utility/utility.htm
Douglas Gregor 6e26a5bbe7 boost/utility.hpp:
- Include boost/utility/addressof.hpp

boost/utility/addressof.hpp:
  - addressof() implementation

libs/utility/utility.htm:
  - Document addressof

libs/utility/addressof_test.cpp:
  - Test addressof()


[SVN r13414]
2002-04-10 03:31:18 +00:00

194 lines
7.8 KiB
HTML
Raw Blame History

<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<title>Header boost/utility.hpp Documentation</title>
</head>
<body bgcolor="#FFFFFF" text="#000000">
<h1><img src="../../c++boost.gif" alt="c++boost.gif (8819 bytes)" align="center" WIDTH="277" HEIGHT="86">Header
<a href="../../boost/utility.hpp">boost/utility.hpp</a></h1>
<p>The entire contents of the header <code><a href="../../boost/utility.hpp">&lt;boost/utility.hpp&gt;</a></code>
are in <code>namespace boost</code>.</p>
<h2>Contents</h2>
<ul>
<li>Class templates supporting the <a href="base_from_member.html">base-from-member
idiom</a></li>
<li>Function templates <a href="#checked_delete">checked_delete() and
checked_array_delete()</a></li>
<li>Function templates <a href="#functions next">next() and prior()</a></li>
<li>Class <a href="#Class noncopyable">noncopyable</a></li>
<li>Function template <a href="#addressof">addressof()</a></li>
<li>Function template <a href="tie.html">tie()</a> and supporting class tied.</li>
</ul>
<h2> Function templates <a name="checked_delete">checked_delete</a>() and
checked_array_delete()</h2>
<p>Deletion of a pointer to an incomplete type is an unsafe programming practice
because there is no way for the compiler to verify that the destructor is indeed
trivial.&nbsp; The checked_delete() and checked_array_delete() function
templates simply <b>delete</b> or <b>delete[]</b> their argument, but also
require that their argument be a complete type.&nbsp; They issue an appropriate
compiler error diagnostic if that requirement is not met.&nbsp; A typical
implementation is shown; other implementations may vary:</p>
<pre> template&lt; typename T &gt;
inline void checked_delete(T const volatile * x)
{
BOOST_STATIC_ASSERT( sizeof(T) ); // assert type complete at point
// of instantiation
delete x;
}
template&lt; typename T &gt;
inline void checked_array_delete(T const volatile * x)
{
BOOST_STATIC_ASSERT( sizeof(T) ); // assert type complete at point
// of instantiation
delete [] x;
}</pre>
<p>Contributed by Beman Dawes, based on a suggestion from Dave Abrahams,
generalizing an idea from Vladimir Prus, with comments from Rainer Deyke, John
Maddock, and others.</p>
<h3>Background</h3>
<p>The C++ Standard specifies that delete on a pointer to an incomplete types is
undefined behavior if the type has a non-trivial destructor in&nbsp; [expr.delete]
5.3.5 paragraph.&nbsp; No diagnostic is required.&nbsp; Some but not all
compilers issue warnings if the type is incomplete at point of deletion.</p>
<h2> <a name="functions next">Function</a> templates next() and prior()</h2>
<p>Certain data types, such as the C++ Standard Library's forward and
bidirectional iterators, do not provide addition and subtraction via operator+()
or operator-().&nbsp; This means that non-modifying computation of the next or
prior value requires a temporary, even though operator++() or operator--() is
provided.&nbsp; It also means that writing code like <code>itr+1</code> inside a
template restricts the iterator category to random access iterators.</p>
<p>The next() and prior() functions provide a simple way around these problems:</p>
<blockquote>
<pre>template &lt;class T&gt;
T next(T x) { return ++x; }
template &lt;class X&gt;
T prior(T x) { return --x; }</pre>
</blockquote>
<p>Usage is simple:</p>
<blockquote>
<pre>const std::list&lt;T&gt;::iterator p = get_some_iterator();
const std::list&lt;T&gt;::iterator prev = boost::prior(p);</pre>
</blockquote>
<p>Contributed by <a href="../../people/dave_abrahams.htm">Dave Abrahams</a>.</p>
<h2><a name="Class noncopyable">Class noncopyable</a></h2>
<p>Class <strong>noncopyable</strong> is a base class.&nbsp; Derive your own class from <strong>noncopyable</strong>
when you want to prohibit copy construction and copy assignment.</p>
<p>Some objects, particularly those which hold complex resources like files or
network connections, have no sensible copy semantics.&nbsp; Sometimes there are
possible copy semantics, but these would be of very limited usefulness and be
very difficult to implement correctly.&nbsp; Sometimes you're implementing a class that doesn't need to be copied
just yet and you don't want to take the time to write the appropriate functions.&nbsp;
Deriving from <b> noncopyable</b> will prevent the otherwise implicitly-generated
functions (which don't have the proper semantics) from becoming a trap for other programmers.</p>
<p>The traditional way to deal with these is to declare a private copy constructor and copy assignment, and then
document why this is done.&nbsp; But deriving from <b>noncopyable</b> is simpler
and clearer, and doesn't require additional documentation.</p>
<p>The program <a href="noncopyable_test.cpp">noncopyable_test.cpp</a> can be
used to verify class <b>noncopyable</b> works as expected. It has have been run successfully under
GCC 2.95, Metrowerks
CodeWarrior 5.0, and Microsoft Visual C++ 6.0 sp 3.</p>
<p>Contributed by <a href="../../people/dave_abrahams.htm">Dave Abrahams</a>.</p>
<h3>Example</h3>
<blockquote>
<pre>// inside one of your own headers ...
#include &lt;boost/utility.hpp&gt;
class ResourceLadenFileSystem : boost::noncopyable {
...</pre>
</blockquote>
<h3>Rationale</h3>
<p>Class noncopyable has protected constructor and destructor members to
emphasize that it is to be used only as a base class.&nbsp; Dave Abrahams notes
concern about the effect on compiler optimization of adding (even trivial inline)
destructor declarations. He says &quot;Probably this concern is misplaced, because
noncopyable will be used mostly for classes which own resources and thus have non-trivial destruction semantics.&quot;</p>
<h2><a name="addressof">Function template addressof()</a></h2>
<p>Function <strong>addressof()</strong> returns the address of an object.</p>
<blockquote>
<pre>
template &lt;typename T&gt; inline T* addressof(T& v);
template &lt;typename T&gt; inline const T* addressof(const T& v);
template &lt;typename T&gt; inline volatile T* addressof(volatile T& v);
template &lt;typename T&gt; inline const volatile T* addressof(const volatile T& v);
</pre>
</blockquote>
<p>C++ allows programmers to replace the unary
<strong>operator&()</strong> class member used to get the address of
an object. Getting the real address of an object requires ugly
casting tricks to avoid invoking the overloaded
<strong>operator&()</strong>. Function <strong>addressof()</strong>
provides a wrapper around the necessary code to make it easy to get an
object's real address.
</p>
<p>The program <a href="addressof_test.cpp">addressof_test.cpp</a> can be
used to verify that <b>addressof()</b> works as expected.</p>
<p>Contributed by Brad King based on ideas from discussion with Doug Gregor.</p>
<h3>Example</h3>
<blockquote>
<pre>#include &lt;boost/utility.hpp&gt;
struct useless_type {};
class nonaddressable {
useless_type operator&() const;
};
void f() {
nonaddressable x;
nonaddressable* xp = boost::addressof(x);
// nonaddressable* xpe = &amp;x; /* error */
}</pre>
</blockquote>
<h2>Class templates for the Base-from-Member Idiom</h2>
<p>See <a href="base_from_member.html">separate documentation</a>.</p>
<h2>Function template tie()</h2>
<p>See <a href="tie.html">separate documentation</a>.</p>
<hr>
<p>Revised&nbsp; <!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan
-->10 September, 2001<!--webbot bot="Timestamp" endspan i-checksum="39328"
-->
</p>
<p><EFBFBD> Copyright boost.org 1999. Permission to copy, use, modify, sell and
distribute this document is granted provided this copyright notice appears in
all copies. This document is provided &quot;as is&quot; without express or
implied warranty, and with no claim as to its suitability for any purpose.</p>
</body>
</html>