Re: Bug: liblua use of undeclared type off_t causes build failure on OSX10.4PPC

[Patrick Donnelly <batrick () batbytes com>]

On Fri, Oct 31, 2014 at 8:28 PM, Geoff Down <geoffdown () fastmail net> wrote:
> Hi Patrick,
> On Fri, Oct 31, 2014, at 08:26 PM, Patrick Donnelly wrote:
> > Hi Geoff,
> >
> > On Thu, Oct 30, 2014 at 9:05 PM, Geoff Down <geoffdown () fastmail net>
> > wrote:
> > > Hello all,
> > >  I experienced this bug when trying to install Macports Nmap @6.47_0 .
> > > https://trac.macports.org/ticket/38822
> > > Seems off_t isn't native to the compiler apple-gcc42 (which is the
> > > latest available from Macports at least) and the legacy sys/types.h,
> > > where it is in fact defined, isn't included by liolib.c .
> > > Can you fix this so that other non C-programmers don't have to dive into
> > > the code please? I assume my kludge won't cause me any problems.
> > > Regards,
> > > Geoff Down
> >
> > That bug seems to be a duplicate of [1]. Also, this has been brought
> > up on Lua's mailing list here [2].
> >
> > Summary: this is Apple's fault since off_t is required to be defined
> > in stdio.h as per [3] (POSIX.1-2008).
>
>  Does it matter whose fault it is?

No it doesn't matter but...

> > Please forward the bug to Apple and/or have downstream macports (?)
> > maintain a patch.
>
> Apple don't maintain 10.4 at all. Macports pass the buck to you.

Are you saying it's fixed after 10.4? I don't have a Mac so I can't
check. I also don't understand why macports is passing this to us when
they are perfectly capable of patching Nmap. It's their distribution
that's broken. Lua (where the problem is) didn't fix it so I don't see
why we should.

> Meanwhile the poor users have broken software.
> Can't you decide between you who is going to fix it?

If this were a problem *in Nmap* I would be inclined to make the
change. However, since it's in Lua, I think it's a bad idea since it
would need to be maintained as long as we distribute Lua (i.e.
reapplied whenever Lua is updated).

-- 
Patrick Donnelly
_______________________________________________
Sent through the dev mailing list
http://nmap.org/mailman/listinfo/dev
Archived at http://seclists.org/nmap-dev/