Documentation

TauCeti.FieldTheory.FunctionField.Place.OfValuationSubring

Valuation rings of an algebraic function field are the rings of places #

Stichtenoth defines a valuation ring of F / k to be a subring š’Ŗ with k ⊊ š’Ŗ ⊊ F such that z ∈ š’Ŗ or z⁻¹ ∈ š’Ŗ for every z : F (Definition 1.1.4), and proves that every such ring is a discrete valuation ring (Theorem 1.1.6), so that the places of F / k are exactly the proper valuation subrings of F containing the constants (Theorem 1.1.13). This file proves that recognition theorem: TauCeti.Place.ofValuationSubring turns a proper ValuationSubring F containing k into a place whose valuation ring is the given one, and TauCeti.Place.existsUnique_integers_eq says the place is unique with that property.

The mathematical content is the discreteness, TauCeti.isDiscreteValuationRing_of_isFunctionField, and the argument is Stichtenoth's. The engine is Lemma 1.1.7: if x is a nonzero nonunit of š’Ŗ, then a family y i of nonunits whose valuations increase strictly and stay above v x is linearly independent over k(x), hence has at most [F : k(x)] members. Because x is transcendental — a nonunit cannot be algebraic over k — that degree is finite, so š’Ŗ admits no infinite chain of nonunits of strictly increasing valuation. Two applications finish the proof: the valuations of the nonzero nonunits attain a maximum, at an element t, and every nonzero z : š’Ŗ is t ^ n times a unit. That is exactly Mathlib's HasUnitMulPowIrreducibleFactorization.

Main results #

Implementation notes #

Everything is phrased through ValuationSubring.valuation, the tautological valuation of a valuation subring, rather than through membership in the subring: x ∈ A is A.valuation x ≤ 1 and x is a nonunit of A exactly when A.valuation x < 1, and in this vocabulary the estimates of Lemma 1.1.7 are one-line applications of Valuation.map_sum_eq_of_lt. The value group of A.valuation is only known to be a linearly ordered commutative group with zero; that it is ℤᵐ⁰ is the conclusion, not a hypothesis, and it is obtained by handing the discrete valuation ring back to Mathlib's IsDiscreteValuationRing.maximalIdeal and the adic valuation of that height-one prime.

References #

Stichtenoth's chain estimate #

theorem TauCeti.linearIndependent_of_strictMono_valuation {k : Type u} {F : Type v} [Field k] [Field F] [Algebra k F] {A : ValuationSubring F} (hk : āˆ€ (c : k), (algebraMap k F) c ∈ A) {x : F} (hx0 : x ≠ 0) (hx : A.valuation x < 1) {ι : Type u_1} [LinearOrder ι] {y : ι → F} (hy : āˆ€ (i : ι), A.valuation (y i) < 1) (hmono : StrictMono fun (i : ι) => A.valuation (y i)) (hxy : āˆ€ (i : ι), A.valuation x ≤ A.valuation (y i)) :
LinearIndependent (ↄk⟮x⟯) y

Stichtenoth, Lemma 1.1.7. Let x be a nonzero nonunit of a valuation subring A of F containing the constants. A family of nonunits of A whose valuations increase strictly along a linear order, and are all at least A.valuation x — hence nonzero — is linearly independent over k(x).

In Stichtenoth's additive notation the hypothesis reads ord x ≄ ord (y iā‚€) > ord (y i₁) > ⋯ > 0, and the conclusion bounds the length of such a chain by [F : k(x)].

Discreteness #

theorem TauCeti.isDiscreteValuationRing_of_isFunctionField {k : Type u} {F : Type v} [Field k] [Field F] [Algebra k F] {A : ValuationSubring F} (hF : IsFunctionField k F) (hk : āˆ€ (c : k), (algebraMap k F) c ∈ A) (hA : A ≠ ⊤) :

Stichtenoth, Theorem 1.1.6. A proper valuation subring of an algebraic function field that contains the constants is a discrete valuation ring.

The place of a valuation subring #

The valuation subring of the adic valuation of the maximal ideal of a discrete valuation subring A of F is A itself. This is Mathlib's IsDiscreteValuationRing.map_algebraMap_eq_valuationSubring, read through the fact that the structure map of a valuation subring is the inclusion.

noncomputable def TauCeti.Place.ofValuationSubring {k : Type u} {F : Type v} [Field k] [Field F] [Algebra k F] {A : ValuationSubring F} (hF : IsFunctionField k F) (hk : āˆ€ (c : k), (algebraMap k F) c ∈ A) (hA : A ≠ ⊤) :
Place k F

Stichtenoth, Theorem 1.1.13. The place of F / k attached to a proper valuation subring of F containing the constants: its valuation is the adic valuation of the maximal ideal of the subring, which is a discrete valuation ring by TauCeti.isDiscreteValuationRing_of_isFunctionField.

Equations
  • One or more equations did not get rendered due to their size.
Instances For
    theorem TauCeti.Place.valuation_ofValuationSubring {k : Type u} {F : Type v} [Field k] [Field F] [Algebra k F] {A : ValuationSubring F} (hF : IsFunctionField k F) (hk : āˆ€ (c : k), (algebraMap k F) c ∈ A) (hA : A ≠ ⊤) :

    The valuation of TauCeti.Place.ofValuationSubring is the adic valuation of the maximal ideal of the subring.

    @[simp]
    theorem TauCeti.Place.integers_ofValuationSubring {k : Type u} {F : Type v} [Field k] [Field F] [Algebra k F] {A : ValuationSubring F} (hF : IsFunctionField k F) (hk : āˆ€ (c : k), (algebraMap k F) c ∈ A) (hA : A ≠ ⊤) :

    The valuation ring of TauCeti.Place.ofValuationSubring is the subring one started from.

    theorem TauCeti.Place.existsUnique_integers_eq {k : Type u} {F : Type v} [Field k] [Field F] [Algebra k F] {A : ValuationSubring F} (hF : IsFunctionField k F) (hk : āˆ€ (c : k), (algebraMap k F) c ∈ A) (hA : A ≠ ⊤) :

    Stichtenoth, Theorem 1.1.13. A proper valuation subring of an algebraic function field containing the constants is the valuation ring of exactly one place.