Users have requested toolchain tags on `lean4-cli`, so let's add it to
the release checklist to make sure these get added regularly.
Previously, `lean4-cli` has used more complicated tags, but going
forward we're going to just use the simple `v4.16.0` style tags, with no
repository-specific versioning.
---------
Co-authored-by: Markus Himmel <markus@lean-fro.org>
This PR adds a `toFin` and `msb` lemma for unsigned bitvector modulus.
Similar to #6402, we don't provide a general `toInt_umod` lemmas, but
instead choose to provide more specialized rewrites, with extra
side-conditions.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
This PR adds a `toFin` and `msb` lemma for unsigned bitvector division.
We *don't* have `toInt_udiv`, since the only truly general statement we
can make does no better than unfolding the definition, and it's not
uncontroversially clear how to unfold `toInt` (see
`toInt_eq_msb_cond`/`toInt_eq_toNat_cond`/`toInt_eq_toNat_bmod` for a
few options currently provided). Instead, we do have `toInt_udiv_of_msb`
that's able to provide a more meaningful rewrite given an extra
side-condition (that `x.msb = false`).
This PR also upstreams a minor `Nat` theorem (`Nat.div_le_div_left`)
needed for the above from Mathlib.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
This PR improves the theorems used to justify the steps performed by the
inequality offset module. See new test for examples of how they are
going to be used.
This PR implements `Std.Net.Addr` which contains structures around IP
and socket addresses.
While we could implement our own parser instead of going through the
`addr_in`/`addr_in6` route we will need to implement these conversions
to make proper system calls anyway. Hence this is likely the approach
with the least amount of non trivial code overall. The only thing I am
uncertain about is whether `ofString` should return `Option` or
`Except`, unfortunately `libuv` doesn't hand out error messages on IP
parsing.
This PR adds support for creating local E-matching theorems for
universal propositions known to be true. It allows `grind` to
automatically solve examples such as:
```lean
example (b : List α) (p : α → Prop) (h₁ : ∀ a ∈ b, p a) (h₂ : ∃ a ∈ b, ¬p a) : False := by
grind
```
This PR fixes the location of the error emitted when the `rintro` and
`intro` tactics cannot introduce the requested number of binders.
This patch adds a few `withRef` wrappers to invocations of
`MVarId.intro` to fix error locations. Perhaps `MVarId.intro` should
take a syntax object to set the location itself in the future; however
there are a couple other call sites which would need non-trivial fixup.
Closes #5659.
This PR adds support for case splitting on `match`-expressions in
`grind`.
We still need to add support for resolving the antecedents of
`match`-conditional equations.
This PR modifies the `induction`/`cases` syntax so that the `with`
clause does not need to be followed by any alternatives. This improves
friendliness of these tactics, since this lets them surface the names of
the missing alternatives:
```lean
example (n : Nat) : True := by
induction n with
/- ~~~~
alternative 'zero' has not been provided
alternative 'succ' has not been provided
-/
```
Related to issue #3555
This PR adds additional tests for `grind`, demonstrating that we can
automate some manual proofs from Mathlib's basic category theory
library, with less reliance on Mathlib's `@[reassoc]` trick.
In several places I've added bidirectional patterns for equational
lemmas.
I've updated some other files to use the new `@[grind_eq]` attribute
(but left as is all cases where we are inspecting the info messages from
`grind_pattern`).
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
This PR introduces a script that automates checking whether major
downstream repositories have been updated for a new toolchain release.
Sample output:
```
% ./release_checklist.py v4.16.0-rc1
Repository: Batteries
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: lean4checker
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: doc-gen4
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: Verso
❌ Not on target toolchain (needs ≥ v4.16.0-rc1, but main is on leanprover/lean4:v4.15.0)
Repository: ProofWidgets4
✅ On compatible toolchain (>= v4.16.0-rc1)
Repository: Aesop
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: import-graph
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: plausible
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: Mathlib
✅ On compatible toolchain (>= v4.16.0-rc1)
✅ Tag v4.16.0-rc1 exists
Repository: REPL
❌ Not on target toolchain (needs ≥ v4.16.0-rc1, but master is on leanprover/lean4:v4.14.0)
```
This PR introduces the parametric attribute `[grind]` for annotating
theorems and definitions. It also replaces `[grind_eq]` with `[grind
=]`. For definitions, `[grind]` is equivalent to `[grind =]`.
The new attribute supports the following variants:
- **`[grind =]`**: Uses the left-hand side of the theorem's conclusion
as the pattern for E-matching.
- **`[grind =_]`**: Uses the right-hand side of the theorem's conclusion
as the pattern for E-matching.
- **`[grind _=_]`**: Creates two patterns. One for the left-hand side
and one for the right-hand side.
- **`[grind →]`**: Searches for (multi-)patterns in the theorem's
antecedents, stopping once a usable multi-pattern is found.
- **`[grind ←]`**: Searches for (multi-)patterns in the theorem's
conclusion, stopping once a usable multi-pattern is found.
- **`[grind]`**: Searches for (multi-)patterns in both the theorem's
conclusion and antecedents. It starts with the conclusion and stops once
a usable multi-pattern is found.
The `grind_pattern` command remains available for cases where these
attributes do not yield the desired result.
This PR introduces the `[grind_eq]` attribute, designed to annotate
equational theorems and functions for heuristic instantiations in the
`grind` tactic.
When applied to an equational theorem, the `[grind_eq]` attribute
instructs the `grind` tactic to automatically use the annotated theorem
to instantiate patterns during proof search. If applied to a function,
it marks all equational theorems associated with that function.
```lean
@[grind_eq]
theorem foo_idempotent : foo (foo x) = foo x := ...
@[grind_eq] def f (a : Nat) :=
match a with
| 0 => 10
| x+1 => g (f x)
```
In the example above, the `grind` tactic will add instances of the
theorem `foo_idempotent` to the local context whenever it encounters the
pattern `foo (foo x)`. Similarly, functions annotated with `[grind_eq]`
will propagate this annotation to their associated equational theorems.
This PR splits a definition out of `Lean.Lsp.Basic`, with the effect
that material about JSON is not needed for `Lean.Meta.Sorry` and its
dependencies.
This PR adds a script to automatically generate release notes using the
new `changelog-*` labels and "This PR ..." conventions.
Usage:
```
script/release_notes.py v4.X.0
```
where `v4.X.0` is the **previous** release, i.e. the script will process
all commits *since* that tag.
This PR fixes a slight bug that was created in the reflection of `bif`
in `bv_decide`.
Tagged as changelog-no as the code in question isn't in an RC yet.
This PR proves the basic theorems about the functions `Int.bdiv` and
`Int.bmod`.
For all integers `x` and all natural numbers `m`, we have:
- `Int.bdiv_add_bmod`: `m * bdiv x m + bmod x m = x` (which is stated in
the docstring for docs#Int.bdiv)
- `Int.bmod_add_bdiv`: `bmod x m + m * bdiv x m = x`
- `Int.bdiv_add_bmod'`: `bdiv x m * m + bmod x m = x`
- `Int.bmod_add_bdiv'`: `bmod x m + bdiv x m * m = x`
- `Int.bmod_eq_self_sub_mul_bdiv`: `bmod x m = x - m * bdiv x m`
- `Int.bmod_eq_self_sub_bdiv_mul`: `bmod x m = x - bdiv x m * m`
These theorems are all equivalent to each other by the basic properties
of addition, multiplication, and subtraction of integers.
The names `Int.bdiv_add_bmod`, `Int.bmod_add_bdiv`,
`Int.bdiv_add_bmod'`, and `Int.bmod_add_bdiv'` are meant to parallel the
names of the existing theorems docs#Int.tmod_add_tdiv,
docs#Int.tdiv_add_tmod, docs#Int.tmod_add_tdiv', and
docs#Int.tdiv_add_tmod'.
The names `Int.bmod_eq_self_sub_mul_bdiv` and
`Int.bmod_eq_self_sub_bdiv_mul` follow mathlib's naming conventions.
Note that there is already a theorem called docs#Int.bmod_def, so it
would not have been possible to parallel the name of the existing
theorem docs#Int.tmod_def.
See
https://leanprover.zulipchat.com/#narrow/channel/217875-Is-there-code-for-X.3F/topic/bdiv.20and.20bmod.
Closes#6493.
This PR introduces support for user-defined fallback code in the `grind`
tactic. The fallback code can be utilized to inspect the state of
failing `grind` subgoals and/or invoke user-defined automation. Users
can now write `grind on_failure <code>`, where `<code>` should have the
type `GoalM Unit`. See the modified tests in this PR for examples.
This PR adds a custom congruence rule for equality in `grind`. The new
rule takes into account that `Eq` is a symmetric relation. In the
future, we will add support for arbitrary symmetric relations. The
current rule is important for propagating disequalities effectively in
`grind`.
This PR fixes a bug in the congruence closure data structure used in the
`grind` tactic. The new test includes an example that previously caused
a panic. A similar panic was also occurring in the test
`grind_nested_proofs.lean`.
This PR ensures `norm_cast` doesn't fail to act in the presence of
`no_index` annotations
While leanprover/lean4#2867 exists, it is necessary to put `no_index`
around `OfNat.ofNat` in simp lemmas.
This results in extra `Expr.mdata` nodes, which must be removed before
checking for `ofNat` numerals.
This PR adds a simple strategy to the (WIP) `grind` tactic. It just
keeps internalizing new theorem instances found by E-matching. The
simple strategy can solve examples such as:
```lean
grind_pattern Array.size_set => Array.set a i v h
grind_pattern Array.get_set_eq => a.set i v h
grind_pattern Array.get_set_ne => (a.set i v hi)[j]
example (as bs : Array α) (v : α)
(i : Nat)
(h₁ : i < as.size)
(h₂ : bs = as.set i v)
: as.size = bs.size := by
grind
example (as bs cs : Array α) (v : α)
(i : Nat)
(h₁ : i < as.size)
(h₂ : bs = as.set i v)
(h₃ : cs = bs)
(h₄ : i ≠ j)
(h₅ : j < cs.size)
(h₆ : j < as.size)
: cs[j] = as[j] := by
grind
opaque R : Nat → Nat → Prop
theorem Rtrans (a b c : Nat) : R a b → R b c → R a c := sorry
grind_pattern Rtrans => R a b, R b c
example : R a b → R b c → R c d → R d e → R a d := by
grind
```
This PR fixes a bug in the theorem instantiation procedure in the (WIP)
`grind` tactic. For example, it was missing the following instance in
one of the tests:
```lean
[grind.ematch.instance] Array.get_set_ne: ∀ (hj : i < bs.size), j ≠ i → (bs.set j w ⋯)[i] = bs[i]
```
This PR also renames the `grind` base monad to `GrindCoreM`.
This PR adds a deriving handler for the `ToExpr` class. It can handle
mutual and nested inductive types, however it falls back to creating
`partial` instances in such cases. This is upstreamed from the Mathlib
deriving handler written by @kmill, but has fixes to handle autoimplicit
universe level variables.
This is a followup to #6285 (adding the `ToLevel` class). This PR
supersedes #5906.
Co-authored-by: Alex Keizer <alex@keizer.dev>
---------
Co-authored-by: Alex Keizer <alex@keizer.dev>
This PR adds support for activating relevant theorems for the (WIP)
`grind` tactic. We say a theorem is relevant to a `grind` goal if the
symbols occurring in its patterns also occur in the goal.
This PR adds pattern validation to the `grind_pattern` command. The new
`checkCoverage` function will also be used to implement the attributes
`@[grind_eq]`, `@[grind_fwd]`, and `@[grind_bwd]`.
This PR implements the command `grind_pattern`. The new command allows
users to associate patterns with theorems. These patterns are used for
performing heuristic instantiation with e-matching. In the future, we
will add the attributes `@[grind_eq]`, `@[grind_fwd]`, and
`@[grind_bwd]` to compute the patterns automatically for theorems.
This PR introduces a command for specifying patterns used in the
heuristic instantiation of global theorems in the `grind` tactic. Note
that this PR only adds the parser.
This PR completes the implementation of `addCongrTable` in the (WIP)
`grind` tactic. It also adds a new test to demonstrate why the extra
check is needed. It also updates the field `cgRoot` (congruence root).
This PR completes support for literal values in the (WIP) `grind`
tactic. `grind` now closes the goal whenever it merges two equivalence
classes with distinct literal values.
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 adds the ability to override package entries in a Lake manifest
via a separate JSON file. This file can be specified on the command line
with `--packages` or applied persistently by placing it at
`.lake/package-overrides.json`.
The overrides file is a subset of `lake-manifest.json` with just a
version and a `packages` field. The entries in the package share the
syntax of the manifest file and take precedence over the entries there.
Lake loads the entries from the manifest, then overrides them with those
in `.lake/package-overrides.json` (if any) and then those in any file
passed to `--packages`.
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 adds basic lemmas about lexicographic order on Array and Vector,
achieving parity with List.
Many lemmas are still missing for all three, particularly about how
order interacts with `++`.
This PR fixes a bug in the `sharecommon` module, which was returning
incorrect results for objects that had already been processed by
`sharecommon`. See the new test for an example that triggered the bug.
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 merges `BuildJob` and `Job`, deprecating the former. `Job` now
contains a trace as part of its state which can be interacted with
monadically. This PR also simplifies the implementation of `OpaqueJob`.
This merger removes the need in Lake to distinguish between different
kinds of jobs, which helps enable the overall goal of making all targets
return a `Job` (and therefore make it easer for the frontend to
manipulate them in, e.g., #6323).
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>
To avoid user confusion, there should be just one manual.
This PR deletes the old manual, adding a link to the new one; the
website config will redirect these pages to the corresponding new manual
content.
This PR adds lemmas reducing for loops over `Std.Range` to for loops
over `List.range'`.
Equivalent theorems previously existed in Batteries, but the underlying
definitions have changed so these are written from scratch.
This PR adds a dockerfile for use with Gitpod.
This provides all the dependencies, and kicks off a build once the
editor is opened for the first time.
It can be tested by going to
https://gitpod.io/#https://github.com/leanprover/lean4/pull/6382
This should make it less painful for users hoping to contribute small
lemmas to `Init/` and `Std/`; they can open gitpod and wait, rather than
having to read the docs to run a series of commands.
This PR generalizes the panic functions to a type of `Sort u` rather
than `Type u`. This better supports universe polymorphic types and
avoids confusing errors.
An minimal (but somewhat contrived) example of such a confusing error
is:
```lean
/-
stuck at solving universe constraint
?u.59+1 =?= max 1 ?u.7
while trying to unify
Subtype.{?u.7} P : Sort (max 1 ?u.7)
with
Subtype.{?u.7} P : Sort (max 1 ?u.7)
-/
def assertSubtype! {P : α → Prop} [Inhabited (Subtype P)] (a : α) [Decidable (P a)] : Subtype P := -- errors on :=
if h : P a then
⟨a, h⟩
else
panic! "Property not satisified"
```
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 adds lemmas about `Vector.set`, `anyM`, `any`, `allM`, and
`all`.
With these additions, `Vector` is now as in-sync with the `List` API as
`Array` is, and in future I'll be updating both simultaneously.
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).
This PR adds `Nat` theorems for distributing `>>>` over bitwise
operations, paralleling those of `BitVec`.
This enables closing goals like the following using `simp`:
```lean
example (n : Nat) : (n <<< 2 ||| 3) >>> 2 = n := by simp [Nat.shiftRight_or_distrib]
```
It might be nice for these theorems to be `simp` lemmas, but they are
not currently in order to be consistent with the existing `BitVec` and
`div_two` theorems.
This PR makes all message constructors handle pretty printer errors.
Prior to this change, pretty printer errors in messages were not
uniformly handled. In core, some printers capture their errors (e.g.,
`ppExprWithInfos` and `ppTerm` ) and some do not (e.g., `ppGoal` and
`ppSignature`) propagate them to whatever serializes the message (e.g.,
the frontend).
To resolve this inconsistency and uniformly handle errors, the signature
for `ofLazy` now uses `BaseIO`. As such, all printers been adapted to
capture any errors within them and print similar messages to
`ppExprWithInfos` and `ppTerm` on such errors.
This PR adjusts the way the pretty printer unresolves names. It used to
make use of all `export`s when pretty printing, but now it only uses
`export`s that put names into parent namespaces (heuristic: these are
"API exports" that are intended by the library author), rather than
"horizontal exports" that put the names into an unrelated namespace,
which the dot notation feature in #6189 now incentivizes.
Closes the already closed#2524
This PR adds `BitVec.[toFin|getMsbD]_setWidth` and
`[getMsb|msb]_signExtend` as well as `ofInt_toInt`.
Also correct renamed the misnamed theorem for
`signExtend_eq_setWidth_of_msb_false`.
---------
Co-authored-by: Siddharth <siddu.druid@gmail.com>
This PR adds docstrings to `Expr.hasLooseBVars` and `Expr.hasLooseBVar`,
to clarify the difference between these functions, and to document that
the former traverses the expression, while the latter is constant-time,
using cached information.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
This PR removes the deprecated aliases `Int.div := Int.tdiv` and
`Int.mod := Int.tmod`. Later we will rename `Int.ediv` to `Int.div` and
`Int.emod` to `Int.mod`.
This PR removes unnecessary parameters from the funcion induction
principles. This is a breaking change; broken code can typically be adjusted
simply by passing fewer parameters.
Part 2, adjusting after stage0 update.
Closes#6320
This PR removes unnecessary parameters from the funcion induction
principles. This is a breaking change; broken code can typically be adjusted
simply by passing fewer parameters.
Part 1, before stage0 update.
Closes#6320
This PR adds `protected` to `Fin.cast` and `BitVec.cast`, to avoid
confusion with `_root_.cast`. These should mostly be used via
dot-notation in any case.
This PR upstreams the `ToLevel` typeclass from mathlib and uses it to
fix the existing `ToExpr` instances so that they are truly universe
polymorphic (previously it generated malformed expressions when the
universe level was nonzero). We improve on the mathlib definition of
`ToLevel` to ensure the class always lives in `Type`, irrespective of
the universe parameter.
This implements part one of the plan to upstream a derive handler for
`ToExpr`, as discussed in #5906 and #5909.
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
This PR adds lemmas simplifying `for` loops over `Option` into
`Option.pelim`, giving parity with lemmas simplifying `for` loops of
`List` into `List.fold`.
This PR removes an unused import in the time library that can yield to
import cycles when building stuff that gets imported by `Std.Internal`
but also wants to import `Std.Time`.
This PR adds `BitVec.[toInt|toFin]_concat` and moves a couple of
theorems into the concat section, as `BitVec.msb_concat` is needed for
the `toInt_concat` proof.
We also add `Bool.toInt`.
This PR adds theorems characterizing the value of the unsigned shift
right of a bitvector in terms of its 2s complement interpretation as an
integer.
Unsigned shift right by at least one bit makes the value of the
bitvector less than or equal to `2^(w-1)`,
makes the interpretation of the bitvector `Int` and `Nat` agree.
In the case when `n = 0`, then the shift right value equals the integer
interpretation.
```lean
theorem toInt_ushiftRight_eq_ite {x : BitVec w} {n : Nat} :
(x >>> n).toInt = if n = 0 then x.toInt else x.toNat >>> n
```
```lean
theorem toFin_uShiftRight {x : BitVec w} {n : Nat} :
(x >>> n).toFin = x.toFin / (Fin.ofNat' (2^w) (2^n))
```
---------
Co-authored-by: Harun Khan <harun19@stanford.edu>
Co-authored-by: Tobias Grosser <github@grosser.es>
This PR changes the implementation of `HashMap.toList`, so the ordering
agrees with `HashMap.toArray`.
Currently there are no verification lemmas about `HashMap.toList`, so no
contract is being broken yet!
This PR moves `IO.Channel` and `IO.Mutex` from `Init` to `Std.Sync` and
renames them to `Std.Channel` and `Std.Mutex`.
Note that the original files are retained and the deprecation is written
manually as we cannot import `Std` from `Init` so this is the only way
to deprecate without a hard breaking change. In particular we do not yet
move `Std.Queue` from `Init` to `Std` both because it needs to be
retained for this deprecation to work but also because it is already
within the `Std` namespace and as such we cannot maintain two copies of
the file at once. After the deprecation period is finished `Std.Queue`
will find a new home in `Std.Data.Queue`.
This PR upstreams `List.length_flatMap`, `countP_flatMap` and
`count_flatMap` from Mathlib. These were not possible to state before we
upstreamed `List.sum`.
This PR makes some proofs more robust so they will still work with
`byAsSorry`. Unfortunately, they are not a complete fix and there are
remaining problems building with `byAsSorry`.
This PR ensures the the log error position is properly preserved when
prepending stray log entries to the job log. It also adds comparison
support for `Log.Pos`.
This PR uses Lean.RArray in bv_decide's reflection proofs. Giving
speedups on problems with lots of variables.
Implement like #6068, speedup:
```
# before
λ hyperfine "lean +nightly-2024-12-02 tests/lean/run/bv_reflection_stress.lean"
Benchmark 1: lean +nightly-2024-12-02 tests/lean/run/bv_reflection_stress.lean
Time (mean ± σ): 1.939 s ± 0.007 s [User: 1.549 s, System: 0.104 s]
Range (min … max): 1.928 s … 1.947 s 10 runs
# after
λ hyperfine "lean tests/lean/run/bv_reflection_stress.lean"
Benchmark 1: lean tests/lean/run/bv_reflection_stress.lean
Time (mean ± σ): 1.409 s ± 0.006 s [User: 1.058 s, System: 0.073 s]
Range (min … max): 1.401 s … 1.419 s 10 runs
```
This PR changes Lake's build process to no longer use `leanc` for
compiling C files or linking shared libraries and executables. Instead,
it directly invokes the bundled compiler (or the native compiler if
none) using the necessary flags.
This PR runs all linters for a single command (together) on a separate
thread from further elaboration, making a first step towards
parallelizing the elaborator.
This PR ensure `bv_decide` uses definitional equality in its reflection
procedure as much as possible. Previously it would build up explicit
congruence proofs for the kernel to check. This reduces the size of
proof terms passed to kernel speeds up checking of large reflection
proofs.
This PR reduces the import closure of `Std.Time` such that it doesn't
have to be rebuilt on every change in `Init.Data`.
Noticed while working on `Init` refactorings.
This PR fixes a bug in structure instance field completion that caused
it to not function correctly for bracketed structure instances written
in Mathlib style.
This PR completes the `toNat` theorems for the bitwise operations
(`and`, `or`, `xor`, `shiftLeft`, `shiftRight`) of the UInt types and
adds `toBitVec` theorems as well. It also renames `and_toNat` to
`toNat_and` to fit with the current naming convention.
This PR fixes a bug that could cause the `injectivity` tactic to fail in
reducible mode, which could cause unfolding lemma generation to fail
(used by tactics such as `unfold`). In particular,
`Lean.Meta.isConstructorApp'?` was not aware that `n + 1` is equivalent
to `Nat.succ n`.
Closes#5064
This PR introduces the basic theory of permutations of `Array`s and
proves `Array.swap_perm`.
The API falls well short of what is available for `List` at this point.
This PR refactors `Array.qsort` to remove runtime array bounds checks,
and avoids the use of `partial`. We use the `Vector` API, along with
auto_params, to avoid having to write any proofs. The new code
benchmarks indistinguishably from the old.
This PR puts code in terms of syntax quotations now that there has been
a stage0 update. Fixes a lingering bug in StructInst where some
intermediate syntax was malformed, but this had no observable effects
outside of some debug messages.
This PR modifies structure instance notation and `where` notation to use
the same notation for fields. Structure instance notation now admits
binders, type ascriptions, and equations, and `where` notation admits
full structure lvals. Examples of these for structure instance notation:
```lean
structure PosFun where
f : Nat → Nat
pos : ∀ n, 0 < f n
def p : PosFun :=
{ f n := n + 1
pos := by simp }
def p' : PosFun :=
{ f | 0 => 1
| n + 1 => n + 1
pos := by rintro (_|_) <;> simp }
```
Just like for the structure `where` notation, a field `f x y z : ty :=
val` expands to `f := fun x y z => (val : ty)`. The type ascription is
optional.
The PR also is setting things up for future expansion. Pending some
discussion, in the future structure/`where` notation could have have
embedded `where` clauses; rather than `{ a := { x := 1, y := z } }` one
could write `{ a where x := 1; y := z }`.
This PR implements `Simp.Config.implicitDefEqsProofs`. When `true`
(default: `true`), `simp` will **not** create a proof term for a
rewriting rule associated with an `rfl`-theorem. Rewriting rules are
provided by users by annotating theorems with the attribute `@[simp]`.
If the proof of the theorem is just `rfl` (reflexivity), and
`implicitDefEqProofs := true`, `simp` will **not** create a proof term
which is an application of the annotated theorem.
The default setting does change the existing behavior. Users can use
`simp -implicitDefEqProofs` to force `simp` to create a proof term for
`rfl`-theorems. This can positively impact proof checking time in the
kernel.
This PR also fixes an issue in the `split` tactic that has been exposed
by this feature. It was looking for `split` candidates in proofs and
implicit arguments. See new test for issue exposed by the previous
feature.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
This PR fixes a performance issue where the Lean language server would
walk the full project file tree every time a file was saved, blocking
the processing of all other requests and notifications and significantly
increasing overall language server latency after saving.
This issue was originally reported at
https://leanprover.zulipchat.com/#narrow/channel/270676-lean4/topic/Compiling.20too.20slow/near/484386515
and uncovered in a lengthy investigation. The performance bug that
causes the Lean language server to walk the full project file tree when
the file watcher for .ilean files is triggered was introduced when the
.ileans were first introduced, whereas the specific issue of file saving
also triggering the walk was introduced by #3247 in 4.8.0 and the use of
the file watcher for .lean files, which would then also trigger the
directory walk. Combining this with VS Code's auto-save feature causes
the language server to walk the full project file tree on every change
of the document.
It somehow hasn't really been much of an issue until now, but we still
do way too much work in the watchdog main loop. I'll look into resolving
that more general issue in the future.
This PR ensures that nesting trace nodes are annotated with timing
information iff `trace.profiler` is active.
The previous connection to the otherwise unrelated `profiler` option was
a remnant from before `trace.profiler` existed; if users want to
annotate explicitly activated trace classes only, they can instead
increase `trace.profiler.threshold`.
This PR adds the builtin simproc `USize.reduceToNat` which reduces the
`USize.toNat` operation on literals less than `UInt32.size` (i.e.,
`4294967296`).
This PR upstreams some UInt theorems from Batteries and adds more
`toNat`-related theorems. It also adds the missing `UInt8` and `UInt16`
to/from `USize` conversions so that the the interface is uniform across
the UInt types.
**Summary of all changes:**
* Upstreamed and added `toNat` constructors lemmas: `toNat_mk`,
`ofNat_toNat`, `toNat_ofNat`, `toNat_ofNatCore`, and
`USize.toNat_ofNat32`
* Upstreamed and added `toNat` canonicalization; `val_val_eq_toNat` and
`toNat_toBitVec_eq_toNat`
* Added injectivity iffs: `toBitVec_inj`, `toNat_inj`, and `val_inj`
* Added inequality iffs: `le_iff_toNat_le` and `lt_iff_toNat_lt`
* Upstreamed antisymmetry lemmas: `le_antisymm` and `le_antisymm_iff`
* Upstreamed missing `toNat` lemmas on arithmetic operations:
`toNat_add`, `toNat_sub`, `toNat_mul`
* Upstreamed and added missing conversion lemmas: `toNat_toUInt*` and
`toNat_USize`
* Added missing `USize` conversions: `USize.toUInt8`, `UInt8.toUSize`,
`USize.toUInt16`, `UInt16.toUSize`
This PR deprecates `Fin.ofNat` in favour of `Fin.ofNat'` (which takes an
`[NeZero]` instance, rather than returning an element of `Fin (n+1)`).
After leaving the deprecation warning in place for some time, we will
then rename `ofNat'` back to `ofNat`.
This PR fixes:
- Problems in other linux distributions that the default `tzdata`
directory is not the same as previously defined by ensuring it with a
fallback behavior when directory is missing.
- Trim unnecessary characters from local time identifier.
This PR adds a cmake knob to allow turning off installing a copy of
`cadical`.
This can be useful for custom builds/installs where cadical is already
available in the system.
Closes: #5603
This PR adds `Lean.loadPlugin` which exposes functionality similar to
the `lean` executable's `--plugin` option to Lean code.
This will allow custom Lean frontends (e.g., Lake, the Lean language
server) to also load plugins.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
This PR improves the errors Lake produces when it fails to fetch a
dependency from Reservoir. If the package is not indexed, it will
produce a suggestion about how to require it from GitHub.
Closes#5330.
This PR upstreams the definition and basic lemmas about `List.finRange`
from Batteries.
Thanks for contributors to Batteries and Mathlib who've previously
worked on this material. Further PRs are welcome here. I'll be adding
more API later.
This PR upstreams lemmas about `Vector` from Batteries.
I'll be adding more soon, and PRs are welcome, particularly from those
who have previously contributed to `Vector` in Batteries.
This PR implements `BitVec.toInt_abs`.
The absolute value of `x : BitVec w` is naively a case split on the sign
of `x`.
However, recall that when `x = intMin w`, `-x = x`.
Thus, the full value of `abs x` is computed by the case split:
- If `x : BitVec w` is `intMin`, then its absolute value is also `intMin
w`, and
thus `toInt` will equal `intMin.toInt`.
- Otherwise, if `x` is negative, then `x.abs.toInt = (-x).toInt`.
- Finally, when `x` is nonnegative, then `x.abs.toInt = x.toInt`.
```lean
theorem toInt_abs {x : BitVec w} :
x.abs.toInt =
if x = intMin w then (intMin w).toInt
else if x.msb then -x.toInt
else x.toInt
```
We also provide a variant of `toInt_abs` that
hides the case split for `x` being positive or negative by using
`natAbs`.
```lean
theorem toInt_abs_eq_natAbs {x : BitVec w} : x.abs.toInt =
if x = intMin w then (intMin w).toInt else x.toInt.natAbs
```
Supercedes https://github.com/leanprover/lean4/pull/5787
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
This PR adds `toNat` theorems for `BitVec.signExtend.`
Sign extending to a larger bitwidth depends on the msb. If the msb is
false, then the result equals the original value. If the msb is true,
then we add a value of `(2^v - 2^w)`, which arises from the sign
extension.
```lean
theorem toNat_signExtend (x : BitVec w) {v : Nat} :
(x.signExtend v).toNat = (x.setWidth v).toNat + if x.msb then 2^v - 2^w else 0
```
Co-authored-by: Harun Khan <harun19@stanford.edu>
This PR adds theorem `mod_eq_sub`, makes theorem
`sub_mul_eq_mod_of_lt_of_le` not private anymore and moves its location
within the `rotate*` section to use it in other proofs.
This PR upstreams `Nat.lt_pow_self` and `Nat.lt_two_pow` from Mathlib
and uses them to prove the simp theorem `Nat.mod_two_pow`.
This simplifies expressions like `System.Platform.numBits % 2 ^
System.Platform.numBits = System.Platform.numBits`, which is needed for
#6188.
This PR adds the theorems `le_usize_size` and `usize_size_le`, which
make proving inequalities about `USize.size` easier.
It also deprecates `usize_size_gt_zero` in favor of `usize_size_pos` (as
that seems more consistent with our naming covention) and adds
`USize.toNat_ofNat_of_lt_32` for dealing with small USize literals.
It also moves `USize.ofNat32` and `USize.toUInt64` to
`Init.Data.UInt.Basic` as neither are used in `Init.Prelude` anymore.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
This PR makes Lake no longer automatically fetch GitHub cloud releases
if the package build directory is already present (mirroring the
behavior of the Reservoir cache). This prevents the cache from
clobbering existing prebuilt artifacts. Users can still manually fetch
the cache and clobber the build directory by running `lake build
<pkg>:release`.
This PR changes the definition of `HashSet.insertMany` and
`HashSet.Raw.insertMany` so that it is equivalent to repeatedly calling
`HashSet.insert`/`HashSet.Raw.insert`. It also clarifies the docstrings
of all the `insert` and `insertMany` functions.
---------
Co-authored-by: Kim Morrison <scott.morrison@gmail.com>
This PR makes stricter requirements for the `@[deprecated]` attribute,
requiring either a replacement identifier as `@[deprecated bar]` or
suggestion text `@[deprecated "Past its use by date"]`, and also
requires a `since := "..."` field.
This PR changes how generalized field notation ("dot notation") resolves
the function. The new resolution rule is that if `x : S`, then `x.f`
resolves the name `S.f` relative to the root namespace (hence it now
affected by `export` and `open`). Breaking change: aliases now resolve
differently. Before, if `x : S`, and if `S.f` is an alias for `S'.f`,
then `x.f` would use `S'.f` and look for an argument of type `S'`. Now,
it looks for an argument of type `S`, which is more generally useful
behavior. Code making use of the old behavior should consider defining
`S` or `S'` in terms of the other, since dot notation can unfold
definitions during resolution.
This also fixes a bug in explicit-mode generalized field notation
(`@x.f`) where `x` could be passed as the wrong argument. This was not a
bug for explicit-mode structure projections.
Closes#3031. Addresses the `Function` namespace issue in #1629.
This PR makes it possible to write `rw (occs := [1,2]) ...` instead of
`rw (occs := .pos [1,2]) ...` by adding a coercion from `List.Nat` to
`Lean.Meta.Occurrences`.
This PR makes `USize.toUInt64` a regular non-opaque definition.
It also moves it to `Init.Data.UInt.Basic`, as it is not actually used
in `Init.Prelude` anymore.
This PR changes the signature of `Array.swap`, so it takes `Nat`
arguments with tactic provided bounds checking. It also renames
`Array.swap!` to `Array.swapIfInBounds`.
This PR completes the TODO in `Init.Data.Array.BinSearch`, removing the
`partial` keyword and converting runtime bounds checks to compile time
bounds checks.
This PR fixes a bug with the `structure`/`class` command where if there
are parents that are not represented as subobjects but which used other
parents as instances, then there would be a kernel error. Closes#2611.
Note: there is still the limitation that parents that are not
represented as subobjects do not themselves provide instances to other
parents.
This PR adds toInt theorems for BitVec.signExtend.
If the current width `w` is larger than the extended width `v`,
then the value when interpreted as an integer is truncated,
and we compute a modulo by `2^v`.
```lean
theorem toInt_signExtend_of_le (x : BitVec w) (hv : v ≤ w) :
(x.signExtend v).toInt = Int.bmod (x.toNat) (2^v)
```
Co-authored-by: Siddharth Bhat <siddu.druid@gmail.com>
Co-authored-by: Harun Khan <harun19@stanford.edu>
Stacked on top of #6155
---------
Co-authored-by: Harun Khan <harun19@stanford.edu>
This PR uses `Array.findFinIdx?` in preference to `Array.findIdx?` where
it allows converting a runtime bounds check to a compile time bounds
check.
(and some other minor cleanup)
This PR fixes a bug where the signature pretty printer would ignore the
current setting of `pp.raw`. This fixes an issue where `#check ident`
would not heed `pp.raw`. Closes#6090.
This PR fixes a non-termination bug that occurred when generating the
match-expression equation theorems. The bug was triggered when the proof
automation for the equation theorem repeatedly applied `injection(` to
the same local declaration, as it could not be removed due to forward
dependencies. See issue #6067 for an example that reproduces this issue.
closes#6067
This PR adds core metaprogramming functions for forking off background
tasks from elaboration such that their results are visible to reporting
and the language server
This PR adds support for `structure` in `mutual` blocks, allowing
inductive types defined by `inductive` and `structure` to be mutually
recursive. The limitations are (1) that the parents in the `extends`
clause must be defined before the `mutual` block and (2) mutually
recursive classes are not allowed (a limitation shared by `class
inductive`). There are also improvements to universe level inference for
inductive types and structures. Breaking change: structure parents now
elaborate with the structure in scope (fix: use qualified names or
rename the structure to avoid shadowing), and structure parents no
longer elaborate with autoimplicits enabled.
Internally, this is a large refactor of both the `inductive` and
`structure` commands. Common material is now in
`Lean.Elab.MutualInductive`, and each command plugs into this mutual
inductive elaboration framework with the logic specific to the
respective command. For example, `structure` has code to add projections
after the inductive types are added to the environment.
Closes#4182
This PR modifies the signature of the functions `Nat.fold`,
`Nat.foldRev`, `Nat.any`, `Nat.all`, so that the function is passed the
upper bound. This allows us to change runtime array bounds checks to
compile time checks in many places.
This file was upstreamed from batteries; I just got bitten by the
invalid reference and it took quite a while to figure out that this one
had been moved!
This PR adds a slight performance improvement to reflection of `if`
statements that I noticed by profiling Leanwuzla against SMTCOMP's
`non-incremental/QF_BV/fft/Sz256_6616.smt2`.
In particular:
1. The profile showed about 4 percent of the total run time were spent
constructing these decidable instances in reflection of `if` statements.
We can construct them much quicker by hand as they always have the same
structure
2. This delays construction of these statements until we actually
generate the reflection proof that we wish to submit to the kernel. Thus
if we encounter a SAT instad of an UNSAT problem we will not spend time
generating these expressions anymore.
```
baseline
Time (mean ± σ): 31.236 s ± 0.258 s
Range (min … max): 30.899 s … 31.661 s 10 runs
after
Time (mean ± σ): 30.671 s ± 0.288 s
Range (min … max): 30.350 s … 31.156 s 10 runs
```
This PR fixes a non-termination bug that occurred when generating the
match-expression splitter theorem. The bug was triggered when the proof
automation for the splitter theorem repeatedly applied `injection` to
the same local declaration, as it could not be removed due to forward
dependencies. See issue #6065 for an example that reproduces this issue.
closes#6065
This PR adds lemmas for extracting a given bit of a `BitVec` obtained
via `sub`/`neg`/`sshiftRight'`/`abs`.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
This PR avoids runtime array bounds checks in places where it can
trivially be done at compile time.
None of these changes are of particular consequence: I mostly wanted to
learn how much we do this, and what the obstacles are to doing it less.
This PR does the same fix as #6104, but such that it doesn't break the
test/the file in `Plausible`. This is done by not creating unused let
binders in metavariable types that are made by `elimMVar`. (This is also
a positive thing for users looking at metavariable types, for example in
error messages)
We get rid of `skipAtMostNumBinders`. This function was originally
defined for the purpose of making this test work, but it is a hack
because it allows cycles in the metavariable context.
It would make sense to split these changes into 2 PRs, but I combined
them here to show that the combination of them closes#6013 without
breaking anything
Closes#6013
This PR adds support for displaying multiple threads in the trace
profiler output.
`TraceState.tid` needs to be adjusted for this purpose, which is not
done yet by the Lean elaborator as it is still single-threaded.
This PR replaces `Array.feraseIdx` and `Array.insertAt` with
`Array.eraseIdx` and `Array.insertIdx`, both of which take a `Nat`
argument and a tactic-provided proof that it is in bounds. We also have
`eraseIdxIfInBounds` and `insertIdxIfInBounds` which are noops if the
index is out of bounds. We also provide a `Fin` valued version of
`Array.findIdx?`. Together, these quite ergonomically improve the array
indexing safety at a number of places in the compiler/elaborator.
This PR adds theorems `BitVec.(getMsbD, msb)_(rotateLeft, rotateRight)`.
We follow the same strategy taken for `getLsbD`, constructing the
necessary auxilliary theorems first (relying on different hypotheses)
and then generalizing.
---------
Co-authored-by: Siddharth <siddu.druid@gmail.com>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
This PR improves the `#print` command for structures to show all fields
and which parents the fields were inherited from, hiding internal
details such as which parents are represented as subobjects. This
information is still present in the constructor if needed. The pretty
printer for private constants is also improved, and it now handles
private names from the current module like any other name; private names
from other modules are made hygienic.
Example output for `#print Monad`:
```
class Monad.{u, v} (m : Type u → Type v) : Type (max (u + 1) v)
number of parameters: 1
parents:
Monad.toApplicative : Applicative m
Monad.toBind : Bind m
fields:
Functor.map : {α β : Type u} → (α → β) → m α → m β
Functor.mapConst : {α β : Type u} → α → m β → m α
Pure.pure : {α : Type u} → α → m α
Seq.seq : {α β : Type u} → m (α → β) → (Unit → m α) → m β
SeqLeft.seqLeft : {α β : Type u} → m α → (Unit → m β) → m α
SeqRight.seqRight : {α β : Type u} → m α → (Unit → m β) → m β
Bind.bind : {α β : Type u} → m α → (α → m β) → m β
constructor:
Monad.mk.{u, v} {m : Type u → Type v} [toApplicative : Applicative m] [toBind : Bind m] : Monad m
resolution order:
Monad, Applicative, Bind, Functor, Pure, Seq, SeqLeft, SeqRight
```
Suggested by Floris van Doorn [on
Zulip](https://leanprover.zulipchat.com/#narrow/channel/270676-lean4/topic/.23print.20command.20for.20structures/near/482503637).
This PR fixes a bug at the definitional equality test (`isDefEq`). At
unification constraints of the form `c.{u} =?= c.{v}`, it was not trying
to unfold `c`. This bug did not affect the kernel.
closes#6117
This PR adds a case to `Level.geq` that is present in the kernel's level
`is_geq` procedure, making them consistent with one another.
This came up during testing of `lean4lean`. Currently `Level.geq`
differs from `level::is_geq` in the case of `max u v >= imax u v`. The
elaborator function is overly pessimistic and yields `false` on this
while the kernel function yields true. This comes up concretely in the
`Trans` class:
```lean
class Trans (r : α → β → Sort u) (s : β → γ → Sort v) (t : outParam (α → γ → Sort w)) where
trans : r a b → s b c → t a c
```
The type of this class is `Sort (max (max (max (max (max (max 1 u) u_1)
u_2) u_3) v) w)` (where `u_1 u_2 u_3` are the levels of `α β γ`), but if
you try writing that type explicitly then the `class` command fails.
Omitting the type leaves the `class` to infer the universe level (the
command assumes the level is correct, and the kernel agrees it is), but
including the type then the elaborator checks the level inequality with
`Level.geq` and fails.
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
This PR changes `lean_sharecommon_{eq,hash}` to only consider the
salient bytes of an object, and not any bytes of any
unspecified/uninitialized unused capacity.
Accessing uninitialized storage results in undefined behaviour.
This does not seem to have any semantics disadvantages: If objects
compare equal after this change, their salient bytes are still equal. By
contrast, if the actual identity of allocations needs to be
distinguished, that can be done by just comparing pointers to the
storage.
If we wanted to retain the current logic, we would need initialize the
otherwise unused parts to some specific value to avoid the undefined
behaviour.
Closes#5831
This PR fixes a bug where structural recursion did not work when indices
of the recursive argument appeared as function parameters in a different
order than in the argument's type's definition.
Fixes#6015.
This PR liberalizes atom rules by allowing `''` to be a prefix of an
atom, after #6012 only added an exception for `''` alone, and also adds
some unit tests for atom validation.
This PR fixes an issue in the `injection` tactic. This tactic may
execute multiple sub-tactics. If any of them fail, we must backtrack the
partial assignment. This issue was causing the error: "`mvarId` is
already assigned" in issue #6066. The issue is not yet resolved, as the
equation generator for the match expressions is failing in the example
provided in this issue.
This PR fixes the caching infrastructure for `whnf` and `isDefEq`,
ensuring the cache accounts for all relevant configuration flags. It
also cleans up the `WHNF.lean` module and improves the configuration of
`whnf`.
This PR fixes a stack overflow caused by a cyclic assignment in the
metavariable context. The cycle is unintentionally introduced by the
structure instance elaborator.
closes#3150
This PR makes the `change` tactic and conv tactic use the same
elaboration strategy. It works uniformly for both the target and local
hypotheses. Now `change` can assign metavariables, for example:
```lean
example (x y z : Nat) : x + y = z := by
change ?a = _
let w := ?a
-- now `w : Nat := x + y`
```
This PR modifies `Lean.MVarId.replaceTargetDefEq` and
`Lean.MVarId.replaceLocalDeclDefEq` to use `Expr.equal` instead of
`Expr.eqv` when determining whether the expression has changed. This is
justified on the grounds that binder names and binder infos are
user-visible and affect elaboration.
This PR adds raw transmutation of floating-point numbers to and from
`UInt64`. Floats and UInts share the same endianness across all
supported platforms. The IEEE 754 standard precisely specifies the bit
layout of floats. Note that `Float.toBits` is distinct from
`Float.toUInt64`, which attempts to preserve the numeric value rather
than the bitwise value.
closes#6071
This PR adds the option `pp.parens` (default: false) that causes the
pretty printer to eagerly insert parentheses, which can be useful for
teaching and for understanding the structure of expressions. For
example, it causes `p → q → r` to pretty print as `p → (q → r)`.
Any notations with precedence greater than or equal to `maxPrec` do not
receive such discretionary parentheses, since this precedence level is
considered to be infinity.
This option was a feature in the Lean 3 community edition.
This PR fixes a bug in the constant folding for the `Nat.ble` and
`Nat.blt` function in the old code generator, leading to a
miscompilation.
Closes#6086
This PR improves the term info for coercions marked with
`CoeFnType.coeFun` (such as `DFunLike.coe` in Mathlib), making "go to
definition" on the function name work. Hovering over such a coerced
function will show the coercee rather than the coercion expression. The
coercion expression can still be seen by hovering over the whitespace in
the function application.
This PR introduces date and time functionality to the Lean 4 Std.
Breaking Changes:
- `Lean.Data.Rat` is now `Std.Internal.Rat` because it's used by the
DateTime library.
---------
Co-authored-by: Markus Himmel <markus@himmel-villmar.de>
Co-authored-by: Mac Malone <tydeu@hatpress.net>
This PR prepares #6068 by using the `RArray` data structure in
`simp_arith` the simp-arith meta code.
After the subsequent stage0 we can change the simp-arith theorems in
`Init`.
This PR adds the Lean.RArray data structure.
This data structure is equivalent to `Fin n → α` or `Array α`, but
optimized for a fast kernel-reduction `get` operation.
It is not suitable as a general-purpose data structure. The primary
intended use case is the “denote” function of a typical proof by
reflection proof, where only the `get` operation is necessary, and where
using `List.get` unnecessarily slows down proofs with more than a
hand-full of atomic expressions.
There is no well-formedness invariant attached to this data structure,
to keep it concise; it's semantics is given through `RArray.get`. In
that way one can also view an `RArray` as a decision tree implementing
`Nat → α`.
In #6068 this data structure is used in `simp_arith`.
This PR improves the validation of new syntactic tokens. Previously, the
validation code had inconsistencies: some atoms would be accepted only
if they had a leading space as a pretty printer hint. Additionally,
atoms with internal whitespace are no longer allowed.
Closes#6011
This PR adds a newline at end of each Lean file generated by `lake new`
templates.
I have tested it with a locally compiled Lean with this commit. I hope
these changes make `lake new`'s behavior more consistent with the Lean 4
plugins and libraries newlines convention.
This PR adds a new definition `Message.kind` which returns the top-level
tag of a message. This is serialized as the new field `kind` in
`SerialMessaege` so that i can be used by external consumers (e.g.,
Lake) to identify messages via `lean --json`.
The tag of trace messages has also been changed from `_traceMsg` to the
more friendly `trace`.
Not a huge benefit, but actually reduces the code complexity (no need
for the `.fuse` function), and can help with problems with many repeated
varibles.
This PR fixes a bug where the monad lift coercion elaborator would
partially unify expressions even if they were not monads. This could be
taken advantage of to propagate information that could help elaboration
make progress, for example the first `change` worked because the monad
lift coercion elaborator was unifying `@Eq _ _` with `@Eq (Nat × Nat)
p`:
```lean
example (p : Nat × Nat) : p = p := by
change _ = ⟨_, _⟩ -- used to work (yielding `p = (p.fst, p.snd)`), now it doesn't
change ⟨_, _⟩ = _ -- never worked
```
As such, this is a breaking change; you may need to adjust expressions
to include additional implicit arguments.
This PR implements conversion functions from `Bool` to all `UIntX` and
`IntX` types.
Note that `Bool.toUInt64` already existed in previous versions of Lean.
This PR simplifies the implementation of `omega`.
When constructing the proof, `omega` is using MVars only for the purpose
of doing case analysis on `Or`. We can simplify the implementation a
fair bit if we just produce the proof directly using `Or.elim`.
While it didn’t yield the performance benefits I was hoping for, this
still seems a worthwhile simplification, now that we already have it.
This PR modifies the order of arguments for higher-order `Array`
functions, preferring to put the `Array` last (besides positional
arguments with defaults). This is more consistent with the `List` API,
and is more flexible, as dot notation allows two different partially
applied versions.
This PR changes the signature of `Array.get` to take a Nat and a proof,
rather than a `Fin`, for consistency with the rest of the (planned)
Array API. Note that because of bootstrapping issues we can't provide
`get_elem_tactic` as an autoparameter for the proof. As users will
mostly use the `xs[i]` notation provided by `GetElem`, this hopefully
isn't a problem.
We may restore `Fin` based versions, either here or downstream, as
needed, but they won't be the "main" functions.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
This PR changes the rule for which projections become instances. Before,
all parents along with all indirect ancestors that were represented as
subobject fields would have their projections become instances. Now only
projections for direct parents become instances.
Features:
- Only parents that are not ancestors of other parents get instances.
This allows "discretionary" indirect parents to be inserted for the
purpose of computing strict resolution orders when
`structure.strictResolutionOrder` is enabled, without having an impact
on typeclass synthesis.
- Non-subobject projections are now theorems if the parent is a
proposition. These are also no longer `@[reducible]`.
Closes#2905
This PR fixes `bv_decide`'s embedded constraint substitution to generate
correct counter examples in the corner case where duplicate theorems are
in the local context.
This PR introduces the and flattening pre processing pass from Bitwuzla
to `bv_decide`. It splits hypotheses of the form `(a && b) = true` into
`a = true` and `b = true` which has synergy potential with the already
existing embedded constraint substitution pass.
Beyond this I also added some profiling infra structure for the passes.
This PR adds a normalization rule to `bv_normalize` (which is used by
`bv_decide`) that converts `x / 2^k` into `x >>> k` under suitable
conditions. This allows us to simplify the expensive division circuits
that are used for bitblasting into much cheaper shifting circuits.
Concretely, it allows for the following canonicalization:
```lean
example {x : BitVec 16} : x / (BitVec.twoPow 16 2) = x >>> 2 := by bv_normalize
example {x : BitVec 16} : x / (BitVec.ofNat 16 8) = x >>> 3 := by bv_normalize
```
This PR changes the signature of `Array.set` to take a `Nat`, and a
tactic-provided bound, rather than a `Fin`.
Corresponding changes (but without the auto-param) for `Array.get` will
arrive shortly, after which I'll go more pervasively through the Array
API.
This PR is a follow-up to https://github.com/leanprover/lean4/pull/5609,
where we add lemmas characterizing `smtUDiv` and `smtSDiv`'s behavior
when the denominator is zero.
We build some `slt` theory, connecting it to `msb` for a clean proof. I
chose not to characterize `slt` in terms of `msb` a `simp` lemma, since
I anticipate use cases where we want to keep the arithmetic
interpretation of `slt`.
This PR removes
- a duplicate `MonadMCtx` instance in `MetavarContext.lean`
- `:= return ←` that I had left there accidentally in a previous PR.
- the unnecessary application of `mapMetaM` in `withTransparency`.
This PR adds a feature to the the mutual def elaborator where the
`instance` command yields theorems instead of definitions when the class
is a `Prop`.
Closes#5672
This PR adds configuration options for
`decide`/`decide!`/`native_decide` and refactors the tactics to be
frontends to the same backend. Adds a `+revert` option that cleans up
the local context and reverts all local variables the goal depends on,
along with indirect propositional hypotheses. Makes `native_decide` fail
at elaboration time on failure without sacrificing performance (the
decision procedure is still evaluated just once). Now `native_decide`
supports universe polymorphism.
Closes#2072
This now occurs for some large completions downstream of `import
Mathlib`. I'd like to get rid of this `whnf` call entirely in the
future, but this is a decent quick mitigation.
We add a new `Meta.instantiateMVars` trace node to the
`instantiateMVarsProfiling` definition used in `elabMutualDef`, and we
replace various uses of plain `instantiateMVars` with the profiled
version (which necessitated pulling up the definition to be higher in
the file).
This fixes a "time leak" when profiling large proofs, where
instantiating the goal metavariable can take a significant amount of
time, that previously would not be accounted for when using the trace
profiler.
This PR changes `bv_decide`'s configuration from lots of `set_option` to
an elaborated config like `simp` or `omega`. The notable exception is
`sat.solver` which is still a `set_option` such that users can configure
a custom SAT solver globally for an entire project or file. Additionally
it introduces the ability to set `maxSteps` for the simp preprocessing
run through the new config.
The latter feature was requested by people using `bv_decide` on SMTLIB
which has ginormous terms that exceed the default.
This PR verifies the `keys` function on `Std.HashMap`.
---
Initial discussions have already happend with @TwoFX and we are
collaborating on this matter.
This will remain a draft as long as not all desired results have been
added.
If we should still create an issue for the topic of this PR, let us
know.
Of course, any other feedback is appreciated as well :)
---------
Co-authored-by: Markus Himmel <markus@lean-fro.org>
Co-authored-by: monsterkrampe <monsterkrampe@users.noreply.github.com>
Co-authored-by: jt0202 <johannes.tantow@gmail.com>
Enables us to auto-generate the changelog from the list of PRs for a
modicum of summarizing/categorizing work on PR creation.
Does not (yet) allow external contributors to set category labels by
themselves as this creates issues with triggering one workflow from
another, it is not clear whether they should be allowed to create new
categories, and the reviewer/triage team likely is in a better position
to do the categorization anyway.
This PR names the default SizeOf instance `instSizeOfDefault`
I regularly have to debug termination checking failures where I end up
hovering over some termination measure, and seeing `instSizeOfDefault`
is more likely to tell me that the default instance is used than
`instSizeOf`.
This PR relates the operations `findSomeM?`, `findM?`, `findSome?`, and
`find?` on `Array` with the corresponding operations on `List`, and also
provides simp lemmas for the `Array` operations `findSomeRevM?`,
`findRevM?`, `findSomeRev?`, `findRev?` (in terms of `reverse` and the
usual forward find operations).
Bumps
[mymindstorm/setup-emsdk](https://github.com/mymindstorm/setup-emsdk)
from 12 to 14.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/mymindstorm/setup-emsdk/releases">mymindstorm/setup-emsdk's
releases</a>.</em></p>
<blockquote>
<h2>Version 14</h2>
<h1>Breaking Changes</h1>
<p>The default cache key naming scheme was changed from
<code>{Emscripten version}-{OS type}-${CPU architecture}-master</code>
to <code>{Github workflow name}-{Emscripten version}-{OS type}-${CPU
architecture}</code>. If <code>actions-cache-folder</code> is defined,
ensure that there are no conflicts with other caches to prevent
issues.</p>
<h1>Changelog</h1>
<ul>
<li>Add option to override cache key naming scheme (<a
href="https://redirect.github.com/mymindstorm/setup-emsdk/issues/20">#20</a>)</li>
<li>Add workflow name to cache key naming scheme (<a
href="https://redirect.github.com/mymindstorm/setup-emsdk/issues/20">#20</a>)</li>
<li>Updated dependencies to latest versions</li>
</ul>
<h2>Version 13</h2>
<ul>
<li>Updated to Node 20</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="6ab9eb1bda"><code>6ab9eb1</code></a>
v13 -> v14</li>
<li><a
href="bb630c3bf4"><code>bb630c3</code></a>
Update all dependencies to latest versions</li>
<li><a
href="74881103d0"><code>7488110</code></a>
Add workflow ID to cache key and cache key override option (<a
href="https://redirect.github.com/mymindstorm/setup-emsdk/issues/40">#40</a>)</li>
<li><a
href="d233ac12b0"><code>d233ac1</code></a>
v13</li>
<li><a
href="1749b22b40"><code>1749b22</code></a>
npm audit fix + update runtime to node20</li>
<li>See full diff in <a
href="https://github.com/mymindstorm/setup-emsdk/compare/v12...v14">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps
[nwtgck/actions-netlify](https://github.com/nwtgck/actions-netlify) from
2.0 to 3.0.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/nwtgck/actions-netlify/releases">nwtgck/actions-netlify's
releases</a>.</em></p>
<blockquote>
<h2>v3.0.0</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
<li>Updates the default runtime to node20</li>
</ul>
<h2>v2.1.0</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
</ul>
<h3>Added</h3>
<ul>
<li>Add "enable-github-deployment" input <a
href="https://redirect.github.com/nwtgck/actions-netlify/pull/901">#901</a>
by <a href="https://github.com/a-tokyo"><code>@a-tokyo</code></a></li>
</ul>
</blockquote>
</details>
<details>
<summary>Changelog</summary>
<p><em>Sourced from <a
href="https://github.com/nwtgck/actions-netlify/blob/develop/CHANGELOG.md">nwtgck/actions-netlify's
changelog</a>.</em></p>
<blockquote>
<h1>Changelog</h1>
<p>All notable changes to this project will be documented in this
file.</p>
<p>The format is based on <a
href="https://keepachangelog.com/en/1.0.0/">Keep a Changelog</a></p>
<h2>[Unreleased]</h2>
<h2>[3.0.0] - 2024-03-10</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
<li>Updates the default runtime to node20</li>
</ul>
<h2>[2.1.0] - 2023-08-18</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
</ul>
<h3>Added</h3>
<ul>
<li>Add "enable-github-deployment" input <a
href="https://redirect.github.com/nwtgck/actions-netlify/pull/901">#901</a>
by <a href="https://github.com/a-tokyo"><code>@a-tokyo</code></a></li>
</ul>
<h2>[2.0.0] - 2022-12-08</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
<li>Updates the default runtime to node16</li>
</ul>
<h2>[1.2.4] - 2022-10-14</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
</ul>
<h2>[1.2.3] - 2021-12-20</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
</ul>
<h2>[1.2.2] - 2021-05-08</h2>
<h3>Fixed</h3>
<ul>
<li>Fix GitHub deployment description</li>
</ul>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
</ul>
<h2>[1.2.1] - 2021-05-05</h2>
<h3>Added</h3>
<ul>
<li>Add "fails-without-credentials" input to fail if the
credentials not provided <a
href="https://redirect.github.com/nwtgck/actions-netlify/pull/532">#532</a></li>
</ul>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
</ul>
<h2>[1.2.0] - 2021-04-29</h2>
<h3>Changed</h3>
<ul>
<li>Update dependencies</li>
<li>(breaking change for <code>overwrites-pull-request-comment:
true</code>): Support multiple app deploys in a single PR <a
href="https://redirect.github.com/nwtgck/actions-netlify/pull/484">#484</a>
by <a
href="https://github.com/kaisermann"><code>@kaisermann</code></a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="4cbaf4c08f"><code>4cbaf4c</code></a>
Merge branch 'release/3.0.0'</li>
<li><a
href="6b45669baf"><code>6b45669</code></a>
bump: 3.0.0</li>
<li><a
href="8d5d80bf73"><code>8d5d80b</code></a>
Merge pull request <a
href="https://redirect.github.com/nwtgck/actions-netlify/issues/1151">#1151</a>
from nwtgck/actions-build/nwtgck-dependabot/npm_and_...</li>
<li><a
href="85c2e8e35a"><code>85c2e8e</code></a>
build</li>
<li><a
href="ea3c314fcd"><code>ea3c314</code></a>
Build(deps): bump <code>@actions/github</code> from 5.1.1 to 6.0.0</li>
<li><a
href="333815eadd"><code>333815e</code></a>
updates the default runtime to node20</li>
<li><a
href="06de7de77b"><code>06de7de</code></a>
Build(deps-dev): bump <code>@vercel/ncc</code> from 0.36.1 to 0.38.1
(<a
href="https://redirect.github.com/nwtgck/actions-netlify/issues/1121">#1121</a>)</li>
<li><a
href="a7f64ad4e2"><code>a7f64ad</code></a>
deps: update</li>
<li><a
href="fee801f039"><code>fee801f</code></a>
Build(deps): bump actions/setup-node from 3 to 4 (<a
href="https://redirect.github.com/nwtgck/actions-netlify/issues/1124">#1124</a>)</li>
<li><a
href="e4998d22a0"><code>e4998d2</code></a>
README.md, sample workflow: bump to latest action versions (<a
href="https://redirect.github.com/nwtgck/actions-netlify/issues/1149">#1149</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/nwtgck/actions-netlify/compare/v2.0...v3.0">compare
view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after
your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge
and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating
it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
New behavior: when in recovery mode, if any tactic fails in `all_goals`
then the metacontext is restored and all goals are admitted.
Without this, it can leave partially-solved metavariables and incomplete
goal lists.
Lake will now update a package's `lean-toolchain` file on `lake update`
if it finds the package's direct dependencies use a newer compatible
toolchain. To skip this step, use the `--keep-toolchain` CLI option.
Closes#2582. Closes#2752. Closes#5615.
### Toolchain update details
To determine "newest compatible" toolchain, Lake parses the toolchain
listed in the packages' `lean-toolchain` files into four categories:
release , nightly, PR, and other. For newness, release toolchains are
compared by semantic version (e.g., `"v4.4.0" < "v4.8.0"` and
`"v4.6.0-rc1" < "v4.6.0"`) and nightlies are compared by date (e.g.,
`"nightly-2024-01-10" < "nightly-2014-10-01"`). All other toolchain
types and mixtures are incompatible. If there is not a single newest
toolchain, Lake will print a warning and continue updating without
changing the toolchain.
If Lake does find a new toolchain, Lake updates the workspace's
`lean-toolchain` file accordingly and restarts the update process on the
new Lake. If Elan is detected, it will spawn the new Lake process via
`elan run` with the same arguments Lake was initially run with. If Elan
is missing, it will prompt the user to restart Lake manually and exit
with a special error code (4).
### Other changes
To implement this new logic, various other refactors were needed. Here
are some key highlights:
* Logs emitted during package and workspace loading are now eagerly
printed.
* The Elan executable used by Lake is now configurable by the `ELAN`
environment variable.
* The `--lean` CLI option was removed. Use the `LEAN` environment
variable instead.
* `Package.deps` / `Package.opaqueDeps` have been removed. Use
`findPackage?` with a dependency's name instead.
* The dependency resolver now uses a pure breadth-first traversal to
resolve dependencies. It also resolves dependencies in reverse order,
which is done for consistency with targets. Latter targets shadow
earlier ones and latter dependencies take precedence over earlier ones.
**These changes mean the order of dependencies in a Lake manifest will
change after the first `lake update` on this version of Lake.**
This introduces a notion of synthetic atoms into `bv_decide`'s
reflection framework. An atom can be declared synthetic if its behavior
is fully specified by additional lemmas that are added in the process of
creating it. This is for example useful in the code that handles `if` as
the entire `if` block is abstracted as an atom and then two lemmas to
describe either branch are added. Previously this had the effect of
creating error messages about potentially unsound counterexamples, now
the synthetic atoms get filtered from the counter example generation.
In patterns, ellipsis should always fill in each remaining argument as
an implicit argument, even if it is an optparam or autoparam. This
prevents examples such as the one in #4555 from failing:
```lean
match e with
| .internal .. => sorry
| .error .. => sorry
```
The `internal` constructor has an optparam (`| internal (id :
InternalExceptionId) (extra : KVMap := {})`).
We may consider having ellipsis suppress optparams and autoparams in
general. We avoid doing so for now since it's possible to opt-out of
them individually (for example with `.internal (extra := _) ..`) but
it's not possible to opt-in, and it is plausible that `..` with
optparams is useful in contexts such as the `refine` tactic. With
patterns however, it is hard to imagine a use case that offsets the
inconvenience of optparams being eagerly supplied.
Closes#4555
Following up #5928, updates the syntax for `omega` and `solve_by_elim`
and restores the syntax quotations in their implementations.
Following up #5898, uses the new tactic syntax in the library, replacing
all uses of `(config := ...)`.
The tactic elaborators match a too-restrictive syntax for the migration
to the new configuration syntax. This generalizes what they accept, and
the code will return to using quotations after a stage0 update and
syntax change.
Adds an optional `text` argument to the `fetchFile*` and `buildFile*`
definitions that can be used to hash built files as text files (with
normalized line endings) instead of as binary files (the previous
default).
Separately, this change also significantly expands the documentation in
the `Lake.Build.Trace` module and preforms minor touchups of some build
job signatures.
Simplifies the definition of `MapDeclarationExtension` so that it only
contains a `NameMap` without an additional `List (Name × α)`. Uses the
`NameMap`'s natural ordering during export rather than sorting.
This fixes issues from inserting into a `MapDeclarationExtension`
multiple times with the same key. Inside a module it appears that each
insertion overwrites the data, since those queries access the `NameMap`.
But across modules, only the first insertion is accessible, since each
insertion was actually pushed to the front of a `List`.
Mathlib needs this for a documentation extension feature, and [they are
considering a PR with a
workaround](https://github.com/leanprover-community/mathlib4/pull/17043)
that digs into the `MapDeclarationExtension` data structures.
As far as I can tell, the ability to pass a structure instance to a
deriving handler is not actually used in practice. It didn't seem to be
used in the test suite, at least.
Do we want to remove this, or do we want to use and document it? This PR
removes it, but that's not something I feel strongly about - but seeing
if it breaks Mathlib is a useful data point.
Example: Normally subtype notation pretty prints as `{ x // x > 0 }`,
but now the difference in domains is exposed:
```lean
example (h : {x : Int // x > 0}) : {x : Nat // x > 0} := h
/-
error: type mismatch
h
has type
{ x : Int // x > 0 } : Type
but is expected to have type
{ x : Nat // x > 0 } : Type
-/
```
Example:
```lean
example : 0 = (0 : Nat) := by
exact Eq.refl (0 : Int)
/-
error: type mismatch
Eq.refl 0
has type
(0 : Int) = 0 : Prop
but is expected to have type
(0 : Nat) = 0 : Prop
-/
```
`bv_normalize` would just silently drop other goals if called while not
focused on a singular goal, for example:
```lean
theorem mvarid (x y : Bool) (h : x ∨ y) : y ∨ x := by
cases h
bv_normalize
-- we want to write another bv_normalize here but all goals are gone
```
Would make the second subgoal disappear and then throw an error about
meta variables in the kernel.
There are many more lemmas about `foldlM`, so this may be useful for
reasoning about for loops by transforming them into folds.
The transformation includes accounting for monad effects, but does have
a mild performance difference in that short-circuiting on
`ForInStep.done` is replaced by traversing the rest of the list with a
noop.
Specializes the congr lemma generated for the `arg` conv tactic to only
rewrite the chosen argument. This makes it much more likely that the
chosen argument is able to be accessed.
Lets `arg` access the domain and codomain of pi types via `arg 1` and
`arg 2` in more situations. Upstreams `pi_congr` for this from mathlib.
Adds a negative indexing option, where `arg -2` accesses the
second-to-last argument for example, making the behavior of `lhs`
available to `arg`. This works for `enter` as well.
Other improvement: when there is an error in the `enter [...]` tactic,
individual locations get underlined with the error. The tactic info now
also is like `rw`, so you can see the intermediate conv states.
Closes#5871
PR #5883 added a new syntax for tactic configuration, and this PR
enables it in most tactics. Example: `simp +contextual`.
There will be followup PRs to modify the remaining ones.
Breaking change: Tactics that are macros for `simp` or other core
tactics need to adapt. The easiest way is to replace `(config)?` with
`optConfig` and then in the syntax quotations replace `$[$cfg]?` by
`$cfg:optConfig`. For tactics that manipulate the configuration, see
`erw` for an example:
```lean
macro "erw" c:optConfig s:rwRuleSeq loc:(location)? : tactic => do
`(tactic| rw $[$(getConfigItems c)]* (transparency := .default) $s:rwRuleSeq $(loc)?)
```
Configuration options are processed left-to-right, so this forces the
`transparency` to always be `.default`.
These implementations could be made more efficient by promoting them to
primitive operations, but I propose installing these in the meantime to
encourage users to avoid non-linearity problems.
* Now `getPathToBaseStructure?` can navigate to all parent structures,
not just through subobjects.
* Adds a "resolution order" for methods. This is the order that
generalized field notation visits parent structures when trying to
resolve names. The algorithm to compute a resolution order is the
commonly used C3 (used for instance by Python). By default we use a
relaxed version of the algorithm that tolerates inconsistencies. Using
`set_option structure.strictResolutionOrder true` makes inconsistent
parent orderings into warnings.
* This makes generalized field notation be able to resolve names for all
parent structures, not just those that are embedded as subobjects.
Closes#3467. (And addresses side note in #1881.)
* Modifies `getAllParentStructures` to return *all* parents. This
improves dot completion in the editor.
I'd previously added an instance from `ForIn'` to `ForIn`, but this then
caused some non-defeq duplication. It seems fine to just remove the
concrete `ForIn` instances in cases where the `ForIn'` instance exists
too. We can even remove a number of type-specific lemmas in favour of
the general ones.
Now that the elaborator supports primitive projections for recursive
inductive types (#5822), enable defining recursive inductive types with
the `structure` command, which was set up in #5842.
Example:
```lean
structure Tree where
n : Nat
children : Fin n → Tree
def Tree.size : Tree → Nat
| {n, children} => Id.run do
let mut s := 0
for h : i in [0 : n] do
s := s + (children ⟨i, h.2⟩).size
pure s
```
Note for kernel re-implementors: recursive structures are exercising the
kernel feature where primitive projections are valid for one-constructor
inductive types in general, so long as the structure isn't a `Prop` and
doesn't have any non-`Prop` fields, not just ones that are non-indexed
and non-recursive.
Closes#2512
The kernel supports primitive projections for all inductive types with
one construtor. The elaborator was assuming primitive projections only
work for "structure-likes", non-recursive inductive types with no
indices.
Enables numeric projection notation for general one-constructor
inductives.
Extracted from #5783.
Modifies the `structureExt` from being a `SimplePersistentEnvExtension`
to a `PersistentEnvExtension`. The simple version contains a `List` of
all added entries, which we do not need since we already have a
`PersistentHashMap` of them in the state. The oversight was that this
`List` contained duplicate entries due to `setStructureParents`
re-adding entries.
This hasn't affected release candidates or stables, but I realised that
I haven't been updating `LEAN_VERSION_MINOR` on `master` the last two
months, so it still says v4.12.0. This advances it to v4.14.0.
This PR adds a new syntax for tactic and command configurations. It also
updates the elaborator construction command to be able to process this
new syntax.
We do not update core tactics yet. Once tactics switch over to it,
rather than (for example) writing `simp (config := { contextual := true,
maxSteps := 22})`, one can write `simp +contextual (maxSteps := 22)`.
The new syntax is reverse compatible in the sense that `(config := ...)`
still sets the entire configuration.
Note to metaprogrammers: Use `optConfig` instead of `(config)?`. The
elaborator generated by `declare_config_elab` accepts both old and new
configurations. The elaborator has also been written to be tolerant to
null nodes, so adapting to `optConfig` should be as easy as changing
just the syntax for your tactics and deleting `mkOptionalNode`.
Breaking change: The new system is mostly reverse compatible, however
the type of the generated elaborator now lands in `TacticM` to make use
of the current recovery state. Commands that wish to elaborate
configurations should now use `declare_command_config_elab` instead of
`declare_config_elab` to get an elaborator landing in `CommandElabM`.
This command comes from Lean 3, which I had previously ported and
contributed to Batteries (née Std). In this new version, `#where`
produces actual command Syntax for all features of a top-level scope
(rather than splicing together strings), and it also now reports
included variables.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
This adds the embedded constraint substitution preprocessing pass from
Bitwuzla to `bv_decide`.
It looks for hypotheses of the form `h : x = true` and then attempts to
find occurrences of
`x` within other hypotheses to replace them with true.
Fixes a serious issue where Lake would delete path dependencies when
attempting to cleanup a dependency required with an incorrect name.
Closes#5876. Originally part of #5684, but also independently
discovered by François.
Makes `MessageData.ofConstName` available without needing to import the
pretty printer. Any code making use of `MessageData` can write `m!" ...
{.ofConstName n} ... "` to have the name print with hover information.
More error messages now have hover information.
* Now `.ofConstName` also has a boolean flag to make names print fully
qualified. Default: false.
* Now `.ofConstName` will sanitize names that aren't constants. It is OK
to use it in `"unknown constant '{.ofConstName constName}'"` errors.
Usability note: it is more user-friendly to have "has already been
declared" errors report the fully qualified name. For this, write
`m!"{.ofConstName n true} has already been declared"`.
An important part of the interface of a function is the parameter names,
for making used of named arguments. This PR makes the parameter names
print in a reliable way. The parameters of the type now appear as
hygienic names if they cannot be used as named arguments.
Modifies the heuristic for how parameters are chosen to appear before or
after the colon. The rule is now that parameters start appearing after
the colon at the first non-dependent non-instance-implicit parameter
that has a name unusable as a named argument. This is a refinement of
#2846.
Fixes the issue where consecutive hygienic names pretty print without a
space separating them, so we now have `(x✝ y✝ : Nat)` rather than `(x✝y✝
: Nat)`.
Breaking change: `Lean.PrettyPrinter.Formatter.pushToken` now takes an
additional boolean `ident` argument, which should be `true` for
identifiers. Used to insert discretionary space between consecutive
identifiers.
Closes#5810
This adds the ability to add the converse direction of a rewrite rule
not just in simp arguments `simp [← thm]`, but also as a global
attribute
```lean
attribute [simp ←] thm
```
This fixes#5828.
This can be undone with `attribute [-simp]`, although note that
`[-simp]` wins and cannot be undone at the moment (#5868).
Like `simp [← thm]` (see #4290), this will do an implicit `attribute
[-simp] thm` if the other direction is already defined.
Type mismatch errors have a nice feature where expressions are annotated
with `pp.explicit` to expose differences via `isDefEq` checking.
However, this procedure has side effects since `isDefEq` may assign
metavariables. This PR wraps the procedure with `withoutModifyingState`
to prevent assignments from escaping.
Assignments can lead to confusing behavior. For example, in the
following a higher-order unification fails, but the difference-finding
procedure unifies metavariables in a naive way, producing a baffling
error message:
```lean
theorem test {f g : Nat → Nat} (n : Nat) (hfg : ∀a, f (g a) = a) :
f (g n) = n := hfg n
example {g2 : ℕ → ℕ} (n2 : ℕ) : (λx => x * 2) (g2 n2) = n2 := by
with_reducible refine test n2 ?_
/-
type mismatch
test n2 ?m.648
has type
(fun x ↦ x * 2) (g2 n2) = n2 : Prop
but is expected to have type
(fun x ↦ x * 2) (g2 n2) = n2 : Prop
-/
```
With the change, it now says `has type ?m.153 (?m.154 n2) = n2`.
Note: this uses `withoutModifyingState` instead of `withNewMCtxDepth`
because we want to know something about where `isDefEq` failed — we are
trying to simulate a very basic version of `isDefEq` for function
applications, and we want the state at the point of failure to know
which argument is "at fault".
Modifies `simp` to elaborate all simp arguments without disabling error
recovery. Like in #4177, simp arguments with elaboration errors are not
added to the simp set. Error recovery is still disabled when `simp` is
used in combinators such as `first`.
This enables better term info and features like tab completion when
there are elaboration errors.
Also included is a fix to the `all_goals` and `<;>` tactic combinators.
Recall that `try`/`catch` for the Tactic monad restores the state on
failure. This meant that all messages were being cleared on tactic
failure. The fix is to use `Tactic.tryCatch` instead, which doesn't
restore state.
Part of addressing #3831Closes#4888
The assumptions behind disabling error recovery for the `apply` tactic
no longer seem to hold, since tactic combinators like `first` themselves
disable error recovery when it makes sense.
This addresses part of #3831
Breaking change: `elabTermForApply` no longer uses `withoutRecover`.
Tactics using `elabTermForApply` should evaluate whether it makes sense
to wrap it with `withoutRecover`, which is generally speaking when it's
used to elaborate identifiers.
Makes the error messages report on RHSs and LHSs that do not match the
expected values when the relations are defeq. If the relations are not
defeq, the error message now no longer mentions the value of the whole
`calc` expression.
Adds a field to `mkCoe` with an optional callback to use to generate
error messages.
Note: it is tempting to try to make use of expected types when
elaborating the `calc` expression, but this runs into issue #2073.
Closes#4318
Adds ability to chain congruence lemmas when a function's arity is less
than the number of supplied arguments. This improves `congr` as well as
all conv tactics implemented using `congr`, like `arg` and `enter`.
(The non-conv `congr` tactic still needs to be fixed.)
Toward #2942.
Followup to #5841. Makes the `structure` command populate the new
`parentInfo` field with all the structures in the `extends` clause.
This will require a stage0 update to fully take effect.
Breaking change: now it's a warning if a structure extends a parent
multiple times.
Breaking change: now `getParentStructures` is `getStructureSubobjects`.
Adds `getStructureParentInfo` for getting all the immediate parents.
Note that the set of subobjects is neither a subset nor a superset of
the immediate parents.
Closes#1881
This default instance makes it possible to write things like `m!"the
constant is {.ofConstName n}"`.
Breaking change: This weakly causes terms to have a type of
`MessageData` if their type is otherwise unknown. For example:
* `m!"... {x} ..."` can cause `x` to have type `MessageData`, causing
the `let` definition of `x` to fail to elaborate. Fix: give `x` an
explicit type.
* Arithmetic expressions in `m!` strings may need a type ascription. For
example, if the type of `i` is unknown at the time the arithmetic
expression is elaborated, then `m!"... {i + 1} ..."` can fail saying
that it cannot find an `HAdd Nat Nat MessageData` instance. Two fixes:
either ensure that the type of `i` is known, or add a type ascription to
guide the `MessageData` coercion, like `m!"... {(i + 1 : Nat)} ..."`.
Using the same strategy as #5852 this provides `bv_decide` support for
`Bool` and `BitVec` ifs
this in turn instantly enables support for:
- `sdiv`
- `smod`
- `abs`
and thus closes our last discrepancies to QF_BV!
This is the first step towards fixing the issue of not having mutual
recursion between the `Bool` and `BitVec` fragment of `QF_BV` in
`bv_decide`. This PR adds support for `BitVec.ofBool` by doing the
following:
1. Introduce a new mechanism into the reification engine that allows us
to add additional lemmas to the top level on the fly as we are
traversing the expression tree.
2. If we encounter an expression `BitVec.ofBool boolExpr` we reify
`boolExpr` and then abstract `BitVec.ofBool boolExpr` as some atom `a`
3. We add two lemmas `boolExpr = true -> a = 1#1` and `boolExpr = false
-> a = 0#1`. This mirrors the full behavior of `BitVec.ofBool` and thus
makes our atom `a` correctly interpreted again.
In order to do the reification in step 2 mutual recursion in the
reification engine is required. For this reason I started pulling out
logic from the, now rather large, mutual block into other files and
document the invariants that they assume explicitly.
A step of expanding structure instances is to determine all the default
values, and part of this is reducing projections that appear in the
default values so that they get replaced with the user-provided values.
Binder types in foralls, lambdas, and lets have to be reduced too.
Closes#2186
Refactors the `structure` command to support recursive structures. These
are disabled for now, pending additional elaborator support in #5822.
This refactor is also a step toward `structure` appearing in `mutual`
blocks.
Error reporting is now more precise, and this fixes an issue where
general errors could appear on the last field. Adds "don't know how to
synthesize placeholder" errors for default values.
Closes#2512
This adds a `parentInfo` field to the `StructureInfo`, which will
eventually be populated with the actual parents of a structure. This is
work toward #1881. Also documents fields of the structure info data
structures.
Requires a stage0 update before the next steps.
`generalize ... at *` sometimes will try to modify the recursive
hypothesis corresponding to the current theorem being defined, which may
not be the expected behaviour. It should only try to `generalize`
hypotheses that it can actually modify and are visible, not
implementation details. Otherwise this means that there are
discrepancies between `generalize ... at *` and `generalize ... at H`,
even though `H` is the only hypothesis in the context.
This commit uses `getLocalHyps` instead of `getFVarIds` to get the
current valid `FVarIds` in the context. This uses
`isImplementationDetail` to filter out `FVarIds` that are implementation
details in the context and are not visible to the user and should not be
manipulated by `generalize`.
Closes#4845
Closes#3146
Reduction doesn't trigger correctly on the bodies of `let`-expressions
in `StructInst`, leading some meta-variables to linger in the terms of
some fields. Because of this, default fields may try multiple times (and
fail) to be generated, leading to an unexpected error.
The solution implemented here is to modify the values of the introduced
variables in the local context so as to reduce them correctly.
The `liftCommandElabM : CommandElabM α -> CoreM α` function now carries
over macro scopes, the name generator, info trees, and messages.
Adds a flag `throwOnError`, which is true by default. When it is true,
then if the messages contain an error message, it is converted into an
exception. In this case, the infotrees and messages are not carried
over; the motivation is that `throwOnError` is likely used for synthetic
syntax, and so the info and messages on errors will just be noise.
Cleanup of #5650
* default `Modifiers.stx` to missing
* rename and clarify `addDeclarationRangesFromSyntax` as the main
convenience function for user metaprograms
Add an example Lean file that includes an unusually large definition
that takes a long time to elaborate.
It may be that it's difficult to process it more efficiently, but
perhaps someone will discover a way to improve it if it's in the
benchmark suite. Improved performance on this benchmark will likely make
some program analysis and verification tasks within Lean more feasible.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Example new output:
```text
failed to compile 'partial' definition 'checkMyList', could not prove that the type
ListNode → Bool × ListNode
is nonempty.
This process uses multiple strategies:
- It looks for a parameter that matches the return type.
- It tries synthesizing 'Inhabited' and 'Nonempty' instances for the return type.
- It tries unfolding the return type.
If the return type is defined using the 'structure' or 'inductive' command, you can try
adding a 'deriving Nonempty' clause to it.
```
The inhabitation prover now also unfolds definitions when trying to
prove inhabitation. For example,
```lean
def T (α : Type) := α × α
partial def f (n : Nat) : T Nat := f n
```
Motivated [by
Zulip](https://leanprover.zulipchat.com/#narrow/channel/113489-new-members/topic/Why.20return.20type.20of.20partial.20function.20MUST.20.60inhabited.60.3F/near/477905312)
Refactors `inductive` elaborator to keep track of universe level
parameters created during elaboration of `variable`s and binders. This
fixes an issue in Mathlib where its `Type*` elaborator can result in
unexpected universe levels.
For example, in
```lean4
variable {F : Type*}
inductive I1 (A B : Type*) (x : F) : Type
```
before this change the signature would be
```
I1.{u_1, u_2} {F : Type u_1} (A : Type u_1) (B : Type u_2) (x : F) : Type
```
but now it is
```
I1.{u_1, u_2, u_3} {F : Type u_1} (A : Type u_2) (B : Type u_3) (x : F) : Type
```
Fixes this for the `axiom` elaborator too.
Adds more accurate universe level validation for mutual inductives.
Breaking change: removes `Lean.Elab.Command.expandDeclId`. Use
`Lean.Elab.Term.expandDeclId` from within `runCommandElabM`.
Breaking changes:
To build Lean from source on Windows, it is now necessary to install the
[Windows
SDK](https://developer.microsoft.com/en-us/windows/downloads/windows-sdk/).
The build instructions have been updated to reflect this. Note that the
Windows SDK is **not** needed to compile Lean programs using a Lean
toolchain obtained using `elan`. The Windows SDK is only needed to build
Lean itself from source.
Furthermore, we are dropping support for Windows versions older than
Windows 10 1903 (released in May 2019).
No Windows version that is still supported by Microsoft as part of
mainstream support is affected by this.
The following Windows versions are still supported by Microsoft as part
of commercial extended support but are no longer supported by Lean:
- Windows 10 Enterprise LTSC 2015
- Windows 10 Enterprise LTSC 2016
- Windows 10 Enterprise LTSC 2019
- Windows Server 2019
It's difficult to distinguish theorems from regular definitions in the
completion menu, which is annoying when using completion for searching
one or the other. This PR makes theorem completions use the "Eureka!"
icon ()
to distinguish them more clearly from other completions.
NB: We are very limited in terms of which icons we can pick here since
[the completion kinds provided by LSP / VS
Code](https://code.visualstudio.com/docs/editor/intellisense#_types-of-completions)
are optimized for object-oriented programming languages, but I think
this choice strikes a nice balance between being easy to identify,
having some visual connection to theorem proving and not being used a
lot in other languages and thus not clashing with pre-existing
associations.
Between #3106 and this, it was possible that reparsing the file up to
the current position was stuck waiting in the threadpool queue,
displaying a yellow bar and not displaying any info on the unchanged
prefix.
`instantiate_mvars` is now implemented in C/C++, and makes many calls to
`has_fvar`, `has_mvar`. The new C/C++ implementations are inlined and
avoid unnecessary RC inc/decs.
Previously `RecursorVal.getInduct` would return the prefix of the
recursor’s name, which is unlikely the right value for the “derived”
recursors in nested recursion. The code using `RecursorVal.getInduct`
seems to expect the name of the inductive type of major argument here.
If we return that name, this fixes#5661.
This bug becomes more visible now that we have structural mutual
recursion.
Also, to avoid confusion, renames the function to ``getMajorInduct`.
This PR simplifies the signature of `Array.mapIdx`, to take a function
`f : Nat \to \a \to \b` rather than a function `f : Fin as.size \to \a
\to \b`.
Lean doesn't actually use the extra generality anywhere (so in fact this
change *simplifies* all the call sites of `Array.mapIdx`, since we no
longer need to throw away the proof).
This change would make the function signature equivalent to
`List.mapIdx`, hence making it easier to write verification lemmas.
We keep the original behaviour as `Array.mapFinIdx`.
This replaces `export Lean (Name NameMap)` and `export System
(SearchPath FilePath)` with the relevant `open` commands. This fixes
docgen output so that it can refer to, for example, `Lean.Name` instead
of `Lake.Name`.
The reason for these `export`s was convenience: by doing `open Lake` you
could get these aliases for free. However, aliases affect pretty
printing, and the Lake aliases took precedence. We don't want to disable
pretty printing re-exported names because this can be a valid pattern
(names could incrementally get re-exported from namespace to parent
namespace).
In the future we might implement a feature to be able to `scoped open`
some names.
Breaking change: Lakefiles that refer to `FilePath` may need to change
this to `System.FilePath` or otherwise add `open System (FilePath)`.
Closes#2524
This PR resolves the following issues related to goal state display:
1. In a new line after a `case` tactic with a completed proof, the state
of the proof in the `case` would be displayed, not the proof state after
the `case`
1. In the range of `next =>` / `case' ... =>`, the state of the proof in
the corresponding case would not be displayed, whereas this is true for
`case`
1. In the `suffices ... by` tactic, the tactic state of the `by` block
was not displayed after the `by` and before the first tactic
The incorrect goal state after `case` was caused by `evalCase` adding a
`TacticInfo` with the full block proof state for the full range of the
`case` block that the goal state selection has no means of
distinguishing from the `TacticInfo` with the same range that contains
the state after the whole `case` block. Narrowing the range of this
`TacticInfo` to `case ... =>` fixed this issue.
The lack of a case proof state on `next =>` was caused by the `case`
syntax that `next` expands to receiving noncanonical synthetic
`SourceInfo`, which is usually ignored by the language server. Adding a
token antiquotation for `next` fixed this issue.
The lack of a case proof state on `case' ... =>` was caused by
`evalCase'` not adding a `TacticInfo` with the full block state to the
range of `case' ... =>`. Adding this `TacticInfo` fixed this issue.
The tactic state of the block not being displayed after the `by` was
caused by the macro expansion of `suffices` to `have` not transferring
the trailing whitespace of the `by`. Ensuring that this trailing
whitespace information is transferred fixed this issue.
Fixes#2881.
Should ensure we visit at most as many expr nodes as in the final expr
instead of many possibly overlapping mvar assignments. This is likely
the only way we can ensure acceptable performance in all cases.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
this option was added in fb97275dcb to
prepare for #4595, due to boostrapping issues, but #4595 has not landed
yet. This is be very confusing when people discover this option and try
to use it (as I did).
So let's clearly mark this as not yet implemented on `master`, and add
the
docstring only with #4595.
Since `getMsbD_add`, `getMsbD_sub`, `getLsbD_sub`, `msb_sub` , `msb_add`
depend on `getLsbD_add` (which lives in`BitBlast.lean`) and on each
other, I put all of these in `BitBlast.lean`.
It bothered me that inferring instances of the shape `Decidable (∀ (x : Fin _), _)`
will go linearly through all instances of that shape, even those that are
about `∀ (x : Nat), …`. And that `Decidable (∃ (x : Fin _), _)` gets better
indexing than `Decidable (∀ (x : Fin _), _)`.
Judging from code comments, the discr tree used to index arrow types
with two arguments (domain and body), and that led to bugs due to the
dependency, so the arguments were removed. But it seems that indexing
the domain is completely simple and innocent.
So let’s see what happens…
Mostly only insignificant perf improvements, unfortunately (~Mathlib.Data.Matroid.IndepAxioms — instructions -11.4B, overall build instructions -0.097 %):
http://speed.lean-fro.org/mathlib4/compare/dd333cc1-fa26-42f2-96c6-b0e66047d0b6/to/6875ff8f-a17c-431d-8b8b-2f00799be794
This is just a small baby step compared to the more invasive improvements
done in the [`RefinedDiscrTree` by J. W. Gerbscheid](https://leanprover-community.github.io/mathlib4_docs/Mathlib/Tactic/FunProp/RefinedDiscrTree.html) in mathlib.
I made a few choices so far that can probably be discussed:
- got rid of `modn` on `UInt`, nobody seems to use it apart from the
definition of `shift` which can use normal `mod`
- removed the previous defeq optimized definition of `USize.size` in
favor for a normal one. The motivation was to allow `OfNat` to work
which doesn't seem to be necessary anymore afaict.
- Minimized uses of `.val`, should we maybe mark it deprecated?
- Mostly got rid of `.val` in basically all theorems as the proper next
level of API would now be `.toBitVec`. We could probably re-prove them
but it would be more annoying given the change of definition.
- Did not yet redefine `log2` in terms of `BitVec` as this would require
a `log2` in `BitVec` as well, do we want this?
- I added a couple of theorems around the relation of `<` on `UInt` and
`Nat`. These were previously not needed because defeq was used all over
the place to save us. I did not yet generalize these to all types as I
wasn't sure if they are the appropriate lemma that we want to have.
Adds `pushGoal`/`pushGoals` and `popGoal` for manipulating the goal
state. These are an alternative to `replaceMainGoal` and `getMainGoal`,
and with them you don't need to worry about making sure nothing clears
assigned metavariables from the goal list between assigning the main
goal and using `replaceMainGoal`.
Modifies `closeMainGoalUsing`, which is like a `TacticM` version of
`liftMetaTactic`. Now the callback is run in a context where the main
goal is removed from the goal list, and the callback is free to modify
the goal list. Furthermore, the `checkUnassigned` argument has been
replaced with `checkNewUnassigned`, which checks whether the value
assigned to the goal has any *new* metavariables, relative to the start
of execution of the callback. This API is sufficient for the `exact`
tactic for example.
Modifies `withCollectingNewGoalsFrom` to take the `parentTag` argument
explicitly rather than indirectly via `getMainTag`. This is needed when
used under `closeMainGoalUsing`.
Modifies `elabTermWithHoles` to optionally take `parentTag?`. It
defaults to `getMainTag` if it is `none`.
Renames `Tactic.tryCatch` to `Tactic.tryCatchRestore`, and adds a
`Tactic.tryCatch` that doesn't do backtracking.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
`getNumHeadForalls` and `getNumHeadLambdas` were both duplicated
downstream with different names; I'll clean up those next.
Also adds `getAppNumArgs'`.
it seems to be unused, arguably even for kernel recursors their type
should be usable with `mkRecursorInfo`, and removing this will help
understand the impact of #5679.
Mathlib has a duplicate of this instance as `Quotient.decidableEq` (with
the same implementation) and refers to it by name a few times, so let's
just rename our version to the mathlib name so that the copy in mathlib
can be dropped.
This takes a few standalone bitvector problems, about inequalties, from
LNSym, and adds them as a benchmark to prevent further regressions with
bv_decide.
These problems are particularly interesting, because they've previously
had a bad interaction with bv_decides normalization pass, see
https://github.com/leanprover/lean4/issues/5664.
---------
Co-authored-by: Henrik Böving <hargonix@gmail.com>
Projects like mathlib like to define projection functions with extra
structure, for example one could imagine defining `Multiset.card :
Multiset α →+ Nat`, which bundles the fact that `Multiset.card (m1 + m2)
= Multiset.card m1 + Multiset.card m2` for all `m1 m2 : Multiset α`. A
problem though is that so far this has prevented dot notation from
working: you can't write `(m1 + m2).card = m1.card + m2.card`.
With this PR, now you can. The way it works is that "LValue resolution"
will apply CoeFun instances when trying to resolve which argument should
receive the object of dot notation.
A contrived-yet-representative example:
```lean
structure Equiv (α β : Sort _) where
toFun : α → β
invFun : β → α
infixl:25 " ≃ " => Equiv
instance: CoeFun (α ≃ β) fun _ => α → β where
coe := Equiv.toFun
structure Foo where
n : Nat
def Foo.n' : Foo ≃ Nat := ⟨Foo.n, Foo.mk⟩
variable (f : Foo)
#check f.n'
-- Foo.n'.toFun f : Nat
```
Design note 1: While LValue resolution attempts to make use of named
arguments when positional arguments cannot be used, when we apply CoeFun
instances we disallow making use of named arguments. The rationale is
that argument names for CoeFun instances tend to be random, which could
lead dot notation randomly succeeding or failing. It is better to be
uniform, and so it uniformly fails in this case.
Design note 2: There is a limitation in that this will *not* make use of
the values of any of the provided arguments when synthesizing the CoeFun
instances (see the tests for an example), since argument elaboration
takes place after LValue resolution. However, we make sure that
synthesis will fail rather than choose the wrong CoeFun instance.
Performance note: Such instances will be synthesized twice, once during
LValue resolution, and again when applying arguments.
This also adds in a small optimization to the parameter list computation
in LValue resolution so that it lazily reduces when a relevant parameter
hasn't been found yet, rather than using `forallTelescopeReducing`. It
also switches to using `forallMetaTelescope` to make sure the CoeFun
synthesis will fail if multiple instances could apply.
Getting this to pretty print will be deferred to future work.
Closes#1910
Gives more control over pretty printing metavariables.
- When `pp.mvars.levels` is false, then universe level metavariables
pretty print as `_` rather than `?u.22`
- When `pp.mvars.anonymous` is false, then anonymous metavariables
pretty print as `?_` rather than `?m.22`. Named metavariables still
pretty print with their names. When this is false, it also sets
`pp.mvars.levels` to false, since every level metavariable is anonymous.
- When `pp.mvars` is false, then all metavariables pretty print as `?_`
or `_`.
Modifies TryThis to use `pp.mvars.anonymous` rather than doing a
post-delaboration modification. This incidentally improves TryThis since
it now prints universe level metavariables as `_` rather than `?u.22`.
We trust that the users read the error messages or tactic docs to
discover the option.
AWS problems have shown that this can be too eager of an operation to
do.
Given that we have the luxury of interactivity let's go for an approach
where the users
can optionally enable it.
This PR ensures that deprecated declarations are displayed with a
strikethrough markup in the completion popup of VS Code and that the
docstring of a completion item denotes the meta-data of the deprecation.
These lemmas are peeled from `leanprover/lnsym`.
Moreover, note that these lemmas only hold when we do not have overflow
in their operands, and thus, we are able to treat the operands as if
they were 'regular' natural numbers.
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
Co-authored-by: Kim Morrison <scott@tqft.net>
Divison proofs are more likely to depend on add/sub/mul proofs than the
other way around. This cleans up
https://github.com/leanprover/lean4/pull/5609, which added division
proofs that rely on negation to already be defined.
Closes#5682
- Removes the broken `-f` flag from the help message which doesn't
behave as expected as an alternative to `--features`.
- Adds the `-g` flag to the help message which is a working alternative
to the `--githash` flag.
Lake will now only automatically fetch Reservoir build caches for
package in the the `leanprover` and `leanprover-community`
organizations. We are not planning to expand the Reservoir build cache
to other packages until farther in the future.
Makes `#eval` use the `elabMutualDef` machinery to process all the `let
rec`s that might appear in the expression. This now works:
```lean
#eval
let rec fact (n : Nat) : Nat :=
match n with
| 0 => 1
| n' + 1 => n * fact n'
fact 5
```
Closes#2374
The `decide!` tactic is like `decide`, but when it tries reducing the
`Decidable` instance it uses kernel reduction rather than the
elaborator's reduction.
The kernel ignores transparency, so it can unfold all definitions (for
better or for worse). Furthermore, by using kernel reduction we can
cache the result as an auxiliary lemma — this is more efficient than
`decide`, which needs to reduce the instance twice: once in the
elaborator to check whether the tactic succeeds, and once again in the
kernel during final typechecking.
While RFC #5629 proposes a `decide!` that skips checking altogether
during elaboration, with this PR's `decide!` we can use `decide!` as
more-or-less a drop-in replacement for `decide`, since the tactic will
fail if kernel reduction fails.
This PR also includes two small fixes:
- `blameDecideReductionFailure` now uses `withIncRecDepth`.
- `Lean.Meta.zetaReduce` now instantiates metavariables while zeta
reducing.
Some profiling:
```lean
set_option maxRecDepth 2000
set_option trace.profiler true
set_option trace.profiler.threshold 0
theorem thm1 : 0 < 1 := by decide!
theorem thm1' : 0 < 1 := by decide
theorem thm2 : ∀ x < 400, x * x ≤ 160000 := by decide!
theorem thm2' : ∀ x < 400, x * x ≤ 160000 := by decide
/-
[Elab.command] [0.003655] theorem thm1 : 0 < 1 := by decide!
[Elab.command] [0.003164] theorem thm1' : 0 < 1 := by decide
[Elab.command] [0.133223] theorem thm2 : ∀ x < 400, x * x ≤ 160000 := by decide!
[Elab.command] [0.252310] theorem thm2' : ∀ x < 400, x * x ≤ 160000 := by decide
-/
```
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
Deprecates `inductive ... :=`, `structure ... :=`, and `class ... :=` in
favor of the `... where` variant. Currently this syntax produces a
warning, controlled by the `linter.deprecated` option.
Breaking change: modifies `Lean.Linter.logLintIf` to use
`Lean.Linter.getLinterValue` to determine if a linter value is set. This
means that the `linter.all` option now is taken into account when the
linter option is not set.
Part of #5236
This PR enables tactic completion in the whitespace of a tactic proof
and adds tactic docstrings to the completion menu.
Future work:
- A couple of broken tactic completions: This is due to tactic
completion now using @david-christiansen's `Tactic.Doc.allTacticDocs` to
obtain the tactic docstrings and should be fixed soon.
- Whitespace tactic completion in tactic combinators: This requires
changing the syntax of tactic combinators to produce a syntax node that
makes it clear that a tactic is expected at the given position.
Closes#1651.
When named arguments introduce eta arguments, the full application
contains fvars for these eta arguments, so `MVarErrorKind.implicitArg`
needs to keep a local context for its error messages. This is because
the local context of the mvar associated to the `MVarErrorKind` is not
sufficient, since when an eta argument come after an implicit argument,
the implicit argument's mvar doesn't contain the eta argument's fvar in
its local context.
Closes#5475
Now one can write `@x.f`, `@(x).f`, `@x.1`, `@(x).1`, and so on.
This fixes an issue where structure instance update notation (like `{x
with a := a'}`) could fail if the field `a` had a type with implicit,
optional, or auto parameters.
Closes#5406
Fixes#5565, by using tags instead of trying to string match on a
`MessageData`. This ends up reverting some unwanted test output changes
from #4781 too.
This changes `isMaxRecDepth` for good measure too.
This was a regression in Lean 4.11.0, so may be worth backporting to
4.12.x, if not also 4.11.x.
A Lake build of target within a a package will no longer build a
package's dependencies package-level extra targets dependencies. At the
technical level, a package's `extraDep` facet no longer transitively
builds its dependencies' `extraDep` facet.
Closes#5633.
Closes#5634. Before assigning the simplified `using` clause expression
to the goal, this adds a check that the expression has no new
metavariables. It also adjusts how new hypotheses are added to the goal
to prevent spurious "don't know how to synthesize placeholder" errors on
that goal metavariable. We also throw in an occurs check immediately
after elaboration to avoid some counterintuitive behavior when
simplifying such a term closes the goal.
Closes#4101. This also improves the type mismatch error message,
showing the elaborated `using` clause rather than `h✝`:
```lean
example : False := by
simpa using (fun x : True => x)
/-
error: type mismatch, term
fun x => x
after simplification has type
True : Prop
but is expected to have type
False : Prop
-/
```
A `Prop`-valued inductive type is a syntactic subsingleton if it has at
most one constructor and all the arguments to the constructor are in
`Prop`. Such types have large elimination, so they could be defined in
`Type` or `Prop` without any trouble, though users tend to expect that
such types define a `Prop` and need to learn to insert `: Prop`.
Currently, the default universe for types is `Type`. This PR adds a
heuristic: if a type is a syntactic subsingleton with exactly one
constructor, and the constructor has at least one parameter, then the
`inductive` command will prefer creating a `Prop` instead of a `Type`.
For `structure`, we ask for at least one field.
More generally, for mutual inductives, each type needs to be a syntactic
subsingleton, at least one type must have one constructor, and at least
one constructor must have at least one parameter. The motivation for
this restriction is that every inductive type starts with a zero
constructors and each constructor starts with zero fields, and
stubbed-out types shouldn't be `Prop`.
Thanks to @arthur-adjedj for the investigation in #2695 and to @digama0
for formulating the heuristic.
Closes#2690
This refactors and improves the `#eval` command, introducing some new
features.
* Now evaluated results can be represented using `ToExpr` and pretty
printing. This means **hoverable output**. If `ToExpr` fails, it then
tries `Repr` and then `ToString`. The `eval.pp` option controls whether
or not to try `ToExpr`.
* There is now **auto-derivation** of `Repr` instances, enabled with the
`pp.derive.repr` option (default to **true**). For example:
```lean
inductive Baz
| a | b
#eval Baz.a
-- Baz.a
```
It simply does `deriving instance Repr for Baz` when there's no way to
represent `Baz`. If core Lean gets `ToExpr` derive handlers, they could
be used here as well.
* The option `eval.type` controls whether or not to include the type in
the output. For now the default is false.
* Now things like `#eval do return 2` work. It tries using
`CommandElabM`, `TermElabM`, or `IO` when the monad is unknown.
* Now there is no longer `Lean.Eval` or `Lean.MetaEval`. These each used
to be responsible for both adapting monads and printing results. The
concerns have been split into two. (1) The `MonadEval` class is
responsible for adapting monads for evaluation (it is similar to
`MonadLift`, but instances are allowed to use default data when
initializing state) and (2) finding a way to represent results is
handled separately.
* Error messages about failed instance synthesis are now more precise.
Once it detects that a `MonadEval` class applies, then the error message
will be specific about missing `ToExpr`/`Repr`/`ToString` instances.
* Fixes a bug where `Repr`/`ToString` instances can't be found by
unfolding types "under the monad". For example, this works now:
```lean
def Foo := List Nat
def Foo.mk (l : List Nat) : Foo := l
#eval show Lean.CoreM Foo from do return Foo.mk [1,2,3]
```
* Elaboration errors now abort evaluation. This eliminates some
not-so-relevant error messages.
* Now evaluating a value of type `m Unit` never prints a blank message.
* Fixes bugs where evaluating `MetaM` and `CoreM` wouldn't collect log
messages.
The `run_cmd`, `run_elab`, and `run_meta` commands are now frontends for
`#eval`.
This verifies a bit hack from here:
https://en.wikipedia.org/wiki/Lehmer_random_number_generator#Sample_C99_code
I previously ran the SMTLIB equivalent this with Bitwuzla in my crypto
class and got the following numbers:
- 22s with Bitwuzla
- Z3 and CVC5 don't yet terminate after > 2min
Now with`bv_decide` the overall timing is 33.7s, consisting of:
- 5s of checking the LRAT cert
- 5s of trimming the LRAT cert from 800k to 300k proof steps
- remainder actual solving time
So running `bv_decide` like a normal SMT solver without verifying the
result of the SAT solver would yield approximately ~24s.
Where before we had
```lean
#check fun x : Nat => ?a
-- fun x ↦ ?m.7 x : (x : Nat) → ?m.6 x
```
Now by default we have
```lean
#check fun x : Nat => ?a
-- fun x => ?a : (x : Nat) → ?m.6 x
```
In particular, delayed assignment metavariables such as `?m.7` pretty
print using the name of the metavariable they are delayed assigned to,
suppressing the bound variables used in the delayed assignment (hence
`?a` rather than `?a x`). Hovering over `?a` shows `?m.7 x`.
The benefit is that users can see the user-provided name in local
contexts. A justification for this pretty printing choice is that `?m.7
x` is supposed to stand for `?a`, and furthermore it is just as opaque
to assignment in defeq as `?a` is (however, when synthetic opaque
metavariables are made assignable, delayed assignments can be a little
less assignable than true synthetic opaque metavariables).
The original pretty printing behavior can be recovered using `set_option
pp.mvars.delayed true`.
This PR also extends the documentation for holes and synthetic holes,
with some technical details about what delayed assignments are. This
likely should be moved to the reference manual, but for now it is
included in this docstring.
(This PR is a simplified version of #3494, which has a round-trippable
notation for delayed assignments. The pretty printing in this PR is
unlikely to round trip, but it is better than the current situation,
which is that delayed assignment metavariables never round trip, and
plus it does not require introducing a new notation.)
The app unexpanders for `Name.mkStr1` through `Name.mkStr8` weren't
respecting the escaping rules for names. For example, ``#check `«a.b»``
would show `` `a.b``.
This PR folds the unexpanders into the name literal delaborator, where
escaping is already handled.
The `#guard_msgs` command runs the command it is attached to as if it
were a top-level command. This is because the top-level command
elaborator runs linters, and we are interested in capturing linter
warnings using `#guard_msgs`. However, the linters will run on
`#guard_msgs` itself, leading sometimes to duplicate warnings (like for
the unused variable linter).
Rather than special-casing `#guard_msgs` in every affected linter, this
PR special-cases it in the top-level command elaborator itself. **Now
linters are only run if the command doesn't contain `#guard_msgs`.**
This way, the linters are only run on the sub-command that `#guard_msgs`
runs itself. This rule also keeps linters from running multiple times in
cases such as `set_option pp.mvars false in /-- ... -/ #guard_msgs in
...`.
This follows the norm for all other Bitvector operations, and makes the
symbols `/` and `%` the simp normal form.
I'd imagine that @hargonix would prefer that this be merged after
https://github.com/leanprover/lean4/pull/5628, so as to prevent churn
for his PR. I'm happy to rebase the PR once the other PR lands.
---------
Co-authored-by: Henrik Böving <hargonix@gmail.com>
These lemmas explain what happens when the denominator is zero with
`udiv`, `umod`, `sdiv`, `smod`. A follow-up PR will show what happens
with `smtUDiv` and `smtSMod`, since these need some more bitvector
theory.
These lemmas will be used by `bv_decide` for bitblasting.
The theorems `{sdiv, smod}_zero` are located after `neg` theory has been
built for the purpose of writing terse proofs.
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
This PR fixes three problems:
- When the language server is being stopped in a non-normal way without
going through the regular LSP shutdown protocol (e.g. by closing VS
Code), it could sometimes happen that both the watchdog and the file
worker were not properly terminated and lingered around forever,
resulting in zombie processes (#5296)
- When the file worker crashes and the user restarts it by making a
change to the document, the file worker would produce incorrect
diagnostics for the document until the file is restarted.
- (Minor) When the file worker would crash during initialization, the
error diagnostic would be reported on stderr instead of stdout
The deadlock-induced termination issue from #5296 should be resolved by
the following measures:
- The watchdog main task is always terminated with `IO.Process.exit` to
ensure that it terminates even if some other tasks in the process are
still running.
- The file worker communication task in the watchdog no longer waits for
the file worker process to terminate when writing to the client fails,
only when reading from the file worker fails.
- When the watchdog shuts down (either as a result of an orderly or a
non-normal shutdown), instead of waiting for the file worker
communication tasks to complete, it kills the file worker process. The
rationale behind this is that the file worker currently should have no
essential work to complete if the server is being stopped anyways, and
so waiting for the communication task is not necessary.
The file worker diagnostic desync after a crash was caused by us
tracking changes to the document of a crashed file worker twice: Once as
part of the document, and once as part of the queued messages to the
file worker. This meant that when the file worker was restarted, it
would receive the changes made to the document while the file worker was
crashed twice, leading to a desynced document state.
(Probably) fixes#5296.
... while at it also call `trivial` to close goals that can be trivially
closed.
---------
Co-authored-by: Siddharth <siddu.druid@gmail.com>
Co-authored-by: Henrik Böving <hargonix@gmail.com>
when the transparency mode is `.all`, then one expects `getFunInfo` and
`inferType` to also work with that transparency mode.
Fixes#5562Fixes#2975Fixes#2194
While `initialize` pretended it had the declaration name of the constant
to be initialized, missing declaration ranges for the latter led call
hierarchy etc. to ignore the definition
Lake no longer attempts to fetch the Reservoir build cache if the build
directory is already present. Plus, failure of the automatic fetch now
only produces a trace message, not a warning.
@kim-em, I'm happy to keep any subset of `foldl_min`, `foldl_min_right`,
`foldl_min_le`, `foldl_min_min_of_le` (should that one have been called
`foldl_min_le_of_le`?). Which ones do you like?
I think the overhead (runtime/later proving) of using `for` is paid off
by being able to short-circuit.
These functions are needed downstream to switch over the Std.HashSet.
On Windows, shared libraries must be removed before linking. Otherwise,
linking can fail with "Permission denied" when the libraries are in use.
This ensures such removal is done for the new `libLake_shared.dll` and
both parts of `libleanshared`.
Lake no longer attempts to fetch Reservoir build caches (barrels) for
non-Reservoir dependencies, and it will only fetch them for Reservoir
dependencies in the presence of a known toolchain.
Also, optional build job failures are now only displayed in verbose
mode.
In C, these are supported only as a vendor extension; they should
instead use proper C99 flexible array members.
In C++, both `[]` and `[0]` are vendor extensions.
Co-authored-by: Thomas Köppe <tkoeppe@google.com>
The constructor `AttributeExtensionOLeanEntry.decl` and related code
seems to be unused, and has been unused since its introduction in
a77598f7cf three years ago. Probably worth
removing (and changing the now one-constructor inductive into a
structure).
ac_nf is a counterpart to ac_rfl, which normalizes bitvector expressions
with respect to associativity and commutativity.
While there, also add test coverage for ac_rfl and ac_nf for BitVec,
complementing the existing test coverage.
The lemma `exists_const` already handles all real cases of `(∃ _ : α, p)
↔ p` for general types `α`. If there are no `Nonempty` instances and
this lemma cannot apply, it seems unlikely that simp could make more
progress with `(∃ _ : α, p) ↔ Nonempty α ∧ p`.
However, it is still worth simplifying `(∃ _ : p, q)` to `p ∧ q`.
Also adds a `Nonempty (Decidable a)` instance, which is used by Mathlib.
…|twoPow|one|replicate]
... and mark `getElem_setWidth` as `@[simp]`.
`getElem_rotateLeft` and `getElem_rotateRight` have a non-trivial rhs
but we follow `getLsbD_[rotateLeft|rotateRight]`for consistency.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
Adds Reservoir as another possible source of build caches in addition to
GitHub releases. If a GitHub release is not configured for a Reservoir
dependency, it will attempt download a build cache from Reservoir. Like
with GitHub releases, failure will not stop the build and instead issue
a warning. Many of the Lake API calls related to these build caches were
refactored and renamed, with the old names remaining around as
deprecated aliases.
Build cache downloads (from Reservoir or GitHub) can now be disabled via
the `--no-cache` CLI option or the `LAKE_NO_CACHE` environment variable.
A disabled cache can be re-enable with the `--try-cache` CLI option.
Macros sometimes create auxiliary types and instances about them, and
they rely on the instance name generate to create unique names in that
case.
This modifies the automatic name generator to add a fresh macro scope to
the generated name if any of the constants in the type of the instance
themselves have macro scopes.
Closes#2044
Generally works best to pick up the proofs by unification with the lhs.
pinging @hargoniX as this goes by, as it changes some proofs in
bv_decide (nothing interesting, just a bit simpler)
@bollu, it would be good to have confirmation from you, but presumably
this was not meant to be `@[simp]`? It competes with `divRec_succ`, and
has a terrible RHS.
after this change, `simp` will be able to discharge side-goals that,
after simplification, are of the form `∀ …, a = b` with `a =?= b`.
Usually these side-goals are solved by simplification using `eq_self`,
but that does not work when there are metavariables involved.
This enables us to have rewrite rules like
```
theorem List.foldl_subtype (p : α → Prop) (l : List (Subtype p)) (f : β → Subtype p → β)
(g : β → α → β) (b : β)
(hf : ∀ b x h, f b ⟨x, h⟩ = g b x) :
l.foldl f b = (l.map (·.val)).foldl g b := by
```
where the parameter `g` does not appear on the lhs, but can be solved
for using the `hf` equation. See `tests/lean/run/simpHigherOrder.lean`
for more examples.
The motivating use-case is that `simp` should be able to clean up the
usual
```
l.attach.map (fun <x, _> => x)
```
idiom often seen in well-founded recursive functions with nested
recursion.
Care needs to be taken with adding such rules to the default simp set if
the lhs is very general, and thus causes them to be tried everywhere.
Performance impact of just this PR (no additional simp rules) on mathlib
is unsuspicious:
http://speed.lean-fro.org/mathlib4/compare/b5bc44c7-e53c-4b6c-9184-bbfea54c4f80/to/ae1d769b-2ff2-4894-940c-042d5a698353
I tried a few alternatives, e.g. letting `simp` apply `eq_self` without
bumping the mvar depth, or just solve equalities directly, but that
broke too much things, and adding code to the default discharger seemed
simpler.
The formatter was using `tk ++ " "` to separate tokens from tokens they
would merge with, but `" "` is not whitespace that could merge. This
affected large binder lists, which wouldn't pretty print with any line
breaks. Now they can be flowed across multiple lines.
Closes#5424
Just an `Array` version of `List.eraseReps`. These functions are for now
outside of scope for verification, so there's just a simple `example` in
the tests.
Now the elab-as-elim procedure allows eliminators whose result is an
arbitrary application of the motive. For example, the following is now
accepted. It will generalize `Int.natAbs _` from the expected type.
```lean
@[elab_as_elim]
theorem natAbs_elim {motive : Nat → Prop} (i : Int)
(hpos : ∀ (n : Nat), i = n → motive n)
(hneg : ∀ (n : Nat), i = -↑n → motive n) :
motive (Int.natAbs i) := by sorry
```
This change simplifies the elaborator, since it no longer needs to keep
track of discriminants (which can easily be read off from the return
type of the eliminator) or the difference between "targets" and "extra
arguments" (which are now both "major arguments" that should be eagerly
elaborated).
Closes#4086
`BitVec.Lemmas` contained a couple of non-terminal simps. We turn
non-terminal `simp$`, `simp [`, and `simp at` expressions into `simp
only` to improve code maintainability.
This was upstreamed from Mathlib in #5478, but leaving off the `@[simp]`
attribute, thereby breaking Mathlib. (We could of course add the simp
attribute back in Mathlib, but wherever it lives it should have been in
place at the time we merged -- this way I have to add it temporarily in
Mathlib and then remove it again once it is redundant.)
Recall that currently named arguments suppress all explicit parameters
that are dependencies. This PR limits this feature to only apply to true
structure projections, except in the case where it is triggered when
there are no more positional arguments. This preserves the primary
reason for generalizing this feature (issue #1851), while removing the
generalized feature, which has led to numerous confusions (issue #1867).
This also fixes a bug pointed out [on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/.40foo.20.28A.20.3A.3D.20bar.29.20_.20_/near/468564862)
where in `@` mode, instance implicit parameter dependencies to named
arguments would be suppressed unless the next positional argument was
`_`.
More detail:
* The `NamedArg` structure now has a `suppressDeps : Bool` field. It is
set to `true` for the `self` argument in structure projections. If there
is such a `NamedArg`, explicit parameters that are dependencies to the
named argument are turned into implicit arguments. The consequence is
that *all* structure projections are treated as if their type parameters
are implicit, even for class projections. This flag is *not* used for
generalized field notation.
* We preserve the suppression feature when there are no positional
arguments remaining. This feature pre-dates the fix to issue #1851, and
it is useful when combining named arguments and the eta expansion
feature, since dependencies of named arguments cannot be turned into eta
arguments. Plus, there are examples of the form `rw [lem (h := foo)]`
where `lem` has explicit arguments that `h` depends on.
* For instance implicit parameters in explicit mode, now `_` arguments
register terminfo and are hoverable.
* Now `..` is respected in explicit mode.
This implements RFC #5397. The `suppressDeps` flag suggests a future
possibility of a named argument syntax that can suppress dependencies.
Adds a mechanism where when an autoparam tactic fails to synthesize a
parameter, the associated parameter name or field name for the autoparam
is reported in an error.
Examples:
```text
could not synthesize default value for parameter 'h' using tactics
could not synthesize default value for field 'inv' of 'S' using tactics
```
Notes:
* Autoparams now run their tactics without any error recovery or
error-to-sorry enabled. This enables catching the error and reporting
the contextual information. This is justified on the grounds that
autoparams are not interactive.
* Autoparams for applications now cleanup the autoParam annotation,
bringing it in line with autoparams for structure fields.
* This preserves the old behavior that autoparams leave terminfo, but we
will revisit this after some imminent improvements to the unused
variable linter.
Closes#2950
`elabEvalUnsafe` already does something similar: it also instantiates
universe metavariables, but it is not clear to me whether that is
sensible here.
To be conservative, I leave it out of this PR.
See https://github.com/leanprover/lean4/pull/3090#discussion_r1432007590
for a comparison between `#eval` and `Meta.evalExpr`. This PR is not
trying to fully align them, but just to fix one particular misalignment
that I am impacted by.
Closes#3091
This PR adds the theorems
```
@[simp]
theorem divRec_zero (qr : DivModState w) :
divRec w w 0 n d qr = qr
@[simp]
theorem divRec_succ' (wn : Nat) (qr : DivModState w) :
divRec w wr (wn + 1) n d qr =
let r' := shiftConcat qr.r (n.getLsbD wn)
let input : DivModState w :=
if r' < d then ⟨qr.q.shiftConcat false, r'⟩ else ⟨qr.q.shiftConcat true, r' - d⟩
divRec w (wr + 1) wn n d input
```
The final statements may need some masasging to interoperate with
`bv_decide`. We prove the recurrence for unsigned division by building a
shift-subtract circuit, and then showing that this circuit obeys the
division algorithm's invariant.
---
A `DivModState` is lawful if the remainder width `wr` plus the dividend
width `wn` equals `w`,
and the bitvectors `r` and `n` have values in the bounds given by
bitwidths `wr`, resp. `wn`.
This is a proof engineering choice: An alternative world could have
`r : BitVec wr` and `n : BitVec wn`, but this required much more
dependent typing coercions.
Instead, we choose to declare all involved bitvectors as length `w`, and
then prove that
the values are within their respective bounds.
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
Co-authored-by: Alex Keizer <alex@keizer.dev>
Co-authored-by: Kim Morrison <scott@tqft.net>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
There's a comment on `withHeartbeats` that says "See also
Lean.withSeconds", but his definition does not seem to actually exist.
Hence, I've removed the comment.
Add iff version of `List.IsPrefix.getElem`, and `eq_of_length_le`
variants of `List.IsInfix.eq_of_length, List.IsPrefix.eq_of_length,
List.IsSuffix.eq_of_length`
We make sure that we can pull `List.toArray` out through all operations
(well, for now "most" rather than "all"). As we also push `Array.toList`
inwards, this hopefully has the effect of them cancelling as they meet,
and `simp` naturally rewriting Array operations into List operations
wherever possible.
This is not at all complete yet.
building upon #3714, this (almost) implements the second half of #3302.
The main effect is that we now get a better error message when `rfl`
fails. For
```lean
example : n+1+m = n + (1+m) := by rfl
```
instead of the wall of text
```
The rfl tactic failed. Possible reasons:
- The goal is not a reflexive relation (neither `=` nor a relation with a @[refl] lemma).
- The arguments of the relation are not equal.
Try using the reflexivity lemma for your relation explicitly, e.g. `exact Eq.refl _` or
`exact HEq.rfl` etc.
n m : Nat
⊢ n + 1 + m = n + (1 + m)
```
we now get
```
error: tactic 'rfl' failed, the left-hand side
n + 1 + m
is not definitionally equal to the right-hand side
n + (1 + m)
n m : Nat
⊢ n + 1 + m = n + (1 + m)
```
Unfortunately, because of very subtle differences in semantics (which
transparency setting is used when reducing the goal and whether the
“implicit lambda” feature applies) I could not make this simply the only
`rfl` implementation. So `rfl` remains a macro and is still expanded to
`eq_refl` (difference transparency setting) and `exact Iff.rfl` and
`exact HEq.rfl` (implicit lambda) to not break existing code. This can
be revised later, so this still closes: #3302.
A user might still be puzzled *why* to terms are not defeq. Explaining
that better (“reduced to… and reduces to… etc.”) would also be great,
but that’s not specific to `rfl`, so better left for some other time.
Previously the formatter was using the builtin token table rather that
the one in the current environment. This could lead to round-tripping
failures for user-defined notations.
For an illustrative example, given the following notation
```lean
infixl:65 "+'" => Int.add
notation:65 a:65 "+'-" b:66 => Int.add a (id b)
```
then `5 +' -1` would parse as `Int.add 5 (-1)` and incorrectly pretty
print as `5+'-1`, which in turn would parse as `Int.add 5 (id 1)`. Now
it pretty prints as `5+' -1`.
Modifies how the declaration command elaborator reports when there are
unassigned metavariables. The visible effects are that (1) now errors
like "don't know how to synthesize implicit argument" and "failed to
infer 'let' declaration type" take precedence over universe level
issues, (2) universe level metavariables are reported as metavariables
(rather than as `u_1`, `u_2`, etc.), and (3) if the universe level
metavariables appear in `let` binding types or `fun` binder types, the
error is localized there.
Motivation: Reporting unsolved expression metavariables is more
important than universe level issues (typically universe issues are from
unsolved expression metavariables). Furthermore, `let` and `fun` binders
can't introduce universe polymorphism, so we can "blame" such bindings
for universe metavariables, if possible.
Example 1: Now the errors are on `x` and `none` (reporting expression
metavariables) rather than on `example` (which reported universe level
metavariables).
```lean
example : IO Unit := do
let x := none
pure ()
```
Example 2: Now there is a "failed to infer universe levels in 'let'
declaration type" error on `PUnit`.
```lean
def foo : IO Unit := do
let x : PUnit := PUnit.unit
pure ()
```
In more detail:
* `elabMutualDef` used to turn all level mvars into fresh level
parameters before doing an analysis for "hidden levels". This analysis
turns out to be exactly the same as instead creating fresh parameters
for level mvars in only pre-definitions' types and then looking for
level metavariables in their bodies. With this PR, error messages refer
to the same level metavariables in the Infoview, rather than obscure
generated `u_1`, `u_2`, ... level parameters.
* This PR made it possible to push the "hidden levels" check into
`addPreDefinitions`, after the checks for unassigned expression mvars.
It used to be that if the "hidden levels" check produced an "invalid
occurrence of universe level" error it would suppress errors for
unassigned expression mvars, and now it is the other way around.
* There is now a list of `LevelMVarErrorInfo` objects in the `TermElabM`
state. These record expressions that should receive a localized error if
they still contain level metavariables. Currently `let` expressions and
binder types in general register such info. Error messages make use of a
new `exposeLevelMVars` function that adds pretty printer annotations
that try to expose all universe level metavariables.
* When there are universe level metavariables, for error recovery the
definition is still added to the environment after assigning each
metavariable to level 0.
* There's a new `Lean.Util.CollectLevelMVars` module for collecting
level metavariables from expressions.
Closes#2058
These theorems are useful when one wants to simplify the goal state,
under knowledge that the bitvector operations don't overflow. This can
produce much smaller goal states that eventually allows `bv_omega` to
quickly close the goal.
Note that the LHS of the theorem is *not* in `simp` normal form, since
e.g. `(x + y).toNat` is normalized to `(x.toNat + y.toNat) % 2^w`. It's
not immediately clear to me what should be done about this.
Co-authored-by: Kim Morrison <scott.morrison@gmail.com>
Resolve cases when the `To/FromJSON` type classes are used with `Empty`,
e.g. in the following motivating example.
```
import Lean
structure Foo (α : Type) where
y : Option α
deriving Lean.ToJson
#eval Lean.toJson (⟨none⟩ : Foo Empty) -- fails
```
This is a follow-up to this PR
https://github.com/leanprover/lean4/pull/5415, as suggested by
@eric-wieser. It expands on the original suggestion by also handling
`FromJSON`.
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
---
Correct some stray spelling mistakes. I think the typo count is
asymptotically approaching zero.
Co-authored-by: euprunin <euprunin@users.noreply.github.com>
The problem here was that in Mathlib's `lean-pr-testing-NNNN` branches,
we were setting Batteries to a `nightly-testing-YYYY-MM-DD` branch. This
means that when we merge or rebase a new `nightly-with-mathlib` into a
Lean PR, the corresponding Mathlib testing branch would keep using an
old version of Batteries.
We also make sure to bump Batteries if Mathlib's `lean-pr-testing-NNNN`
branch already exists.
On a document edit, it may be the case that the first nontrivial
snapshot is e.g. for a macro-generated tactic call that does not have
range information. In that case, instead of just displaying nothing, we
should fall back to a previous range, in this case of the original
tactic macro.
Previously, it was not possible to use `decide` with most Array
functions (including `==`).
Later, we may replace some of these functions with defeqs that go via
the `List` operations, and use `csimp` lemmas for fast runtime
behaviour. In the meantime, this allows using `decide`.
Given the derived `Repr` instance for types with parameters, the absence
of `Repr Empty` can cause `Repr` instance synthesis to fail. For
example, given
```lean
inductive Prim (special : Type) where
| plus
| other : special → Prim special
deriving Repr
```
this works:
```lean
#eval (Prim.plus : Prim Int)
```
but this fails:
```lean
#eval (Prim.plus : Prim Empty)
```
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
Co-authored-by: Eric Wieser <wieser.eric@gmail.com>
This implements the first half of #3302: It improves the extensible
`apply_rfl` tactic (the one that looks at `refl` attributes, part of
the `rfl` macro) to
* Check itself and ahead of time that the lhs and rhs are defEq, and
give
a nice consistent error message when they don't (instead of just passing
on
the less helpful error message from `apply Foo.refl`), and using the
machinery that `apply` uses to elaborate expressions to highlight diffs
in implicit arguments.
* Also handle `Eq` and `HEq` (built in) and `Iff` (using the attribute)
Care is taken that, as before, the current transparency setting affects
comparing the lhs and rhs, but not the reduction of the relation
So before we had
```lean
opaque P : Nat → Nat → Prop
@[refl] axiom P.refl (n : Nat) : P n n
/--
error: tactic 'apply' failed, failed to unify
P ?n ?n
with
P 42 23
⊢ P 42 23
-/
#guard_msgs in
example : P 42 23 := by apply_rfl
opaque withImplicitNat {n : Nat} : Nat
/--
error: tactic 'apply' failed, failed to unify
P ?n ?n
with
P withImplicitNat withImplicitNat
⊢ P withImplicitNat withImplicitNat
-/
#guard_msgs in
example : P (@withImplicitNat 42) (@withImplicitNat 23) := by apply_rfl
```
and with this PR the messages we get are
```
error: tactic 'apply_rfl' failed, The lhs
42
is not definitionally equal to rhs
23
⊢ P 42 23
```
resp.
```
error: tactic 'apply_rfl' failed, The lhs
@withImplicitNat 42
is not definitionally equal to rhs
@withImplicitNat 23
⊢ P withImplicitNat withImplicitNat
```
A test file checks the various failure modes and error messages.
I believe this `apply_rfl` can serve as the only implementation of
`rfl`, which would then complete #3302, and actually expose these
improved
error messages to the user. But as that seems to require a
non-trivial bootstrapping dance, it’ll be separate.
Provide an instance `Inhabited (TacticM α)`, even when `α` is not known
to be inhabited.
The default value is just the default value of `TermElabM α`, which
already has a similar instance.
closes#5333
This PR tries to address issue #5333.
My conjecture is that the binder annotations for `C.toB` and
`Algebra.toSMul` are not ideal. `Algebra.toSMul` is one of declarations
where the new command `set_synth_order` was used. Both classes, `C` and
`Algebra`, are parametric over instances, and in both cases, the issue
arises due to projection instances: `C.toB` and `Algebra.toSMul`. Let's
focus on the binder annotations for `C.toB`. They are as follows:
```
C.toB [inst : A 20000] [self : @C inst] : @B ...
```
As a projection, it seems odd that `inst` is an instance-implicit
argument instead of an implicit one, given that its value is fixed by
`self`. We observe the same issue in `Algebra.toSMul`:
```
Algebra.toSMul {R : Type u} {A : Type v} [inst1 : CommSemiring R] [inst2 : Semiring A]
[self : @Algebra R A inst1 inst2] : SMul R A
```
The PR changes the binder annotations as follows:
```
C.toB {inst : A 20000} [self : @C inst] : @B ...
```
and
```
Algebra.toSMul {R : Type u} {A : Type v} {inst1 : CommSemiring R} {inst2 : Semiring A}
[self : @Algebra R A inst1 inst2] : SMul R A
```
In both cases, the `set_synth_order` is used to force `self` to be
processed first.
In the MWE, there is no instance for `C ...`, and `C.toB` is quickly
discarded. I suspect a similar issue occurs when trying to use
`Algebra.toSMul`, where there is no `@Algebra R A ... ...`, but Lean
spends unnecessary time trying to synthesize `CommSemiring R` and
`Semiring A` instances. I believe the new binder annotations make sense,
as if there is a way to synthesize `Algebra R A ... ...`, it will tell
us how to retrieve the instance-implicit arguments.
TODO:
- Impact on Mathlib.
- Document changes.
---------
Co-authored-by: Kim Morrison <scott.morrison@gmail.com>
Co-authored-by: Johan Commelin <johan@commelin.net>
We need to follow the fvar aliases registered by `match` in both
directions
Fixes#4714, fixes#2837
---------
Co-authored-by: Mario Carneiro <di.gama@gmail.com>
After #5270, `partial` functions that use products of sums no longer
compile with only `Nonempty` constraints on their arguments. These
instances allow the compilation to work.
In LNSym we often use the pattern `ofBool (a.getLsbD i)` to pick out a
specific bit (`i`) from a bitvector (`a`).
By adding a rewrite to `extractLsb` to `bv_decide`s normalization set,
we can still automatically close goals that have this pattern. In the
process, I also added a simp-lemma about the value of a `Fin 1`.
Obviously a link to the web docs isn't ideal, but having hovers
available on the symbol is much better than nothing.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
A round of clean-up for the context of the functional induction
principle cases.
* Already previously, with `match e with | p => …`, functional induction
would ensure that `h : e = p` is in scope, but it wouldn’t work in
dependent cases. Now it introduces heterogeneous equality where needed
(fixes#4146)
* These equalities are now added always (previously we omitted them when
the discriminant was a variable that occurred in the goal, on the
grounds that the goal gets refined through the match, but it’s more
consistent to introduce the equality in any case)
* We no longer use `MVarId.cleanup` to clean up the goal; it was
sometimes too aggressive (fixes#5347)
* Instead, we clean up more carefully and with a custom strategy:
* First, we substitute all variables without a user-accessible name, if
we can.
* Then, we substitute all variable, if we can, outside in.
* As we do that, we look for `HEq`s that we can turn into `Eq`s to
substitute some more
* We substitute unused `let`s.
**Breaking change**: In some cases leads to a different functional
induction principle (different names and order of assumptions, for
example).
Fixes a workflow bug where the `check-level` was not always set
correctly. Arguments to a `gh` call used to determine the `check_level`
were accidentally outside of the relevant command substitution (`$(gh
...)`).
-----
This can be observed in [these
logs](https://github.com/leanprover/lean4/actions/runs/10859763037/job/30139540920),
where the check level (shown first under "configure build matrix") is
`2`, but the PR does not have the `release-ci` tag. As a "test", run the
script for "set check level" printed in those logs (with some lines
omitted):
```
check_level=0
labels="$(gh api repos/leanprover/lean4/pulls/5343) --jq '.labels'"
if echo "$labels" | grep -q "release-ci"; then
check_level=2
elif echo "$labels" | grep -q "merge-ci"; then
check_level=1
fi
echo "check_level=$check_level"
```
Note that this prints `check_level=2`, but changing `labels` to
`labels="$(gh api repos/leanprover/lean4/pulls/5343 --jq '.labels')"`
prints `check_level=0`.
This PR fixes an issue reported a while ago at
https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/.60Monad.2Emap.60.20is.20a.20namespace.3F/near/425662846
where `Monad.map` was incorrectly reported by the autocompletion as a
namespace.
The underlying issue is that `Monad.map` contains an internal
declaration `_default`. This PR ensures that no namespaces are
registered that only contain internal declarations.
This also means that `open`ing namespaces that only contain internal
declarations will now fail.
The Mathlib adaption for this is a minor change where a declaration
(i.e. a namespace that only contains internal declarations) was `open`ed
by accident.
This solves the issue where certain subexpressions are lacking syntax
hovers because the hover text is not "builtin" - it only shows up if the
`Parser` constant is imported in the environment. For top level syntaxes
this is not a problem because `builtin_term_parser` will automatically
add this doc information, but nested syntaxes don't get the same
treatment.
We could walk the expression and add builtin docs recursively, but this
is somewhat expensive and unnecessary given that it's a fixed list of
declarations in lean core. Moreover, there are reasons to want to
control which syntax nodes actually get hovers, and while a better
system for that is forthcoming, for now it can be achieved by
strategically not applying the `@[builtin_doc]` attribute.
Fixes#3842
When the elaborator doesn't provide us with any `CompletionInfo`, we
currently provide no completions whatsoever. But in many cases, we can
still provide some helpful identifier completions without elaborator
information. This PR adds a fallback mode for this situation.
There is more potential here, but this should be a good start.
In principle, this issue alleviates #5172 (since we now provide
completions in these contexts). I'll leave it up to an elaboration
maintainer whether we also want to ensure that the completion infos are
provided correctly in these cases.
This adds a simplification lemma for `(x - y).toNat` when the
subtraction is known to not overflow (i.e., `y ≤ x`).
We make a new section for this for two reasons:
1. Definitions of subtraction occur before the definition of
`BitVec.le_def`, so we cannot directly place this lemma at `sub`.
2. There are other theorems of this kind, for addition and
multiplication, which can morally live in the same section.
The theorem
```lean
namespace Int
theorem ediv_nonneg_of_nonpos_of_nonpos {a b : Int} (Ha : a ≤ 0) (Hb : b ≤ 0) : 0 ≤ a / b := by
match a, b with
| ofNat a, b =>
match Int.le_antisymm Ha (ofNat_zero_le a) with
| h1 =>
rw [h1, zero_ediv,]
exact Int.le_refl 0
| a, ofNat b =>
match Int.le_antisymm Hb (ofNat_zero_le b) with
| h1 =>
rw [h1, Int.ediv_zero]
exact Int.le_refl 0
| negSucc a, negSucc b =>
rw [Int.div_def, ediv]
have le_succ {a: Int} : a ≤ a+1 := (le_add_one (Int.le_refl a))
have h2: 0 ≤ ((↑b:Int) + 1) := Int.le_trans (ofNat_zero_le b) le_succ
have h3: (0:Int) ≤ ↑a / (↑b + 1) := (ediv_nonneg (ofNat_zero_le a) h2)
exact Int.le_trans h3 le_succ
```
is nontrivial to prove from existing theorems and would be nice to add
as standard theorem in DivModLemmas.
See the zullip conversation
[here](https://leanprover.zulipchat.com/#narrow/stream/113488-general/topic/Adding.20theorem.20theorem.20ediv_nonneg'.20for.20negative.20a.20and.20b)
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
From the new doc-string:
```quote
In early versions of Lean, the typeclasses provided by `/` and `%`
were defined in terms of `tdiv` and `tmod`, and these were named simply as `div` and `mod`.
However we decided it was better to use `ediv` and `emod`,
as they are consistent with the conventions used in SMTLib, Mathlib,
and often mathematical reasoning is easier with these conventions.
At that time, we did not rename `div` and `mod` to `tdiv` and `tmod` (along with all their lemma).
In September 2024, we decided to do this rename (with deprecations in place),
and later we intend to rename `ediv` and `emod` to `div` and `mod`, as nearly all users will only
ever need to use these functions and their associated lemmas.
```
Proves that `<` and `<=` on `BitVec` are (strict) (total) partial
orders. This is required for the `UInt` as `BitVec` refactor.
This does open the question how to state these theorems "correctly" for
`BitVec`, we have both `<` living in `Prop` and `BitVec.ult` living in
`Bool`. We might of course say to always use `<` but: Once we start
adding `IntX` we need to prove the same results for `BitVec.slt` to
provide an equivalent API. So it would appear that it is unavoidable to
have a `= true` variant of these theorems there?
Question answered: Use `<` and `slt`.
Refactors the derive handlers for `ToJson` and `FromJson` in preparation
for #3160.
This splits up the different parts of the handler according to how other
similar handlers are implemented while keeping the original logic
intact. This makes the changes necessary to adapt the file in #3160 much
easier.
Fixes#4455, fixes#4705, fixes#5219
Also fixes a minor bug where a dot in brackets would report incorrect
completions instead of no completions.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
I found that the kernel has special support for `e =?= true`, and will
in this case aggressively whnf `e`. This explains the following behavior
(for a `sqrt` function with fuel):
```lean
theorem foo : sqrt 100000000000000000002 == 10000000000 := rfl -- fast
theorem foo : sqrt 100000000000000000002 = 10000000000 := rfl -- slow
theorem foo : sqrt 100000000000000000002 = 10000000000 := by decide -- fast
```
The special support in the kernel only applies for closed `e` and `true`
on the RHS. It could be generlized (also open terms, also `false`, other
data type's constructors, different orientation). But maybe I should
wait for evidence that this generaziation really matters, or whether
all applications (proof by reflection) can be made to have this form.
This PR enables the use of incrementality for completion in tactic
blocks. Consider the following example:
```lean
example : True := by
have : True := T
sleep 10000
```
Before this PR, in order to respond to a completion request after `T`,
`sleep 10000` has to complete first since the command must be fully
elaborated. After this PR, the completion request is responded to
immediately.
Currently, `ll_infer_type` is responsible for telling the user about
`noncomputable` when a definition depends on one without executable
code. However, this is imperfect because type inference does not check
every subexpression. This leads to errors later on that users find to be
hard to interpret.
Now, `Lean.IR.checkDecls` has a friendlier error message when it
encounters constants without compiled definitions, suggesting to
consider using `noncomputable`. While this function is an internal IR
consistency check, it is also reasonable to have it give an informative
error message in this particular case. The suggestion to use
`noncomputable` is limited to just unknown constants.
Some alternatives would be to either (1) create another checker just for
missing constants, (2) change `ll_infer_type` to always visit every
subexpression no matter if they are necessary for inferring the type, or
(3) investigate whether `tests/lean/run/1785.lean` is due to a deeper
issue.
Closes#1785
This is "upstreaming" mathlib's `unfold_let` tactic by incorporating its
functionality into `unfold`. Now `unfold` can, in addition to unfolding
global definitions, unfold local definitions. The PR also updates the
`conv` version of the tactic.
An improvement over `unfold_let` is that it beta reduces unfolded local
functions.
Two features not present in `unfold` are that (1) `unfold_let` with no
arguments does zeta delta reduction of *all* local definitions, and also
(2) `unfold_let` can interleave unfoldings (in contrast, `unfold a b c`
is exactly the same as `unfold a; unfold b; unfold c`).
Closes RFC #4090
When an eliminator was overapplied with more than one additional
argument, elaboration produced an incorrect term because the list of
processed arguments was being reversed. Now these arguments are not
reversed.
1. Remove the need to allocate an intermediate `String` for literally
every character in a JSON `String`.
2. Use a single `String` buffer in the entire `Json.compress` machinery.
3. Use `toListAppend`
Number 1 is doing most of the lifting in the perf diff, the rest are
some minor but measurable improvements.
We change the `bv_decide` to understand `BitVec.extractLsb'` as a
primitive, and add a normalization lemma for `extractLsb`.
It's important to pick the primed version as a primitive, because it is
not always possible to rewrite `extractLsb'` back into `extractLsb` (see
#5007 for that direction, and the relevant side-conditions).
That is, with this PR, `bv_decide` is able to bitblast both versions of
extracting bits.
I don't think we gain anything from having them as `abbrev` here, and
the simpNF linter complains:
```
-- Init.Data.BitVec.Lemmas
#check @BitVec.toNat_intMin /- simp can prove this:
by simp only [BitVec.toNat_twoPow]
One of the lemmas above could be a duplicate.
If that's not the case try reordering lemmas or adding @[priority].
-/
#check @BitVec.toNat_intMax /- Left-hand side simplifies from
(BitVec.intMax w).toNat
to
(2 ^ w - 1 % 2 ^ w + 2 ^ (w - 1)) % 2 ^ w
using
simp only [@BitVec.toNat_sub, @BitVec.ofNat_eq_ofNat, BitVec.toNat_ofNat, BitVec.toNat_twoPow, Nat.add_mod_mod]
Try to change the left-hand side to the simplified term!
-/
```
```
#lint only simpNF in all
```
reports (amongst others):
```
-- Init.Data.Int.Order
#check @Int.toNat_of_nonneg /- Left-hand side simplifies from
↑a.toNat
to
max a 0
using
simp only [Int.ofNat_toNat]
Try to change the left-hand side to the simplified term!
-/
#check Int.toNat_sub_toNat_neg /- Left-hand side simplifies from
↑n.toNat - ↑(-n).toNat
to
max n 0 - max (-n) 0
using
simp only [Int.ofNat_toNat]
Try to change the left-hand side to the simplified term!
-/
```
This doesn't completely resolve the danger (only relevant in `prelude`
files) of importing `Init.Data.List.Basic` but not `Init.Data.List.Impl`
and thereby not having `@[csimp]` lemmas installed for some list
operations.
I'm going to address this better while working on `Array`.
Sebastian mentioned that the use of the kernel defeq was to work around
a performance issue that was fixed since. Let's see if we can do
without.
This is also a semantic change: Ground terms (no free vars, no mvars)
are reduced at
“all” transparency even if the the transparency setting is default. This
was the case
even before 03f6b87647 switched to the
kernel defeq
checking for performance. It seems that this is rather surprising
behavior from the user
point of view. The fallout on batteries and mathlib is rather limited,
only a few
`rfl` proofs seem to have (inadvertently or not) have relied on this.
The speedcenter reports no significant regressions on core or mathlib.
Remark: declarations like `sizeWithSharing` must be in `IO` since they
are not functions.
The commit also uses the more efficient `ShareCommon.shareCommon'`.
Adds additional fields to the package configuration which will be used
by Reservoir:
* `version`: The version of the package. Follows Lean's model of
`<major>.<minor>.<patch>[-<specialDescr>]`.
* `versionTags`: A pattern matching the set of Git tags Reservoir should
consider package version revisions.
* `description`: A short description for the package. Takes precedence
over the GitHub's description.
* `keywords`: An array of package keywords that will be used to group
packages into categories on Reservoir. Takes precedence over labels on
the repository.
* `homepage`: A URL to a website for the package. Takes precedence over
GitHub's homepage.
* `license`: An SPFX license identifier for the package's license (not
verified to be well-formed).
* `licenseFiles`: An array of (relative) files the contain license
information (e.g., `#["LICENSE", "NOTICE"]` for Apache 2.0).
* `readmeFile`: Relative path to the package's readme (enables
non-standard README locations).
* `reservoir`: Reservoir will use this setting to determine whether to
include packages in its index.
Also adds two new CLI commands:
* `lake reservoir-config`: Used by Reservoir to extract a package's
configuration.
* `lake check-build`: Determines whether the package has any default
build targets configured.
The Reservoir configuration also makes uses of the exiting `name` and
`platformIndependent` fields.
These commands were trusting that elaboration resulted in type-correct
terms, but users testing custom elaborators have found it to be
surprising that they do not do typechecking. This adds a `Meta.check`
step.
This renames `BitVec.getLsb` to `getLsbD` (`D` for "default" value, i.e.
false), and introduces `getLsb?` and `getLsb'` (which we can rename to
`getLsb` after a deprecation cycle).
(Similarly for `getMsb`.)
Also adds a `GetElem` class so we can use `x[i]` and `x[i]?` notation.
Later, we will turn
```
theorem getLsbD_eq_getElem?_getD (x : BitVec w) (i : Nat) (h : i < w) :
x.getLsbD i = x[i]?.getD false
```
on as a `@[simp]` lemma.
This PR doesn't attempt to demonstrate the benefits, but I think both
arguments are going to get easier, and this will bring the BitVec API
closer in line to List/Array, etc.
---------
Co-authored-by: Markus Himmel <markus@lean-fro.org>
in #4154 and #5129 the rules for equational lemmas have changed, and new
options were introduced that can be used to revert to the pre-4.12
behavior. Hopefully nobody really needs these options besides for
backwards compatibility, therefore we put these options in the
`backward` option name space.
So the previous behavior can be achieved by setting
```lean
set_option backward.eqns.nonrecursive false
set_option backward.eqns.deepRecursiveSplit false
```
With this, lean produces the following zoo of rewrite rules:
```
Option.map.eq_1 : Option.map f none = none
Option.map.eq_2 : Option.map f (some x) = some (f x)
Option.map.eq_def : Option.map f p = match o with | none => none | (some x) => some (f x)
Option.map.eq_unfold : Option.map = fun f p => match o with | none => none | (some x) => some (f x)
```
The `f.eq_unfold` variant is especially useful to rewrite with `rw`
under
binders.
This implements and fixes#5110
This PR propagates the `AttributeKind` to `SimpleScopedEnvExtension.add`
in attributes created with `register_label_attr`.
This also fixes a nearby stale docstring which referenced `Std`.
---
Closes#3697
This PR roughly halves the time needed to load the .ilean files by
optimizing the JSON parser and the conversion from JSON to Lean data
structures.
The code is optimized roughly as follows:
- String operations are inlined more aggressively
- Parsers are changed to use new `String.Iterator` functions `curr'` and
`next'` that receive a proof and hence do not need to perform an
additional check
- The `RefIdent` of .ilean files now uses a `String` instead of a `Name`
to avoid the expensive parse step from `String` to `Name` (despite the
fact that we only very rarely actually need a `Name` in downstream code)
- Instead of `List`s and `Subarray`s, the JSON to Lean conversion now
directly passes around arrays and array indices to avoid redundant
boxing
- Parsec's `peek?` sometimes generates redundant `Option` wrappers
because the generation of basic blocks interferes with the ctor-match
optimization, so it is changed to use an `isEof` check where possible
- Early returns and inline-do-blocks cause the code generator to
generate new functions, which then interfere with optimizations, so they
are now avoided
- Mutual defs are used instead of unspecialized passing of higher-order
functions to generate faster code
- The object parser is made tail-recursive
This PR also fixes a stack overflow in `Lean.Json.compress` that would
occur with long lists and adds a benchmark for the .ilean roundtrip
(compressed pretty-printing -> parsing).
This PR fixes a small bug where over time, "import out of data" messages
would accumulate in files when their size changed before restarting its
file worker.
https://github.com/leanprover/vscode-lean4/pull/521 changed the display
name of the VS Code extension so that it can be found more easily when
searching for "Lean" (before it would appear far down in the list). This
PR updates the quickstart guide to reflect this fact.
This was not a great simp lemma, and hurts simp confluence. Better to
just use it locally where it is useful.
Similarly `List.head_eq_iff_head?_eq_some`.
This PR also pulls in some mathlib theorems on testBit and Nat and establishes facts about 2^w that are needed here and which are generally useful for bitvector reasoning.
The following theorem is not generalized to arbitrary x instead of 2, as this would require a condition to be added for x > 1 which would have to be passed to simp each time this theorem should fire.
chore: derive from testBit_two_pow
chore: convert first to prop and then decide
chore: move intMax down as well
chore: add simp set
Add simp-set into this PR
chore: fix simp extension
Move file to src/Lean to fix build
Add prelude
update date
Add university of cambridge as copyright holder
improve naming
use whitespace uniformly
use decide (n = m)
Drop the 'Nat.' namespace
Update src/Init/Data/BitVec/Lemmas.lean
Co-authored-by: Siddharth <siddu.druid@gmail.com>
Update src/Init/Data/BitVec/Lemmas.lean
Co-authored-by: Siddharth <siddu.druid@gmail.com>
Fix build
add some theorems
Revert "add some theorems"
This reverts commit fb97bc2007e371854b40badb3d6014da034c1f5e.
WIP
Shorten proof
Update src/Init/Data/Nat/Lemmas.lean
finish proofs
Update src/Init/Data/BitVec/Lemmas.lean
Co-authored-by: Kim Morrison <scott@tqft.net>
Update src/Init/Data/Nat/Lemmas.lean
Co-authored-by: Kim Morrison <scott@tqft.net>
chore: move BoolToPropSimps
`simp only` will not apply this simproc anymore. Users must now write
`simp only [reduceCtorEq]`. See RFC #5046 for motivation.
This PR also renames simproc to `reduceCtorEq`.
close#5046
@semorrison A few `simp only ...` tactics will probably break in
Mathlib. Fix: include `reduceCtorEq`.
We use `no_index` to work around special-handling of `OfNat.ofNat` in
`DiscrTree`, which has been reported as an issue in
https://github.com/leanprover/lean4/issues/2867 and is currently in the
process of being fixed in https://github.com/leanprover/lean4/pull/3684.
As the potential fix seems non-trivial and might need some time to
arrive in-tree, we meanwhile add the `no_index` keyword to the
problematic subterm.
---------
Co-authored-by: Eric Wieser <wieser.eric@gmail.com>
We swap the arguments for `Membership.mem` so that when proceeded by a
`SetLike` coercion, as is often the case in Mathlib, the resulting
expression is recognized as eta expanded and reduce for many
computations. The most beneficial outcome is that the discrimination
tree keys for instances and simp lemmas concerning subsets become more
robust resulting in more efficient searches.
Closes `RFC` #4932
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
Co-authored-by: Henrik Böving <hargonix@gmail.com>
This is part of #3983.
After #4154 introduced equational lemmas for non-recursive functions and
#5055
unififed the lemmas for structural and wf recursive funcitons, this now
disables the special handling of recursive functions in
`findMatchToSplit?`, so that the equational lemmas should be the same no
matter how the function was defined.
The new option `eqns.deepRecursiveSplit` can be disabled to get the old
behavior.
### Breaking change
This can break existing code, as there now can be extra equational
lemmas:
* Explicit uses of `f.eq_2` might have to be adjusted if the numbering
changed.
* Uses of `rw [f]` or `simp [f]` may no longer apply if they previously
matched (and introduced a `match` statement), when the equational
lemmas got more fine-grained.
In this case either case analysis on the parameters before rewriting
helps, or setting the option `opt.deepRecursiveSplit false` while
defining the function
Updates the user widget manual to account for more recent changes. One
issue is that the samples no longer work on https://live.lean-lang.org/
because it uses an outdated version of the `@leanprover/infoview` NPM
package. They work on https://lean.math.hhu.de/ and in recent versions
of the VSCode extension.
This is part of #3983.
Fine-grained equational lemmas are useful even for non-recursive
functions, so this adds them.
The new option `eqns.nonrecursive` can be set to `false` to have the old
behavior.
### Breaking channge
This is a breaking change: Previously, `rw [Option.map]` would rewrite
`Option.map f o` to `match o with … `. Now this rewrite will fail
because the equational lemmas require constructors here (like they do
for, say, `List.map`).
Remedies:
* Split on `o` before rewriting.
* Use `rw [Option.map.eq_def]`, which rewrites any (saturated)
application of `Option.map`
* Use `set_option eqns.nonrecursive false` when *defining* the function
in question.
### Interaction with simp
The `simp` tactic so far had a special provision for non-recursive
functions so that `simp [f]` will try to use the equational lemmas, but
will also unfold `f` else, so less breakage here (but maybe performance
improvements with functions with many cases when applied to a
constructor, as the simplifier will no longer unfold to a large
`match`-statement and then collapse it right away).
For projection functions and functions marked `[reducible]`, `simp [f]`
won’t use the equational theorems, and will only use its internal
unfolding machinery.
### Implementation notes
It uses the same `mkEqnTypes` function as for recursive functions, so we
are close to a consistency here. There is still the wrinkle that for
recursive functions we don't split matches without an interesting
recursive call inside. Unifying that is future work.
in principle we'd like to use the existing parser
```
"?" >> (ident <|> hole)
```
but somehow annotate it so that hovering the `hole` will not show the
hole's hover. But for now it was easier to just change the parser to
```
"?" >> (ident <|> "_")
```
and be done with it.
Fixes#5021
The goal at the crucial step is
```
a : Array Nat
i : Fin ?m.27
⊢ ↑i < a.size
```
and after the `apply Fin.val_lt_of_le;` we have
```
a : Array Nat
i : Fin ?m.27
⊢ ?m.27 ≤ a.size
```
and now `apply Fin.val_lt_of_le` applies again, due to accidential
defeq. Adding `with_reducible` helps here.
fixes#5061
Defines `mergeSort`, a naive stable merge sort algorithm, replaces it
via a `@[csimp]` lemma with something faster at runtime, and proves the
following results:
* `mergeSort_sorted`: `mergeSort` produces a sorted list.
* `mergeSort_perm`: `mergeSort` is a permutation of the input list.
* `mergeSort_of_sorted`: `mergeSort` does not change a sorted list.
* `mergeSort_cons`: proves `mergeSort le (x :: xs) = l₁ ++ x :: l₂` for
some `l₁, l₂`
so that `mergeSort le xs = l₁ ++ l₂`, and no `a ∈ l₁` satisfies `le a
x`.
* `mergeSort_stable`: if `c` is a sorted sublist of `l`, then `c` is
still a sublist of `mergeSort le l`.
@arthur-adjedj was very confused when a mutually recursive definition
didn't work as expected, and the reason was that he used different names
for the fixed parameters.
It seems plausible to simply allow that and calculate the fixed-prefix
up to alpha renaming.
It does mean, though, that, for example, termination proof goals will
mention the names as used by the first function. But probably better
than simply failing. And we could even fix that later (by passing down
the
actual names, and renmaing the variables in the context of the mvar,
depending on the “current function”) should it bother our users.
This PR imports LeanSAT's LRAT module as step 4/~6 (step 7 could go
after I did some refactorings to import this) of the LeanSAT
upstreaming. It is the last large component, after this only the LRAT
parser and the reflection tactic that hooks everything up to the meta
level remains. In particular it is the last component that contains
notable proofs, yay!
Again a few remarks:
1. Why is this not in `Std`? I'm not quite sure whether it should be
there. At the current level of code/proof quality we can certainly not
import the checker itself into `Std` but maybe having the data type as
well as the trimming algorithm there might be of interested? I'm hoping
that as we refactor the checker in the future its quality will be high
enough to be also put into `Std`. At this point we would have a full AIG
-> CNF -> LRAT verification pipeline in `Std` for everyone to use. One
additional blocker in this is that we cannot provide the parsers for the
format in `Std` as of today because `Parsec` is still in `Lean` so that
would also have to change.
2. There do exist two abstraction levels to make sure we can swap out
the LRAT implementation at any time:
- The public interface is just all files in the top level `LRAT`
directory. It basically only contains the LRAT format itself, the
checker + soundness proof and the trimming algorithm. As long as we
don't need to change their API (which we shouldn't have to I think) we
can always swap out the entire `Internal` directory without breaking
anything else in LeanSAT.
- The `Internal` module itself contains another layer of abstraction in
the form of the `Formula` class. This allows us to swap out the most
complex component in `Internal` as well, without having to touch any of
the infrastructure that is built around it either.
3. I mostly performed stylistic cleanups on the `Internal` module. In my
experience over upgrading to many nightlies during the course of LeanSAT
development, I have gotten these proofs cleaned up to the point, where
they only break if we change the `List` or `Array` proof API
significantly. Given that we are currently in the process of stabilizing
it I'm hoping that these proofs do not have to be touched anymore unless
we do something crazy. All of the custom theory that the LRAT component
developed around various basic data types has been upstreamed into Lean
over the course of various other PRs.
4. If there are some simple tricks that we can pull off to increase the
code / proof quality in `Internal` and in particular `Internal.Formula`
(this module is not for the light-hearted Lean reviewer) I'm all for it.
Otherwise the best course of action to provide LeanSAT to our users soon
would probably be to merge it as is and do a cut + rewrite at one of the
two interface points described above.
When elaborating the headers of mutual indexed inductive types, mvars
have to be synthesized and instantiated before replacing the fvars
present there. Otherwise, some fvars present in uninstantiated mvars may
be missed and lead to an error later.
Closes#3242 (again)
As discussed with @semorrison, feel free to do whatever to the branch.
---------
Co-authored-by: Kim Morrison <scott.morrison@gmail.com>
Co-authored-by: Eric Wieser <wieser.eric@gmail.com>
Co-authored-by: Kim Morrison <kim@tqft.net>
Previously, the tactic state shown at `decreasing_by` would leak lots of
details about the translation, and mention `invImage`, `PSigma` etc.
This is not nice.
So this introduces `clean_wf`, which is like `simp_wf` but using
`simp`'s `only` mode, and runs this unconditionally. This should clean
up the goal to a reasonable extent.
Previously `simp_wf` was an unrestricted `simp […]` call, but we
probably don’t want arbitrary simplification to happen at this point, so
this now became `simp only` call. For backwards compatibility,
`decreasing_with` begins with `try simp`. The `simp_wf` tactic
is still available to not break too much existing code; it’s docstring
suggests to no longer use it.
With `set_option cleanDecreasingByGoal false` one can disable the use of
`clean_wf`. I hope this is only needed for debugging and understanding.
Migration advise: If your `decreasing_by` proof begins with `simp_wf`,
either remove that (if the proof still goes through), or replace with
`simp`.
I am a bit anxious about running even `simp only` unconditionally here,
as it may do more than some user might want, e.g. because of options
like `zetaDelta := true`. We'll see if we need to reign in this tactic
some more.
I wonder if in corner cases the `simp_wf` tactic might be able to close
the goal, and if that is a problem. If so, we may have to promote simp’s
internal `mayCloseGoal` parameter to a simp configuration option and use
that here.
fixes#4928
Using `Nat.lt_trans` is too restrictive, and using `Nat.lt_of_lt_of_le`
should make this tactic prove more goals.
This fixes a regression probably introduced by #3991; at least in some
cases before that `apply sizeOf_get` would have solved the goal here.
And it’s true that this is now subsumed by `simp`, but because of the
order that `macro_rules` are tried, the too restrictive variant with
`Nat.lt_trans` would be tried before `simp`, without backtracking.
Fixes#5027
Step 3/~7 in upstreaming LeanSAT.
A few thoughts:
- Why is this not in `Std.Sat`? LeanSAT's bitblaster operates on a
limited internal language. For example it has no idea that signed
comparision operators even exist. This is because it relies on a
normalization pass before being given the goal. For this reason I would
not classify the bitblaster as an API that we should publicly advertise
at this abstraction level
- Sometimes I slightly rebuild parts of the LawfulOperator
infrastructure for operators that work non-tail-recursively. This is
because they do not return an `Entrypoint` but instead an
`ExtendingEntrypoint` in order to even be defined in the first place
(casting Ref's and all that). Given the fact that this barely happens
and I never actually commit to rebuilding the full API I'm hoping that
this is indeed a fine decision?
- The single explicit `decreasing_by` that has a simp only which
*almost* looks like `simp_wf` is missing a singular lemma from `simp_wf`
because it doesn't terminate otherwise.
- I am not using functional induction because it basically always fails
at some generalization step, that is also the reason that there is lots
of explicit `generalize` and manually recursive proofs.
---------
Co-authored-by: Markus Himmel <markus@lean-fro.org>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
In #4976, I forgot that we do need info trees eventually on the cmdline
for .ilean generation. Unfortunately, not reporting them incrementally
would require an API change, so let's see what the impact of incremental
reporting is
I'm experimenting with changing the signature of `Ord.arrayOrd`; rather
than make a local synonym here, let's make a local instance so it
doesn't interact with the experiments.
Adds a lemma to rewrite `BitVec.extractLsb'` to `extractLsb` plus a
cast.
Note that `extractLsb'` with a length of 0 returns `BitVec 0`, while
`extractLsb` will never return an empty bitvector (because of the `+ 1`
in it `hi - lo + 1`). Hence, this lemma needs a side condition that the
length is non-zero.
Also adds `getLsb_extractLsb'`
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
Since https://github.com/curl/curl/pull/4465 curl adheres to the
`Retry-After` header, so maybe this fixes the issues with
```
jq: error (at <stdin>:5): Cannot index string with string "body"
```
that sometimes make this workflow fail.
#4976 moved resolution of a promise to an earlier point, but that led to
object being marked MT earlier, so we need to move the code that
minimizes those objects earlier too to revert the performance
regression.
For structure projections, the pretty printer assumed that the
expression was type correct. Now it checks that the object being
projected is of the correct type. Such terms appear in type mismatch
errors.
Also, fixes and improves `#print` for structures. The types of
projections now use MessageData (so are now hoverable), and the type of
`self` is now the correct type.
Closes#4670
...unless we are about to kill the process anyway (which is not the
default)
Ensures panics are visible as regular messages in the language server
and properly ordered in relation to other messages on the cmdline
The prior default of 1000000 could not be achieved in practice, because
the stack would overflow after around 5000 recursive invocations. This
meant that a poorly-chosen @[ext] lemma could crash Lean.
Talking to Mathlib users, it seems that 10 would be a very large number
in practice, so a default limit of 100 should not change successful
uses. But it does make it much easier to diagnose and recover from poor
choices of @[ext] lemmas.
Without this change, a stack overflow on Mac OS during tactic execution
can lead to the message:
terminated by signal SIGBUS (Misaligned address error)
This comes from `lean_alloc_small`. With the change, the process instead
terminates with the more accurate and actionable:
Stack overflow detected. Aborting.
After having added already `BitVec.ushiftRight_*_distrib`in
https://github.com/leanprover/lean4/pull/4667 for ushiftRight, this PR
now completes the `*_distrib` theorems for shift.
Adds `@[app_delab ident]` as a macro for `@[delab app.ident]`. Resolves
the identifier when expanding the macro, saving needing to use the fully
qualified identifiers that `@[delab]` requires. Also, unlike `@[delab]`,
throws an error if the identifier cannot be resolved.
Closes#4899
Fixes an issue where each alternative in choice nodes would get their
own arguments. Now cdot function expansion is aware of choice nodes.
Also modifies the variable naming so that multi-argument functions like
`(· + ·)` expand as `fun x1 x2 => x1 + x2` rather than `fun x x_1 => x +
x_1`.
Closes#4832
This restores all of the imports of `Lean.Data.HashMap` and
`Lean.Data.HashSet` so that users actually see the deprecation warnings
instead of a "declaration not found" error.
This implements a naive version of `getline` because Windows does not
have `getline`. Given the fact that `FILE` has buffered IO, calling
`fgetc` in a loop is not as big of a performance hazard as it might seem
at first glance.
The proper solution to this would of course be to have our own buffered
IO so we are fully in charge of the buffer. In this situation we could
check the entire buffer for a newline at once instead of char by char.
However that is not going to happen for the near future so I propose we
stay with this implementation. If reading individual lines of a file
does truly end up being the performance bottle neck we have already
won^^.
Step 1 out of approximately 7 to upstream LeanSAT.
---------
Co-authored-by: Tobias Grosser <tobias@grosser.es>
Co-authored-by: Markus Himmel <markus@lean-fro.org>
This change canonicalizes the BitVec variable names to `x y z : BitVec`
instead of alternative namings such as `s t : BitVec` or `a b : BitVec`.
Variable names that carry semantic meaning such as `(msbs : BitVec w)
(lsb : Bool)` remain untouched.
This is purely a naming change to make our bitvector proofs more
consistent and polish the (auto-generated) documentation as a very small
step towards polishing the documentation of the BitVec library in Lean.
---------
Co-authored-by: AnotherAlexHere <153999274+AnotherAlexHere@users.noreply.github.com>
#4917 will expose users of the `Lean` API to the renaming of the hash
map query methods. This PR aims to make the transition easier by adding
deprecated functions with the old names.
With the recent unification of server and cmdline processing,
`IO.Process` tests that previously broke the server because they
directly wrote to stdout are now flaky on the cmdline because
elaboration and reporting are happening in separate threads. By removing
direct writes to stdout, the race condition is removed and the file can
actually be edited in the language server as well again.
This PR:
- changes the implementation of `readBinFile` and `readFile` to only
require two system calls (`stat` + `read`) instead of one `read` per
1024 byte chunk.
- fixes a bug where `Handle.getLine` would get tripped up by a NUL
character in the line and cut the string off. This is caused by the fact
that the original implementation uses `strlen` and `lean_mk_string`
which is the backer of `mk_string` does so as well.
- fixes a bug where `Handle.putStr` and thus by extension `writeFile`
would get tripped up by a NUL char in the line and cut the string off.
Cause here is the use of `fputs` when a NUL char is possible.
Closes: #4891Closes: #3546Closes: #3741
This PR resolves two language server bugs that especially affect Windows
users:
1. Editing the header could result in the watchdog not correctly
restarting the file worker (#3786, #3787), which would lead to the file
seemingly being processed forever.
- The cause of this issue was a race condition in the watchdog that was
accidentally introduced as far back as #1884: In specific circumstances,
the watchdog will attempt forwarding a message to the file worker after
the process has exited due to a changed header, but before the file
worker exiting has been noticed by the watchdog (which will then restart
the file worker). In this case, the watchdog would mark the file worker
as having crashed and not look at its exit code to restart the file
worker, but instead treat it like a crashed file worker that will only
be restarted when editing the file again. Not inspecting the exit code
of the file worker when it crashed from forwarding a message from the
file worker is necessary since we do not restart the file worker until
another notification from the client arrives, and so we would read the
same crash exit code over and over again in the main loop of the
watchdog if we did not remove it from our list of file workers that we
listen to.
- This PR resolves this issue by distinguishing between "crashes when
forwarding messages to the file worker" and "crashes when forwarding
messages from the file worker". In the former case, we still inspect the
exit code of the file worker and potentially restart it if the imports
changed, whereas in the latter case, we stop inspecting the exit code of
the file worker. This is correct because the latter case is exactly the
one where we need to stop inspecting the exit code but where a crash
cannot occur as a result of a changed header, whereas the former case is
exactly the one where we still need to inspect the exit code after a
crash to ensure that we restart the file worker in case it exited
because the header changed.
- At some point in the future, it would be nice to revamp the
concurrency model of the watchdog entirely now that we have all those
fancy concurrency primitives that were not available four years ago when
the watchdog was first written.
2. On an especially slow Windows machine, we found that starting the
language server would sometimes not succeed at all because reading from
the stdin pipe in the watchdog produced an EINVAL error, which was in
turn caused by an NT "pipe empty" error.
- After lots of debugging, @Kha found that Lake accidentally passes its
stdin to Git because it does not explicitly set the `stdin` field to
`null` when spawning the process.
- Changing this fixes the issue, which suggests that Git may mutate the
pipe we pass to it to be non-blocking, which then causes a "pipe empty"
error in the watchdog when we also attempt to read from that same pipe.
- I'm still very uncertain why we only saw this issue on one
particularly slow machine and not across the whole eco system.
This PR also resolves an issue where we would not correctly emit
messages that we received while the file worker is being restarted to
the corresponding file worker after the restart.
Closes#3786, closes#3787.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Autoparam tactic scripts have no source positions, which until recently
made it so that any errors or messages would be logged at the current
ref, which was the application or structure instance being elaborated.
However, with the new incrementality features the ref is now carefully
managed to avoid leakage of outside data. This inhibits the elaborator's
ref from being used for the tactic's ref, causing messages to be placed
at the beginning of the file rather than on the syntax that triggered
the autoparam.
To fix this, now the elaborators insert the ref's source position
everywhere into the autoparam tactic script.
If in the future messages for synthetic tactics appear at the tops of
files in other contexts, we should consider an approach where
`Lean.Elab.Term.withReuseContext` uses something like `replaceRef` to
set the ref while disabling incrementality when the tactic does not
contain source position information.
Closes#4880
Currently, the messages in the diagnostic summaries are created by
appending interpolated strings. We wrap these in `.trace`'s, and the
results are better formatted when expanding child nodes in the info
view. Particularly, the latter diagnostic summaries remain on their own
lines flush to the left instead of on the same line directly adjacent to
the last child node.
For experimentation by @the-sofi-uwu.
I also have an efficient number parser in LeanSAT that I am planning to
upstream after we have sufficiently bikeshed this change.
When `set_option diagnostics true`, for each theorem with size >
`diagnostics.threshold.proofSize`, display proof size, and the number of
applications for each constant symbol.
TODO:
- Support for `zeta := true` at `apply_beta`.
- Investigate test failure.
- Break PR in pieces because of bootstrapping issues. The current PR
updates a stage0 file to workaround the issue.
Motivation: significant performance improvement at
https://github.com/leanprover/LNSym/blob/proof_size_expt/Proofs/SHA512/Experiments/Sym30.lean
With M1 Pro:
- Before: 4.56 secs
- After: 3.16 secs
Successfully built stage2 using this PR
This modification improves the performance of the example in issue
#4861. It no longer times out but is still expensive.
Here is the analysis of the performance issue: Given `(x : Int)`, to
elaborate `x ^ 1`, a few default instances have to be tried.
First, the homogeneous instance is tried and fails since `Int` does not
implement `Pow Int`. Then, the `NatPow` instance is tried, and it also
fails. The same process is performed for each term of the form `p ^ 1`.
There are seveal of them at #4861. After all of these fail, the lower
priority default instance for numerals is tried, and `x ^ 1` becomes `x
^ (1 : Nat)`. Then, `HPow Int Nat Int` can be applied, and the
elaboration succeeds. However, this process has to be repeated for every
single term of the form `p ^ 1`. The elaborator tries all homogeneous
`HPow` and `NatPow` instances for all `p ^ 1` terms before trying the
lower priority default instance `OfNat`.
This commit ensures `Int` has a `NatPow` instance instead of `HPow Int
Nat Int`. This change shortcuts the process, but it still first tries
the homogeneous `HPow` instance, fails, and then tries `NatPow`. The
elaboration can be made much more efficient by writing `p ^ (1 : Nat)`.
Those represent ~13% of the time spent in `save_result`,
even though `r` is a temporary in all cases but one.
See #4698 for details.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
…rators
Right now those constructors result in a copy instead of the desired
move. We've measured that expr copying and assignment by itself uses
around 10% of total runtime on our workloads.
See #4698 for details.
Initial options are now re-parsed and validated after importing. Cmdline
option assignments prefixed with `weak.` are silently discarded if the
option name without the prefix does not exist.
Fixes#3403
This allows bitblasting `BitVec.replicate`.
I changed the definition of `BitVec.replicate` to use `BitVec.cast` in
order to make the proof smoother, since it's an easier time simplifying
away terms with `BitVec.cast`.
---------
Co-authored-by: Tobias Grosser <tobias@grosser.es>
This message is often incorporated into source files via `#guard_msgs`.
This change ensures it won't go over the 100 character ruler, and I
think is equally grammatical. :-)
It is confusing that the message suggesting to use the `diagnostics`
option is given even when the option is already set. This PR makes use
of lazy message data to make the message contingent on the option being
false.
It also tones down the promise that there is any diagonostic information
available, since sometimes there is nothing to report.
Suggested by Johan Commelin.
now that we support structural mutual recursion, I expect that every
`DecidableEq` instance be implemented using structural recursion, so
let's be explicit about it.
Some eliminators (such as `False.rec`) have an explicit motive argument.
The `elabAsElim` elaborator assumed that all motives are implicit.
If the explicit motive argument is `_`, then it uses the elab-as-elim
procedure, and otherwise it falls back to the standard app elaborator.
Furthermore, if an explicit elaborator is not provided, it falls back to
treating the elaborator as being implicit, which is convenient for
writing `h.rec` rather than `h.rec _`. Rationale: for `False.rec`, this
simulates it having an implicit motive, and also motives are generally
not going to be available in the expected type.
Closes#4347
Before, the delaborator was conservative about omitting optional
arguments, only omitting the very last one. Now it can omit arbitrarily
long sequences of optional arguments from the end.
For simplicity of implementation, every optional argument is delaborated
and then potentially discarded. It could save state and lazily
delaborate, but we're running under the hypothesis that most optional
arguments are for very simple values (like `true`, `false`, or a numeric
literal), so it is unlikely that efficiency gains, if any, are worth it.
In particular, in the future structure constructors will have optional
arguments, but `unexpandStructureInstance` assumes none of the optional
fields are omitted.
Closes#4812
when transforming the `match` statements in `IndPredBelow`, given a
local variable `x : T`, we need to search for `hx : T.below x`.
Previously this was done using the custom `backwardsChaining` method,
although my hypothesis is that we don’t need to chain anything here, and
can use `apply_assumption`.
this improves support for structural recursion over inductive
*predicates* when there are reflexive arguments.
Consider
```lean
inductive F: Prop where
| base
| step (fn: Nat → F)
-- set_option trace.Meta.IndPredBelow.search true
set_option pp.proofs true
def F.asdf1 : (f : F) → True
| base => trivial
| step f => F.asdf1 (f 0)
termination_by structural f => f`
```
Previously the search for the right induction hypothesis would fail with
```
could not solve using backwards chaining x✝¹ : F
x✝ : x✝¹.below
f : Nat → F
a✝¹ : ∀ (a : Nat), (f a).below
a✝ : Nat → True
⊢ True
```
The backchaining process will try to use `a✝ : Nat → True`, but then has
no idea what to use for `Nat`.
There are three steps here to fix this.
1. We let-bind the function's type before the whole process. Now the
goal is
```
funType : F → Prop := fun x => True
x✝ : x✝¹.below
f : Nat → F
a✝¹ : ∀ (a : Nat), (f a).below
a✝ : ∀ (a : Nat), funType (f a)
⊢ funType (f 0)
```
2. Instead of using the general purpose backchaining proof search, which
is more
powerful than we need here (we need on recursive search and no
backtracking),
we have a custom search that looks for local assumptions that
provide evidence of `funType`, and extracts the arguments from that
“type” application to construct the recursive call.
Above, it will thus unify `f a =?= f 0`.
3. In order to make progress here, we also turn on use
`withoutProofIrrelevance`,
because else `isDefEq` is happy to say “they are equal” without actually
looking
at the terms and thus assigning `?a := 0`.
This idea of let-binding the function's motive may also be useful for
the other recursion compilers, as it may simplify the FunInd
construction. This is to be investigated.
fixes#4751
The function `locationLinksFromDecl` could throw an error if the name it
is provided doesn't exist in the environment, which is possible if for
example an elaborator is a builtin.
Closes#3789
Adds the `--log-level=<lv>` CLI option for controlling the minimum log
level Lake should output. For instance, `--log-level=error` will only
print errors (not warnings or info).
Also, adds the parallel `--fail-level` CLI option to control what the
minimum log level of build failures is. The existing `--iofail` and
`--wfail` options are equivalent to `--fail-level=info` and
`--fail-level=warning` , respectively.
Closes#4805,
Due to nested recursion, we do two passes of `getRecArgInfo`: One on
each argument in isolation, to see which inductive types are around
(e.g. `Tree` and `List`), and
then we later refine/replace this result with the data for the nested
type former (the implicit `ListTree`).
If we have nested recursion through a non-recursive data type like
`Array` or `Prod` then arguemnts of these types should survive the first
phase, so that we can still use them when looking for, say, `Array
Tree`.
This was helpfully reported by @arthur-adjedj.
For every parenthesized expression `(foo)`, the InfoView produces an
interactive component both for `(foo)` itself and its subexpression
`foo` because the corresponding `TaggedText` in the language server is
duplicated as well. Both of these subexpressions have the same
subexpression position and so they are identical w.r.t. interactive
features.
Removing this duplication would help reduce the size of the DOM of the
InfoView and ensure that the UI for InfoView features is consistent for
`(foo)` and `foo` (e.g. hovers would always highlight `(foo)`, not
either `(foo)` or `foo` depending on whether the mouse cursor is on the
bracket or not). It would also help resolve a bug where selecting a
subexpression will yield selection highlighting both for `(foo)` and
`foo`, as we use the subexpression position to identify which terms to
highlight.
This PR adjusts the parenthesizer to move the corresponding info instead
of duplicating it.
Draft of adding ci workflow using lean-action on `lake new/init`
This PR is currently missing lake options for the user to control this
feature.
Closes#4606
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
When resolving anonymous dot notation (`.ident x y z`), it would reduce
the expected type to whnf. Now, it unfolds definitions step-by-step,
even if the type synonym is for a pi type like so
```lean
def Foo : Prop := ∀ a : Nat, a = a
protected theorem Foo.intro : Foo := sorry
example : Foo := .intro
```
Closes#4761
After each tactic step, we save the info tree created by it together
with an appropriate info tree context that makes it stand-alone (which
we already did before to some degree, see `Info.updateContext?`). Then,
in the adjusted request handlers, we first search for a snapshot task
containing the required position, if so wait on it, and if it yielded an
info tree, use it to answer the request, or else continue searching and
waiting, falling back to the full info tree, which should be unchanged
by this PR.
The definition header does *not* report info trees early as in general
it is not stand-alone in the tactic sense but may contain e.g.
metavariables solved by the body in which case we do want to show the
ultimate state as before. This could be refined in the future in case
there are no unsolved mvars.
The adjusted request handlers are exactly the ones waited on together by
the info view, so they all have to be adjusted to have any effect on the
UX. Further request handlers may be adjusted in the future.
No new tests as "replies early" is not something we can test with our
current framework but the existing test suite did help in uncovering
functional regressions.
previously, `#eval` would happily evaluate expressions that contain
`sorry`, either explicitly or because of failing tactics. In conjunction
with operations like array access this can lead to the lean process
crashing, which isn't particularly great.
So how `#eval` will refuse to run code that (transitively) depends on
the `sorry` axiom (using the same code as `#print axioms`).
If the user really wants to run it, they can use `#eval!`.
Closes#1697
This PR updates the screenshots and instructions in the quickstart guide
for the most recent Lean 4 VS Code extension version and makes a small
stylistic change suggested by @semorrison.
The `elab_as_elim` elaborator eagerly elaborates arguments that can help
with elaborating the motive, however it does not include the transitive
closure of parameters appearing in types of parameters appearing in ...
types of targets.
This leads to counter-intuitive behavior where arguments supplied to the
eliminator may unexpectedly have postponed elaboration, causing motives
to be type incorrect for under-applied eliminators such as the
following:
```lean
class IsEmpty (α : Sort u) : Prop where
protected false : α → False
@[elab_as_elim]
def isEmptyElim [IsEmpty α] {p : α → Sort _} (a : α) : p a :=
(IsEmpty.false a).elim
example {α : Type _} [IsEmpty α] :
id (α → False) := isEmptyElim (α := α)
```
The issue is that when `isEmptyElim (α := α)` is computing its motive,
the value of the postponed argument `α` is still an unassignable
metavariable. With this PR, this argument is now among those that are
eagerly elaborated since it appears as the type of the target `a`.
This PR also contains some other fixes:
* When underapplied, does unification when instantiating foralls in the
expected type.
* When overapplied, type checks the generalized-and-reverted expected
type.
* When collecting targets, collects them in the correct order.
Adds trace class `trace.Elab.app.elab_as_elim`.
This is a followup to #4722, which added motive type checking but
exposed the eagerness issue.
Also extends existing definition for `getScope`/`getScopes` and
clarifies that the `end` command is optional at the end of a file.
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
The original idea was to use `bif` in computation contexts and `if` in
propositional contexts, but this turned out to be really inconvenient in
practice.
code to create nested `PProd`s, and project out, and related functions
were scattered in variuos places. This unifies them in
`Lean.Meta.PProdN`.
It also consistently avoids the terminal `True` or `PUnit`, for slightly
easier to read constructions.
This refactoring PR changes the structure of the `FunInd` module, with
the main purpose to make it easier to support mutual structural
recursion.
In particular the recursive calls are now longer recognized by their
terms (simple for well-founded recursion, `.app oldIH [arg, proof]`, but
tedious for structural recursion and even more so for mutual structural
recursion), but the type after replacing `oldIH` with `newIH`, where the
type will be simply and plainly `mkAppN motive args`).
We also no longer try to guess whether we deal with well-founded or
structural recursion but instead rely on the `EqnInfo` environment
extensions. The previous code tried to handle both variants, but they
differ too much, so having separate top-level functions is easier.
This also fuses the `foldCalls` and `collectIHs` traversals and
introduces a suitable monad for collecting the inductive hypotheses.
This is part 2 of 2 of #4801 (which closes#4654). That PR was split in
two to allow a stage0 update between declaring the `usize` functions and
using them where they are needed.
Add efficient `usize` functions for `Array`, `ByteArray`, `FloatArray`.
This is part 1 of 2 since there is a need to update stage0 between the
two parts. (See discussion below.)
Closes#4654
Changes:
- We avoid the thread local storage.
- We use a hash map to ensure that cached values are not lost.
- We remove `check_system`. If this becomes an issue in the future we
should precompute the remaining amount of stack space, and use a cheaper
check.
- We add a `Expr.replaceImpl`, and will use it to implement
`Expr.replace` after update-stage0
Declarations with `@[elab_as_elim]` could elaborate as type-incorrect
expressions. Reported by Jireh Loreaux [on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/287929-mathlib4/topic/bug.20in.20revert/near/450522157).
(In principle the elabAsElim routine could revert fvars appearing in the
expected type that depend on the discriminants (if the discriminants are
fvars) to increase the likelihood of type correctness, but that's at the
cost of some complexity to both the elaborator and to the user.)
Now it suggests using `@[ext (iff := false)]` to disable generating the
`ext_iff` lemma.
This PR also adjusts error messages and attribute documentation.
Additionally, to simplify the code now the `x` and `y` arguments can't
come in reverse order (this feature was was added in the refactor
#4543).
Closes#4758
A more restrictive but efficient max sharing primitive.
**Motivation:** Some software verification proofs may contain
significant redundancy that can be eliminated using hash-consing (also
known as `shareCommon`). For example, [theorem
`sha512_block_armv8_test_4_sym`](460fe5d74c/Proofs/SHA512/SHA512Sym.lean (L29))
took a few seconds at [`addPreDefinitions`
](1a12f63f74/src/Lean/Elab/PreDefinition/Main.lean (L155))
and one second at `fixLevelParams` on a MacBook Pro (with M1 Pro). The
proof term initially had over 16 million subterms, but the redundancy
was indirectly and inefficiently eliminated using `Core.transform` at
`addPreDefinitions`. I tried to use `shareCommon` method to fix the
performance issue, but it was too inefficient. This PR introduces a new
`shareCommon'` method that, although less flexible (e.g., it uses only a
local cache and hash-consing table), is much more efficient. The new
procedure minimizes the number of RC operations and optimizes the
caching strategy. It is 20 times faster than the old `shareCommon`
procedure for theorem `sha512_block_armv8_test_4_sym`.
I noticed that a change to `Lean.PrettyPrinter.Delaborator.Builtins`
rebuilt more modules than I expected, so I moved a definition and
reduced some dependcies.
More reduction would be possible to move const-delaboration out of the
big `Lean.PrettyPrinter`, and import from `Lean.PrettyPrinter`
selectively.
Add helper function for computing the number of allocated
sub-expressions in a given expression. Note: Use this function primarily
for diagnosing performance issues.
This PR addresses the absence of the `profileitM` function in two
auxiliary functions. The added `profileitM` instances are particularly
useful for diagnosing performance issues in declarations that contain
many repeated sub-terms.
The name `remove` was chosen because it is more popular in mainstream
programming languages, but being consistent with other Lean container
types (including `Lean.HashMap` and `Batteries.HashMap`) is more
important, so let's change the name while we still can.
the internal constructions for structural and well-founded recursion
use plenty of `PProd` and `MProd`, and reading these, deeply
nested and in prefix notation, is unnecessarily troublesome.
Therefore this introduces notations
```
a ×ₚ b -- PProd a b
a ×ₘ b -- MProd a b
()ₚ -- PUnit.unit
(x,y,z)ₚ -- PProd.mk x (PProd.mk y z)
(x,y,z)ₘ -- MProd.mk x (MProd.mk y z)
```
(This is the post-stage0-part 2.)
the internal constructions for structural and well-founded recursion
use plenty of `PProd` and `MProd`, and reading these, deeply
nested and in prefix notation, is unnecessarily troublesome.
Therefore this introduces notations
```
a ×ₚ b -- PProd a b
a ×ₘ b -- MProd a b
()ₚ -- PUnit.unit
(x,y,z)ₚ -- PProd.mk x (PProd.mk y z)
(x,y,z)ₘ -- MProd.mk x (MProd.mk y z)
```
(This is part 1, the rest will follow in #4730 after a stage0 update.)
This now works:
```lean
inductive Tree where | node : List Tree → Tree
mutual
def Tree.size : Tree → Nat
| node ts => list_size ts
def Tree.list_size : List Tree → Nat
| [] => 0
| t::ts => t.size + list_size ts
end
```
It is still out of scope to expect to be able to use nested recursion
(e.g. through `List.map` or `List.foldl`) here.
Depends on #4718.
---------
Co-authored-by: Tobias Grosser <tobias@grosser.es>
the support for mutual structural recursion (new since #4575) is
extended so that Lean tries to infer it even without annotations.
* The error message when termination checking fails looks quite
different now. Maybe a bit better, maybe with more room for
improvements.
* If there are too many combinations (with an arbitrary cut-off) for a
given argument type, it will just give up and ask the user to use
`termination_by structural`.
* It is now legal to specify `termination_by structural` on not
necessarily all functions of a clique; this simply restricts the
combinations of arguments that Lean considers.
---------
Co-authored-by: Tobias Grosser <tobias@grosser.es>
This adds the types
* `IndGroupInfo`, a variant of `InductiveVal` with information that
applies to a whole group of mutual inductives and
* `IndGroupInst` which extends `IndGroupInfo` with levels and parameters
to indicate a instantiation of the group.
One purpose of this abstraction is to make it clear when a fuction
operates on a group as a whole, rather than a specific inductive within
the group.
This is extracted from #4718 and #4733 to reduce PR size and improve
bisectability.
Improves a number of elements related to Git checkouts, cloud releases,
and related error handling.
* On error, Lake now prints all top-level logs. Top-level logs are those
produced by Lake outside of the job monitor (e.g., when cloning
dependencies).
* When fetching a remote for a dependency, Lake now forcibly fetches
tags. This prevents potential errors caused by a repository recreating
tags already fetched.
* Tweaked Git error handling to hopefully be more informative.
* The builtin package facets `release`, `optRelease`, `extraDep` are now
caption in the same manner as other facets. Previously, they were
attempting to be too clever.
* `afterReleaseSync` and `afterReleaseAsync` now fetch `optRelease`
rather than `release`.
* Added support for optional jobs, whose failure does not cause the
whole build to failure (and made `optRelease` such a job).
Closes#4302.
We now get `.below` and `.brecOn` definitions for nested inductives.
No surprises in the implementation: the kernel already gives us suitable
`.rec_1` etc. recursors, and our construction follows the structure of
this recursor.
---------
Co-authored-by: Tobias Grosser <tobias@grosser.es>
Adds a command and tactic to print the `Array <| DiscrTree.Key` for
equalities helping the user to debug perceived `simp` failures.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
this idiom shows up multiple times, is non-trivial (in the sense that
the `localInsts` has to be updated, and I am about to use it once more.
Hence time to abstract this out.
When the `decide` tactic fails, it can try to give hints about the
failure:
- It tells you which `Decidable` instances it unfolded, by making use of
the diagnostics feature.
- If it encounters `Eq.rec`, it gives you a hint that one of these
instances was likely defined using tactics.
- If it encounters `Classical.choice`, it hints that you might have
classical instances in scope.
- During this, it tries to process `Decidable.rec`s and matchers to pin
blame on a particular instance that failed to reduce.
This idea comes from discussion with Heather Macbeth [on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/Decidable.20with.20structures/near/449409870).
this code
```
inductive N where
| cons : (Nat -> N) -> N
mutual
def f : N -> Nat
| .cons a => g (a 32) + 1
termination_by structural n => n
def g : N -> Nat
| .cons a => f (a 42) + 1
termination_by structural n => n
end
```
would break. When searching for the right `belowDict` we now have to,
evne after instantiating the paramters for a reflexive argument, again
search through a bunch of `PProd`s.
(Instead of searching we could pass down the index, but since we are
searching anyways in this function let's just re-use.)
Fixes: #4726
Matchers usually have implicit arguments, and even if they don't the
notation hides the name of the matcher function.
Now when hovering over `match` expressions you can see the actual
underlying matcher expression.
if will fail otherwise, but with a worse error message, and it's helpful
in later transformation to know that the parameters are the same for the
whole group.
This makes it reflect how we are writing release notes for 4.9.0,
including how to handle the `releases_drafts` folder and how and when to
update `RELEASES.md`.
Co-authored-by: Kim Morrison <kim@tqft.net>
Upstreaming of basic material on `List.Pairwise` and `List.Nodup`. More
complete API to follow later, this is just a first approximation of what
leansat will need.
When a definition is redeclared, the original code would clobber the
value of `const2ModIdx` every time, meaning that a constant would be
attributed to a module which occurs later than the modules for constants
referencing this one. Preferring the original module ensures that these
module indexes are dependency-ordered. This originally came up as a bug
in `shake`, which assumes this property, see
[Zulip](https://leanprover.zulipchat.com/#narrow/stream/287929-mathlib4/topic/check.20for.20unused.20imports.20doesn't.20stop/near/449139309).
I'll update `list_simp.lean` (simp normal form testing) and add missing
lemmas in follow-up PRs.
This just upstreams the material, and reorders the lemmas to match the
other sections.
Adds syntactic sugar specifying a git revision as a dependency version
in a `require` command. For example:
```
require "leanprover-community" / "proofwidgets" @ git "v0.0.39"
```
This PR refactors the 'ext' attribute and implements the following
features:
- The 'local' and 'scoped' attribute kinds are now usable.
- The attribute realizes the `ext`/`ext_iff` lemmas when they do not
already exist, rather than always generating them. This is useful in
conjunction with `@[local ext]`.
- Adding `@[ext]` to a user ext lemma now realizes an `ext_iff` lemma as
well; formerly this was only for structures. The name of the generated
`ext_iff` theorem for a user `ext` theorem named `A.B.myext` is
`A.B.myext_iff`. If this process leads to an error, the user can write
`@[ext (iff := false)]` to disable this feature.
Breaking changes:
- Now the "x" and "y" term arguments to the realized `ext` and `ext_iff`
lemmas are implicit.
- Now the realized `ext` and `ext_iff` lemmas are protected.
Bootstrapping notes:
- There are a few `ext_iff` lemmas to address after the next stage0
update.
Closes https://github.com/leanprover/lean4/issues/3643
Suggested by Floris [on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/113488-general/topic/.22Missing.20Tactics.22.20list/near/446267660).
right now, in order to find out how many auxilary datatype are in a
mutual group of inductive with nested data type, one has to jump
through hoops like this:
```
private def numNestedInducts (indName : Name) : MetaM Nat := do
let .inductInfo indVal ← getConstInfo indName | panic! "{indName} is an inductive"
let .recInfo recVal ← getConstInfo (mkRecName indName) | panic! "{indName} has a recursor"
return recVal.numMotives - indVal.all.lengt
```
The `InductiveVal` data structure already has `.isNested : Bool`, so it
seems to be a natural extension to beef that up to `.numNested: Nat`.
This touched kernel code.
This adds support for mutual structural recursive functions.
For now this is opt-in: The functions must have a `termination_by
structural …` annotation (new since #4542) for this to work:
```lean
mutual
inductive A
| self : A → A
| other : B → A
| empty
inductive B
| self : B → B
| other : A → B
| empty
end
mutual
def A.size : A → Nat
| .self a => a.size + 1
| .other b => b.size + 1
| .empty => 0
termination_by structural x => x
def B.size : B → Nat
| .self b => b.size + 1
| .other a => a.size + 1
| .empty => 0
termination_by structural x => x
end
```
The recursive functions don’t have to be in a one-to-one relation to a
set of mutually recursive inductive data types. It is possible to ignore
some of the types:
```lean
def A.self_size : A → Nat
| .self a => a.self_size + 1
| .other _ => 0
| .empty => 0
termination_by structural x => x
```
or have more than one function per argument type:
```lean
def isEven : Nat → Prop
| 0 => True
| n+1 => ¬ isOdd n
termination_by structural x => x
def isOdd : Nat → Prop
| 0 => False
| n+1 => ¬ isEven n
termination_by structural x => x
```
This does not include
* Support for nested inductive data types or nested recursion
* Inferring mutual structural recursion in the absence of
`termination_by`.
* Functional induction principles for these.
* Mutually recursive functions that live in different universes. This
may be possible,
maybe after beefing up the `.below` and `.brecOn` functions; we can look
into this some
other time, maybe when there are concrete use cases.
---------
Co-authored-by: Richard Kiss <him@richardkiss.com>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
This file has comments that recall the data type definitions in Lean.
Most of them were still using lean3 syntax, and at least one of them was
out of date (one field missing), so I updated them.
I took the liberty to shorten the comments from the original file, or
omit them if they don’t add much over the field names.
Generalizes #3556 to not suppressing errors in tactic steps either when
the parse error is in a later step, as otherwise changes to the end of a
proof would affect (correctness or effectiveness of) incrementality of
preceding steps.
Fixes#4623, in combination with #4643
As we do not build multiple shared libraries on non-Windows anymore,
count the max exported symbols per static library instead.
Unfortunately, this still does seem to match the number on Windows.
The previous check, looking only at the type of the parameter, was too
permissive and led to ill-typed terms later on.
This fixes#4671.
In some cases the previous code might have worked by accident, in this
sense this is a breaking change. Affected functions can be fixed by
reordering their parameters to that all the function parameters that
occur in the parameter of the inductive type of the parameter that the
function recurses on come first.
Fixes the Windows build. As libLean is by far the biggest component,
there is no need for a separate libStd_shared for now.
```
$ find build/release/stage1/lib/lean -name '*.a' -exec bash -c 'echo -n "{} " ; nm {} | grep " T " | wc -l' \;
build/release/stage1/lib/lean/libleanrt.a 497
build/release/stage1/lib/lean/libleancpp.a 1320
build/release/stage1/lib/lean/libInit.a 7476
build/release/stage1/lib/lean/libStd.a 1696
build/release/stage1/lib/lean/libLean.a 64339
build/release/stage1/lib/lean/libLake.a 5722
```
In #3911, a refactor to share `MessageData` code between `ppConst` and
the signature pretty printer unintentionally caused the signature pretty
printer to use the `pp.tagAppFns` option. This causes, for example, `+`
in `a + b` to independently have its own hover information due to the
fact that `notation` app unexpanders use the head function's syntax as
the `ref` when constructing the notation syntax. This behavior of
`pp.tagAppFns` is intentional, and it is used by docgen, but it should
not be activated for signatures.
This affects `#check` and was reported by Kevin Buzzard [on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/degraded.20hover.20experience.20on.20.23check/near/449380674).
This PR also makes sure the initial `ref` when applying app unexpanders
is `.missing`, rather than whatever random value might be present in the
`CoreM` context.
### Preliminary PRs:
- [x] #4597
- [x] #4599
- [x] #4600
- [x] #4602
- [x] #4603
- [x] #4604
- [x] #4605
- [x] #4607
- [x] #4627
- [x] #4629
### Quick overview over API/naming changes compared to `Lean.HashMap`
and `Batteries.HashMap`:
#### Lean
* `find?` -> `get?`/`getElem?`
* `find!` -> `get!`/`gtetElem!`
* `findD` -> `getD`
* `findEntry?` -> not implemented for now
* `insert'` -> `containsThenInsert` (order reversed in result)
* `insertIfNew` -> `getThenInsertIfNew?` (order reversed in result)
* `numBuckets` -> `Internal.numBuckets`
* `ofListWith` -> not implemented for now
* `Array.groupByKey` -> not implemented for now
* `merge` -> not implemented for now, but you can use `insertMany`
#### Batteries
* `modify` -> not implemented for now
* `mergeWith` -> not implemented for now
* `mergeWithM` -> not implemented for now
I made a mistake in #4517, fixed here, so about time to add a test.
I wonder if this generic level optimization should be moved into
`mkLevelMax'`, but not today.
fixes#4650
Split from #4583
There are two open questions, opinions appreciated:
- Should this material be part of `Init` or `Std`?
- Should the typeclasses be in the `Std` namespace?
This is an auxiliary procedured used by `rw` and `apply` tactics. It
synthesizes pending type class instances.
The new test contains an example where it failed. The comment at
`synthAppInstances.step` explains why, and the fix.
we have a `forallBoundedTelescope`, and for a long while I was
wondering why we also don't have `lambdaBoundedTelescope`, and every now
and then felt the need for it. So let's just add it.
this job sometimes fails, maybe a race condition with the `gh run
cancel` not happenign quickly enough. Maybe more verbose output will
help understand this better.
Now syntax nodes have their formatters run even if the parsers they wrap
are all arity zero. This fixes an issue where if `ppSpace` appears in a
`macro`/`elab` then it does not format with a space due to the fact that
macro argument processing wraps this as `group(ppSpace)`, and `ppSpace`
has arity zero.
Implementation note: the fix is to make the `visitArgs` formatter
combinator always visit the last child, even if it does not exist (in
which case the visited node will be `Syntax.missing`). To compensate,
parser combinators like many and optional need to be sure to keep track
of whether there any children. Only optional's needed to be modified.
Closes#4561
Summary:
- Adds configuration option `exponentiation.threshold`
- An expression `b^n` where `b` and `n` are literals is not reduced by
`whnf`, `simp`, and `isDefEq` if `n > exponentiation.threshold`.
Motivation: prevents system from becoming irresponsive and/or crashing
without memory.
TODO: improve support in the kernel. It is using a hard-coded limit for
now.
Before, `pp.instantiateMVars` generally had no effect because most call
sites for the pretty printer instantiated metavariables first, but now
this functionality is entrusted upon the `pp.instantiateMVars` option.
This also has an effect in hovers, where metavariables can be unfolded
one assignment at a time. However, the goal state still sees all
metavariables instantiated due to the fact that the algorithm relies on
expression equality post-instantiation (see
`Lean.Widget.goalToInteractive`).
Closes#4406
Closes#2736
See comment at `ExprDefEq.lean` for explanation.
Side effects:
- Improved error messages in two tests.
- Had to improve `getSuccesses` procedure at `App.lean`. It now
discards candidates that contain postponed elaboration problems.
If it is too disruptive for Mathlib, we should try to discard the
ones that have postponed metavariables.
Split from #4583
`exists_of_set` appears in Batteries as `exists_of_set'`. The
`exists_of_set` version is unused in batteries and mathlib at least and
I would argue that the primed version (i.e., the one added in this PR)
is always better anyway.
`isEmpty_iff` appears in mathlib as `isEmpty_iff_eq_nil`.
Fixes#4591. The extra code already existed in the only other user of
`unresolveNameGlobal` (in the pretty printer), although I did not make
it use this function because it has some additional behavior around
universes and in pattern position.
This implements the recurrence theorems `getLsb_mul`, `mulRec_zero_eq`,
`mulRec_succ_eq` to allow bitblasting multiplication.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
This implements the `termination_by structural` syntax proposed in
#3909.
I went with `termination_by structural` over, say,
`termination_by (config := {method := .structural})` mainly because it
was
easier to get going (otherwise I’d have to look into how to define
recursive
parsers, as `Parser.config` depends on `term` and `termination_by` is
part of
term. But also because I find it more ergonomic and aesthetic as a user.
But syntax can still change.
The `termination_by?` syntax will no longer force well-founded
recursion,
and instead the inferred `termination_by structurally` annotation will
be shown
if structural termination is possible.
While I was it, this fixes#4546 the easy way (log errors about but
otherwise
ignore incomplete `termination_by` sets for mutual recursion). Maybe we
get
multiple replacements (#4551), but even then this this good behavior.
Involves a bit of shuffling around `TerimationHints` (now validated for
a
clique already by `PreDefinition.main`) and `TerminationArguments` (now
lifted
out of the `WF` namespace, and a bit simplified).
Fixes#3909
---------
Co-authored-by: Richard Kiss <him@richardkiss.com>
using the order as it comes out of the `HashMap` led to annying test
suite output variations. Moreover, sorting by the canonical order leads
to messages that are probably easier to digest as a user.
Adds linkage to `Std` so the build behaviour on darwin is in line with
linux
I'm not sure why linking with `Std` is needed. I deleted it in the
previous patch https://github.com/leanprover/lean4/pull/3811 and Lean
still builds and runs. @tydeu mentioned this issue so I created this PR.
Adds a new type of `require` which fetches package metadata from a
registry API endpoint (i.e., Reservoir) and then clones a Git package
using the information provided. To require such a dependency, the new
syntax is:
```lean
require <scope> / <pkg-name> [@ "git#<rev>"] -- e.g., require "leanprover" / "doc-gen4"
```
Or in TOML:
```toml
[[require]]
name = "<pkg-name>"
scope = "<scope>"
rev = "<rev>"
```
Unlike with Git dependencies, Lake can make use of the richer
information provided by the registry to determine the default branch of
the package. This means for repositories of packages like `doc-gen4`
which have a default branch that is not `master`, Lake will now use said
default branch (e.g., in `doc-gen4`'s case, `main`).
Lake also supports configuring the registry endpoint via an environment
variable: `RESERVIOR_API_URL`. Thus, any server providing a similar
interface to Reservoir can be used as the registry. Further
configuration options paralleling those of Cargo's [Alternative
Registries](https://doc.rust-lang.org/cargo/reference/registries.html)
and [Source
Replacement](https://doc.rust-lang.org/cargo/reference/source-replacement.html)
will come in the future.
Updated and split from #3174.
This example, reported from LNSym, started failing when we changed the
definition of `Fin.sub` in
https://github.com/leanprover/lean4/pull/4421.
When we use the new definition, `omega` produces a proof term that the
kernel is very slow on.
To work around this for now, I've removed `BitVec.toNat_sub` from the
`bv_toNat` simp set,
and replaced it with `BitVec.toNat_sub'` which uses the old definition
for subtraction.
This is only a workaround, and I would like to understand why the term
chokes the kernel.
```
example
(n : Nat)
(addr2 addr1 : BitVec 64)
(h0 : n ≤ 18446744073709551616)
(h1 : addr2 + 18446744073709551615#64 - addr1 ≤ BitVec.ofNat 64 (n - 1))
(h2 : addr2 - addr1 ≤ addr2 + 18446744073709551615#64 - addr1) :
n = 18446744073709551616 := by
bv_omega
```
The new option `set_option debug.skipKernelTC true` is meant for
temporarily working around kernel performance issues.
It compromises soundness because a buggy tactic may produce an invalid
proof, and the kernel will not catch it if the new option is set to true.
Remark: I had to comment
```
if debug.skipKernelTC.get opts then
addDeclWithoutChecking env decl
else
```
because the build was crashing when trying to compile Lake.
Going to perform `update-stage0` and try again.
Addresses a few issues with precompile library computation.
* Fixes a bug where Lake would always precompile the package of a
module.
* If a module is precompiled, it now precompiles its imports.
Previously, it would only do this if imported.
Closes#4565.
This appears to have been a semantic merge conflict between #3940 and
#4129. The effect on the language server is that if two edits are
sufficiently close in time to create an interrupt, some elaboration
steps like `simp` may accidentally catch the exception when it is
triggered during their execution, which makes incrementality assume that
elaboration of the body was successful, which can lead to incorrect
reuse, presenting the interrupted state to the user with symptoms such
as "uses sorry" without accompanying errors and incorrect lints.
When the type of a definition or example is a proposition,
we should elaborate on them as we elaborate on theorems.
This is particularly important for examples that are often
used in educational material.
Recall that when elaborating theorem headers, we convert unassigned
universe metavariables into universe parameters. The motivation is
that the proof of a theorem should not influence its statement.
However, before this commit, this was not the case for definitions and
examples when their type was a proposition. This discrepancy often
confused users.
Additionally, we considered extending the above behavior whenever
the type of a definition is provided. That is, we would keep the
current behavior only if `: <type>` was omitted in a definition.
However, this proved to be too restrictive.
For example, the following instance in `Core.lean` would fail:
```
instance {α : Sort u} [Setoid α] : HasEquiv α :=
⟨Setoid.r⟩
```
and we would have to write instead:
```
instance {α : Sort u} [Setoid α] : HasEquiv.{u, 0} α :=
⟨Setoid.r⟩
```
There are other failures like this in the core, and we assume many more
in Mathlib.
closes#4398
@semorrison @jcommelin: what do you think?
this is in preparation for #4542, and extracts from `findRecArg` the
functionality for trying one particular argument.
It also refactors the code a bit. In particular
* It reports errors in the order of the parameters, not the order of
in which they are tried (it tries non-indices first).
* For every argument it will say why it wasn't tried, even if the
reason is quite obviously (fixed prefix, or `Prop`-typed etc.)
Therefore there is some error message churn.
This ports the `.below` and `.brecOn` constructions to lean.
I kept them in the same file, as they were in the C code, because they
are
highly coupled and the constructions are very analogous.
For validation I developed this in a separate repository at
https://github.com/nomeata/lean-constructions/tree/fad715e
and checked that all declarations found in Lean and Mathlib are
equivalent, up to
def canon (e : Expr) : CoreM Expr := do
Core.transform (← Core.betaReduce e) (pre := fun
| .const n ls => return .done (.const n (ls.map (·.normalize)))
| .sort l => return .done (.sort l.normalize)
| _ => return .continue)
It was not feasible to make them completely equal, because the kernel's
type inference code seem to optimize level expressions a bit less
aggressively, and beta-reduces less in inference.
The private helper functions about `PProd` can later move into their own
file, used by these constructions as well as the structural recursion
module.
Fixes some issues with the executable build and bad imports.
**Release notes:**
* A bad import in an executable no longer prevents the executable's root
module from being built., This also fixes a problem where the location
of a transitive bad import would not been shown.
* The root module of the executable now respects `nativeFacets`.
**Technical touchups:**
* Expanded and better documented `tests/badImport`.
* Use `ensureJob` in `recBuildDeps` to catch import errors instead of
individual `try ... catch` blocks.
Issue #4535 is being affected by a bug in the structural inductive
predicate termination checker (`IndPred.lean`). This module did not
exist in Lean 3, and it is buggy in Lean 4. In the given example, it
introduces an auxiliary declaration containing a `sorry`, and the fails.
This PR ensures this kind of declaration is not added to the
environment.
Closes#4535
TODO: we need a new maintainer for the `IndPred.lean`.
The `pp.maxSteps` option is a hard limit on the complexity of pretty
printer output, which is necessary to prevent the LSP from crashing when
there are accidental large terms. We're using the default value from the
corresponding Lean 3 option.
This PR also sets `pp.deepTerms` to `false` by default.
When the type of an `example` is a proposition,
we should elaborate on them as we elaborate on theorems.
This is particularly important for examples that are often
used in educational material.
Recall that when elaborating theorem headers, we convert unassigned
universe metavariables into universe parameters. The motivation is
that the proof of a theorem should not influence its statement.
However, before this commit, this was not the case for examples when
their type was a proposition.
This discrepancy often confused users.
Additionally, we considered extending the above behavior to definitions
when
1- When their type is a proposition. However, it still caused disruption
in Mathlib.
2- When their type is provided. That is, we would keep the current
behavior only if `: <type>` was omitted. This would make the elaborator
for `def` much closer to the one for `theorem`, but it proved to be too
restrictive.
For example, the following instance in `Core.lean` would fail:
```
instance {α : Sort u} [Setoid α] : HasEquiv α :=
⟨Setoid.r⟩
```
and we would have to write instead:
```
instance {α : Sort u} [Setoid α] : HasEquiv.{u, 0} α :=
⟨Setoid.r⟩
```
There are other failures like this in the core, and we assume many more
in Mathlib.
closes#4398closes#4482 Remark: PR #4482 implements option 1 above. We may consider
it again in the future.
We add a new definition `BitVec.twoPow w i` to represent `(1#w <<< i)`.
This expression is used to test bits when building the multiplication
bitblaster.
Patch 1/?, being peeled from https://github.com/opencompl/lean4/pull/6.
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
Fixes typo "reflexivitiy" to "reflexivity", and changes exact Eq.rfl to
exact rfl, since Eq.rfl does not exist.
(I got something confused wrt the bot message on #4367 and accidentally
closed that one, so making this one instead, which I think satisfies the
requirements it wanted.)
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
this is the simplest of the constructions to be ported from C++ to Lean,
so I’ll PR this one first.
This begins to put each construction into its own file, as it was the
case with C++.
For validation I developed this in a separate repository at
https://github.com/nomeata/lean-constructions/tree/fad715e
and checked that all `.recOn` declarations found in Lean and Mathlib are
identical (per `==`) to the ones produced by the C code.
Fixes a bug where Lake incorrectly included the module dynlib in a
platform-independent trace. It was incorrectly excluded only external
native libraries from the trace. Also adds a test.
as #4527 describes there is inconsistency between `by`, `case` and
`next` on the one hand who, if the goal isn’t closed, put squiggly
underlines on the first line, and `.`, which so far only squiggled the
dot (which is a very short symbol!)
With this change the same mechanism as used by `case`, namely
`withCaseRef`, is also used for `.`.
There is an argument for the status quo: The `.` tactic is more commonly
used
with further tactics on the same line, and thus there is now a higher
risk that
the user might think that the first tactic is broken. But
* the same argument does apply to `by` and `case` where there was an
intentional
choice to do it this way
* consistency and
* a squiggly line just under the short `.` is easy to miss, so it is
actually
better to underlining more here (at least until we have a better way to
indicate incomplete proofs, which I have hopes for)
Fixes#4527, at least most of it.
This is the groundwork for a tactic index in generated documentation, as
there was in Lean 3. There are a few challenges to getting this to work
well in Lean 4:
* There's no natural notion of *tactic identity* - a tactic may be
specified by multiple syntax rules (e.g. the pattern-matching version of
`intro` is specified apart from the default version, but both are the
same from a user perspective)
* There's no natural notion of *tactic name* - here, we take the
pragmatic choice of using the first keyword atom in the tactic's syntax
specification, but this may need to be overridable someday.
* Tactics are extensible, but we don't want to allow arbitrary imports
to clobber existing tactic docstrings, which could become unpredictable
in practice.
For tactic identity, this PR introduces the notion of a *tactic
alternative*, which is a `syntax` specification that is really "the same
as" an existing tactic, but needs to be separate for technical reasons.
This provides a notion of tactic identity, which we can use as the basis
of a tactic index in generated documentation. Alternative forms of
tactics are specified using a new `@[tactic_alt IDENT]` attribute,
applied to the new tactic syntax. It is an error to declare a tactic
syntax rule to be an alternative of another one that is itself an
alternative. Documentation hovers now take alternatives into account,
and display the docs for the canonical name.
*Tactic tags*, created with the `register_tactic_tag` command, specify
tags that may be applied to tactics. This is intended to be used by
doc-gen and Verso. Tags may be applied using the `@[tactic_tag TAG1 TAG2
...]` attribute on a canonical tactic parser, which may be used in any
module to facilitate downstream projects introducing tags that apply to
pre-existing tactics. Tags may not be removed, but it's fine to
redundantly add them. The collection of tags, and the tactics to which
they're applied, can be seen using the `#print tactic tags` command.
*Extension documentation* provides a structured way to document
extensions to tactics. The resulting documentation is gathered into a
bulleted list at the bottom of the tactic's docstring. Extensions are
added using the `tactic_extension TAC` command. This can be used when
adding new interpretations of a tactic via `macro_rules`, when extending
some table or search index used by the tactic, or in any other way. It
is a command to facilitate its flexible use with various extension
mechanisms.
Describes the intended modes of use, potential performance tradeoffs,
and data representation in more detail.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Co-authored-by: Mac Malone <tydeu@hatpress.net>
The recent change of the trace format exposed some unexpected issues
with Lake's tracing handling. This aims to fix that.
Lake will now perform a rebuild if the trace file is invalid/unreadable.
However, it will still fall back to modification times if the trace file
is missing. Also, Lake is now backwards compatible with the previous
pure numeric traces (and tolerates the absence of a `log` field in the
JSON trace).
This PR introduces complete simprocs for all the Int versions of
div/mod, and makes some small refactoring of Int lemmas and
library_search.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
#3850 included a commit that added an extra test for `exact?`, but was
otherwise unrelated the to PR. It also removed a test. I've
cherry-picked that test over, and restored the deleted test, and next
will remove the commit from #3850.
The linters in Batteries can be used to spot mistakes in Lean. See the
message on
[Zulip](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/Go-to-def.20on.20typeclass.20fields.20and.20type-dependent.20notation/near/442613564).
These are the different linters with errors:
- unusedArguments:
There are many unused instance arguments, especially a redundant `[Monad
m]` is very common
- checkUnivs:
There was a problem with universes in a definition in
`Init.Control.StateCps`. I fixed it by adding a `variable` statement for
the implicit arguments in the file.
- defLemma:
many proofs are written as `def` instead of `theorem`, most notably
`rfl`. Because `rfl` is used as a match pattern, it must be a def. Is
this desirable?
The keyword `abbrev` is sometimes used for an alias of a theorem, which
also results in a def. I would want to replace it with the `alias`
keyword to fix this, but it isn't available.
- dupNamespace:
I fixed some of these, but left `Tactic.Tactic` and `Parser.Parser` as
they are as these seem intended.
- unusedHaveSuffices:
I cleaned up a few proofs with unused `have` or `suffices`
- explicitVarsOfIff:
I didn't fix any of these, because that would be a breaking change.
- simpNF:
I didn't fix any of these, because I think that requires knowing the
intended simplification order.
Continuation of #3958. To ensure that lean code is able to uphold the
invariant that `String`s are valid UTF-8 (which is assumed by the lean
model), we have to make sure that no lean objects are created with
invalid UTF-8. #3958 covers the case of lean code creating strings via
`fromUTF8Unchecked`, but there are still many cases where C++ code
constructs strings from a `const char *` or `std::string` with unclear
UTF-8 status.
To address this and minimize accidental missed validation, the
`(lean_)mk_string` function is modified to validate UTF-8. The original
function is renamed to `mk_string_unchecked`, with several other
variants depending on whether we know the string is UTF-8 or ASCII and
whether we have the length and/or utf8 char count on hand. I reviewed
every function which leads to `mk_string` or its variants in the C code,
and used the appropriate validation function, defaulting to `mk_string`
if the provenance is unclear.
This PR adds no new error handling paths, meaning that incorrect UTF-8
will still produce incorrect results in e.g. IO functions, they are just
not causing unsound behavior anymore. A subsequent PR will handle adding
better error reporting for bad UTF-8.
this is a first step towards porting the code `constructions.cpp` to
Lean: It leaves the construction of the `Declaration` untouched, but
moves adding the declarations to the environment, and setting various
attributes, to the Lean world.
This allows the remaining logic (the construction of the `Declaration`)
to be implemented in Lean separately and easily compared to the C++
implementation, before we replace that too.
To that end, `Declaraion` gains an `BEq` instance.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
Co-authored-by: Arthur Adjedj <arthur.adjedj@ens-paris-saclay.fr>
Fixes a bug in #4371 where the version of a package used by a dependency
would take precedence over that of a the same package as a direct
dependency if that package had a a manifest. This was because the direct
dependency's manifest entries were added before all the direct
dependencies were visited.
A set of general tweaks of the `require` syntax and docs that provide a
base for #4495.
The sole significant behavioral change is that the `name` field of a
`require` in TOML now falls back to being interpreted as a simple string
name if the value is not a valid Lean identifier. This means that a
require for a package like `doc-gen4` can be written without French
quotes.
I removed a redundant `if tFn.isMVar || sFn.isMVar then ... else return
LBool.undef` in the `else` clause of
```
if !tFn.isMVar && !sFn.isMVar then
return LBool.undef
else
```
I made a modification to the `mkLambdaFVars` function, adding a
`etaReduce : Bool` parameter that determines whether a new lambda of the
form `fun x => f x` should be replaced by `f`. I then set this option to
true at `isDefEq` when processing metavariable assignments.
This means that many unnecessary eta unreduced expression are now
reduced. This is beneficial for users, so that they do not have to deal
with such unreduced expressions. It is also beneficial for performance,
leading to a 0.6% improvement in build instructions. Most notably,
`Mathlib.Algebra.DirectLimit`, previously a top 50 slowest file, has
sped up by 40%.
Quite a number of proof in mathlib broke. Many of these involve removing
a now unnecessary `simp only`. In other cases, a simp or rewrite doesn't
work anymore, such as a `simp_rw [mul_comm]` that was used to rewrite
`fun x => 2*x`, but now this term has turned into `HMul.hMul 2`.
Closes#4386
This is not the most exciting place to start, but I started here to:
* pick a function with little development in Batteries and Mathlib, so I
wouldn't have conflicts
* that is easy!
* to see how much effort it is to get fairly complete coverage
* and to set up some infrastructure to be used later, i.e.
`tests/lean/run/list_simp.lean`
This assigns priorities to the equational lemmas so that more specific
ones
are tried first before a possible catch-all with possible
side-conditions.
We assign very low priorities to match the simplifiers behavior when
unfolding
a definition, which happens in `simpLoop`’ `visitPreContinue` after
applying
rewrite rules.
Definitions with more than 100 equational theorems will use priority 1
for all
but the last (a heuristic, not perfect).
fixes#4173, to some extent.
`Nat.succ_eq_add_one` and `Nat.pred_eq_sub_one` are now simp lemmas. For
theorems about `Nat.succ` or `Nat.pred` without corresponding theorem
for `+ 1` or `- 1`, this adds the corresponding theorem.
This PR neither adds nor removes material, but improves the organization
of `Init/Data/List/*`.
These files are essentially completely re-ordered, to ensure that
material is developed in a consistent order between `List.Basic`,
`List.Impl`, `List.BasicAux`, and `List.Lemmas`.
Everything is organised in subsections, and I've added some module docs.
presumably this avoids unnecessary work when `omega` is used in tactic
combinators where the error message is never seen. Measurement did not
show
any significant changes, though.
With an artificial sleep in
```diff
diff --git a/src/Lean/Elab/Tactic/Omega/Frontend.lean b/src/Lean/Elab/Tactic/Omega/Frontend.lean
index fd297eef60..31ea3f6bd0 100644
--- a/src/Lean/Elab/Tactic/Omega/Frontend.lean
+++ b/src/Lean/Elab/Tactic/Omega/Frontend.lean
@@ -538,6 +538,7 @@ def formatErrorMessage (p : Problem) : OmegaM MessageData := do
else
let as ← atoms
return .ofLazyM (es := as) do
+ IO.sleep 10000
let mask ← mentioned as p.constraints
let names ← varNames mask
return m!"a possible counterexample may satisfy the constraints\n" ++
```
I can observe that `omega` is slow and `try omega` fast, so it seems to
work at least.
When an implicit argument cannot be inferred, the error should show the
name of the argument.
Showing the argument name in the error message for an uninstantiated
metavariable was introduced in da33f498f5,
but this implementation causes some argument names to get lost.
The modules `CollectMVars` and `FindMVars` only search for expression
metavariables and not level metavariables, so we should use
`Expr.hasExprMVar` instead of `Expr.hasMVar`.
This came up when watching new Lean users in a class situation. A number
of them were confused when they omitted a namespace on a constructor
name, and Lean treated the variable as a pattern that matches anything.
For example, this program is accepted but may not do what the user
thinks:
```
inductive Tree (α : Type) where
| leaf
| branch (left : Tree α) (val : α) (right : Tree α)
def depth : Tree α → Nat
| leaf => 0
```
Adding a `branch` case to `depth` results in a confusing message.
With this linter, Lean marks `leaf` with:
```
Local variable 'leaf' resembles constructor 'Tree.leaf' - write '.leaf' (with a dot) or 'Tree.leaf' to use the constructor.
note: this linter can be disabled with `set_option linter.constructorNameAsVariable false`
```
Additionally, the error message that occurs when invalid names are
applied in patterns now suggests similar names. This means that:
```
def length (list : List α) : Nat :=
match list with
| nil => 0
| cons x xs => length xs + 1
```
now results in the following warning on `nil`:
```
warning: Local variable 'nil' resembles constructor 'List.nil' - write '.nil' (with a dot) or 'List.nil' to use the constructor.
note: this linter can be disabled with `set_option linter.constructorNameAsVariable false`
```
and error on `cons`:
```
invalid pattern, constructor or constant marked with '[match_pattern]' expected
Suggestion: 'List.cons' is similar
```
The list of suggested constructors is generated before the type of the
pattern is known, so it's less accurate, but it truncates the list to
ten elements to avoid being overwhelming. This mostly comes up with
`mk`.
This restores the behavior prior to
9f6bbfa106
for `MessageData.ofSyntax` `MessageData.ofExpr`, and
`MessageData.ofLevel` while staying within the new `.ofLazy` paradigm.
Also adds some documentation to help developers understand the missing
context issue.
Closes#4432
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
We recently discovered inconsistencies in Mathlib and Std over the
ordering of the arguments for `==`.
The most common usage puts the "more variable" term on the LHS, and the
"more constant" term on the RHS, however there are plenty of exceptions,
and they cause unnecessary pain when switching (particularly, sometimes
requiring otherwise unneeded `LawfulBEq` hypotheses).
This convention is consistent with the (obvious) preference for `x == 0`
over `0 == x` when one term is a literal.
We recently updated Std to use this convention
https://github.com/leanprover/std4/pull/430
This PR changes the two major places in Lean that use the opposite
convention, and adds a suggestion to the docstring for `BEq` about the
preferred convention.
This incorporates many general Lake DSL changes from #2439 and adds some
new related changes.
* Rework configuration names (e.g., `package <name>`)
* String literals ca now be used instead of identifiers for names.
* The name syntax is now optional and can instead be set via the `name`
field.
* Avoid French quotes in `lake new` / `lake init` templates (except in
`lean_lib` names). This is not done for `lean_lib` because it needs a
proper identifier for its root. It could use a string and reparse it as
an identifier, but this seems liable to produce confusion.
* The `exe` templates now names it main module `Main` like the `std`
template.
* Improve `math` template error if `lean-toolchain` fails to download.
* Lake now logs a warning rather than an error on unknown configuration
fields. This increases the Lake DSL's cross-version compatibility.
Closes#3385.
Moves the cached log into the trace file (no more `.log.json`). This
means logs are no longer cached on fatal errors and this ensures that an
out-of-date log is not associated with an up-to-date trace. Separately,
`.hash` file generation was changed to be more reliable as well. `.hash`
files are deleted as part of the build and always regenerate with
`--rehash`.
Closes#2751.
Use a TOML file for the Lake configuration of the `src/lake` directory
instead of a Lean file. This avoids having to load a version of the Lake
library to build Lake.
This is from a ~~pair~~triple programming session with @tydeu and
@mhuisi.
If stage 1 is built with `-DUSE_LAKE=ON`, the CMake run will generate
`lakefile.toml` files for the root, `src`, and `tests`. These Lake
configuration files can then be used to build core oleans. While they do
not yet allow Lake to be used to build the Lean binaries. they do allow
Lake to be used for working interactively with the Lean source. In our
preliminary experiments, this allowed updates to `Init.Data.Nat` to be
noticed automatically when reloading downstream files, rather than
requiring a full manual compiler rebuild. This will make it easier to
work on the system.
As part of this change, Lake is added to stage 0. This allows Lake to
function in `src`, which uses the stage 0 toolchain.
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
This PR addresses some non-critical but annoying issues that sometimes
cause the language server to report an error:
- When using global search and replace in VS Code, the language client
sends `textDocument/didChange` notifications for documents that it never
told the server to open first. Instead of emitting an error and crashing
the language server when this occurs, we now instead ignore the
notification. Fixes#4435.
- When terminating the language server, VS Code sometimes still sends
request to the language server even after emitting a `shutdown` request.
The LSP spec explicitly forbids this, but instead of emitting an error
when this occurs, we now error requests and ignore all other messages
until receiving the final `exit` notification. Reported on Zulip several
times over the years but never materialized as an issue, e.g.
https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/Got.20.60shutdown.60.20request.2C.20expected.20an.20.60exit.60.20notification/near/441914289.
- Some language clients attempt to reply to the file watcher
registration request before completing the LSP initialization dance. To
fix this, we now only send this request after the initialization dance
has completed. Fixes#3904.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
A pending tactic mvar managed to escape into an unexpected context in
specific circumstances.
```lean
example : True := by
· rw [show 0 = 0 by rfl]
```
* Term elaboration of the `show` creates a pending mvar for the `by rfl`
proof
* `rw` fails with an exception because the pattern does not occur in the
target
* `cdot` catches the exception and admits the goal
* `Term.runTactic` [synthesizes all pending mvars from the tactic's
execution](5f9dedfe5e/src/Lean/Elab/SyntheticMVars.lean (L350)),
including the `by rfl` proof. But this would not have happened without
`cdot` as the exception would have skipped that invocation!
* Now incrementality is confused because the nested `by rfl` proof is
unexpectedly run in the same context as the top-level proof, writing to
the wrong promise, and the error message is lost
Solution: disable incrementality for these pending mvars
The performance issue at #4413 is due to our `Fin.sub` definition.
```
def sub : Fin n → Fin n → Fin n
| ⟨a, h⟩, ⟨b, _⟩ => ⟨(a + (n - b)) % n, mlt h⟩
```
Thus, the following runs out of stack space
```
example (a : UInt64) : a - 1 = a :=
rfl
```
at the `isDefEq` test
```
(a.val.val + 18446744073709551615) % 18446744073709551616 =?= a.val.val
```
From the user's perspective, this timeout is unexpected since they are
using small numerals, and none of the other `Fin` basic operations (such
as `Fin.add` and `Fin.mul`) suffer from this problem.
This PR implements an inelegant solution for the performance issue. It
redefines `Fin.sub` as
```
def sub : Fin n → Fin n → Fin n
| ⟨a, h⟩, ⟨b, _⟩ => ⟨((n - b) + a) % n, mlt h⟩
```
This approach is unattractive because it relies on the fact that
`Nat.add` is defined using recursion on the second argument.
The impact on this repo was small, but we want to evaluate the impact on
Mathlib.
closes#4413
It seems:
* there was no actual need for the UInt32 valued version
* downstream we were getting duplicative lemmas about both
* so lets reduce the API surface area!
If anyone would prefer the remaining function is still called
`Char.utf8Size` I will happily change it. (`size` is hopefully still
unambiguous, and it's helpful to rename here so we can give a
deprecation warning that explains the type signature change.)
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
Before this commit, the `theorem` and `def` declarations had different
universe parameter orders.
For example, the following `theorem`:
```
theorem f (a : α) (f : α → β) : f a = f a := by
rfl
```
was elaborated as
```
theorem f.{u_2, u_1} : ∀ {α : Sort u_1} {β : Sort u_2} (a : α) (f : α → β), f a = f a :=
fun {α} {β} a f => Eq.refl (f a)
```
However, if we declare `f` as a `def`, the expected order is produced.
```
def f.{u_1, u_2} : ∀ {α : Sort u_1} {β : Sort u_2} (a : α) (f : α → β), f a = f a :=
fun {α} {β} a f => Eq.refl (f a)
```
This commit fixes this discrepancy.
@semorrison @jcommelin: This might be a disruptive change to Mathlib,
but it is better to fix the issue asap. I am surprised nobody has
complained about this issue before. I discovered it while trying to
reduce discrepancies between `theorem` and `def` elaboration.
Closes#4375
The following example raises `error: (kernel) declaration has free
variables '_example'`:
```lean
example: Nat → Nat :=
let a : Nat := Nat.zero
fun (_ : Nat) =>
let b : Nat := Nat.zero
(fun (_ : a = b) => 0) (Eq.refl a)
```
During elaboration of `0`, `elabNumLit` creates a synthetic mvar
`?_uniq.16` which gets abstracted by `elabFun` to `?_uniq.16 :=
?_uniq.50 _uniq.6 _uniq.12`. The `isDefEq` to `instOfNatNat 0` results
in:
```
?_uniq.50 :=
fun (x._@.4375._hyg.13 : Nat) =>
let b : Nat := Nat.zero
fun (x._@.4375._hyg.23 : Eq.{1} Nat _uniq.4 b) =>
instOfNatNat 0
```
This has a free variable `_uniq.4` which was `a`.
When the application of `?_uniq.50` to `#[#2, #0]` is instantiated, the
`let b : Nat := Nat.zero` blocks the beta-reduction and `_uniq.4`
remains in the expression.
fix: add `(useZeta := true)` here:
ea46bf2839/src/Lean/MetavarContext.lean (L567)
As [reported on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/113488-general/topic/maybe.20a.20cache.20bug.3F).
We expected that for sound reuse of elaboration results, it is
sufficient to compare the old and new syntax tree's structure and atoms
including position info, but not the whitespace in between them.
However, we have at least one request handler, the goal view, that
inspects the whitespace after a tactic and thus could return incorrect
results on reuse. For now we implement the straightforward fix of
checking the whitespace as well. Alternatives like updating the
whitespace stored in the reused info tree are tbd.
This has the slight disadvantage that adding whitespace at the end of a
tactic will re-execute it (or the entire body, but not the header, if
the body is not a tactic block), but only up to typing the first
character of the next tactic or command.
Deprecates `inputFile` and replaces it with `inputBinFile` and
`inputTextFile`. `inputTextFile` normalizes line endings, which helps
ensure text file traces are platform-independent.
The type uses `PUnit`, but the `pure ()` in the body was forcing the
implicit universe level at `PUnit` to be `1`.
We should probably elaborate `def`s like we elaborate theorems when the
resulting type is provided. This kind of mistake is hard to spot.
This `@[inline]` causes Lean to respecialize `RBMap.find?` to `NameMap`
at each call site of `NameMap.find?`, creating lots of unnecessary
duplicate IR.
so that the pretty-printed origin is clickable, and avoid the
unnecessary `@`.
Particularly nice is this fix:
```diff
/--
-info: [Meta.Tactic.simp.discharge] @bar discharge ✅
+info: [Meta.Tactic.simp.discharge] bar discharge ✅
autoParam T _auto✝
- [Meta.Tactic.simp.rewrite] { }:1000, T ==> True
-[Meta.Tactic.simp.rewrite] @bar:1000, U ==> True
+ [Meta.Tactic.simp.rewrite] T.mk:1000, T ==> True
+[Meta.Tactic.simp.rewrite] bar:1000, U ==> True
-/
```
this is an amendment to #4177, after @kmill pointed out an issue:
Users might expect that within a tactic combinator like `first`, `simp
[h]` fails if `h` does not exist. Therefore the behavior introduced in
PR #4177, which is really most useful in mormal interactive use of
`skip`, is restricted to when `recover := true`.
types like
```
inductive Many (α : Type u) where
| none : Many α
| more : α → (Unit → Many α) → Many α
```
have a `.brecOn` only supports motives producing `Type u`, but not `Sort
u`, but our induction principles produce `Prop`. So the previous
implementation of functional induction would fail for functions that
structurally recurse over such types.
We recognize this case now and, rather hazardously, replace `.brecOn`
with `.binductionOn` (and thus `.below ` with `.ibelow` and `PProd` with
`And`). This assumes that these definitions are highly analogous.
This also improves the error message when realizing a reserved name
fails with an exception, by prepending
```
Failed to realize constant {id}:
```
to the error message.
Fixes#4320
The key idea is to notice that `signExtend` behavior is controlled by
the `msb`. When `msb = false`, `sext` behaves the same as `trunc`. When
`msb = true`, `sext` behaves like `trunc` but adds high 1-bits. This is
expressed using the negate-truncate-negate pattern. Lemma statements
below:
```lean
theorem signExtend_eq_neg_truncate_neg_of_msb_false {x : BitVec w} {v : Nat} (hmsb : x.msb = false) :
(x.signExtend v) = x.truncate v := by
theorem signExtend_eq_neg_truncate_neg_of_msb_true {x : BitVec w} {v : Nat} (hmsb : x.msb = true) :
(x.signExtend v) = ~~~((~~~x).truncate v) := by
```
These give the final theorem statement:
```lean
theorem getLsb_signExtend {x : BitVec w} {v i : Nat} :
(x.signExtend v).getLsb i = (decide (i < v) && if i < w then x.getLsb i else x.msb) := by
```
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
Co-authored-by: Alex Keizer <alex@keizer.dev>
Co-authored-by: Kim Morrison <scott@tqft.net>
Remark: when splitting an `if-then-else` term, the subgoals now have
tags `isTrue` and `isFalse` instead of `inl` and `inr`.
closes#4313
---------
Co-authored-by: Mario Carneiro <di.gama@gmail.com>
The current manner of lifting `LogIO` into `CliM` produces excessive
specializations (due to a nested inlined `forM`). There was also a bug
where `IO` was lifted into `CliM` via `LogIO` rather than directly
through `MainM`.
Stores the dependency trace for a build in the cached build log and then
verifies that it matches the trace of the current build before replaying
the log. Includes test.
Closes#4303.
The `save` happened in a slightly different context from the restore,
which a refinement of the `saveOrRestoreFull` signature now makes
impossible.
Fixes#4328
this fixes a usability paper cut that just annoyed me. When editing a
larger simp proof, I usually want to see the goal state after the simp,
and this is what I see while the `simp` command is complete. But then,
when I start typing, and necessarily type incomplete lemma names, that
error makes `simp` do nothing again and I see the original goal state.
In fact, if a prefix of the simp theorem name I am typing is a valid
identifier, it jumps even more around.
With this PR, using `logException`, I still get the red squiggly lines
for the unknown identifer, but `simp` just ignores that argument and
still shows me the final goal. Much nicer.
I also demoted the message for `[-foo]` when `foo` isn’t `simp` to a
warning and gave it the correct `ref`.
See it in action here: (in the middle, when you suddenly see the
terminal,
I am switching lean versions.)
https://github.com/leanprover/lean4/assets/148037/8cb3c563-1354-4c2d-bcee-26dfa1005ae0
In the course of the development, I grabbed facts about right shifting
over integers [from
`mathlib4`](https://github.com/leanprover-community/mathlib4/blob/master/Mathlib/Data/Int/Bitwise.lean).
The core proof strategy is to perform a case analysis of the msb:
- If `msb = false`, then `sshiftRight = ushiftRight`.
- If `msb = true`. then `x >>>s i = ~~~(~~~(x >>>u i))`. The double
negation introduces the high `1` bits that one expects of the arithmetic
shift.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
These will be used by LeanSAT for bitblasting rotations by constant
distances.
We first reduce the case when the rotation amount is larger than the
width to the case where the rotation amount is less than the width
(`x.rotateLeft/Right r = x.rotateLeft/Right (r%w)`).
Then, we case analyze on the low bits versus the high bits of the
rotation, where we prove equality by extensionality.
---------
Co-authored-by: Alex Keizer <alex@keizer.dev>
Co-authored-by: Tobias Grosser <github@grosser.es>
These lemmas are morally equivalent to Mathlib lemmas which are proposed
to be deleted from Mathlib in
[#13286](https://github.com/leanprover-community/mathlib4/pull/13286).
It is only morally equivalent, because the Mathlib lemmas are stated in
terms of Mathlib-defined things: `toFin_natCast` uses a coercion from
`Nat` to `Fin (2^w)` which relies on `NeZero` machinery available only
in Mathlib. Thus, I've rephrased the rhs in terms of the def-eq
`Fin.ofNat'` with an explicit proof that `2^w` is non-zero.
Similarly, the RHS of `toFin_neg` was phrased in terms of negation on
`Fin`s, which is only defined in Mathlib, so I've unfolded the
definition.
Allows embedding user widgets in structured messages. Companion PR is
leanprover/vscode-lean4#449.
Some technical choices:
- The `MessageData.ofWidget` constructor might not be strictly necessary
as we already have `MessageData.ofFormatWithInfos`, and there is
`Info.ofUserWidget`. However, `.ofUserWidget` also requires a `Syntax`
object (as it is normally produced when widgets are saved at a piece of
syntax during elaboration) which we do not have in this case. More
generally, it continues to be a bit cursed that `Elab.Info` nodes are
used both for elaboration and delaboration (pretty-printing), so
entrenching that approach seems wrong. The better approach would be to
have a separate notion of pretty-printer annotation; but such a refactor
would not be clearly beneficial right now.
- To support non-JS-based environments such as
https://github.com/Julian/lean.nvim, `.ofWidget` requires also providing
another message which approximates the widget in a textual form.
However, in practice these environments might still want to support a
few specific user widgets such as "Try this".
---
Closes#2064.
In `v4.8.0-rc2`, due to additional build refactor changes, `JobM` no
longer cleanly lifts in `FetchM`. Generally, a `JobM` action should not
be run `FetchM` directly but spawned asynchronously as job (e.g., via
`Job.async`). However, there may be some edge cases were this is
necessary and it is a backwards compatibility break, so this change adds
back the lift. This change also includes an `example` definition to
ensure the lift works in order to prevent similar accidental breakages
in the future.
This breakage was first reported by Mario on
[Zulip](https://leanprover.zulipchat.com/#narrow/stream/113488-general/topic/v4.2E8.2E0-rc2.20discussion/near/440407037).
Switches the manifest format to use `major.minor.patch` semantic
versions. Major version increments indicate breaking changes (e.g., new
required fields and semantic changes to existing fields). Minor version
increments (after `0.x`) indicate backwards-compatible extensions (e.g.,
adding optional fields, removing fields). This change is backwards
compatible. Lake will still successfully read old manifest with numeric
versions. It will treat the numeric version `N` as semantic version
`0.N.0`. Lake will also accept manifest versions with `-` suffixes
(e.g., `x.y.z-foo`) and then ignore the suffix.
This change also includes the general cleanup/refactoring of the
manifest code and data structures that was part of #3174.
Adds two new Lake commands, `lake pack` and `lake unpack`, which pack
and unpack, respectively, Lake build artifacts from an archive. If a
path argument is given, creates the archive specified, otherwise uses
the information in a package's `buildArchive` configuration as the
default.
The pack command will be used by Reservoir to prepare crate-style build
archives for packages. In the future, the command will also be
extensible through configuration file hooks.
Extends the functionality of `lake test` and adds a parallel command in
`lake lint`.
* Rename `@[test_runner]` / `testRunner` to `@[test_driver]` /
`testDriver`. The old names are kept as deprecated aliases.
* Extend help page for `lake test` and adds one for `lake check-test`.
* Add `lake lint` and its parallel tag `@[lint_driver]` , setting
`lintDriver`, and checker `lake check-lint`.
* Add support for specifying test / lint drivers from dependencies.
* Add `testDriverArgs` / `lintDriverArgs` for fixing additional
arguments to the invocation of a driver script or executable.
* Add support for library test drivers (but not library lint drivers).
* `lake check-test` / `lake check-lint` only load the package (without
dependencies), not the whole workspace.
Closes#4116. Closes#4121. Closes#4142.
The type class `MonadStore1` and friends have an outParam, which should
not be an outParam, because there are multiple possible values for this
parameter. At this function
[fetchOrCreate](1382e9fbc4/src/lake/Lake/Load/Main.lean (L196C49-L196C63)),
there are multiple stacked `StateT` monad transformers that each give a
different instance to `MonadStore1`. It is an implementation detail of
type class synthesis which instance is found. This particular type class
synthesis fails when the unused instance
`Lake.instMonadStore1OfMonadDStoreOfFamilyOut` is set to a lower
priority, because then the synthesis order happens to go differently, so
the wrong instance is found.
Replacing the outParam with a semiOutParam solves this issue. Thus, we
make a new type class `MonadStore1Of`, which is the same, but with a
semiOutParam. This follows the design of `MonadState` and
`MonadStateOf`.
However, then it turns out that the instance cannot anymore be
synthesised.
There are two instances for `MonadStore1`:
```
instance [MonadDStore κ β m] : MonadStore1 k (β k) m
instance [MonadDStore κ β m] [FamilyOut β k α] : MonadStore1 k α m
```
The first one is problematic during unification, especially when `β`
should be instantiated as a constant function. We make the second one
sufficient by adding an instance for the general type family:
```
/-- The general type family -/
instance (priority := low) : FamilyDef Fam a (Fam a) where
family_key_eq_type := rfl
```
So then we can get rid of the first instance.
Without this, it would not easy but perhaps be feasible to break
incrementality when editing command prefixes such as `set_option ... in
theorem` or also `theorem namesp.name ...` (which is a macro),
especially if at some later point we support incrementality in input
shifted by an edit. Explicit, sound support for these common cases will
be brought back soon.
Changing document string in `Attributes.lean`, in order to consistent
with code in `Lean.Parser.Attr`.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
We are switching to a new system for preparing release notes.
* Release notes will be compiled when creating a release candidate from
all the commits that are part of that release.
* PRs can include suggestions for release notes in PR messages. Please
use language such as "release notes" and "breaking changes" to call
attention to the suggestions. Release notes are user-centric rather than
developer-centric.
* For more complicated release notes, these can be put into the
`releases_drafts` folder.
This solves an issue where PRs that include release notes can, when
merged, have those notes appear under the wrong Lean version, since they
might have been created before a release but not merged until after. It
also solves merge conflicts due to multiple PRs updating the release
notes.
Extends Lean's incremental reporting and reuse between commands into
various steps inside declarations:
* headers and bodies of each (mutual) definition/theorem
* `theorem ... := by` for each contained tactic step, including
recursively inside supported combinators currently consisting of
* `·` (cdot), `case`, `next`
* `induction`, `cases`
* macros such as `next` unfolding to the above
*Incremental reuse* means not recomputing any such steps if they are not
affected by a document change. *Incremental reporting* includes the
parts seen in the recording above: the progress bar and messages. Other
language server features such as hover etc. are *not yet* supported
incrementally, i.e. they are shown only when the declaration has been
fully processed as before.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
This PR extracts `msb_eq_false_iff_two_mul_lt` and
`msb_eq_true_iff_two_mul_ge` from #4179, and uses them to prove a
theorem that characterizes `BitVec.toInt` in terms of `BitVec.msb`. This
lemma will be useful to prove a bit-blasting theorem for `BitVec.slt`
and `BitVec.sle`.
Also cleans up an existing proof (`toInt_eq_toNat_cond `), which turns
out to be provable by `rfl`.
---------
Co-authored-by: Kim Morrison <scott@tqft.net>
this fixes#4078. It is an alternative fix to the one in #4137,
suggested
by @kmill.
Incidentially, it makes the unused variable linter better. My theory is
that
if we don’t reset the info when backtracking, the binder shows up more
than
once in the info tree, and then it is considered “used”, although there
are
just multiple binders.
Add docstrings, usage examples, and doc tests for `String.prev`,
`.front`, `.back`, `.atEnd`.
Improve docstring examples for `String.next` based on discussion
examples for `String.prev`.
---------
Co-authored-by: Kim Morrison <kim@tqft.net>
This ensures that rotateLeft/Right behave correctly even when the
rotation amount is larger than the bitwidth.
This shall be followed up with `getLsb` theorems for rotations for
LeanSAT.
We choose to write `aux` definitions since it is cleaner to reason about
the `aux` theorems with the assumption that `rotation-amount <
bit-width`, followed by auxiliary lemmas that link the behavior of
rotation to the canonical case when `rotation-amount < bit-width`.
Proof strategy we will execute based on these definitions: [Link to
proof of
`getLsb_rotateLeft`](a0b18ec0f4/src/Init/Data/BitVec/Lemmas.lean (L1129-L1204))
---------
Co-authored-by: Tobias Grosser <github@grosser.es>
The ANSI mode build monitor now now longer displays built jobs (instead
only those that print info or failed). Also upgrades the progress ticker
with a spinner icon and information on the number of running jobs.
To eliminate parsing differences between Windows and other platforms,
the frontend now normalizes all CRLF line endings to LF, like [in
Rust](https://github.com/rust-lang/rust/issues/62865).
Effects:
- This makes Lake hashes be faithful to what Lean sees (Lake already
normalizes line endings before computing hashes).
- Docstrings now have normalized line endings. In particular, this fixes
`#guard_msgs` failing multiline tests for Windows users using CRLF.
- Now strings don't have different lengths depending on the platform.
Before this PR, the following theorem is true for LF and false for CRLF
files.
```lean
example : "
".length = 1 := rfl
```
Note: the normalization will take `\r\r\n` and turn it into `\r\n`. In
the elaborator, we reject loose `\r`'s that appear in whitespace. Rust
instead takes the approach of making the normalization routine fail.
They do this so that there's no downstream confusion about any `\r\n`
that appears.
Implementation note: the LSP maintains its own copy of a source file
that it updates when edit operations are applied. We are assuming that
edit operations never split or join CRLFs. If this assumption is not
correct, then the LSP copy of a source file can become slightly out of
sync. If this is an issue, there is some discussion
[here](https://github.com/leanprover/lean4/pull/3903#discussion_r1592930085).
Given `h` with type `x + k = y + k'` (or `h : k = k')`, `cases h`
produced a proof of size linear in `min k k'`. `isDefEq` has support for
offset, but `unifyEq?` did not have it, and a stack overflow occurred
while processing the resulting proof. This PR fixes this issue.
closes#4219
Show that shifting a natural number left and then shifting right by the
same amount is a no-op.
I originally proved this in a different PR, ended up not needing the
fact after all, but it still seemed like a generally useful simp lemma
to have.
Fixes two output bugs with cloud releases: (1) the fetch as part of an
`extraDep` was not properly isolated in a job, and (2) the release job
would be shown even if the release had already been successfully
fetched.
Also includes some related touchups, including the addition of show all
jobs on `-v` which helps with debugging job counts.
### Explanation
In the case that `assignSyntheticOpaque := true` and the given
metavariable is `syntheticOpaque` and the depth of the metavariable is
not the current depth, `isReadOnlyOrSyntheticOpaque` returns false, even
though the metavariable is read-only because of being declared at a
smaller depth. This causes the metavariable to (wrongly) be able to be
instantiated by `isDefEq`.
This bug was found at the proof of
[RingHom.PropertyIsLocal.sourceAffineLocally_of_source_openCover](https://leanprover-community.github.io/mathlib4_docs/Mathlib/AlgebraicGeometry/Morphisms/RingHomProperties.html#RingHom.PropertyIsLocal.sourceAffineLocally_of_source_openCover),
which involves a type class synthesis for `CommRing ?m.2404`, and the
synthesis manages to instantiate this metavariable into different
values, even though `synthInstance?` increases the metavariable depth.
This synthesis fails after 1 second.
I found the bug while modifying the instance synthesis code: the
modified code spent several minutes on this failed synthesis.
### Test
The problem can be verified with the test:
```
run_meta do
let m ← mkFreshExprMVar (Expr.sort levelOne) MetavarKind.syntheticOpaque
withAssignableSyntheticOpaque do
withNewMCtxDepth do
let eq ← isDefEq m (.const ``Nat [])
Lean.logInfo m! "{eq}"
```
this unification used to succeed, giving `true`, and this fix makes it
return `false`.
### Impact on Mathlib
This fix causes a change in the behaviour of `congr`, `convert` and
friends, which breaks a couple of proofs in mathlib. Most of these are
fixed by supplying more arguments.
I fixed these proofs, and
[benched](http://speed.lean-fro.org/mathlib4/compare/b821bfd9-3769-4930-b77f-0adc6f9d218f/to/e7b27246-a3e6-496a-b552-ff4b45c7236e?hash2=4f3c460cc1668820c9af8418a87a23db44c7acab)
mathlib. The result is that most files are unaffected, but some files
are significantly improved. This is most prominent in
Mathlib.RingTheory.Jacobson, where the number of instructions has
decreased by 28%. The overall improvement is a 0.3% reduction in
instructions.
[Zulip
message](https://leanprover.zulipchat.com/#narrow/stream/287929-mathlib4/topic/Ways.20to.20speed.20up.20Mathlib/near/439218960)
The expression tree elaborator computes a "maxType" that every leaf term
can be coerced to, but the elaborator was not ensuring that the entire
expression tree would have maxType as its type. This led to unexpected
errors in examples such as
```lean
example (a : Nat) (b : Int) :
a = id (a * b^2) := sorry
```
where it would say it could not synthesize an `HMul Int Int Nat`
instance (the `Nat` would propagate from the `a` on the LHS of the
equality). The issue in this case is that `HPow` uses default instances,
so while the expression tree elaborator decides that `a * b^2` should be
referring to an `Int`, the actual elaborated type is temporarily a
metavariable. Then, when the binrel elaborator is looking at both sides
of the equality, it decides that `Nat` will work and coercions don't
need to be inserted.
The fix is to unify the type of the resulting elaborated expression with
the computed maxType. One wrinkle is that `hasUncomparable` being false
is a valid test only if there are no leaf terms with unknown types (if
they become known, it could change `hasUncomparable` to true), so this
unification is only performed if the leaf terms all have known types.
Fixes issue described by Floris van Doorn on
[Zulip](https://leanprover.zulipchat.com/#narrow/stream/287929-mathlib4/topic/elaboration.20issue.20involving.20powers.20and.20sums/near/439243587).
luckily the necessary functionality already exists in the form of
`addPPExplicitToExposeDiff`. But it is not cheap, and we should not run
this code
when the error message isn’t shown, so we should do this lazily.
We already had `MessageData.ofPPFormat` to assemble the error message
lazily, but it
was restricted to returning `FormatWithInfo`, a data type that doesn’t
admit a nice
API to compose more complex messages (like `Format` or `MessageData`
has; an attempt to
fix that is in #3926).
Therefore we split the functionality of `.ofPPFormat` into
`.ofFormatWithInfo` and `.ofLazy`,
and use `.ofLazy` to compute the more complex error message of `apply`.
Fixes#3232.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
Co-authored-by: Wojciech Nawrocki <wjnawrocki@protonmail.com>
Messaged @tydeu about adding a README.md to new lake projects. I decided
to add it with the help of GPT.
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
The new Lake build monitor is now more selective, accurate, and prettier
in what it prints.
**Key Changes:**
* Poll jobs at a fixed frequency (100ms), updating the caption and
finished job count.
* Add `action` field to jobs to record information about what jobs do.
It enables distinguishing between jobs which build something, fetch from
a store, or reload logs from the cache.
* At standard verbosity, print build captions only when a job is know to
build or fetch something (i.e., `action >= .fetch`).
* Add an icon and color to job captions based on their log-level / build
status. Also add color to levels in logs.
* Add `--ansi`/`--no-ansi` to toggle Lake's use of ANSI escape codes.
* Fix some `v4.8.0-rc1` bugs and `--old`.
Closes#2822.
The `simp` tactic uses a discrimination tree to select candidate
theorems that will be used to rewrite an expression. This indexing data
structure minimizes the number of theorems that need to be tried and
improves performance. However, indexing modulo reducibility is
challenging, and a theorem that could be applied, when taking reduction
into account, may be missed. For example, suppose we have a `simp`
theorem `foo : forall x y, f x (x, y).2 = y`, and we are trying to
simplify the expression `f a b <= b`. `foo` will not be tried by `simp`
because the second argument of `f a b` is not a projection of a pair.
However, `f a b` is definitionally equal to `f a (a, b).2` since we can
reduce `(a, b).2`.
In Lean 3, we had a much simpler indexing data structure where only the
head symbol was taken into account. For the theorem `foo`, the head
symbol is `f`. Thus, the theorem would be considered by `simp`.
This commit adds the option `Simp.Config.index`. When `simp (config := {
index := false })`, only the head symbol is considered when retrieving
theorems, as in Lean 3. Moreover, if `set_option diagnostics true`,
`simp` will check whether every applied theorem would also have been
applied if `index := true`, and report them. This feature can help users
diagnose tricky issues in code that has been ported from libraries
developed using Lean 3 and then ported to Lean 4. In the following
example, it will report that `foo` is a problematic theorem.
```lean
opaque f : Nat → Nat → Nat
@[simp] theorem foo : f x (x, y).2 = y := by sorry
example : f a b ≤ b := by
set_option diagnostics true in
simp (config := { index := false })
```
In the example above, the following diagnostic message is produced.
```lean
[simp] theorems with bad keys
foo, key: [f, *, Prod.1, Prod.mk, Nat, Nat, *, *]
```
With the information above, users can annotate theorems such as `foo`
using `no_index` for problematic subterms.
Example:
```lean
opaque f : Nat → Nat → Nat
@[simp] theorem foo : f x (no_index (x, y).2) = y := by sorry
example : f a b ≤ b := by
simp -- `foo` is still applied
```
cc @semorrison
cc @PatrickMassot
Many of our tests in `tests/lean/run/` produce output from `#eval` (or
`#check`) statements, that is then ignored.
This PR tries to capture all the useful output using `#guard_msgs`. I've
only done a cursory check that the output is still sane --- there is a
chance that some "unchecked" tests have already accumulated regressions
and this just cements them!
In the other direction, I did identify two rotten tests:
* a minor one in `setStructInstNotation.lean`, where a comment says `Set
Nat`, but `#check` actually prints `?_`. Weird?
* `CompilerProbe.lean` is generating empty output, apparently indicating
that something is broken, but I don't know the signficance of this file.
In any case, I'll ask about these elsewhere.
(This started by noticing that a recent `grind` test file had an
untested `trace_state`, and then got carried away.)
This PR adds theorems that relate unsigned bitvector comparisons
`BitVec.ult` and `BitVec.ule` to `BitVec.carry`. These lemmas are a
prerequisite to bit-blasting these comparisons in LeanSAT.
The trace class Meta.synthInstance.answer isn't registered, so it can't
be used.
I set `inherited := true`, because I think it is a useful trace to have.
In particular it tells you when an instance has been found that has a
too large size. This is very useful information.
as usually, just a matter of using the `WithInfo` variant.
Also simplifying the code a bit, it seems we can use
`realizeGlobalConstNoOverloadWithInfo` here.
(It's somehwatdubious API design that of all the functions in
the `{resolve/realise}GlobalConst{NoOverload,}{WithInfo,}` family
the one with the longest name is the one that should be used
unless one has a reason to use another one.)
Fixes: #4174
in #4158 I was experimenting with a change to the simplifier that
affectes the order in which lemmas were tried, and of course it breaks
proofs all over the place whenever we have a non-confluent simp set.
Among the first breakages encountered, a large fraction was due to
`simp` rewriting with `List.length_pos : 0 < length l ↔ l ≠ []`.
This does not strike me a as a good simp lemma: If `l` is a manifest
constructor, the simplifier will reduce `length` and solve it anyways,
and if it isn't then an inequality usually isn’t very simp friendly. It
is also highly non-confluent with any kind of `length`-lemma we might
have.
This therefore removes it from the standard simp set.
Summary:
- Take `synthPendingDepth` into account when caching TC results
- Add `maxSynthPendingDepth` option with default := 2.
- Add support for tracking `synthPending` failures when using
`set_option diagnostics true`
closes#2522closes#3313closes#3927
Identical to #4114 but with `maxSynthPendingDepth := 1`
closes#4114
cc @semorrison
This improves job captions, the grouping of logs underneath them, and
the handling of import errors. It also adds a number of log-related
utilities to help achieve this.
**Key Changes:**
* Job captions for facets now include the name of the object (e.g.,
module, library, facet). A caption has also been added to the top-level
build of imports (e.g., for the server and `lake lean`).
* Stray I/O and errors outside the build job in a build function
captioned with `withRegisterJob` (e.g., user-defined targets) will now
be properly grouped under that caption instead of ending up under
"Computing build jobs". Stray I/O will be converted to a single
informational log entry.
* Builds no longer fail immediately on erroneous imports. Lake will now
attempt to recover as best as possible from any import errors.
Information on the import error will appear both in the build of the
erroneous import and in the files which transitive import it. For
example, uf `Lib.B` imports a missing module `Lib.A`, then the build of
`Lib.A` will mention that the file does not exist, and the build of
`Lib.B` will mention the bad import of `Lib.A`.
Closes#3351. Closes#3809.
Previously, the CI would run upon every label addition, including things
like `builds-mathlib`
or `will-merge-soon`, possibly triggering a new PR release, new mathlib
builds etc. Very wasteful!
Unfortunately (but not surprisingly) Github does not offer a nice way of
saying
“this workflow depends on that label, please re-run if changed”. Not
enough
functional programmer or nix enthusiasts there, I guess…
So here is the next iteration trying to work with what we have from
Github:
A new workflow watches for (only) `full-ci` label addition or deletion,
and then re-runs
the CI job for the current PR.
Sounds simple? But remember, this is github!
* `github.event.pull_request.labels.*.name` is *not* updated when a job
is re-run.
(This is actually a reasonable step towards determinism, but doesn't
help us
constructing this work-around.)
Ok, so let’s use the API to fetch the current state of the label.
* There is no good way to say “find the latest run of workflow `"CI"` on
PR `$n`”.
The best approximation seems to search by branch and triggering event.
This can
probably go wrong if there are multiple PRs from different repos with
the same
head ref name (`patch-1` anyone?). Let’s hope that it doesn’t happen too
often.
* You cannot just rerun a workflow. You can only rerun a finished
workflow. So cancel
it first. And `sleep` a bit…
So let’s see how well this will work. It’s plausibly an improvement.
I did not introduce `inductTheoremSuffix` etc, it seems more direct to
just spell out the suffix here. If we ever change it there are many
occurrences where they need to be changed anyways, so the definition
doesn't seem to save much work or add that much robustness.
Adds `IO.getTaskState` which returns the state of a `Task` in the Lean
runtime's task manager. The `TaskState` inductive has 3 constructors:
`waiting`, `running`, and `finished`. The `waiting` constructor
encompasses the waiting and queued states within the C task object
documentation, because the task object does not provide a low cost way
to distinguish these different forms of waiting. Furthermore, it seems
unlikely for consumers to wish to distinguish between these internal
states. The `running` constructor encompasses both the running and
promised states in C docs. While not ideal, the C implementation does
not provide a way to distinguish between a running `Task` and a waiting
`Promise.result` (they both have null closures).
The main loop logic could be simplified, and `if let` could be used to
make control flow more obvious.
Also adds a check for macro scopes to prevent `unresolveNameGlobal` from
returning names with macro scopes in the event there's an alias with
one.
This is a follow up to #3946.
This reverts commit 706a4cfd73 introduced
in #3970
As explained in #4124, `findM?` can become a footgun if used in monads
which induce side-effects such as caching. This PR removes that
function, and fixes the code introduced by #3398 for which the function
was first added.
cc @semorrison.
we keep running into examples where working with well-founded recursion
is slow because defeq checks (which are all over the place, including
failing ones that are back-tracked) unfold well-founded definitions.
The definition of a function defined by well-founded recursion should be
an implementation detail that should only be peeked inside by the
equation generator and the functional induction generator.
We now mark the mutual recursive function as irreducible (if the user
did not
set a flag explicitly), and use `withAtLeastTransparency .all` when
producing
the equations.
Proofs can be fixed by using rewriting, or – a bit blunt, but nice for
adjusting
existing proofs – using `unseal` (a.k.a. `attribute [local
semireducible]`).
Mathlib performance does not change a whole lot:
http://speed.lean-fro.org/mathlib4/compare/08b82265-75db-4a28-b12b-08751b9ad04a/to/16f46d5e-28b1-41c4-a107-a6f6594841f8
Build instructions -0.126 %, four modules with significant instructions
decrease.
To reduce impact, these definitions were changed:
* `Nat.mod`, to make `1 % n` reduce definitionally, so that `1` as a
`Fin 2` literal
works nicely. Theorems with larger `Fin` literals tend to need a `unseal
Nat.modCore`
https://github.com/leanprover/lean4/pull/4098
* `List.ofFn` rewritten to be structurally recursive and not go via
`Array.ofFn`:
https://github.com/leanprover-community/batteries/pull/784
Alternative designs explored were
* Making `WellFounded.fix` irreducible.
One benefit is that recursive functions with equal definitions (possibly
after
instantiating fixed parameters) are defeq; this is used in mathlib to
relate
[`OrdinalApprox.gfpApprox`](https://leanprover-community.github.io/mathlib4_docs/Mathlib/SetTheory/Ordinal/FixedPointApproximants.html#OrdinalApprox.gfpApprox)
with `.lfpApprox`.
But the downside is that one cannot use `unseal` in a
targeted way, being explicit in which recursive function needs to be
reducible here.
And in cases where Lean does unwanted unfolding, we’d still unfold the
recursive
definition once to expose `WellFounded.fix`, leading to large terms for
often no good
reason.
* Defining `WellFounded.fix` to unroll defintionally once before hitting
a irreducible
`WellFounded.fixF`. This was explored in #4002. It shares most of the
ups and downs
with the previous variant, with the additional neat benefit that
function calls that
do not lead to recursive cases (e.g. a `[]` base case) reduce nicely.
This means that
the majority of existing `rfl` proofs continue to work.
Issue #4051, which demonstrates how badly things can go if wf recursive
functions can be
unrolled, showed that making the recursive function irreducible there
leads to noticeably
faster elaboration than making `WellFounded.fix` irreducible; this is
good evidence that
the present PR is the way to go.
This fixes https://github.com/leanprover/lean4/issues/3988
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
This fixes#2901, a bug in the old compiler which causes a segfault. The
issue is that when visiting `noConfusion` applications, it assumes that
each constructor case has `nfields` arguments, e.g. `head1 = head2 ->
tail1 = tail2 -> P` has two arguments because `List.cons` has 2 fields,
but in fact propositional fields are skipped by the noConfusion type
generator, so for example `Subtype.noConfusionType` is:
```lean
@[reducible] protected def Subtype.noConfusionType.{u_1, u} : {α : Sort u} →
{p : α → Prop} → Sort u_1 → Subtype p → Subtype p → Sort u_1 :=
fun {α} {p} P v1 v2 ↦
Subtype.casesOn v1 fun val property ↦ Subtype.casesOn v2 fun val_1 property ↦
(val = val_1 → P) → P
```
where `val = val_1 → P` only has the one argument even though
`Subtype.mk` has two fields, presumably because it is useless to have an
equality of propositions. Unfortunately there isn't any easy cache or
getter to use here to get the number of non-propositional fields, so we
just calculate it on the spot.
Fixes#3270 by moving the deprecation check from
`Lean.Elab.Term.mkConsts` to `Lean.Elab.Term.mkConst`, so
`Lean.Elab.Term.mkBaseProjections`, `.elabAppLValsAux`, `.elabAppFn`,
and `.elabForIn` also hit the check. Not all of these really need to hit
the check, so I'll run `!bench` to see if it's a problem.
this is in preparation for #4061. Once that lands, `1 % 42 = 1` will no
longer hold definitionally (at least not without an ungly `unseal
Nat.modCore in` around). This affects mathlib in a few places,
essentially every time a `1 : Fin (n+1)` literal is written.
So this extends the existing special case for `0 % n = 0` to `1 % n`.
https://live.lean-lang.org/#project=lean-nightly now allows users to
play around with the latest lean nightly, and it seems prudent to ask
them to test bug reports, if possible, there, and not just with whatever
release they use.
Also reformatted the descriptions to look well in a text area. Users
will not see this as rendered markdown, but as plain text.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
As [reported on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/341532-lean4-dev/topic/Find.20references.20broken.20in.20lean.20core/near/437051935).
The `mainModuleName` was being set incorrectly when browsing lean core
sources, resulting in failure of cross-file server requests like "Find
References". Because the `srcSearchPath` is generated asynchronously, we
store it as a `Task Name` which is resolved some time before the header
is finished parsing. (I don't think the `.get` here will ever block,
because the srcSearchPath will be ready by the time the initial command
snap is requested.)
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Add docstrings and usage examples for `String.length`, `.push`,
`.append`, `.get?`, `.set`, `.modyify`, and `.next`. Update docstrings
and add usage examples for `String.toList`, `.get`, and `.get!`.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
This issue was affecting several Mathlib files.
@mattrobball @semorrison This is a different solution for the issue. The
comment at `Extra.lean` describes the new solution and documents the new
issues found with the previous one.
closes#4085
This makes the `leanArts` in `library_data leanArts : BuildJob Unit` get
a hover for the generated axiom. It also simplifies the `quoteFrom`
function so that it delaborates properly by using a name literal (which
elaborates to `mkStr1`, `mkStr2` etc) instead of a `mkStr` application.
otherwise it remains in the equational theorem and may cause the
“unused have linter” to trigger. By moving the proof into
`decreasing_by`, the equational theorems are unencumbered by termination
arguments.
see also
https://github.com/leanprover/std4/pull/690#issuecomment-2095378609
Closes#3386
Currently, when generating the signature of an injectivity lemma for a
certain constructor `c : forall xs, Foo a_1 ... a_n`,
`mkInjectiveTheoremTypeCore?` will differentiate between variables which
are bound to stay the same between the two equal values (i.e inductive
indices), and non-fixed ones. To do that, the function currently checks
whether a variable `x ∈ xs` appears in the final co-domain `Foo a_1 ...
a_n` of the constructor. This condition isn't enough however. As shown
in the linked issue, the codomain may also depend on variables which
appears in the type of free vars contained in `Foo a_1 ... a_n`, but not
in the term itself. This PR fixes the issue by also checking the types
of any free variable occuring in the final codomain, so as to ensure
injectivity lemmas are well-typed.
This PR upstreams lemmas about List/Array operations already defined in
Lean from std/batteries.
Happy to take suggestions about increasing or decreasing scope.
---------
Co-authored-by: Mario Carneiro <di.gama@gmail.com>
On
[Zulip](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/Notation.20in.20namespace.20not.20showing.20in.20pp/near/437016468),
Peter Nelson reported that notations that could be pretty printed with
generalized field notation did not pretty print using the intended
notation.
This PR makes it so that app unexpanders are considered before
generalized field notation. The complexity before was that we wanted to
do parent projection collapse, and since we did the collapse before
pretty printing that argument, it meant it wasn't possible to do app
unexpanders when there was a field notation candidate. The new solution
is to collapse parent projections only when actually considering field
notation, which can be done because we can safely strip off projection
syntax in an expression-directed way.
The layout algorithm, while somewhat finicky, is (unfortunately)
necessary for C code to interface with lean structures. This adds a
(AFAIK) complete description of the layout algorithm, including a worked
example large enough to make it possible to reconstruct the whole
decision diagram.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
This is still experimental, but it implements identifier support in auto
tactics "in the obvious way". It also converts `quoteAutoTactic` to
generate Expr directly instead of going via syntax (this doesn't have
any effect other than increasing compile cost AFAICT).
Adds `IO.Process.getCurrentDir` and `IO.Process.setCurrentDir` for
retrieving and setting, respectively, the current working directory of a
process. The names of the functions are inspired by Rust (e.g.,
[`set_current_dir`](https://doc.rust-lang.org/std/env/fn.set_current_dir.html)).
even when rewriting the type of `h` becuase there is no expected type.
(When there is an expected type, it already tried both orientations.)
Also feeble attempt to include this information in the docstring without
writing half a manual chapter.
when dealing with well-founded recursive definitions, `tryURefl` isn't
going to be that useful and possibly slow. So disable that code path
when doing well-founded recursion.
(This is a variant of #4025 where I tried using `with_reducible` to
limit the impact of slow unfolding, but if we can get away with
disabling it complete, then even better.)
In the following, hovering over `true` in the infoview was showing
`Nat.succ y`.
```lean
#check fun (x : Nat) =>
match h : x with
| 0 => false
| y + 1 => true
```
Now hovering over `true` shows `true`.
The issue was that SubExpr positions were not being tracked for
patterns, and the position for a pattern could coincide with the
position for a RHS, putting overwriting terminfo. Now the position given
to a pattern is correct and unique.
Refactors the `match` delaborator, makes it handle shadowing of `h :`
discriminant annotations correctly, and makes it use the standard
`withOverApp` combinator to handle overapplication.
Fixes some bugs with the log refactor (#3835). Namely, quiet mode
progress printing and missing string interpolation in the fetching cloud
release caption.
This is a major refactor of Lake's build code. The key changes:
* **Job Registration**: Significant build jobs are now registered by
build functions. The DSL inserts this registration automatically into
user-defined targets and facets, so this change should require no
end-user adaption. Registered jobs are incrementally awaited by the main
build function and the progress counter now indicates how many of these
jobs are completed and left-to-await. On the positive side, this means
the counter is now always accurate. On the negative side, this means
that jobs are displayed even if they are no-ops (i.e., if the target is
already up-to-date).
* **Log Retention**: Logs are now part of a Lake monad's state instead
of being eagerly printed. As a result, build jobs retain their logs.
Using this change, logs are are now always printed after their
associated caption (e.g., `[X/Y] Building Foo`) and are not arbitrarily
interleaved with the output of other jobs.
* **Simplify the build monad stack**: Previously, there was a lot of
confused mixing between the various build monads in the codebase (i.e.,
`JobM`, `ScedulerM`, `BuildM`, `RecBuildM`, and `IndexBuildM` ). This
refactor attempts to make there use more consistent and straightforward:
* `FetchM` (formerly `IndexBuildM`) is the top-level build monad used by
targets and facets and is now uniformly used in the codebase for all
top-level build functions.
* `JobM` is the monad of asynchronous build jobs. It is more limited
than `FetchM` due to the fact that the build cache can not be modified
asynchronously.
* `SpawnM` (formerly `SchedulerM`) is the monad used to spawn build
jobs. It lifts into `FetchM`.
* `RecBuildM` and `CoreBuildM` (formerly `BuildM`) have been relegated
to internal details of how `FetchM` / `JobM` are implemented / run and
are no longer used outside of that context.
* **Pretty progress.** Build progress (e.g., `[X/Y] Building Foo`) is
now updated on a single line via ANSI escape sequences when Lake is
outputting to a terminal. Redirected Lake output still sees progress on
separate lines.
* **Warnings-as-error option.** Adds a `--wfail` option to Lake that
will cause a build to fail if Lake logs any warnings doing a build.
Unlike some systems, this does not convert warnings into errors and it
does not abort jobs which log warnings. Instead, only the top-level
build fails.
* **Build log cache.** Logs from builds are now cached to a file and
replayed when the build is revisited. For example, this means multiple
runs of a `--wfail` Lean build (without changes) will still produce the
same warnings even though there is now an up-to-date `.olean` for the
module.
Closes#2349. Closes#2764.
Lake now errors instead of warns on a mismatch between a package name
and what is required as. This avoids sometimes confusing downstream
errors. Also, this change provides additional information for errors
that may be caused by the upcoming Std rename.
Because of the last-added-tried-first rule for macros, all the special
purpose `decreasing_trivial` rules are tried for most recursive
definitions out there, and because they use `apply` and `assumption`
with default transparency may cause some definitoins to be unfolded over
and over again.
A quick test with one of the functions in the leansat project shows that
elaboration time goes down from 600ms to 375ms when using
```
decreasing_by all_goals decreasing_with with_reducible decreasing_trivial
```
instead of
```
decreasing_by all_goals decreasing_with decreasing_trivial
```
This change uses `with_reducible` in most of these macros.
This means that these tactics will no longer work when the
relations/definitions they look for is hidden behind a definition.
This affected in particular `Array.sizeOf_get`, which now has a
companion `sizeOf_getElem`.
In addition, there were three tactics using `apply` to apply Nat-related
lemmas
that we now expect `omega` to solve. We still need them when building
`Init` modules
that don’t have access to `omega`, but they now live in
`decreasing_trivial_pre_omega`,
meant to be only used internally.
- We can set `[irreducible]`, `[semireducible]`, and `[reducible]` for
imported declarations.
- Support for `scoped` and `local` versions
TODO: discuss whether we need all this power after we add the module
system.
It currently only reports how many times each declaration has been
unfolded, and how often the `isDefEq` heuristic for `f a =?= f b` has
been used. Only counters above the threshold are reported.
The subst notation substitues in the expected type, if present, or in
the type of the argument, if no expected type is known.
If there is an expected type it already fails if it cannot find the
equations' left hand side or right hand side. But if the expected type
is not known and the equation's lhs is not present in the second
argument's type, it will happily do a no-op-substitution.
This is inconsistent and unlikely what the user intended to do, so we
now print an error message now.
This still only looks for the lhs; search for the rhs as well seems
prudent, but I’ll leave that for a separate PR, to better diagnose the
impact on mathlib.
This triggers a small number of pointless uses of subst in mathlib, see
https://github.com/leanprover-community/mathlib4/pull/12451
macos-latest changed to arm64. It should not be hard to switch our setup
to cross-compiling x64 instead of arm64 but let's get master green again
first.
We add a new configuration flag for `isDefEq`:
`Meta.Config.univApprox`.
When it is true, we approximate the solution for universe constraints
such as
- `u =?= max u ?v`, we use `?v := u`, and ignore the solution `?v := 0`.
- `max u v =?= max u ?w`, we use `?w := v`, and ignore the solution `?w
:= max u v`.
We only apply these approximations when there the contraints cannot be
postponed anymore. These approximations prevent error messages such as
```
error: stuck at solving universe constraint
max u ?u.3430 =?= u
```
This kind of error seems to appear in several Mathlib files.
We currently do not use these approximations while synthesizing type
class instances.
`Name.append` has special handling of macro scopes, and it would cause
`unresolveNameGlobal` to panic. Using `Name.appendCore` to append name
parts is justified by the fact that it's being used to reassemble a
disassembled name.
Closes#2291
Adds `ppLevel` to the `PPFns` extension so that the coercion can pass
the pretty printing context (including the `pp.mvars` option setting) to
the `Level` formatter.
Expands on #3971 to do something useful even before the PR enters the
queue:
If stage0 changes are detected in the PR, set the changes-stage0 label
(which
has a tooltip to explain what this entail), and also remove the label if
it no
longer applies.
Previously the `ac_rfl` tactic was only really usable when depending on
mathlib. With these instances, `ac_rfl` can deal with the various
operations defined in Lean.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
This makes `exact?%` behave like `by exact?` rather than `by apply?`.
If the underlying function `librarySearch` finds a suggestion which
closes the goal, use it (and add a code action). Otherwise log an error
and use `sorry`. The error is either
```text
`exact?%` didn't find any relevant lemmas
```
or
```text
`exact?%` could not close the goal. Try `by apply` to see partial suggestions.
```
---
[Zulip](https://leanprover.zulipchat.com/#narrow/stream/287929-mathlib4/topic/Useful.20term.20elaborators/near/434863856)
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
<!--
# Read this section before submitting
* Ensure your PR follows the [External Contribution
Guidelines](https://github.com/leanprover/lean4/blob/master/CONTRIBUTING.md).
* Please make sure the PR has excellent documentation and tests. If we
label it `missing documentation` or `missing tests` then it needs
fixing!
* Include the link to your `RFC` or `bug` issue in the description.
* If the issue does not already have approval from a developer, submit
the PR as draft.
* The PR title/description will become the commit message. Keep it
up-to-date as the PR evolves.
* If you rebase your PR onto `nightly-with-mathlib` then CI will test
Mathlib against your PR.
* You can manage the `awaiting-review`, `awaiting-author`, and `WIP`
labels yourself, by writing a comment containing one of these labels on
its own line.
* Remove this section, up to and including the `---` before submitting.
-->
See RFC #3644 for a discussion of design choices.
Closes#3644
Currently this will fail in two tests, because of changes in #3965.
* Sometimes we need to add an additional universe annotation, or we get
a `stuck at solving universe constraint max u ?u =?= u`.
* Sometimes we need to specify arguments that could previously be found
by unification.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
Complement to #3967 , adds a `(since := "<date>")` field to
`@[deprecated]` so that metaprogramming code has access to the
deprecation date for e.g. bulk removals. Also adds `@[deprecated
"deprecation message"]` to optionally replace the default text
"`{declName}` has been deprecated, use `{newName}` instead".
these need manual rebase merges by an admin, so lets prevent accidential
merges via the squashing merge queue.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Adds a `--json` option to the `lean` CLI. When used, the Lean frontend
will print messages as JSON objects using the default `ToJson` encoding
for the `Message` structure. This allows consumers (such as Lake) to
handle Lean output in a more intelligent, well-structured way.
`Message` has been refactored into `BaseMessage`, `Message`, and
`SerialMessage` to enable deriving `ToJson`/ `FromJson` instances
automatically for `BaseMessage` / `SerialMessage`. `SerialMessage` is a
`Message` with its `MessageData` eagerly serialized to a `String`.
- [x] Depends on: #3958
- [x] Depends on: #3960
This makes the UTF-8 encode and decode functions have lean definitions,
so that we can prove properties about them downstream.
previously, the empty `MessageData` (`m!""`) was used to indicate “no
message”, and `throwTacticEx` would format the message differently then.
But the semantics of `MessageData.isEmpty` isn't entirely clear in the
presence of lazy message data (e.g. `.ofPPFormat`).
So to avoid wondering what `isEmpty` should do there, let's simply use
an optional argument to `throwTacticEx` and get rid of
`MessageData.isEmpty`.
This reduces the number of reimplemented functions which complicate
proofs. After inlining it ends up the same as before.
`ltOfOrd` is also changed to use `compare a b = .lt` instead of
`(compare a b == .lt) = true`, for consistency with the normal form in
std.
Rather than adding symm hypotheses to the local context, it now adds
them to the list of hypotheses derived from the local context.
This is not ideal for performance reasons, but it at least closes#3922.
In the future, solveByElim could maintain its own cache of facts that it
updates whenever it does intro.
This PR partly addresses #3458, by adding an option `autoPromoteIndices`
to turn off the promotion of fixed indices to parameters. The actual fix
for the issue is in a separate PR #3591.
Because nested inductive datatypes parameters cannot contain local
variables, it is often desirable for a fixed index to not be promoted,
as to allow free variables in that place. See example in `3458_1.lean`
This is intended to fail at present: it just adds a test case containing
a minimization of a Mathlib slowdown from #3807.
Prior to #3807, the declaration `exists_algHom_adjoin_of_splits'''` at
the end of the file would take around 16,000 heartbeats. Now it takes
around 210,000 heartbeats.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
Previously, there was a function `opaque fromUTF8Unchecked : ByteArray
-> String` which would convert a list of bytes into a string, but as the
name implies it does not validate that the string is UTF-8 before doing
so and as a result it produces unsound results in the compiler (because
the lean model of `String` indirectly asserts UTF-8 validity). This PR
replaces that function by
```lean
opaque validateUTF8 (a : @& ByteArray) : Bool
opaque fromUTF8 (a : @& ByteArray) (h : validateUTF8 a) : String
```
so that while the function is still "unchecked", we have a proof witness
that the string is valid. To recover the original, actually unchecked
version, use `lcProof` or other unsafe methods to produce the proof
witness.
Because this was the only `ByteArray -> String` conversion function, it
was used in several places in an unsound way (e.g. reading untrusted
input from IO and treating it as UTF-8). These have been replaced by
`fromUTF8?` or `fromUTF8!` as appropriate.
* Replaces the unused `Lean.PrettyPrinter.ppConst` with
`MessageData.ofConst` (which similarly avoids an unnecessary `@`) and
that further generates a hover for the constant
* Uses this in `TryThis.addRewriteSuggestion`, so that `rw?` suggestions
don't have unnecessary `@`s.
* Add `MessageData.signature`, as a wrapper around
`PrettyPrinter.signature`, using the same machinery to generate hovers
for constants, improving the hover behaviour in #check so that we get
second order pop-up for constants in the signature. (Not sure how to
write tests for second order hovers, so there is no test for this.)
Adds the ability to show a diff when `guard_msgs` fails, using the
histogram diff algorithm pioneered in jgit. This algorithm tends to
produce more user-friendly diffs, but it can be quadratic in the worst
case. Empirically, the quadratic case of this implementation doesn't
seem to be slow enough to matter for messages smaller than hundreds of
megabytes, but if it's ever a problem, we can mitigate it the same way
jgit does by falling back to Myers diff.
See lean/run/guard_msgs.lean in the tests directory for some examples of
its output.
Adds `IO.FS.Handle.isTty` to check whether a handle is a Windows console
or Unix terminal. Also adds an `isTty` field to `IO.FS.Stream`, so that
this can be checked on, e.g., `stdout`.
Go-to-def on `@[builtin_term_parser]` should go to the line
```lean
builtin_initialize registerBuiltinParserAttribute `builtin_term_parser ``Category.term
```
not
```lean
/-- `term` is the builtin syntax category for terms. ... -/
def term : Category := {}
```
While implementing #3925, I noticed that the performance of the
`textDocument/semanticTokens/full` request is *extremely* bad due to a
quadratic implementation. Specifically, on my machine, computing the
full semantic tokens for `Lean/Elab/Do.lean` took a full 5s. In
practice, this means that while elaborating the file, one core is
entirely busy with computing the semantic tokens for the file.
This PR fixes this performance bug by re-implementing the semantic token
handling, reducing the latency for `Lean/Elab/Do.lean` from 5s to 60ms.
As a result, the overly cautious refresh latency of 5s in #3925 can
easily be reduced to 2s again.
Since the previous semantic tokens implementation used a very brittle
hack to identify projections, this PR also changes the projection
notation elaboration to augment the `InfoTree` syntax for the field of a
projection with a special syntax node of kind
`Lean.Parser.Term.identProjKind`. With this syntax kind, projection
fields can now easily be identified in the `InfoTree`.
`Nat.repr` was implemented by generating a list of `Chars`, each created
by a 10-way if-then-else. This can cause significant slow down in some
particular use cases.
Now `Nat.repr` is `implemented_by` a faster implementation that uses
C++’s `std::to_string` on small numbers (< USize.size) and maintains an
array of pre-allocated strings for the first 128 numbers.
The handling of big numbers (≥ USize.size) remains as before.
The `#guard_msgs` command already runs linters by virtue of using
`elabCommandTopLevel`, so linters should *not* be run on `#guard_msgs`
itself. While we could use a more general solution, of the linters the
unused variables linter is the noisiest one, and it's easy enough to
make it not report messages for `#guard_msgs`.
Just a lemma that we noticed is missing when working on #3880 at the
retreat. We also noticed that there are naming inconsistencies in the
lemmas for `bmod` and `emod`, we should fix that in the future.
@semorrison, does this include all the answers to the questions I asked
in our thread? I think so!
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
while trying to help a user who was facing an unhelpful
```
omega did not find a contradiction:
[0, 0, 0, 0, 1, -1] ∈ [1, ∞)
[0, 0, 0, 0, 0, 1] ∈ [0, ∞)
[0, 0, 0, 0, 1] ∈ [0, ∞)
[1, -1] ∈ [1, ∞)
[0, 0, 0, 1] ∈ [0, ∞)
[0, 1] ∈ [0, ∞)
[1] ∈ [0, ∞)
[0, 0, 0, 1, 1] ∈ [-1, ∞)
```
I couldn’t resist and wrote a pretty-printer for these problem that
shows the linear combination as such, and includes the recognized atoms.
This is especially useful since oftem `omega` failures stem from failure
to recognize atoms as equal. In this case, we now get:
```
omega-failure.lean:19:2-19:7: error: omega could not prove the goal:
a possible counterexample may satisfy the constraints
d - e ≥ 1
e ≥ 0
d ≥ 0
a - b ≥ 1
c ≥ 0
b ≥ 0
a ≥ 0
c + d ≥ -1
where
a := ↑(sizeOf xs)
b := ↑(sizeOf x)
c := ↑(sizeOf x.fst)
d := ↑(sizeOf x.snd)
e := ↑(sizeOf xs)
```
and this might help the user make progress (e.g. by using `case x`
first, and investingating why `sizeOf xs` shows up twice)
Reusing the best profiling UI out there
Usage:
```
lean -Dtrace.profiler=true -Dtrace.profiler.output=profile.json foo.lean ...
```
then open `profile.json` in https://profiler.firefox.com/.
See also `script/collideProfiles.lean` for minimizing and merging
profiles.
Implements a new method to generate instance names for anonymous
instances that uses a heuristic that tends to produce shorter names. A
design goal is to make them relatively unique within projects and
definitely unique across projects, while also using accessible names so
that they can be referred to as needed, both in Lean code and in
discussions.
The new method also takes into account binders provided to the instance,
and it adds project-based suffixes. Despite this, a median new name is
73% its original auto-generated length. (Compare: [old generated
names](https://gist.github.com/kmill/b72bb43f5b01dafef41eb1d2e57a8237)
and [new generated
names](https://gist.github.com/kmill/393acc82e7a8d67fc7387829f4ed547e).)
Some notes:
* The naming is sensitive to what is explicitly provided as a binder vs
what is provided via a `variable`. It does not make use of `variable`s
since, when names are generated, it is not yet known which variables are
used in the body of the instance.
* If the instance name refers to declarations in the current "project"
(given by the root module), then it does not add a suffix. Otherwise, it
adds the project name as a suffix to protect against cross-project
collisions.
* `set_option trace.Elab.instance.mkInstanceName true` can be used to
see what name the auto-generator would give, even if the instance
already has an explicit name.
There were a number of instances that were referred to explicitly in
meta code, and these have been given explicit names.
Removes the unused `Lean.Elab.mkFreshInstanceName` along with the
Command state's `nextInstIdx`.
Fixes#2343
As a special case, makes the `rcases` machinery use `Nat.casesAuxOn` so
that goal states see `0` and `n + 1` rather than `Nat.zero` and
`Nat.succ n`. This is a followup to enabling custom eliminators for
`cases` and `induction`.
This doesn't use custom eliminators in general since `rcases` uses
`Lean.MVarId.cases`, which is completely different from what `cases` and
`induction` use.
Adds options to control whitespace normalization and message ordering in
`#guard_msgs`.
Examples:
1. `#guard_msgs (whitespace := lax)` ignores differences in whitespace
completely.
2. `#guard_msgs (whitespace := exact)` requires an exact match for
whitespace (after trimming).
3. `#guard_msgs (ordering := sorted)` sorts the list of messages, to
make it insensitive to message order.
This should improve the performance of the deriving a bit since it
doesn't have to generate so many matchers. The main motivation though is
to make it easier to prove properties about the expression by using more
standard functions. The generated implementation should end up the same,
since `Ordering.then` is `@[macro_inline]`.
This adds some basic lemmas to support commuting ofInt/toInt and
add/mul.
It also removes the simp annotation on `ofNat_add_ofNat` as in some
contexts the other direction or conversion to Int may be desired.
The Canonicalizer creates a “key” expression eliding certain information
(implicit parameters, levels), and `getFunInfo` can be
confused by these terms (in particular, wrong number of level
parameters).
By running `getFunInfo` on the original expression we avoid this, and
can just put `[]` as the level list in the key.
This changes how Nat typeclass checks in offset terms from syntactic
equality to definitional equality with "instances" transparency.
This may have a negative performance penalty in `isOffset?`, but it
should be small in common cases since the relevant instances are small
terms.
This closes#3836
This removes simp attributes from `Nat.succ.injEq` and
`Nat.succ_sub_succ_eq_sub` to replace them with simprocs. This is
because any reductions involving `Nat.succ` has a high risk of leading
proof performance problems when dealing with even moderately large
numbers.
Here are a couple examples that will both report a maximum recursion
depth error currently. These examples are fixed by this PR.
```
example : (123456: Nat) = 12345667 := by
simp
example (x : Nat) (p : x = 0) : 1000 - (x + 1000) = 0 := by
simp
```
Possibly the more principled fix is to not have `simp` invoke
dischargers under `withReducible`.
In the meantime, this ensures that `falseOrByContra` still succeeds with
`intro1` on a `Not` goal, which previously was breaking `omega` as a
simp discharger.
Closes#3805.
This fixes an issue where the completion would use info nodes before the
cursor for computing completions.
Fixes https://github.com/leanprover/lean4/issues/3462.
ToDo:
- [x] Fix test failures for completions that previously worked by
accident (cc: @Kha)
- [x] stage0 update
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
The matches returned by the lazy discriminator tree are partially
constrained by a priority, but ties are broken by the order in which
keys are traversed and the order of declarations.
This PR changes the match key traversal to use an explicit stack rather
than recursion and implicitly changes the order in which results are
returned to favor left-matches first e.g., given the term `f a b` with
constants `f a b`, and a tree with patterns `f a x -> 1` `f x b -> 2`
that have the same priority, this will return `#[1, 2]` since the early
matches for the key `a` are returned before the match for `x` which has
a star.
This appears to address the [lower quality results mentioned on
zulip](https://leanprover.zulipchat.com/#narrow/stream/428973-nightly-testing/topic/Mathlib.20status.20updates/near/429955747).
When `discharge?` failed, the `usedSimps` was being restored, but the
cache wasn't. This bug was exposed by issue #3710.
This PR makes the following changes:
- We restore the `cache` at `discharge?`. We use `SMap` to ensure the
operation is efficient.
- We don't need the field `dischargeDepth` anymore at `Simp.Result`.
- `UsedSimps` should use `PHashMap` since it is not used linearly.
closes#3710
---------
Co-authored-by: Mario Carneiro <di.gama@gmail.com>
Now, only `(<- ...)`s occurring in the condition of a pure if-then-else
are lifted.
That is, `if (<- foo) then ... else ...` is ok, but `if ... then (<-
foo) else ...` is not. See #3713closes#3713
This PR also adjusts this repo. Note that some of the `(<- ...)` were
harmless since they were just accessing some
read-only state.
It have to keep it as a private definition for now. We currently only
support duplicate theorems in different modules. Splitters are generated
on demand, and are also used to write code.
Adds a `lake lean` CLI command that builds the imports of a Lean file
and then elaborates it via `lean` with the workspace's configuration
(i.e., adding the `moreLeanArgs` / `leanOptions` of the root `package`
and Lake's environment).
* Setting `pp.mvars` to false causes metavariables to pretty print as
`?_`.
* Setting `pp.mvars.withType` to true causes metavariables to pretty
print with type ascriptions.
Motivation: when making tests, it is inconvenient using `#guard_msgs`
when there are metavariables, since the unique numbering is subject to
change.
This feature does not use `⋯` omissions since a metavariable is already
in a sense an omitted term. If repeated metavariables do not appear in
an expression, there is a chance that a term pretty printed with
`pp.mvars` set to false can still elaborate to the correct term, unlike
for other omissions.
(In the future we could consider an option that pretty prints uniquely
numbered metavariables as `?m✝`, `?m✝¹`, `?m✝²`, etc. to be able to tell
them apart, at least in the same pretty printed expression. It would
take care to make sure that these names are stable across different
hovers.)
Closes#3781
The `conv` tactic tries to close “trivial” goals after itself. As of
now, it uses
`try rfl`, which means it can close goals that are only trivial after
reducing with
default transparency. This is suboptimal
* this can require a fair amount of unfolding, and possibly slow down
the proof
a lot. And the user cannot even prevent it.
* it does not match what `rw` does, and a user might expect the two to
behave the
same.
So this PR changes it to `with_reducible rfl`, matching `rw`’s behavior.
I considered `with_reducible eq_refl` to only solve trivial goals that
involve equality,
but not other relations (e.g. `Perm xs xs`), but a discussion on mathlib
pointed out
that it’s expected and desirable to solve more general reflexive goals:
https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/Closing.20after.20.60rw.60.2C.20.60conv.60.3A.20.60eq_refl.60.20instead.20of.20.60rfl.60/near/429851605
Modifies `withBindingBodyUnusedName` to annotate the syntax for the
variable with its corresponding fvar. Now, for example, you can hover
over the variables in `fun x y => ...` in the infoview to see their
types. This change affects notations such as `∃ n, n = 1`, where
hovering over `n` shows that `n : Nat`.
Also adds such annotations for the variables in `let` and `let_fun`.
Implementation note: the variables are annotated with fresh positions
using `nextExtraPos`.
Removes the unused and unnecessary
`Lean.PrettyPrinter.Delaborator.liftMetaM`.
Closes#1618, closes#2737
This PR includes the following fixes:
- Reserved name resolution inside namespaces
- Equation theorems for `match`er declarations are not private anymore
- Equation theorems for `match`er declarations are realizable
- `foo.match_<idx>.splitter` is now a reserved name
Normalize the relative packages directory paths in the pre-rename check
to avoid renames if the difference in paths is only in the path
separators. Also adds a log message on rename.
Commands that can optionally parse an `ident` or parse any number of
`ident`s generally should require that the `ident` use `colGt`. This
keeps typos in commands from being interpreted as identifiers.
For example, without this rule,
```
universe u
Open Lean
````
parses the same as `universe u Open Lean`. It would be better to get an
error on `Open`.
This PR adds `checkColGt` to `section`, `namespace`, `end`, `variable`,
and `universe`.
Closes#2684
Adds an alternative TOML configuration format to Lake.
* Uses TOML v1.0.0 and is fully specification compliant (tested via
[toml-test v1.4.0](https://github.com/toml-lang/toml-test/tree/v1.4.0)).
* Supports package configuration options, Lean libraries, Lean
executables, and dependencies.
* TOML configurations can be generated for new projects via `lake
new|init <pkg> <template>.toml`.
* Supported configurations can be converted to/from TOML via `lake
translate-config <lang>`.
This makes changes to the `GetElem` class so that it does not lead to
unnecessary overhead in container like `RBMap`.
The changes are to:
1. Make `getElem?` and `getElem!` part of the `GetElem` class so they
can be overridden in instances.
2. Introduce a `LawfulGetElem` class that contains correctness theorems
for `getElem?` and `getElem!` using the original definitions.
3. Reorganize definitions (e.g, by moving `GetElem` out of
`Init.Prelude`) so that the `GetElem` changes are feasible.
4. Provide `LawfulGetElem` instances to complement all existing
`GetElem` instances in Lean core.
To reduce the size of the PR, this doesn't do the work of providing new
`GetElem` instances for `RBMap`, `HashMap` etc. That will be done in a
separate PR (#3688) that depends on this.
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
fixes#3657
These functions are mostly not used by lean itself, but it does affect
two occurrences of `ByteArray.toUInt64LE! <$> IO.getRandomBytes 8` which
I left as is instead of switching them to use `toUInt64BE!` to preserve
behavior; but they are random bytes anyway seeded by the OS so it's
unlikely any use of them depending on particular values was sound to
begin with.
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
When it was upstreamed, it lost the mention of "revert/intro pattern",
which is helpful for finding this function. Also extended the
description of the function and clarified some points.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Previously we were suggesting rebasing onto the most recently nightly in
the branches history, but that is incorrect and we should *always*
suggest rebasing on `origin/nightly-with-mathlib`.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
This avoids printing the entire docstring for `⋯` when hovering over it,
which is rather long, and instead it gives a brief reason for omission
and what option to set to pretty print the omitted term.
Previously:
If the `rfl` macro was going to fail, it would:
1. expand to `eq_refl`, which is implemented by
`Lean.Elab.Tactic.evalRefl`, and call `Lean.MVarId.refl` which would:
* either try kernel defeq (if in `.default` or `.all` transparency mode)
* otherwise try `IsDefEq`
* then fail.
2. Next expand to the `apply_rfl` tactic, which is implemented by
`Lean.Elab.Tactic.Rfl.evalApplyRfl`, and call `Lean.MVarId.applyRefl`
which would look for lemmas labelled `@[refl]`, and unfortunately in
Mathlib find `Eq.refl`, so try applying that (resulting in another
`IsDefEq`)
3. Because of an accidental duplication, if `Lean.Elab.Tactic.Rfl` was
imported, it would *again* expand to `apply_rfl`.
Now:
1. Same behaviour in `eq_refl`.
2. The `@[refl]` attribute will reject `Eq.refl`, and `MVarId.applyRefl`
will fail when applied to equality goals.
3. The duplication has been removed.
[Before](https://github.com/leanprover/lean4/files/14772220/oi.pdf) and
[after](https://github.com/leanprover/lean4/files/14772226/oi2.pdf).
This gets `ByteArray`, `String.Extra`, `ToString.Macro` and `RCases` out
of the imports of `omega`. I'd hoped to get `Array.Subarray` too, but
it's tangled up in the list literal syntax. Further progress could come
from make `split` use available `Decidable` instances, so we could pull
out `Classical` (and possibly some of `PropLemmas`).
I think this was in error in my original Mathlib implementation. We're
not interested in relations other than `=`, so there is no point uses
`MVarId.applyRfl`, which just looks up `@[refl]` tagged lemmas and tries
those.
In a separate PR, I will change `MVarId.applyRfl` so it has a flag to
control whether on `=` it should just hand-off to `MVarId.refl`, or
fail. Failure is appropriate in the version we call from the `rfl`
macro, to avoid doing a double `IsDefEq` check on every `rfl`!
This makes several changes to rw? and lazy discrimination trees based on
test failures in rewrite search.
Changes include:
1. Reverting to Mathlib function for candidate lemma priority in rw?
2. Introducing additional filters for auto-generated named in lazy
discriminator tree.
3. Refactoring lazy discriminator values to clarify what is stored.
4. Including star keys in calculation of match closeness in
prioritization.
5. Using more fields in current core context when initializing lazy
discriminator tree and avoiding max heartbeat issues.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
no need to enter `derive_functional_induction` anymore.
(Will remove the support for `derive_functional_induction` after the
next stage0 update, since we are already using it in Init.)
fixes#3770
Also start `rfl` with a `fail` message that is hopefully more helpful
than what we get now (see updated test output). This would be a cheaper
way to address #3302 without changing the implementation of rfl (as
tried in #3714).
This extends `derive_functional_induction` to work with structural
recursion as well.
It produces the less general, more concrete induction rule where the
induction hypothesis is
specialized for every argument of the recursive call, not just the the
one that the function
is recursing on.
Care is taken so that the induction principle and it's motive take the
arguments in the same
order as the original function.
While I was it, also makes sure that the order of the cases in the
induction principle matches
the order of recursive calls in the function better.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
This shortens `Array.findIdx?` code, by using termination_by (and
well-founded recursion) instead of a structural recursion trick, with
the intent to make it more proof friendly.
One motivation is that it makes it easier to write a proof that
`Array.findIdx?` and `List.findIdx?` are equivalent. Furthermore, this
will be useful to prove that more complex functions are equivalent.
Closes#3646
System.FilePath.parent did not return the correct parent path in the
case of absolute file paths
Example of previous behavior
```
(FilePath.mk "/foo").parent -> some (FilePath.mk "")
(System.FilePath.mk "/").parent -> some (FilePath.mk "")
```
The new behavior is based on rust's std::path::Path::parent function (as
previously described in comment in System.FilePath)
Example of updated behavior
```
(System.FilePath.mk "/foo").parent -> some (FilePath.mk "/")
(System.FilePath.mk "/").parent -> none
```
Behavior for relative file paths is unchanged
Closes#3618
Given a definition `foo`, they were previously called `foo._unfold`
until 4.7.0. We tried to rename them to `foo.def`, but it created too
many issues in the Mathlib repo. We decided to rename it again to
`foo.eq_def`. The new name is also consistent with the `eq_<idx>`
theorems generated for different "cases". That is, `foo.eq_def` is the
equality theorem for the whole definition, and `foo.eq_<idx>` is the
equality theorem for case `<idx>`.
cc @semorrison
`FileMap.lines` is an array that seems to be manually managed to have
the form `#[1, 2, ..., n-1, n-1]` with same length as
`FileMap.positions`. Remove this structure field in favour of
calculating the line number as `min(x+1, positions.size-1)` when needed.
Follow-up on #3221
Given
```lean
structure A where
x : Nat
structure B extends A where
y : Nat
```
rather than pretty printing `{ x := 1, y := 2 : B }` as `{ toA := { x :=
1 }, y := 2 }`, it now pretty prints as `{ x := 1, y := 2 }`.
The option `pp.structureInstances.flatten` controls whether to flatten
structure instances like this.
This updates the rw? tactic from Mathlib to use lazy discriminator trees
and upstreams it.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Sets the default value to `pp.fieldNotation.generalized` to `true`.
Updates tests, and fixes some minor flaws in the implementation of the
generalized field notation pretty printer.
Now generalized field notation won't be used for any function that has a
`motive` argument. This is intended to prevent recursors from pretty
printing using it as (1) recursors are more like control flow structures
than actual functions and (2) generalized field notation tends to cause
elaboration problems for recursors.
Note: be sure functions that have an `@[app_unexpander]` use
`@[pp_nodot]` if applicable. For example, `List.toArray` needs
`@[pp_nodot]` to ensure the unexpander prints it using `#[...]`
notation.
The concrete dependency that is stale isn't really actionable
information for users (ideally we'd like something like "amount of
dependencies that will be rebuilt when you restart file"). This also
makes the diagnostic an "information" diagnostic so that non-infoview
users can still see it.
Since we are moving away from using notifications for stale dependency
information, we don't need to provide an ID anymore, either.
Refactors app delaborator, merging in the projection delaborator, to
support pretty printing with generalized field notation.
Renames option `pp.structureProjections` to `pp.fieldNotation` and adds
sub-option `pp.fieldNotation.generalized` to enable/disable generalized
field notation. Adds `@[pp_nodot]` attribute to permanently disable
using field notation for a given declaration.
For now, the default value of `pp.fieldNotation.generalized` is false
since we need a stage0 update to add `@[pp_nodot]` to some core
definitions (such as `List.toArray`) before updating the tests.
[Zulip
discussion](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/.60pp.2EgeneralizedFieldNotation.60/near/425856054)
This attribute, which was implemented in #3640, is applied to the
following structures: `Sigma`, `PSigma`, `PProd`, `And`, `Subtype`, and
`Fin`. These were given this attribute in Lean 3.
This coercion caused difficult-to-diagnose bugs sometimes. Because there
are some situations where converting a string to a name should be done
by parsing the string, and others where it should not, an explicit
choice seems better here.
---------
Co-authored-by: Mac Malone <tydeu@hatpress.net>
I forgot to use the sticky diagnostics in `getInteractiveDiagnostics` in
#3247, leading to them not consistently showing up in the "Messages"
panel of the InfoView.
This lead to incorrect diagnostic spans in the editor and resulted in
header errors that did not show up under "Messages" everywhere in the
file because the `fullRange?` property was missing.
Also changes the "Import out of date" warning diagnostic severity to
"Hint" so that it doesn't show up in the "Problems" view.
This is a rewrite of the `UnusedVariables` lint to inline and simplify
many of the dependent functions to try to improve the performance of
this lint, which quite often shows up in perf reports.
* The mvar assignment scanning is one of the most expensive parts of the
process, so we do two things to improve this:
* Lazily perform the scan only if we need it
* Use an object-pointer hashmap to ensure that we don't have quadratic
behavior when there are many mvar assignments with slight differences.
* The dependency on `Lean.Server` is removed, meaning we don't need to
do the LSP conversion stuff anymore. The main logic of reference finding
is inlined.
* We take `fvarAliases` into account, and union together fvars which are
aliases of a base fvar. (It would be great if we had `UnionFind` here.)
More docs will be added once we confirm an actual perf improvement.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Closes#3706
This derive handler's implementation is very similar to `BEq`'s, which
already ignores unused alternative so as to work correctly on indexed
inductive types. This PR simply implements the same solution as the one
present in
[`BEq.lean`](2c15cdda04/src/Lean/Elab/Deriving/BEq.lean (L94)).
After some tests, it doesn't seem like any other derive handler present
in Core suffers from the same issue (though some handlers don't work on
indexed inductives for other reasons).
Sends a diagnostic informing the user to run Restart File when a file
dependency is saved.
Based on #3014 because this feature was easier to implement with the new
architecture.
ToDo:
- [x] Adjust vscode-lean4 to display a notification when this diagnostic
appears in a non-annoying way
(https://github.com/leanprover/vscode-lean4/pull/393)
- [x] Use a file watcher to identify changes to files not tracked by VS
Code
- [x] Rebase onto master when #3014 is merged
These are used in Mathlib's `congr!` and `convert` tactics, which will
be upstreamed soon.
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
- Removes the public definitions `Array.eraseIdxAux` and
`Array.eraseIdxSzAux` which were implementation details.
- Motivation: `Array.eraseIdxAux` and `Array.eraseIdxSzAux` were clearly
not intended to remain public, but simply making them private would make
it inconvenient to unfold them when writing proofs in Std.
- Adds documentation comments to the public `Array.eraseIdx`-related
definitions which remain.
- Removes `Array.eraseIdx'` which was just `Array.feraseIdx` wrapped in
a subtype and adds `Array.size_feraseIdx` to prove the subtype property
as a standalone theorem.
Co-Authored-By: Daniel Windham <daniel@atlascomputing.org>
Previously, if there was a `nightly-testing-YYYY-MM-DD` tag at Std, but
not Mathlib, we were erroneously proceeding with Mathlib CI, and hence
using a probably-broken version of Mathlib.
using the `substVars` tactic on the goal can remove too much
information, as it does not take into account that the `motive` may
depend on the fixed parameters.
This is fixed by etracting `substVar` from `subst` which expects the
`x`, not the `h : x = rhs`, and then using this tactic on the local
declarations _after_ the `motive` exclusively.
a common pattern for recursive functions is
```
def countUp (n i acc : Nat) : Nat :=
if i < n then
countUp n (i+1) (acc + i)
else
acc
```
where we increase a value `i` until it hits an upper bound. This is
particularly common with array processing functions:
```
$ git grep 'termination_by.*size.*-' src/|wc -l
26
```
GuessLex now recognizes this pattern. The general approach is:
For every recursive call, check if the context contains hypotheses of
the form `e₁ < e₂` (or similar comparisions), and then consider `e₂ -
e₁` as a termination argument.
Currently, this only fires when `e₁` and `e₂` only depend on the
functions parameters, but not local let-bindings or variables bound in
local pattern matches.
Duplicates are removed.
In the table showing the termination argument failures, long termination
arguments are now given a number and abbreviated as e.g. `#4` in the
table headers.
More examples in the test file, here as some highlights:
```
def distinct (xs : Array Nat) : Bool :=
let rec loop (i j : Nat) : Bool :=
if _ : i < xs.size then
if _ : j < i then
if xs[j] = xs[i] then
false
else
loop i (j+1)
else
loop (i+1) 0
else
true
loop 0 0
```
infers
```
termination_by (Array.size xs - i, i - j)
```
and the weird functions where `i` goes up or down
```
def weird (xs : Array Nat) (i : Nat) : Bool :=
if _ : i < xs.size then
if _ : 0 < i then
if xs[i] = 42 then
weird xs.pop (i - 1)
else
weird xs (i+1)
else
weird xs (i+1)
else
true
decreasing_by all_goals simp_wf; omega
```
infers
```
termination_by (Array.size xs - i, i)
```
but unfortunately needs `decreasing_by` pending the “big
decreasing_tactic refactor” that
I expect we’ll want to do at some point.
this refactor prepares GuessLex to be able to infer more complex
termination arguments.
As a side-effect it fixes an (obscure) bug where `sizeOf` would be
applied to a term of the wrong type and thus a wrong `SizeOf` instance
could be inferred.
On Windows, we now compile all core `.o`s twice, once with and without
`dllexport`, for use in the shipped dynamic and static libraries,
respectively. On other platforms, we export always as before to avoid
the duplicate work.
---------
Co-authored-by: tydeu <tydeu@hatpress.net>
This reverts commit 4e3a8468c3 for PR
#3619. It looks like the CI in that commit didn't inform me that a test
was broken by the PR, so I managed to commit it despite the broken test.
- Add support for reserved declaration names. We use them for theorems
generated on demand.
- Equation theorems are not private declarations anymore.
- Generate equation theorems on demand when resolving symbols.
- Prevent users from creating declarations using reserved names. Users
can bypass it using meta-programming.
See next test for examples.
Before, the termination argument as inferred by `GuessLex` was passed
further
on as `Syntax`, to be elaborated later in `WF.Rel`.
This didn’t feel quite right anymore. In particular if we want to teach
`GuessLex` about guessing more complex termination arguments like
`xs.size -
i`, using `Expr` here is more natural.
So this introduces `TerminationArgument` based on an `Expr` to be used
here.
A side-effect of how the termination arguments are elaborated is that
the unused
variables linter will now look at `termination_by` variables, and that
parameters
past the colon are not even invisibly in scope, so `‹_›` will not find
them
See https://github.com/leanprover-community/mathlib4/pull/11370/files
for examples
of fixing these changes.
This PR fixes an issue where the file worker would not provide the
client with semantic tokens until the file had been elaborated
completely. The file worker now also tells the client to refresh its
semantic tokens after running "Restart File". This PR is based on #3271.
Fixes#1170.
This PR adds the module name to `RefIdent` in order to distinguish
conflicting names from different files. This also fixes related issues
in find-references or the call hierarchy feature.
It also adds some docstrings and stylistically refactors a bunch of
code.
This PR adds support for requests from the server to the client in the
language server. It is based on #3014 and was developed during an
experiment for #3247 that unfortunately did not go anywhere.
This replaces a few uses of initialize with builtin_initialize, and
removes some unneeded functionality added when it was unclear if lazy
discriminator trees would be efficient enough.
This introduces the `ArgsPacker` module and abstraction, to replace the
exising `PackDomain`/`PackMutual` code. The motivation was that we now
have more uses besides `Fix.lean` (`GuessLex` and `FunInd`), and the
code was spread in various places.
The goals are
* consistent function naming withing the the `PSigma` handling, the
`PSum` handling, and the combined interface
* avoid taking a type apart just based on the `PSigma`/`PSum` nesting,
to be robust in case the user happens to be using `PSigma`/`PSum`
somewhere. Therefore, always pass an `arity` or `numFuncs` or `varNames`
around.
* keep all the `PSigma`/`PSum` encoding logic contained within one
module (`ArgsPacker`), and keep that module independent of its users (so
no `EqnInfos` visible here).
* pick good variable names when matching on a packed argument
* the unary function now is either called `fun1._unary` or
`fun1._mutual`, never `fun1._unary._mutual`.
This file has less heavy dependencies than `PackMutual` had, so build
parallelism is improved as well.
This makes trailing whitespace visible and protectes them against
trimming by the editor, by appending the symbol ⏎ to such a line (and
also to any line that ends with such a symbol, to avoid ambiguities in
the case the message already had that symbol).
(Only the code action output / docstring parsing is affected; the error
message as sent
to the InfoView is unaffected.)
Fixes#3571
Modifies `dsimpLocation'` (which implements `dsimp?`) to take a
`simprocs : SimprocsArray` argument, like `simpLocation` and
`dsimpLocation`. This ensures that the behavior of `dsimp` matches
`dsimp?`.
---
Closes#3653
Enables the combination of `noncomputable unsafe` to be used for
definitions. Outside of pure theory, `noncomputable` is also useful to
prevent Lean from compiling a definition which will be implemented with
external code later. Such definitions may also wish to be marked
`unsafe` if they perform morally impure or memory-unsafe functions.
this makes `termination_by?` even slicker.
The heuristics is agressive in the non-mutual case (will omit `sizeOf`
if the argument is non-dependent and the `WellFoundedRelation` relation
is via `sizeOfWFRel`.
In the mutual case we'd also have to check the arguments, as they line
up in the termination argument, have the same types. I did not bother at
this point; in the mutual case we omit `sizeOf` only if the argument
type is `Nat`.
As a drive-by fix, `termination_by?` now also works on functions that
have only one plausible measure.
This is a temporary workaround for a limitation on Windows shared
libraries. We are getting errors of the form:
```
ld.lld: error: too many exported symbols (got 65572, max 65535)
```
Replaces `@[eliminator]` with two attributes `@[induction_eliminator]`
and `@[cases_eliminator]` for defining custom eliminators for the
`induction` and `cases` tactics, respectively.
Adds `Nat.recAux` and `Nat.casesAuxOn`, which are eliminators that are
defeq to `Nat.rec` and `Nat.casesOn`, but these use `0` and `n + 1`
rather than `Nat.zero` and `Nat.succ n`.
For example, using `induction` to prove that the factorial function is
positive now has the following goal states (thanks also to #3616 for the
goal state after unfolding).
```lean
example : 0 < fact x := by
induction x with
| zero => decide
| succ x ih =>
/-
x : Nat
ih : 0 < fact x
⊢ 0 < fact (x + 1)
-/
unfold fact
/-
...
⊢ 0 < (x + 1) * fact x
-/
simpa using ih
```
Thanks to @adamtopaz for initial work on splitting the `@[eliminator]`
attribute.
Floris van Doorn [reported on
Zulip](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/have.20tactic.20error.20recovery/near/425283053)
that it is confusing that the `have : T := e` tactic completely fails if
the body `e` is not of type `T`. This is in contrast to `have : T := by
exact e`, which does not completely fail when `e` is not of type `T`.
This ends up being caused by `elabTermEnsuringType` throwing an error
when it fails to insert a coercion. Now, it detects this case, and it
checks the `errToSorry` flag to decide whether to throw the error or to
log the error and insert a `sorry`.
This is justified by `elabTermEnsuringType` being a frontend to
`elabTerm`, which inserts `sorry` on error.
An alternative would be to make `ensureType` respect `errToSorry`, but
there exists code that expects being able to catch when `ensureType`
fails. Making such code manipulate `errToSorry` seems error prone, and
this function is not a main entry point to the term elaborator, unlike
`elabTermEnsuringType`.
Remark: this commit removes the `jason1.lean` test. Motivation: It
breaks all the time due to changes we make, and it is not clear anymore
what it is testing.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
This migrates lemmas about Nat `compare`, `min`, `max`, `dvd`, `gcd`,
`lcm` and `div`/`mod` from Std to Lean itself.
Std still has some additional recursors, `CoPrime` and a few additional
definitions that might merit further discussion prior to upstreaming.
If the `LEAN_GITHASH` environment variable is set, Lake will now use it
instead of the detected Lean's githash when computing traces for builds
and the elaborated Lake configuration. This override allows one to
replace the Lean version used by a library
(e.g., Mathlib) without completely rebuilding it, which is useful for
testing custom builds of Lean.
The `delabConstWithSignature` delaborator is responsible for pretty
printing constants with a declaration-like signature, with binders, a
colon, and a type. This is used by the `#check` command when it is given
just an identifier.
It used to accumulate binders from pi types indiscriminately, but this
led to unfriendly behavior. For example, `#check String.append` would
give
```
String.append (a✝ : String) (a✝¹ : String) : String
```
with inaccessible names. These appear because `String.append` is defined
using patterns, so it never names these parameters.
Now the delaborator stops accumulating binders once it reaches an
inaccessible name, and for example `#check String.append` now gives
```
String.append : String → String → String
```
We do not synthesize names for the sake of enabling binder syntax
because the binder names are part of the API of a function — one can use
`(arg := ...)` syntax to pass arguments by name. The delaborator also
now stops accumulating binders once it reaches a parameter with a name
already seen before — we then rely on the main delaborator to provide
that parameter with a fresh name when pretty printing the pi type.
As a special case, instance parameters with inaccessible names are
included as binders, pretty printing like `[LT α]`, rather than
relegating them (and all the remaining parameters) to after the colon.
It would be more accurate to pretty print this as `[inst✝ : LT α]`, but
we make the simplifying assumption that such instance parameters are
generally used via typeclass inference. Likely `inst✝` would not
directly appear in pretty printer output, and even if it appears in a
hover, users can likely figure out what is going on. (We may consider
making such `inst✝` variables pretty print as `‹LT α›` or
`infer_instance` in the future, to make this more consistent.)
Something we note here is that we do not do anything to make sure
parameters that can be used as named arguments actually appear named
after the colon (nor do we assure that the names are the correct names).
For example, one sees `foo : String → String → String` rather than `foo
: String → (baz : String) → String`. We can investigate this later if it
is wanted.
We also give `delabConstWithSignature` a `universes` flag to enable
turning off pretty printing universe levels parameters.
Closes#2846
this makes the ugly `fst`/`snd` variable names in the functional
induction principles go away.
Ironically I thought in order to fix these name, I should touch the
mutual/n-ary argument packing code used for well-founded recursion, and
embarked on a big refactor/rewrite of that code, only to find that at
least this particular instance of the issue was somewhere else. Hence
breaking this into its own PR; the refactoring will follow (and will
also improve some other variable names.)
closes#3022
With this commit, given the declaration
```
def foo : Nat → Nat
| 0 => 2
| n + 1 => foo n
```
when we unfold `foo (n+1)`, we now obtain `foo n` instead of `foo
(Nat.add n 0)`.
This fixes an issue discovered in Mathlib with the meta cache being
poisoned by using a name generator. It is difficult to reproduce due to
the name collisions being rare, but here is a minimal module with
definitions that result in an error:
```lean
prelude
universe u
inductive Unit2 : Type where
| unit : Unit2
inductive Eq2 {α : Sort u} : α → α → Prop where
| refl (a : α) : Eq2 a a
structure Subtype2 {α : Sort u} (p : α → Prop) where
val : α
def End (α) := α → α
theorem end_app_eq (α : Type u) (f : End α) (a : α) : Eq2 (f a) (f a) := Eq2.refl _
theorem Set.coe_eq_subtype {α : Type u} (s : α → Prop) : Eq2 (Subtype2 s) (Subtype2 s) := Eq2.refl _
def succAboveCases {_ : Unit2} {α : Unit2 → Sort u} (i : Unit2) (v : α i) : α i := v
theorem succAbove_cases_eq_insertNth : Eq2 @succAboveCases.{u + 1} @succAboveCases.{u + 1} := Eq2.refl _
```
Removing any of thee last 5 definitions avoids the error. Testing
against Mathlib shows this PR fixes the issue.
This bug is the real cause of leanprover/vscode-lean4#392.
At the end of a tactic state, the client calls
`getInteractiveDiagnostics` with a range `[last line of proof, last line
of proof + 1)`. The `fullRange` span of the `unresolved goals` error
however is something like `[(first line of proof, start character),
(last line of proof, nonzero end character)).
Since it operates on line numbers, `getInteractiveDiagnostics` would
then check whether `[last line of proof, last line of proof + 1)` and
`[first line of proof, last line of proof)` intersect, which is false
because of the excluded upper bound on the latter interval, despite the
fact that the end character in the last line may be nonzero.
This fix adjusts the intersection logic to use `[first line of proof,
last line of proof]` if the end character is nonzero.
Closesleanprover/vscode-lean4#392.
This PR enables import auto-completion to complete partial words in
imports.
Other inconsistencies that I've found in import completion already seem
to be fixed by #3014. Since it will be merged soon, there is no need to
invest time to fix these issues on master.
This adds the concept of **functional induction** to lean.
Derived from the definition of a (possibly mutually) recursive function,
a **functional
induction principle** is tailored to proofs about that function. For
example from:
```
def ackermann : Nat → Nat → Nat
| 0, m => m + 1
| n+1, 0 => ackermann n 1
| n+1, m+1 => ackermann n (ackermann (n + 1) m)
derive_functional_induction ackermann
```
we get
```
ackermann.induct (motive : Nat → Nat → Prop) (case1 : ∀ (m : Nat), motive 0 m)
(case2 : ∀ (n : Nat), motive n 1 → motive (Nat.succ n) 0)
(case3 : ∀ (n m : Nat), motive (n + 1) m → motive n (ackermann (n + 1) m) → motive (Nat.succ n) (Nat.succ m))
(x x : Nat) : motive x x
```
At the moment, the user has to ask for the functional induction
principle explicitly using
```
derive_functional_induction ackermann
```
The module docstring of `Lean/Meta/Tactic/FunInd.lean` contains more
details on the
design and implementation of this command.
More convenience around this (e.g. a `functional induction` tactic) will
follow eventually.
This PR includes a bunch of `PSum`/`PSigma` related functions in the
`Lean.Tactic.FunInd`
namespace. I plan to move these to `PackArgs`/`PackMutual` afterwards,
and do some cleaning
up as I do that.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
This is still WIP: the checklist for release candidates will get
finished as I do the release of `v4.7.0-rc1`.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
This adds a number of lemmas for simplification of `Bool` and `Prop`
terms. It pulls lemmas from Mathlib and adds additional lemmas where
confluence or consistency suggested they are needed.
It has been tested against Mathlib using some automated test
infrastructure.
That testing module is not yet included in this PR, but will be included
as part of this.
Note. There are currently some comments saying the origin of the simp
rule. These will be removed prior to merging, but are added to clarify
where the rule came from during review.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Proves
`Nat.mod_mul : x % (a * b) = x % a + a * (x / a % b)` and
`Nat.mod_pow_succ : x % b ^ (k + 1) = x % b ^ k + b ^ k * ((x / b ^ k) %
b)`, helpful for bitblasting.
We use `let_delayed` to elaborate `match_expr` join points, which
elaborate the body of the `let` before its value. Thus, there is a
difference between:
- `let_delayed f (x : Expr) := <val>; <body>`
- `let_delayed f := fun (x : Expr) => <val>; <body>`
In the latter, when `<body>` is elaborated, the elaborator does not know
that `f` takes an argument of type `Expr`, and that `f` is a function.
Before this commit ensures the former representation is used.
In v4.6.0, there was a significant regression in initial server startup
performance because the .ilean files got bigger in #3082 and we load the
information stored in all .ilean files synchronously when the server
starts up.
This PR makes this loading asynchronous. The trade-off is that requests
that are issued right after the initial server start when the references
are not fully loaded yet may yield incomplete results.
Benchmark for this in a separate PR soon after this one.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Else the `case` will now allow introducing all necessary variables.
Induction principles with `let` in the types of the cases will be more
common with #3432.
This implementation no longer reduces the type as it goes, but really
only counts
manifest foralls and lets. I find this more sensible and predictable: If
you have
```
theorem induction₂_symm {P : EReal → EReal → Prop} (symm : Symmetric P) …
```
then previously, writing
```
case symm =>
```
would actually bring a fresh `x` and `y` and variable `h : P x y` into
scope and produce a
goal of `P y x`, because `Symmetric P` happens to be
```
def Symmetric := ∀ ⦃x y⦄, x ≺ y → y ≺ x
```
After this change, after `case symm =>` will leave `Symmetric P` as the
goal.
This gives more control to the author of the induction hypothesis about
the actual
goal of the cases. This shows up in mathlib in two places; fixes in
https://github.com/leanprover-community/mathlib4/pull/11023.
I consider these improvements.
the user can now write `termination_by?` to see the termination argument
inferred by GuessLex, and turn it into `termination_by …` using the “Try
this” widget or a code action.
To be done later, maybe: Avoid writing `sizeOf` if it's not necessary.
When editing core Lean, the `pp.proofs` feature causes goal states to fail to display in the Infoview, instead showing only "error when printing message: unknown constant '«term⋯»'". This PR moves the `⋯` syntax from Init.NotationExtra to Lean.Elab.BuiltinTerm
It also makes it so that `⋯` elaborates as `_` while logging a warning, rather than throwing an error, which should be somewhat more friendly when copy/pasting from the Infoview.
Closes#3476
When using `set_option tactic.skipAssignedInstances false`, `simp` and
`rw` will synthesize instance implicit arguments even if they have
assigned by unification. If the synthesized argument does not match the
assigned one the rewrite is not performed. This option has been added
for backward compatibility.
```
example (a : Nat) :
(((a + (2 ^ 64 - 1)) % 2 ^ 64 + 1) * 8 - 1 - (a + (2 ^ 64 - 1)) % 2 ^ 64 * 8 + 1) = 8 := by
omega
```
used to time out, and now is fast.
(We will probably make separate changes later so the defeq checks would
be fast in any case here.)
During the switch to `.lake`, I overlooked updating the paths in
`LakeInstall`. This fixes that and helps prevent further mistakes by
using the same default definitions as the package configuration itself.
The `decide` tactic produces error messages that users find to be
obscure. Now:
1. If the `Decidable` instance reduces to `isFalse`, it reports that
`decide` failed because the proposition is false.
2. If the `Decidable` instance fails to reduce, it explains what
proposition it failed for, and it shows the reduced `Decidable` instance
rather than the `Decidable.decide` expression. That expression tends to
be less useful since it shows the unreduced `Decidable` argument (plus
it's a lot longer!)
Examples:
```lean
example : 1 ≠ 1 := by decide
/-
tactic 'decide' proved that the proposition
1 ≠ 1
is false
-/
opaque unknownProp : Prop
open scoped Classical in
example : unknownProp := by decide
/-
tactic 'decide' failed for proposition
unknownProp
since its 'Decidable' instance reduced to
Classical.choice ⋯
rather than to the 'isTrue' constructor.
-/
```
When reporting the error, `decide` only shows the whnf of the
`Decidable` instance. In the future we could consider having it reduce
all decidable instances present in the term, which can help with
determining the cause of failure (this was explored in
8cede580690faa5ce18683f168838b08b372bacb).
The elaboration function `Lean.Meta.coerceMonadLift?` inserts these
coercion helper functions into a term and tries to unfolded them with
`expandCoe`, but because that function only unfolds up to
reducible-and-instance transparency, these functions were not being
unfolded. The fix here is to give them the `@[reducible]` attribute.
with this, more functions will be proven terminating automatically,
namely those where after `simp_wf`, lexicographic order handling,
possibly `subst_vars` the remaining goal can be solved by `omega`.
Note that `simp_wf` already does simplification of the goal, so
this adds `omega`, not `(try simp) <;> omega` here.
There are certainly cases where `(try simp) <;> omega` will solve more
goals (e.g. due to the `subst_vars` in `decreasing_with`), and
`(try simp at *) <;> omega` even more. This PR errs on the side of
taking
smaller steps.
Just appending `<;> omega` to the existing
`simp (config := { arith := true, failIfUnchanged := false })` call
doesn’t work nicely, as that leaves forms like `Nat.sub` in the goal
that
`omega` does not seem to recognize.
This does *not* remove any of the existing ad-hoc `decreasing_trivial`
rules based on `apply` and `assumption`, to not regress over the status
quo (these rules may apply in cases where `omega` wouldn't “see”
everything, but `apply` due to defeq works).
Additionally, just extending makes bootstrapping easier; early in `Init`
where
`omega` does not work yet these other tactics can still be used.
(Using a single `omega`-based tactic was tried in #3478 but isn’t quite
possible yet, and will be postponed until we have better automation
including forward reasoning.)
with this, hopefully more obvious array accesses will be handled
automatically.
Just like #3503, this PR does not investiate which of the exitsting
tactics in `get_elem_tactic_trivial` are subsumed now and could be
dropped without (too much) breakage.
Before, app unexpanders would only be applied to entire applications.
However, some notations produce functions, and these functions can be
given additional arguments. The solution so far has been to write app
unexpanders so that they can take an arbitrary number of additional
arguments. However, as reported in [this Zulip
thread](https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/pretty.20printer.20bug/near/420662236),
this leads to misleading hover information in the Infoview. For example,
while `HAdd.hAdd f g 1` pretty prints as `(f + g) 1`, hovering over `f +
g` shows `f`. There is no way to fix the situation from within an app
unexpander; the expression position for `HAdd.hAdd f g` is absent, and
app unexpanders cannot register TermInfo.
This commit changes the app delaborator to try running app unexpanders
on every prefix of an application, from longest to shortest prefix. For
efficiency, it is careful to only try this when app delaborators do in
fact exist for the head constant, and it also ensures arguments are only
delaborated once. Then, in `(f + g) 1`, the `f + g` gets TermInfo
registered for that subexpression, making it properly hoverable.
The app delaborator is also refactored, and there are some bug fixes:
- app unexpanders only run when `pp.explicit` is false
- trailing parameters in under-applied applications are now only
considered up to reducible & instance transparency, which lets, for
example, optional arguments for `IO`-valued functions to be omitted.
(`IO` is a reader monad, so it's hiding a pi type)
- app unexpanders will no longer run for delaborators that use
`withOverApp`
- auto parameters now always pretty print, since we are not verifying
that the provided argument equals the result of evaluating the tactic
Furthermore, the `notation` command has been modified to generate an app
unexpander that relies on the app delaborator's new behavior.
The change to app unexpanders is reverse-compatible, but it's
recommended to update `@[app_unexpander]`s in downstream projects so
that they no longer handle overapplication themselves.
If Lake fails to download a cloud release, it will now print a warning
indicating that it is falling back to a local build. For example:
```
[0/2] Downloading cloud_test cloud release
[0/2] Building CloudTest
error: > curl -s -f -o [...] -L [...]
error: external command `curl` exited with code 22
warning: fetching cloud release failed; falling back to local build
```
This PR addresses several performance issues in the auto-completion
implementation. It also fixes a number of smaller bugs related to
auto-completion.
In a file with `import Mathlib`, the performance of various kinds of
completions has improved as follows:
- Completing `C`: 49000ms -> 1400ms
- Completing `Cat`: 14300ms -> 1000ms
- Completing `x.` for `x : Nat`: 3700ms -> 220ms
- Completing `.` for an expected type of `Nat`: 11000ms -> 180ms
The following bugs have been fixed as well:
- VS Code never used our custom completion order. Now, the server fuzzy
completion score decides the order that completions appear in.
- Dot auto-completion for private types did not work at all. It does
now.
- Completing `.<identifier>` (where the expected type is used to infer
the namespace) did not filter by the expected type and instead displayed
all matching constants in the respective namespace. Now, it uses the
expected type for filtering. Note that this is not perfect because
sub-namespaces are technically correct completions as well (e.g.
`.Foo.foobar`). Implementing this is future work.
- Completing `.` was often not possible at all. Now, as long as the `.`
is not used in a bracket (where it may be used for the anonymous lambda
feature, e.g. `(. + 1)`), it triggers the correct completion.
- Fixes#3228.
- The auto-completion in `#check` commands would always try to complete
identifiers using the full declaration name (including namespaces) if it
could be resolved. Now it simply uses the identifier itself in case
users want to complete this identifier to another identifier.
## Details
Regarding completion performance, I have more ideas on how to improve it
further in the future.
Other changes:
- The feature that completions with a matching expected type are sorted
to the top of the server-side ordering was removed. This was never
enabled in VS Code because it would use its own completion item order
and when testing it I found it to be more confusing than useful.
- In the server-side ordering, we would always display keywords at the
top of the list. They are now displayed according to their fuzzy match
score as well.
The following approaches have been used to improve performance:
- Pretty-printing the type for every single completion made up a
significant amount of the time needed to compute the completions. We now
do not pretty-print the type for every single completion that is offered
to the user anymore. Instead, the language server now supports
`completionItem/resolve` requests to compute the type lazily when the
user selects a completion item.
- Note that we need to keep the amount of properties that we compute in
a resolve request to a minimum. When the server receives the resolve
request, the document state may have changed from the state it was in
when the initial auto-completion request was received. LSP doesn't tell
us when it will stop sending resolve requests, so we cannot keep this
state around, as we would have to keep it around forever.
LSP's solution for this dilemma is to have servers send all the state
they need to compute a response to a resolve request to the client as
part of the initial auto completion response (which then sends it back
as part of the resolve request), but this is clearly infeasible for all
real language servers where the amount of state needed to resolve a
request is massive.
This means that the only practical solution is to use the current state
to compute a response to the resolve request, which may yield an
incorrect result. This scenario can especially occur when using
LiveShare where the document is edited by another person while cycling
through available completions.
- Request handlers can now specify a "header caching handler" that is
called after elaborating the header of a file. Request handlers can use
this caching handler to compute caches for information stored in the
header. The auto-completion uses this to pre-compute non-blacklisted
imported declarations, which in turn allow us to iterate only over
non-blacklisted imported declarations where we would before iterate over
all declarations in the environment. This is significant because
blacklisted declarations make up about 4/5 of all declarations.
- Dot completion now looks up names modulo private prefixes to figure
out whether a declaration is in the namespace of the type to the left of
the dot instead of first stripping the private prefix from the name and
then comparing it. This has the benefit that we do not need to scan the
full name in most cases.
This PR also adds a couple of regression tests for fixed bugs, but *no
benchmarks*. We will add these in the future when we add proper support
for benchmarking server interaction sessions to our benchmarking
architecture.
All tests that were broken by producing different completion output
(empty `detail` field, added `sortText?` and `data?` fields) have been
manually checked by me to be still correct before replacing their
expected output.
This is still a draft PR, but includes the core exact? and apply?
tactics.
Still need to convert to builtin syntax and test on Std.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
The current `ToExpr Int` instance produces `@Int.ofNat (@OfNat.ofNat Nat
i ...)` for nonnegative `i` and `@Int.negSucc (@OfNat.ofNat Nat (-i+1)
...)` for negative `i`.
However it should be producing `@OfNat.ofNat Int i ...` for nonnegative
`i`, and `@Neg.neg ... (@OfNat.ofNat Int (-i) ...)` for negative `i`.
Make `x.toNat * 2 + b.toNat` the simp normal form of `(concat x
b).toNat`.
The choice for multiplication and addition was inspired by `Nat.bit_val`
from Mathlib.
Also, because we have considerably more lemmas about multiplication and
`_ + 1` than about shifts and `_ ||| 1`.
This is very helpful when dealing with bitvectors, where a case analysis
on the bitwidth leaves one with hypotheses of the form `x<2^(Nat.succ
w)`.
Design decisions I am unsure about:
- Is creating a helper `succ?` the correct way to match on the exponent
`e+1`?
- I'm not certain why the prior call to `Int.ofNat_pow` also checked
that the exponent was a ground natural. I removed this, since we now
explicitly handle cases where the exponent is a term of the form `e+1`.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Co-authored-by: Joe Hendrix <joe@lean-fro.org>
Co-authored-by: Alex Keizer <alex@keizer.dev>
First (baby)-step to a `concat`-based `bitblast`: a characterization of
`concat` in terms of `getLsb`.
The proof might benefit slightly from a `toNat_concat` lemma, but I
wasn't sure what the normal form there should be, so I avoided it.
---------
Co-authored-by: Scott Morrison <scott@tqft.net>
Every usage of `carry` followed the pattern: `carry _ x.toNat y.toNat`,
so we've refactorod `carry` to take the `BitVec`s as arguments, and made
the `toNat` part of its definition.
PR #3432 will introduce more operations on `MatcherApp`, including somet
that have more dependencies.
This change prepares by introducing `Lean.Meta.Match.MatcherApp.Basic`
for the basic definition, and `Lean.Meta.MatcherApp.Transform` for the
transformations, currently `addArg` and `refineThrough`, but more to
come.
This PR is an effort to improve reasoning at the Nat level about
bitvectors and reduce of Fin and Nat.
It slightly tightens some proofs, but is generally aimed at reducing
inconsistencies between definitions at the Nat and Fin types in favor of
more consistently using Nat operations.
This ports leanprover/std4#664 to Lean core.
Here was the rational I provided in the discussion for
leanprover/std4#664:
It's mostly about consistency. If we use the same types and style in
definitions and proofs, there is less surprise when unfolding or
otherwise using definitions. We use some Nat based operations that
haven't been extended to Fin such as the bitwise operations, and I don't
want to pay the overhead of introducing a Fin version of every Bitvector
operation.
So this basically means Nat is preferred.
One argument potentially in favor of Fin is that we could reuse results
proven there, but that doesn't really seem to be the case so far.
A second argument is that we want to simplify expression to use more
canonical forms and we currently can pretty-print those operations
better using ofNat than ofFin. We could define the notations using ofFin
of course though, but that's additional operators that will show up in
expressions.
Adds a simple error-recovery mechanism to Lean's parser, similar to
those used in other combinator parsing libraries.
Lean itself isn't very amenable to error recovery with this mechanism,
as it requires global knowledge of the grammar in question to write
recovery rules that don't break backtracking or `<|>`. I only found a
few opportunities.
But for DSLs, this is really important. In particular, Verso parse
errors interacted very badly with Lean parse errors in a way that
required frequent "restart file" commands, but this mechanism allows me
to both recover from Verso parse errors and to have Lean skip the rest
of the file rather than repeatedly trying to parse it as Lean commands.
#3408 was somewhat large and didn't properly test the symm and label
attribute code after edits to the builtin versions.
This migrates the code for generating labeled attributes from Init back
to Lean so that the required definitions are in scope.
This also addresses a mistake in the symm elaborator that prevented symm
without location information from elaborating.
Both fixes have been tested on the Std test suite and successfully
passed.
Adds documentation to the `String.Iterator` API, mentored by
@eric-wieser and @david-christiansen
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
This is a quite substantial tactic.
It also includes the infamour `NatCast` typeclass (which I've equipped
with a module-doc). I wasn't at all sure where that should live, so it
is currently randomly in `Lean/Elan/Tactic/NatCast.lean`: presumably if
we're doing this it will go somewhere in `Init`.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
When updating Std, be careful that not every lemma has been upstreamed,
so we need to be careful to only delete things that have already been
declared.
Before the `zeta` / `zetaDelta` split, `dsimp` was performing `zeta`
by going inside of a `let`-expression, performing `zetaDelta`, and
then removing the unused `let`-expression.
Loose fvars are never supposed to be pretty printed, but having them
print with "fvar" in the name can help with debugging broken tactics and
elaborators.
Metaprogramming users often do not realize at first that `_uniq.???` in
pretty printing output refers to fvars not in the current local context.
in all uses of `CasesOnApp`, we treat `MatcherApp`s the same way,
dupliating a fair amount of relatively hairy code (and there is more to
come).
However, the `MatcherApp` abstraction is perfectly capable of
also representing `casesOn` applications, at least for the use cases
encountered so far.
So lets just (optionally) include `casesOn` applications when looking
for matchers,
and remove the `CasesOnApp` abstraction completely.
Incrementally unveil trace children for excessively large nodes to
improve infoview rendering time, adjust particularly chatty
`simp.ground` trace to make use of it.
Lake previously opened the configuration trace as read-write even if it
does not update the configuration. This meant it failed if the trace was
read-only. With this change, it now first acquires a read-only handle
and then, if and only if it determines the need for a reconfigure, does
it re-open the file with a read-write handle. Also, this change fixes a
potential deadlock (Lake will error instead) and generally clarifies the
trace locking code.
Fixes a bug with Lake cloud releases where a cloud release would produce
a different trace if the package was the root of the workspace versus a
dependency. Also, an explicit fetch of a cloud release (e.g., via `lake
build :release`) will now error out with a non-zero exit code if it
fails to find, download, and unpack a release.
Previously, `CasesOn.addArg?` would do that check inline, while
`MatcherApp.addArg?` would do it after the fact.
Now `MatcherApp.addArg?` uses the same idiom.
Also, makes both `addArg?` always fail if the argument was not refined.
The work on functional induction principles calls for more unification
between the handling of `CasesOnApp` and `MatcherApp`, so this is a step
in that direction.
This is pretty big PR that upstreams all of Std.Data.Int.Init in one go.
So far lemmas have seen minimal changes needed to adapt to Lean core
environment.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Changes the goal to `False`, retaining as much information as possible:
* If the goal is `False`, do nothing.
* If the goal is an implication or a function type, introduce the
argument and restart.
(In particular, if the goal is `x ≠ y`, introduce `x = y`.)
* Otherwise, for a propositional goal `P`, replace it with `¬ ¬ P`
(attempting to find a `Decidable` instance, but otherwise falling back
to working classically)
and introduce `¬ P`.
* For a non-propositional goal use `False.elim`.
`nat?` checks if an expression is a "natural number in normal form",
i.e. of the form `OfNat n`, where `n` matches `.lit (.natVal n)` for
some `n`.
and if so returns `n`.
This is a widely used helper function in Std/Mathlib when matching on
expressions.
I've reordered some definitions to keep things together. This
introduces:
```
/-- Return the function (name) and arguments of an application. -/
def getAppFnArgs (e : Expr) : Name × Array Expr :=
withApp e λ e a => (e.constName, a)
```
and
```
/-- If the expression is a constant, return that name. Otherwise return `Name.anonymous`. -/
def constName (e : Expr) : Name :=
e.constName?.getD Name.anonymous
```
This upstreams NatCast and IntCast alone independent of norm_cast in
#3322.
This will allow more efficiently upstreaming parts of Std.Data.Int
relevant for omega.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
By having the `pp.proofs` feature use `⋯` when omitting proofs, when
users copy/paste terms from the InfoView the elaborator can give an
error message explaining why the term cannot be elaborated.
Also adds `pp.proofs.threshold` option to allow users to pretty print
shallow proof terms. By default, only atomic proof terms are pretty
printed.
This adjustment was suggested in PR #3201, which added `⋯` and the
related `pp.deepTerms` option.
This makes it so that when `withOverApp` is handling overapplied
functions, the term produced by the supplied delaborator is hoverable in
the Infoview.
When projection functions are delaborated, intermediate parent
projections are no longer printed. For example, rather than pretty
printing as `o.toB.toA.x` with these `toB` and `toA` parent projections,
it pretty prints as `o.x`.
This feature is being upstreamed from mathlib.
This is not a complete upstreaming of that file (it also supports `∀ᵉ (x
< 2) (y < 3), p x y` as shorthand for `∀ x < 2, ∀ y < 3, p x y`, but I
don't think we need this; it is used in Mathlib).
Syntaxes still need to be made built-in.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
This causes problems when used in conjunction with `#guard_msgs` (which
checks whitespace) and trailing whitespace removal. Discovered by
@PatrickMassot in verbose-lean4.
Again co-developed with @bollu.
Based on top of: #3225
While hunting down the performance discrepancy on qsort.lean between C
and LLVM we noticed there was a single, trivially optimizeable, alloca
(LLVM's stack memory allocation instruction) that had load/stores in the
hot code path. We then found:
https://groups.google.com/g/llvm-dev/c/e90HiFcFF7Y.
TLDR: `mem2reg`, the pass responsible for getting rid of allocas if
possible, only triggers on an alloca if it is in the first BB. The
allocas of the current implementation get put right at the location
where they are needed -> they are ignored by mem2reg.
Thus we decided to add functionality that allows us to push all allocas
up into the first BB.
We initially wanted to write `buildPrologueAlloca` in a `withReader`
style so:
1. get the current position of the builder
2. jump to first BB and do the thing
3. revert position to the original
However the LLVM C API does not expose an option to obtain the current
position of an IR builder. Thus we ended up at the current
implementation which resets the builder position to the end of the BB
that the function was called from. This is valid because we never
operate anywhere but the end of the current BB in the LLVM emitter.
The numbers on the qsort benchmark got improved by the change as
expected, however we are not fully there yet:
```
C:
Benchmark 1: ./qsort.lean.out 400
Time (mean ± σ): 2.005 s ± 0.013 s [User: 1.996 s, System: 0.003 s]
Range (min … max): 1.993 s … 2.036 s 10 runs
LLVM before aligning the types
Benchmark 1: ./qsort.lean.out 400
Time (mean ± σ): 2.151 s ± 0.007 s [User: 2.146 s, System: 0.001 s]
Range (min … max): 2.142 s … 2.161 s 10 runs
LLVM after aligning the types
Benchmark 1: ./qsort.lean.out 400
Time (mean ± σ): 2.073 s ± 0.011 s [User: 2.067 s, System: 0.002 s]
Range (min … max): 2.060 s … 2.097 s 10 runs
LLVM after this
Benchmark 1: ./qsort.lean.out 400
Time (mean ± σ): 2.038 s ± 0.009 s [User: 2.032 s, System: 0.001 s]
Range (min … max): 2.027 s … 2.052 s 10 runs
```
Note: If you wish to merge this PR independently from its predecessor,
there is no technical dependency between the two, I'm merely stacking
them so we can see the performance impacts of each more clearly.
Debugged and authored in collaboration with @bollu.
This PR fixes several performance regressions of the LLVM backend
compared to the C backend
as described in #3192. We are now at the point where some benchmarks
from `tests/bench` achieve consistently equal and sometimes ever so
slightly better performance when using LLVM instead of C. However there
are still a few testcases where we are lacking behind ever so slightly.
The PR contains two changes:
1. Using the same types for `lean.h` runtime functions in the LLVM
backend as in `lean.h` it turns out that:
a) LLVM does not throw an error if we declare a function with a
different type than it actually has. This happened on multiple occasions
here, in particular when the function used `unsigned`, as it was
wrongfully assumed to be `size_t` sized.
b) Refuses to inline a function to the call site if such a type mismatch
occurs. This means that we did not inline important functionality such
as `lean_ctor_set` and were thus slowed down compared to the C backend
which did this correctly.
2. While developing this change we noticed that LLVM does treat the
following as invalid: Having a function declared with a certain type but
called with integers of a different type. However this will manifest in
completely nonsensical errors upon optimizing the bitcode file through
`leanc` such as:
```
error: Invalid record (Producer: 'LLVM15.0.7' Reader: 'LLVM 15.0.7')
```
Presumably because the generate .bc file is invalid in the first place.
Thus we added a call to `LLVMVerifyModule` before serializing the module
into a bitcode file. This ended producing the expected type errors from
LLVM an aborting the bitcode file generation as expected.
We manually checked each function in `lean.h` that is mentioned in
`EmitLLVM.lean` to make sure that all of their types align correctly
now.
Quick overview of the fast benchmarks as measured on my machine, 2 runs
of LLVM and 2 runs of C to get a feeling for how far the averages move:
- binarytrees: basically equal performance
- binarytrees.st: basically equal performance
- const_fold: equal if not slightly better for LLVM
- deriv: LLVM has 8% more instructions than C but same wall clock time
- liasolver: basically equal performance
- qsort: LLVM is slower by 7% instructions, 4% time. We have identified
why the generated code is slower (there is a store/load in a hot loop in
LLVM that is not in C) but not figured out why that happens/how to
address it.
- rbmap: LLVM has 3% less instructions and 13% less wall-clock time than
C (woop woop)
- rbmap_1 and rbmap_10 show similar behavior
- rbmap_fbip: LLVM has 2% more instructions but 2% better wall time
- rbmap_library: equal if not slightly better for LLVM
- unionfind: LLVM has 5% more instructions but 4% better wall time
Leaving out benchmarks related to the compiler itself as I was too lazy
to keep recompiling it from scratch until we are on a level with C.
Summing things up, it appears that LLVM has now caught up or surpassed
the C backend in the microbenchmarks for the most part. Next steps from
our side are:
- trying to win the qsort benchmark
- figuring out why/how LLVM runs more instructions for less wall-clock
time. My current guesses would be measurement noise and/or better use of
micro architecture?
- measuring the larger benchmarks as well
We previously had the syntax for `change` and `change at`, but no
implementation.
This moves Kyle's implementation from Std.
This also changes the `changeLocalDecl` function to push nodes to the
infotree about FVar aliases.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
A small fix to the `DiscrTree` documentation to reflect the fact that
implicit type arguments *are* indexed and do not become `star` or
`other`. The following is a reproduction:
```lean
import Lean
open Lean Meta Elab Tactic
elab "test_tac" t:term : tactic => do
Tactic.withMainContext do
let e ← Term.elabTerm t none
let a : DiscrTree Nat ← DiscrTree.empty.insert e 1 {}
logInfo m!"{a}"
example (α : Type) (ringAdd : Add α) : True := by
/- (Add.add => (node (Nat => (node (* => (node (0 => (node (1 => (node #[1])))))))))) -/
test_tac @Add.add Nat instAddNat 0 1
/- (Add.add => (node (_uniq.1154 => (node (* => (node (◾ => (node (◾ => (node #[1])))))))))) -/
test_tac @Add.add α ringAdd ?_ ?_
```
This moves the `rcases` and `obtain` tactics from Std, and makes them
built-in tactics.
We will separately move the test cases from Std after #3297
(`guard_expr`).
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
Moves the `@[coe]` attribute and associated elaborators/delaborators
from Std to Lean.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
The `push_cast` tactic in Std currently uses a copy-paste version of
`mkSimpContext` that allows overriding `getSimpTheorems`. However it has
been diverging from the version in Lean.
This is one way of generalizing `mkSimpContext` in Lean to allow what is
needed downstream., but I'm not at all set on this one. As far as I can
see there are no other tactics currently using this.
`push_cast` itself just replaces `getSimpTheorems` with
`pushCastExt.getTheorems`, where `pushCastExt` is a simp extension. If
there is another approach that suits that situation it would be fine.
I've tested that the change in this PR works downstream.
This is used in the "Try this:" widget machinery powering `simp?`.
There is a test file in Std, which I am not upstreaming at the same
time, as that relies on more code actions / #guard_msgs material. That
test file will still of course test things from Std, and later it can be
reunited with the code it is testing.
---------
Co-authored-by: Leonardo de Moura <leomoura@amazon.com>
These additional options are currently implemented in Std in a function
`Format.prettyExtra` (via `open private`), and used to implement the
`simp?` functionality.
This just adds the options to the core function.
This does not completely empty `Std.Lean.Name`, as working out how to
document the difference between `Name.isInternalDetail` and
`Name.isImplementationDetail` requires further thought.
The induction principle used by `induction` may have explicit parameters
that are
not motive, target or “real” alternatives (that have the `motive` as
conclusion), e.g. restrictions on the `motive` or other parameters.
Previously, `induction` would treat them as normal alternatives, and try
to re-introduce the automatically reverted hypotheses. But this only
works when the `motive` is actually the conclusion in the type of that
alternative.
We now pay attention to that, thread that information through, and only
revert when needed.
Fixes#3212.
This PR adds links to some folder references in the docs, making them
easier to navigate.
Please advise if these need to be made to be full URIs rather than
relative paths in order to work correctly with the doc generation
tooling that is in place.
Implements the pretty printer option `pp.numericTypes` for including a
type ascription for numeric literals. For example, `(2 : Nat)`, `(-2 :
Int)`, and `(-2 / 3 : Rat)`. This is useful for debugging how arithmetic
expressions have elaborated or have been otherwise transformed. For
example, with exponentiation is is helpful knowing whether it is `x ^ (2
: Nat)` or `x ^ (2 : Real)`. This is like the Lean 3 option
`pp.numeralTypes` but it has a wider notion of a numeric literal.
Also implements the pretty printer option `pp.natLit` for including the
`nat_lit` prefix for raw natural number literals.
Closes#3021
When we declare a `simp` set using `register_simp_attr`, we
automatically create `simproc` set. However, users may create `simp`
sets programmatically, and the associated `simproc` set may be missing
and vice-versa.
Before this commit, `Simproc`s were defined as `Expr -> SimpM (Option Step)`, where `Step` is inductively defined as follows:
```
inductive Step where
| visit : Result → Step
| done : Result → Step
```
Here, `Result` is a structure containing the resulting expression and a proof demonstrating its equality to the input. Notably, the proof is optional; in its absence, `simp` assumes reflexivity.
A simproc can:
- Fail by returning `none`, indicating its inapplicability. In this case, the next suitable simproc is attempted, along with other simp extensions.
- Succeed and invoke further simplifications using the `.visit`
constructor. This action returns control to the beginning of the
simplification loop.
- Succeed and indicate that the result should not undergo further
simplifications. However, I find the current approach unsatisfactory, as it does not align with the methodology employed in `Transform.lean`, where we have the type:
```
inductive TransformStep where
/-- Return expression without visiting any subexpressions. -/
| done (e : Expr)
/--
Visit expression (which should be different from current expression) instead.
The new expression `e` is passed to `pre` again.
-/
| visit (e : Expr)
/--
Continue transformation with the given expression (defaults to current expression).
For `pre`, this means visiting the children of the expression.
For `post`, this is equivalent to returning `done`. -/
| continue (e? : Option Expr := none)
```
This type makes it clearer what is going on. The new `Simp.Step` type is similar but use `Result` instead of `Expr` because we need a proof.
Modifies the structure instance elaborator to
1. Fill in missing fields from sources in strict left-to-right order. In
`{a, b with}`, sometimes the elaborator
would ignore `a` even if both `a` and `b` provided the same field,
depending on what subobject fields they had.
2. Use the sources, or subobjects of the sources, to fill in entire
subobjects of the target structure as much as possible.
Currently, a field cannot be filled directly by a source itself
resulting in the term being eta expanded.
This change avoids this unnecessary and surprisingly costly extra eta
expansion.
Adds two new tests to illustrate the performance benefit (one courtesy
@semorrison). These are currently failing on master and succeed on this
branch.
There is one additional test to exercise the changes to the elaboration
of structure instances.
Changes to make mathlib build are in leanprover-community/mathlib4#9843
Closes#2451
This combines a few platform-related changes:
* Add a ternary `platformIndependent` Lean configuration option to
assert whether Lake should assume Lean code is platform-independent. If
`true`, Lake will exclude platform-independent objects like external
libraries or dynlibs created through `precompileModules` from module
traces. If `false`, Lake will add the platform to module traces. If
`none` (the default), Lake will retain the current behavior (modules are
platform-dependent if and only if it depends on native objects).
* Use `System.Platform.target` from #3207 as the platform descriptor in
Lake for the configuration file trace, the cloud release archive, and as
the platform trace in Lean modules and native artifacts (e.g., object
files, and static and shared libraries).
* Do not add the platform descriptor into custom build archive names
(i.e., a user-set `buildArchive` configuration). This allows users to
create cross-platform / platform-independent archives via a name
override should they so desire.
Closes#2754.
This replaces the no-op `unusedVariablesIgnoreFnsExt` environment
extension with an actual environment extension which can be extended
using either `@[unused_variables_ignore_fn]` or
`@[builtin_unused_variables_ignore_fn]` (although for the present all
the builtin `unused_variables_ignore_fn`s are being added using direct
calls to `builtin_initialize addBuiltinUnusedVariablesIgnoreFn`, because
this also works and a stage0 update is required before the attribute can
be used).
We would like to use this attribute to disable unused variables in
syntaxes defined in std and mathlib, like
[`proof_wanted`](https://leanprover.zulipchat.com/#narrow/stream/113488-general/topic/Unused.20variables.20and.20proof_wanted/near/408554690).
This PR adds two new delaboration settings: `pp.deepTerms : Bool`
(default: `true`) and `pp.deepTerms.threshold : Nat` (default: `20`).
Setting `pp.deepTerms` to `false` will make the delaborator terminate
early after `pp.deepTerms.threshold` layers of recursion and replace the
omitted subterm with the symbol `⋯` if the subterm is deeper than
`pp.deepTerms.threshold / 4` (i.e. it is not shallow). To display the
omitted subterm in the InfoView, `⋯` can be clicked to open a popup with
the delaborated subterm.
<details>
<summary>InfoView with pp.deepTerms set to false (click to show
image)</summary>
</details>
### Implementation
- The delaborator is adjusted to use the new configuration settings and
terminate early if the threshold is exceeded and the corresponding term
to omit is shallow.
- To be able to distinguish `⋯` from regular terms, a new constructor
`Lean.Elab.Info.ofOmissionInfo` is added to `Lean.Elab.Info` that takes
a value of a new type `Lean.Elab.OmissionInfo`.
- `ofOmissionInfo` is needed in `Lean.Widget.makePopup` for the
`Lean.Widget.InteractiveDiagnostics.infoToInteractive` RPC procedure
that is used to display popups when clicking on terms in the InfoView.
It ensures that the expansion of an omitted subterm is delaborated using
`explicit := false`, which is typically set to `true` in popups for
regular terms.
- Several `Info` widget utility functions are adjusted to support
`ofOmissionInfo`.
- The list delaborator is adjusted with special support for `⋯` so that
long lists `[x₁, ..., xₖ, ..., xₙ]` are shortened to `[x₁, ..., xₖ, ⋯]`.
this way this function does not have to peek at the `altType` to see
when there are no more arguments, which makes it a bit more explicit,
and also a bit more robust should one apply this function to the type of
an alternative with the motive already instantiated.
It seems this uncovered a variable shadow bug, where the counter `i` was
accidentially reset after removing the `i`’th entry in `ys`.
Adds support for `let_fun` to the `intro` and `intros` tactics. Also
adds support to `intro` for anonymous binder names, since the default
variable name for a `letFun` with an eta reduced body is anonymous.
Encouraged by the performance gains from making `rewrite` produce
smaller proof objects
(#3121) I am here looking for low-hanging fruit in `simp`.
Consider this typical example:
```
set_option pp.explicit true
theorem test
(a : Nat)
(b : Nat)
(c : Nat)
(heq : a = b)
(h : (c.add (c.add ((c.add b).add c))).add c = c)
: (c.add (c.add ((c.add a).add c))).add c = c
```
We get a rather nice proof term when using
```
:= by rw [heq]; assumption
```
namely
```
theorem test : ∀ (a b c : Nat),
@Eq Nat a b →
@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c →
@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c :=
fun a b c heq h =>
@Eq.mpr (@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c)
(@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c)
(@congrArg Nat Prop a b (fun _a => @Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c _a) c))) c) c) heq) h
```
(this is with #3121).
But with `by simp only [heq]; assumption`, it looks rather different:
```
theorem test : ∀ (a b c : Nat),
@Eq Nat a b →
@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c →
@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c :=
fun a b c heq h =>
@Eq.mpr (@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c)
(@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c)
(@id
(@Eq Prop (@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c)
(@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c))
(@congrFun Nat (fun a => Prop) (@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c))
(@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c))
(@congrArg Nat (Nat → Prop) (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c)
(Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) (@Eq Nat)
(@congrFun Nat (fun a => Nat) (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))))
(Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))))
(@congrArg Nat (Nat → Nat) (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c)))
(Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) Nat.add
(@congrArg Nat Nat (Nat.add c (Nat.add (Nat.add c a) c)) (Nat.add c (Nat.add (Nat.add c b) c)) (Nat.add c)
(@congrArg Nat Nat (Nat.add (Nat.add c a) c) (Nat.add (Nat.add c b) c) (Nat.add c)
(@congrFun Nat (fun a => Nat) (Nat.add (Nat.add c a)) (Nat.add (Nat.add c b))
(@congrArg Nat (Nat → Nat) (Nat.add c a) (Nat.add c b) Nat.add
(@congrArg Nat Nat a b (Nat.add c) heq))
c))))
c))
c))
h
```
Since simp uses only single-step `congrArg`/`congrFun` congruence lemmas
here, the proof
term grows very large, likely quadratic in this case.
Can we do better? Every nesting of `congrArg` (and it's little brother
`congrFun`) can be
turned into a single `congrArg` call.
In this PR I make making the smart app builders `Meta.mkCongrArg` and
`Meta.mkCongrFun` a bit
smarter and not only fuse with `Eq.refl`, but also with
`congrArg`/`congrFun`.
Now we get, in this simple example,
```
theorem test : ∀ (a b c : Nat),
@Eq Nat a b →
@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c →
@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c :=
fun a b c heq h =>
@Eq.mpr (@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c a) c))) c) c)
(@Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c b) c))) c) c)
(@congrArg Nat Prop a b (fun x => @Eq Nat (Nat.add (Nat.add c (Nat.add c (Nat.add (Nat.add c x) c))) c) c) heq) h
```
Let’s see if it works and how much we gain.
right now, the `induction` tactic accepts a custom eliminator using the
`using <ident>` syntax, but is restricted to identifiers. This
limitation becomes annoying when the elminator has explicit parameters
that are not targets, and the user (naturally) wants to be able to write
```
induction a, b, c using foo (x := …)
```
This generalizes the syntax to expressions and changes the code
accordingly.
This can be used to instantiate a multi-motive induction:
```
example (a : A) : True := by
induction a using A.rec (motive_2 := fun b => True)
case mkA b IH => exact trivial
case A => exact trivial
case mkB b IH => exact trivial
```
For this to work the term elaborator learned the `heedElabAsElim` flag,
`true` by default. But in the default setting, `A.rec (motive_2 := fun b
=> True)`
would fail to elaborate, because there is no expected type. So the
induction
tactic will elaborate in a mode where that attribute is simply ignored.
As a side effect, the “failed to infer implicit target” error message
is improved and prints the name of the implicit target that could not be
instantiated.
This PR adds support for the "call hierarchy" feature of LSP that allows
quickly navigating both inbound and outbound call sites of functions. In
this PR, "call" is taken to mean "usage", so inbound and outbound
references of all kinds of identifiers (e.g. functions or types) can be
navigated. To implement the call hierarchy feature, this PR implements
the LSP requests `textDocument/prepareCallHierarchy`,
`callHierarchy/incomingCalls` and `callHierarchy/outgoingCalls`.
<details>
<summary>Showing the call hierarchy (click to show image)</summary>
</details>
<details>
<summary>Incoming calls (click to show image)</summary>
</details>
<details>
<summary>Outgoing calls (click to show image)</summary>
</details>
It is based on #3159, which should be merged before this PR.
To route the parent declaration name through to the language server, the
`.ilean` format is adjusted, breaking backwards compatibility with
version 1 of the ILean format and yielding version 2.
This PR also makes the following more minor adjustments:
- `Lean.Server.findModuleRefs` now also combines the identifiers of
constants and FVars and prefers constant over FVars for the combined
identifier. This is necessary because e.g. declarations declared using
`where` yield both a constant (for usage outside of the function) and an
FVar (for usage inside of the function) with the same range, whereas we
would typically like all references to refer to the former. This also
fixes a bug introduced in #2462 where renaming a declaration declared
using `where` would not rename usages outside of the function, as well
as a bug in the unused variable linter where `where` declarations would
be reported as unused even if they were being used outside of the
function.
- The function converting `Lean.Server.RefInfo` to `Lean.Lsp.RefInfo`
now also computes the `Lean.DeclarationRanges` for parent declaration
names via `MetaM` and must hence be in `IO` now.
- Add a utility function `Array.groupByKey` to `HashMap.lean`.
- Stylistic refactoring of `Watchdog.lean` and `LanguageFeatures.lean`.
In the new snapshot design, we have a tree of `Task`s that represents
the asynchronously processed document structure. When transforming this
tree in response to a user edit, we want to quickly run through
reusable, already computed nodes of the tree synchronously and then
spawn new tasks for the new parts. The new flag allows us to do such
mixed sync/async tree transformations uniformly. This flag exists as
e.g.
[`ExecuteSynchronously`](https://learn.microsoft.com/en-us/dotnet/api/system.threading.tasks.taskcontinuationoptions?view=net-8.0)
in other runtimes.
I deleted internal links that seemed to have the character of "TODO". I
think that the residual TODO is of little value, given that we plan a
big revamp and revision soon anyway, but I could do it some other way as
well.
This makes changes to the definitions of Associativity, Commutativity,
Idempotence and Identity classes to be more aligned with Mathlib's
versions.
The changes are:
* Move classes are moved from `Lean` to root namespace.
* Drop `Is` prefix from names.
* Rename `IsNeutral` to `LawfulIdentity` and add Left and Right
subclasses.
* Change neutral/identity element to outParam.
* Introduce `HasIdentity` for operations not intended for proofs to
implement
The identity changes are to make this compatible with
[Mathlib](718042db9d/Mathlib/Init/Algebra/Classes.lean)
and to enable nicer fold operations in Std that can use type classes to
infer the identity/initial element on binary operations.
---------
Co-authored-by: Kyle Miller <kmill31415@gmail.com>
Makes the LLVM triple of the current platform available to Lean code
towards a solution for #2754.
Defaults to the empty string if the compiler is not clang, which can
introduce some divergence between CI and local builds but should not be
noticeable in most cases and is not really possible to avoid.
Recursive predefinitions contains “rec app” markers as mdata in the
predefinitions,
but sometimes these get in the way of termination checking, when you
have
```
[mdata (fun x => f)] arg
```
Therefore, the `preprocess` pass floats them out of applications
(originally
only for structural recursion, since #2818 also for well-founded
recursion).
But the code was incomplete: Because `Meta.transform` calls `post` on `f
x y` only
once (and not also on `f x`) one has to float out of nested applications
as well.
A consequence of this can be that in a recursive proof, `rw [foo]` does
not work
although `rw [foo _ _]` does.
Also adding the testcase where @david-christiansen and I stumbled over
this
(Maybe the two preprocess modules can be combined, now that #2973 is
landed, will try that
in a follow-up).
As suggested by @kmill, removing an unnecessary `let` (possibly only
there in the first place for copy/paste reasons) seems to fix the
included test.
This makes `~q()` matching in quote4 noticeably more useful in things
like `norm_num` (as it fixes
https://github.com/leanprover-community/quote4/issues/29)
It also makes a quote4 bug slightly more visible
(https://github.com/leanprover-community/quote4/issues/30), but the bug
there already existed anyway, and isn't caused by this patch.
Fixes#3065
Give n-ary `Expr.app` constructors such as `mkApp2`, `mkApp3`, ...,
`mkApp10` the `@[match_pattern]` attribute so that it is easier to read
and write pattern matching for applications.
This PR facilitates augmenting the context of an `InfoTree` with
*partial* contexts while elaborating a command. Using partial contexts,
this PR also adds support for tracking the parent declaration name of a
term in the `InfoTree`. The parent declaration name is needed to compute
the call hierarchy in #3082.
Specifically, the `Lean.Elab.InfoTree.context` constructor is refactored
to take a value of the new type `Lean.Elab.PartialContextInfo` instead
of a `Lean.Elab.ContextInfo`, which now refers to a full `InfoTree`
context. The `PartialContextInfo` is then merged into a `ContextInfo`
while traversing the tree using
`Lean.Elab.PartialContextInfo.mergeIntoOuter?`. The partial context
after executing `liftTermElabM` is stored in values of a new type
`Lean.Elab.CommandContextInfo`.
As a result of this, `Lean.Elab.ContextInfo.save` moves to
`Lean.Elab.CommandContextInfo.save`.
For obtaining the parent declaration for a term, a new typeclass
`MonadParentDecl` is introduced to save the parent declaration in
`Lean.Elab.withSaveParentDeclInfoContext`. `Lean.Elab.Term.withDeclName
x` now calls `withSaveParentDeclInfoContext x` to save the declaration
name.
### Migration
**The changes to the `InfoTree.context` constructor break backwards
compatibility with all downstream users that traverse the `InfoTree`
manually instead of going through the functions in `InfoUtils.lean`.**
To fix this, you can merge the outer `ContextInfo` in a traversal with
the `PartialContextInfo` of an `InfoTree.context` node using
`PartialContextInfo.mergeIntoOuter?`. See e.g.
`Lean.Elab.InfoTree.foldInfo` for an example:
```lean
partial def InfoTree.foldInfo (f : ContextInfo → Info → α → α) (init : α) : InfoTree → α :=
go none init
where go ctx? a
| context ctx t => go (ctx.mergeIntoOuter? ctx?) a t
| node i ts =>
let a := match ctx? with
| none => a
| some ctx => f ctx i a
ts.foldl (init := a) (go <| i.updateContext? ctx?)
| _ => a
```
Downstream users that manually save `InfoTree`s may need to adjust calls
to `ContextInfo.save` to use `CommandContextInfo.save` instead and
potentially wrap their `CommandContextInfo` in a
`PartialContextInfo.commandCtx` constructor when storing it in an
`InfoTree` or `ContextInfo.mk` when creating a full context.
### Motivation
As of now, `ContextInfo`s are always *full* contexts, constructed as if
they were always created in `liftTermElabM` after running the
`TermElabM` action. This is not strictly true; we already create
`ContextInfo`s in several places other than `liftTermElabM` and work
around the limitation that `ContextInfo`s are always full contexts in
certain places (e.g. `Info.updateContext?` is a crux that we need
because we can't always create partial contexts at the term-level), but
it has mostly worked out so far. Note that one must be very careful when
saving a `ContextInfo` in places other than `liftTermElabM` because the
context may not be as complete as we would like (e.g. it may lack
meta-variable assignments, potentially leading to a language server
panic).
Unfortunately, the parent declaration of a term is another example of a
context that cannot be provided in `liftTermElabM`: The parent
declaration is usually set via `withDeclName`, which itself lives in
`TermElabM`. So by the time we are trying to save the full
`ContextInfo`, the declaration name is already gone. There is no easy
fix for this like in the other cases where we would really just like to
augment the context with an extra field.
The refactor that we decided on to resolve the issue is to refactor the
`InfoTree` to take a `PartialContextInfo` instead of a `ContextInfo` and
have code that traverses the `InfoTree` merge inner contexts with outer
contexts to produce a full `ContextInfo` value.
### Bumps for downstream projects
- `lean-pr-testing-3159` branch at Std, not yet opened as a PR
- `lean-pr-testing-3159` branch at Mathlib, not yet opened as a PR
- https://github.com/leanprover/LeanInk/pull/57
- https://github.com/hargoniX/LeanInk/pull/1
- https://github.com/tydeu/lean4-alloy/pull/7
- https://github.com/leanprover-community/repl/pull/29
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Currently we create `lean-pr-testing-NNNN` branches at Mathlib
automatically for each Lean PR.
We don't automatically create one at Std; mostly simply because Std
fails less often, so it has been okay to do this manually as needed. It
is conceptually simpler, however, if this is done uniformly.
This PR:
* does not proceed with Std/Mathlib CI unless the appropriate
`nightly-testing-YYYY-MM-DD` tag exists at Std (like it already doesn't
proceed if that tag is missing at Mathlib)
* creates `lean-pr-testing-NNNN` branches at Std
* when it creates `lean-pr-testing-NNNN` branches at Mathlib, updates
the Std dependency to use the `lean-pr-testing-NNNN` branch at Std
- [x] depends on #3199
Note that because most users do not have write access at Std, in order
to make updates to `lean-pr-testing-NNNN` branches there they will need
to make PRs. These will be merged with a very low bar, and feel free to
ping me for assistance on this. If this is annoying we will automate.
Also, frequent contributors to Lean may ask @digama0 or @joehendrix for
write access in order to easily work on these branches.
This PR requires that we have a secret here with write access at Std.
I'm arranging that [on
zulip](https://leanprover.zulipchat.com/#narrow/stream/348111-std4/topic/bot.20access/near/416686090).
I will update the documentation at
https://leanprover-community.github.io/contribute/tags_and_branches.html
to reflect these changes when they are merged.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
As discussed during the FRO meeting 2024-01-18, we are changing the
`nightly-testing-YYYY-MM-DD` branches at Std and Mathlib from branches
to tags, in:
* https://github.com/leanprover/std4/pull/545
* https://github.com/leanprover-community/mathlib4/pull/9842
This PR updates the script that creates the `lean-pr-testing-NNNN`
branches at Mathlib so it is agnostic about whether
`nightly-testing-YYYY-MM-DD` will be a branch or a tag.
---------
Co-authored-by: Joachim Breitner <mail@joachim-breitner.de>
`Array.set!` and `Array.swap!` are fairly similar operations, both
modify an array, both take an index that it out of bounds.
But they behave different; all of these return `true`
```
#eval #[1,2].set! 2 42 == #[1,2] -- with panic
#reduce #[1,2].set! 2 42 == #[1,2] -- no panic
#eval #[1,2].swap! 0 2 == #[1,2] -- with panic
#reduce #[1,2].swap! 0 2 == default -- no panic
```
The implementations are
```
@[extern "lean_array_set"]
def Array.set! (a : Array α) (i : @& Nat) (v : α) : Array α :=
Array.setD a i v
```
but
```
@[extern "lean_array_swap"]
def swap! (a : Array α) (i j : @& Nat) : Array α :=
if h₁ : i < a.size then
if h₂ : j < a.size then swap a ⟨i, h₁⟩ ⟨j, h₂⟩
else panic! "index out of bounds"
else panic! "index out of bounds"
```
It seems to be more consistent to unify the behaviors, and define
```
@[extern "lean_array_swap"]
def swap! (a : Array α) (i j : @& Nat) : Array α :=
if h₁ : i < a.size then
if h₂ : j < a.size then swap a ⟨i, h₁⟩ ⟨j, h₂⟩
else a
else a
```
Also adds docstrings.
Fixes#3196
Consider
```
import Std.Tactic.ShowTerm
opaque a : Nat
opaque b : Nat
axiom a_eq_b : a = b
opaque P : Nat → Prop
set_option pp.explicit true
-- Using rw
example (h : P b) : P a := by show_term rw [a_eq_b]; assumption
```
Before, a typical proof term for `rewrite` looked like this:
```
-- Using the proof term that rw produces
example (h : P b) : P a :=
@Eq.mpr (P a) (P b)
(@id (@Eq Prop (P a) (P b))
(@Eq.ndrec Nat a (fun _a => @Eq Prop (P a) (P _a))
(@Eq.refl Prop (P a)) b a_eq_b))
h
```
which is rather round-about, applying `ndrec` to `refl`. It would be
more direct to write
```
example (h : P b) : P a :=
@Eq.mpr (P a) (P b)
(@id (@Eq Prop (P a) (P b))
(@congrArg Nat Prop a b (fun _a => (P _a)) a_eq_b))
h
```
which this change does.
This makes proof terms smaller, causing mild general speed up throughout
the code; if the brenchmarks don’t lie the highlights are
* olean size -2.034 %
* lint wall-clock -3.401 %
* buildtactic execution s -10.462 %
H'T to @digama0 for advice and help.
NB: One might even expect the even simpler
```
-- Using the proof term that I would have expected
example (h : P b) : P a :=
@Eq.ndrec Nat b (fun _a => P _a) h a a_eq_b.symm
```
but that would require non-local changes to the source code, so one step
at a time.
The `checkTargets` function introduced in 4a0f8bf2 as
```
checkTargets (targets : Array Expr) : MetaM Unit := do
let mut foundFVars : FVarIdSet := {}
for target in targets do
unless target.isFVar do
throwError "index in target's type is not a variable (consider using the `cases` tactic instead){indentExpr target}"
if foundFVars.contains target.fvarId! then
throwError "target (or one of its indices) occurs more than once{indentExpr target}"
```
looks like it tries to check for duplicate indices, but it doesn’t
actually, as `foundFVars` is never written to.
This adds
```
foundFVars := foundFVars.insert target.fvarId!
```
and a test case.
Maybe a linter that warns about `let mut` that are never writen to would
be useful?
I keep messing things up, so time for some guard rails, so check them
using
[actionlint](https://github.com/raven-actions/actionlint).
This also runs [shellcheck](https://www.shellcheck.net/) on the files.
Shellcheck
is a bit picky about putting double quotes around variables, and will
flag many
cases where we know it’s safe, but why not simply always write the safer
variant.
Unfortunately, actionlint does not (yet) check `actions/github-script`
scripts, which is
unfortunate. Maybe they will in the future
(https://github.com/rhysd/actionlint/issues/389)
there was a check
if !Structural.recArgHasLooseBVarsAt recFnName fixedPrefixSize e then
that would avoid going through `.refineThrough`/`.addArg` for
matcher/casesOn applications. It seems it tries to detect when refining
the motive/param is pointless, but it was too eager, and cause confusion
with, for example, this reasonably reasonable function:
def foo : (n : Nat) → (i : Fin n) → Bool
| 0, _ => false
| 1, _ => false
| _+2, _ => foo 1 ⟨0, Nat.zero_lt_one⟩
decreasing_by simp_wf; simp_arith
In particular, the `GuessLex` code later expects that the (implict)
`PProd.casesOn` in the implementation of `foo._unary` will refine the
paramter, because else the (rather picky) `unpackArg` fails. But it also
prevents this from being provable.
So let's try without this shortcut.
Fixing this also revealed that `withRecApps` wasn’t looking in all
corners
of a matcherApp/casesOnApp.
Fixes#3175
this didn’t recognize the new comments with an intro, and thus the bot
would post multiple comments.
The code was also out of sync with mathlib, fixing.
The `first(…)` in the `jq` program makes it more robust in case this
went wrong once (as on #3171) and there are now multiple PRs matching.
This uses the improved termination_by syntax to give Nat.gcd a cleaner
definition. It removes the last explicit use of WellFounded.fix in Init.
This was also partly motivated by leanprover/std4#520 so that unfold
Nat.gcd gives a sensible definition.
If the current manifest is from unsupported (or has errors), a bare
`lake update` will now discard it and create a new one from scratch
rather than erroring and requiring you to manually delete the manifest.
Lake will produce warnings noting it is ignoring such invalid manifests.
so far, our reference manual did not mention these at all, this takes
the discussion of recursive definition out of the “equation compiler”
section, put it into its own section, and expands it a bit.
This is more a MVP doc change to at least mention the features briefly,
and not the most polished and thought through didactic exposition. But
it provides a start for more improvements.
---------
Co-authored-by: Arthur Adjedj <arthur.adjedj@gmail.com>
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
This change
* moves `termination_by` and `decreasing_by` next to the function they
apply to
* simplify the syntax of `termination_by`
* apply the `decreasing_by` goal to all goals at once, for better
interactive use.
See the section in `RELEASES.md` for more details and migration advise.
This is a hard breaking change, requiring developers to touch every
`termination_by` in their code base. We decided to still do it as a
hard-breaking change, because supporting both old and new syntax at the
same time would be non-trivial, and not save that much. Moreover, this
requires changes to some metaprograms that developers might have
written, and supporting both syntaxes at the same time would make
_their_ migration harder.
To handle delaborating notations that are functions that can be applied
to arguments, extracts the core function application delaborator as a
separate function that accepts the number of arguments to process and a
delaborator to apply to the "head" of the expression.
Defines `withOverApp`, which has the same interface as the combinator of
the same name from std4, but it uses this core function application
delaborator.
Uses `withOverApp` to improve a number of application delaborators,
notably projections. This means Mathlib can stop using `pp_dot` for
structure fields that have function types.
Incidentally fixes `getParamKinds` to specialize default values to use
supplied arguments, which impacts how default arguments are delaborated.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Allow `simproc`s to be declared without setting the `[simproc]`
attribute. A `simproc` declaration is function + pattern.
Motivation: allow them to be provided as arguments to `simp` **and** `simp only`.
TODO: track their use in `simp`.
TODO: builtin simprocs
Motivations:
- We can simplify the big mutual recursion and the implementation.
- We can implement the support for `match`-expressions in the `pre` method.
- It is easier to define and simplify `Simprocs`.
The example was looping with the new `simp` reduction strategy. Here
is the looping trace.
```
List.reverseAux (List.reverseAux as []) bs
==> rewrite using reverseAux_reverseAux
List.reverseAux [] (List.reverseAux (List.reverseAux as []) bs)
==> unfold reverseAux
List.reverseAux (List.reverseAux as []) bs
==> rewrite using reverseAux_reverseAux
List.reverseAux [] (List.reverseAux (List.reverseAux as []) bs)
==> ...
```
See new test for example that takes exponential time without new simp
theorems.
TODO: replace auxiliary theorems with simprocs as soon as we implement them.
I was about to to address the TODO
/- TODO: check arity of the given function. If it takes a PSigma as the
last argument,
this function will produce incorrect results. -/
because we now have an arity-observing variant of `decodePackedArg?` in
`unpackArg` in `PackMutual`, and it would be prudent to use it here.
But I first wanted to create a test case that would actually exhibit
this corner case, and failed.
This code was added in 096e4eb6d0 and it had a test case, but not even
that test case seems to be actually using the `decodePackedArg?`
function, neither back then nor now.
Also, mathlib works without this code.
So this seems to be dead code, possibly due to other changes to the
system, and thus can be removed. A strategically place comments points
back to this PR in case we need to resurrect that code.
The pattern
```
for h : i in [:xs.size] do
let x := xs[i]'h.2
```
is occassionally useful to iterate over an array with the index in
hand. This PR extends the `get_elem_tactic_trivial` so that one can
simply write
```
for h : i in [:xs.size] do
let x := xs[i]
```
fixes#3032.
When looking at a PR I sometimes wonder which `nightly` release is this
based on, and is used for the mathlib testing.
Right now, the action uses a label (`toolchain-available`) for this, but
a label cannot easily carry more information.
It seems a rather simple way to communicate extra information is by
setting [commit
statuses](https://docs.github.com/en/rest/commits/statuses?apiVersion=2022-11-28#create-a-commit-status);
with this change the following statuses will appear in the PR:
One could also use
[checks](https://docs.github.com/en/rest/checks/runs?apiVersion=2022-11-28#create-a-check-run)
to add more information, even with a nicely formatted markdown
description as in [this
example](https://github.com/nomeata/lean4/pull/1/checks?check_run_id=20165137082),
but it seems there you can’t set a summary that’s visible without an
extra click, and Github seems to associate these checks to “the first
workflow”, which is odd. So using statuses seems fine here.
Often one uses bots writing PR comments for this purpose, but that's a
bit noisy (extra notifications etc.), especially for stuff that happens
on every PR, but isn’t always interesting/actionable
If this works well, we can use this for more pieces of information, and
a link can be added as well.
This removes checks in `Lean.Meta.reduceNat?` that caused it to fail on
terms it could handle because they contain meta variables in arguments.
This lead to those operations being reduced using their equational
definitions and slow performance on large patterns:
```
set_option profiler true
set_option profiler.threshold 1
def testMod (x:Nat) :=
match x with
| 128 % 1024 => true
| _ => false
-- elaboration took 3.02ms
def testMul (x:Nat) :=
match x with
| 128 * 1 => true
| _ => false
-- type checking took 11.1ms
-- compilation of testMul.match_1 took 313ms
-- compilation of testMul took 65.7ms
-- elaboration took 58.9ms
```
Performance is slower on `testMul` than `testMod` because `whnf` ends up
evaluateing `128 * 1` using Peano arithmetic while `128 % 1024` is able
to avoid that treatment since `128 < 1024`.
This makes hover info, go to definition, etc work for the `h` in `cases
h : e`. The implementation is similar to that used for the `generalize h
: e = x` tactic.
#2966 was the `@[extern]` bug that prompted development of the
`test_extern` command, but then we merged the fix to #2966 without
updating the tests to use `test_extern`.
# Summary
This makes a small addition to our take on the LSP protocol
in the form of supporting snippet text edits.
It has been discussed
[here](https://github.com/microsoft/language-server-protocol/issues/592)
on the LSP issue tracker for a while,
but seems unlikely to be added anytime soon.
This feature was requested by @PatrickMassot for the purposes
of supporting Lean code templates in code actions and widgets.
---------
Co-authored-by: David Thrane Christiansen <david@davidchristiansen.dk>
This PR fixes the documentation error in "Extended Setup Notes", where
the path of builded binary is pointed to
`./build/bin/foo`, but the truly path is `./lake/build/bin/foo`.
---
Closes#3094 (`RFC` or `bug` issue number fixed by this PR, if any)
There were no `quot_precheck` instances registered for the expression
tree elaborators, which prevented them from being usable in a `notation`
expansion without turning off the quotation prechecker.
Users can evaluate whether `set_option quotPrecheck false` is still
necessary for their `notation` definitions.
given that we now use the PR description as the commit message, the PR
template should point that out. Also, a `# Summary` is relatively
strange in a commit message, so removed it.
---------
Co-authored-by: Scott Morrison <scott.morrison@gmail.com>
Switches from encoding `let_fun` using an annotated `(fun x : t => b) v`
expression to a function application `letFun v (fun x : t => b)`.
---------
Co-authored-by: Sebastian Ullrich <sebasti@nullri.ch>
Getting the original PR number from a `workflow_run` cleanly and
reliably seems to be
basically impossible. See
<https://github.com/orgs/community/discussions/25220> for a discussion.
So for now let’s go back to the working state, even though it’s
deprecated and throws warnings.
This prompts users opening the workspace (on a new device) for the first
time to install the lean extension
# Summary
Link to `RFC` or `bug` issue: N/A
#3066 is causing CI failures, e.g.
[here](https://github.com/leanprover/lean4/actions/runs/7202184616/job/19619827364).
Although there are plenty of examples of using `await` in a Github
workflow script block, the error *seems* to be about this. This refactor
hopefully works around that, but I'm still uncertain of a root cause.
This fixe a surprisingly embarrassing bug introduced by me in
fa26d222cb (maybe while testing).
Enable more debug output while we are at it, to find out why sometimes
`context.payload.workflow_run.pull_requests[0]` is undefined.
If a user deleted `lakefile.olean` manually without deleting
`lakefile.olean.lock`, Lake would still attempt to load it and thus
produce an error. Now it should properly re-elaborate the configuration
file.
This adds a `test_extern` command.
Usage:
```
import Lean.Util.TestExtern
test_extern Nat.add 17 37
```
This:
* Checks that the head symbol has an `@[extern]` attribute.
* Writes down `t == t'`, where `t` is the term provided, and `t'` is the
reference implementation (specifically, `t` with the head symbol
unfolded).
* Tries to reduce this to `true`, and complains if this fails.
Note that the type of the term must have a `BEq` instance for this to
work: there's a self-explanatory error message if it isn't available.
the workflow is triggered not only by pull-request-CI-runs but also by
others. These should be skipped.
Also, no need to query the Github API to get the pull request number and
head sha, they are part of the payload, it seems.
Since the vscode-lean4 setup guide allows us to provide information on
setting up Lean 4 tailored to the user's operating system, this PR
adjusts the quickstart guide to reference the vscode-lean4 setup guide
instead.
This definition was clearly meant to be in the `List` namespace, but it
is also in a `namespace Lean` so it ended up as `Lean.List.toSMap`
instead of `List.toSMap`. It would be nice if #3031 made this
unnecessary, but for now this seems to be the convention.
I noticed this because of another side effect: it defines `Lean.List` as
a namespace, which means that
```lean
import Std
namespace Lean
open List
#check [1] <+ [2]
```
does not work as expected, it opens the `Lean.List` namespace instead of
the `List` namespace. Should there be a regression test to ensure that
the `Lean.List` namespace (and maybe others) are not accidentally
created? (Unfortunately this puts a bit of a damper on #3031.)
In particular:
* Do not use deprecated `potiuk/get-workflow-origin`.
* Use a bare checkout to push PR to `pr-releases`
* Replace `script/most-recent-nightly-tag.sh` by a one-liner inside the
workflow, so that th workflow is self-contained
Fixes an issue reported on Zulip; see the test case.
* Modifies the `MonadBacktrack` instance for `SimpM` to also backtrack
the `UsedSimps` field.
* When calling the discharger, `saveState`, and then `restoreState` if
something goes wrong.
I'm not certain that it makes sense to restore the `MetaM` state if
discharging fails. I can easily change this to more conservatively just
backtrack the `UsedSimps` after failed discharging.
Changes the implementation of `List.all` and `List.any` so they
short-circuit. The implementations are tail-recursive.
This replaces https://github.com/leanprover/std4/pull/392, which was
going to do this with `@[csimp]`.
until around 7fe6881 the way to define well-founded recursions was to
specify a `WellFoundedRelation` on the argument explicitly. This was
rather low-level, for example one had to predict the packing of multiple
arguments into `PProd`s, the packing of mutual functions into `PSum`s,
and the cliques that were calculated.
Then the current `termination_by` syntax was introduced, where you
specify the termination argument at a higher level (one clause per
functions, unpacked arguments), and the `WellFoundedRelation` is found
using type class resolution.
The old syntax was kept around as `termination_by'`. This is not used
anywhere in the lean, std, mathlib or the theorem-proving-in-lean
repositories,
and three occurrences I found in the wild can do without
In particular, it should be possible to express anything that the old
syntax
supported also with the new one, possibly requiring a helper type with a
suitable instance, or the following generic wrapper that now lives in
std
```
def wrap {α : Sort u} {r : α → α → Prop} (h : WellFounded r) (x : α) : {x : α // Acc r x}
```
Since the old syntax is unused, has an unhelpful name and relies on
internals, this removes the support. Now is a good time before the
refactoring that's planned in #2921.
The test suite was updated without particular surprises.
The parametric `terminationHint` parser is gone, which means we can
match on syntax more easily now, in `expandDecreasingBy?`.
This Github action automatically updates `stage0` on `master` if
`src/stdlib_flags.h` and `stage0/src/stdlib_flags.h`
are out of sync there.
It bypasses the merge queue to be quick, this way, an out-of-date stage0
on on
master should only exist for a few minutes.
Needs access to a _deploy SSH key_ with write permission.
The elaborator is prone to duplicate terms, including recursive calls,
even if the user only wrote a single one. This duplication is wasteful
if we run the tactics on duplicated calls, and confusing in the output
of GuessLex. So prune the list of recursive calls, and remove those
where another call exists that has the same goal and context that is no
more specific.
Implements "gaps" in string literals. These are escape sequences of the
form `"\" newline whitespace+` that have the interpretation of an empty
string. For example,
```
"this is \
a string"
```
is equivalent to `"this is a string"`. These are modeled after string
continuations in
[Rust](https://doc.rust-lang.org/beta/reference/tokens.html#string-literals).
Implements RFC #2838
before code like
def dup (a : Nat) (b : Nat := a) := a + b
def rec : Nat → Nat
| 0 => 1
| n+1 => dup (dup (dup (rec n)))
decreasing_by decreasing_tactic
would run the `decreasing_tactic` 8 tims, because the recursive call
`rec n` gets duplicate due to the default paramter. Similar effects can
be observed due to dependent types or tactics like `cases`.
This is wasteful, and is confusing to the user when they use
`decreasing_by` interactively. Therfore, we now go through the proof
obligations (MVars) and if solving one would imply solving another one,
we assign the mvars to each other accordingly.
This PR is a sibling of #3004.
I found the documentation page hard to parse, so I figured I should fix
this. It's mostly indentation (e.g. in lists), some line breaks and
making URLs clickable.
With
set_option showInferredTerminationBy true
this prints a message like
Inferred termination argument:
termination_by
ackermann n m => (sizeOf n, sizeOf m)
it tries hard to use names that
* match the names that the user used, if present
* have no daggers (so that it can be copied)
* do not shadow each other
* do not shadow anything from the environment (just to be nice)
it does so by appending sufficient `'` to the name.
Some of the emitted `sizeOf` calls are unnecessary, but they are needed
sometimes with dependent parameters. A follow-up PR will not emit them
for non-dependent arguments, so that in most cases the output is pretty.
Somewhen down the road we also want a code action, maybe triggered by
`termination_by?`. This should come after #2921, as that simplifies that
feature (no need to merge termination arguments from different cliques
for example.)
by showing the matrix of calls and measures, and what we know about that
call (=, <, ≤, ?), e.g.
guessLexFailures.lean:27:0-33:31: error: Could not find a decreasing
measure.
The arguments relate at each recursive call as follows:
(<, ≤, =: relation proved, ? all proofs failed, _: no proof attempted)
x1 x2 x3
1) 29:6-25 = = =
2) 30:6-23 = ? <
3) 31:6-23 < _ _
Please use `termination_by` to specify a decreasing measure
It’s a bit more verbose for mutual functions.
It will use the user-specified argument names for functions written
```
foo (n : Nat) := …
```
but not with pattern matching like
```
foo : Nat → …
| n => …
```
This can be refined later and separately (and maybe right away in
`expandMatchAltsWhereDecls`).
This is pure refactoring: Instead of solving each subgoal as we
encounter it while traversing the syntax tree, we leave the `MVar`
there, at the end collect them all using `getMVarsNoDelayed`, and then
solve them.
This is a refactoring preparing for two upcoming changes:
* removing unexpected duplicate goals that can arise from term
duplication
* running interactive tactics on all, not each goal (#2921)
In order to not regress with error locations, we have to associated the
`TermElabM`’s syntax refernce with the `MVar` somehow. I do this using
the existing `mkRecAppWithSyntax` expression annotation, on the `MVar`’s
type. Alternatives would be stack another `StateT` on the traversal
and accumulate `Array (MVarId, Syntax)` explicitly, but that did not
seem to be more appealing.
This sets the build directory to `build/release` for the "CMake Tools
for Visual Studio Code" extension documented at
https://vector-of-bool.github.io/docs/vscode-cmake-tools/settings.html#cmake-builddirectory.
It also sets the generator to `make`, since otherwise it tries `Ninja`
which doesn't work.
Without these settings, the extension runs configure in a bad place at
startup.
This does *not* add the cmake tools extension to the default workspace
configuration; the goal is simply to prevent bad behavior for users who
already have the extension enabled.
# Summary
Screenshot of this in action:
Link to `RFC` or `bug` issue: N/A, this is not a bug nor a user-visible
feature.
Now that we're, at least temporarily, relying more on the Nix CI,
replace some old hacks of mine with better solutions people have figured
out in the meantime.
Cachix support could probably be dropped at this point but it doesn't
really hurt.
In order to familiarize myself with this code, and so that the next
person has an easier time, I
* added docstrings explaining what I found out these things to
* rewrote the syntax expansion functions using syntax pattern matches,
to the extend possible
else I see
```
[ 69%] Building CXX object runtime/CMakeFiles/leanrt.dir/platform.cpp.o
/home/jojo/build/lean/lean4/src/runtime/io.cpp:509:75: warning: 'static_assert' with no message is a C++17 extension [-Wc++17-extensions]
static_assert(sizeof(std::chrono::milliseconds::rep) <= sizeof(uint64));
^
, ""
/home/jojo/build/lean/lean4/src/runtime/io.cpp:517:74: warning: 'static_assert' with no message is a C++17 extension [-Wc++17-extensions]
static_assert(sizeof(std::chrono::nanoseconds::rep) <= sizeof(uint64));
^
, ""
2 warnings generated.
```
when building
CI will now run on _any_ manually added label; hard to avoid.
Fun fact: Because the `toolchain-available` label is added by a github
action with the default token, it will _not_ trigger the workflow. Lucky
coincidence.
Following up on #2986, stop running the test suite in ci.yml in quick
mode; the test suite is run in the Nix job, and we do not need to run it
twice.
With a cold nix cache, when `lean` is rebuilt, not much changes, as both
jobs take ~20mins. But when `lean` is unchanged, the nix build should
be faster, and shaving off the (currently) 4mins in the CI.yaml run
should get us to a green PR sooner.
Another benefit is that we get the PR release sooner and even get it
when the test suite fails, which can be useful if you want to test
mathlib or other things before fixing the lean test suite.
which also removes an error condition at the use site.
While I am at it, I rename a parameter in `GuessLex` that I forgot to
rename earlier.
The effect will be user-visible (in obscure corner cases) with #2960, so
I’ll have the test there.
A few places would benefit from a `lambdaTelescopeBounded` that
garantees the result has the right length (eta-expanding when
necessary). I’ll look into that separately, and left TODOs here.
The goal of this change is to run a trimmed-down CI on PRs by default,
but allows opt-in the full CI as necessary.
### Specification
The CI workflow runs in “quick” mode if it was triggered from a pull
request, and that pull request does not have the `full-ci` label set.
In “quick” mode the build matrix contains fewer jobs. At the moment
only:
* Linux-release, to get the PR releases.
In non-quick mode everything should be as before.
### Implementation notes
I created a `configure` job that combines all the previous `set-` jobs,
I guess this is faster than firing up separate jobs.
The matrix is calculated in this job; this seems to be the cleanest way
to get a dynamic matrix going (experiments using `exclude` failed). The
downside is that the matrix is now in JSON rather than Yaml syntax. The
upside is that we can (later) make it’s calculation simpler, e.g. set
default `shell` values etc.
I was not able to make it so that CI runs when the `full-ci` label is
added, but don’t do anything otherwise. I think it can be done with
another workflow listening to `labeled` and then triggering this one,
but let’s do that separately. For now, add the label and then push (or
close and reopen).
The checks
```
if: matrix.build-stage2 || matrix.check-stage3
if: matrix.check-stage3
```
were dead code, we did not have these fields in the matrix anymore, so I
replaced them with
```
if: matrix.test-speedcenter
```
Now that there is a helpful message at the point of use when
`supportInterpreter` is required, we don't need to clutter every
`lakefile` with the advice.
there is a little dance with `if: success()` because otherwise a failed
`build` job would make this new job skipped, not failed, and I fear
skipped means ok when it is a required job.
So let’s make sure this job actually fails.
Also turn this into a proper check, run when a PR is opened or edited.
I took the liberty to rename the workflow file and name, so that one
doesn't have to look inside to guess what the workflow is doing.
If here is only one plausible measure, there is no point having the
`GuessLex` code see if it
is terminating, running all the tactics, only for the `MkFix` code then
run the tactics again.
So if there is only one plausible measure (non-mutual recursion with
only one varying
parameter), just use that measure.
Side benefit: If the function isn’t terminating, more detailed error
messages are shown
(failing proof goals), located at the recursive calls.
Removes the `CI` option from the `math` template. Since the template
does not currently generate a GitHub workflow, it does not do anything
out of the box except add unnecessary complexity.
The `math` template is also now tested in `tests/init` (minus the
Mathlib `require`).
This improves Lean’s capabilities to guess the termination measure for
well-founded
recursion, by also trying lexicographic orders. For example:
def ackermann (n m : Nat) := match n, m with
| 0, m => m + 1
| .succ n, 0 => ackermann n 1
| .succ n, .succ m => ackermann n (ackermann (n + 1) m)
now just works.
The module docstring of `Lean.Elab.PreDefinition.WF.GuessLex` tells the
technical story.
Fixes#2837
Closes#2548.
Later packages and libraries in the dependency tree are now preferred
over earlier ones. That is, the later ones "shadow" the earlier ones.
Such an ordering is more consistent with how declarations generally work
in programming languages.
This will break any package that relied on the previous ordering.
Also includes a related fix to `findModule?` that mistakenly treated
executable roots as importable.
Improves executable handling in `lake exe` and `lake init`:
* `lake exe <target>` now parses `target` like a build target (as the
help text states it should) rather than as a basic name.
* `lake new foo.bar [std]` now generates executables named `foo-bar`.
* `lake new foo.bar exe` now properly creates `foo/bar.lean`.
these are compagnions to `MatcherApp.addArg` and `CasesOnApp.addArg`
when one only has an
expression (which may not be a type) to transform, but not a concret
values.
This is a prerequisite for guessing lexicographic order (#2874). Keeping
this on a separate PR because it’s sizable, and has a clear independent
specification.
This is an additional safety net on top of #2749: it protects users that
circumvent the build system (e.g. with `lake env`) as well as obviates
the need for TOCTOU-like race condition checks in the build system.
The check is activated by `CHECK_OLEAN_VERSION=ON`, which now defaults
to `OFF` as the sensible default for local development. When activated,
`USE_GITHASH=ON` is also force-enabled for stage 0 in order to make sure
that stage 1 can load its own core library.
This PR adds per-package server options to resolve#2455. It is based on
the previous work in #2456, but takes a different approach: options are
loaded for the specific file in the file worker when `print-paths` is
called, instead of loading them in the watchdog with a separate Lake
command. This change addresses review comments made in #2456.
In doing so, it introduces two new Lake config fields:
- `leanOptions`: `-D` flag options that are passed to both the language
server and `lean` when building.
- `moreServerOptions`: `-D` flag options that are passed to the language
server.
Since `print-paths` must also accept a file path to compute the options
for that file, this PR is changing the API for `print-paths`. As there
have been numerous complaints about the name `print-paths`, I also
decided to change it to `setup-file` in this PR, since it would break
compatibility with the old Lake API anyways.
This PR deprecates the Lakefile field `moreServerArgs` in favor of
`moreGlobalServerArgs`, as suggested in the review for #2456.
Fixes#2455
---------
Co-authored-by: digama0 <mcarneir@andrew.cmu.edu>
This was a Lean 3 pretty printer option. While this pretty printer
option tends to lead to confusing situations when set, it has been
frequently requested. [It is
possible](https://github.com/leanprover-community/mathlib4/pull/7910) to
implement this pretty printer option as a user, but it comes with some
artifacts -- for instance, expressions in hovers are not beta reduced.
Adding this as a core pp option is cleanest.
(We should consider having hooks into the tactic evaluator to allow
users to transform the tactic state between tactics. This would enable
beta reducing the entire local context for real, which would be useful
for teaching.)
Closes#715
We noticed at
https://github.com/leanprover/lean4/pull/2923#discussion_r1400468371
that this instance is not used. It's arguably also incorrect (as it
doesn't backtrack the `usedTheorems` field).
Seems better to just remove to avoid confusion.
Evidence that this is dead code:
* After deleting the instance, calling `saveState` in the `SimpM` monad
raises an error `failed to synthesize instance MonadBacktrack PUnit
SimpM`.
* Understanding the `MonadBacktrack` monad leads one to believe that
would have happened, via the fact that the only instances for
`MonadBacktrack` are either concrete instances (e.g. for `MetaM`,
`TacticM`, etc), or a single lifting instance `instance [MonadBacktrack
s m] [Monad m] : MonadBacktrack s (ExceptT ε m)`. (This is good and
correct behaviour: lifting instances for `MonadBacktrack` would be hard
to model.)
* Mathlib builds after the instance is removed.
Potential evidence that I have not sought, because we don't have
sufficient tooling:
* Compiling Lean/Std/Mathlib with a debugger, breaking on entering this
code.
This PR adds basic auto-completion support for imports. Since it still
lacks Lake support for accurate completion suggestions (cc @tydeu - we
already know what needs to be done), it falls back to traversing the
`LEAN_SRC_PATH` for available imports.
Three kinds of import completion requests are supported:
- Completion of the full `import` command. Triggered when requesting
completions in an empty space within the header.
- Known issue: It is possible to trigger this completion within a
comment in the header. Fixing this would require architecture for
parsing some kind of sub-syntax between individual commands.
- Completion of the full module name after an incomplete `import`
command.
- Completion of a partial module name with a trailing dot.
Since the set of imports is potentially expensive to compute, they are
cached for 10 seconds after the last import auto-completion request.
Closes#2655.
### Changes
This PR also makes the following changes:
- To support completions on the trailing dot, the `import` syntax was
adjusted to provide partial syntax when a trailing dot is used.
- `FileWorker.lean` was refactored lightly with some larger definitions
being broken apart.
- The `WorkerState` gained two new fields:
- `currHeaderStx` tracks the current header syntax, as opposed to
tracking only the initial header syntax in `initHeaderStx`. When the
header syntax changes, a task is launched that restarts the file worker
after a certain delay to avoid constant restarts while editing the
header. During this time period, we may still want to serve import
auto-completion requests, so we need to know the up-to-date header
syntax.
- `importCachingTask?` contains a task that computes the set of
available imports.
- `determineLakePath` has moved to a new file `Lean/Util/LakePath.lean`
as it is now needed both in `ImportCompletion.lean` and
`FileWorker.lean`.
- `forEachModuleIn` from `Lake/Config/Blob.lean` has moved to
`Lean/Util/Path.lean` as it is a generally useful utility function that
was useful for traversing the `LEAN_SRC_PATH` as well.
### Tests
Unfortunately, this PR lacks tests since the set of imports available in
`tests/lean/interactive` will not be stable. In the future, I will add
support for testing LSP requests in full project setups, which is when
tests for import auto-completion will be added as well.
there were wrong italics, missing backticks, missing indentation and I
took the liberty to replace `[here]` links with link targets that better
tell the reader what to expect when clicking there.
The `packMutual` code ought to reliably replace all recursive calls to
the functions in `preDefs`, even when they are under- or over-applied.
Therefore eta-expand if need rsp. keep extra arguments around.
Needs a tweak to `Meta.transform` to avoid mistaking the `f` in
`f x1 x2` as a zero-arity application.
Includes a test case.
This fixes#2628 and #2883.
This didn't work before
```
def f (n : Nat) : Nat :=
match n with
| 0 => 0
| n + 1 => (f) n
```
because the `RecApp` metadata marker gets in the way. More practically
relevant, such code is to be produced when using `rw` or `simp` in
recursive theorems (see included test case).
We can fix this by preprocessing the definitions and floating the
`.mdata` marker out of applications.
For structural recursion, there already exists a `preprocess` function;
this now also floats out `.mdata` markers.
For well-founded recursion, this introduces an analogous `preprocess`
function.
Fixes#2810.
One test case output changes: With the `.mdata` out of the way, we get a
different error message. Seems fine.
Alternative approaches are:
* Leaving the `.mdata` marker where it is, and looking around it.
Tried in #2813, but not nice (many many places where `withApp` etc.
need to be adjusted).
* Moving the `.mdata` _inside_ the application, so that `withApp` still
works. Tried in #2814. Also not nice, the invariant that the `.mdata`
is around the `.const` is tedious to maintain.
the code stumbled over recursive functions whose type doesn’t have
enough manifest foralls, like:
```
def FunType := Nat → Nat
mutual
def foo : FunType
| .zero => 0
| .succ n => bar n
def bar : FunType
| .zero => 0
| .succ n => foo n
end
termination_by foo n => n; bar n => n
```
This can be fixed by using `whnf` in appropriate places, to expose the
`.forall` constructor.
Fixes#2925, comes with test case.
`script/most-recent-nightly-tag.sh` determines the most recent nightly
release in your current git history.
Previously it was assuming that you had a `nightly` remote, to pull tags
from. Now it just pulls directly from the repository by URL.
This only really shows up when the `decreasing_tactic` fails with
multiple goals, as in
```
macro_rules
| `(tactic|decreasing_tactic) => `(tactic| by_cases (2 > 1))
def foo (n : Nat) : Nat := foo (n - 1)
termination_by foo n => n
```
where we now get
```
unsolved goals
case inl
n: Nat
h✝: 2 > 1
⊢ (invImage (fun a => a) instWellFoundedRelation).1 (n - 1) n
case inr
n: Nat
h✝: ¬2 > 1
⊢ (invImage (fun a => a) instWellFoundedRelation).1 (n - 1) n
```
rather than
```
LeanProject.lean:3:27
unsolved goals
case inl
n: Nat
h✝: 2 > 1
⊢ (invImage (fun a => a) instWellFoundedRelation).1 (n - 1) n
LeanProject.lean:3:27
unsolved goals
case inr
n: Nat
h✝: ¬2 > 1
⊢ (invImage (fun a => a) instWellFoundedRelation).1 (n - 1) n
```
The effect is neglectible, but the code is a bit nicer, so why not,
before someone looks at it again and wonders whether the goals are
reported separately for a reason.
Some beginners have trouble finding the `if h : c then t else e`
(`dite`) version of `ite`. This augments `ite`'s docstring to mention
the dependent version.
This implements a request handler for the `textDocument/rename` LSP
request, enabling renames via F2. It handles both local renames (e.g.
`let x := 1; x` to `let y := 1; y`) as well as global renames
(definitions).
Unfortunately it does not work for "orphan" files outside a project, as
it uses ilean data for the current file and this does not seem to be
saved for orphan files. As a result, the test file does not work,
although one can manually test the implementation against a project such
as mathlib. (This issue already exists for the "references" request,
e.g. ctrl click on the first `x` in `let x := 1; x` takes you to the
second one only if you are not in an orphan file.)
* Fixesleanprover-community/mathlib4#7124
Modifies `cleanup` so that it takes (1) an array of additional fvarids
to preserve and (2) a flag to control whether to include indirect
propositions.
(This is wanted in mathlib for the `extract_goal` tactic.)
previously, it would ignore a recursive call that has extra arguments,
which can happen when the recursive functions return something of
function type. Therefore just leave them extra arguments in place.
Fixes#2883.
In the previous doc-string, the sentence
> "If any of the selected tactic applications fail, it will call
`failed` with the main goal mvar."
was false both for `Location.wildcard` (where it should have said "If
all", not "If any") or for `Location.targets` (where `failed` is never
called).
Add documentation comments with examples to `universe`, `open`,
`export`, and `variable`.
The documentation shows up when hovering over keywords, hopefully
improving the experience for beginners.
Because `Decidable` carries data,
when writing `@[simp]` lemmas which include a `Decidable` instance on the LHS,
it is best to use `{_ : Decidable p}` rather than `[Decidable p]`
so that non-canonical instances can be found via unification rather than
typeclass search.
(Previously this behaviour was often being hidden by the default `decide :=
true` in `simp`.)
We were checking out the synthetic merge commit between the PR `HEAD`
and `master`, and this was then breaking the logic to determine which
nightly-testing branches to use in Mathlib and Std.
previously, only the WellFounded code was making use of the error
location in the RecApp-metadata. We can do the same for structural
recursion. This way,
```
def f (n : Nat) : Nat :=
match n with
| 0 => 0
| n + 1 => f (n + 1)
```
will show the error with squiggly lines under `f (n + 1)`, and not at
`def f`.
`simp` was previously swallowing runtime exceptions and masking an
issue with this example.
`runT` is defined by well-founded recursion, but reducing the ground
term `runT x` takes a long time when `decide := true`.
Remark PR #2722 changes the `decide` default value to `false`.
When `decide := true`, we should probably have better diagnostics /
error messages for this kind of situation.
This is the same flag that the C test uses. Previously this was hidden
in the Lean compiler itself but now that the optimization pass is phased
out of the compiler we need to put it here.
Co-authored-by: Henrik Böving <hargonix@gmail.com>
In LLVM builds the Cmake CC is necessarily clang -> leanc will be able
to act on LLVM bitcode files if configured this way.
Co-authored-by: Siddharth <siddu.druid@gmail.com>
* [ ] Put an X between the brackets on this line if you have done all of the following:
* Check that your issue is not already [filed](https://github.com/leanprover/lean4/issues).
* Reduce the issue to a minimal, self-contained, reproducible test case. Avoid dependencies to mathlib4 or std4.
Please put an X between the brackets as you perform the following steps:
* [ ] Check that your issue is not already filed:
https://github.com/leanprover/lean4/issues
* [ ] Reduce the issue to a minimal, self-contained, reproducible test case.
Avoid dependencies to Mathlib or Batteries.
* [ ] Test your test case against the latest nightly release, for example on
https://live.lean-lang.org/#project=lean-nightly
(You can also use the settings there to switch to “Lean nightly”)
### Description
@@ -19,7 +25,7 @@ assignees: ''
### Context
[Broader context that the issue occured in. If there was any prior discussion on [the Lean Zulip](https://leanprover.zulipchat.com), link it here as well.]
[Broader context that the issue occurred in. If there was any prior discussion on [the Lean Zulip](https://leanprover.zulipchat.com), link it here as well.]
### Steps to Reproduce
@@ -33,8 +39,8 @@ assignees: ''
### Versions
[Output of `lean --version`in the folder that the issue occured in]
[OS version]
[Output of `#version`or `#eval Lean.versionString`]
* Ensure your PR follows the [External Contribution Guidelines](https://github.com/leanprover/lean4/blob/master/CONTRIBUTING.md).
* Please make sure the PR has excellent documentation and tests. If we label it `missing documentation` or `missing tests` then it needs fixing!
*Add the link to your `RFC` or `bug` issue below.
*Include the link to your `RFC` or `bug` issue in the description.
* If the issue does not already have approval from a developer, submit the PR as draft.
*Remove this section before submitting.
*The PR title/description will become the commit message. Keep it up-to-date as the PR evolves.
* For `feat/fix` PRs, the first paragraph starting with "This PR" must be present and will become a
changelog entry unless the PR is labeled with `no-changelog`. If the PR does not have this label,
it must instead be categorized with one of the `changelog-*` labels (which will be done by a
reviewer for external PRs).
* A toolchain of the form `leanprover/lean4-pr-releases:pr-release-NNNN` for Linux and M-series Macs will be generated upon build. To generate binaries for Windows and Intel-based Macs as well, write a comment containing `release-ci` on its own line.
* If you rebase your PR onto `nightly-with-mathlib` then CI will test Mathlib against your PR.
* You can manage the `awaiting-review`, `awaiting-author`, and `WIP` labels yourself, by writing a comment containing one of these labels on its own line.
* Remove this section, up to and including the `---` before submitting.
You can manage the `awaiting-review`, `awaiting-author`, and `WIP` labels yourself, by writing a comment containing one of these labels on its own line.
---
# Summary
This PR <short changelog summary for feat/fix, see above>.
Link to `RFC` or `bug` issue:
Closes <`RFC` or `bug` issue number fixed by this PR, if any>
echo "... and Batteries has a 'nightly-testing-$MOST_RECENT_NIGHTLY' tag."
MESSAGE=""
if [[ -n "$MATHLIB_REMOTE_TAGS" ]]; then
echo "... and Mathlib has a 'nightly-testing-$MOST_RECENT_NIGHTLY' tag."
else
echo "... but Mathlib does not yet have a 'nightly-testing-$MOST_RECENT_NIGHTLY' tag."
MESSAGE="- ❗ Mathlib CI can not be attempted yet, as the \`nightly-testing-$MOST_RECENT_NIGHTLY\` tag does not exist there yet. We will retry when you push more commits. If you rebase your branch onto \`nightly-with-mathlib\`, Mathlib CI should run now."
fi
else
echo "... but Batteries does not yet have a 'nightly-testing-$MOST_RECENT_NIGHTLY' tag."
MESSAGE="- ❗ Batteries CI can not be attempted yet, as the \`nightly-testing-$MOST_RECENT_NIGHTLY\` tag does not exist there yet. We will retry when you push more commits. If you rebase your branch onto \`nightly-with-mathlib\`, Batteries CI should run now."
fi
else
echo "The most recently nightly tag on this branch has SHA: $NIGHTLY_SHA"
MESSAGE="- ❗ Batteries/Mathlib CI will not be attempted unless your PR branches off the \`nightly-with-mathlib\` branch. Try \`git rebase $MERGE_BASE_SHA --onto $NIGHTLY_WITH_MATHLIB_SHA\`."
fi
if [[ -n "$MESSAGE" ]]; then
echo "Checking existing messages"
# The code for updating comments is duplicated in mathlib's
# scripts/lean-pr-testing-comments.sh
# so keep in sync
# Use GitHub API to check if a comment already exists
// check first commit only (and only once) since later commits might be intended to be squashed away
if (!/^(feat|fix|doc|style|refactor|test|chore|perf): .*[^.]($|\n\n)/.test(commits[0].commit.message)) {
await github.rest.issues.createComment({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
body: 'Thanks for your contribution! Please make sure to follow our [Commit Convention](https://leanprover.github.io/lean4/doc/dev/commit_convention.html).',
RUN curl https://raw.githubusercontent.com/leanprover/elan/master/elan-init.sh -sSf | sh -s -- -y --default-toolchain none
ENVPATH="/home/gitpod/.elan/bin:${PATH}"
# Create a dummy toolchain so that we can pre-register it with elan
RUN mkdir -p /workspace/lean4/build/release/stage1/bin && touch /workspace/lean4/build/release/stage1/bin/lean && elan toolchain link lean4 /workspace/lean4/build/release/stage1
RUN mkdir -p /workspace/lean4/build/release/stage0/bin && touch /workspace/lean4/build/release/stage0/bin/lean && elan toolchain link lean4-stage0 /workspace/lean4/build/release/stage0
**Descriptive Title and Summary**: The PR title should briefly explain the purpose of the PR. The summary should give more detailed information on what changes are made and why. Links to Zulip threads are not acceptable as a summary. You are responsible for summarizing the discussion, and getting support for it.
**Follow the commit convention**: Pull requests are squash merged, and the
commit message is taken from the pull request title and body, so make sure they adhere to the [commit convention](https://github.com/leanprover/lean4/blob/master/doc/dev/commit_convention.md). Put questions and extra information, which should not be part of the final commit message, into a first comment rather than the Pull Request description.
Because the change will be squashed, there is no need to polish the commit messages and history on the branch.
**Link to Relevant Issues**: Reference any issues that your PR addresses to provide context.
**Stay Responsive**: Once the PR is submitted, stay responsive to feedback and be prepared to make necessary revisions. We will close any PR that has been inactive (no response or updates from the submitter) for more than a month.
@@ -59,6 +63,20 @@ PR Submission:
Reviews and Feedback:
----
The lean4 repo is managed by the Lean FRO's *triage team* that aims to provide initial feedback on new bug reports, PRs, and RFCs weekly.
This feedback generally consists of prioritizing the ticket using one of the following categories:
* label `P-high`: We will work on this issue
* label `P-medium`: We may work on this issue if we find the time
* label `P-low`: We are not planning to work on this issue
* *closed*: This issue is already fixed, it is not an issue, or is not sufficiently compatible with our roadmap for the project and we will not work on it nor accept external contributions on it
For *bug reports*, the listed priority reflects our commitment to fixing the issue.
It is generally indicative but not necessarily identical to the priority an external contribution addressing this bug would receive.
For *PRs* and *RFCs*, the priority reflects our commitment to reviewing them and getting them to an acceptable state.
Accepted RFCs are marked with the label `RFC accepted` and afterwards assigned a new "implementation" priority as with bug reports.
General guidelines for interacting with reviews and feedback:
**Be Patient**: Given the limited number of full-time maintainers and the volume of PRs, reviews may take some time.
**Engage Constructively**: Always approach feedback positively and constructively. Remember, reviews are about ensuring the best quality for the project, not personal criticism.
Lean supports the basic mathematical operations you’d expect for all of the number types: addition, subtraction, multiplication, division, and remainder.
The following code shows how you’d use each one in a `def` commands:
```lean
-- addition
defsum:=5+10
-- subtraction
defdifference:=95.5-4.3
-- multiplication
defproduct:=4*30
-- division
defquotient:=53.7/32.2
-- remainder/modulo
defmodulo:=43%5
```
Each expression in these statements uses a mathematical operator and evaluates to a single value.
If any of the terms ``tᵢ`` in the template above contain a recursive call to ``foo``, the equation compiler tries to interpret the definition as a structural recursion. In order for that to succeed, the recursive arguments must be subterms of the corresponding arguments on the left-hand side. The function is then defined using a *course of values* recursion, using automatically generated functions ``below`` and ``brec`` in the namespace corresponding to the inductive type of the recursive argument. In this case the defining equations hold definitionally, possibly with additional case splits.
The case where patterns are matched against an argument whose type is an inductive family is known as *dependent pattern matching*. This is more complicated, because the type of the function being defined can impose constraints on the patterns that are matched. In this case, the equation compiler will detect inconsistent cases and rule them out.
```lean
universe u
inductive Vector (α : Type u) : Nat → Type u
| nil : Vector α 0
| cons : α → Vector α n → Vector α (n+1)
namespace Vector
def head : Vector α (n+1) → α
| cons h t => h
def tail : Vector α (n+1) → Vector α n
| cons h t => t
def map (f : α → β → γ) : Vector α n → Vector β n → Vector γ n
| nil, nil => nil
| cons a va, cons b vb => cons (f a b) (map f va vb)
end Vector
```
.. _recursive_functions:
Recursive functions
===================
Lean must ensure that a recursive function terminates, for which there are two strategies: _structural recursion_, in which all recursive calls are made on smaller parts of the input data, and _well-founded recursion_, in which recursive calls are justified by showing that arguments to recursive calls are smaller according to some other measure.
Structural recursion
--------------------
If the definition of a function contains recursive calls, Lean first tries to interpret the definition as a structural recursion. In order for that to succeed, the recursive arguments must be subterms of the corresponding arguments on the left-hand side.
The function is then defined using a *course of values* recursion, using automatically generated functions ``below`` and ``brec`` in the namespace corresponding to the inductive type of the recursive argument. In this case the defining equations hold definitionally, possibly with additional case splits.
If structural recursion fails, the equation compiler falls back on well-founded recursion. It tries to infer an instance of ``SizeOf`` for the type of each argument, and then show that each recursive call is decreasing under the lexicographic order of the arguments with respect to ``sizeOf`` measure. If it fails, the error message provides information as to the goal that Lean tried to prove. Lean uses information in the local context, so you can often provide the relevant proof manually using ``have`` in the body of the definition. In this case of well-founded recursion, the defining equations hold only propositionally, and can be accessed using ``simp`` and ``rewrite`` with the name ``foo``.
Well-founded recursion
---------------------
If structural recursion fails, the equation compiler falls back on well-founded recursion. It tries to infer an instance of ``SizeOf`` for the type of each argument, and then tries to find a permutation of the arguments such that each recursive call is decreasing under the lexicographic order with respect to ``sizeOf`` measures. Lean uses information in the local context, so you can often provide the relevant proof manually using ``have`` in the body of the definition.
In the case of well-founded recursion, the equation used to declare the function holds only propositionally, but not definitionally, and can be accessed using ``unfold``, ``simp`` and ``rewrite`` with the function name (for example ``unfold foo`` or ``simp [foo]``, where ``foo`` is the function defined with well-founded recursion).
```lean
namespace Hide
@@ -528,9 +569,53 @@ by rw [div]; rfl
end Hide
```
If Lean cannot find a permutation of the arguments for which all recursive calls are decreasing, it will print a table that contains, for every recursive call, which arguments Lean could prove to be decreasing. For example, a function with three recursive calls and four parameters might cause the following message to be printed
```
example.lean:37:0-43:31: error: Could not find a decreasing measure.
The arguments relate at each recursive call as follows:
(<, ≤, =: relation proved, ? all proofs failed, _: no proof attempted)
x1 x2 x3 x4
1) 39:6-27 = = _ =
2) 40:6-25 = ? _ <
3) 41:6-25 < _ _ _
Please use `termination_by` to specify a decreasing measure.
```
This table should be read as follows:
* In the first recursive call, in line 39, arguments 1, 2 and 4 are equal to the function's parameters.
* The second recursive call, in line 40, has an equal first argument, a smaller fourth argument, and nothing could be inferred for the second argument.
* The third recursive call, in line 41, has a decreasing first argument.
* No other proofs were attempted, either because the parameter has a type without a non-trivial ``WellFounded`` instance (parameter 3), or because it is already clear that no decreasing measure can be found.
Lean will print the termination argument it found if ``set_option showInferredTerminationBy true`` is set.
If Lean does not find the termination argument, or if you want to be explicit, you can append a `termination_by` clause to the function definition, after the function's body, but before the `where` clause if present. It is of the form
```
termination_by e
```
where ``e`` is an expression that depends on the parameters of the function and should be decreasing at each recursive call. The type of `e` should be an instance of the class ``WellFoundedRelation``, which determines how to compare two values of that type.
If ``f`` has parameters “after the ``:``” (for example when defining functions via patterns using `|`), then these can be brought into scope using the syntax
```
termination_by a₁ … aₙ => e
```
By default, Lean uses the tactic ``decreasing_tactic`` when proving that an argument is decreasing; see its documentation for how to globally extend it. You can also choose to use a different tactic for a given function definition with the clause
```
decreasing_by <tac>
```
which should come after ``termination_by`, if present.
Note that recursive definitions can in general require nested recursions, that is, recursion on different arguments of ``foo`` in the template above. The equation compiler handles this by abstracting later arguments, and recursively defining higher-order functions to meet the specification.
The equation compiler also allows mutual recursive definitions, with a syntax similar to that of [Mutual and Nested Inductive Definitions](#mutual-and-nested-inductive-definitions). They are compiled using well-founded recursion, and so once again the defining equations hold only propositionally.
Mutual recursion
----------------
The equation compiler also allows mutual recursive definitions, with a syntax similar to that of [Mutual and Nested Inductive Definitions](#mutual-and-nested-inductive-definitions). Mutual definitions are always compiled using well-founded recursion, and so once again the defining equations hold only propositionally.
```lean
mutual
@@ -587,29 +672,31 @@ def num_consts_lst : List Term → Nat
end
```
The case where patterns are matched against an argument whose type is an inductive family is known as *dependent pattern matching*. This is more complicated, because the type of the function being defined can impose constraints on the patterns that are matched. In this case, the equation compiler will detect inconsistent cases and rule them out.
In a set of mutually recursive function, either all or no functions must have an explicit termination argument (``termination_by``). A change of the default termination tactic (``decreasing_by``) only affects the proofs about the recursive calls of that function, not the other functions in the group.
```lean
universe u
```
mutual
theorem even_of_odd_succ : ∀ n, Odd (n + 1) → Even n
| _, odd_succ n h => h
termination_by n h => h
decreasing_by decreasing_tactic
inductive Vector (α : Type u) : Nat → Type u
| nil : Vector α 0
| cons : α → Vector α n → Vector α (n+1)
theorem odd_of_even_succ : ∀ n, Even (n + 1) → Odd n
| _, even_succ n h => h
termination_by n h => h
end
```
namespace Vector
Another way to express mutual recursion is using local function definitions in ``where`` or ``let rec`` clauses: these can be mutually recursive with each other and their containing function:
def head {α : Type} : Vector α (n+1) → α
| cons h t => h
def tail {α : Type} : Vector α (n+1) → Vector α n
| cons h t => t
def map {α β γ : Type} (f : α → β → γ) :
∀ {n}, Vector α n → Vector β n → Vector γ n
| 0, nil, nil => nil
| n+1, cons a va, cons b vb => cons (f a b) (map f va vb)
end Vector
```
theorem even_of_odd_succ : ∀ n, Odd (n + 1) → Even n
| _, odd_succ n h => h
termination_by n h => h
where
theorem odd_of_even_succ : ∀ n, Even (n + 1) → Odd n
@@ -65,26 +65,59 @@ You now have a Lean binary and library that include your changes, though their
own compilation was not influenced by them, that you can use to test your
changes on test programs whose compilation *will* be influenced by the changes.
Finally, when we want to use new language features in the library, we need to
update the stage 0 compiler, which can be done via `make -C stageN update-stage0`.
`make update-stage0` without `-C` defaults to stage1.
## Updating stage0
Updates to `stage0` should be their own commits in the Git history. In
other words, before running `make update-stage0`, please commit your
work. Then, commit the updated `stage0` compiler code with the commit message:
Finally, when we want to use new language features in the library, we need to
update the archived C source code of the stage 0 compiler in `stage0/src`.
The github repository will automatically update stage0 on `master` once
`src/stdlib_flags.h` and `stage0/src/stdlib_flags.h` are out of sync.
If you have write access to the lean4 repository, you can also manually
trigger that process, for example to be able to use new features in the compiler itself.
You can do that on <https://github.com/leanprover/lean4/actions/workflows/update-stage0.yml>
or using Github CLI with
```
chore: updatestage0
gh workflow run update-stage0.yml
```
Leaving stage0 updates to the CI automation is preferable, but should you need
to do it locally, you can use `make update-stage0-commit` in `build/release` to
update `stage0` from `stage1` or `make -C stageN update-stage0-commit` to
update from another stage. This command will automatically stage the updated files
and introduce a commit,so make sure to commit your work before that.
If you rebased the branch (either onto a newer version of `master`, or fixing
up some commits prior to the stage0 update, recreate the stage0 update commits.
The script `script/rebase-stage0.sh` can be used for that.
The CI should prevent PRs with changes to stage0 (besides `stdlib_flags.h`)
from entering `master` through the (squashing!) merge queue, and label such PRs
with the `changes-stage0` label. Such PRs should have a cleaned up history,
with separate stage0 update commits; then coordinate with the admins to merge
your PR using rebase merge, bypassing the merge queue.
## Further Bootstrapping Complications
As written above, changes in meta code in the current stage usually will only
affect later stages. This is an issue in two specific cases.
* For the special case of *quotations*, it is desirable to have changes in builtin parsers affect them immediately: when the changes in the parser become active in the next stage, builtin macros implemented via quotations should generate syntax trees compatible with the new parser, and quotation patterns in builtin macros and elaborators should be able to match syntax created by the new parser and macros.
Since quotations capture the syntax tree structure during execution of the current stage and turn it into code for the next stage, we need to run the current stage's builtin parsers in quotations via the interpreter for this to work.
Caveats:
* We activate this behavior by default when building stage 1 by setting `-Dinternal.parseQuotWithCurrentStage=true`.
We force-disable it inside `macro/macro_rules/elab/elab_rules` via `suppressInsideQuot` as they are guaranteed not to run in the next stage and may need to be run in the current one, so the stage 0 parser is the correct one to use for them.
It may be necessary to extend this disabling to functions that contain quotations and are (exclusively) used by one of the mentioned commands. A function using quotations should never be used by both builtin and non-builtin macros/elaborators. Example: https://github.com/leanprover/lean4/blob/f70b7e5722da6101572869d87832494e2f8534b7/src/Lean/Elab/Tactic/Config.lean#L118-L122
* The parser needs to be reachable via an `import` statement, otherwise the version of the previous stage will silently be used.
* Only the parser code (`Parser.fn`) is affected; all metadata such as leading tokens is taken from the previous stage.
For an example, see https://github.com/leanprover/lean4/commit/f9dcbbddc48ccab22c7674ba20c5f409823b4cc1#diff-371387aed38bb02bf7761084fd9460e4168ae16d1ffe5de041b47d3ad2d22422R13
* For *non-builtin* meta code such as `notation`s or `macro`s in
`Notation.lean`, we expect changes to affect the current file and all later
files of the same stage immediately, just like outside the stdlib. To ensure
this, we need to build the stage using `-Dinterpreter.prefer_native=false` -
this, we build stage 1 using `-Dinterpreter.prefer_native=false` -
otherwise, when executing a macro, the interpreter would notice that there is
already a native symbol available for this function and run it instead of the
new IR, but the symbol is from the previous stage!
@@ -102,26 +135,11 @@ affect later stages. This is an issue in two specific cases.
further stages (e.g. after an `update-stage0`) will then need to be compiled
with the flag set to `false` again since they will expect the new signature.
For an example, see https://github.com/leanprover/lean4/commit/da4c46370d85add64ef7ca5e7cc4638b62823fbb.
When enabling `prefer_native`, we usually want to *disable*`parseQuotWithCurrentStage` as it would otherwise make quotations use the interpreter after all.
However, there is a specific case where we want to set both options to `true`: when we make changes to a non-builtin parser like `simp` that has a builtin elaborator, we cannot have the new parser be active outside of quotations in stage 1 as the builtin elaborator from stage 0 would not understand them; on the other hand, we need quotations in e.g. the builtin `simp` elaborator to produce the new syntax in the next stage.
As this issue usually affects only tactics, enabling `debug.byAsSorry` instead of `prefer_native` can be a simpler solution.
* For the special case of *quotations*, it is desirable to have changes in
built-in parsers affect them immediately: when the changes in the parser
become active in the next stage, macros implemented via quotations should
generate syntax trees compatible with the new parser, and quotation patterns
in macro and elaborators should be able to match syntax created by the new
parser and macros. Since quotations capture the syntax tree structure during
execution of the current stage and turn it into code for the next stage, we
need to run the current stage's built-in parsers in quotation via the
interpreter for this to work. Caveats:
* Since interpreting full parsers is not nearly as cheap and we rarely change
built-in syntax, this needs to be opted in using `-Dinternal.parseQuotWithCurrentStage=true`.
* The parser needs to be reachable via an `import` statement, otherwise the
version of the previous stage will silently be used.
* Only the parser code (`Parser.fn`) is affected; all metadata such as leading
tokens is taken from the previous stage.
For an example, see https://github.com/leanprover/lean4/commit/f9dcbbddc48ccab22c7674ba20c5f409823b4cc1#diff-371387aed38bb02bf7761084fd9460e4168ae16d1ffe5de041b47d3ad2d22422
(from before the flag defaulted to `false`).
For a `prefer_native` example, see https://github.com/leanprover/lean4/commit/da4c46370d85add64ef7ca5e7cc4638b62823fbb.
To modify either of these flags both for building and editing the stdlib, adjust
the code in `stage0/src/stdlib_flags.h`. The flags will automatically be reset
@@ -5,7 +5,7 @@ Some notes on how to debug Lean, which may also be applicable to debugging Lean
## Tracing
In `CoreM` and derived monads, we use `trace![traceCls] "msg with {interpolations}"` to fill the structured trace viewable with `set_option trace.traceCls true`.
In `CoreM` and derived monads, we use `trace[traceCls] "msg with {interpolations}"` to fill the structured trace viewable with `set_option trace.traceCls true`.
New trace classes have to be registered using `registerTraceClass` first.
Notable trace classes:
@@ -22,7 +22,9 @@ Notable trace classes:
In pure contexts or when execution is aborted before the messages are finally printed, one can instead use the term `dbg_trace "msg with {interpolations}"; val` (`;` can also be replaced by a newline), which will print the message to stderr before evaluating `val`. `dbgTraceVal val` can be used as a shorthand for `dbg_trace "{val}"; val`.
Note that if the return value is not actually used, the trace code is silently dropped as well.
In the language server, stderr output is buffered and shown as messages after a command has been elaborated, unless the option `server.stderrAsMessages` is deactivated.
By default, such stderr output is buffered and shown as messages after a command has been elaborated, which is necessary to ensure deterministic ordering of messages under parallelism.
If Lean aborts the process before it can finish the command or takes too long to do that, using `-DstderrAsMessages=false` avoids this buffering and shows `dbg_trace` output (but not `trace`s or other diagnostics) immediately.
@@ -49,14 +49,64 @@ In the case of `@[extern]` all *irrelevant* types are removed first; see next se
is represented by the representation of that parameter's type.
For example, `{ x : α // p }`, the `Subtype` structure of a value of type `α` and an irrelevant proof, is represented by the representation of `α`.
* `Nat` is represented by `lean_object *`.
Its runtime value is either a pointer to an opaque bignum object or, if the lowest bit of the "pointer" is 1 (`lean_is_scalar`), an encoded unboxed natural number (`lean_box`/`lean_unbox`).
Similarly, the signed integer types `Int8`, ..., `Int64`, `ISize` are also represented by the unsigned C types `uint8_t`, ..., `uint64_t`, `size_t`, respectively, because they have a trivial structure.
* `Nat` and `Int` are represented by `lean_object *`.
Their runtime values is either a pointer to an opaque bignum object or, if the lowest bit of the "pointer" is 1 (`lean_is_scalar`), an encoded unboxed natural number or integer (`lean_box`/`lean_unbox`).
* A universe `Sort u`, type constructor `... → Sort u`, or proposition `p : Prop` is *irrelevant* and is either statically erased (see above) or represented as a `lean_object *` with the runtime value `lean_box(0)`
* Any other type is represented by `lean_object *`.
Its runtime value is a pointer to an object of a subtype of `lean_object` (see respective declarations in `lean.h`) or the unboxed value `lean_box(cidx)` for the `cidx`th constructor of an inductive type if this constructor does not have any relevant parameters.
Its runtime value is a pointer to an object of a subtype of `lean_object` (see the "Inductive types" section below) or the unboxed value `lean_box(cidx)` for the `cidx`th constructor of an inductive type if this constructor does not have any relevant parameters.
Example: the runtime value of `u : Unit` is always `lean_box(0)`.
#### Inductive types
For inductive types which are in the fallback `lean_object *` case above and not trivial constructors, the type is stored as a `lean_ctor_object`, and `lean_is_ctor` will return true. A `lean_ctor_object` stores the constructor index in the header, and the fields are stored in the `m_objs` portion of the object.
The memory order of the fields is derived from the types and order of the fields in the declaration. They are ordered as follows:
* Non-scalar fields stored as `lean_object *`
* Fields of type `USize`
* Other scalar fields, in decreasing order by size
Within each group the fields are ordered in declaration order. **Warning**: Trivial wrapper types still count toward a field being treated as non-scalar for this purpose.
* To access fields of the first kind, use `lean_ctor_get(val, i)` to get the `i`th non-scalar field.
* To access `USize` fields, use `lean_ctor_get_usize(val, n+i)` to get the `i`th usize field and `n` is the total number of fields of the first kind.
* To access other scalar fields, use `lean_ctor_get_uintN(val, off)` or `lean_ctor_get_usize(val, off)` as appropriate. Here `off` is the byte offset of the field in the structure, starting at `n*sizeof(void*)` where `n` is the number of fields of the first two kinds.
For example, a structure such as
```lean
structure S where
ptr_1 : Array Nat
usize_1 : USize
sc64_1 : UInt64
ptr_2 : { x : UInt64 // x > 0 } -- wrappers don't count as scalars
sc64_2 : Float -- `Float` is 64 bit
sc8_1 : Bool
sc16_1 : UInt16
sc8_2 : UInt8
sc64_3 : UInt64
usize_2 : USize
ptr_3 : Char -- trivial wrapper around `UInt32`
sc32_1 : UInt32
sc16_2 : UInt16
```
would get re-sorted into the following memory order:
By default, all `lean_object *` parameters of an `@[extern]` function are considered *owned*, i.e. the external code is passed a "virtual RC token" and is responsible for passing this token along to another consuming function (exactly once) or freeing it via `lean_dec`.
@@ -111,6 +161,15 @@ if (lean_io_result_is_ok(res)) {
lean_io_mark_end_initialization();
```
In addition, any other thread not spawned by the Lean runtime itself must be initialized for Lean use by calling
```c
void lean_initialize_thread();
```
and should be finalized in order to free all thread-local resources by calling
```c
void lean_finalize_thread();
```
## `@[extern]` in the Interpreter
The interpreter can run Lean declarations for which symbols are available in loaded shared libraries, which includes `@[extern]` declarations.
@@ -121,4 +180,4 @@ Thus to e.g. run `#eval` on such a declaration, you need to
Note that it is not sufficient to load the foreign library containing the external symbol because the interpreter depends on code that is emitted for each `@[extern]` declaration.
Thus it is not possible to interpret an `@[extern]` declaration in the same file.
See `tests/compiler/foreign` for an example.
See [`tests/compiler/foreign`](https://github.com/leanprover/lean4/tree/master/tests/compiler/foreign/) for an example.
You can use `elantoolchain link` to give a specific stage build
directory a reference name, then use `elan override set` to associate
such a name to the current directory. We usually want to use `stage0`
for editing files in `src` and `stage1`for everything else (e.g.
tests).
The `lean-toolchain` files in the Lean 4 repository are set up to use the `lean4-stage0`
toolchain for editing files in `src` and the `lean4` toolchain for editing files in `tests`.
Run the following commands to make `lean4` point at `stage1` and `lean4-stage0`point at `stage0`:
```bash
# in the Lean rootdir
elan toolchain link lean4 build/release/stage1
elan toolchain link lean4-stage0 build/release/stage0
# make `lean` etc. point to stage1 in the rootdir and subdirs
elan override set lean4
cd src
# make `lean` etc. point to stage0 anywhere inside `src`
elan override set lean4-stage0
```
You can also use the `+toolchain` shorthand (e.g. `lean +lean4-debug`) to switch
@@ -65,9 +59,31 @@ If you push `my-tag` to a fork in your github account `my_name`,
you can then put `my_name/lean4:my-tag` in your `lean-toolchain` file in a project using `lake`.
(You must use a tag name that does not start with a numeral, or contain `_`).
### VS Code
There is a `lean.code-workspace` file that correctly sets up VS Code with workspace roots for the stage0/stage1 setup described above as well as with other settings.
You should always load it when working on Lean, such as by invoking
```
code lean.code-workspace
```
on the command line.
### `ccache`
Lean's build process uses [`ccache`](https://ccache.dev/) if it is
installed to speed up recompilation of the generated C code. Without
`ccache`, you'll likely spend more time than necessary waiting on
rebuilds - it's a good idea to make sure it's installed.
### `prelude`
Unlike most Lean projects, all submodules of the `Lean` module begin with the
`prelude` keyword. This disables the automated import of `Init`, meaning that
developers need to figure out their own subset of `Init` to import. This is done
such that changing files in `Init` doesn't force a full rebuild of `Lean`.
### Testing against Mathlib/Batteries
You can test a Lean PR against Mathlib and Batteries by rebasing your PR
on to `nightly-with-mathlib` branch. (It is fine to force push after rebasing.)
CI will generate a branch of Mathlib and Batteries called `lean-pr-testing-NNNN`
that uses the toolchain for your PR, and will report back to the Lean PR with results from Mathlib CI.
See https://leanprover-community.github.io/contribute/tags_and_branches.html for more details.
This checklist walks you through releasing a stable version.
See below for the checklist for release candidates.
We'll use `v4.6.0` as the intended release version as a running example.
-`git checkout releases/v4.6.0`
(This branch should already exist, from the release candidates.)
-`git pull`
- In `src/CMakeLists.txt`, verify you see
-`set(LEAN_VERSION_MINOR 6)` (for whichever `6` is appropriate)
-`set(LEAN_VERSION_IS_RELEASE 1)`
- (both of these should already be in place from the release candidates)
-`git tag v4.6.0`
-`git push $REMOTE v4.6.0`, where `$REMOTE` is the upstream Lean repository (e.g., `origin`, `upstream`)
- Now wait, while CI runs.
- You can monitor this at `https://github.com/leanprover/lean4/actions/workflows/ci.yml`,
looking for the `v4.6.0` tag.
- This step can take up to an hour.
- If you are intending to cut the next release candidate on the same day,
you may want to start on the release candidate checklist now.
- Go to https://github.com/leanprover/lean4/releases and verify that the `v4.6.0` release appears.
- Edit the release notes on Github to select the "Set as the latest release".
- Follow the instructions in creating a release candidate for the "GitHub release notes" step,
now that we have a written `RELEASES.md` section.
Do a quick sanity check.
- Next, we will move a curated list of downstream repos to the latest stable release.
- For each of the repositories listed below:
- Make a PR to `master`/`main` changing the toolchain to `v4.6.0`
- Update the toolchain file
- In the Lakefile, if there are dependencies on specific version tags of dependencies that you've already pushed as part of this process, update them to the new tag.
If they depend on `main` or `master`, don't change this; you've just updated the dependency, so it will work and be saved in the manifest
- Run `lake update`
- The PR title should be "chore: bump toolchain to v4.6.0".
- Merge the PR once CI completes.
- Create the tag `v4.6.0` from `master`/`main` and push it.
- Merge the tag `v4.6.0` into the `stable` branch and push it.
- In `RELEASES.md` replace `Development in progress` in the `v4.7.0` section with `Release notes to be written.`
- It is essential to choose the nightly that will become the release candidate as early as possible, to avoid confusion.
- In `src/CMakeLists.txt`,
- verify that you see `set(LEAN_VERSION_MINOR 7)` (for whichever `7` is appropriate); this should already have been updated when the development cycle began.
- `set(LEAN_VERSION_IS_RELEASE 1)` (this should be a change; on `master` and nightly releases it is always `0`).
- Commit your changes to `src/CMakeLists.txt`, and push.
- `git tag v4.7.0-rc1`
- `git push origin v4.7.0-rc1`
- Now wait, while CI runs.
- You can monitor this at `https://github.com/leanprover/lean4/actions/workflows/ci.yml`, looking for the `v4.7.0-rc1` tag.
- This step can take up to an hour.
- (GitHub release notes) Once the release appears at https://github.com/leanprover/lean4/releases/
- Verify that the release is marked as a prerelease (this should have been done automatically by the CI release job).
- In the "previous tag" dropdown, select `v4.6.0`, and click "Generate release notes".
This will add a list of all the commits since the last stable version.
- Delete "update stage0" commits, and anything with a completely inscrutable commit message.
- Next, we will move a curated list of downstream repos to the release candidate.
- This assumes that for each repository either:
* There is already a *reviewed* branch `bump/v4.7.0` containing the required adaptations.
The preparation of this branch is beyond the scope of this document.
* The repository does not need any changes to move to the new version.
- For each of the target repositories:
- If the repository does not need any changes (i.e. `bump/v4.7.0` does not exist) then create
a new PR updating `lean-toolchain` to `leanprover/lean4:v4.7.0-rc1` and running `lake update`.
- Otherwise:
- Checkout the `bump/v4.7.0` branch.
- Verify that the `lean-toolchain` is set to the nightly from which the release candidate was created.
- `git merge origin/master`
- Change the `lean-toolchain` to `leanprover/lean4:v4.7.0-rc1`
- In `lakefile.lean`, change any dependencies which were using `nightly-testing` or `bump/v4.7.0` branches
back to `master` or `main`, and run `lake update` for those dependencies.
- Run `lake build` to ensure that dependencies are found (but it's okay to stop it after a moment).
- `git commit`
- `git push`
- Open a PR from `bump/v4.7.0` to `master`, and either merge it yourself after CI, if appropriate,
or notify the maintainers that it is ready to go.
- Once the PR has been merged, tag `master` with `v4.7.0-rc1` and push this tag.
- We do this for the same list of repositories as for stable releases, see above.
As above, there are dependencies between these, and so the process above is iterative.
It greatly helps if you can merge the `bump/v4.7.0` PRs yourself!
It is essential for Mathlib CI that you then create the next `bump/v4.8.0` branch
for the next development cycle.
Set the `lean-toolchain` file on this branch to same `nightly` you used for this release.
- For Batteries/Aesop/Mathlib, which maintain a `nightly-testing` branch, make sure there is a tag
`nightly-testing-2024-02-29` with date corresponding to the nightly used for the release
(create it if not), and then on the `nightly-testing` branch `git reset --hard master`, and force push.
- Make an announcement!
This should go in https://leanprover.zulipchat.com/#narrow/stream/113486-announce, with topic `v4.7.0-rc1`.
Please see previous announcements for suggested language.
You will want a few bullet points for main topics from the release notes.
Please also make sure that whoever is handling social media knows the release is out.
- Begin the next development cycle (i.e. for `v4.8.0`) on the Lean repository, by making a PR that:
- Updates `src/CMakeLists.txt` to say `set(LEAN_VERSION_MINOR 8)`
- Replaces the "release notes will be copied" text in the `v4.6.0` section of `RELEASES.md` with the
finalized release notes from the `releases/v4.6.0` branch.
- Replaces the "development in progress" in the `v4.7.0` section of `RELEASES.md` with
```
Release candidate, release notes will be copied from the branch `releases/v4.7.0` once completed.
```
and inserts the following section before that section:
```
v4.8.0
----------
Development in progress.
```
- Removes all the entries from the `./releases_drafts/` folder.
- Titled "chore: begin development cycle for v4.8.0"
## Time estimates:
Slightly longer than the corresponding steps for a stable release.
Similar process, but more things go wrong.
In particular, updating the downstream repositories is significantly more work
(because we need to merge existing `bump/v4.7.0` branches, not just update a toolchain).
# Preparing `bump/v4.7.0` branches
While not part of the release process per se,
this is a brief summary of the work that goes into updating Batteries/Aesop/Mathlib to new versions.
@@ -5,7 +5,6 @@ After [building Lean](../make/index.md) you can run all the tests using
cd build/release
make test ARGS=-j4
```
Change the 4 to the maximum number of parallel tests you want to
allow. The best choice is the number of CPU cores on your machine as
the tests are mostly CPU bound. You can find the number of processors
@@ -17,6 +16,12 @@ adding the `-C stageN` argument. The default when run as above is stage 1. The
Lean tests will automatically use that stage's corresponding Lean
executables
Running `make test` will not pick up new test files; run
```bash
cmake build/release/stage1
```
to update the list of tests.
You can also use `ctest` directly if you are in the right folder. So
to run stage1 tests with a 300 second timeout run this:
@@ -24,6 +29,9 @@ to run stage1 tests with a 300 second timeout run this:
cd build/release/stage1
ctest -j 4 --output-on-failure --timeout 300
```
Useful `ctest` flags are `-R <name of test>` to run a single test, and
`--rerun-failed` to run all tests that failed during the last run.
You can also pass `ctest` flags via `make test ARGS="--rerun-failed"`.
To get verbose output from ctest pass the `--verbose` command line
option. Test output is normally suppressed and only summary
@@ -33,17 +41,17 @@ information is displayed. This option will show all test output.
All these tests are included by [src/shell/CMakeLists.txt](https://github.com/leanprover/lean4/blob/master/src/shell/CMakeLists.txt):
-`tests/lean`: contains tests that come equipped with a
.lean.expected.out file. The driver script `test_single.sh` runs
-[`tests/lean`](https://github.com/leanprover/lean4/tree/master/tests/lean/): contains tests that come equipped with a
.lean.expected.out file. The driver script [`test_single.sh`](https://github.com/leanprover/lean4/tree/master/tests/lean/test_single.sh) runs
each test and checks the actual output (*.produced.out) with the
checked in expected output.
-`tests/lean/run`: contains tests that are run through the lean
-[`tests/lean/run`](https://github.com/leanprover/lean4/tree/master/tests/lean/run/): contains tests that are run through the lean
command line one file at a time. These tests only look for error
codes and do not check the expected output even though output is
produced, it is ignored.
-`tests/lean/interactive`: are designed to test server requests at a
-[`tests/lean/interactive`](https://github.com/leanprover/lean4/tree/master/tests/lean/interactive/): are designed to test server requests at a
given position in the input file. Each .lean file contains comments
that indicate how to simulate a client request at that position.
using a `--^` point to the line position. Example:
@@ -53,7 +61,7 @@ All these tests are included by [src/shell/CMakeLists.txt](https://github.com/le
Bla.
--^ textDocument/completion
```
In this example, the test driver `test_single.sh` will simulate an
In this example, the test driver [`test_single.sh`](https://github.com/leanprover/lean4/tree/master/tests/lean/interactive/test_single.sh) will simulate an
auto-completion request at `Bla.`. The expected output is stored in
a .lean.expected.out in the json format that is part of the
[Language Server
@@ -70,21 +78,21 @@ All these tests are included by [src/shell/CMakeLists.txt](https://github.com/le
--^ collectDiagnostics
```
- `tests/lean/server`: Tests more of the Lean `--server` protocol.
- [`tests/lean/server`](https://github.com/leanprover/lean4/tree/master/tests/lean/server/): Tests more of the Lean `--server` protocol.
There are just a few of them, and it uses .log files containing
JSON.
- `tests/compiler`: contains tests that will run the Lean compiler and
- [`tests/compiler`](https://github.com/leanprover/lean4/tree/master/tests/compiler/): contains tests that will run the Lean compiler and
build an executable that is executed and the output is compared to
the .lean.expected.out file. This test also contains a subfolder
`foreign` which shows how to extend Lean using C++.
[`foreign`](https://github.com/leanprover/lean4/tree/master/tests/compiler/foreign/) which shows how to extend Lean using C++.
- `tests/lean/trust0`: tests that run Lean in a mode that Lean doesn't
- [`tests/lean/trust0`](https://github.com/leanprover/lean4/tree/master/tests/lean/trust0): tests that run Lean in a mode that Lean doesn't
- `tests/plugin`: tests that compiled Lean code can be loaded into
- [`tests/plugin`](https://github.com/leanprover/lean4/tree/master/tests/plugin/): tests that compiled Lean code can be loaded into
`lean` via the `--plugin` command line option.
## Writing Good Tests
@@ -95,7 +103,7 @@ Every test file should contain:
and, if not 100% clear, why that is the desirable behavior
At the time of writing, most tests do not follow these new guidelines yet.
For an example of a conforming test, see `tests/lean/1971.lean`.
For an example of a conforming test, see [`tests/lean/1971.lean`](https://github.com/leanprover/lean4/tree/master/tests/lean/1971.lean).
## Fixing Tests
@@ -111,7 +119,7 @@ First, we must install [meld](http://meldmerge.org/). On Ubuntu, we can do it by
sudo apt-get install meld
```
Now, suppose `bad_class.lean` test is broken. We can see the problem by going to `tests/lean` directory and
Now, suppose `bad_class.lean` test is broken. We can see the problem by going to [`tests/lean`](https://github.com/leanprover/lean4/tree/master/tests/lean) directory and
executing
```
@@ -124,8 +132,3 @@ outputs. `meld` can also be used to repair the problems.
In Emacs, we can also execute `M-x lean4-diff-test-file` to check/diff the file of the current buffer.
To mass-copy all `.produced.out` files to the respective `.expected.out` file, use `tests/lean/copy-produced`.
When using the Nix setup, add `--keep-failed` to the `nix build` call and then call
Lean is a pure functional programming language, but you can write effectful code using the `do` embedded domain specific language (DSL). The following simple program prints two strings "hello" and "world" in the standard output and terminates with exit code 0. Note that the type of the program is `IO UInt32`. You can read this type as the type of values that perform input-output effects and produce a value of type `UInt32`.
```lean
defmain:IOUInt32:=do
IO.println"hello"
IO.println"world"
return0
```
The type of `IO.println` is `String → IO Unit`. That is, it is a function from `String` to `IO Unit` which indicates it may perform input-output effects and produce a value of type `Unit`. We often say that functions that may perform effects are *methods*.
We also say a method application, such as `IO.println "hello"` is an *action*.
Note that the examples above also demonstrates that braceless `do` blocks are whitespace sensitive.
If you like `;`s and curly braces, you can write the example above as
```lean
defmain:IOUInt32:=do{
IO.println"hello";
IO.println"world";
return0;
}
```
Semicolons can be used even when curly braces are not used. They are particularly useful when you want to "pack" more than one action in a single line.
```lean
defmain:IOUInt32:=do
IO.println"hello";IO.println"world"
return0
```
Whitespace sensitivity in programming languages is a controversial topic
among programmers. You should use your own style. We, the Lean developers, **love** the
braceless and semicolon-free style.
We believe it is clean and beautiful.
The `do` DSL expands into the core Lean language. Let's inspect the different components using the commands `#print` and `#check`.
```lean
#defmain:IOUInt32:=do
#IO.println"hello"
#IO.println"world"
#return0
#checkIO.println"hello"
-- IO Unit
#printmain
-- Output contains the infix operator `>>=` and `pure`
-- The following `set_option` disables notation such as `>>=` in the output
set_optionpp.notationfalsein
#printmain
-- Output contains `bind` and `pure`
#printbind
-- bind : {m : Type u → Type v} → [self : Bind m] → {α β : Type u} →
-- m α → (α → m β) → m β
#printpure
-- pure : {m : Type u → Type v} → [self : Pure m] → {α : Type u} →
-- α → m α
-- IO implements the type classes `Bind` and `Pure`.
#check(inferInstance:BindIO)
#check(inferInstance:PureIO)
```
The types of `bind` and `pure` may look daunting at first sight.
They both have many implicit arguments. Let's focus first on the explicit arguments.
`bind` has two explicit arguments `m α` and `α → m β`. The first one should
be viewed as an action with effects `m` and producing a value of type `α`.
The second is a function that takes a value of type `α` and produces an action
with effects `m` and a value of type `β`. The result is `m β`. The method `bind` is composing
these two actions. We often say `bind` is an abstract semicolon. The method `pure` converts
a value `α` into an action that produces an action `m α`.
Here is the same function being defined using `bind` and `pure` without the `do` DSL.
```lean
defmain:IOUInt32:=
bind(IO.println"hello")fun_=>
bind(IO.println"world")fun_=>
pure0
```
The notations `let x <- action1; action2` and `let x ← action1; action2` are just syntax sugar for `bind action1 fun x => action2`.
Here is a small example using it.
```lean
defisGreaterThan0(x:Nat):IOBool:=do
IO.printlns!"value: {x}"
returnx>0
deff(x:Nat):IOUnit:=do
letc<-isGreaterThan0x
ifcthen
IO.printlns!"{x} is greater than 0"
else
pure()
#evalf10
-- value: 10
-- 10 is greater than 0
```
## Nested actions
Note that we cannot write `if isGreaterThan0 x then ... else ...` because the condition in a `if-then-else` is a **pure** value without effects, but `isGreaterThan0 x` has type `IO Bool`. You can use the nested action notation to avoid this annoyance. Here is an equivalent definition for `f` using a nested action.
```lean
#defisGreaterThan0(x:Nat):IOBool:=do
#IO.printlns!"x: {x}"
#returnx>0
deff(x:Nat):IOUnit:=do
if(<-isGreaterThan0x)then
IO.printlns!"{x} is greater than 0"
else
pure()
#printf
```
Lean "lifts" the nested actions and introduces the `bind` for us.
Here is an example with two nested actions. Note that both actions are executed
even if `x = 0`.
```lean
#defisGreaterThan0(x:Nat):IOBool:=do
#IO.printlns!"x: {x}"
#returnx>0
deff(xy:Nat):IOUnit:=do
if(<-isGreaterThan0x)&&(<-isGreaterThan0y)then
IO.printlns!"{x} and {y} are greater than 0"
else
pure()
#evalf010
-- value: 0
-- value: 10
-- The function `f` above is equivalent to
defg(xy:Nat):IOUnit:=do
letc1<-isGreaterThan0x
letc2<-isGreaterThan0y
ifc1&&c2then
IO.printlns!"{x} and {y} are greater than 0"
else
pure()
theoremfgEqual:f=g:=
rfl-- proof by reflexivity
```
Here are two ways to achieve the short-circuit semantics in the example above
```lean
#defisGreaterThan0(x:Nat):IOBool:=do
#IO.printlns!"x: {x}"
#returnx>0
deff1(xy:Nat):IOUnit:=do
if(<-isGreaterThan0x<&&>isGreaterThan0y)then
IO.printlns!"{x} and {y} are greater than 0"
else
pure()
-- `<&&>` is the effectful version of `&&`
-- Given `x y : IO Bool`, `x <&&> y` : m Bool`
-- It only executes `y` if `x` returns `true`.
#evalf1010
-- value: 0
#evalf1110
-- value: 1
-- value: 10
-- 1 and 10 are greater than 0
deff2(xy:Nat):IOUnit:=do
if(<-isGreaterThan0x)then
if(<-isGreaterThan0y)then
IO.printlns!"{x} and {y} are greater than 0"
else
pure()
else
pure()
```
## `if-then` notation
In the `do` DSL, we can write `if c then action` as a shorthand for `if c then action else pure ()`. Here is the method `f2` using this shorthand.
```lean
#defisGreaterThan0(x:Nat):IOBool:=do
#IO.printlns!"x: {x}"
#returnx>0
deff2(xy:Nat):IOUnit:=do
if(<-isGreaterThan0x)then
if(<-isGreaterThan0y)then
IO.printlns!"{x} and {y} are greater than 0"
```
## Reassignments
When writing effectful code, it is natural to think imperatively.
For example, suppose we want to create an empty array `xs`,
add `0` if some condition holds, add `1` if another condition holds,
and then print it. In the following example, we use variable
"shadowing" to simulate this kind of "update".
```lean
deff(b1b2:Bool):IOUnit:=do
letxs:=#[]
letxs:=ifb1thenxs.push0elsexs
letxs:=ifb2thenxs.push1elsexs
IO.printlnxs
#evalftruetrue
-- #[0, 1]
#evalffalsetrue
-- #[1]
#evalftruefalse
-- #[0]
#evalffalsefalse
-- #[]
```
We can use tuples to simulate updates on multiple variables.
You can capture complex control-flow using join-points.
The `do` DSL offers the variable reassignment feature to make this kind of code more comfortable to write. In the following example, the `mut` modifier at `let mut xs := #[]` indicates that variable `xs` can be reassigned. The example contains two reassignments `xs := xs.push 0` and `xs := xs.push 1`. The reassignments are compiled using join-points. There is no hidden state being updated.
```lean
deff(b1b2:Bool):IOUnit:=do
letmutxs:=#[]
ifb1thenxs:=xs.push0
ifb2thenxs:=xs.push1
IO.printlnxs
#evalftruetrue
-- #[0, 1]
```
The notation `x <- action` reassigns `x` with the value produced by the action. It is equivalent to `x := (<- action)`
## Iteration
The `do` DSL provides a unified notation for iterating over datastructures. Here are a few examples.
```lean
defsum(xs:ArrayNat):IONat:=do
letmuts:=0
forxinxsdo
IO.printlns!"x: {x}"
s:=s+x
returns
#evalsum#[1,2,3]
-- x: 1
-- x: 2
-- x: 3
-- 6
-- We can write pure code using the `Id.run <| do` DSL too.
continue-- it behaves like the `continue` statement in imperative languages
IO.printlns!"x: {x}"
s:=s+x
ifs>thresholdthen
break-- it behaves like the `break` statement in imperative languages
IO.printlns!"result: {s}"
returns
#evalsumOddUpTo[2,3,4,11,20,31,41,51,107]40
-- x: 3
-- x: 11
-- x: 31
-- result: 45
-- 45
```
TODO: describe `forIn`
## Try-catch
TODO
## Returning early from a failed match
Inside a `do` block, the pattern `let _ ← <success> | <fail>` will continue with the rest of the block if the match on the left hand side succeeds, but will execute the right hand side and exit the block on failure:
Now, we prove that if `Expr.typeCheck e` returns `Maybe.unknown`, then forall `ty`, `HasType e ty` does not hold.
The notation `e.typeCheck` is sugar for `Expr.typeCheck e`. Lean can infer this because we explicitly said that `e` has type `Expr`.
The proof is by induction on `e` and case analysis. The tactic `rename_i` is used to to rename "inaccessible" variables.
We say a variable is inaccessible if it is introduced by a tactic (e.g., `cases`) or has been shadowed by another variable introduced
by the user. Note that the tactic `simp [typeCheck]` is applied to all goal generated by the `induction` tactic, and closes
The proof is by induction on `e` and case analysis. Note that the tactic `simp [typeCheck]` is applied to all goal generated by the `induction` tactic, and closes
the cases corresponding to the constructors `Expr.nat` and `Expr.bool`.
Proving and programming are inherently interactive tasks. Lots of mathematical objects and data
structures are visual in nature. *User widgets* let you associate custom interactive UIs with
sections of a Lean document. User widgets are rendered in the Lean infoview.
Proving and programming are inherently interactive tasks.
Lots of mathematical objects and data structures are visual in nature.
*User widgets* let you associate custom interactive UIs
with sections of a Lean document.
User widgets are rendered in the Lean infoview.
## Trying it out
To try it out, simply type in the following code and place your cursor over the `#widget` command.
To try it out, type in the following code and place your cursor over the `#widget` command.
You can also [view this manual entry in the online editor](https://live.lean-lang.org/#url=https%3A%2F%2Fraw.githubusercontent.com%2Fleanprover%2Flean4%2Fmaster%2Fdoc%2Fexamples%2Fwidgets.lean).
-/
@[widget]
defhelloWidget:UserWidgetDefinitionwhere
name:="Hello"
@[widget_module]
defhelloWidget:Widget.Modulewhere
javascript:="
import * as React from 'react';
export default function(props) {
const name = props.name || 'world'
return React.createElement('p', {}, name + '!')
return React.createElement('p', {}, 'Hello ' + name + '!')
}"
#widgethelloWidget.null
#widgethelloWidget
/-!
If you want to dive into a full sample right away, check out
@@ -396,7 +396,7 @@ Every expression in Lean has a natural computational interpretation, unless it i
* *β-reduction* : An expression ``(λ x, t) s`` β-reduces to ``t[s/x]``, that is, the result of replacing ``x`` by ``s`` in ``t``.
* *ζ-reduction* : An expression ``let x := s in t`` ζ-reduces to ``t[s/x]``.
* *δ-reduction* : If ``c`` is a defined constant with definition ``t``, then ``c`` δ-reduces to to ``t``.
* *δ-reduction* : If ``c`` is a defined constant with definition ``t``, then ``c`` δ-reduces to ``t``.
* *ι-reduction* : When a function defined by recursion on an inductive type is applied to an element given by an explicit constructor, the result ι-reduces to the specified function value, as described in [Inductive Types](inductive.md).
The reduction relation is transitive, which is to say, is ``s`` reduces to ``s'`` and ``t`` reduces to ``t'``, then ``s t`` reduces to ``s' t'``, ``λ x, s`` reduces to ``λ x, s'``, and so on. If ``s`` and ``t`` reduce to a common term, they are said to be *definitionally equal*. Definitional equality is defined to be the smallest equivalence relation that satisfies all these properties and also includes α-equivalence and the following two relations:
Because `compose` is polymorphic over types ``α``, ``β``, and ``γ``, we have to provide them in the examples above.
But this information is redundant: one can infer the types from the arguments ``g`` and ``f``.
This is a central feature of dependent type theory: terms carry a lot of information, and often some of that information can be inferred from the context.
In Lean, one uses an underscore, ``_``, to specify that the system should fill in the information automatically.
All that has changed are the braces around ``α β γ: Type``.
It makes these three arguments implicit. Notationally, this hides the specification of the type,
making it look as though ``compose`` simply takes 3 arguments.
Variables can also be specified as implicit when they are declared with
the ``variable`` command:
```lean
universe u
section
variable {α : Type u}
variable (x : α)
def ident := x
end
variable (α β : Type u)
variable (a : α) (b : β)
#check ident
#check ident a
#check ident b
```
This definition of ``ident`` here has the same effect as the one above.
Lean has very complex mechanisms for instantiating implicit arguments, and we will see that they can be used to infer function types, predicates, and even proofs.
The process of instantiating these "holes," or "placeholders," in a term is part of a bigger process called *elaboration*.
The presence of implicit arguments means that at times there may be insufficient information to fix the meaning of an expression precisely.
An expression like ``ident`` is said to be *polymorphic*, because it can take on different meanings in different contexts.
One can always specify the type ``T`` of an expression ``e`` by writing ``(e : T)``.
This instructs Lean's elaborator to use the value ``T`` as the type of ``e`` when trying to elaborate it.
In the following example, this mechanism is used to specify the desired types of the expressions ``ident``.
```lean
def ident {α : Type u} (a : α) : α := a
#check (ident : Nat → Nat) -- Nat → Nat
```
Numerals are overloaded in Lean, but when the type of a numeral cannot be inferred, Lean assumes, by default, that it is a natural number.
So the expressions in the first two ``#check`` commands below are elaborated in the same way, whereas the third ``#check`` command interprets ``2`` as an integer.
```lean
#check 2 -- Nat
#check (2 : Nat) -- Nat
#check (2 : Int) -- Int
```
Sometimes, however, we may find ourselves in a situation where we have declared an argument to a function to be implicit,
but now want to provide the argument explicitly. If ``foo`` is such a function, the notation ``@foo`` denotes the same function with all
the arguments made explicit.
```lean
# def ident {α : Type u} (a : α) : α := a
variable (α β : Type)
#check @ident -- {α : Type u} → α → α
#check @ident α -- α → α
#check @ident β -- β → β
#check @ident Nat -- Nat → Nat
#check @ident Bool true -- Bool
```
Notice that now the first ``#check`` command gives the type of the identifier, ``ident``, without inserting any placeholders.
Moreover, the output indicates that the first argument is implicit.
Named arguments enable you to specify an argument for a parameter by matching the argument with
its name rather than with its position in the parameter list. You can use them to specify explicit *and* implicit arguments.
If you don't remember the order of the parameters but know their names, you can send the arguments in any order.
You may also provide the value for an implicit parameter when
Lean failed to infer it. Named arguments also improve the readability of your code by identifying what
These are instructions to set up a working development environment for those who wish to make changes to Lean itself. It is part of the [Development Guide](../dev/index.md).
We strongly suggest that new users instead follow the [Quickstart](../quickstart.md) to get started using Lean, since this sets up an environment that can automatically manage multiple Lean toolchain versions, which is necessary when working within the Lean ecosystem.
make -C build/release -j$(nproc || sysctl -n hw.logicalcpu)
```
For regular development, we recommend running
```bash
git config submodule.recurse true
```
in the checkout so that `--recurse-submodules` doesn't have to be
specified with `git pull/checkout/...`.
You can replace `$(nproc || sysctl -n hw.logicalcpu)` with the desired parallelism amount.
The above commands will compile the Lean library and binaries into the
`stage1` subfolder; see below for details. Add `-j N` for an
appropriate `N` to `make` for a parallel build.
`stage1` subfolder; see below for details.
For example, on an AMD Ryzen 9`make` takes 00:04:55, whereas `make -j 10`
takes 00:01:38. Your results may vary depending on the speed of your hard
drive.
You should not usually run `make install` after a successful build.
You should not usually run`cmake --install` after a successful build.
See [Dev setup using elan](../dev/index.md#dev-setup-using-elan) on how to properly set up your editor to use the correct stage depending on the source directory.
Useful CMake Configuration Settings
-----------------------------------
Pass these along with the `cmake ../..` command.
Pass these along with the `cmake --preset release` command.
There are also two alternative presets that combine some of these options you can use instead of `release`: `debug` and `sandebug` (sanitize + debug).
*`-D CMAKE_BUILD_TYPE=`\
Select the build type. Valid values are `RELEASE` (default), `DEBUG`,
While [Nix](https://nixos.org/nix/) can be used to quickly open a shell with all dependencies for the [standard setup](index.md) installed, the user-facing [Nix Setup](../setup.md#nix-setup) can also be used to work *on* Lean.
## Setup
Follow the setup in the link above; to open the Lean shell inside a Lean checkout, you can also use
```bash
# in the Lean root directory
$ nix-shell -A nix
```
On top of the local and remote Nix cache, we do still rely on CCache as well to make C/C++ build steps incremental, which are atomic steps from Nix's point of view.
To enable CCache, add the following line to the config file mentioned in the setup:
```bash
extra-sandbox-paths = /nix/var/cache/ccache
```
Then set up that directory as follows:
```bash
sudo mkdir -m0770 -p /nix/var/cache/ccache
# macOS standard chown doesn't support --reference
From the Lean root directory inside the Lean shell:
```bash
nix build .#stage1 # build this stage's stdlib & executable
nix build .#stage1.test # run all tests
nix run .#stage1.update-stage0 # update ./stage0 from this stage
nix run .#stage1.update-stage0-commit # ...and commit the results
```
The `stage1.` part in each command is optional:
```bash
nix build .#test # run tests for stage 1
nix build . # build stage 1
nix build # ditto
```
## Build Process Description
The Nix build process conceptually works the same as described in [Lean Build Pipeline](index.md#lean-build-pipeline).
However, there are two important differences in practice apart from the standard Nix properties (hermeneutic, reproducible builds stored in a global hash-indexed store etc.):
* Only files tracked by git (using `git add` or at least `git add --intent-to-add`) are compiled.
This is actually a general property of Nix flakes, and has the benefit of making it basically impossible to forget to commit a file (at least in `src/`).
* Only files reachable from `src/Lean.lean` are compiled.
This is because modules are discovered not from a directory listing anymore but by recursively compiling all dependencies of that top module.
## Editor Integration
As in the standard Nix setup.
After adding `src/` as an LSP workspace, it should automatically fall back to using stage 0 in there.
Note that the UX of `{emacs,vscode}-dev` is quite different from the Make-based setup regarding the compilation of dependencies:
there is no mutable directory incrementally filled by the build that we could point the editor at for .olean files.
Instead, `emacs-dev` will gather the individual dependency outputs from the Nix store when checking a file -- and build them on the fly when necessary.
However, it will only ever load changes saved to disk, not ones opened in other buffers.
The absence of a mutable output directory also means that the Lean server will not automatically pick up `.ilean` metadata from newly compiled files.
Instead, you can run `nix run .#link-ilean` to symlink the `.ilean` tree of the stdlib state at that point in time to `src/build/lib`, where the server should automatically find them.
## Other Fun Stuff to Do with Nix
Open Emacs with Lean set up from an arbitrary commit (without even cloning Lean beforehand... if your Nix is new enough):
```bash
nix run github:leanprover/lean4/7e4edeb#emacs-package
```
Open a shell with `lean` and `LEAN_PATH` set up for compiling a specific module (this is exactly what `emacs-dev` is doing internally):
```bash
nix develop .#mods.\"Lean.Parser.Basic\"
# alternatively, directly pass a command to execute:
In this section we introduce some organizational features of Lean that are not a part of its kernel per se,
but make it possible to work in the framework more efficiently.
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
