NEWS
author berghofe
Wed Apr 23 18:09:48 2003 +0200 (2003-04-23)
changeset 13921 69c627b6b28d
parent 13899 12c7029d278b
child 13938 b033b53d0c1e
permissions -rw-r--r--
Fixed problem in add_elim_realizer which caused bound variables to
get mixed up.
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle release
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Provers/simplifier:
    10 
    11   - Completely reimplemented method simp (ML: Asm_full_simp_tac):
    12     Assumptions are now subject to complete mutual simplification,
    13     not just from left to right. The simplifier now preserves
    14     the order of assumptions.
    15 
    16     Potential INCOMPATIBILITY:
    17 
    18     -- simp sometimes diverges where the old version did
    19        not, e.g. invoking simp on the goal
    20 
    21         [| P (f x); y = x; f x = f y |] ==> Q
    22 
    23        now gives rise to the infinite reduction sequence
    24 
    25         P(f x) --(f x = f y)--> P(f y) --(y = x)--> P(f x) --(f x = f y)--> ...
    26 
    27        Using "simp (asm_lr)" (ML: Asm_lr_simp_tac) instead often solves this
    28        kind of problem.
    29 
    30     -- Tactics combining classical reasoner and simplification (such as auto)
    31        are also affected by this change, because many of them rely on
    32        simp. They may sometimes diverge as well or yield a different numbers
    33        of subgoals. Try to use e.g. force, fastsimp, or safe instead of auto
    34        in case of problems. Sometimes subsequent calls to the classical
    35        reasoner will fail because a preceeding call to the simplifier too
    36        eagerly simplified the goal, e.g. deleted redundant premises.
    37 
    38   - The simplifier trace now shows the names of the applied rewrite rules
    39 
    40   - You can limit the number of recursive invocations of the simplifier
    41     during conditional rewriting (where the simplifie tries to solve the
    42     conditions before applying the rewrite rule):
    43     ML "simp_depth_limit := n"
    44     where n is an integer. Thus you can force termination where previously
    45     the simplifier would diverge.
    46 
    47   - Accepts free variables as head terms in congruence rules.  Useful in Isar.
    48 
    49 * Pure: The main goal of the proof state is no longer shown by default, only
    50 the subgoals. This behaviour is controlled by a new flag.
    51    PG menu: Isabelle/Isar -> Settings -> Show Main Goal
    52 (ML: Proof.show_main_goal).
    53 
    54 * Pure: You can find all matching introduction rules for subgoal 1, i.e. all
    55 rules whose conclusion matches subgoal 1:
    56       PG menu: Isabelle/Isar -> Show me -> matching rules
    57 The rules are ordered by how closely they match the subgoal.
    58 In particular, rules that solve a subgoal outright are displayed first
    59 (or rather last, the way they are printed).
    60 (ML: ProofGeneral.print_intros())
    61 
    62 * Pure: New flag trace_unify_fail causes unification to print
    63 diagnostic information (PG: in trace buffer) when it fails. This is
    64 useful for figuring out why single step proofs like rule, erule or
    65 assumption failed.
    66 
    67 * Pure: Locale specifications now produce predicate definitions
    68 according to the body of text (covering assumptions modulo local
    69 definitions); predicate "loc_axioms" covers newly introduced text,
    70 while "loc" is cumulative wrt. all included locale expressions; the
    71 latter view is presented only on export into the global theory
    72 context; potential INCOMPATIBILITY, use "(open)" option to fall back
    73 on the old view without predicates;
    74 
    75 * Pure: predefined locales "var" and "struct" are useful for sharing
    76 parameters (as in CASL, for example); just specify something like
    77 ``var x + var y + struct M'' as import;
    78 
    79 * Pure: improved thms_containing: proper indexing of facts instead of
    80 raw theorems; check validity of results wrt. current name space;
    81 include local facts of proof configuration (also covers active
    82 locales), cover fixed variables in index; may use "_" in term
    83 specification; an optional limit for the number of printed facts may
    84 be given (the default is 40);
    85 
    86 * Pure: disallow duplicate fact bindings within new-style theory files
    87 (batch-mode only);
    88 
    89 * Provers: improved induct method: assumptions introduced by case
    90 "foo" are split into "foo.hyps" (from the rule) and "foo.prems" (from
    91 the goal statement); "foo" still refers to all facts collectively;
    92 
    93 * Provers: the function blast.overloaded has been removed: all constants
    94 are regarded as potentially overloaded, which improves robustness in exchange
    95 for slight decrease in efficiency;
    96 
    97 * Provers/linorder: New generic prover for transitivity reasoning over
    98 linear orders.  Note: this prover is not efficient!
    99 
   100 * Isar: preview of problems to finish 'show' now produce an error
   101 rather than just a warning (in interactive mode);
   102 
   103 
   104 *** HOL ***
   105 
   106 * arith(_tac)
   107 
   108  - Produces a counter example if it cannot prove a goal.
   109    Note that the counter example may be spurious if the goal is not a formula
   110    of quantifier-free linear arithmetic.
   111    In ProofGeneral the counter example appears in the trace buffer.
   112 
   113  - Knows about div k and mod k where k is a numeral of type nat or int.
   114 
   115  - Calls full Presburger arithmetic (by Amine Chaieb) if quantifier-free
   116    linear arithmetic fails. This takes account of quantifiers and divisibility.
   117    Presburger arithmetic can also be called explicitly via presburger(_tac). 
   118 
   119 * simp's arithmetic capabilities have been enhanced a bit: it now
   120 takes ~= in premises into account (by performing a case split);
   121 
   122 * simp reduces "m*(n div m) + n mod m" to n, even if the two summands
   123 are distributed over a sum of terms;
   124 
   125 * New tactic "trans_tac" and method "trans" instantiate
   126 Provers/linorder.ML for axclasses "order" and "linorder" (predicates
   127 "<=", "<" and "="). 
   128 
   129 * function INCOMPATIBILITIES: Pi-sets have been redefined and moved from main 
   130 HOL to Library/FuncSet; constant "Fun.op o" is now called "Fun.comp";
   131 
   132 * 'typedef' command has new option "open" to suppress the set
   133 definition;
   134 
   135 * functions Min and Max on finite sets have been introduced (theory
   136 Finite_Set);
   137 
   138 * attribute [symmetric] now works for relations as well; it turns
   139 (x,y) : R^-1 into (y,x) : R, and vice versa;
   140 
   141 * induct over a !!-quantified statement (say !!x1..xn):
   142   each "case" automatically performs "fix x1 .. xn" with exactly those names.
   143 
   144 * Map: `empty' is no longer a constant but a syntactic abbreviation for
   145 %x. None. Warning: empty_def now refers to the previously hidden definition
   146 of the empty set.
   147 
   148 * Real/HahnBanach: updated and adapted to locales;
   149 
   150 * GroupTheory: converted to Isar theories, using locales with implicit
   151 structures.  Also a new, experimental summation operator for abelian groups;
   152 
   153 * NumberTheory: added Gauss's law of quadratic reciprocity (by Avigad, Gray and
   154 Kramer);
   155 
   156 * UNITY: added the Meier-Sanders theory of progress sets;
   157 
   158 
   159 *** ZF ***
   160 
   161 * ZF/Constructible: consistency proof for AC (Gödel's constructible
   162 universe, etc.);
   163 
   164 * Main ZF: virtually all theories converted to new-style format;
   165 
   166 
   167 *** ML ***
   168 
   169 * Pure: Tactic.prove provides sane interface for internal proofs;
   170 omits the infamous "standard" operation, so this is more appropriate
   171 than prove_goalw_cterm in many situations (e.g. in simprocs);
   172 
   173 * Pure: improved error reporting of simprocs;
   174 
   175 * Provers: Simplifier.simproc(_i) provides sane interface for setting
   176 up simprocs;
   177 
   178 
   179 
   180 New in Isabelle2002 (March 2002)
   181 --------------------------------
   182 
   183 *** Document preparation ***
   184 
   185 * greatly simplified document preparation setup, including more
   186 graceful interpretation of isatool usedir -i/-d/-D options, and more
   187 instructive isatool mkdir; users should basically be able to get
   188 started with "isatool mkdir HOL Test && isatool make"; alternatively,
   189 users may run a separate document processing stage manually like this:
   190 "isatool usedir -D output HOL Test && isatool document Test/output";
   191 
   192 * theory dependency graph may now be incorporated into documents;
   193 isatool usedir -g true will produce session_graph.eps/.pdf for use
   194 with \includegraphics of LaTeX;
   195 
   196 * proper spacing of consecutive markup elements, especially text
   197 blocks after section headings;
   198 
   199 * support bold style (for single symbols only), input syntax is like
   200 this: "\<^bold>\<alpha>" or "\<^bold>A";
   201 
   202 * \<bullet> is now output as bold \cdot by default, which looks much
   203 better in printed text;
   204 
   205 * added default LaTeX bindings for \<tturnstile> and \<TTurnstile>;
   206 note that these symbols are currently unavailable in Proof General /
   207 X-Symbol; new symbols \<zero>, \<one>, ..., \<nine>, and \<euro>;
   208 
   209 * isatool latex no longer depends on changed TEXINPUTS, instead
   210 isatool document copies the Isabelle style files to the target
   211 location;
   212 
   213 
   214 *** Isar ***
   215 
   216 * Pure/Provers: improved proof by cases and induction;
   217   - 'case' command admits impromptu naming of parameters (such as
   218     "case (Suc n)");
   219   - 'induct' method divinates rule instantiation from the inductive
   220     claim; no longer requires excessive ?P bindings for proper
   221     instantiation of cases;
   222   - 'induct' method properly enumerates all possibilities of set/type
   223     rules; as a consequence facts may be also passed through *type*
   224     rules without further ado;
   225   - 'induct' method now derives symbolic cases from the *rulified*
   226     rule (before it used to rulify cases stemming from the internal
   227     atomized version); this means that the context of a non-atomic
   228     statement becomes is included in the hypothesis, avoiding the
   229     slightly cumbersome show "PROP ?case" form;
   230   - 'induct' may now use elim-style induction rules without chaining
   231     facts, using ``missing'' premises from the goal state; this allows
   232     rules stemming from inductive sets to be applied in unstructured
   233     scripts, while still benefitting from proper handling of non-atomic
   234     statements; NB: major inductive premises need to be put first, all
   235     the rest of the goal is passed through the induction;
   236   - 'induct' proper support for mutual induction involving non-atomic
   237     rule statements (uses the new concept of simultaneous goals, see
   238     below);
   239   - append all possible rule selections, but only use the first
   240     success (no backtracking);
   241   - removed obsolete "(simplified)" and "(stripped)" options of methods;
   242   - undeclared rule case names default to numbers 1, 2, 3, ...;
   243   - added 'print_induct_rules' (covered by help item in recent Proof
   244     General versions);
   245   - moved induct/cases attributes to Pure, methods to Provers;
   246   - generic method setup instantiated for FOL and HOL;
   247 
   248 * Pure: support multiple simultaneous goal statements, for example
   249 "have a: A and b: B" (same for 'theorem' etc.); being a pure
   250 meta-level mechanism, this acts as if several individual goals had
   251 been stated separately; in particular common proof methods need to be
   252 repeated in order to cover all claims; note that a single elimination
   253 step is *not* sufficient to establish the two conjunctions, so this
   254 fails:
   255 
   256   assume "A & B" then have A and B ..   (*".." fails*)
   257 
   258 better use "obtain" in situations as above; alternative refer to
   259 multi-step methods like 'auto', 'simp_all', 'blast+' etc.;
   260 
   261 * Pure: proper integration with ``locales''; unlike the original
   262 version by Florian Kammüller, Isar locales package high-level proof
   263 contexts rather than raw logical ones (e.g. we admit to include
   264 attributes everywhere); operations on locales include merge and
   265 rename; support for implicit arguments (``structures''); simultaneous
   266 type-inference over imports and text; see also HOL/ex/Locales.thy for
   267 some examples;
   268 
   269 * Pure: the following commands have been ``localized'', supporting a
   270 target locale specification "(in name)": 'lemma', 'theorem',
   271 'corollary', 'lemmas', 'theorems', 'declare'; the results will be
   272 stored both within the locale and at the theory level (exported and
   273 qualified by the locale name);
   274 
   275 * Pure: theory goals may now be specified in ``long'' form, with
   276 ad-hoc contexts consisting of arbitrary locale elements. for example
   277 ``lemma foo: fixes x assumes "A x" shows "B x"'' (local syntax and
   278 definitions may be given, too); the result is a meta-level rule with
   279 the context elements being discharged in the obvious way;
   280 
   281 * Pure: new proof command 'using' allows to augment currently used
   282 facts after a goal statement ('using' is syntactically analogous to
   283 'apply', but acts on the goal's facts only); this allows chained facts
   284 to be separated into parts given before and after a claim, as in
   285 ``from a and b have C using d and e <proof>'';
   286 
   287 * Pure: renamed "antecedent" case to "rule_context";
   288 
   289 * Pure: new 'judgment' command records explicit information about the
   290 object-logic embedding (used by several tools internally); no longer
   291 use hard-wired "Trueprop";
   292 
   293 * Pure: added 'corollary' command;
   294 
   295 * Pure: fixed 'token_translation' command;
   296 
   297 * Pure: removed obsolete 'exported' attribute;
   298 
   299 * Pure: dummy pattern "_" in is/let is now automatically lifted over
   300 bound variables: "ALL x. P x --> Q x" (is "ALL x. _ --> ?C x")
   301 supersedes more cumbersome ... (is "ALL x. _ x --> ?C x");
   302 
   303 * Pure: method 'atomize' presents local goal premises as object-level
   304 statements (atomic meta-level propositions); setup controlled via
   305 rewrite rules declarations of 'atomize' attribute; example
   306 application: 'induct' method with proper rule statements in improper
   307 proof *scripts*;
   308 
   309 * Pure: emulation of instantiation tactics (rule_tac, cut_tac, etc.)
   310 now consider the syntactic context of assumptions, giving a better
   311 chance to get type-inference of the arguments right (this is
   312 especially important for locales);
   313 
   314 * Pure: "sorry" no longer requires quick_and_dirty in interactive
   315 mode;
   316 
   317 * Pure/obtain: the formal conclusion "thesis", being marked as
   318 ``internal'', may no longer be reference directly in the text;
   319 potential INCOMPATIBILITY, may need to use "?thesis" in rare
   320 situations;
   321 
   322 * Pure: generic 'sym' attribute which declares a rule both as pure
   323 'elim?' and for the 'symmetric' operation;
   324 
   325 * Pure: marginal comments ``--'' may now occur just anywhere in the
   326 text; the fixed correlation with particular command syntax has been
   327 discontinued;
   328 
   329 * Pure: new method 'rules' is particularly well-suited for proof
   330 search in intuitionistic logic; a bit slower than 'blast' or 'fast',
   331 but often produces more compact proof terms with less detours;
   332 
   333 * Pure/Provers/classical: simplified integration with pure rule
   334 attributes and methods; the classical "intro?/elim?/dest?"
   335 declarations coincide with the pure ones; the "rule" method no longer
   336 includes classically swapped intros; "intro" and "elim" methods no
   337 longer pick rules from the context; also got rid of ML declarations
   338 AddXIs/AddXEs/AddXDs; all of this has some potential for
   339 INCOMPATIBILITY;
   340 
   341 * Provers/classical: attribute 'swapped' produces classical inversions
   342 of introduction rules;
   343 
   344 * Provers/simplifier: 'simplified' attribute may refer to explicit
   345 rules instead of full simplifier context; 'iff' attribute handles
   346 conditional rules;
   347 
   348 * HOL: 'typedef' now allows alternative names for Rep/Abs morphisms;
   349 
   350 * HOL: 'recdef' now fails on unfinished automated proofs, use
   351 "(permissive)" option to recover old behavior;
   352 
   353 * HOL: 'inductive' no longer features separate (collective) attributes
   354 for 'intros' (was found too confusing);
   355 
   356 * HOL: properly declared induction rules less_induct and
   357 wf_induct_rule;
   358 
   359 
   360 *** HOL ***
   361 
   362 * HOL: moved over to sane numeral syntax; the new policy is as
   363 follows:
   364 
   365   - 0 and 1 are polymorphic constants, which are defined on any
   366   numeric type (nat, int, real etc.);
   367 
   368   - 2, 3, 4, ... and -1, -2, -3, ... are polymorphic numerals, based
   369   binary representation internally;
   370 
   371   - type nat has special constructor Suc, and generally prefers Suc 0
   372   over 1::nat and Suc (Suc 0) over 2::nat;
   373 
   374 This change may cause significant problems of INCOMPATIBILITY; here
   375 are some hints on converting existing sources:
   376 
   377   - due to the new "num" token, "-0" and "-1" etc. are now atomic
   378   entities, so expressions involving "-" (unary or binary minus) need
   379   to be spaced properly;
   380 
   381   - existing occurrences of "1" may need to be constraint "1::nat" or
   382   even replaced by Suc 0; similar for old "2";
   383 
   384   - replace "#nnn" by "nnn", and "#-nnn" by "-nnn";
   385 
   386   - remove all special provisions on numerals in proofs;
   387 
   388 * HOL: simp rules nat_number expand numerals on nat to Suc/0
   389 representation (depends on bin_arith_simps in the default context);
   390 
   391 * HOL: symbolic syntax for x^2 (numeral 2);
   392 
   393 * HOL: the class of all HOL types is now called "type" rather than
   394 "term"; INCOMPATIBILITY, need to adapt references to this type class
   395 in axclass/classes, instance/arities, and (usually rare) occurrences
   396 in typings (of consts etc.); internally the class is called
   397 "HOL.type", ML programs should refer to HOLogic.typeS;
   398 
   399 * HOL/record package improvements:
   400   - new derived operations "fields" to build a partial record section,
   401     "extend" to promote a fixed record to a record scheme, and
   402     "truncate" for the reverse; cf. theorems "xxx.defs", which are *not*
   403     declared as simp by default;
   404   - shared operations ("more", "fields", etc.) now need to be always
   405     qualified) --- potential INCOMPATIBILITY;
   406   - removed "make_scheme" operations (use "make" with "extend") --
   407     INCOMPATIBILITY;
   408   - removed "more" class (simply use "term") -- INCOMPATIBILITY;
   409   - provides cases/induct rules for use with corresponding Isar
   410     methods (for concrete records, record schemes, concrete more
   411     parts, and schematic more parts -- in that order);
   412   - internal definitions directly based on a light-weight abstract
   413     theory of product types over typedef rather than datatype;
   414 
   415 * HOL: generic code generator for generating executable ML code from
   416 specifications; specific support for HOL constructs such as inductive
   417 datatypes and sets, as well as recursive functions; can be invoked
   418 via 'generate_code' theory section;
   419 
   420 * HOL: canonical cases/induct rules for n-tuples (n = 3..7);
   421 
   422 * HOL: consolidated and renamed several theories.  In particular:
   423 	Ord.thy has been absorbed into HOL.thy
   424 	String.thy has been absorbed into List.thy
   425  
   426 * HOL: concrete setsum syntax "\<Sum>i:A. b" == "setsum (%i. b) A"
   427 (beware of argument permutation!);
   428 
   429 * HOL: linorder_less_split superseded by linorder_cases;
   430 
   431 * HOL/List: "nodups" renamed to "distinct";
   432 
   433 * HOL: added "The" definite description operator; move Hilbert's "Eps"
   434 to peripheral theory "Hilbert_Choice"; some INCOMPATIBILITIES:
   435   - Ex_def has changed, now need to use some_eq_ex
   436 
   437 * HOL: made split_all_tac safe; EXISTING PROOFS MAY FAIL OR LOOP, so
   438 in this (rare) case use:
   439 
   440   delSWrapper "split_all_tac"
   441   addSbefore ("unsafe_split_all_tac", unsafe_split_all_tac)
   442 
   443 * HOL: added safe wrapper "split_conv_tac" to claset; EXISTING PROOFS
   444 MAY FAIL;
   445 
   446 * HOL: introduced f^n = f o ... o f; warning: due to the limits of
   447 Isabelle's type classes, ^ on functions and relations has too general
   448 a domain, namely ('a * 'b) set and 'a => 'b; this means that it may be
   449 necessary to attach explicit type constraints;
   450 
   451 * HOL/Relation: the prefix name of the infix "O" has been changed from
   452 "comp" to "rel_comp"; INCOMPATIBILITY: a few theorems have been
   453 renamed accordingly (eg "compI" -> "rel_compI").
   454 
   455 * HOL: syntax translations now work properly with numerals and records
   456 expressions;
   457 
   458 * HOL: bounded abstraction now uses syntax "%" / "\<lambda>" instead
   459 of "lam" -- INCOMPATIBILITY;
   460 
   461 * HOL: got rid of some global declarations (potential INCOMPATIBILITY
   462 for ML tools): const "()" renamed "Product_Type.Unity", type "unit"
   463 renamed "Product_Type.unit";
   464 
   465 * HOL: renamed rtrancl_into_rtrancl2 to converse_rtrancl_into_rtrancl
   466 
   467 * HOL: removed obsolete theorem "optionE" (use "option.exhaust", or
   468 the "cases" method);
   469 
   470 * HOL/GroupTheory: group theory examples including Sylow's theorem (by
   471 Florian Kammüller);
   472 
   473 * HOL/IMP: updated and converted to new-style theory format; several
   474 parts turned into readable document, with proper Isar proof texts and
   475 some explanations (by Gerwin Klein);
   476 
   477 * HOL-Real: added Complex_Numbers (by Gertrud Bauer);
   478 
   479 * HOL-Hyperreal is now a logic image;
   480 
   481 
   482 *** HOLCF ***
   483 
   484 * Isar: consts/constdefs supports mixfix syntax for continuous
   485 operations;
   486 
   487 * Isar: domain package adapted to new-style theory format, e.g. see
   488 HOLCF/ex/Dnat.thy;
   489 
   490 * theory Lift: proper use of rep_datatype lift instead of ML hacks --
   491 potential INCOMPATIBILITY; now use plain induct_tac instead of former
   492 lift.induct_tac, always use UU instead of Undef;
   493 
   494 * HOLCF/IMP: updated and converted to new-style theory;
   495 
   496 
   497 *** ZF ***
   498 
   499 * Isar: proper integration of logic-specific tools and packages,
   500 including theory commands '(co)inductive', '(co)datatype',
   501 'rep_datatype', 'inductive_cases', as well as methods 'ind_cases',
   502 'induct_tac', 'case_tac', and 'typecheck' (with attribute 'TC');
   503 
   504 * theory Main no longer includes AC; for the Axiom of Choice, base
   505 your theory on Main_ZFC;
   506 
   507 * the integer library now covers quotients and remainders, with many
   508 laws relating division to addition, multiplication, etc.;
   509 
   510 * ZF/UNITY: Chandy and Misra's UNITY is now available in ZF, giving a
   511 typeless version of the formalism;
   512 
   513 * ZF/AC, Coind, IMP, Resid: updated and converted to new-style theory
   514 format;
   515 
   516 * ZF/Induct: new directory for examples of inductive definitions,
   517 including theory Multiset for multiset orderings; converted to
   518 new-style theory format;
   519 
   520 * ZF: many new theorems about lists, ordinals, etc.;
   521 
   522 
   523 *** General ***
   524 
   525 * Pure/kernel: meta-level proof terms (by Stefan Berghofer); reference
   526 variable proof controls level of detail: 0 = no proofs (only oracle
   527 dependencies), 1 = lemma dependencies, 2 = compact proof terms; see
   528 also ref manual for further ML interfaces;
   529 
   530 * Pure/axclass: removed obsolete ML interface
   531 goal_subclass/goal_arity;
   532 
   533 * Pure/syntax: new token syntax "num" for plain numerals (without "#"
   534 of "xnum"); potential INCOMPATIBILITY, since -0, -1 etc. are now
   535 separate tokens, so expressions involving minus need to be spaced
   536 properly;
   537 
   538 * Pure/syntax: support non-oriented infixes, using keyword "infix"
   539 rather than "infixl" or "infixr";
   540 
   541 * Pure/syntax: concrete syntax for dummy type variables admits genuine
   542 sort constraint specifications in type inference; e.g. "x::_::foo"
   543 ensures that the type of "x" is of sort "foo" (but not necessarily a
   544 type variable);
   545 
   546 * Pure/syntax: print modes "type_brackets" and "no_type_brackets"
   547 control output of nested => (types); the default behavior is
   548 "type_brackets";
   549 
   550 * Pure/syntax: builtin parse translation for "_constify" turns valued
   551 tokens into AST constants;
   552 
   553 * Pure/syntax: prefer later declarations of translations and print
   554 translation functions; potential INCOMPATIBILITY: need to reverse
   555 multiple declarations for same syntax element constant;
   556 
   557 * Pure/show_hyps reset by default (in accordance to existing Isar
   558 practice);
   559 
   560 * Provers/classical: renamed addaltern to addafter, addSaltern to
   561 addSafter;
   562 
   563 * Provers/clasimp: ``iff'' declarations now handle conditional rules
   564 as well;
   565 
   566 * system: tested support for MacOS X; should be able to get Isabelle +
   567 Proof General to work in a plain Terminal after installing Poly/ML
   568 (e.g. from the Isabelle distribution area) and GNU bash alone
   569 (e.g. from http://www.apple.com); full X11, XEmacs and X-Symbol
   570 support requires further installations, e.g. from
   571 http://fink.sourceforge.net/);
   572 
   573 * system: support Poly/ML 4.1.1 (able to manage larger heaps);
   574 
   575 * system: reduced base memory usage by Poly/ML (approx. 20 MB instead
   576 of 40 MB), cf. ML_OPTIONS;
   577 
   578 * system: Proof General keywords specification is now part of the
   579 Isabelle distribution (see etc/isar-keywords.el);
   580 
   581 * system: support for persistent Proof General sessions (refrain from
   582 outdating all loaded theories on startup); user may create writable
   583 logic images like this: ``isabelle -q HOL Test'';
   584 
   585 * system: smart selection of Isabelle process versus Isabelle
   586 interface, accommodates case-insensitive file systems (e.g. HFS+); may
   587 run both "isabelle" and "Isabelle" even if file names are badly
   588 damaged (executable inspects the case of the first letter of its own
   589 name); added separate "isabelle-process" and "isabelle-interface";
   590 
   591 * system: refrain from any attempt at filtering input streams; no
   592 longer support ``8bit'' encoding of old isabelle font, instead proper
   593 iso-latin characters may now be used; the related isatools
   594 "symbolinput" and "nonascii" have disappeared as well;
   595 
   596 * system: removed old "xterm" interface (the print modes "xterm" and
   597 "xterm_color" are still available for direct use in a suitable
   598 terminal);
   599 
   600 
   601 
   602 New in Isabelle99-2 (February 2001)
   603 -----------------------------------
   604 
   605 *** Overview of INCOMPATIBILITIES ***
   606 
   607 * HOL: please note that theories in the Library and elsewhere often use the
   608 new-style (Isar) format; to refer to their theorems in an ML script you must
   609 bind them to ML identifers by e.g.      val thm_name = thm "thm_name";
   610 
   611 * HOL: inductive package no longer splits induction rule aggressively,
   612 but only as far as specified by the introductions given; the old
   613 format may be recovered via ML function complete_split_rule or attribute
   614 'split_rule (complete)';
   615 
   616 * HOL: induct renamed to lfp_induct, lfp_Tarski to lfp_unfold,
   617 gfp_Tarski to gfp_unfold;
   618 
   619 * HOL: contrapos, contrapos2 renamed to contrapos_nn, contrapos_pp;
   620 
   621 * HOL: infix "dvd" now has priority 50 rather than 70 (because it is a
   622 relation); infix "^^" has been renamed "``"; infix "``" has been
   623 renamed "`"; "univalent" has been renamed "single_valued";
   624 
   625 * HOL/Real: "rinv" and "hrinv" replaced by overloaded "inverse"
   626 operation;
   627 
   628 * HOLCF: infix "`" has been renamed "$"; the symbol syntax is \<cdot>;
   629 
   630 * Isar: 'obtain' no longer declares "that" fact as simp/intro;
   631 
   632 * Isar/HOL: method 'induct' now handles non-atomic goals; as a
   633 consequence, it is no longer monotonic wrt. the local goal context
   634 (which is now passed through the inductive cases);
   635 
   636 * Document preparation: renamed standard symbols \<ll> to \<lless> and
   637 \<gg> to \<ggreater>;
   638 
   639 
   640 *** Document preparation ***
   641 
   642 * \isabellestyle{NAME} selects version of Isabelle output (currently
   643 available: are "it" for near math-mode best-style output, "sl" for
   644 slanted text style, and "tt" for plain type-writer; if no
   645 \isabellestyle command is given, output is according to slanted
   646 type-writer);
   647 
   648 * support sub/super scripts (for single symbols only), input syntax is
   649 like this: "A\<^sup>*" or "A\<^sup>\<star>";
   650 
   651 * some more standard symbols; see Appendix A of the system manual for
   652 the complete list of symbols defined in isabellesym.sty;
   653 
   654 * improved isabelle style files; more abstract symbol implementation
   655 (should now use \isamath{...} and \isatext{...} in custom symbol
   656 definitions);
   657 
   658 * antiquotation @{goals} and @{subgoals} for output of *dynamic* goals
   659 state; Note that presentation of goal states does not conform to
   660 actual human-readable proof documents.  Please do not include goal
   661 states into document output unless you really know what you are doing!
   662 
   663 * proper indentation of antiquoted output with proportional LaTeX
   664 fonts;
   665 
   666 * no_document ML operator temporarily disables LaTeX document
   667 generation;
   668 
   669 * isatool unsymbolize tunes sources for plain ASCII communication;
   670 
   671 
   672 *** Isar ***
   673 
   674 * Pure: Isar now suffers initial goal statements to contain unbound
   675 schematic variables (this does not conform to actual readable proof
   676 documents, due to unpredictable outcome and non-compositional proof
   677 checking); users who know what they are doing may use schematic goals
   678 for Prolog-style synthesis of proven results;
   679 
   680 * Pure: assumption method (an implicit finishing) now handles actual
   681 rules as well;
   682 
   683 * Pure: improved 'obtain' --- moved to Pure, insert "that" into
   684 initial goal, declare "that" only as Pure intro (only for single
   685 steps); the "that" rule assumption may now be involved in implicit
   686 finishing, thus ".." becomes a feasible for trivial obtains;
   687 
   688 * Pure: default proof step now includes 'intro_classes'; thus trivial
   689 instance proofs may be performed by "..";
   690 
   691 * Pure: ?thesis / ?this / "..." now work for pure meta-level
   692 statements as well;
   693 
   694 * Pure: more robust selection of calculational rules;
   695 
   696 * Pure: the builtin notion of 'finished' goal now includes the ==-refl
   697 rule (as well as the assumption rule);
   698 
   699 * Pure: 'thm_deps' command visualizes dependencies of theorems and
   700 lemmas, using the graph browser tool;
   701 
   702 * Pure: predict failure of "show" in interactive mode;
   703 
   704 * Pure: 'thms_containing' now takes actual terms as arguments;
   705 
   706 * HOL: improved method 'induct' --- now handles non-atomic goals
   707 (potential INCOMPATIBILITY); tuned error handling;
   708 
   709 * HOL: cases and induct rules now provide explicit hints about the
   710 number of facts to be consumed (0 for "type" and 1 for "set" rules);
   711 any remaining facts are inserted into the goal verbatim;
   712 
   713 * HOL: local contexts (aka cases) may now contain term bindings as
   714 well; the 'cases' and 'induct' methods new provide a ?case binding for
   715 the result to be shown in each case;
   716 
   717 * HOL: added 'recdef_tc' command;
   718 
   719 * isatool convert assists in eliminating legacy ML scripts;
   720 
   721 
   722 *** HOL ***
   723 
   724 * HOL/Library: a collection of generic theories to be used together
   725 with main HOL; the theory loader path already includes this directory
   726 by default; the following existing theories have been moved here:
   727 HOL/Induct/Multiset, HOL/Induct/Acc (as Accessible_Part), HOL/While
   728 (as While_Combinator), HOL/Lex/Prefix (as List_Prefix);
   729 
   730 * HOL/Unix: "Some aspects of Unix file-system security", a typical
   731 modelling and verification task performed in Isabelle/HOL +
   732 Isabelle/Isar + Isabelle document preparation (by Markus Wenzel).
   733 
   734 * HOL/Algebra: special summation operator SUM no longer exists, it has
   735 been replaced by setsum; infix 'assoc' now has priority 50 (like
   736 'dvd'); axiom 'one_not_zero' has been moved from axclass 'ring' to
   737 'domain', this makes the theory consistent with mathematical
   738 literature;
   739 
   740 * HOL basics: added overloaded operations "inverse" and "divide"
   741 (infix "/"), syntax for generic "abs" operation, generic summation
   742 operator \<Sum>;
   743 
   744 * HOL/typedef: simplified package, provide more useful rules (see also
   745 HOL/subset.thy);
   746 
   747 * HOL/datatype: induction rule for arbitrarily branching datatypes is
   748 now expressed as a proper nested rule (old-style tactic scripts may
   749 require atomize_strip_tac to cope with non-atomic premises);
   750 
   751 * HOL: renamed theory "Prod" to "Product_Type", renamed "split" rule
   752 to "split_conv" (old name still available for compatibility);
   753 
   754 * HOL: improved concrete syntax for strings (e.g. allows translation
   755 rules with string literals);
   756 
   757 * HOL-Real-Hyperreal: this extends HOL-Real with the hyperreals
   758  and Fleuriot's mechanization of analysis, including the transcendental
   759  functions for the reals;
   760 
   761 * HOL/Real, HOL/Hyperreal: improved arithmetic simplification;
   762 
   763 
   764 *** CTT ***
   765 
   766 * CTT: x-symbol support for Pi, Sigma, -->, : (membership); note that
   767 "lam" is displayed as TWO lambda-symbols
   768 
   769 * CTT: theory Main now available, containing everything (that is, Bool
   770 and Arith);
   771 
   772 
   773 *** General ***
   774 
   775 * Pure: the Simplifier has been implemented properly as a derived rule
   776 outside of the actual kernel (at last!); the overall performance
   777 penalty in practical applications is about 50%, while reliability of
   778 the Isabelle inference kernel has been greatly improved;
   779 
   780 * print modes "brackets" and "no_brackets" control output of nested =>
   781 (types) and ==> (props); the default behaviour is "brackets";
   782 
   783 * Provers: fast_tac (and friends) now handle actual object-logic rules
   784 as assumptions as well;
   785 
   786 * system: support Poly/ML 4.0;
   787 
   788 * system: isatool install handles KDE version 1 or 2;
   789 
   790 
   791 
   792 New in Isabelle99-1 (October 2000)
   793 ----------------------------------
   794 
   795 *** Overview of INCOMPATIBILITIES ***
   796 
   797 * HOL: simplification of natural numbers is much changed; to partly
   798 recover the old behaviour (e.g. to prevent n+n rewriting to #2*n)
   799 issue the following ML commands:
   800 
   801   Delsimprocs Nat_Numeral_Simprocs.cancel_numerals;
   802   Delsimprocs [Nat_Numeral_Simprocs.combine_numerals];
   803 
   804 * HOL: simplification no longer dives into case-expressions; this is
   805 controlled by "t.weak_case_cong" for each datatype t;
   806 
   807 * HOL: nat_less_induct renamed to less_induct;
   808 
   809 * HOL: systematic renaming of the SOME (Eps) rules, may use isatool
   810 fixsome to patch .thy and .ML sources automatically;
   811 
   812   select_equality  -> some_equality
   813   select_eq_Ex     -> some_eq_ex
   814   selectI2EX       -> someI2_ex
   815   selectI2         -> someI2
   816   selectI          -> someI
   817   select1_equality -> some1_equality
   818   Eps_sym_eq       -> some_sym_eq_trivial
   819   Eps_eq           -> some_eq_trivial
   820 
   821 * HOL: exhaust_tac on datatypes superceded by new generic case_tac;
   822 
   823 * HOL: removed obsolete theorem binding expand_if (refer to split_if
   824 instead);
   825 
   826 * HOL: the recursion equations generated by 'recdef' are now called
   827 f.simps instead of f.rules;
   828 
   829 * HOL: qed_spec_mp now also handles bounded ALL as well;
   830 
   831 * HOL: 0 is now overloaded, so the type constraint ":: nat" may
   832 sometimes be needed;
   833 
   834 * HOL: the constant for "f``x" is now "image" rather than "op ``";
   835 
   836 * HOL: the constant for "f-``x" is now "vimage" rather than "op -``";
   837 
   838 * HOL: the disjoint sum is now "<+>" instead of "Plus"; the cartesian
   839 product is now "<*>" instead of "Times"; the lexicographic product is
   840 now "<*lex*>" instead of "**";
   841 
   842 * HOL: theory Sexp is now in HOL/Induct examples (it used to be part
   843 of main HOL, but was unused); better use HOL's datatype package;
   844 
   845 * HOL: removed "symbols" syntax for constant "override" of theory Map;
   846 the old syntax may be recovered as follows:
   847 
   848   syntax (symbols)
   849     override  :: "('a ~=> 'b) => ('a ~=> 'b) => ('a ~=> 'b)"
   850       (infixl "\\<oplus>" 100)
   851 
   852 * HOL/Real: "rabs" replaced by overloaded "abs" function;
   853 
   854 * HOL/ML: even fewer consts are declared as global (see theories Ord,
   855 Lfp, Gfp, WF); this only affects ML packages that refer to const names
   856 internally;
   857 
   858 * HOL and ZF: syntax for quotienting wrt an equivalence relation
   859 changed from A/r to A//r;
   860 
   861 * ZF: new treatment of arithmetic (nat & int) may break some old
   862 proofs;
   863 
   864 * Isar: renamed some attributes (RS -> THEN, simplify -> simplified,
   865 rulify -> rule_format, elimify -> elim_format, ...);
   866 
   867 * Isar/Provers: intro/elim/dest attributes changed; renamed
   868 intro/intro!/intro!! flags to intro!/intro/intro? (in most cases, one
   869 should have to change intro!! to intro? only); replaced "delrule" by
   870 "rule del";
   871 
   872 * Isar/HOL: renamed "intrs" to "intros" in inductive definitions;
   873 
   874 * Provers: strengthened force_tac by using new first_best_tac;
   875 
   876 * LaTeX document preparation: several changes of isabelle.sty (see
   877 lib/texinputs);
   878 
   879 
   880 *** Document preparation ***
   881 
   882 * formal comments (text blocks etc.) in new-style theories may now
   883 contain antiquotations of thm/prop/term/typ/text to be presented
   884 according to latex print mode; concrete syntax is like this:
   885 @{term[show_types] "f(x) = a + x"};
   886 
   887 * isatool mkdir provides easy setup of Isabelle session directories,
   888 including proper document sources;
   889 
   890 * generated LaTeX sources are now deleted after successful run
   891 (isatool document -c); may retain a copy somewhere else via -D option
   892 of isatool usedir;
   893 
   894 * isatool usedir -D now lets isatool latex -o sty update the Isabelle
   895 style files, achieving self-contained LaTeX sources and simplifying
   896 LaTeX debugging;
   897 
   898 * old-style theories now produce (crude) LaTeX output as well;
   899 
   900 * browser info session directories are now self-contained (may be put
   901 on WWW server seperately); improved graphs of nested sessions; removed
   902 graph for 'all sessions';
   903 
   904 * several improvements in isabelle style files; \isabellestyle{it}
   905 produces fake math mode output; \isamarkupheader is now \section by
   906 default; see lib/texinputs/isabelle.sty etc.;
   907 
   908 
   909 *** Isar ***
   910 
   911 * Isar/Pure: local results and corresponding term bindings are now
   912 subject to Hindley-Milner polymorphism (similar to ML); this
   913 accommodates incremental type-inference very nicely;
   914 
   915 * Isar/Pure: new derived language element 'obtain' supports
   916 generalized existence reasoning;
   917 
   918 * Isar/Pure: new calculational elements 'moreover' and 'ultimately'
   919 support accumulation of results, without applying any rules yet;
   920 useful to collect intermediate results without explicit name
   921 references, and for use with transitivity rules with more than 2
   922 premises;
   923 
   924 * Isar/Pure: scalable support for case-analysis type proofs: new
   925 'case' language element refers to local contexts symbolically, as
   926 produced by certain proof methods; internally, case names are attached
   927 to theorems as "tags";
   928 
   929 * Isar/Pure: theory command 'hide' removes declarations from
   930 class/type/const name spaces;
   931 
   932 * Isar/Pure: theory command 'defs' supports option "(overloaded)" to
   933 indicate potential overloading;
   934 
   935 * Isar/Pure: changed syntax of local blocks from {{ }} to { };
   936 
   937 * Isar/Pure: syntax of sorts made 'inner', i.e. have to write
   938 "{a,b,c}" instead of {a,b,c};
   939 
   940 * Isar/Pure now provides its own version of intro/elim/dest
   941 attributes; useful for building new logics, but beware of confusion
   942 with the version in Provers/classical;
   943 
   944 * Isar/Pure: the local context of (non-atomic) goals is provided via
   945 case name 'antecedent';
   946 
   947 * Isar/Pure: removed obsolete 'transfer' attribute (transfer of thms
   948 to the current context is now done automatically);
   949 
   950 * Isar/Pure: theory command 'method_setup' provides a simple interface
   951 for definining proof methods in ML;
   952 
   953 * Isar/Provers: intro/elim/dest attributes changed; renamed
   954 intro/intro!/intro!! flags to intro!/intro/intro? (INCOMPATIBILITY, in
   955 most cases, one should have to change intro!! to intro? only);
   956 replaced "delrule" by "rule del";
   957 
   958 * Isar/Provers: new 'hypsubst' method, plain 'subst' method and
   959 'symmetric' attribute (the latter supercedes [RS sym]);
   960 
   961 * Isar/Provers: splitter support (via 'split' attribute and 'simp'
   962 method modifier); 'simp' method: 'only:' modifier removes loopers as
   963 well (including splits);
   964 
   965 * Isar/Provers: Simplifier and Classical methods now support all kind
   966 of modifiers used in the past, including 'cong', 'iff', etc.
   967 
   968 * Isar/Provers: added 'fastsimp' and 'clarsimp' methods (combination
   969 of Simplifier and Classical reasoner);
   970 
   971 * Isar/HOL: new proof method 'cases' and improved version of 'induct'
   972 now support named cases; major packages (inductive, datatype, primrec,
   973 recdef) support case names and properly name parameters;
   974 
   975 * Isar/HOL: new transitivity rules for substitution in inequalities --
   976 monotonicity conditions are extracted to be proven at end of
   977 calculations;
   978 
   979 * Isar/HOL: removed 'case_split' thm binding, should use 'cases' proof
   980 method anyway;
   981 
   982 * Isar/HOL: removed old expand_if = split_if; theorems if_splits =
   983 split_if split_if_asm; datatype package provides theorems foo.splits =
   984 foo.split foo.split_asm for each datatype;
   985 
   986 * Isar/HOL: tuned inductive package, rename "intrs" to "intros"
   987 (potential INCOMPATIBILITY), emulation of mk_cases feature for proof
   988 scripts: new 'inductive_cases' command and 'ind_cases' method; (Note:
   989 use "(cases (simplified))" method in proper proof texts);
   990 
   991 * Isar/HOL: added global 'arith_split' attribute for 'arith' method;
   992 
   993 * Isar: names of theorems etc. may be natural numbers as well;
   994 
   995 * Isar: 'pr' command: optional arguments for goals_limit and
   996 ProofContext.prems_limit; no longer prints theory contexts, but only
   997 proof states;
   998 
   999 * Isar: diagnostic commands 'pr', 'thm', 'prop', 'term', 'typ' admit
  1000 additional print modes to be specified; e.g. "pr(latex)" will print
  1001 proof state according to the Isabelle LaTeX style;
  1002 
  1003 * Isar: improved support for emulating tactic scripts, including proof
  1004 methods 'rule_tac' etc., 'cut_tac', 'thin_tac', 'subgoal_tac',
  1005 'rename_tac', 'rotate_tac', 'tactic', and 'case_tac' / 'induct_tac'
  1006 (for HOL datatypes);
  1007 
  1008 * Isar: simplified (more robust) goal selection of proof methods: 1st
  1009 goal, all goals, or explicit goal specifier (tactic emulation); thus
  1010 'proof method scripts' have to be in depth-first order;
  1011 
  1012 * Isar: tuned 'let' syntax: replaced 'as' keyword by 'and';
  1013 
  1014 * Isar: removed 'help' command, which hasn't been too helpful anyway;
  1015 should instead use individual commands for printing items
  1016 (print_commands, print_methods etc.);
  1017 
  1018 * Isar: added 'nothing' --- the empty list of theorems;
  1019 
  1020 
  1021 *** HOL ***
  1022 
  1023 * HOL/MicroJava: formalization of a fragment of Java, together with a
  1024 corresponding virtual machine and a specification of its bytecode
  1025 verifier and a lightweight bytecode verifier, including proofs of
  1026 type-safety; by Gerwin Klein, Tobias Nipkow, David von Oheimb, and
  1027 Cornelia Pusch (see also the homepage of project Bali at
  1028 http://isabelle.in.tum.de/Bali/);
  1029 
  1030 * HOL/Algebra: new theory of rings and univariate polynomials, by
  1031 Clemens Ballarin;
  1032 
  1033 * HOL/NumberTheory: fundamental Theorem of Arithmetic, Chinese
  1034 Remainder Theorem, Fermat/Euler Theorem, Wilson's Theorem, by Thomas M
  1035 Rasmussen;
  1036 
  1037 * HOL/Lattice: fundamental concepts of lattice theory and order
  1038 structures, including duals, properties of bounds versus algebraic
  1039 laws, lattice operations versus set-theoretic ones, the Knaster-Tarski
  1040 Theorem for complete lattices etc.; may also serve as a demonstration
  1041 for abstract algebraic reasoning using axiomatic type classes, and
  1042 mathematics-style proof in Isabelle/Isar; by Markus Wenzel;
  1043 
  1044 * HOL/Prolog: a (bare-bones) implementation of Lambda-Prolog, by David
  1045 von Oheimb;
  1046 
  1047 * HOL/IMPP: extension of IMP with local variables and mutually
  1048 recursive procedures, by David von Oheimb;
  1049 
  1050 * HOL/Lambda: converted into new-style theory and document;
  1051 
  1052 * HOL/ex/Multiquote: example of multiple nested quotations and
  1053 anti-quotations -- basically a generalized version of de-Bruijn
  1054 representation; very useful in avoiding lifting of operations;
  1055 
  1056 * HOL/record: added general record equality rule to simpset; fixed
  1057 select-update simplification procedure to handle extended records as
  1058 well; admit "r" as field name;
  1059 
  1060 * HOL: 0 is now overloaded over the new sort "zero", allowing its use with
  1061 other numeric types and also as the identity of groups, rings, etc.;
  1062 
  1063 * HOL: new axclass plus_ac0 for addition with the AC-laws and 0 as identity.
  1064 Types nat and int belong to this axclass;
  1065 
  1066 * HOL: greatly improved simplification involving numerals of type nat, int, real:
  1067    (i + #8 + j) = Suc k simplifies to  #7 + (i + j) = k
  1068    i*j + k + j*#3*i     simplifies to  #4*(i*j) + k
  1069   two terms #m*u and #n*u are replaced by #(m+n)*u
  1070     (where #m, #n and u can implicitly be 1; this is simproc combine_numerals)
  1071   and the term/formula #m*u+x ~~ #n*u+y simplifies simplifies to #(m-n)+x ~~ y
  1072     or x ~~ #(n-m)+y, where ~~ is one of = < <= or - (simproc cancel_numerals);
  1073 
  1074 * HOL: meson_tac is available (previously in ex/meson.ML); it is a
  1075 powerful prover for predicate logic but knows nothing of clasets; see
  1076 ex/mesontest.ML and ex/mesontest2.ML for example applications;
  1077 
  1078 * HOL: new version of "case_tac" subsumes both boolean case split and
  1079 "exhaust_tac" on datatypes; INCOMPATIBILITY: exhaust_tac no longer
  1080 exists, may define val exhaust_tac = case_tac for ad-hoc portability;
  1081 
  1082 * HOL: simplification no longer dives into case-expressions: only the
  1083 selector expression is simplified, but not the remaining arms; to
  1084 enable full simplification of case-expressions for datatype t, you may
  1085 remove t.weak_case_cong from the simpset, either globally (Delcongs
  1086 [thm"t.weak_case_cong"];) or locally (delcongs [...]).
  1087 
  1088 * HOL/recdef: the recursion equations generated by 'recdef' for
  1089 function 'f' are now called f.simps instead of f.rules; if all
  1090 termination conditions are proved automatically, these simplification
  1091 rules are added to the simpset, as in primrec; rules may be named
  1092 individually as well, resulting in a separate list of theorems for
  1093 each equation;
  1094 
  1095 * HOL/While is a new theory that provides a while-combinator. It
  1096 permits the definition of tail-recursive functions without the
  1097 provision of a termination measure. The latter is necessary once the
  1098 invariant proof rule for while is applied.
  1099 
  1100 * HOL: new (overloaded) notation for the set of elements below/above
  1101 some element: {..u}, {..u(}, {l..}, {)l..}. See theory SetInterval.
  1102 
  1103 * HOL: theorems impI, allI, ballI bound as "strip";
  1104 
  1105 * HOL: new tactic induct_thm_tac: thm -> string -> int -> tactic
  1106 induct_tac th "x1 ... xn" expects th to have a conclusion of the form
  1107 P v1 ... vn and abbreviates res_inst_tac [("v1","x1"),...,("vn","xn")] th;
  1108 
  1109 * HOL/Real: "rabs" replaced by overloaded "abs" function;
  1110 
  1111 * HOL: theory Sexp now in HOL/Induct examples (it used to be part of
  1112 main HOL, but was unused);
  1113 
  1114 * HOL: fewer consts declared as global (e.g. have to refer to
  1115 "Lfp.lfp" instead of "lfp" internally; affects ML packages only);
  1116 
  1117 * HOL: tuned AST representation of nested pairs, avoiding bogus output
  1118 in case of overlap with user translations (e.g. judgements over
  1119 tuples); (note that the underlying logical represenation is still
  1120 bogus);
  1121 
  1122 
  1123 *** ZF ***
  1124 
  1125 * ZF: simplification automatically cancels common terms in arithmetic
  1126 expressions over nat and int;
  1127 
  1128 * ZF: new treatment of nat to minimize type-checking: all operators
  1129 coerce their operands to a natural number using the function natify,
  1130 making the algebraic laws unconditional;
  1131 
  1132 * ZF: as above, for int: operators coerce their operands to an integer
  1133 using the function intify;
  1134 
  1135 * ZF: the integer library now contains many of the usual laws for the
  1136 orderings, including $<=, and monotonicity laws for $+ and $*;
  1137 
  1138 * ZF: new example ZF/ex/NatSum to demonstrate integer arithmetic
  1139 simplification;
  1140 
  1141 * FOL and ZF: AddIffs now available, giving theorems of the form P<->Q
  1142 to the simplifier and classical reasoner simultaneously;
  1143 
  1144 
  1145 *** General ***
  1146 
  1147 * Provers: blast_tac now handles actual object-logic rules as
  1148 assumptions; note that auto_tac uses blast_tac internally as well;
  1149 
  1150 * Provers: new functions rulify/rulify_no_asm: thm -> thm for turning
  1151 outer -->/All/Ball into ==>/!!; qed_spec_mp now uses rulify_no_asm;
  1152 
  1153 * Provers: delrules now handles destruct rules as well (no longer need
  1154 explicit make_elim);
  1155 
  1156 * Provers: Blast_tac now warns of and ignores "weak elimination rules" e.g.
  1157   [| inj ?f;          ?f ?x = ?f ?y; ?x = ?y ==> ?W |] ==> ?W
  1158 use instead the strong form,
  1159   [| inj ?f; ~ ?W ==> ?f ?x = ?f ?y; ?x = ?y ==> ?W |] ==> ?W
  1160 in HOL, FOL and ZF the function cla_make_elim will create such rules
  1161 from destruct-rules;
  1162 
  1163 * Provers: Simplifier.easy_setup provides a fast path to basic
  1164 Simplifier setup for new object-logics;
  1165 
  1166 * Pure: AST translation rules no longer require constant head on LHS;
  1167 
  1168 * Pure: improved name spaces: ambiguous output is qualified; support
  1169 for hiding of names;
  1170 
  1171 * system: smart setup of canonical ML_HOME, ISABELLE_INTERFACE, and
  1172 XSYMBOL_HOME; no longer need to do manual configuration in most
  1173 situations;
  1174 
  1175 * system: compression of ML heaps images may now be controlled via -c
  1176 option of isabelle and isatool usedir (currently only observed by
  1177 Poly/ML);
  1178 
  1179 * system: isatool installfonts may handle X-Symbol fonts as well (very
  1180 useful for remote X11);
  1181 
  1182 * system: provide TAGS file for Isabelle sources;
  1183 
  1184 * ML: infix 'OF' is a version of 'MRS' with more appropriate argument
  1185 order;
  1186 
  1187 * ML: renamed flags Syntax.trace_norm_ast to Syntax.trace_ast; global
  1188 timing flag supersedes proof_timing and Toplevel.trace;
  1189 
  1190 * ML: new combinators |>> and |>>> for incremental transformations
  1191 with secondary results (e.g. certain theory extensions):
  1192 
  1193 * ML: PureThy.add_defs gets additional argument to indicate potential
  1194 overloading (usually false);
  1195 
  1196 * ML: PureThy.add_thms/add_axioms/add_defs now return theorems as
  1197 results;
  1198 
  1199 
  1200 
  1201 New in Isabelle99 (October 1999)
  1202 --------------------------------
  1203 
  1204 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  1205 
  1206 * HOL: The THEN and ELSE parts of conditional expressions (if P then x else y)
  1207 are no longer simplified.  (This allows the simplifier to unfold recursive
  1208 functional programs.)  To restore the old behaviour, declare
  1209 
  1210     Delcongs [if_weak_cong];
  1211 
  1212 * HOL: Removed the obsolete syntax "Compl A"; use -A for set
  1213 complement;
  1214 
  1215 * HOL: the predicate "inj" is now defined by translation to "inj_on";
  1216 
  1217 * HOL/datatype: mutual_induct_tac no longer exists --
  1218   use induct_tac "x_1 ... x_n" instead of mutual_induct_tac ["x_1", ..., "x_n"]
  1219 
  1220 * HOL/typedef: fixed type inference for representing set; type
  1221 arguments now have to occur explicitly on the rhs as type constraints;
  1222 
  1223 * ZF: The con_defs part of an inductive definition may no longer refer
  1224 to constants declared in the same theory;
  1225 
  1226 * HOL, ZF: the function mk_cases, generated by the inductive
  1227 definition package, has lost an argument.  To simplify its result, it
  1228 uses the default simpset instead of a supplied list of theorems.
  1229 
  1230 * HOL/List: the constructors of type list are now Nil and Cons;
  1231 
  1232 * Simplifier: the type of the infix ML functions
  1233         setSSolver addSSolver setSolver addSolver
  1234 is now  simpset * solver -> simpset  where `solver' is a new abstract type
  1235 for packaging solvers. A solver is created via
  1236         mk_solver: string -> (thm list -> int -> tactic) -> solver
  1237 where the string argument is only a comment.
  1238 
  1239 
  1240 *** Proof tools ***
  1241 
  1242 * Provers/Arith/fast_lin_arith.ML contains a functor for creating a
  1243 decision procedure for linear arithmetic. Currently it is used for
  1244 types `nat', `int', and `real' in HOL (see below); it can, should and
  1245 will be instantiated for other types and logics as well.
  1246 
  1247 * The simplifier now accepts rewrite rules with flexible heads, eg
  1248      hom ?f ==> ?f(?x+?y) = ?f ?x + ?f ?y
  1249   They are applied like any rule with a non-pattern lhs, i.e. by first-order
  1250   matching.
  1251 
  1252 
  1253 *** General ***
  1254 
  1255 * New Isabelle/Isar subsystem provides an alternative to traditional
  1256 tactical theorem proving; together with the ProofGeneral/isar user
  1257 interface it offers an interactive environment for developing human
  1258 readable proof documents (Isar == Intelligible semi-automated
  1259 reasoning); for further information see isatool doc isar-ref,
  1260 src/HOL/Isar_examples and http://isabelle.in.tum.de/Isar/
  1261 
  1262 * improved and simplified presentation of theories: better HTML markup
  1263 (including colors), graph views in several sizes; isatool usedir now
  1264 provides a proper interface for user theories (via -P option); actual
  1265 document preparation based on (PDF)LaTeX is available as well (for
  1266 new-style theories only); see isatool doc system for more information;
  1267 
  1268 * native support for Proof General, both for classic Isabelle and
  1269 Isabelle/Isar;
  1270 
  1271 * ML function thm_deps visualizes dependencies of theorems and lemmas,
  1272 using the graph browser tool;
  1273 
  1274 * Isabelle manuals now also available as PDF;
  1275 
  1276 * theory loader rewritten from scratch (may not be fully
  1277 bug-compatible); old loadpath variable has been replaced by show_path,
  1278 add_path, del_path, reset_path functions; new operations such as
  1279 update_thy, touch_thy, remove_thy, use/update_thy_only (see also
  1280 isatool doc ref);
  1281 
  1282 * improved isatool install: option -k creates KDE application icon,
  1283 option -p DIR installs standalone binaries;
  1284 
  1285 * added ML_PLATFORM setting (useful for cross-platform installations);
  1286 more robust handling of platform specific ML images for SML/NJ;
  1287 
  1288 * the settings environment is now statically scoped, i.e. it is never
  1289 created again in sub-processes invoked from isabelle, isatool, or
  1290 Isabelle;
  1291 
  1292 * path element specification '~~' refers to '$ISABELLE_HOME';
  1293 
  1294 * in locales, the "assumes" and "defines" parts may be omitted if
  1295 empty;
  1296 
  1297 * new print_mode "xsymbols" for extended symbol support (e.g. genuine
  1298 long arrows);
  1299 
  1300 * new print_mode "HTML";
  1301 
  1302 * new flag show_tags controls display of tags of theorems (which are
  1303 basically just comments that may be attached by some tools);
  1304 
  1305 * Isamode 2.6 requires patch to accomodate change of Isabelle font
  1306 mode and goal output format:
  1307 
  1308 diff -r Isamode-2.6/elisp/isa-load.el Isamode/elisp/isa-load.el
  1309 244c244
  1310 <       (list (isa-getenv "ISABELLE") "-msymbols" logic-name)
  1311 ---
  1312 >       (list (isa-getenv "ISABELLE") "-misabelle_font" "-msymbols" logic-name)
  1313 diff -r Isabelle-2.6/elisp/isa-proofstate.el Isamode/elisp/isa-proofstate.el
  1314 181c181
  1315 < (defconst proofstate-proofstart-regexp "^Level [0-9]+$"
  1316 ---
  1317 > (defconst proofstate-proofstart-regexp "^Level [0-9]+"
  1318 
  1319 * function bind_thms stores lists of theorems (cf. bind_thm);
  1320 
  1321 * new shorthand tactics ftac, eatac, datac, fatac;
  1322 
  1323 * qed (and friends) now accept "" as result name; in that case the
  1324 theorem is not stored, but proper checks and presentation of the
  1325 result still apply;
  1326 
  1327 * theorem database now also indexes constants "Trueprop", "all",
  1328 "==>", "=="; thus thms_containing, findI etc. may retrieve more rules;
  1329 
  1330 
  1331 *** HOL ***
  1332 
  1333 ** HOL arithmetic **
  1334 
  1335 * There are now decision procedures for linear arithmetic over nat and
  1336 int:
  1337 
  1338 1. arith_tac copes with arbitrary formulae involving `=', `<', `<=',
  1339 `+', `-', `Suc', `min', `max' and numerical constants; other subterms
  1340 are treated as atomic; subformulae not involving type `nat' or `int'
  1341 are ignored; quantified subformulae are ignored unless they are
  1342 positive universal or negative existential. The tactic has to be
  1343 invoked by hand and can be a little bit slow. In particular, the
  1344 running time is exponential in the number of occurrences of `min' and
  1345 `max', and `-' on `nat'.
  1346 
  1347 2. fast_arith_tac is a cut-down version of arith_tac: it only takes
  1348 (negated) (in)equalities among the premises and the conclusion into
  1349 account (i.e. no compound formulae) and does not know about `min' and
  1350 `max', and `-' on `nat'. It is fast and is used automatically by the
  1351 simplifier.
  1352 
  1353 NB: At the moment, these decision procedures do not cope with mixed
  1354 nat/int formulae where the two parts interact, such as `m < n ==>
  1355 int(m) < int(n)'.
  1356 
  1357 * HOL/Numeral provides a generic theory of numerals (encoded
  1358 efficiently as bit strings); setup for types nat/int/real is in place;
  1359 INCOMPATIBILITY: since numeral syntax is now polymorphic, rather than
  1360 int, existing theories and proof scripts may require a few additional
  1361 type constraints;
  1362 
  1363 * integer division and remainder can now be performed on constant
  1364 arguments;
  1365 
  1366 * many properties of integer multiplication, division and remainder
  1367 are now available;
  1368 
  1369 * An interface to the Stanford Validity Checker (SVC) is available through the
  1370 tactic svc_tac.  Propositional tautologies and theorems of linear arithmetic
  1371 are proved automatically.  SVC must be installed separately, and its results
  1372 must be TAKEN ON TRUST (Isabelle does not check the proofs, but tags any
  1373 invocation of the underlying oracle).  For SVC see
  1374   http://verify.stanford.edu/SVC
  1375 
  1376 * IsaMakefile: the HOL-Real target now builds an actual image;
  1377 
  1378 
  1379 ** HOL misc **
  1380 
  1381 * HOL/Real/HahnBanach: the Hahn-Banach theorem for real vector spaces
  1382 (in Isabelle/Isar) -- by Gertrud Bauer;
  1383 
  1384 * HOL/BCV: generic model of bytecode verification, i.e. data-flow
  1385 analysis for assembly languages with subtypes;
  1386 
  1387 * HOL/TLA (Lamport's Temporal Logic of Actions): major reorganization
  1388 -- avoids syntactic ambiguities and treats state, transition, and
  1389 temporal levels more uniformly; introduces INCOMPATIBILITIES due to
  1390 changed syntax and (many) tactics;
  1391 
  1392 * HOL/inductive: Now also handles more general introduction rules such
  1393   as "ALL y. (y, x) : r --> y : acc r ==> x : acc r"; monotonicity
  1394   theorems are now maintained within the theory (maintained via the
  1395   "mono" attribute);
  1396 
  1397 * HOL/datatype: Now also handles arbitrarily branching datatypes
  1398   (using function types) such as
  1399 
  1400   datatype 'a tree = Atom 'a | Branch "nat => 'a tree"
  1401 
  1402 * HOL/record: record_simproc (part of the default simpset) takes care
  1403 of selectors applied to updated records; record_split_tac is no longer
  1404 part of the default claset; update_defs may now be removed from the
  1405 simpset in many cases; COMPATIBILITY: old behavior achieved by
  1406 
  1407   claset_ref () := claset() addSWrapper record_split_wrapper;
  1408   Delsimprocs [record_simproc]
  1409 
  1410 * HOL/typedef: fixed type inference for representing set; type
  1411 arguments now have to occur explicitly on the rhs as type constraints;
  1412 
  1413 * HOL/recdef (TFL): 'congs' syntax now expects comma separated list of theorem
  1414 names rather than an ML expression;
  1415 
  1416 * HOL/defer_recdef (TFL): like recdef but the well-founded relation can be
  1417 supplied later.  Program schemes can be defined, such as
  1418     "While B C s = (if B s then While B C (C s) else s)"
  1419 where the well-founded relation can be chosen after B and C have been given.
  1420 
  1421 * HOL/List: the constructors of type list are now Nil and Cons;
  1422 INCOMPATIBILITY: while [] and infix # syntax is still there, of
  1423 course, ML tools referring to List.list.op # etc. have to be adapted;
  1424 
  1425 * HOL_quantifiers flag superseded by "HOL" print mode, which is
  1426 disabled by default; run isabelle with option -m HOL to get back to
  1427 the original Gordon/HOL-style output;
  1428 
  1429 * HOL/Ord.thy: new bounded quantifier syntax (input only): ALL x<y. P,
  1430 ALL x<=y. P, EX x<y. P, EX x<=y. P;
  1431 
  1432 * HOL basic syntax simplified (more orthogonal): all variants of
  1433 All/Ex now support plain / symbolic / HOL notation; plain syntax for
  1434 Eps operator is provided as well: "SOME x. P[x]";
  1435 
  1436 * HOL/Sum.thy: sum_case has been moved to HOL/Datatype;
  1437 
  1438 * HOL/Univ.thy: infix syntax <*>, <+>, <**>, <+> eliminated and made
  1439 thus available for user theories;
  1440 
  1441 * HOLCF/IOA/Sequents: renamed 'Cons' to 'Consq' to avoid clash with
  1442 HOL/List; hardly an INCOMPATIBILITY since '>>' syntax is used all the
  1443 time;
  1444 
  1445 * HOL: new tactic smp_tac: int -> int -> tactic, which applies spec
  1446 several times and then mp;
  1447 
  1448 
  1449 *** LK ***
  1450 
  1451 * the notation <<...>> is now available as a notation for sequences of
  1452 formulas;
  1453 
  1454 * the simplifier is now installed
  1455 
  1456 * the axiom system has been generalized (thanks to Soren Heilmann)
  1457 
  1458 * the classical reasoner now has a default rule database
  1459 
  1460 
  1461 *** ZF ***
  1462 
  1463 * new primrec section allows primitive recursive functions to be given
  1464 directly (as in HOL) over datatypes and the natural numbers;
  1465 
  1466 * new tactics induct_tac and exhaust_tac for induction (or case
  1467 analysis) over datatypes and the natural numbers;
  1468 
  1469 * the datatype declaration of type T now defines the recursor T_rec;
  1470 
  1471 * simplification automatically does freeness reasoning for datatype
  1472 constructors;
  1473 
  1474 * automatic type-inference, with AddTCs command to insert new
  1475 type-checking rules;
  1476 
  1477 * datatype introduction rules are now added as Safe Introduction rules
  1478 to the claset;
  1479 
  1480 * the syntax "if P then x else y" is now available in addition to
  1481 if(P,x,y);
  1482 
  1483 
  1484 *** Internal programming interfaces ***
  1485 
  1486 * tuned simplifier trace output; new flag debug_simp;
  1487 
  1488 * structures Vartab / Termtab (instances of TableFun) offer efficient
  1489 tables indexed by indexname_ord / term_ord (compatible with aconv);
  1490 
  1491 * AxClass.axclass_tac lost the theory argument;
  1492 
  1493 * tuned current_goals_markers semantics: begin / end goal avoids
  1494 printing empty lines;
  1495 
  1496 * removed prs and prs_fn hook, which was broken because it did not
  1497 include \n in its semantics, forcing writeln to add one
  1498 uncoditionally; replaced prs_fn by writeln_fn; consider std_output:
  1499 string -> unit if you really want to output text without newline;
  1500 
  1501 * Symbol.output subject to print mode; INCOMPATIBILITY: defaults to
  1502 plain output, interface builders may have to enable 'isabelle_font'
  1503 mode to get Isabelle font glyphs as before;
  1504 
  1505 * refined token_translation interface; INCOMPATIBILITY: output length
  1506 now of type real instead of int;
  1507 
  1508 * theory loader actions may be traced via new ThyInfo.add_hook
  1509 interface (see src/Pure/Thy/thy_info.ML); example application: keep
  1510 your own database of information attached to *whole* theories -- as
  1511 opposed to intra-theory data slots offered via TheoryDataFun;
  1512 
  1513 * proper handling of dangling sort hypotheses (at last!);
  1514 Thm.strip_shyps and Drule.strip_shyps_warning take care of removing
  1515 extra sort hypotheses that can be witnessed from the type signature;
  1516 the force_strip_shyps flag is gone, any remaining shyps are simply
  1517 left in the theorem (with a warning issued by strip_shyps_warning);
  1518 
  1519 
  1520 
  1521 New in Isabelle98-1 (October 1998)
  1522 ----------------------------------
  1523 
  1524 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  1525 
  1526 * several changes of automated proof tools;
  1527 
  1528 * HOL: major changes to the inductive and datatype packages, including
  1529 some minor incompatibilities of theory syntax;
  1530 
  1531 * HOL: renamed r^-1 to 'converse' from 'inverse'; 'inj_onto' is now
  1532 called `inj_on';
  1533 
  1534 * HOL: removed duplicate thms in Arith:
  1535   less_imp_add_less  should be replaced by  trans_less_add1
  1536   le_imp_add_le      should be replaced by  trans_le_add1
  1537 
  1538 * HOL: unary minus is now overloaded (new type constraints may be
  1539 required);
  1540 
  1541 * HOL and ZF: unary minus for integers is now #- instead of #~.  In
  1542 ZF, expressions such as n#-1 must be changed to n#- 1, since #-1 is
  1543 now taken as an integer constant.
  1544 
  1545 * Pure: ML function 'theory_of' renamed to 'theory';
  1546 
  1547 
  1548 *** Proof tools ***
  1549 
  1550 * Simplifier:
  1551   1. Asm_full_simp_tac is now more aggressive.
  1552      1. It will sometimes reorient premises if that increases their power to
  1553         simplify.
  1554      2. It does no longer proceed strictly from left to right but may also
  1555         rotate premises to achieve further simplification.
  1556      For compatibility reasons there is now Asm_lr_simp_tac which is like the
  1557      old Asm_full_simp_tac in that it does not rotate premises.
  1558   2. The simplifier now knows a little bit about nat-arithmetic.
  1559 
  1560 * Classical reasoner: wrapper mechanism for the classical reasoner now
  1561 allows for selected deletion of wrappers, by introduction of names for
  1562 wrapper functionals.  This implies that addbefore, addSbefore,
  1563 addaltern, and addSaltern now take a pair (name, tactic) as argument,
  1564 and that adding two tactics with the same name overwrites the first
  1565 one (emitting a warning).
  1566   type wrapper = (int -> tactic) -> (int -> tactic)
  1567   setWrapper, setSWrapper, compWrapper and compSWrapper are replaced by
  1568   addWrapper, addSWrapper: claset * (string * wrapper) -> claset
  1569   delWrapper, delSWrapper: claset *  string            -> claset
  1570   getWrapper is renamed to appWrappers, getSWrapper to appSWrappers;
  1571 
  1572 * Classical reasoner: addbefore/addSbefore now have APPEND/ORELSE
  1573 semantics; addbefore now affects only the unsafe part of step_tac
  1574 etc.; this affects addss/auto_tac/force_tac, so EXISTING PROOFS MAY
  1575 FAIL, but proofs should be fixable easily, e.g. by replacing Auto_tac
  1576 by Force_tac;
  1577 
  1578 * Classical reasoner: setwrapper to setWrapper and compwrapper to
  1579 compWrapper; added safe wrapper (and access functions for it);
  1580 
  1581 * HOL/split_all_tac is now much faster and fails if there is nothing
  1582 to split.  Some EXISTING PROOFS MAY REQUIRE ADAPTION because the order
  1583 and the names of the automatically generated variables have changed.
  1584 split_all_tac has moved within claset() from unsafe wrappers to safe
  1585 wrappers, which means that !!-bound variables are split much more
  1586 aggressively, and safe_tac and clarify_tac now split such variables.
  1587 If this splitting is not appropriate, use delSWrapper "split_all_tac".
  1588 Note: the same holds for record_split_tac, which does the job of
  1589 split_all_tac for record fields.
  1590 
  1591 * HOL/Simplifier: Rewrite rules for case distinctions can now be added
  1592 permanently to the default simpset using Addsplits just like
  1593 Addsimps. They can be removed via Delsplits just like
  1594 Delsimps. Lower-case versions are also available.
  1595 
  1596 * HOL/Simplifier: The rule split_if is now part of the default
  1597 simpset. This means that the simplifier will eliminate all occurrences
  1598 of if-then-else in the conclusion of a goal. To prevent this, you can
  1599 either remove split_if completely from the default simpset by
  1600 `Delsplits [split_if]' or remove it in a specific call of the
  1601 simplifier using `... delsplits [split_if]'.  You can also add/delete
  1602 other case splitting rules to/from the default simpset: every datatype
  1603 generates suitable rules `split_t_case' and `split_t_case_asm' (where
  1604 t is the name of the datatype).
  1605 
  1606 * Classical reasoner / Simplifier combination: new force_tac (and
  1607 derivatives Force_tac, force) combines rewriting and classical
  1608 reasoning (and whatever other tools) similarly to auto_tac, but is
  1609 aimed to solve the given subgoal completely.
  1610 
  1611 
  1612 *** General ***
  1613 
  1614 * new top-level commands `Goal' and `Goalw' that improve upon `goal'
  1615 and `goalw': the theory is no longer needed as an explicit argument -
  1616 the current theory context is used; assumptions are no longer returned
  1617 at the ML-level unless one of them starts with ==> or !!; it is
  1618 recommended to convert to these new commands using isatool fixgoal
  1619 (backup your sources first!);
  1620 
  1621 * new top-level commands 'thm' and 'thms' for retrieving theorems from
  1622 the current theory context, and 'theory' to lookup stored theories;
  1623 
  1624 * new theory section 'locale' for declaring constants, assumptions and
  1625 definitions that have local scope;
  1626 
  1627 * new theory section 'nonterminals' for purely syntactic types;
  1628 
  1629 * new theory section 'setup' for generic ML setup functions
  1630 (e.g. package initialization);
  1631 
  1632 * the distribution now includes Isabelle icons: see
  1633 lib/logo/isabelle-{small,tiny}.xpm;
  1634 
  1635 * isatool install - install binaries with absolute references to
  1636 ISABELLE_HOME/bin;
  1637 
  1638 * isatool logo -- create instances of the Isabelle logo (as EPS);
  1639 
  1640 * print mode 'emacs' reserved for Isamode;
  1641 
  1642 * support multiple print (ast) translations per constant name;
  1643 
  1644 * theorems involving oracles are now printed with a suffixed [!];
  1645 
  1646 
  1647 *** HOL ***
  1648 
  1649 * there is now a tutorial on Isabelle/HOL (do 'isatool doc tutorial');
  1650 
  1651 * HOL/inductive package reorganized and improved: now supports mutual
  1652 definitions such as
  1653 
  1654   inductive EVEN ODD
  1655     intrs
  1656       null "0 : EVEN"
  1657       oddI "n : EVEN ==> Suc n : ODD"
  1658       evenI "n : ODD ==> Suc n : EVEN"
  1659 
  1660 new theorem list "elims" contains an elimination rule for each of the
  1661 recursive sets; inductive definitions now handle disjunctive premises
  1662 correctly (also ZF);
  1663 
  1664 INCOMPATIBILITIES: requires Inductive as an ancestor; component
  1665 "mutual_induct" no longer exists - the induction rule is always
  1666 contained in "induct";
  1667 
  1668 
  1669 * HOL/datatype package re-implemented and greatly improved: now
  1670 supports mutually recursive datatypes such as
  1671 
  1672   datatype
  1673     'a aexp = IF_THEN_ELSE ('a bexp) ('a aexp) ('a aexp)
  1674             | SUM ('a aexp) ('a aexp)
  1675             | DIFF ('a aexp) ('a aexp)
  1676             | NUM 'a
  1677   and
  1678     'a bexp = LESS ('a aexp) ('a aexp)
  1679             | AND ('a bexp) ('a bexp)
  1680             | OR ('a bexp) ('a bexp)
  1681 
  1682 as well as indirectly recursive datatypes such as
  1683 
  1684   datatype
  1685     ('a, 'b) term = Var 'a
  1686                   | App 'b ((('a, 'b) term) list)
  1687 
  1688 The new tactic  mutual_induct_tac [<var_1>, ..., <var_n>] i  performs
  1689 induction on mutually / indirectly recursive datatypes.
  1690 
  1691 Primrec equations are now stored in theory and can be accessed via
  1692 <function_name>.simps.
  1693 
  1694 INCOMPATIBILITIES:
  1695 
  1696   - Theories using datatypes must now have theory Datatype as an
  1697     ancestor.
  1698   - The specific <typename>.induct_tac no longer exists - use the
  1699     generic induct_tac instead.
  1700   - natE has been renamed to nat.exhaust - use exhaust_tac
  1701     instead of res_inst_tac ... natE. Note that the variable
  1702     names in nat.exhaust differ from the names in natE, this
  1703     may cause some "fragile" proofs to fail.
  1704   - The theorems split_<typename>_case and split_<typename>_case_asm
  1705     have been renamed to <typename>.split and <typename>.split_asm.
  1706   - Since default sorts of type variables are now handled correctly,
  1707     some datatype definitions may have to be annotated with explicit
  1708     sort constraints.
  1709   - Primrec definitions no longer require function name and type
  1710     of recursive argument.
  1711 
  1712 Consider using isatool fixdatatype to adapt your theories and proof
  1713 scripts to the new package (backup your sources first!).
  1714 
  1715 
  1716 * HOL/record package: considerably improved implementation; now
  1717 includes concrete syntax for record types, terms, updates; theorems
  1718 for surjective pairing and splitting !!-bound record variables; proof
  1719 support is as follows:
  1720 
  1721   1) standard conversions (selectors or updates applied to record
  1722 constructor terms) are part of the standard simpset;
  1723 
  1724   2) inject equations of the form ((x, y) = (x', y')) == x=x' & y=y' are
  1725 made part of standard simpset and claset via addIffs;
  1726 
  1727   3) a tactic for record field splitting (record_split_tac) is part of
  1728 the standard claset (addSWrapper);
  1729 
  1730 To get a better idea about these rules you may retrieve them via
  1731 something like 'thms "foo.simps"' or 'thms "foo.iffs"', where "foo" is
  1732 the name of your record type.
  1733 
  1734 The split tactic 3) conceptually simplifies by the following rule:
  1735 
  1736   "(!!x. PROP ?P x) == (!!a b. PROP ?P (a, b))"
  1737 
  1738 Thus any record variable that is bound by meta-all will automatically
  1739 blow up into some record constructor term, consequently the
  1740 simplifications of 1), 2) apply.  Thus force_tac, auto_tac etc. shall
  1741 solve record problems automatically.
  1742 
  1743 
  1744 * reorganized the main HOL image: HOL/Integ and String loaded by
  1745 default; theory Main includes everything;
  1746 
  1747 * automatic simplification of integer sums and comparisons, using cancellation;
  1748 
  1749 * added option_map_eq_Some and not_Some_eq to the default simpset and claset;
  1750 
  1751 * added disj_not1 = "(~P | Q) = (P --> Q)" to the default simpset;
  1752 
  1753 * many new identities for unions, intersections, set difference, etc.;
  1754 
  1755 * expand_if, expand_split, expand_sum_case and expand_nat_case are now
  1756 called split_if, split_split, split_sum_case and split_nat_case (to go
  1757 with add/delsplits);
  1758 
  1759 * HOL/Prod introduces simplification procedure unit_eq_proc rewriting
  1760 (?x::unit) = (); this is made part of the default simpset, which COULD
  1761 MAKE EXISTING PROOFS FAIL under rare circumstances (consider
  1762 'Delsimprocs [unit_eq_proc];' as last resort); also note that
  1763 unit_abs_eta_conv is added in order to counter the effect of
  1764 unit_eq_proc on (%u::unit. f u), replacing it by f rather than by
  1765 %u.f();
  1766 
  1767 * HOL/Fun INCOMPATIBILITY: `inj_onto' is now called `inj_on' (which
  1768 makes more sense);
  1769 
  1770 * HOL/Set INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
  1771   It and 'sym RS equals0D' are now in the default  claset, giving automatic
  1772   disjointness reasoning but breaking a few old proofs.
  1773 
  1774 * HOL/Relation INCOMPATIBILITY: renamed the relational operator r^-1
  1775 to 'converse' from 'inverse' (for compatibility with ZF and some
  1776 literature);
  1777 
  1778 * HOL/recdef can now declare non-recursive functions, with {} supplied as
  1779 the well-founded relation;
  1780 
  1781 * HOL/Set INCOMPATIBILITY: the complement of set A is now written -A instead of
  1782     Compl A.  The "Compl" syntax remains available as input syntax for this
  1783     release ONLY.
  1784 
  1785 * HOL/Update: new theory of function updates:
  1786     f(a:=b) == %x. if x=a then b else f x
  1787 may also be iterated as in f(a:=b,c:=d,...);
  1788 
  1789 * HOL/Vimage: new theory for inverse image of a function, syntax f-``B;
  1790 
  1791 * HOL/List:
  1792   - new function list_update written xs[i:=v] that updates the i-th
  1793     list position. May also be iterated as in xs[i:=a,j:=b,...].
  1794   - new function `upt' written [i..j(] which generates the list
  1795     [i,i+1,...,j-1], i.e. the upper bound is excluded. To include the upper
  1796     bound write [i..j], which is a shorthand for [i..j+1(].
  1797   - new lexicographic orderings and corresponding wellfoundedness theorems.
  1798 
  1799 * HOL/Arith:
  1800   - removed 'pred' (predecessor) function;
  1801   - generalized some theorems about n-1;
  1802   - many new laws about "div" and "mod";
  1803   - new laws about greatest common divisors (see theory ex/Primes);
  1804 
  1805 * HOL/Relation: renamed the relational operator r^-1 "converse"
  1806 instead of "inverse";
  1807 
  1808 * HOL/Induct/Multiset: a theory of multisets, including the wellfoundedness
  1809   of the multiset ordering;
  1810 
  1811 * directory HOL/Real: a construction of the reals using Dedekind cuts
  1812   (not included by default);
  1813 
  1814 * directory HOL/UNITY: Chandy and Misra's UNITY formalism;
  1815 
  1816 * directory HOL/Hoare: a new version of Hoare logic which permits many-sorted
  1817   programs, i.e. different program variables may have different types.
  1818 
  1819 * calling (stac rew i) now fails if "rew" has no effect on the goal
  1820   [previously, this check worked only if the rewrite rule was unconditional]
  1821   Now rew can involve either definitions or equalities (either == or =).
  1822 
  1823 
  1824 *** ZF ***
  1825 
  1826 * theory Main includes everything; INCOMPATIBILITY: theory ZF.thy contains
  1827   only the theorems proved on ZF.ML;
  1828 
  1829 * ZF INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
  1830   It and 'sym RS equals0D' are now in the default  claset, giving automatic
  1831   disjointness reasoning but breaking a few old proofs.
  1832 
  1833 * ZF/Update: new theory of function updates
  1834     with default rewrite rule  f(x:=y) ` z = if(z=x, y, f`z)
  1835   may also be iterated as in f(a:=b,c:=d,...);
  1836 
  1837 * in  let x=t in u(x), neither t nor u(x) has to be an FOL term.
  1838 
  1839 * calling (stac rew i) now fails if "rew" has no effect on the goal
  1840   [previously, this check worked only if the rewrite rule was unconditional]
  1841   Now rew can involve either definitions or equalities (either == or =).
  1842 
  1843 * case_tac provided for compatibility with HOL
  1844     (like the old excluded_middle_tac, but with subgoals swapped)
  1845 
  1846 
  1847 *** Internal programming interfaces ***
  1848 
  1849 * Pure: several new basic modules made available for general use, see
  1850 also src/Pure/README;
  1851 
  1852 * improved the theory data mechanism to support encapsulation (data
  1853 kind name replaced by private Object.kind, acting as authorization
  1854 key); new type-safe user interface via functor TheoryDataFun; generic
  1855 print_data function becomes basically useless;
  1856 
  1857 * removed global_names compatibility flag -- all theory declarations
  1858 are qualified by default;
  1859 
  1860 * module Pure/Syntax now offers quote / antiquote translation
  1861 functions (useful for Hoare logic etc. with implicit dependencies);
  1862 see HOL/ex/Antiquote for an example use;
  1863 
  1864 * Simplifier now offers conversions (asm_)(full_)rewrite: simpset ->
  1865 cterm -> thm;
  1866 
  1867 * new tactical CHANGED_GOAL for checking that a tactic modifies a
  1868 subgoal;
  1869 
  1870 * Display.print_goals function moved to Locale.print_goals;
  1871 
  1872 * standard print function for goals supports current_goals_markers
  1873 variable for marking begin of proof, end of proof, start of goal; the
  1874 default is ("", "", ""); setting current_goals_markers := ("<proof>",
  1875 "</proof>", "<goal>") causes SGML like tagged proof state printing,
  1876 for example;
  1877 
  1878 
  1879 
  1880 New in Isabelle98 (January 1998)
  1881 --------------------------------
  1882 
  1883 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  1884 
  1885 * changed lexical syntax of terms / types: dots made part of long
  1886 identifiers, e.g. "%x.x" no longer possible, should be "%x. x";
  1887 
  1888 * simpset (and claset) reference variable replaced by functions
  1889 simpset / simpset_ref;
  1890 
  1891 * no longer supports theory aliases (via merge) and non-trivial
  1892 implicit merge of thms' signatures;
  1893 
  1894 * most internal names of constants changed due to qualified names;
  1895 
  1896 * changed Pure/Sequence interface (see Pure/seq.ML);
  1897 
  1898 
  1899 *** General Changes ***
  1900 
  1901 * hierachically structured name spaces (for consts, types, axms, thms
  1902 etc.); new lexical class 'longid' (e.g. Foo.bar.x) may render much of
  1903 old input syntactically incorrect (e.g. "%x.x"); COMPATIBILITY:
  1904 isatool fixdots ensures space after dots (e.g. "%x. x"); set
  1905 long_names for fully qualified output names; NOTE: ML programs
  1906 (special tactics, packages etc.) referring to internal names may have
  1907 to be adapted to cope with fully qualified names; in case of severe
  1908 backward campatibility problems try setting 'global_names' at compile
  1909 time to have enrything declared within a flat name space; one may also
  1910 fine tune name declarations in theories via the 'global' and 'local'
  1911 section;
  1912 
  1913 * reimplemented the implicit simpset and claset using the new anytype
  1914 data filed in signatures; references simpset:simpset ref etc. are
  1915 replaced by functions simpset:unit->simpset and
  1916 simpset_ref:unit->simpset ref; COMPATIBILITY: use isatool fixclasimp
  1917 to patch your ML files accordingly;
  1918 
  1919 * HTML output now includes theory graph data for display with Java
  1920 applet or isatool browser; data generated automatically via isatool
  1921 usedir (see -i option, ISABELLE_USEDIR_OPTIONS);
  1922 
  1923 * defs may now be conditional; improved rewrite_goals_tac to handle
  1924 conditional equations;
  1925 
  1926 * defs now admits additional type arguments, using TYPE('a) syntax;
  1927 
  1928 * theory aliases via merge (e.g. M=A+B+C) no longer supported, always
  1929 creates a new theory node; implicit merge of thms' signatures is
  1930 restricted to 'trivial' ones; COMPATIBILITY: one may have to use
  1931 transfer:theory->thm->thm in (rare) cases;
  1932 
  1933 * improved handling of draft signatures / theories; draft thms (and
  1934 ctyps, cterms) are automatically promoted to real ones;
  1935 
  1936 * slightly changed interfaces for oracles: admit many per theory, named
  1937 (e.g. oracle foo = mlfun), additional name argument for invoke_oracle;
  1938 
  1939 * print_goals: optional output of const types (set show_consts and
  1940 show_types);
  1941 
  1942 * improved output of warnings (###) and errors (***);
  1943 
  1944 * subgoal_tac displays a warning if the new subgoal has type variables;
  1945 
  1946 * removed old README and Makefiles;
  1947 
  1948 * replaced print_goals_ref hook by print_current_goals_fn and result_error_fn;
  1949 
  1950 * removed obsolete init_pps and init_database;
  1951 
  1952 * deleted the obsolete tactical STATE, which was declared by
  1953     fun STATE tacfun st = tacfun st st;
  1954 
  1955 * cd and use now support path variables, e.g. $ISABELLE_HOME, or ~
  1956 (which abbreviates $HOME);
  1957 
  1958 * changed Pure/Sequence interface (see Pure/seq.ML); COMPATIBILITY:
  1959 use isatool fixseq to adapt your ML programs (this works for fully
  1960 qualified references to the Sequence structure only!);
  1961 
  1962 * use_thy no longer requires writable current directory; it always
  1963 reloads .ML *and* .thy file, if either one is out of date;
  1964 
  1965 
  1966 *** Classical Reasoner ***
  1967 
  1968 * Clarify_tac, clarify_tac, clarify_step_tac, Clarify_step_tac: new
  1969 tactics that use classical reasoning to simplify a subgoal without
  1970 splitting it into several subgoals;
  1971 
  1972 * Safe_tac: like safe_tac but uses the default claset;
  1973 
  1974 
  1975 *** Simplifier ***
  1976 
  1977 * added simplification meta rules:
  1978     (asm_)(full_)simplify: simpset -> thm -> thm;
  1979 
  1980 * simplifier.ML no longer part of Pure -- has to be loaded by object
  1981 logics (again);
  1982 
  1983 * added prems argument to simplification procedures;
  1984 
  1985 * HOL, FOL, ZF: added infix function `addsplits':
  1986   instead of `<simpset> setloop (split_tac <thms>)'
  1987   you can simply write `<simpset> addsplits <thms>'
  1988 
  1989 
  1990 *** Syntax ***
  1991 
  1992 * TYPE('a) syntax for type reflection terms;
  1993 
  1994 * no longer handles consts with name "" -- declare as 'syntax' instead;
  1995 
  1996 * pretty printer: changed order of mixfix annotation preference (again!);
  1997 
  1998 * Pure: fixed idt/idts vs. pttrn/pttrns syntactic categories;
  1999 
  2000 
  2001 *** HOL ***
  2002 
  2003 * HOL: there is a new splitter `split_asm_tac' that can be used e.g.
  2004   with `addloop' of the simplifier to faciliate case splitting in premises.
  2005 
  2006 * HOL/TLA: Stephan Merz's formalization of Lamport's Temporal Logic of Actions;
  2007 
  2008 * HOL/Auth: new protocol proofs including some for the Internet
  2009   protocol TLS;
  2010 
  2011 * HOL/Map: new theory of `maps' a la VDM;
  2012 
  2013 * HOL/simplifier: simplification procedures nat_cancel_sums for
  2014 cancelling out common nat summands from =, <, <= (in)equalities, or
  2015 differences; simplification procedures nat_cancel_factor for
  2016 cancelling common factor from =, <, <= (in)equalities over natural
  2017 sums; nat_cancel contains both kinds of procedures, it is installed by
  2018 default in Arith.thy -- this COULD MAKE EXISTING PROOFS FAIL;
  2019 
  2020 * HOL/simplifier: terms of the form
  2021   `? x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x)'  (or t=x)
  2022   are rewritten to
  2023   `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t)',
  2024   and those of the form
  2025   `! x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x) --> R(x)'  (or t=x)
  2026   are rewritten to
  2027   `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t) --> R(t)',
  2028 
  2029 * HOL/datatype
  2030   Each datatype `t' now comes with a theorem `split_t_case' of the form
  2031 
  2032   P(t_case f1 ... fn x) =
  2033      ( (!y1 ... ym1. x = C1 y1 ... ym1 --> P(f1 y1 ... ym1)) &
  2034         ...
  2035        (!y1 ... ymn. x = Cn y1 ... ymn --> P(f1 y1 ... ymn))
  2036      )
  2037 
  2038   and a theorem `split_t_case_asm' of the form
  2039 
  2040   P(t_case f1 ... fn x) =
  2041     ~( (? y1 ... ym1. x = C1 y1 ... ym1 & ~P(f1 y1 ... ym1)) |
  2042         ...
  2043        (? y1 ... ymn. x = Cn y1 ... ymn & ~P(f1 y1 ... ymn))
  2044      )
  2045   which can be added to a simpset via `addsplits'. The existing theorems
  2046   expand_list_case and expand_option_case have been renamed to
  2047   split_list_case and split_option_case.
  2048 
  2049 * HOL/Arithmetic:
  2050   - `pred n' is automatically converted to `n-1'.
  2051     Users are strongly encouraged not to use `pred' any longer,
  2052     because it will disappear altogether at some point.
  2053   - Users are strongly encouraged to write "0 < n" rather than
  2054     "n ~= 0". Theorems and proof tools have been modified towards this
  2055     `standard'.
  2056 
  2057 * HOL/Lists:
  2058   the function "set_of_list" has been renamed "set" (and its theorems too);
  2059   the function "nth" now takes its arguments in the reverse order and
  2060   has acquired the infix notation "!" as in "xs!n".
  2061 
  2062 * HOL/Set: UNIV is now a constant and is no longer translated to Compl{};
  2063 
  2064 * HOL/Set: The operator (UN x.B x) now abbreviates (UN x:UNIV. B x) and its
  2065   specialist theorems (like UN1_I) are gone.  Similarly for (INT x.B x);
  2066 
  2067 * HOL/record: extensible records with schematic structural subtyping
  2068 (single inheritance); EXPERIMENTAL version demonstrating the encoding,
  2069 still lacks various theorems and concrete record syntax;
  2070 
  2071 
  2072 *** HOLCF ***
  2073 
  2074 * removed "axioms" and "generated by" sections;
  2075 
  2076 * replaced "ops" section by extended "consts" section, which is capable of
  2077   handling the continuous function space "->" directly;
  2078 
  2079 * domain package:
  2080   . proves theorems immediately and stores them in the theory,
  2081   . creates hierachical name space,
  2082   . now uses normal mixfix annotations (instead of cinfix...),
  2083   . minor changes to some names and values (for consistency),
  2084   . e.g. cases -> casedist, dists_eq -> dist_eqs, [take_lemma] -> take_lemmas,
  2085   . separator between mutual domain defs: changed "," to "and",
  2086   . improved handling of sort constraints;  now they have to
  2087     appear on the left-hand side of the equations only;
  2088 
  2089 * fixed LAM <x,y,zs>.b syntax;
  2090 
  2091 * added extended adm_tac to simplifier in HOLCF -- can now discharge
  2092 adm (%x. P (t x)), where P is chainfinite and t continuous;
  2093 
  2094 
  2095 *** FOL and ZF ***
  2096 
  2097 * FOL: there is a new splitter `split_asm_tac' that can be used e.g.
  2098   with `addloop' of the simplifier to faciliate case splitting in premises.
  2099 
  2100 * qed_spec_mp, qed_goal_spec_mp, qed_goalw_spec_mp are available, as
  2101 in HOL, they strip ALL and --> from proved theorems;
  2102 
  2103 
  2104 
  2105 New in Isabelle94-8 (May 1997)
  2106 ------------------------------
  2107 
  2108 *** General Changes ***
  2109 
  2110 * new utilities to build / run / maintain Isabelle etc. (in parts
  2111 still somewhat experimental); old Makefiles etc. still functional;
  2112 
  2113 * new 'Isabelle System Manual';
  2114 
  2115 * INSTALL text, together with ./configure and ./build scripts;
  2116 
  2117 * reimplemented type inference for greater efficiency, better error
  2118 messages and clean internal interface;
  2119 
  2120 * prlim command for dealing with lots of subgoals (an easier way of
  2121 setting goals_limit);
  2122 
  2123 
  2124 *** Syntax ***
  2125 
  2126 * supports alternative (named) syntax tables (parser and pretty
  2127 printer); internal interface is provided by add_modesyntax(_i);
  2128 
  2129 * Pure, FOL, ZF, HOL, HOLCF now support symbolic input and output; to
  2130 be used in conjunction with the Isabelle symbol font; uses the
  2131 "symbols" syntax table;
  2132 
  2133 * added token_translation interface (may translate name tokens in
  2134 arbitrary ways, dependent on their type (free, bound, tfree, ...) and
  2135 the current print_mode); IMPORTANT: user print translation functions
  2136 are responsible for marking newly introduced bounds
  2137 (Syntax.mark_boundT);
  2138 
  2139 * token translations for modes "xterm" and "xterm_color" that display
  2140 names in bold, underline etc. or colors (which requires a color
  2141 version of xterm);
  2142 
  2143 * infixes may now be declared with names independent of their syntax;
  2144 
  2145 * added typed_print_translation (like print_translation, but may
  2146 access type of constant);
  2147 
  2148 
  2149 *** Classical Reasoner ***
  2150 
  2151 Blast_tac: a new tactic!  It is often more powerful than fast_tac, but has
  2152 some limitations.  Blast_tac...
  2153   + ignores addss, addbefore, addafter; this restriction is intrinsic
  2154   + ignores elimination rules that don't have the correct format
  2155         (the conclusion MUST be a formula variable)
  2156   + ignores types, which can make HOL proofs fail
  2157   + rules must not require higher-order unification, e.g. apply_type in ZF
  2158     [message "Function Var's argument not a bound variable" relates to this]
  2159   + its proof strategy is more general but can actually be slower
  2160 
  2161 * substitution with equality assumptions no longer permutes other
  2162 assumptions;
  2163 
  2164 * minor changes in semantics of addafter (now called addaltern); renamed
  2165 setwrapper to setWrapper and compwrapper to compWrapper; added safe wrapper
  2166 (and access functions for it);
  2167 
  2168 * improved combination of classical reasoner and simplifier:
  2169   + functions for handling clasimpsets
  2170   + improvement of addss: now the simplifier is called _after_ the
  2171     safe steps.
  2172   + safe variant of addss called addSss: uses safe simplifications
  2173     _during_ the safe steps. It is more complete as it allows multiple
  2174     instantiations of unknowns (e.g. with slow_tac).
  2175 
  2176 *** Simplifier ***
  2177 
  2178 * added interface for simplification procedures (functions that
  2179 produce *proven* rewrite rules on the fly, depending on current
  2180 redex);
  2181 
  2182 * ordering on terms as parameter (used for ordered rewriting);
  2183 
  2184 * new functions delcongs, deleqcongs, and Delcongs. richer rep_ss;
  2185 
  2186 * the solver is now split into a safe and an unsafe part.
  2187 This should be invisible for the normal user, except that the
  2188 functions setsolver and addsolver have been renamed to setSolver and
  2189 addSolver; added safe_asm_full_simp_tac;
  2190 
  2191 
  2192 *** HOL ***
  2193 
  2194 * a generic induction tactic `induct_tac' which works for all datatypes and
  2195 also for type `nat';
  2196 
  2197 * a generic case distinction tactic `exhaust_tac' which works for all
  2198 datatypes and also for type `nat';
  2199 
  2200 * each datatype comes with a function `size';
  2201 
  2202 * patterns in case expressions allow tuple patterns as arguments to
  2203 constructors, for example `case x of [] => ... | (x,y,z)#ps => ...';
  2204 
  2205 * primrec now also works with type nat;
  2206 
  2207 * recdef: a new declaration form, allows general recursive functions to be
  2208 defined in theory files.  See HOL/ex/Fib, HOL/ex/Primes, HOL/Subst/Unify.
  2209 
  2210 * the constant for negation has been renamed from "not" to "Not" to
  2211 harmonize with FOL, ZF, LK, etc.;
  2212 
  2213 * HOL/ex/LFilter theory of a corecursive "filter" functional for
  2214 infinite lists;
  2215 
  2216 * HOL/Modelcheck demonstrates invocation of model checker oracle;
  2217 
  2218 * HOL/ex/Ring.thy declares cring_simp, which solves equational
  2219 problems in commutative rings, using axiomatic type classes for + and *;
  2220 
  2221 * more examples in HOL/MiniML and HOL/Auth;
  2222 
  2223 * more default rewrite rules for quantifiers, union/intersection;
  2224 
  2225 * a new constant `arbitrary == @x.False';
  2226 
  2227 * HOLCF/IOA replaces old HOL/IOA;
  2228 
  2229 * HOLCF changes: derived all rules and arities
  2230   + axiomatic type classes instead of classes
  2231   + typedef instead of faking type definitions
  2232   + eliminated the internal constants less_fun, less_cfun, UU_fun, UU_cfun etc.
  2233   + new axclasses cpo, chfin, flat with flat < chfin < pcpo < cpo < po
  2234   + eliminated the types void, one, tr
  2235   + use unit lift and bool lift (with translations) instead of one and tr
  2236   + eliminated blift from Lift3.thy (use Def instead of blift)
  2237   all eliminated rules are derived as theorems --> no visible changes ;
  2238 
  2239 
  2240 *** ZF ***
  2241 
  2242 * ZF now has Fast_tac, Simp_tac and Auto_tac.  Union_iff is a now a default
  2243 rewrite rule; this may affect some proofs.  eq_cs is gone but can be put back
  2244 as ZF_cs addSIs [equalityI];
  2245 
  2246 
  2247 
  2248 New in Isabelle94-7 (November 96)
  2249 ---------------------------------
  2250 
  2251 * allowing negative levels (as offsets) in prlev and choplev;
  2252 
  2253 * super-linear speedup for large simplifications;
  2254 
  2255 * FOL, ZF and HOL now use miniscoping: rewriting pushes
  2256 quantifications in as far as possible (COULD MAKE EXISTING PROOFS
  2257 FAIL); can suppress it using the command Delsimps (ex_simps @
  2258 all_simps); De Morgan laws are also now included, by default;
  2259 
  2260 * improved printing of ==>  :  ~:
  2261 
  2262 * new object-logic "Sequents" adds linear logic, while replacing LK
  2263 and Modal (thanks to Sara Kalvala);
  2264 
  2265 * HOL/Auth: correctness proofs for authentication protocols;
  2266 
  2267 * HOL: new auto_tac combines rewriting and classical reasoning (many
  2268 examples on HOL/Auth);
  2269 
  2270 * HOL: new command AddIffs for declaring theorems of the form P=Q to
  2271 the rewriter and classical reasoner simultaneously;
  2272 
  2273 * function uresult no longer returns theorems in "standard" format;
  2274 regain previous version by: val uresult = standard o uresult;
  2275 
  2276 
  2277 
  2278 New in Isabelle94-6
  2279 -------------------
  2280 
  2281 * oracles -- these establish an interface between Isabelle and trusted
  2282 external reasoners, which may deliver results as theorems;
  2283 
  2284 * proof objects (in particular record all uses of oracles);
  2285 
  2286 * Simp_tac, Fast_tac, etc. that refer to implicit simpset / claset;
  2287 
  2288 * "constdefs" section in theory files;
  2289 
  2290 * "primrec" section (HOL) no longer requires names;
  2291 
  2292 * internal type "tactic" now simply "thm -> thm Sequence.seq";
  2293 
  2294 
  2295 
  2296 New in Isabelle94-5
  2297 -------------------
  2298 
  2299 * reduced space requirements;
  2300 
  2301 * automatic HTML generation from theories;
  2302 
  2303 * theory files no longer require "..." (quotes) around most types;
  2304 
  2305 * new examples, including two proofs of the Church-Rosser theorem;
  2306 
  2307 * non-curried (1994) version of HOL is no longer distributed;
  2308 
  2309 
  2310 
  2311 New in Isabelle94-4
  2312 -------------------
  2313 
  2314 * greatly reduced space requirements;
  2315 
  2316 * theory files (.thy) no longer require \...\ escapes at line breaks;
  2317 
  2318 * searchable theorem database (see the section "Retrieving theorems" on
  2319 page 8 of the Reference Manual);
  2320 
  2321 * new examples, including Grabczewski's monumental case study of the
  2322 Axiom of Choice;
  2323 
  2324 * The previous version of HOL renamed to Old_HOL;
  2325 
  2326 * The new version of HOL (previously called CHOL) uses a curried syntax
  2327 for functions.  Application looks like f a b instead of f(a,b);
  2328 
  2329 * Mutually recursive inductive definitions finally work in HOL;
  2330 
  2331 * In ZF, pattern-matching on tuples is now available in all abstractions and
  2332 translates to the operator "split";
  2333 
  2334 
  2335 
  2336 New in Isabelle94-3
  2337 -------------------
  2338 
  2339 * new infix operator, addss, allowing the classical reasoner to
  2340 perform simplification at each step of its search.  Example:
  2341         fast_tac (cs addss ss)
  2342 
  2343 * a new logic, CHOL, the same as HOL, but with a curried syntax
  2344 for functions.  Application looks like f a b instead of f(a,b).  Also pairs
  2345 look like (a,b) instead of <a,b>;
  2346 
  2347 * PLEASE NOTE: CHOL will eventually replace HOL!
  2348 
  2349 * In CHOL, pattern-matching on tuples is now available in all abstractions.
  2350 It translates to the operator "split".  A new theory of integers is available;
  2351 
  2352 * In ZF, integer numerals now denote two's-complement binary integers.
  2353 Arithmetic operations can be performed by rewriting.  See ZF/ex/Bin.ML;
  2354 
  2355 * Many new examples: I/O automata, Church-Rosser theorem, equivalents
  2356 of the Axiom of Choice;
  2357 
  2358 
  2359 
  2360 New in Isabelle94-2
  2361 -------------------
  2362 
  2363 * Significantly faster resolution;
  2364 
  2365 * the different sections in a .thy file can now be mixed and repeated
  2366 freely;
  2367 
  2368 * Database of theorems for FOL, HOL and ZF.  New
  2369 commands including qed, qed_goal and bind_thm store theorems in the database.
  2370 
  2371 * Simple database queries: return a named theorem (get_thm) or all theorems of
  2372 a given theory (thms_of), or find out what theory a theorem was proved in
  2373 (theory_of_thm);
  2374 
  2375 * Bugs fixed in the inductive definition and datatype packages;
  2376 
  2377 * The classical reasoner provides deepen_tac and depth_tac, making FOL_dup_cs
  2378 and HOL_dup_cs obsolete;
  2379 
  2380 * Syntactic ambiguities caused by the new treatment of syntax in Isabelle94-1
  2381 have been removed;
  2382 
  2383 * Simpler definition of function space in ZF;
  2384 
  2385 * new results about cardinal and ordinal arithmetic in ZF;
  2386 
  2387 * 'subtype' facility in HOL for introducing new types as subsets of existing
  2388 types;
  2389 
  2390 
  2391 $Id$