File size: 8,654 Bytes
4a56cd6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
import OAI.AlgebraicGeometry.ComplexMultiplication.CM.ImaginaryElement
import Mathlib.NumberTheory.NumberField.CanonicalEmbedding.Basic
import Mathlib.Topology.Algebra.Group.Quotient
import Mathlib.Topology.Algebra.IsUniformGroup.Basic

/-! # Period Space -/

namespace OAI.AlgebraicGeometry.ComplexMultiplication.CM

noncomputable section

/-!
# The lattice in the selected CM embedding space

Choosing one embedding at every infinite place identifies the usual mixed
embedding space with a complex vector space. The image of the ring of
integers is a discrete full lattice there. Its additive quotient is compact.
This constructs the underlying topological torus; projectivity and the
identification of its geometric cohomology are separate steps.
-/

open scoped Classical _root_.NumberField

open _root_.NumberField

variable (E : Type*) [Field E] [Algebra ℚ E]

/-- One complex coordinate for each conjugate pair of embeddings. -/
abbrev PeriodSpace.CMPeriodSpace := InfinitePlace E → ℂ

variable {E} (ε : TypeHodgeStructure.CMEmbeddingType E)

/-- Adjust the canonical embedding at a place to the embedding selected by the CM type. -/
def TypeHodgeStructure.CMEmbeddingType.periodConjugation (v : InfinitePlace E) : ℂ ≃ₗ[ℝ] ℂ :=
  if ε.val v.embedding.toRatAlgHom then LinearEquiv.refl ℝ ℂ
  else Complex.conjAe.toLinearEquiv

theorem TypeHodgeStructure.CMEmbeddingType.periodConjugation_involutive (v : InfinitePlace E) (z : ℂ) :
    ε.periodConjugation v (ε.periodConjugation v z) = z := by
  by_cases hv : ε.val v.embedding.toRatAlgHom = true
  · simp [TypeHodgeStructure.CMEmbeddingType.periodConjugation, hv]
  · simp [TypeHodgeStructure.CMEmbeddingType.periodConjugation, hv]

theorem TypeHodgeStructure.CMEmbeddingType.periodConjugation_embedding (v : InfinitePlace E) (x : E) :
    ε.periodConjugation v (v.embedding x) = ε.placeEmbedding v x := by
  by_cases hv : ε.val v.embedding.toRatAlgHom = true
  · simp [TypeHodgeStructure.CMEmbeddingType.periodConjugation, TypeHodgeStructure.CMEmbeddingType.placeEmbedding, hv]
  · simp [TypeHodgeStructure.CMEmbeddingType.periodConjugation, TypeHodgeStructure.CMEmbeddingType.placeEmbedding, hv, EmbeddingHodgeType.CM.conjugateEmbedding_apply]

/-- The actual field embedding in the selected complex coordinates. -/
def TypeHodgeStructure.CMEmbeddingType.periodEmbedding : E →+* PeriodSpace.CMPeriodSpace E :=
  RingHom.pi fun v => (ε.placeEmbedding v).toRingHom

@[simp]
theorem TypeHodgeStructure.CMEmbeddingType.periodEmbedding_apply (x : E) (v : InfinitePlace E) :
    ε.periodEmbedding x v = ε.placeEmbedding v x := rfl

/-- All embeddings are the selected embeddings and their conjugates, without repetition. -/
def TypeHodgeStructure.CMEmbeddingType.embeddingPairEquiv : (InfinitePlace E × Bool) ≃ (E →ₐ[ℚ] ℂ) where
  toFun p := if p.2 then ε.placeEmbedding p.1 else EmbeddingHodgeType.CM.conjugateEmbedding (ε.placeEmbedding p.1)
  invFun σ := (InfinitePlace.mk σ.toRingHom, ε.val σ)
  left_inv p := by
    rcases p with ⟨v, b⟩
    cases b
    · apply Prod.ext
      · change InfinitePlace.mk (ComplexEmbedding.conjugate (ε.placeEmbedding v).toRingHom) = v
        rw [InfinitePlace.mk_conjugate_eq, ε.mk_placeEmbedding]
      · simp [ε.property, ε.placeEmbedding_positive]
    · apply Prod.ext
      · exact ε.mk_placeEmbedding v
      · exact ε.placeEmbedding_positive v
  right_inv σ := by
    by_cases hσ : ε.val σ = true
    · simpa only [hσ, ↓reduceIte] using ε.placeEmbedding_mk σ hσ
    · have hc : ε.val (EmbeddingHodgeType.CM.conjugateEmbedding σ) = true := by
        rw [ε.property, Bool.eq_false_iff.mpr hσ, Bool.not_false]
      have hp := ε.placeEmbedding_mk (EmbeddingHodgeType.CM.conjugateEmbedding σ) hc
      have hm : InfinitePlace.mk (EmbeddingHodgeType.CM.conjugateEmbedding σ).toRingHom =
          InfinitePlace.mk σ.toRingHom := InfinitePlace.mk_conjugate_eq σ.toRingHom
      rw [hm] at hp
      simp only [Bool.eq_false_iff.mpr hσ, Bool.false_eq_true, ↓reduceIte, hp,
        EmbeddingHodgeType.CM.conjugateEmbedding_involutive]

variable [NumberField E] [IsCMField E]

omit [IsCMField E] in
/-- A sum over embeddings groups into the two embeddings at each infinite place. -/
theorem TypeHodgeStructure.CMEmbeddingType.sum_embeddings_eq_sum_places (f : (E →ₐ[ℚ] ℂ) → ℂ) :
    ∑ σ, f σ = ∑ v : InfinitePlace E,
      (f (ε.placeEmbedding v) + f (EmbeddingHodgeType.CM.conjugateEmbedding (ε.placeEmbedding v))) := by
  rw [← Equiv.sum_comp ε.embeddingPairEquiv f, Fintype.sum_prod_type]
  simp only [Fintype.sum_bool, TypeHodgeStructure.CMEmbeddingType.embeddingPairEquiv, Equiv.coe_fn_mk, Bool.false_eq_true, ↓reduceIte]

/-- Transport from the mixed embedding space, using total complexity to eliminate real places. -/
def TypeHodgeStructure.CMEmbeddingType.mixedPeriodEquiv : mixedEmbedding.mixedSpace E ≃ₗ[ℝ] PeriodSpace.CMPeriodSpace E where
  toFun x v := ε.periodConjugation v (x.2 ⟨v, IsTotallyComplex.isComplex v⟩)
  invFun y := (fun _ => 0, fun w => ε.periodConjugation w.val (y w.val))
  left_inv x := by
    apply Prod.ext
    · funext w
      exact ((InfinitePlace.not_isReal_iff_isComplex.mpr
        (IsTotallyComplex.isComplex w.val)) w.property).elim
    · funext w
      exact ε.periodConjugation_involutive w.val (x.2 w)
  right_inv y := by
    funext v
    exact ε.periodConjugation_involutive v (y v)
  map_add' x y := by
    funext v
    exact map_add (ε.periodConjugation v) _ _
  map_smul' r x := by
    funext v
    exact map_smul (ε.periodConjugation v) r _

theorem TypeHodgeStructure.CMEmbeddingType.mixedPeriodEquiv_mixedEmbedding (x : E) :
    ε.mixedPeriodEquiv (mixedEmbedding E x) = ε.periodEmbedding x := by
  funext v
  exact ε.periodConjugation_embedding v x

/-- The real-linear equivalence is continuous in the usual finite-dimensional topologies. -/
def TypeHodgeStructure.CMEmbeddingType.mixedPeriodContinuousEquiv : mixedEmbedding.mixedSpace E ≃L[ℝ] PeriodSpace.CMPeriodSpace E :=
  ε.mixedPeriodEquiv.toContinuousLinearEquiv

/-- The lattice transported to the selected embedding space. -/
def TypeHodgeStructure.CMEmbeddingType.periodIntegerLattice : Submodule ℤ (PeriodSpace.CMPeriodSpace E) :=
  ZLattice.comap ℝ (mixedEmbedding.integerLattice E)
    ε.mixedPeriodContinuousEquiv.symm.toLinearMap

instance TypeHodgeStructure.CMEmbeddingType.periodIntegerLattice_discrete : DiscreteTopology ε.periodIntegerLattice := by
  unfold TypeHodgeStructure.CMEmbeddingType.periodIntegerLattice
  infer_instance

instance TypeHodgeStructure.CMEmbeddingType.periodIntegerLattice_isZLattice : IsZLattice ℝ ε.periodIntegerLattice := by
  unfold TypeHodgeStructure.CMEmbeddingType.periodIntegerLattice
  infer_instance

/-- The transported lattice is precisely the image of the actual ring of integers. -/
theorem TypeHodgeStructure.CMEmbeddingType.mem_periodIntegerLattice (z : PeriodSpace.CMPeriodSpace E) :
    z ∈ ε.periodIntegerLattice ↔ ∃ a : 𝓞 E, ε.periodEmbedding (a : E) = z := by
  change (∃ a : 𝓞 E, mixedEmbedding E (a : E) = ε.mixedPeriodEquiv.symm z) ↔ _
  constructor
  · rintro ⟨a, ha⟩
    refine ⟨a, ?_⟩
    rw [← ε.mixedPeriodEquiv_mixedEmbedding, ha, LinearEquiv.apply_symm_apply]
  · rintro ⟨a, ha⟩
    refine ⟨a, ?_⟩
    apply ε.mixedPeriodEquiv.injective
    rw [ε.mixedPeriodEquiv_mixedEmbedding, ha, LinearEquiv.apply_symm_apply]

/-- The additive quotient by the full CM integer lattice. -/
abbrev TypeHodgeStructure.CMEmbeddingType.periodTorus := PeriodSpace.CMPeriodSpace E ⧸ ε.periodIntegerLattice.toAddSubgroup

/-- A fundamental parallelepiped maps onto the quotient, so the CM torus is compact. -/
instance TypeHodgeStructure.CMEmbeddingType.periodTorus_compact : CompactSpace ε.periodTorus := by
  apply isCompact_univ_iff.mp
  have hc := IsZLattice.isCompact_range_of_periodic ε.periodIntegerLattice
    (QuotientAddGroup.mk : PeriodSpace.CMPeriodSpace E → ε.periodTorus)
    QuotientAddGroup.continuous_mk (by
      intro z w hw
      rw [QuotientAddGroup.mk_add, (QuotientAddGroup.eq_zero_iff w).mpr hw, add_zero])
  rwa [Set.range_eq_univ.mpr QuotientAddGroup.mk_surjective] at hc

instance TypeHodgeStructure.CMEmbeddingType.periodTorus_t2 : T2Space ε.periodTorus := by
  let : DiscreteTopology ε.periodIntegerLattice.toAddSubgroup :=
    ε.periodIntegerLattice_discrete
  let : IsClosed (ε.periodIntegerLattice.toAddSubgroup : Set (PeriodSpace.CMPeriodSpace E)) :=
    AddSubgroup.isClosed_of_discreteTopology
  infer_instance

end

end OAI.AlgebraicGeometry.ComplexMultiplication.CM