Zulip Chat Archive
Stream: general
Topic: Tool for extracting current goal into a dedicated lemma?
aprilgrimoire (Dec 08 2024 at 12:19):
Hi! Sometimes I need to write long proofs due to things with many parameters, for example, Subfield.closure_induction
. In such cases, I need to split some of the proof into a dedicated lemma to prevent timeout at whnf
. Just like simp?
, is there a tactic or tool that could help me formulate such lemmas?
Luigi Massacci (Dec 08 2024 at 12:34):
Yes, it's called extract_goal
; )
Luigi Massacci (Dec 08 2024 at 12:35):
https://leanprover-community.github.io/mathlib4_docs/Mathlib/Tactic/ExtractGoal.html
Eric Wieser (Dec 08 2024 at 12:40):
Note that when using Subfield.closure_induction
, you probably want to write it using induction hx using Subfield.closure_induction
, which might also help simplify your proof / make it easier to follow.
aprilgrimoire (Dec 08 2024 at 12:41):
Thanks!
image.png
Why did it contain omits? Where can I see the full version?
Eric Wieser (Dec 08 2024 at 12:43):
I think if you hover over the ...
, lean will tell you the setting to make it show that detail
aprilgrimoire (Dec 08 2024 at 12:44):
Yeah, but then I can't simply copy the statement out. How can I invoke the linter in a terminal?
aprilgrimoire (Dec 08 2024 at 12:46):
Oh sorry seems the link you provided contained info about it
Eric Wieser (Dec 08 2024 at 13:19):
I don't really understand your question; did you solve it?
Eric Wieser (Dec 08 2024 at 13:20):
In the particular example you gave above, I think running dsimp
before extract_goal
will help
aprilgrimoire (Dec 10 2024 at 18:21):
Yeah I solved it. set_option pp.proofs true in extract_goal
worked. Sorry for the late reply, I was working on an assignment yesterday.
Last updated: May 02 2025 at 03:31 UTC