Re: Problems Building ppl-0.10.2 with g++-4.4.0 on Open Solaris 2008.11

[Marc Glisse <marc dot glisse at normalesup dot org>]

On Sat, 25 Apr 2009, Tom Browder wrote:

> /bin/sh ../libtool --tag=CXX   --mode=compile g++-4.4.0t
> -DHAVE_CONFIG_H -I. -I..  -I.. -I../src  -fexceptions
> -I/usr/local/include  -g -O2 -frounding-math  -W -Wall -MT
> C_Polyhedron.lo -MD -MP -MF .deps/C_Polyhedron.Tpo -c -o
> C_Polyhedron.lo C_Polyhedron.cc
> libtool: compile:  g++-4.4.0t -DHAVE_CONFIG_H -I. -I.. -I.. -I../src
> -fexceptions -I/usr/local/include -g -O2 -frounding-math -W -Wall -MT
> C_Polyhedron.lo -MD -MP -MF .deps/C_Polyhedron.Tpo -c C_Polyhedron.cc
> -fPIC -DPIC -o .libs/C_Polyhedron.o
> In file included from algorithms.hh:27,
>                 from C_Polyhedron.cc:28:
> Pointset_Powerset.defs.hh:68: error: expected unqualified-id before
> numeric constant

This is a classic problem with solaris. Some of its standard headers 
(sys/regset.h for instance) have a tendancy to #define every short name on 
earth. Since the name "CS" is private, it probably doesn't break anything 
to replace it with something else in this code.

I complained about it some time ago:
http://www.opensolaris.org/jive/thread.jspa?threadID=67402&tstart=45

but as you can read my comment wasn't really taken into consideration...

--
Marc Glisse