Re: [sigc] libsigc++ 1.9.13 fails on Solaris 9 sparc

[Martin Schulze <martin-ml hippogriff de>]

Hi Damien!

I'm afraid we are gradually loosing track of what is working and what  
is causing problems. Please check whether the attached test compiles  
for you. Based on the result I'll do another release which we can use  
for further discussion.

Thanks for your effort!

 Martin


Am 2004.03.16 18:21 schrieb(en) Damien Carbery:
> Martin Schulze wrote:
> > Am 2004.03.15 20:03 schrieb(en) Murray Cumming:
> >
> > > On Mon, 2004-03-15 at 12:23, Damien Carbery wrote:
> > > > Murray Cumming wrote:
> > > > > On Sat, 2004-03-06 at 00:10, Martin Schulze wrote:
> > > > >
> > > > >>Am 2004.03.04 17:22 schrieb(en) Damien Carbery:
> > > > >>
> > > > >>>Apologies for not doing this build earlier in the week. Busy.
> > > > >>>I updated tests/test_mem_fun.cc and tests/test_ptr_fun.cc and
> > > rebuilt
> > > > >>>(make clean; make).
> > > > >>>Log is attached.
> > > > >>
> > > > >>Please try again with the version of file sigc++/adaptors/ 
> > > macros/
> > > > >>adaptor_trait.h.m4 that is attached to this email. The problem  
> > > is
> > > > >>the syntax for explicit method template specializations. gcc
> > > expects:
> > > > >
> > > > >
> > > > > Martin, is all this stuff in cvs so far? I now have access to a
> > > SUN
> > > > > Forte compiler, though it might not be the same as Damien's:
> > > > >
> > > > > bash-2.03$ CC -V
> > > > > CC: Sun WorkShop 6 update 2 C++ 5.3 Patch 111685-11 2002/10/31
> > > > >
> > > > > Damien, what does CC -V give you?
> > > > >
> > > >
> > > > CC: Sun C++ 5.5 Patch 113817-03 2003/10/14
> > > > It is from the Sun ONE Studio 8 Compiler Collection.
> > > > http://wwws.sun.com/software/products/studio/index.html
> > >
> > > Hmm, maybe 5.5 is much more capable than 5.3 - but I'd be happy if  
> > > at
> > > least one of them can build libsigc++ 2. I'm not sure at what point
> > > you
> > > are, and exactly what sources you are both investigating - could
> > > someone
> > > explain?
> >
> >
> > First, sorry for my absence. I was at a conference in Kiel last  
> > week.
> >
> > Great that we have access to two SUN boxes now :)
> >
> > I'm attaching the "corrected" test cases and adaptor_trait.h.m4 I   
> > already sent to Demian.
> > I didn't want to check in this stuff until it is clear that the  
> > changes  really help - they are quite experimental and Demian  
> > couldn't work with  CVS anyway until now because of his outdated  
> > autoconf version (has this  changed, by the way?).
> >
> > Demain has been working with a patched libsigc++-1.9.14. All patches
> > but the most recent ones affecting the test cases and adaptor_trait. 
> > h. m4 (-> attachement) are checked in cvs.
> >
> > Regards,
> >
> >  Martin
>
> I copied the 3 files into the build, did 'make clean' and then  
> 'make'.
> Error output attached.
#include <iostream>

class Thing
{
public:
  Thing()
    {}

  template <class T>
  void operator()(T a, T b)
  {
    T c = a + b;
    std::cout << c << std::endl;
  }
};

int main(int, char*[])
{
  Thing thing_;
  thing_./*template */operator()<int>(1, 2);
}