Re: recursive shapes in SHACL

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

If recursive shapes are forbidden, then it is possible to do away with
query-execution-time treatment of shape inclusions in favour of a
macro-expansion treatment of shape inclusion.

peter


On 03/24/2015 08:26 PM, Holger Knublauch wrote:
> On 3/25/2015 12:52, Peter F. Patel-Schneider wrote:
>> One way forward is to forbid recursive shapes.
> 
> Do you suggest to remove sh:valueShape completely, or only mark it
> recursion illegal in the specification? In the latter case, we would
> still need to give advice to implementers on what to do if it encounters
> an illegal situation (to prevent infinite loops etc).
> 
> Holger
> 
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJVEoeQAAoJECjN6+QThfjz71QH/0WKzBeFMJ4DtN5dI3/c7x49
1zcPt1JpaLhR+6yr0TbF2lhG+JNFL9qRSJGW+4pfB0cpgObG/0X3wP1dFXzbwVh8
6vhbgjfLOSaJ+V+hYVa5e6ODkSXBNmH4neXxA0iXYY6KbFQT2kswk9WCbCGrmogc
nOH9KRbJaGebc88qrgRhzGfp2hxUnuXptmczzB9E9U9NWfEzCzf6atS/RVp3QKv+
yufEt1BcqzfZK8N6Xfovrsejt48QU89f5PYPCkf/4kcmd09WgmKSkn168Dpe2Exk
pNLleIseE3US45R8uTuFqLlSgcTCn0HEFncDScF8E36Hmr0u1eB0v3vx7PZ1I/U=
=zoAx
-----END PGP SIGNATURE-----

Received on Wednesday, 25 March 2015 10:02:22 UTC