Sat, 14 Feb 2015 10:24:15 +0100 haftmann tuned
Sat, 14 Feb 2015 10:24:15 +0100 haftmann more consistent teminology
Sat, 14 Feb 2015 10:24:15 +0100 haftmann fact consolidation
Sat, 14 Feb 2015 10:24:15 +0100 haftmann dropped redundancy
Sat, 14 Feb 2015 10:24:15 +0100 haftmann less warnings
Fri, 13 Feb 2015 23:39:25 +0100 wenzelm merged
Wed, 11 Feb 2015 18:16:33 +0100 wenzelm proper context;
Wed, 11 Feb 2015 11:42:39 +0100 wenzelm proper context;
Wed, 11 Feb 2015 11:26:17 +0100 wenzelm proper context;
Wed, 11 Feb 2015 11:18:36 +0100 wenzelm proper context;
Wed, 11 Feb 2015 10:54:53 +0100 wenzelm proper context and method setup;
Thu, 12 Feb 2015 12:46:50 +0000 paulson new lemmas re refinement of one equivalence relation WRT another
Wed, 11 Feb 2015 18:39:56 +0100 Andreas Lochbihler rel_pmf preserves orders
Wed, 11 Feb 2015 18:38:34 +0100 Andreas Lochbihler generalise lemma
Wed, 11 Feb 2015 15:22:37 +0100 Andreas Lochbihler more lemmas
Wed, 11 Feb 2015 15:04:23 +0100 Andreas Lochbihler merged
Wed, 11 Feb 2015 15:03:21 +0100 Andreas Lochbihler more transfer rules
Wed, 11 Feb 2015 14:45:10 +0100 Andreas Lochbihler add lemmas about functions on option
Wed, 11 Feb 2015 14:19:46 +0100 Andreas Lochbihler tuned proof
Wed, 11 Feb 2015 14:15:59 +0100 Andreas Lochbihler add lema about card and vimage
Wed, 11 Feb 2015 14:12:30 +0100 Andreas Lochbihler add more general version of finite_vimageD
Wed, 11 Feb 2015 14:07:28 +0100 Andreas Lochbihler add lemma
Wed, 11 Feb 2015 14:03:05 +0100 Andreas Lochbihler add lemmas about flat_ord
Wed, 11 Feb 2015 13:58:51 +0100 Andreas Lochbihler add parametricity rules for monotone, fun_lub, and fun_ord
Wed, 11 Feb 2015 13:52:12 +0100 Andreas Lochbihler add parametricity rule for Ex1
Wed, 11 Feb 2015 13:51:16 +0100 Andreas Lochbihler add intro and elim rules for right_total
Wed, 11 Feb 2015 13:50:11 +0100 Andreas Lochbihler add monotonicity lemmas for rel_fun
Wed, 11 Feb 2015 13:47:48 +0100 Andreas Lochbihler add lemmas about bind and image
Wed, 11 Feb 2015 14:53:56 +0100 blanchet updated NEWS
Wed, 11 Feb 2015 14:51:36 +0100 blanchet updated Sledgehammer docs
Wed, 11 Feb 2015 14:48:07 +0100 blanchet added CVC4 component (and took out CVC3 from main components)
Wed, 11 Feb 2015 14:48:06 +0100 blanchet tuned default provers
Wed, 11 Feb 2015 12:01:56 +0000 paulson Merge
Tue, 10 Feb 2015 17:37:06 +0000 paulson Not a simprule, as it complicates proofs
Tue, 10 Feb 2015 16:09:30 +0000 paulson Merge
Tue, 10 Feb 2015 16:08:11 +0000 paulson New lemmas and a bit of tidying up.
Tue, 10 Feb 2015 23:02:39 +0100 wenzelm check unused theory;
Tue, 10 Feb 2015 22:52:44 +0100 wenzelm tuned;
Tue, 10 Feb 2015 20:51:43 +0100 wenzelm more accurate context;
Tue, 10 Feb 2015 17:13:23 +0100 wenzelm merged
Tue, 10 Feb 2015 16:46:21 +0100 wenzelm misc tuning;
Tue, 10 Feb 2015 14:48:26 +0100 wenzelm proper context for resolve_tac, eresolve_tac, dresolve_tac, forward_tac etc.;
Tue, 10 Feb 2015 14:29:36 +0100 wenzelm indicate slow proof (approx. 20s);
Tue, 10 Feb 2015 14:06:57 +0100 hoelzl merged
Tue, 10 Feb 2015 13:50:30 +0100 hoelzl add bind_cond_pmf_cancel
Tue, 10 Feb 2015 12:15:05 +0100 hoelzl add cond_map_pmf
Tue, 10 Feb 2015 12:09:32 +0100 hoelzl introduce discrete conditional probabilities, use it to simplify bnf proof of pmf
Tue, 10 Feb 2015 12:27:30 +0100 Andreas Lochbihler tuned proof
Tue, 10 Feb 2015 12:17:22 +0100 Andreas Lochbihler add another lemma to split nn_integral over product count_space
Tue, 10 Feb 2015 12:10:26 +0100 Andreas Lochbihler tune proof
Tue, 10 Feb 2015 12:05:21 +0100 Andreas Lochbihler nn_integral can be split over arbitrary product count_spaces
Tue, 10 Feb 2015 12:04:24 +0100 Andreas Lochbihler add stronger version of lemma
Fri, 06 Feb 2015 17:57:03 +0100 haftmann default abstypes and default abstract equations make technical (no_code) annotation superfluous
Fri, 06 Feb 2015 19:17:17 +0100 blanchet careful about visibility of facts that have the same 'theory' in optimization
Fri, 06 Feb 2015 08:47:48 +0100 haftmann non-intrusive default code setup for mappings
Thu, 05 Feb 2015 19:44:14 +0100 haftmann slightly more standard code setup for String.literal, with explicit special case in predicate compiler
Thu, 05 Feb 2015 19:44:13 +0100 haftmann explicit type annotation avoids problems with Haskell type inference
Thu, 05 Feb 2015 13:01:12 +0100 haftmann more explicit hint on import order
Thu, 05 Feb 2015 13:01:12 +0100 haftmann dropped dead code;
Thu, 05 Feb 2015 13:01:12 +0100 haftmann dropped obsolete external entrance point
(0) -30000 -10000 -3000 -1000 -300 -100 -60 +60 +100 +300 +1000 +3000 +10000 tip