Newton polygon of polynomials over power series with infinite precision

[Bruno Grenet]

Computing the Newton polygon of a polynomial with coefficients in a ring of power series fails if the coefficients have infinite precision:

sage: S.<x> = PowerSeriesRing(GF(5))
sage: R.<y> = S[]
sage: p = x^2+y+x*y^2
sage: p.newton_polygon()
Traceback (most recent call last):
...
IndexError: list index out of range

[Xavier Caruso]

I attach a small patch fixing this issue. I've also updated the method _factor_of_degree in order to make the slope factorization work correctly. (Otherwise, at infinite precision, it may never stop.)

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​61cb00c

Newton polygon and slope factorization fixed

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​ac0d975

Replace TESTS: by TESTS::

[Julian Rüth]

Looks good, except for the TESTS:. Positive review if the patchbot is happy.

---

New commits:

​ac0d975

Replace TESTS: by TESTS::

[Bruno Grenet]

Two One comments:

• You changed too many TESTS: by TESTS::: There should be double colons only when TESTS: is followed by an actual test, not when it is followed by a sentence that explains the test.

• It would be nice to add a test for slope factorization. (The test already exists...)

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​144898d	Merge branch 'newton_polygon_infinite_prec' into a
​f5fbc72	Better writing for TESTS::?

[Bruno Grenet]

LGTM, thanks for the patch!