A very weak reference (cannot be turned into shared) - c++

I'd like to know if there is any smart pointer type concept that implements the "very weak reference" idea.
This would be basically a weak_ptr but that cannot be turned into a shared_ptr, basically, when you have very_weak_refs out-there, you are sure that the strong-ref count can never go up.
This would allow better "strong ownership" of memory by managers, and delivering very weak references in the wild would still allow clients to access the data by using good old raw pointer through a .lock_get() function or equivalent... (name designed to mirror what you would have doing .lock().get() usually).
You don't have the same security on data, because your object may be destroyed while you use it, but if your environment is controlled enough so that you know that the manager's cannot clean its data while you are processing, you're still good to use raw-pointers locally after having checked against null_ptr after lock_get().
Do any of you wished for similar thing, more info/intelligence/thoughts on that ?
thanks.
rationale : the motivation behind is that weak_ptr has the "security flaw" of being turnable to shared and therefore, after distributing weak references in the wild, you basically did the same than distributing shared ones because anybody can keep very long lived shared refs on your data effectively preventing correct cleaning by the entities that were suppsed to be strong (the manager).
This is solved by very-weak-refs, when you distribute that kind of objects in your manager's public interface, you are sure that when you delete your last shared ref, your data is deleted.
For me, the whole concept of weak references works only with well behaved clients; who understands that they should promotes their weak refs into shareds for only small amounts of time.

Unfortunately, what you are asking for is impossible with the traditional interface of smart pointers.
The issue is one of lifetime. A weak_ptr cannot be use to access the object directly, because it does not guarantee that said object will live long enough: the object might be pulled right from under your feet.
Example:
int main() {
std::shared_ptr<int> sp(new int(4));
std::weak_ptr<int> wp(sp);
if (not wp.expired()) {
sp.reset();
std::cout << *wp << "\n"; // Access WP ? But there is nothing there!
}
}
Thus, for better or worse, there is no other choice than recovering a shared pointer from the weak pointer any time you actually need to access the object without controlling the duration of this access.
This last point, however, is our clue. A simple idea being to write a well-behaved client of weak_ptr yourself and change the way it allows the external world to access the data. For example:
template <typename T>
class very_weak_ptr {
public:
very_weak_ptr() {}
explicit very_weak_ptr(std::weak_ptr<T> wp): _wp(wp) {}
template <typename F>
void apply(F&& f) {
std::shared_ptr<T> sp = _wp.lock();
f(sp.get());
}
private:
std::weak_ptr<T> _wp;
}; // class very_weak_ptr
Note: there is one remaining flaw, enable_shared_from_this allows to recover a std::shared_ptr<T> from the very instance of T; you can add a compile time check on T to prevent usage of this class with such objects.

What you ask for is functionally equivalent to, starting with a std::shared_ptr<>:
initially creating a std::weakptr<> thereto + .get()-ting a raw pointer,
before use of the raw pointer, checking the weak_ptr<> hasn't .expired().
Do any of you wished for similar thing, more info/intelligence/thoughts on that ? thanks.
No... if you need to check the weak_ptr<>.expired() anyway, you might as well get a valid shared_ptr<>. What do you really think this is going to achieve? 'Better "strong ownership"' when you know somehow / require that the manager can't release the object during the period you're using the raw pointer - doesn't add up....

Related

How to enable Rust Ownership paradigm in C++

The system programming language Rust uses the ownership paradigm to ensure at compile time with zero cost for the runtime when a resource has to be freed.
In C++ we commonly use smart pointers to achieve the same goal of hiding the complexity of managing resource allocation. There are a couple of differences though:
In Rust there is always only one owner, whereas C++ shared_ptr can easily leak ownership.
In Rust we can borrow references we do not own, whereas C++ unique_ptr cannot be shared in a safe way via weak_ptr and lock().
Reference counting of shared_ptr is costly.
My question is: How can we emulate the ownership paradigm in C++ within the following constraints:
Only one owner at any time
Possibility to borrow a pointer and use it temporarily without fear of the resource going out of scope (observer_ptr is useless for this)
As much compile-time checks as possible.
Edit: Given the comments so far, we can conclude:
No compile-time support for this (I was hoping for some decltype/template magic unknown to me) in the compilers. Might be possible using static analysis elsewhere (taint?)
No way to get this without reference counting.
No standard implementation to distinguish shared_ptrs with owning or borrowing semantic
Could roll your own by creating wrapper types around shared_ptr and weak_ptr:
owned_ptr: non-copyable, move-semantics, encapsulates shared_ptr, access to borrowed_ptr
borrowed_ptr: copyable, encapsulates weak_ptr, lock method
locked_ptr: non-copyable, move-semantics, encapsulates shared_ptr from locking weak_ptr
You can't do this with compile-time checks at all. The C++ type system is lacking any way to reason about when an object goes out of scope, is moved, or is destroyed — much less turn this into a type constraint.
What you could do is have a variant of unique_ptr that keeps a counter of how many "borrows" are active at run time. Instead of get() returning a raw pointer, it would return a smart pointer that increments this counter on construction and decrements it on destruction. If the unique_ptr is destroyed while the count is non-zero, at least you know someone somewhere did something wrong.
However, this is not a fool-proof solution. Regardless of how hard you try to prevent it, there will always be ways to get a raw pointer to the underlying object, and then it's game over, since that raw pointer can easily outlive the smart pointer and the unique_ptr. It will even sometimes be necessary to get a raw pointer, to interact with an API that requires raw pointers.
Moreover, ownership is not about pointers. Box/unique_ptr allows you to heap allocate an object, but it changes nothing about ownership, life time, etc. compared to putting the same object on the stack (or inside another object, or anywhere else really). To get the same mileage out of such a system in C++, you'd have to make such "borrow counting" wrappers for all objects everywhere, not just for unique_ptrs. And that is pretty impractical.
So let's revisit the compile time option. The C++ compiler can't help us, but maybe lints can? Theoretically, if you implement the whole life time part of the type system and add annotations to all APIs you use (in addition to your own code), that may work.
But it requires annotations for all functions used in the whole program. Including private helper function of third party libraries. And those for which no source code is available. And for those whose implementation that are too complicated for the linter to understand (from Rust experience, sometimes the reason something is safe are too subtle to express in the static model of lifetimes and it has to be written slightly differently to help the compiler). For the last two, the linter can't verify that the annotation is indeed correct, so you're back to trusting the programmer. Additionally, some APIs (or rather, the conditions for when they are safe) can't really be expressed very well in the lifetime system as Rust uses it.
In other words, a complete and practically useful linter for this this would be substantial original research with the associated risk of failure.
Maybe there is a middle ground that gets 80% of the benefits with 20% of the cost, but since you want a hard guarantee (and honestly, I'd like that too), tough luck. Existing "good practices" in C++ already go a long way to minimizing the risks, by essentially thinking (and documenting) the way a Rust programmer does, just without compiler aid. I'm not sure if there is much improvement over that to be had considering the state of C++ and its ecosystem.
tl;dr Just use Rust ;-)
What follows are some examples of ways people have tried to emulate parts of Rust's ownership paradigm in C++, with limited success:
Lifetime safety: Preventing common dangling. The most thorough and rigorous approach, involving several additions to the language to support the necessary annotations. If the effort is still alive (last commit was in 2019), getting this analysis added to a mainstream compiler is probably the most likely route to "borrow checked" C++. Discussed on IRLO.
Borrowing Trouble: The Difficulties Of A C++ Borrow-Checker
Is it possible to achieve Rust's ownership model with a generic C++ wrapper?
C++Now 2017: Jonathan Müller “Emulating Rust's borrow checker in C++" (video) and associated code, about which the author says, "You're not actually supposed to use that, if you need such a feature, you should use Rust."
Emulating the Rust borrow checker with C++ move-only types and part II (which is actually more like emulating RefCell than the borrow checker, per se)
I believe you can get some of the benefits of Rust by enforcing some strict coding conventions (which is after all what you'd have to do anyway, since there's no way with "template magic" to tell the compiler not to compile code that doesn't use said "magic"). Off the top of my head, the following could get you...well...kind of close, but only for single-threaded applications:
Never use new directly; instead, use make_unique. This goes partway toward ensuring that heap-allocated objects are "owned" in a Rust-like manner.
"Borrowing" should always be represented via reference parameters to function calls. Functions that take a reference should never create any sort of pointer to the refered-to object. (It may in some cases be necessary to use a raw pointer as a paramter instead of a reference, but the same rule should apply.)
Note that this works for objects on the stack or on the heap; the function shouldn't care.
Transfer of ownership is, of course, represented via R-value references (&&) and/or R-value references to unique_ptrs.
Unfortunately, I can't think of any way to enforce Rust's rule that mutable references can only exist anywhere in the system when there are no other extant references.
Also, for any kind of parallelism, you would need to start dealing with lifetimes, and the only way I can think of to permit cross-thread lifetime management (or cross-process lifetime management using shared memory) would be to implement your own "ptr-with-lifetime" wrapper. This could be implemented using shared_ptr, because here, reference-counting would actually be important; it's still a bit of unnecessary overhead, though, because reference-count blocks actually have two reference counters (one for all the shared_ptrs pointing to the object, another for all the weak_ptrs). It's also a little... odd, because in a shared_ptr scenario, everybody with a shared_ptr has "equal" ownership, whereas in a "borrowing with lifetime" scenario, only one thread/process should actually "own" the memory.
I think one could add a degree of compile-time introspection and custom sanitisation by introducing custom wrapper classes that track ownership and borrowing.
The code below is a hypothetical sketch, and not a production solution which would need a lot more tooling, e.g. #def out the checks when not sanitising. It uses a very naive lifetime checker to 'count' borrow errors in ints, in this instance during compilation. static_asserts are not possible as the ints are not constexpr, but the values are there and can be interrogated before runtime. I believe this answers your 3 constraints, regardless of whether these are heap allocations, so I'm using a simple int type to demo the idea, rather than a smart pointer.
Try uncommenting the use cases in main() below (run in compiler explorer with -O3 to see boilerplate optimise away), and you'll see the warning counters change.
https://godbolt.org/z/Pj4WMr
// Hypothetical Rust-like owner / borrow wrappers in C++
// This wraps types with data which is compiled away in release
// It is not possible to static_assert, so this uses static ints to count errors.
#include <utility>
// Statics to track errors. Ideally these would be static_asserts
// but they depen on Owner::has_been_moved which changes during compilation.
static int owner_already_moved = 0;
static int owner_use_after_move = 0;
static int owner_already_borrowed = 0;
// This method exists to ensure static errors are reported in compiler explorer
int get_fault_count() {
return owner_already_moved + owner_use_after_move + owner_already_borrowed;
}
// Storage for ownership of a type T.
// Equivalent to mut usage in Rust
// Disallows move by value, instead ownership must be explicitly moved.
template <typename T>
struct Owner {
Owner(T v) : value(v) {}
Owner(Owner<T>& ov) = delete;
Owner(Owner<T>&& ov) {
if (ov.has_been_moved) {
owner_already_moved++;
}
value = std::move(ov.value);
ov.has_been_moved = true;
}
T& operator*() {
if (has_been_moved) {
owner_use_after_move++;
}
return value;
}
T value;
bool has_been_moved{false};
};
// Safely borrow a value of type T
// Implicit constuction from Owner of same type to check borrow is safe
template <typename T>
struct Borrower {
Borrower(Owner<T>& v) : value(v.value) {
if (v.has_been_moved) {
owner_already_borrowed++;
}
}
const T& operator*() const {
return value;
}
T value;
};
// Example of function borrowing a value, can only read const ref
static void use(Borrower<int> v) {
(void)*v;
}
// Example of function taking ownership of value, can mutate via owner ref
static void use_mut(Owner<int> v) {
*v = 5;
}
int main() {
// Rather than just 'int', Owner<int> tracks the lifetime of the value
Owner<int> x{3};
// Borrowing value before mutating causes no problems
use(x);
// Mutating value passes ownership, has_been_moved set on original x
use_mut(std::move(x));
// Uncomment for owner_already_borrowed = 1
//use(x);
// Uncomment for owner_already_moved = 1
//use_mut(std::move(x));
// Uncomment for another owner_already_borrowed++
//Borrower<int> y = x;
// Uncomment for owner_use_after_move = 1;
//return *x;
}
The use of static counters is obviously not desirable, but it is not possible to use static_assert as owner_already_moved is non-const. The idea is these statics give hints to errors appearing, and in final production code they could be #defed out.
You can use an enhanced version of a unique_ptr (to enforce a unique owner) together with an enhanced version of observer_ptr (to get a nice runtime exception for dangling pointers, i.e. if the original object maintained through unique_ptr went out of scope). The Trilinos package implements this enhanced observer_ptr, they call it Ptr. I have implemented the enhanced version of unique_ptr here (I call it UniquePtr): https://github.com/certik/trilinos/pull/1
Finally, if you want the object to be stack allocated, but still be able to pass safe references around, you need to use the Viewable class, see my initial implementation here: https://github.com/certik/trilinos/pull/2
This should allow you to use C++ just like Rust for pointers, except that in Rust you get a compile time error, while in C++ you get a runtime exception. Also, it should be noted, that you only get a runtime exception in Debug mode. In Release mode, the classes do not do these checks, so they are as fast as in Rust (essentially as fast as raw pointers), but then they can segfault. So one has to make sure the whole test suite runs in Debug mode.

Enforce safe use of class containing reference or raw pointer

Suppose we have a class that looks like the following.
class DoStuffWithRef
{
DoStuffWithRef(LargeObject& lo) : lo_(lo) {}
// a bunch of member functions, some of them useful
// [...]
private:
LargeObject& lo_;
};
The class is designed such that the only sane use is when at least one other entity has ownership of the object that lo_ references. If client code uses the class appropriately, there's no need for DoStuffWithRef to have ownership of the LargeObject.
Is there a way to enforce this usage or signal an error if the class is being misused? Can anything be done besides documenting the intended use of DoStuffWithRef?
For example, an automatically stored DoStuffWithRef might refer to an automatically stored LargeObject.
void foo()
{
LargeObject lo;
DoStuffWithRef dswr(lo);
// some code that makes use of the DoStuffWithRef instance
return;
}
This is the primary usage I have in mind although there are other possible cases where someone else might be guaranteed to have ownership.
The problem is that it's very possible for client code to create a DoStuffWithRef instance that will end up with a dangling reference. If no one else has ownership of a LargeObject referred to by a DoStuffWithRef, then chaos ensues when the the DoStuffWithRef tries to access the LargeObject. Worse yet, chaos may ensue only long after the error has been made.
When this has come up in the past, I've used a boost::shared_ptr<> instead of a reference (this was in C++03). That doesn't quite express the semantics of the class appropriately. The premise is that DoStuffWithRef doesn't need ownership -- it is meant to act on objects owned by others and if no one else owns the object then the functionality provided by DoStuffWithRef makes no sense. Giving ownership to DoStuffWithRef has unnecessary overhead and forces shared ownership semantics. weak_ptr<> would also have the wrong semantics because we shouldn't be asking at runtime if the pointer is valid.
I've never had this scenario come up in a performance sensitive section of code or in a case where shared ownership was a significant burden so it just hasn't really mattered but I wish I knew of a way to accurately express this in C++. The costs are small but they're also unnecessary. More importantly, a shared_ptr<> will hide incorrect use of the class. If a DoStuffWithRef, modified to use a shared_ptr<>, is the only remaining owner of a LargeObject, then the dangling reference has been averted in the short term but the client code ends up in uncharted territory.
Is DoStuffWithRef otherwise stateful? If not, it seems like it is a bucket of utility functions. The reference to LargeObject could just as well be passed as an argument to each function. That also resolves the question of ownership.
Otherwise, if it is stateful, this is a clear case of shared ownership, since DoStuffWithRef really does need to extend the lifetime of a single live instance of LargeObject to that of itself. Or, use a weak pointer and eat the cost of the runtime check...

Suicide object implementation leveraging `std::weak_ptr`

I'm considering using "suicide objects" to model entities in a game, that is, objects able to delete themselves. Now, the usual C++03 implementation (plain old delete this) does nothing for other objects potentially refering to the suicide object, which is why I'm using std::shared_ptr and std::weak_ptr.
Now for the code dump :
#include <memory>
#include <iostream>
#include <cassert>
struct SuObj {
SuObj() { std::cout << __func__ << '\n'; }
~SuObj() { std::cout << __func__ << '\n'; }
void die() {
ptr.reset();
}
static std::weak_ptr<SuObj> create() {
std::shared_ptr<SuObj> obj = std::make_shared<SuObj>();
return (obj->ptr = std::move(obj));
}
private:
std::shared_ptr<SuObj> ptr;
};
int main() {
std::weak_ptr<SuObj> obj = SuObj::create();
assert(!obj.expired());
std::cout << "Still alive\n";
obj.lock()->die();
assert(obj.expired());
std::cout << "Deleted\n";
return 0;
}
Question
This code appears to work fine. However, I'd like to have someone else's eye to gauge it. Does this code make sense ? Did I blindly sail into undefined lands ? Should I drop my keyboard and begin art studies right now ?
I hope this question is sufficiently narrowed down for SO. Seemed a bit tiny and low-level for CR.
Minor precision
I do not intend to use this in multithreaded code. If the need ever arises, I'll be sure to reconsider the whole thing.
When you have shared_ptr based object lifetime, the lifetime of your object is the "lifetime" of the union of the shared_ptrs who own it collectively.
In your case, you have an internal shared_ptr, and your object will not die until that internal shared_ptr expires.
However, this does not mean you can commit suicide. If you remove that last reference, your object continues to exist if anyone has .lock()'d the weak_ptr and stored the result. As this is the only way you can access the object externally, it may happen1.
In short, die() can fail to kill the object. It might better be called remove_life_support(), as something else could keep the object alive after said life support is removed.
Other than that, your design works.
1
You could say "well, then callers should just not keep the shared_ptr around" -- but that doesn't work, as the check that the object is valid is only valid as long as the shared_ptr persists. Plus, by exposing the way to create shared_ptr, you have no type guarantees that the client code won't store them (accidentally or on purpose).
A transaction based model (where you pass a lambda in, and it operates on it internally) could help with this if you want seriously paranoid robustness.
Or you can live with the object sometimes living too long.
Consider hiding these messy details behind a Regular Type (or almost-regular) that has a pImpl to the nasty memory management problem. That pImpl could be a weak_ptr with the above semantics.
Then users of your code need only interact with the Regular (or pseudoRegular) wrapper.
If you don't want cloning to be easy, disable copy construction/assignment and only expose move.
Now your nasty memory management is hiding behind a fascade, and if you decide you did it all wrong the external pseudoRegular interface can have different guts.
Regular type in C++11
Not a direct answer but potentially useful information.
In Chromium codebase there is a concept of exactly what you are trying to achieve. They call it WeakPtrFactory. Their solution cannot be directly taken into your code since they have their own implementation of e.g. shared_ptr and weak_ptr but design wise it can be of use to you.
I made a try to implement it and found out that the problem of double deletion can be solved by passing into inner shared_ptr custom empty deleter - from this moment on neither shared_ptrs created from weak_ptr not inner shared_ptr will be able to call destructor (again) on your object.
The only problem to solve is what if your object is deleted and somewhere else you keep shared_ptr to it? But from what I see it cannot be simply solved by any magic mean and require designing that whole project the way that it simply never happens e.g. by using shared_ptr only in local scope and ensuring that some set of operations (creating suicide object, using it, ordering its suicide) could be performed only in the same thread.
I understand you're trying to create a minimal example for SO, but I see a few challenges you'll want to consider:
You have a public constructor and destructor, so technically there's no guarantee that the create() method is always used.
You could make those protected or private but that decision would interfere with use with std algorithms and containers.
This doesn't guarantee that the object will actually destruct because as long as someone has a shared_ptr it's going to exist. That may or may not be a problem for your use case, but because of that I don't think this will add as much value as you're heading.
This is likely going to be confusing and counter-intuitive to other developers. It might make maintenance harder, even if your intent is to make it easier. That's a bit of a value judgement, but I'd encourage you to consider if it's truly easier to manage.
I commend you for putting thought into memory management up front. Disciplined use of shared_ptr and weak_ptr will help with your memory management issues -- I'd counsel against trying to have the instance try to manage its own lifecycle.
As for art studies... I'd only recommend that if that's truly your passion! Good luck!

delete this & private destructor

I've been thinking about the possible use of delete this in c++, and I've seen one use.
Because you can say delete this only when an object is on heap, I can make the destructor private and stop objects from being created on stack altogether. In the end I can just delete the object on heap by saying delete this in a random public member function that acts as a destructor. My questions:
1) Why would I want to force the object to be made on the heap instead of on the stack?
2) Is there another use of delete this apart from this? (supposing that this is a legitimate use of it :) )
Any scheme that uses delete this is somewhat dangerous, since whoever called the function that does that is left with a dangling pointer. (Of course, that's also the case when you delete an object normally, but in that case, it's clear that the object has been deleted). Nevertheless, there are somewhat legitimate cases for wanting an object to manage its own lifetime.
It could be used to implement a nasty, intrusive reference-counting scheme. You would have functions to "acquire" a reference to the object, preventing it from being deleted, and then "release" it once you've finished, deleting it if noone else has acquired it, along the lines of:
class Nasty {
public:
Nasty() : references(1) {}
void acquire() {
++references;
}
void release() {
if (--references == 0) {
delete this;
}
}
private:
~Nasty() {}
size_t references;
};
// Usage
Nasty * nasty = new Nasty; // 1 reference
nasty->acquire(); // get a second reference
nasty->release(); // back to one
nasty->release(); // deleted
nasty->acquire(); // BOOM!
I would prefer to use std::shared_ptr for this purpose, since it's thread-safe, exception-safe, works for any type without needing any explicit support, and prevents access after deleting.
More usefully, it could be used in an event-driven system, where objects are created, and then manage themselves until they receive an event that tells them that they're no longer needed:
class Worker : EventReceiver {
public:
Worker() {
start_receiving_events(this);
}
virtual void on(WorkEvent) {
do_work();
}
virtual void on(DeleteEvent) {
stop_receiving_events(this);
delete this;
}
private:
~Worker() {}
void do_work();
};
1) Why would I want to force the object to be made on the heap instead of on the stack?
1) Because the object's lifetime is not logically tied to a scope (e.g., function body, etc.). Either because it must manage its own lifespan, or because it is inherently a shared object (and thus, its lifespan must be attached to those of its co-dependent objects). Some people here have pointed out some examples like event handlers, task objects (in a scheduler), and just general objects in a complex object hierarchy.
2) Because you want to control the exact location where code is executed for the allocation / deallocation and construction / destruction. The typical use-case here is that of cross-module code (spread across executables and DLLs (or .so files)). Because of issues of binary compatibility and separate heaps between modules, it is often a requirement that you strictly control in what module these allocation-construction operations happen. And that implies the use of heap-based objects only.
2) Is there another use of delete this apart from this? (supposing that this is a legitimate use of it :) )
Well, your use-case is really just a "how-to" not a "why". Of course, if you are going to use a delete this; statement within a member function, then you must have controls in place to force all creations to occur with new (and in the same translation unit as the delete this; statement occurs). Not doing this would just be very very poor style and dangerous. But that doesn't address the "why" you would use this.
1) As others have pointed out, one legitimate use-case is where you have an object that can determine when its job is over and consequently destroy itself. For example, an event handler deleting itself when the event has been handled, a network communication object that deletes itself once the transaction it was appointed to do is over, or a task object in a scheduler deleting itself when the task is done. However, this leaves a big problem: signaling to the outside world that it no longer exists. That's why many have mentioned the "intrusive reference counting" scheme, which is one way to ensure that the object is only deleted when there are no more references to it. Another solution is to use a global (singleton-like) repository of "valid" objects, in which case any accesses to the object must go through a check in the repository and the object must also add/remove itself from the repository at the same time as it makes the new and delete this; calls (either as part of an overloaded new/delete, or alongside every new/delete calls).
However, there is a much simpler and less intrusive way to achieve the same behavior, albeit less economical. One can use a self-referencing shared_ptr scheme. As so:
class AutonomousObject {
private:
std::shared_ptr<AutonomousObject> m_shared_this;
protected:
AutonomousObject(/* some params */);
public:
virtual ~AutonomousObject() { };
template <typename... Args>
static std::weak_ptr<AutonomousObject> Create(Args&&... args) {
std::shared_ptr<AutonomousObject> result(new AutonomousObject(std::forward<Args>(args)...));
result->m_shared_this = result; // link the self-reference.
return result; // return a weak-pointer.
};
// this is the function called when the life-time should be terminated:
void OnTerminate() {
m_shared_this.reset( NULL ); // do not use reset(), but use reset( NULL ).
};
};
With the above (or some variations upon this crude example, depending on your needs), the object will be alive for as long as it deems necessary and that no-one else is using it. The weak-pointer mechanism serves as the proxy to query for the existence of the object, by possible outside users of the object. This scheme makes the object a bit heavier (has a shared-pointer in it) but it is easier and safer to implement. Of course, you have to make sure that the object eventually deletes itself, but that's a given in this kind of scenario.
2) The second use-case I can think of ties in to the second motivation for restricting an object to be heap-only (see above), however, it applies also for when you don't restrict it as such. If you want to make sure that both the deallocation and the destruction are dispatched to the correct module (the module from which the object was allocated and constructed), you must use a dynamic dispatching method. And for that, the easiest is to just use a virtual function. However, a virtual destructor is not going to cut it because it only dispatches the destruction, not the deallocation. The solution is to use a virtual "destroy" function that calls delete this; on the object in question. Here is a simple scheme to achieve this:
struct CrossModuleDeleter; //forward-declare.
class CrossModuleObject {
private:
virtual void Destroy() /* final */;
public:
CrossModuleObject(/* some params */); //constructor can be public.
virtual ~CrossModuleObject() { }; //destructor can be public.
//.... whatever...
friend struct CrossModuleDeleter;
template <typename... Args>
static std::shared_ptr< CrossModuleObject > Create(Args&&... args);
};
struct CrossModuleDeleter {
void operator()(CrossModuleObject* p) const {
p->Destroy(); // do a virtual dispatch to reach the correct deallocator.
};
};
// In the cpp file:
// Note: This function should not be inlined, so stash it into a cpp file.
void CrossModuleObject::Destroy() {
delete this;
};
template <typename... Args>
std::shared_ptr< CrossModuleObject > CrossModuleObject::Create(Args&&... args) {
return std::shared_ptr< CrossModuleObject >( new CrossModuleObject(std::forward<Args>(args)...), CrossModuleDeleter() );
};
The above kind of scheme works well in practice, and it has the nice advantage that the class can act as a base-class with no additional intrusion by this virtual-destroy mechanism in the derived classes. And, you can also modify it for the purpose of allowing only heap-based objects (as usually, making constructors-destructors private or protected). Without the heap-based restriction, the advantage is that you can still use the object as a local variable or data member (by value) if you want, but, of course, there will be loop-holes left to avoid by whoever uses the class.
As far as I know, these are the only legitimate use-cases I have ever seen anywhere or heard of (and the first one is easily avoidable, as I have shown, and often should be).
The general reason is that the lifetime of the object is determined by some factor internal to the class, at least from an application viewpoint. Hence, it may very well be a private method which calls delete this;.
Obviously, when the object is the only one to know how long it's needed, you can't put it on a random thread stack. It's necessary to create such objects on the heap.
It's generally an exceptionally bad idea. There are a very few cases- for example, COM objects have enforced intrusive reference counting. You'd only ever do this with a very specific situational reason- never for a general-purpose class.
1) Why would I want to force the object to be made on the heap instead of on the stack?
Because its life span isn't determined by the scoping rule.
2) Is there another use of delete this apart from this? (supposing that this is a legitimate use of it :) )
You use delete this when the object is the best placed one to be responsible for its own life span. One of the simplest example I know of is a window in a GUI. The window reacts to events, a subset of which means that the window has to be closed and thus deleted. In the event handler the window does a delete this. (You may delegate the handling to a controller class. But the situation "window forwards event to controller class which decides to delete the window" isn't much different of delete this, the window event handler will be left with the window deleted. You may also need to decouple the close from the delete, but your rationale won't be related to the desirability of delete this).
delete this;
can be useful at times and is usually used for a control class that also controls the lifetime of another object. With intrusive reference counting, the class it is controlling is one that derives from it.
The outcome of using such a class should be to make lifetime handling easier for users or creators of your class. If it doesn't achieve this, it is bad practice.
A legitimate example may be where you need a class to clean up all references to itself before it is destructed. In such a case, you "tell" the class whenever you are storing a reference to it (in your model, presumably) and then on exit, your class goes around nulling out these references or whatever before it calls delete this on itself.
This should all happen "behind the scenes" for users of your class.
"Why would I want to force the object to be made on the heap instead of on the stack?"
Generally when you force that it's not because you want to as such, it's because the class is part of some polymorphic hierarchy, and the only legitimate way to get one is from a factory function that returns an instance of a different derived class according to the parameters you pass it, or according to some configuration that it knows about. Then it's easy to arrange that the factory function creates them with new. There's no way that users of those classes could have them on the stack even if they wanted to, because they don't know in advance the derived type of the object they're using, only the base type.
Once you have objects like that, you know that they're destroyed with delete, and you can consider managing their lifecycle in a way that ultimately ends in delete this. You'd only do this if the object is somehow capable of knowing when it's no longer needed, which usually would be (as Mike says) because it's part of some framework that doesn't manage object lifetime explicitly, but does tell its components that they've been detached/deregistered/whatever[*].
If I remember correctly, James Kanze is your man for this. I may have misremembered, but I think he occasionally mentions that in his designs delete this isn't just used but is common. Such designs avoid shared ownership and external lifecycle management, in favour of networks of entity objects managing their own lifecycles. And where necessary, deregistering themselves from anything that knows about them prior to destroying themselves. So if you have several "tools" in a "toolbelt" then you wouldn't construe that as the toolbelt "owning" references to each of the tools, you think of the tools putting themselves in and out of the belt.
[*] Otherwise you'd have your factory return a unique_ptr or auto_ptr to encourage callers to stuff the object straight into the memory management type of their choice, or you'd return a raw pointer but provide the same encouragement via documentation. All the stuff you're used to seeing.
A good rule of thumb is not to use delete this.
Simply put, the thing that uses new should be responsible enough to use the delete when done with the object. This also avoids the problems with is on the stack/heap.
Once upon a time i was writing some plugin code. I believe i mixed build (debug for plugin, release for main code or maybe the other way around) because one part should be fast. Or maybe another situation happened. Such main is already released built on gcc and plugin is being debugged/tested on VC. When the main code deleted something from the plugin or plugin deleted something a memory issue would occur. It was because they both used different memory pools or malloc implementations. So i had a private dtor and a virtual function called deleteThis().
-edit- Now i may consider overloading the delete operator or using a smart pointer or simply just state never delete a function. It will depend and usually overloading new/delete should never be done unless you really know what your doing (dont do it). I decide to use deleteThis() because i found it easier then the C like way of thing_alloc and thing_free as deleteThis() felt like the more OOP way of doing it

What is the best way to implement smart pointers in C++?

I've been evaluating various smart pointer implementations (wow, there are a LOT out there) and it seems to me that most of them can be categorized into two broad classifications:
1) This category uses inheritance on the objects referenced so that they have reference counts and usually up() and down() (or their equivalents) implemented. IE, to use the smart pointer, the objects you're pointing at must inherit from some class the ref implementation provides.
2) This category uses a secondary object to hold the reference counts. For example, instead of pointing the smart pointer right at an object, it actually points at this meta data object... Who has a reference count and up() and down() implementations (and who usually provides a mechanism for the pointer to get at the actual object being pointed to, so that the smart pointer can properly implement operator ->()).
Now, 1 has the downside that it forces all of the objects you'd like to reference count to inherit from a common ancestor, and this means that you cannot use this to reference count objects that you don't have control over the source code to.
2 has the problem that since the count is stored in another object, if you ever have a situation that a pointer to an existing reference counted object is being converted into a reference, you probably have a bug (I.E., since the count is not in the actual object, there is no way for the new reference to get the count... ref to ref copy construction or assignment is fine, because they can share the count object, but if you ever have to convert from a pointer, you're totally hosed)...
Now, as I understand it, boost::shared_pointer uses mechanism 2, or something like it... That said, I can't quite make up my mind which is worse! I have only ever used mechanism 1, in production code... Does anyone have experience with both styles? Or perhaps there is another way thats better than both of these?
"What is the best way to implement smart pointers in C++"
Don't! Use an existing, well tested smart pointer, such as boost::shared_ptr or std::tr1::shared_ptr (std::unique_ptr and std::shared_ptr with C++ 11)
If you have to, then remember to:
use safe-bool idiom
provide an operator->
provide the strong exception guarantee
document the exception requirements your class makes on the deleter
use copy-modify-swap where possible to implement the strong exception guarantee
document whether you handle multithreading correctly
write extensive unit tests
implement conversion-to-base in such a way that it will delete on the derived pointer type (policied smart pointers / dynamic deleter smart pointers)
support getting access to raw pointer
consider cost/benifit of providing weak pointers to break cycles
provide appropriate casting operators for your smart pointers
make your constructor templated to handle constructing base pointer from derived.
And don't forget anything I may have forgotten in the above incomplete list.
Just to supply a different view to the ubiquitous Boost answer (even though it is the right answer for many uses), take a look at Loki's implementation of smart pointers. For a discourse on the design philosophy, the original creator of Loki wrote the book Modern C++ Design.
I've been using boost::shared_ptr for several years now and while you are right about the downside (no assignment via pointer possible), I think it was definitely worth it because of the huge amount of pointer-related bugs it saved me from.
In my homebrew game engine I've replaced normal pointers with shared_ptr as much as possible. The performance hit this causes is actually not so bad if you are calling most functions by reference so that the compiler does not have to create too many temporary shared_ptr instances.
Boost also has an intrusive pointer (like solution 1), that doesn't require inheriting from anything. It does require changing the pointer to class to store the reference count and provide appropriate member functions. I've used this in cases where memory efficiency was important, and didn't want the overhead of another object for each shared pointer used.
Example:
class Event {
public:
typedef boost::intrusive_ptr<Event> Ptr;
void addRef();
unsigned release();
\\ ...
private:
unsigned fRefCount;
};
inline void Event::addRef()
{
fRefCount++;
}
inline unsigned Event::release(){
fRefCount--;
return fRefCount;
}
inline void intrusive_ptr_add_ref(Event* e)
{
e->addRef();
}
inline void intrusive_ptr_release(Event* e)
{
if (e->release() == 0)
delete e;
}
The Ptr typedef is used so that I can easily switcth between boost::shared_ptr<> and boost::intrusive_ptr<> without changing any client code
If you stick with the ones that are in the standard library you will be fine.
Though there are a few other types than the ones you specified.
Shared: Where the ownership is shared between multiple objects
Owned: Where one object owns the object but transfer is allowed.
Unmovable: Where one object owns the object and it can not be transferred.
The standard library has:
std::auto_ptr
Boost has a couple more than have been adapted by tr1 (next version of the standard)
std::tr1::shared_ptr
std::tr1::weak_ptr
And those still in boost (which in relatively is a must have anyway) that hopefully make it into tr2.
boost::scoped_ptr
boost::scoped_array
boost::shared_array
boost::intrusive_ptr
See:
Smart Pointers: Or who owns you baby?
It seems to me this question is kind of like asking "Which is the best sort algorithm?" There is no one answer, it depends on your circumstances.
For my own purposes, I'm using your type 1. I don't have access to the TR1 library. I do have complete control over all the classes I need to have shared pointers to. The additional memory and time efficiency of type 1 might be pretty slight, but memory usage and speed are big issues for my code, so type 1 was a slam dunk.
On the other hand, for anyone who can use TR1, I'd think the type 2 std::tr1::shared_ptr class would be a sensible default choice, to be used whenever there isn't some pressing reason not to use it.
The problem with 2 can be worked around. Boost offers boost::shared_from_this for this same reason. In practice, it's not a big problem.
But the reason they went with your option #2 is that it can be used in all cases. Relying on inheritance isn't always an option, and then you're left with a smart pointer you can't use for half your code.
I'd have to say #2 is best, simply because it can be used in any circumstances.
Our project uses smart pointers extensively. In the beginning there was uncertainty about which pointer to use, and so one of the main authors chose an intrusive pointer in his module and the other a non-intrusive version.
In general, the differences between the two pointer types were not significant. The only exception being that early versions of our non-intrusive pointer implicitly converted from a raw pointer and this can easily lead to memory problems if the pointers are used incorrectly:
void doSomething (NIPtr<int> const &);
void foo () {
NIPtr<int> i = new int;
int & j = *i;
doSomething (&j); // Ooops - owned by two pointers! :(
}
A while ago, some refactoring resulted in some parts of the code being merged, and so a choice had to be made about which pointer type to use. The non-intrusive pointer now had the converting constructor declared as explicit and so it was decided to go with the intrusive pointer to save on the amount of code change that was required.
To our great surprise one thing we did notice was that we had an immediate performance improvement by using the intrusive pointer. We did not put much research into this, and just assumed that the difference was the cost of maintaining the count object. It is possible that other implementations of non-intrusive shared pointer have solved this problem by now.
What you are talking about are intrusive and non-intrusive smart pointers. Boost has both. boost::intrusive_ptr calls a function to decrease and increase the reference count of your object, everytime it needs to change the reference count. It's not calling member functions, but free functions. So it allows managing objects without the need to change the definition of their types. And as you say, boost::shared_ptr is non-intrusive, your category 2.
I have an answer explaining intrusive_ptr: Making shared_ptr not use delete. In short, you use it if you have an object that has already reference counting, or need (as you explain) an object that is already referenced to be owned by an intrusive_ptr.