[Lemon-devel] I can't believe
Alpár Jüttner
alpar at cs.elte.hu
Mon Apr 21 15:15:01 CEST 2008
> #define NOMINMAX
> #include <windows.h>
>
> does this particular problem,
I mean: _solves_ this particular problem...
Regards,
Alpar
> but I see no good solution to the other
> problem of my previos mail (i.e. in time_measure.h, pushing an
> std:string to an ostream does not work)
>
>
> Regards,
> Alpar
>
>
> >
> > Best, Balazs
> >
> > On Friday 18 April 2008 17.43.23 Alpár Jüttner wrote:
> > > Hi,
> > >
> > > MSVC compiler complains about line
> > >
> > > max_w=std::max(double(_arcWidths[e]),max_w);
> > >
> > > in graph_to_eps.h saying that
> > >
> > > C:\Work\lemon\cmake\lemon/graph_to_eps.h(748) : error C2589: '(' :
> > > illegal token on right side of '::'
> > >
> > > You won't believe the reason: There seems to exist a
> > >
> > > #define max(a,b) (((a)>(b))?(a):(b))
> > >
> > > line somewhere in the windows.h header file.
> > >
> > > How to handle this situation? We use std::max in several places. In fact
> > > it typically do not cause any problem because we don't use windows.h.
> > > Even that special case can be solved if we move the windows.h dependent
> > > stuff into a separate file including windows.h but not using std::max().
> > > However this is still very dangerous, as a user of lemon can also
> > > include windows.h.
> > >
> > >
> > > Another 'nice' feature of MSVC:
> > > In time_measure.h, will not compile the line
> > >
> > > _os << _title;
> > >
> > > but it compiles this:
> > >
> > > _os << _title.c_str();
> > >
> > > Yes, you see it right, MSVC does not allow you to push an std::string to
> > > an std::ostream, but only a char *.
> > >
> > > Regards,
> > > Alpar
> > >
> > >
> > > _______________________________________________
> > > Lemon-devel mailing list
> > > Lemon-devel at lemon.cs.elte.hu
> > > http://lemon.cs.elte.hu/mailman/listinfo/lemon-devel
> >
> >
>
> _______________________________________________
> Lemon-devel mailing list
> Lemon-devel at lemon.cs.elte.hu
> http://lemon.cs.elte.hu/mailman/listinfo/lemon-devel
More information about the Lemon-devel
mailing list