Closed GoogleCodeExporter closed 9 years ago
Original comment by bparsia
on 18 Apr 2007 at 1:03
Original comment by bparsia
on 18 Apr 2007 at 1:10
All class constructors that accept a set of classes must contain at least two
classes, and the DisjointUnion follows that pattern.
Please note that it makes no sense to define a class A as a disjoint union of a
class B: you can simply then define A as being equal to B.
Original comment by borismo...@gmail.com
on 23 May 2007 at 5:57
Hmm, I'm a little bit puzzled.
As far as I understand the structure of the DisjointUnion construct, the
mentioned restriction is reflected as a
minOccurs="2" within the UnionOf element (line 448). Why do we have to express
DisjointUnion(A C1 C2 C3)
as
<DisjointUnion>
<OWLClass "A"/>
<OWLClass "C1"/>
<UnionOf>
<OWLClass "C2"/>
<OWLClass "C3"/>
</UnionOf>
</DisjointUnion>
Furthermore, with the current schema it is not possible to define a class as
the disjoint union of two
other classes (i.e. DisjointUnion(A C1 C2)).
Is there something I miss?
Original comment by thorsten...@uni-ulm.de
on 29 May 2007 at 10:00
OK, now I understood what your original comment refers to.
The original idea was to rerpesent DisjointUnion(A C1 C2 C3) in XML as
<DisjointUnion>
<OWLClass "A"/>
<UnionOf>
<OWLClass "C1"/>
<OWLClass "C2"/>
<OWLClass "C3"/>
</UnionOf>
</DisjointUnion>
Now I agree that the XML schema was confusing: it allowed one to put several
descriptions before the <UnionOf> tag.
In fact, after looking at it, there probably is no reason to have the UnionOf
tag at
all. For example, in the textual syntax, there is no such tag. Hence, in the
new
schema version I've removed it. (The schema should be available online within
24
hours - there is a cron job that periodically updates the Web site with the
contents
of SVN.)
Now, the above example is serialized simply as
<DisjointUnion>
<OWLClass "A"/>
<OWLClass "C1"/>
<OWLClass "C2"/>
<OWLClass "C3"/>
</DisjointUnion>
Original comment by borismo...@gmail.com
on 31 May 2007 at 9:49
Original issue reported on code.google.com by
thorsten...@uni-ulm.de
on 18 Apr 2007 at 12:40