boost_1_45_0/libs/serialization/doc/derivation.html - nest-learning-thermostat/5.0/boost - Git at Google

 <!doctype HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
 <html>
 <!--
 (C) Copyright 2002-4 Robert Ramey - http://www.rrsd.com .
 Use, modification and distribution is subject to the Boost Software
 License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
 http://www.boost.org/LICENSE_1_0.txt)
 -->
 <head>
 <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
 <link rel="stylesheet" type="text/css" href="../../../boost.css">
 <link rel="stylesheet" type="text/css" href="style.css">
 <title>Serialization - Derivation from an Existing Archive</title>
 </head>
 <body link="#0000ff" vlink="#800080">
 <table border="0" cellpadding="7" cellspacing="0" width="100%" summary="header">
   <tr>
     <td valign="top" width="300">
       <h3><a href="../../../index.htm"><img height="86" width="277" alt="C++ Boost" src="../../../boost.png" border="0"></a></h3>
     </td>
     <td valign="top">
       <h1 align="center">Serialization</h1>
       <h2 align="center">Derivation from an Existing Archive</h2>
     </td>
   </tr>
 </table>
 <hr>
 <dl class="page-index">
 </dl>

 <h3>Log Archive</h3>
 It may happen that one wants to create a new archive class by derivation from one
 of the included ones. Included is a sample program that shows how to derive a
 new archive from one of the ones included with the library.  The first example is
 <a href="../example/demo_log.cpp" target="demo_log_cpp">
 demo_log.cpp</a>.
 <p>
 This derivation from the xml archive writes output in xml without the extra
 information required to read the data back into the application.  It might be
 used to export one's data as simple xml for other applications or for logging
 data while debugging.
 <p>
 To this end it is derived from the included xml archive and the save functions for
 some types are specialized for this application.
 <p>
 The serialization library is
 implemented using the <b>C</b>uriously <b>R</b>ecurring <b>T</b>emplate
 <b>P</b>attern (<b>CRTP</b>).  Also, all common code is factored out into
 separate modules to minimize code repetition. This makes derivation from
 an existing archive less straight forward than it would otherwise be.
 <p>
 This example illustrates several issues that have to be addressed when doing
 something like this
 <ol>
     <li><i>It is derived from</i> <code style="white-space: normal">xml_oarchive_impl<log_archive></code>
         <b>NOT</b> <code style="white-space: normal">xml_oarchive</code> <br>
 As described in the comments in
 <a href="../../../boost/archive/xml_oarchive.hpp" target="xml_oarchive_hpp">xml_oarchive.hpp</a>.
 <code style="white-space: normal">xml_oarchive</code> really a shorthand name for
 <code style="white-space: normal">xml_oarchive_impl&lt;xml_oarchive&gt;</code>.  So we should derive
 from <code style="white-space: normal">xml_oarchive_impl&lt;log_archive&gt;</code> rather
 than <code style="white-space: normal">xml_oarchive</code>.
 <pre><code>
 class log_archive :
     // don't derive from xml_oarchive !!!
     public xml_oarchive_impl&lt;log_archive&gt;
 {
 ...
 </code></pre>
     <li><i>Note the</i> <code style="white-space: normal">log_archive</code> <i>between the</i> &lt;&gt;
 This is required so that base classes can downcast their <code style="white-space: normal">this</code> pointer
 to the most derived class.  This is referred to as <b>C</b>uriously <b>R</b>ecurring
 <b>T</b>emplate <b>P</b>attern (<b>CRTP</b>) <a href="bibliography.html#11">[11]</a>.
 It is used to implement static polymorphism.
     <li><i>Base classes need to be explicitly given access to the derived class.</i>
 This can be done by making members public or by including friend declarations for
 the base classes.
 <pre><code>
     friend class detail::common_oarchive&lt;log_archive&gt;;
     friend class basic_xml_oarchive&lt;log_archive&gt;;
     friend class boost::serialization::save_access;
 </code></pre>

     <li><i></i>Reread <a target="detail" href="headers.html#archiveinternals">Archive Internals</a>.
 This describes the class hierarchy so that you know what to override.
     <li><i>Note the usage of PFTO.</i>  Some compilers fail to provide support for
 partial function template ordering.  The serialization library works around by
 using <a target="detail" href="implementation.html#functiontemplateordering">
 <b>P</b>artial <b>F</b>unction <b>T</b>emplate <b>O</b>rdering</a> in several places.
 In archive classes, this takes
 . This is done
 in several places, including the archive classes themselves.
     <li><i>Base class functions will usually need to be explicitly invoked</i>
 We commonly specialize the function name <code style="white-space: normal">save_override</code>
 for saving primitives. Usage of a function name in a derived class
 "hides" similarly named functions of the base class.  That is,
 function name overloading doesn't automatically
 include base classes.  To address this, we can use:
 <pre><code>
     using xml_oarchive_impl&lt;derived_t&gt;::save;
     void save(const unsigned int t);
     ...
 </code></pre>
 which should work on conforming compilers. However, I have found
 that the following equivalent works on more compilers.
 <pre><code>
     // default fall through for any types not specified here
     template&lt;class T&gt;
     void save(const T & t){
         xml_oarchive_impl&lt;derived_t&gt;::save(t);
     }
     void save(const unsigned int t);
     ...
 </code></pre>
 so it's what I use.
     <li><i>Template definitions of base classes may have to be explicitly instantiated.</i>
     The demo includes
 <pre><code>
 // explicitly instantiate for this type of binary stream
 #include &lt;boost/archive/basic_binary_oprimitive.ipp&gt;
 </code></pre>
 for just this purpose.  Failure to include required template definitions
 will result in undefined symbol errors when the program is linked.
     <li><i>Without alteration, this class cannot be further derived from</i><br>
 Base classes using <b>CRTP</b> must be templates with a parameter corresponding to
 the most derived class.  As presented here, this class doesn't qualify, so
 it cannot be used as a base class.  In order to derive further from this class,
 it would have to be reorganized along the lines of the original <code style="white-space: normal">xml_oarchive</code>.
 Specifically, it would look something like:
 <pre><code>
 template&lt;class Archive&gt;
 class log_archive_impl :
     // don't derive from xml_oarchive !!!
     public xml_oarchive_impl&lt;Archive&gt;
 {
     ...
 );

 // do not derived from this class !!!
 class log_archive :
     public log_archive_impl&lt;log_archive&gt;
 {
 public:
     log_archive(std::ostream & os, unsigned int flags = 0) :
         log_archive_impl&lt;xml_oarchive&gt;(os, flags)
     {}
 };
 </code></pre>

 </ol>

 <hr>
 <p><i>&copy; Copyright <a href="http://www.rrsd.com">Robert Ramey</a> 2002-2004.
 Distributed under the Boost Software License, Version 1.0. (See
 accompanying file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
 </i></p>
 </body>
 </html>
	<!doctype HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
	<html>
	<!--
	(C) Copyright 2002-4 Robert Ramey - http://www.rrsd.com .
	Use, modification and distribution is subject to the Boost Software
	License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
	http://www.boost.org/LICENSE_1_0.txt)
	-->
	<head>
	<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
	<link rel="stylesheet" type="text/css" href="../../../boost.css">
	<link rel="stylesheet" type="text/css" href="style.css">
	<title>Serialization - Derivation from an Existing Archive</title>
	</head>
	<body link="#0000ff" vlink="#800080">
	<table border="0" cellpadding="7" cellspacing="0" width="100%" summary="header">
	<tr>
	<td valign="top" width="300">
	<h3><a href="../../../index.htm"><img height="86" width="277" alt="C++ Boost" src="../../../boost.png" border="0"></a></h3>
	</td>
	<td valign="top">
	<h1 align="center">Serialization</h1>
	<h2 align="center">Derivation from an Existing Archive</h2>
	</td>
	</tr>
	</table>
	<hr>
	<dl class="page-index">
	</dl>

	<h3>Log Archive</h3>
	It may happen that one wants to create a new archive class by derivation from one
	of the included ones. Included is a sample program that shows how to derive a
	new archive from one of the ones included with the library. The first example is
	<a href="../example/demo_log.cpp" target="demo_log_cpp">
	demo_log.cpp</a>.
	<p>
	This derivation from the xml archive writes output in xml without the extra
	information required to read the data back into the application. It might be
	used to export one's data as simple xml for other applications or for logging
	data while debugging.
	<p>
	To this end it is derived from the included xml archive and the save functions for
	some types are specialized for this application.
	<p>
	The serialization library is
	implemented using the <b>C</b>uriously <b>R</b>ecurring <b>T</b>emplate
	<b>P</b>attern (<b>CRTP</b>). Also, all common code is factored out into
	separate modules to minimize code repetition. This makes derivation from
	an existing archive less straight forward than it would otherwise be.
	<p>
	This example illustrates several issues that have to be addressed when doing
	something like this
	<ol>
	<li><i>It is derived from</i> <code style="white-space: normal">xml_oarchive_impl<log_archive></code>
	<b>NOT</b> <code style="white-space: normal">xml_oarchive</code> <br>
	As described in the comments in
	<a href="../../../boost/archive/xml_oarchive.hpp" target="xml_oarchive_hpp">xml_oarchive.hpp</a>.
	<code style="white-space: normal">xml_oarchive</code> really a shorthand name for
	<code style="white-space: normal">xml_oarchive_impl<xml_oarchive></code>. So we should derive
	from <code style="white-space: normal">xml_oarchive_impl<log_archive></code> rather
	than <code style="white-space: normal">xml_oarchive</code>.
	<pre><code>
	class log_archive :
	// don't derive from xml_oarchive !!!
	public xml_oarchive_impl<log_archive>
	{
	...
	</code></pre>
	<li><i>Note the</i> <code style="white-space: normal">log_archive</code> <i>between the</i> <>
	This is required so that base classes can downcast their <code style="white-space: normal">this</code> pointer
	to the most derived class. This is referred to as <b>C</b>uriously <b>R</b>ecurring
	<b>T</b>emplate <b>P</b>attern (<b>CRTP</b>) <a href="bibliography.html#11">[11]</a>.
	It is used to implement static polymorphism.
	<li><i>Base classes need to be explicitly given access to the derived class.</i>
	This can be done by making members public or by including friend declarations for
	the base classes.
	<pre><code>
	friend class detail::common_oarchive<log_archive>;
	friend class basic_xml_oarchive<log_archive>;
	friend class boost::serialization::save_access;
	</code></pre>

	<li><i></i>Reread <a target="detail" href="headers.html#archiveinternals">Archive Internals</a>.
	This describes the class hierarchy so that you know what to override.
	<li><i>Note the usage of PFTO.</i> Some compilers fail to provide support for
	partial function template ordering. The serialization library works around by
	using <a target="detail" href="implementation.html#functiontemplateordering">
	<b>P</b>artial <b>F</b>unction <b>T</b>emplate <b>O</b>rdering</a> in several places.
	In archive classes, this takes
	. This is done
	in several places, including the archive classes themselves.
	<li><i>Base class functions will usually need to be explicitly invoked</i>
	We commonly specialize the function name <code style="white-space: normal">save_override</code>
	for saving primitives. Usage of a function name in a derived class
	"hides" similarly named functions of the base class. That is,
	function name overloading doesn't automatically
	include base classes. To address this, we can use:
	<pre><code>
	using xml_oarchive_impl<derived_t>::save;
	void save(const unsigned int t);
	...
	</code></pre>
	which should work on conforming compilers. However, I have found
	that the following equivalent works on more compilers.
	<pre><code>
	// default fall through for any types not specified here
	template<class T>
	void save(const T & t){
	xml_oarchive_impl<derived_t>::save(t);
	}
	void save(const unsigned int t);
	...
	</code></pre>
	so it's what I use.
	<li><i>Template definitions of base classes may have to be explicitly instantiated.</i>
	The demo includes
	<pre><code>
	// explicitly instantiate for this type of binary stream
	#include <boost/archive/basic_binary_oprimitive.ipp>
	</code></pre>
	for just this purpose. Failure to include required template definitions
	will result in undefined symbol errors when the program is linked.
	<li><i>Without alteration, this class cannot be further derived from</i><br>
	Base classes using <b>CRTP</b> must be templates with a parameter corresponding to
	the most derived class. As presented here, this class doesn't qualify, so
	it cannot be used as a base class. In order to derive further from this class,
	it would have to be reorganized along the lines of the original <code style="white-space: normal">xml_oarchive</code>.
	Specifically, it would look something like:
	<pre><code>
	template<class Archive>
	class log_archive_impl :
	// don't derive from xml_oarchive !!!
	public xml_oarchive_impl<Archive>
	{
	...
	);

	// do not derived from this class !!!
	class log_archive :
	public log_archive_impl<log_archive>
	{
	public:
	log_archive(std::ostream & os, unsigned int flags = 0) :
	log_archive_impl<xml_oarchive>(os, flags)
	{}
	};
	</code></pre>

	</ol>

	<hr>
	<p><i>© Copyright <a href="http://www.rrsd.com">Robert Ramey</a> 2002-2004.
	Distributed under the Boost Software License, Version 1.0. (See
	accompanying file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
	</i></p>
	</body>
	</html>