2026-06-09 20:14:53 -07:00
|
|
|
|
/-
|
|
|
|
|
|
Port of `Analysis/Forward/Lattices.agda`.
|
|
|
|
|
|
|
|
|
|
|
|
The Agda module instantiates `Lattice.FiniteMap` twice (variables ↦ abstract
|
|
|
|
|
|
values, states ↦ variable maps) and re-exports everything with ᵛ/ᵐ suffixes.
|
|
|
|
|
|
In Lean the two instantiations are `abbrev`s and the FiniteMap API is used
|
|
|
|
|
|
directly; the module parameters (the finite-height lattice `L`, the program)
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
become section variables, with the finite-height structure and the lattice
|
|
|
|
|
|
interpretation arriving by instance resolution as in Agda.
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
Correspondence:
|
|
|
|
|
|
VariableValues, StateVariables ↦ VariableValues, StateVariables
|
|
|
|
|
|
isLatticeᵛ/isLatticeᵐ, ⊔ᵛ, ≼ᵛ … ↦ (the FiniteMap Lattice instances)
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
fixedHeightᵛ, fixedHeightᵐ ↦ (the FiniteMap FiniteHeightLattice instance)
|
2026-06-09 20:14:53 -07:00
|
|
|
|
⊥ᵛ, ⊥ᵛ-contains-bottoms ↦ botV, FiniteMap.bot_contains_bots
|
|
|
|
|
|
states-in-Map ↦ states_memKey
|
|
|
|
|
|
variablesAt ↦ variablesAt
|
|
|
|
|
|
variablesAt-∈ ↦ variablesAt_mem
|
|
|
|
|
|
variablesAt-≈ ↦ (congruence, trivial with `=`)
|
|
|
|
|
|
joinForKey, joinForKey-Mono ↦ joinForKey, joinForKey_mono
|
|
|
|
|
|
joinAll, joinAll-Mono,
|
|
|
|
|
|
joinAll-k∈ks-≡ ↦ joinAll, joinAll_mono, joinAll_mem_eq
|
|
|
|
|
|
variablesAt-joinAll ↦ variablesAt_joinAll
|
|
|
|
|
|
⟦_⟧ᵛ ↦ interpV
|
|
|
|
|
|
⟦⊥ᵛ⟧ᵛ∅ ↦ interpV_botV_nil
|
|
|
|
|
|
⟦⟧ᵛ-respects-≈ᵛ ↦ (trivial with `=`)
|
|
|
|
|
|
⟦⟧ᵛ-⊔ᵛ-∨ ↦ interpV_sup
|
|
|
|
|
|
⟦⟧ᵛ-foldr ↦ interpV_foldr
|
|
|
|
|
|
-/
|
|
|
|
|
|
import Spa.Language
|
|
|
|
|
|
import Spa.Lattice.FiniteMap
|
|
|
|
|
|
|
|
|
|
|
|
namespace Spa
|
|
|
|
|
|
|
|
|
|
|
|
variable (L : Type) [Lattice L] (prog : Program)
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `VariableValues`. -/
|
|
|
|
|
|
abbrev VariableValues : Type := FiniteMap String L prog.vars
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `StateVariables`. -/
|
|
|
|
|
|
abbrev StateVariables : Type := FiniteMap prog.State (VariableValues L prog) prog.states
|
|
|
|
|
|
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
/-- Agda: `⊥ᵛ` (the bottom of `fixedHeightᵛ`, now found by instance search). -/
|
|
|
|
|
|
def botV [FiniteHeightLattice L] : VariableValues L prog :=
|
|
|
|
|
|
FiniteHeightLattice.bot (VariableValues L prog)
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
variable {L prog}
|
|
|
|
|
|
|
|
|
|
|
|
omit [Lattice L] in
|
|
|
|
|
|
/-- Agda: `states-in-Map`. -/
|
|
|
|
|
|
theorem states_memKey (s : prog.State) (sv : StateVariables L prog) :
|
|
|
|
|
|
FiniteMap.MemKey s sv :=
|
|
|
|
|
|
FiniteMap.memKey_iff.mpr (prog.states_complete s)
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `variablesAt`. -/
|
|
|
|
|
|
def variablesAt (s : prog.State) (sv : StateVariables L prog) :
|
|
|
|
|
|
VariableValues L prog :=
|
|
|
|
|
|
(FiniteMap.locate (states_memKey s sv)).1
|
|
|
|
|
|
|
|
|
|
|
|
omit [Lattice L] in
|
|
|
|
|
|
/-- Agda: `variablesAt-∈`. -/
|
|
|
|
|
|
theorem variablesAt_mem (s : prog.State) (sv : StateVariables L prog) :
|
|
|
|
|
|
(s, variablesAt s sv) ∈ sv :=
|
|
|
|
|
|
(FiniteMap.locate (states_memKey s sv)).2
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `m₁≼m₂⇒m₁[k]ᵐ≼m₂[k]ᵐ`, specialized the way `Forward.agda` uses it. -/
|
|
|
|
|
|
theorem variablesAt_le {sv₁ sv₂ : StateVariables L prog} (hle : sv₁ ≤ sv₂)
|
|
|
|
|
|
(s : prog.State) : variablesAt s sv₁ ≤ variablesAt s sv₂ :=
|
|
|
|
|
|
FiniteMap.le_of_mem_mem prog.states_nodup hle
|
|
|
|
|
|
(variablesAt_mem s sv₁) (variablesAt_mem s sv₂)
|
|
|
|
|
|
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
variable [FiniteHeightLattice L]
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
/-- Agda: `joinForKey`. -/
|
|
|
|
|
|
def joinForKey (k : prog.State) (sv : StateVariables L prog) :
|
|
|
|
|
|
VariableValues L prog :=
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
(sv.valuesAt (prog.incoming k)).foldr (· ⊔ ·) (botV L prog)
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
/-- Agda: `joinForKey-Mono`. -/
|
|
|
|
|
|
theorem joinForKey_mono (k : prog.State) :
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
Monotone (joinForKey (L := L) k) := by
|
2026-06-09 20:14:53 -07:00
|
|
|
|
intro sv₁ sv₂ hle
|
|
|
|
|
|
exact foldr_mono _ (FiniteMap.valuesAt_le hle (prog.incoming k)) (le_refl _)
|
|
|
|
|
|
(fun b _ _ hab => sup_le_sup_right hab b)
|
|
|
|
|
|
(fun a _ _ hab => sup_le_sup_left hab a)
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `joinAll` (the "Exercise 4.26" generalized update with `f = id`). -/
|
|
|
|
|
|
def joinAll (sv : StateVariables L prog) : StateVariables L prog :=
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
FiniteMap.generalizedUpdate id joinForKey prog.states sv
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
/-- Agda: `joinAll-Mono`. -/
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
theorem joinAll_mono : Monotone (joinAll (L := L) (prog := prog)) :=
|
|
|
|
|
|
FiniteMap.generalizedUpdate_monotone monotone_id joinForKey_mono
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
/-- Agda: `joinAll-k∈ks-≡`. -/
|
|
|
|
|
|
theorem joinAll_mem_eq {s : prog.State} {vs : VariableValues L prog}
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
{sv : StateVariables L prog} (h : (s, vs) ∈ joinAll sv) :
|
|
|
|
|
|
vs = joinForKey s sv :=
|
2026-06-09 20:14:53 -07:00
|
|
|
|
FiniteMap.generalizedUpdate_mem_eq (prog.states_complete s) h
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `variablesAt-joinAll`. -/
|
|
|
|
|
|
theorem variablesAt_joinAll (s : prog.State) (sv : StateVariables L prog) :
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
variablesAt s (joinAll sv) = joinForKey s sv :=
|
|
|
|
|
|
joinAll_mem_eq (variablesAt_mem s (joinAll sv))
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
/-! ### Lifting an interpretation to variable maps -/
|
|
|
|
|
|
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
variable [I : LatticeInterpretation L]
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
omit [FiniteHeightLattice L] in
|
2026-06-09 20:14:53 -07:00
|
|
|
|
/-- Agda: `⟦_⟧ᵛ`. -/
|
|
|
|
|
|
def interpV (vs : VariableValues L prog) (ρ : Env) : Prop :=
|
|
|
|
|
|
∀ (k : String) (l : L), (k, l) ∈ vs →
|
|
|
|
|
|
∀ (v : Value), Env.Mem (k, v) ρ → I.interp l v
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `⟦⊥ᵛ⟧ᵛ∅`. -/
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
theorem interpV_botV_nil : interpV (botV L prog) [] := by
|
2026-06-09 20:14:53 -07:00
|
|
|
|
intro k l _ v hmem
|
|
|
|
|
|
cases hmem
|
|
|
|
|
|
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
omit [FiniteHeightLattice L] in
|
2026-06-09 20:14:53 -07:00
|
|
|
|
/-- Agda: `⟦⟧ᵛ-⊔ᵛ-∨`. -/
|
|
|
|
|
|
theorem interpV_sup {vs₁ vs₂ : VariableValues L prog} {ρ : Env}
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
(h : interpV vs₁ ρ ∨ interpV vs₂ ρ) : interpV (vs₁ ⊔ vs₂) ρ := by
|
2026-06-09 20:14:53 -07:00
|
|
|
|
intro k l hmem v hv
|
|
|
|
|
|
obtain ⟨l₁, l₂, rfl, h₁, h₂⟩ := FiniteMap.mem_sup hmem
|
|
|
|
|
|
rcases h with h | h
|
|
|
|
|
|
· exact I.interp_sup v (Or.inl (h _ _ h₁ _ hv))
|
|
|
|
|
|
· exact I.interp_sup v (Or.inr (h _ _ h₂ _ hv))
|
|
|
|
|
|
|
|
|
|
|
|
/-- Agda: `⟦⟧ᵛ-foldr`. -/
|
|
|
|
|
|
theorem interpV_foldr {vs : VariableValues L prog}
|
|
|
|
|
|
{vss : List (VariableValues L prog)} {ρ : Env}
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
(hvs : interpV vs ρ) (hmem : vs ∈ vss) :
|
|
|
|
|
|
interpV (vss.foldr (· ⊔ ·) (botV L prog)) ρ := by
|
2026-06-09 20:14:53 -07:00
|
|
|
|
induction vss with
|
|
|
|
|
|
| nil => cases hmem
|
|
|
|
|
|
| cons vs' vss' ih =>
|
|
|
|
|
|
rcases List.mem_cons.mp hmem with rfl | hmem'
|
Lean migration: typeclass-based parameter passing, as in the Agda original
The port had flattened Agda's instance arguments ({{flA}}, {{evaluator}},
{{latticeInterpretation}}, {{validEvaluator}}) into explicitly threaded
values (fhL, E, I, hE). Restore them as typeclasses:
- Spa.FiniteHeightLattice: now actually used — Fixedpoint takes the
instance instead of a FixedHeight value; FiniteMap gets the missing
instance (height = ks.length * height B), so varsFixedHeight /
statesFixedHeight / signFixedHeight / constFixedHeight plumbing
disappears (instance bottoms are defeq to the old ones)
- Spa.Analysis.Forward.Evaluation: StmtEvaluator/ExprEvaluator become
classes; the Valid* Props become Prop-classes, as in Agda
- Spa.Analysis.Forward.Adapters: the expr→stmt adapter and its validity
are instances (Agda: the ExprToStmtAdapter instances)
- LatticeInterpretation is a class; sign/const interpretations,
evaluators and validity proofs are instances; use sites read like the
Agda module applications: result SignLattice prog
Proof simplifications (same theorems, proofs factored):
- Spa.Lattice.AboveBelow.monotone₂_of_strict: any ⊥-strict/⊤-dominated
operation on a flat lattice is monotone — replaces the four near-
identical case bashes per analysis (postulates in Agda)
- Spa.Lattice.AboveBelow.interp_sup_of/interp_inf_of: the shared flat-
lattice interpretation case analysis, making interpSign_sup/inf and
interpConst_sup/inf one-liners
lake build green with zero warnings; lake exe spa output verified
byte-identical (diff) to the previous, Agda-verified output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-09 23:32:38 -07:00
|
|
|
|
· exact interpV_sup (Or.inl hvs)
|
|
|
|
|
|
· exact interpV_sup (Or.inr (ih hmem'))
|
2026-06-09 20:14:53 -07:00
|
|
|
|
|
|
|
|
|
|
end Spa
|