diff --git "a/leanbench_tasks.csv" "b/leanbench_tasks.csv" --- "a/leanbench_tasks.csv" +++ "b/leanbench_tasks.csv" @@ -1,4 +1,4 @@ -task_id,task_type,difficulty,difficulty_score,repo,pr_number,pr_title,pr_url,pr_labels,base_commit,merge_commit,files_changed,primary_file,problem_statement,golden_patch,loc_added,loc_deleted,loc,task_quality_score,statement_strategy,has_linked_issue,pr_body_length,files_changed_count,new_theorems_count,new_lemmas_count,new_definitions_count,proof_source,human_verified,lean_toolchain,docker_image,verification_command,timeout_seconds,verification_method,domain,created_at,created_by +task_id,task_type,difficulty,difficulty_score,repo,pr_number,pr_title,pr_url,pr_labels,base_commit,merge_commit,files_changed,primary_file,problem_statement,golden_patch,loc_added,loc_deleted,loc,task_quality_score,statement_strategy,has_linked_issue,pr_body_length,files_changed_count,new_theorems_count,new_lemmas_count,new_definitions_count,proof_source,human_verified,lean_toolchain,docker_image,verification_command,timeout_seconds,verification_method,domain,created_at,created_by LB-0001,pr_completion,easy,0.44,ImperialCollegeLondon/FLT,910,fix(CI): Blueprint test driver,https://github.com/ImperialCollegeLondon/FLT/pull/910,,ed4af4822818ad935393f983051e3bf27308e855,47c2d775b991e9f8dacf77a375b02e4c0edf613a,"FLTTest.lean,FLTTest/FLTTest.lean",FLTTest.lean,"# Task: fix(CI): Blueprint test driver ## Context @@ -19,7 +19,7 @@ diff --git a/FLTTest/FLTTest.lean b/FLTTest/FLTTest.lean --- /dev/null +++ b/FLTTest/FLTTest.lean @@ -0,0 +1 @@ -+import FLTTest.MathlibCompatibility",2,0,2,6.0,diff_derived,false,148,2,0,0,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-27T20:20:48Z,extraction_pipeline_v2 ++import FLTTest.MathlibCompatibility",2,0,2,6.0,diff_derived,False,148,2,0,0,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-28T12:39:47Z,extraction_pipeline_v2 LB-0002,pr_completion,easy,3.1,ImperialCollegeLondon/FLT,898,feat(Patching/Utils/CompactHausdorffRings): add partial proof that compact Hausdorff rings are totally disconnected,https://github.com/ImperialCollegeLondon/FLT/pull/898,,d346976c67e8437079ff1fd2b7878506e9ad0f37,e8c35753f741cf30ffc59dcec88b5ce4e36f37b7,"FLT.lean,FLT/Patching/Utils/AdicTopology.lean,FLT/Patching/Utils/CompactHausdorffRings.lean",FLT.lean,"# Task: feat(Patching/Utils/CompactHausdorffRings): add partial proof that compact Hausdorff rings are totally disconnected ## Context @@ -211,7 +211,7 @@ diff --git a/FLT/Patching/Utils/CompactHausdorffRings.lean b/FLT/Patching/Utils/ + rw [AddMonoidHom.range_eq_bot_iff.mp key, AddMonoidHom.zero_apply] + exact (one_smul R c).symm.trans (Subtype.ext_iff.mp key) + -+end CompactHausdorff",137,0,137,8.0,diff_derived,false,224,3,3,0,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:20:48Z,extraction_pipeline_v2 ++end CompactHausdorff",137,0,137,8.0,diff_derived,False,224,3,3,0,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:39:47Z,extraction_pipeline_v2 LB-0003,pr_completion,easy,1.6,ImperialCollegeLondon/FLT,887,fix: finiteness files in `4.29` bump,https://github.com/ImperialCollegeLondon/FLT/pull/887,,4554fdf0c9ff937384bda4692e603572c6adb7f0,504213abdafa05a5508d31f25b03dc9fc3c09e99,"FLT/AutomorphicForm/QuaternionAlgebra/FiniteDimensional.lean,FLT/DivisionAlgebra/Finiteness.lean,FLT/Mathlib/MeasureTheory/Haar/Extension.lean",FLT/AutomorphicForm/QuaternionAlgebra/FiniteDimensional.lean,"# Task: fix: finiteness files in `4.29` bump ## Context @@ -1099,7 +1099,7 @@ diff --git a/FLT/Mathlib/MeasureTheory/Haar/Extension.lean b/FLT/Mathlib/Measure + haar haar) haar hU] · apply ENNReal.nnreal_smul_ne_top apply ENNReal.mul_ne_top - · exact isCompact_univ.measure_ne_top",85,298,383,5.0,diff_derived,false,106,3,0,0,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:20:48Z,extraction_pipeline_v2 + · exact isCompact_univ.measure_ne_top",85,298,383,5.0,diff_derived,False,106,3,0,0,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:39:47Z,extraction_pipeline_v2 LB-0004,pr_completion,easy,1.06,ImperialCollegeLondon/FLT,886,fix: `4.29` fixes for `HaarChar.AdeleRing`,https://github.com/ImperialCollegeLondon/FLT/pull/886,,7c8ac5865d778d3c2eafd3fe7dfb83803d8dc41c,4554fdf0c9ff937384bda4692e603572c6adb7f0,FLT/HaarMeasure/HaarChar/AdeleRing.lean,FLT/HaarMeasure/HaarChar/AdeleRing.lean,"# Task: fix: `4.29` fixes for `HaarChar.AdeleRing` ## Files affected @@ -1208,7 +1208,7 @@ The patched repository must compile with `lake build`.","diff --git a/FLT/HaarMe +set_option backward.isDefEq.respectTransparency false in open TensorProduct.RightActions in /-- Right multiplication by an element of Bˣ on B ⊗ 𝔸_K does not scale additive - Haar measure.",22,21,43,6.0,diff_derived,false,0,1,0,0,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:20:48Z,extraction_pipeline_v2 + Haar measure.",22,21,43,6.0,diff_derived,False,0,1,0,0,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:39:47Z,extraction_pipeline_v2 LB-0005,pr_completion,medium,5.0,ImperialCollegeLondon/FLT,885,fix: `4.29` fixes for finite adele files,https://github.com/ImperialCollegeLondon/FLT/pull/885,,166dbe151303e083f2d18a2478f7d3bb218ee511,b30cfb53c2a30eb96fb19c2b62b670075dd44fd1,"FLT/DedekindDomain/Completion/BaseChange.lean,FLT/DedekindDomain/FiniteAdeleRing/BaseChange.lean,FLT/DedekindDomain/IntegralClosure.lean,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/NumberField/AdeleRing.lean",FLT/DedekindDomain/Completion/BaseChange.lean,"# Task: fix: `4.29` fixes for finite adele files ## Context @@ -2552,7 +2552,7 @@ diff --git a/FLT/NumberField/AdeleRing.lean b/FLT/NumberField/AdeleRing.lean +set_option backward.isDefEq.respectTransparency false in -- this isn't rfl. Explanation below example (x : L ⊗[K] 𝔸 K) : baseChangeEquiv K L x = baseChangeEquiv' K L x := by - induction x with",267,167,434,6.5,diff_derived,false,129,5,1,3,2,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:20:48Z,extraction_pipeline_v2 + induction x with",267,167,434,6.5,diff_derived,False,129,5,1,3,2,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:39:48Z,extraction_pipeline_v2 LB-0006,pr_completion,easy,2.4,ImperialCollegeLondon/FLT,883,fix: `WithZeroMulInt` issues for bump to 4.29.0,https://github.com/ImperialCollegeLondon/FLT/pull/883,,6e04ca51c224b2ef37db017617e56bf917858a68,df1bd1124322a770996e7cb67e747e6802c25ea8,"FLT/Mathlib/Topology/Algebra/Valued/WithZeroMulInt.lean,FLT/NumberField/Completion/Finite.lean",FLT/Mathlib/Topology/Algebra/Valued/WithZeroMulInt.lean,"# Task: fix: `WithZeroMulInt` issues for bump to 4.29.0 ## Files affected @@ -2691,7 +2691,7 @@ diff --git a/FLT/NumberField/Completion/Finite.lean b/FLT/NumberField/Completion + (v.valuedAdicCompletion_surjective) lemma NumberField.isCompactAdicCompletionIntegers : - IsCompact (v.adicCompletionIntegers K : Set (v.adicCompletion K)) := by",22,46,68,7.0,diff_derived,false,0,2,2,0,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:20:48Z,extraction_pipeline_v2 + IsCompact (v.adicCompletionIntegers K : Set (v.adicCompletion K)) := by",22,46,68,7.0,diff_derived,False,0,2,2,0,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:39:48Z,extraction_pipeline_v2 LB-0007,pr_completion,hard,7.4,ImperialCollegeLondon/FLT,882,fix: `WithAbs` issues in 4.29.0 bump,https://github.com/ImperialCollegeLondon/FLT/pull/882,,bd1033acf6742eb55c375c2f5a7fafb9e0ef6d25,5b0ecd9f4770fe3425cff1af11a8ad5bf1d7e243,"FLT.lean,FLT/Mathlib/Algebra/Algebra/Hom.lean,FLT/Mathlib/Analysis/Normed/Ring/WithAbs.lean,FLT/NumberField/Completion/Infinite.lean,FLT/NumberField/InfiniteAdeleRing.lean,FLT/NumberField/InfinitePlace/Dimension.lean,FLT/NumberField/InfinitePlace/WeakApproximation.lean",FLT.lean,"# Task: fix: `WithAbs` issues in 4.29.0 bump ## Context @@ -4200,8 +4200,275 @@ diff --git a/FLT/NumberField/InfinitePlace/WeakApproximation.lean b/FLT/NumberFi - · exact InfiniteAdeleRing.denseRange_algebraMap K - · exact continuous_pi (fun _ => continuous_apply _) - --end NumberField.InfinitePlace.Completion",210,972,1182,4.5,diff_derived,false,450,7,9,0,2,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:20:49Z,extraction_pipeline_v2 -LB-0008,pr_completion,easy,1.7,leanprover-community/mathlib4,36880,review: MessageData.ForExprs,https://github.com/leanprover-community/mathlib4/pull/36880,t-meta,35d565bba7a120c88be1e421e84ff70e8778dc3c,94acb1345e3767c673cd2cf656f86e25d5f4e6ec,Mathlib/Lean/MessageData/ForExprs.lean,Mathlib/Lean/MessageData/ForExprs.lean,"# Task: review: MessageData.ForExprs +-end NumberField.InfinitePlace.Completion",210,972,1182,4.5,diff_derived,False,450,7,9,0,2,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:39:48Z,extraction_pipeline_v2 +LB-0008,pr_completion,easy,2.4,leanprover-community/mathlib4,37059,review: #36709,https://github.com/leanprover-community/mathlib4/pull/37059,"ready-to-merge,t-meta",9450d934e6c23ce226e8ddfa16ebd28f9cab1df4,23a8036a7a7f35719b883ebf20f5fa25c4b2a65e,"Mathlib/Tactic/DefEqAbuse.lean,MathlibTest/DefEqAbuse.lean",Mathlib/Tactic/DefEqAbuse.lean,"# Task: review: #36709 + +## Context +This review suggests +- various improvements to the code structure + readability; pruning of unnecessary code +- using `withCurrHeartbeats` instead of setting max heartbeats to 0 (and nothing at all in the `elabCommand` case, which should effectively start heartbeats afresh anyway) +- not re-throwing `.internal` errors (does not touch unmodified sections of code that do the same) +- hovers on names in the logged message + message newline fixes +- another test, this time at the command level +- `Name.lt` instead of `.quickLt` for human-readable sorting + +## Files affected +- Mathlib/Tactic/DefEqAbuse.lean (+55/-62) +- MathlibTest/DefEqAbuse.lean (+22/-4) + +## Definitions to add +- `minimizeCandidates` in `Mathlib/Tactic/DefEqAbuse.lean`: Given a function `succeeds : Array Name → m Bool` and an initial array of `candidates` such +- `somewhere.` in `Mathlib/Tactic/DefEqAbuse.lean` + ```lean + instance somewhere. \n\n\ + ``` + +## Verification +The patched repository must compile with `lake build`.","diff --git a/Mathlib/Tactic/DefEqAbuse.lean b/Mathlib/Tactic/DefEqAbuse.lean +--- a/Mathlib/Tactic/DefEqAbuse.lean ++++ b/Mathlib/Tactic/DefEqAbuse.lean +@@ -362,48 +362,40 @@ def reportDefEqAbuse {m : Type → Type} [Monad m] [MonadLog m] [AddMessageConte + `backward.isDefEq.respectTransparency true` but succeeds with `false`.\n\ + The following isDefEq checks are the root causes of the failure:\n{failureList}"" + +-/-- Collect candidate semireducible definition names by transitively following definition values ++/-- Collect candidate semireducible definition names by transitively following declaration values + reachable from `roots`, up to a bounded depth. -/ + def collectCandidates (env : Environment) (roots : Array Name) : Array Name := Id.run do +- let mut visited : Std.HashSet Name := {} ++ let mut visited : NameSet := {} + let mut queue := roots +- for _ in List.range 6 do -- depth limit ++ let mut candidates : Array Name := #[] ++ for _ in (0 : Nat)...6 do -- depth limit + if queue.isEmpty then break + let current := queue + queue := #[] + for name in current do + if visited.contains name then continue + visited := visited.insert name + if let some info := env.find? name then +- if let some val := info.value? then +- for c in val.getUsedConstants do ++ if wasOriginallyDefn env name && getReducibilityStatusCore env name == .semireducible then ++ candidates := candidates.push name ++ -- Only `defs`, `theorem`s and `opaque`s can have values, and we only care about the first. ++ if let .defnInfo { value .. } := info then ++ for c in value.getUsedConstants do + if !visited.contains c then + queue := queue.push c +- let mut candidates : Array Name := #[] +- for name in visited.toArray do +- if getReducibilityStatusCore env name == .semireducible then +- if let some (.defnInfo _) := env.find? name then +- candidates := candidates.push name + return candidates + +-/-- Apply `@[implicit_reducible]` marks to the given constants. -/ +-private def markImplicitReducible {m : Type → Type} [Monad m] [MonadEnv m] +- (names : Array Name) : m Unit := do +- for name in names do setReducibilityStatus name .implicitReducible +- + /-- Temporarily mark constants as `@[implicit_reducible]`, run an action, then restore all state. + +-Both the environment (which carries the reducibility marks) and the full tactic state (metavar +-context, goals) are saved before the marks are applied and restored via `finally`, so this +-helper is self-contained: callers do not need additional save/restore wrappers. -/ ++The full tactic state (which includes the environment carrying recuibility marks) are saved before ++the marks are applied and restored via `finally`, so this helper is self-contained: callers do not ++need additional save/restore wrappers. -/ + def withTempImplicitReducible {α : Type} (names : Array Name) (k : TacticM α) : TacticM α := do + let s ← saveState +- let savedEnv ← getEnv + try +- markImplicitReducible names ++ for name in names do setReducibilityStatus name .implicitReducible + k + finally +- setEnv savedEnv + s.restore (restoreInfo := true) + + /-- Temporarily mark constants as `@[implicit_reducible]`, run an action, then restore all state. +@@ -414,11 +406,27 @@ def withTempImplicitReducibleCmd {α : Type} (names : Array Name) + (k : CommandElabM α) : CommandElabM α := do + let saved ← get + try +- markImplicitReducible names ++ for name in names do setReducibilityStatus name .implicitReducible + k + finally + set saved + ++/-- Given a function `succeeds : Array Name → m Bool` and an initial array of `candidates` such ++that `succeeds candidates` is `true` (returning `none` if this is not the case), tries removing ++elements from `candidates` in order such that the resulting array still `succeeds`. Sorts the ++result. ++ ++This minimization may not be unique. It also is only minimal under the assumption that removing a ++later element from the array cannot cause an earlier element to become removable. -/ ++def minimizeCandidates {m} [Monad m] (succeeds : Array Name → m Bool) (candidates : Array Name) : ++ m (Option (Array Name)) := do ++ unless ← succeeds candidates do return none ++ let mut minimal := candidates ++ for name in minimal do ++ let without := minimal.filter (· != name) ++ if ← succeeds without then minimal := without ++ return minimal.qsort Name.lt ++ + /-- Try to find a (possibly non-unique) minimal set of semireducible constants that, when marked + `@[implicit_reducible]`, make the tactic succeed with `backward.isDefEq.respectTransparency true`. + +@@ -429,20 +437,15 @@ def suggestAnnotationsTac (tac : Syntax) : TacticM (Option (Array Name)) := do + let goalType ← getMainTarget + let candidates := collectCandidates (← getEnv) goalType.getUsedConstants + if candidates.isEmpty then return none +- let tryWith (names : Array Name) : TacticM Bool := +- withTempImplicitReducible names do +- withTheReader Core.Context (fun c => { c with maxHeartbeats := 0 }) do +- withOptions (fun o => o.setBool `backward.isDefEq.respectTransparency true) do +- try evalTactic tac; pure true +- catch | .internal id ref => throw (.internal id ref) | _ => pure false +- -- Verify that marking ALL candidates fixes the issue. +- unless ← tryWith candidates do return none +- -- Minimize by greedy removal. +- let mut minimal := candidates +- for name in candidates do +- let without := minimal.filter (· != name) +- if ← tryWith without then minimal := without +- return some (minimal.qsort Name.quickLt) ++ let succeedsWith (names : Array Name) : TacticM Bool := ++ withTempImplicitReducible names do withCurrHeartbeats do ++ withOptions (·.setBool `backward.isDefEq.respectTransparency true) do ++ try ++ Core.resetMessageLog ++ Term.withoutErrToSorry <| evalTactic tac ++ notM MonadLog.hasErrors ++ catch _ => pure false ++ minimizeCandidates succeedsWith candidates + + /-- Try to find a (possibly non-unique) minimal set of semireducible constants that, when marked + `@[implicit_reducible]`, make the command succeed with `backward.isDefEq.respectTransparency true`. +@@ -456,44 +459,34 @@ def suggestAnnotationsCmd (cmd : Syntax) : CommandElabM (Option (Array Name)) := + let saved ← get + let roots ← try + withScope (fun scope => +- { scope with opts := (scope.opts.setBool `Elab.async false) ++ { scope with opts := scope.opts.setBool `Elab.async false + |>.setBool `backward.isDefEq.respectTransparency false }) do + elabCommand cmd + let newEnv ← getEnv +- let mut names : Array Name := #[] ++ let mut names : NameSet := {} + -- Collect constants from any new declarations added by this command +- for (name, _) in newEnv.constants.map₂.toList do +- if env.constants.map₂.find? name |>.isNone then +- if let some info := newEnv.find? name then +- names := names ++ info.type.getUsedConstants +- if let some val := info.value? then +- names := names ++ val.getUsedConstants ++ for (name, info) in newEnv.constants.map₂ do ++ if !env.constants.map₂.contains name then ++ names := names ∪ info.getUsedConstantsAsSet + return names +- catch _ => pure #[] ++ catch _ => pure {} + set saved +- let candidates := collectCandidates (← getEnv) roots ++ let candidates := collectCandidates (← getEnv) roots.toArray + if candidates.isEmpty then return none +- let tryWith (names : Array Name) : CommandElabM Bool := ++ let succeedsWith (names : Array Name) : CommandElabM Bool := + withTempImplicitReducibleCmd names do + withScope (fun scope => +- { scope with opts := ((scope.opts.setBool `Elab.async false) +- |>.setBool `backward.isDefEq.respectTransparency true) +- |>.insert `maxHeartbeats (.ofNat 0) }) do ++ { scope with opts := scope.opts.setBool `Elab.async false ++ |>.setBool `backward.isDefEq.respectTransparency true }) do + try + elabCommand cmd +- let hasErrors := (← get).messages.hasErrors +- return !hasErrors +- catch | .internal id ref => throw (.internal id ref) | _ => return false +- unless ← tryWith candidates do return none +- let mut minimal := candidates +- for name in candidates do +- let without := minimal.filter (· != name) +- if ← tryWith without then minimal := without +- return some (minimal.qsort Name.quickLt) ++ notM MonadLog.hasErrors ++ catch _ => return false ++ minimizeCandidates succeedsWith candidates + + /-- Format the annotation workaround as a Lean code snippet. -/ + def formatAnnotations (names : Array Name) : MessageData := +- let namesList := joinSep (names.toList.map fun n => m!"" {n}"") ""\n"" ++ let namesList := joinSep (names.toList.map (m!"" {.ofConstName · (fullNames := true)}"")) ""\n"" + m!""set_option allowUnsafeReducibility true\nattribute [implicit_reducible]\n{namesList}"" + + /-- Log annotation suggestions as info. -/ +@@ -502,8 +495,8 @@ def logAnnotationSuggestions {m : Type → Type} [Monad m] [MonadLog m] [AddMess + let some names := names | return + if names.isEmpty then return + logInfo m!""Workaround: the following `@[implicit_reducible]` annotations (a possibly \ +- non-unique minimal set) would paper over this problem,\n\ +- but the real issue is likely a leaky instance somewhere.\n\ ++ non-unique minimal set) would paper over this problem, but the real issue is likely a leaky \ ++ instance somewhere.\n\n\ + {formatAnnotations names}"" + + /-- +diff --git a/MathlibTest/DefEqAbuse.lean b/MathlibTest/DefEqAbuse.lean +--- a/MathlibTest/DefEqAbuse.lean ++++ b/MathlibTest/DefEqAbuse.lean +@@ -68,8 +68,8 @@ warning: #defeq_abuse: tactic fails with `backward.isDefEq.respectTransparency t + The following isDefEq checks are the root causes of the failure: + ❌️ (i : ℕ) → (fun a => Prop) i =?= MyPred ℕ + --- +-info: Workaround: the following `@[implicit_reducible]` annotations (a possibly non-unique minimal set) would paper over this problem, +-but the real issue is likely a leaky instance somewhere. ++info: Workaround: the following `@[implicit_reducible]` annotations (a possibly non-unique minimal set) would paper over this problem, but the real issue is likely a leaky instance somewhere. ++ + set_option allowUnsafeReducibility true + attribute [implicit_reducible] + MyPred +@@ -79,6 +79,24 @@ noncomputable example (s : MyPred ℕ) (a : ℕ) (ha : a ∉ s) : Disjoint s {a} + #defeq_abuse in rw [myPred_disjoint_singleton_right] + exact ha + ++/-- ++warning: #defeq_abuse: command fails with `backward.isDefEq.respectTransparency true` but succeeds with `false`. ++The following isDefEq checks are the root causes of the failure: ++ ❌️ MyPred ℕ =?= (i : ℕ) → (fun a => Prop) i ++ ❌️ (i : ℕ) → (fun a => Prop) i =?= MyPred ℕ ++--- ++info: Workaround: the following `@[implicit_reducible]` annotations (a possibly non-unique minimal set) would paper over this problem, but the real issue is likely a leaky instance somewhere. ++ ++set_option allowUnsafeReducibility true ++attribute [implicit_reducible] ++ MyPred ++-/ ++#guard_msgs in ++#defeq_abuse in ++noncomputable def fooDependingOnMyPred (s : MyPred ℕ) (a : ℕ) (ha : a ∉ s) : Disjoint s {a} := by ++ rw [myPred_disjoint_singleton_right] ++ exact ha ++ + end SetAliasAbuse + + section VirtualParentAbuse +@@ -196,8 +214,8 @@ warning: #defeq_abuse: tactic fails with `backward.isDefEq.respectTransparency t + The following isDefEq checks are the root causes of the failure: + ❌️ @ZoC.zo Int instZoCInt =?= @ZoC.zo Int (@GrC.toZoC Int ?m.11) + --- +-info: Workaround: the following `@[implicit_reducible]` annotations (a possibly non-unique minimal set) would paper over this problem, +-but the real issue is likely a leaky instance somewhere. ++info: Workaround: the following `@[implicit_reducible]` annotations (a possibly non-unique minimal set) would paper over this problem, but the real issue is likely a leaky instance somewhere. ++ + set_option allowUnsafeReducibility true + attribute [implicit_reducible] + zoDirectC",77,66,143,8.0,diff_derived,False,557,2,0,0,2,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-28T12:39:49Z,extraction_pipeline_v2 +LB-0009,pr_completion,easy,1.7,leanprover-community/mathlib4,36880,review: MessageData.ForExprs,https://github.com/leanprover-community/mathlib4/pull/36880,t-meta,35d565bba7a120c88be1e421e84ff70e8778dc3c,94acb1345e3767c673cd2cf656f86e25d5f4e6ec,Mathlib/Lean/MessageData/ForExprs.lean,Mathlib/Lean/MessageData/ForExprs.lean,"# Task: review: MessageData.ForExprs ## Context This review of #36796 suggests implementing `for (ppCtx, e) in msgs.exprs do` notation by letting the caller handle `ppCtx` themselves, via an approach patterned after `Syntax.topDown`. This also means we can avoid running `MetaM` when we're simply getting the expressions. @@ -4369,8 +4636,8 @@ The patched repository must compile with `lake build`.","diff --git a/Mathlib/Le + arr := arr.push e + return arr + -+end Lean.MessageData",71,37,108,8.5,diff_derived,false,729,1,0,0,1,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-27T20:20:55Z,extraction_pipeline_v2 -LB-0009,pr_completion,easy,3.4,leanprover-community/mathlib4,35843,review: just the basics,https://github.com/leanprover-community/mathlib4/pull/35843,"delegated,t-meta",be3f919ded3ff56ae14edfbbe776bb5542a929f0,93584a945344b3c995cac4249c4e1ae747998b02,"Mathlib/Lean/MessageData/Trace.lean,Mathlib/Tactic/DefEqAbuse.lean",Mathlib/Lean/MessageData/Trace.lean,"# Task: review: just the basics ++end Lean.MessageData",71,37,108,8.5,diff_derived,False,729,1,0,0,1,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-28T12:39:53Z,extraction_pipeline_v2 +LB-0010,pr_completion,easy,3.4,leanprover-community/mathlib4,35843,review: just the basics,https://github.com/leanprover-community/mathlib4/pull/35843,"delegated,t-meta",be3f919ded3ff56ae14edfbbe776bb5542a929f0,93584a945344b3c995cac4249c4e1ae747998b02,"Mathlib/Lean/MessageData/Trace.lean,Mathlib/Tactic/DefEqAbuse.lean",Mathlib/Lean/MessageData/Trace.lean,"# Task: review: just the basics ## Context This ultra-quick review of #35750 doesn't check for good behavior or maintainable code; instead, as per the [zulip thread](https://leanprover.zulipchat.com/#narrow/channel/113488-general/topic/backward.2EisDefEq.2ErespectTransparency/with/575687635), it merely @@ -4814,8 +5081,8 @@ diff --git a/Mathlib/Tactic/DefEqAbuse.lean b/Mathlib/Tactic/DefEqAbuse.lean + { scope with opts := scope.opts.setBool `backward.isDefEq.respectTransparency false }) do elabCommand cmd - end Mathlib.Tactic.DefEqAbuse",139,128,267,8.5,diff_derived,false,421,2,0,0,4,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-27T20:20:55Z,extraction_pipeline_v2 -LB-0010,pr_completion,medium,4.4,ImperialCollegeLondon/FLT,841,Complete `localcomponent_matrix` and `f_g_local_global`,https://github.com/ImperialCollegeLondon/FLT/pull/841,,e36185c7b9eaf3ecb0ce044a0536490f4007e0f1,712e8a20d2647895feed1cbf5518dce4056f58cb,"FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/Mathlib/Topology/Algebra/RestrictedProduct/TopologicalSpace.lean",FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,"# Task: Complete `localcomponent_matrix` and `f_g_local_global` + end Mathlib.Tactic.DefEqAbuse",139,128,267,8.5,diff_derived,False,421,2,0,0,4,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,general,2026-04-28T12:39:53Z,extraction_pipeline_v2 +LB-0011,pr_completion,medium,4.4,ImperialCollegeLondon/FLT,841,Complete `localcomponent_matrix` and `f_g_local_global`,https://github.com/ImperialCollegeLondon/FLT/pull/841,,e36185c7b9eaf3ecb0ce044a0536490f4007e0f1,712e8a20d2647895feed1cbf5518dce4056f58cb,"FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/Mathlib/Topology/Algebra/RestrictedProduct/TopologicalSpace.lean",FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,"# Task: Complete `localcomponent_matrix` and `f_g_local_global` ## Files affected - FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean (+112/-12) @@ -5058,8 +5325,8 @@ diff --git a/FLT/Mathlib/Topology/Algebra/RestrictedProduct/TopologicalSpace.lea + theorem Homeomorph.restrictedProductMatrix_aux {ι n : Type*} [Fintype n] {A : ι → Type*} [(i : ι) → TopologicalSpace (A i)] {C : (i : ι) → Set (A i)} - (i : ι) (hCopen : ∀ (i : ι), IsOpen (C i)) :",130,12,142,7.5,diff_derived,false,0,2,0,6,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:01Z,extraction_pipeline_v2 -LB-0011,pr_completion,medium,6.6,ImperialCollegeLondon/FLT,836,feat: FiniteAdeleRing.Aux.almost_always_bijOn,https://github.com/ImperialCollegeLondon/FLT/pull/836,awaiting-CI,20c7c4548a6a67fdcfb585a83a46b273f45843ca,799f7d54d7e3f096b3037189f9713abc28be81f0,"FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/Mathlib/LinearAlgebra/TensorProduct/Algebra.lean,FLT/Mathlib/RingTheory/DedekindDomain/FiniteAdeleRing.lean",FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,"# Task: feat: FiniteAdeleRing.Aux.almost_always_bijOn + (i : ι) (hCopen : ∀ (i : ι), IsOpen (C i)) :",130,12,142,7.5,diff_derived,False,0,2,0,6,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:01Z,extraction_pipeline_v2 +LB-0012,pr_completion,medium,6.6,ImperialCollegeLondon/FLT,836,feat: FiniteAdeleRing.Aux.almost_always_bijOn,https://github.com/ImperialCollegeLondon/FLT/pull/836,awaiting-CI,20c7c4548a6a67fdcfb585a83a46b273f45843ca,799f7d54d7e3f096b3037189f9713abc28be81f0,"FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/Mathlib/LinearAlgebra/TensorProduct/Algebra.lean,FLT/Mathlib/RingTheory/DedekindDomain/FiniteAdeleRing.lean",FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,"# Task: feat: FiniteAdeleRing.Aux.almost_always_bijOn ## Files affected - FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean (+198/-6) @@ -5444,8 +5711,8 @@ diff --git a/FLT/Mathlib/RingTheory/DedekindDomain/FiniteAdeleRing.lean b/FLT/Ma + variable [DecidableEq (HeightOneSpectrum R)] in /-- - `localIdempotent R K p` is the finite adele which is 1 at p and 0 elsewhere.",257,10,267,7.5,diff_derived,false,14,3,0,8,2,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:01Z,extraction_pipeline_v2 -LB-0012,pr_completion,easy,1.0,ImperialCollegeLondon/FLT,834,feat(HaarChar/FiniteAdeleRing): Fill in two sorries,https://github.com/ImperialCollegeLondon/FLT/pull/834,,52dd0fdbe9ccb9b3a874773c3bc44c8b946e3d46,20c7c4548a6a67fdcfb585a83a46b273f45843ca,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,"# Task: feat(HaarChar/FiniteAdeleRing): Fill in two sorries + `localIdempotent R K p` is the finite adele which is 1 at p and 0 elsewhere.",257,10,267,7.5,diff_derived,False,14,3,0,8,2,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:01Z,extraction_pipeline_v2 +LB-0013,pr_completion,easy,1.0,ImperialCollegeLondon/FLT,834,feat(HaarChar/FiniteAdeleRing): Fill in two sorries,https://github.com/ImperialCollegeLondon/FLT/pull/834,,52dd0fdbe9ccb9b3a874773c3bc44c8b946e3d46,20c7c4548a6a67fdcfb585a83a46b273f45843ca,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,"# Task: feat(HaarChar/FiniteAdeleRing): Fill in two sorries ## Context This PR fills in two sorres in `HaarChar/FiniteAdeleRing`. @@ -5510,8 +5777,8 @@ The patched repository must compile with `lake build`.","diff --git a/FLT/HaarMe + congr /-- left multiplication and right multiplication by a unit have the same Haar character - on `𝔸_K^f ⊗ B`. See also",34,6,40,6.0,diff_derived,false,58,1,0,0,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:01Z,extraction_pipeline_v2 -LB-0013,pr_completion,easy,1.98,ImperialCollegeLondon/FLT,835,"localcomponent_id, localcomponent_comp, IsCompact (v.adicCompletionIntegers K)",https://github.com/ImperialCollegeLondon/FLT/pull/835,,52dd0fdbe9ccb9b3a874773c3bc44c8b946e3d46,ba672bce4061ac4782b7d959d87020de5d59e60d,"FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/NumberField/Completion/Finite.lean",FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,"# Task: localcomponent_id, localcomponent_comp, IsCompact (v.adicCompletionIntegers K) + on `𝔸_K^f ⊗ B`. See also",34,6,40,6.0,diff_derived,False,58,1,0,0,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:02Z,extraction_pipeline_v2 +LB-0014,pr_completion,easy,1.98,ImperialCollegeLondon/FLT,835,"localcomponent_id, localcomponent_comp, IsCompact (v.adicCompletionIntegers K)",https://github.com/ImperialCollegeLondon/FLT/pull/835,,52dd0fdbe9ccb9b3a874773c3bc44c8b946e3d46,ba672bce4061ac4782b7d959d87020de5d59e60d,"FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,FLT/HaarMeasure/HaarChar/FiniteAdeleRing.lean,FLT/NumberField/Completion/Finite.lean",FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,"# Task: localcomponent_id, localcomponent_comp, IsCompact (v.adicCompletionIntegers K) ## Context The first 3 tasks in [#FLT > Outstanding tasks, V10](https://leanprover.zulipchat.com/#narrow/channel/416277-FLT/topic/Outstanding.20tasks.2C.20V10) @@ -5629,8 +5896,8 @@ diff --git a/FLT/NumberField/Completion/Finite.lean b/FLT/NumberField/Completion + lemma NumberField.isOpenAdicCompletionIntegers : IsOpen (v.adicCompletionIntegers K : Set (v.adicCompletion K)) := - Valued.isOpen_valuationSubring _",37,7,44,7.5,diff_derived,false,148,3,0,1,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:01Z,extraction_pipeline_v2 -LB-0014,pr_completion,easy,1.88,ImperialCollegeLondon/FLT,833,feat: state id and comp for localcomponent,https://github.com/ImperialCollegeLondon/FLT/pull/833,,5c2aa9fd9120d195a0fb0178fe221ac0bab617cd,52dd0fdbe9ccb9b3a874773c3bc44c8b946e3d46,FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,"# Task: feat: state id and comp for localcomponent + Valued.isOpen_valuationSubring _",37,7,44,7.5,diff_derived,False,148,3,0,1,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:02Z,extraction_pipeline_v2 +LB-0015,pr_completion,easy,1.88,ImperialCollegeLondon/FLT,833,feat: state id and comp for localcomponent,https://github.com/ImperialCollegeLondon/FLT/pull/833,,5c2aa9fd9120d195a0fb0178fe221ac0bab617cd,52dd0fdbe9ccb9b3a874773c3bc44c8b946e3d46,FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,"# Task: feat: state id and comp for localcomponent ## Files affected - FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean (+21/-13) @@ -5696,8 +5963,8 @@ The patched repository must compile with `lake build`.","diff --git a/FLT/Dedeki + simp [TensorProduct.localcomponent_id_apply] end FiniteAdeleRing - ",21,13,34,7.5,diff_derived,false,0,1,0,2,0,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:02Z,extraction_pipeline_v2 -LB-0015,pr_completion,hard,7.2,ImperialCollegeLondon/FLT,823,isCentralSimple_addHaarScalarFactor_left_mul_eq_right_mul,https://github.com/ImperialCollegeLondon/FLT/pull/823,awaiting-review,0b0f52f7516a7baab568686949b3ce3f952e444a,3c9d5604540b3b9f50857802d85ea6230103e764,"FLT.lean,FLT/DivisionAlgebra/Finiteness.lean,FLT/HaarMeasure/HaarChar/AdeleRing.lean,FLT/Mathlib/Algebra/Central/TensorProduct.lean,FLT/Mathlib/LinearAlgebra/Determinant.lean,FLT/NumberField/Completion/Finite.lean",FLT.lean,"# Task: isCentralSimple_addHaarScalarFactor_left_mul_eq_right_mul + ",21,13,34,7.5,diff_derived,False,0,1,0,2,0,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:02Z,extraction_pipeline_v2 +LB-0016,pr_completion,hard,7.2,ImperialCollegeLondon/FLT,823,isCentralSimple_addHaarScalarFactor_left_mul_eq_right_mul,https://github.com/ImperialCollegeLondon/FLT/pull/823,awaiting-review,0b0f52f7516a7baab568686949b3ce3f952e444a,3c9d5604540b3b9f50857802d85ea6230103e764,"FLT.lean,FLT/DivisionAlgebra/Finiteness.lean,FLT/HaarMeasure/HaarChar/AdeleRing.lean,FLT/Mathlib/Algebra/Central/TensorProduct.lean,FLT/Mathlib/LinearAlgebra/Determinant.lean,FLT/NumberField/Completion/Finite.lean",FLT.lean,"# Task: isCentralSimple_addHaarScalarFactor_left_mul_eq_right_mul ## Context Current progress: reduce to the finite adeles, i.e. instead of `isCentralSimple_addHaarScalarFactor_left_mul_eq_right_mul`, we now have `isCentralSimple_finite_addHaarScalarFactor_left_mul_eq_right_mul` in the same place, with `AdeleRing` replaced by `FiniteAdeleRing`. @@ -6244,8 +6511,8 @@ diff --git a/FLT/NumberField/Completion/Finite.lean b/FLT/NumberField/Completion + -- does this exist upstream? Should do. example (v : HeightOneSpectrum (𝓞 K)) : SecondCountableTopology (v.adicCompletion K) := - inferInstance",386,5,391,5.5,diff_derived,false,760,6,0,12,4,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:02Z,extraction_pipeline_v2 -LB-0016,pr_completion,hard,7.6,ImperialCollegeLondon/FLT,825,feat: API for finite adele ring tensor vector space,https://github.com/ImperialCollegeLondon/FLT/pull/825,awaiting-CI,68062db94b8fdbdc8d306532d7b0e99f4e0046b1,0b0f52f7516a7baab568686949b3ce3f952e444a,"FLT.lean,FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,FLT/Mathlib/LinearAlgebra/TensorProduct/Algebra.lean,FLT/Mathlib/LinearAlgebra/TensorProduct/FiniteFree.lean,FLT/Mathlib/RingTheory/DedekindDomain/FiniteAdeleRing.lean,FLT/Mathlib/Topology/Algebra/Module/TensorProduct.lean,FLT/Mathlib/Topology/Algebra/RestrictedProduct/Basic.lean,FLT/Mathlib/Topology/Algebra/RestrictedProduct/TopologicalSpace.lean",FLT.lean,"# Task: feat: API for finite adele ring tensor vector space + inferInstance",386,5,391,5.5,diff_derived,False,760,6,0,12,4,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:02Z,extraction_pipeline_v2 +LB-0017,pr_completion,hard,7.6,ImperialCollegeLondon/FLT,825,feat: API for finite adele ring tensor vector space,https://github.com/ImperialCollegeLondon/FLT/pull/825,awaiting-CI,68062db94b8fdbdc8d306532d7b0e99f4e0046b1,0b0f52f7516a7baab568686949b3ce3f952e444a,"FLT.lean,FLT/DedekindDomain/FiniteAdeleRing/TensorProduct.lean,FLT/Mathlib/LinearAlgebra/TensorProduct/Algebra.lean,FLT/Mathlib/LinearAlgebra/TensorProduct/FiniteFree.lean,FLT/Mathlib/RingTheory/DedekindDomain/FiniteAdeleRing.lean,FLT/Mathlib/Topology/Algebra/Module/TensorProduct.lean,FLT/Mathlib/Topology/Algebra/RestrictedProduct/Basic.lean,FLT/Mathlib/Topology/Algebra/RestrictedProduct/TopologicalSpace.lean",FLT.lean,"# Task: feat: API for finite adele ring tensor vector space ## Files affected - FLT.lean (+4/-0) @@ -6801,1525 +7068,1107 @@ diff --git a/FLT/Mathlib/Topology/Algebra/RestrictedProduct/TopologicalSpace.lea + +end eval + -+end RestrictedProduct",409,4,413,4.5,diff_derived,false,0,8,0,10,4,pr_merge,false,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-27T20:21:03Z,extraction_pipeline_v2 -LB-0017,pr_completion,easy,2.08,google-deepmind/formal-conjectures,3744,feat(Wikipedia/EulerBrick): proof for `cuboidOne`,https://github.com/google-deepmind/formal-conjectures/pull/3744,wikipedia,42bece60b362c401b7d8abe318e0b6d9aecaa646,92302ba152fc7c9dcca397e98e72f87987fbb952,FormalConjectures/Wikipedia/EulerBrick.lean,FormalConjectures/Wikipedia/EulerBrick.lean,"# Task: feat(Wikipedia/EulerBrick): proof for `cuboidOne` ++end RestrictedProduct",409,4,413,4.5,diff_derived,False,0,8,0,10,4,pr_merge,False,leanprover/lean4:v4.29.0,leanbench/lean4-env:4.29.0,lake build,600,lean_compiler,algebra,2026-04-28T12:40:02Z,extraction_pipeline_v2 +LB-0018,pr_completion,easy,3.5,leanprover/lean4,13301,feat: add `try? => tac` syntax and parallel cancellation test,https://github.com/leanprover/lean4/pull/13301,"builds-mathlib,toolchain-available,changelog-tactics,mathlib4-nightly-available",ec727859274c18a82f26dd66028e36f380fdd35c,432d11541ba76fb474d5817769f8a9b9a5fd5c4c,"src/Init/Try.lean,src/Lean/Elab/Tactic/Try.lean,src/Lean/Server/Test/Cancel.lean,tests/elab/try_eval_suggest.lean,tests/server_interactive/cancellation_par.lean",src/Init/Try.lean,"# Task: feat: add `try? => tac` syntax and parallel cancellation test + +## Context +This PR adds a `try? => tac` syntax that runs `evalSuggest` directly on a given tactic, useful for testing the `try?` machinery in isolation. It also adds a server_interactive test (`cancellation_par.lean`) that demonstrates a cancellation bug with parallel tactic combinators. + +The test contrasts three combinators: +- **`first`** (sequential): cancellation works correctly — the tactic runs on the main elaboration thread and shares its cancel token. +- **`attempt_all_par`** (parallel): cancellation is broken — the subtask spawned via `asTask` gets a fresh cancel token that is never set on re-elaboration. +- **`first_par`** (parallel): same bug as `attempt_all_par`. + +The test uses a `check_cancel