tuned

split timeout among ATPs in and add Metis to the mix as backup

more portable cp options, e.g. for non-GNU version on Mac OS X Leopard;

tuned positions of ambiguity messages -- less intrusive in IDE view;

tuned

merged

avoid loading List_Cset and Dlist_Cet at the same time

corrected slip

Cset, Dlist_Cset, List_Cset: restructured

Cset, Dlist_Cset, List_Cset: restructured

adapted to changes in Cset.thy

adapted to changes in Cset.thy

separating predicates and sets syntactically

merged

avoid intermixing set and predicates; dropped lemmas mem_rsp and mem_prs (now in Quotient_Set.thy)

updated generated files

merged

updated generated files

explicit markup for legacy warnings;

updated generated files;

less aggressive warning icon;

tuned colors;

transparent foreground color for quoted entities;

more precise treatment of nodes that are fully required for partially visible ones;

de-assigned commands also count as changed;

beef up sledgehammer_tac in Mirabelle some more

merged

change default for generation of tag idempotence and tag argument equations

merged

NEWS entry for setsum_norm ~> norm_setsum

make HOL-Probability respect set/pred distinction

merged

use rename_tac to make proof script more robust (with separate set type, 'clarify' yields different variable names)

merged

generalize and simplify proof of continuous_within_sequentially

add lemma sequentially_imp_eventually_within; rename LIMSEQ_SEQ_conv2_lemma to sequentially_imp_eventually_at;

replace some continuous_on lemmas with more general versions

remove legacy theorem Lim_inner

arrange everything related to ordered_euclidean_space class together

generalize and shorten proof of basis_orthogonal

remove dot_lsum and dot_rsum in favor of inner_setsum_{left,right}

merged

generalize lemma finite_imp_compact_convex_hull and related lemmas

generalize some lemmas

generalize lemma convex_cone_hull

rename subset_{interior,closure} to {interior,closure}_mono; remove some legacy theorem names;

simplify many proofs about subspace and span; reorder premises in rule span_induct;

remove duplicate simp declaration

simplify definition of 'interior'; add lemmas interiorI and interiorE; change lemmas interior_unique and closure_unique to rule_format; tidy some proofs;

add lemma closure_union; simplify some proofs;

minimize imports

move everything related to 'norm' method into new theory file Norm_Arith.thy

remove unused lemmas dimensionI, dimension_eq

move geometric progression lemmas from Linear_Algebra.thy to Integration.thy where they are used

merge

FSet: Explicit proof without mem_def

merged

added lemma

added a component in generated file names reflecting whether the minimizer is used -- needed for evaluation to keep these files separated from the main problem files

comment

disable TFF for Vampire 1.8 until they've fixed the soundness issues and it's back on SystemOnTPTP

improve completeness of polymorphic encodings

mangle tag bound declarations properly

fixed inverted logic and improve precision when handling monotonic types in polymorphic encodings

make sure that if slicing is disabled, a non-SOS slice is chosen

honor TFF Implicit

make polymorphic encodings more complete

make default unsound mode less unsound

make TFF output less explicit where possible

use more appropriate encoding for Z3 TPTP, as confirmed by evaluation

added one more known Z3 failure

added config options to control two aspects of the translation, for evaluation purposes

added choice operator output for Satallax

rationalized option names -- mono becomes raw_mono and mangled becomes mono

handle nonmangled monomorphich the same way as mangled monomorphic when it comes to helper -- otherwise we can end up generating too tight type guards

avoid using ":" for anything but systematic type tag annotations, because Hurd's Metis gives it that special semantics

fixed bang encoding detection of which types to encode

lemma Compl_insert: "- insert x A = (-A) - {x}"

avoid variable clashes by properly incrementing indices

improved completeness and efficiency of Z3 proof reconstruction, especially by an improved handling of Skolemization

include chained facts for minimizer, otherwise it won't work

remove Vampire imconplete proof detection -- the bug it was trying to work around has been fixed in version 1.8, and the check is too sensitive anyway

back to tradition Scratch.thy default -- execution wrt. perspective overcomes the main problems of 226563829580;

tuned Session.edit_node: update_perspective based on last_exec_offset;

tuned signature -- iterate subsumes both fold and get_first;

further clarification of Document.updated, based on last_common and after_entry; tuned;

tuned signature;

improved Document.edit: more accurate update_start and no_execs; tuned;

refined document state assignment: observe perspective, more explicit assignment message; misc tuning and clarification;

tuned signature -- emphasize traditional read/eval/print terminology, which is still relevant here;

maintain last_execs assignment on Scala side; prefer tables over IDs instead of objects;

propagate information about last command with exec state assignment through document model;

tuned signature;

slightly more abstract Command.Perspective;

slightly more abstract Text.Perspective;

tuned proofs;

tuned syntax -- avoid ambiguities;

more accurate treatment of index syntax constants, for proper entity references in concrete notation (e.g. infix "\<oplus>\<index>");

delete commented-out dead code

merged

change some subsection headings to subsubsection

remove unnecessary lemma card_ge1

move connected_real_lemma to the one place it is used

merged

make sure that all facts are passed to ATP from minimizer

more reliable "sledgehammer\_tac" reconstruction, by avoiding "insert_tac"

specify timeout for "sledgehammer_tac"

tuning

Quotient Package: add mem_rsp, mem_prs, tune proofs.

merged

declare euclidean_simps [simp] at the point they are proved; avoid duplicate rule warnings;

merged

merged

avoid warnings

comment out dead code to avoid compiler warnings

legacy theorem names

remove duplicate lemma

fixed "hBOOL" of existential variables, and generate more helpers

don't select facts when using sledgehammer_tac for reconstruction

don't perform a triviality check if the goal is skipped anyway

optional reconstructor

misc tuning and simplification;

tuned pri: prefer purging of canceled execution;

tuned Document.node: maintain "touched" flag to indicate changes in entries etc.;

clarified Document.Node.clear -- retain header (cf. ML version);

clarified norm_header/header_edit -- disallow update of loaded theories;

misc tuning and simplification;

ignore irrelevant timings;

print state only for visible command, to avoid wasting resources for the larger part of the text;

early filtering of unchanged perspective;

more reliable update_perspective handler based on actual text visibility (e.g. on startup or when resizing without scrolling);

update_perspective without actual edits, bypassing the full state assignment protocol; edit_nodes/Perspective: do not touch_descendants here; propagate editor scroll events via update_perspective; misc tuning;

tuned;

handle potentially more approriate BufferUpdate.LOADED event;

special treatment of structure index 1 in Pure, including legacy warning;

compile

added "max_calls" option to get a fixed number of Sledgehammer calls per theory

beef up "sledgehammer_tac" reconstructor

clean up Sledgehammer tactic

fixed document;

tuned signature;

merged

always use TFF if possible

clearer separator in generated file names

exploit TFF format in Z3 used as ATP, and renamed it "z3_tptp"

updated known failures for Z3 3.0 TPTP

updated Z3 docs

avoid TFF format with older Vampire versions

update the Vampire related parts of the documentation

fixed TFF slicing

kindly ask Vampire to output axiom names

added formats to the slice and use TFF for remote Vampire

tuned specifications, syntax and proofs

tuned specifications and syntax

Quotient Package: some infrastructure for lifting inside sets

include all encodings in tests, now that the incompleteness of some encodings has been addressed

change Metis's default settings if type information axioms are generated

we must tag any type whose ground types intersect a nonmonotonic type

prefer the lighter, slightly unsound monotonicity-based encodings for Metis

made reconstruction of type tag equalities "\?x = \?x" reliable

tuning ATP problem output

revert guard logic -- make sure that typing information is generated for existentials

generate tag equations for existential variables

tuning, plus started implementing tag equation generation for existential variables

precisely distinguish between universal and existential quantifiers, instead of assuming the worst (universal), for monotonicity analysis

clearer terminology

renamed "heavy" to "uniform", based on discussion with Nick Smallbone

removed unused configuration option

added caching for (in)finiteness checks

remove needless typing information

cleaner handling of polymorphic monotonicity inference

started cleaning up polymorphic monotonicity-based encodings, based on discussions with Nick Smallbone

more precise warning

added option to control soundness of encodings more precisely, for evaluation purposes

make sound mode more sound (and clean up code)

reintroduced slightly unsound optimization taken out in 717880e98e6b, but only if "sound" is false

gracefully handle empty SPASS problems

pass sound option to Sledgehammer tactic

tuned signature -- contrast physical output primitives versus Output.raw_message;

tuned signature;

discontinued slightly odd Future/Lazy.get_finished, which do not really fit into the execution model of Future.cancel/join_tasks (canceled tasks need to be dequeued and terminated explicitly);

some support for toplevel printing wrt. editor perspective (still inactive);

propagate editor perspective through document model;

some support for editor perspective;

discontinued redundant Edit_Command_ID;

reduced warnings;

tuned message;

old-style numbered structure index is legacy feature (hardly ever used now);

added official Text.Range.Ordering; some support for text perspective;

tuned signature;

reverted some odd changes to HOL/Import (cf. 74c08021ab2e);

merged

modernized specifications

removed session HOL/Subst -- now subsumed my more modern HOL/ex/Unification.thy

removed technical or trivial unused facts

more precise authors and comments; tuned order and headers

added proof of idempotence, roughly after HOL/Subst/Unify.thy

tuned proofs, sledgehammering overly verbose parts

tuned notation

ported some lemmas from HOL/Subst/*; tuned order

changed constant names and notation to match HOL/Subst/*.thy, from which this theory is a clone.

merged

add lemmas interior_Times and closure_Times

avoid pred/set mixture

avoid pred/set mixture

merged

remove unnecessary euclidean_space class constraints

section -> subsection

scale dependency graph to fit on page

more robust initialization of token marker and line context wrt. session startup;

tuned Parse.group: delayed failure message;

avoid actual Color.white, which would be turned into Color.black by org.gjt.sp.jedit.print.BufferPrintable;

default style for user fonts -- to prevent org.gjt.sp.jedit.print.BufferPrintable from choking on null;

tuned;

discontinued somewhat pointless Par_List.map_name -- most of the time is spent in tokenization;

discontinued obsolete Thy_Syntax.report_span -- information can be reproduced in Isabelle/Scala;

merged

replace lemma realpow_two_diff with new lemma square_diff_square_factored

remove redundant lemma real_0_le_divide_iff in favor or zero_le_divide_iff

move lemma add_eq_0_iff to Groups.thy

remove redundant lemma realpow_two_disj, use square_eq_iff or power2_eq_iff instead

rename real_squared_diff_one_factored to square_diff_one_factored and move to Rings.thy

add lemma power2_eq_iff

remove some over-specific rules from the simpset

merged

redefine constant 'trivial_limit' as an abbreviation

purely functional task_queue.ML -- moved actual interrupt_unsynchronized to future.ML;

refined Task_Queue.cancel: passive tasks are considered running due to pending abort operation;

odd workaround for odd problem of load order in HOL/ex/ROOT.ML (!??);

refined Graph implementation: more abstract/scalable Graph.Keys instead of plain lists -- order of adjacency is now standardized wrt. Key.ord;

clarified get_imports -- should not rely on accidental order within graph;

tuned Table.delete_safe: avoid potentially expensive attempt of delete;

discontinued "Interrupt", which could disturb administrative tasks of the document model;

more direct balanced version Ord_List.unions;

reverted to join_bodies/join_proofs based on fold_body_thms to regain performance (escpecially of HOL-Proofs) -- see also aa9c1e9ef2ce and 4e2abb045eac;

tuned future priorities (again);

clarified fulfill_norm_proof: no join_thms yet; clarified priority of fulfill_proof_future, which is followed by explicit join_thms; explicit Thm.future_body_of without join yet; tuned Thm.future_result: join_promises without fulfill_norm_proof;

added Future.joins convenience; clarified Future.map: based on Future.cond_forks;

merged

deactivated »unknown« nitpick example