On 12 May 2012 02:22, Matthew Rocklin <[email protected]> wrote:
> You're right in that they both redundantly serve as immutable storage
> classes. I can envision a system where they're merged.
>
> On the other hand one can think about them quite differently. Tuples contain
> Basics, IMs contain Exprs (not sure this is actually enforced.) Most
> importantly IMs have all the matrix computation stuff attached to them,
> Tuples are simpler. Were you thinking of subclassing IM from Tuple,
> subclassing Tuple from IM? Creating NDArray?
>
> What is the situation? Why does Tuple/IM require a special case?
>

I do not see a meaningful difference between a column vector and a
tuple, If someone needs to input coordinates in a function at the
moment he can use list, tuple, Tuple and Matrix. That does not seem
right.

More generally I do not see why iterables are not sympified into
column matrices. This will make so many special cases in the code go
away.

And I am not speaking about some kind of Arrays. Matrices are arrays
and that seems good enough to me. For something more fancy one can
always use numpy arrays with dtype=object.

I will try to remove Tuple from the code base.

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sympy?hl=en.

Reply via email to