f3cb8a6c2d
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.
57 lines
1.1 KiB
Text
57 lines
1.1 KiB
Text
/-!
|
|
Comments after a command may become part of the next command on edits.
|
|
(Note that this module doc is a command on its own)
|
|
-/
|
|
|
|
--v sync
|
|
--v insert: "-"
|
|
/-
|
|
info: "3.14"
|
|
-/
|
|
#guard_msgs in
|
|
#eval "3.14"
|
|
--^ collectDiagnostics
|
|
-- (should be empty if edit was handled correctly)
|
|
|
|
/-! Same, after header -/
|
|
-- RESET
|
|
import Init
|
|
|
|
--v sync
|
|
--v insert: "-"
|
|
/-
|
|
info: "3.14"
|
|
-/
|
|
#guard_msgs in
|
|
#eval "3.14"
|
|
--^ collectDiagnostics
|
|
-- (should be empty if edit was handled correctly)
|
|
|
|
/-! Commands not marked as `[incremental]` should not allow accidental reuse in unknown contexts. -/
|
|
-- RESET
|
|
import Lean
|
|
|
|
open Lean Elab Command in
|
|
elab "wrap" num c:command : command => do
|
|
elabCommand c
|
|
|
|
--v change: "1" "2"
|
|
wrap 1 def wrapped := by
|
|
dbg_trace "w"
|
|
|
|
/-!
|
|
The example used to result in nothing but "declaration uses 'sorry'" (and using the downstream
|
|
"unreachable tactic" linter, the `simp` would be flagged) as `simp` among other elaborators
|
|
accidentally swallowed the interrupt exception.
|
|
-/
|
|
-- RESET
|
|
import Lean
|
|
|
|
open Lean Elab Core in
|
|
elab "interrupt" : tactic =>
|
|
throw <| .internal interruptExceptionId
|
|
|
|
example : False := by
|
|
interrupt
|
|
simp
|
|
--^ collectDiagnostics
|