--08ATZu8fEq0x2T3M
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sat, Mar 27, 2004 at 02:25:49AM +0100, Klaus Heinz wrote:

> Hi,
>=20
> shouldn't the use of RPATH_FLAG be protected by a test for _USE_RPATH?
> mk/buildlink?/bsd.buildlink?.mk does this, although through a comparison
> against "yes" instead of !empty(_USE_RPATH:M[yY][eE][sS]).
>=20
> bsd.pkg.mk uses RPATH_FLAG without this protection:
>=20
>   .if defined(USE_X11)
>   X11_LDFLAGS+=3D           -Wl,${RPATH_FLAG}${X11BASE}/lib
>   ...
>   .if !empty(USE_BUILDLINK2:M[nN][oO]) && !empty(USE_BUILDLINK3:M[nN][oO])
>   LDFLAGS+=3D               -Wl,${RPATH_FLAG}${LOCALBASE}/lib=20
>   ...
>   PKGCONFIG_OVERRIDE_SED=3D \
>         -e 's|^\(Libs:.*[       ]\)-L\([        ]*[^ ]*\)\(.*\)$$|\1${_CO=
MPILER_LD_FLAG}${RPATH_FLAG}\2 -L\2\3|'
>   ...
>=20
> xaw.buildlink2.mk does the same:
>=20
>   LIBXAW?=3D                -L${X11BASE}/lib${ABI} -Wl,${RPATH_FLAG}${X11=
BASE}/lib${ABI} -lXaw
>=20
> The stumbled over this while trying to use pkgsrc on a platform where
> _USE_RPATH=3Dno and _OPSYS_RPATH_NAME/RPATH_FLAG is empty.

which one? I thought the only platform which had _USE_RPATH=3Dno was
Darwin, and it's a special case..

grant.


--08ATZu8fEq0x2T3M
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (NetBSD)

iD8DBQFAZNwiluYOb9yiFXoRAjNGAJ9FGXecZYaASMyRsUB/yr5H9p7TiwCeKrlY
7dkIxol71RVIVuGYuy5BtFU=
=434Y
-----END PGP SIGNATURE-----

--08ATZu8fEq0x2T3M--