[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Porting KDE 2 (1.90) to other Unixes ...
From:       Stephan Kulow <coolo () kde ! org>
Date:       2000-05-11 8:03:15
[Download RAW message or body]

Thomas Leitner wrote:
> 
> On Thu, 11 May 2000, Michael Koch wrote:
> 
> > > However, "setenv" was only an example. I've stumbled accross many
> > > similar problems which could be solved easily without modifying the
> > > source files when I had kind of an "os.h" include file which would
> > > be included in each and every source file.
> > >
> > > Any ideas?
> >
> > I think we need a header file for kdelibs/kdecore/fakes.cpp. this can be
> > included in such cases then. But it should only be included if one plaftform
> > has problems with a file not in everyone.
> 
> Sounds like a good idea. So I would have to include "fakes.h" (or similar)
> only in the source files which I have troubles with.
> 
> BTW: I've just solved the particular problem with setenv. "setenv" *is*
> defined somewhere in the Tru64 libraries, so the "configure" program
> picks it up and defines HAVE_SETENV in config.h. However the prototype
> in the GCC stdlib.h is only declared if AES_SOURCE is defined. I'm
> now recompiling with AES_SOURCE (whatever this means) defined.
> 
Well, we should tweak the test then. It was once a test of it's own to
be
a C++ test, but we changed that, but can tweak it again.

Greetings, Stephan

-- 
... but you ain't had mine

[prev in list] [next in list] [prev in thread] [next in thread]