This is really easy to add to Abella. See attached patch and example.
The patch can probably be refactored with a small amount of extra work.
Unfortunately, it confuses the ProofGeneral mode even more than usual.
-- Kaustuv
(PS, sorry for duplicates.)
You may also want to give some thought to how these two specification
methods interact. For example, what if you Import a .thc file that was
compiled from a .thm file that has an inline specification? Can inline
specifications extend .mod/.sig and/or other inline specifications? And
so on.
--Todd
I think the idea of inline specifications is fine. The concerns Todd
raises about correctness and interactions with other module mechanics
shouldn't be an issue. My concerns are
1. Right now the code isn't refactored and I don't have time to do it.
If I integrate the patch without it, there's a fair chance that
inconsistencies get introduced by later code changes to only one of
the specification mechanisms.
2. The HTML formatter for thm files is fairly dumb and just breaks
things up based on the periods at the end of each command. It would
require a decent amount of work to update it to handle inline
specifications, so I probably won't do that.
-Andrew
> --
> You received this message because you are subscribed to the Google Groups "Abella" group.
> To post to this group, send email to abella-the...@googlegroups.com.
> To unsubscribe from this group, send email to abella-theorem-p...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/abella-theorem-prover?hl=en.
>
>