I am working on design a wrapper class to provide RAII function.
The original use case is as follows:
void* tid(NULL);
OpenFunc(&tid);
CloseFunc(&tid);
After I introduce a new wrapper class, I expect the future usage will be as follows:
void* tid(NULL);
TTTA(tid);
or
TTTB(tid);
Question:
Which implementation TTTA or TTTB is better? Or they are all bad and please introduce a better one.
One thing I have concern is that after the resource is allocated, the id will be accessed outside of class TTTA or TTTB until the id is destroyed. Based on my understanding, my design should not have side-effect for that.
Thank you
class TTTA : boost::noncopyable
{
public:
explicit TTTA(void *id)
: m_id(id)
{
OpenFunc(&m_id); // third-party allocate resource API
}
~TTTA()
{
CloseFunc(&m_id); // third-party release resource API
}
private:
void* &m_id; // have to store the value in order to release in destructor
}
class TTTB : boost::noncopyable
{
public:
explicit TTTB(void *id)
: m_id(&id)
{
OpenFunc(m_id); // third-party allocate resource API
}
~TTTB()
{
CloseFunc(m_id); // third-party release resource API
}
private:
void** m_id; // have to store the value in order to release in destructor
}
// pass-in pointers comparison
class TTTD
{
public:
TTTD(int* id) // Take as reference, do not copy to stack.
: m_id(&id)
{
*m_id = new int(40);
}
private:
int** m_id;
};
class TTTC
{
public:
TTTC(int* &id)
: m_id(id)
{
m_id = new int(30);
}
private:
int* &m_id;
};
class TTTB
{
public:
TTTB(int* id)
: m_id(id)
{
m_id = new int(20);
}
private:
int* &m_id;
};
class TTTA
{
public:
TTTA(int** id)
: m_id(id)
{
*m_id = new int(10);
}
private:
int** m_id;
};
int main()
{
//////////////////////////////////////////////////////////////////////////
int *pA(NULL);
TTTA a(&pA);
cout << *pA << endl; // 10
//////////////////////////////////////////////////////////////////////////
int *pB(NULL);
TTTB b(pB);
//cout << *pB << endl; // wrong
//////////////////////////////////////////////////////////////////////////
int *pC(NULL);
TTTC c(pC);
cout << *pC << endl; // 30
//////////////////////////////////////////////////////////////////////////
int *pD(NULL);
TTTD d(pD);
cout << *pD << endl; // wrong
}
Both break in bad ways.
TTTA stores a reference to a variable (the parameter id) that's stored on the stack.
TTTB stores a pointer to a variable that's stored on the stack.
Both times, the variable goes out of scope when the constructor returns.
EDIT: Since you want the values modifiable, the simplest fix is to take the pointer as a reference; that will make TTTC reference the actual pointer instead of the local copy made when taking the pointer as a non reference parameter;
class TTTC : boost::noncopyable
{
public:
explicit TTTA(void *&id) // Take as reference, do not copy to stack.
: m_id(id)
...
private:
void* &m_id; // have to store the value in order to release in destructor
}
The simple test that breaks your versions is to add a print method to the classes to print the pointer value and do;
int main() {
void* a = (void*)0x200;
void* b = (void*)0x300;
{
TTTA ta(a);
TTTA tb(b);
ta.print();
tb.print();
}
}
Both TTTA and TTTB print both values as 0x300 on my machine. Of course, the result is really UB; so your result may vary.
Why do you tid at all? It’s leaking information to the client and makes the usage twice as long (two lines instead of one):
class tttc {
void* id;
public:
tttc() {
OpenFunc(&id);
}
~tttc() {
CloseFunc(&id);
}
tttc(tttc const&) = delete;
tttc& operator =(tttc const&) = delete;
};
Note that this class forbids copying – your solutions break the rule of three.
If you require access to id from the outside, provide a conversion inside tttc:
void* get() const { return id; }
Or, if absolutely necessary, via an implicit conversion:
operator void*() const { return id; }
(But use that one judiciously since implicit conversions weaken the type system and may lead to hard to diagnose bugs.)
And then there’s std::unique_ptr in the standard library which, with a custom deleter, actually achieves the same and additionally implements the rule of three properly.
What about wrapping it completely? This way you do not have to worry about managing the lifecycles of two variables, but only one.
class TTTC
{
void* m_id;
public:
TTTC()
: m_id(nullptr)
{
OpenFunc(&m_id); // third-party allocate resource API
}
TTTC(TTTC const&) = delete; // or ensure copying does what you expect
void*const& tid() const { return m_id; }
~TTTC()
{
CloseFunc(&m_id); // third-party release resource API
}
};
Using it is simplicity itself:
TTTC wrapped;
DoSomethingWithTid(wrapped.tid());
Related
I can't figure out why I get error for the code below.
The instances of object A will be pushed into a vector (vectorA.push_back(A a)) continuously. So sometimes, vectorA needs to be reallocated; the destructor will be called, which is where the destructor of A gets called, then the error message appears.
class A
{
long filePos;
union {
Recording* recording;
UINT64 timeStamp;
};
public:
inline A(long fpos, UINT64 ts) : filePos(fpos), timeStamp(ts) {}
~A()
{
if (getDetailedType() == RECORDING_TYPE)
if (recording)
delete recording; // error: scalar deleting destructor ???
}
inline short getDetailedType() const { return (short)(timeStamp % 5); }
A(const A& edi)
{
filePos = edi.filePos;
if (getDetailedType() == RECORDING_INFO)
recording = edi.recording;
else
timeStamp = edi.timeStamp;
}
}
class Recording : protected RECORDINGS
{
UINT64 timestamp;
float scalar;
public:
~Recording() // with or without this dtor, got same error
{
}
inline Recording()
{
timestamp = 0;
scalar = 2.0;
time = 0;
rate = 30;
type = 1;
side = 1;
}
}
typedef struct
{
UINT32 time;
float rate;
int type;
int side;
} RECORDINGS;
Your copy constructor does a shallow copy. So, now you have two objects that both have the same recording pointer.
You should either do a deep copy, or ensure the ownership is properly transferred (by using something like std::unique_ptr<Recording> if C++11 is available.
See This question on the difference between deep and shallow copies.
Let's look at some examples:
class ABadCopyingClass
{
public:
ABadCopyingClass()
{
a_ = new int(5);
}
~ABadCopyingClass()
{
delete a_;
}
private:
int* a_;
};
The above class is bad because the default copy constructor and assignment operator will perform a shallow copy, and lead to two objects both thinking that they own the underlying a_ object. When one of them goes out of scope, the a_ will be deleted, and the other one will be left with a dangling pointer that will eventually lead to a crash.
class ABetterCopyingClass
{
public:
ABetterCopyingClass()
a_(new int(5))
{
}
ABetterCopyingClass(const ABetterCopyingClass& r)
{
a_ = new int(*r.a_);
}
ABetterCopyingClass& operator=(const ABetterCopyingClass& r)
{
// in the case of reassignment...
delete a_;
a_ = new int(*r.a_);
return *this;
}
~ABetterCopyingClass()
{
delete a_;
}
private:
int* a_;
};
This class improved our situation a little (note, that the normal error checking is left out in this simple example). Now the copy constructor and assignment operator properly perform the necessary deep copying. The drawback here is the amount of boilerplate code we had to add -- it's easy to get that wrong.
class ACannonicalCopyingClass
{
public:
ACannonicalCopyingClass()
: a_(new int(5))
{
}
ACannonicalCopyingClass(ACannonicalCopyingClass&& moved_from)
{
a_ = std::move(moved_from.a_);
}
private:
std::unique_ptr<int> a_;
};
This example (C++11 only) is even better. We've removed a significant amount of boilerplate code, however the semantics here are a bit different. Instead of deep copying, we get in this case transfer of ownership of the underlying a_ object.
The easiest version (C++11 only) to implement is the version that provides shared ownership of the underlying a_ object. This is the version that is most similar to your provided example, with the added bonus that it does not cause a crash.
class ASharedCopyingClass
{
public:
ASharedCopyingClass()
: a_(std::make_shared<int>(5))
{
}
private:
std::shared_ptr<int> a_;
};
This version can be copied at will, and the underlying a_ object will happily be reference counted. The last copy to go out of scope will set the reference count to 0, which will trigger the memory deallocation.
My psychic debugging skills tell me that you forgot to implement a copy constructor for A which then results in a double deletion of a Recording when the copy is destroyed.
The growth of the vector would trigger the copy-destroy pairs.
I have this class:
class CComputer {
public:
// constructor
CComputer(string name) {
this->name = name;
};
// overloaded operator << for printing
friend ostream& operator<<(ostream& os, const CComputer& c);
// adds some component for this computer
CComputer & AddComponent(Component const & component) {
this->listOfComponents.push_back(component);
return *this;
};
// sets address for this computer
CComputer & AddAddress(const string & address) {
this->address = address;
return *this;
};
string name;
string address;
list<Component> listOfComponents;
};
and then these classes:
// ancestor for other classes...It's really dummy yet, but I dunno what to add there
class Component {
public:
Component() {};
~Component() {};
};
class CCPU : public Component {
public:
CCPU(int cores, int freq) {
this->cores = cores;
this->freq = freq;
};
int cores;
int freq;
};
class CMemory : public Component {
public:
CMemory(int mem) {
this->mem = mem;
};
int mem;
};
Now I feed my CComputer class with some values:
CComputer c("test.com");
c . AddAddress("123.45.678.910") .
AddComponent(CCPU(8, 2400)) .
AddComponent(CCPU(8, 1200)).
AddComponent(CMemory(2000)).
AddComponent(CMemory(2000)));
And now I would like to print it out with all the info I've put in there (CCPU & CMemory details including)
but how to implement it, to be able to iterate through CComputer::listOfComponents and don't care if I acctually access CCPU or CMemory ? I can add it to that list, but I have really no idea, how to make it, to be able to access the variables of those components.
So the output should look like:
##### STARTING #####
CComputer:
name:test.com
address:123.45.678.910
CCPU:
cores:8,freq:2400
CCPU:
cores:8, freq:1200
CMemory:
mem:2000
CMemory:
mem:2000
###### FINISHED! #####
As others have mentioned, you need to implement a virtual function (e.g. virtual std::string ToString() const = 0;) in the base class that is inherited and overridden by each child class.
However, that isn’t enough. Your code exhibits slicing which happens when you copy your child class instances into the list: the list contains objects of type Component, not of the relevant child class.
What you need to do is store polymorphic instances. Values themselves are never polymorphic, you need to use (smart) pointers or references for this. References are out, however, since you cannot store them in a standard container (such as std::list). Using raw pointers is considered bad style nowadays, but judging from the naming conventions of your classes you don’t learn modern C++ in your class (sorry!).
Therefore, raw pointers is probably the way to go. Change your code accordingly:
Store a list of pointers:
list<Component*> listOfComponents;
Make the argument type of AddComponent a pointer instead of const&.
Call the function by passing a newed object, e.g.:
AddComponent(new CCPU(8, 2400))
Now your code leaks memory left, right and center. You need to implement a destructor to free the memory:
~CComputer() {
typedef std::list<Component*>::iterator iter_t;
for (iter_t i = listOfComponents.begin(); i != listOfComponents.end(); ++i)
delete *i;
}
But now your code violates the Rule of Three (read this article! It’s important, and it may be the most useful thing about C++ you’re going to learn in this programming class) and consequently you also need to implement the copy constructor and copy assignment operator. However, we can’t. Sorry. In order to implement copying for your class, you would have to implement another virtual function in your Component class, namely one that clones an object (virtual Component* Clone() const = 0;). Only then can we proceed.
Here’s a sample implementation in CCPU:
Component* Clone() const {
return new CCPU(cores, freq);
}
… this needs to be done in all classes deriving from Component, otherwise we cannot correctly copy an object of a type that derives from Component and is hidden behind a pointer.
And now we can implement copying in the CComputer class:
CComputer(CComputer const& other)
: name(name)
, address(addess) {
typedef std::list<Component*>::iterator iter_t;
for (iter_t i = other.listOfComponents.begin(); i != other.listOfComponents.end(); ++i)
listOfComponents.push_back((*i)->Clone());
}
CComputer& operator =(CComputer const& other) {
if (this == &other)
return *this;
name = other.name;
address = other.address;
listOfComponents.clear();
for (iter_t i = other.listOfComponents.begin(); i != other.listOfComponents.end(); ++i)
listOfComponents.push_back((*i)->Clone());
return *this;
}
This code is brittle, not thread-safe and error-prone and no competent C++ programmer would ever write this1. Real code would for instance use smart pointers instead – but as mentioned before I’m pretty sure that this would be beyond the scope of the class.
1 What does this make me now, I wonder?
Just add a virtual method to Class Component called e.g. toString(), which returns a string describing the component. Then you can iterate through all components and call toString() without worrying about exactly what each component is. If you do that, then for each computer you would be able to print out the values of all the components.
However, as pointed out in one of the comments, the example output you give in the question outputs the CCPU for all computers, then all the memory for all computers. To order the output like that, you'll need to add another virtual method to Component called e.g. getType() which returns an enum or integer that represents the type of the information. You can then have two for-next loops, one nested inside the other, where the outer loop iterates through all the types and the inner loop iterating through all the computers calling the toString() on all components which match the type specified in the outer for loop.
Here's something that implements this idea.
#include <iostream>
#include <string>
#include <list>
using namespace std;
int const TYPE_CCPU = 1;
int const TYPE_MEMORY = 2;
class Component {
public:
virtual int GetType() { return -1; }
virtual std::string ToString() const {
return "OOPS! Default `ToString` called";
}
};
class CComputer {
public:
typedef std::list<Component*>::iterator iter_t;
// constructor
CComputer(string name) {
this->name = name;
};
~CComputer() {
for (iter_t i = listOfComponents.begin(); i != listOfComponents.end(); ++i) {
delete *i;
}
}
// overloaded operator << for printing
friend ostream& operator<<(ostream& os, const CComputer& c);
// adds some component for this computer
CComputer & AddComponent(Component *component) {
this->listOfComponents.push_back(component);
return *this;
};
// sets address for this computer
CComputer & AddAddress(const string & address) {
this->address = address;
return *this;
};
void PrintType(int type) {
for (iter_t i = listOfComponents.begin(); i != listOfComponents.end(); ++i) {
if ((*i)->GetType() == type)
std::cout << (*i)->ToString() << '\n';
}
}
string name;
string address;
list<Component*> listOfComponents;
};
class CCPU : public Component {
public:
CCPU(int cores, int freq) {
this->cores = cores;
this->freq = freq;
};
int GetType() { return TYPE_CCPU; }
std::string ToString() const {
return "CCPU::ToString()";
}
int cores;
int freq;
};
class CMemory : public Component {
public:
CMemory(int mem) { this->mem = mem; };
int GetType() { return TYPE_MEMORY; }
std::string ToString() const {
return "CMemory::ToString()";
}
int mem;
};
typedef std::list<CComputer*>::iterator iter_c;
int main() {
list<CComputer*> computerlist;
CComputer *c1 = new CComputer("test.com"), *c2 = new CComputer("test2.com");
c1->AddAddress("123.45.678.910").
AddComponent(new CCPU(8, 1200)).
AddComponent(new CMemory(2000));
computerlist.push_back(c1);
c2->AddAddress("987.65.432.10").
AddComponent(new CCPU(8, 2400)).
AddComponent(new CMemory(4000));
computerlist.push_back(c2);
for(int t=TYPE_CCPU; t<=TYPE_MEMORY; t++)
for (iter_c i = computerlist.begin(); i != computerlist.end(); ++i) {
(*i)->PrintType(t);
}
for (iter_c i = computerlist.begin(); i != computerlist.end(); ++i) {
delete (*i);
}
}
Implement ToString() in each of your classes. In .NET this is a standard even the "object" type implements.
I need to pass a pointer to a class so some code I don't control. This code automatically free()s the pointer when it is done, but I need the class later. I hoped I could just make a 'wrapper' class that would keep the class from being deallocated without actually preventing the code from accessing it, but virtual calls don't work.
template <class T>
class PointerWrapper:public T
{
public:
T* p;
PointerWrapper(T *ptr)
{
p=ptr;
}
~PointerWrapper(void)
{
}
T* operator->() const
{
return p;
}
T& operator*() const
{
return *p;
}
};
void codeIDontControl(Example *ex)
{
ex->virtualfunction();
delete ex;
}
void myCode()
{
Example *ex=new Example();
codeIDontControl(ex);
do something with ex //doesn't work because ex has been freed
codeIDontControl(new PointerWrapper<Example>(ex));
do something with ex //ex hasn't been freed, but the changes made to it via
// Example::virtualfunction() in codeIDontControl() aren't there anymore
}
Basically, ex->virtualfunction() calls the virtual function in PointerWrapper itself instead of the virtual function in PointerWrapper->p. It seems that it's ignoring the -> operator?
Now, I don't need to use a PointerWrapper-esque class if there's a different way to do this, but it was all I could think of...
I can't modify Example either, but I can subclass it
You should provide Forwarder class - which redirects virtual calls to stored pointer. Freeing of forwarder class will not cause releasing of pointee. This approach does NOT need to do copy (which can be expensive/may be not implemented/or even not make sense):
struct Forwarder : Example
{
Example *impl;
Forwarder(Example *i) : impl(i) {}
void virtualfunction()
{
impl->virtualfunction();
}
};
Full code:
live demo:
#include <iostream>
#include <ostream>
using namespace std;
struct Example
{
virtual void virtualfunction()=0;
virtual ~Example() {}
};
struct Implmenetation : Example
{
bool alive;
Implmenetation() : alive(true) {}
void virtualfunction()
{
cout << "Implmenetation::virtualfunction alive=" << alive << endl;
}
~Implmenetation()
{
alive=false;
cout << "Implmenetation::~Implmenetation" << endl;
}
};
struct Forwarder : Example
{
Example *impl;
Forwarder(Example *i) : impl(i) {}
void virtualfunction()
{
impl->virtualfunction();
}
};
void codeIDontControl(Example *ex)
{
ex->virtualfunction();
delete ex;
}
void myCode()
{
Implmenetation impl;
codeIDontControl(new Forwarder(&impl));
//do something with ex //doesn't work because ex has been freed
impl.virtualfunction();
}
int main()
{
myCode();
}
Output is:
Implmenetation::virtualfunction alive=1
Implmenetation::virtualfunction alive=1
Implmenetation::~Implmenetation
It's bad design, really. Only the allocator should be allowed to free memory. Functions like this are dangerous, as they leave with with dangling pointers.
This is just off the top of my head, maybe you could try something like this? It's not a safe idea, but if someone implemented it I would be interested to know what happens.
class Foo
{
Foo(Foo* copy) : m_copy(copy) {}
~Foo() { if(m_copy) { *m_copy = *this; } } // Use copy constructor to create copy on destuction.
Foo* m_copy;
}
Foo copy(NULL);
Foo* original = new Foo(©);
MethodThatDeletes(original);
// Original should be destroyed, and made a copy in the process.
original = NULL;
// Copy should be a copy of the original at it's last know state.
copy;
You are providing a Example* to codeIDontControl. The overloaded operator-> on PointerWrapper is an for the PointerWrapper type not the Example* type or even the PointerWrapper* type (i.e. for a value or reference of that type not a pointer to that type).
Since the function you need to call isn't controlled by you, you will need to provide a complete wrapper of the type it expects as a wrapper over the instance you wish to control the lifetime of.
In C++, the T q = dynamic_cast<T>(p); construction performs a runtime cast of a pointer p to some other pointer type T that must appear in the inheritance hierarchy of the dynamic type of *p in order to succeed. That is all fine and well.
However, it is also possible to perform dynamic_cast<void*>(p), which will simply return a pointer to the "most derived object" (see 5.2.7::7 in C++11). I understand that this feature probably comes out for free in the implementation of the dynamic cast, but is it useful in practice? After all, its return type is at best void*, so what good is this?
The dynamic_cast<void*>() can indeed be used to check for identity, even if dealing with multiple inheritance.
Try this code:
#include <iostream>
class B {
public:
virtual ~B() {}
};
class D1 : public B {
};
class D2 : public B {
};
class DD : public D1, public D2 {
};
namespace {
bool eq(B* b1, B* b2) {
return b1 == b2;
}
bool eqdc(B* b1, B *b2) {
return dynamic_cast<void*>(b1) == dynamic_cast<void*>(b2);
}
};
int
main() {
DD *dd = new DD();
D1 *d1 = dynamic_cast<D1*>(dd);
D2 *d2 = dynamic_cast<D2*>(dd);
std::cout << "eq: " << eq(d1, d2) << ", eqdc: " << eqdc(d1, d2) << "\n";
return 0;
}
Output:
eq: 0, eqdc: 1
Bear in mind that C++ lets you do things the old C way.
Suppose I have some API in which I'm forced to smuggle an object pointer through the type void*, but where the callback it's eventually passed to will know its dynamic type:
struct BaseClass {
typedef void(*callback_type)(void*);
virtual callback_type get_callback(void) = 0;
virtual ~BaseClass() {}
};
struct ActualType: BaseClass {
callback_type get_callback(void) { return my_callback; }
static void my_callback(void *p) {
ActualType *self = static_cast<ActualType*>(p);
...
}
};
void register_callback(BaseClass *p) {
// service.register_listener(p->get_callback(), p); // WRONG!
service.register_listener(p->get_callback(), dynamic_cast<void*>(p));
}
The WRONG! code is wrong because it fails in the presence of multiple inheritance (and isn't guaranteed to work in the absence, either).
Of course, the API isn't very C++-style, and even the "right" code can go wrong if I inherit from ActualType. So I wouldn't claim that this is a brilliant use of dynamic_cast<void*>, but it's a use.
Casting pointers to void* has its importance since way back in C days.
Most suitable place is inside the memory manager of Operating System. It has to store all the pointer and the object of what you create. By storing it in void* they generalize it to store any object on to the memory manager data structure which could be heap/B+Tree or simple arraylist.
For simplicity take example of creating a list of generic items(List contains items of completely different classes). That would be possible only using void*.
standard says that dynamic_cast should return null for illegal type casting and standard also guarantees that any pointer should be able to type cast it to void* and back from it with only exception of function pointers.
Normal application level practical usage is very less for void* typecasting but it is used extensively in low level/embedded systems.
Normally you would want to use reinterpret_cast for low level stuff, like in 8086 it is used to offset pointer of same base to get the address but not restricted to this.
Edit:
Standard says that you can convert any pointer to void* even with dynamic_cast<> but it no where states that you can not convert the void* back to the object.
For most usage, its a one way street but there are some unavoidable usage.
It just says that dynamic_cast<> needs type information for converting it back to the requested type.
There are many API's that require you to pass void* to some object eg. java/Jni Code passes the object as void*.
Without type info you cannot do the casting.If you are confident enough that type requested is correct you can ask compiler to do the dynmaic_cast<> with a trick.
Look at this code:
class Base_Class {public : virtual void dummy() { cout<<"Base\n";} };
class Derived_Class: public Base_Class { int a; public: void dummy() { cout<<"Derived\n";} };
class MostDerivedObject : public Derived_Class {int b; public: void dummy() { cout<<"Most\n";} };
class AnotherMostDerivedObject : public Derived_Class {int c; public: void dummy() { cout<<"AnotherMost\n";} };
int main () {
try {
Base_Class * ptr_a = new Derived_Class;
Base_Class * ptr_b = new MostDerivedObject;
Derived_Class * ptr_c,*ptr_d;
ptr_c = dynamic_cast< Derived_Class *>(ptr_a);
ptr_d = dynamic_cast< Derived_Class *>(ptr_b);
void* testDerived = dynamic_cast<void*>(ptr_c);
void* testMost = dynamic_cast<void*>(ptr_d);
Base_Class* tptrDerived = dynamic_cast<Derived_Class*>(static_cast<Base_Class*>(testDerived));
tptrDerived->dummy();
Base_Class* tptrMost = dynamic_cast<Derived_Class*>(static_cast<Base_Class*>(testMost));
tptrMost->dummy();
//tptrMost = dynamic_cast<AnotherMostDerivedObject*>(static_cast<Base_Class*>(testMost));
//tptrMost->dummy(); //fails
} catch (exception& my_ex) {cout << "Exception: " << my_ex.what();}
system("pause");
return 0;
}
Please correct me if this is not correct in any way.
it is usefull when we put the storage back to memory pool but we only keep a pointer to the base class. This case we should figure out the original address.
Expanding on #BruceAdi's answer and inspired by this discussion, here's a polymorphic situation which may require pointer adjustment. Suppose we have this factory-type setup:
struct Base { virtual ~Base() = default; /* ... */ };
struct Derived : Base { /* ... */ };
template <typename ...Args>
Base * Factory(Args &&... args)
{
return ::new Derived(std::forward<Args>(args)...);
}
template <typename ...Args>
Base * InplaceFactory(void * location, Args &&... args)
{
return ::new (location) Derived(std::forward<Args>(args)...);
}
Now I could say:
Base * p = Factory();
But how would I clean this up manually? I need the actual memory address to call ::operator delete:
void * addr = dynamic_cast<void*>(p);
p->~Base(); // OK thanks to virtual destructor
// ::operator delete(p); // Error, wrong address!
::operator delete(addr); // OK
Or I could re-use the memory:
void * addr = dynamic_cast<void*>(p);
p->~Base();
p = InplaceFactory(addr, "some", "arguments");
delete p; // OK now
Don't do that at home
struct Base {
virtual ~Base ();
};
struct D : Base {};
Base *create () {
D *p = new D;
return p;
}
void *destroy1 (Base *b) {
void *p = dynamic_cast<void*> (b);
b->~Base ();
return p;
}
void destroy2 (void *p) {
operator delete (p);
}
int i = (destroy2 (destroy1 (create ())), i);
Warning: This will not work if D is defined as:
struct D : Base {
void* operator new (size_t);
void operator delete (void*);
};
and there is no way to make it work.
This might be one way to provide an Opaque Pointer through an ABI. Opaque Pointers -- and, more generally, Opaque Data Types -- are used to pass objects and other resources around between library code and client code in such a way that the client code can be isolated from the implementation details of the library. There are other ways to accomplish this, to be sure, and maybe some of them would be better for a particular use case.
Windows makes a lot of use of Opaque Pointers in its API. HANDLE is, I believe, generally an opaque pointer to the actual resource you have a HANDLE to, for example. HANDLEs can be Kernel Objects like files, GDI objects, and all sorts of User Objects of various kinds -- all of which must be vastly different in implementation, but all are returned as a HANDLE to the user.
#include <iostream>
#include <string>
#include <iomanip>
using namespace std;
/*** LIBRARY.H ***/
namespace lib
{
typedef void* MYHANDLE;
void ShowObject(MYHANDLE h);
MYHANDLE CreateObject();
void DestroyObject(MYHANDLE);
};
/*** CLIENT CODE ***/
int main()
{
for( int i = 0; i < 25; ++i )
{
cout << "[" << setw(2) << i << "] :";
lib::MYHANDLE h = lib::CreateObject();
lib::ShowObject(h);
lib::DestroyObject(h);
cout << "\n";
}
}
/*** LIBRARY.CPP ***/
namespace impl
{
class Base { public: virtual ~Base() { cout << "[~Base]"; } };
class Foo : public Base { public: virtual ~Foo() { cout << "[~Foo]"; } };
class Bar : public Base { public: virtual ~Bar() { cout << "[~Bar]"; } };
};
lib::MYHANDLE lib::CreateObject()
{
static bool init = false;
if( !init )
{
srand((unsigned)time(0));
init = true;
}
if( rand() % 2 )
return static_cast<impl::Base*>(new impl::Foo);
else
return static_cast<impl::Base*>(new impl::Bar);
}
void lib::DestroyObject(lib::MYHANDLE h)
{
delete static_cast<impl::Base*>(h);
}
void lib::ShowObject(lib::MYHANDLE h)
{
impl::Foo* foo = dynamic_cast<impl::Foo*>(static_cast<impl::Base*>(h));
impl::Bar* bar = dynamic_cast<impl::Bar*>(static_cast<impl::Base*>(h));
if( foo )
cout << "FOO";
if( bar )
cout << "BAR";
}
Consider the following code:
class A
{
B* b; // an A object owns a B object
A() : b(NULL) { } // we don't know what b will be when constructing A
void calledVeryOften(…)
{
if (b)
delete b;
b = new B(param1, param2, param3, param4);
}
};
My goal: I need to maximize performance, which, in this case, means minimizing the amount of memory allocations.
The obvious thing to do here is to change B* b; to B b;. I see two problems with this approach:
I need to initialize b in the constructor. Since I don't know what b will be, this means I need to pass dummy values to B's constructor. Which, IMO, is ugly.
In calledVeryOften(), I'll have to do something like this: b = B(…), which is wrong for two reasons:
The destructor of b won't be called.
A temporary instance of B will be constructed, then copied into b, then the destructor of the temporary instance will be called. The copy and the destructor call could be avoided. Worse, calling the destructor could very well result in undesired behavior.
So what solutions do I have to avoid using new? Please keep in mind that:
I only have control over A. I don't have control over B, and I don't have control over the users of A.
I want to keep the code as clean and readable as possible.
I liked Klaim's answer, so I wrote this up real fast. I don't claim perfect correctness but it looks pretty good to me. (i.e., the only testing it has is the sample main below)
It's a generic lazy-initializer. The space for the object is allocated once, and the object starts at null. You can then create, over-writing previous objects, with no new memory allocations.
It implements all the necessary constructors, destructor, copy/assignment, swap, yadda-yadda. Here you go:
#include <cassert>
#include <new>
template <typename T>
class lazy_object
{
public:
// types
typedef T value_type;
typedef const T const_value_type;
typedef value_type& reference;
typedef const_value_type& const_reference;
typedef value_type* pointer;
typedef const_value_type* const_pointer;
// creation
lazy_object(void) :
mObject(0),
mBuffer(::operator new(sizeof(T)))
{
}
lazy_object(const lazy_object& pRhs) :
mObject(0),
mBuffer(::operator new(sizeof(T)))
{
if (pRhs.exists())
{
mObject = new (buffer()) T(pRhs.get());
}
}
lazy_object& operator=(lazy_object pRhs)
{
pRhs.swap(*this);
return *this;
}
~lazy_object(void)
{
destroy();
::operator delete(mBuffer);
}
// need to make multiple versions of this.
// variadic templates/Boost.PreProccesor
// would help immensely. For now, I give
// two, but it's easy to make more.
void create(void)
{
destroy();
mObject = new (buffer()) T();
}
template <typename A1>
void create(const A1 pA1)
{
destroy();
mObject = new (buffer()) T(pA1);
}
void destroy(void)
{
if (exists())
{
mObject->~T();
mObject = 0;
}
}
void swap(lazy_object& pRhs)
{
std::swap(mObject, pRhs.mObject);
std::swap(mBuffer, pRhs.mBuffer);
}
// access
reference get(void)
{
return *get_ptr();
}
const_reference get(void) const
{
return *get_ptr();
}
pointer get_ptr(void)
{
assert(exists());
return mObject;
}
const_pointer get_ptr(void) const
{
assert(exists());
return mObject;
}
void* buffer(void)
{
return mBuffer;
}
// query
const bool exists(void) const
{
return mObject != 0;
}
private:
// members
pointer mObject;
void* mBuffer;
};
// explicit swaps for generality
template <typename T>
void swap(lazy_object<T>& pLhs, lazy_object<T>& pRhs)
{
pLhs.swap(pRhs);
}
// if the above code is in a namespace, don't put this in it!
// specializations in global namespace std are allowed.
namespace std
{
template <typename T>
void swap(lazy_object<T>& pLhs, lazy_object<T>& pRhs)
{
pLhs.swap(pRhs);
}
}
// test use
#include <iostream>
int main(void)
{
// basic usage
lazy_object<int> i;
i.create();
i.get() = 5;
std::cout << i.get() << std::endl;
// asserts (not created yet)
lazy_object<double> d;
std::cout << d.get() << std::endl;
}
In your case, just create a member in your class: lazy_object<B> and you're done. No manual releases or making copy-constructors, destructors, etc. Everything is taken care of in your nice, small re-usable class. :)
EDIT
Removed the need for vector, should save a bit of space and what-not.
EDIT2
This uses aligned_storage and alignment_of to use the stack instead of heap. I used boost, but this functionality exists in both TR1 and C++0x. We lose the ability to copy, and therefore swap.
#include <boost/type_traits/aligned_storage.hpp>
#include <cassert>
#include <new>
template <typename T>
class lazy_object_stack
{
public:
// types
typedef T value_type;
typedef const T const_value_type;
typedef value_type& reference;
typedef const_value_type& const_reference;
typedef value_type* pointer;
typedef const_value_type* const_pointer;
// creation
lazy_object_stack(void) :
mObject(0)
{
}
~lazy_object_stack(void)
{
destroy();
}
// need to make multiple versions of this.
// variadic templates/Boost.PreProccesor
// would help immensely. For now, I give
// two, but it's easy to make more.
void create(void)
{
destroy();
mObject = new (buffer()) T();
}
template <typename A1>
void create(const A1 pA1)
{
destroy();
mObject = new (buffer()) T(pA1);
}
void destroy(void)
{
if (exists())
{
mObject->~T();
mObject = 0;
}
}
// access
reference get(void)
{
return *get_ptr();
}
const_reference get(void) const
{
return *get_ptr();
}
pointer get_ptr(void)
{
assert(exists());
return mObject;
}
const_pointer get_ptr(void) const
{
assert(exists());
return mObject;
}
void* buffer(void)
{
return mBuffer.address();
}
// query
const bool exists(void) const
{
return mObject != 0;
}
private:
// types
typedef boost::aligned_storage<sizeof(T),
boost::alignment_of<T>::value> storage_type;
// members
pointer mObject;
storage_type mBuffer;
// non-copyable
lazy_object_stack(const lazy_object_stack& pRhs);
lazy_object_stack& operator=(lazy_object_stack pRhs);
};
// test use
#include <iostream>
int main(void)
{
// basic usage
lazy_object_stack<int> i;
i.create();
i.get() = 5;
std::cout << i.get() << std::endl;
// asserts (not created yet)
lazy_object_stack<double> d;
std::cout << d.get() << std::endl;
}
And there we go.
Simply reserve the memory required for b (via a pool or by hand) and reuse it each time you delete/new instead of reallocating each time.
Example :
class A
{
B* b; // an A object owns a B object
bool initialized;
public:
A() : b( malloc( sizeof(B) ) ), initialized(false) { } // We reserve memory for b
~A() { if(initialized) destroy(); free(b); } // release memory only once we don't use it anymore
void calledVeryOften(…)
{
if (initialized)
destroy();
create();
}
private:
void destroy() { b->~B(); initialized = false; } // hand call to the destructor
void create( param1, param2, param3, param4 )
{
b = new (b) B( param1, param2, param3, param4 ); // in place new : only construct, don't allocate but use the memory that the provided pointer point to
initialized = true;
}
};
In some cases a Pool or ObjectPool could be a better implementation of the same idea.
The construction/destruction cost will then only be dependante on the constructor and destructor of the B class.
How about allocating the memory for B once (or for it's biggest possible variant) and using placement new?
A would store char memB[sizeof(BiggestB)]; and a B*. Sure, you'd need to manually call the destructors, but no memory would be allocated/deallocated.
void* p = memB;
B* b = new(p) SomeB();
...
b->~B(); // explicit destructor call when needed.
If B correctly implements its copy assignment operator then b = B(...) should not call any destructor on b. It is the most obvious solution to your problem.
If, however, B cannot be appropriately 'default' initialized you could do something like this. I would only recommend this approach as a last resort as it is very hard to get safe. Untested, and very probably with corner case exception bugs:
// Used to clean up raw memory of construction of B fails
struct PlacementHelper
{
PlacementHelper() : placement(NULL)
{
}
~PlacementHelper()
{
operator delete(placement);
}
void* placement;
};
void calledVeryOften(....)
{
PlacementHelper hp;
if (b == NULL)
{
hp.placement = operator new(sizeof(B));
}
else
{
hp.placement = b;
b->~B();
b = NULL; // We can't let b be non-null but point at an invalid B
}
// If construction throws, hp will clean up the raw memory
b = new (placement) B(param1, param2, param3, param4);
// Stop hp from cleaning up; b points at a valid object
hp.placement = NULL;
}
A quick test of Martin York's assertion that this is a premature optimisation, and that new/delete are optimised well beyond the ability of mere programmers to improve. Obviously the questioner will have to time his own code to see whether avoiding new/delete helps him, but it seems to me that for certain classes and uses it will make a big difference:
#include <iostream>
#include <vector>
int g_construct = 0;
int g_destruct = 0;
struct A {
std::vector<int> vec;
A (int a, int b) : vec((a*b) % 2) { ++g_construct; }
~A() {
++g_destruct;
}
};
int main() {
const int times = 10*1000*1000;
#if DYNAMIC
std::cout << "dynamic\n";
A *x = new A(1,3);
for (int i = 0; i < times; ++i) {
delete x;
x = new A(i,3);
}
#else
std::cout << "automatic\n";
char x[sizeof(A)];
A* yzz = new (x) A(1,3);
for (int i = 0; i < times; ++i) {
yzz->~A();
new (x) A(i,3);
}
#endif
std::cout << g_construct << " constructors and " << g_destruct << " destructors\n";
}
$ g++ allocperf.cpp -oallocperf -O3 -DDYNAMIC=0 -g && time ./allocperf
automatic
10000001 constructors and 10000000 destructors
real 0m7.718s
user 0m7.671s
sys 0m0.030s
$ g++ allocperf.cpp -oallocperf -O3 -DDYNAMIC=1 -g && time ./allocperf
dynamic
10000001 constructors and 10000000 destructors
real 0m15.188s
user 0m15.077s
sys 0m0.047s
This is roughly what I expected: the GMan-style (destruct/placement new) code takes twice as long, and is presumably doing twice as much allocation. If the vector member of A is replaced with an int, then the GMan-style code takes a fraction of a second. That's GCC 3.
$ g++-4 allocperf.cpp -oallocperf -O3 -DDYNAMIC=1 -g && time ./allocperf
dynamic
10000001 constructors and 10000000 destructors
real 0m5.969s
user 0m5.905s
sys 0m0.030s
$ g++-4 allocperf.cpp -oallocperf -O3 -DDYNAMIC=0 -g && time ./allocperf
automatic
10000001 constructors and 10000000 destructors
real 0m2.047s
user 0m1.983s
sys 0m0.000s
This I'm not so sure about, though: now the delete/new takes three times as long as the destruct/placement new version.
[Edit: I think I've figured it out - GCC 4 is faster on the 0-sized vectors, in effect subtracting a constant time from both versions of the code. Changing (a*b)%2 to (a*b)%2+1 restores the 2:1 time ratio, with 3.7s vs 7.5]
Note that I've not taken any special steps to correctly align the stack array, but printing the address shows it's 16-aligned.
Also, -g doesn't affect the timings. I left it in accidentally after I was looking at the objdump to check that -O3 hadn't completely removed the loop. That pointers called yzz because searching for "y" didn't go quite as well as I'd hoped. But I've just re-run without it.
Are you sure that memory allocation is the bottleneck you think it is? Is B's constructor trivially fast?
If memory allocation is the real problem, then placement new or some of the other solutions here might well help.
If the types and ranges of the param[1..4] are reasonable, and the B constructor "heavy", you might also consider using a cached set of B. This presumes you are actually allowed to have more than one at a time, that it does not front a resource for example.
Like the others have already suggested: Try placement new..
Here is a complete example:
#include <new>
#include <stdio.h>
class B
{
public:
int dummy;
B (int arg)
{
dummy = arg;
printf ("C'Tor called\n");
}
~B ()
{
printf ("D'tor called\n");
}
};
void called_often (B * arg)
{
// call D'tor without freeing memory:
arg->~B();
// call C'tor without allocating memory:
arg = new(arg) B(10);
}
int main (int argc, char **args)
{
B test(1);
called_often (&test);
}
I'd go with boost::scoped_ptr here:
class A: boost::noncopyable
{
typedef boost::scoped_ptr<B> b_ptr;
b_ptr pb_;
public:
A() : pb_() {}
void calledVeryOften( /*…*/ )
{
pb_.reset( new B( params )); // old instance deallocated
// safely use *pb_ as reference to instance of B
}
};
No need for hand-crafted destructor, A is non-copyable, as it should be in your original code, not to leak memory on copy/assignment.
I'd suggest to re-think the design though if you need to re-allocate some inner state object very often. Look into Flyweight and State patterns.
Erm, is there some reason you can't do this?
A() : b(new B()) { }
void calledVeryOften(…)
{
b->setValues(param1, param2, param3, param4);
}
(or set them individually, since you don't have access to the B class - those values do have mutator-methods, right?)
Just have a pile of previously used Bs, and re-use them.