executable true liveness analysis incl an approximating version

now that sets are executable again, no more special treatment of variable sets

handle non-unit clauses gracefully

several improvements to Isar proof reconstruction, by Steffen Smolka (step merging in case splits, time measurements, etc.)

use measurability prover

add measurability prover; add support for Borel sets

add syntax and a.e.-rules for (conditional) probability on predicates

infinite product measure is invariant under adding prefixes

for the product measure it is enough if only one measure is sigma-finite

Allow parentheses around left-hand sides of array associations

made MaSh more robust in the face of duplicate "nicknames" (which can happen e.g. if you have a lemma called foo(1) and another called foo_1 in the same theory)

regenerated SMT certificates

regenerated "SMT_Examples" certificates after soft-timeout change + removed a few needless oracles

fixed bool vs. prop mismatch

removed "refute" command from Isar manual, now that it has been moved outside "Main"

repaired "Mutabelle" after Refute move

less verbose -- the warning will reach the users anyway by other means

tuned messages

moved "SAT" before "FunDef" and moved back all SAT-related ML files to where they belong

fixes related to Refute's move

added a timeout around script that relies on the network

took out "using only ..." comments in Sledgehammer generated metis/smt calls, until these can be generated soundly

moved Refute to "HOL/Library" to speed up building "Main" even more

tuning

use metaquantification when possible in Isar proofs

tuned code

tuning

soft SMT timeout

added function store_termination_rule to the signature, as it is used in Nominal2

longer log, to accomodate final status line of isabelle build;

transfer package: error message if preprocessing goal to object-logic formula fails

transfer package: add test to prevent trying to make cterms from open terms

transfer package: more flexible handling of equality relations using is_equality predicate

ensured that rewr_conv rule t = "t == u" literally not just modulo beta-eta

new theorems

incorporated constant chars into instantiation proof for enum; tuned proofs for properties on enum of chars; swapped theory dependency of Enum.thy and String.thy

close code theorems explicitly after preprocessing

tuned proofs;

further attempts to cope with large files via option jedit_text_overview_limit;

more detailed Prover IDE NEWS;

recovered explicit error message, which was lost in b8570ea1ce25;

removed dead code;

proper signatures;

tuned;

merged

Renamed {left,right}_distrib to distrib_{right,left}.

Tuned.

more conventional argument order; dropped dead code

another refinement in the comprehension conversion

refined simplifier call in comprehension_conv

passing around the simpset instead of the context; rewriting tactics to avoid the 'renamed bound variable' warnings in nested simplifier calls

avoid STIX font, which tends to render badly;

extra jar for scala-2.10.0-RC1;

more explicit auxiliary classes to avoid warning "reflective access of structural type member method" of scala-2.10.0-RC1;

avoid duplicate build of jars_fresh;

obsolete, cf. README_REPOSITORY;

accomodate scala-2.10.0-RC1;

tailored enum specification towards simple instantiation; tuned enum instantiations

refined internal structure of Enum.thy

moved quite generic material from theory Enum to more appropriate places

adding another test case for the set_comprehension_simproc to the theory in HOL/ex

improving tactic in setcomprehension_simproc

adjusting proofs

tuned tactic in set_comprehension_pointfree simproc to handle composition of negation and vimage

passing names and types of all bounds around in the simproc

locally inverting previously applied simplifications with ex_simps in set_comprehension_pointfree

more precise pixel_range: avoid popup when pointing into empty space after actual end-of-line;

merged

don't include Quotient_Option - workaround to a transfer bug

ignore old stuff and thus speed up the script greatly;

proper find -mtime (file data) instead of -ctime (meta data);

made SML/NJ happy;

clarified Future.map (again): finished value is mapped in-place, which saves task structures and changes error behaviour slightly (tolerance against canceled group of old value etc.);

back to polyml-5.4.1 (cf. b3110dec1a32) -- no cause of spurious interrupts;

merged

back to parallel HOL-BNF-Examples, which seems to have suffered from Future.map on canceled persistent futures;

more basic Goal.reset_futures as snapshot of implicit state; more robust Session.finish_futures: do not cancel here, to allow later Future.map of persistent futures (notably proof terms);

tuned proof;

update RBT_Mapping, AList_Mapping and Mapping to use lifting/transfer

tuned proofs

new theorem

merged

merged

merged

merged

fixed proof (cf. a81f95693c68);

tuned Isar output

tuned

updated docs

renamed Isar-proof related options + changed semantics of Isar shrinking

tuning

fixed theorem lookup code in Isar proof reconstruction

tuning

refactor code

tuning

more robust future_proof result with specific error message (e.g. relevant for incomplete proof of non-registered theorem);

collective errors from use_thys and Session.finish/Goal.finish_futures -- avoid uninformative interrupts stemming from failure of goal forks that are not registered in the theory (e.g. unnamed theorems);

more uniform group for map_future, which is relevant for cancel in worker_task vs. future_job -- prefer peer group despite 81d03a29980c;

tuned message;

tuned comment;

avoid spurious "bad" markup for show/test_proof; tuned;

more official Future.terminate; tuned signature;

simp results for simplification results of Inf/Sup expressions on bool; tuned proofs

no sort constraints on datatype constructors in internal bookkeeping

HOL-BNF-Examples is sequential for now, due to spurious interrupts (again);

merged

comprehension conversion reuses suggested names for bound variables instead of invented fresh ones; tuned tactic

checking for bound variables in the set expression; handling negation more generally

set_comprehension_pointfree simproc now handles the complicated test case; tuned

refined conversion to only react on proper set comprehensions; tuned

moving Pair_inject from legacy and duplicate section to general section, as Pair_inject was considered a duplicate in e8400e31528a by mistake (cf. communication on dev mailing list)

employing a preprocessing conversion that rewrites {(x1, ..., xn). P x1 ... xn} to {(x1, ..., xn) | x1 ... xn. P x1 ... xn} in set_comprehension_pointfree simproc

another Future.shutdown after Future.cancel_groups (cf. 0d4106850eb2);

more robust cancel_now: avoid shooting yourself in the foot;

more robust Session.finish (batch mode): use Goal.finish_futures to exhibit remaining failures of disconnected goal forks (e.g. from unnamed theorems) and Goal.cancel_futures the purge the persistent state;

proper 'oops' to force sequential checking here, and avoid spurious *** Interrupt stemming from crash of forked outer syntax element;

added Output "Detach" button;

skipped proofs appear as "bad" without counting as error;

more method position information, notably finished_pos after end of previous text;

more formal markup;