Why can not I use new [ ] with smart_pointers?
Actually I can not understand this piece of text.
Caution You should use an auto_prt or shared_ptr object only for
memory allocated by new, not for memory allocated by new []. You
should not use auto_ptr, shared_ptr,orunique_ptr for memory not
allocated via new or, in the case of unique_ptr, via new or new[].
Why can not I use new[] with smart pointers?
In general you can, but that smart pointer must be aware of the fact that it stores a dynamically allocated array, not a single object. This is because objects allocated with operator new[] should be deallocated with operator delete[], not delete. How can a smart pointer know which operator should be applied?
The distinction is made by providing a specialization of smart pointer class templates for array types, like it is currently done in std::unique_ptr<T>:
std::unique_ptr<int> ptr(new int); // will call delete
std::unique_ptr<int[]> arr(new int[5]); // will call delete[]
↑↑
DEMO
However, that syntax does not (yet) apply to all smart pointer types available in the Standard Library.
For comparison, the Boost Smart Pointers library provides separate class templates for storing pointers to dynamically allocated arrays:
boost::shared_array<int> arr1(new int[5]); // will call delete[]
// ~~~~^
boost::scoped_array<int> arr2(new int[5]); // will call delete[]
// ~~~~^
DEMO 2
You should use an auto_ptr or shared_ptr object only for memory allocated by new, not for memory allocated by new [].
std::auto_ptr<T>(† 2017)1 applies a plain delete operator to a pointer it stores, and there is no way to change that behavior. As such, storing a pointer to an array is not an option.
As far as std::shared_ptr<T> is concerned, by default it does the same (calls operator delete). To change that behavior, and properly deallocate a memory area of a stored array, you could use a custom deleter, like std::default_delete<T[]>:
std::shared_ptr<int> arr(new int[5], std::default_delete<int[]>{});
↑↑
or some other provided by yourself:
std::shared_ptr<int> arr(new int[5], [](int* ptr) { delete[] ptr; } );
// ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
DEMO 3
However, a missing specialization for std::shared_ptr<T[]> implies no operator[] that could let you easily access the elements of a stored array, which leads to unintuitive syntax like arr.get()[0].
With proposal N4077 introduced, there will be a specialization for array type pointers:
std::shared_ptr<int[]> arr(new int[5]); // will call delete[]
↑↑
You should not use auto_ptr, shared_ptr, or unique_ptr for memory not allocated via new or, in the case of unique_ptr, via new or new[].
This excerpt simply states that one should not construct a smart pointer from a pointer to an object that was not allocated dynamically, as (by default) it would result in calling delete on something that was not allocated with new (ditto new[]/delete[]).
What is the difference between unique_ptr<double[]> p1(new double[2]);, unique_ptr<double> p2(new double[2]);, unique_ptr<double[]> p3(new double(2)); ?
std::unique_ptr<double[]> p1(new double[2]);
OK: Constructs a unique_ptr from (and takes ownership of) a pointer to an array of two doubles. It will call delete[] to deallocate the memory pointed.
std::unique_ptr<double> p2(new double[2]);
Wrong: Constructs a unique_ptr from (and takes ownership of) a pointer to an array of two doubles. It will call delete (!) to deallocate the memory pointed. (possibly undefined behavior - a mismatch between new[] and delete).
std::unique_ptr<double[]> p3(new double(2));
Wrong: Constructs a unique_ptr from (and takes ownership of) a pointer to a single double initialized to value 2. It will call delete[] (!) to deallocate the memory pointed. (possibly undefined behavior - a mismatch between new and delete[]).
1 std::auto_ptr<T> is deemed deprecated in favor of std::unique_ptr<T> since C++11, and will be removed from the Standard Library in C++1z according to N4168.
Examples:
#include <memory>
int
main()
{
auto p1 = std::unique_ptr<char[]>{new char[3]}; // C++11
auto p2 = std::shared_ptr<char>{new char[3], [](char* p) {delete [] p;}}; // C++11
auto p3 = std::make_unique<char[]>(3); // C++14
}
The first and second are good for C++11 and forward. The 3rd was introduced in C++14. The first and third represent unique ownership of the new, and the second has shared ownership of the new.
Related
Consider the following class:
class Vector{
int dim; //dimension of array v
Complex* v; //Complex is another class
public:
Vector(int dim = 0):dim(dim){(dim)?(v=new Complex[dim]):(v=nullptr);}
Vector(int dim, const Complex* c):dim(dim),v(new Complex[dim]){
for(int i=0;i<dim;i++) v[i]=c[i];}
Vector(const Vector& a):dim(a.dim),v(new Complex[a.dim]){
for(int i=0;i<dim;i++) v[i]=a.v[i];}
~Vector(){if(dim)delete [] v,v=nullptr;}
friend Vector& operator >> (Vector& is,Complex& z){
Vector copie(is);
is.~Vector();
is.Vector::Vector(is.dim+1);}
};
I try to overload the >> operator in order to add elements to v.
My idea was to create a copy, then call dctor and the ctor for the object to
be modified via >> operator.
I'm stuck after getting this error:
In function ‘Vector& operator>>(Vector&, Complex&)’:
main.cc:56:20: error: cannot call constructor ‘Vector::Vector’ directly
is.Vector::Vector(is.dim+1);
I'm not allowed to use containers!!
Please help me!
That's right, you can't call the constructor directly. Probably you want to use placement new.
friend Vector& operator >> (Vector& is,Complex& z){
Vector copie(is);
is.~Vector();
// is.Vector::Vector(is.dim+1);
new(&is) Vector(is.dim + 1);
return is;
}
Even then the code may not be semantically correct.
Having said that, this is not the recommended way to do it for
the last 20 years. Watch this Jon Kalb "Exception-Safe Code, Part
I" for an explanation (the example is almost the same). The
recommended way is to implement this in terms of other operations like
copy or swap.
Minor syntactic detail, operator>> is confusing, use operator<< at worst.
There is no need for calling the destructor and calling the constructor. Steps you can take to make your function work:
Allocate memory to hold the current objects plus the additional object.
Copy the objects from the old memory location to the new memory location.
Delete the old memory.
Associate the newly allocated memory with the input object.
friend Vector& operator>>(Vector& is, Complex& z){
// Allocate memory
Complex* vnew = new Complex[dim+1];
// Copy objects to new memory.
std::copy(is.v, is.v + is.dim, vnew);
vnew[is.dim] = z;
// Delete the old memory.
delete [] is.v;
// Use the new memory
is.v = vnew;
// Increment dim.
is.dim++;
return is;
}
Having said that, I think you are using the wrong function to insert an element to Vector. operator>> is for extracting data from. operator<< is for inserting data to. You should use operator<< to insert an element to a Vector.
friend Vector& operator<<(Vector& is, Complex const& z){
...
}
I am trying to write my own Allocator which can be used in STL. That far I could almost successfully finish but with one function I have my problem:
The Allocator used in STL should provide the function construct [example from a standard allocator using new & delete]:
// initialize elements of allocated storage p with value value
void construct (T* p, const T& value)
{
::new((void*)p)T(value);
}
I am stuck how to rewrite this using my own function which replaces the new keyword initializing it with the value.
This function construct is for example used in this code: fileLines.push_back( fileLine );
where
MyVector<MyString> fileLines;
MyString fileLine;
These are my typedefs where I use my own Allocator:
template <typename T> using MyVector = std::vector<T, Allocator<T>>;
using MyString = std::basic_string<char, std::char_traits<char>, Allocator<char>>;
I am confused because here is allocated a pointer to T which can be for example [when I understood it correctly] MySstring.
Do I understand it correctly that the pointer - allocated by new - will have 10 bytes, when value is 123456789 and then the provided value is copied to the new pointer?
My question:
How to rewrite the one line of code using my own function? For me the difficult point is how to get the length of value [which can have any type] in order I can correctly determinate the length of the allocated block and how to copy it in order it works for all possible types T?
The new operator in the construct function does not allocate anything at all, it's a placement new call, which takes an already allocated chunk of memory (which needs to have been previously allocated some way, and at least as large as sizeof(T)) and initializes it as a T object by calling the constructor of T, pretending that the memory pointed to by p is a T object.
::new int(7) calls the default new operator, gets some memory big enough for an int, and constructs an int with the value 7 in it.
::new(ptr) int(7) takes a void* called ptr, and constructs an int with the value 7 in it. This is called "placement new".
The important part is what is missing from the second paragraph. It does not create space, but rather constructs an object in some existing space.
It is like saying ptr->T::T() ptr->int::int(7) where we "call the constructorofinton the memory location ofptr, exceptptr->T::T()orptr->int::int(7)are not valid syntaxes. Placementnew` is the way to explicitly call a constructor in C++.
Similarly, ptr->~T() or ptr->~int() will call the destructor on the object located at ptr (however, there is no ~int so that is an error, unless int is a template or dependent type, in which case it is a pseudo-destructor and the call it ignored instead of generating an error).
It is very rare that you want to change construct from the default implementation in an allocator. You might do this if your allocator wants to track creation of objects and attach the information about their arguments, without intrusively modifying the constructor. But this is a corner case.
I have the following member variable in a class:
std::vector<std::unique_ptr<Object>> objects_;
I explicitly want the vector to maintain ownership at all times. I've seen suggestions that in order for a member function to access a pointer in the vector and make changes to the object T wrapped in the std::unique_ptr, we must move the pointer to calling code, i.e:
void foo(int i) {
auto object = std::move( vector.at( i ) ); // move object to caller (caller owns)
object->dosomething();
vector.at(i) = std::move(object); // move back into vector (vector owns)
}
Another method was to work with raw pointers:
void foo(int i) {
Object* object = vector.at( i ).get();
object->doSomething();
}
However, I've been working with this:
void foo(int i) {
auto& object = vector.at( i );
object->doSomething();
}
Which is the correct and most robust method for my case? Does this function ever take ownership of the data in the std::unique_ptr? Is there a way to access Object without playing with the std::unique_ptr?
(excuse me if my methods are incorrect, I hope I got the point across)
The first approach will not retain ownership of the object if object->dosomething() throws an exception (i.e. it is not exception safe) since the second std::move() statement will not be executed.
Assuming C++11, both of the other approaches are effectively equivalent, subject to the assumption that the owned pointer is not null. Under the same assumption, the code can be simplified to
void foo(int i)
{
vector.at(i)->doSomething();
}
which will work with all C++ standards (not just C++11 or later).
It is possible to access the object without monkeying with the unique_ptr - simply store the pointer elsewhere and use that. However, that does compromise the purpose of using std::unique_ptr in the first place. And is error-prone - for example, the std::unique_ptr can destroy the object, and leave those other pointers dangling.
If you are really that worried about the potential of your vector losing ownership, consider using a shared_ptr instead.
Problem:
Let's assume I have an algorithm that takes a unique_ptr to some type:
void FancyAlgo(unique_ptr<SomeType>& ptr);
Now I have shared_ptr sPtr to SomeType, and I need to apply the same algorithm on sPtr. Does this mean I have to duplicate the algorithm just for the shared_ptr?
void FancyAlgo(shared_ptr<SomeType>& sPtr);
I know smart pointers come with ownership of the underlying managed object on the heap. Here in my FancyAlgo, ownership is usually not an issue. I thought about stripping off the smart pointer layer and do something like:
void FancyAlgo(SomeType& value);
and when I need to call it with unique_ptr:
FancyAlgo(*ptr);
likewise for shared_ptr.
1, Is this an acceptable style in PRODUCTION code?(I saw somewhere that in a context of smart pointers, you should NOT manipulate raw pointers in a similar way. It has the danger of introducing mysterious bugs.)
2, Can you suggest any better way (without code duplication) if 1 is not a good idea.
Thanks.
Smart pointers are about ownership. Asking for a smart pointer is asking for ownership information or control.
Asking for a non-const lvalue reference to a smart pointer is asking for permission to change the ownership status of that value.
Asking for a const lvalue reference to a smart pointer is asking for permission to query the ownership status of that value.
Asking for an rvalue reference to a smart pointer is being a "sink", and promising to take that ownership away from the caller.
Asking for a const rvalue reference is a bad idea.
If you are accessing the pointed to value, and you want it to be non-nullable, a reference to the underlying type is good.
If you want it to be nullable, a boost::optional<T&> or a T* are acceptable, as is the std::experimental "world's dumbest smart pointer" (or an equivalent hand-written one). All of these are non-owning nullable references to some variable.
In an interface, don't ask for things you don't need and won't need in the future. That makes reasoning about what the function does harder, and leads to problems like you have in the OP. A function that reseats a reference is a very different function from one that reads a value.
Now, a more interesting question based off yours is one where you want the function to reseat the smart pointer, but you want to be able to do it to both shared and unique pointer inputs. This is sort of a strange case, but I could imagine writing a type-erase-down-to-emplace type (a emplace_sink<T>).
template<class T>
using later_ctor = std::function<T*(void*)>;
template<class T, class...Args>
later_ctor<T> delayed_emplace(Args&&...args) {
// relies on C++1z lambda reference reference binding, write manually
// if that doesn't get in, or don't want to rely on it:
return [&](void* ptr)->T* {
return new T(ptr)(std::forward<Args>(args));
};
}
namespace details {
template<class T>
struct emplace_target {
virtual ~emplace_target() {}
virtual T* emplace( later_ctor<T> ctor ) = 0;
};
}
template<class T>
struct emplacer {
std::unique_ptr<emplace_target<T>> pImpl;
template<class...Args>
T* emplace( Args&&... args ) {
return pImpl->emplace( delayed_emplace<T>(std::forward<Args>(args)...) );
}
template<class D>
emplacer( std::shared_ptr<T, D>& target ):
pImpl( new details::emplace_shared_ptr<T,D>(&target) ) // TODO
{}
template<class D>
emplacer( std::unique_ptr<T, D>& target ):
pImpl( new details::emplace_unique_ptr<T,D>(&target) ) // TODO
{}
};
etc. Lots of polish needed. The idea is to type-erase construction of an object T into an arbitrary context. We might need to special case shared_ptr so we can call make_shared, as a void*->T* delayed ctor is not good enough to pull that off (not fundamentally, but because of lack of API hooks).
Aha! I can do a make shared shared ptr without special casing it much.
We allocate a block of memory (char[sizeof(T)]) with a destructor that converts the buffer to T then calls delete, in-place construct in that buffer (getting the T*), then convert to a shared_ptr<T> via the shared_ptr<T>( shared_ptr<char[sizeof(T)]>, T* ) constructor. With careful exception catching this should be safe, and we can emplace using our emplacement function into a make_shared combined buffer.
I am trying to use a FILE pointer multiple times through out my application
for this I though I create a function and pass the pointer through that. Basically I have this bit of code
FILE* fp;
_wfopen_s (&fp, L"ftest.txt", L"r");
_setmode (_fileno(fp), _O_U8TEXT);
wifstream file(fp);
which is repeated and now instead I want to have something like this:
wifstream file(SetFilePointer(L"ftest.txt",L"r"));
....
wofstream output(SetFilePointer(L"flist.txt",L"w"));
and for the function :
FILE* SetFilePointer(const wchar_t* filePath, const wchar_t * openMode)
{
shared_ptr<FILE> fp = make_shared<FILE>();
_wfopen_s (fp.get(), L"ftest.txt", L"r");
_setmode (_fileno(fp.get()), _O_U8TEXT);
return fp.get();
}
this doesn't simply work. I tried using &*fp instead of fp.get() but still no luck.
You aren't supposed to create FILE instances with new and destroy them with delete, like make_shared does. Instead, FILEs are created with fopen (or in this case, _wfopen_s) and destroyed with fclose. These functions do the allocating and deallocating internally using some unspecified means.
Note that _wfopen_s does not take a pointer but a pointer to pointer - it changes the pointer you gave it to point to the new FILE object it allocates. You cannot get the address of the pointer contained in shared_ptr to form a pointer-to-pointer to it, and this is a very good thing - it would horribly break the ownership semantics of shared_ptr and lead to memory leaks or worse.
However, you can use shared_ptr to manage arbitrary "handle"-like types, as it can take a custom deleter object or function:
FILE* tmp;
shared_ptr<FILE> fp;
if(_wfopen_s(&tmp, L"ftest.txt", L"r") == 0) {
// Note that we use the shared_ptr constructor, not make_shared
fp = shared_ptr<FILE>(tmp, std::fclose);
} else {
// Remember to handle errors somehow!
}
Please do take a look at the link #KerrekSB gave, it covers this same idea with more detail.