Closed daviddelaat closed 10 years ago
For now I've created a new package for multivariate polynomials, so that I can experiment with the design a bit more. I've tried to use the same naming conventions as Polynomial.jl
so that it will feel natural to use both package side by side, and for a possible future merger.
Package url: https://github.com/daviddelaat/MultiPoly.jl
I'll close this issue for now.
I wasn't trying to ignore you, but your pull request caused renewed interest in fixing #5 first
No problem. I've put the multivariate code in a separate package so that I can experiment with the API a bit. When it stabilizes I'll do another pull request. Issue #5 indeed seems very important for this.
This adds support for sparse multivariate polynomials. (As multivariate polynomials tend to be sparse I think that dense univariate and sparse multivariate are the most important types for polynomials)
I'm not sure if there is interest in this functionality for this package, but I think it could be a good fit. I could also make a separate
SpareMultiPol.jl
package if that is preferred.The idea here would be that for instance the addition of two dense univariate polynomials with different variables would yield a sparse multivariate polynomial (not implemented).
Documentation is still missing but see
test/testmpol.jl
for some examples.