Hello !

> Because after thinking this it kind of feels wrong. A class should not
> "know" other classes deriving from it.

Yes yes. Well, "the correct way" would be to implement .relabel() in
every superclass. But like for "disjoint union" that is a lot of code
(+doc) for almost nothing, so listing the known superclasses in the
function as you did is a good way out I believe.

> And in theory someone could also
> extend class hierarchy, for example to have distributed lattices to be own
> class (with, for example, function giving the unique complement of an
> element). After that one should also modify posets.py.

Yep. Well, either implement it 4 times or centralize it all in
posets.py, as you prefer.

> It is #17142 and you already are at cc field.

Oh okay. The only tickets I have in mind are those on needs_review
that I should work on. I will review it when it will be implemented
:-)

Nathann

-- 
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/d/optout.

Reply via email to