mirror of
https://github.com/GTBarkley/comm_alg.git
synced 2024-12-26 07:38:36 -06:00
commit
b832cecf59
1 changed files with 16 additions and 3 deletions
|
@ -77,12 +77,25 @@ lemma containment_radical_power_containment :
|
||||||
rintro ⟨RisNoetherian, containment⟩
|
rintro ⟨RisNoetherian, containment⟩
|
||||||
rw [isNoetherianRing_iff_ideal_fg] at RisNoetherian
|
rw [isNoetherianRing_iff_ideal_fg] at RisNoetherian
|
||||||
specialize RisNoetherian (Ideal.radical I)
|
specialize RisNoetherian (Ideal.radical I)
|
||||||
rcases RisNoetherian with ⟨S, Sgenerates⟩
|
-- rcases RisNoetherian with ⟨S, Sgenerates⟩
|
||||||
|
have containment2 : ∃ n : ℕ, (Ideal.radical I) ^ n ≤ I := by
|
||||||
|
apply Ideal.exists_radical_pow_le_of_fg I RisNoetherian
|
||||||
|
cases' containment2 with n containment2'
|
||||||
|
have containment3 : J ^ n ≤ (Ideal.radical I) ^ n := by
|
||||||
|
apply Ideal.pow_mono containment
|
||||||
|
use n
|
||||||
|
apply le_trans containment3 containment2'
|
||||||
|
-- The above can be proven using the following quicker theorem that is in the wrong place.
|
||||||
|
-- Ideal.exists_pow_le_of_le_radical_of_fG
|
||||||
|
|
||||||
-- how to I get a generating set?
|
|
||||||
|
|
||||||
-- Stacks Lemma 10.52.6: I is a maximal ideal and IM = 0. Then length of M is
|
-- Stacks Lemma 10.52.6: I is a maximal ideal and IM = 0. Then length of M is
|
||||||
--
|
-- the same as the dimension as a vector space over R/I,
|
||||||
|
lemma length_eq_dim_if_maximal_annihilates {I : Ideal R} [Ideal.IsMaximal I]
|
||||||
|
: I • (⊤ : Submodule R M) = 0
|
||||||
|
→ Module.length R M = Module.rank R⧸I M⧸(I • (⊤ : Submodule R M)) := by sorry
|
||||||
|
|
||||||
|
-- Does lean know that M/IM is a R/I module?
|
||||||
|
|
||||||
-- Stacks Lemma 10.52.8: I is a finitely generated maximal ideal of R.
|
-- Stacks Lemma 10.52.8: I is a finitely generated maximal ideal of R.
|
||||||
-- M is a finite R-mod and I^nM=0. Then length of M is finite.
|
-- M is a finite R-mod and I^nM=0. Then length of M is finite.
|
||||||
|
|
Loading…
Reference in a new issue