanism based on static types.
If you want to change behaviour dynamically based on the dynamic type of an object, C++ provides another built-in language feature for that: Virtual functions. Use them like this:
struct Int2TypeBase
{
virtual void do_f() = 0;
};
template <int v> struct Int2Type : Int2TypeBase
{
void do_f() override
{
// specific behaviour for Int2Type<v> goes here
}
/* ... */
};
void f(Int2TypeBase & x) { x.do_f(); }
Now you can call f on any base subobject and the correct behaviour is selected at runtime. In particular, f(x0.get()) and f(x1.get()) now select and dispatch to Int2Type<0>::do_f and Int2Type<1>::do_f at runtime, respectively.

Implicitly cast wrapper class to supperclass in templated call

In designing a DSL (which compiles into C++), I found it convenient to define a wrapper class that, uppon destruction, would call a .free() method on the contained class:
template<class T>
class freeOnDestroy : public T {
using T::T;
public:
operator T&() const { return *this; }
~freeOnDestroy() { T::free(); }
};
The wrapper is designed to be completely transparent: All methods, overloads and constructors are inherited from T (at least to my knowledge), but when included in the wrapper, the free() method is called uppon destruction. Note that I explicitly avoid using T's destructor for this since T::free() and ~T() may have different semantics!
All this works fine, untill a wrapped class gets used as a member to a non-reference templated call, at which point freeOnDestroy is instantiated, calling free on the wrapped object. What I would like to happen is for the tempated method to use T instead of freeOnDestroy<T>, and to implicitly cast the parameter into the supperclass. The following code sample illustrates this problem:
// First class that has a free (and will be used in foo)
class C{
int * arr;
public:
C(int size){
arr = new int[size];
for (int i = 0; i < size; i++) arr[i] = i;
}
int operator[] (int idx) { return arr[idx]; }
void free(){ cout << "free called!\n"; delete []arr; }
};
// Second class that has a free (and is also used in foo)
class V{
int cval;
public:
V(int cval) : cval(cval) {}
int operator[] (int idx) { return cval; }
void free(){}
};
// Foo: in this case, accepts anything with operator[int]
// Foo cannot be assumed to be written as T &in!
// Foo in actuality may have many differently-templated parameters, not just one
template<typename T>
void foo(T in){
for(int i = 0; i < 5; i++) cout << in[i] << ' ';
cout << '\n';
}
int main(void){
C c(15);
V v(1);
freeOnDestroy<C> f_c(15);
foo(c); // OK!
foo(v); // OK!
foo<C>(f_c); // OK, but the base (C) of f_c may not be explicitly known at the call site, for example, if f_c is itself received as a template
foo(f_c); // BAD: Creates a new freeOnDestroy<C> by implicit copy constructor, and uppon completion calls C::free, deleting arr! Would prefer it call foo<C>
foo(f_c); // OH NO! Tries to print arr, but it has been deleted by previous call! Segmentation fault :(
return 0;
}
A few non solutions I should mention are:
Making freeOnDestroy::freeOnDestroy(const freeOnDestroy &src) explicit and private, but this seems to override T's constructor. I'd hoped it would try to implicitly convert it to T and use that as the template argument.
Assume foo receives a reference of its templated arguments (as in void foo(T &in): This is neither the case, nor desirable in some cases
Always explicitly template the call to foo, as in foo<C>(f_c): f_c itself may be templated, so it's hard to know to instantiate foo with C (yes, this could be done with creating multiple versions of foo, to remove the wrappers one by one, but I can't find a way of doing that without creating a different overload for each templated argument of foo).
In summary, my question is: Is there a clean(ish) method to ensure a base class will be casted to its superclass when resolving a template? Or, if not, is there some way of using SFINAE, by causing a substitution failure when the template argument is an instance of the wrapper class, and thus force it to use the implicit cast to the wrapped class (without duplicating each foo-like method signature possibly dozens of times)?
I presently have a work-arround that involves changes in the DSL, but I'm not entirely happy with it, and was curious if it was at all possible to design a wrapper class that works as described.
The problem here not when "wrapped class gets used as a member to a non-reference templated call".
The problem here is that the template wrapper -- and likely its superclass too -- has violated the Rule Of Three.
Passing an instance of the class as a non-reference parameter is just another way of saying "passing by value". Passing by value makes a copy of the instance of the class. Neither your template class -- nor its wrapped class, most likely -- has an explicit copy constructor; as such the copied instance of the class has no knowledge that it is a copy, hence the destructor does what it thinks it should do.
The correct solution here is not to hack something up that makes passing an instance of freeOnDestroy<T> by value end up copying T, rather than freeOnDestroy<T>. The correct solution is to add a proper copy-constructor and the assignment operator to both the freeOnDestroy template, and possibly any superclass that uses it, so that everything complies with the Rule Of Three.
You can use a properly defined detector and a sfinaed function, as it follows:
#include<iostream>
#include<type_traits>
template<class T>
class freeOnDestroy : public T {
using T::T;
public:
operator T&() const { return *this; }
~freeOnDestroy() { T::free(); }
};
template<typename T>
struct FreeOnDestroyDetector: std::false_type { };
template<typename T>
struct FreeOnDestroyDetector<freeOnDestroy<T>>: std::true_type { };
class C{
int * arr;
public:
C(int size){
arr = new int[size];
for (int i = 0; i < size; i++) arr[i] = i;
}
int operator[] (int idx) { return arr[idx]; }
void free(){ std::cout << "free called!\n"; delete []arr; }
};
class V{
int cval;
public:
V(int cval) : cval(cval) {}
int operator[] (int idx) { return cval; }
void free(){}
};
template<typename..., typename T>
std::enable_if_t<not FreeOnDestroyDetector<std::decay_t<T>>::value>
foo(T in) {
std::cout << "here you have not a freeOnDestroy based class" << std::endl;
}
template<typename..., typename T>
std::enable_if_t<FreeOnDestroyDetector<std::decay_t<T>>::value>
foo(T &in) {
std::cout << "here you have a freeOnDestroy based class" << std::endl;
}
int main(void){
C c(15);
V v(1);
freeOnDestroy<C> f_c(15);
foo(c);
foo(v);
foo<C>(f_c);
foo(f_c);
foo(f_c);
return 0;
}
As you can see by running the example, free is called only once, that is for the freeOnDestroy created in the main function.
If you want to forbid definitely freeOnDestroy as a parameter, you can use a single function as the following one:
template<typename..., typename T>
void foo(T &in) {
static_assert(not FreeOnDestroyDetector<std::decay_t<T>>::value, "!");
std::cout << "here you have a freeOnDestroy based class" << std::endl;
}
Note that I added a variadic parameter as a guard, so that one can no longer use foo<C>(f_c); to force a type to be used.
Remove it if you want to allow such an expression. It was not clear from the question.
One solution, which, although a little ugly, seems to work, is to use an overloaded unwrapping method, such as:
template<typename T> T freeOnDestroyUnwrapper(const T &in){ return in; }
template<typename T> T freeOnDestroyUnwrapper(const freeOnDestroy<T> &in){ return in; }
template<typename T> T freeOnDestroyUnwrapper(const freeOnDestroy<typename std::decay<T>::type> &in){ return in; }
template<typename T> T& freeOnDestroyUnwrapper(T &in){ return in; }
template<typename T> T& freeOnDestroyUnwrapper(freeOnDestroy<T> &in){ return in; }
template<typename T> T& freeOnDestroyUnwrapper(freeOnDestroy<typename std::decay<T>::type> &in){ return in; }
Then, calls can be made using the unwrapper:
int main(void){
C c(15);
V v(1);
freeOnDestroy<C> f_c(15);
foo(freeOnDestroyUnwrapper(c));
foo(freeOnDestroyUnwrapper(v));
foo<C>(freeOnDestroyUnwrapper(f_c));
foo(freeOnDestroyUnwrapper(f_c));
foo(freeOnDestroyUnwrapper(f_c));
return 0;
}
Or, to make this less verbose, we can alter foo so it does this for us:
template<typename T>
void _foo(T in){
for(int i = 0; i < 5; i++) cout << in[i] << ' ';
cout << '\n';
}
template<typename... Ts>
void foo(Ts&&... args){
_foo(freeOnDestroyUnwrapper(args)...);
}
And then call it as normal:
int main(void){
C c(15);
V v(1);
freeOnDestroy<C> f_c(15);
foo(c);
foo(v);
//foo<C>(f_c); // This now doesn't work!
foo(f_c);
foo(f_c);
return 0;
}
This seems to work for any number of arguments foo may have (of different templates, if needed), and seems to behave appropriately when foos input is a reference (which does not occur in my context, but would be good for the sake of making this solution generic).
I'm not convinced that this is the best solution, or that it generalizes to every case, plus, having to double all declarations is a bit cumbersome, and opaque to most IDEs autocomplete features. Better solutions and improvements are welcome!

Use of rvalue references in function parameter of overloaded function creates too many combinations

Imagine you have a number of overloaded methods that (before C++11) looked like this:
class MyClass {
public:
void f(const MyBigType& a, int id);
void f(const MyBigType& a, string name);
void f(const MyBigType& a, int b, int c, int d);
// ...
};
This function makes a copy of a (MyBigType), so I want to add an optimization by providing a version of f that moves a instead of copying it.
My problem is that now the number of f overloads will duplicate:
class MyClass {
public:
void f(const MyBigType& a, int id);
void f(const MyBigType& a, string name);
void f(const MyBigType& a, int b, int c, int d);
// ...
void f(MyBigType&& a, int id);
void f(MyBigType&& a, string name);
void f(MyBigType&& a, int b, int c, int d);
// ...
};
If I had more parameters that could be moved, it would be unpractical to provide all the overloads.
Has anyone dealt with this issue? Is there a good solution/pattern to solve this problem?
Thanks!
Herb Sutter talks about something similar in a cppcon talk
This can be done but probably shouldn't. You can get the effect out using universal references and templates, but you want to constrain the type to MyBigType and things that are implicitly convertible to MyBigType. With some tmp tricks, you can do this:
class MyClass {
public:
template <typename T>
typename std::enable_if<std::is_convertible<T, MyBigType>::value, void>::type
f(T&& a, int id);
};
The only template parameter will match against the actual type of the parameter, the enable_if return type disallows incompatible types. I'll take it apart piece by piece
std::is_convertible<T, MyBigType>::value
This compile time expression will evaluate to true if T can be converted implicitly to a MyBigType. For example, if MyBigType were a std::string and T were a char* the expression would be true, but if T were an int it would be false.
typename std::enable_if<..., void>::type // where the ... is the above
this expression will result in void in the case that the is_convertible expression is true. When it's false, the expression will be malformed, so the template will be thrown out.
Inside the body of the function you'll need to use perfect forwarding, if you are planning on copy assigning or move assigning, the body would be something like
{
this->a_ = std::forward<T>(a);
}
Here's a coliru live example with a using MyBigType = std::string. As Herb says, this function can't be virtual and must be implemented in the header. The error messages you get from calling with a wrong type will be pretty rough compared to the non-templated overloads.
Thanks to Barry's comment for this suggestion, to reduce repetition, it's probably a good idea to create a template alias for the SFINAE mechanism. If you declare in your class
template <typename T>
using EnableIfIsMyBigType = typename std::enable_if<std::is_convertible<T, MyBigType>::value, void>::type;
then you could reduce the declarations to
template <typename T>
EnableIfIsMyBigType<T>
f(T&& a, int id);
However, this assumes all of your overloads have a void return type. If the return type differs you could use a two-argument alias instead
template <typename T, typename R>
using EnableIfIsMyBigType = typename std::enable_if<std::is_convertible<T, MyBigType>::value,R>::type;
Then declare with the return type specified
template <typename T>
EnableIfIsMyBigType<T, void> // void is the return type
f(T&& a, int id);
The slightly slower option is to take the argument by value. If you do
class MyClass {
public:
void f(MyBigType a, int id) {
this->a_ = std::move(a); // move assignment
}
};
In the case where f is passed an lvalue, it will copy construct a from its argument, then move assign it into this->a_. In the case that f is passed an rvalue, it will move construct a from the argument and then move assign. A live example of this behavior is here. Note that I use -fno-elide-constructors, without that flag, the rvalue cases elides the move construction and only the move assignment takes place.
If the object is expensive to move (std::array for example) this approach will be noticeably slower than the super-optimized first version. Also, consider watching this part of Herb's talk that Chris Drew links to in the comments to understand when it could be slower than using references. If you have a copy of Effective Modern C++ by Scott Meyers, he discusses the ups and downs in item 41.
You may do something like the following.
class MyClass {
public:
void f(MyBigType a, int id) { this->a = std::move(a); /*...*/ }
void f(MyBigType a, string name);
void f(MyBigType a, int b, int c, int d);
// ...
};
You just have an extra move (which may be optimized).
My first thought is that you should change the parameters to pass by value. This covers the existing need to copy, except the copy happens at the call point rather than explicitly in the function. It also allows the parameters to be created by move construction in a move-able context (either unnamed temporaries or by using std::move).
Why you would do that
These extra overloads only make sense, if modifying the function paramers in the implementation of the function really gives you a signigicant performance gain (or some kind of guarantee). This is hardly ever the case except for the case of constructors or assignment operators. Therefore, I would advise you to rethink, whether putting these overloads there is really necessary.
If the implementations are almost identical...
From my experience this modification is simply passing the parameter to another function wrapped in std::move() and the rest of the function is identical to the const & version. In that case you might turn your function into a template of this kind:
template <typename T> void f(T && a, int id);
Then in the function implementation you just replace the std::move(a) operation with std::forward<T>(a) and it should work. You can constrain the parameter type T with std::enable_if, if you like.
In the const ref case: Don't create a temporary, just to to modify it
If in the case of constant references you create a copy of your parameter and then continue the same way the move version works, then you may as well just pass the parameter by value and use the same implementation you used for the move version.
void f( MyBigData a, int id );
This will usually give you the same performance in both cases and you only need one overload and implementation. Lots of plusses!
Significantly different implementations
In case the two implementations differ significantly, there is no generic solution as far as I know. And I believe there can be none. This is also the only case, where doing this really makes sense, if profiling the performance shows you adequate improvements.
You might introduce a mutable object:
#include <memory>
#include <type_traits>
// Mutable
// =======
template <typename T>
class Mutable
{
public:
Mutable(const T& value) : m_ptr(new(m_storage) T(value)) {}
Mutable(T& value) : m_ptr(&value) {}
Mutable(T&& value) : m_ptr(new(m_storage) T(std::move(value))) {}
~Mutable() {
auto storage = reinterpret_cast<T*>(m_storage);
if(m_ptr == storage)
m_ptr->~T();
}
Mutable(const Mutable&) = delete;
Mutable& operator = (const Mutable&) = delete;
const T* operator -> () const { return m_ptr; }
T* operator -> () { return m_ptr; }
const T& operator * () const { return *m_ptr; }
T& operator * () { return *m_ptr; }
private:
T* m_ptr;
char m_storage[sizeof(T)];
};
// Usage
// =====
#include <iostream>
struct X
{
int value = 0;
X() { std::cout << "default\n"; }
X(const X&) { std::cout << "copy\n"; }
X(X&&) { std::cout << "move\n"; }
X& operator = (const X&) { std::cout << "assign copy\n"; return *this; }
X& operator = (X&&) { std::cout << "assign move\n"; return *this; }
~X() { std::cout << "destruct " << value << "\n"; }
};
X make_x() { return X(); }
void fn(Mutable<X>&& x) {
x->value = 1;
}
int main()
{
const X x0;
std::cout << "0:\n";
fn(x0);
std::cout << "1:\n";
X x1;
fn(x1);
std::cout << "2:\n";
fn(make_x());
std::cout << "End\n";
}
This is the critical part of the question:
This function makes a copy of a (MyBigType),
Unfortunately, it is a little ambiguous. We would like to know what is the ultimate target of the data in the parameter. Is it:
1) to be assigned to an object that existing before f was called?
2) or instead, stored in a local variable:
i.e:
void f(??? a, int id) {
this->x = ??? a ???;
...
}
or
void f(??? a, int id) {
MyBigType a_copy = ??? a ???;
...
}
Sometimes, the first version (the assignment) can be done without any copies or moves. If this->x is already long string, and if a is short, then it can efficiently reuse the existing capacity. No copy-construction, and no moves. In short, sometimes assignment can be faster because we can skip the copy contruction.
Anyway, here goes:
template<typename T>
void f(T&& a, int id) {
this->x = std::forward<T>(a); // is assigning
MyBigType local = std::forward<T>(a); // if move/copy constructing
}
If the move version will provide any optimization then the implementation of the move overloaded function and the copy one must be really different. I don't see a way to get around this without providing implementations for both.

Run-time polymorphic class with value semantics

I would like a class Value, which both has a run-time polymorphic behaviour, and a value semantics. For instance, I would like to be able to do things like:
// create polymorphic data
Value v1 = IntValue(42);
Value v2 = DoubleValue(12.3);
// copy-by-value semantics
Value v3 = v1;
v3.increments();
Value v4;
v4 = v2;
v4.increments();
// possibly put them in my favourite container
MyList<Value> l;
l << v1 << v2 << v3 << v4;
// print them: "Int(42) Double(12.0) Int(43) Double(13.0) "
for(int i=0; i<l.size(); i++) l[i].print();
Is it possible, and if yes, how?
Note: Using boost or C++11 smart pointers as here is not desired: they make the caller code verbose, use -> instead of ., and do not have copy constructors or assignment operators implementing a true value semantics. Also, this question doesn't target specifically containers.
polymorphic_value has been proposed for standardisation and has some of the semantics you require. You'll have to define your own operator << though.
A polymorphic_value<T> may hold a an object of a class publicly derived from T, and copying the polymorphic_value will copy the object of the derived type.
polymorphic_value<T> is implemented with type erasure and uses the compiler-generated copy-constructor of the derived objects to correctly copy objects stored as polymorphic_value<BaseType>.
Copy constructors and assignment operators are defined so that the objects are value-like. There is no need to use or define a custom clone method.
In brief:
template <class T>
struct control_block
{
virtual ~control_block() = default;
virtual T* ptr() = 0;
virtual std::unique_ptr<control_block> clone() const = 0;
};
template <class T>
class polymorphic_value {
std::unique_ptr<control_block<T>> cb_;
T* ptr_ = nullptr;
public:
polymorphic_value() = default;
polymorphic_value(const polymorphic_value& p) :
cb_(p.cb_->clone())
{
ptr_ = cb_->ptr();
}
T* operator->() { return ptr_; }
const T* operator->() const { return ptr_; }
T& operator*() { return *ptr_; }
const T& operator*() const { return *ptr_; }
// Some methods omitted/deferred.
};
Specializations of the control block allow other constructors to be defined.
Motivation and design is discussed here :
https://github.com/jbcoe/polymorphic_value/blob/master/talks/2017_1_25_cxx_london.md
and here
https://github.com/jbcoe/polymorphic_value/blob/master/draft.md
A full implementation with tests can be found here:
https://github.com/jbcoe/polymorphic_value
It's hard to know what you're trying to achieve here, but at first guess it seems that the (upcoming) Boost Type Erasure library might be suitable?
any<
mpl::vector<
copy_constructible<>,
typeid_<>,
incrementable<>,
ostreamable<>
>
> x(10);
++x;
std::cout << x << std::endl; // prints 11
(Example from docs).
Yes, it is possible, but of course there must be some hidden pointer, and the actual data must be stored on the heap. The reason is that the actual size of the data cannot be known at compile-time, and then can't be on the stack.
The idea is to store the actual implementation through a pointer of a polymorphic class ValueImpl, that provides any virtual method you need, like increments() or print(), and in addition a method clone(), so that your class Data is able to implement the value semantics:
class ValueImpl
{
public:
virtual ~ValueImpl() {};
virtual std::unique_ptr<ValueImpl> clone() const { return new ValueImpl(); }
virtual void increments() {}
virtual void print() const { std::cout << "VoidValue "; }
};
class Value
{
private:
ValueImpl * p_; // The underlying pointer
public:
// Default constructor, allocating a "void" value
Value() : p_(new ValueImpl) {}
// Construct a Value given an actual implementation:
// This allocates memory on the heap, hidden in clone()
// This memory is automatically deallocated by unique_ptr
Value(const ValueImpl & derived) : p_(derived.clone()) {}
// Destruct the data (unique_ptr automatically deallocates the memory)
~Value() {}
// Copy constructor and assignment operator:
// Implements a value semantics by allocating new memory
Value(const Value & other) : p_(other.p_->clone()) {}
Value & operator=(const Value & other)
{
if(&other != this)
{
p_ = std::move(other.p_->clone());
}
return *this;
}
// Custom "polymorphic" methods
void increments() { p_->increments(); }
void print() { p_->print(); }
};
The contained pointer is stored inside a C++11 std::unique_ptr<ValueImpl> to ensure the memory is released when destroyed or assigned a new value.
The derived implementations can finally be defined the following way:
class IntValue : public ValueImpl
{
public:
IntValue(int k) : k_(k) {}
std::unique_ptr<IntValue> clone() const
{
return std::unique_ptr<IntValue>(new IntValue(k_));
}
void increments() { k_++; }
void print() const { std::cout << "Int(" << k_ << ") "; }
private:
int k_;
};
class DoubleValue : public ValueImpl
{
public:
DoubleValue(double x) : x_(x) {}
std::unique_ptr<DoubleValue> clone() const
{
return std::unique_ptr<DoubleValue>(new DoubleValue(k_));
}
void increments() { x_ += 1.0; }
void print() const { std::cout << "Double(" << x_ << ") "; }
private:
int x_;
};
Which is enough to make the code snippet in the question works without any modification. This provides run-time polymorphism with value semantics, instead of the traditional run-time polymorphism with pointer semantics provided built-in by the C++ language. In fact, the concept of polymorphism (handling generic objects that behave differently according to their true "type") is independent from the concept of pointers (being able to share memory and optimize function calls by using the address of an object), and IMHO it is more for implementation details that polymorphism is only provided via pointers in C++. The code above is a work-around to take advantage of polymorphism when using pointers is not "philosophically required", and hence ease memory management.
Note: Thanks for CaptainObvlious for the contribution and his evolved code available here that I partially integrated. Not integrated are:
To ease the creation of derived implementations, you may want to create an intermediate templated class
You may prefer to use an abstract interface instead of my non-abstract base class

How to wrap calls of every member function of a class in C++11?

Herb Sutter asked this question in a talk about C++11 and concurrency (See this video)
The key idea here is to have a non locking class X where every function call should be decorated with a lock that is unlocked after a function.
However, Herb Sutter drifts then off and presents a functor based approach. I'm wondering if it is even possible with C++11 to wrap each function call with lock and unlock of a class in a generic way (not wrapping every function call manually).
class X {
public:
X() = default;
void somefunc(arg1 x1, arg2 x2, ...);
void somefunc2(arg1 x1, arg2 x2, ...);
/* and more */
};
// herb admits one way to make all functions *available*
// in another class is by derivation
class XX : public X {
public:
XX() = default;
// all functions available in NON overloaded form...
};
there is also the decorator pattern
class XXX {
public:
XXX(X &x) : m_x(x) {}
// explicitly call each wrapped function ... done for each class separately.
void somefunc(arg1 x1, arg2 x2, ...);
void somefunc2(arg1 x1, arg2 x2, ...);
private:
class X& m_x;
};
but is there something like this possible:
template<>
class wrap_everything;
wrap_everything<X> x;
x.somefunc(x1,x2,...); // this is then locked.
for the sake of completeness this is herb sutter's functor based approach:
template <class T> class locker {
private:
mutable T m_t;
mutable std::mutex m_m;
public:
locker( T t = T{} ) : m_t(t) {}
template <typename F>
auto operator()(F f) const -> decltype(f(m_t)) {
std::lock_guard<mutex> _{m_m};
return f(t);
}
};
// usage
locker<std::string> s;
s([](string &s) {
s += "foobar";
s += "barfoo";
});
The question is about the EXECUTE-AROUND pattern. I made a generic (but only barely tested) implementation of EXECUTE-AROUND POINTER at https://gitlab.com/redistd/redistd/blob/master/include/redi/exec_around.h
This allows:
struct X { void f() { } };
auto x = mutex_around<X>();
x->f(); // locks a mutex for duration of call to X::f
A more in depth explaination on how the family of execute around patterns work can be found here (pdf)
I don't believe there is a portable generic way to do this in current C++. If templates were capable of taking an overload set as a template parameter (which I'd very much like to see in C++14 for many reasons), and the call site could be changed from x.y(z) to x->y(z), I think it could probably be done with a proxy and an overloaded operator->. Otherwise, the best generic way of doing something like this is using Aspect Oriented Programming frameworks for C++ (such as AspectC++).
Being able to wrap each member function call is only really half the story on this, though. According to the Interface Principle, the interface of a class is the functions that mention a class and are supplied with a class. This includes public member functions, friend functions, and free functions in the same namespace as the class. Being able to pass instances to such functions in a wrapped way is a much more subtle problem than merely wrapping member function calls, which is where Sutter's approach shows real power and flexibility.
It is not possible to do exactly what you want, but something close is doable.
#include <iostream>
class Foo {
public:
void one (int x) {
std::cout << "Called Foo::one(" << x << ")\n";
}
void two (int x, double y) {
std::cout << "Called Foo::two(" << x << ", " << y << ")\n";
}
};
class ScopeDecorator {
public:
ScopeDecorator() {
std::cout << "Enter scope\n";
}
~ScopeDecorator() {
std::cout << "Exit scope\n";
}
};
template <class Wrappee, class Wrapper>
class Wrap {
public:
Wrap (Wrappee& w) : wrappee(w) {}
template <typename rettype, typename... argtype>
rettype call (rettype (Wrappee::*func)(argtype...), argtype... args)
{
Wrapper wrapper;
return (wrappee.*func)(args...);
}
private:
Wrappee& wrappee;
};
int main ()
{
Foo foo;
Wrap<Foo, ScopeDecorator> wfoo(foo);
wfoo.call(&Foo::one, 42);
wfoo.call(&Foo::two, 32, 3.1415);
}
For anyone interested, I also wrote a generic implementation of the execute around idom:
https://github.com/ArnaudBienner/ExecuteAround
https://github.com/ArnaudBienner/ExecuteAround/blob/master/ExecuteAround.h
with an example on how to make a thread safe object from it:
https://github.com/ArnaudBienner/ExecuteAround/blob/master/main.cpp#L78
Just for the record, since the one provided by Jonathan already looks great, and mine probably needs some clean up.
It is entirely possible and it was proposed long back by none other than Stroustrup and his original proposal is still available. See www.stroustrup.com/wrapper.pdf
Basically the idea is to override operator -> at 2 levels and lock/unlock mutex in the constructor and destructor of temporary object that is returned by first operator ->.
The second operator -> will return the pointer of object on which the method will be invoked.