NEWS
author wenzelm
Fri Oct 08 11:10:07 1999 +0200 (1999-10-08)
changeset 7791 66d3b64dbf49
parent 7780 099742c562aa
child 7805 0ae9ddc36fe0
permissions -rw-r--r--
tuned;
wenzelm@5363
     1
Isabelle NEWS -- history user-relevant changes
wenzelm@5363
     2
==============================================
wenzelm@2553
     3
wenzelm@4981
     4
New in this Isabelle version
wenzelm@4981
     5
----------------------------
oheimb@4649
     6
paulson@5931
     7
*** Overview of INCOMPATIBILITIES (see below for more details) ***
paulson@5931
     8
paulson@6922
     9
* HOL: The THEN and ELSE parts of conditional expressions (if P then x else y)
paulson@6922
    10
are no longer simplified.  (This allows the simplifier to unfold recursive
paulson@6922
    11
functional programs.)  To restore the old behaviour, declare
wenzelm@7215
    12
wenzelm@7215
    13
    Delcongs [if_weak_cong];
paulson@6922
    14
wenzelm@6269
    15
* HOL: Removed the obsolete syntax "Compl A"; use -A for set
wenzelm@6269
    16
complement;
paulson@5931
    17
wenzelm@6269
    18
* HOL: the predicate "inj" is now defined by translation to "inj_on";
paulson@6174
    19
wenzelm@6386
    20
* HOL/typedef: fixed type inference for representing set; type
wenzelm@6386
    21
arguments now have to occur explicitly on the rhs as type constraints;
wenzelm@6386
    22
wenzelm@6269
    23
* ZF: The con_defs part of an inductive definition may no longer refer
wenzelm@6269
    24
to constants declared in the same theory;
nipkow@6057
    25
wenzelm@6269
    26
* HOL, ZF: the function mk_cases, generated by the inductive
wenzelm@6269
    27
definition package, has lost an argument.  To simplify its result, it
wenzelm@6269
    28
uses the default simpset instead of a supplied list of theorems.
paulson@6141
    29
wenzelm@7215
    30
* HOL/List: the constructors of type list are now Nil and Cons;
wenzelm@7215
    31
nipkow@7619
    32
* Simplifier: the type of the infix ML functions
nipkow@7619
    33
	setSSolver addSSolver setSolver addSolver
nipkow@7619
    34
is now  simpset * solver -> simpset  where `solver' is a new abstract type
nipkow@7619
    35
for packaging solvers. A solver is created via
nipkow@7619
    36
	mk_solver: string -> (thm list -> int -> tactic) -> solver
nipkow@7619
    37
where the string argument is only a comment.
nipkow@6057
    38
wenzelm@7647
    39
paulson@6069
    40
*** Proof tools ***
paulson@6069
    41
wenzelm@6343
    42
* Provers/Arith/fast_lin_arith.ML contains a functor for creating a
wenzelm@6343
    43
decision procedure for linear arithmetic. Currently it is used for
wenzelm@7593
    44
types `nat', `int', and `real' in HOL (see below); it can, should and
wenzelm@7593
    45
will be instantiated for other types and logics as well.
paulson@6069
    46
nipkow@7324
    47
* The simplifier now accepts rewrite rules with flexible heads, eg
nipkow@7324
    48
     hom ?f ==> ?f(?x+?y) = ?f ?x + ?f ?y
nipkow@7324
    49
  They are applied like any rule with a non-pattern lhs, i.e. by first-order
nipkow@7324
    50
  matching.
paulson@6069
    51
wenzelm@7593
    52
paulson@6014
    53
*** General ***
paulson@6014
    54
wenzelm@7215
    55
* new Isabelle/Isar subsystem provides an alternative to traditional
wenzelm@7215
    56
tactical theorem proving; together with the ProofGeneral/isar user
wenzelm@7215
    57
interface it offers an interactive environment for developing human
wenzelm@7215
    58
readable proof documents (Isar == Intelligible semi-automated
wenzelm@7215
    59
reasoning); see isatool doc isar-ref and
wenzelm@7215
    60
http://isabelle.in.tum.de/Isar/ for more information;
wenzelm@7215
    61
wenzelm@7252
    62
* native support for Proof General, both for classic Isabelle and
wenzelm@7215
    63
Isabelle/Isar (the latter is slightly better supported and more
wenzelm@7215
    64
robust);
wenzelm@7215
    65
wenzelm@7791
    66
* ML function thm_deps visualizes dependencies of theorems and lemmas,
wenzelm@7791
    67
using the graph browser tool;
wenzelm@7791
    68
wenzelm@6751
    69
* Isabelle manuals now also available as PDF;
wenzelm@6751
    70
wenzelm@6671
    71
* improved browser info generation: better HTML markup (including
wenzelm@6671
    72
colors), graph views in several sizes; isatool usedir now provides a
wenzelm@6671
    73
proper interface for user theories (via -P option);
wenzelm@6671
    74
wenzelm@6449
    75
* theory loader rewritten from scratch (may not be fully
wenzelm@6449
    76
bug-compatible); old loadpath variable has been replaced by show_path,
wenzelm@6671
    77
add_path, del_path, reset_path functions; new operations such as
wenzelm@7593
    78
update_thy, touch_thy, remove_thy, use/update_thy_only (see also
wenzelm@7593
    79
isatool doc ref);
wenzelm@6449
    80
wenzelm@7215
    81
* improved isatool install: option -k creates KDE application icon,
wenzelm@7215
    82
option -p DIR installs standalone binaries;
wenzelm@7215
    83
wenzelm@7215
    84
* added ML_PLATFORM setting (useful for cross-platform installations);
wenzelm@7215
    85
more robust handling of platform specific ML images for SML/NJ;
wenzelm@7215
    86
wenzelm@7215
    87
* path element specification '~~' refers to '$ISABELLE_HOME';
wenzelm@7215
    88
wenzelm@6343
    89
* in locales, the "assumes" and "defines" parts may be omitted if
wenzelm@6343
    90
empty;
wenzelm@5973
    91
wenzelm@6269
    92
* new print_mode "xsymbols" for extended symbol support (e.g. genuine
wenzelm@6269
    93
long arrows);
wenzelm@6259
    94
wenzelm@6343
    95
* new print_mode "HTML";
wenzelm@6343
    96
wenzelm@6343
    97
* new flag show_tags controls display of tags of theorems (which are
wenzelm@6343
    98
basically just comments that may be attached by some tools);
wenzelm@6343
    99
wenzelm@6461
   100
* Isamode 2.6 requires patch to accomodate change of Isabelle font
wenzelm@6461
   101
mode and goal output format:
wenzelm@6461
   102
wenzelm@6461
   103
diff -r Isamode-2.6/elisp/isa-load.el Isamode/elisp/isa-load.el
wenzelm@6461
   104
244c244
wenzelm@6461
   105
<       (list (isa-getenv "ISABELLE") "-msymbols" logic-name)
wenzelm@6461
   106
---
wenzelm@6533
   107
>       (list (isa-getenv "ISABELLE") "-misabelle_font" "-msymbols" logic-name)
wenzelm@6461
   108
diff -r Isabelle-2.6/elisp/isa-proofstate.el Isamode/elisp/isa-proofstate.el
wenzelm@6461
   109
181c181
wenzelm@6461
   110
< (defconst proofstate-proofstart-regexp "^Level [0-9]+$"
wenzelm@6461
   111
---
wenzelm@6461
   112
> (defconst proofstate-proofstart-regexp "^Level [0-9]+"
wenzelm@6461
   113
wenzelm@7450
   114
* function bind_thms stores lists of theorems (cf. bind_thm);
wenzelm@7450
   115
wenzelm@7593
   116
* new shorthand tactics ftac, eatac, datac, fatac;
wenzelm@7593
   117
wenzelm@7593
   118
* qed (and friends) now accept "" as result name; in that case the
wenzelm@7593
   119
result is not stored, but proper checks and presentation of the result
wenzelm@7593
   120
still apply;
wenzelm@7593
   121
oheimb@6028
   122
nipkow@6057
   123
*** HOL ***
nipkow@6057
   124
wenzelm@7215
   125
** HOL arithmetic **
wenzelm@7215
   126
wenzelm@6343
   127
* There are now decision procedures for linear arithmetic over nat and
wenzelm@6343
   128
int:
nipkow@6131
   129
wenzelm@6343
   130
1. arith_tac copes with arbitrary formulae involving `=', `<', `<=',
wenzelm@6343
   131
`+', `-', `Suc', `min', `max' and numerical constants; other subterms
wenzelm@6343
   132
are treated as atomic; subformulae not involving type `nat' or `int'
wenzelm@6343
   133
are ignored; quantified subformulae are ignored unless they are
wenzelm@6343
   134
positive universal or negative existential. The tactic has to be
wenzelm@6343
   135
invoked by hand and can be a little bit slow. In particular, the
wenzelm@6343
   136
running time is exponential in the number of occurrences of `min' and
wenzelm@6343
   137
`max', and `-' on `nat'.
nipkow@6131
   138
wenzelm@6343
   139
2. fast_arith_tac is a cut-down version of arith_tac: it only takes
wenzelm@6343
   140
(negated) (in)equalities among the premises and the conclusion into
wenzelm@6343
   141
account (i.e. no compound formulae) and does not know about `min' and
wenzelm@6343
   142
`max', and `-' on `nat'. It is fast and is used automatically by the
wenzelm@6343
   143
simplifier.
nipkow@6131
   144
wenzelm@6343
   145
NB: At the moment, these decision procedures do not cope with mixed
wenzelm@6343
   146
nat/int formulae where the two parts interact, such as `m < n ==>
wenzelm@6343
   147
int(m) < int(n)'.
oheimb@6028
   148
wenzelm@7215
   149
* HOL/Numeral provides a generic theory of numerals (encoded
wenzelm@7313
   150
efficiently as bit strings); setup for types nat/int/real is in place;
wenzelm@7215
   151
INCOMPATIBILITY: since numeral syntax is now polymorphic, rather than
wenzelm@7215
   152
int, existing theories and proof scripts may require a few additional
wenzelm@7215
   153
type constraints;
wenzelm@7215
   154
wenzelm@7215
   155
* integer division and remainder can now be performed on constant
wenzelm@7215
   156
arguments;
paulson@7157
   157
wenzelm@7215
   158
* many properties of integer multiplication, division and remainder
wenzelm@7215
   159
are now available;
paulson@6922
   160
paulson@7287
   161
* An interface to the Stanford Validity Checker (SVC) is available through the
paulson@7287
   162
tactic svc_tac.  Propositional tautologies and theorems of linear arithmetic
paulson@7287
   163
are proved automatically.  SVC must be installed separately, and its results
paulson@7287
   164
must be TAKEN ON TRUST (Isabelle does not check the proofs, but tags any
paulson@7287
   165
invocation of the underlying oracle).  For SVC see
paulson@7444
   166
  http://verify.stanford.edu/SVC
paulson@6922
   167
wenzelm@7125
   168
* IsaMakefile: the HOL-Real target now builds an actual image;
wenzelm@7125
   169
wenzelm@7215
   170
wenzelm@7215
   171
** HOL misc **
wenzelm@7215
   172
wenzelm@7595
   173
* HOL/Real/HahnBanach: the Hahn-Banach theorem for real vector spaces
wenzelm@7595
   174
(in Isabelle/Isar) -- by Gertrud Bauer;
wenzelm@7595
   175
wenzelm@7691
   176
* HOL/BCV: generic model of bytecode verification, i.e. data-flow
wenzelm@7691
   177
analysis for assembly languages with subtypes;
wenzelm@7691
   178
wenzelm@6278
   179
* HOL/TLA (Lamport's Temporal Logic of Actions): major reorganization
wenzelm@6278
   180
-- avoids syntactic ambiguities and treats state, transition, and
wenzelm@6278
   181
temporal levels more uniformly; introduces INCOMPATIBILITIES due to
wenzelm@6278
   182
changed syntax and (many) tactics;
wenzelm@6278
   183
wenzelm@7791
   184
* HOL/inductive: Now also handles more general introduction rules such
wenzelm@7791
   185
  as "ALL y. (y, x) : r --> y : acc r ==> x : acc r"; monotonicity
wenzelm@7791
   186
  theorems are now maintained within the theory (maintained via the
wenzelm@7791
   187
  "mono" attribute);
berghofe@7780
   188
wenzelm@7238
   189
* HOL/datatype: Now also handles arbitrarily branching datatypes
wenzelm@7238
   190
  (using function types) such as
wenzelm@7238
   191
wenzelm@7238
   192
  datatype 'a tree = Atom 'a | Branch "nat => 'a tree"
berghofe@7047
   193
wenzelm@7326
   194
* HOL/record: record_simproc (part of the default simpset) takes care
wenzelm@7326
   195
of selectors applied to updated records; record_split_tac is no longer
wenzelm@7327
   196
part of the default claset; update_defs may now be removed from the
wenzelm@7327
   197
simpset in many cases; COMPATIBILITY: old behavior achieved by
wenzelm@7326
   198
wenzelm@7326
   199
  claset_ref () := claset() addSWrapper record_split_wrapper;
wenzelm@7326
   200
  Delsimprocs [record_simproc]
wenzelm@7326
   201
wenzelm@6386
   202
* HOL/typedef: fixed type inference for representing set; type
wenzelm@6386
   203
arguments now have to occur explicitly on the rhs as type constraints;
wenzelm@6386
   204
paulson@7287
   205
* HOL/recdef (TFL): 'congs' syntax now expects comma separated list of theorem
paulson@7287
   206
names rather than an ML expression;
paulson@7287
   207
paulson@7287
   208
* HOL/defer_recdef (TFL): like recdef but the well-founded relation can be
paulson@7287
   209
supplied later.  Program schemes can be defined, such as
paulson@7287
   210
    "While B C s = (if B s then While B C (C s) else s)"
paulson@7287
   211
where the well-founded relation can be chosen after B and C have been given.
wenzelm@6563
   212
wenzelm@7215
   213
* HOL/List: the constructors of type list are now Nil and Cons;
wenzelm@7215
   214
INCOMPATIBILITY: while [] and infix # syntax is still there, of
wenzelm@7215
   215
course, ML tools referring to List.list.op # etc. have to be adapted;
wenzelm@7215
   216
wenzelm@7238
   217
* HOL_quantifiers flag superseded by "HOL" print mode, which is
wenzelm@7238
   218
disabled by default; run isabelle with option -m HOL to get back to
wenzelm@7238
   219
the original Gordon/HOL-style output;
wenzelm@7238
   220
wenzelm@7238
   221
* HOL/Ord.thy: new bounded quantifier syntax (input only): ALL x<y. P,
wenzelm@7238
   222
ALL x<=y. P, EX x<y. P, EX x<=y. P;
wenzelm@7238
   223
wenzelm@7238
   224
* HOL basic syntax simplified (more orthogonal): all variants of
wenzelm@7238
   225
All/Ex now support plain / symbolic / HOL notation; plain syntax for
wenzelm@7238
   226
Eps operator is provided as well: "SOME x. P[x]";
wenzelm@7238
   227
berghofe@7320
   228
* HOL/Sum.thy: sum_case has been moved to HOL/Datatype;
wenzelm@7261
   229
wenzelm@7280
   230
* HOL/Univ.thy: infix syntax <*>, <+>, <**>, <+> eliminated and made
wenzelm@7280
   231
thus available for user theories;
wenzelm@7280
   232
wenzelm@7300
   233
* HOLCF/IOA/Sequents: renamed 'Cons' to 'Consq' to avoid clash with
wenzelm@7300
   234
HOL/List; hardly an INCOMPATIBILITY since '>>' syntax is used all the
wenzelm@7300
   235
time;
wenzelm@7300
   236
oheimb@7492
   237
* HOL: new tactic smp_tac: int -> int -> tactic, which applies spec several
oheimb@7492
   238
  times and then mp
oheimb@7492
   239
wenzelm@7215
   240
paulson@7113
   241
*** LK ***
paulson@7113
   242
wenzelm@7215
   243
* the notation <<...>> is now available as a notation for sequences of
wenzelm@7215
   244
formulas;
paulson@7113
   245
paulson@7113
   246
* the simplifier is now installed
paulson@7113
   247
paulson@7113
   248
* the axiom system has been generalized (thanks to Soren Heilmann) 
paulson@7113
   249
paulson@7113
   250
* the classical reasoner now has a default rule database
paulson@7113
   251
paulson@7113
   252
paulson@6064
   253
*** ZF ***
paulson@6064
   254
paulson@6064
   255
* new primrec section allows primitive recursive functions to be given
wenzelm@6269
   256
directly (as in HOL) over datatypes and the natural numbers;
paulson@6064
   257
wenzelm@6269
   258
* new tactics induct_tac and exhaust_tac for induction (or case
wenzelm@6269
   259
analysis) over datatypes and the natural numbers;
paulson@6064
   260
paulson@6064
   261
* the datatype declaration of type T now defines the recursor T_rec;
paulson@6064
   262
paulson@6141
   263
* simplification automatically does freeness reasoning for datatype
wenzelm@6269
   264
constructors;
paulson@6141
   265
wenzelm@6269
   266
* automatic type-inference, with AddTCs command to insert new
wenzelm@6269
   267
type-checking rules;
paulson@6155
   268
wenzelm@6269
   269
* datatype introduction rules are now added as Safe Introduction rules
wenzelm@6269
   270
to the claset;
paulson@6155
   271
wenzelm@6269
   272
* the syntax "if P then x else y" is now available in addition to
wenzelm@6269
   273
if(P,x,y);
wenzelm@6269
   274
paulson@6069
   275
wenzelm@6343
   276
*** Internal programming interfaces ***
wenzelm@6343
   277
wenzelm@7420
   278
* structures Vartab / Termtab (instances of TableFun) offer efficient
wenzelm@7420
   279
tables indexed by indexname_ord / term_ord (compatible with aconv);
wenzelm@7420
   280
wenzelm@6386
   281
* AxClass.axclass_tac lost the theory argument;
wenzelm@6386
   282
wenzelm@6343
   283
* tuned current_goals_markers semantics: begin / end goal avoids
wenzelm@6343
   284
printing empty lines;
wenzelm@6343
   285
wenzelm@6343
   286
* removed prs and prs_fn hook, which was broken because it did not
wenzelm@6343
   287
include \n in its semantics, forcing writeln to add one
wenzelm@6343
   288
uncoditionally; replaced prs_fn by writeln_fn; consider std_output:
wenzelm@6343
   289
string -> unit if you really want to output text without newline;
wenzelm@6343
   290
wenzelm@6343
   291
* Symbol.output subject to print mode; INCOMPATIBILITY: defaults to
wenzelm@6343
   292
plain output, interface builders may have to enable 'isabelle_font'
wenzelm@6343
   293
mode to get Isabelle font glyphs as before;
wenzelm@6343
   294
wenzelm@6343
   295
* refined token_translation interface; INCOMPATIBILITY: output length
wenzelm@6343
   296
now of type real instead of int;
wenzelm@6343
   297
wenzelm@7196
   298
* theory loader actions may be traced via new ThyInfo.add_hook
wenzelm@7196
   299
interface (see src/Pure/Thy/thy_info.ML); example application: keep
wenzelm@7196
   300
your own database of information attached to *whole* theories -- as
wenzelm@7196
   301
opposed to intra-theory data slots offered via TheoryDataFun;
wenzelm@7196
   302
wenzelm@7647
   303
* proper handling of dangling sort hypotheses (at last!);
wenzelm@7647
   304
Thm.strip_shyps and Drule.strip_shyps_warning take care of removing
wenzelm@7647
   305
extra sort hypotheses that can be witnessed from the type signature;
wenzelm@7647
   306
the force_strip_shyps is gone, any remaining shyps are simply left in
wenzelm@7647
   307
the theorem (with a warning issued by strip_shyps_warning);
wenzelm@7647
   308
wenzelm@6343
   309
paulson@6064
   310
wenzelm@5781
   311
New in Isabelle98-1 (October 1998)
wenzelm@5781
   312
----------------------------------
wenzelm@5781
   313
wenzelm@5127
   314
*** Overview of INCOMPATIBILITIES (see below for more details) ***
wenzelm@4842
   315
wenzelm@5726
   316
* several changes of automated proof tools;
wenzelm@5373
   317
wenzelm@5726
   318
* HOL: major changes to the inductive and datatype packages, including
wenzelm@5726
   319
some minor incompatibilities of theory syntax;
berghofe@5214
   320
wenzelm@5726
   321
* HOL: renamed r^-1 to 'converse' from 'inverse'; 'inj_onto' is now
wenzelm@5217
   322
called `inj_on';
paulson@5160
   323
nipkow@5275
   324
* HOL: removed duplicate thms in Arith:
nipkow@5275
   325
  less_imp_add_less  should be replaced by  trans_less_add1
nipkow@5275
   326
  le_imp_add_le      should be replaced by  trans_le_add1
paulson@5160
   327
wenzelm@5726
   328
* HOL: unary minus is now overloaded (new type constraints may be
wenzelm@5726
   329
required);
paulson@5490
   330
wenzelm@5726
   331
* HOL and ZF: unary minus for integers is now #- instead of #~.  In
wenzelm@5726
   332
ZF, expressions such as n#-1 must be changed to n#- 1, since #-1 is
wenzelm@5726
   333
now taken as an integer constant.
paulson@5541
   334
wenzelm@5726
   335
* Pure: ML function 'theory_of' renamed to 'theory';
wenzelm@5397
   336
wenzelm@5363
   337
wenzelm@5127
   338
*** Proof tools ***
nipkow@4880
   339
nipkow@5657
   340
* Simplifier:
nipkow@5657
   341
  1. Asm_full_simp_tac is now more aggressive.
nipkow@5657
   342
     1. It will sometimes reorient premises if that increases their power to
nipkow@5657
   343
        simplify.
nipkow@5657
   344
     2. It does no longer proceed strictly from left to right but may also
nipkow@5657
   345
        rotate premises to achieve further simplification.
nipkow@5657
   346
     For compatibility reasons there is now Asm_lr_simp_tac which is like the
nipkow@5657
   347
     old Asm_full_simp_tac in that it does not rotate premises.
nipkow@5657
   348
  2. The simplifier now knows a little bit about nat-arithmetic.
nipkow@4880
   349
wenzelm@5127
   350
* Classical reasoner: wrapper mechanism for the classical reasoner now
wenzelm@5127
   351
allows for selected deletion of wrappers, by introduction of names for
wenzelm@5127
   352
wrapper functionals.  This implies that addbefore, addSbefore,
wenzelm@5127
   353
addaltern, and addSaltern now take a pair (name, tactic) as argument,
wenzelm@5127
   354
and that adding two tactics with the same name overwrites the first
wenzelm@5127
   355
one (emitting a warning).
oheimb@4824
   356
  type wrapper = (int -> tactic) -> (int -> tactic)
oheimb@4649
   357
  setWrapper, setSWrapper, compWrapper and compSWrapper are replaced by
oheimb@4824
   358
  addWrapper, addSWrapper: claset * (string * wrapper) -> claset
oheimb@4824
   359
  delWrapper, delSWrapper: claset *  string            -> claset
oheimb@4649
   360
  getWrapper is renamed to appWrappers, getSWrapper to appSWrappers;
oheimb@4649
   361
wenzelm@5705
   362
* Classical reasoner: addbefore/addSbefore now have APPEND/ORELSE
wenzelm@5726
   363
semantics; addbefore now affects only the unsafe part of step_tac
wenzelm@5726
   364
etc.; this affects addss/auto_tac/force_tac, so EXISTING PROOFS MAY
wenzelm@5726
   365
FAIL, but proofs should be fixable easily, e.g. by replacing Auto_tac
wenzelm@5726
   366
by Force_tac;
oheimb@5524
   367
wenzelm@5726
   368
* Classical reasoner: setwrapper to setWrapper and compwrapper to
wenzelm@5726
   369
compWrapper; added safe wrapper (and access functions for it);
oheimb@5524
   370
wenzelm@5127
   371
* HOL/split_all_tac is now much faster and fails if there is nothing
wenzelm@5726
   372
to split.  Some EXISTING PROOFS MAY REQUIRE ADAPTION because the order
wenzelm@5726
   373
and the names of the automatically generated variables have changed.
wenzelm@5726
   374
split_all_tac has moved within claset() from unsafe wrappers to safe
wenzelm@5726
   375
wrappers, which means that !!-bound variables are split much more
wenzelm@5726
   376
aggressively, and safe_tac and clarify_tac now split such variables.
wenzelm@5726
   377
If this splitting is not appropriate, use delSWrapper "split_all_tac".
wenzelm@5726
   378
Note: the same holds for record_split_tac, which does the job of
wenzelm@5726
   379
split_all_tac for record fields.
wenzelm@5127
   380
wenzelm@5726
   381
* HOL/Simplifier: Rewrite rules for case distinctions can now be added
wenzelm@5726
   382
permanently to the default simpset using Addsplits just like
wenzelm@5726
   383
Addsimps. They can be removed via Delsplits just like
wenzelm@5726
   384
Delsimps. Lower-case versions are also available.
wenzelm@5127
   385
wenzelm@5726
   386
* HOL/Simplifier: The rule split_if is now part of the default
wenzelm@5726
   387
simpset. This means that the simplifier will eliminate all occurrences
wenzelm@5726
   388
of if-then-else in the conclusion of a goal. To prevent this, you can
wenzelm@5726
   389
either remove split_if completely from the default simpset by
wenzelm@5726
   390
`Delsplits [split_if]' or remove it in a specific call of the
wenzelm@5726
   391
simplifier using `... delsplits [split_if]'.  You can also add/delete
wenzelm@5726
   392
other case splitting rules to/from the default simpset: every datatype
wenzelm@5726
   393
generates suitable rules `split_t_case' and `split_t_case_asm' (where
wenzelm@5726
   394
t is the name of the datatype).
wenzelm@5127
   395
wenzelm@5726
   396
* Classical reasoner / Simplifier combination: new force_tac (and
wenzelm@5127
   397
derivatives Force_tac, force) combines rewriting and classical
wenzelm@5127
   398
reasoning (and whatever other tools) similarly to auto_tac, but is
wenzelm@5726
   399
aimed to solve the given subgoal completely.
wenzelm@5127
   400
wenzelm@5127
   401
wenzelm@5127
   402
*** General ***
wenzelm@5127
   403
wenzelm@5217
   404
* new top-level commands `Goal' and `Goalw' that improve upon `goal'
wenzelm@5127
   405
and `goalw': the theory is no longer needed as an explicit argument -
wenzelm@5127
   406
the current theory context is used; assumptions are no longer returned
wenzelm@5127
   407
at the ML-level unless one of them starts with ==> or !!; it is
wenzelm@5217
   408
recommended to convert to these new commands using isatool fixgoal
wenzelm@5217
   409
(backup your sources first!);
wenzelm@4842
   410
wenzelm@5217
   411
* new top-level commands 'thm' and 'thms' for retrieving theorems from
wenzelm@5207
   412
the current theory context, and 'theory' to lookup stored theories;
paulson@4806
   413
paulson@5722
   414
* new theory section 'locale' for declaring constants, assumptions and
paulson@5722
   415
definitions that have local scope;
paulson@5722
   416
wenzelm@5127
   417
* new theory section 'nonterminals' for purely syntactic types;
wenzelm@4858
   418
wenzelm@5127
   419
* new theory section 'setup' for generic ML setup functions
wenzelm@5127
   420
(e.g. package initialization);
wenzelm@4869
   421
wenzelm@5131
   422
* the distribution now includes Isabelle icons: see
wenzelm@5131
   423
lib/logo/isabelle-{small,tiny}.xpm;
wenzelm@5131
   424
wenzelm@5363
   425
* isatool install - install binaries with absolute references to
wenzelm@5363
   426
ISABELLE_HOME/bin;
wenzelm@5363
   427
wenzelm@5572
   428
* isatool logo -- create instances of the Isabelle logo (as EPS);
wenzelm@5572
   429
wenzelm@5407
   430
* print mode 'emacs' reserved for Isamode;
wenzelm@5407
   431
wenzelm@5726
   432
* support multiple print (ast) translations per constant name;
wenzelm@5726
   433
wenzelm@6925
   434
* theorems involving oracles are now printed with a suffixed [!];
wenzelm@6925
   435
wenzelm@4711
   436
paulson@4661
   437
*** HOL ***
paulson@4661
   438
wenzelm@5710
   439
* there is now a tutorial on Isabelle/HOL (do 'isatool doc tutorial');
nipkow@5709
   440
wenzelm@5217
   441
* HOL/inductive package reorganized and improved: now supports mutual
paulson@5267
   442
definitions such as
wenzelm@5217
   443
wenzelm@5217
   444
  inductive EVEN ODD
wenzelm@5217
   445
    intrs
wenzelm@5217
   446
      null "0 : EVEN"
wenzelm@5217
   447
      oddI "n : EVEN ==> Suc n : ODD"
wenzelm@5217
   448
      evenI "n : ODD ==> Suc n : EVEN"
wenzelm@5217
   449
wenzelm@5217
   450
new theorem list "elims" contains an elimination rule for each of the
wenzelm@5217
   451
recursive sets; inductive definitions now handle disjunctive premises
wenzelm@5217
   452
correctly (also ZF);
berghofe@5214
   453
wenzelm@5217
   454
INCOMPATIBILITIES: requires Inductive as an ancestor; component
wenzelm@5217
   455
"mutual_induct" no longer exists - the induction rule is always
wenzelm@5217
   456
contained in "induct";
wenzelm@5217
   457
wenzelm@5217
   458
wenzelm@5217
   459
* HOL/datatype package re-implemented and greatly improved: now
paulson@5267
   460
supports mutually recursive datatypes such as
wenzelm@5217
   461
wenzelm@5217
   462
  datatype
wenzelm@5217
   463
    'a aexp = IF_THEN_ELSE ('a bexp) ('a aexp) ('a aexp)
wenzelm@5217
   464
            | SUM ('a aexp) ('a aexp)
wenzelm@5217
   465
            | DIFF ('a aexp) ('a aexp)
wenzelm@5217
   466
            | NUM 'a
wenzelm@5217
   467
  and
wenzelm@5217
   468
    'a bexp = LESS ('a aexp) ('a aexp)
wenzelm@5217
   469
            | AND ('a bexp) ('a bexp)
wenzelm@5217
   470
            | OR ('a bexp) ('a bexp)
wenzelm@5217
   471
paulson@5267
   472
as well as indirectly recursive datatypes such as
berghofe@5214
   473
wenzelm@5217
   474
  datatype
wenzelm@5217
   475
    ('a, 'b) term = Var 'a
wenzelm@5217
   476
                  | App 'b ((('a, 'b) term) list)
berghofe@5214
   477
wenzelm@5217
   478
The new tactic  mutual_induct_tac [<var_1>, ..., <var_n>] i  performs
wenzelm@5217
   479
induction on mutually / indirectly recursive datatypes.
wenzelm@5217
   480
wenzelm@5217
   481
Primrec equations are now stored in theory and can be accessed via
wenzelm@5217
   482
<function_name>.simps.
wenzelm@5217
   483
wenzelm@5217
   484
INCOMPATIBILITIES:
berghofe@5214
   485
wenzelm@5217
   486
  - Theories using datatypes must now have theory Datatype as an
wenzelm@5217
   487
    ancestor.
wenzelm@5217
   488
  - The specific <typename>.induct_tac no longer exists - use the
wenzelm@5217
   489
    generic induct_tac instead.
berghofe@5226
   490
  - natE has been renamed to nat.exhaust - use exhaust_tac
wenzelm@5217
   491
    instead of res_inst_tac ... natE. Note that the variable
berghofe@5226
   492
    names in nat.exhaust differ from the names in natE, this
wenzelm@5217
   493
    may cause some "fragile" proofs to fail.
wenzelm@5217
   494
  - The theorems split_<typename>_case and split_<typename>_case_asm
wenzelm@5217
   495
    have been renamed to <typename>.split and <typename>.split_asm.
wenzelm@5217
   496
  - Since default sorts of type variables are now handled correctly,
wenzelm@5217
   497
    some datatype definitions may have to be annotated with explicit
wenzelm@5217
   498
    sort constraints.
wenzelm@5217
   499
  - Primrec definitions no longer require function name and type
wenzelm@5217
   500
    of recursive argument.
berghofe@5214
   501
wenzelm@5217
   502
Consider using isatool fixdatatype to adapt your theories and proof
wenzelm@5217
   503
scripts to the new package (backup your sources first!).
wenzelm@5217
   504
wenzelm@5217
   505
wenzelm@5726
   506
* HOL/record package: considerably improved implementation; now
wenzelm@5726
   507
includes concrete syntax for record types, terms, updates; theorems
wenzelm@5726
   508
for surjective pairing and splitting !!-bound record variables; proof
wenzelm@5726
   509
support is as follows:
wenzelm@5726
   510
wenzelm@5726
   511
  1) standard conversions (selectors or updates applied to record
wenzelm@5726
   512
constructor terms) are part of the standard simpset;
wenzelm@5726
   513
wenzelm@5726
   514
  2) inject equations of the form ((x, y) = (x', y')) == x=x' & y=y' are
wenzelm@5726
   515
made part of standard simpset and claset via addIffs;
wenzelm@5726
   516
wenzelm@5726
   517
  3) a tactic for record field splitting (record_split_tac) is part of
wenzelm@5726
   518
the standard claset (addSWrapper);
wenzelm@5726
   519
wenzelm@5726
   520
To get a better idea about these rules you may retrieve them via
wenzelm@5726
   521
something like 'thms "foo.simps"' or 'thms "foo.iffs"', where "foo" is
wenzelm@5726
   522
the name of your record type.
wenzelm@5726
   523
wenzelm@5726
   524
The split tactic 3) conceptually simplifies by the following rule:
wenzelm@5726
   525
wenzelm@5726
   526
  "(!!x. PROP ?P x) == (!!a b. PROP ?P (a, b))"
wenzelm@5726
   527
wenzelm@5726
   528
Thus any record variable that is bound by meta-all will automatically
wenzelm@5726
   529
blow up into some record constructor term, consequently the
wenzelm@5726
   530
simplifications of 1), 2) apply.  Thus force_tac, auto_tac etc. shall
wenzelm@5726
   531
solve record problems automatically.
wenzelm@5726
   532
berghofe@5214
   533
wenzelm@5125
   534
* reorganized the main HOL image: HOL/Integ and String loaded by
wenzelm@5125
   535
default; theory Main includes everything;
wenzelm@5125
   536
paulson@5650
   537
* automatic simplification of integer sums and comparisons, using cancellation;
paulson@5650
   538
oheimb@5526
   539
* added option_map_eq_Some and not_Some_eq to the default simpset and claset;
wenzelm@5127
   540
wenzelm@5127
   541
* added disj_not1 = "(~P | Q) = (P --> Q)" to the default simpset;
wenzelm@5127
   542
wenzelm@5127
   543
* many new identities for unions, intersections, set difference, etc.;
wenzelm@5127
   544
wenzelm@5127
   545
* expand_if, expand_split, expand_sum_case and expand_nat_case are now
wenzelm@5127
   546
called split_if, split_split, split_sum_case and split_nat_case (to go
wenzelm@5127
   547
with add/delsplits);
wenzelm@5125
   548
wenzelm@5127
   549
* HOL/Prod introduces simplification procedure unit_eq_proc rewriting
wenzelm@5127
   550
(?x::unit) = (); this is made part of the default simpset, which COULD
wenzelm@5127
   551
MAKE EXISTING PROOFS FAIL under rare circumstances (consider
wenzelm@5207
   552
'Delsimprocs [unit_eq_proc];' as last resort); also note that
wenzelm@5207
   553
unit_abs_eta_conv is added in order to counter the effect of
wenzelm@5207
   554
unit_eq_proc on (%u::unit. f u), replacing it by f rather than by
wenzelm@5207
   555
%u.f();
wenzelm@5125
   556
wenzelm@5217
   557
* HOL/Fun INCOMPATIBILITY: `inj_onto' is now called `inj_on' (which
wenzelm@5217
   558
makes more sense);
berghofe@5109
   559
paulson@5475
   560
* HOL/Set INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
paulson@5475
   561
  It and 'sym RS equals0D' are now in the default  claset, giving automatic
paulson@5475
   562
  disjointness reasoning but breaking a few old proofs.
paulson@5267
   563
wenzelm@5217
   564
* HOL/Relation INCOMPATIBILITY: renamed the relational operator r^-1
wenzelm@5217
   565
to 'converse' from 'inverse' (for compatibility with ZF and some
wenzelm@5217
   566
literature);
wenzelm@5085
   567
wenzelm@5127
   568
* HOL/recdef can now declare non-recursive functions, with {} supplied as
wenzelm@5127
   569
the well-founded relation;
oheimb@4838
   570
paulson@5490
   571
* HOL/Set INCOMPATIBILITY: the complement of set A is now written -A instead of
paulson@5490
   572
    Compl A.  The "Compl" syntax remains available as input syntax for this
paulson@5490
   573
    release ONLY.
paulson@5490
   574
wenzelm@5127
   575
* HOL/Update: new theory of function updates:
wenzelm@5127
   576
    f(a:=b) == %x. if x=a then b else f x
wenzelm@5127
   577
may also be iterated as in f(a:=b,c:=d,...);
nipkow@5077
   578
wenzelm@5127
   579
* HOL/Vimage: new theory for inverse image of a function, syntax f-``B;
paulson@4899
   580
nipkow@5282
   581
* HOL/List:
nipkow@5282
   582
  - new function list_update written xs[i:=v] that updates the i-th
nipkow@5282
   583
    list position. May also be iterated as in xs[i:=a,j:=b,...].
nipkow@5428
   584
  - new function `upt' written [i..j(] which generates the list
nipkow@5428
   585
    [i,i+1,...,j-1], i.e. the upper bound is excluded. To include the upper
nipkow@5428
   586
    bound write [i..j], which is a shorthand for [i..j+1(].
nipkow@5282
   587
  - new lexicographic orderings and corresponding wellfoundedness theorems.
paulson@4779
   588
wenzelm@5127
   589
* HOL/Arith:
wenzelm@5127
   590
  - removed 'pred' (predecessor) function;
wenzelm@5127
   591
  - generalized some theorems about n-1;
wenzelm@5127
   592
  - many new laws about "div" and "mod";
wenzelm@5127
   593
  - new laws about greatest common divisors (see theory ex/Primes);
oheimb@4766
   594
wenzelm@5127
   595
* HOL/Relation: renamed the relational operator r^-1 "converse"
wenzelm@4842
   596
instead of "inverse";
wenzelm@4711
   597
nipkow@5651
   598
* HOL/Induct/Multiset: a theory of multisets, including the wellfoundedness
nipkow@5651
   599
  of the multiset ordering;
nipkow@5651
   600
wenzelm@5127
   601
* directory HOL/Real: a construction of the reals using Dedekind cuts
nipkow@5651
   602
  (not included by default);
nipkow@4835
   603
wenzelm@5127
   604
* directory HOL/UNITY: Chandy and Misra's UNITY formalism;
wenzelm@4711
   605
nipkow@5651
   606
* directory HOL/Hoare: a new version of Hoare logic which permits many-sorted
nipkow@5651
   607
  programs, i.e. different program variables may have different types.
nipkow@5651
   608
paulson@5142
   609
* calling (stac rew i) now fails if "rew" has no effect on the goal
paulson@5142
   610
  [previously, this check worked only if the rewrite rule was unconditional]
paulson@5308
   611
  Now rew can involve either definitions or equalities (either == or =).
wenzelm@5002
   612
wenzelm@5363
   613
paulson@4879
   614
*** ZF ***
paulson@4879
   615
paulson@5332
   616
* theory Main includes everything; INCOMPATIBILITY: theory ZF.thy contains
paulson@5332
   617
  only the theorems proved on ZF.ML;
paulson@5160
   618
paulson@5475
   619
* ZF INCOMPATIBILITY: rule `equals0D' is now a well-formed destruct rule;
paulson@5475
   620
  It and 'sym RS equals0D' are now in the default  claset, giving automatic
paulson@5475
   621
  disjointness reasoning but breaking a few old proofs.
paulson@5267
   622
paulson@5160
   623
* ZF/Update: new theory of function updates
paulson@5160
   624
    with default rewrite rule  f(x:=y) ` z = if(z=x, y, f`z)
paulson@5160
   625
  may also be iterated as in f(a:=b,c:=d,...);
paulson@5160
   626
paulson@4879
   627
* in  let x=t in u(x), neither t nor u(x) has to be an FOL term.
oheimb@4649
   628
paulson@5142
   629
* calling (stac rew i) now fails if "rew" has no effect on the goal
paulson@5142
   630
  [previously, this check worked only if the rewrite rule was unconditional]
paulson@5308
   631
  Now rew can involve either definitions or equalities (either == or =).
paulson@5142
   632
paulson@5160
   633
* case_tac provided for compatibility with HOL
paulson@5160
   634
    (like the old excluded_middle_tac, but with subgoals swapped)
paulson@5160
   635
wenzelm@4842
   636
wenzelm@5127
   637
*** Internal programming interfaces ***
wenzelm@5002
   638
wenzelm@5251
   639
* Pure: several new basic modules made available for general use, see
wenzelm@5251
   640
also src/Pure/README;
wenzelm@5207
   641
wenzelm@5008
   642
* improved the theory data mechanism to support encapsulation (data
wenzelm@5008
   643
kind name replaced by private Object.kind, acting as authorization
wenzelm@5373
   644
key); new type-safe user interface via functor TheoryDataFun; generic
wenzelm@5373
   645
print_data function becomes basically useless;
wenzelm@5002
   646
wenzelm@5251
   647
* removed global_names compatibility flag -- all theory declarations
wenzelm@5251
   648
are qualified by default;
wenzelm@5251
   649
wenzelm@5085
   650
* module Pure/Syntax now offers quote / antiquote translation
wenzelm@5085
   651
functions (useful for Hoare logic etc. with implicit dependencies);
wenzelm@5373
   652
see HOL/ex/Antiquote for an example use;
wenzelm@5085
   653
wenzelm@5127
   654
* Simplifier now offers conversions (asm_)(full_)rewrite: simpset ->
wenzelm@5127
   655
cterm -> thm;
wenzelm@5127
   656
wenzelm@5207
   657
* new tactical CHANGED_GOAL for checking that a tactic modifies a
wenzelm@5207
   658
subgoal;
paulson@5142
   659
wenzelm@5251
   660
* Display.print_goals function moved to Locale.print_goals;
wenzelm@5251
   661
wenzelm@5731
   662
* standard print function for goals supports current_goals_markers
wenzelm@5731
   663
variable for marking begin of proof, end of proof, start of goal; the
wenzelm@5731
   664
default is ("", "", ""); setting current_goals_markers := ("<proof>",
wenzelm@5731
   665
"</proof>", "<goal>") causes SGML like tagged proof state printing,
wenzelm@5731
   666
for example;
wenzelm@5731
   667
wenzelm@5002
   668
wenzelm@5002
   669
wenzelm@4410
   670
New in Isabelle98 (January 1998)
wenzelm@4410
   671
--------------------------------
wenzelm@4410
   672
wenzelm@4410
   673
*** Overview of INCOMPATIBILITIES (see below for more details) ***
wenzelm@4410
   674
wenzelm@4410
   675
* changed lexical syntax of terms / types: dots made part of long
wenzelm@4410
   676
identifiers, e.g. "%x.x" no longer possible, should be "%x. x";
wenzelm@4410
   677
wenzelm@4410
   678
* simpset (and claset) reference variable replaced by functions
wenzelm@4410
   679
simpset / simpset_ref;
wenzelm@4410
   680
wenzelm@4410
   681
* no longer supports theory aliases (via merge) and non-trivial
wenzelm@4410
   682
implicit merge of thms' signatures;
wenzelm@4410
   683
wenzelm@4410
   684
* most internal names of constants changed due to qualified names;
wenzelm@4410
   685
wenzelm@4410
   686
* changed Pure/Sequence interface (see Pure/seq.ML);
wenzelm@4410
   687
wenzelm@3454
   688
paulson@3715
   689
*** General Changes ***
paulson@3715
   690
wenzelm@4174
   691
* hierachically structured name spaces (for consts, types, axms, thms
wenzelm@3943
   692
etc.); new lexical class 'longid' (e.g. Foo.bar.x) may render much of
wenzelm@4108
   693
old input syntactically incorrect (e.g. "%x.x"); COMPATIBILITY:
wenzelm@4108
   694
isatool fixdots ensures space after dots (e.g. "%x. x"); set
wenzelm@4174
   695
long_names for fully qualified output names; NOTE: ML programs
wenzelm@4174
   696
(special tactics, packages etc.) referring to internal names may have
wenzelm@4174
   697
to be adapted to cope with fully qualified names; in case of severe
wenzelm@4174
   698
backward campatibility problems try setting 'global_names' at compile
wenzelm@4174
   699
time to have enrything declared within a flat name space; one may also
wenzelm@4174
   700
fine tune name declarations in theories via the 'global' and 'local'
wenzelm@4174
   701
section;
wenzelm@4108
   702
wenzelm@4108
   703
* reimplemented the implicit simpset and claset using the new anytype
wenzelm@4108
   704
data filed in signatures; references simpset:simpset ref etc. are
wenzelm@4108
   705
replaced by functions simpset:unit->simpset and
wenzelm@4108
   706
simpset_ref:unit->simpset ref; COMPATIBILITY: use isatool fixclasimp
wenzelm@4108
   707
to patch your ML files accordingly;
paulson@3856
   708
wenzelm@3857
   709
* HTML output now includes theory graph data for display with Java
wenzelm@3857
   710
applet or isatool browser; data generated automatically via isatool
wenzelm@3901
   711
usedir (see -i option, ISABELLE_USEDIR_OPTIONS);
wenzelm@3857
   712
paulson@3856
   713
* defs may now be conditional; improved rewrite_goals_tac to handle
paulson@3856
   714
conditional equations;
paulson@3856
   715
wenzelm@4174
   716
* defs now admits additional type arguments, using TYPE('a) syntax;
wenzelm@4174
   717
wenzelm@3901
   718
* theory aliases via merge (e.g. M=A+B+C) no longer supported, always
wenzelm@3901
   719
creates a new theory node; implicit merge of thms' signatures is
wenzelm@4112
   720
restricted to 'trivial' ones; COMPATIBILITY: one may have to use
wenzelm@3901
   721
transfer:theory->thm->thm in (rare) cases;
wenzelm@3901
   722
wenzelm@3968
   723
* improved handling of draft signatures / theories; draft thms (and
wenzelm@3968
   724
ctyps, cterms) are automatically promoted to real ones;
wenzelm@3968
   725
wenzelm@3901
   726
* slightly changed interfaces for oracles: admit many per theory, named
wenzelm@3901
   727
(e.g. oracle foo = mlfun), additional name argument for invoke_oracle;
wenzelm@3901
   728
wenzelm@3901
   729
* print_goals: optional output of const types (set show_consts and
wenzelm@3901
   730
show_types);
wenzelm@3851
   731
wenzelm@4388
   732
* improved output of warnings (###) and errors (***);
wenzelm@3697
   733
paulson@4178
   734
* subgoal_tac displays a warning if the new subgoal has type variables;
paulson@4178
   735
paulson@3715
   736
* removed old README and Makefiles;
wenzelm@3697
   737
paulson@3856
   738
* replaced print_goals_ref hook by print_current_goals_fn and result_error_fn;
wenzelm@3670
   739
paulson@3715
   740
* removed obsolete init_pps and init_database;
paulson@3715
   741
paulson@3715
   742
* deleted the obsolete tactical STATE, which was declared by
paulson@3715
   743
    fun STATE tacfun st = tacfun st st;
paulson@3715
   744
wenzelm@4388
   745
* cd and use now support path variables, e.g. $ISABELLE_HOME, or ~
wenzelm@4388
   746
(which abbreviates $HOME);
wenzelm@4269
   747
wenzelm@4269
   748
* changed Pure/Sequence interface (see Pure/seq.ML); COMPATIBILITY:
wenzelm@4269
   749
use isatool fixseq to adapt your ML programs (this works for fully
wenzelm@4269
   750
qualified references to the Sequence structure only!);
wenzelm@4269
   751
wenzelm@4381
   752
* use_thy no longer requires writable current directory; it always
wenzelm@4381
   753
reloads .ML *and* .thy file, if either one is out of date;
wenzelm@4269
   754
paulson@3715
   755
paulson@3715
   756
*** Classical Reasoner ***
paulson@3715
   757
wenzelm@3744
   758
* Clarify_tac, clarify_tac, clarify_step_tac, Clarify_step_tac: new
wenzelm@3744
   759
tactics that use classical reasoning to simplify a subgoal without
wenzelm@3744
   760
splitting it into several subgoals;
paulson@3715
   761
paulson@3719
   762
* Safe_tac: like safe_tac but uses the default claset;
paulson@3719
   763
paulson@3715
   764
paulson@3715
   765
*** Simplifier ***
paulson@3715
   766
paulson@3715
   767
* added simplification meta rules:
paulson@3715
   768
    (asm_)(full_)simplify: simpset -> thm -> thm;
paulson@3715
   769
paulson@3715
   770
* simplifier.ML no longer part of Pure -- has to be loaded by object
paulson@3715
   771
logics (again);
paulson@3715
   772
paulson@3715
   773
* added prems argument to simplification procedures;
paulson@3715
   774
paulson@4325
   775
* HOL, FOL, ZF: added infix function `addsplits':
paulson@4325
   776
  instead of `<simpset> setloop (split_tac <thms>)'
paulson@4325
   777
  you can simply write `<simpset> addsplits <thms>'
paulson@4325
   778
paulson@3715
   779
paulson@3715
   780
*** Syntax ***
paulson@3715
   781
wenzelm@4174
   782
* TYPE('a) syntax for type reflection terms;
wenzelm@4174
   783
wenzelm@3985
   784
* no longer handles consts with name "" -- declare as 'syntax' instead;
paulson@3856
   785
paulson@3856
   786
* pretty printer: changed order of mixfix annotation preference (again!);
wenzelm@3846
   787
paulson@3715
   788
* Pure: fixed idt/idts vs. pttrn/pttrns syntactic categories;
paulson@3715
   789
paulson@3715
   790
paulson@3715
   791
*** HOL ***
paulson@3715
   792
wenzelm@5726
   793
* HOL: there is a new splitter `split_asm_tac' that can be used e.g.
oheimb@4189
   794
  with `addloop' of the simplifier to faciliate case splitting in premises.
oheimb@4189
   795
nipkow@4035
   796
* HOL/TLA: Stephan Merz's formalization of Lamport's Temporal Logic of Actions;
wenzelm@3985
   797
wenzelm@3985
   798
* HOL/Auth: new protocol proofs including some for the Internet
nipkow@4035
   799
  protocol TLS;
wenzelm@3985
   800
wenzelm@4125
   801
* HOL/Map: new theory of `maps' a la VDM;
nipkow@3982
   802
wenzelm@4335
   803
* HOL/simplifier: simplification procedures nat_cancel_sums for
wenzelm@4335
   804
cancelling out common nat summands from =, <, <= (in)equalities, or
wenzelm@4335
   805
differences; simplification procedures nat_cancel_factor for
wenzelm@4335
   806
cancelling common factor from =, <, <= (in)equalities over natural
wenzelm@4373
   807
sums; nat_cancel contains both kinds of procedures, it is installed by
wenzelm@4373
   808
default in Arith.thy -- this COULD MAKE EXISTING PROOFS FAIL;
wenzelm@4335
   809
nipkow@3580
   810
* HOL/simplifier: terms of the form
paulson@4325
   811
  `? x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x)'  (or t=x)
nipkow@3580
   812
  are rewritten to
nipkow@4035
   813
  `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t)',
nipkow@4035
   814
  and those of the form
paulson@4325
   815
  `! x. P1(x) & ... & Pn(x) & x=t & Q1(x) & ... Qn(x) --> R(x)'  (or t=x)
nipkow@4035
   816
  are rewritten to
nipkow@4035
   817
  `P1(t) & ... & Pn(t) & Q1(t) & ... Qn(t) --> R(t)',
nipkow@4035
   818
nipkow@4035
   819
* HOL/datatype
nipkow@4035
   820
  Each datatype `t' now comes with a theorem `split_t_case' of the form
nipkow@3580
   821
nipkow@4035
   822
  P(t_case f1 ... fn x) =
nipkow@4035
   823
     ( (!y1 ... ym1. x = C1 y1 ... ym1 --> P(f1 y1 ... ym1)) &
nipkow@4035
   824
        ...
oheimb@4189
   825
       (!y1 ... ymn. x = Cn y1 ... ymn --> P(f1 y1 ... ymn))
nipkow@4035
   826
     )
nipkow@4035
   827
oheimb@4930
   828
  and a theorem `split_t_case_asm' of the form
oheimb@4189
   829
oheimb@4189
   830
  P(t_case f1 ... fn x) =
oheimb@4189
   831
    ~( (? y1 ... ym1. x = C1 y1 ... ym1 & ~P(f1 y1 ... ym1)) |
oheimb@4189
   832
        ...
oheimb@4189
   833
       (? y1 ... ymn. x = Cn y1 ... ymn & ~P(f1 y1 ... ymn))
oheimb@4189
   834
     )
oheimb@4930
   835
  which can be added to a simpset via `addsplits'. The existing theorems
oheimb@4930
   836
  expand_list_case and expand_option_case have been renamed to
oheimb@4930
   837
  split_list_case and split_option_case.
oheimb@4189
   838
nipkow@4361
   839
* HOL/Arithmetic:
nipkow@4361
   840
  - `pred n' is automatically converted to `n-1'.
nipkow@4361
   841
    Users are strongly encouraged not to use `pred' any longer,
nipkow@4361
   842
    because it will disappear altogether at some point.
nipkow@4361
   843
  - Users are strongly encouraged to write "0 < n" rather than
nipkow@4361
   844
    "n ~= 0". Theorems and proof tools have been modified towards this
nipkow@4361
   845
    `standard'.
nipkow@4357
   846
nipkow@4502
   847
* HOL/Lists:
nipkow@4502
   848
  the function "set_of_list" has been renamed "set" (and its theorems too);
nipkow@4502
   849
  the function "nth" now takes its arguments in the reverse order and
nipkow@4502
   850
  has acquired the infix notation "!" as in "xs!n".
paulson@3570
   851
paulson@4154
   852
* HOL/Set: UNIV is now a constant and is no longer translated to Compl{};
paulson@4154
   853
paulson@4154
   854
* HOL/Set: The operator (UN x.B x) now abbreviates (UN x:UNIV. B x) and its
paulson@4154
   855
  specialist theorems (like UN1_I) are gone.  Similarly for (INT x.B x);
paulson@4154
   856
wenzelm@4575
   857
* HOL/record: extensible records with schematic structural subtyping
wenzelm@4575
   858
(single inheritance); EXPERIMENTAL version demonstrating the encoding,
wenzelm@4575
   859
still lacks various theorems and concrete record syntax;
wenzelm@4575
   860
wenzelm@4125
   861
paulson@3715
   862
*** HOLCF ***
wenzelm@3535
   863
wenzelm@4125
   864
* removed "axioms" and "generated by" sections;
wenzelm@4125
   865
oheimb@4123
   866
* replaced "ops" section by extended "consts" section, which is capable of
wenzelm@4125
   867
  handling the continuous function space "->" directly;
wenzelm@4125
   868
wenzelm@4125
   869
* domain package:
wenzelm@4125
   870
  . proves theorems immediately and stores them in the theory,
wenzelm@4125
   871
  . creates hierachical name space,
wenzelm@4125
   872
  . now uses normal mixfix annotations (instead of cinfix...),
wenzelm@4125
   873
  . minor changes to some names and values (for consistency),
wenzelm@4125
   874
  . e.g. cases -> casedist, dists_eq -> dist_eqs, [take_lemma] -> take_lemmas,
wenzelm@4125
   875
  . separator between mutual domain defs: changed "," to "and",
wenzelm@4125
   876
  . improved handling of sort constraints;  now they have to
wenzelm@4125
   877
    appear on the left-hand side of the equations only;
oheimb@4123
   878
oheimb@4123
   879
* fixed LAM <x,y,zs>.b syntax;
wenzelm@3567
   880
wenzelm@3744
   881
* added extended adm_tac to simplifier in HOLCF -- can now discharge
wenzelm@3744
   882
adm (%x. P (t x)), where P is chainfinite and t continuous;
wenzelm@3579
   883
wenzelm@3579
   884
paulson@3719
   885
*** FOL and ZF ***
paulson@3719
   886
wenzelm@5726
   887
* FOL: there is a new splitter `split_asm_tac' that can be used e.g.
oheimb@4189
   888
  with `addloop' of the simplifier to faciliate case splitting in premises.
oheimb@4189
   889
wenzelm@3744
   890
* qed_spec_mp, qed_goal_spec_mp, qed_goalw_spec_mp are available, as
wenzelm@3744
   891
in HOL, they strip ALL and --> from proved theorems;
wenzelm@3744
   892
paulson@3719
   893
wenzelm@3579
   894
wenzelm@3006
   895
New in Isabelle94-8 (May 1997)
wenzelm@3006
   896
------------------------------
wenzelm@2654
   897
paulson@3002
   898
*** General Changes ***
paulson@3002
   899
paulson@3002
   900
* new utilities to build / run / maintain Isabelle etc. (in parts
paulson@3002
   901
still somewhat experimental); old Makefiles etc. still functional;
wenzelm@2971
   902
wenzelm@3205
   903
* new 'Isabelle System Manual';
wenzelm@3205
   904
wenzelm@2825
   905
* INSTALL text, together with ./configure and ./build scripts;
wenzelm@2773
   906
wenzelm@3006
   907
* reimplemented type inference for greater efficiency, better error
wenzelm@3006
   908
messages and clean internal interface;
paulson@3002
   909
paulson@3002
   910
* prlim command for dealing with lots of subgoals (an easier way of
paulson@3002
   911
setting goals_limit);
paulson@3002
   912
wenzelm@3006
   913
wenzelm@3006
   914
*** Syntax ***
paulson@3002
   915
wenzelm@3116
   916
* supports alternative (named) syntax tables (parser and pretty
wenzelm@3116
   917
printer); internal interface is provided by add_modesyntax(_i);
wenzelm@3116
   918
paulson@3002
   919
* Pure, FOL, ZF, HOL, HOLCF now support symbolic input and output; to
paulson@3002
   920
be used in conjunction with the Isabelle symbol font; uses the
paulson@3002
   921
"symbols" syntax table;
paulson@3002
   922
wenzelm@2705
   923
* added token_translation interface (may translate name tokens in
wenzelm@2756
   924
arbitrary ways, dependent on their type (free, bound, tfree, ...) and
wenzelm@3116
   925
the current print_mode); IMPORTANT: user print translation functions
wenzelm@3116
   926
are responsible for marking newly introduced bounds
wenzelm@3116
   927
(Syntax.mark_boundT);
wenzelm@2705
   928
wenzelm@2730
   929
* token translations for modes "xterm" and "xterm_color" that display
wenzelm@3006
   930
names in bold, underline etc. or colors (which requires a color
wenzelm@3006
   931
version of xterm);
wenzelm@2730
   932
paulson@3002
   933
* infixes may now be declared with names independent of their syntax;
paulson@3002
   934
paulson@3002
   935
* added typed_print_translation (like print_translation, but may
paulson@3002
   936
access type of constant);
paulson@3002
   937
wenzelm@3006
   938
paulson@3002
   939
*** Classical Reasoner ***
paulson@3002
   940
paulson@3002
   941
Blast_tac: a new tactic!  It is often more powerful than fast_tac, but has
paulson@3002
   942
some limitations.  Blast_tac...
paulson@3002
   943
  + ignores addss, addbefore, addafter; this restriction is intrinsic
paulson@3002
   944
  + ignores elimination rules that don't have the correct format
wenzelm@5726
   945
        (the conclusion MUST be a formula variable)
paulson@3002
   946
  + ignores types, which can make HOL proofs fail
paulson@3002
   947
  + rules must not require higher-order unification, e.g. apply_type in ZF
paulson@3002
   948
    [message "Function Var's argument not a bound variable" relates to this]
paulson@3002
   949
  + its proof strategy is more general but can actually be slower
paulson@3002
   950
wenzelm@3107
   951
* substitution with equality assumptions no longer permutes other
wenzelm@3107
   952
assumptions;
paulson@3002
   953
paulson@3002
   954
* minor changes in semantics of addafter (now called addaltern); renamed
paulson@3002
   955
setwrapper to setWrapper and compwrapper to compWrapper; added safe wrapper
wenzelm@3107
   956
(and access functions for it);
paulson@3002
   957
wenzelm@5726
   958
* improved combination of classical reasoner and simplifier:
oheimb@3317
   959
  + functions for handling clasimpsets
oheimb@3317
   960
  + improvement of addss: now the simplifier is called _after_ the
oheimb@3317
   961
    safe steps.
oheimb@3317
   962
  + safe variant of addss called addSss: uses safe simplifications
wenzelm@5726
   963
    _during_ the safe steps. It is more complete as it allows multiple
oheimb@3317
   964
    instantiations of unknowns (e.g. with slow_tac).
wenzelm@3006
   965
paulson@3002
   966
*** Simplifier ***
paulson@3002
   967
wenzelm@3006
   968
* added interface for simplification procedures (functions that
wenzelm@3006
   969
produce *proven* rewrite rules on the fly, depending on current
wenzelm@3006
   970
redex);
wenzelm@3006
   971
wenzelm@3006
   972
* ordering on terms as parameter (used for ordered rewriting);
wenzelm@3006
   973
wenzelm@3107
   974
* new functions delcongs, deleqcongs, and Delcongs. richer rep_ss;
paulson@3002
   975
paulson@3002
   976
* the solver is now split into a safe and an unsafe part.
paulson@3002
   977
This should be invisible for the normal user, except that the
paulson@3002
   978
functions setsolver and addsolver have been renamed to setSolver and
wenzelm@3107
   979
addSolver; added safe_asm_full_simp_tac;
paulson@3002
   980
paulson@3002
   981
paulson@3002
   982
*** HOL ***
paulson@3002
   983
nipkow@3042
   984
* a generic induction tactic `induct_tac' which works for all datatypes and
wenzelm@3107
   985
also for type `nat';
nipkow@3042
   986
nipkow@3316
   987
* a generic case distinction tactic `exhaust_tac' which works for all
nipkow@3316
   988
datatypes and also for type `nat';
nipkow@3316
   989
nipkow@3316
   990
* each datatype comes with a function `size';
nipkow@3316
   991
paulson@3002
   992
* patterns in case expressions allow tuple patterns as arguments to
wenzelm@3107
   993
constructors, for example `case x of [] => ... | (x,y,z)#ps => ...';
paulson@3002
   994
paulson@3002
   995
* primrec now also works with type nat;
paulson@3002
   996
paulson@3338
   997
* recdef: a new declaration form, allows general recursive functions to be
paulson@3338
   998
defined in theory files.  See HOL/ex/Fib, HOL/ex/Primes, HOL/Subst/Unify.
paulson@3338
   999
paulson@3002
  1000
* the constant for negation has been renamed from "not" to "Not" to
wenzelm@3107
  1001
harmonize with FOL, ZF, LK, etc.;
paulson@3002
  1002
wenzelm@3107
  1003
* HOL/ex/LFilter theory of a corecursive "filter" functional for
wenzelm@3107
  1004
infinite lists;
paulson@3002
  1005
wenzelm@3227
  1006
* HOL/Modelcheck demonstrates invocation of model checker oracle;
wenzelm@3227
  1007
paulson@3002
  1008
* HOL/ex/Ring.thy declares cring_simp, which solves equational
paulson@3002
  1009
problems in commutative rings, using axiomatic type classes for + and *;
paulson@3002
  1010
paulson@3002
  1011
* more examples in HOL/MiniML and HOL/Auth;
paulson@3002
  1012
paulson@3002
  1013
* more default rewrite rules for quantifiers, union/intersection;
paulson@3002
  1014
nipkow@3321
  1015
* a new constant `arbitrary == @x.False';
nipkow@3321
  1016
wenzelm@3107
  1017
* HOLCF/IOA replaces old HOL/IOA;
wenzelm@3107
  1018
wenzelm@5726
  1019
* HOLCF changes: derived all rules and arities
wenzelm@5726
  1020
  + axiomatic type classes instead of classes
slotosch@2653
  1021
  + typedef instead of faking type definitions
paulson@2747
  1022
  + eliminated the internal constants less_fun, less_cfun, UU_fun, UU_cfun etc.
wenzelm@2730
  1023
  + new axclasses cpo, chfin, flat with flat < chfin < pcpo < cpo < po
slotosch@2653
  1024
  + eliminated the types void, one, tr
slotosch@2653
  1025
  + use unit lift and bool lift (with translations) instead of one and tr
slotosch@2653
  1026
  + eliminated blift from Lift3.thy (use Def instead of blift)
wenzelm@3107
  1027
  all eliminated rules are derived as theorems --> no visible changes ;
oheimb@2649
  1028
wenzelm@3006
  1029
paulson@3002
  1030
*** ZF ***
wenzelm@2553
  1031
paulson@2865
  1032
* ZF now has Fast_tac, Simp_tac and Auto_tac.  Union_iff is a now a default
paulson@2865
  1033
rewrite rule; this may affect some proofs.  eq_cs is gone but can be put back
paulson@2865
  1034
as ZF_cs addSIs [equalityI];
wenzelm@2553
  1035
wenzelm@2554
  1036
wenzelm@2732
  1037
wenzelm@2553
  1038
New in Isabelle94-7 (November 96)
wenzelm@2553
  1039
---------------------------------
wenzelm@2553
  1040
wenzelm@2553
  1041
* allowing negative levels (as offsets) in prlev and choplev;
wenzelm@2553
  1042
wenzelm@2554
  1043
* super-linear speedup for large simplifications;
wenzelm@2554
  1044
wenzelm@2554
  1045
* FOL, ZF and HOL now use miniscoping: rewriting pushes
wenzelm@2554
  1046
quantifications in as far as possible (COULD MAKE EXISTING PROOFS
wenzelm@2554
  1047
FAIL); can suppress it using the command Delsimps (ex_simps @
wenzelm@2554
  1048
all_simps); De Morgan laws are also now included, by default;
wenzelm@2554
  1049
wenzelm@2554
  1050
* improved printing of ==>  :  ~:
wenzelm@2554
  1051
wenzelm@2554
  1052
* new object-logic "Sequents" adds linear logic, while replacing LK
wenzelm@2554
  1053
and Modal (thanks to Sara Kalvala);
wenzelm@2554
  1054
wenzelm@2554
  1055
* HOL/Auth: correctness proofs for authentication protocols;
wenzelm@2554
  1056
wenzelm@2554
  1057
* HOL: new auto_tac combines rewriting and classical reasoning (many
wenzelm@2554
  1058
examples on HOL/Auth);
wenzelm@2554
  1059
wenzelm@2554
  1060
* HOL: new command AddIffs for declaring theorems of the form P=Q to
wenzelm@2554
  1061
the rewriter and classical reasoner simultaneously;
wenzelm@2554
  1062
wenzelm@2554
  1063
* function uresult no longer returns theorems in "standard" format;
wenzelm@2554
  1064
regain previous version by: val uresult = standard o uresult;
wenzelm@2554
  1065
wenzelm@2554
  1066
wenzelm@2554
  1067
wenzelm@2554
  1068
New in Isabelle94-6
wenzelm@2554
  1069
-------------------
wenzelm@2554
  1070
wenzelm@2554
  1071
* oracles -- these establish an interface between Isabelle and trusted
wenzelm@2554
  1072
external reasoners, which may deliver results as theorems;
wenzelm@2554
  1073
wenzelm@2554
  1074
* proof objects (in particular record all uses of oracles);
wenzelm@2554
  1075
wenzelm@2554
  1076
* Simp_tac, Fast_tac, etc. that refer to implicit simpset / claset;
wenzelm@2554
  1077
wenzelm@2554
  1078
* "constdefs" section in theory files;
wenzelm@2554
  1079
wenzelm@2554
  1080
* "primrec" section (HOL) no longer requires names;
wenzelm@2554
  1081
wenzelm@2554
  1082
* internal type "tactic" now simply "thm -> thm Sequence.seq";
wenzelm@2554
  1083
wenzelm@2554
  1084
wenzelm@2554
  1085
wenzelm@2554
  1086
New in Isabelle94-5
wenzelm@2554
  1087
-------------------
wenzelm@2554
  1088
wenzelm@2554
  1089
* reduced space requirements;
wenzelm@2554
  1090
wenzelm@2554
  1091
* automatic HTML generation from theories;
wenzelm@2554
  1092
wenzelm@2554
  1093
* theory files no longer require "..." (quotes) around most types;
wenzelm@2554
  1094
wenzelm@2554
  1095
* new examples, including two proofs of the Church-Rosser theorem;
wenzelm@2554
  1096
wenzelm@2554
  1097
* non-curried (1994) version of HOL is no longer distributed;
wenzelm@2554
  1098
wenzelm@2553
  1099
paulson@2557
  1100
paulson@2557
  1101
New in Isabelle94-4
paulson@2557
  1102
-------------------
paulson@2557
  1103
paulson@2747
  1104
* greatly reduced space requirements;
paulson@2557
  1105
paulson@2557
  1106
* theory files (.thy) no longer require \...\ escapes at line breaks;
paulson@2557
  1107
wenzelm@5726
  1108
* searchable theorem database (see the section "Retrieving theorems" on
paulson@2557
  1109
page 8 of the Reference Manual);
paulson@2557
  1110
paulson@2557
  1111
* new examples, including Grabczewski's monumental case study of the
paulson@2557
  1112
Axiom of Choice;
paulson@2557
  1113
paulson@2557
  1114
* The previous version of HOL renamed to Old_HOL;
paulson@2557
  1115
wenzelm@5726
  1116
* The new version of HOL (previously called CHOL) uses a curried syntax
paulson@2557
  1117
for functions.  Application looks like f a b instead of f(a,b);
paulson@2557
  1118
paulson@2557
  1119
* Mutually recursive inductive definitions finally work in HOL;
paulson@2557
  1120
paulson@2557
  1121
* In ZF, pattern-matching on tuples is now available in all abstractions and
paulson@2557
  1122
translates to the operator "split";
paulson@2557
  1123
paulson@2557
  1124
paulson@2557
  1125
paulson@2557
  1126
New in Isabelle94-3
paulson@2557
  1127
-------------------
paulson@2557
  1128
wenzelm@5726
  1129
* new infix operator, addss, allowing the classical reasoner to
paulson@2557
  1130
perform simplification at each step of its search.  Example:
wenzelm@5726
  1131
        fast_tac (cs addss ss)
paulson@2557
  1132
wenzelm@5726
  1133
* a new logic, CHOL, the same as HOL, but with a curried syntax
wenzelm@5726
  1134
for functions.  Application looks like f a b instead of f(a,b).  Also pairs
paulson@2557
  1135
look like (a,b) instead of <a,b>;
paulson@2557
  1136
paulson@2557
  1137
* PLEASE NOTE: CHOL will eventually replace HOL!
paulson@2557
  1138
paulson@2557
  1139
* In CHOL, pattern-matching on tuples is now available in all abstractions.
paulson@2557
  1140
It translates to the operator "split".  A new theory of integers is available;
paulson@2557
  1141
paulson@2557
  1142
* In ZF, integer numerals now denote two's-complement binary integers.
paulson@2557
  1143
Arithmetic operations can be performed by rewriting.  See ZF/ex/Bin.ML;
paulson@2557
  1144
wenzelm@5726
  1145
* Many new examples: I/O automata, Church-Rosser theorem, equivalents
paulson@2557
  1146
of the Axiom of Choice;
paulson@2557
  1147
paulson@2557
  1148
paulson@2557
  1149
paulson@2557
  1150
New in Isabelle94-2
paulson@2557
  1151
-------------------
paulson@2557
  1152
wenzelm@5726
  1153
* Significantly faster resolution;
paulson@2557
  1154
paulson@2557
  1155
* the different sections in a .thy file can now be mixed and repeated
paulson@2557
  1156
freely;
paulson@2557
  1157
paulson@2557
  1158
* Database of theorems for FOL, HOL and ZF.  New
paulson@2557
  1159
commands including qed, qed_goal and bind_thm store theorems in the database.
paulson@2557
  1160
paulson@2557
  1161
* Simple database queries: return a named theorem (get_thm) or all theorems of
paulson@2557
  1162
a given theory (thms_of), or find out what theory a theorem was proved in
paulson@2557
  1163
(theory_of_thm);
paulson@2557
  1164
paulson@2557
  1165
* Bugs fixed in the inductive definition and datatype packages;
paulson@2557
  1166
paulson@2557
  1167
* The classical reasoner provides deepen_tac and depth_tac, making FOL_dup_cs
paulson@2557
  1168
and HOL_dup_cs obsolete;
paulson@2557
  1169
paulson@2557
  1170
* Syntactic ambiguities caused by the new treatment of syntax in Isabelle94-1
paulson@2557
  1171
have been removed;
paulson@2557
  1172
paulson@2557
  1173
* Simpler definition of function space in ZF;
paulson@2557
  1174
paulson@2557
  1175
* new results about cardinal and ordinal arithmetic in ZF;
paulson@2557
  1176
paulson@2557
  1177
* 'subtype' facility in HOL for introducing new types as subsets of existing
paulson@2557
  1178
types;
paulson@2557
  1179
paulson@2557
  1180
wenzelm@2553
  1181
$Id$