Re: [sigc] Unwanted pass-by-value behavior with MSVC++ 7.1

[Martin Schulze <martin-ml hippogriff de>]

Am 23.07.2004 15:25:05 schrieb(en) Martin Schulze:
> Am 11.07.2004 22:27:49 schrieb(en) Martin Schulze:
> > Am 08.07.2004 00:00:03 schrieb(en) James Lin:
> > > Murray Cumming wrote:
> > > >
> > > > > Basically sigc::bind somehow was causing arguments to be
> > > > > passed by value even when explicitly told to pass by
> > > > > reference. (I verified this by passing an object by
> > > > > reference that had a private copy-constructor and
> > > > > getting a compile-time error.  Also, test_lambda was
> > > > > producing incorrect results [....]
> > > >
> > > > A test case for this would be really useful. We could add
> > > > it to the tests directory. In bugzilla please.
> > >
> > > Okay, will do sometime this week.
> > >
> > > To elaborate on the test_lambda case, before I disabled the
> > > sun_forte_workaround calls with MSVC++, lines 69-71 in test_lambda. 
> > > cc
> > > produced (with optimizations enabled):
> > >
> > >   ((++_1)*2)(ref(a)): 4; a: 1
> > >   ((++(*_1))*2)(&a):  4; a: 2
> > >   ((--(*(&_1)))*2)(ref(a)): 2; a: 2
> > >
> > > (AFAICT, the correct values should be (4, 2), (6, 3), and (4, 2)
> > > respectively.)
> > >
> > > > > and that SIGC_WORKAROUND_OPERATOR_PARENTHESES and
> > > > > LIBSIGC_TEMPLATE_PREFIX really do need their own,
> > > > > independent tests.
> > > >
> > > > Yes, and I guess you are the best person to implement them.
> > > > It's really quite simple, because there is already an example.
> > > > You need to
> > > >
> > > > 1. Add an .m4 macro in scripts/cxx.m4. Copy the existing one
> > > > and change
> > > > [...]
> > >
> > > Okay, but I'm not sure how useful that would be to Windows MSVC++
> > > users, the
> > > people who'd encounter the problem.  Are MSVC++ users expected to  
> > > run
> > > the
> > > configure script?
> > >
> > > > Please tell me if you have any problems. Unfortunately, I
> > > > don't think a check for WIN32 will be good enough, because there
> > > > are now several versions of the MSVC++ compiler in common use,
> > > > each with different capabilities.
> > >
> > > That's true.  On the plus side, Microsoft has made the
> > > VC++7.1/VC++2003
> > > compiler available for free, and it can be installed alongside an
> > > existing
> > > Visual Studio installation.  Plus, the Visual Studio project files
> > > included
> > > with libsigc++ 2.0.3 already require VS.net 2003 anyway.
> >
> > I've added the additional configure check. However through the lines  
> > I can see that this doesn't help you because you don't execute the  
> > configure script with MSVC, do you? Can we define some macros in the  
> > project file instead? We need to #define  
> > SIGC_CXX_MSVC_TEMPLATE_SPECIALIZATION_OPERATOR_OVERLOAD. Could you  
> > please modify the project file accordingly and attach a patch  
> > against cvs to http://bugzilla.gnome.org/show_bug.cgi?id=147311 ?
>
> James, could you please comment on this? I'd like to make that bugfix  
> release by Friday next week at the latest.

Whoops, my mistake - I was barking nonsense. In fact, sigc++config.h  
needs to be configured according to the platform (-> for the MSVC, only  
'SIGC_MSVC_TEMPLATE_SPECIALIZATION_OPERATOR_OVERLOAD' should be  
'#define'd there). Normally, this is done by the configure script. What  
options do we have with MSVC?

Regards,

 Martin