Boost logo

Boost :

Subject: Re: [boost] [predef] missing file On MacOS 10.7.5.
From: ÷ÌÁÄÉÓÌÁ× ýÁÐÏ× (phprus_at_[hidden])
Date: 2013-10-06 02:56:31


Not work, because file <boost/predef/os/bsd.h> included before file
<boost/predef/os/macos.h> in <boost/predef/os.h>.

Please, check test code with next #include list:
#include <sys/param.h>
#include <boost/predef.h>

2013/10/6 Rene Rivera <grafikrobot_at_[hidden]>

> Of course.. Thanks for finding that. I've fixed both problems now. Now only
> one OS is allowed to be detected at once. And the endian definition uses
> the OSX specific endian header. Test added for both cases on OSX.
>
>
> On Sat, Oct 5, 2013 at 8:14 AM, ÷ÌÁÄÉÓÌÁ× ýÁÐÏ× <phprus_at_[hidden]> wrote:
>
> > Hello!
> >
> > On OSX "BSD" has defined in file <sys/param.h>
> >
> > If this file has been used in source code, then Boost.Predef make wrong
> > detect OS (OS X as BSD).
> >
> >
> > 2013/9/30 Rene Rivera <grafikrobot_at_[hidden]>
> >
> > > On Fri, Sep 27, 2013 at 1:30 AM, Tim Blechmann <tim_at_[hidden]> wrote:
> > >
> > > > > Something interesting about your output that jumps out at me is
> that
> > > it's
> > > > > considerably more than the 166 lines of #defines I get on my Xcode
> > > setup.
> > > > > And from looking at the defs it seems that your setup is including
> a
> > > > bunch
> > > > > of system headers. I.e. as if the empty.cpp file was not actually
> > > empty.
> > > > Is
> > > > > there something I'm missing in your setup?
> > > >
> > > > aj sorry ... seems that i appended an empty string to an existing
> file,
> > > > my bad ...
> > > >
> > >
> > > I still see nothing that would cause OS_BSD to get defined :-( And
> after
> > > some upgrades and more testing I can't repo this.. Nor do any of the
> OSX
> > > testers in the Boost results reproduce the problem.
> > >
> > > iac, this is what apple's clang gives me on a truely empty file. btw,
> it
> > > > also defines __BYTE_ORDER__ without including any header.
> > > >
> > >
> > > And under usual circumstances the byte order macros are used in the
> > > OSX/clang/gcc case.
> > >
> > > I've tried building all the libraries that directly depend on the
> endian
> > > header and they all build/work fine for me. Hence I'm still looking for
> > > suggestions/help on what could be going on in your case.
> > >
> > >
> > > --
> > > --
> > > -- Grafik - Don't Assume Anything
> > > -- Redshift Software, Inc. - http://redshift-software.com
> > > -- rrivera/acm.org - grafik/redshift-software.com
> > > -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
> > >
> > > _______________________________________________
> > > Unsubscribe & other changes:
> > > http://lists.boost.org/mailman/listinfo.cgi/boost
> > >
> >
> >
> >
> > --
> > ó Õ×ÁÖÅÎÉÅÍ, ÷ÌÁÄÉÓÌÁ× ýÁÐÏ×
> > ICQ: 223-695-654
> >
> > _______________________________________________
> > Unsubscribe & other changes:
> > http://lists.boost.org/mailman/listinfo.cgi/boost
> >
>
>
>
> --
> --
> -- Grafik - Don't Assume Anything
> -- Redshift Software, Inc. - http://redshift-software.com
> -- rrivera/acm.org - grafik/redshift-software.com
> -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
>
> _______________________________________________
> Unsubscribe & other changes:
> http://lists.boost.org/mailman/listinfo.cgi/boost
>

-- 
ó Õ×ÁÖÅÎÉÅÍ, ÷ÌÁÄÉÓÌÁ× ýÁÐÏ×
ICQ: 223-695-654

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk