David Faure wrote:
> 
> On Sun, Dec 19, 1999 at 09:22:49PM +0100, Dirk A. Mueller wrote:
> > On Son, 19 Dez 1999, Mirko Sucker wrote:
> >
> > > - implement the kdebug-based message handler for Nana logging output and
> >
> > Well, this sounds like inverse logik to me. From what I've read up to now,
> > the Nana logs seems to contain the source file name and line from where the
> > logging was called
> Even when compiled without debug info ?
Yes, it uses preprocessor statements. 
Nana lacks one main feature of kdebug, the debug areas. So there are two
ways of integrating:
° rewriting a part of nana or wrapping it to support areas - lots of
work
° redirecting nana messages in the kdebug buffers - nearly no work
BTW, adding file and line number to kdebug statements is a matter of
seconds, see a file of KDE 1.1.2, AssertDialog.cpp (I think, search for
evaluate_assertion),
where I did it. Should be a good idea, anyway.
> > It seems to be more reasonable to do the opposite - make kdebug a subset of
> > nana.
...
> I think we should forget about making any relation between kdebug and nana.
> They are two completely different things.
> kdebug is for verbose debug output, just checking where I am and
> what the values of some variables are.
> nana is for assertions, and any other kind of checks.
Nana remains independant from kdebug. Its message handler is defineable,
that is what I will do. We need integration of it in KDE anyway.
It is a matter of taste, in the end. We may simply forget about the
messages from nana, too. 
Greetings,
--Mirko.
-- 
Denn der  Mensch  liebt und ehrt den  Menschen,  solange er ihn
nicht zu beurteilen vermag, und die Sehnsucht ist ein Erzeugnis
mangelhafter Erkenntnis. (Thomas Mann)