Zulip Chat Archive

Stream: Is there code for X?

Topic: a tactic that does not need to backtrack


Paige Thomas (Jun 08 2024 at 20:11):

Is there a nonclosing tactic that will apply all of the tactics that would never need to backtrack? For example, splitting and, or and iff in the hypotheses, existential elimination of existential formulas in the hypothesis, calling intro on implication and forall in the goal, substituting equalities in the hypotheses, splitting if, then, else, etc.? And maybe a way to tag tactics for its use?

Yaël Dillies (Jun 08 2024 at 20:12):

That sounds like aesop configured to be safe

Paige Thomas (Jun 08 2024 at 20:12):

oh?

Paige Thomas (Jun 08 2024 at 20:13):

I'll take a look. Thank you!

Yaël Dillies (Jun 08 2024 at 20:14):

Something like aesop (config := { warnOnNonterminal := false, enableSimp := false })

Yaël Dillies (Jun 08 2024 at 20:14):

You will want to have a look at docs#Aesop.Options

Paige Thomas (Jun 08 2024 at 20:15):

Is there an intro guide to aesop in general?

Yaël Dillies (Jun 08 2024 at 20:16):

Eh, not really. Best it to ask Jannis Limperg

Yaël Dillies (Jun 08 2024 at 20:16):

There is a paper about it and also talks, eg https://limperg.de/talk/bicmr-202404/slides.pdf

Paige Thomas (Jun 08 2024 at 20:18):

Ok. Thank you.

Kevin Buzzard (Jun 09 2024 at 06:24):

I have found the project README (linked to in the module docstring) very informative as an intro guide


Last updated: May 02 2025 at 03:31 UTC