Fri, 25 Jul 2014 11:31:20 +0200 blanchet added missing facts to proof method
Fri, 25 Jul 2014 11:26:23 +0200 blanchet don't lose 'minimize' flag before it reaches Isar proof text generation
Fri, 25 Jul 2014 11:26:19 +0200 blanchet tuning
Fri, 25 Jul 2014 11:26:17 +0200 blanchet avoid 'eproof' and 'eproof_ram' scripts if possible (i.e. if 'eprover' can produce reasonable enough proofs for one-liner reconstruction)
Fri, 25 Jul 2014 11:26:11 +0200 blanchet compile
Fri, 25 Jul 2014 11:26:11 +0200 blanchet more robustness in Isar proof construction
Fri, 25 Jul 2014 11:26:10 +0200 blanchet tuning
Thu, 24 Jul 2014 23:18:01 +0200 wenzelm merged
Thu, 24 Jul 2014 23:17:26 +0200 wenzelm more elementary exception handling: evade hard crash of (Runtime.thread true undefined) on Poly/ML 5.5.1 and 5.5.2;
Thu, 24 Jul 2014 23:01:23 +0200 blanchet tuned code
Thu, 24 Jul 2014 20:21:59 +0200 kuncar having extra assumptions (typically from a context) means there is no chance to have a valid code equation => skip decoding and registration of the code equations
Thu, 24 Jul 2014 20:21:34 +0200 kuncar store explicitly quotient types with no_code => more precise registration of code equations
Thu, 24 Jul 2014 19:01:06 +0200 blanchet don't needlessly regenerate entire file when the time stamps are equal
Thu, 24 Jul 2014 18:53:14 +0200 blanchet eliminated source of 'DUP's in MaSh
Thu, 24 Jul 2014 18:46:38 +0200 blanchet fixed sorting (broken since 9cc802a8ab06)
Thu, 24 Jul 2014 18:46:38 +0200 blanchet reenabled MaSh for Isabelle2014 release (hopefully)
Thu, 24 Jul 2014 18:46:38 +0200 blanchet beware of duplicate fact names
Thu, 24 Jul 2014 18:46:38 +0200 blanchet refined filter for ATP steps to avoid 'have True' steps in E proofs
Thu, 24 Jul 2014 18:46:38 +0200 blanchet filter out 'theory(...)' from dependencies early on
Thu, 24 Jul 2014 18:46:38 +0200 blanchet introduce fact chaining also under first step
Thu, 24 Jul 2014 18:46:38 +0200 blanchet 'shift_quantors' is not an E skolemization rule (cf. 3ab503b04bdb)
Thu, 24 Jul 2014 17:58:29 +0200 wenzelm merged
Thu, 24 Jul 2014 17:13:26 +0200 wenzelm proper perl;
Thu, 24 Jul 2014 17:11:40 +0200 wenzelm less warnings -- ignore potential prover startup/shutdown races;
Thu, 24 Jul 2014 16:21:50 +0200 wenzelm tuned spelling;
Thu, 24 Jul 2014 15:54:56 +0200 wenzelm further distinction of Isabelle distribution: alert for identified release candidates;
Thu, 24 Jul 2014 15:13:37 +0200 wenzelm updated to scala-2.11.2;
Thu, 24 Jul 2014 15:01:17 +0200 wenzelm clarified file names;
Thu, 24 Jul 2014 14:56:30 +0200 wenzelm less ambitious isatest, avoid "Exception- InternalError: Backing up too far (32bit) raised while compiling" in polyml-5.4.1;
Thu, 24 Jul 2014 14:08:29 +0200 wenzelm tuned imports;
Thu, 24 Jul 2014 14:04:55 +0200 wenzelm proper scope of comments;
Thu, 24 Jul 2014 13:48:00 +0200 wenzelm make SML/NJ happy;
Thu, 24 Jul 2014 13:01:49 +0200 kuncar prevent beta-contraction in proving extra assumptions for abs_eq
Thu, 24 Jul 2014 11:54:15 +0200 wenzelm more robust notation BNF_Def.convol, which is private to main HOL, but may cause syntax ambiguities nonetheless (e.g. List.thy);
Thu, 24 Jul 2014 11:51:22 +0200 wenzelm reconfirm continuous checking on startup, to address common trap of disabling it accidentally;
Thu, 24 Jul 2014 11:46:40 +0200 wenzelm tuned;
Thu, 24 Jul 2014 10:38:46 +0200 wenzelm less authentic SHA1.digest: trust Scala side on blobs and avoid re-calculation via Foreign Language Interface, which might be a cause of problems;
Thu, 24 Jul 2014 10:22:34 +0200 wenzelm updated NEWS according to d38a98f496dd (see also bdc2c6b40bf2);
Thu, 24 Jul 2014 00:24:00 +0200 blanchet stick to external proofs when invoking E, because they are more detailed and do not merge steps
Thu, 24 Jul 2014 00:24:00 +0200 blanchet more robust handling of types for skolems (modeled as Frees)
Thu, 24 Jul 2014 00:24:00 +0200 blanchet tuning
Thu, 24 Jul 2014 00:24:00 +0200 blanchet repaired named derivations
Thu, 24 Jul 2014 00:24:00 +0200 blanchet use the noted theorems in 'BNF_Comp'
Thu, 24 Jul 2014 00:24:00 +0200 blanchet use the noted theorem whenever possible, also in 'BNF_Def'
Thu, 24 Jul 2014 00:24:00 +0200 blanchet use termtab instead of (perhaps overly sensitive) thmtab
Thu, 24 Jul 2014 00:24:00 +0200 blanchet use the noted theorem whenever possible, because it has a named derivation (leading to cleaner proof terms)
Wed, 23 Jul 2014 23:16:44 +0200 wenzelm tuned message;
Wed, 23 Jul 2014 23:08:22 +0200 wenzelm added action "isabelle.options" (despite problems with initial window size);
Wed, 23 Jul 2014 21:02:45 +0200 wenzelm more official Thy_Info.script_thy;
Wed, 23 Jul 2014 21:01:28 +0200 wenzelm more frugal edits;
Wed, 23 Jul 2014 18:16:04 +0200 wenzelm enable hires explictly, as seen for other high-end Java applications on the Web;
Wed, 23 Jul 2014 18:14:59 +0200 wenzelm more markup;
Wed, 23 Jul 2014 18:04:16 +0200 wenzelm another attempt at more aggressive auto-loading (amending af28fdd50690) -- hidden buffers are now suppressed;
Wed, 23 Jul 2014 16:56:03 +0200 wenzelm more frugal edits;
Wed, 23 Jul 2014 16:20:07 +0200 wenzelm more explicit treatment of cleared nodes (removal is implicit);
Wed, 23 Jul 2014 15:32:05 +0200 wenzelm clarified display;
Wed, 23 Jul 2014 15:11:42 +0200 wenzelm more workarounds for scalac;
Wed, 23 Jul 2014 15:00:46 +0200 wenzelm clarified display;
Wed, 23 Jul 2014 14:50:20 +0200 wenzelm avoid redundant data structure;
Wed, 23 Jul 2014 13:01:30 +0200 wenzelm more explicit discrimination of empty nodes -- suppress from Theories panel;
(0) -30000 -10000 -3000 -1000 -300 -100 -60 +60 +100 +300 +1000 +3000 +10000 tip