Addition of new presentations to Chow rings of matroids

[25shriya]

This PR focuses on adding the atom-free and simplicial presentation of Chow rings of matroids. see relevant issue
The former is taken from Definition 4.1 of this paper and the latter from Definition 3.2.1 of this paper.

@tscrim

📝 Checklist

• The title is concise and informative.
• The description explains in detail what this PR is about.
• I have linked a relevant issue or discussion.
• I have created tests covering the changes.
• I have updated the documentation and checked the documentation preview.

⌛ Dependencies

[github-actions]

Documentation preview for this PR (built with commit 2feb4d6; changes) is ready! 🎉
This preview will update shortly after each push to this PR.

[tscrim]

You might consider using register_unpickle_override for backwards compatibility. (Technically speaking, the old class should be deprecated.)

[25shriya]

I'm a little confused about the issue here, primarily because the version in my branch defines the FY presentation as ChowRingIdeal_nonaug_fy (I made that change long ago)

[tscrim]

If I created a pickle with the old version, then tried to unpickle it with the new version, it wouldn’t know how to deal with it because the class wouldn’t exist anymore (since its name has changed). This is a backwards incompatible change.

[25shriya]

But isn't register_unpickle_override used only when both versions exist?

[tscrim]

No, it redirects the unpickling.

[25shriya]

So, would I need to import all of the following from sage.misc.persist import (unpickle_override, register_unpickle_override, SageUnpickler) and implement another method or something?
Sorry I'm quite new to dealing with pickling

[tscrim]

No, just thr unpickle_override. The path is set by strings. There are some examples around the code base, but i don't remember where offhand.

[25shriya]

I hope the new commit fixes this

[25shriya]

Why does build-doc-pdf fail when build-doc passes?
I've checked every corner of the documentation. Not sure what I'm missing

[tscrim]

I hope the new commit fixes this

I think so, but you can check it by creating a pickle on the develop branch (using dumps() and saving the output string), then loading it with your PR (via loads(saved_string)).

[tscrim]

The pdf build is more sensitive than the html. You can usually find the issue by looking through the html doc.

[tscrim]

Indeed, I found it. \superseteq should be \supseteq as I recall.

[25shriya]

Seems like linting is failing due to some other code not authored by me in this PR

[tscrim]

One failure in the kahler_algebras.py file due to a basis order change (I think).

[25shriya]

What's wrong with the doctests check? The doctests pass locally in all edited files for me

[tscrim]

Looks like somehow the ordering of the output is different. I am not sure why ar this point though.

[25shriya]

Do the tests pass locally for you?

[tscrim]

It doesn't really matter. They should pass on all machines and this indicating that there might be something in the ordering that is machine dependent.

[25shriya]

Is it worth running a sorted() on groebner_basis(), normal_basis() and basis() methods?

[tscrim]

We could, but it's overkill to do that in the code. It's a bit annoying to do it in the output too. I can only guess that it comes from iterating over the lattice of flats. Still not sure yet what the best thing to do now is. I will think about it.

[tscrim]

Also I get a different order in the doctests too.