polyPrimeCost_nonneg

formal source

 115
 116/-- `polyPrimeCost` is nonnegative on any polynomial (it is always `J`
 117    of a positive real). -/
 118theorem polyPrimeCost_nonneg (P : Polynomial F) : 0 ≤ polyPrimeCost P :=
 119  Jcost_nonneg (polyNorm_pos P)
 120
 121/-- `polyCost` is nonnegative on any polynomial. -/
 122theorem polyCost_nonneg (f : Polynomial F) : 0 ≤ polyCost f := by
 123  unfold polyCost
 124  apply Multiset.sum_nonneg
 125  intro x hx
 126  obtain ⟨P, _, hP_eq⟩ := Multiset.mem_map.mp hx
 127  rw [← hP_eq]
 128  exact polyPrimeCost_nonneg P
 129
 130/-! ## Multiplicativity over factorization -/
 131
 132/-- The total cost is additive under multiplication of nonzero polynomials.
 133    This is the function-field analog of `costSpectrumValue_mul`. -/
 134theorem polyCost_mul {f g : Polynomial F} (hf : f ≠ 0) (hg : g ≠ 0) :
 135    polyCost (f * g) = polyCost f + polyCost g := by
 136  unfold polyCost
 137  rw [normalizedFactors_mul hf hg]
 138  rw [Multiset.map_add, Multiset.sum_add]
 139
 140/-! ## Cost on irreducibles and powers -/
 141
 142/-- For a monic irreducible polynomial `P`, the cost equals `J(‖P‖)`.
 143    Mathlib's `normalizedFactors` returns the multiset of monic
 144    irreducible factors. -/
 145theorem polyCost_irreducible {P : Polynomial F}
 146    (hP_irr : Irreducible P) (hP_monic : P.Monic) :
 147    polyCost P = polyPrimeCost P := by
 148  unfold polyCost