coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: Thery Laurent <thery AT ns.di.univaq.it>
- To: coq-club AT pauillac.inria.fr
- Subject: [Coq-Club] Tactics with "with" extension
- Date: Mon, 3 Nov 2003 12:35:23 +0100 (CET)
- List-archive: <http://pauillac.inria.fr/pipermail/coq-club/>
Hi,
I'm findind the with extension for writing tactics very useful.
For example
Repeat Rewrite My_theorem with a := a
instantiate my theorem so to restrict the possible rewriting
Apply My_theorem with 1:= H
instantiante first a theorem with the local context before applying
it.
I'm wondering why this is not allowed for all tactics?
For example
Generalize My_theorem with 1:= H.
would shorten the ugly Generalize [x,y,z:?] (My_theorem x y z H).
As a matter of fact, the best would be that the with extension
would be considered not a the tactic level but at the term level.
So
Apply My_theorem with 1:= H.
would mean
Apply (My_theorem with 1:= H).
and in particular
(My_theorem with 1:= H)
would be a licite term. This would have the advantage to make
the with extension available for homemade tactics.
--
Laurent Thery
- [Coq-Club] Tactics with "with" extension, Thery Laurent
Archive powered by MhonArc 2.6.16.