On Nov 23, 2013, at 3:36 AM, Sergey Kirpichev <[email protected]> wrote:
> > I think we should rather represent series as a Series class, roughly > equivalent to SeriesData in Mathematica: > > http://reference.wolfram.com/mathematica/ref/SeriesData.html<http://www.google.com/url?q=http%3A%2F%2Freference.wolfram.com%2Fmathematica%2Fref%2FSeriesData.html&sa=D&sntz=1&usg=AFQjCNGUrewTF7qnSvMywwXBrTqnr9qneg> > This is a bit undocumented (as all in Mathematica, usually). For example, it's not so clear what kind of coefficients are allowed. My impression was that Mathematica had excellent documentation. In fact, it seems to be the best in the field. But I've never used the actual functions outside of wolfram alpha, so I guess I missed things like under specification. But definitely if you consider Mathematica docs, plus the functions site, plus mathworld, plus the more or less interactive documentation that is wolfram alpha, the Mathematica documentation is the best in the field. Aaron Meurer > The Series class would interact with SymPy similar to how Poly works in > SymPy. > The Series would represent a series expansion in one variable (I don't > have opinions about multiple variables yet), so it would remember the > terms in some suitable format (see below, just like for Polys, there > are more options on the internal representation) and the order of the > series. Then, when you print it, it will print it just like we > currently do, with the "O(x^2)" term, though that's just how it is > printed. > Not sure if this is a Series class job. I mean, to remember terms. The Series should be build on (or to be itself) the ground of some stream-like entity, that can do actual arithmetic. Memoization/caching - another issue. > If you look at how GiNaC does it > It looks, like it can only formal power series and laurent (just like sage?). We want more? Open ideas: > > * I don't like that Order requires special handling in Add: > > https://github.com/sympy/sympy/blob/master/sympy/core/add.py#L237<https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2Fsympy%2Fsympy%2Fblob%2Fmaster%2Fsympy%2Fcore%2Fadd.py%23L237&sa=D&sntz=1&usg=AFQjCNFg_9hCCBrkVjHKxInryZlS4ADe6g> > That's another issue. But arithmetics (not differential calculus!) with O makes sense. -- 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. For more options, visit https://groups.google.com/groups/opt_out. -- 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. For more options, visit https://groups.google.com/groups/opt_out.
