> I took a look at the line_integrate() function. The Curve class is > something of a start. Though, I like very much Raoul's idea for a > single class to represent everything (path/surface/volume). I think > this is doable.
The idea is to push this as far as possible before switching to the heavy diffgeo stuff. As mentioned, this breaks down when we need multiple patches to cover the manifold. > > would then represent an integral over an N-dimensional hypervolume > > I have not had a relevant course that would acquaint me with > 'N-dimensional hypervolumes'. Nevertheless, I am willing to read up > on it. I hope that just implementing integrals in such hypervolumes > wouldn't be too different from the 2 or 3 dimensional case. No, the principle is mostly the same, just there are more variables. There is nothing fundamentally new in higher dimensions. > > How are they numerical in nature? Integrals are by their nature > > symbolic. > They take symbolic functions and > return a symbolic result. > Numerics are useful if the answer is not expressible in terms of nice > functions. Well, numerics is another topic. If we can not solve the integrals, return them symbolically as unevaluated placeholder objects. In the best case, the users then can call .evalf() and get a number. I'd not make numerics part of your proposal. There is more than enough work with the vector calc stuff. -- 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 sympy+unsubscr...@googlegroups.com. To post to this group, send email to sympy@googlegroups.com. Visit this group at http://groups.google.com/group/sympy?hl=en-US. For more options, visit https://groups.google.com/groups/opt_out.
