Re: The C++ article in April issue of DDJ
On Mar 22, 11:36 am, "Andrei Iltchenko" <andrei.ilche...@gmail.com>
wrote:
On Mar 22, 4:16 am, Sebastian Redl <e0226...@stud3.tuwien.ac.at>
wrote:
On Mon, 19 Mar 2007, Andrei Iltchenko wrote:
3. The language is clear that an implicitly-declared destructor "shall
allow no exceptions if every function it directly invokes allows no
exceptions" (See 15.4/11-13)
std::basic_string::~basic_string has no exception specifier according to
21.3/4 of the 1997 public review document (don't have access to a final,
sorry). It allows all exceptions, therefore according to 15.4/13 an
implicitly created destructor for the exception shall allow all
exceptions, too.
This is not true and is a misconception. All C++ Standard Library
classes must have destructors that don't throw. The Standard Library
(especially its STL part) has been designed based on that assumption.
This is carefully reflected in the wording of Section 17.4.4.8/3
"Restrictions on exceptions handling". I'll quote:
Whether the destructor (or anything else) actually throws is
irrelevant here. All that counts is the exception
specification.
No destructor operation defined in the C++ Standard Library will throw
an exception. Any other functions defined in the C++ Standard Library
that do not have an exception-specification may throw implementation-
defined exceptions unless otherwise specified. An implementation may
strengthen this implicit exception-specification by adding an explicit
one.
The last statement means that in practice you always need a user
defined destructor when deriving from a class in the standard
library. Since a legal implementation can always do something
like:
namespace std
{
template< typename charT, typename Traits >
class basic_streambuf
{
// ...
~basic_streambuf() throw( BetYouHaventDeclaredThisOne )
{
// ...
}
} ;
}
Of course, most of the time, it will be OK, because
implementations don't do something this stupid in practice.
--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orient?e objet/
Beratung in objektorientierter Datenverarbeitung
9 place S?mard, 78210 St.-Cyr-l'?cole, France, +33 (0)1 30 23 00 34
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]