// Copyright (C) 2004, 2006 International Business Machines and others. // All Rights Reserved. // This code is published under the Common Public License. // // $Id: IpReferenced.hpp 735 2006-06-04 06:10:05Z andreasw $ // // Authors: Carl Laird, Andreas Waechter IBM 2004-08-13 // Removed lots of debugging stuff and reformatted: Laszlo Ladanyi, IBM #ifndef CoinSmartPtr_hpp #define CoinSmartPtr_hpp #include #include namespace Coin { //######################################################################### /** ReferencedObject class. * This is part of the implementation of an intrusive smart pointer * design. This class stores the reference count of all the smart * pointers that currently reference it. See the documentation for * the SmartPtr class for more details. * * A SmartPtr behaves much like a raw pointer, but manages the lifetime * of an object, deleting the object automatically. This class implements * a reference-counting, intrusive smart pointer design, where all * objects pointed to must inherit off of ReferencedObject, which * stores the reference count. Although this is intrusive (native types * and externally authored classes require wrappers to be referenced * by smart pointers), it is a safer design. A more detailed discussion of * these issues follows after the usage information. * * Usage Example: * Note: to use the SmartPtr, all objects to which you point MUST * inherit off of ReferencedObject. * * \verbatim * * In MyClass.hpp... * * #include "CoinSmartPtr.hpp" * * class MyClass : public Coin::ReferencedObject // must derive from ReferencedObject * { * ... * } * * In my_usage.cpp... * * #include "CoinSmartPtr.hpp" * #include "MyClass.hpp" * * void func(AnyObject& obj) * { * Coin::SmartPtr ptr_to_myclass = new MyClass(...); * // ptr_to_myclass now points to a new MyClass, * // and the reference count is 1 * * ... * * obj.SetMyClass(ptr_to_myclass); * // Here, let's assume that AnyObject uses a * // SmartPtr internally here. * // Now, both ptr_to_myclass and the internal * // SmartPtr in obj point to the same MyClass object * // and its reference count is 2. * * ... * * // No need to delete ptr_to_myclass, this * // will be done automatically when the * // reference count drops to zero. * * } * * \endverbatim * * Other Notes: * The SmartPtr implements both dereference operators -> & *. * The SmartPtr does NOT implement a conversion operator to * the raw pointer. Use the GetRawPtr() method when this * is necessary. Make sure that the raw pointer is NOT * deleted. * The SmartPtr implements the comparison operators == & != * for a variety of types. Use these instead of * \verbatim * if (GetRawPtr(smrt_ptr) == ptr) // Don't use this * \endverbatim * SmartPtr's, as currently implemented, do NOT handle circular references. * For example: consider a higher level object using SmartPtrs to point * to A and B, but A and B also point to each other (i.e. A has a * SmartPtr to B and B has a SmartPtr to A). In this scenario, when the * higher level object is finished with A and B, their reference counts * will never drop to zero (since they reference each other) and they * will not be deleted. This can be detected by memory leak tools like * valgrind. If the circular reference is necessary, the problem can be * overcome by a number of techniques: * * 1) A and B can have a method that "releases" each other, that is * they set their internal SmartPtrs to NULL. * \verbatim * void AClass::ReleaseCircularReferences() * { * smart_ptr_to_B = NULL; * } * \endverbatim * Then, the higher level class can call these methods before * it is done using A & B. * * 2) Raw pointers can be used in A and B to reference each other. * Here, an implicit assumption is made that the lifetime is * controlled by the higher level object and that A and B will * both exist in a controlled manner. Although this seems * dangerous, in many situations, this type of referencing * is very controlled and this is reasonably safe. * * 3) This SmartPtr class could be redesigned with the Weak/Strong * design concept. Here, the SmartPtr is identified as being * Strong (controls lifetime of the object) or Weak (merely * referencing the object). The Strong SmartPtr increments * (and decrements) the reference count in ReferencedObject * but the Weak SmartPtr does not. In the example above, * the higher level object would have Strong SmartPtrs to * A and B, but A and B would have Weak SmartPtrs to each * other. Then, when the higher level object was done with * A and B, they would be deleted. The Weak SmartPtrs in A * and B would not decrement the reference count and would, * of course, not delete the object. This idea is very similar * to item (2), where it is implied that the sequence of events * is controlled such that A and B will not call anything using * their pointers following the higher level delete (i.e. in * their destructors!). This is somehow safer, however, because * code can be written (however expensive) to perform run-time * detection of this situation. For example, the ReferencedObject * could store pointers to all Weak SmartPtrs that are referencing * it and, in its destructor, tell these pointers that it is * dying. They could then set themselves to NULL, or set an * internal flag to detect usage past this point. * * Comments on Non-Intrusive Design: * In a non-intrusive design, the reference count is stored somewhere other * than the object being referenced. This means, unless the reference * counting pointer is the first referencer, it must get a pointer to the * referenced object from another smart pointer (so it has access to the * reference count location). In this non-intrusive design, if we are * pointing to an object with a smart pointer (or a number of smart * pointers), and we then give another smart pointer the address through * a RAW pointer, we will have two independent, AND INCORRECT, reference * counts. To avoid this pitfall, we use an intrusive reference counting * technique where the reference count is stored in the object being * referenced. */ class ReferencedObject { public: ReferencedObject() : reference_count_(0) {} virtual ~ReferencedObject() { assert(reference_count_ == 0); } inline int ReferenceCount() const { return reference_count_; } inline void AddRef() const { ++reference_count_; } inline void ReleaseRef() const { --reference_count_; } private: mutable int reference_count_; }; //######################################################################### //#define IP_DEBUG_SMARTPTR #if COIN_IPOPT_CHECKLEVEL > 2 # define IP_DEBUG_SMARTPTR #endif #ifdef IP_DEBUG_SMARTPTR # include "IpDebug.hpp" #endif /** Template class for Smart Pointers. * A SmartPtr behaves much like a raw pointer, but manages the lifetime * of an object, deleting the object automatically. This class implements * a reference-counting, intrusive smart pointer design, where all * objects pointed to must inherit off of ReferencedObject, which * stores the reference count. Although this is intrusive (native types * and externally authored classes require wrappers to be referenced * by smart pointers), it is a safer design. A more detailed discussion of * these issues follows after the usage information. * * Usage Example: * Note: to use the SmartPtr, all objects to which you point MUST * inherit off of ReferencedObject. * * \verbatim * * In MyClass.hpp... * * #include "CoinSmartPtr.hpp" * * class MyClass : public Coin::ReferencedObject // must derive from ReferencedObject * { * ... * } * * In my_usage.cpp... * * #include "CoinSmartPtr.hpp" * #include "MyClass.hpp" * * void func(AnyObject& obj) * { * SmartPtr ptr_to_myclass = new MyClass(...); * // ptr_to_myclass now points to a new MyClass, * // and the reference count is 1 * * ... * * obj.SetMyClass(ptr_to_myclass); * // Here, let's assume that AnyObject uses a * // SmartPtr internally here. * // Now, both ptr_to_myclass and the internal * // SmartPtr in obj point to the same MyClass object * // and its reference count is 2. * * ... * * // No need to delete ptr_to_myclass, this * // will be done automatically when the * // reference count drops to zero. * * } * * \endverbatim * * It is not necessary to use SmartPtr's in all cases where an * object is used that has been allocated "into" a SmartPtr. It is * possible to just pass objects by reference or regular pointers, * even if lower down in the stack a SmartPtr is to be held on to. * Everything should work fine as long as a pointer created by "new" * is immediately passed into a SmartPtr, and if SmartPtr's are used * to hold on to objects. * * Other Notes: * The SmartPtr implements both dereference operators -> & *. * The SmartPtr does NOT implement a conversion operator to * the raw pointer. Use the GetRawPtr() method when this * is necessary. Make sure that the raw pointer is NOT * deleted. * The SmartPtr implements the comparison operators == & != * for a variety of types. Use these instead of * \verbatim * if (GetRawPtr(smrt_ptr) == ptr) // Don't use this * \endverbatim * SmartPtr's, as currently implemented, do NOT handle circular references. * For example: consider a higher level object using SmartPtrs to point to * A and B, but A and B also point to each other (i.e. A has a SmartPtr * to B and B has a SmartPtr to A). In this scenario, when the higher * level object is finished with A and B, their reference counts will * never drop to zero (since they reference each other) and they * will not be deleted. This can be detected by memory leak tools like * valgrind. If the circular reference is necessary, the problem can be * overcome by a number of techniques: * * 1) A and B can have a method that "releases" each other, that is * they set their internal SmartPtrs to NULL. * \verbatim * void AClass::ReleaseCircularReferences() * { * smart_ptr_to_B = NULL; * } * \endverbatim * Then, the higher level class can call these methods before * it is done using A & B. * * 2) Raw pointers can be used in A and B to reference each other. * Here, an implicit assumption is made that the lifetime is * controlled by the higher level object and that A and B will * both exist in a controlled manner. Although this seems * dangerous, in many situations, this type of referencing * is very controlled and this is reasonably safe. * * 3) This SmartPtr class could be redesigned with the Weak/Strong * design concept. Here, the SmartPtr is identified as being * Strong (controls lifetime of the object) or Weak (merely * referencing the object). The Strong SmartPtr increments * (and decrements) the reference count in ReferencedObject * but the Weak SmartPtr does not. In the example above, * the higher level object would have Strong SmartPtrs to * A and B, but A and B would have Weak SmartPtrs to each * other. Then, when the higher level object was done with * A and B, they would be deleted. The Weak SmartPtrs in A * and B would not decrement the reference count and would, * of course, not delete the object. This idea is very similar * to item (2), where it is implied that the sequence of events * is controlled such that A and B will not call anything using * their pointers following the higher level delete (i.e. in * their destructors!). This is somehow safer, however, because * code can be written (however expensive) to perform run-time * detection of this situation. For example, the ReferencedObject * could store pointers to all Weak SmartPtrs that are referencing * it and, in its destructor, tell these pointers that it is * dying. They could then set themselves to NULL, or set an * internal flag to detect usage past this point. * * Comments on Non-Intrusive Design: * In a non-intrusive design, the reference count is stored somewhere other * than the object being referenced. This means, unless the reference * counting pointer is the first referencer, it must get a pointer to the * referenced object from another smart pointer (so it has access to the * reference count location). In this non-intrusive design, if we are * pointing to an object with a smart pointer (or a number of smart * pointers), and we then give another smart pointer the address through * a RAW pointer, we will have two independent, AND INCORRECT, reference * counts. To avoid this pitfall, we use an intrusive reference counting * technique where the reference count is stored in the object being * referenced. */ template class SmartPtr { public: /** Returns the raw pointer contained. Use to get the value of the * raw ptr (i.e. to pass to other methods/functions, etc.) Note: This * method does NOT copy, therefore, modifications using this value * modify the underlying object contained by the SmartPtr, NEVER * delete this returned value. */ T* GetRawPtr() const { return ptr_; } /** Returns true if the SmartPtr is NOT NULL. * Use this to check if the SmartPtr is not null * This is preferred to if(GetRawPtr(sp) != NULL) */ bool IsValid() const { return ptr_ != NULL; } /** Returns true if the SmartPtr is NULL. * Use this to check if the SmartPtr IsNull. * This is preferred to if(GetRawPtr(sp) == NULL) */ bool IsNull() const { return ptr_ == NULL; } private: /**@name Private Data/Methods */ //@{ /** Actual raw pointer to the object. */ T* ptr_; /** Release the currently referenced object. */ void ReleasePointer_() { if (ptr_) { ptr_->ReleaseRef(); if (ptr_->ReferenceCount() == 0) { delete ptr_; } ptr_ = NULL; } } /** Set the value of the internal raw pointer from another raw * pointer, releasing the previously referenced object if necessary. */ SmartPtr& SetFromRawPtr_(T* rhs){ ReleasePointer_(); // Release any old pointer if (rhs != NULL) { rhs->AddRef(); ptr_ = rhs; } return *this; } /** Set the value of the internal raw pointer from a SmartPtr, * releasing the previously referenced object if necessary. */ inline SmartPtr& SetFromSmartPtr_(const SmartPtr& rhs) { SetFromRawPtr_(rhs.GetRawPtr()); return (*this); } //@} public: #define dbg_smartptr_verbosity 0 /**@name Constructors/Destructors */ //@{ /** Default constructor, initialized to NULL */ SmartPtr() : ptr_(NULL) {} /** Copy constructor, initialized from copy */ SmartPtr(const SmartPtr& copy) : ptr_(NULL) { (void) SetFromSmartPtr_(copy); } /** Constructor, initialized from T* ptr */ SmartPtr(T* ptr) : ptr_(NULL) { (void) SetFromRawPtr_(ptr); } /** Destructor, automatically decrements the reference count, deletes * the object if necessary.*/ ~SmartPtr() { ReleasePointer_(); } //@} /**@name Overloaded operators. */ //@{ /** Overloaded arrow operator, allows the user to call * methods using the contained pointer. */ T* operator->() const { #if COIN_COINUTILS_CHECKLEVEL > 0 assert(ptr_); #endif return ptr_; } /** Overloaded dereference operator, allows the user * to dereference the contained pointer. */ T& operator*() const { #if COIN_IPOPT_CHECKLEVEL > 0 assert(ptr_); #endif return *ptr_; } /** Overloaded equals operator, allows the user to * set the value of the SmartPtr from a raw pointer */ SmartPtr& operator=(T* rhs) { return SetFromRawPtr_(rhs); } /** Overloaded equals operator, allows the user to * set the value of the SmartPtr from another * SmartPtr */ SmartPtr& operator=(const SmartPtr& rhs) { return SetFromSmartPtr_(rhs); } /** Overloaded equality comparison operator, allows the * user to compare the value of two SmartPtrs */ template friend bool operator==(const SmartPtr& lhs, const SmartPtr& rhs); /** Overloaded equality comparison operator, allows the * user to compare the value of a SmartPtr with a raw pointer. */ template friend bool operator==(const SmartPtr& lhs, U2* raw_rhs); /** Overloaded equality comparison operator, allows the * user to compare the value of a raw pointer with a SmartPtr. */ template friend bool operator==(U1* lhs, const SmartPtr& raw_rhs); /** Overloaded in-equality comparison operator, allows the * user to compare the value of two SmartPtrs */ template friend bool operator!=(const SmartPtr& lhs, const SmartPtr& rhs); /** Overloaded in-equality comparison operator, allows the * user to compare the value of a SmartPtr with a raw pointer. */ template friend bool operator!=(const SmartPtr& lhs, U2* raw_rhs); /** Overloaded in-equality comparison operator, allows the * user to compare the value of a SmartPtr with a raw pointer. */ template friend bool operator!=(U1* lhs, const SmartPtr& raw_rhs); //@} }; template bool ComparePointers(const U1* lhs, const U2* rhs) { if (lhs == rhs) { return true; } // If lhs and rhs point to the same object with different interfaces // U1 and U2, we cannot guarantee that the value of the pointers will // be equivalent. We can guarantee this if we convert to void*. return static_cast(lhs) == static_cast(rhs); } } // namespace Coin //############################################################################# /**@name SmartPtr friends that are overloaded operators, so they are not in the Coin namespace. */ //@{ template bool operator==(const Coin::SmartPtr& lhs, const Coin::SmartPtr& rhs) { return Coin::ComparePointers(lhs.GetRawPtr(), rhs.GetRawPtr()); } template bool operator==(const Coin::SmartPtr& lhs, U2* raw_rhs) { return Coin::ComparePointers(lhs.GetRawPtr(), raw_rhs); } template bool operator==(U1* raw_lhs, const Coin::SmartPtr& rhs) { return Coin::ComparePointers(raw_lhs, rhs.GetRawPtr()); } template bool operator!=(const Coin::SmartPtr& lhs, const Coin::SmartPtr& rhs) { return ! operator==(lhs, rhs); } template bool operator!=(const Coin::SmartPtr& lhs, U2* raw_rhs) { return ! operator==(lhs, raw_rhs); } template bool operator!=(U1* raw_lhs, const Coin::SmartPtr& rhs) { return ! operator==(raw_lhs, rhs); } //@} #define CoinReferencedObject Coin::ReferencedObject #define CoinSmartPtr Coin::SmartPtr #define CoinComparePointers Coin::ComparePointers #endif