boost_1_45_0/doc/html/function/faq.html

<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=US-ASCII">
<title>Frequently Asked Questions</title>
<link rel="stylesheet" href="../../../doc/src/boostbook.css" type="text/css">
<meta name="generator" content="DocBook XSL Stylesheets V1.75.2">
<link rel="home" href="../index.html" title="The Boost C++ Libraries BoostBook Documentation Subset">
<link rel="up" href="../function.html" title="Chapter&#160;7.&#160;Boost.Function">
<link rel="prev" href="../boost/function_equal.html" title="Function template function_equal">
<link rel="next" href="misc.html" title="Miscellaneous Notes">
</head>
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
<table cellpadding="2" width="100%"><tr>
<td valign="top"><img alt="Boost C++ Libraries" width="277" height="86" src="../../../boost.png"></td>
<td align="center"><a href="../../../index.html">Home</a></td>
<td align="center"><a href="../../../libs/libraries.htm">Libraries</a></td>
<td align="center"><a href="http://www.boost.org/users/people.html">People</a></td>
<td align="center"><a href="http://www.boost.org/users/faq.html">FAQ</a></td>
<td align="center"><a href="../../../more/index.htm">More</a></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="../boost/function_equal.html"><img src="../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../function.html"><img src="../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../index.html"><img src="../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="misc.html"><img src="../../../doc/src/images/next.png" alt="Next"></a>
</div>
<div class="section">
<div class="titlepage"><div><div><h2 class="title" style="clear: both">
<a name="function.faq"></a>Frequently Asked Questions</h2></div></div></div>
<div class="qandaset">
<a name="id1284468"></a><dl>
<dt>1. <a href="faq.html#id1284470">Why can't I compare
    boost::function objects with
    operator== or
    operator!=?</a>
</dt>
<dt>2. <a href="faq.html#id1284823">I see void pointers; is this [mess] type safe?</a>
</dt>
<dt>3. <a href="faq.html#id1284844">Why are there workarounds for void returns? C++ allows them!</a>
</dt>
<dt>4. <a href="faq.html#id1284895">Why (function) cloning?</a>
</dt>
<dt>5. <a href="faq.html#id1284915">How much overhead does a call through boost::function incur?</a>
</dt>
</dl>
<table border="0" width="100%" summary="Q and A Set">
<col align="left" width="1%">
<col>
<tbody>
<tr class="question">
<td align="left" valign="top">
<a name="id1284470"></a><a name="id1284472"></a><p><b>1.</b></p>
</td>
<td align="left" valign="top"><p>Why can't I compare
    <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code> objects with
    <code class="computeroutput">operator==</code> or
    <code class="computeroutput">operator!=</code>?</p></td>
</tr>
<tr class="answer">
<td align="left" valign="top"></td>
<td align="left" valign="top">
<p>Comparison between <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code>
      objects cannot be implemented "well", and therefore will not be
      implemented. The typical semantics requested for <code class="computeroutput">f ==
      g</code> given <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code> objects
      <code class="computeroutput">f</code> and <code class="computeroutput">g</code> are:</p>
<div class="itemizedlist"><ul class="itemizedlist" type="disc">
<li class="listitem">If <code class="computeroutput">f</code> and <code class="computeroutput">g</code>
          store function objects of the same type, use that type's
          <code class="computeroutput">operator==</code> to compare
          them.</li>
<li class="listitem">If <code class="computeroutput">f</code> and <code class="computeroutput">g</code>
          store function objects of different types, return
          <code class="computeroutput">false</code>.</li>
</ul></div>
<p>The problem occurs when the type of the function objects
      stored by both <code class="computeroutput">f</code> and <code class="computeroutput">g</code> doesn't have an
      <code class="computeroutput">operator==</code>: we would like the expression <code class="computeroutput">f ==
      g</code> to fail to compile, as occurs with, e.g., the standard
      containers. However, this is not implementable for
      <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code> because it necessarily
      "erases" some type information after it has been assigned a
      function object, so it cannot try to call
      <code class="computeroutput">operator==</code> later: it must either find a way to call
      <code class="computeroutput">operator==</code> now, or it will never be able to call it
      later. Note, for instance, what happens if you try to put a
      <code class="computeroutput">float</code> value into a
      <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code> object: you will get an
      error at the assignment operator or constructor, not in
      <code class="computeroutput">operator()</code>, because the function-call expression
      must be bound in the constructor or assignment operator.</p>
<p>The most promising approach is to find a method of
      determining if <code class="computeroutput">operator==</code> can be called for a
      particular type, and then supporting it only when it is
      available; in other situations, an exception would be
      thrown. However, to date there is no known way to detect if an
      arbitrary operator expression <code class="computeroutput">f == g</code> is suitably
      defined. The best solution known has the following undesirable
      qualities:</p>
<div class="orderedlist"><ol class="orderedlist" type="1">
<li class="listitem">Fails at compile-time for objects where
        <code class="computeroutput">operator==</code> is not accessible (e.g., because it is
        <code class="computeroutput">private</code>).</li>
<li class="listitem">Fails at compile-time if calling
        <code class="computeroutput">operator==</code> is ambiguous.</li>
<li class="listitem">Appears to be correct if the
        <code class="computeroutput">operator==</code> declaration is correct, even though
        <code class="computeroutput">operator==</code> may not compile.</li>
</ol></div>
<p>All of these problems translate into failures in the
      <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code> constructors or
      assignment operator, <span class="emphasis"><em>even if the user never invokes
      operator==</em></span>. We can't do that to users.</p>
<p>The other option is to place the burden on users that want
      to use <code class="computeroutput">operator==</code>, e.g., by providing an
      <code class="computeroutput">is_equality_comparable</code> trait they may
      specialize. This is a workable solution, but is dangerous in
      practice, because forgetting to specialize the trait will result
      in unexpected exceptions being thrown from
      <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code>'s
      <code class="computeroutput">operator==</code>. This essentially negates the usefulness
      of <code class="computeroutput">operator==</code> in the context in which it is most
      desired: multitarget callbacks. The
      <a class="link" href="../signals.html" title="Chapter&#160;18.&#160;Boost.Signals">Signals</a> library has a way around
      this.</p>
</td>
</tr>
<tr class="question">
<td align="left" valign="top">
<a name="id1284823"></a><a name="id1284825"></a><p><b>2.</b></p>
</td>
<td align="left" valign="top"><p>I see void pointers; is this [mess] type safe?</p></td>
</tr>
<tr class="answer">
<td align="left" valign="top"></td>
<td align="left" valign="top"><p>Yes, <code class="computeroutput">boost::function</code> is type
safe even though it uses void pointers and pointers to functions
returning void and taking no arguments. Essentially, all type
information is encoded in the functions that manage and invoke
function pointers and function objects. Only these functions are
instantiated with the exact type that is pointed to by the void
pointer or pointer to void function. The reason that both are required
is that one may cast between void pointers and object pointers safely
or between different types of function pointers (provided you don't
invoke a function pointer with the wrong type).  </p></td>
</tr>
<tr class="question">
<td align="left" valign="top">
<a name="id1284844"></a><a name="id1284846"></a><p><b>3.</b></p>
</td>
<td align="left" valign="top"><p>Why are there workarounds for void returns? C++ allows them!</p></td>
</tr>
<tr class="answer">
<td align="left" valign="top"></td>
<td align="left" valign="top">
<p>Void returns are permitted by the C++ standard, as in this code snippet:
</p>
<pre class="programlisting">void f();
void g() { return f(); }</pre>
<p>
    </p>
<p> This is a valid usage of <code class="computeroutput">boost::function</code> because void returns are not used. With void returns, we would attempting to compile ill-formed code similar to:
</p>
<pre class="programlisting">int f();
void g() { return f(); }</pre>
<p>
</p>
<p> In essence, not using void returns allows
<code class="computeroutput">boost::function</code> to swallow a return value. This is
consistent with allowing the user to assign and invoke functions and
function objects with parameters that don't exactly match.</p>
</td>
</tr>
<tr class="question">
<td align="left" valign="top">
<a name="id1284895"></a><a name="id1284897"></a><p><b>4.</b></p>
</td>
<td align="left" valign="top"><p>Why (function) cloning?</p></td>
</tr>
<tr class="answer">
<td align="left" valign="top"></td>
<td align="left" valign="top"><p>In November and December of 2000, the issue of cloning
      vs. reference counting was debated at length and it was decided
      that cloning gave more predictable semantics. I won't rehash the
      discussion here, but if it cloning is incorrect for a particular
      application a reference-counting allocator could be used.</p></td>
</tr>
<tr class="question">
<td align="left" valign="top">
<a name="id1284915"></a><a name="id1284917"></a><p><b>5.</b></p>
</td>
<td align="left" valign="top"><p>How much overhead does a call through <code class="computeroutput"><a class="link" href="../boost/function.html" title="Class template function">boost::function</a></code> incur?</p></td>
</tr>
<tr class="answer">
<td align="left" valign="top"></td>
<td align="left" valign="top">
<p>The cost of <code class="computeroutput">boost::function</code> can be reasonably
      consistently measured at around 20ns +/- 10 ns on a modern &gt;2GHz
      platform versus directly inlining the code.</p>
<p>However, the performance of your application may benefit
      from or be disadvantaged by <code class="computeroutput">boost::function</code>
      depending on how your C++ optimiser optimises.  Similar to a
      standard function pointer, differences of order of 10% have been
      noted to the benefit or disadvantage of using
      <code class="computeroutput">boost::function</code> to call a function that contains a
      tight loop depending on your compilation circumstances.</p>
<p>[Answer provided by Matt Hurd. See <a href="http://article.gmane.org/gmane.comp.lib.boost.devel/33278" target="_top">http://article.gmane.org/gmane.comp.lib.boost.devel/33278</a>]</p>
</td>
</tr>
</tbody>
</table>
</div>
</div>
<table xmlns:rev="http://www.cs.rpi.edu/~gregod/boost/tools/doc/revision" width="100%"><tr>
<td align="left"><p><small>Last revised: November 03, 2006 at 19:41:10 +0000</small></p></td>
<td align="right"><div class="copyright-footer">Copyright &#169; 2001-2004 Douglas Gregor<p>Use, modification and distribution is subject to the Boost
    Software License, Version 1.0. (See accompanying file
    <code class="filename">LICENSE_1_0.txt</code> or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>)</p>
</div></td>
</tr></table>
<hr>
<div class="spirit-nav">
<a accesskey="p" href="../boost/function_equal.html"><img src="../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../function.html"><img src="../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../index.html"><img src="../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="misc.html"><img src="../../../doc/src/images/next.png" alt="Next"></a>
</div>
</body>
</html>