Em Terça-feira 24 Novembro 2009, Dâniel Fraga escreveu:
> On Tue, 24 Nov 2009 14:07:38 -0200
>
> Lamarque Vieira Souza <lamar...@gmail.com> wrote:
> > I do not think that is the problem, if LIBMSN_INBOX_URL_ENABLED is not
> > defined when including wlmlibmsn.h the error message would be different.
> > The linker thinks inboxUrl is a class method, not a signal, that is the
> > problem. Dâniel, could you try recompiling again now just removing the
> > #ifdef LIBMSN_INBOX_URL_ENABLED and the #endif but keeping inboxUrl at
> > the same place (after newMailNotification signal). If it does not work,
> > try adding a fake
>
> Ok, I tried that (removing the #ifdef) and it worked.
>
> > signal after inboxUrl to see what happens. I still think it is a
> > compiler/liker bug, we need more test to figure out what is really
> > happening.
>
> So the whole problem is with the #ifdef? I mean, the "parser"
> is not recognizing the ifdef and generates the error message?
This is weird, if the #ifdef is resolving to false the error message
should
be different (like ‘inboxUrl’ was not declared in this scope). You can confirm
if the #ifdef is resolving to true by adding this line inside the #ifdef:
#warning Inside ifdef
and search for a line like this in make's output:
protocols/wlm/wlmlibmsn.h:463:2: warning: #warning Inside ifdef
> Would it worth to write a small sample code to test those ifdef
> statements? If so, and if it's really a compiler bug we could send to
> gcc developers...
#ifdef is a pre-processor statment, not compiler's, we need to know if
the
problem is in the pre-processor (cpp command) or in the compiler (g++
command).
--
Lamarque V. Souza
http://www.geographicguide.com/brazil.htm
Linux User #57137 - http://counter.li.org/
_______________________________________________
kopete-devel mailing list
kopete-devel@kde.org
https://mail.kde.org/mailman/listinfo/kopete-devel
