Re: [Boost-users] Bug report in date_time, 1.33.0, not fixed in CVS

From: Eric Teutsch (eteutsch_at_[hidden])
Date: 2006-11-20 15:58:21

Hmm.  I already had removed all the using namespaces in my code, but I dug
into the code again, and found a "using namespace" in another header file
which was called AFTER the boost header -- and when I removed that header
and expressly added System::Runtime::InteropServices:: to all the methods,
then it worked without modifying the header file.

Thanks for pointing out that there must be another using namespace
somewhere.  Don't you think that FILETIME should be preceded by :: though?
I.e. be explicit in which namespace it exists?

Eric T.
> -----Original Message-----
> From: boost-users-bounces_at_[hidden] [mailto:boost-users-
> bounces_at_[hidden]] On Behalf Of Martin Adrian
> Sent: Monday, November 20, 2006 15:33
> To: boost-users_at_[hidden]
> Subject: Re: [Boost-users]Bug report in date_time, 1.33.0, not fixed in
> CVS
> 
> Eric Teutsch <eteutsch <at> intldata.ca> writes:
> 
> > When using date_time in a managed C++ environment, the compiler
> > (VC7.1) can't differentiate between the windows global FILETIME, and the
> > .NET System::Runtime::InteropServices::FILETIME
> 
> I know that it is popular in .Net to have "using" declarations for
> everything
> (maybe because C# lacks namespace aliases) but why do you put them before
> the
> include files?
> 
> 
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users