<!DOCTYPE html PUBLIC '-//W3C//DTD XHTML 1.0 Strict//EN' 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd'> <html xmlns='http://www.w3.org/1999/xhtml' xml:lang='en' lang='en'> <head> <meta http-equiv='Content-Type' content='text/html; charset=utf-8'/> <title>using virtual inheritance in exception types</title> <link href='reno.css' type='text/css' rel='stylesheet'/> </head> <body> <div class="body-0"> <div class="body-1"> <div class="body-2"> <div> <div id="boost_logo"> <a href="http://www.boost.org"><img style="border:0" src="../../../boost.png" alt="Boost" width="277" height="86"/></a> </div> <h1>Boost Exception</h1> </div> <!-- Copyright (c) 2006-2009 Emil Dotchevski and Reverge Studios, Inc. --> <!-- 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) --> <div class="RenoIncludeDIV"><div class="RenoAutoDIV"><h3>Using Virtual Inheritance in Exception Types</h3> </div> <p>Exception types should use virtual inheritance when deriving from other exception types. This insight is due to Andrew Koenig. Using virtual inheritance prevents ambiguity problems in the exception handler:</p> <pre>#include <iostream> 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& e) {} catch(...) { std::cout << "whoops!" << std::endl; } }</pre> <p>The program above outputs "whoops!" because the conversion to std::exception is ambiguous.</p> <p>The overhead introduced by virtual inheritance is always negligible in the context of exception handling. Note that virtual bases are initialized directly by the constructor of the most-derived-type (the type passed to the throw statement, in case of exceptions.) However, typically this detail is of no concern when boost::<span class="RenoLink"><a href="exception.html">exception</a></span> is used, because it enables exception types to be trivial structs with no members (there's nothing to initialize.) See <span class="RenoLink"><a href="exception_types_as_simple_semantic_tags.html">Exception Types as Simple Semantic Tags</a></span>.</p> </div><div class="RenoAutoDIV"><div class="RenoHR"><hr/></div> See also: <span class="RenoPageList"><a href="boost-exception.html">Boost Exception</a> | <a href="frequently_asked_questions.html">Frequently Asked Questions</a></span> </div> <!-- Copyright (c) 2006-2009 Emil Dotchevski and Reverge Studios, Inc. --> <!-- 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) --> <div id="footer"> <p> <a class="logo" href="http://jigsaw.w3.org/css-validator/check/referer"><img class="logo_pic" src="valid-css.png" alt="Valid CSS" height="31" width="88"/></a> <a class="logo" href="http://validator.w3.org/check?uri=referer"><img class="logo_pic" src="valid-xhtml.png" alt="Valid XHTML 1.0" height="31" width="88"/></a> <small>Copyright (c) 2006-2009 by Emil Dotchevski and Reverge Studios, Inc.<br/> Distributed under the <a href="http://www.boost.org/LICENSE_1_0.txt">Boost Software License, Version 1.0</a>.</small> </p> </div> </div> </div> </div> </body> </html>