Le jeudi 28 avril 2011 à 19:17 -0700, Ondrej Certik a écrit :
> On Thu, Apr 28, 2011 at 6:59 PM, Ronan Lamy <[email protected]> wrote:
> > Le jeudi 28 avril 2011 à 17:43 -0700, Mateusz Paprocki a écrit :
> >> Hi,
> >>
> >> On 28 April 2011 17:06, Ondrej Certik <[email protected]> wrote:
> >>         Hi,
> >>
> >>         Should TableForm go into SymPy or not? Here is the pull
> >>         request:
> >>
> >>         https://github.com/sympy/sympy/pull/268
> >>
> >>         here is the discussion at length:
> >>
> >>         https://groups.google.com/d/topic/sympy/RSOo5cZNd2E/discussion
> >>
> >>         and here are the examples of usage of TableForm:
> >>
> >>         https://groups.google.com/d/msg/sympy/RSOo5cZNd2E/8bLSGHqcdU4J
> >>         https://groups.google.com/d/msg/sympy/RSOo5cZNd2E/TWmxeR-0A3gJ
> >>
> >>         I am +1 to have it in SymPy, Ronan is -1. What do other people
> >>         think?
> >>
> >>
> >>         I guess the discussion is whether we want SymPy to contain all
> >>         useful
> >>         code regarding symbolic manipulation, just like Mathematica
> >>         has
> >>         (Mathematica has TableForm in by default). My vision is
> >>         clearly yes.
> >>         But I would like to know the opinion of the sympy community.
> >>
> > But printing tables isn't symbolic manipulation. The comparison with
> > Mathematica isn't terribly helpful since it isn't a library but a
> > full-fledged programming language, and the IDE that goes with it, and
> > more things besides.
> 
> My vision has always been to take numpy, scipy, matplotlib and sympy,
> put an IDE around it and get something like Mathematica (or better).
> 
> >
> >>
> >> I didn't look into the implementation yet, but anyway +1 for the idea.
> >> The question you formulated it broader than this and includes OEIS,
> >> physics, code generators etc. In my opinion all this should be in
> >> SymPy and moreover, this is how we (at least I, but I'm sure it's not
> >> only me) developed SymPy from the very beginning, so I don't see why
> >> we should change this direction now.
> 
> 
> I still want to continue going in this direction, but Ronan gave -1 to
> my patch which implements another little thing on the road (the reason
> being that it doesn't fit into sympy, not that the patch quality is
> bad --- which it might be, but that's not what we are discussing now),
> and as such, I want to open the public discussion about this and see
> what people think.
> 
> >
> > I don't understand how you define "all this". Code generation is
> > perfectly in scope for a symbolic manipulation library and making sympy
> > usable for physics calculations is basically a core design goal (though
> > sympy.physics should ideally be split off into a separate project having
> > sympy as its major dependency). On the other hand, printing a table and
> > calling up a bookmark seem rather peripheral to the goal of building a
> > library for symbolic mathematics.
> 
> TableForm is extremely useful, people who use Mathematica use it. So
> it needs to go somewhere.
> 
But it's just a wrapper used but Mathematica's printing system. It only
makes sense in relation with InputForm, OutputForm, StandardForm,
MatrixForm, ...

> It might go into SciPy, but given all the printing system that we have
> in sympy, it nicely fits into the SymPy infrastructure.
> 
It certainly doesn't fit as a subclass of Basic. What's TableForm.args?
TableForm.subs()? 

> Ronan, if not into SymPy, where should TableForm go?

I'd say nowhere, because tablib[http://tablib.org] already exists.



-- 
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