[Merged by Bors] - feat(AlgebraicGeometry): Zariski's main theorem#35093
[Merged by Bors] - feat(AlgebraicGeometry): Zariski's main theorem#35093erdOne wants to merge 30 commits intoleanprover-community:masterfrom
Conversation
erdOne
commented
Feb 10, 2026
Co-authored-by: Joël Riou <37772949+joelriou@users.noreply.github.com>
…lib4 into erd1/unramifiedLocalStructure3
Co-authored-by: Christian Merten <christian@merten.dev>
…rdOne/mathlib4 into erd1/unramifiedLocalStructure3
PR summary 2598404fe9Import changes for modified filesNo significant changes to the import graph Import changes for all files
|
…lib4 into erd1/ZMTScheme
d4dc886 to
f2f819d
Compare
786cbd4 to
f2f819d
Compare
| lemma Scheme.Hom.QuasiFiniteAt.quasiFiniteAt | ||
| {x : X} (hx : f.QuasiFiniteAt x) {V : X.Opens} (hV : IsAffineOpen V) {U : Y.Opens} | ||
| (hU : IsAffineOpen U) (hVU : V ≤ f ⁻¹ᵁ U) (hxV : x ∈ V.1) : | ||
| letI := (f.appLE U V hVU).hom.toAlgebra |
There was a problem hiding this comment.
Can we avoid this? Should we have a RingHom.QuasiFiniteAt?
There was a problem hiding this comment.
I think the issue here is that Algebra Γ(Y, U) Γ(X, V) should be an instance but couldn't because of hVU : V ≤ f ⁻¹ᵁ U. Maybe we can experiment with having Fact (V ≤ f ⁻¹ᵁ U).
As for RingHom.QuasiFiniteAt, I don't see any applications beyond applying to (f.appLE U V hVU).hom for now, and even if one adds such a predicate, the first thing one would do when applying this lemma is to algebraize[(f.appLE U V hVU).hom] and turn this back to Algebra.QuasiFiniteAt.
So I don't think this justifies having a bunch of API added for this. Unless you are satisfied by a barebones abbrev with virtually no API on it.
There was a problem hiding this comment.
I have added a barebones abbrev with virtually no API on it.
Co-authored-by: Christian Merten <christian@merten.dev>
Co-authored-by: Christian Merten <christian@merten.dev>
e56ce45 to
b97aead
Compare
|
Thanks! bors merge |
|
Pull request successfully merged into master. Build succeeded: |