In addition, the library contains the following supporting utility functions and classes:. Revision History Changes in 1. Changes in 1. Since the previous object needs to be deleted, T must be a complete type.
Returns the stored pointer. T need not be a complete type. Exchanges the contents of the two smart pointers. Best Practices A simple guideline that nearly eliminates the possibility of memory leaks is: always use a named smart pointer variable to hold the result of new.
Code Example 3. Requires Y must be a complete type. Exception safety If an exception is thrown, the constructor calls delete[] p , when T is an array type, or delete p , when T is not an array type. Requires D must be CopyConstructible. Exception safety If an exception is thrown, d p is called. Exception safety If an exception is thrown, the constructor has no effect. Effects When r. Requires T should not be an array type. Returns get. Requires T should be an array type. Returns get [i]. Returns The stored pointer.
Returns get! Effects Exchanges the contents of the two smart pointers. Returns An unspecified hash value such that two instances that share ownership have the same hash value. Returns a. Returns p. Effects Equivalent to a. Returns os. Atomic Access Note. Returns false. Code Example 4. Code Example 5. Code Example 6. Code Example 7. Code Example 8. Frequently Asked Questions There are several variations of shared pointers, with different tradeoffs; why does the smart pointer library supply only a single implementation?
Template parameters affect the type. See the answer to the first question above. Automatic conversion is believed to be too error prone. Postconditions r is empty. Returns expired? Requires A shall be an allocator. Effects Allocates memory for an object of type T or n objects of U if T is an array type of the form U[] and n is determined by arguments, as specified by the concrete overload.
Postconditions r. Remarks Performs no more than one memory allocation. Array elements are initialized in ascending order of their addresses. Constraints T is not an array. Constraints T is an array of unknown bounds.
Constraints T is an array of known bounds. Constraints T is not an array, or is an array of known bounds. Throws An exception thrown from A::allocate , or from the initialization of the object.
Remarks When an object of an array type is specified to be initialized to a value of the same type v , this shall be interpreted to mean that each array element of the object is initialized to the corresponding element from v. A type that satisfies NullablePointer.
Effects Initializes the stored allocator from a. Effects Destroys the objects and deallocates the storage referenced by p , using the stored allocator. Effects if p! Postconditions get equals the old value of r. Effects if get!
Requirements get! Returns a! Effects Destroys the counter object. Effects Does nothing, reference counter is not modified. Returns The current value of the reference counter. Effects Increments the reference counter. Effects Decrements the reference counter. Code Example 9. Code Example Here p2. Rationale Boost smart pointers usually overload those functions to provide a mechanism to emulate pointers casts.
Example The following example demonstrates how the generic pointer casts help us create pointer independent code.
Rationale When building pointer independent classes, like memory managers, allocators, or containers, there is often a need to define pointers generically, so that if a template parameter represents a pointer for example, a raw or smart pointer to an int , we can define another pointer of the same type to another pointee a raw or smart pointer to a float.
Remarks The old value of w is not preserved in either case. Returns u. Appendix A: Smart Pointer Programming Techniques Using incomplete classes for implementation hiding A proven technique that works in C, too for separating interface from implementation is to use a pointer to an incomplete class as an opaque handle:. Preventing delete px. The same technique works for any object known to outlive the pointer.
The solution is to keep a weak pointer to this as a member in impl :. Delayed deallocation In some situations, a single px. Embedded attached: the count is a member of the object pointed to.
Jonathan Sternberg Jonathan Sternberg 5, 7 7 gold badges 33 33 silver badges 57 57 bronze badges. Post the code, it's impossible to tell for sure what's wrong without that. What compiler are you using? Which compilation options? There are ways to disable RTTI on almost every compiler, are you doing it by chance? Add a comment. Active Oldest Votes. Improve this answer. Ross Smith Ross Smith 3, 1 1 gold badge 26 26 silver badges 22 22 bronze badges. It doesn't need a virtual destructor highly recommended though!
Sign up or log in Sign up using Google. Sign up using Facebook. Sign up using Email and Password. Post as a guest Name. Email Required, but never shown. Notes: Matches the interface of std::swap. Provided as an aid to generic programming. Notes: Provided as an aid to generic programming. More complicated yet, the container operations may throw exceptions under a variety of circumstances.
Getting the memory management and exception handling in this example right without a smart pointer would be a nightmare. Note that there is no need for an explicit destructor.
Nevertheless, the implementation attempts to do its best to accommodate the request to use non-atomic updates in those translation units. No guarantees, though. There are several variations of shared pointers, with different tradeoffs; why does the smart pointer library supply only a single implementation?
It would be useful to be able to experiment with each type so as to find the most suitable for the job at hand? Having a single pointer type is important for stable library interfaces, since different shared pointers typically cannot interoperate, i. Parameterization discourages users. Some day a highly configurable smart pointer may be invented that is also very easy to use and very hard to misuse.
I am not convinced. Default parameters can be used where appropriate to hide the complexity. Again, why not policies? Template parameters affect the type. See the answer to the first question above. A linked list implementation does not offer enough advantages to offset the added cost of an extra pointer. See timings page. In addition, it is expensive to make a linked list implementation thread safe.
0コメント