- From: Richard Cyganiak <richard@cyganiak.de>
- Date: Tue, 8 Jun 2010 09:56:21 +0100
- To: Eric Prud'hommeaux <eric@w3.org>
- Cc: RDB2RDF WG <public-rdb2rdf-wg@w3.org>
Hi Eric,
That's a good start.
On 7 Jun 2010, at 16:49, Eric Prud'hommeaux wrote:
> In order to get some common terminology, I've created a draft of a
> Direct Mapping.
Maybe start by defining what a database is in your notation? "A
database db is a set of relations R_1...R_i" etc?
> This defines a Direct Graph and demonstrates how this
> definition can be extended.
>
> The crux of it is still:
>
> directDB(db) ≝ { directR(r) ∀ r ∈ db }
> directR(R) ≝ { directT(R, T) ∀ T ∈ R.Body }
> directT(R, T) ≝ { directL(R, S, A) ∀ A ∈ scalar(T) }
> ∪ { directN(R, S, A) ∀ A ∈ reference(T) }
What are scalar(T) and reference(T)?
> ∣ S = nodemap(R, pk(T))
> directL(R, S, A) ≝ triple(S, predicatemap(R, A), literalmap(A))
What's literalmap?
> directN(R, S, A) ≝ triple(S, predicatemap(R, A), nodemap(R, A))
>
> nodemap(R, A) ≝ IRI(stem + "/" + R.name "/" A.name + "." +
> A.value + "#_")
> predicatemap(R, A) ≝ IRI(stem + "/" + R.name "#" A.name)
What's the logic for uses hashes in some places and slashes in others?
Shouldn't there be an rdf:type triple somewhere? If not, then how do
you SPARQL for all records in a single DB?
To be really useful, this direct mapping should define a URI for the
DB itself, for each relation, and for each attribute in the relations.
> I'm still playing with the notation.
I'd prefer classical mathematical sets, so where you have:
directDB(db) ≝ { directR(r) ∀ r ∈ db }
I'd rather see:
directDB(db) = { directR(r) | r ∈ db }
General note: "succinct" and "clear" are correlated, but not the same.
The latter should be the goal, not the former.
Best,
Richard
> It's currently a pretty classic
> notation:
>
> 3.1 Notation for Types
> A : a type
> A ⊔ B : disjoint union of A and B
> ( A, B ) : tuple (Cartesian product) of types A and B
> [ A ] : list of elements of type A
> { A } : set of elements of type A
> { A→B } : map of elements of type A to elements of type B
> 3.2 Notation for Injectors
> a : an instance of an A
> ( a1, b1 ) : a tuple with elements a1 and b1
> [ a1, a2 ] : list with elements a1 and a2
> { a1, a2 } : set with elements a1 and a2
> { a1→b1, a2→b2 } : map with elements with key a1 mapped to b1
> and key a2 mapped to b2
> 3.3 Supporting Functions
> AB[a] : in a map of A to B, the instance of B for a given A*
>
> We can get more type-safety if we use something like a scala notation,
> but I'm not sure how to tersely express things like disjoint union.
>
> 3.1 Notation for Types
> x:X : x is an element in the set X
> A ?? B : disjoint union of A and B (normally case classes
> extending an abstract class, e.g.:
> abstract class AB; A extends AB; B extends AB;
> )
> ( A, B ) : tuple (Cartesian product) of types A and B
> List[ A ] : list of elements of type A
> Set[ A ] : set of elements of type A
> Map[ A, B ] : map of elements of type A to elements of type B
> 3.2 Notation for Injectors
> a : an instance of an A
> ( a1, b1 ) : a tuple with elements a1 and b1
> List( a1, a2 ) : list with elements a1 and a2
> Set( a1, a2 ) : set with elements a1 and a2
> Map( a1→b1, a2→b2 ) : map with elements with key a1 mapped to
> b1 and key a2 mapped to b2
> 3.3 Supporting Functions
> AB(a) : in a map of A to B, the instance of B for a given A*
>
> --
> -ericP
>
Received on Tuesday, 8 June 2010 09:18:07 UTC