bug or problem with gcc and include files

Robert Marcano robert at marcanoonline.com
Thu Mar 25 15:48:58 UTC 2004


On Thu, 2004-03-25 at 11:38, Nils Philippsen wrote:
> On Thu, 2004-03-25 at 16:30, Robert Marcano wrote:
> > On Thu, 2004-03-25 at 11:14, Steve Wampler wrote:
> > > On Thu, 2004-03-25 at 06:59, jludwig wrote:
> > > > If include is a preprocessor directive how can it be in the body of a
> > > > function?
> > > 
> > > Uh, how can it not be ok there?  The preprocessor doesn't know
> > > anything about function bodies - it's just processing text, not
> > > C code.
> > > 
> > > In fact, you can do some amazing things with the preprocessor,
> > > most of which shouldn't be done.
> > 
> > A good example of this is located at
> > http://lxr.mozilla.org/security/source/security/nss/lib/softoken/pkcs11.c#110
> > 
> > this #include allows this code to define a function pointers structure
> > with the same function definitions previously included
> 
> Nevertheless, usually headers need to be "aware" of being included at
> random places. Most system headers aren't.

Most? I think that no respectable system headers will use those tricks.
They must be used for internal code, not APIs

> 
> Nils
> -- 
>      Nils Philippsen    /    Red Hat    /    nphilipp at redhat.com
> "They that can give up essential liberty to obtain a little temporary
>  safety deserve neither liberty nor safety."     -- B. Franklin, 1759
>  PGP fingerprint:  C4A8 9474 5C4C ADE3 2B8F  656D 47D8 9B65 6951 3011
-- 
Robert Marcano





More information about the users mailing list