Monday, June 4, 2012

Class structure

The class structure of the upcoming lattice module is now as follows: Lattice_with_basis inherits from FreeModule_submodule_with_basis_pid (with some tweaks as described in my previous post) and is the general class for all lattices with bases. This is what Robert Miller called SubLatticeModule in a rather old discussion about lattices, although it is more general as it does not restrict to ZZ-lattices. Would it make sense to implement a general LatticeModule (in Robert Miller's notion) that does not have a concrete basis? How would such a LatticeModule be constructed and what could it offer?

Inheriting from Lattice_with_basis there is Lattice_ZZ that specializes on lattices with integer coefficients (ZZ-Lattices), probably the most usual form of lattices.

Furthermore, there is the specialization on lattices embeddable in the real vector space, Lattice_ZZ_in_RR (again, a very common form of lattices). The actual type of the vectors in the lattice could still be ZZ^n or QQ^n (not necessarily RR^n). Most of the algorithms implemented in this project will probably lie in this class.

I have already added a few lines of documentation and some examples/doctests.

You can see the current state of the code in the main file in the code repository on github.

No comments:

Post a Comment