There might be serious issues with this approach. Most of sympy tries
to have selfcontained objects. This is quite important because of
(among other things):

- having consistent hashes and __eq__ behavior
- actually rebuilding these objects on tree traversal

The `mechanics` module uses the approach that you suggested
successfully, however I do not think it was ever discussed whether it
is a good style within sympy as a whole. It certainly clashes with the
style of the rest of the codebase.

For the mechanics module itself this is not a problem, however if you
are going to extend it to other modules more thought should be given
to it.

A way to refactor this would be for the "world" object to be an
argument for the "entities within the world" objects. Like for
instance the relation between hilbert spaces and kets in quantum or
between manifold and field in diffgeom.

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" 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/sympy?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to