blanchet [Thu, 24 May 2012 17:46:35 +0200] rev 47990
fixed soundness bug in Nitpick related to unfolding -- the unfolding criterion must at least as strict when looking at a definitional axiom as elsewhere, otherwise we end up unfolding a constant's definition in its own definition, yielding a trivial equality
wenzelm [Thu, 24 May 2012 17:42:47 +0200] rev 47989
more robust Swing_Thread.now: allow body to fail;
wenzelm [Thu, 24 May 2012 17:25:53 +0200] rev 47988
tuned proofs;
wenzelm [Thu, 24 May 2012 17:05:30 +0200] rev 47987
reparse descendants after update of imports, e.g. required for 'primrec' (line 17 of "src/HOL/Power.thy");
wenzelm [Thu, 24 May 2012 15:54:38 +0200] rev 47986
merged
blanchet [Thu, 24 May 2012 15:11:53 +0200] rev 47985
update Satallax setup based on evaluation
kuncar [Thu, 24 May 2012 15:03:06 +0200] rev 47984
merged
kuncar [Thu, 24 May 2012 14:20:25 +0200] rev 47983
drop the feature that more than one quotient type can be defined by quotient_type -> it causes problems
kuncar [Thu, 24 May 2012 14:20:23 +0200] rev 47982
prove reflexivity also for the quotient composition relation; reflp_preserve renamed to reflexivity_rule
blanchet [Thu, 24 May 2012 15:01:29 +0200] rev 47981
gracefully handle definition-looking premises