Commit graph

20 commits

Author SHA1 Message Date
JovanGerb
c96797eb93
fix: show argument name in implicit argument error (#4426)
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.
2024-06-14 18:08:42 +00:00
Sebastian Ullrich
bddb2152e5
chore: default compiler.enableNew to false until development restarts (#3034) 2023-12-21 07:48:25 +00:00
Leonardo de Moura
ec42581d1f perf: use Core.transform instead of Meta.transform at betaReduceLetRecApps
Observed big performance improvement on files containing big proofs
generated using tactics.
2023-06-28 12:29:12 -07:00
Gabriel Ebner
a67a5080e9 chore: fix tests after hash change 2022-12-01 20:18:14 -08:00
Leonardo de Moura
5a151ca64c chore: fix tests 2022-11-30 17:52:37 -08:00
Leonardo de Moura
dfd95c712e chore: fix tests 2022-10-13 06:16:56 -07:00
Gabriel Ebner
15d7744cca chore: fix tests 2022-10-11 17:24:35 -07:00
Leonardo de Moura
ce90e98648 feat: activate new compiler first phase 2022-09-24 14:20:21 -07:00
Leonardo de Moura
5d8b4f33c0 feat: improve binder names introduced by the match discriminant refinement feature 2022-04-08 06:49:09 -07:00
Leonardo de Moura
fa9b0f6c7e feat: remove space before propositions with inaccessible names 2022-04-07 07:54:50 -07:00
Leonardo de Moura
9fe5458077 feat: do not display inaccessible proposition names if they do not have forward dependencies
Even if `pp.inaccessibleNames = true`
2022-04-02 13:15:17 -07:00
Leonardo de Moura
46ce3013d0 feat: cleanup local context before elaborating match alternatives RHS 2022-03-29 18:52:07 -07:00
Leonardo de Moura
ca253c43e1 refactor: pattern elaboration
We don't use the following hack anymore:
-  /- HACK: `fvarId` is not in the scope of `mvarId`
-     If this generates problems in the future, we should update the metavariable declarations. -/
-  assignExprMVar mvarId (mkFVar fvarId)

This hack was corrupting the `InfoTree`.
2022-03-09 18:19:14 -08:00
Leonardo de Moura
b6dc3dda04 chore: fix tests 2022-03-09 14:02:13 -08:00
Leonardo de Moura
c93982d10c chore: disable pp.analyze for now
see #651
2021-09-07 07:51:43 -07:00
Daniel Selsam
c3d62c1076 chore: patch tests for pp.analyze default 2021-08-03 09:13:18 +02:00
Leonardo de Moura
030f53fa43 fix: closes #421
The unifier used to implement the `cases` tactic should not discard
equations of the form `x = t` and `t = x` using proof irrelanvance.
The new test demonstrates the issue. The unifier was reaching the
state
```
x : Conw Con.nil
|- x = Conw.nilw -> x = Conw.nilw
```
and discarding the equality instead of substituting `x`
because `x` and `Conw.nilw` are definionally equal due to
proof irrelevance.

@javra Do you have more complicated examples that were being
affected by this issue?
2021-04-23 12:27:39 -07:00
Leonardo de Moura
51e7f45af2 refactor: cases
remove dead `induction/cases` code
2021-03-07 12:37:02 -08:00
Sebastian Ullrich
b2b78eb222 test: use printMessageEndPos for leantests 2021-01-15 16:27:59 +01:00
Leonardo de Moura
975271e1a3 feat: hide inaccessible names and add pp.inaccessibleNames
@Kha We will probably have to refine the heuristic for hiding the
inaccessible names, but the first version is already useful.
Here is the error message for a `_` before this commit
```
error: don't know how to synthesize placeholder
context:
x✝⁴ : List Nat
x✝³ : Nat
x✝² : x✝⁴ ≠ []
a b x✝¹ : Nat
x✝ : [a, b] ≠ []
⊢ Nat
```
After
```
error: don't know how to synthesize placeholder
a b : Nat
 : [a, b] ≠ []
⊢ Nat
```
2020-11-25 17:22:09 -08:00