Meta-ticket: Split sage.geometry.polyhedron.base

[mkoeppe]

This is a pretty large module, consisting of 11 kLOC.

We could split it into several modules, each defining an intermediate class (like it is done in sage.matrix.matrix0, matrix1, matrix2).

For example,

• a module sage.geometry.polyhedron.base0 contains the very basic methods (initialization etc) Split initialization and basic access of polyhedra out as a separate module #32767.
• a module sage.geometry.polyhedron.base1 could contain the implementation of the methods defined by the ConvexSet_base API, split convex set methods of polyhedra into base1.py #32776
• a module sage.geometry.polyhedron.base2 could contain all methods that have to do with lattice points in polyhedra, Move lattice point methods from Polyhedron_base to Polyhedron_base2 #32876
• a module sage.geometry.polyhedron.base3 contains the combinatorial methods, Move combinatorial methods of Polyhedron_base to Polyhedron_base3 #32884
• a module sage.geometry.polyhedron.base4 could contain all methods whose implementation needs sage.graphs.graphs, Move graphs methods to Polyhedron_base4 #32889 -- this would help with modularization because then we could conditionalize the doctests via a file-level directive (sage.doctest.control: Exclude doctests in files via file directives ''# sage.doctest: optional - xyz' #30778)
  • vertex_graph, graph, vertex_digraph, vertex_facet_graph
  • hasse_diagram
  • face_lattice, flag_f_vector
  • {combinatorial,restricted}_automorphism_group
  • is_combinatorially_isomorphic
• a module sage.geometry.polyhedron.base5 could contain all constructions of new polyhedra, Move construction methods to Polyhedron_base5 #33123
• a module sage.geometry.polyhedron.base6 could contain all methods related to plotting and affine hull (projection), Move plotting and affine hull to Polyhedron_base6 #33449
• a module sage.geometry.polyhedron.base7 could contain all methods related to triangulation and volume, Move integration and volume to Polyhedron_base7 #33583

This could also lead to an improvement of the documentation.

CC: @kliem @jplab @tscrim

Component: refactoring

Issue created by migration from https://trac.sagemath.org/ticket/32651

[kliem]

comment:13

Is this done with #33583? What did you have in mind with "This could also lead to an improvement of the documentation."?

[jplab]

comment:14

That's a great ticket! ... I'm not dead. I expect to be "back on trac" soon!

I would be up for improving the doc and complement it further... I'll slowly go through all this new stuff in polyhedron this Summer.

Thanks for all the work!

[jhpalmieri]

comment:17

Is this done?

[mkoeppe]

comment:18

It's done except we didn't get around to do "an improvement of the documentation" (comment:13) -- in the course of which we would surely have noticed that the documentation for all of this is no longer being built (#25675).