"Paul Mensonides" <[EMAIL PROTECTED]> writes:

> ----- Original Message -----
> From: "David Abrahams" <[EMAIL PROTECTED]>
>
>> > ----- Original Message -----
>> > From: "Russell Yanofsky" <[EMAIL PROTECTED]>
>> >
>> >
>> >> A preprocessor bug in codewarrior 8.3 causes various parts of the mpl
>> >> library to fail to compile. There was already a workaround in place,
> but
>> >> it was disabled for the latest compiler version. A patch which fixes
> the
>> >> problem is below.
>>
>>
>> Hum. I've been testing with CWPro8.3 for a LONG time and have never
>> noticed a problem.  I think I use the PP library extensively.
>>
>> Are we sure this patch is needed?
>
> I think that Aleksey is using his own custom stuff for certain things for
> optimization purposes.  It isn't the pp-lib that has the problem, it is
> parts of the MPL.

Hum, hum.
I *am* seeing one error from the MPL regressions with cwpro8.3:

metrowerks-C++-action 
c:\build\libs\mpl\test\bin\as_sequence.test\cwpro8.3\debug\runtime-link-dynamic\as_sequence.obj
### mwcc Compiler:
#      In: ..\..\..\boost\mpl\is_sequence.hpp
#    From: as_sequence.cpp
# ------------------------
#      35:      : logical_not< is_same< typename begin<T>::type, non_sequence_tag > >
#   Error:                                                                          ^
#   illegal struct/union/enum/class definition
#    (point of instantiation: 'main()')
#     (instantiating: 'boost::mpl::as_sequence<UDT>')
#      (instantiating: 'boost::mpl::if_<boost::mpl::is_sequence<UDT>, UDT, 
boost::mpl::single_view<UDT>>')
#       (instantiating: 'boost::mpl::is_sequence<UDT>')
### mwcc Compiler:
#      In: C:\boost\boost\mpl\if.hpp
# ----------------------------------
#      60:          >::type type;
#   Error:                      ^
#   'value' is not a member of class 'boost::mpl::is_sequence<UDT>'
#    (point of instantiation: 'main()')
#     (instantiating: 'boost::mpl::as_sequence<UDT>')
#      (instantiating: 'boost::mpl::if_<boost::mpl::is_sequence<UDT>, UDT, 
boost::mpl::single_view<UDT>>')


However, it doesn't look like it's preprocessor-related.

-- 
                       David Abrahams
   [EMAIL PROTECTED] * http://www.boost-consulting.com
Boost support, enhancements, training, and commercial distribution

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Reply via email to