boost/more
1
0
Fork 0
mirror of https://github.com/boostorg/more.git synced 2024-12-25 23:20:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

Move several C++ related articles to the new site.

Browse Source 
Fixes #1343, #1344, #1347, #1352, #1353.


[SVN r41012]
This commit is contained in:
Daniel James 2007-11-11 19:14:34 +00:00
parent 570af02790
commit cd710d8e34
5 changed files with 0 additions and 2273 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

774
count_bdy.htm
View File

@ -1,774 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Language" content="en-us">
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="GENERATOR" content="Microsoft FrontPage 6.0">
<meta name="Author" content="Kevlin Henney">
<meta name="KeyWords" content=
"C++, Reference Counting, Advanced Techniques, Smart Pointers, Patterns">
<title>Counted Body Techniques</title>
</head>
<body bgcolor="#FFFFFF" text="#000000">
<h1 align="center"><i><font size="+4">Counted Body Techniques</font></i></h1>
<center>
<p><b><font size="+1"><a href="../people/kevlin_henney.htm">Kevlin Henney</a><br></font>
(<a href=
"mailto:kevlin@acm.org">kevlin@acm.org</a>, <a href=
"mailto:khenney@qatraining.com">khenney@qatraining.com</a>)</b></p>
</center>
<div style="margin-left: 2em">
<p>Reference counting techniques? Nothing new, you might think. Every good
C++ text that takes you to an intermediate or advanced level will
introduce the concept. It has been explored with such thoroughness in the
past that you might be forgiven for thinking that everything that can be
said has been said. Well, let's start from first principles and see if we
can unearth something new....</p>
</div>
<hr width="100%">
<h2>And then there were none...</h2>
<div style="margin-left: 2em">
<p>The principle behind reference counting is to keep a running usage
count of an object so that when it falls to zero we know the object is
unused. This is normally used to simplify the memory management for
dynamically allocated objects: keep a count of the number of references
held to that object and, on zero, delete the object.</p>
<p>How to keep a track of the number of users of an object? Well, normal
pointers are quite dumb, and so an extra level of indirection is required
to manage the count. This is essentially the P<font size="-1">ROXY</font>
pattern described in <i>Design Patterns</i> [Gamma, Helm, Johnson &amp;
Vlissides, Addison-Wesley, <font size="-1">ISBN</font> 0-201-63361-2]. The
intent is given as</p>
<div style="margin-left: 2em">
<p><i>Provide a surrogate or placeholder for another object to control
access to it.</i></p>
</div>
<p>Coplien [<i>Advanced C++ Programming Styles and Idioms</i>,
Addison-Wesley, <font size="-1">ISBN</font> 0-201-56365-7] defines a set
of idioms related to this essential separation of a handle and a body
part. The <i>Taligent Guide to Designing Programs</i> [Addison-Wesley,
<font size="-1">ISBN</font> 0-201-40888-0] identifies a number of specific
categories for proxies (aka surrogates). Broadly speaking they fall into
two general categories:</p>
<ul>
<li><i>Hidden</i>: The handle is the object of interest, hiding the body
itself. The functionality of the handle is obtained by delegation to the
body, and the user of the handle is unaware of the body. Reference
counted strings offer a transparent optimisation. The body is shared
between copies of a string until such a time as a change is needed, at
which point a copy is made. Such a C<font size=
"-1">OPY</font> <font size="-1">ON</font> W<font size="-1">RITE</font>
pattern (a specialisation of L<font size="-1">AZY</font> E<font size=
"-1">VALUATION</font>) requires the use of a hidden reference counted
body.</li>
<li><i>Explicit</i>: Here the body is of interest and the handle merely
provides intelligence for its access and housekeeping. In C++ this is
often implemented as the S<font size="-1">MART</font> P<font size=
"-1">OINTER</font> idiom. One such application is that of reference
counted smart pointers that collaborate to keep a count of an object,
deleting it when the count falls to zero.</li>
</ul>
</div>
<hr width="100%">
<h2>Attached vs detached</h2>
<div style="margin-left: 2em">
<p>For reference counted smart pointers there are two places the count can
exist, resulting in two different patterns, both outlined in
<i>Software Patterns</i> [Coplien, SIGS, <font size="-1">ISBN</font>
0-884842-50-X]:</p>
<ul>
<li>C<font size="-1">OUNTED</font> B<font size="-1">ODY</font> or A<font size="-1">TTACHED</font>
C<font size="-1">OUNTED</font>
H<font size="-1">ANDLE</font>/B<font size="-1">ODY</font> places the
count within the object being counted. The benefits are that
countability is a part of the object being counted, and that reference
counting does not require an additional object. The drawbacks are
clearly that this is intrusive, and that the space for the reference
count is wasted when the object is not heap based. Therefore the
reference counting ties you to a particular implementation and style of
use.</li>
<li>D<font size="-1">ETACHED</font> C<font size="-1">OUNTED</font>
H<font size="-1">ANDLE</font>/B<font size="-1">ODY</font> places the
count outside the object being counted, such that they are handled
together. The clear benefit of this is that this technique is completely
unintrusive, with all of the intelligence and support apparatus in the
smart pointer, and therefore can be used on classes created
independently of the reference counted pointer. The main disadvantage is
that frequent use of this can lead to a proliferation of small objects,
i.e. the counter, being created on the heap.</li>
</ul>
<p>Even with this simple analysis, it seems that the D<font size=
"-1">ETACHED</font> C<font size="-1">OUNTED</font> H<font size=
"-1">ANDLE</font>/B<font size="-1">ODY</font> approach is ahead. Indeed,
with the increasing use of templates this is often the favourite, and is
the principle behind the common - but not standard - <tt><font size=
"+1">counted_ptr</font></tt>. <i>[The Boost name is <a href=
"../libs/smart_ptr/shared_ptr.htm"><tt><font size=
"+1">shared_ptr</font></tt></a> rather than <tt><font size=
"+1">counted_ptr</font></tt>.]</i></p>
<p>A common implementation of C<font size="-1">OUNTED</font> B<font size=
"-1">ODY</font> is to provide the counting mechanism in a base class that
the counted type is derived from. Either that, or the reference counting
mechanism is provided anew for each class that needs it. Both of these
approaches are unsatisfactory because they are quite closed, coupling a
class into a particular framework. Added to this the non-cohesiveness of
having the count lying dormant in a non-counted object, and you get the
feeling that excepting its use in widespread object models such as COM and
CORBA the C<font size="-1">OUNTED</font> B<font size="-1">ODY</font>
approach is perhaps only of use in specialised situations.</p>
</div>
<hr width="100%">
<h2>A requirements based approach</h2>
<div style="margin-left: 2em">
<p>It is the question of openness that convinced me to revisit the
problems with the C<font size="-1">OUNTED</font> B<font size=
"-1">ODY</font> idiom. Yes, there is a certain degree of intrusion
expected when using this idiom, but is there anyway to minimise this and
decouple the choice of counting mechanism from the smart pointer type
used?</p>
<p>In recent years the most instructive body of code and specification for
constructing open general purpose components has been the Stepanov and
Lee's STL (Standard Template Library), now part of the C++ standard
library. The STL approach makes extensive use of compile time polymorphism
based on well defined operational requirements for types. For instance,
each container, contained and iterator type is defined by the operations
that should be performable on an object of that type, often with
annotations describing additional constraints. Compile time polymorphism,
as its name suggests, resolves functions at compile time based on function
name and argument usage, i.e. overloading. This is less intrusive,
although less easily diagnosed if incorrect, than runtime poymorphism that
is based on types, names and function signatures.</p>
<p>This requirements based approach can be applied to reference counting.
The operations we need for a type to be <i>Countable</i> are loosely:</p>
<ul>
<li>An <tt><font size="+1">acquire</font></tt> operation that registers
interest in a <i>Countable</i> object.</li>
<li>A <tt><font size="+1">release</font></tt> operation unregisters
interest in a <i>Countable</i> object.</li>
<li>An <tt><font size="+1">acquired</font></tt> query that returns
whether or not a <i>Countable</i> object is currently acquired.</li>
<li>A <tt><font size="+1">dispose</font></tt> operation that is
responsible for disposing of an object that is no longer acquired.</li>
</ul>
<p>Note that the count is deduced as a part of the abstract state of this
type, and is not mentioned or defined in any other way. The openness of
this approach derives in part from the use of global functions, meaning
that no particular member functions are implied; a perfect way to wrap up
an existing counted body class without modifying the class itself. The
other aspect to the openness comes from a more precise specification of
the operations.</p>
<p>For a type to be <i>Countable</i> it must satisfy the following
requirements, where <tt><font size="+1">ptr</font></tt> is a non-null
pointer to a single object (i.e. not an array) of the type, and
<i><tt><font size="+1">#function</font></tt></i> indicates number of calls
to <tt><font size="+1"><i>function(</i>ptr<i>)</i></font></tt>:</p>
<center>
<table border="1" cellspacing="2" cellpadding="2" summary="">
<tr>
<td><i>Expression</i></td>
<td><i>Return type</i></td>
<td><i>Semantics and notes</i></td>
</tr>
<tr>
<td><tt><font size="+1">acquire(ptr)</font></tt></td>
<td>no requirement</td>
<td><i>post</i>: <tt><font size="+1">acquired(ptr)</font></tt></td>
</tr>
<tr>
<td><tt><font size="+1">release(ptr)</font></tt></td>
<td>no requirement</td>
<td><i>pre</i>: <tt><font size="+1">acquired(ptr)<br></font></tt>
<i>post</i>: <tt><font size="+1">acquired(ptr) == #acquire -
#release</font></tt></td>
</tr>
<tr>
<td><tt><font size="+1">acquired(ptr)</font></tt></td>
<td>convertible to <tt><font size="+1">bool</font></tt></td>
<td><i>return</i>: <tt><font size="+1">#acquire &gt; #release</font></tt></td>
</tr>
<tr>
<td><tt><font size="+1">dispose(ptr, ptr)</font></tt></td>
<td>no requirement</td>
<td><i>pre</i>: <tt><font size="+1">!acquired(ptr)<br></font></tt>
<i>post</i>: <tt><font size="+1">*ptr</font></tt> no longer usable</td>
</tr>
</table>
</center>
<p>Note that the two arguments to <tt><font size="+1">dispose</font></tt>
are to support selection of the appropriate type safe version of the
function to be called. In the general case the intent is that the first
argument determines the type to be deleted, and would typically be
templated, while the second selects which template to use, e.g. by
conforming to a specific base class.</p>
<p>In addition the following requirements must also be satisfied, where
<tt><font size="+1">null</font></tt> is a null pointer to the
<i>Countable</i> type:</p>
<center>
<table border="1" summary="">
<tr>
<td><i>Expression</i></td>
<td><i>Return type</i></td>
<td><i>Semantics and notes</i></td>
</tr>
<tr>
<td><tt><font size="+1">acquire(null)</font></tt></td>
<td>no requirement</td>
<td><i>action</i>: none</td>
</tr>
<tr>
<td><tt><font size="+1">release(null)</font></tt></td>
<td>no requirement</td>
<td><i>action</i>: none</td>
</tr>
<tr>
<td><tt><font size="+1">acquired(null)</font></tt></td>
<td>convertible to <tt><font size="+1">bool</font></tt></td>
<td><i>return</i>: <tt><font size="+1">false</font></tt></td>
</tr>
<tr>
<td><tt><font size="+1">dispose(null, null)</font></tt></td>
<td>no requirement</td>
<td><i>action</i>: none</td>
</tr>
</table>
</center>
<p>Note that there are no requirements on these functions in terms of
exceptions thrown or not thrown, except that if exceptions are thrown the
functions themselves should be exception safe.</p>
</div>
<hr width="100%">
<h2>Getting smart</h2>
<div style="margin-left: 2em">
<p>Given the <i>Countable</i> requirements for a type, it is possible to
define a generic smart pointer type that uses them for reference counting:</p>
<div style="margin-left: 2em">
<pre>
<tt>template&lt;typename countable_type&gt;
class countable_ptr
{
public: // construction and destruction
explicit countable_ptr(countable_type *);
countable_ptr(const countable_ptr &amp;);
~countable_ptr();
public: // access
countable_type *operator-&gt;() const;
countable_type &amp;operator*() const;
countable_type *get() const;
public: // modification
countable_ptr &amp;clear();
countable_ptr &amp;assign(countable_type *);
countable_ptr &amp;assign(const countable_ptr &amp;);
countable_ptr &amp;operator=(const countable_ptr &amp;);
private: // representation
countable_type *body;
};
</tt>
</pre>
</div>
<p>The interface to this class has been kept intentionally simple, e.g.
member templates and <tt><font size="+1">throw</font></tt> specs have been
omitted, for exposition. The majority of the functions are quite simple in
implementation, relying very much on the <tt><font size=
"+1">assign</font></tt> member as a keystone function:</p>
<div style="margin-left: 2em">
<pre>
<tt>template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt;::countable_ptr(countable_type *initial)
: body(initial)
{
acquire(body);
}
template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt;::countable_ptr(const countable_ptr &amp;other)
: body(other.body)
{
acquire(body);
}
template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt;::~countable_ptr()
{
clear();
}
template&lt;typename countable_type&gt;
countable_type *countable_ptr&lt;countable_type&gt;::operator-&gt;() const
{
return body;
}
template&lt;typename countable_type&gt;
countable_type &amp;countable_ptr&lt;countable_type&gt;::operator*() const
{
return *body;
}
template&lt;typename countable_type&gt;
countable_type *countable_ptr&lt;countable_type&gt;::get() const
{
return body;
}
template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt; &amp;countable_ptr&lt;countable_type&gt;::clear()
{
return assign(0);
}
template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt; &amp;countable_ptr&lt;countable_type&gt;::assign(countable_type *rhs)
{
// set to rhs (uses Copy Before Release idiom which is self assignment safe)
acquire(rhs);
countable_type *old_body = body;
body = rhs;
// tidy up
release(old_body);
if(!acquired(old_body))
{
dispose(old_body, old_body);
}
return *this;
}
template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt; &amp;countable_ptr&lt;countable_type&gt;::assign(const countable_ptr &amp;rhs)
{
return assign(rhs.body);
}
template&lt;typename countable_type&gt;
countable_ptr&lt;countable_type&gt; &amp;countable_ptr&lt;countable_type&gt;::operator=(const countable_ptr &amp;rhs)
{
return assign(rhs);
}
</tt>
</pre>
</div>
</div>
<hr width="100%">
<h2>Public accountability</h2>
<div style="margin-left: 2em">
<p>Conformance to the requirements means that a type can be used with
<tt><font size="+1">countable_ptr</font></tt>. Here is an implementation
mix-in class (<i>mix-imp</i>) that confers countability on its derived
classes through member functions. This class can be used as a class
adaptor:</p>
<div style="margin-left: 2em">
<pre>
<tt>class countability
{
public: // manipulation
void acquire() const;
void release() const;
size_t acquired() const;
protected: // construction and destruction
countability();
~countability();
private: // representation
mutable size_t count;
private: // prevention
countability(const countability &amp;);
countability &amp;operator=(const countability &amp;);
};
</tt>
</pre>
</div>
<p>Notice that the manipulation functions are <tt><font size=
"+1">const</font></tt> and that the <tt><font size="+1">count</font></tt>
member itself is <tt><font size="+1">mutable</font></tt>. This is because
countability is not a part of an object's abstract state: memory
management does not depend on the <tt><font size=
"+1">const</font></tt>-ness or otherwise of an object. I won't include the
definitions of the member functions here as you can probably guess them:
increment, decrement and return the current count, respectively for the
manipulation functions. In a multithreaded environment you should ensure
that such read and write operations are atomic.</p>
<p>So how do we make this class <i>Countable</i>? A simple set of
forwarding functions does the job:</p>
<div style="margin-left: 2em">
<pre>
<tt>void acquire(const countability *ptr)
{
if(ptr)
{
ptr-&gt;acquire();
}
}
void release(const countability *ptr)
{
if(ptr)
{
ptr-&gt;release();
}
}
size_t acquired(const countability *ptr)
{
return ptr ? ptr-&gt;acquired() : 0;
}
template&lt;class countability_derived&gt;
void dispose(const countability_derived *ptr, const countability *)
{
delete ptr;
}
</tt>
</pre>
</div>
<p>Any type that now derives from <tt><font size=
"+1">countability</font></tt> may now be used with <tt><font size=
"+1">countable_ptr</font></tt>:</p>
<div style="margin-left: 2em">
<pre>
<tt>class example : public countability
{
...
};
void simple()
{
countable_ptr&lt;example&gt; ptr(new example);
countable_ptr&lt;example&gt; qtr(ptr);
ptr.clear(); // set ptr to point to null
} // allocated object deleted when qtr destructs
</tt>
</pre>
</div>
</div>
<hr width="100%">
<h2>Runtime mixin</h2>
<div style="margin-left: 2em">
<p>The challenge is to apply C<font size="-1">OUNTED</font> B<font size=
"-1">ODY</font> in a non-intrusive fashion, such that there is no overhead
when an object is not counted. What we would like to do is confer this
capability on a per object rather than on a per class basis. Effectively
we are after <i>Countability</i> on any object, i.e. anything pointed to
by a <tt><font size="+1">void *</font></tt>! It goes without saying that <tt><font size="+1">
void</font></tt> is perhaps the least committed of any type.</p>
<p>The forces to resolve on this are quite interesting, to say the least.
Interesting, but not insurmountable. Given that the class of a runtime
object cannot change dynamically in any well defined manner, and the
layout of the object must be fixed, we have to find a new place and time
to add the counting state. The fact that this must be added only on heap
creation suggests the following solution:</p>
<div style="margin-left: 2em">
<pre>
<tt>struct countable_new;
extern const countable_new countable;
void *operator new(size_t, const countable_new &amp;);
void operator delete(void *, const countable_new &amp;);</tt>
</pre>
</div>
<p>We have overloaded <tt><font size="+1">operator new</font></tt> with a
dummy argument to distinguish it from the regular global <tt><font size=
"+1">operator new</font></tt>. This is comparable to the use of the
<tt><font size="+1">std::nothrow_t</font></tt> type and <tt><font size=
"+1">std::nothrow</font></tt> object in the standard library. The
placement <tt><font size="+1">operator delete</font></tt> is there to
perform any tidy up in the event of failed construction. Note that this is
not yet supported on all that many compilers.</p>
<p>The result of a <tt><font size="+1">new</font></tt> expression using
<tt><font size="+1">countable</font></tt> is an object allocated on the
heap that has a header block that holds the count, i.e. we have extended
the object by prefixing it. We can provide a couple of features in an
anonymous namespace (not shown) in the implementation file for for
supporting the count and its access from a raw pointer:</p>
<div style="margin-left: 2em">
<pre>
<tt>struct count
{
size_t value;
};
count *header(const void *ptr)
{
return const_cast&lt;count *&gt;(static_cast&lt;const count *&gt;(ptr) - 1);
}
</tt>
</pre>
</div>
<p>An important constraint to observe here is the alignment of
<tt><font size="+1">count</font></tt> should be such that it is suitably
aligned for any type. For the definition shown this will be the case on
almost all platforms. However, you may need to add a padding member for
those that don't, e.g. using an anonymous <tt><font size=
"+1">union</font></tt> to coalign <tt><font size="+1">count</font></tt>
and the most aligned type. Unfortunately, there is no portable way of
specifying this such that the minimum alignment is also observed - this is
a common problem when specifying your own allocators that do not directly
use the results of either <tt><font size="+1">new</font></tt> or
<tt><font size="+1">malloc</font></tt>.</p>
<p>Again, note that the count is not considered to be a part of the
logical state of the object, and hence the conversion from
<tt><font size="+1">const</font></tt> to non-<tt><font size=
"+1">const</font></tt> - <tt><font size="+1">count</font></tt> is in
effect a <tt><font size="+1">mutable</font></tt> type.</p>
<p>The allocator functions themselves are fairly straightforward:</p>
<div style="margin-left: 2em">
<pre>
<tt>void *operator new(size_t size, const countable_new &amp;)
{
count *allocated = static_cast&lt;count *&gt;(::operator new(sizeof(count) + size));
*allocated = count(); // initialise the header
return allocated + 1; // adjust result to point to the body
}
void operator delete(void *ptr, const countable_new &amp;)
{
::operator delete(header(ptr));
}
</tt>
</pre>
</div>
<p>Given a correctly allocated header, we now need the <i>Countable</i>
functions to operate on <tt><font size="+1">const void *</font></tt> to
complete the picture:</p>
<div style="margin-left: 2em">
<pre>
<tt>void acquire(const void *ptr)
{
if(ptr)
{
++header(ptr)-&gt;value;
}
}
void release(const void *ptr)
{
if(ptr)
{
--header(ptr)-&gt;value;
}
}
size_t acquired(const void *ptr)
{
return ptr ? header(ptr)-&gt;value : 0;
}
template&lt;typename countable_type&gt;
void dispose(const countable_type *ptr, const void *)
{
ptr-&gt;~countable_type();
operator delete(const_cast&lt;countable_type *&gt;(ptr), countable);
}
</tt>
</pre>
</div>
<p>The most complex of these is the <tt><font size=
"+1">dispose</font></tt> function that must ensure that the correct type
is destructed and also that the memory is collected from the correct
offset. It uses the value and type of first argument to perform this
correctly, and the second argument merely acts as a strategy selector,
i.e. the use of <tt><font size="+1">const void *</font></tt>
distinguishes it from the earlier dispose shown for <tt><font size=
"+1">const countability *</font></tt>.</p>
</div>
<hr width="100%">
<h2>Getting smarter</h2>
<div style="margin-left: 2em">
<p>Now that we have a way of adding countability at creation for objects
of any type, what extra is needed to make this work with the
<tt><font size="+1">countable_ptr</font></tt> we defined earlier? Good
news: nothing!</p>
<div style="margin-left: 2em">
<pre>
<tt>class example
{
...
};
void simple()
{
countable_ptr&lt;example&gt; ptr(new(countable) example);
countable_ptr&lt;example&gt; qtr(ptr);
ptr.clear(); // set ptr to point to null
} // allocated object deleted when qtr destructs
</tt>
</pre>
</div>
<p>The <tt><font size="+1">new(countable)</font></tt> expression defines a
different policy for allocation and deallocation and, in common with other
allocators, any attempt to mix your allocation policies, e.g. call
<tt><font size="+1">delete</font></tt> on an object allocated with
<tt><font size="+1">new(countable)</font></tt>, results in undefined
behaviour. This is similar to what happens when you mix <tt><font size=
"+1">new[]</font></tt> with <tt><font size="+1">delete</font></tt> or
<tt><font size="+1">malloc</font></tt> with <tt><font size=
"+1">delete</font></tt>. The whole point of <i>Countable</i> conformance
is that <i>Countable</i> objects are used with <tt><font size=
"+1">countable_ptr</font></tt>, and this ensures the correct use.</p>
<p>However, accidents will happen, and inevitably you may forget to
allocate using <tt><font size="+1">new(countable)</font></tt> and instead
use <tt><font size="+1">new</font></tt>. This error and others can be
detected in most cases by extending the code shown here to add a check
member to the <tt><font size="+1">count</font></tt>, validating the check
on every access. A benefit of ensuring clear separation between header and
implementation source files means that you can introduce a checking
version of this allocator without having to recompile your code.</p>
</div>
<hr width="100%">
<h2>Conclusion</h2>
<div style="margin-left: 2em">
<p>There are two key concepts that this article has introduced:</p>
<ul>
<li>The use of a generic requirements based approach to simplify and
adapt the use of the C<font size="-1">OUNTED</font> B<font size=
"-1">ODY</font> pattern.</li>
<li>The ability, through control of allocation, to dynamically and
non-intrusively add capabilities to fixed types using the R<font size=
"-1">UNTIME</font> M<font size="-1">IXIN</font> pattern.</li>
</ul>
<p>The application of the two together gives rise to a new variant of the
essential C<font size="-1">OUNTED</font> B<font size="-1">ODY</font>
pattern, U<font size="-1">NINTRUSIVE</font> C<font size=
"-1">OUNTED</font> B<font size="-1">ODY</font>. You can take this theme
even further and contrive a simple garbage collection system for C++.</p>
<p>The complete code for <tt><font size="+1">countable_ptr</font></tt>,
<tt><font size="+1">countability</font></tt>, and the <tt><font size=
"+1">countable new</font></tt> is also available.</p>
</div>
<div align="right">
<hr width="100%">
<font size="-1"><i>First published in</i> <a href=
"http://www.accu.org/c++sig/public/Overload.html">Overload</a> <i>25,
April 1998, ISSN 1354-3172</i></font>
</div>
<p><a href="http://validator.w3.org/check?uri=referer"><img border="0" src=
"http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01 Transitional"
height="31" width="88"></a></p>
<p>Revised
<!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %B, %Y" startspan -->04 December, 2006<!--webbot bot="Timestamp" endspan i-checksum="38514" --></p>
<p><i>Copyright &copy; 1998-1999 Kevlin Henney</i></p>
<p><i>Distributed under the Boost Software License, Version 1.0. (See
accompanying file <a href="../LICENSE_1_0.txt">LICENSE_1_0.txt</a> or copy
at <a href=
"http://www.boost.org/LICENSE_1_0.txt">http://www.boost.org/LICENSE_1_0.txt</a>)</i></p>
</body>
</html>

125
cpp_committee_meetings.html
View File

@ -1,125 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Language" content="en-us">
<meta name="GENERATOR" content="Microsoft FrontPage 5.0">
<meta name="ProgId" content="FrontPage.Editor.Document">
<meta http-equiv="Content-Type" content="text/html; charset=windows-1252">
<title>C++ Committee Meetings</title>
</head>
<body bgcolor="#FFFFFF">
<h1>C++ Committee Meeting FAQ for Boost Members</h1>
<p><b>Who can attend C++ Committee meetings?</b> Members of
J16 (the INCITS/ANSI committee) or of a WG21 (ISO) member country committee
(&quot;national body&quot; in
ISO-speak). <a href="http://www.ncits.org/">
INCITS</a> has broadened&nbsp; J16 membership requirements so anyone can
join, regardless of nationality or employer.</p>
<p>In addition, a small number of &quot;technical experts&quot; who are not committee
members can also attend meetings. The &quot;technical expert&quot; umbrella is broad enough to cover
the
Boost members who attend meetings.</p>
<p><b>When and where is the next meeting?</b> There are two meetings a year. The
Fall meeting is usually in North America, and the Spring meeting is usually
outside North America. See a general
<a href="http://www.open-std.org/jtc1/sc22/wg21/docs/meetings">list of meeting locations and
dates</a>. Detailed information about a particular meeting, including hotel
information, is usually provided in a paper appearing in one of
<a href="#Mailing">mailings</a> for the prior meeting. If there isn't a link to
it on the <a href="http://www.open-std.org/jtc1/sc22/wg21/docs/meetings">
Meetings</a> web page, you will have to go to
the committee's <a href="http://www.open-std.org/jtc1/sc22/wg21/docs/papers/">
Papers</a> page and search a bit.</p>
<p><b>Is there a fee for attending meetings?</b> No, but there can be a lot of
incidental expenses like travel, lodging, and meals, and there is a $US 800 a
year INCITS fee to become a voting member.</p>
<p><b>What is the schedule?</b>&nbsp; The meetings start at 9:00AM on
Monday, and 8:30AM other days, unless otherwise announced. It is best to arrive
a half-hour early to grab a good seat, some coffee, tea, or donuts, and to say
hello to people. (There is also a Sunday evening a WG21 administrative meeting,
which is closed except to delegates from national bodies.)</p>
<p>The meetings generally end on Friday, although there is discussion of
extending them one extra day until the next standard ships. The last day the meeting&nbsp; is generally over by 11:00AM. Because
the last day's meeting is for formal votes only, it is primarily of interest only to
actual committee
members.</p>
<p>Sometimes there are evening technical sessions; the details aren't
usually available until the Monday morning meeting.&nbsp; There may be a
reception one evening, and, yes, significant others are
invited. Again, details usually&nbsp;become available Monday morning.</p>
<p><b>What actually happens at the meetings?</b> Monday morning an hour or two
is spent in full committee on administrivia, and then the committee breaks up
into working groups (Core, Library, and Enhancements). The full committee also
gets together later in the week to hear working group progress reports.</p>
<p>The working groups are where most technical activities take place.&nbsp; Each
active issue that appears on an issues list is discussed, as are papers from the
mailing. Most issues are non-controversial and disposed of in a few minutes.
Technical discussions are often led by long-term committee members, often
referring to past decisions or longstanding working group practice. Sometimes a
controversy erupts. It takes first-time attendees awhile to understand the
discussions and how decisions are actually made. The working group chairperson
moderates.</p>
<p>Sometimes straw polls are taken. In a straw poll anyone attending can vote,
in contrast to the formal votes taken by the full committee, where only voting
members can vote.</p>
<p>Lunch break is an hour and a half.&nbsp; Informal subgroups often lunch
together; a lot of technical problems are discussed or actually solved at lunch,
or later at dinner. In many ways these discussions involving only a few people
are the most interesting. Sometimes during the regular meetings, a working group
chair will break off a sub-group to tackle a difficult problem. </p>
<p><b>Do I have to stay at the main hotel?</b> No, and committee members on
tight budgets often stay at other, cheaper, hotels. (The main hotels are usually
chosen because they have large meeting rooms available, and thus tend to be pricey.)
The advantage of staying at the main hotel is that it is then easier to
participate in the off-line discussions which can be at least as interesting
as what actually happens in the scheduled meetings.</p>
<p><b>What do people wear at meetings?</b>&nbsp; Programmer casual. No neckties
to be seen. </p>
<p><b>What should I bring to a meeting?</b> It is almost essential to have a
laptop computer along. There is a committee LAN with a wiki and Internet connectivity.
Wireless connectivity has become the norm, although there is usually a wired hub
or two for those needed wired access.</p>
<p><b>What should I do to prepare for a meeting?</b> It is helpful to have
downloaded the mailing or individual papers for the
meeting, and read any papers you are interested in. Familiarize yourself with
the issues lists if you haven't done so already. Decide which of the working
groups you want to attend.</p>
<p><b>What is a &quot;<a name="Paper">Paper</a>&quot;?</b> An electronic document containing issues,
proposals, or anything else the committee is interested in. Very little gets
discussed at a meeting, much less acted upon, unless it is presented in a paper.&nbsp;
<a href="http://std.dkuug.dk/jtc1/sc22/wg21/docs/papers/">Papers are available</a>
to anyone. Papers don't just appear randomly; they become available four (lately
six) times a
year, before and after each meeting. Committee members often refer to a paper by
saying what mailing it was in: &quot;See the pre-Redmond mailing.&quot;</p>
<p><b>What is a &quot;<a name="Mailing">Mailing</a>&quot;?</b> A mailing is the
set of papers prepared four to six times a year before and after each meeting,
or between meetings.&nbsp; It
is physically just a
<a href="http://www.open-std.org/jtc1/sc22/wg21/docs/mailings/">.zip or .gz</a>
archive of
all the papers for a meeting. Although the mailing's archive file itself is only available to committee members and technical
experts, the contents (except copies of the standard) are available to the
general public as individual papers. The ways of ISO are
inscrutable.</p>
<p><b>What is a &quot;Reflector&quot;?</b> The committee's mailing lists are
called &quot;reflectors&quot;. There are a number of them; &quot;all&quot;, &quot;core&quot;, &quot;lib&quot;, and &quot;ext&quot;
are the main ones. As a courtesy, Boost technical experts can be added to
committee reflectors at the request of a committee member. </p>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%B %d, %Y" startspan -->April 17, 2005<!--webbot bot="Timestamp" endspan i-checksum="17669" --></p>
<p>© Copyright Beman Dawes, 2002</p>
<p>
Distributed under the Boost Software License, Version 1.0. (See
accompanying file <a href="../LICENSE_1_0.txt">LICENSE_1_0.txt</a> or copy
at <a href=
"http://www.boost.org/LICENSE_1_0.txt">http://www.boost.org/LICENSE_1_0.txt</a>)
</p>
</body>
</html>

240
error_handling.html
View File

@ -1,240 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Cygwin (vers 1st April 2002), see www.w3.org">
<meta http-equiv="Content-Type" content=
"text/html; charset=windows-1252">
<title>Error and Exception Handling</title>
</head>
<body>
<h1>Error and Exception Handling</h1>
<h2>References</h2>
<p>The following paper is a good introduction to some of the issues of
writing robust generic components:</p>
<blockquote>
<a href="generic_exception_safety.html">D. Abrahams: ``Exception Safety
in Generic Components''</a>, originally published in <a href=
"http://www.springer.de/cgi-bin/search_book.pl?isbn=3-540-41090-2">M.
Jazayeri, R. Loos, D. Musser (eds.): Generic Programming, Proc. of a
Dagstuhl Seminar, Lecture Notes on Computer Science. Volume. 1766</a>
</blockquote>
<h2>Guidelines</h2>
<h3>When should I use exceptions?</h3>
<p>The simple answer is: ``whenever the semantic and performance
characteristics of exceptions are appropriate.''</p>
<p>An oft-cited guideline is to ask yourself the question ``is this an
exceptional (or unexpected) situation?'' This guideline has an attractive
ring to it, but is usually a mistake. The problem is that one person's
``exceptional'' is another's ``expected'': when you really look at the
terms carefully, the distinction evaporates and you're left with no
guideline. After all, if you check for an error condition, then in some
sense you expect it to happen, or the check is wasted code.</p>
<p>A more appropriate question to ask is: ``do we want stack
unwinding here?'' Because actually handling an exception is likely
to be significantly slower than executing mainline code, you
should also ask: ``Can I afford stack unwinding here?'' For
example, a desktop application performing a long computation might
periodically check to see whether the user had pressed a cancel
button. Throwing an exception could allow the operation to be
cancelled gracefully. On the other hand, it would probably be
inappropriate to throw and <i>handle</i> exceptions in the inner
loop of this computation because that could have a significant
performance impact. The guideline mentioned above has a grain of
truth in it: in time critical code, throwing an exception
should <em>be</em> the exception, not the rule.</p>
<h3>How should I design my exception classes?</h3>
<ol>
<li><b>Derive your exception class
from <code>std::exception</code></b>. Except in *very* rare
circumstances where you can't afford the cost of a virtual
table,
<code>std::exception</code> makes a reasonable exception base class,
and when used universally, allows programmers to catch "everything"
without resorting to <code>catch(...)</code>. For more about
<code>catch(...)</code>, see below.
<li><b>Use <i>virtual</i> inheritance.</b> This insight is due
to Andrew Koenig. Using virtual inheritance from your
exception's base class(es) prevents ambiguity problems at the
catch-site in case someone throws an exception derived from
multiple bases which have a base class in common:
<pre>
#include &lt;iostream&gt;
struct my_exc1 : std::exception { char const* what() const throw(); };
struct my_exc2 : std::exception { char const* what() const throw(); };
struct your_exc3 : my_exc1, my_exc2 {};
int main()
{
try { throw your_exc3(); }
catch(std::exception const&amp; e) {}
catch(...) { std::cout &lt;&lt; &quot;whoops!&quot; &lt;&lt; std::endl; }
}
</pre>
The program above prints <code>&quot;whoops&quot;</code> because the
C++ runtime can't resolve which <code>exception</code> instance to
match in the first catch clause.
</li>
<li>
<b><i>Don't</i> embed a std::string object</b> or any other data
member or base class whose copy constructor could throw an exception.
That could lead directly to std::terminate() at the throw point.
Similarly, it's a bad idea to use a base or member whose ordinary
constructor(s) might throw, because, though not necessarily fatal to
your program, you may report a different exception than intended from
a <i>throw-expression</i> that includes construction such as:
<blockquote>
<pre>
throw some_exception();
</pre>
</blockquote>
<p>There are various ways to avoid copying string objects when
exceptions are copied, including embedding a fixed-length buffer in
the exception object, or managing strings via reference-counting.
However, consider the next point before pursuing either of these
approaches.</p>
</li>
<li><b>Format the <code>what()</code> message on demand</b>, if you
feel you really must format the message. Formatting an exception error
message is typically a memory-intensive operation that could
potentially throw an exception. This is an operation best delayed until
after stack unwinding has occurred, and presumably, released some
resources. It's a good idea in this case to protect your
<code>what()</code> function with a <code>catch(...)</code> block so
that you have a fallback in case the formatting code throws</li>
<li><b>Don't worry <i>too</i> much about the <code>what()</code>
message</b>. It's nice to have a message that a programmer stands a
chance of figuring out, but you're very unlikely to be able to compose
a relevant and <i>user</i>-comprehensible error message at the point an
exception is thrown. Certainly, internationalization is beyond the
scope of the exception class author. <a href=
"../people/peter_dimov.htm">Peter Dimov</a> makes an excellent argument
that the proper use of a <code>what()</code> string is to serve as a
key into a table of error message formatters. Now if only we could get
standardized <code>what()</code> strings for exceptions thrown by the
standard library...</li>
<li><b>Expose relevant information about the cause of the error</b> in
your exception class' public interface. A fixation on the
<code>what()</code> message is likely to mean that you neglect to
expose information someone might need in order to make a coherent
message for users. For example, if your exception reports a numeric
range error, it's important to have the actual numbers involved
available <i>as numbers</i> in the exception class' public interface
where error reporting code can do something intelligent with them. If
you only expose a textual representation of those numbers in the
<code>what()</code> string, you will make life very difficult for
programmers who need to do something more (e.g. subtraction) with them
than dumb output.</li>
<li><b>Make your exception class immune to double-destruction</b> if
possible. Unfortunately, several popular compilers occasionally cause
exception objects to be destroyed twice. If you can arrange for that to
be harmless (e.g. by zeroing deleted pointers) your code will be more
robust.</li>
</ol>
<h3>What About Programmer Errors?</h3>
<p>As a developer, if I have violated a precondition of a library I'm
using, I don't want stack unwinding. What I want is a core dump or the
equivalent - a way to inspect the state of the program at the exact point
where the problem was detected. That usually means <tt>assert()</tt> or
something like it.</p>
<p>Sometimes it is necessary to have resilient APIs which can stand up to
nearly any kind of client abuse, but there is usually a significant cost
to this approach. For example, it usually requires that each object used
by a client be tracked so that it can be checked for validity. If you
need that sort of protection, it can usually be provided as a layer on
top of a simpler API. Beware half-measures, though. An API which promises
resilience against some, but not all abuse is an invitation to disaster.
Clients will begin to rely on the protection and their expectations will
grow to cover unprotected parts of the interface.</p>
<p><b>Note for Windows developers</b>: unfortunately, the native
exception-handling used by most Windows compilers actually throws an
exception when you use <tt>assert()</tt>. Actually, this is true of other
programmer errors such as segmentation faults and divide-by-zero errors.
One problem with this is that if you use JIT (Just In Time) debugging,
there will be collateral exception-unwinding before the debugger comes up
because <code>catch(...)</code> will catch these not-really-C++
exceptions. Fortunately, there is a simple but little-known workaround,
which is to use the following incantation:</p>
<blockquote>
<pre>
extern "C" void straight_to_debugger(unsigned int, EXCEPTION_POINTERS*)
{
throw;
}
extern "C" void (*old_translator)(unsigned, EXCEPTION_POINTERS*)
= _set_se_translator(straight_to_debugger);
</pre>
</blockquote>
This technique doesn't work if the SEH is raised from within a catch
block (or a function called from within a catch block), but it still
eliminates the vast majority of JIT-masking problems.
<h3>How should I handle exceptions?</h3>
<p>Often the best way to deal with exceptions is to not handle them at
all. If you can let them pass through your code and allow destructors to
handle cleanup, your code will be cleaner.</p>
<h4>Avoid <code>catch(...)</code> when possible</h4>
Unfortunately, operating systems other than Windows also wind non-C++
"exceptions" (such as thread cancellation) into the C++ EH machinery, and
there is sometimes no workaround corresponding to the
<code>_set_se_translator</code> hack described above. The result is that
<code>catch(...)</code> can have the effect of making some unexpected
system notification at a point where recovery is impossible look just
like a C++ exception thrown from a reasonable place, invalidating the
usual safe assumptions that destructors and catch blocks have taken valid
steps to ensure program invariants during unwinding.
<p>I reluctantly concede this point to Hillel Y. Sims, after many
long debates in the newsgroups: until all OSes are "fixed", if
every exception were derived from <code>std::exception</code> and
everyone substituted
<code>catch(std::exception&amp;)</code> for <code>catch(...)</code>, the
world would be a better place.</p>
<p>Sometimes, <code>catch(...)</code>, is still the most appropriate
pattern, in spite of bad interactions with OS/platform design choices. If
you have no idea what kind of exception might be thrown and you really
<i>must</i> stop unwinding it's probably still your best bet. One obvious
place where this occurs is at language boundaries.</p>
<hr>
<p>&copy; Copyright David Abrahams 2001-2003. All rights reserved.</p>
<p>Revised
<!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %B, %Y" startspan -->
21 August, 2003<!--webbot bot="Timestamp" endspan i-checksum="34359" -->
</p>
</body>
</html>

683
generic_exception_safety.html
View File

@ -1,683 +0,0 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<!-- saved from url=(0052)http://people.ne.mediaone.net/abrahams/abrahams.html -->
<meta name="generator" content="Microsoft FrontPage 5.0">
<title>Exception-Safety in Generic Components</title>
<meta http-equiv="Content-Type" content="text/html; charset=windows-1252">
<meta content="MSHTML 5.50.4522.1800" name="GENERATOR">
<h1 align="center">Exception-Safety in Generic Components</h1>
<p align="center"><i><b>Lessons Learned from Specifying Exception-Safety
for the C++ Standard Library</b></i>
<h3 align="center">David Abrahams</h3>
<h3 align="center"><a href="mailto:david.abrahams@rcn.com">
david.abrahams@rcn.com</a></h3>
<p><b>Abstract.</b> This paper represents the knowledge accumulated in
response to a real-world need: that the C++ Standard Template Library
exhibit useful and well-defined interactions with exceptions, the
error-handling mechanism built-in to the core C++ language. It explores the
meaning of exception-safety, reveals surprising myths about exceptions and
genericity, describes valuable tools for reasoning about program
correctness, and outlines an automated testing procedure for verifying
exception-safety.
<p><b>Keywords:</b> exception-safety, exceptions, STL, C++
<h2>1 What is exception-safety?</h2>
<p>Informally, exception-safety in a component means that it exhibits
reasonable behavior when an exception is thrown during its execution. For
most people, the term ``reasonable'' includes all the usual
expectations for error-handling: that resources should not be leaked, and
that the program should remain in a well-defined state so that execution
can continue. For most components, it also includes the expectation that
when an error is encountered, it is reported to the caller.
<p>More formally, we can describe a component as minimally exception-safe
if, when exceptions are thrown from within that component, its invariants
are intact. Later on we'll see that at least three different levels of
exception-safety can be usefully distinguished. These distinctions can help
us to describe and reason about the behavior of large systems.
<p>In a generic component, we usually have an additional expectation of
<i>exception-neutrality</i>, which means that exceptions thrown by a
component's type parameters should be propagated, unchanged, to the
component's caller.
<h2>2 Myths and Superstitions</h2>
<p>Exception-safety seems straightforward so far: it doesn't constitute
anything more than we'd expect from code using more traditional
error-handling techniques. It might be worthwhile, however, to examine the
term from a psychological viewpoint. Nobody ever spoke of
``error-safety'' before C++ had exceptions.
<p>It's almost as though exceptions are viewed as a <i>mysterious
attack</i> on otherwise correct code, from which we must protect ourselves.
Needless to say, this doesn't lead to a healthy relationship with error
handling! During standardization, a democratic process which requires broad
support for changes, I encountered many widely-held superstitions. In order
to even begin the discussion of exception-safety in generic components, it
may be worthwhile confronting a few of them.
<p><i>``Interactions between templates and exceptions are not
well-understood.''</i> This myth, often heard from those who consider
these both new language features, is easily disposed of: there simply are
no interactions. A template, once instantiated, works in all respects like
an ordinary class or function. A simple way to reason about the behavior of
a template with exceptions is to think of how a specific instantiation of
that template works. Finally, the genericity of templates should not cause
special concern. Although the component's client supplies part of the
operation (which may, unless otherwise specified, throw arbitrary
exceptions), the same is true of operations using familiar virtual
functions or simple function pointers.
<p><i>``It is well known to be impossible to write an exception-safe
generic container.''</i> This claim is often heard with reference to
an article by Tom Cargill <a title=
"Tom Cargill, ``Exception Handling: A False Sense of Security'', C++ Report, Nov-Dec 1994"
href=
"#reference4"><sup>[4]</sup></a>
in which he explores the problem of exception-safety for a generic stack
template. In his article, Cargill raises many useful questions, but
unfortunately fails to present a solution to his problem.<a title=
"Probably the greatest impediment to a solution in Cargill's case was an unfortunate combination of choices on his part: the interface he chose for his container was incompatible with his particular demands for safety. By changing either one he might have solved the problem."
href=
"#footnote1"><sup>1</sup></a>
He concludes by suggesting that a solution may not be possible.
Unfortunately, his article was read by many as ``proof'' of that
speculation. Since it was published there have been many examples of
exception-safe generic components, among them the C++ standard library
containers.
<p><i>``Dealing with exceptions will slow code down, and templates are
used specifically to get the best possible performance.''</i> A good
implementation of C++ will not devote a single instruction cycle to dealing
with exceptions until one is thrown, and then it can be handled at a speed
comparable with that of calling a function <a title=
"D. R. Musser, ``Introspective Sorting and Selection Algorithms'', Software-Practice and Experience 27(8):983-993, 1997."
href=
"#reference7"><sup>[7]</sup></a>.
That alone gives programs using exceptions performance equivalent to that
of a program which ignores the possibility of errors. Using exceptions can
actually result in faster programs than ``traditional'' error
handling methods for other reasons. First, a catch block clearly indicates
to the compiler which code is devoted to error-handling; it can then be
separated from the usual execution path, improving locality of reference.
Second, code using ``traditional'' error handling must typically
test a return value for errors after every single function call; using
exceptions completely eliminates that overhead.
<p><i>``Exceptions make it more difficult to reason about a program's
behavior.''</i> Usually cited in support of this myth is the way
``hidden'' execution paths are followed during stack-unwinding.
Hidden execution paths are nothing new to any C++ programmer who expects
local variables to be destroyed upon returning from a function:
<blockquote>
<pre>ErrorCode f( int&amp; result ) // 1
{ // 2
X x; // 3
ErrorCode err = x.g( result ); // 4
if ( err != kNoError ) // 5
return err; // 6
// ...More code here...
return kNoError; // 7
}
</pre>
</blockquote>
<p>In the example above, there is a ``hidden'' call to
<code>X::~X()</code> in lines 6 and 7. Granted, using exceptions, there is
no code devoted to error handling visible:
<blockquote>
<pre>int f() // 1
{ // 2
X x; // 3
int result = x.g(); // 4
// ...More code here...
return result; // 5
}
</pre>
</blockquote>
<p>For many programmers more familiar with exceptions, the second example
is actually more readable and understandable than the first. The
``hidden'' code paths include the same calls to destructors of
local variables. In addition, they follow a simple pattern which acts
<i>exactly</i> as though there were a potential return statement after each
function call in case of an exception. Readability is enhanced because the
normal path of execution is unobscured by error-handling, and return values
are freed up to be used in a natural way.
<p>There is an even more important way in which exceptions can enhance
correctness: by allowing simple class invariants. In the first example, if
<code>x</code>'s constructor should need to allocate resources, it has no
way to report a failure: in C++, constructors have no return values. The
usual result when exceptions are avoided is that classes requiring
resources must include a separate initializer function which finishes the
job of construction. The programmer can therefore never be sure, when an
object of class <code>X</code> is used, whether he is handling a
full-fledged <code>X</code> or some abortive attempt to construct one (or
worse: someone simply forgot to call the initializer!)
<h2>3 A contractual basis for exception-safety</h2>
<p>A non-generic component can be described as exception-safe in isolation,
but because of its configurability by client code, exception-safety in a
generic component usually depends on a contract between the component and
its clients. For example, the designer of a generic component might require
that an operation which is used in the component's destructor not throw any
exceptions.<a title=
" It is usually inadvisable to throw an exception from a destructor in C++, since the destructor may itself be called during the stack-unwinding caused by another exception. If the second exception is allowed to propagate beyond the destructor, the program is immediately terminated."
href=
"#footnote2"><sup>2</sup></a>
The generic component might, in return, provide one of the following
guarantees:
<ul>
<li>The <i>basic</i> guarantee: that the invariants of the component are
preserved, and no resources are leaked.
<li>The <i>strong</i> guarantee: that the operation has either completed
successfully or thrown an exception, leaving the program state exactly as
it was before the operation started.
<li>The <i>no-throw</i> guarantee: that the operation will not throw an
exception.
</ul>
<p>The basic guarantee is a simple minimum standard for exception-safety to
which we can hold all components. It says simply that after an exception,
the component can still be used as before. Importantly, the preservation of
invariants allows the component to be destroyed, potentially as part of
stack-unwinding. This guarantee is actually less useful than it might at
first appear. If a component has many valid states, after an exception we
have no idea what state the component is in|only that the state is valid.
The options for recovery in this case are limited: either destruction or
resetting the component to some known state before further use. Consider
the following example:
<blockquote>
<pre>template &lt;class X&gt;
void print_random_sequence()
{
std::vector&lt;X&gt; v(10); // A vector of 10 items
try {
// Provides only the <i>basic</i> guarantee
v.insert( v.begin(), X() );
}
catch(...) {} // ignore any exceptions above
// print the vector's contents
std::cout "(" &lt;&lt; v.size() &lt;&lt; ") ";
std::copy( v.begin(), v.end(),
std::ostream_iterator&lt;X&gt;( std::cout, " " ) );
}
</pre>
</blockquote>
<p>Since all we know about v after an exception is that it is valid, the
function is allowed to print any random sequence of <code>X</code>s.<a
title=
"In practice of course, this function would make an extremely poor random sequence generator!"
href=
"#footnote3"><sup>3</sup></a>
It is ``safe'' in the sense that it is not allowed to crash, but
its output may be unpredictable.
<p>The <i>strong</i> guarantee provides full
``commit-or-rollback'' semantics. In the case of C++ standard
containers, this means, for example, that if an exception is thrown all
iterators remain valid. We also know that the container has exactly the
same elements as before the exception was thrown. A transaction that has no
effects if it fails has obvious benefits: the program state is simple and
predictable in case of an exception. In the C++ standard library, nearly
all of the operations on the node-based containers list, set, multiset,
map, and multimap provide the <i>strong</i> guarantee.<a title=
"It is worth noting that mutating algorithms usually cannot provide the strong guarantee: to roll back a modified element of a range, it must be set back to its previous value using operator=, which itself might throw. In the C++ standard library, there are a few exceptions to this rule, whose rollback behavior consists only of destruction: uninitialized_copy, uninitialized_fill, and uninitialized_fill_n."
href=
"#footnote4"><sup>4</sup></a>).
<p>The <i>no-throw</i> guarantee is the strongest of all, and it says that
an operation is guaranteed not to throw an exception: it always completes
successfully. This guarantee is necessary for most destructors, and indeed
the destructors of C++ standard library components are all guaranteed not
to throw exceptions. The <i>no-throw</i> guarantee turns out to be
important for other reasons, as we shall see.<a title=
"All type parameters supplied by clients of the C++ standard library are required not to throw from their destructors. In return, all components of the C++ standard library provide at least the basic guarantee."
href=
"#footnote5"><sup>5</sup></a>
<h2>4 Legal Wrangling</h2>
<p>Inevitably, the contract can get more complicated: a quid pro quo
arrangement is possible. Some components in the C++ Standard Library give
one guarantee for arbitrary type parameters, but give a stronger guarantee
in exchange for additional promises from the client type that no exceptions
will be thrown. For example, the standard container operation
<code>vector&lt;T&gt;::erase</code> gives the <i>basic</i> guarantee for
any <code>T</code>, but for types whose copy constructor and copy
assignment operator do not throw, it gives the <i>no-throw</i> guarantee.<a
title=
"Similar arrangements might have been made in the C++ standard for many of the mutating algorithms, but were never considered due to time constraints on the standardization process."
href=
"#footnote6"><sup>6</sup></a>
<h2>5 What level of exception-safety should a component specify?</h2>
<p>From a client's point-of-view, the strongest possible level of safety
would be ideal. Of course, the <i>no-throw</i> guarantee is simply
impossible for many operations, but what about the <i>strong</i> guarantee?
For example, suppose we wanted atomic behavior for
<code>vector&lt;T&gt;::insert</code>. Insertion into the middle of a vector
requires copying elements after the insertion point into later positions,
to make room for the new element. If copying an element can fail, rolling
back the operation would require ``undoing'' the previous
copies...which depends on copying again. If copying back should fail (as it
likely would), we have failed to meet our guarantee.
<p>One possible alternative would be to redefine <code>insert</code> to
build the new array contents in a fresh piece of memory each time, and only
destroy the old contents when that has succeeded. Unfortunately, there is a
non-trivial cost if this approach is followed: insertions near the end of a
vector which might have previously caused only a few copies would now cause
every element to be copied. The <i>basic</i> guarantee is a
``natural'' level of safety for this operation, which it can
provide without violating its performance guarantees. In fact all of the
operations in the library appear to have such a ``natural'' level
of safety.
<p>Because performance requirements were already a well-established part of
the draft standard and because performance is a primary goal of the STL,
there was no attempt to specify more safety than could be provided within
those requirements. Although not all of the library gives the <i>strong</i>
guarantee, almost any operation on a standard container which gives the
<i>basic</i> guarantee can be made <i>strong</i> using the ``make a
new copy'' strategy described above:
<blockquote>
<pre>template &lt;class Container, class BasicOp&gt;
void MakeOperationStrong( Container&amp; c, const BasicOp&amp; op )
{
Container tmp(c); // Copy c
op(tmp); // Work on the copy
c.swap(tmp); // Cannot fail<a title=
"Associative containers whose Compare object might throw an exception when copied cannot use this technique, since the swap function might fail."
href=
"#footnote7"><sup>7</sup></a>
}
</pre>
</blockquote>
<p>This technique can be folded into a wrapper class to make a similar
container which provides stronger guarantees (and different performance
characteristics).<a title=
"This suggests another potential use for the oft-wished-for but as yet unseen container traits&lt;&gt; template: automated container selection to meet exceptionsafety constraints."
href=
"#footnote8"><sup>8</sup></a>
<h2>6 Should we take everything we can get?</h2>
<p>By considering a particular implementation, we can hope to discern a
natural level of safety. The danger in using this to establish requirements
for a component is that the implementation might be restricted. If someone
should come up with a more-efficient implementation which we'd like to use,
we may find that it's incompatible with our exception-safety requirements.
One might expect this to be of no concern in the well-explored domains of
data structures and algorithms covered by the STL, but even there, advances
are being made. A good example is the recent <i>introsort</i> algorithm <a
title=
"D. R. Musser, ``Introspective Sorting and Selection Algorithms'', Software-Practice and Experience 27(8):983-993, 1997."
href=
"#reference6"><sup>[6]</sup></a>,
which represents a substantial improvement in worst-case complexity over
the well-established <i>quicksort</i>.
<p>To determine exactly how much to demand of the standard components, I
looked at a typical real-world scenario. The chosen test case was a
``composite container.'' Such a container, built of two or more
standard container components, is not only commonly needed, but serves as a
simple representative case for maintaining invariants in larger systems:
<blockquote>
<pre>// SearchableStack - A stack which can be efficiently searched
// for any value.
template &lt;class T&gt;
class SearchableStack
{
public:
void push(const T&amp; t); // O(log n)
void pop(); // O(log n)
bool contains(const T&amp; t) const; // O(log n)
const T&amp; top() const; // O(1)
private:
std::set&lt;T&gt; set_impl;
std::list&lt;std::set&lt;T&gt;::iterator&gt; list_impl;
};
</pre>
</blockquote>
<p>The idea is that the list acts as a stack of set iterators: every
element goes into the set first, and the resulting position is pushed onto
the list. The invariant is straightforward: the set and the list should
always have the same number of elements, and every element of the set
should be referenced by an element of the list. The following
implementation of the push function is designed to give the <i>strong</i>
guarantee within the natural levels of safety provided by set and list:
<blockquote>
<pre>template &lt;class T&gt; // 1
void SearchableStack&lt;T&gt;::push(const T&amp; t) // 2
{ // 3
set&lt;T&gt;::iterator i = set_impl.insert(t); // 4
try // 5
{ // 6
list_impl.push_back(i); // 7
} // 8
catch(...) // 9
{ // 10
set_impl.erase(i); // 11
throw; // 12
} // 13
} // 14
</pre>
</blockquote>
<p>What does our code actually require of the library? We need to examine
the lines where non-const operations occur:
<ul>
<li>Line 4: if the insertion fails but <code>set_impl</code> is modified
in the process, our invariant is violated. We need to be able to rely on
the <i>strong</i> guarantee from <code>set&lt;T&gt;::insert</code>.
<li>Line 7: likewise, if <code>push_back</code> fails, but
<code>list_impl</code> is modified in the process, our invariant is
violated, so we need to be able to rely on the <i>strong</i> guarantee
from list&lt;T&gt;::insert.
<li>Line 11: here we are ``rolling back'' the insertion on line
4. If this operation should fail, we will be unable to restore our
invariant. We absolutely depend on the <i>no-throw</i> guarantee from
<code>set&lt;T&gt;::erase</code>.<a title=
"One might be tempted to surround the erase operation with a try/catch block to reduce the requirements on set&lt;T&gt; and the problems that arise in case of an exception, but in the end that just begs the question. First, erase just failed and in this case there are no viable alternative ways to produce the necessary result. Second and more generally, because of the variability of its type parameters a generic component can seldom be assured that any alternatives will succeed."
href=
"#footnote9"><sup>9</sup></a>
<li>Line 11: for the same reasons, we also depend on being able to pass
the <code>i</code> to the <code>erase</code> function: we need the
<i>no-throw</i> guarantee from the copy constructor of
<code>set&lt;T&gt;::iterator</code>.
</ul>
<p>I learned a great deal by approaching the question this way during
standardization. First, the guarantee specified for the composite container
actually depends on stronger guarantees from its components (the
<i>no-throw</i> guarantees in line 11). Also, I took advantage of all of
the natural level of safety to implement this simple example. Finally, the
analysis revealed a requirement on iterators which I had previously
overlooked when operations were considered on their own. The conclusion was
that we should provide as much of the natural level of safety as possible.
Faster but less-safe implementations could always be provided as extensions
to the standard components. <sup><a title=
"The prevalent philosophy in the design of STL was that functionality that wasn't essential to all uses should be left out in favor of efficiency, as long as that functionality could be obtained when needed by adapting the base components. This departs from that philosophy, but it would be difficult or impossible to obtain even the basic guarantee by adapting a base component that doesn't already have it."
name="#footnote10">10</a></sup>
<h2>7 Automated testing for exception-safety</h2>
<p>As part of the standardization process, I produced an exception-safe
reference implementation of the STL. Error-handling code is seldom
rigorously tested in real life, in part because it is difficult to cause
error conditions to occur. It is very common to see error-handling code
which crashes the first time it is executed ...in a shipping product! To
bolster confidence that the implementation actually worked as advertised, I
designed an automated test suite, based on an exhaustive technique due to
my colleague Matt Arnold.
<p>The test program started with the basics: reinforcement and
instrumentation, especially of the global operators <code>new</code> and
<code>delete</code>.<sup><a title=
"An excellent discussion on how to fortify memory subsystems can be found in: Steve Maguire, Writing Solid Code, Microsoft Press, Redmond, WA, 1993, ISBN 1-55615- 551-4."
name="#footnote11">11</a></sup>Instances of the components (containers and
algorithms) were created, with type parameters chosen to reveal as many
potential problems as possible. For example, all type parameters were given
a pointer to heap-allocated memory, so that leaking a contained object
would be detected as a memory leak.
<p>Finally, a scheme was designed that could cause an operation to throw an
exception at each possible point of failure. At the beginning of every
client-supplied operation which is allowed to throw an exception, a call to
<code>ThisCanThrow</code> was added. A call to <code>ThisCanThrow</code>
also had to be added everywhere that the generic operation being tested
might throw an exception, for example in the global operator
<code>new</code>, for which an instrumented replacement was supplied.
<blockquote>
<pre>// Use this as a type parameter, e.g. vector&lt;TestClass&gt;
struct TestClass
{
TestClass( int v = 0 )
: p( ThisCanThrow(), new int( v ) ) {}
TestClass( const TestClass&amp; rhs )
: p( ThisCanThrow(), new int( *rhs.p ) ) {}
const TestClass&amp; operator=( const TestClass&amp; rhs )
{ ThisCanThrow(); *p = *rhs.p; }
bool operator==( const TestClass&amp; rhs ) const
{ ThisCanThrow(); return *p == *rhs.p; }
...etc...
~TestClass() { delete p; }
};
</pre>
</blockquote>
<p><code>ThisCanThrow</code> simply decrements a ``throw
counter'' and, if it has reached zero, throws an exception. Each test
takes a form which begins the counter at successively higher values in an
outer loop and repeatedly attempts to complete the operation being tested.
The result is that the operation throws an exception at each successive
step along its execution path that can possibly fail. For example, here is
a simplified version of the function used to test the <i>strong</i>
guarantee: <a title=
"Note that this technique requires that the operation being tested be exception-neutral. If the operation ever tries to recover from an exception and proceed, the throw counter will be negative, and subsequent operations that might fail will not be tested for exception-safety."
href=
"#footnote12"><sup>12</sup></a>
<blockquote>
<pre>extern int gThrowCounter; // The throw counter
void ThisCanThrow()
{
if (gThrowCounter-- == 0)
throw 0;
}
template &lt;class Value, class Operation&gt;
void StrongCheck(const Value&amp; v, const Operation&amp; op)
{
bool succeeded = false;
for (long nextThrowCount = 0; !succeeded; ++nextThrowCount)
{
Value duplicate = v;
try
{
gThrowCounter = nextThrowCount;
op( duplicate ); // Try the operation
succeeded = true;
}
catch(...) // Catch all exceptions
{
bool unchanged = duplicate == v; // Test <i>strong</i> guarantee
assert( unchanged );
}
// Specialize as desired for each container type, to check
// integrity. For example, size() == distance(begin(),end())
CheckInvariant(v); // Check any invariant
}
}
</pre>
</blockquote>
<p>Notably, this kind of testing is much easier and less intrusive with a
generic component than with non-generics, because testing-specific type
parameters can be used without modifying the source code of the component
being tested. Also, generic functions like <code>StrongCheck</code> above
were instrumental in performing the tests on a wide range of values and
operations.
<h2>8 Further Reading</h2>
To my knowledge, there are currently only two descriptions of STL
exception-safety available. The original specification <a title=
"D. Abrahams, Exception Safety in STLport" href=
"#reference2"><sup>[2]</sup></a>
for the reference exception-safe implementation of the STL is an informal
specification, simple and self-explanatory (also verbose), and uses the
<i>basic-</i> and <i>strong-</i>guarantee distinctions outlined in this
article. It explicitly forbids leaks, and differs substantively from the
final C++ standard in the guarantees it makes, though they are largely
identical. I hope to produce an updated version of this document soon.
<p>The description of exception-safety in the C++ Standard <a title=
"International Standard ISO/IEC 14882, Information Technology-Programming Languages-C++, Document Number ISO/IEC 14882-1998"
href=
"#reference1"><sup>[1]</sup></a>
is only slightly more formal, but relies on hard-to-read
``standardese'' and an occasionally subtle web of implication.<a
title=
"The changes to the draft standard which introduced exception-safety were made late in the process, when amendments were likely to be rejected solely on the basis of the number of altered words. Unfortunately, the result compromises clarity somewhat in favor of brevity. Greg Colvin was responsible for the clever language-lawyering needed to minimize the extent of these changes."
href=
"#footnote13"><sup>13</sup></a>
In particular, leaks are not treated directly at all. It does have the
advantage that it <i>is</i> the standard.
<p>The original reference implementation <a title=
"B. Fomitchev, Adapted SGI STL Version 1.0, with exception handling code by D. Abrahams"
href=
"#reference5"><sup>[5]</sup></a>
of the exception-safe STL is an adaptation of an old version of the SGI
STL, designed for C++ compilers with limited features. Although it is not a
complete STL implementation, the code may be easier to read, and it
illustrates a useful base-class technique for eliminating
exception-handling code in constructors. The full test suite <a title=
"D. Abrahams and B. Fomitchev, Exception Handling Test Suite" href=
"#reference3"><sup>[3]</sup></a>
used to validate the reference implementation has been used successfully to
validate all recent versions of the SGI STL, and has been adapted to test
one other vendor's implementation (which failed). As noted on the
documentation page, it also seems to have the power to reveal hidden
compiler bugs, particularly where optimizers interact with
exception-handling code.
<h2>References</h2>
<ol>
<li><a name="reference1">International</a> Standard ISO/IEC 14882,
<i>Information Technology-Programming Languages-C++</i>, Document Number
ISO/IEC 14882-1998, available from <a href=
"http://webstore.ansi.org/ansidocstore/default.asp">http://webstore.ansi.org/ansidocstore/default.asp</a>.
<li><a name="reference2">D.</a> Abrahams, <i>Exception Safety in
STLport</i>, available at <a href=
"http://www.stlport.org/doc/exception_safety.html">http://www.stlport.org/doc/exception_safety.html</a>.
<li><a name="reference3">D.</a> Abrahams and B. Fomitchev, <i>Exception
Handling Test Suite</i>, available at <a href=
"http://www.stlport.org/doc/eh_testsuite.html">http://www.stlport.org/doc/eh_testsuite.html</a>.
<li><a name="reference4">Tom</a> Cargill, ``Exception Handling:
A False Sense of Security,'' C++ Report, Nov-Dec 1994, also
available at <a href=
"http://www.awprofessional.com/content/images/020163371x/supplements/Exception_Handling_Article.html">http://www.awprofessional.com/content/images/020163371x/supplements/Exception_Handling_Article.html</a>.
<li><a name="reference5">B.</a> Fomitchev, <i>Adapted SGI STL Version
1.0</i>, with exception handling code by D. Abrahams, available at <a
href="http://www.stlport.org">http://www.stlport.org</a>.
<li><a name="reference6">D.</a> R. Musser, ``Introspective Sorting
and Selection Algorithms,'' <i>Software-Practice and Experience</i>
27(8):983-993, 1997.
<li><a name="reference7">Bjarne</a> Stroustrup, <i>The Design And
Evolution of C++</i>. Addison Wesley, Reading, MA, 1995, ISBN
0-201-54330-3, Section 16.9.1.
</ol>
<h2>Footnotes</h2>
<p><a name="footnote1">1</a> Probably the greatest impediment to a solution
in Cargill's case was an unfortunate combination of choices on his part:
the interface he chose for his container was incompatible with his
particular demands for safety. By changing either one he might have solved
the problem.
<p><a name="footnote2">2</a> It is usually inadvisable to throw an
exception from a destructor in C++, since the destructor may itself be
called during the stack-unwinding caused by another exception. If the
second exception is allowed to propagate beyond the destructor, the program
is immediately terminated.
<p><a name="footnote3">3</a> In practice of course, this function would
make an extremely poor random sequence generator!
<p><a name="footnote4">4</a> It is worth noting that mutating algorithms
usually cannot provide the <i>strong</i> guarantee: to roll back a modified
element of a range, it must be set back to its previous value using
<code>operator=</code>, which itself might throw. In the C++ standard
library, there are a few exceptions to this rule, whose rollback behavior
consists only of destruction: <code>uninitialized_copy</code>,
<code>uninitialized_fill</code>, and <code>uninitialized_fill_n</code>.
<p><a name="footnote5">5</a> All type parameters supplied by clients of the
C++ standard library are required not to throw from their destructors. In
return, all components of the C++ standard library provide at least the
<i>basic</i> guarantee.
<p><a name="footnote6">6</a> Similar arrangements might have been made in
the C++ standard for many of the mutating algorithms, but were never
considered due to time constraints on the standardization process.
<p><a name="footnote7">7</a> Associative containers whose
<code>Compare</code> object might throw an exception when copied cannot use
this technique, since the swap function might fail.
<p><a name="footnote8">8</a> This suggests another potential use for the
oft-wished-for but as yet unseen <code>container_traits&lt;&gt;</code>
template: automated container selection to meet exception-safety
constraints.
<p><a name="footnote9">9</a> One might be tempted to surround the erase
operation with a <code>try</code>/<code>catch</code> block to reduce the
requirements on <code>set&lt;T&gt;</code> and the problems that arise in
case of an exception, but in the end that just begs the question. First,
erase just failed and in this case there are no viable alternative ways to
produce the necessary result. Second and more generally, because of the
variability of its type parameters a generic component can seldom be
assured that any alternatives will succeed.
<p><a name="footnote10">10</a> The prevalent philosophy in the design of
STL was that functionality that wasn't essential to all uses should be left
out in favor of efficiency, as long as that functionality could be obtained
when needed by adapting the base components. This departs from that
philosophy, but it would be difficult or impossible to obtain even the
<i>basic</i> guarantee by adapting a base component that doesn't already
have it.
<p><a name="footnote11">11</a> An excellent discussion on how to fortify
memory subsystems can be found in: Steve Maguire, Writing Solid Code,
Microsoft Press, Redmond, WA, 1993, ISBN 1-55615- 551-4.
<p><a name="footnote12">12</a> Note that this technique requires that the
operation being tested be exception-neutral. If the operation ever tries to
recover from an exception and proceed, the throw counter will be negative,
and subsequent operations that might fail will not be tested for
exception-safety.
<p><a name="footnote13">13</a> The changes to the draft standard which
introduced exception-safety were made late in the process, when amendments
were likely to be rejected solely on the basis of the number of altered
words. Unfortunately, the result compromises clarity somewhat in favor of
brevity. Greg Colvin was responsible for the clever language-lawyering
needed to minimize the extent of these changes.

451
generic_programming.html
View File

@ -1,451 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head>
<meta name="generator" content=
"Microsoft FrontPage 5.0">
<meta http-equiv="Content-Type" content=
"text/html; charset=windows-1252">
<meta name="GENERATOR" content="Microsoft FrontPage 4.0">
<meta name="ProgId" content="FrontPage.Editor.Document">
<title>Generic Programming Techniques</title>
</head>
<body bgcolor="#FFFFFF" text="#000000">
<img src="../boost.png" alt="boost.png (6897 bytes)" align="center"
width="277" height="86">
<h1>Generic Programming Techniques</h1>
<p>This is an incomplete survey of some of the generic programming
techniques used in the <a href="../index.htm">boost</a> libraries.</p>
<h2>Table of Contents</h2>
<ul>
<li><a href="#introduction">Introduction</a></li>
<li><a href="#concept">The Anatomy of a Concept</a></li>
<li><a href="#traits">Traits</a></li>
<li><a href="#tag_dispatching">Tag Dispatching</a></li>
<li><a href="#adaptors">Adaptors</a></li>
<li><a href="#type_generator">Type Generators</a></li>
<li><a href="#object_generator">Object Generators</a></li>
<li><a href="#policy">Policy Classes</a></li>
</ul>
<h2><a name="introduction">Introduction</a></h2>
<p>Generic programming is about generalizing software components so that
they can be easily reused in a wide variety of situations. In C++, class
and function templates are particularly effective mechanisms for generic
programming because they make the generalization possible without
sacrificing efficiency.</p>
<p>As a simple example of generic programming, we will look at how one
might generalize the <tt>memcpy()</tt> function of the C standard
library. An implementation of <tt>memcpy()</tt> might look like the
following:<br>
<br>
</p>
<blockquote>
<pre>void* memcpy(void* region1, const void* region2, size_t n)
{
const char* first = (const char*)region2;
const char* last = ((const char*)region2) + n;
char* result = (char*)region1;
while (first != last)
*result++ = *first++;
return result;
}
</pre>
</blockquote>
The <tt>memcpy()</tt> function is already generalized to some extent by
the use of <tt>void*</tt> so that the function can be used to copy arrays
of different kinds of data. But what if the data we would like to copy is
not in an array? Perhaps it is in a linked list. Can we generalize the
notion of copy to any sequence of elements? Looking at the body of
<tt>memcpy()</tt>, the function's <b><i>minimal requirements</i></b> are
that it needs to <i>traverse</i> through the sequence using some sort
of pointer, <i>access</i> elements pointed to, <i>write</i> the elements
to the destination, and <i>compare</i> pointers to know when to stop. The
C++ standard library groups requirements such as these into
<b><i>concepts</i></b>, in this case the <a href=
"http://www.sgi.com/tech/stl/InputIterator.html">Input Iterator</a>
concept (for <tt>region2</tt>) and the <a href=
"http://www.sgi.com/tech/stl/OutputIterator.html">Output Iterator</a>
concept (for <tt>region1</tt>).
<p>If we rewrite the <tt>memcpy()</tt> as a function template, and use
the <a href="http://www.sgi.com/tech/stl/InputIterator.html">Input
Iterator</a> and <a href=
"http://www.sgi.com/tech/stl/OutputIterator.html">Output Iterator</a>
concepts to describe the requirements on the template parameters, we can
implement a highly reusable <tt>copy()</tt> function in the following
way:<br>
<br>
</p>
<blockquote>
<pre>template &lt;typename InputIterator, typename OutputIterator&gt;
OutputIterator
copy(InputIterator first, InputIterator last, OutputIterator result)
{
while (first != last)
*result++ = *first++;
return result;
}
</pre>
</blockquote>
<p>Using the generic <tt>copy()</tt> function, we can now copy elements
from any kind of sequence, including a linked list that exports iterators
such as <tt>std::<a href=
"http://www.sgi.com/tech/stl/List.html">list</a></tt>.<br>
<br>
</p>
<blockquote>
<pre>#include &lt;list&gt;
#include &lt;vector&gt;
#include &lt;iostream&gt;
int main()
{
const int N = 3;
std::vector&lt;int&gt; region1(N);
std::list&lt;int&gt; region2;
region2.push_back(1);
region2.push_back(0);
region2.push_back(3);
std::copy(region2.begin(), region2.end(), region1.begin());
for (int i = 0; i &lt; N; ++i)
std::cout &lt;&lt; region1[i] &lt;&lt; " ";
std::cout &lt;&lt; std::endl;
}
</pre>
</blockquote>
<h2><a name="concept">Anatomy of a Concept</a></h2>
A <b><i>concept</i></b> is a set of requirements
consisting of valid expressions, associated types, invariants, and
complexity guarantees. A type that satisfies the requirements is
said to <b><i>model</i></b> the concept. A concept can extend the
requirements of another concept, which is called
<b><i>refinement</i></b>.
<ul>
<li><a name="valid_expression"><b>Valid Expressions</b></a> are C++
expressions which must compile successfully for the objects involved in
the expression to be considered <i>models</i> of the concept.</li>
<li><a name="associated_type"><b>Associated Types</b></a> are types
that are related to the modeling type in that they participate in one
or more of the valid expressions. Typically associated types can be
accessed either through typedefs nested within a class definition for
the modeling type, or they are accessed through a <a href=
"#traits">traits class</a>.</li>
<li><b>Invariants</b> are run-time characteristics of the objects that
must always be true, that is, the functions involving the objects must
preserve these characteristics. The invariants often take the form of
pre-conditions and post-conditions.</li>
<li><b>Complexity Guarantees</b> are maximum limits on how long the
execution of one of the valid expressions will take, or how much of
various resources its computation will use.</li>
</ul>
<p>The concepts used in the C++ Standard Library are documented at the <a
href="http://www.sgi.com/tech/stl/table_of_contents.html">SGI STL
site</a>.</p>
<h2><a name="traits">Traits</a></h2>
<p>A traits class provides a way of associating information with a
compile-time entity (a type, integral constant, or address). For example,
the class template <tt><a href=
"http://www.sgi.com/tech/stl/iterator_traits.html">std::iterator_traits&lt;T&gt;</a></tt>
looks something like this:</p>
<blockquote>
<pre>template &lt;class Iterator&gt;
struct iterator_traits {
typedef ... iterator_category;
typedef ... value_type;
typedef ... difference_type;
typedef ... pointer;
typedef ... reference;
};
</pre>
</blockquote>
The traits' <tt>value_type</tt> gives generic code the type which the
iterator is "pointing at", while the <tt>iterator_category</tt> can be
used to select more efficient algorithms depending on the iterator's
capabilities.
<p>A key feature of traits templates is that they're
<i>non-intrusive</i>: they allow us to associate information with
arbitrary types, including built-in types and types defined in
third-party libraries, Normally, traits are specified for a particular
type by (partially) specializing the traits template.</p>
<p>For an in-depth description of <tt>std::iterator_traits</tt>, see <a
href="http://www.sgi.com/tech/stl/iterator_traits.html">this page</a>
provided by SGI. Another very different expression of the traits idiom in
the standard is <tt>std::numeric_limits&lt;T&gt;</tt> which provides
constants describing the range and capabilities of numeric types.</p>
<h2><a name="tag_dispatching">Tag Dispatching</a></h2>
<p>Tag dispatching is a way of using function overloading to
dispatch based on properties of a type, and is often used hand in
hand with traits classes. A good example of this synergy is the
implementation of the <a href=
"http://www.sgi.com/tech/stl/advance.html"><tt>std::advance()</tt></a>
function in the C++ Standard Library, which increments an iterator
<tt>n</tt> times. Depending on the kind of iterator, there are different
optimizations that can be applied in the implementation. If the iterator
is <a href="http://www.sgi.com/tech/stl/RandomAccessIterator.html">random
access</a> (can jump forward and backward arbitrary distances), then the
<tt>advance()</tt> function can simply be implemented with <tt>i +=
n</tt>, and is very efficient: constant time. Other iterators must be
<tt>advance</tt>d in steps, making the operation linear in n. If the
iterator is <a href=
"http://www.sgi.com/tech/stl/BidirectionalIterator.html">bidirectional</a>,
then it makes sense for <tt>n</tt> to be negative, so we must decide
whether to increment or decrement the iterator.</p>
<p>The relation between tag dispatching and traits classes is that the
property used for dispatching (in this case the
<tt>iterator_category</tt>) is often accessed through a traits class. The
main <tt>advance()</tt> function uses the <a href=
"http://www.sgi.com/tech/stl/iterator_traits.html"><tt>iterator_traits</tt></a>
class to get the <tt>iterator_category</tt>. It then makes a call the the
overloaded <tt>advance_dispatch()</tt> function. The appropriate
<tt>advance_dispatch()</tt> is selected by the compiler based on whatever
type the <tt>iterator_category</tt> resolves to, either <a href=
"http://www.sgi.com/tech/stl/input_iterator_tag.html"><tt>input_iterator_tag</tt></a>,
<a href=
"http://www.sgi.com/tech/stl/bidirectional_iterator_tag.html"><tt>bidirectional_iterator_tag</tt></a>,
or <a href=
"http://www.sgi.com/tech/stl/random_access_iterator_tag.html"><tt>random_access_iterator_tag</tt></a>.
A <b><i>tag</i></b> is simply a class whose only purpose is to convey
some property for use in tag dispatching and similar techniques. Refer to
<a href="http://www.sgi.com/tech/stl/iterator_tags.html">this page</a>
for a more detailed description of iterator tags.</p>
<blockquote>
<pre>namespace std {
struct input_iterator_tag { };
struct bidirectional_iterator_tag { };
struct random_access_iterator_tag { };
namespace detail {
template &lt;class InputIterator, class Distance&gt;
void advance_dispatch(InputIterator&amp; i, Distance n, <b>input_iterator_tag</b>) {
while (n--) ++i;
}
template &lt;class BidirectionalIterator, class Distance&gt;
void advance_dispatch(BidirectionalIterator&amp; i, Distance n,
<b>bidirectional_iterator_tag</b>) {
if (n &gt;= 0)
while (n--) ++i;
else
while (n++) --i;
}
template &lt;class RandomAccessIterator, class Distance&gt;
void advance_dispatch(RandomAccessIterator&amp; i, Distance n,
<b>random_access_iterator_tag</b>) {
i += n;
}
}
template &lt;class InputIterator, class Distance&gt;
void advance(InputIterator&amp; i, Distance n) {
typename <b>iterator_traits&lt;InputIterator&gt;::iterator_category</b> category;
detail::advance_dispatch(i, n, <b>category</b>);
}
}
</pre>
</blockquote>
<h2><a name="adaptors">Adaptors</a></h2>
<p>An <i>adaptor</i> is a class template which builds on another type or
types to provide a new interface or behavioral variant. Examples of
standard adaptors are <a href=
"http://www.sgi.com/tech/stl/ReverseIterator.html">std::reverse_iterator</a>,
which adapts an iterator type by reversing its motion upon
increment/decrement, and <a href=
"http://www.sgi.com/tech/stl/stack.html">std::stack</a>, which adapts a
container to provide a simple stack interface.</p>
<p>A more comprehensive review of the adaptors in the standard can be
found <a href="http://portal.acm.org/citation.cfm?id=249118.249120">
here</a>.</p>
<h2><a name="type_generator">Type Generators</a></h2>
<p><b>Note:</b> The <i>type generator</i> concept has largely been
superseded by the more refined notion of a <a href=
"../libs/mpl/doc/refmanual/metafunction.html"><i>metafunction</i></a>. See
<i><a href="http://www.boost-consulting.com/mplbook">C++ Template
Metaprogramming</a></i> for an in-depth discussion of metafunctions.</p>
<p>A <i>type generator</i> is a template whose only purpose is to
synthesize a new type or types based on its template argument(s)<a href=
"#1">[1]</a>. The generated type is usually expressed as a nested typedef
named, appropriately <tt>type</tt>. A type generator is usually used to
consolidate a complicated type expression into a simple one. This example
uses an old version of <tt><a href=
"../libs/iterator/doc/iterator_adaptor.html">iterator_adaptor</a></tt>
whose design didn't allow derived iterator types. As a result, every
adapted iterator had to be a specialization of <tt>iterator_adaptor</tt>
itself and generators were a convenient way to produce those types.</p>
<blockquote>
<pre>template &lt;class Predicate, class Iterator,
class Value = <i>complicated default</i>,
class Reference = <i>complicated default</i>,
class Pointer = <i>complicated default</i>,
class Category = <i>complicated default</i>,
class Distance = <i>complicated default</i>
&gt;
struct filter_iterator_generator {
typedef iterator_adaptor&lt;
Iterator,filter_iterator_policies&lt;Predicate,Iterator&gt;,
Value,Reference,Pointer,Category,Distance&gt; <b>type</b>;
};
</pre>
</blockquote>
<p>Now, that's complicated, but producing an adapted filter iterator
using the generator is much easier. You can usually just write:</p>
<blockquote>
<pre>boost::filter_iterator_generator&lt;my_predicate,my_base_iterator&gt;::type
</pre>
</blockquote>
<h2><a name="object_generator">Object Generators</a></h2>
<p>An <i>object generator</i> is a function template whose only purpose
is to construct a new object out of its arguments. Think of it as a kind
of generic constructor. An object generator may be more useful than a
plain constructor when the exact type to be generated is difficult or
impossible to express and the result of the generator can be passed
directly to a function rather than stored in a variable. Most Boost
object generators are named with the prefix "<tt>make_</tt>", after
<tt>std::<a href=
"http://www.sgi.com/tech/stl/pair.html">make_pair</a>(const&nbsp;T&amp;,&nbsp;const&nbsp;U&amp;)</tt>.</p>
<p>For example, given:</p>
<blockquote>
<pre>struct widget {
void tweak(int);
};
std::vector&lt;widget *&gt; widget_ptrs;
</pre>
</blockquote>
By chaining two standard object generators, <tt>std::<a href=
"http://www.dinkumware.com/htm_cpl/functio2.html#bind2nd">bind2nd</a>()</tt>
and <tt>std::<a href=
"http://www.dinkumware.com/htm_cpl/functio2.html#mem_fun">mem_fun</a>()</tt>,
we can easily tweak all widgets:
<blockquote>
<pre>void tweak_all_widgets1(int arg)
{
for_each(widget_ptrs.begin(), widget_ptrs.end(),
<b>bind2nd</b>(std::<b>mem_fun</b>(&amp;widget::tweak), arg));
}
</pre>
</blockquote>
<p>Without using object generators the example above would look like
this:</p>
<blockquote>
<pre>void tweak_all_widgets2(int arg)
{
for_each(struct_ptrs.begin(), struct_ptrs.end(),
<b>std::binder2nd&lt;std::mem_fun1_t&lt;void, widget, int&gt; &gt;</b>(
std::<b>mem_fun1_t&lt;void, widget, int&gt;</b>(&amp;widget::tweak), arg));
}
</pre>
</blockquote>
<p>As expressions get more complicated the need to reduce the verbosity
of type specification gets more compelling.</p>
<h2><a name="policy">Policy Classes</a></h2>
<p>A policy class is a template parameter used to transmit behavior. An
example from the standard library is <tt>std::<a href=
"http://www.dinkumware.com/htm_cpl/memory.html#allocator">allocator</a></tt>,
which supplies memory management behaviors to standard <a href=
"http://www.sgi.com/tech/stl/Container.html">containers</a>.</p>
<p>Policy classes have been explored in detail by <a href=
"http://www.moderncppdesign.com/">Andrei Alexandrescu</a> in <a href=
"http://www.informit.com/articles/article.asp?p=167842">this chapter</a>
of his book, <i>Modern C++ Design</i>. He writes:</p>
<blockquote>
<p>In brief, policy-based class design fosters assembling a class with
complex behavior out of many little classes (called policies), each of
which takes care of only one behavioral or structural aspect. As the
name suggests, a policy establishes an interface pertaining to a
specific issue. You can implement policies in various ways as long as
you respect the policy interface.</p>
<p>Because you can mix and match policies, you can achieve a
combinatorial set of behaviors by using a small core of elementary
components.</p>
</blockquote>
<p>Andrei's description of policy classes suggests that their power is
derived from granularity and orthogonality. Less-granular policy
interfaces have been shown to work well in practice, though. <a href=
"http://cvs.sourceforge.net/viewcvs.py/*checkout*/boost/boost/libs/utility/Attic/iterator_adaptors.pdf">
This paper</a> describes an old version of <tt><a href=
"../libs/iterator/doc/iterator_adaptor.html">iterator_adaptor</a></tt>
that used non-orthogonal policies. There is also precedent in the
standard library: <tt><a href=
"http://www.dinkumware.com/htm_cpl/string2.html#char_traits">std::char_traits</a></tt>,
despite its name, acts as a policies class that determines the behaviors
of <a href=
"http://www.dinkumware.com/htm_cpl/string2.html#basic_string">std::basic_string</a>.</p>
<h2>Notes</h2>
<a name="1">[1]</a> Type generators are sometimes viewed as a workaround
for the lack of ``templated typedefs'' in C++.
<hr>
<p>Revised
<!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %b %Y" startspan -->06 Nov 2007<!--webbot bot="Timestamp" endspan i-checksum="15272" -->
</p>
<p>© Copyright David Abrahams 2001.</p>
<p>Distributed under the Boost Software License, Version 1.0. See
<a href="http://www.boost.org/LICENSE_1_0.txt">www.boost.org/LICENSE_1_0.txt</a></p>
</body>
</html>