On Monday, July 22, 2013 6:58:11 AM UTC-7, Volker Braun wrote:
>
> On Sunday, July 21, 2013 8:35:24 PM UTC-4, davidp wrote:
> > At the moment, I just have Hyperplane as a class deriving from 
> AffineSpace (my version of AffineSpace, that is, which I will change, as
> > noted above), and HyperplaneArrangement deriving from object.  Can you 
> spell out your idea a bit more?
>
> All the data that define the ambient space, the base ring, etc goes into 
> the Parent. The element refers to the parent for constant data. Or, by 
> analogy:
>
> Element : Parent
> Polynomial : PolynomialRing
> Hyperplane : HyperplaneArrangement
>
> There isn't really a simple introduction, but 
> http://www.sagemath.org/doc/thematic_tutorials/coercion_and_categories.htmlspells
>  out all the details.
>
> I don't really see a need for another affine space. It would be very 
> confusing to present the user with multiple affine space implementations. 
>

I need a class to model an affine subspace of K^n where K is a finite field 
or QQ.  An affine subspace has the form p + W where p is a point and W is 
a linear subspace of K^n.  Besides testing for equality and containment, 
one main method I need is intersection: the output of intersect(self, 
other) is another affine space.  Another useful method provides an 
isomorphism of an affine space of dimension d in K^n with K^d, and in the 
case K = QQ, one that is approximately orthogonal.  Thus, AffineSpace does 
not really help.  So maybe the problem is finding a better name for what I 
have in mind. Perhaps "AffineSubspace"?  Or can you suggest a better idea?

Thanks for your help!
 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to