--=-OuC5p8VVSjMv/lTcs0p0 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable > > > is this the only place to add it ? > >=20 > > it would seem more effective to use some variant on: > >=20 > > grep ^arch_compa /usr/lib/rpm/rpmrc > >=20 ok, so looking at this some more. A sample from that output: arch_compat: alphaev67: alphaev6 arch_compat: alphaev6: alphapca56 arch_compat: alphapca56: alphaev56 arch_compat: alphaev56: alphaev5 arch_compat: alphaev5: alpha arch_compat: alpha: axp noarch arch_compat: athlon: i686 arch_compat: i686: i586 arch_compat: i586: i486 arch_compat: i486: i386 arch_compat: i386: noarch Now is it reasonable to read that as: alphaev67 compat w/ alphaev6 alphaev6 compat w/ alphaca56 and are they therefore transitive? is alphaev67 compat with alphaev6? I would say yes based on what the athlon -> i386 say. And if that's true can I look through this list, grab things where the line ends in noarch and call those the "base classes" - doing a quick grep of the one on my system that gets me:=20 alpha i386 rs6000 sparc mips mipsel parisc armv4b armv3l atarist atariste ataritt falcon atariclone milan hades i370 s390 s390x ia64 x86_64 but then there is ppc and ppc64 which appear to not support noarch packages. Is that an endian issue? -sv --=-OuC5p8VVSjMv/lTcs0p0 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQA+Uq6x1Aj3x2mIbMcRAjJxAJ9N2zmfXIuy3I+l1L34wRIMctO5TQCfa/Lv aq3owUluPSJinLW0CsTGFKU= =RtNc -----END PGP SIGNATURE----- --=-OuC5p8VVSjMv/lTcs0p0--
