cleanup, expand explanation

This commit is contained in:
Floris van Doorn 2018-11-13 19:36:35 -05:00
parent 2913de520d
commit 02f5c54b77

View file

@ -9,16 +9,24 @@ open eq pointed is_trunc is_conn is_equiv equiv sphere fiber chain_complex left_
namespace cohomology namespace cohomology
universe variable u universe variable u
/- /-
We have maps: Given a pointed map E →* B with as fiber the sphere S^{n+1} and an abelian group A.
The only nontrivial differentials in the spectral sequence of this map are the following
differentials on page n:
d_m = d_(m-1,n+1)^n : E_(m-1,n+1)^n → E_(m+n+1,0)^n d_m = d_(m-1,n+1)^n : E_(m-1,n+1)^n → E_(m+n+1,0)^n
Note that ker d_m = E_(m-1,n+1)^∞ and coker d_m = E_(m+n+1,0)^∞. Note that ker d_m = E_(m-1,n+1)^∞ and coker d_m = E_(m+n+1,0)^∞.
We have short exact sequences Each diagonal on the ∞-page has at most two nontrivial groups, which means that
coker d_{m-1} → D_{m+n}^∞ → ker d_m coker d_{m-1} and ker d_m are the only two nontrivial groups building up D_{m+n}^∞,
where D^∞ is the abutment of the spectral sequence. where D^∞ is the abutment of the spectral sequence.
This comes from the spectral sequence, using the fact that coker d_{m-1} and ker d_m are the only This gives the short exact sequences:
two nontrivial groups building up D_{m+n}^∞ (in the filtration of D_{m+n}^∞). coker d_{m-1} → D_{m+n}^∞ → ker d_m
We can splice these SESs together to get a LES We can splice these SESs together to get a LES
... E_(m+n,0)^n → D_{m+n}^∞ → E_(m-1,n+1)^n → E_(m+n+1,0)^n → D_{m+n+1}^∞ ... ... E_(m+n,0)^n → D_{m+n}^∞ → E_(m-1,n+1)^n → E_(m+n+1,0)^n → D_{m+n+1}^∞ ...
Now we have
E_(p,q)^n = E_(p,q)^0 = H^p(B; H^q(S^{n+1}; A)) = H^p(B; A) if q = n+1 or q = 0
and
D_{n}^∞ = H^n(E; A)
This gives the Gysin sequence
... H^{m+n}(B; A) → H^{m+n}(E; A) → H^{m-1}(B; A) → H^{m+n+1}(B; A) → H^{m+n+1}(E; A) ...
-/ -/
@ -111,11 +119,6 @@ left_module.LES_of_SESs _ _ _ (λm, convergent_spectral_sequence.d c n (m - 1, n
refine ap (λx, x + _) !add.right_inv ⬝ !zero_add }} refine ap (λx, x + _) !add.right_inv ⬝ !zero_add }}
end end
-- set_option pp.universes true
-- print unreduced_ordinary_cohomology_sphere_zero
-- print unreduced_ordinary_cohomology_zero
-- print ordinary_cohomology_sphere_of_neq
-- set_option formatter.hide_full_terms false
definition gysin_sequence'_zero {E B : Type*} {n : } (HB : is_conn 1 B) (f : E →* B) definition gysin_sequence'_zero {E B : Type*} {n : } (HB : is_conn 1 B) (f : E →* B)
(e : pfiber f ≃* sphere (n+1)) (A : AbGroup) (m : ) : (e : pfiber f ≃* sphere (n+1)) (A : AbGroup) (m : ) :
gysin_sequence' HB f e A (m, 0) ≃lm LeftModule_int_of_AbGroup (uoH^m+n+1[B, A]) := gysin_sequence' HB f e A (m, 0) ≃lm LeftModule_int_of_AbGroup (uoH^m+n+1[B, A]) :=
@ -170,6 +173,7 @@ begin
{ refine uoH^≃n+1[e⁻¹ᵉ*, A] ⬝g unreduced_ordinary_cohomology_sphere _ _ (succ_ne_zero n) } { refine uoH^≃n+1[e⁻¹ᵉ*, A] ⬝g unreduced_ordinary_cohomology_sphere _ _ (succ_ne_zero n) }
end end
-- todo: maybe rewrite n+m to m+n (or above rewrite m+n+1 to n+m+1 or n+(m+1))?
definition gysin_sequence'_two {E B : Type*} {n : } (HB : is_conn 1 B) (f : E →* B) definition gysin_sequence'_two {E B : Type*} {n : } (HB : is_conn 1 B) (f : E →* B)
(e : pfiber f ≃* sphere (n+1)) (A : AbGroup) (m : ) : (e : pfiber f ≃* sphere (n+1)) (A : AbGroup) (m : ) :
gysin_sequence' HB f e A (m, 2) ≃lm LeftModule_int_of_AbGroup (uoH^n+m[E, A]) := gysin_sequence' HB f e A (m, 2) ≃lm LeftModule_int_of_AbGroup (uoH^n+m[E, A]) :=