Module

FractionalIdealNormPrimeExponent.lean

Nyll.ClassFieldTheory.Theorems.ConductorsAndRayClassFields.FractionalIdealNormPrimeExponent
1 declaration indexed from sourceView source ↗

Prime exponents of a fractional-ideal norm

At a finite prime of the base, the exponent of the norm is the sum of the upstairs exponents, each weighted by its inertia degree. This is the calculation needed when passing from ideals to norm-defined ray subgroups.

Declarations

ClassFieldTheory.fractionalIdealNorm_primeExponent

The exponent at v of an ideal norm is the inertia-degree-weighted sum of the exponents at the primes lying above v.

Source context 6 commands

Surrounding namespaces, section variables, and local commands at this declaration.

open scoped NumberField
open NumberField IsDedekindDomain
noncomputable section
namespace ClassFieldTheory
universe u v
open scoped Classical in
theorem fractionalIdealNorm_primeExponent
    (K : Type u) (L : Type v)
    [Field K] [NumberField K]
    [Field L] [NumberField L] [Algebra K L]
    [FiniteDimensional K L]
    (I : NumberFieldFractionalIdealGroup L)
    (v : HeightOneSpectrum (𝓞 K)) :
    FractionalIdeal.count K v
        ((fractionalIdealNorm K L I : NumberFieldFractionalIdealGroup K) :
          FractionalIdeal (nonZeroDivisors (𝓞 K)) K) =
      (NumberFieldFractionalIdealGroup.countVector I).sum fun W n =>
        if fractionalIdealNormPrimeBelow K L W = v then
          (W.asIdeal.inertiaDeg (𝓞 K) : ℤ) * n
        else 0 := by
  exact fractionalIdealNorm_count K L I v