This PR adds support for constructors to the (WIP) `grind` tactic. When
merging equivalence classes, `grind` checks for equalities between
constructors. If they are distinct, it closes the goal; if they are the
same, it applies injectivity.
This PR adds support for compact congruence proofs in the (WIP) `grind`
tactic. The `mkCongrProof` function now verifies whether the congruence
proof can be constructed using only `congr`, `congrFun`, and `congrArg`,
avoiding the need to generate the more complex `hcongr` auxiliary
theorems.
This PR improves bv_decide's performance in the presence of large
literals.
The core change of this PR is the reformulation of the reflection code
for literals to:
```diff
def eval (assign : Assignment) : BVExpr w → BitVec w
| .var idx =>
- let ⟨bv⟩ := assign.get idx
- bv.truncate w
+ let packedBv := assign.get idx
+ /-
+ This formulation improves performance, as in a well formed expression the condition always holds
+ so there is no need for the more involved `BitVec.truncate` logic.
+ -/
+ if h : packedBv.w = w then
+ h ▸ packedBv.bv
+ else
+ packedBv.bv.truncate w
```
The remainder is merely further simplifications that make the terms
smaller and easier to deal with in general. This change is motivated by
applying the following diff to the kernel:
```diff
diff --git a/src/kernel/type_checker.cpp b/src/kernel/type_checker.cpp
index b0e6844dca..f13bb96bd4 100644
--- a/src/kernel/type_checker.cpp
+++ b/src/kernel/type_checker.cpp
@@ -518,6 +518,7 @@ optional<constant_info> type_checker::is_delta(expr const & e) const {
optional<expr> type_checker::unfold_definition_core(expr const & e) {
if (is_constant(e)) {
if (auto d = is_delta(e)) {
+// std::cout << "Working on unfolding: " << d->get_name() << std::endl;
if (length(const_levels(e)) == d->get_num_lparams()) {
if (m_diag) {
m_diag->record_unfold(d->get_name());
```
and observing that in the test case from #6043 we see a long series of
```
Working on unfolding: Bool.decEq
Working on unfolding: Bool.decEq.match_1
Working on unfolding: Bool.casesOn
Working on unfolding: Nat.ble
Working on unfolding: Nat.brecOn
Working on unfolding: Nat.beq.match_1
Working on unfolding: Nat.casesOn
Working on unfolding: Nat.casesOn
Working on unfolding: Nat.beq.match_1
Working on unfolding: Nat.casesOn
Working on unfolding: Nat.casesOn
```
the chain begins with `BitVec.truncate`, works through a few
abstractions and then continues like above forever, so I avoid the call
to truncate like this. It is not quite clear to me why removing `ofBool`
helps so much here, maybe some other kernel heuristic kicks in to rescue
us.
Either way this diff is a general improvement for reflection of `BitVec`
constants as we should never have to run `BitVec.truncate` again!
Fixes: #6043
This PR adds support for detecting congruent terms in the (WIP) `grind`
tactic. It also introduces the `grind.debug` option, which, when set to
`true`, checks many invariants after each equivalence class is merged.
This option is intended solely for debugging purposes.
This PR adds a custom type and instance canonicalizer for the (WIP)
`grind` tactic. The `grind` tactic uses congruence closure but
disregards types, type formers, instances, and proofs. Proofs are
ignored due to proof irrelevance. Types, type formers, and instances are
considered supporting elements and are not factored into congruence
detection. Instead, `grind` only checks whether elements are
structurally equal, which, in the context of the `grind` tactic, is
equivalent to pointer equality. See new tests for examples where the
canonicalizer is important.
This PR adds an explanation to the error message when `cases` and
`induction` are applied to a term whose type is not an inductive type.
For `Prop`, these tactics now suggest the `by_cases` tactic. Example:
```
tactic 'cases' failed, major premise type is not an inductive type
Prop
Explanation: the 'cases' tactic is for constructor-based reasoning as well as for applying
custom cases principles with a 'using' clause or a registered '@[cases_eliminator]' theorem.
The above type neither is an inductive type nor has a registered theorem.
Consider using the 'by_cases' tactic, which does true/false reasoning for propositions.
```
[Zulip
discussion](https://leanprover.zulipchat.com/#narrow/channel/270676-lean4/topic/Improving.20the.20error.20for.20.60cases.20p.60.20when.20.60p.60.20is.20a.20proposition/near/488882682)
This PR ensures that `simp` and `dsimp` do not unfold definitions that
are not intended to be unfolded by the user. See issue #5755 for an
example affected by this issue.
Closes#5755
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
This PR adds the predicate `Expr.fvarsSet a b`, which returns `true` if
and only if the free variables in `a` are a subset of the free variables
in `b`.
This PR adds a new preprocessing step to the `grind` tactic:
universe-level normalization. The goal is to avoid missing equalities in
the congruence closure module.
This PR fixes a bug in `Lean.Meta.Closure` that would introduce
under-applied delayed assignment metavariables, which would keep them
from ever getting instantiated. This bug affected `match` elaboration
when the expected type contained postponed elaboration problems, for
example tactic blocks.
Closes#5925, closes#6354
This PR introduces the following features to the WIP `grind` tactic:
- `Expr` internalization.
- Congruence theorem cache.
- Procedure for adding new facts
- New tracing options
- New preprocessing steps: fold projections and eliminate dangling
`Expr.mdata`
This PR adds reserved names for congruence theorems used in the
simplifier and `grind` tactics. The idea is prevent the same congruence
theorems to be generated over and over again.
After update stage0, we must use the new API in the simplifier.
This PR fixes a regression where goals that don't exist were being
displayed. The regression was triggered by #5835 and originally caused
by #4926.
Bug originally reported at
https://leanprover.zulipchat.com/#narrow/channel/270676-lean4/topic/tactic.20doesn't.20change.20primary.20goal.20state/near/488957772.
The cause of this issue was that #5835 made certain `SourceInfo`s
canonical, which was directly transferred to several `TacticInfo`s by
#4926. The goal state selection mechanism would then pick up these extra
`TacticInfo`s.
The approach taken by this PR is to ensure that the `SourceInfo` that is
being transferred by #4926 is noncanonical.
This PR adds support for erasure of `Decidable.decide` to the new code
generator. It also adds a new `Probe.runOnDeclsNamed` function, which is
helpful for writing targeted single-file tests of compiler internals.
---------
Co-authored-by: Cameron Zwarich <cameron@lean-fro.org>
This PR replaces `List.lt` with `List.Lex`, from Mathlib, and adds the
new `Bool` valued lexicographic comparatory function `List.lex`. This
subtly changes the definition of `<` on Lists in some situations.
`List.lt` was a weaker relation: in particular if `l₁ < l₂`, then
`a :: l₁ < b :: l₂` may hold according to `List.lt` even if `a` and `b`
are merely incomparable
(either neither `a < b` nor `b < a`), whereas according to `List.Lex`
this would require `a = b`.
When `<` is total, in the sense that `¬ · < ·` is antisymmetric, then
the two relations coincide.
Mathlib was already overriding the order instances for `List α`,
so this change should not be noticed by anyone already using Mathlib.
We simultaneously add the boolean valued `List.lex` function,
parameterised by a `BEq` typeclass
and an arbitrary `lt` function. This will support the flexibility
previously provided for `List.lt`,
via a `==` function which is weaker than strict equality.
This PR ensures the new code generator produces code for `opaque`
definitions that are not tagged as `@[extern]`.
Remark: This is the behavior of the old code generator.
This PR adds the `--error=kind` option (shorthand: `-Ekind`) to the
`lean` CLI. When set, messages of `kind` (e.g.,
`linter.unusedVariables`) will be reported as errors. This setting does
nothing in interactive contexts (e.g., the server).
Closes#5194.
The spelling `--error` was chosen instead of the common `-Werror` both
for practical and behavioral reasons. Behaviorally, this option effects
not just warnings, but informational messages as well. Practically,
`-Werror` conflicts with the existing `-W` option for the worker and
`lean` also does not currently use long single-hyphen option names.
This PR ensures that the configuration in `Simp.Config` is used when
reducing terms and checking definitional equality in `simp`.
closes#5455
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
This PR fixes a bug in the simplifier. It was producing terms with loose
bound variables when eliminating unused `let_fun` expressions.
This issue was affecting the example at #6374. The example is now timing
out.
This PR makes it harder to create "fake" theorems about definitions that
are stubbed-out with `sorry` by ensuring that each `sorry` is not
definitionally equal to any other. For example, this now fails:
```lean
example : (sorry : Nat) = sorry := rfl -- fails
```
However, this still succeeds, since the `sorry` is a single
indeterminate `Nat`:
```lean
def f (n : Nat) : Nat := sorry
example : f 0 = f 1 := rfl -- succeeds
```
One can be more careful by putting parameters to the right of the colon:
```lean
def f : (n : Nat) → Nat := sorry
example : f 0 = f 1 := rfl -- fails
```
Most sources of synthetic sorries (recall: a sorry that originates from
the elaborator) are now unique, except for elaboration errors, since
making these unique tends to cause a confusing cascade of errors. In
general, however, such sorries are labeled. This enables "go to
definition" on `sorry` in the Infoview, which brings you to its origin.
The option `set_option pp.sorrySource true` causes the pretty printer to
show source position information on sorries.
**Details:**
* Adds `Lean.Meta.mkLabeledSorry`, which creates a sorry that is labeled
with its source position. For example, `(sorry : Nat)` might elaborate
to
```
sorryAx (Lean.Name → Nat) false
`lean.foo.12.8.12.13.8.13._sorry._@.lean.foo._hyg.153
```
It can either be made unique (like the above) or merely labeled. Labeled
sorries use an encoding that does not impact defeq:
```
sorryAx (Unit → Nat) false (Function.const Lean.Name ()
`lean.foo.14.7.13.7.13.69._sorry._@.lean.foo._hyg.174)
```
* Makes the `sorry` term, the `sorry` tactic, and every elaboration
failure create labeled sorries. Most are unique sorries, but some
elaboration errors are labeled sorries.
* Renames `OmissionInfo` to `DelabTermInfo` and adds configuration
options to control LSP interactions. One field is a source position to
use for "go to definition". This is used to implement "go to definition"
on labeled sorries.
* Makes hovering over a labeled `sorry` show something friendlier than
that full `sorryAx` expression. Instead, the first hover shows the
simplified ``sorry `«lean.foo:48:11»``. Hovering over that hover shows
the full `sorryAx`. Setting `set_option pp.sorrySource true` makes
`sorry` always start with printing with this source position
information.
* Removes `Lean.Meta.mkSyntheticSorry` in favor of `Lean.Meta.mkSorry`
and `Lean.Meta.mkLabeledSorry`.
* Changes `sorryAx` so that the `synthetic` argument is no longer
optional.
* Gives `addPPExplicitToExposeDiff` awareness of labeled sorries. It can
set `pp.sorrySource` when source positions differ.
* Modifies the delaborator framework so that delaborators can set Info
themselves without it being overwritten.
Incidentally closes#4972.
Inspired by [this Zulip
thread](https://leanprover.zulipchat.com/#narrow/channel/287929-mathlib4/topic/Is.20a.20.60definition_wanted.60.20keyword.20possible.3F/near/477260277).
