NEWS
author wenzelm
Tue Dec 18 16:14:50 2001 +0100 (2001-12-18)
changeset 12538 150af0a4bb11
parent 12489 c92e38c3cbaa
child 12563 3813bcab493d
permissions -rw-r--r--
* system: tested support for MacOS X;
     1 
     2 Isabelle NEWS -- history user-relevant changes
     3 ==============================================
     4 
     5 New in Isabelle2001 (December 2001)
     6 -----------------------------------
     7 
     8 *** Document preparation ***
     9 
    10 * greatly simplified document preparation setup, including more
    11 graceful interpretation of isatool usedir -i/-d/-D options, and more
    12 instructive isatool mkdir; users should basically be able to get
    13 started with "isatool mkdir Test && isatool make";
    14 
    15 * theory dependency graph may now be incorporated into documents;
    16 isatool usedir -g true will produce session_graph.eps/.pdf for use
    17 with \includegraphics of LaTeX;
    18 
    19 * proper spacing of consecutive markup elements, especially text
    20 blocks after section headings;
    21 
    22 * support bold style (for single symbols only), input syntax is like
    23 this: "\<^bold>\<alpha>" or "\<^bold>A";
    24 
    25 * \<bullet> is now output as bold \cdot by default, which looks much
    26 better in printed text;
    27 
    28 * added default LaTeX bindings for \<tturnstile> and \<TTurnstile>;
    29 note that these symbols are currently unavailable in Proof General /
    30 X-Symbol;
    31 
    32 
    33 *** Isar ***
    34 
    35 * Pure/Provers: improved proof by cases and induction;
    36   - 'case' command admits impromptu naming of parameters (such as
    37     "case (Suc n)");
    38   - 'induct' method divinates rule instantiation from the inductive
    39     claim; no longer requires excessive ?P bindings for proper
    40     instantiation of cases;
    41   - 'induct' method properly enumerates all possibilities of set/type
    42     rules; as a consequence facts may be also passed through *type*
    43     rules without further ado;
    44   - 'induct' method now derives symbolic cases from the *rulified*
    45     rule (before it used to rulify cases stemming from the internal
    46     atomized version); this means that the context of a non-atomic
    47     statement becomes is included in the hypothesis, avoiding the
    48     slightly cumbersome show "PROP ?case" form;
    49   - 'induct' may now use elim-style induction rules without chaining
    50     facts, using ``missing'' premises from the goal state; this allows
    51     rules stemming from inductive sets to be applied in unstructured
    52     scripts, while still benefitting from proper handling of non-atomic
    53     statements; NB: major inductive premises need to be put first, all
    54     the rest of the goal is passed through the induction;
    55   - 'induct' proper support for mutual induction involving non-atomic
    56     rule statements (uses the new concept of simultaneous goals, see
    57     below);
    58   - removed obsolete "(simplified)" and "(stripped)" options of methods;
    59   - added 'print_induct_rules' (covered by help item in Proof General > 3.3);
    60   - moved induct/cases attributes to Pure, methods to Provers;
    61   - generic method setup instantiated for FOL and HOL;
    62 
    63 * Pure: support multiple simultaneous goal statements, for example
    64 "have a: A and b: B" (same for 'theorem' etc.); being a pure
    65 meta-level mechanism, this acts as if several individual goals had
    66 been stated separately; in particular common proof methods need to be
    67 repeated in order to cover all claims; note that a single elimination
    68 step is *not* sufficient to establish the two conjunctions, so this
    69 fails:
    70 
    71   assume "A & B" then have A and B ..   (*".." fails*)
    72 
    73 better use "obtain" in situations as above; alternative refer to
    74 multi-step methods like 'auto', 'simp_all', 'blast+' etc.;
    75 
    76 * Pure: proper integration with ``locales''; unlike the original
    77 version by Florian Kammüller, Isar locales package high-level proof
    78 contexts rather than raw logical ones (e.g. we admit to include
    79 attributes everywhere); operations on locales include merge and
    80 rename; e.g. see HOL/ex/Locales.thy;
    81 
    82 * Pure: theory goals now support ad-hoc contexts, which are discharged
    83 in the result, as in ``lemma (assumes A and B) K: A .''; syntax
    84 coincides with that of a locale body;
    85 
    86 * Pure: renamed "antecedent" case to "rule_context";
    87 
    88 * Pure: added 'corollary' command;
    89 
    90 * Pure: fixed 'token_translation' command;
    91 
    92 * Pure: removed obsolete 'exported' attribute;
    93 
    94 * Pure: dummy pattern "_" in is/let is now automatically lifted over
    95 bound variables: "ALL x. P x --> Q x" (is "ALL x. _ --> ?C x")
    96 supersedes more cumbersome ... (is "ALL x. _ x --> ?C x");
    97 
    98 * Pure: method 'atomize' presents local goal premises as object-level
    99 statements (atomic meta-level propositions); setup controlled via
   100 rewrite rules declarations of 'atomize' attribute; example
   101 application: 'induct' method with proper rule statements in improper
   102 proof *scripts*;
   103 
   104 * Pure: emulation of instantiation tactics (rule_tac, cut_tac, etc.)
   105 now consider the syntactic context of assumptions, giving a better
   106 chance to get type-inference of the arguments right (this is
   107 especially important for locales);
   108 
   109 * Pure: "sorry" no longer requires quick_and_dirty in interactive
   110 mode;
   111 
   112 * Pure/obtain: the formal conclusion "thesis", being marked as
   113 ``internal'', may no longer be reference directly in the text;
   114 potential INCOMPATIBILITY, may need to use "?thesis" in rare
   115 situations;
   116 
   117 * Pure: generic 'sym' attribute which declares a rule both as pure
   118 'elim?' and for the 'symmetric' operation;
   119 
   120 * Pure/Provers/classical: simplified integration with pure rule
   121 attributes and methods; the classical "intro?/elim?/dest?"
   122 declarations coincide with the pure ones; the "rule" method no longer
   123 includes classically swapped intros; "intro" and "elim" methods no
   124 longer pick rules from the context; also got rid of ML declarations
   125 AddXIs/AddXEs/AddXDs; all of this has some potential for
   126 INCOMPATIBILITY;
   127 
   128 * Provers/classical: attribute 'swapped' produces classical inversions
   129 of introduction rules;
   130 
   131 * Provers/simplifier: 'simplified' attribute may refer to explicit
   132 rules instead of full simplifier context; 'iff' attribute handles
   133 conditional rules;
   134 
   135 * HOL: 'typedef' now allows alternative names for Rep/Abs morphisms;
   136 
   137 * HOL: 'recdef' now fails on unfinished automated proofs, use
   138 "(permissive)" option to recover old behavior;
   139 
   140 * HOL: 'inductive' no longer features separate (collective) attributes
   141 for 'intros' (was found too confusing);
   142 
   143 * HOL: properly declared induction rules less_induct and
   144 wf_induct_rule;
   145 
   146 * HOLCF: domain package adapted to new-style theories, e.g. see
   147 HOLCF/ex/Dnat.thy;
   148 
   149 * ZF: proper integration of logic-specific tools and packages,
   150 including theory commands '(co)inductive', '(co)datatype',
   151 'rep_datatype', 'inductive_cases', as well as methods 'ind_cases',
   152 'induct_tac', 'case_tac', and 'typecheck' (with attribute 'TC');
   153 
   154 
   155 *** HOL ***
   156 
   157 * HOL: moved over to sane numeral syntax; the new policy is as
   158 follows:
   159 
   160   - 0 and 1 are polymorphic constants, which are defined on any
   161   numeric type (nat, int, real etc.);
   162 
   163   - 2, 3, 4, ... and -1, -2, -3, ... are polymorphic numerals, based
   164   binary representation internally;
   165 
   166   - type nat has special constructor Suc, and generally prefers Suc 0
   167   over 1::nat and Suc (Suc 0) over 2::nat;
   168 
   169 This change may cause significant problems of INCOMPATIBILITY; here
   170 are some hints on converting existing sources:
   171 
   172   - due to the new "num" token, "-0" and "-1" etc. are now atomic
   173   entities, so expressions involving "-" (unary or binary minus) need
   174   to be spaced properly;
   175 
   176   - existing occurrences of "1" may need to be constraint "1::nat" or
   177   even replaced by Suc 0; similar for old "2";
   178 
   179   - replace "#nnn" by "nnn", and "#-nnn" by "-nnn";
   180 
   181   - remove all special provisions on numerals in proofs;
   182 
   183 * HOL: the class of all HOL types is now called "type" rather than
   184 "term"; INCOMPATIBILITY, need to adapt references to this type class
   185 in axclass/classes, instance/arities, and (usually rare) occurrences
   186 in typings (of consts etc.); internally the class is called
   187 "HOL.type", ML programs should refer to HOLogic.typeS;
   188 
   189 * HOL/record package improvements:
   190   - new derived operations "fields" to build a partial record section,
   191     "extend" to promote a fixed record to a record scheme, and
   192     "truncate" for the reverse; cf. theorems "xxx.defs", which are *not*
   193     declared as simp by default;
   194   - removed "make_scheme" operations (use "make" with "extend") --
   195     INCOMPATIBILITY;
   196   - removed "more" class (simply use "term") -- INCOMPATIBILITY;
   197   - provides cases/induct rules for use with corresponding Isar
   198     methods (for concrete records, record schemes, concrete more
   199     parts, and schematic more parts -- in that order);
   200   - internal definitions directly based on a light-weight abstract
   201     theory of product types over typedef rather than datatype;
   202 
   203 * HOL: canonical cases/induct rules for n-tuples (n = 3..7);
   204 
   205 * HOL: concrete setsum syntax "\<Sum>i:A. b" == "setsum (%i. b) A"
   206 (beware of argument permutation!);
   207 
   208 * HOL: linorder_less_split superseded by linorder_cases;
   209 
   210 * HOL: added "The" definite description operator; move Hilbert's "Eps"
   211 to peripheral theory "Hilbert_Choice";
   212 
   213 * HOL: made split_all_tac safe; EXISTING PROOFS MAY FAIL OR LOOP, so
   214 in this (rare) case use:
   215 
   216   delSWrapper "split_all_tac"
   217   addSbefore ("unsafe_split_all_tac", unsafe_split_all_tac)
   218 
   219 * HOL: added safe wrapper "split_conv_tac" to claset; EXISTING PROOFS
   220 MAY FAIL;
   221 
   222 * HOL: introduced f^n = f o ... o f; warning: due to the limits of
   223 Isabelle's type classes, ^ on functions and relations has too general
   224 a domain, namely ('a * 'b) set and 'a => 'b; this means that it may be
   225 necessary to attach explicit type constraints;
   226 
   227 * HOL/Relation: the prefix name of the infix "O" has been changed from "comp"
   228 to "rel_comp"; INCOMPATIBILITY: a few theorems have been renamed accordingly
   229 (eg "compI" -> "rel_compI").
   230 
   231 * HOL: syntax translations now work properly with numerals and records
   232 expressions;
   233 
   234 * HOL: bounded abstraction now uses syntax "%" / "\<lambda>" instead
   235 of "lam" -- INCOMPATIBILITY;
   236 
   237 * HOL: got rid of some global declarations (potential INCOMPATIBILITY
   238 for ML tools): const "()" renamed "Product_Type.Unity", type "unit"
   239 renamed "Product_Type.unit";
   240 
   241 * HOL/GroupTheory: group theory examples including Sylow's theorem, by
   242 Florian Kammüller;
   243 
   244 
   245 *** HOLCF ***
   246 
   247 * proper rep_datatype lift (see theory Lift) instead of ML hacks --
   248 potential INCOMPATIBILITY; now use plain induct_tac instead of former
   249 lift.induct_tac, always use UU instead of Undef;
   250 
   251 
   252 *** ZF ***
   253 
   254 * ZF/UNITY: Chandy and Misra's UNITY is now available in ZF, giving a
   255 typeless version of the formalism;
   256 
   257 * ZF/Induct: new directory for examples of inductive definitions,
   258 including theory Multiset for multiset orderings;
   259 
   260 * ZF: the integer library now covers quotients and remainders, with
   261 many laws relating division to addition, multiplication, etc.;
   262 
   263 
   264 *** General ***
   265 
   266 * Pure/kernel: meta-level proof terms (by Stefan Berghofer); reference
   267 variable proof controls level of detail: 0 = no proofs (only oracle
   268 dependencies), 1 = lemma dependencies, 2 = compact proof terms; see
   269 also ref manual for further ML interfaces;
   270 
   271 * Pure/axclass: removed obsolete ML interface
   272 goal_subclass/goal_arity;
   273 
   274 * Pure/syntax: new token syntax "num" for plain numerals (without "#"
   275 of "xnum"); potential INCOMPATIBILITY, since -0, -1 etc. are now
   276 separate tokens, so expressions involving minus need to be spaced
   277 properly;
   278 
   279 * Pure/syntax: support non-oriented infixes, using keyword "infix"
   280 rather than "infixl" or "infixr";
   281 
   282 * Pure/syntax: concrete syntax for dummy type variables admits genuine
   283 sort constraint specifications in type inference; e.g. "x::_::foo"
   284 ensures that the type of "x" is of sort "foo" (but not necessarily a
   285 type variable);
   286 
   287 * Pure/syntax: print modes "type_brackets" and "no_type_brackets"
   288 control output of nested => (types); the default behavior is
   289 "type_brackets";
   290 
   291 * Pure/syntax: builtin parse translation for "_constify" turns valued
   292 tokens into AST constants;
   293 
   294 * Pure/syntax: prefer later declarations of translations and print
   295 translation functions; potential INCOMPATIBILITY: need to reverse
   296 multiple declarations for same syntax element constant;
   297 
   298 * Provers/classical: renamed addaltern to addafter, addSaltern to
   299 addSafter;
   300 
   301 * Provers/clasimp: ``iff'' declarations now handle conditional rules
   302 as well;
   303 
   304 * system: tested support for MacOS X; should be able to get Isabelle +
   305 Proof General to work in a plain Terminal after installing Poly/ML
   306 (e.g. from the Isabelle distribution area) and GNU bash alone
   307 (e.g. from http://www.apple.com); full X11, XEmacs and X-Symbol
   308 support requires further installations, e.g. from
   309 http://fink.sourceforge.net/);
   310 
   311 * system: support Poly/ML 4.1.1 (able to manage larger heaps);
   312 
   313 * system: Proof General keywords specification is now part of the
   314 Isabelle distribution (see etc/isar-keywords.el);
   315 
   316 * system: smart selection of Isabelle process versus Isabelle
   317 interface, accommodates case-insensitive file systems (e.g. HFS+); may
   318 run both "isabelle" and "Isabelle" even if file names are badly
   319 damaged (executable inspects the case of the first letter of its own
   320 name); added separate "isabelle-process" and "isabelle-interface";
   321 
   322 * system: refrain from any attempt at filtering input streams; no
   323 longer support ``8bit'' encoding of old isabelle font, instead proper
   324 iso-latin characters may now be used; the related isatools
   325 "symbolinput" and "nonascii" have disappeared as well;
   326 
   327 * system: removed old "xterm" interface (the print modes "xterm" and
   328 "xterm_color" are still available for direct use in a suitable
   329 terminal);
   330 
   331 
   332 
   333 New in Isabelle99-2 (February 2001)
   334 -----------------------------------
   335 
   336 *** Overview of INCOMPATIBILITIES ***
   337 
   338 * HOL: please note that theories in the Library and elsewhere often use the
   339 new-style (Isar) format; to refer to their theorems in an ML script you must
   340 bind them to ML identifers by e.g.	val thm_name = thm "thm_name";
   341 
   342 * HOL: inductive package no longer splits induction rule aggressively,
   343 but only as far as specified by the introductions given; the old
   344 format may be recovered via ML function complete_split_rule or attribute
   345 'split_rule (complete)';
   346 
   347 * HOL: induct renamed to lfp_induct, lfp_Tarski to lfp_unfold,
   348 gfp_Tarski to gfp_unfold;
   349 
   350 * HOL: contrapos, contrapos2 renamed to contrapos_nn, contrapos_pp;
   351 
   352 * HOL: infix "dvd" now has priority 50 rather than 70 (because it is a
   353 relation); infix "^^" has been renamed "``"; infix "``" has been
   354 renamed "`"; "univalent" has been renamed "single_valued";
   355 
   356 * HOL/Real: "rinv" and "hrinv" replaced by overloaded "inverse"
   357 operation;
   358 
   359 * HOLCF: infix "`" has been renamed "$"; the symbol syntax is \<cdot>;
   360 
   361 * Isar: 'obtain' no longer declares "that" fact as simp/intro;
   362 
   363 * Isar/HOL: method 'induct' now handles non-atomic goals; as a
   364 consequence, it is no longer monotonic wrt. the local goal context
   365 (which is now passed through the inductive cases);
   366 
   367 * Document preparation: renamed standard symbols \<ll> to \<lless> and
   368 \<gg> to \<ggreater>;
   369 
   370 
   371 *** Document preparation ***
   372 
   373 * \isabellestyle{NAME} selects version of Isabelle output (currently
   374 available: are "it" for near math-mode best-style output, "sl" for
   375 slanted text style, and "tt" for plain type-writer; if no
   376 \isabellestyle command is given, output is according to slanted
   377 type-writer);
   378 
   379 * support sub/super scripts (for single symbols only), input syntax is
   380 like this: "A\<^sup>*" or "A\<^sup>\<star>";
   381 
   382 * some more standard symbols; see Appendix A of the system manual for
   383 the complete list of symbols defined in isabellesym.sty;
   384 
   385 * improved isabelle style files; more abstract symbol implementation
   386 (should now use \isamath{...} and \isatext{...} in custom symbol
   387 definitions);
   388 
   389 * antiquotation @{goals} and @{subgoals} for output of *dynamic* goals
   390 state; Note that presentation of goal states does not conform to
   391 actual human-readable proof documents.  Please do not include goal
   392 states into document output unless you really know what you are doing!
   393 
   394 * proper indentation of antiquoted output with proportional LaTeX
   395 fonts;
   396 
   397 * no_document ML operator temporarily disables LaTeX document
   398 generation;
   399 
   400 * isatool unsymbolize tunes sources for plain ASCII communication;
   401 
   402 
   403 *** Isar ***
   404 
   405 * Pure: Isar now suffers initial goal statements to contain unbound
   406 schematic variables (this does not conform to actual readable proof
   407 documents, due to unpredictable outcome and non-compositional proof
   408 checking); users who know what they are doing may use schematic goals
   409 for Prolog-style synthesis of proven results;
   410 
   411 * Pure: assumption method (an implicit finishing) now handles actual
   412 rules as well;
   413 
   414 * Pure: improved 'obtain' --- moved to Pure, insert "that" into
   415 initial goal, declare "that" only as Pure intro (only for single
   416 steps); the "that" rule assumption may now be involved in implicit
   417 finishing, thus ".." becomes a feasible for trivial obtains;
   418 
   419 * Pure: default proof step now includes 'intro_classes'; thus trivial
   420 instance proofs may be performed by "..";
   421 
   422 * Pure: ?thesis / ?this / "..." now work for pure meta-level
   423 statements as well;
   424 
   425 * Pure: more robust selection of calculational rules;
   426 
   427 * Pure: the builtin notion of 'finished' goal now includes the ==-refl
   428 rule (as well as the assumption rule);
   429 
   430 * Pure: 'thm_deps' command visualizes dependencies of theorems and
   431 lemmas, using the graph browser tool;
   432 
   433 * Pure: predict failure of "show" in interactive mode;
   434 
   435 * Pure: 'thms_containing' now takes actual terms as arguments;
   436 
   437 * HOL: improved method 'induct' --- now handles non-atomic goals
   438 (potential INCOMPATIBILITY); tuned error handling;
   439 
   440 * HOL: cases and induct rules now provide explicit hints about the
   441 number of facts to be consumed (0 for "type" and 1 for "set" rules);
   442 any remaining facts are inserted into the goal verbatim;
   443 
   444 * HOL: local contexts (aka cases) may now contain term bindings as
   445 well; the 'cases' and 'induct' methods new provide a ?case binding for
   446 the result to be shown in each case;
   447 
   448 * HOL: added 'recdef_tc' command;
   449 
   450 * isatool convert assists in eliminating legacy ML scripts;
   451 
   452 
   453 *** HOL ***
   454 
   455 * HOL/Library: a collection of generic theories to be used together
   456 with main HOL; the theory loader path already includes this directory
   457 by default; the following existing theories have been moved here:
   458 HOL/Induct/Multiset, HOL/Induct/Acc (as Accessible_Part), HOL/While
   459 (as While_Combinator), HOL/Lex/Prefix (as List_Prefix);
   460 
   461 * HOL/Unix: "Some aspects of Unix file-system security", a typical
   462 modelling and verification task performed in Isabelle/HOL +
   463 Isabelle/Isar + Isabelle document preparation (by Markus Wenzel).
   464 
   465 * HOL/Algebra: special summation operator SUM no longer exists, it has
   466 been replaced by setsum; infix 'assoc' now has priority 50 (like
   467 'dvd'); axiom 'one_not_zero' has been moved from axclass 'ring' to
   468 'domain', this makes the theory consistent with mathematical
   469 literature;
   470 
   471 * HOL basics: added overloaded operations "inverse" and "divide"
   472 (infix "/"), syntax for generic "abs" operation, generic summation
   473 operator \<Sum>;
   474 
   475 * HOL/typedef: simplified package, provide more useful rules (see also
   476 HOL/subset.thy);
   477 
   478 * HOL/datatype: induction rule for arbitrarily branching datatypes is
   479 now expressed as a proper nested rule (old-style tactic scripts may
   480 require atomize_strip_tac to cope with non-atomic premises);
   481 
   482 * HOL: renamed theory "Prod" to "Product_Type", renamed "split" rule
   483 to "split_conv" (old name still available for compatibility);
   484 
   485 * HOL: improved concrete syntax for strings (e.g. allows translation
   486 rules with string literals);
   487 
   488 * HOL-Real-Hyperreal: this extends HOL-Real with the hyperreals
   489  and Fleuriot's mechanization of analysis, including the transcendental
   490  functions for the reals;
   491 
   492 * HOL/Real, HOL/Hyperreal: improved arithmetic simplification;
   493 
   494 
   495 *** CTT ***
   496 
   497 * CTT: x-symbol support for Pi, Sigma, -->, : (membership); note that
   498 "lam" is displayed as TWO lambda-symbols
   499 
   500 * CTT: theory Main now available, containing everything (that is, Bool
   501 and Arith);
   502 
   503 
   504 *** General ***
   505 
   506 * Pure: the Simplifier has been implemented properly as a derived rule
   507 outside of the actual kernel (at last!); the overall performance
   508 penalty in practical applications is about 50%, while reliability of
   509 the Isabelle inference kernel has been greatly improved;
   510 
   511 * print modes "brackets" and "no_brackets" control output of nested =>
   512 (types) and ==> (props); the default behaviour is "brackets";
   513 
   514 * Provers: fast_tac (and friends) now handle actual object-logic rules
   515 as assumptions as well;
   516 
   517 * system: support Poly/ML 4.0;
   518 
   519 * system: isatool install handles KDE version 1 or 2;
   520 
   521 
   522 
   523 New in Isabelle99-1 (October 2000)
   524 ----------------------------------
   525 
   526 *** Overview of INCOMPATIBILITIES ***
   527 
   528 * HOL: simplification of natural numbers is much changed; to partly
   529 recover the old behaviour (e.g. to prevent n+n rewriting to #2*n)
   530 issue the following ML commands:
   531 
   532   Delsimprocs Nat_Numeral_Simprocs.cancel_numerals;
   533   Delsimprocs [Nat_Numeral_Simprocs.combine_numerals];
   534 
   535 * HOL: simplification no longer dives into case-expressions; this is
   536 controlled by "t.weak_case_cong" for each datatype t;
   537 
   538 * HOL: nat_less_induct renamed to less_induct;
   539 
   540 * HOL: systematic renaming of the SOME (Eps) rules, may use isatool
   541 fixsome to patch .thy and .ML sources automatically;
   542 
   543   select_equality  -> some_equality
   544   select_eq_Ex     -> some_eq_ex
   545   selectI2EX       -> someI2_ex
   546   selectI2         -> someI2
   547   selectI          -> someI
   548   select1_equality -> some1_equality
   549   Eps_sym_eq       -> some_sym_eq_trivial
   550   Eps_eq           -> some_eq_trivial
   551 
   552 * HOL: exhaust_tac on datatypes superceded by new generic case_tac;
   553 
   554 * HOL: removed obsolete theorem binding expand_if (refer to split_if
   555 instead);
   556 
   557 * HOL: the recursion equations generated by 'recdef' are now called
   558 f.simps instead of f.rules;
   559 
   560 * HOL: qed_spec_mp now also handles bounded ALL as well;
   561 
   562 * HOL: 0 is now overloaded, so the type constraint ":: nat" may
   563 sometimes be needed;
   564 
   565 * HOL: the constant for "f``x" is now "image" rather than "op ``";
   566 
   567 * HOL: the constant for "f-``x" is now "vimage" rather than "op -``";
   568 
   569 * HOL: the disjoint sum is now "<+>" instead of "Plus"; the cartesian
   570 product is now "<*>" instead of "Times"; the lexicographic product is
   571 now "<*lex*>" instead of "**";
   572 
   573 * HOL: theory Sexp is now in HOL/Induct examples (it used to be part
   574 of main HOL, but was unused); better use HOL's datatype package;
   575 
   576 * HOL: removed "symbols" syntax for constant "override" of theory Map;
   577 the old syntax may be recovered as follows:
   578 
   579   syntax (symbols)
   580     override  :: "('a ~=> 'b) => ('a ~=> 'b) => ('a ~=> 'b)"
   581       (infixl "\\<oplus>" 100)
   582 
   583 * HOL/Real: "rabs" replaced by overloaded "abs" function;
   584 
   585 * HOL/ML: even fewer consts are declared as global (see theories Ord,
   586 Lfp, Gfp, WF); this only affects ML packages that refer to const names
   587 internally;
   588 
   589 * HOL and ZF: syntax for quotienting wrt an equivalence relation
   590 changed from A/r to A//r;
   591 
   592 * ZF: new treatment of arithmetic (nat & int) may break some old
   593 proofs;
   594 
   595 * Isar: renamed some attributes (RS -> THEN, simplify -> simplified,
   596 rulify -> rule_format, elimify -> elim_format, ...);
   597 
   598 * Isar/Provers: intro/elim/dest attributes changed; renamed
   599 intro/intro!/intro!! flags to intro!/intro/intro? (in most cases, one
   600 should have to change intro!! to intro? only); replaced "delrule" by
   601 "rule del";
   602 
   603 * Isar/HOL: renamed "intrs" to "intros" in inductive definitions;
   604 
   605 * Provers: strengthened force_tac by using new first_best_tac;
   606 
   607 * LaTeX document preparation: several changes of isabelle.sty (see
   608 lib/texinputs);
   609 
   610 
   611 *** Document preparation ***
   612 
   613 * formal comments (text blocks etc.) in new-style theories may now
   614 contain antiquotations of thm/prop/term/typ/text to be presented
   615 according to latex print mode; concrete syntax is like this:
   616 @{term[show_types] "f(x) = a + x"};
   617 
   618 * isatool mkdir provides easy setup of Isabelle session directories,
   619 including proper document sources;
   620 
   621 * generated LaTeX sources are now deleted after successful run
   622 (isatool document -c); may retain a copy somewhere else via -D option
   623 of isatool usedir;
   624 
   625 * isatool usedir -D now lets isatool latex -o sty update the Isabelle
   626 style files, achieving self-contained LaTeX sources and simplifying
   627 LaTeX debugging;
   628 
   629 * old-style theories now produce (crude) LaTeX output as well;
   630 
   631 * browser info session directories are now self-contained (may be put
   632 on WWW server seperately); improved graphs of nested sessions; removed
   633 graph for 'all sessions';
   634 
   635 * several improvements in isabelle style files; \isabellestyle{it}
   636 produces fake math mode output; \isamarkupheader is now \section by
   637 default; see lib/texinputs/isabelle.sty etc.;
   638 
   639 
   640 *** Isar ***
   641 
   642 * Isar/Pure: local results and corresponding term bindings are now
   643 subject to Hindley-Milner polymorphism (similar to ML); this
   644 accommodates incremental type-inference very nicely;
   645 
   646 * Isar/Pure: new derived language element 'obtain' supports
   647 generalized existence reasoning;
   648 
   649 * Isar/Pure: new calculational elements 'moreover' and 'ultimately'
   650 support accumulation of results, without applying any rules yet;
   651 useful to collect intermediate results without explicit name
   652 references, and for use with transitivity rules with more than 2
   653 premises;
   654 
   655 * Isar/Pure: scalable support for case-analysis type proofs: new
   656 'case' language element refers to local contexts symbolically, as
   657 produced by certain proof methods; internally, case names are attached
   658 to theorems as "tags";
   659 
   660 * Isar/Pure: theory command 'hide' removes declarations from
   661 class/type/const name spaces;
   662 
   663 * Isar/Pure: theory command 'defs' supports option "(overloaded)" to
   664 indicate potential overloading;
   665 
   666 * Isar/Pure: changed syntax of local blocks from {{ }} to { };
   667 
   668 * Isar/Pure: syntax of sorts made 'inner', i.e. have to write
   669 "{a,b,c}" instead of {a,b,c};
   670 
   671 * Isar/Pure now provides its own version of intro/elim/dest
   672 attributes; useful for building new logics, but beware of confusion
   673 with the version in Provers/classical;
   674 
   675 * Isar/Pure: the local context of (non-atomic) goals is provided via
   676 case name 'antecedent';
   677 
   678 * Isar/Pure: removed obsolete 'transfer' attribute (transfer of thms
   679 to the current context is now done automatically);
   680 
   681 * Isar/Pure: theory command 'method_setup' provides a simple interface
   682 for definining proof methods in ML;
   683 
   684 * Isar/Provers: intro/elim/dest attributes changed; renamed
   685 intro/intro!/intro!! flags to intro!/intro/intro? (INCOMPATIBILITY, in
   686 most cases, one should have to change intro!! to intro? only);
   687 replaced "delrule" by "rule del";
   688 
   689 * Isar/Provers: new 'hypsubst' method, plain 'subst' method and
   690 'symmetric' attribute (the latter supercedes [RS sym]);
   691 
   692 * Isar/Provers: splitter support (via 'split' attribute and 'simp'
   693 method modifier); 'simp' method: 'only:' modifier removes loopers as
   694 well (including splits);
   695 
   696 * Isar/Provers: Simplifier and Classical methods now support all kind
   697 of modifiers used in the past, including 'cong', 'iff', etc.
   698 
   699 * Isar/Provers: added 'fastsimp' and 'clarsimp' methods (combination
   700 of Simplifier and Classical reasoner);
   701 
   702 * Isar/HOL: new proof method 'cases' and improved version of 'induct'
   703 now support named cases; major packages (inductive, datatype, primrec,
   704 recdef) support case names and properly name parameters;
   705 
   706 * Isar/HOL: new transitivity rules for substitution in inequalities --
   707 monotonicity conditions are extracted to be proven at end of
   708 calculations;
   709 
   710 * Isar/HOL: removed 'case_split' thm binding, should use 'cases' proof
   711 method anyway;
   712 
   713 * Isar/HOL: removed old expand_if = split_if; theorems if_splits =
   714 split_if split_if_asm; datatype package provides theorems foo.splits =
   715 foo.split foo.split_asm for each datatype;
   716 
   717 * Isar/HOL: tuned inductive package, rename "intrs" to "intros"
   718 (potential INCOMPATIBILITY), emulation of mk_cases feature for proof
   719 scripts: new 'inductive_cases' command and 'ind_cases' method; (Note:
   720 use "(cases (simplified))" method in proper proof texts);
   721 
   722 * Isar/HOL: added global 'arith_split' attribute for 'arith' method;
   723 
   724 * Isar: names of theorems etc. may be natural numbers as well;
   725 
   726 * Isar: 'pr' command: optional arguments for goals_limit and
   727 ProofContext.prems_limit; no longer prints theory contexts, but only
   728 proof states;
   729 
   730 * Isar: diagnostic commands 'pr', 'thm', 'prop', 'term', 'typ' admit
   731 additional print modes to be specified; e.g. "pr(latex)" will print
   732 proof state according to the Isabelle LaTeX style;
   733 
   734 * Isar: improved support for emulating tactic scripts, including proof
   735 methods 'rule_tac' etc., 'cut_tac', 'thin_tac', 'subgoal_tac',
   736 'rename_tac', 'rotate_tac', 'tactic', and 'case_tac' / 'induct_tac'
   737 (for HOL datatypes);
   738 
   739 * Isar: simplified (more robust) goal selection of proof methods: 1st
   740 goal, all goals, or explicit goal specifier (tactic emulation); thus
   741 'proof method scripts' have to be in depth-first order;
   742 
   743 * Isar: tuned 'let' syntax: replaced 'as' keyword by 'and';
   744 
   745 * Isar: removed 'help' command, which hasn't been too helpful anyway;
   746 should instead use individual commands for printing items
   747 (print_commands, print_methods etc.);
   748 
   749 * Isar: added 'nothing' --- the empty list of theorems;
   750 
   751 
   752 *** HOL ***
   753 
   754 * HOL/MicroJava: formalization of a fragment of Java, together with a
   755 corresponding virtual machine and a specification of its bytecode
   756 verifier and a lightweight bytecode verifier, including proofs of
   757 type-safety; by Gerwin Klein, Tobias Nipkow, David von Oheimb, and
   758 Cornelia Pusch (see also the homepage of project Bali at
   759 http://isabelle.in.tum.de/Bali/);
   760 
   761 * HOL/Algebra: new theory of rings and univariate polynomials, by
   762 Clemens Ballarin;
   763 
   764 * HOL/NumberTheory: fundamental Theorem of Arithmetic, Chinese
   765 Remainder Theorem, Fermat/Euler Theorem, Wilson's Theorem, by Thomas M
   766 Rasmussen;
   767 
   768 * HOL/Lattice: fundamental concepts of lattice theory and order
   769 structures, including duals, properties of bounds versus algebraic
   770 laws, lattice operations versus set-theoretic ones, the Knaster-Tarski
   771 Theorem for complete lattices etc.; may also serve as a demonstration
   772 for abstract algebraic reasoning using axiomatic type classes, and
   773 mathematics-style proof in Isabelle/Isar; by Markus Wenzel;
   774 
   775 * HOL/Prolog: a (bare-bones) implementation of Lambda-Prolog, by David
   776 von Oheimb;
   777 
   778 * HOL/IMPP: extension of IMP with local variables and mutually
   779 recursive procedures, by David von Oheimb;
   780 
   781 * HOL/Lambda: converted into new-style theory and document;
   782 
   783 * HOL/ex/Multiquote: example of multiple nested quotations and
   784 anti-quotations -- basically a generalized version of de-Bruijn
   785 representation; very useful in avoiding lifting of operations;
   786 
   787 * HOL/record: added general record equality rule to simpset; fixed
   788 select-update simplification procedure to handle extended records as
   789 well; admit "r" as field name;
   790 
   791 * HOL: 0 is now overloaded over the new sort "zero", allowing its use with
   792 other numeric types and also as the identity of groups, rings, etc.;
   793 
   794 * HOL: new axclass plus_ac0 for addition with the AC-laws and 0 as identity.
   795 Types nat and int belong to this axclass;
   796 
   797 * HOL: greatly improved simplification involving numerals of type nat, int, real:
   798    (i + #8 + j) = Suc k simplifies to  #7 + (i + j) = k
   799    i*j + k + j*#3*i     simplifies to  #4*(i*j) + k
   800   two terms #m*u and #n*u are replaced by #(m+n)*u
   801     (where #m, #n and u can implicitly be 1; this is simproc combine_numerals)
   802   and the term/formula #m*u+x ~~ #n*u+y simplifies simplifies to #(m-n)+x ~~ y
   803     or x ~~ #(n-m)+y, where ~~ is one of = < <= or - (simproc cancel_numerals);
   804 
   805 * HOL: meson_tac is available (previously in ex/meson.ML); it is a
   806 powerful prover for predicate logic but knows nothing of clasets; see
   807 ex/mesontest.ML and ex/mesontest2.ML for example applications;
   808 
   809 * HOL: new version of "case_tac" subsumes both boolean case split and
   810 "exhaust_tac" on datatypes; INCOMPATIBILITY: exhaust_tac no longer
   811 exists, may define val exhaust_tac = case_tac for ad-hoc portability;
   812 
   813 * HOL: simplification no longer dives into case-expressions: only the
   814 selector expression is simplified, but not the remaining arms; to
   815 enable full simplification of case-expressions for datatype t, you may
   816 remove t.weak_case_cong from the simpset, either globally (Delcongs
   817 [thm"t.weak_case_cong"];) or locally (delcongs [...]).
   818 
   819 * HOL/recdef: the recursion equations generated by 'recdef' for
   820 function 'f' are now called f.simps instead of f.rules; if all
   821 termination conditions are proved automatically, these simplification
   822 rules are added to the simpset, as in primrec; rules may be named
   823 individually as well, resulting in a separate list of theorems for
   824 each equation;
   825 
   826 * HOL/While is a new theory that provides a while-combinator. It
   827 permits the definition of tail-recursive functions without the
   828 provision of a termination measure. The latter is necessary once the
   829 invariant proof rule for while is applied.
   830 
   831 * HOL: new (overloaded) notation for the set of elements below/above
   832 some element: {..u}, {..u(}, {l..}, {)l..}. See theory SetInterval.
   833 
   834 * HOL: theorems impI, allI, ballI bound as "strip";
   835 
   836 * HOL: new tactic induct_thm_tac: thm -> string -> int -> tactic
   837 induct_tac th "x1 ... xn" expects th to have a conclusion of the form
   838 P v1 ... vn and abbreviates res_inst_tac [("v1","x1"),...,("vn","xn")] th;
   839 
   840 * HOL/Real: "rabs" replaced by overloaded "abs" function;
   841 
   842 * HOL: theory Sexp now in HOL/Induct examples (it used to be part of
   843 main HOL, but was unused);
   844 
   845 * HOL: fewer consts declared as global (e.g. have to refer to
   846 "Lfp.lfp" instead of "lfp" internally; affects ML packages only);
   847 
   848 * HOL: tuned AST representation of nested pairs, avoiding bogus output
   849 in case of overlap with user translations (e.g. judgements over
   850 tuples); (note that the underlying logical represenation is still
   851 bogus);
   852 
   853 
   854 *** ZF ***
   855 
   856 * ZF: simplification automatically cancels common terms in arithmetic
   857 expressions over nat and int;
   858 
   859 * ZF: new treatment of nat to minimize type-checking: all operators
   860 coerce their operands to a natural number using the function natify,
   861 making the algebraic laws unconditional;
   862 
   863 * ZF: as above, for int: operators coerce their operands to an integer
   864 using the function intify;
   865 
   866 * ZF: the integer library now contains many of the usual laws for the
   867 orderings, including $<=, and monotonicity laws for $+ and $*;
   868 
   869 * ZF: new example ZF/ex/NatSum to demonstrate integer arithmetic
   870 simplification;
   871 
   872 * FOL and ZF: AddIffs now available, giving theorems of the form P<->Q
   873 to the simplifier and classical reasoner simultaneously;
   874 
   875 
   876 *** General ***
   877 
   878 * Provers: blast_tac now handles actual object-logic rules as
   879 assumptions; note that auto_tac uses blast_tac internally as well;
   880 
   881 * Provers: new functions rulify/rulify_no_asm: thm -> thm for turning
   882 outer -->/All/Ball into ==>/!!; qed_spec_mp now uses rulify_no_asm;
   883 
   884 * Provers: delrules now handles destruct rules as well (no longer need
   885 explicit make_elim);
   886 
   887 * Provers: Blast_tac now warns of and ignores "weak elimination rules" e.g.
   888   [| inj ?f;          ?f ?x = ?f ?y; ?x = ?y ==> ?W |] ==> ?W
   889 use instead the strong form,
   890   [| inj ?f; ~ ?W ==> ?f ?x = ?f ?y; ?x = ?y ==> ?W |] ==> ?W
   891 in HOL, FOL and ZF the function cla_make_elim will create such rules
   892 from destruct-rules;
   893 
   894 * Provers: Simplifier.easy_setup provides a fast path to basic
   895 Simplifier setup for new object-logics;
   896 
   897 * Pure: AST translation rules no longer require constant head on LHS;
   898 
   899 * Pure: improved name spaces: ambiguous output is qualified; support
   900 for hiding of names;
   901 
   902 * system: smart setup of canonical ML_HOME, ISABELLE_INTERFACE, and
   903 XSYMBOL_HOME; no longer need to do manual configuration in most
   904 situations;
   905 
   906 * system: compression of ML heaps images may now be controlled via -c
   907 option of isabelle and isatool usedir (currently only observed by
   908 Poly/ML);
   909 
   910 * system: isatool installfonts may handle X-Symbol fonts as well (very
   911 useful for remote X11);
   912 
   913 * system: provide TAGS file for Isabelle sources;
   914 
   915 * ML: infix 'OF' is a version of 'MRS' with more appropriate argument
   916 order;
   917 
   918 * ML: renamed flags Syntax.trace_norm_ast to Syntax.trace_ast; global
   919 timing flag supersedes proof_timing and Toplevel.trace;
   920 
   921 * ML: new combinators |>> and |>>> for incremental transformations
   922 with secondary results (e.g. certain theory extensions):
   923 
   924 * ML: PureThy.add_defs gets additional argument to indicate potential
   925 overloading (usually false);
   926 
   927 * ML: PureThy.add_thms/add_axioms/add_defs now return theorems as
   928 results;
   929 
   930 
   931 
   932 New in Isabelle99 (October 1999)
   933 --------------------------------
   934 
   935 *** Overview of INCOMPATIBILITIES (see below for more details) ***
   936 
   937 * HOL: The THEN and ELSE parts of conditional expressions (if P then x else y)
   938 are no longer simplified.  (This allows the simplifier to unfold recursive
   939 functional programs.)  To restore the old behaviour, declare
   940 
   941     Delcongs [if_weak_cong];
   942 
   943 * HOL: Removed the obsolete syntax "Compl A"; use -A for set
   944 complement;
   945 
   946 * HOL: the predicate "inj" is now defined by translation to "inj_on";
   947 
   948 * HOL/datatype: mutual_induct_tac no longer exists --
   949   use induct_tac "x_1 ... x_n" instead of mutual_induct_tac ["x_1", ..., "x_n"]
   950 
   951 * HOL/typedef: fixed type inference for representing set; type
   952 arguments now have to occur explicitly on the rhs as type constraints;
   953 
   954 * ZF: The con_defs part of an inductive definition may no longer refer
   955 to constants declared in the same theory;
   956 
   957 * HOL, ZF: the function mk_cases, generated by the inductive
   958 definition package, has lost an argument.  To simplify its result, it
   959 uses the default simpset instead of a supplied list of theorems.
   960 
   961 * HOL/List: the constructors of type list are now Nil and Cons;
   962 
   963 * Simplifier: the type of the infix ML functions
   964         setSSolver addSSolver setSolver addSolver
   965 is now  simpset * solver -> simpset  where `solver' is a new abstract type
   966 for packaging solvers. A solver is created via
   967         mk_solver: string -> (thm list -> int -> tactic) -> solver
   968 where the string argument is only a comment.
   969 
   970 
   971 *** Proof tools ***
   972 
   973 * Provers/Arith/fast_lin_arith.ML contains a functor for creating a
   974 decision procedure for linear arithmetic. Currently it is used for
   975 types `nat', `int', and `real' in HOL (see below); it can, should and
   976 will be instantiated for other types and logics as well.
   977 
   978 * The simplifier now accepts rewrite rules with flexible heads, eg
   979      hom ?f ==> ?f(?x+?y) = ?f ?x + ?f ?y
   980   They are applied like any rule with a non-pattern lhs, i.e. by first-order
   981   matching.
   982 
   983 
   984 *** General ***
   985 
   986 * New Isabelle/Isar subsystem provides an alternative to traditional
   987 tactical theorem proving; together with the ProofGeneral/isar user
   988 interface it offers an interactive environment for developing human
   989 readable proof documents (Isar == Intelligible semi-automated
   990 reasoning); for further information see isatool doc isar-ref,
   991 src/HOL/Isar_examples and http://isabelle.in.tum.de/Isar/
   992 
   993 * improved and simplified presentation of theories: better HTML markup
   994 (including colors), graph views in several sizes; isatool usedir now
   995 provides a proper interface for user theories (via -P option); actual
   996 document preparation based on (PDF)LaTeX is available as well (for
   997 new-style theories only); see isatool doc system for more information;
   998 
   999 * native support for Proof General, both for classic Isabelle and
  1000 Isabelle/Isar;
  1001 
  1002 * ML function thm_deps visualizes dependencies of theorems and lemmas,
  1003 using the graph browser tool;
  1004 
  1005 * Isabelle manuals now also available as PDF;
  1006 
  1007 * theory loader rewritten from scratch (may not be fully
  1008 bug-compatible); old loadpath variable has been replaced by show_path,
  1009 add_path, del_path, reset_path functions; new operations such as
  1010 update_thy, touch_thy, remove_thy, use/update_thy_only (see also
  1011 isatool doc ref);
  1012 
  1013 * improved isatool install: option -k creates KDE application icon,
  1014 option -p DIR installs standalone binaries;
  1015 
  1016 * added ML_PLATFORM setting (useful for cross-platform installations);
  1017 more robust handling of platform specific ML images for SML/NJ;
  1018 
  1019 * the settings environment is now statically scoped, i.e. it is never
  1020 created again in sub-processes invoked from isabelle, isatool, or
  1021 Isabelle;
  1022 
  1023 * path element specification '~~' refers to '$ISABELLE_HOME';
  1024 
  1025 * in locales, the "assumes" and "defines" parts may be omitted if
  1026 empty;
  1027 
  1028 * new print_mode "xsymbols" for extended symbol support (e.g. genuine
  1029 long arrows);
  1030 
  1031 * new print_mode "HTML";
  1032 
  1033 * new flag show_tags controls display of tags of theorems (which are
  1034 basically just comments that may be attached by some tools);
  1035 
  1036 * Isamode 2.6 requires patch to accomodate change of Isabelle font
  1037 mode and goal output format:
  1038 
  1039 diff -r Isamode-2.6/elisp/isa-load.el Isamode/elisp/isa-load.el
  1040 244c244
  1041 <       (list (isa-getenv "ISABELLE") "-msymbols" logic-name)
  1042 ---
  1043 >       (list (isa-getenv "ISABELLE") "-misabelle_font" "-msymbols" logic-name)
  1044 diff -r Isabelle-2.6/elisp/isa-proofstate.el Isamode/elisp/isa-proofstate.el
  1045 181c181
  1046 < (defconst proofstate-proofstart-regexp "^Level [0-9]+$"
  1047 ---
  1048 > (defconst proofstate-proofstart-regexp "^Level [0-9]+"
  1049 
  1050 * function bind_thms stores lists of theorems (cf. bind_thm);
  1051 
  1052 * new shorthand tactics ftac, eatac, datac, fatac;
  1053 
  1054 * qed (and friends) now accept "" as result name; in that case the
  1055 theorem is not stored, but proper checks and presentation of the
  1056 result still apply;
  1057 
  1058 * theorem database now also indexes constants "Trueprop", "all",
  1059 "==>", "=="; thus thms_containing, findI etc. may retrieve more rules;
  1060 
  1061 
  1062 *** HOL ***
  1063 
  1064 ** HOL arithmetic **
  1065 
  1066 * There are now decision procedures for linear arithmetic over nat and
  1067 int:
  1068 
  1069 1. arith_tac copes with arbitrary formulae involving `=', `<', `<=',
  1070 `+', `-', `Suc', `min', `max' and numerical constants; other subterms
  1071 are treated as atomic; subformulae not involving type `nat' or `int'
  1072 are ignored; quantified subformulae are ignored unless they are
  1073 positive universal or negative existential. The tactic has to be
  1074 invoked by hand and can be a little bit slow. In particular, the
  1075 running time is exponential in the number of occurrences of `min' and
  1076 `max', and `-' on `nat'.
  1077 
  1078 2. fast_arith_tac is a cut-down version of arith_tac: it only takes
  1079 (negated) (in)equalities among the premises and the conclusion into
  1080 account (i.e. no compound formulae) and does not know about `min' and
  1081 `max', and `-' on `nat'. It is fast and is used automatically by the
  1082 simplifier.
  1083 
  1084 NB: At the moment, these decision procedures do not cope with mixed
  1085 nat/int formulae where the two parts interact, such as `m < n ==>
  1086 int(m) < int(n)'.
  1087 
  1088 * HOL/Numeral provides a generic theory of numerals (encoded
  1089 efficiently as bit strings); setup for types nat/int/real is in place;
  1090 INCOMPATIBILITY: since numeral syntax is now polymorphic, rather than
  1091 int, existing theories and proof scripts may require a few additional
  1092 type constraints;
  1093 
  1094 * integer division and remainder can now be performed on constant
  1095 arguments;
  1096 
  1097 * many properties of integer multiplication, division and remainder
  1098 are now available;
  1099 
  1100 * An interface to the Stanford Validity Checker (SVC) is available through the
  1101 tactic svc_tac.  Propositional tautologies and theorems of linear arithmetic
  1102 are proved automatically.  SVC must be installed separately, and its results
  1103 must be TAKEN ON TRUST (Isabelle does not check the proofs, but tags any
  1104 invocation of the underlying oracle).  For SVC see
  1105   http://verify.stanford.edu/SVC
  1106 
  1107 * IsaMakefile: the HOL-Real target now builds an actual image;
  1108 
  1109 
  1110 ** HOL misc **
  1111 
  1112 * HOL/Real/HahnBanach: the Hahn-Banach theorem for real vector spaces
  1113 (in Isabelle/Isar) -- by Gertrud Bauer;
  1114 
  1115 * HOL/BCV: generic model of bytecode verification, i.e. data-flow
  1116 analysis for assembly languages with subtypes;
  1117 
  1118 * HOL/TLA (Lamport's Temporal Logic of Actions): major reorganization
  1119 -- avoids syntactic ambiguities and treats state, transition, and
  1120 temporal levels more uniformly; introduces INCOMPATIBILITIES due to
  1121 changed syntax and (many) tactics;
  1122 
  1123 * HOL/inductive: Now also handles more general introduction rules such
  1124   as "ALL y. (y, x) : r --> y : acc r ==> x : acc r"; monotonicity
  1125   theorems are now maintained within the theory (maintained via the
  1126   "mono" attribute);
  1127 
  1128 * HOL/datatype: Now also handles arbitrarily branching datatypes
  1129   (using function types) such as
  1130 
  1131   datatype 'a tree = Atom 'a | Branch "nat => 'a tree"
  1132 
  1133 * HOL/record: record_simproc (part of the default simpset) takes care
  1134 of selectors applied to updated records; record_split_tac is no longer
  1135 part of the default claset; update_defs may now be removed from the
  1136 simpset in many cases; COMPATIBILITY: old behavior achieved by
  1137 
  1138   claset_ref () := claset() addSWrapper record_split_wrapper;
  1139   Delsimprocs [record_simproc]
  1140 
  1141 * HOL/typedef: fixed type inference for representing set; type
  1142 arguments now have to occur explicitly on the rhs as type constraints;
  1143 
  1144 * HOL/recdef (TFL): 'congs' syntax now expects comma separated list of theorem
  1145 names rather than an ML expression;
  1146 
  1147 * HOL/defer_recdef (TFL): like recdef but the well-founded relation can be
  1148 supplied later.  Program schemes can be defined, such as
  1149     "While B C s = (if B s then While B C (C s) else s)"
  1150 where the well-founded relation can be chosen after B and C have been given.
  1151 
  1152 * HOL/List: the constructors of type list are now Nil and Cons;
  1153 INCOMPATIBILITY: while [] and infix # syntax is still there, of
  1154 course, ML tools referring to List.list.op # etc. have to be adapted;
  1155 
  1156 * HOL_quantifiers flag superseded by "HOL" print mode, which is
  1157 disabled by default; run isabelle with option -m HOL to get back to
  1158 the original Gordon/HOL-style output;
  1159 
  1160 * HOL/Ord.thy: new bounded quantifier syntax (input only): ALL x<y. P,
  1161 ALL x<=y. P, EX x<y. P, EX x<=y. P;
  1162 
  1163 * HOL basic syntax simplified (more orthogonal): all variants of
  1164 All/Ex now support plain / symbolic / HOL notation; plain syntax for
  1165 Eps operator is provided as well: "SOME x. P[x]";
  1166 
  1167 * HOL/Sum.thy: sum_case has been moved to HOL/Datatype;
  1168 
  1169 * HOL/Univ.thy: infix syntax <*>, <+>, <**>, <+> eliminated and made
  1170 thus available for user theories;
  1171 
  1172 * HOLCF/IOA/Sequents: renamed 'Cons' to 'Consq' to avoid clash with
  1173 HOL/List; hardly an INCOMPATIBILITY since '>>' syntax is used all the
  1174 time;
  1175 
  1176 * HOL: new tactic smp_tac: int -> int -> tactic, which applies spec
  1177 several times and then mp;
  1178 
  1179 
  1180 *** LK ***
  1181 
  1182 * the notation <<...>> is now available as a notation for sequences of
  1183 formulas;
  1184 
  1185 * the simplifier is now installed
  1186 
  1187 * the axiom system has been generalized (thanks to Soren Heilmann)
  1188 
  1189 * the classical reasoner now has a default rule database
  1190 
  1191 
  1192 *** ZF ***
  1193 
  1194 * new primrec section allows primitive recursive functions to be given
  1195 directly (as in HOL) over datatypes and the natural numbers;
  1196 
  1197 * new tactics induct_tac and exhaust_tac for induction (or case
  1198 analysis) over datatypes and the natural numbers;
  1199 
  1200 * the datatype declaration of type T now defines the recursor T_rec;
  1201 
  1202 * simplification automatically does freeness reasoning for datatype
  1203 constructors;
  1204 
  1205 * automatic type-inference, with AddTCs command to insert new
  1206 type-checking rules;
  1207 
  1208 * datatype introduction rules are now added as Safe Introduction rules
  1209 to the claset;
  1210 
  1211 * the syntax "if P then x else y" is now available in addition to
  1212 if(P,x,y);
  1213 
  1214 
  1215 *** Internal programming interfaces ***
  1216 
  1217 * tuned simplifier trace output; new flag debug_simp;
  1218 
  1219 * structures Vartab / Termtab (instances of TableFun) offer efficient
  1220 tables indexed by indexname_ord / term_ord (compatible with aconv);
  1221 
  1222 * AxClass.axclass_tac lost the theory argument;
  1223 
  1224 * tuned current_goals_markers semantics: begin / end goal avoids
  1225 printing empty lines;
  1226 
  1227 * removed prs and prs_fn hook, which was broken because it did not
  1228 include \n in its semantics, forcing writeln to add one
  1229 uncoditionally; replaced prs_fn by writeln_fn; consider std_output:
  1230 string -> unit if you really want to output text without newline;
  1231 
  1232 * Symbol.output subject to print mode; INCOMPATIBILITY: defaults to
  1233 plain output, interface builders may have to enable 'isabelle_font'
  1234 mode to get Isabelle font glyphs as before;
  1235 
  1236 * refined token_translation interface; INCOMPATIBILITY: output length
  1237 now of type real instead of int;
  1238 
  1239 * theory loader actions may be traced via new ThyInfo.add_hook
  1240 interface (see src/Pure/Thy/thy_info.ML); example application: keep
  1241 your own database of information attached to *whole* theories -- as
  1242 opposed to intra-theory data slots offered via TheoryDataFun;
  1243 
  1244 * proper handling of dangling sort hypotheses (at last!);
  1245 Thm.strip_shyps and Drule.strip_shyps_warning take care of removing
  1246 extra sort hypotheses that can be witnessed from the type signature;
  1247 the force_strip_shyps flag is gone, any remaining shyps are simply
  1248 left in the theorem (with a warning issued by strip_shyps_warning);
  1249 
  1250 
  1251 
  1252 New in Isabelle98-1 (October 1998)
  1253 ----------------------------------
  1254 
  1255 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  1256 
  1257 * several changes of automated proof tools;
  1258 
  1259 * HOL: major changes to the inductive and datatype packages, including
  1260 some minor incompatibilities of theory syntax;
  1261 
  1262 * HOL: renamed r^-1 to 'converse' from 'inverse'; 'inj_onto' is now
  1263 called `inj_on';
  1264 
  1265 * HOL: removed duplicate thms in Arith:
  1266   less_imp_add_less  should be replaced by  trans_less_add1
  1267   le_imp_add_le      should be replaced by  trans_le_add1
  1268 
  1269 * HOL: unary minus is now overloaded (new type constraints may be
  1270 required);
  1271 
  1272 * HOL and ZF: unary minus for integers is now #- instead of #~.  In
  1273 ZF, expressions such as n#-1 must be changed to n#- 1, since #-1 is
  1274 now taken as an integer constant.
  1275 
  1276 * Pure: ML function 'theory_of' renamed to 'theory';
  1277 
  1278 
  1279 *** Proof tools ***
  1280 
  1281 * Simplifier:
  1282   1. Asm_full_simp_tac is now more aggressive.
  1283      1. It will sometimes reorient premises if that increases their power to
  1284         simplify.
  1285      2. It does no longer proceed strictly from left to right but may also
  1286         rotate premises to achieve further simplification.
  1287      For compatibility reasons there is now Asm_lr_simp_tac which is like the
  1288      old Asm_full_simp_tac in that it does not rotate premises.
  1289   2. The simplifier now knows a little bit about nat-arithmetic.
  1290 
  1291 * Classical reasoner: wrapper mechanism for the classical reasoner now
  1292 allows for selected deletion of wrappers, by introduction of names for
  1293 wrapper functionals.  This implies that addbefore, addSbefore,
  1294 addaltern, and addSaltern now take a pair (name, tactic) as argument,
  1295 and that adding two tactics with the same name overwrites the first
  1296 one (emitting a warning).
  1297   type wrapper = (int -> tactic) -> (int -> tactic)
  1298   setWrapper, setSWrapper, compWrapper and compSWrapper are replaced by
  1299   addWrapper, addSWrapper: claset * (string * wrapper) -> claset
  1300   delWrapper, delSWrapper: claset *  string            -> claset
  1301   getWrapper is renamed to appWrappers, getSWrapper to appSWrappers;
  1302 
  1303 * Classical reasoner: addbefore/addSbefore now have APPEND/ORELSE
  1304 semantics; addbefore now affects only the unsafe part of step_tac
  1305 etc.; this affects addss/auto_tac/force_tac, so EXISTING PROOFS MAY
  1306 FAIL, but proofs should be fixable easily, e.g. by replacing Auto_tac
  1307 by Force_tac;
  1308 
  1309 * Classical reasoner: setwrapper to setWrapper and compwrapper to
  1310 compWrapper; added safe wrapper (and access functions for it);
  1311 
  1312 * HOL/split_all_tac is now much faster and fails if there is nothing
  1313 to split.  Some EXISTING PROOFS MAY REQUIRE ADAPTION because the order
  1314 and the names of the automatically generated variables have changed.
  1315 split_all_tac has moved within claset() from unsafe wrappers to safe
  1316 wrappers, which means that !!-bound variables are split much more
  1317 aggressively, and safe_tac and clarify_tac now split such variables.
  1318 If this splitting is not appropriate, use delSWrapper "split_all_tac".
  1319 Note: the same holds for record_split_tac, which does the job of
  1320 split_all_tac for record fields.
  1321 
  1322 * HOL/Simplifier: Rewrite rules for case distinctions can now be added
  1323 permanently to the default simpset using Addsplits just like
  1324 Addsimps. They can be removed via Delsplits just like
  1325 Delsimps. Lower-case versions are also available.
  1326 
  1327 * HOL/Simplifier: The rule split_if is now part of the default
  1328 simpset. This means that the simplifier will eliminate all occurrences
  1329 of if-then-else in the conclusion of a goal. To prevent this, you can
  1330 either remove split_if completely from the default simpset by
  1331 `Delsplits [split_if]' or remove it in a specific call of the
  1332 simplifier using `... delsplits [split_if]'.  You can also add/delete
  1333 other case splitting rules to/from the default simpset: every datatype
  1334 generates suitable rules `split_t_case' and `split_t_case_asm' (where
  1335 t is the name of the datatype).
  1336 
  1337 * Classical reasoner / Simplifier combination: new force_tac (and
  1338 derivatives Force_tac, force) combines rewriting and classical
  1339 reasoning (and whatever other tools) similarly to auto_tac, but is
  1340 aimed to solve the given subgoal completely.
  1341 
  1342 
  1343 *** General ***
  1344 
  1345 * new top-level commands `Goal' and `Goalw' that improve upon `goal'
  1346 and `goalw': the theory is no longer needed as an explicit argument -
  1347 the current theory context is used; assumptions are no longer returned
  1348 at the ML-level unless one of them starts with ==> or !!; it is
  1349 recommended to convert to these new commands using isatool fixgoal
  1350 (backup your sources first!);
  1351 
  1352 * new top-level commands 'thm' and 'thms' for retrieving theorems from
  1353 the current theory context, and 'theory' to lookup stored theories;
  1354 
  1355 * new theory section 'locale' for declaring constants, assumptions and
  1356 definitions that have local scope;
  1357 
  1358 * new theory section 'nonterminals' for purely syntactic types;
  1359 
  1360 * new theory section 'setup' for generic ML setup functions
  1361 (e.g. package initialization);
  1362 
  1363 * the distribution now includes Isabelle icons: see
  1364 lib/logo/isabelle-{small,tiny}.xpm;
  1365 
  1366 * isatool install - install binaries with absolute references to
  1367 ISABELLE_HOME/bin;
  1368 
  1369 * isatool logo -- create instances of the Isabelle logo (as EPS);
  1370 
  1371 * print mode 'emacs' reserved for Isamode;
  1372 
  1373 * support multiple print (ast) translations per constant name;
  1374 
  1375 * theorems involving oracles are now printed with a suffixed [!];
  1376 
  1377 
  1378 *** HOL ***
  1379 
  1380 * there is now a tutorial on Isabelle/HOL (do 'isatool doc tutorial');
  1381 
  1382 * HOL/inductive package reorganized and improved: now supports mutual
  1383 definitions such as
  1384 
  1385   inductive EVEN ODD
  1386     intrs
  1387       null "0 : EVEN"
  1388       oddI "n : EVEN ==> Suc n : ODD"
  1389       evenI "n : ODD ==> Suc n : EVEN"
  1390 
  1391 new theorem list "elims" contains an elimination rule for each of the
  1392 recursive sets; inductive definitions now handle disjunctive premises
  1393 correctly (also ZF);
  1394 
  1395 INCOMPATIBILITIES: requires Inductive as an ancestor; component
  1396 "mutual_induct" no longer exists - the induction rule is always
  1397 contained in "induct";
  1398 
  1399 
  1400 * HOL/datatype package re-implemented and greatly improved: now
  1401 supports mutually recursive datatypes such as
  1402 
  1403   datatype
  1404     'a aexp = IF_THEN_ELSE ('a bexp) ('a aexp) ('a aexp)
  1405             | SUM ('a aexp) ('a aexp)
  1406             | DIFF ('a aexp) ('a aexp)
  1407             | NUM 'a
  1408   and
  1409     'a bexp = LESS ('a aexp) ('a aexp)
  1410             | AND ('a bexp) ('a bexp)
  1411             | OR ('a bexp) ('a bexp)
  1412 
  1413 as well as indirectly recursive datatypes such as
  1414 
  1415   datatype
  1416     ('a, 'b) term = Var 'a
  1417                   | App 'b ((('a, 'b) term) list)
  1418 
  1419 The new tactic  mutual_induct_tac [<var_1>, ..., <var_n>] i  performs
  1420 induction on mutually / indirectly recursive datatypes.
  1421 
  1422 Primrec equations are now stored in theory and can be accessed via
  1423 <function_name>.simps.
  1424 
  1425 INCOMPATIBILITIES:
  1426 
  1427   - Theories using datatypes must now have theory Datatype as an
  1428     ancestor.
  1429   - The specific <typename>.induct_tac no longer exists - use the
  1430     generic induct_tac instead.
  1431   - natE has been renamed to nat.exhaust - use exhaust_tac
  1432     instead of res_inst_tac ... natE. Note that the variable
  1433     names in nat.exhaust differ from the names in natE, this
  1434     may cause some "fragile" proofs to fail.
  1435   - The theorems split_<typename>_case and split_<typename>_case_asm
  1436     have been renamed to <typename>.split and <typename>.split_asm.
  1437   - Since default sorts of type variables are now handled correctly,
  1438     some datatype definitions may have to be annotated with explicit
  1439     sort constraints.
  1440   - Primrec definitions no longer require function name and type
  1441     of recursive argument.
  1442 
  1443 Consider using isatool fixdatatype to adapt your theories and proof
  1444 scripts to the new package (backup your sources first!).
  1445 
  1446 
  1447 * HOL/record package: considerably improved implementation; now
  1448 includes concrete syntax for record types, terms, updates; theorems
  1449 for surjective pairing and splitting !!-bound record variables; proof
  1450 support is as follows:
  1451 
  1452   1) standard conversions (selectors or updates applied to record
  1453 constructor terms) are part of the standard simpset;
  1454 
  1455   2) inject equations of the form ((x, y) = (x', y')) == x=x' & y=y' are
  1456 made part of standard simpset and claset via addIffs;
  1457 
  1458   3) a tactic for record field splitting (record_split_tac) is part of
  1459 the standard claset (addSWrapper);
  1460 
  1461 To get a better idea about these rules you may retrieve them via
  1462 something like 'thms "foo.simps"' or 'thms "foo.iffs"', where "foo" is
  1463 the name of your record type.
  1464 
  1465 The split tactic 3) conceptually simplifies by the following rule:
  1466 
  1467   "(!!x. PROP ?P x) == (!!a b. PROP ?P (a, b))"
  1468 
  1469 Thus any record variable that is bound by meta-all will automatically
  1470 blow up into some record constructor term, consequently the
  1471 simplifications of 1), 2) apply.  Thus force_tac, auto_tac etc. shall
  1472 solve record problems automatically.
  1473 
  1474 
  1475 * reorganized the main HOL image: HOL/Integ and String loaded by
  1476 default; theory Main includes everything;
  1477 
  1478 * automatic simplification of integer sums and comparisons, using cancellation;
  1479 
  1480 * added option_map_eq_Some and not_Some_eq to the default simpset and claset;
  1481 
  1482 * added disj_not1 = "(~P | Q) = (P --> Q)" to the default simpset;
  1483 
  1484 * many new identities for unions, intersections, set difference, etc.;
  1485 
  1486 * expand_if, expand_split, expand_sum_case and expand_nat_case are now
  1487 called split_if, split_split, split_sum_case and split_nat_case (to go
  1488 with add/delsplits);
  1489 
  1490 * HOL/Prod introduces simplification procedure unit_eq_proc rewriting
  1491 (?x::unit) = (); this is made part of the default simpset, which COULD
  1492 MAKE EXISTING PROOFS FAIL under rare circumstances (consider
  1493 'Delsimprocs [unit_eq_proc];' as last resort); also note that
  1494 unit_abs_eta_conv is added in order to counter the effect of
  1495 unit_eq_proc on (%u::unit. f u), replacing it by f rather than by
  1496 %u.f();
  1497 
  1498 * HOL/Fun INCOMPATIBILITY: `inj_onto' is now called `inj_on' (which
  1499 makes more sense);
  1500 
  1501 * HOL/Set INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
  1502   It and 'sym RS equals0D' are now in the default  claset, giving automatic
  1503   disjointness reasoning but breaking a few old proofs.
  1504 
  1505 * HOL/Relation INCOMPATIBILITY: renamed the relational operator r^-1
  1506 to 'converse' from 'inverse' (for compatibility with ZF and some
  1507 literature);
  1508 
  1509 * HOL/recdef can now declare non-recursive functions, with {} supplied as
  1510 the well-founded relation;
  1511 
  1512 * HOL/Set INCOMPATIBILITY: the complement of set A is now written -A instead of
  1513     Compl A.  The "Compl" syntax remains available as input syntax for this
  1514     release ONLY.
  1515 
  1516 * HOL/Update: new theory of function updates:
  1517     f(a:=b) == %x. if x=a then b else f x
  1518 may also be iterated as in f(a:=b,c:=d,...);
  1519 
  1520 * HOL/Vimage: new theory for inverse image of a function, syntax f-``B;
  1521 
  1522 * HOL/List:
  1523   - new function list_update written xs[i:=v] that updates the i-th
  1524     list position. May also be iterated as in xs[i:=a,j:=b,...].
  1525   - new function `upt' written [i..j(] which generates the list
  1526     [i,i+1,...,j-1], i.e. the upper bound is excluded. To include the upper
  1527     bound write [i..j], which is a shorthand for [i..j+1(].
  1528   - new lexicographic orderings and corresponding wellfoundedness theorems.
  1529 
  1530 * HOL/Arith:
  1531   - removed 'pred' (predecessor) function;
  1532   - generalized some theorems about n-1;
  1533   - many new laws about "div" and "mod";
  1534   - new laws about greatest common divisors (see theory ex/Primes);
  1535 
  1536 * HOL/Relation: renamed the relational operator r^-1 "converse"
  1537 instead of "inverse";
  1538 
  1539 * HOL/Induct/Multiset: a theory of multisets, including the wellfoundedness
  1540   of the multiset ordering;
  1541 
  1542 * directory HOL/Real: a construction of the reals using Dedekind cuts
  1543   (not included by default);
  1544 
  1545 * directory HOL/UNITY: Chandy and Misra's UNITY formalism;
  1546 
  1547 * directory HOL/Hoare: a new version of Hoare logic which permits many-sorted
  1548   programs, i.e. different program variables may have different types.
  1549 
  1550 * calling (stac rew i) now fails if "rew" has no effect on the goal
  1551   [previously, this check worked only if the rewrite rule was unconditional]
  1552   Now rew can involve either definitions or equalities (either == or =).
  1553 
  1554 
  1555 *** ZF ***
  1556 
  1557 * theory Main includes everything; INCOMPATIBILITY: theory ZF.thy contains
  1558   only the theorems proved on ZF.ML;
  1559 
  1560 * ZF INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
  1561   It and 'sym RS equals0D' are now in the default  claset, giving automatic
  1562   disjointness reasoning but breaking a few old proofs.
  1563 
  1564 * ZF/Update: new theory of function updates
  1565     with default rewrite rule  f(x:=y) ` z = if(z=x, y, f`z)
  1566   may also be iterated as in f(a:=b,c:=d,...);
  1567 
  1568 * in  let x=t in u(x), neither t nor u(x) has to be an FOL term.
  1569 
  1570 * calling (stac rew i) now fails if "rew" has no effect on the goal
  1571   [previously, this check worked only if the rewrite rule was unconditional]
  1572   Now rew can involve either definitions or equalities (either == or =).
  1573 
  1574 * case_tac provided for compatibility with HOL
  1575     (like the old excluded_middle_tac, but with subgoals swapped)
  1576 
  1577 
  1578 *** Internal programming interfaces ***
  1579 
  1580 * Pure: several new basic modules made available for general use, see
  1581 also src/Pure/README;
  1582 
  1583 * improved the theory data mechanism to support encapsulation (data
  1584 kind name replaced by private Object.kind, acting as authorization
  1585 key); new type-safe user interface via functor TheoryDataFun; generic
  1586 print_data function becomes basically useless;
  1587 
  1588 * removed global_names compatibility flag -- all theory declarations
  1589 are qualified by default;
  1590 
  1591 * module Pure/Syntax now offers quote / antiquote translation
  1592 functions (useful for Hoare logic etc. with implicit dependencies);
  1593 see HOL/ex/Antiquote for an example use;
  1594 
  1595 * Simplifier now offers conversions (asm_)(full_)rewrite: simpset ->
  1596 cterm -> thm;
  1597 
  1598 * new tactical CHANGED_GOAL for checking that a tactic modifies a
  1599 subgoal;
  1600 
  1601 * Display.print_goals function moved to Locale.print_goals;
  1602 
  1603 * standard print function for goals supports current_goals_markers
  1604 variable for marking begin of proof, end of proof, start of goal; the
  1605 default is ("", "", ""); setting current_goals_markers := ("<proof>",
  1606 "</proof>", "<goal>") causes SGML like tagged proof state printing,
  1607 for example;
  1608 
  1609 
  1610 
  1611 New in Isabelle98 (January 1998)
  1612 --------------------------------
  1613 
  1614 *** Overview of INCOMPATIBILITIES (see below for more details) ***
  1615 
  1616 * changed lexical syntax of terms / types: dots made part of long
  1617 identifiers, e.g. "%x.x" no longer possible, should be "%x. x";
  1618 
  1619 * simpset (and claset) reference variable replaced by functions
  1620 simpset / simpset_ref;
  1621 
  1622 * no longer supports theory aliases (via merge) and non-trivial
  1623 implicit merge of thms' signatures;
  1624 
  1625 * most internal names of constants changed due to qualified names;
  1626 
  1627 * changed Pure/Sequence interface (see Pure/seq.ML);
  1628 
  1629 
  1630 *** General Changes ***
  1631 
  1632 * hierachically structured name spaces (for consts, types, axms, thms
  1633 etc.); new lexical class 'longid' (e.g. Foo.bar.x) may render much of
  1634 old input syntactically incorrect (e.g. "%x.x"); COMPATIBILITY:
  1635 isatool fixdots ensures space after dots (e.g. "%x. x"); set
  1636 long_names for fully qualified output names; NOTE: ML programs
  1637 (special tactics, packages etc.) referring to internal names may have
  1638 to be adapted to cope with fully qualified names; in case of severe
  1639 backward campatibility problems try setting 'global_names' at compile
  1640 time to have enrything declared within a flat name space; one may also
  1641 fine tune name declarations in theories via the 'global' and 'local'
  1642 section;
  1643 
  1644 * reimplemented the implicit simpset and claset using the new anytype
  1645 data filed in signatures; references simpset:simpset ref etc. are
  1646 replaced by functions simpset:unit->simpset and
  1647 simpset_ref:unit->simpset ref; COMPATIBILITY: use isatool fixclasimp
  1648 to patch your ML files accordingly;
  1649 
  1650 * HTML output now includes theory graph data for display with Java
  1651 applet or isatool browser; data generated automatically via isatool
  1652 usedir (see -i option, ISABELLE_USEDIR_OPTIONS);
  1653 
  1654 * defs may now be conditional; improved rewrite_goals_tac to handle
  1655 conditional equations;
  1656 
  1657 * defs now admits additional type arguments, using TYPE('a) syntax;
  1658 
  1659 * theory aliases via merge (e.g. M=A+B+C) no longer supported, always
  1660 creates a new theory node; implicit merge of thms' signatures is
  1661 restricted to 'trivial' ones; COMPATIBILITY: one may have to use
  1662 transfer:theory->thm->thm in (rare) cases;
  1663 
  1664 * improved handling of draft signatures / theories; draft thms (and
  1665 ctyps, cterms) are automatically promoted to real ones;
  1666 
  1667 * slightly changed interfaces for oracles: admit many per theory, named
  1668 (e.g. oracle foo = mlfun), additional name argument for invoke_oracle;
  1669 
  1670 * print_goals: optional output of const types (set show_consts and
  1671 show_types);
  1672 
  1673 * improved output of warnings (###) and errors (***);
  1674 
  1675 * subgoal_tac displays a warning if the new subgoal has type variables;
  1676 
  1677 * removed old README and Makefiles;
  1678 
  1679 * replaced print_goals_ref hook by print_current_goals_fn and result_error_fn;
  1680 
  1681 * removed obsolete init_pps and init_database;
  1682 
  1683 * deleted the obsolete tactical STATE, which was declared by
  1684     fun STATE tacfun st = tacfun st st;
  1685 
  1686 * cd and use now support path variables, e.g. $ISABELLE_HOME, or ~
  1687 (which abbreviates $HOME);
  1688 
  1689 * changed Pure/Sequence interface (see Pure/seq.ML); COMPATIBILITY:
  1690 use isatool fixseq to adapt your ML programs (this works for fully
  1691 qualified references to the Sequence structure only!);
  1692 
  1693 * use_thy no longer requires writable current directory; it always
  1694 reloads .ML *and* .thy file, if either one is out of date;
  1695 
  1696 
  1697 *** Classical Reasoner ***
  1698 
  1699 * Clarify_tac, clarify_tac, clarify_step_tac, Clarify_step_tac: new
  1700 tactics that use classical reasoning to simplify a subgoal without
  1701 splitting it into several subgoals;
  1702 
  1703 * Safe_tac: like safe_tac but uses the default claset;
  1704 
  1705 
  1706 *** Simplifier ***
  1707 
  1708 * added simplification meta rules:
  1709     (asm_)(full_)simplify: simpset -> thm -> thm;
  1710 
  1711 * simplifier.ML no longer part of Pure -- has to be loaded by object
  1712 logics (again);
  1713 
  1714 * added prems argument to simplification procedures;
  1715 
  1716 * HOL, FOL, ZF: added infix function `addsplits':
  1717   instead of `<simpset> setloop (split_tac <thms>)'
  1718   you can simply write `<simpset> addsplits <thms>'
  1719 
  1720 
  1721 *** Syntax ***
  1722 
  1723 * TYPE('a) syntax for type reflection terms;
  1724 
  1725 * no longer handles consts with name "" -- declare as 'syntax' instead;
  1726 
  1727 * pretty printer: changed order of mixfix annotation preference (again!);
  1728 
  1729 * Pure: fixed idt/idts vs. pttrn/pttrns syntactic categories;
  1730 
  1731 
  1732 *** HOL ***
  1733 
  1734 * HOL: there is a new splitter `split_asm_tac' that can be used e.g.
  1735   with `addloop' of the simplifier to faciliate case splitting in premises.
  1736 
  1737 * HOL/TLA: Stephan Merz's formalization of Lamport's Temporal Logic of Actions;
  1738 
  1739 * HOL/Auth: new protocol proofs including some for the Internet
  1740   protocol TLS;
  1741 
  1742 * HOL/Map: new theory of `maps' a la VDM;
  1743 
  1744 * HOL/simplifier: simplification procedures nat_cancel_sums for
  1745 cancelling out common nat summands from =, <, <= (in)equalities, or
  1746 differences; simplification procedures nat_cancel_factor for
  1747 cancelling common factor from =, <, <= (in)equalities over natural
  1748 sums; nat_cancel contains both kinds of procedures, it is installed by
  1749 default in Arith.thy -- this COULD MAKE EXISTING PROOFS FAIL;
  1750 
  1751 * HOL/simplifier: terms of the form
  1752   `? x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x)'  (or t=x)
  1753   are rewritten to
  1754   `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t)',
  1755   and those of the form
  1756   `! x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x) --> R(x)'  (or t=x)
  1757   are rewritten to
  1758   `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t) --> R(t)',
  1759 
  1760 * HOL/datatype
  1761   Each datatype `t' now comes with a theorem `split_t_case' of the form
  1762 
  1763   P(t_case f1 ... fn x) =
  1764      ( (!y1 ... ym1. x = C1 y1 ... ym1 --> P(f1 y1 ... ym1)) &
  1765         ...
  1766        (!y1 ... ymn. x = Cn y1 ... ymn --> P(f1 y1 ... ymn))
  1767      )
  1768 
  1769   and a theorem `split_t_case_asm' of the form
  1770 
  1771   P(t_case f1 ... fn x) =
  1772     ~( (? y1 ... ym1. x = C1 y1 ... ym1 & ~P(f1 y1 ... ym1)) |
  1773         ...
  1774        (? y1 ... ymn. x = Cn y1 ... ymn & ~P(f1 y1 ... ymn))
  1775      )
  1776   which can be added to a simpset via `addsplits'. The existing theorems
  1777   expand_list_case and expand_option_case have been renamed to
  1778   split_list_case and split_option_case.
  1779 
  1780 * HOL/Arithmetic:
  1781   - `pred n' is automatically converted to `n-1'.
  1782     Users are strongly encouraged not to use `pred' any longer,
  1783     because it will disappear altogether at some point.
  1784   - Users are strongly encouraged to write "0 < n" rather than
  1785     "n ~= 0". Theorems and proof tools have been modified towards this
  1786     `standard'.
  1787 
  1788 * HOL/Lists:
  1789   the function "set_of_list" has been renamed "set" (and its theorems too);
  1790   the function "nth" now takes its arguments in the reverse order and
  1791   has acquired the infix notation "!" as in "xs!n".
  1792 
  1793 * HOL/Set: UNIV is now a constant and is no longer translated to Compl{};
  1794 
  1795 * HOL/Set: The operator (UN x.B x) now abbreviates (UN x:UNIV. B x) and its
  1796   specialist theorems (like UN1_I) are gone.  Similarly for (INT x.B x);
  1797 
  1798 * HOL/record: extensible records with schematic structural subtyping
  1799 (single inheritance); EXPERIMENTAL version demonstrating the encoding,
  1800 still lacks various theorems and concrete record syntax;
  1801 
  1802 
  1803 *** HOLCF ***
  1804 
  1805 * removed "axioms" and "generated by" sections;
  1806 
  1807 * replaced "ops" section by extended "consts" section, which is capable of
  1808   handling the continuous function space "->" directly;
  1809 
  1810 * domain package:
  1811   . proves theorems immediately and stores them in the theory,
  1812   . creates hierachical name space,
  1813   . now uses normal mixfix annotations (instead of cinfix...),
  1814   . minor changes to some names and values (for consistency),
  1815   . e.g. cases -> casedist, dists_eq -> dist_eqs, [take_lemma] -> take_lemmas,
  1816   . separator between mutual domain defs: changed "," to "and",
  1817   . improved handling of sort constraints;  now they have to
  1818     appear on the left-hand side of the equations only;
  1819 
  1820 * fixed LAM <x,y,zs>.b syntax;
  1821 
  1822 * added extended adm_tac to simplifier in HOLCF -- can now discharge
  1823 adm (%x. P (t x)), where P is chainfinite and t continuous;
  1824 
  1825 
  1826 *** FOL and ZF ***
  1827 
  1828 * FOL: there is a new splitter `split_asm_tac' that can be used e.g.
  1829   with `addloop' of the simplifier to faciliate case splitting in premises.
  1830 
  1831 * qed_spec_mp, qed_goal_spec_mp, qed_goalw_spec_mp are available, as
  1832 in HOL, they strip ALL and --> from proved theorems;
  1833 
  1834 
  1835 
  1836 New in Isabelle94-8 (May 1997)
  1837 ------------------------------
  1838 
  1839 *** General Changes ***
  1840 
  1841 * new utilities to build / run / maintain Isabelle etc. (in parts
  1842 still somewhat experimental); old Makefiles etc. still functional;
  1843 
  1844 * new 'Isabelle System Manual';
  1845 
  1846 * INSTALL text, together with ./configure and ./build scripts;
  1847 
  1848 * reimplemented type inference for greater efficiency, better error
  1849 messages and clean internal interface;
  1850 
  1851 * prlim command for dealing with lots of subgoals (an easier way of
  1852 setting goals_limit);
  1853 
  1854 
  1855 *** Syntax ***
  1856 
  1857 * supports alternative (named) syntax tables (parser and pretty
  1858 printer); internal interface is provided by add_modesyntax(_i);
  1859 
  1860 * Pure, FOL, ZF, HOL, HOLCF now support symbolic input and output; to
  1861 be used in conjunction with the Isabelle symbol font; uses the
  1862 "symbols" syntax table;
  1863 
  1864 * added token_translation interface (may translate name tokens in
  1865 arbitrary ways, dependent on their type (free, bound, tfree, ...) and
  1866 the current print_mode); IMPORTANT: user print translation functions
  1867 are responsible for marking newly introduced bounds
  1868 (Syntax.mark_boundT);
  1869 
  1870 * token translations for modes "xterm" and "xterm_color" that display
  1871 names in bold, underline etc. or colors (which requires a color
  1872 version of xterm);
  1873 
  1874 * infixes may now be declared with names independent of their syntax;
  1875 
  1876 * added typed_print_translation (like print_translation, but may
  1877 access type of constant);
  1878 
  1879 
  1880 *** Classical Reasoner ***
  1881 
  1882 Blast_tac: a new tactic!  It is often more powerful than fast_tac, but has
  1883 some limitations.  Blast_tac...
  1884   + ignores addss, addbefore, addafter; this restriction is intrinsic
  1885   + ignores elimination rules that don't have the correct format
  1886         (the conclusion MUST be a formula variable)
  1887   + ignores types, which can make HOL proofs fail
  1888   + rules must not require higher-order unification, e.g. apply_type in ZF
  1889     [message "Function Var's argument not a bound variable" relates to this]
  1890   + its proof strategy is more general but can actually be slower
  1891 
  1892 * substitution with equality assumptions no longer permutes other
  1893 assumptions;
  1894 
  1895 * minor changes in semantics of addafter (now called addaltern); renamed
  1896 setwrapper to setWrapper and compwrapper to compWrapper; added safe wrapper
  1897 (and access functions for it);
  1898 
  1899 * improved combination of classical reasoner and simplifier:
  1900   + functions for handling clasimpsets
  1901   + improvement of addss: now the simplifier is called _after_ the
  1902     safe steps.
  1903   + safe variant of addss called addSss: uses safe simplifications
  1904     _during_ the safe steps. It is more complete as it allows multiple
  1905     instantiations of unknowns (e.g. with slow_tac).
  1906 
  1907 *** Simplifier ***
  1908 
  1909 * added interface for simplification procedures (functions that
  1910 produce *proven* rewrite rules on the fly, depending on current
  1911 redex);
  1912 
  1913 * ordering on terms as parameter (used for ordered rewriting);
  1914 
  1915 * new functions delcongs, deleqcongs, and Delcongs. richer rep_ss;
  1916 
  1917 * the solver is now split into a safe and an unsafe part.
  1918 This should be invisible for the normal user, except that the
  1919 functions setsolver and addsolver have been renamed to setSolver and
  1920 addSolver; added safe_asm_full_simp_tac;
  1921 
  1922 
  1923 *** HOL ***
  1924 
  1925 * a generic induction tactic `induct_tac' which works for all datatypes and
  1926 also for type `nat';
  1927 
  1928 * a generic case distinction tactic `exhaust_tac' which works for all
  1929 datatypes and also for type `nat';
  1930 
  1931 * each datatype comes with a function `size';
  1932 
  1933 * patterns in case expressions allow tuple patterns as arguments to
  1934 constructors, for example `case x of [] => ... | (x,y,z)#ps => ...';
  1935 
  1936 * primrec now also works with type nat;
  1937 
  1938 * recdef: a new declaration form, allows general recursive functions to be
  1939 defined in theory files.  See HOL/ex/Fib, HOL/ex/Primes, HOL/Subst/Unify.
  1940 
  1941 * the constant for negation has been renamed from "not" to "Not" to
  1942 harmonize with FOL, ZF, LK, etc.;
  1943 
  1944 * HOL/ex/LFilter theory of a corecursive "filter" functional for
  1945 infinite lists;
  1946 
  1947 * HOL/Modelcheck demonstrates invocation of model checker oracle;
  1948 
  1949 * HOL/ex/Ring.thy declares cring_simp, which solves equational
  1950 problems in commutative rings, using axiomatic type classes for + and *;
  1951 
  1952 * more examples in HOL/MiniML and HOL/Auth;
  1953 
  1954 * more default rewrite rules for quantifiers, union/intersection;
  1955 
  1956 * a new constant `arbitrary == @x.False';
  1957 
  1958 * HOLCF/IOA replaces old HOL/IOA;
  1959 
  1960 * HOLCF changes: derived all rules and arities
  1961   + axiomatic type classes instead of classes
  1962   + typedef instead of faking type definitions
  1963   + eliminated the internal constants less_fun, less_cfun, UU_fun, UU_cfun etc.
  1964   + new axclasses cpo, chfin, flat with flat < chfin < pcpo < cpo < po
  1965   + eliminated the types void, one, tr
  1966   + use unit lift and bool lift (with translations) instead of one and tr
  1967   + eliminated blift from Lift3.thy (use Def instead of blift)
  1968   all eliminated rules are derived as theorems --> no visible changes ;
  1969 
  1970 
  1971 *** ZF ***
  1972 
  1973 * ZF now has Fast_tac, Simp_tac and Auto_tac.  Union_iff is a now a default
  1974 rewrite rule; this may affect some proofs.  eq_cs is gone but can be put back
  1975 as ZF_cs addSIs [equalityI];
  1976 
  1977 
  1978 
  1979 New in Isabelle94-7 (November 96)
  1980 ---------------------------------
  1981 
  1982 * allowing negative levels (as offsets) in prlev and choplev;
  1983 
  1984 * super-linear speedup for large simplifications;
  1985 
  1986 * FOL, ZF and HOL now use miniscoping: rewriting pushes
  1987 quantifications in as far as possible (COULD MAKE EXISTING PROOFS
  1988 FAIL); can suppress it using the command Delsimps (ex_simps @
  1989 all_simps); De Morgan laws are also now included, by default;
  1990 
  1991 * improved printing of ==>  :  ~:
  1992 
  1993 * new object-logic "Sequents" adds linear logic, while replacing LK
  1994 and Modal (thanks to Sara Kalvala);
  1995 
  1996 * HOL/Auth: correctness proofs for authentication protocols;
  1997 
  1998 * HOL: new auto_tac combines rewriting and classical reasoning (many
  1999 examples on HOL/Auth);
  2000 
  2001 * HOL: new command AddIffs for declaring theorems of the form P=Q to
  2002 the rewriter and classical reasoner simultaneously;
  2003 
  2004 * function uresult no longer returns theorems in "standard" format;
  2005 regain previous version by: val uresult = standard o uresult;
  2006 
  2007 
  2008 
  2009 New in Isabelle94-6
  2010 -------------------
  2011 
  2012 * oracles -- these establish an interface between Isabelle and trusted
  2013 external reasoners, which may deliver results as theorems;
  2014 
  2015 * proof objects (in particular record all uses of oracles);
  2016 
  2017 * Simp_tac, Fast_tac, etc. that refer to implicit simpset / claset;
  2018 
  2019 * "constdefs" section in theory files;
  2020 
  2021 * "primrec" section (HOL) no longer requires names;
  2022 
  2023 * internal type "tactic" now simply "thm -> thm Sequence.seq";
  2024 
  2025 
  2026 
  2027 New in Isabelle94-5
  2028 -------------------
  2029 
  2030 * reduced space requirements;
  2031 
  2032 * automatic HTML generation from theories;
  2033 
  2034 * theory files no longer require "..." (quotes) around most types;
  2035 
  2036 * new examples, including two proofs of the Church-Rosser theorem;
  2037 
  2038 * non-curried (1994) version of HOL is no longer distributed;
  2039 
  2040 
  2041 
  2042 New in Isabelle94-4
  2043 -------------------
  2044 
  2045 * greatly reduced space requirements;
  2046 
  2047 * theory files (.thy) no longer require \...\ escapes at line breaks;
  2048 
  2049 * searchable theorem database (see the section "Retrieving theorems" on
  2050 page 8 of the Reference Manual);
  2051 
  2052 * new examples, including Grabczewski's monumental case study of the
  2053 Axiom of Choice;
  2054 
  2055 * The previous version of HOL renamed to Old_HOL;
  2056 
  2057 * The new version of HOL (previously called CHOL) uses a curried syntax
  2058 for functions.  Application looks like f a b instead of f(a,b);
  2059 
  2060 * Mutually recursive inductive definitions finally work in HOL;
  2061 
  2062 * In ZF, pattern-matching on tuples is now available in all abstractions and
  2063 translates to the operator "split";
  2064 
  2065 
  2066 
  2067 New in Isabelle94-3
  2068 -------------------
  2069 
  2070 * new infix operator, addss, allowing the classical reasoner to
  2071 perform simplification at each step of its search.  Example:
  2072         fast_tac (cs addss ss)
  2073 
  2074 * a new logic, CHOL, the same as HOL, but with a curried syntax
  2075 for functions.  Application looks like f a b instead of f(a,b).  Also pairs
  2076 look like (a,b) instead of <a,b>;
  2077 
  2078 * PLEASE NOTE: CHOL will eventually replace HOL!
  2079 
  2080 * In CHOL, pattern-matching on tuples is now available in all abstractions.
  2081 It translates to the operator "split".  A new theory of integers is available;
  2082 
  2083 * In ZF, integer numerals now denote two's-complement binary integers.
  2084 Arithmetic operations can be performed by rewriting.  See ZF/ex/Bin.ML;
  2085 
  2086 * Many new examples: I/O automata, Church-Rosser theorem, equivalents
  2087 of the Axiom of Choice;
  2088 
  2089 
  2090 
  2091 New in Isabelle94-2
  2092 -------------------
  2093 
  2094 * Significantly faster resolution;
  2095 
  2096 * the different sections in a .thy file can now be mixed and repeated
  2097 freely;
  2098 
  2099 * Database of theorems for FOL, HOL and ZF.  New
  2100 commands including qed, qed_goal and bind_thm store theorems in the database.
  2101 
  2102 * Simple database queries: return a named theorem (get_thm) or all theorems of
  2103 a given theory (thms_of), or find out what theory a theorem was proved in
  2104 (theory_of_thm);
  2105 
  2106 * Bugs fixed in the inductive definition and datatype packages;
  2107 
  2108 * The classical reasoner provides deepen_tac and depth_tac, making FOL_dup_cs
  2109 and HOL_dup_cs obsolete;
  2110 
  2111 * Syntactic ambiguities caused by the new treatment of syntax in Isabelle94-1
  2112 have been removed;
  2113 
  2114 * Simpler definition of function space in ZF;
  2115 
  2116 * new results about cardinal and ordinal arithmetic in ZF;
  2117 
  2118 * 'subtype' facility in HOL for introducing new types as subsets of existing
  2119 types;
  2120 
  2121 
  2122 $Id$