Re: [Vxl-users] conflict between VXL and windows.h

[Wheeler, F. W (GE G. Research) <wh...@ge...>]

There are few other macros like VC_EXTRALEAN and WIN32_LEAN_AND_MEAN that when set limit what is #defined and declared by windows.h.  WIN32_LEAN_AND_MEAN is used in a few places in vcl.  It is possible that you are getting WIN32_LEAN_AND_MEAN defined when you include vil/vil_convert.h.  Might be easier to simply #define WIN32_LEAN_AND_MEAN before including windows.h.

-Fred

> -----Original Message-----
> From: Gehua Yang [mailto:yan...@gm...]
> Sent: Monday, January 03, 2011 8:28 PM
> To: Roddy Collins
> Cc: Vxl-Users
> Subject: Re: [Vxl-users] conflict between VXL and windows.h
> 
> I confirmed the trick does work.
> 
> Thank you for pointing it out, Roddy!
> 
> Gehua.
> 
> 
> On Mon, Jan 3, 2011 at 8:20 PM, Roddy Collins <rod...@ki...>
> wrote:
> > On Mon, Jan 3, 2011 at 8:13 PM, Gehua Yang <yan...@gm...>
> wrote:
> >>
> >> I got the same error / warning.  I ended up undefining min and max
> >> right after each include of windows.h, which works well for me:
> >>
> >
> > Just curious-- does setting NOMINMAX in the cmake compile flags work?
> >
> > e.g. http://www.quantlib.co.uk/blog/what-does-nominmax-do
> >
> > -Roddy
> >
> 
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
> to consolidate database storage, standardize their database environment,
> and,
> should the need arise, upgrade to a full multi-node Oracle RAC database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> Vxl-users mailing list
> Vxl...@li...
> https://lists.sourceforge.net/lists/listinfo/vxl-users