Boost-Commit :

Date view	Thread view	Subject view	Author view

Subject: [Boost-commit] svn:boost r81731 - trunk/libs/smart_ptr
From: pdimov_at_[hidden]
Date: 2012-12-05 22:20:47

Next message: eldiener_at_[hidden]: "[Boost-commit] svn:boost r81732 - trunk/boost/tti"
Previous message: pdimov_at_[hidden]: "[Boost-commit] svn:boost r81730 - in trunk: boost/smart_ptr libs/smart_ptr/test"

Author: pdimov
Date: 2012-12-05 22:20:46 EST (Wed, 05 Dec 2012)
New Revision: 81731
URL: http://svn.boost.org/trac/boost/changeset/81731

Log:
Update shared_ptr.htm.
Text files modified:
trunk/libs/smart_ptr/shared_ptr.htm | 982 +++++++++++++++++++++------------------
1 files changed, 525 insertions(+), 457 deletions(-)

Modified: trunk/libs/smart_ptr/shared_ptr.htm
==============================================================================
--- trunk/libs/smart_ptr/shared_ptr.htm (original)
+++ trunk/libs/smart_ptr/shared_ptr.htm 2012-12-05 22:20:46 EST (Wed, 05 Dec 2012)
@@ -2,410 +2,496 @@
<html>
 <head>
 <title>shared_ptr</title>
- <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
+ <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
 </head>
- <body text="#000000" bgColor="#ffffff">
- <h1><A href="../../index.htm"><IMG height="86" alt="boost.png (6897 bytes)" src="../../boost.png" width="277" align="middle"
- border="0"></A>shared_ptr class template</h1>
- Introduction 
- Best Practices 
- Synopsis 
- Members 
- Free Functions 
- Example 
- Handle/Body Idiom 
- Thread Safety 
- Frequently Asked Questions 
- Smart Pointer Timings 
- Programming Techniques
- <h2><a name="Introduction">Introduction</a></h2>
- The shared_ptr class template stores a pointer to a dynamically allocated
- object, typically with a C++ new-expression. The object pointed to is
- guaranteed to be deleted when the last shared_ptr pointing to it is
- destroyed or reset. See the example.
- Every shared_ptr meets the CopyConstructible and Assignable
- requirements of the C++ Standard Library, and so can be used in standard
- library containers. Comparison operators are supplied so that shared_ptr
+ <body>
+ <h1>boost::shared_ptr class template</h1>
+ Introduction 
+ Best Practices 
+ Synopsis 
+ Members 
+ Free Functions 
+ Example 
+ Handle/Body Idiom 
+ Thread Safety 
+ Frequently Asked Questions 
+ Smart Pointer Timings 
+ Programming Techniques
+ <h2 id="Introduction">Introduction</h2>
+ The <code>shared_ptr</code> class template stores a pointer to a dynamically allocated
+ object, typically with a C++ new-expression. The object pointed to is
+ guaranteed to be deleted when the last <code>shared_ptr</code> pointing to it is
+ destroyed or reset.
+ <blockquote>Example: <pre>shared_ptr<X> p1( new X );
+shared_ptr<void> p2( new int(5) );
+</pre></blockquote>
+
+ <code>shared_ptr</code> deletes the exact pointer that has been passed at construction time,
+ complete with its original type, regardless of the template parameter. In the second example above,
+ when <code>p2</code> is destroyed or reset, it will call <code>delete</code> on the original <code>int*</code>
+ that has been passed to the constructor, even though <code>p2</code> itself is of type
+ <code>shared_ptr<void></code> and stores a pointer of type <code>void*</code>.
+
+ Every <code>shared_ptr</code> meets the <code>CopyConstructible</code>, <code>MoveConstructible</code>,
+ <code>CopyAssignable</code> and <code>MoveAssignable</code>
+ requirements of the C++ Standard Library, and can be used in standard
+ library containers. Comparison operators are supplied so that <code>shared_ptr</code>
 works with the standard library's associative containers.
- Normally, a shared_ptr cannot correctly hold a pointer to a dynamically
- allocated array. See shared_array for
- that usage.
- Because the implementation uses reference counting, cycles of shared_ptr instances
- will not be reclaimed. For example, if main() holds a shared_ptr to
- A, which directly or indirectly holds a shared_ptr back to A,
- A's use count will be 2. Destruction of the original shared_ptr will
- leave A dangling with a use count of 1. Use weak_ptr
+ Because the implementation uses reference counting, cycles of <code>shared_ptr</code> instances
+ will not be reclaimed. For example, if <code>main()</code> holds a <code>shared_ptr</code> to
+ <code>A</code>, which directly or indirectly holds a <code>shared_ptr</code> back to <code>A</code>,
+ <code>A</code>'s use count will be 2. Destruction of the original <code>shared_ptr</code> will
+ leave <code>A</code> dangling with a use count of 1. Use weak_ptr
 to "break cycles."
- The class template is parameterized on T, the type of the object pointed
- to. shared_ptr and most of its member functions place no
- requirements on T; it is allowed to be an incomplete type, or 
- void. Member functions that do place additional requirements (constructors,
- reset) are explicitly documented below.
- shared_ptr<T> can be implicitly converted to shared_ptr
- whenever T* can be implicitly converted to U*.
- In particular, shared_ptr<T> is implicitly convertible
- to shared_ptr<T const>, to shared_ptr
- where U is an accessible base of T, and to 
- shared_ptr<void>.
- shared_ptr is now part of TR1, the first C++
- Library Technical Report. The latest draft of TR1 is available
- at the following location:
- http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2005/n1745.pdf
- (1.36Mb PDF)
- This implementation conforms to the TR1 specification, with the only exception
- that it resides in namespace <code>boost</code> instead of <code>std::tr1</code>.
- <h2><a name="BestPractices">Best Practices</a></h2>
- 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. 
- Every occurence of the new keyword in the code should have the
- form:
- <PRE>shared_ptr<T> p(new Y);</PRE>
- It is, of course, acceptable to use another smart pointer in place of shared_ptr
- above; having T and Y be the same type, or
- passing arguments to Y's constructor is also OK.
- If you observe this guideline, it naturally follows that you will have no
- explicit deletes; try/catch constructs will
- be rare.
- Avoid using unnamed shared_ptr temporaries to save typing; to
- see why this is dangerous, consider this example:
- <PRE>void f(shared_ptr<int>, int);
+ The class template is parameterized on <code>T</code>, the type of the object pointed
+ to. <code>shared_ptr</code> and most of its member functions place no
+ requirements on <code>T</code>; it is allowed to be an incomplete type, or <code>
+ void</code>. Member functions that do place additional requirements (constructors,
+ reset) are explicitly documented below.
+ <code>shared_ptr<T></code> can be implicitly converted to <code>shared_ptr</code>
+ whenever <code>T*</code> can be implicitly converted to <code>U*</code>.
+ In particular, <code>shared_ptr<T></code> is implicitly convertible
+ to <code>shared_ptr<T const></code>, to <code>shared_ptr</code>
+ where <code>U</code> is an accessible base of <code>T</code>, and to <code>
+ shared_ptr<void></code>.
+ <code>shared_ptr</code> is now part of the C++11 Standard, as <code>std::shared_ptr</code>.
+ Starting with Boost release 1.53, <code>shared_ptr</code> can be used to hold a pointer to a dynamically
+ allocated array. This is accomplished by using an array type (<code>T[]</code> or <code>T[N]</code>) as
+ the template parameter. There is almost no difference between using an unsized array, <code>T[]</code>,
+ and a sized array, <code>T[N]</code>; the latter just enables <code>operator[]</code> to perform a range check
+ on the index.
+ <blockquote>Example: <pre>shared_ptr<double[1024]> p1( new double(1024) );
+shared_ptr<double[]> p2( new double(n) );
+</pre></blockquote>
+
+ <h2 id="BestPractices">Best Practices</h2>
+ A simple guideline that nearly eliminates the possibility of memory leaks is:
+ always use a named smart pointer variable to hold the result of <code>new</code>.
+ Every occurence of the <code>new</code> keyword in the code should have the
+ form:
+ <pre>shared_ptr<T> p(new Y);</pre>
+ It is, of course, acceptable to use another smart pointer in place of <code>shared_ptr</code>
+ above; having <code>T</code> and <code>Y</code> be the same type, or
+ passing arguments to <code>Y</code>'s constructor is also OK.
+ If you observe this guideline, it naturally follows that you will have no
+ explicit <code>delete</code> statements; <code>try/catch</code> constructs will
+ be rare.
+ Avoid using unnamed <code>shared_ptr</code> temporaries to save typing; to
+ see why this is dangerous, consider this example:
+ <pre>void f(shared_ptr<int>, int);
int g();

void ok()
{
- shared_ptr<int> p(new int(2));
- f(p, g());
+ shared_ptr<int> p( new int(2) );
+ f( p, g() );
}

void bad()
{
- f(shared_ptr<int>(new int(2)), g());
+ f( shared_ptr<int>( new int(2) ), g() );
}
-</PRE>
- The function ok follows the guideline to the letter, whereas 
- bad constructs the temporary shared_ptr in place,
+</pre>
+ The function <code>ok</code> follows the guideline to the letter, whereas
+ <code>bad</code> constructs the temporary <code>shared_ptr</code> in place,
 admitting the possibility of a memory leak. Since function arguments are
- evaluated in unspecified order, it is possible for new int(2) to
- be evaluated first, g() second, and we may never get to the 
- shared_ptr constructor if g throws an exception.
- See Herb Sutter's treatment (also <A href="http://www.cuj.com/reference/articles/2002/0212/0212_sutter.htm">
- here</A>) of the issue for more information.
- The exception safety problem described above may also be eliminated by using
+ evaluated in unspecified order, it is possible for <code>new int(2)</code> to
+ be evaluated first, <code>g()</code> second, and we may never get to the
+ <code>shared_ptr</code>constructor if <code>g</code> throws an exception.
+ See Herb Sutter's treatment (also <a href="http://www.cuj.com/reference/articles/2002/0212/0212_sutter.htm">
+ here</a>) of the issue for more information.
+ The exception safety problem described above may also be eliminated by using
 the make_shared
 or allocate_shared
- factory functions defined in boost/make_shared.hpp. These factory functions also provide
- an efficiency benefit by consolidating allocations.
- <h2><a name="Synopsis">Synopsis</a></h2>
+ factory functions defined in <code>boost/make_shared.hpp</code>.
+ These factory functions also provide an efficiency benefit by consolidating allocations.
+ <h2 id="Synopsis">Synopsis</h2>
 <pre>namespace boost {

 class bad_weak_ptr: public std::exception;

- template<class T> class weak_ptr;
+ template<class T> class weak_ptr;

 template<class T> class shared_ptr {

 public:

- typedef T element_type;
+ typedef see below element_type;
+
+ shared_ptr(); // never throws
+
+ template<class Y> explicit shared_ptr(Y * p);
+ template<class Y, class D> shared_ptr(Y * p, D d);
+ template<class Y, class D, class A> shared_ptr(Y * p, D d, A a);
+
+ ~shared_ptr(); // never throws
+
+ shared_ptr(shared_ptr const & r); // never throws
+ template<class Y> shared_ptr(shared_ptr<Y> const & r); // never throws
+
+ shared_ptr(shared_ptr && r); // never throws
+ template<class Y> shared_ptr(shared_ptr<Y> && r); // never throws
+
+ template<class Y> shared_ptr(shared_ptr<Y> const & r, element_type * p); // never throws
+
+ template<class Y> explicit shared_ptr(weak_ptr<Y> const & r);
+
+ template<class Y> explicit shared_ptr(std::auto_ptr<Y> & r);
+ template<class Y> shared_ptr(std::auto_ptr<Y> && r);
+
+ template<class Y, class D> shared_ptr(std::unique_ptr<Y, D> && r);
+
+ shared_ptr & operator=(shared_ptr const & r); // never throws
+ template<class Y> shared_ptr & operator=(shared_ptr<Y> const & r); // never throws
+
+ shared_ptr & operator=(shared_ptr const && r); // never throws
+ template<class Y> shared_ptr & operator=(shared_ptr<Y> const && r); // never throws
+
+ template<class Y> shared_ptr & operator=(std::auto_ptr<Y> & r);
+ template<class Y> shared_ptr & operator=(std::auto_ptr<Y> && r);

- shared_ptr(); // never throws
- template<class Y> explicit shared_ptr(Y * p);
- template<class Y, class D> shared_ptr(Y * p, D d);
- template<class Y, class D, class A> shared_ptr(Y * p, D d, A a);
- ~shared_ptr(); // never throws
-
- shared_ptr(shared_ptr const & r); // never throws
- template<class Y> shared_ptr(shared_ptr<Y> const & r); // never throws
- template<class Y> shared_ptr(shared_ptr<Y> const & r, T * p); // never throws
- template<class Y> explicit shared_ptr(weak_ptr<Y> const & r);
- template<class Y> explicit shared_ptr(std::auto_ptr<Y> & r);
-
- shared_ptr & operator=(shared_ptr const & r); // never throws
- template<class Y> shared_ptr & operator=(shared_ptr<Y> const & r); // never throws
- template<class Y> shared_ptr & operator=(std::auto_ptr<Y> & r);
-
- void reset(); // never throws
- template<class Y> void reset(Y * p);
- template<class Y, class D> void reset(Y * p, D d);
- template<class Y, class D, class A> void reset(Y * p, D d, A a);
- template<class Y> void reset(shared_ptr<Y> const & r, T * p); // never throws
-
- T & operator*() const; // never throws
- T * operator->() const; // never throws
- T * get() const; // never throws
+ template<class Y, class D> shared_ptr & operator=(std::unique_ptr<Y, D> && r);

- bool unique() const; // never throws
- long use_count() const; // never throws
+ void reset(); // never throws

- operator unspecified-bool-type() const; // never throws
+ template<class Y> void reset(Y * p);
+ template<class Y, class D> void reset(Y * p, D d);
+ template<class Y, class D, class A> void reset(Y * p, D d, A a);

- void swap(shared_ptr & b); // never throws
+ template<class Y> void reset(shared_ptr<Y> const & r, element_type * p); // never throws
+
+ T & operator*() const; // never throws; only valid when T is not an array type
+ T * operator->() const; // never throws; only valid when T is not an array type
+
+ element_type & operator[]( std::ptrdiff_t i ) const; // never throws; only valid when T is an array type
+
+ element_type * get() const; // never throws
+
+ bool unique() const; // never throws
+ long use_count() const; // never throws
+
+ operator unspecified-bool-type() const; // never throws
+
+ void swap(shared_ptr & b); // never throws
+
+ template<class Y> bool owner_before( shared_ptr<Y> const & rhs ) const; // never throws
+ template<class Y> bool owner_before( weak_ptr<Y> const & rhs ) const; // never throws
 };

 template<class T, class U>
- bool operator==(shared_ptr<T> const & a, shared_ptr const & b); // never throws
+ bool operator==(shared_ptr<T> const & a, shared_ptr const & b); // never throws

 template<class T, class U>
- bool operator!=(shared_ptr<T> const & a, shared_ptr const & b); // never throws
+ bool operator!=(shared_ptr<T> const & a, shared_ptr const & b); // never throws

 template<class T, class U>
- bool operator<(shared_ptr<T> const & a, shared_ptr const & b); // never throws
+ bool operator<(shared_ptr<T> const & a, shared_ptr const & b); // never throws

- template<class T> void swap(shared_ptr<T> & a, shared_ptr<T> & b); // never throws
+ template<class T> void swap(shared_ptr<T> & a, shared_ptr<T> & b); // never throws

- template<class T> T * get_pointer(shared_ptr<T> const & p); // never throws
+ template<class T> typename shared_ptr<T>::element_type * get_pointer(shared_ptr<T> const & p); // never throws

 template<class T, class U>
- shared_ptr<T> static_pointer_cast(shared_ptr const & r); // never throws
+ shared_ptr<T> static_pointer_cast(shared_ptr const & r); // never throws

 template<class T, class U>
- shared_ptr<T> const_pointer_cast(shared_ptr const & r); // never throws
+ shared_ptr<T> const_pointer_cast(shared_ptr const & r); // never throws

 template<class T, class U>
- shared_ptr<T> dynamic_pointer_cast(shared_ptr const & r); // never throws
+ shared_ptr<T> dynamic_pointer_cast(shared_ptr const & r); // never throws
+
+ template<class T, class U>
+ shared_ptr<T> reinterpet_pointer_cast(shared_ptr const & r); // never throws

 template<class E, class T, class Y>
- std::basic_ostream<E, T> & operator<< (std::basic_ostream<E, T> & os, shared_ptr<Y> const & p);
+ std::basic_ostream<E, T> & operator<< (std::basic_ostream<E, T> & os, shared_ptr<Y> const & p);

 template<class D, class T>
- D * get_deleter(shared_ptr<T> const & p);
+ D * get_deleter(shared_ptr<T> const & p);
}</pre>
- <h2><a name="Members">Members</a></h2>
- <h3><a name="element_type">element_type</a></h3>
- <pre>typedef T element_type;</pre>
+ <h2 id="Members">Members</h2>
+ <h3 id="element_type">element_type</h3>
+ <pre>typedef ... element_type;</pre>
 <blockquote>
- Provides the type of the template parameter T.
+ <code>element_type</code> is <code>T</code> when <code>T</code> is not an array type,
+ and <code>U</code> when <code>T</code> is <code>U[]</code> or <code>U[N]</code>.
 </blockquote>
- <h3><a name="constructors">constructors</a></h3>
+ <h3 id="default_constructor">default constructor</h3>
 <pre>shared_ptr(); // never throws</pre>
 <blockquote>
- Effects: Constructs an empty shared_ptr.
+ Effects: Constructs an empty <code>shared_ptr</code>.
 Postconditions: <code>use_count() == 0 && get() == 0</code>.
 Throws: nothing.
 </blockquote>
- [The nothrow guarantee is important, since reset() is specified
+ [The nothrow guarantee is important, since <code>reset()</code> is specified
 in terms of the default constructor; this implies that the constructor must not
- allocate memory.]
+ allocate memory.]
+ <h3 id="pointer_constructor">pointer constructor</h3>
 <pre>template<class Y> explicit shared_ptr(Y * p);</pre>
 <blockquote>
- Requirements: p must be convertible to T *. Y
- must be a complete type. The expression <code>delete p</code> must be
- well-formed, must not invoke undefined behavior, and must not throw exceptions.
+ Requirements:
+ <code>Y</code> must be a complete type.
+ The expression <code>delete[] p</code>, when <code>T</code> is an array type, or <code>delete p</code>,
+ when <code>T</code> is not an array type,
+ must be well-formed, must not invoke undefined behavior, and must not throw exceptions.
+ When <code>T</code> is <code>U[N]</code>, <code>Y (*) [N]</code> must be convertible to <code>T*</code>;
+ when <code>T</code> is <code>U[]</code>, <code>Y (*) []</code> must be convertible to <code>T*</code>;
+ otherwise, <code>Y*</code> must be convertible to <code>T*</code>.
 
- Effects: Constructs a shared_ptr that owns the pointer p.
+ Effects:
+ When <code>T</code> is not an array type, constructs a <code>shared_ptr</code> that owns
+ the pointer <code>p</code>.
+ Otherwise, constructs a <code>shared_ptr</code> that owns
+ <code>p</code> and a deleter of an unspecified type that calls <code>delete[] p</code>.
 Postconditions: <code>use_count() == 1 && get() == p</code>.
- Throws: std::bad_alloc, or an implementation-defined
+ Throws: <code>std::bad_alloc</code>, or an implementation-defined
 exception when a resource other than memory could not be obtained.
- Exception safety: If an exception is thrown, <code>delete p</code> is
- called.
- Notes: p must be a pointer to an object that was
- allocated via a C++ new expression or be 0. The postcondition that <A href="#use_count">
- use count</A> is 1 holds even if p is 0; invoking delete
- on a pointer that has a value of 0 is harmless.
- </blockquote>
- [This constructor has been changed to a template in order to remember the actual
- pointer type passed. The destructor will call delete with the
- same pointer, complete with its original type, even when T does
- not have a virtual destructor, or is void.
- The optional intrusive counting support has been dropped as it exposes too much
- implementation details and doesn't interact well with weak_ptr.
- The current implementation uses a different mechanism, <A href="enable_shared_from_this.html">
- enable_shared_from_this</A>, to solve the "shared_ptr from 
- this" problem.]
- <a name="allocator_constructor"></a>
+ Exception safety: If an exception is thrown, the constructor calls
+ <code>delete[] p</code>, when <code>T</code> is an array type,
+ or <code>delete p</code>, when <code>T</code> is not an array type.
+ Notes: <code>p</code> must be a pointer to an object that was
+ allocated via a C++ <code>new</code> expression or be 0. The postcondition that <a href="#use_count">
+ use count</a> is 1 holds even if <code>p</code> is 0; invoking <code>delete</code>
+ on a pointer that has a value of 0 is harmless.
+ </blockquote>
+ [This constructor is a template in order to remember the actual
+ pointer type passed. The destructor will call <code>delete</code> with the
+ same pointer, complete with its original type, even when <code>T</code> does
+ not have a virtual destructor, or is <code>void</code>.]
+ <h3 id="deleter_constructor">constructors taking a deleter</h3>
 <pre>template<class Y, class D> shared_ptr(Y * p, D d);
template<class Y, class D, class A> shared_ptr(Y * p, D d, A a);</pre>
 <blockquote>
- Requirements: p must be convertible to T *. D
- must be CopyConstructible. The copy constructor and destructor
- of D must not throw. The expression <code>d(p)</code> must be
- well-formed, must not invoke undefined behavior, and must not throw exceptions. 
- A must be an Allocator, as described in section 20.1.5 (Allocator
- requirements) of the C++ Standard.
+ Requirements:
+ <code>D</code> must be <code>CopyConstructible</code>. The copy constructor and destructor
+ of <code>D</code> must not throw. The expression <code>d(p)</code> must be
+ well-formed, must not invoke undefined behavior, and must not throw exceptions.
+ <code>A</code> must be an Allocator, as described in section 20.1.5
+ (<code>Allocator requirements</code>) of the C++ Standard.
+ When <code>T</code> is <code>U[N]</code>, <code>Y (*) [N]</code> must be convertible to <code>T*</code>;
+ when <code>T</code> is <code>U[]</code>, <code>Y (*) []</code> must be convertible to <code>T*</code>;
+ otherwise, <code>Y*</code> must be convertible to <code>T*</code>.
 
- Effects: Constructs a shared_ptr that owns the pointer 
- p and the deleter d. The second constructor allocates
- memory using a copy of a.
+ Effects: Constructs a <code>shared_ptr</code> that owns the pointer <code>
+ p</code> and the deleter <code>d</code>. The second constructor allocates
+ memory using a copy of <code>a</code>.
 Postconditions: <code>use_count() == 1 && get() == p</code>.
- Throws: std::bad_alloc, or an implementation-defined
+ Throws: <code>std::bad_alloc</code>, or an implementation-defined
 exception when a resource other than memory could not be obtained.
 Exception safety: If an exception is thrown, <code>d(p)</code> is called.
- Notes: When the the time comes to delete the object pointed to by p,
- the stored copy of d is invoked with the stored copy of p
+ Notes: When the the time comes to delete the object pointed to by <code>p</code>,
+ the stored copy of <code>d</code> is invoked with the stored copy of <code>p</code>
 as an argument.
 </blockquote>
- [Custom deallocators allow a factory function returning a shared_ptr
+ [Custom deallocators allow a factory function returning a <code>shared_ptr</code>
 to insulate the user from its memory allocation strategy. Since the deallocator
 is not part of the type, changing the allocation strategy does not break source
 or binary compatibility, and does not require a client recompilation. For
- example, a "no-op" deallocator is useful when returning a shared_ptr
- to a statically allocated object, and other variations allow a shared_ptr
- to be used as a wrapper for another smart pointer, easing interoperability.
- The support for custom deallocators does not impose significant overhead. Other 
- shared_ptr features still require a deallocator to be kept.
- The requirement that the copy constructor of D does not throw comes from
- the pass by value. If the copy constructor throws, the pointer is leaked.
- Removing the requirement requires a pass by (const) reference.
- The main problem with pass by reference lies in its interaction with rvalues. A
- const reference may still cause a copy, and will require a const operator(). A
- non-const reference won't bind to an rvalue at all. A good solution to this
- problem is the rvalue reference proposed in <A href="http://std.dkuug.dk/jtc1/sc22/wg21/docs/papers/2002/n1377.htm">
- N1377</A>/N1385.]
+ example, a "no-op" deallocator is useful when returning a <code>shared_ptr</code>
+ to a statically allocated object, and other variations allow a <code>shared_ptr</code>
+ to be used as a wrapper for another smart pointer, easing interoperability.
+ The support for custom deallocators does not impose significant overhead. Other <code>
+ shared_ptr</code> features still require a deallocator to be kept.
+ The requirement that the copy constructor of <code>D</code> does not throw comes from
+ the pass by value. If the copy constructor throws, the pointer would leak.]
+ <h3 id="copy_constructor">copy and converting constructors</h3>
 <pre>shared_ptr(shared_ptr const & r); // never throws
template<class Y> shared_ptr(shared_ptr<Y> const & r); // never throws</pre>
 <blockquote>
- Effects: If r is empty, constructs an empty shared_ptr;
- otherwise, constructs a shared_ptr that shares ownership with r.
+ Requires: <code>Y*</code> should be convertible to <code>T*</code>.
+ Effects: If <code>r</code> is empty, constructs an empty <code>shared_ptr</code>;
+ otherwise, constructs a <code>shared_ptr</code> that shares ownership with <code>r</code>.
 Postconditions: <code>get() == r.get() && use_count() ==
 r.use_count()</code>.
 Throws: nothing.
 </blockquote>
- <pre>template<class Y> shared_ptr(shared_ptr<Y> const & r, T * p); // never throws</pre>
+ <h3 id="move_constructor">move constructors</h3>
+ <pre>shared_ptr(shared_ptr && r); // never throws
+template<class Y> shared_ptr(shared_ptr<Y> && r); // never throws</pre>
 <blockquote>
- Effects: constructs a shared_ptr that shares ownership with
- r and stores p.
+ Requires: <code>Y*</code> should be convertible to <code>T*</code>.
+ Effects: Move-constructs a <code>shared_ptr</code> from <code>r</code>.
+ Postconditions: <code>*this</code> contains the old value of <code>r</code>. <code>r</code> is empty and <code>r.get() == 0</code>.
+ Throws: nothing.
+ </blockquote>
+ <h3 id="aliasing_constructor">aliasing constructor</h3>
+ <pre>template<class Y> shared_ptr(shared_ptr<Y> const & r, element_type * p); // never throws</pre>
+ <blockquote>
+ Effects: constructs a <code>shared_ptr</code> that shares ownership with
+ <code>r</code> and stores <code>p</code>.
 Postconditions: <code>get() == p && use_count() == r.use_count()</code>.
 Throws: nothing.
 </blockquote>
- <pre>template<class Y> explicit shared_ptr(weak_ptr<Y> const & r);</pre>
+ <h3 id="weak_ptr_constructor">weak_ptr constructor</h3>
+ <pre>template<class Y> explicit shared_ptr(weak_ptr<Y> const & r);</pre>
 <blockquote>
- Effects: Constructs a shared_ptr that shares ownership with
- r and stores a copy of the pointer stored in r.
+ Requires: <code>Y*</code> should be convertible to <code>T*</code>.
+ Effects: Constructs a <code>shared_ptr</code> that shares ownership with
+ <code>r</code> and stores a copy of the pointer stored in <code>r</code>.
 Postconditions: <code>use_count() == r.use_count()</code>.
- Throws: bad_weak_ptr when <code>r.use_count() == 0</code>.
+ Throws: <code>bad_weak_ptr</code> when <code>r.use_count() == 0</code>.
+ Exception safety: If an exception is thrown, the constructor has no
+ effect.
+ </blockquote>
+ <h3 id="auto_ptr_constructor">auto_ptr constructors</h3>
+ <pre>template<class Y> shared_ptr(std::auto_ptr<Y> & r);
+template<class Y> shared_ptr(std::auto_ptr<Y> && r);</pre>
+ <blockquote>
+ Requires: <code>Y*</code> should be convertible to <code>T*</code>.
+ Effects: Constructs a <code>shared_ptr</code>, as if by storing a copy of <code>r.release()</code>.
+ Postconditions: <code>use_count() == 1</code>.
+ Throws: <code>std::bad_alloc</code>, or an implementation-defined
+ exception when a resource other than memory could not be obtained.
 Exception safety: If an exception is thrown, the constructor has no
 effect.
 </blockquote>
- <pre>template<class Y> shared_ptr(std::auto_ptr<Y> & r);</pre>
- <BLOCKQUOTE>
- Effects: Constructs a shared_ptr, as if by storing a copy of r.release().
+ <h3 id="unique_ptr_constructor">unique_ptr constructor</h3>
+ <pre>template<class Y, class D> shared_ptr(std::unique_ptr<Y, D> && r);</pre>
+ <blockquote>
+ Requires: <code>Y*</code> should be convertible to <code>T*</code>.
+ Effects:
+ Equivalent to <code>shared_ptr( r.release(), r.get_deleter() )</code> when <code>D</code> is not a reference type.
+ Otherwise, equivalent to <code>shared_ptr( r.release(), del )</code>, where del is a deleter
+ that stores the reference <code>rd</code> returned from <code>r.get_deleter()</code> and <code>del(p)</code> calls <code>rd(p)</code>.
 Postconditions: <code>use_count() == 1</code>.
- Throws: std::bad_alloc, or an implementation-defined
+ Throws: <code>std::bad_alloc</code>, or an implementation-defined
 exception when a resource other than memory could not be obtained.
- Exception safety: If an exception is thrown, the constructor has no
- effect.
- </BLOCKQUOTE>
- [This constructor takes a the source auto_ptr by reference and
- not by value, and cannot accept auto_ptr temporaries. This is
- by design, as the constructor offers the strong guarantee; an rvalue reference
- would solve this problem, too.]
- <h3><a name="destructor">destructor</a></h3>
+ Exception safety: If an exception is thrown, the constructor has no
+ effect.
+ </blockquote>
+ <h3 id="destructor">destructor</h3>
 <pre>~shared_ptr(); // never throws</pre>
- <BLOCKQUOTE>
- Effects:
- <UL>
- <LI>
- If *this is empty, or shares ownership with
- another shared_ptr instance (<code>use_count() > 1</code>),
- there are no side effects.
- <LI>
- Otherwise, if *this owns a pointer p
- and a deleter d, <code>d(p)</code>
- is called.
- <LI>
- Otherwise, *this owns a pointer p,
- and <code>delete p</code> is called.</LI></UL>
- Throws: nothing.
- </BLOCKQUOTE>
- <H3><a name="assignment">assignment</a></H3>
+ <blockquote>
+ Effects:
+ <ul>
+ <li>
+ If <code>*this</code> is empty, or shares ownership with
+ another <code>shared_ptr</code> instance (<code>use_count() > 1</code>),
+ there are no side effects.</li>
+ <li>
+ Otherwise, if <code>*this</code> owns a pointer <code>p</code>
+ and a deleter <code>d</code>, <code>d(p)</code>
+ is called.</li>
+ <li>
+ Otherwise, <code>*this</code> owns a pointer <code>p</code>,
+ and <code>delete p</code> is called.</li>
+ </ul>
+ Throws: nothing.
+ </blockquote>
+ <h3 id="assignment">assignment</h3>
 <pre>shared_ptr & operator=(shared_ptr const & r); // never throws
template<class Y> shared_ptr & operator=(shared_ptr<Y> const & r); // never throws
template<class Y> shared_ptr & operator=(std::auto_ptr<Y> & r);</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>shared_ptr(r).swap(*this)</code>.
- Returns: <code>*this</code>.
- Notes: The use count updates caused by the temporary object construction
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr(r).swap(*this)</code>.
+ Returns: <code>*this</code>.
+ Notes: The use count updates caused by the temporary object construction
 and destruction are not considered observable side effects, and the
 implementation is free to meet the effects (and the implied guarantees) via
- different means, without creating a temporary. In particular, in the example:
+ different means, without creating a temporary. In particular, in the example:
 <pre>shared_ptr<int> p(new int);
shared_ptr<void> q(p);
p = p;
q = p;
</pre>
 both assignments may be no-ops.
- </BLOCKQUOTE>
- <h3><a name="reset">reset</a></h3>
+ </blockquote>
+ <pre>shared_ptr & operator=(shared_ptr && r); // never throws
+template<class Y> shared_ptr & operator=(shared_ptr<Y> && r); // never throws
+template<class Y> shared_ptr & operator=(std::auto_ptr<Y> && r);
+template<class Y, class D> shared_ptr & operator=(std::unique_ptr<Y, D> && r);</pre>
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr( std::move(r) ).swap(*this)</code>.
+ Returns: <code>*this</code>.
+ </blockquote>
+ <h3 id="reset">reset</h3>
 <pre>void reset(); // never throws</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>shared_ptr().swap(*this)</code>.
- </BLOCKQUOTE>
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr().swap(*this)</code>.
+ </blockquote>
 <pre>template<class Y> void reset(Y * p);</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>shared_ptr(p).swap(*this)</code>.
- </BLOCKQUOTE>
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr(p).swap(*this)</code>.
+ </blockquote>
 <pre>template<class Y, class D> void reset(Y * p, D d);</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>shared_ptr(p, d).swap(*this)</code>.
- </BLOCKQUOTE>
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr(p, d).swap(*this)</code>.
+ </blockquote>
 <pre>template<class Y, class D, class A> void reset(Y * p, D d, A a);</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>shared_ptr(p, d, a).swap(*this)</code>.
- </BLOCKQUOTE>
- <pre>template<class Y> void reset(shared_ptr<Y> const & r, T * p); // never throws</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>shared_ptr(r, p).swap(*this)</code>.
- </BLOCKQUOTE>
- <h3><a name="indirection">indirection</a></h3>
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr(p, d, a).swap(*this)</code>.
+ </blockquote>
+ <pre>template<class Y> void reset(shared_ptr<Y> const & r, element_type * p); // never throws</pre>
+ <blockquote>
+ Effects: Equivalent to <code>shared_ptr(r, p).swap(*this)</code>.
+ </blockquote>
+ <h3 id="indirection">indirection</h3>
 <pre>T & operator*() const; // never throws</pre>
 <blockquote>
- Requirements: The stored pointer must not be 0.
+ Requirements: <code>T</code> should not be an array type. The stored pointer must not be 0.
 Returns: a reference to the object pointed to by the stored pointer.
 Throws: nothing.
 </blockquote>
 <pre>T * operator->() const; // never throws</pre>
 <blockquote>
- Requirements: The stored pointer must not be 0.
+ Requirements: <code>T</code> should not be an array type. The stored pointer must not be 0.
 Returns: the stored pointer.
 Throws: nothing.
 </blockquote>
- <h3><a name="get">get</a></h3>
- <pre>T * get() const; // never throws</pre>
+ <pre>element_type & operator[]( std::ptrdiff_t i ) const; // never throws</pre>
+ <blockquote>
+ Requirements: <code>T</code> should be an array type. The stored pointer must not be 0.
+ <code>i >= 0</code>. If <code>T</code> is <code>U[N]</code>, <code>i < N</code>.
+ Returns: <code>get()[ i ]</code>.
+ Throws: nothing.
+ </blockquote>
+ <h3 id="get">get</h3>
+ <pre>element_type * get() const; // never throws</pre>
 <blockquote>
 Returns: the stored pointer.
 Throws: nothing.
 </blockquote>
- <h3><a name="unique">unique</a></h3>
+ <h3 id="unique">unique</h3>
 <pre>bool unique() const; // never throws</pre>
 <blockquote>
 Returns: <code>use_count() == 1</code>.
 Throws: nothing.
- Notes: <code>unique()</code> may be faster than <code>use_count()</code>.
+ Notes: <code>unique()</code> may be faster than <code>use_count()</code>.
 If you are using <code>unique()</code> to implement copy on write, do not rely
- on a specific value when the stored pointer is zero.
+ on a specific value when the stored pointer is zero.
 </blockquote>
- <h3><a name="use_count">use_count</a></h3>
+ <h3 id="use_count">use_count</h3>
 <pre>long use_count() const; // never throws</pre>
 <blockquote>
- Returns: the number of shared_ptr objects, *this included,
- that share ownership with *this, or 0 when *this
- is empty.
+ Returns: the number of <code>shared_ptr</code> objects, <code>*this</code> included,
+ that share ownership with <code>*this</code>, or 0 when <code>*this</code>
+ is empty.
 Throws: nothing.
- Notes: <code>use_count()</code> is not necessarily efficient. Use only
- for debugging and testing purposes, not for production code.
+ Notes: <code>use_count()</code> is not necessarily efficient. Use only
+ for debugging and testing purposes, not for production code.
 </blockquote>
- <h3><a name="conversions">conversions</a></h3>
+ <h3 id="conversions">conversions</h3>
 <pre>operator unspecified-bool-type () const; // never throws</pre>
 <blockquote>
 Returns: an unspecified value that, when used in boolean contexts, is
 equivalent to <code>get() != 0</code>.
 Throws: nothing.
- Notes: This conversion operator allows shared_ptr objects to be
+ Notes: This conversion operator allows <code>shared_ptr</code> objects to be
 used in boolean contexts, like <code>if (p && p->valid()) {}</code>.
 The actual target type is typically a pointer to a member function, avoiding
- many of the implicit conversion pitfalls.
+ many of the implicit conversion pitfalls.
 </blockquote>
- [The conversion to bool is not merely syntactic sugar. It allows shared_ptrs
- to be declared in conditions when using dynamic_pointer_cast
- or weak_ptr::lock.]
- <h3><a name="swap">swap</a></h3>
+ [The conversion to bool is not merely syntactic sugar. It allows <code>shared_ptr</code>s
+ to be declared in conditions when using dynamic_pointer_cast
+ or weak_ptr::lock.]
+ <h3 id="swap">swap</h3>
 <pre>void swap(shared_ptr & b); // never throws</pre>
 <blockquote>
 Effects: Exchanges the contents of the two smart pointers.
 Throws: nothing.
 </blockquote>
- <h2><a name="functions">Free Functions</a></h2>
- <h3><a name="comparison">comparison</a></h3>
+ <h2 id="functions">Free Functions</h2>
+ <h3 id="comparison">comparison</h3>
 <pre>template<class T, class U>
 bool operator==(shared_ptr<T> const & a, shared_ptr const & b); // never throws</pre>
 <blockquote>
@@ -422,147 +508,143 @@
 bool operator<(shared_ptr<T> const & a, shared_ptr const & b); // never throws</pre>
 <blockquote>
 Returns: an unspecified value such that
- <UL>
- <LI>
- operator< is a strict weak ordering as described in section 25.3 <code>[lib.alg.sorting]</code>
- of the C++ standard;
- <LI>
- under the equivalence relation defined by operator<, <code>!(a
- < b) && !(b < a)</code>, two shared_ptr instances
- are equivalent if and only if they share ownership or are both empty.</LI></UL>
- Throws: nothing.
- Notes: Allows shared_ptr objects to be used as keys in
- associative containers.
- </blockquote>
- [Operator< has been preferred over a std::less 
- specialization for consistency and legality reasons, as std::less
- is required to return the results of operator<, and many
- standard algorithms use operator< instead of std::less
- for comparisons when a predicate is not supplied. Composite objects, like std::pair,
- also implement their operator< in terms of their contained
- subobjects' operator<.
- The rest of the comparison operators are omitted by design.]
- <h3><a name="free-swap">swap</a></h3>
+ <ul>
+ <li>
+ <code>operator<</code> is a strict weak ordering as described in section 25.3 <code>[lib.alg.sorting]</code>
+ of the C++ standard;</li>
+ <li>
+ under the equivalence relation defined by <code>operator<</code>, <code>!(a
+ < b) && !(b < a)</code>, two <code>shared_ptr</code> instances
+ are equivalent if and only if they share ownership or are both empty.</li></ul>
+ Throws: nothing.
+ Notes: Allows <code>shared_ptr</code> objects to be used as keys in
+ associative containers.
+ </blockquote>
+ [<code>Operator<</code> has been preferred over a <code>std::less</code>
+ specialization for consistency and legality reasons, as <code>std::less</code>
+ is required to return the results of <code>operator<</code>, and many
+ standard algorithms use <code>operator<</code> instead of <code>std::less</code>
+ for comparisons when a predicate is not supplied. Composite objects, like <code>std::pair</code>,
+ also implement their <code>operator<</code> in terms of their contained
+ subobjects' <code>operator<</code>.
+ The rest of the comparison operators are omitted by design.]
+ <h3 id="free-swap">swap</h3>
 <pre>template<class T>
 void swap(shared_ptr<T> & a, shared_ptr<T> & b); // never throws</pre>
- <BLOCKQUOTE>
- Effects: Equivalent to <code>a.swap(b)</code>.
- Throws: nothing.
- Notes: Matches the interface of std::swap. Provided as an aid to
- generic programming.
- </BLOCKQUOTE>
- [swap is defined in the same namespace as shared_ptr
- as this is currently the only legal way to supply a swap function
- that has a chance to be used by the standard library.]
- <h3><a name="get_pointer">get_pointer</a></h3>
+ <blockquote>
+ Effects: Equivalent to <code>a.swap(b)</code>.
+ Throws: nothing.
+ Notes: Matches the interface of <code>std::swap</code>. Provided as an aid to
+ generic programming.
+ </blockquote>
+ [<code>swap</code> is defined in the same namespace as <code>shared_ptr</code>
+ as this is currently the only legal way to supply a <code>swap</code> function
+ that has a chance to be used by the standard library.]
+ <h3 id="get_pointer">get_pointer</h3>
 <pre>template<class T>
- T * get_pointer(shared_ptr<T> const & p); // never throws</pre>
- <BLOCKQUOTE>
- Returns: <code>p.get()</code>.
- Throws: nothing.
- Notes: Provided as an aid to generic programming. Used by <A href="../bind/mem_fn.html">
- mem_fn</A>.
- </BLOCKQUOTE>
- <h3><a name="static_pointer_cast">static_pointer_cast</a></h3>
+ typename shared_ptr<T>::element_type * get_pointer(shared_ptr<T> const & p); // never throws</pre>
+ <blockquote>
+ Returns: <code>p.get()</code>.
+ Throws: nothing.
+ Notes: Provided as an aid to generic programming. Used by <a href="../bind/mem_fn.html">
+ mem_fn</a>.
+ </blockquote>
+ <h3 id="static_pointer_cast">static_pointer_cast</h3>
 <pre>template<class T, class U>
 shared_ptr<T> static_pointer_cast(shared_ptr const & r); // never throws</pre>
- <BLOCKQUOTE>
- Requires: The expression <code>static_cast<T*>(r.get())</code>
- must be well-formed.
- Returns: If r is empty, an empty shared_ptr<T>;
- otherwise, a shared_ptr<T> object that stores a copy of <code>
- static_cast<T*>(r.get())</code> and shares ownership with r.
- Throws: nothing.
- Notes: the seemingly equivalent expression
- <code>shared_ptr<T>(static_cast<T*>(r.get()))</code>
- will eventually result in undefined behavior, attempting to delete the same
+ <blockquote>
+ Requires: The expression <code>static_cast<T*>( (U*)0 )</code>
+ must be well-formed.
+ Returns: <code>shared_ptr<T>( r, static_cast<typename shared_ptr<T>::element_type*>(r.get()) )</code>.
+ Throws: nothing.
+ Notes: the seemingly equivalent expression
+ <code>shared_ptr<T>(static_cast<T*>(r.get()))</code>
+ will eventually result in undefined behavior, attempting to delete the same
 object twice.
- </BLOCKQUOTE>
- <h3><a name="const_pointer_cast">const_pointer_cast</a></h3>
+ </blockquote>
+ <h3 id="const_pointer_cast">const_pointer_cast</h3>
 <pre>template<class T, class U>
 shared_ptr<T> const_pointer_cast(shared_ptr const & r); // never throws</pre>
- <BLOCKQUOTE>
- Requires: The expression <code>const_cast<T*>(r.get())</code>
- must be well-formed.
- Returns: If r is empty, an empty shared_ptr<T>;
- otherwise, a shared_ptr<T> object that stores a copy of <code>
- const_cast<T*>(r.get())</code> and shares ownership with r.
- Throws: nothing.
- Notes: the seemingly equivalent expression
- <code>shared_ptr<T>(const_cast<T*>(r.get()))</code>
- will eventually result in undefined behavior, attempting to delete the same
- object twice.
- </BLOCKQUOTE>
- <h3><a name="dynamic_pointer_cast">dynamic_pointer_cast</a></h3>
+ <blockquote>
+ Requires: The expression <code>const_cast<T*>( (U*)0 )</code>
+ must be well-formed.
+ Returns: <code>shared_ptr<T>( r, const_cast<typename shared_ptr<T>::element_type*>(r.get()) )</code>.
+ Throws: nothing.
+ </blockquote>
+ <h3 id="dynamic_pointer_cast">dynamic_pointer_cast</h3>
 <pre>template<class T, class U>
 shared_ptr<T> dynamic_pointer_cast(shared_ptr const & r);</pre>
- <BLOCKQUOTE>
- Requires: The expression <CODE>dynamic_cast<T*>(r.get())</CODE>
- must be well-formed and its behavior defined.
- Returns:
- <UL>
- <LI>
- When <CODE>dynamic_cast<T*>(r.get())</CODE> returns a nonzero value, a 
- shared_ptr<T> object that stores a copy of it and shares
- ownership with r;
- <LI>
- Otherwise, an empty shared_ptr<T> object.</LI></UL>
- Throws: nothing.
- Notes: the seemingly equivalent expression
- <CODE>shared_ptr<T>(dynamic_cast<T*>(r.get()))</CODE>
- will eventually result in undefined behavior, attempting to delete the same
- object twice.
- </BLOCKQUOTE>
- <h3><a name="insertion-operator">operator<<</a></h3>
+ <blockquote>
+ Requires: The expression <code>dynamic_cast<T*>( (U*)0 )</code>
+ must be well-formed.
+ Returns:
+ <ul>
+ <li>
+ When <code>dynamic_cast<typename shared_ptr<T>::element_type*>(r.get())</code> returns a nonzero value <code>p</code>,
+ <code>shared_ptr<T>(r, p)</code>;</li>
+ <li>
+ Otherwise, <code>shared_ptr<T>()</code>.</li></ul>
+ Throws: nothing.
+ </blockquote>
+ <h3 id="reinterpret_pointer_cast">reinterpret_pointer_cast</h3>
+ <pre>template<class T, class U>
+ shared_ptr<T> reinterpret_pointer_cast(shared_ptr const & r); // never throws</pre>
+ <blockquote>
+ Requires: The expression <code>reinterpret_cast<T*>( (U*)0 )</code>
+ must be well-formed.
+ Returns: <code>shared_ptr<T>( r, reinterpret_cast<typename shared_ptr<T>::element_type*>(r.get()) )</code>.
+ Throws: nothing.
+ </blockquote>
+ <h3 id="insertion-operator">operator<<</h3>
 <pre>template<class E, class T, class Y>
 std::basic_ostream<E, T> & operator<< (std::basic_ostream<E, T> & os, shared_ptr<Y> const & p);</pre>
- <BLOCKQUOTE>
- Effects: <code>os << p.get();</code>.
- Returns: os.
- </BLOCKQUOTE>
- <h3><a name="get_deleter">get_deleter</a></h3>
+ <blockquote>
+ Effects: <code>os << p.get();</code>.
+ Returns: <code>os</code>.
+ </blockquote>
+ <h3 id="get_deleter">get_deleter</h3>
 <pre>template<class D, class T>
 D * get_deleter(shared_ptr<T> const & p);</pre>
- <BLOCKQUOTE>
- Returns: If *this owns a deleter d
- of type (cv-unqualified) D, returns <code>&d</code>;
- otherwise returns 0.
- Throws: nothing.
- </BLOCKQUOTE>
- <h2><a name="example">Example</a></h2>
- See shared_ptr_example.cpp for a
- complete example program. The program builds a std::vector and std::set
- of shared_ptr objects.
- Note that after the containers have been populated, some of the shared_ptr
+ <blockquote>
+ Returns: If <code>*this</code> owns a deleter <code>d</code>
+ of type (cv-unqualified) <code>D</code>, returns <code>&d</code>;
+ otherwise returns 0.
+ Throws: nothing.
+ </blockquote>
+ <h2 id="example">Example</h2>
+ See shared_ptr_example.cpp for a
+ complete example program. The program builds a <code>std::vector</code> and <code>std::set</code>
+ of <code>shared_ptr</code> objects.
+ Note that after the containers have been populated, some of the <code>shared_ptr</code>
 objects will have a use count of 1 rather than a use count of 2, since the set
- is a std::set rather than a std::multiset, and thus does not
+ is a <code>std::set</code> rather than a <code>std::multiset</code>, and thus does not
 contain duplicate entries. Furthermore, the use count may be even higher at
- various times while push_back and insert container operations are
+ various times while <code>push_back</code> and <code>insert</code> container operations are
 performed. 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.
- <h2><a name="Handle/Body">Handle/Body</a> Idiom</h2>
- One common usage of shared_ptr is to implement a handle/body (also called
+ <h2 id="HandleBody">Handle/Body Idiom</h2>
+ One common usage of <code>shared_ptr</code> is to implement a handle/body (also called
 pimpl) idiom which avoids exposing the body (implementation) in the header
 file.
- The shared_ptr_example2_test.cpp
- sample program includes a header file, shared_ptr_example2.hpp,
- which uses a shared_ptr<> to an incomplete type to hide the
+ The shared_ptr_example2_test.cpp
+ sample program includes a header file, shared_ptr_example2.hpp,
+ which uses a <code>shared_ptr</code> to an incomplete type to hide the
 implementation. The instantiation of member functions which require a complete
- type occurs in the shared_ptr_example2.cpp
+ type occurs in the shared_ptr_example2.cpp
 implementation file. Note that there is no need for an explicit destructor.
- Unlike ~scoped_ptr, ~shared_ptr does not require that T be a complete
+ Unlike <code>~scoped_ptr</code>, <code>~shared_ptr</code> does not require that <code>T</code> be a complete
 type.
- <h2><a name="ThreadSafety">Thread Safety</a></h2>
- shared_ptr objects offer the same level of thread safety as
- built-in types. A shared_ptr instance can be "read" (accessed
- using only const operations) simultaneously by multiple threads. Different shared_ptr
- instances can be "written to" (accessed using mutable operations such as operator=
- or reset) simultaneosly by multiple threads (even
+ <h2 id="ThreadSafety">Thread Safety</h2>
+ <code>shared_ptr</code> objects offer the same level of thread safety as
+ built-in types. A <code>shared_ptr</code> instance can be "read" (accessed
+ using only const operations) simultaneously by multiple threads. Different <code>shared_ptr</code>
+ instances can be "written to" (accessed using mutable operations such as <code>operator=
+ </code>or <code>reset</code>) simultaneosly by multiple threads (even
 when these instances are copies, and share the same reference count
 underneath.)
- Any other simultaneous accesses result in undefined behavior.
- Examples:
+ Any other simultaneous accesses result in undefined behavior.
+ Examples:
 <pre>shared_ptr<int> p(new int(42));

//--- Example 1 ---
@@ -606,89 +688,78 @@
p3.reset(new int(2)); // undefined, multiple writes
</pre>
  
- Starting with Boost release 1.33.0, shared_ptr uses a lock-free
- implementation on the following platforms:
- <UL>
- <LI>
- GNU GCC on x86 or x86-64;
- <LI>
- GNU GCC on IA64;
- <LI>
- Metrowerks CodeWarrior on PowerPC;
- <LI>
- GNU GCC on PowerPC;
- <LI>
- Windows.</LI></UL>
- If your program is single-threaded and does not link to any libraries that might
- have used shared_ptr in its default configuration, you can 
- #define the macro BOOST_SP_DISABLE_THREADS on a
- project-wide basis to switch to ordinary non-atomic reference count updates.
- (Defining BOOST_SP_DISABLE_THREADS in some, but not all,
+ Starting with Boost release 1.33.0, <code>shared_ptr</code> uses a lock-free
+ implementation on most common platforms.
+ If your program is single-threaded and does not link to any libraries that might
+ have used <code>shared_ptr</code> in its default configuration, you can <code>
+ #define</code> the macro <code>BOOST_SP_DISABLE_THREADS</code> on a
+ project-wide basis to switch to ordinary non-atomic reference count updates.
+ (Defining <code>BOOST_SP_DISABLE_THREADS</code> in some, but not all,
 translation units is technically a violation of the One Definition Rule and
 undefined behavior. Nevertheless, the implementation attempts to do its best to
 accommodate the request to use non-atomic updates in those translation units.
- No guarantees, though.)
- You can define the macro BOOST_SP_USE_PTHREADS to turn off the
- lock-free platform-specific implementation and fall back to the generic pthread_mutex_t-based
- code.
- <h2><a name="FAQ">Frequently Asked Questions</a></h2>
- Q. There are several variations of shared pointers, with different
+ No guarantees, though.)
+ You can define the macro <code>BOOST_SP_USE_PTHREADS</code> to turn off the
+ lock-free platform-specific implementation and fall back to the generic
+ <code>pthread_mutex_t</code>-based code.
+ <h2 id="FAQ">Frequently Asked Questions</h2>
+ Q. 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?
- 
- A. An important goal of shared_ptr is to provide a
+ as to find the most suitable for the job at hand?
+ 
+ A. An important goal of <code>shared_ptr</code> is to provide a
 standard shared-ownership pointer. Having a single pointer type is important
 for stable library interfaces, since different shared pointers typically cannot
 interoperate, i.e. a reference counted pointer (used by library A) cannot share
- ownership with a linked pointer (used by library B.) 
- 
- Q. Why doesn't shared_ptr have template parameters supplying
- traits or policies to allow extensive user customization?
- 
- A. Parameterization discourages users. The shared_ptr template is
+ ownership with a linked pointer (used by library B.)
+ 
+ Q. Why doesn't <code>shared_ptr</code> have template parameters supplying
+ traits or policies to allow extensive user customization?
+ 
+ A. Parameterization discourages users. The <code>shared_ptr</code> template is
 carefully crafted to meet common needs without extensive parameterization. Some
 day a highly configurable smart pointer may be invented that is also very easy
- to use and very hard to misuse. Until then, shared_ptr is the smart
+ to use and very hard to misuse. Until then, <code>shared_ptr</code> is the smart
 pointer of choice for a wide range of applications. (Those interested in policy
- based smart pointers should read <A href="http://www.awprofessional.com/bookstore/product.asp?isbn=0201704315&rl=1">
- Modern C++ Design</A> by Andrei Alexandrescu.) 
- 
- Q. I am not convinced. Default parameters can be used where appropriate
- to hide the complexity. Again, why not policies?
- 
- A. Template parameters affect the type. See the answer to the first
- question above. 
- 
- Q. Why doesn't shared_ptr use a linked list implementation?
- 
+ based smart pointers should read <a href="http://www.awprofessional.com/bookstore/product.asp?isbn=0201704315&rl=1">
+ Modern C++ Design</a> by Andrei Alexandrescu.)
+ 
+ Q. I am not convinced. Default parameters can be used where appropriate
+ to hide the complexity. Again, why not policies?
+ 
+ A. Template parameters affect the type. See the answer to the first
+ question above.
+ 
+ Q. Why doesn't <code>shared_ptr</code> use a linked list implementation?
+ 
 A. A linked list implementation does not offer enough advantages to
- offset the added cost of an extra pointer. See timings
+ offset the added cost of an extra pointer. See timings
 page. In addition, it is expensive to make a linked list implementation thread
- safe. 
- 
- Q. Why doesn't shared_ptr (or any of the other Boost smart
- pointers) supply an automatic conversion to T*?
- 
- A. Automatic conversion is believed to be too error prone. 
- 
- Q. Why does shared_ptr supply use_count()?
- 
+ safe.
+ 
+ Q. Why doesn't <code>shared_ptr</code> (or any of the other Boost smart
+ pointers) supply an automatic conversion to <code>T*</code>?
+ 
+ A. Automatic conversion is believed to be too error prone.
+ 
+ Q. Why does <code>shared_ptr</code> supply use_count()?
+ 
 A. As an aid to writing test cases and debugging displays. One of the
- progenitors had use_count(), and it was useful in tracking down bugs in a
- complex project that turned out to have cyclic-dependencies. 
- 
- Q. Why doesn't shared_ptr specify complexity requirements?
- 
+ progenitors had <code>use_count()</code>, and it was useful in tracking down bugs in a
+ complex project that turned out to have cyclic-dependencies.
+ 
+ Q. Why doesn't <code>shared_ptr</code> specify complexity requirements?
+ 
 A. Because complexity requirements limit implementors and complicate the
- specification without apparent benefit to shared_ptr users. For example,
+ specification without apparent benefit to <code>shared_ptr</code> users. For example,
 error-checking implementations might become non-conforming if they had to meet
- stringent complexity requirements. 
- 
- Q. Why doesn't shared_ptr provide a release() function?
- 
- A. shared_ptr cannot give away ownership unless it's unique()
- because the other copy will still destroy the object.
+ stringent complexity requirements.
+ 
+ Q. Why doesn't <code>shared_ptr</code> provide a <code>release()</code> function?
+ 
+ A. <code>shared_ptr</code> cannot give away ownership unless it's <code>unique()</code>
+ because the other copy will still destroy the object.
 Consider:
 <blockquote><pre>shared_ptr<int> a(new int);
shared_ptr<int> b(a); // a.use_count() == b.use_count() == 2
@@ -698,25 +769,22 @@
// Who owns p now? b will still call delete on it in its destructor.</pre>
 </blockquote>
 Furthermore, the pointer returned by <code>release()</code> would be difficult
- to deallocate reliably, as the source shared_ptr could have been created
- with a custom deleter. 
+ to deallocate reliably, as the source <code>shared_ptr</code> could have been created
+ with a custom deleter.
 
- Q. Why is <code>operator->()</code> const, but its return value is a
- non-const pointer to the element type?
- 
+ Q. Why is <code>operator->()</code> const, but its return value is a
+ non-const pointer to the element type?
+ 
 A. Shallow copy pointers, including raw pointers, typically don't
 propagate constness. It makes little sense for them to do so, as you can always
 obtain a non-const pointer from a const one and then proceed to modify the
- object through it.shared_ptr is "as close to raw pointers as possible
- but no closer". 
- 
- 
- <hr>
- 
- $Date$
+ object through it. <code>shared_ptr</code> is "as close to raw pointers as possible
+ but no closer".
+ 
+ <hr />
 Copyright 1999 Greg Colvin and Beman Dawes. Copyright 2002 Darin Adler.
- Copyright 2002-2005 Peter Dimov. Distributed under the Boost Software License,
- Version 1.0. See accompanying file LICENSE_1_0.txt
- or copy at http://www.boost.org/LICENSE_1_0.txt.
+ Copyright 2002-2005, 2012 Peter Dimov. Distributed under the Boost Software License,
+ Version 1.0. See accompanying file LICENSE_1_0.txt
+ or copy at http://www.boost.org/LICENSE_1_0.txt.
 </body>
</html>

Next message: eldiener_at_[hidden]: "[Boost-commit] svn:boost r81732 - trunk/boost/tti"
Previous message: pdimov_at_[hidden]: "[Boost-commit] svn:boost r81730 - in trunk: boost/smart_ptr libs/smart_ptr/test"

Date view	Thread view	Subject view	Author view

Boost-Commit list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk