-
-
Notifications
You must be signed in to change notification settings - Fork 81
Open
Description
There is an odd interaction between Rewrap and VsRocq (https://github.com/rocq-prover/vsrocq):
When I type a comment like this
(** The lemma statement is carefully chosen: we will use this lemma in cases
where [e] is known, so [to_val e] will simplify. We will [eapply] the lemma,
making [v] an evar, which can then be trivially instantiated after
simplification. If we had use [of_val v = e] as the premise, no simplification
would have been possible. *)
and then rewrap 3 times, the result is:
(** The lemma statement is carefully chosen: we will use this lemma in cases
where [e] is known, so [to_val e] will simplify. We will [eapply] the
lemma, making [v] an evar, which can then be trivially instantiated after
simplification. If we had use [of_val v = e] as the premise, no
simplification would have been possible. *)
The indentation increases by 2 spaces every time I rewrap. Rewrapping should be idempotent (wrapping twice should do the same as wrapping once), so something seems off here. However, strangely I only see this happen in .v files (the extension used by Rocq), not with other file types.
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels