NEWS
author berghofe
Thu Jul 30 15:52:33 1998 +0200 (1998-07-30)
changeset 5214 75c6392d1274
parent 5207 dd4f51adfff3
child 5217 3ecd7c952396
permissions -rw-r--r--
Adapted to new datatype package.
wenzelm@5207
     1
wenzelm@2553
     2
Isabelle NEWS -- history of user-visible changes
wenzelm@2553
     3
================================================
wenzelm@2553
     4
wenzelm@4981
     5
New in this Isabelle version
wenzelm@4981
     6
----------------------------
oheimb@4649
     7
wenzelm@5127
     8
*** Overview of INCOMPATIBILITIES (see below for more details) ***
wenzelm@4842
     9
wenzelm@5207
    10
* several changes of proof tools (see next section);
wenzelm@5207
    11
berghofe@5214
    12
* HOL/datatype:
berghofe@5214
    13
  - Theories using datatypes must now have Datatype.thy as an
berghofe@5214
    14
    ancestor.
berghofe@5214
    15
  - The specific <typename>.induct_tac no longer exists - use the
berghofe@5214
    16
    generic induct_tac instead.
berghofe@5214
    17
  - natE has been renamed to nat.exhaustion - use exhaust_tac
berghofe@5214
    18
    instead of res_inst_tac ... natE. Note that the variable
berghofe@5214
    19
    names in nat.exhaustion differ from the names in natE, this
berghofe@5214
    20
    may cause some "fragile" proofs to fail.
berghofe@5214
    21
  - the theorems split_<typename>_case and split_<typename>_case_asm
berghofe@5214
    22
    have been renamed to <typename>.split and <typename>.split_asm.
berghofe@5214
    23
  - Since default sorts are no longer ignored by the package,
berghofe@5214
    24
    some datatype definitions may have to be annotated with
berghofe@5214
    25
    explicit sort constraints.
berghofe@5214
    26
  - Primrec definitions no longer require function name and type
berghofe@5214
    27
    of recursive argument.
berghofe@5214
    28
  Use isatool fixdatatype to adapt your theories and proof scripts
berghofe@5214
    29
  to the new package (as usual, backup your sources first!).
berghofe@5214
    30
wenzelm@5207
    31
* HOL/inductive requires Inductive.thy as an ancestor; `inj_onto' is
wenzelm@5207
    32
now called `inj_on' (which makes more sense);
wenzelm@5127
    33
paulson@5160
    34
* HOL/Relation: renamed the relational operator r^-1 to 'converse'
wenzelm@5207
    35
from 'inverse' (for compatibility with ZF and some literature);
paulson@5160
    36
paulson@5160
    37
wenzelm@5127
    38
*** Proof tools ***
nipkow@4880
    39
paulson@5075
    40
* Simplifier: Asm_full_simp_tac is now more aggressive.
nipkow@4880
    41
  1. It will sometimes reorient premises if that increases their power to
nipkow@4880
    42
     simplify.
nipkow@4880
    43
  2. It does no longer proceed strictly from left to right but may also
nipkow@4880
    44
     rotate premises to achieve further simplification.
nipkow@4880
    45
  For compatibility reasons there is now Asm_lr_simp_tac which is like the
nipkow@4880
    46
  old Asm_full_simp_tac in that it does not rotate premises.
nipkow@4880
    47
wenzelm@5127
    48
* Classical reasoner: wrapper mechanism for the classical reasoner now
wenzelm@5127
    49
allows for selected deletion of wrappers, by introduction of names for
wenzelm@5127
    50
wrapper functionals.  This implies that addbefore, addSbefore,
wenzelm@5127
    51
addaltern, and addSaltern now take a pair (name, tactic) as argument,
wenzelm@5127
    52
and that adding two tactics with the same name overwrites the first
wenzelm@5127
    53
one (emitting a warning).
oheimb@4824
    54
  type wrapper = (int -> tactic) -> (int -> tactic)
oheimb@4649
    55
  setWrapper, setSWrapper, compWrapper and compSWrapper are replaced by
oheimb@4824
    56
  addWrapper, addSWrapper: claset * (string * wrapper) -> claset
oheimb@4824
    57
  delWrapper, delSWrapper: claset *  string            -> claset
oheimb@4649
    58
  getWrapper is renamed to appWrappers, getSWrapper to appSWrappers;
oheimb@4649
    59
wenzelm@5127
    60
* HOL/split_all_tac is now much faster and fails if there is nothing
wenzelm@5127
    61
to split.  Existing (fragile) proofs may require adaption because the
wenzelm@5127
    62
order and the names of the automatically generated variables have
wenzelm@5127
    63
changed.  split_all_tac has moved within claset() from unsafe wrappers
wenzelm@5127
    64
to safe wrappers, which means that !!-bound variables are split much
wenzelm@5127
    65
more aggressively, and safe_tac and clarify_tac now split such
wenzelm@5127
    66
variables.  If this splitting is not appropriate, use delSWrapper
wenzelm@5127
    67
"split_all_tac".
wenzelm@5127
    68
wenzelm@5127
    69
* HOL/Simplifier:
wenzelm@5127
    70
wenzelm@5127
    71
 - Rewrite rules for case distinctions can now be added permanently to
wenzelm@5127
    72
   the default simpset using Addsplits just like Addsimps. They can be
wenzelm@5127
    73
   removed via Delsplits just like Delsimps. Lower-case versions are
wenzelm@5127
    74
   also available.
wenzelm@5127
    75
wenzelm@5127
    76
 - The rule split_if is now part of the default simpset. This means
wenzelm@5127
    77
   that the simplifier will eliminate all occurrences of if-then-else
wenzelm@5127
    78
   in the conclusion of a goal. To prevent this, you can either remove
wenzelm@5127
    79
   split_if completely from the default simpset by `Delsplits
wenzelm@5127
    80
   [split_if]' or remove it in a specific call of the simplifier using
wenzelm@5127
    81
   `... delsplits [split_if]'.  You can also add/delete other case
wenzelm@5127
    82
   splitting rules to/from the default simpset: every datatype
wenzelm@5127
    83
   generates suitable rules `split_t_case' and `split_t_case_asm'
wenzelm@5127
    84
   (where t is the name of the datatype).
wenzelm@5127
    85
wenzelm@5127
    86
* Classical reasoner - Simplifier combination: new force_tac (and
wenzelm@5127
    87
derivatives Force_tac, force) combines rewriting and classical
wenzelm@5127
    88
reasoning (and whatever other tools) similarly to auto_tac, but is
wenzelm@5127
    89
aimed to solve the given subgoal completely;
wenzelm@5127
    90
wenzelm@5127
    91
wenzelm@5127
    92
*** General ***
wenzelm@5127
    93
wenzelm@5127
    94
* new toplevel commands `Goal' and `Goalw' that improve upon `goal'
wenzelm@5127
    95
and `goalw': the theory is no longer needed as an explicit argument -
wenzelm@5127
    96
the current theory context is used; assumptions are no longer returned
wenzelm@5127
    97
at the ML-level unless one of them starts with ==> or !!; it is
wenzelm@5127
    98
recommended to convert to these new commands using isatool fixgoal (as
wenzelm@5127
    99
usual, backup your sources first!);
wenzelm@4842
   100
wenzelm@5059
   101
* new toplevel commands 'thm' and 'thms' for retrieving theorems from
wenzelm@5207
   102
the current theory context, and 'theory' to lookup stored theories;
paulson@4806
   103
wenzelm@5127
   104
* new theory section 'nonterminals' for purely syntactic types;
wenzelm@4858
   105
wenzelm@5127
   106
* new theory section 'setup' for generic ML setup functions
wenzelm@5127
   107
(e.g. package initialization);
wenzelm@4869
   108
wenzelm@5131
   109
* the distribution now includes Isabelle icons: see
wenzelm@5131
   110
lib/logo/isabelle-{small,tiny}.xpm;
wenzelm@5131
   111
wenzelm@4711
   112
paulson@4661
   113
*** HOL ***
paulson@4661
   114
berghofe@5214
   115
* HOL/datatype package reorganized and improved: now supports mutually
berghofe@5214
   116
  recursive datatypes such as
berghofe@5214
   117
berghofe@5214
   118
    datatype
berghofe@5214
   119
      'a aexp = IF_THEN_ELSE ('a bexp) ('a aexp) ('a aexp)
berghofe@5214
   120
              | SUM ('a aexp) ('a aexp)
berghofe@5214
   121
              | DIFF ('a aexp) ('a aexp)
berghofe@5214
   122
              | NUM 'a
berghofe@5214
   123
    and
berghofe@5214
   124
      'a bexp = LESS ('a aexp) ('a aexp)
berghofe@5214
   125
              | AND ('a bexp) ('a bexp)
berghofe@5214
   126
              | OR ('a bexp) ('a bexp)
berghofe@5214
   127
berghofe@5214
   128
  as well as indirectly recursive datatypes such as
berghofe@5214
   129
berghofe@5214
   130
    datatype
berghofe@5214
   131
      ('a, 'b) term = Var 'a
berghofe@5214
   132
                    | App 'b ((('a, 'b) term) list)
berghofe@5214
   133
berghofe@5214
   134
  The new tactic
berghofe@5214
   135
berghofe@5214
   136
    mutual_induct_tac [<var_1>, ..., <var_n>] i
berghofe@5214
   137
berghofe@5214
   138
  performs induction on mutually / indirectly recursive datatypes.
berghofe@5214
   139
  Primrec equations are now stored in theory and can be accessed
berghofe@5214
   140
  via <function_name>.simps
berghofe@5214
   141
wenzelm@5125
   142
* reorganized the main HOL image: HOL/Integ and String loaded by
wenzelm@5125
   143
default; theory Main includes everything;
wenzelm@5125
   144
wenzelm@5127
   145
* added option_map_eq_Some to the default simpset claset;
wenzelm@5127
   146
wenzelm@5127
   147
* added disj_not1 = "(~P | Q) = (P --> Q)" to the default simpset;
wenzelm@5127
   148
wenzelm@5127
   149
* many new identities for unions, intersections, set difference, etc.;
wenzelm@5127
   150
wenzelm@5127
   151
* expand_if, expand_split, expand_sum_case and expand_nat_case are now
wenzelm@5127
   152
called split_if, split_split, split_sum_case and split_nat_case (to go
wenzelm@5127
   153
with add/delsplits);
wenzelm@5125
   154
wenzelm@5127
   155
* HOL/Prod introduces simplification procedure unit_eq_proc rewriting
wenzelm@5127
   156
(?x::unit) = (); this is made part of the default simpset, which COULD
wenzelm@5127
   157
MAKE EXISTING PROOFS FAIL under rare circumstances (consider
wenzelm@5207
   158
'Delsimprocs [unit_eq_proc];' as last resort); also note that
wenzelm@5207
   159
unit_abs_eta_conv is added in order to counter the effect of
wenzelm@5207
   160
unit_eq_proc on (%u::unit. f u), replacing it by f rather than by
wenzelm@5207
   161
%u.f();
wenzelm@5125
   162
wenzelm@5207
   163
* HOL/record package: now includes concrete syntax for record types,
wenzelm@5207
   164
terms, updates; still lacks important theorems, like surjective
wenzelm@5207
   165
pairing and split;
wenzelm@5127
   166
wenzelm@5127
   167
* HOL/inductive package reorganized and improved: now supports mutual
wenzelm@5127
   168
definitions such as
berghofe@5109
   169
berghofe@5109
   170
  inductive EVEN ODD
berghofe@5109
   171
    intrs
berghofe@5214
   172
      null "0 : EVEN"
berghofe@5109
   173
      oddI "n : EVEN ==> Suc n : ODD"
berghofe@5109
   174
      evenI "n : ODD ==> Suc n : EVEN"
berghofe@5109
   175
wenzelm@5127
   176
new component "elims" of the structure created by the package contains
wenzelm@5127
   177
an elimination rule for each of the recursive sets; requires
wenzelm@5127
   178
Inductive.thy as an ancestor; component "mutual_induct" no longer
wenzelm@5127
   179
exists - the induction rule is always contained in "induct"; inductive
wenzelm@5127
   180
definitions now handle disjunctive premises correctly (also ZF);
wenzelm@5085
   181
wenzelm@5127
   182
* HOL/recdef can now declare non-recursive functions, with {} supplied as
wenzelm@5127
   183
the well-founded relation;
oheimb@4838
   184
wenzelm@5127
   185
* HOL/Update: new theory of function updates:
wenzelm@5127
   186
    f(a:=b) == %x. if x=a then b else f x
wenzelm@5127
   187
may also be iterated as in f(a:=b,c:=d,...);
nipkow@5077
   188
wenzelm@5127
   189
* HOL/Vimage: new theory for inverse image of a function, syntax f-``B;
paulson@4899
   190
nipkow@5077
   191
* HOL/List: new function list_update written xs[i:=v] that updates the i-th
nipkow@5077
   192
  list position. May also be iterated as in xs[i:=a,j:=b,...].
paulson@4779
   193
wenzelm@5127
   194
* HOL/Arith:
wenzelm@5127
   195
  - removed 'pred' (predecessor) function;
wenzelm@5127
   196
  - generalized some theorems about n-1;
wenzelm@5127
   197
  - many new laws about "div" and "mod";
wenzelm@5127
   198
  - new laws about greatest common divisors (see theory ex/Primes);
oheimb@4766
   199
wenzelm@5127
   200
* HOL/Relation: renamed the relational operator r^-1 "converse"
wenzelm@4842
   201
instead of "inverse";
wenzelm@4711
   202
nipkow@5145
   203
* HOL/Set: added the predicate "disjoint A B" that stands for "A <= Compl B".
nipkow@5145
   204
  This is much better than "A Int B = {}" for Fast/Blast_tac.
nipkow@5145
   205
wenzelm@5127
   206
* directory HOL/Real: a construction of the reals using Dedekind cuts
wenzelm@5127
   207
(not included by default);
nipkow@4835
   208
wenzelm@5127
   209
* directory HOL/UNITY: Chandy and Misra's UNITY formalism;
wenzelm@4711
   210
paulson@5142
   211
* calling (stac rew i) now fails if "rew" has no effect on the goal
paulson@5142
   212
  [previously, this check worked only if the rewrite rule was unconditional]
paulson@5142
   213
wenzelm@5002
   214
paulson@4879
   215
*** ZF ***
paulson@4879
   216
paulson@5160
   217
* theory Main includes everything;
paulson@5160
   218
paulson@5160
   219
* ZF/Update: new theory of function updates
paulson@5160
   220
    with default rewrite rule  f(x:=y) ` z = if(z=x, y, f`z)
paulson@5160
   221
  may also be iterated as in f(a:=b,c:=d,...);
paulson@5160
   222
paulson@4879
   223
* in  let x=t in u(x), neither t nor u(x) has to be an FOL term.
oheimb@4649
   224
paulson@5142
   225
* calling (stac rew i) now fails if "rew" has no effect on the goal
paulson@5142
   226
  [previously, this check worked only if the rewrite rule was unconditional]
paulson@5142
   227
paulson@5160
   228
* case_tac provided for compatibility with HOL
paulson@5160
   229
    (like the old excluded_middle_tac, but with subgoals swapped)
paulson@5160
   230
wenzelm@4842
   231
wenzelm@5127
   232
*** Internal programming interfaces ***
wenzelm@5002
   233
wenzelm@5207
   234
* removed global_names compatibility flag -- all theory declarations
wenzelm@5207
   235
are qualified by default;
wenzelm@5207
   236
wenzelm@5008
   237
* improved the theory data mechanism to support encapsulation (data
wenzelm@5008
   238
kind name replaced by private Object.kind, acting as authorization
wenzelm@5008
   239
key); new type-safe user interface via functor TheoryDataFun;
wenzelm@5002
   240
wenzelm@5085
   241
* module Pure/Syntax now offers quote / antiquote translation
wenzelm@5085
   242
functions (useful for Hoare logic etc. with implicit dependencies);
wenzelm@5085
   243
wenzelm@5127
   244
* Simplifier now offers conversions (asm_)(full_)rewrite: simpset ->
wenzelm@5127
   245
cterm -> thm;
wenzelm@5127
   246
wenzelm@5128
   247
* Pure: several new basic modules made available for general use, see
wenzelm@5128
   248
also src/Pure/README;
wenzelm@5128
   249
wenzelm@5207
   250
* new tactical CHANGED_GOAL for checking that a tactic modifies a
wenzelm@5207
   251
subgoal;
paulson@5142
   252
wenzelm@5002
   253
wenzelm@5002
   254
wenzelm@4410
   255
New in Isabelle98 (January 1998)
wenzelm@4410
   256
--------------------------------
wenzelm@4410
   257
wenzelm@4410
   258
*** Overview of INCOMPATIBILITIES (see below for more details) ***
wenzelm@4410
   259
wenzelm@4410
   260
* changed lexical syntax of terms / types: dots made part of long
wenzelm@4410
   261
identifiers, e.g. "%x.x" no longer possible, should be "%x. x";
wenzelm@4410
   262
wenzelm@4410
   263
* simpset (and claset) reference variable replaced by functions
wenzelm@4410
   264
simpset / simpset_ref;
wenzelm@4410
   265
wenzelm@4410
   266
* no longer supports theory aliases (via merge) and non-trivial
wenzelm@4410
   267
implicit merge of thms' signatures;
wenzelm@4410
   268
wenzelm@4410
   269
* most internal names of constants changed due to qualified names;
wenzelm@4410
   270
wenzelm@4410
   271
* changed Pure/Sequence interface (see Pure/seq.ML);
wenzelm@4410
   272
wenzelm@3454
   273
paulson@3715
   274
*** General Changes ***
paulson@3715
   275
wenzelm@4174
   276
* hierachically structured name spaces (for consts, types, axms, thms
wenzelm@3943
   277
etc.); new lexical class 'longid' (e.g. Foo.bar.x) may render much of
wenzelm@4108
   278
old input syntactically incorrect (e.g. "%x.x"); COMPATIBILITY:
wenzelm@4108
   279
isatool fixdots ensures space after dots (e.g. "%x. x"); set
wenzelm@4174
   280
long_names for fully qualified output names; NOTE: ML programs
wenzelm@4174
   281
(special tactics, packages etc.) referring to internal names may have
wenzelm@4174
   282
to be adapted to cope with fully qualified names; in case of severe
wenzelm@4174
   283
backward campatibility problems try setting 'global_names' at compile
wenzelm@4174
   284
time to have enrything declared within a flat name space; one may also
wenzelm@4174
   285
fine tune name declarations in theories via the 'global' and 'local'
wenzelm@4174
   286
section;
wenzelm@4108
   287
wenzelm@4108
   288
* reimplemented the implicit simpset and claset using the new anytype
wenzelm@4108
   289
data filed in signatures; references simpset:simpset ref etc. are
wenzelm@4108
   290
replaced by functions simpset:unit->simpset and
wenzelm@4108
   291
simpset_ref:unit->simpset ref; COMPATIBILITY: use isatool fixclasimp
wenzelm@4108
   292
to patch your ML files accordingly;
paulson@3856
   293
wenzelm@3857
   294
* HTML output now includes theory graph data for display with Java
wenzelm@3857
   295
applet or isatool browser; data generated automatically via isatool
wenzelm@3901
   296
usedir (see -i option, ISABELLE_USEDIR_OPTIONS);
wenzelm@3857
   297
paulson@3856
   298
* defs may now be conditional; improved rewrite_goals_tac to handle
paulson@3856
   299
conditional equations;
paulson@3856
   300
wenzelm@4174
   301
* defs now admits additional type arguments, using TYPE('a) syntax;
wenzelm@4174
   302
wenzelm@3901
   303
* theory aliases via merge (e.g. M=A+B+C) no longer supported, always
wenzelm@3901
   304
creates a new theory node; implicit merge of thms' signatures is
wenzelm@4112
   305
restricted to 'trivial' ones; COMPATIBILITY: one may have to use
wenzelm@3901
   306
transfer:theory->thm->thm in (rare) cases;
wenzelm@3901
   307
wenzelm@3968
   308
* improved handling of draft signatures / theories; draft thms (and
wenzelm@3968
   309
ctyps, cterms) are automatically promoted to real ones;
wenzelm@3968
   310
wenzelm@3901
   311
* slightly changed interfaces for oracles: admit many per theory, named
wenzelm@3901
   312
(e.g. oracle foo = mlfun), additional name argument for invoke_oracle;
wenzelm@3901
   313
wenzelm@3901
   314
* print_goals: optional output of const types (set show_consts and
wenzelm@3901
   315
show_types);
wenzelm@3851
   316
wenzelm@4388
   317
* improved output of warnings (###) and errors (***);
wenzelm@3697
   318
paulson@4178
   319
* subgoal_tac displays a warning if the new subgoal has type variables;
paulson@4178
   320
paulson@3715
   321
* removed old README and Makefiles;
wenzelm@3697
   322
paulson@3856
   323
* replaced print_goals_ref hook by print_current_goals_fn and result_error_fn;
wenzelm@3670
   324
paulson@3715
   325
* removed obsolete init_pps and init_database;
paulson@3715
   326
paulson@3715
   327
* deleted the obsolete tactical STATE, which was declared by
paulson@3715
   328
    fun STATE tacfun st = tacfun st st;
paulson@3715
   329
wenzelm@4388
   330
* cd and use now support path variables, e.g. $ISABELLE_HOME, or ~
wenzelm@4388
   331
(which abbreviates $HOME);
wenzelm@4269
   332
wenzelm@4269
   333
* changed Pure/Sequence interface (see Pure/seq.ML); COMPATIBILITY:
wenzelm@4269
   334
use isatool fixseq to adapt your ML programs (this works for fully
wenzelm@4269
   335
qualified references to the Sequence structure only!);
wenzelm@4269
   336
wenzelm@4381
   337
* use_thy no longer requires writable current directory; it always
wenzelm@4381
   338
reloads .ML *and* .thy file, if either one is out of date;
wenzelm@4269
   339
paulson@3715
   340
paulson@3715
   341
*** Classical Reasoner ***
paulson@3715
   342
wenzelm@3744
   343
* Clarify_tac, clarify_tac, clarify_step_tac, Clarify_step_tac: new
wenzelm@3744
   344
tactics that use classical reasoning to simplify a subgoal without
wenzelm@3744
   345
splitting it into several subgoals;
paulson@3715
   346
paulson@3719
   347
* Safe_tac: like safe_tac but uses the default claset;
paulson@3719
   348
paulson@3715
   349
paulson@3715
   350
*** Simplifier ***
paulson@3715
   351
paulson@3715
   352
* added simplification meta rules:
paulson@3715
   353
    (asm_)(full_)simplify: simpset -> thm -> thm;
paulson@3715
   354
paulson@3715
   355
* simplifier.ML no longer part of Pure -- has to be loaded by object
paulson@3715
   356
logics (again);
paulson@3715
   357
paulson@3715
   358
* added prems argument to simplification procedures;
paulson@3715
   359
paulson@4325
   360
* HOL, FOL, ZF: added infix function `addsplits':
paulson@4325
   361
  instead of `<simpset> setloop (split_tac <thms>)'
paulson@4325
   362
  you can simply write `<simpset> addsplits <thms>'
paulson@4325
   363
paulson@3715
   364
paulson@3715
   365
*** Syntax ***
paulson@3715
   366
wenzelm@4174
   367
* TYPE('a) syntax for type reflection terms;
wenzelm@4174
   368
wenzelm@3985
   369
* no longer handles consts with name "" -- declare as 'syntax' instead;
paulson@3856
   370
paulson@3856
   371
* pretty printer: changed order of mixfix annotation preference (again!);
wenzelm@3846
   372
paulson@3715
   373
* Pure: fixed idt/idts vs. pttrn/pttrns syntactic categories;
paulson@3715
   374
paulson@3715
   375
paulson@3715
   376
*** HOL ***
paulson@3715
   377
oheimb@4207
   378
* HOL: there is a new splitter `split_asm_tac' that can be used e.g. 
oheimb@4189
   379
  with `addloop' of the simplifier to faciliate case splitting in premises.
oheimb@4189
   380
nipkow@4035
   381
* HOL/TLA: Stephan Merz's formalization of Lamport's Temporal Logic of Actions;
wenzelm@3985
   382
wenzelm@3985
   383
* HOL/Auth: new protocol proofs including some for the Internet
nipkow@4035
   384
  protocol TLS;
wenzelm@3985
   385
wenzelm@4125
   386
* HOL/Map: new theory of `maps' a la VDM;
nipkow@3982
   387
wenzelm@4335
   388
* HOL/simplifier: simplification procedures nat_cancel_sums for
wenzelm@4335
   389
cancelling out common nat summands from =, <, <= (in)equalities, or
wenzelm@4335
   390
differences; simplification procedures nat_cancel_factor for
wenzelm@4335
   391
cancelling common factor from =, <, <= (in)equalities over natural
wenzelm@4373
   392
sums; nat_cancel contains both kinds of procedures, it is installed by
wenzelm@4373
   393
default in Arith.thy -- this COULD MAKE EXISTING PROOFS FAIL;
wenzelm@4335
   394
nipkow@3580
   395
* HOL/simplifier: terms of the form
paulson@4325
   396
  `? x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x)'  (or t=x)
nipkow@3580
   397
  are rewritten to
nipkow@4035
   398
  `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t)',
nipkow@4035
   399
  and those of the form
paulson@4325
   400
  `! x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x) --> R(x)'  (or t=x)
nipkow@4035
   401
  are rewritten to
nipkow@4035
   402
  `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t) --> R(t)',
nipkow@4035
   403
nipkow@4035
   404
* HOL/datatype
nipkow@4035
   405
  Each datatype `t' now comes with a theorem `split_t_case' of the form
nipkow@3580
   406
nipkow@4035
   407
  P(t_case f1 ... fn x) =
nipkow@4035
   408
     ( (!y1 ... ym1. x = C1 y1 ... ym1 --> P(f1 y1 ... ym1)) &
nipkow@4035
   409
        ...
oheimb@4189
   410
       (!y1 ... ymn. x = Cn y1 ... ymn --> P(f1 y1 ... ymn))
nipkow@4035
   411
     )
nipkow@4035
   412
oheimb@4930
   413
  and a theorem `split_t_case_asm' of the form
oheimb@4189
   414
oheimb@4189
   415
  P(t_case f1 ... fn x) =
oheimb@4189
   416
    ~( (? y1 ... ym1. x = C1 y1 ... ym1 & ~P(f1 y1 ... ym1)) |
oheimb@4189
   417
        ...
oheimb@4189
   418
       (? y1 ... ymn. x = Cn y1 ... ymn & ~P(f1 y1 ... ymn))
oheimb@4189
   419
     )
oheimb@4930
   420
  which can be added to a simpset via `addsplits'. The existing theorems
oheimb@4930
   421
  expand_list_case and expand_option_case have been renamed to
oheimb@4930
   422
  split_list_case and split_option_case.
oheimb@4189
   423
nipkow@4361
   424
* HOL/Arithmetic:
nipkow@4361
   425
  - `pred n' is automatically converted to `n-1'.
nipkow@4361
   426
    Users are strongly encouraged not to use `pred' any longer,
nipkow@4361
   427
    because it will disappear altogether at some point.
nipkow@4361
   428
  - Users are strongly encouraged to write "0 < n" rather than
nipkow@4361
   429
    "n ~= 0". Theorems and proof tools have been modified towards this
nipkow@4361
   430
    `standard'.
nipkow@4357
   431
nipkow@4502
   432
* HOL/Lists:
nipkow@4502
   433
  the function "set_of_list" has been renamed "set" (and its theorems too);
nipkow@4502
   434
  the function "nth" now takes its arguments in the reverse order and
nipkow@4502
   435
  has acquired the infix notation "!" as in "xs!n".
paulson@3570
   436
paulson@4154
   437
* HOL/Set: UNIV is now a constant and is no longer translated to Compl{};
paulson@4154
   438
paulson@4154
   439
* HOL/Set: The operator (UN x.B x) now abbreviates (UN x:UNIV. B x) and its
paulson@4154
   440
  specialist theorems (like UN1_I) are gone.  Similarly for (INT x.B x);
paulson@4154
   441
wenzelm@4575
   442
* HOL/record: extensible records with schematic structural subtyping
wenzelm@4575
   443
(single inheritance); EXPERIMENTAL version demonstrating the encoding,
wenzelm@4575
   444
still lacks various theorems and concrete record syntax;
wenzelm@4575
   445
wenzelm@4125
   446
paulson@3715
   447
*** HOLCF ***
wenzelm@3535
   448
wenzelm@4125
   449
* removed "axioms" and "generated by" sections;
wenzelm@4125
   450
oheimb@4123
   451
* replaced "ops" section by extended "consts" section, which is capable of
wenzelm@4125
   452
  handling the continuous function space "->" directly;
wenzelm@4125
   453
wenzelm@4125
   454
* domain package:
wenzelm@4125
   455
  . proves theorems immediately and stores them in the theory,
wenzelm@4125
   456
  . creates hierachical name space,
wenzelm@4125
   457
  . now uses normal mixfix annotations (instead of cinfix...),
wenzelm@4125
   458
  . minor changes to some names and values (for consistency),
wenzelm@4125
   459
  . e.g. cases -> casedist, dists_eq -> dist_eqs, [take_lemma] -> take_lemmas,
wenzelm@4125
   460
  . separator between mutual domain defs: changed "," to "and",
wenzelm@4125
   461
  . improved handling of sort constraints;  now they have to
wenzelm@4125
   462
    appear on the left-hand side of the equations only;
oheimb@4123
   463
oheimb@4123
   464
* fixed LAM <x,y,zs>.b syntax;
wenzelm@3567
   465
wenzelm@3744
   466
* added extended adm_tac to simplifier in HOLCF -- can now discharge
wenzelm@3744
   467
adm (%x. P (t x)), where P is chainfinite and t continuous;
wenzelm@3579
   468
wenzelm@3579
   469
paulson@3719
   470
*** FOL and ZF ***
paulson@3719
   471
oheimb@4207
   472
* FOL: there is a new splitter `split_asm_tac' that can be used e.g. 
oheimb@4189
   473
  with `addloop' of the simplifier to faciliate case splitting in premises.
oheimb@4189
   474
wenzelm@3744
   475
* qed_spec_mp, qed_goal_spec_mp, qed_goalw_spec_mp are available, as
wenzelm@3744
   476
in HOL, they strip ALL and --> from proved theorems;
wenzelm@3744
   477
paulson@3719
   478
wenzelm@3579
   479
wenzelm@3006
   480
New in Isabelle94-8 (May 1997)
wenzelm@3006
   481
------------------------------
wenzelm@2654
   482
paulson@3002
   483
*** General Changes ***
paulson@3002
   484
paulson@3002
   485
* new utilities to build / run / maintain Isabelle etc. (in parts
paulson@3002
   486
still somewhat experimental); old Makefiles etc. still functional;
wenzelm@2971
   487
wenzelm@3205
   488
* new 'Isabelle System Manual';
wenzelm@3205
   489
wenzelm@2825
   490
* INSTALL text, together with ./configure and ./build scripts;
wenzelm@2773
   491
wenzelm@3006
   492
* reimplemented type inference for greater efficiency, better error
wenzelm@3006
   493
messages and clean internal interface;
paulson@3002
   494
paulson@3002
   495
* prlim command for dealing with lots of subgoals (an easier way of
paulson@3002
   496
setting goals_limit);
paulson@3002
   497
wenzelm@3006
   498
wenzelm@3006
   499
*** Syntax ***
paulson@3002
   500
wenzelm@3116
   501
* supports alternative (named) syntax tables (parser and pretty
wenzelm@3116
   502
printer); internal interface is provided by add_modesyntax(_i);
wenzelm@3116
   503
paulson@3002
   504
* Pure, FOL, ZF, HOL, HOLCF now support symbolic input and output; to
paulson@3002
   505
be used in conjunction with the Isabelle symbol font; uses the
paulson@3002
   506
"symbols" syntax table;
paulson@3002
   507
wenzelm@2705
   508
* added token_translation interface (may translate name tokens in
wenzelm@2756
   509
arbitrary ways, dependent on their type (free, bound, tfree, ...) and
wenzelm@3116
   510
the current print_mode); IMPORTANT: user print translation functions
wenzelm@3116
   511
are responsible for marking newly introduced bounds
wenzelm@3116
   512
(Syntax.mark_boundT);
wenzelm@2705
   513
wenzelm@2730
   514
* token translations for modes "xterm" and "xterm_color" that display
wenzelm@3006
   515
names in bold, underline etc. or colors (which requires a color
wenzelm@3006
   516
version of xterm);
wenzelm@2730
   517
paulson@3002
   518
* infixes may now be declared with names independent of their syntax;
paulson@3002
   519
paulson@3002
   520
* added typed_print_translation (like print_translation, but may
paulson@3002
   521
access type of constant);
paulson@3002
   522
wenzelm@3006
   523
paulson@3002
   524
*** Classical Reasoner ***
paulson@3002
   525
paulson@3002
   526
Blast_tac: a new tactic!  It is often more powerful than fast_tac, but has
paulson@3002
   527
some limitations.  Blast_tac...
paulson@3002
   528
  + ignores addss, addbefore, addafter; this restriction is intrinsic
paulson@3002
   529
  + ignores elimination rules that don't have the correct format
paulson@3002
   530
	(the conclusion MUST be a formula variable)
paulson@3002
   531
  + ignores types, which can make HOL proofs fail
paulson@3002
   532
  + rules must not require higher-order unification, e.g. apply_type in ZF
paulson@3002
   533
    [message "Function Var's argument not a bound variable" relates to this]
paulson@3002
   534
  + its proof strategy is more general but can actually be slower
paulson@3002
   535
wenzelm@3107
   536
* substitution with equality assumptions no longer permutes other
wenzelm@3107
   537
assumptions;
paulson@3002
   538
paulson@3002
   539
* minor changes in semantics of addafter (now called addaltern); renamed
paulson@3002
   540
setwrapper to setWrapper and compwrapper to compWrapper; added safe wrapper
wenzelm@3107
   541
(and access functions for it);
paulson@3002
   542
oheimb@3317
   543
* improved combination of classical reasoner and simplifier: 
oheimb@3317
   544
  + functions for handling clasimpsets
oheimb@3317
   545
  + improvement of addss: now the simplifier is called _after_ the
oheimb@3317
   546
    safe steps.
oheimb@3317
   547
  + safe variant of addss called addSss: uses safe simplifications
oheimb@3317
   548
    _during_ the safe steps. It is more complete as it allows multiple 
oheimb@3317
   549
    instantiations of unknowns (e.g. with slow_tac).
wenzelm@3006
   550
paulson@3002
   551
*** Simplifier ***
paulson@3002
   552
wenzelm@3006
   553
* added interface for simplification procedures (functions that
wenzelm@3006
   554
produce *proven* rewrite rules on the fly, depending on current
wenzelm@3006
   555
redex);
wenzelm@3006
   556
wenzelm@3006
   557
* ordering on terms as parameter (used for ordered rewriting);
wenzelm@3006
   558
wenzelm@3107
   559
* new functions delcongs, deleqcongs, and Delcongs. richer rep_ss;
paulson@3002
   560
paulson@3002
   561
* the solver is now split into a safe and an unsafe part.
paulson@3002
   562
This should be invisible for the normal user, except that the
paulson@3002
   563
functions setsolver and addsolver have been renamed to setSolver and
wenzelm@3107
   564
addSolver; added safe_asm_full_simp_tac;
paulson@3002
   565
paulson@3002
   566
paulson@3002
   567
*** HOL ***
paulson@3002
   568
nipkow@3042
   569
* a generic induction tactic `induct_tac' which works for all datatypes and
wenzelm@3107
   570
also for type `nat';
nipkow@3042
   571
nipkow@3316
   572
* a generic case distinction tactic `exhaust_tac' which works for all
nipkow@3316
   573
datatypes and also for type `nat';
nipkow@3316
   574
nipkow@3316
   575
* each datatype comes with a function `size';
nipkow@3316
   576
paulson@3002
   577
* patterns in case expressions allow tuple patterns as arguments to
wenzelm@3107
   578
constructors, for example `case x of [] => ... | (x,y,z)#ps => ...';
paulson@3002
   579
paulson@3002
   580
* primrec now also works with type nat;
paulson@3002
   581
paulson@3338
   582
* recdef: a new declaration form, allows general recursive functions to be
paulson@3338
   583
defined in theory files.  See HOL/ex/Fib, HOL/ex/Primes, HOL/Subst/Unify.
paulson@3338
   584
paulson@3002
   585
* the constant for negation has been renamed from "not" to "Not" to
wenzelm@3107
   586
harmonize with FOL, ZF, LK, etc.;
paulson@3002
   587
wenzelm@3107
   588
* HOL/ex/LFilter theory of a corecursive "filter" functional for
wenzelm@3107
   589
infinite lists;
paulson@3002
   590
wenzelm@3227
   591
* HOL/Modelcheck demonstrates invocation of model checker oracle;
wenzelm@3227
   592
paulson@3002
   593
* HOL/ex/Ring.thy declares cring_simp, which solves equational
paulson@3002
   594
problems in commutative rings, using axiomatic type classes for + and *;
paulson@3002
   595
paulson@3002
   596
* more examples in HOL/MiniML and HOL/Auth;
paulson@3002
   597
paulson@3002
   598
* more default rewrite rules for quantifiers, union/intersection;
paulson@3002
   599
nipkow@3321
   600
* a new constant `arbitrary == @x.False';
nipkow@3321
   601
wenzelm@3107
   602
* HOLCF/IOA replaces old HOL/IOA;
wenzelm@3107
   603
slotosch@2653
   604
* HOLCF changes: derived all rules and arities 
slotosch@2653
   605
  + axiomatic type classes instead of classes 
slotosch@2653
   606
  + typedef instead of faking type definitions
paulson@2747
   607
  + eliminated the internal constants less_fun, less_cfun, UU_fun, UU_cfun etc.
wenzelm@2730
   608
  + new axclasses cpo, chfin, flat with flat < chfin < pcpo < cpo < po
slotosch@2653
   609
  + eliminated the types void, one, tr
slotosch@2653
   610
  + use unit lift and bool lift (with translations) instead of one and tr
slotosch@2653
   611
  + eliminated blift from Lift3.thy (use Def instead of blift)
wenzelm@3107
   612
  all eliminated rules are derived as theorems --> no visible changes ;
oheimb@2649
   613
wenzelm@3006
   614
paulson@3002
   615
*** ZF ***
wenzelm@2553
   616
paulson@2865
   617
* ZF now has Fast_tac, Simp_tac and Auto_tac.  Union_iff is a now a default
paulson@2865
   618
rewrite rule; this may affect some proofs.  eq_cs is gone but can be put back
paulson@2865
   619
as ZF_cs addSIs [equalityI];
wenzelm@2553
   620
wenzelm@2554
   621
wenzelm@2732
   622
wenzelm@2553
   623
New in Isabelle94-7 (November 96)
wenzelm@2553
   624
---------------------------------
wenzelm@2553
   625
wenzelm@2553
   626
* allowing negative levels (as offsets) in prlev and choplev;
wenzelm@2553
   627
wenzelm@2554
   628
* super-linear speedup for large simplifications;
wenzelm@2554
   629
wenzelm@2554
   630
* FOL, ZF and HOL now use miniscoping: rewriting pushes
wenzelm@2554
   631
quantifications in as far as possible (COULD MAKE EXISTING PROOFS
wenzelm@2554
   632
FAIL); can suppress it using the command Delsimps (ex_simps @
wenzelm@2554
   633
all_simps); De Morgan laws are also now included, by default;
wenzelm@2554
   634
wenzelm@2554
   635
* improved printing of ==>  :  ~:
wenzelm@2554
   636
wenzelm@2554
   637
* new object-logic "Sequents" adds linear logic, while replacing LK
wenzelm@2554
   638
and Modal (thanks to Sara Kalvala);
wenzelm@2554
   639
wenzelm@2554
   640
* HOL/Auth: correctness proofs for authentication protocols;
wenzelm@2554
   641
wenzelm@2554
   642
* HOL: new auto_tac combines rewriting and classical reasoning (many
wenzelm@2554
   643
examples on HOL/Auth);
wenzelm@2554
   644
wenzelm@2554
   645
* HOL: new command AddIffs for declaring theorems of the form P=Q to
wenzelm@2554
   646
the rewriter and classical reasoner simultaneously;
wenzelm@2554
   647
wenzelm@2554
   648
* function uresult no longer returns theorems in "standard" format;
wenzelm@2554
   649
regain previous version by: val uresult = standard o uresult;
wenzelm@2554
   650
wenzelm@2554
   651
wenzelm@2554
   652
wenzelm@2554
   653
New in Isabelle94-6
wenzelm@2554
   654
-------------------
wenzelm@2554
   655
wenzelm@2554
   656
* oracles -- these establish an interface between Isabelle and trusted
wenzelm@2554
   657
external reasoners, which may deliver results as theorems;
wenzelm@2554
   658
wenzelm@2554
   659
* proof objects (in particular record all uses of oracles);
wenzelm@2554
   660
wenzelm@2554
   661
* Simp_tac, Fast_tac, etc. that refer to implicit simpset / claset;
wenzelm@2554
   662
wenzelm@2554
   663
* "constdefs" section in theory files;
wenzelm@2554
   664
wenzelm@2554
   665
* "primrec" section (HOL) no longer requires names;
wenzelm@2554
   666
wenzelm@2554
   667
* internal type "tactic" now simply "thm -> thm Sequence.seq";
wenzelm@2554
   668
wenzelm@2554
   669
wenzelm@2554
   670
wenzelm@2554
   671
New in Isabelle94-5
wenzelm@2554
   672
-------------------
wenzelm@2554
   673
wenzelm@2554
   674
* reduced space requirements;
wenzelm@2554
   675
wenzelm@2554
   676
* automatic HTML generation from theories;
wenzelm@2554
   677
wenzelm@2554
   678
* theory files no longer require "..." (quotes) around most types;
wenzelm@2554
   679
wenzelm@2554
   680
* new examples, including two proofs of the Church-Rosser theorem;
wenzelm@2554
   681
wenzelm@2554
   682
* non-curried (1994) version of HOL is no longer distributed;
wenzelm@2554
   683
wenzelm@2553
   684
paulson@2557
   685
paulson@2557
   686
New in Isabelle94-4
paulson@2557
   687
-------------------
paulson@2557
   688
paulson@2747
   689
* greatly reduced space requirements;
paulson@2557
   690
paulson@2557
   691
* theory files (.thy) no longer require \...\ escapes at line breaks;
paulson@2557
   692
paulson@2557
   693
* searchable theorem database (see the section "Retrieving theorems" on 
paulson@2557
   694
page 8 of the Reference Manual);
paulson@2557
   695
paulson@2557
   696
* new examples, including Grabczewski's monumental case study of the
paulson@2557
   697
Axiom of Choice;
paulson@2557
   698
paulson@2557
   699
* The previous version of HOL renamed to Old_HOL;
paulson@2557
   700
paulson@2557
   701
* The new version of HOL (previously called CHOL) uses a curried syntax 
paulson@2557
   702
for functions.  Application looks like f a b instead of f(a,b);
paulson@2557
   703
paulson@2557
   704
* Mutually recursive inductive definitions finally work in HOL;
paulson@2557
   705
paulson@2557
   706
* In ZF, pattern-matching on tuples is now available in all abstractions and
paulson@2557
   707
translates to the operator "split";
paulson@2557
   708
paulson@2557
   709
paulson@2557
   710
paulson@2557
   711
New in Isabelle94-3
paulson@2557
   712
-------------------
paulson@2557
   713
paulson@2557
   714
* new infix operator, addss, allowing the classical reasoner to 
paulson@2557
   715
perform simplification at each step of its search.  Example:
paulson@2557
   716
	fast_tac (cs addss ss)
paulson@2557
   717
paulson@2557
   718
* a new logic, CHOL, the same as HOL, but with a curried syntax 
paulson@2557
   719
for functions.  Application looks like f a b instead of f(a,b).  Also pairs 
paulson@2557
   720
look like (a,b) instead of <a,b>;
paulson@2557
   721
paulson@2557
   722
* PLEASE NOTE: CHOL will eventually replace HOL!
paulson@2557
   723
paulson@2557
   724
* In CHOL, pattern-matching on tuples is now available in all abstractions.
paulson@2557
   725
It translates to the operator "split".  A new theory of integers is available;
paulson@2557
   726
paulson@2557
   727
* In ZF, integer numerals now denote two's-complement binary integers.
paulson@2557
   728
Arithmetic operations can be performed by rewriting.  See ZF/ex/Bin.ML;
paulson@2557
   729
paulson@2557
   730
* Many new examples: I/O automata, Church-Rosser theorem, equivalents 
paulson@2557
   731
of the Axiom of Choice;
paulson@2557
   732
paulson@2557
   733
paulson@2557
   734
paulson@2557
   735
New in Isabelle94-2
paulson@2557
   736
-------------------
paulson@2557
   737
paulson@2557
   738
* Significantly faster resolution;  
paulson@2557
   739
paulson@2557
   740
* the different sections in a .thy file can now be mixed and repeated
paulson@2557
   741
freely;
paulson@2557
   742
paulson@2557
   743
* Database of theorems for FOL, HOL and ZF.  New
paulson@2557
   744
commands including qed, qed_goal and bind_thm store theorems in the database.
paulson@2557
   745
paulson@2557
   746
* Simple database queries: return a named theorem (get_thm) or all theorems of
paulson@2557
   747
a given theory (thms_of), or find out what theory a theorem was proved in
paulson@2557
   748
(theory_of_thm);
paulson@2557
   749
paulson@2557
   750
* Bugs fixed in the inductive definition and datatype packages;
paulson@2557
   751
paulson@2557
   752
* The classical reasoner provides deepen_tac and depth_tac, making FOL_dup_cs
paulson@2557
   753
and HOL_dup_cs obsolete;
paulson@2557
   754
paulson@2557
   755
* Syntactic ambiguities caused by the new treatment of syntax in Isabelle94-1
paulson@2557
   756
have been removed;
paulson@2557
   757
paulson@2557
   758
* Simpler definition of function space in ZF;
paulson@2557
   759
paulson@2557
   760
* new results about cardinal and ordinal arithmetic in ZF;
paulson@2557
   761
paulson@2557
   762
* 'subtype' facility in HOL for introducing new types as subsets of existing
paulson@2557
   763
types;
paulson@2557
   764
paulson@2557
   765
wenzelm@2553
   766
$Id$