NEWS
author wenzelm
Fri Mar 06 15:58:56 2015 +0100 (2015-03-06)
changeset 59621 291934bac95e
parent 59588 c6f3dbe466b6
child 59648 d1148f0baef5
permissions -rw-r--r--
Thm.cterm_of and Thm.ctyp_of operate on local context;
wenzelm@57491
     1
Isabelle NEWS -- history of user-relevant changes
wenzelm@57491
     2
=================================================
wenzelm@2553
     3
wenzelm@57695
     4
New in this Isabelle version
wenzelm@57695
     5
----------------------------
wenzelm@57695
     6
wenzelm@57941
     7
*** General ***
wenzelm@57941
     8
wenzelm@59569
     9
* Commands 'method_setup' and 'attribute_setup' now work within a local
wenzelm@59569
    10
theory context.
wenzelm@57941
    11
wenzelm@57946
    12
* Command 'named_theorems' declares a dynamic fact within the context,
wenzelm@59569
    13
together with an attribute to maintain the content incrementally. This
wenzelm@59569
    14
supersedes functor Named_Thms, but with a subtle change of semantics due
wenzelm@59569
    15
to external visual order vs. internal reverse order.
wenzelm@57946
    16
wenzelm@58801
    17
* Command 'notepad' requires proper nesting of begin/end and its proof
wenzelm@58801
    18
structure in the body: 'oops' is no longer supported here. Minor
wenzelm@58801
    19
INCOMPATIBILITY, use 'sorry' instead.
wenzelm@58801
    20
wenzelm@58928
    21
* Outer syntax commands are managed authentically within the theory
wenzelm@59569
    22
context, without implicit global state. Potential for accidental
wenzelm@58928
    23
INCOMPATIBILITY, make sure that required theories are really imported.
wenzelm@58928
    24
haftmann@59105
    25
* 'find_theorems': search patterns which are abstractions are
haftmann@59570
    26
schematically expanded before search. Search results match the naive
wenzelm@59569
    27
expectation more closely, particularly wrt. abbreviations.
haftmann@59105
    28
INCOMPATIBILITY.
haftmann@59105
    29
wenzelm@57941
    30
wenzelm@58524
    31
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@58524
    32
wenzelm@59308
    33
* Old graph browser (Java/AWT 1.0) is superseded by improved graphview
wenzelm@59308
    34
panel, which also includes PDF output.
wenzelm@59308
    35
wenzelm@59569
    36
* Improved folding mode "isabelle" based on Isar syntax. Alternatively,
wenzelm@59569
    37
the "sidekick" mode may be used for document structure.
wenzelm@59569
    38
wenzelm@59569
    39
* Extended bracket matching based on Isar language structure. System
wenzelm@59569
    40
option jedit_structure_limit determines maximum number of lines to scan
wenzelm@59569
    41
in the buffer.
wenzelm@58758
    42
wenzelm@58540
    43
* Support for BibTeX files: context menu, context-sensitive token
wenzelm@58540
    44
marker, SideKick parser.
wenzelm@58524
    45
wenzelm@58551
    46
* Document antiquotation @{cite} provides formal markup, which is
wenzelm@58551
    47
interpreted semi-formally based on .bib files that happen to be opened
wenzelm@58592
    48
in the editor (hyperlinks, completion etc.).
wenzelm@58551
    49
wenzelm@58785
    50
* Less waste of vertical space via negative line spacing (see Global
wenzelm@58785
    51
Options / Text Area).
wenzelm@58785
    52
wenzelm@58524
    53
haftmann@58202
    54
*** Pure ***
haftmann@58202
    55
haftmann@58202
    56
* Command "class_deps" takes optional sort arguments constraining
haftmann@58202
    57
the search space.
haftmann@58202
    58
haftmann@58410
    59
* Lexical separation of signed and unsigend numerals: categories "num"
haftmann@58410
    60
and "float" are unsigend.  INCOMPATIBILITY: subtle change in precedence
haftmann@58410
    61
of numeral signs, particulary in expressions involving infix syntax like
haftmann@58410
    62
"(- 1) ^ n".
haftmann@58410
    63
wenzelm@58421
    64
* Old inner token category "xnum" has been discontinued.  Potential
wenzelm@58421
    65
INCOMPATIBILITY for exotic syntax: may use mixfix grammar with "num"
wenzelm@58421
    66
token category instead.
wenzelm@58421
    67
haftmann@58202
    68
blanchet@57737
    69
*** HOL ***
blanchet@57737
    70
nipkow@59587
    71
* removed functions "natfloor" and "natceiling",
nipkow@59587
    72
  use "nat o floor" and "nat o ceiling" instead.
nipkow@59587
    73
  A few of the lemmas have been retained and adapted: in their names
nipkow@59587
    74
  "natfloor"/"natceiling" has been replaced by "nat_floor"/"nat_ceiling".
nipkow@59587
    75
haftmann@59557
    76
* Qualified some duplicated fact names required for boostrapping
haftmann@59557
    77
the type class hierarchy:
haftmann@59557
    78
  ab_add_uminus_conv_diff ~> diff_conv_add_uminus
haftmann@59557
    79
  field_inverse_zero ~> inverse_zero
haftmann@59557
    80
  field_divide_inverse ~> divide_inverse
haftmann@59557
    81
  field_inverse ~> left_inverse
haftmann@59557
    82
Minor INCOMPATIBILITY.
haftmann@59557
    83
haftmann@59555
    84
* Eliminated fact duplicates:
haftmann@59555
    85
  mult_less_imp_less_right ~> mult_right_less_imp_less
haftmann@59555
    86
  mult_less_imp_less_left ~> mult_left_less_imp_less
haftmann@59555
    87
Minor INCOMPATIBILITY.
haftmann@59555
    88
haftmann@59536
    89
* Fact consolidation: even_less_0_iff is subsumed by
haftmann@59536
    90
double_add_less_zero_iff_single_add_less_zero (simp by default anyway).
haftmann@59536
    91
wenzelm@59569
    92
* Discontinued old-fashioned "codegen" tool. Code generation can always
wenzelm@59569
    93
be externally triggered using an appropriate ROOT file plus a
wenzelm@59569
    94
corresponding theory. Parametrization is possible using environment
wenzelm@59569
    95
variables, or ML snippets in the most extreme cases. Minor
wenzelm@59569
    96
INCOMPATIBILITY.
haftmann@59480
    97
hoelzl@58775
    98
* Add NO_MATCH-simproc, allows to check for syntactic non-equality
hoelzl@58775
    99
haftmann@58649
   100
* Generalized and consolidated some theorems concerning divsibility:
haftmann@58649
   101
  dvd_reduce ~> dvd_add_triv_right_iff
haftmann@58649
   102
  dvd_plus_eq_right ~> dvd_add_right_iff
haftmann@58649
   103
  dvd_plus_eq_left ~> dvd_add_left_iff
haftmann@58649
   104
Minor INCOMPATIBILITY.
haftmann@58649
   105
haftmann@58770
   106
* "even" and "odd" are mere abbreviations for "2 dvd _" and "~ 2 dvd _"
haftmann@58770
   107
and part of HOL-Main.
haftmann@58645
   108
  even_def ~> even_iff_mod_2_eq_zero
haftmann@58740
   109
INCOMPATIBILITY.
haftmann@58645
   110
haftmann@58512
   111
* Lemma name consolidation: divide_Numeral1 ~> divide_numeral_1
haftmann@58512
   112
Minor INCOMPATIBILITY.
haftmann@58512
   113
hoelzl@58776
   114
* field_simps: Use NO_MATCH-simproc for distribution rules, to avoid
hoelzl@58776
   115
  non-termination in case of distributing a division. With this change
hoelzl@58776
   116
  field_simps is in some cases slightly less powerful, if it fails try
hoelzl@58776
   117
  to add algebra_simps, or use divide_simps.
hoelzl@58776
   118
Minor INCOMPATIBILITY.
hoelzl@58776
   119
haftmann@58321
   120
* Bootstrap of listsum as special case of abstract product over lists.
haftmann@58321
   121
Fact rename:
haftmann@58321
   122
    listsum_def ~> listsum.eq_foldr
haftmann@58321
   123
INCOMPATIBILITY.
haftmann@58321
   124
wenzelm@59569
   125
* Command and antiquotation "value" provide different evaluation slots
wenzelm@59569
   126
(again), where the previous strategy (nbe after ML) serves as default.
haftmann@58100
   127
Minor INCOMPATIBILITY.
haftmann@58100
   128
blanchet@57983
   129
* New (co)datatype package:
blanchet@58373
   130
  - The 'datatype_new' command has been renamed 'datatype'. The old
blanchet@58373
   131
    command of that name is now called 'old_datatype' and is provided
blanchet@58373
   132
    by "~~/src/HOL/Library/Old_Datatype.thy". See
blanchet@58373
   133
    'isabelle doc datatypes' for information on porting.
blanchet@58373
   134
    INCOMPATIBILITY.
blanchet@57983
   135
  - Renamed theorems:
blanchet@57983
   136
      disc_corec ~> corec_disc
blanchet@57983
   137
      disc_corec_iff ~> corec_disc_iff
blanchet@57983
   138
      disc_exclude ~> distinct_disc
blanchet@57983
   139
      disc_exhaust ~> exhaust_disc
blanchet@57983
   140
      disc_map_iff ~> map_disc_iff
blanchet@57983
   141
      sel_corec ~> corec_sel
blanchet@57983
   142
      sel_exhaust ~> exhaust_sel
blanchet@57983
   143
      sel_map ~> map_sel
blanchet@57983
   144
      sel_set ~> set_sel
blanchet@57983
   145
      sel_split ~> split_sel
blanchet@57983
   146
      sel_split_asm ~> split_sel_asm
blanchet@57983
   147
      strong_coinduct ~> coinduct_strong
blanchet@57983
   148
      weak_case_cong ~> case_cong_weak
blanchet@57983
   149
    INCOMPATIBILITY.
blanchet@58192
   150
  - The "no_code" option to "free_constructors", "datatype_new", and
blanchet@58192
   151
    "codatatype" has been renamed "plugins del: code".
blanchet@58192
   152
    INCOMPATIBILITY.
blanchet@58044
   153
  - The rules "set_empty" have been removed. They are easy
blanchet@58044
   154
    consequences of other set rules "by auto".
blanchet@58044
   155
    INCOMPATIBILITY.
blanchet@58044
   156
  - The rule "set_cases" is now registered with the "[cases set]"
blanchet@57990
   157
    attribute. This can influence the behavior of the "cases" proof
blanchet@57990
   158
    method when more than one case rule is applicable (e.g., an
blanchet@57990
   159
    assumption is of the form "w : set ws" and the method "cases w"
blanchet@57990
   160
    is invoked). The solution is to specify the case rule explicitly
blanchet@57990
   161
    (e.g. "cases w rule: widget.exhaust").
blanchet@57990
   162
    INCOMPATIBILITY.
blanchet@57983
   163
blanchet@57983
   164
* Old datatype package:
blanchet@58310
   165
  - The old 'datatype' command has been renamed 'old_datatype', and
blanchet@58373
   166
    'rep_datatype' has been renamed 'old_rep_datatype'. They are
blanchet@58373
   167
    provided by "~~/src/HOL/Library/Old_Datatype.thy". See
blanchet@58310
   168
    'isabelle doc datatypes' for information on porting.
blanchet@58373
   169
    INCOMPATIBILITY.
blanchet@57983
   170
  - Renamed theorems:
blanchet@57983
   171
      weak_case_cong ~> case_cong_weak
blanchet@57983
   172
    INCOMPATIBILITY.
blanchet@58373
   173
  - Renamed theory:
blanchet@58373
   174
      ~~/src/HOL/Datatype.thy ~> ~~/src/HOL/Library/Old_Datatype.thy
blanchet@58373
   175
    INCOMPATIBILITY.
blanchet@57983
   176
haftmann@58368
   177
* Product over lists via constant "listprod".
haftmann@58368
   178
blanchet@59039
   179
* Nitpick:
blanchet@59039
   180
  - Fixed soundness bug related to the strict and nonstrict subset
blanchet@59039
   181
    operations.
blanchet@59039
   182
blanchet@57737
   183
* Sledgehammer:
blanchet@59511
   184
  - CVC4 is now included with Isabelle instead of CVC3 and run by
blanchet@59511
   185
    default.
blanchet@57737
   186
  - Minimization is now always enabled by default.
blanchet@57737
   187
    Removed subcommand:
blanchet@57737
   188
      min
blanchet@59039
   189
  - The proof reconstruction, both one-liners and Isar, has been
blanchet@59039
   190
    dramatically improved.
blanchet@59039
   191
  - Improved support for CVC4 and veriT.
blanchet@57737
   192
blanchet@58062
   193
* Old and new SMT modules:
blanchet@58067
   194
  - The old 'smt' method has been renamed 'old_smt' and moved to
wenzelm@59569
   195
    'src/HOL/Library/Old_SMT.thy'. It is provided for compatibility,
wenzelm@59569
   196
    until applications have been ported to use the new 'smt' method. For
wenzelm@59569
   197
    the method to work, an older version of Z3 (e.g. Z3 3.2 or 4.0) must
wenzelm@59569
   198
    be installed, and the environment variable "OLD_Z3_SOLVER" must
wenzelm@59569
   199
    point to it.
blanchet@58062
   200
    INCOMPATIBILITY.
blanchet@58067
   201
  - The 'smt2' method has been renamed 'smt'.
blanchet@58060
   202
    INCOMPATIBILITY.
wenzelm@59569
   203
  - New option 'smt_reconstruction_step_timeout' to limit the
wenzelm@59569
   204
    reconstruction time of Z3 proof steps in the new 'smt' method.
boehmes@59216
   205
  - New option 'smt_statistics' to display statistics of the new 'smt'
boehmes@59216
   206
    method, especially runtime statistics of Z3 proof reconstruction.
blanchet@58060
   207
nipkow@58247
   208
* List: renamed drop_Suc_conv_tl and nth_drop' to Cons_nth_drop_Suc
nipkow@58247
   209
Andreas@58626
   210
* New infrastructure for compiling, running, evaluating and testing
wenzelm@59569
   211
generated code in target languages in HOL/Library/Code_Test. See
wenzelm@59569
   212
HOL/Codegenerator_Test/Code_Test* for examples.
wenzelm@58008
   213
wenzelm@58630
   214
* Library/Sum_of_Squares: simplified and improved "sos" method. Always
wenzelm@58630
   215
use local CSDP executable, which is much faster than the NEOS server.
wenzelm@58630
   216
The "sos_cert" functionality is invoked as "sos" with additional
wenzelm@58630
   217
argument. Minor INCOMPATIBILITY.
wenzelm@58630
   218
immler@58990
   219
* HOL-Decision_Procs:
immler@58990
   220
  - New counterexample generator quickcheck[approximation] for
immler@58990
   221
    inequalities of transcendental functions.
immler@58990
   222
    Uses hardware floating point arithmetic to randomly discover
immler@58990
   223
    potential counterexamples. Counterexamples are certified with the
immler@58990
   224
    'approximation' method.
immler@58990
   225
    See HOL/Decision_Procs/ex/Approximation_Quickcheck_Ex.thy for
immler@58990
   226
    examples.
immler@58990
   227
hoelzl@59354
   228
* HOL-Probability: Reworked measurability prover
hoelzl@59354
   229
  - applies destructor rules repeatetly
hoelzl@59354
   230
  - removed application splitting (replaced by destructor rule)
wenzelm@59569
   231
  - added congruence rules to rewrite measure spaces under the sets
wenzelm@59569
   232
    projection
wenzelm@59569
   233
wenzelm@58630
   234
wenzelm@58716
   235
*** Document preparation ***
wenzelm@58716
   236
wenzelm@59446
   237
* Discontinued obsolete option "document_graph": session_graph.pdf is
wenzelm@59450
   238
produced unconditionally for HTML browser_info and PDF-LaTeX document.
wenzelm@59446
   239
wenzelm@58999
   240
* Document markup commands 'chapter', 'section', 'subsection',
wenzelm@58999
   241
'subsubsection', 'text', 'txt', 'text_raw' work uniformly in any
wenzelm@58999
   242
context, even before the initial 'theory' command. Obsolete proof
wenzelm@58999
   243
commands 'sect', 'subsect', 'subsubsect', 'txt_raw' have been
wenzelm@58999
   244
discontinued, use 'section', 'subsection', 'subsubsection', 'text_raw'
wenzelm@58999
   245
instead. The old 'header' command is still retained for some time, but
wenzelm@58999
   246
should be replaced by 'chapter', 'section' etc. (using "isabelle
wenzelm@58999
   247
update_header"). Minor INCOMPATIBILITY.
wenzelm@58999
   248
wenzelm@58999
   249
* Diagnostic commands and document markup commands within a proof do not
wenzelm@58999
   250
affect the command tag for output. Thus commands like 'thm' are subject
wenzelm@58999
   251
to proof document structure, and no longer "stick out" accidentally.
wenzelm@58999
   252
Commands 'text' and 'txt' merely differ in the LaTeX style, not their
wenzelm@58999
   253
tags. Potential INCOMPATIBILITY in exotic situations.
wenzelm@58868
   254
wenzelm@58716
   255
* Official support for "tt" style variants, via \isatt{...} or
wenzelm@58716
   256
\begin{isabellett}...\end{isabellett}. The somewhat fragile \verb or
wenzelm@58716
   257
verbatim environment of LaTeX is no longer used. This allows @{ML} etc.
wenzelm@58716
   258
as argument to other macros (such as footnotes).
wenzelm@58716
   259
wenzelm@58716
   260
* Document antiquotation @{verbatim} prints ASCII text literally in "tt"
wenzelm@58716
   261
style.
wenzelm@58716
   262
wenzelm@58716
   263
blanchet@58066
   264
*** ML ***
blanchet@58066
   265
wenzelm@59621
   266
* The main operations to certify logical entities are Thm.ctyp_of and
wenzelm@59621
   267
Thm.cterm_of with a local context; old-style global theory variants are
wenzelm@59621
   268
available as Thm.global_ctyp_of and Thm.global_cterm_of.
wenzelm@59621
   269
INCOMPATIBILITY.
wenzelm@59621
   270
wenzelm@59582
   271
* Elementary operations in module Thm are no longer pervasive.
wenzelm@59582
   272
INCOMPATIBILITY, need to use qualified Thm.prop_of, Thm.cterm_of,
wenzelm@59582
   273
Thm.term_of etc.
wenzelm@59582
   274
wenzelm@59180
   275
* Former combinators NAMED_CRITICAL and CRITICAL for central critical
wenzelm@59180
   276
sections have been discontinued, in favour of the more elementary
wenzelm@59180
   277
Multithreading.synchronized and its high-level derivative
wenzelm@59180
   278
Synchronized.var (which is usually sufficient in applications). Subtle
wenzelm@59180
   279
INCOMPATIBILITY: synchronized access needs to be atomic and cannot be
wenzelm@59180
   280
nested.
wenzelm@59180
   281
wenzelm@59112
   282
* Cartouches within ML sources are turned into values of type
wenzelm@59112
   283
Input.source (with formal position information).
wenzelm@59112
   284
wenzelm@58963
   285
* Proper context for various elementary tactics: assume_tac,
wenzelm@59498
   286
resolve_tac, eresolve_tac, dresolve_tac, forward_tac, match_tac,
wenzelm@59498
   287
compose_tac, Splitter.split_tac etc. INCOMPATIBILITY.
wenzelm@58956
   288
blanchet@58066
   289
* Tactical PARALLEL_ALLGOALS is the most common way to refer to
blanchet@58066
   290
PARALLEL_GOALS.
blanchet@58066
   291
wenzelm@59057
   292
* Basic combinators map, fold, fold_map, split_list, apply are
wenzelm@59057
   293
available as parameterized antiquotations, e.g. @{map 4} for lists of
wenzelm@59057
   294
quadruples.
wenzelm@58634
   295
wenzelm@59058
   296
* Renamed "pairself" to "apply2", in accordance to @{apply 2}.
wenzelm@59058
   297
INCOMPATIBILITY.
wenzelm@59058
   298
wenzelm@59139
   299
* Synchronized.value (ML) is actually synchronized (as in Scala):
wenzelm@59139
   300
subtle change of semantics with minimal potential for INCOMPATIBILITY.
wenzelm@59139
   301
wenzelm@59564
   302
* Goal.prove_multi is superseded by the fully general Goal.prove_common,
wenzelm@59564
   303
which also allows to specify a fork priority.
wenzelm@59564
   304
blanchet@58066
   305
wenzelm@58610
   306
*** System ***
wenzelm@58610
   307
wenzelm@58846
   308
* Support for Proof General and Isar TTY loop has been discontinued.
wenzelm@58846
   309
Minor INCOMPATIBILITY.
wenzelm@58842
   310
wenzelm@59200
   311
* JVM system property "isabelle.threads" determines size of Scala thread
wenzelm@59200
   312
pool, like Isabelle system option "threads" for ML.
wenzelm@59200
   313
wenzelm@59201
   314
* JVM system property "isabelle.laf" determines the default Swing
wenzelm@59201
   315
look-and-feel, via internal class name or symbolic name as in the jEdit
wenzelm@59201
   316
menu Global Options / Appearance.
wenzelm@59201
   317
wenzelm@59175
   318
* System option "pretty_margin" is superseded by "thy_output_margin",
wenzelm@59175
   319
which is also accessible via document antiquotation option "margin".
wenzelm@59175
   320
Only the margin for document output may be changed, but not the global
wenzelm@59175
   321
pretty printing: that is 76 for plain console output, and adapted
wenzelm@59175
   322
dynamically in GUI front-ends. Implementations of document
wenzelm@59175
   323
antiquotations need to observe the margin explicitly according to
wenzelm@59175
   324
Thy_Output.string_of_margin. Minor INCOMPATIBILITY.
wenzelm@59175
   325
wenzelm@58861
   326
* Historical command-line terminator ";" is no longer accepted.  Minor
wenzelm@58861
   327
INCOMPATIBILITY, use "isabelle update_semicolons" to remove obsolete
wenzelm@58861
   328
semicolons from theory sources.
wenzelm@58861
   329
wenzelm@58610
   330
* The Isabelle tool "update_cartouches" changes theory files to use
wenzelm@58610
   331
cartouches instead of old-style {* verbatim *} or `alt_string` tokens.
wenzelm@58610
   332
wenzelm@58610
   333
wenzelm@57695
   334
wenzelm@57452
   335
New in Isabelle2014 (August 2014)
wenzelm@57452
   336
---------------------------------
wenzelm@54055
   337
wenzelm@54702
   338
*** General ***
wenzelm@54702
   339
wenzelm@57452
   340
* Support for official Standard ML within the Isabelle context.
wenzelm@57452
   341
Command 'SML_file' reads and evaluates the given Standard ML file.
wenzelm@57452
   342
Toplevel bindings are stored within the theory context; the initial
wenzelm@57452
   343
environment is restricted to the Standard ML implementation of
wenzelm@57452
   344
Poly/ML, without the add-ons of Isabelle/ML.  Commands 'SML_import'
wenzelm@57452
   345
and 'SML_export' allow to exchange toplevel bindings between the two
wenzelm@57452
   346
separate environments.  See also ~~/src/Tools/SML/Examples.thy for
wenzelm@57452
   347
some examples.
wenzelm@56499
   348
wenzelm@57504
   349
* Standard tactics and proof methods such as "clarsimp", "auto" and
wenzelm@57504
   350
"safe" now preserve equality hypotheses "x = expr" where x is a free
wenzelm@57504
   351
variable.  Locale assumptions and chained facts containing "x"
wenzelm@57504
   352
continue to be useful.  The new method "hypsubst_thin" and the
wenzelm@57504
   353
configuration option "hypsubst_thin" (within the attribute name space)
wenzelm@57504
   354
restore the previous behavior.  INCOMPATIBILITY, especially where
wenzelm@57504
   355
induction is done after these methods or when the names of free and
wenzelm@57504
   356
bound variables clash.  As first approximation, old proofs may be
wenzelm@57504
   357
repaired by "using [[hypsubst_thin = true]]" in the critical spot.
wenzelm@57504
   358
wenzelm@56232
   359
* More static checking of proof methods, which allows the system to
wenzelm@56232
   360
form a closure over the concrete syntax.  Method arguments should be
wenzelm@56232
   361
processed in the original proof context as far as possible, before
wenzelm@56232
   362
operating on the goal state.  In any case, the standard discipline for
wenzelm@56232
   363
subgoal-addressing needs to be observed: no subgoals or a subgoal
wenzelm@56232
   364
number that is out of range produces an empty result sequence, not an
wenzelm@56232
   365
exception.  Potential INCOMPATIBILITY for non-conformant tactical
wenzelm@56232
   366
proof tools.
wenzelm@56232
   367
wenzelm@57452
   368
* Lexical syntax (inner and outer) supports text cartouches with
wenzelm@57452
   369
arbitrary nesting, and without escapes of quotes etc.  The Prover IDE
wenzelm@57452
   370
supports input via ` (backquote).
wenzelm@57452
   371
wenzelm@57452
   372
* The outer syntax categories "text" (for formal comments and document
wenzelm@57452
   373
markup commands) and "altstring" (for literal fact references) allow
wenzelm@57452
   374
cartouches as well, in addition to the traditional mix of quotations.
wenzelm@57452
   375
wenzelm@57452
   376
* Syntax of document antiquotation @{rail} now uses \<newline> instead
wenzelm@57452
   377
of "\\", to avoid the optical illusion of escaped backslash within
wenzelm@57491
   378
string token.  General renovation of its syntax using text cartouches.
wenzelm@57452
   379
Minor INCOMPATIBILITY.
wenzelm@57452
   380
wenzelm@57452
   381
* Discontinued legacy_isub_isup, which was a temporary workaround for
wenzelm@57452
   382
Isabelle/ML in Isabelle2013-1.  The prover process no longer accepts
wenzelm@57452
   383
old identifier syntax with \<^isub> or \<^isup>.  Potential
wenzelm@57452
   384
INCOMPATIBILITY.
wenzelm@57452
   385
wenzelm@57452
   386
* Document antiquotation @{url} produces markup for the given URL,
wenzelm@57452
   387
which results in an active hyperlink within the text.
wenzelm@57452
   388
wenzelm@57452
   389
* Document antiquotation @{file_unchecked} is like @{file}, but does
wenzelm@57452
   390
not check existence within the file-system.
wenzelm@57452
   391
wenzelm@57452
   392
* Updated and extended manuals: codegen, datatypes, implementation,
wenzelm@57452
   393
isar-ref, jedit, system.
wenzelm@57423
   394
wenzelm@54702
   395
wenzelm@54533
   396
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@54533
   397
wenzelm@57650
   398
* Improved Document panel: simplified interaction where every single
wenzelm@57452
   399
mouse click (re)opens document via desktop environment or as jEdit
wenzelm@57452
   400
buffer.
wenzelm@57452
   401
wenzelm@57452
   402
* Support for Navigator plugin (with toolbar buttons), with connection
wenzelm@57452
   403
to PIDE hyperlinks.
wenzelm@57452
   404
wenzelm@57452
   405
* Auxiliary files ('ML_file' etc.) are managed by the Prover IDE.
wenzelm@57452
   406
Open text buffers take precedence over copies within the file-system.
wenzelm@57452
   407
wenzelm@57452
   408
* Improved support for Isabelle/ML, with jEdit mode "isabelle-ml" for
wenzelm@57452
   409
auxiliary ML files.
wenzelm@57423
   410
wenzelm@57423
   411
* Improved syntactic and semantic completion mechanism, with simple
wenzelm@57423
   412
templates, completion language context, name-space completion,
wenzelm@57423
   413
file-name completion, spell-checker completion.
wenzelm@57423
   414
wenzelm@57423
   415
* Refined GUI popup for completion: more robust key/mouse event
wenzelm@57423
   416
handling and propagation to enclosing text area -- avoid loosing
wenzelm@57423
   417
keystrokes with slow / remote graphics displays.
wenzelm@57423
   418
wenzelm@57833
   419
* Completion popup supports both ENTER and TAB (default) to select an
wenzelm@57833
   420
item, depending on Isabelle options.
wenzelm@57833
   421
wenzelm@57423
   422
* Refined insertion of completion items wrt. jEdit text: multiple
wenzelm@57423
   423
selections, rectangular selections, rectangular selection as "tall
wenzelm@57423
   424
caret".
wenzelm@56342
   425
wenzelm@56580
   426
* Integrated spell-checker for document text, comments etc. with
wenzelm@57423
   427
completion popup and context-menu.
wenzelm@56554
   428
wenzelm@56879
   429
* More general "Query" panel supersedes "Find" panel, with GUI access
wenzelm@56879
   430
to commands 'find_theorems' and 'find_consts', as well as print
wenzelm@56879
   431
operations for the context.  Minor incompatibility in keyboard
wenzelm@56879
   432
shortcuts etc.: replace action isabelle-find by isabelle-query.
wenzelm@56761
   433
wenzelm@56901
   434
* Search field for all output panels ("Output", "Query", "Info" etc.)
wenzelm@56901
   435
to highlight text via regular expression.
wenzelm@56901
   436
wenzelm@54881
   437
* Option "jedit_print_mode" (see also "Plugin Options / Isabelle /
wenzelm@54881
   438
General") allows to specify additional print modes for the prover
wenzelm@54881
   439
process, without requiring old-fashioned command-line invocation of
wenzelm@54881
   440
"isabelle jedit -m MODE".
wenzelm@54881
   441
wenzelm@56505
   442
* More support for remote files (e.g. http) using standard Java
wenzelm@56505
   443
networking operations instead of jEdit virtual file-systems.
wenzelm@56505
   444
wenzelm@57822
   445
* Empty editors buffers that are no longer required (e.g.\ via theory
wenzelm@57822
   446
imports) are automatically removed from the document model.
wenzelm@57822
   447
wenzelm@57869
   448
* Improved monitor panel.
wenzelm@57869
   449
wenzelm@56838
   450
* Improved Console/Scala plugin: more uniform scala.Console output,
wenzelm@56838
   451
more robust treatment of threads and interrupts.
wenzelm@56838
   452
wenzelm@56939
   453
* Improved management of dockable windows: clarified keyboard focus
wenzelm@56939
   454
and window placement wrt. main editor view; optional menu item to
wenzelm@56939
   455
"Detach" a copy where this makes sense.
wenzelm@56939
   456
wenzelm@57452
   457
* New Simplifier Trace panel provides an interactive view of the
wenzelm@57591
   458
simplification process, enabled by the "simp_trace_new" attribute
wenzelm@57452
   459
within the context.
wenzelm@57452
   460
wenzelm@57452
   461
wenzelm@55001
   462
*** Pure ***
wenzelm@55001
   463
wenzelm@57504
   464
* Low-level type-class commands 'classes', 'classrel', 'arities' have
wenzelm@57504
   465
been discontinued to avoid the danger of non-trivial axiomatization
wenzelm@57504
   466
that is not immediately visible.  INCOMPATIBILITY, use regular
wenzelm@57504
   467
'instance' command with proof.  The required OFCLASS(...) theorem
wenzelm@57504
   468
might be postulated via 'axiomatization' beforehand, or the proof
wenzelm@57504
   469
finished trivially if the underlying class definition is made vacuous
wenzelm@57504
   470
(without any assumptions).  See also Isabelle/ML operations
wenzelm@57504
   471
Axclass.class_axiomatization, Axclass.classrel_axiomatization,
wenzelm@57504
   472
Axclass.arity_axiomatization.
wenzelm@57504
   473
wenzelm@56245
   474
* Basic constants of Pure use more conventional names and are always
wenzelm@56245
   475
qualified.  Rare INCOMPATIBILITY, but with potentially serious
wenzelm@56245
   476
consequences, notably for tools in Isabelle/ML.  The following
wenzelm@56245
   477
renaming needs to be applied:
wenzelm@56245
   478
wenzelm@56245
   479
  ==             ~>  Pure.eq
wenzelm@56245
   480
  ==>            ~>  Pure.imp
wenzelm@56245
   481
  all            ~>  Pure.all
wenzelm@56245
   482
  TYPE           ~>  Pure.type
wenzelm@56245
   483
  dummy_pattern  ~>  Pure.dummy_pattern
wenzelm@56245
   484
wenzelm@56245
   485
Systematic porting works by using the following theory setup on a
wenzelm@56245
   486
*previous* Isabelle version to introduce the new name accesses for the
wenzelm@56245
   487
old constants:
wenzelm@56245
   488
wenzelm@56245
   489
setup {*
wenzelm@56245
   490
  fn thy => thy
wenzelm@56245
   491
    |> Sign.root_path
wenzelm@56245
   492
    |> Sign.const_alias (Binding.qualify true "Pure" @{binding eq}) "=="
wenzelm@56245
   493
    |> Sign.const_alias (Binding.qualify true "Pure" @{binding imp}) "==>"
wenzelm@56245
   494
    |> Sign.const_alias (Binding.qualify true "Pure" @{binding all}) "all"
wenzelm@56245
   495
    |> Sign.restore_naming thy
wenzelm@56245
   496
*}
wenzelm@56245
   497
wenzelm@56245
   498
Thus ML antiquotations like @{const_name Pure.eq} may be used already.
wenzelm@56245
   499
Later the application is moved to the current Isabelle version, and
wenzelm@56245
   500
the auxiliary aliases are deleted.
wenzelm@56245
   501
wenzelm@55143
   502
* Attributes "where" and "of" allow an optional context of local
wenzelm@55143
   503
variables ('for' declaration): these variables become schematic in the
wenzelm@55143
   504
instantiated theorem.
wenzelm@55143
   505
wenzelm@55152
   506
* Obsolete attribute "standard" has been discontinued (legacy since
wenzelm@55152
   507
Isabelle2012).  Potential INCOMPATIBILITY, use explicit 'for' context
wenzelm@55152
   508
where instantiations with schematic variables are intended (for
wenzelm@55152
   509
declaration commands like 'lemmas' or attributes like "of").  The
wenzelm@55152
   510
following temporary definition may help to port old applications:
wenzelm@55152
   511
wenzelm@55152
   512
  attribute_setup standard =
wenzelm@55152
   513
    "Scan.succeed (Thm.rule_attribute (K Drule.export_without_context))"
wenzelm@55152
   514
wenzelm@55001
   515
* More thorough check of proof context for goal statements and
wenzelm@55006
   516
attributed fact expressions (concerning background theory, declared
wenzelm@55006
   517
hyps).  Potential INCOMPATIBILITY, tools need to observe standard
wenzelm@55006
   518
context discipline.  See also Assumption.add_assumes and the more
wenzelm@55006
   519
primitive Thm.assume_hyps.
wenzelm@55001
   520
wenzelm@55108
   521
* Inner syntax token language allows regular quoted strings "..."
wenzelm@55108
   522
(only makes sense in practice, if outer syntax is delimited
wenzelm@57452
   523
differently, e.g. via cartouches).
wenzelm@57452
   524
wenzelm@57504
   525
* Command 'print_term_bindings' supersedes 'print_binds' for clarity,
wenzelm@57504
   526
but the latter is retained some time as Proof General legacy.
wenzelm@57504
   527
wenzelm@57452
   528
* Code generator preprocessor: explicit control of simp tracing on a
wenzelm@57452
   529
per-constant basis.  See attribute "code_preproc".
haftmann@57430
   530
wenzelm@55001
   531
haftmann@54227
   532
*** HOL ***
haftmann@54227
   533
wenzelm@57504
   534
* Code generator: enforce case of identifiers only for strict target
wenzelm@57504
   535
language requirements.  INCOMPATIBILITY.
wenzelm@57504
   536
wenzelm@57504
   537
* Code generator: explicit proof contexts in many ML interfaces.
wenzelm@57504
   538
INCOMPATIBILITY.
wenzelm@57504
   539
wenzelm@57504
   540
* Code generator: minimize exported identifiers by default.  Minor
wenzelm@57504
   541
INCOMPATIBILITY.
wenzelm@57504
   542
wenzelm@57504
   543
* Code generation for SML and OCaml: dropped arcane "no_signatures"
wenzelm@57504
   544
option.  Minor INCOMPATIBILITY.
wenzelm@57504
   545
wenzelm@57504
   546
* "declare [[code abort: ...]]" replaces "code_abort ...".
wenzelm@57504
   547
INCOMPATIBILITY.
wenzelm@57504
   548
wenzelm@57504
   549
* "declare [[code drop: ...]]" drops all code equations associated
wenzelm@57504
   550
with the given constants.
wenzelm@57504
   551
wenzelm@57504
   552
* Code generations are provided for make, fields, extend and truncate
wenzelm@57504
   553
operations on records.
haftmann@57437
   554
wenzelm@57452
   555
* Command and antiquotation "value" are now hardcoded against nbe and
wenzelm@57452
   556
ML.  Minor INCOMPATIBILITY.
wenzelm@57452
   557
wenzelm@57504
   558
* Renamed command 'enriched_type' to 'functor'. INCOMPATIBILITY.
wenzelm@57504
   559
wenzelm@57504
   560
* The symbol "\<newline>" may be used within char or string literals
wenzelm@57504
   561
to represent (Char Nibble0 NibbleA), i.e. ASCII newline.
wenzelm@57504
   562
wenzelm@57504
   563
* Qualified String.implode and String.explode.  INCOMPATIBILITY.
haftmann@56923
   564
wenzelm@57452
   565
* Simplifier: Enhanced solver of preconditions of rewrite rules can
wenzelm@57452
   566
now deal with conjunctions.  For help with converting proofs, the old
wenzelm@57452
   567
behaviour of the simplifier can be restored like this: declare/using
wenzelm@57452
   568
[[simp_legacy_precond]].  This configuration option will disappear
wenzelm@57452
   569
again in the future.  INCOMPATIBILITY.
nipkow@56073
   570
wenzelm@55139
   571
* Simproc "finite_Collect" is no longer enabled by default, due to
wenzelm@55139
   572
spurious crashes and other surprises.  Potential INCOMPATIBILITY.
wenzelm@55139
   573
wenzelm@57452
   574
* Moved new (co)datatype package and its dependencies from session
wenzelm@57452
   575
  "HOL-BNF" to "HOL".  The commands 'bnf', 'wrap_free_constructors',
wenzelm@57452
   576
  'datatype_new', 'codatatype', 'primcorec', 'primcorecursive' are now
wenzelm@57452
   577
  part of theory "Main".
wenzelm@57452
   578
blanchet@55098
   579
  Theory renamings:
blanchet@55098
   580
    FunDef.thy ~> Fun_Def.thy (and Fun_Def_Base.thy)
blanchet@55098
   581
    Library/Wfrec.thy ~> Wfrec.thy
blanchet@55098
   582
    Library/Zorn.thy ~> Zorn.thy
blanchet@55098
   583
    Cardinals/Order_Relation.thy ~> Order_Relation.thy
blanchet@55098
   584
    Library/Order_Union.thy ~> Cardinals/Order_Union.thy
blanchet@55098
   585
    Cardinals/Cardinal_Arithmetic_Base.thy ~> BNF_Cardinal_Arithmetic.thy
blanchet@55098
   586
    Cardinals/Cardinal_Order_Relation_Base.thy ~> BNF_Cardinal_Order_Relation.thy
blanchet@55098
   587
    Cardinals/Constructions_on_Wellorders_Base.thy ~> BNF_Constructions_on_Wellorders.thy
blanchet@55098
   588
    Cardinals/Wellorder_Embedding_Base.thy ~> BNF_Wellorder_Embedding.thy
blanchet@55098
   589
    Cardinals/Wellorder_Relation_Base.thy ~> BNF_Wellorder_Relation.thy
blanchet@55098
   590
    BNF/Ctr_Sugar.thy ~> Ctr_Sugar.thy
blanchet@55098
   591
    BNF/Basic_BNFs.thy ~> Basic_BNFs.thy
blanchet@55098
   592
    BNF/BNF_Comp.thy ~> BNF_Comp.thy
blanchet@55098
   593
    BNF/BNF_Def.thy ~> BNF_Def.thy
blanchet@55098
   594
    BNF/BNF_FP_Base.thy ~> BNF_FP_Base.thy
blanchet@55098
   595
    BNF/BNF_GFP.thy ~> BNF_GFP.thy
blanchet@55098
   596
    BNF/BNF_LFP.thy ~> BNF_LFP.thy
blanchet@55098
   597
    BNF/BNF_Util.thy ~> BNF_Util.thy
blanchet@55098
   598
    BNF/Coinduction.thy ~> Coinduction.thy
blanchet@55098
   599
    BNF/More_BNFs.thy ~> Library/More_BNFs.thy
blanchet@55098
   600
    BNF/Countable_Type.thy ~> Library/Countable_Set_Type.thy
blanchet@55098
   601
    BNF/Examples/* ~> BNF_Examples/*
wenzelm@57452
   602
blanchet@55098
   603
  New theories:
blanchet@55098
   604
    Wellorder_Extension.thy (split from Zorn.thy)
blanchet@55098
   605
    Library/Cardinal_Notations.thy
traytel@56942
   606
    Library/BNF_Axomatization.thy
blanchet@55098
   607
    BNF_Examples/Misc_Primcorec.thy
blanchet@55098
   608
    BNF_Examples/Stream_Processor.thy
wenzelm@57452
   609
blanchet@55519
   610
  Discontinued theories:
blanchet@55098
   611
    BNF/BNF.thy
blanchet@55098
   612
    BNF/Equiv_Relations_More.thy
wenzelm@57452
   613
wenzelm@57452
   614
INCOMPATIBILITY.
blanchet@55098
   615
blanchet@56118
   616
* New (co)datatype package:
wenzelm@57452
   617
  - Command 'primcorec' is fully implemented.
wenzelm@57452
   618
  - Command 'datatype_new' generates size functions ("size_xxx" and
wenzelm@57452
   619
    "size") as required by 'fun'.
wenzelm@57452
   620
  - BNFs are integrated with the Lifting tool and new-style
wenzelm@57452
   621
    (co)datatypes with Transfer.
wenzelm@57452
   622
  - Renamed commands:
blanchet@55875
   623
      datatype_new_compat ~> datatype_compat
blanchet@55875
   624
      primrec_new ~> primrec
blanchet@55875
   625
      wrap_free_constructors ~> free_constructors
blanchet@55875
   626
    INCOMPATIBILITY.
wenzelm@57452
   627
  - The generated constants "xxx_case" and "xxx_rec" have been renamed
blanchet@55875
   628
    "case_xxx" and "rec_xxx" (e.g., "prod_case" ~> "case_prod").
blanchet@55875
   629
    INCOMPATIBILITY.
wenzelm@57452
   630
  - The constant "xxx_(un)fold" and related theorems are no longer
wenzelm@57452
   631
    generated.  Use "xxx_(co)rec" or define "xxx_(un)fold" manually
wenzelm@57452
   632
    using "prim(co)rec".
blanchet@55875
   633
    INCOMPATIBILITY.
wenzelm@57452
   634
  - No discriminators are generated for nullary constructors by
wenzelm@57452
   635
    default, eliminating the need for the odd "=:" syntax.
blanchet@57091
   636
    INCOMPATIBILITY.
wenzelm@57452
   637
  - No discriminators or selectors are generated by default by
blanchet@57094
   638
    "datatype_new", unless custom names are specified or the new
blanchet@57094
   639
    "discs_sels" option is passed.
blanchet@57094
   640
    INCOMPATIBILITY.
blanchet@55875
   641
blanchet@55643
   642
* Old datatype package:
wenzelm@57452
   643
  - The generated theorems "xxx.cases" and "xxx.recs" have been
wenzelm@57452
   644
    renamed "xxx.case" and "xxx.rec" (e.g., "sum.cases" ->
wenzelm@57452
   645
    "sum.case").  INCOMPATIBILITY.
wenzelm@57452
   646
  - The generated constants "xxx_case", "xxx_rec", and "xxx_size" have
wenzelm@57452
   647
    been renamed "case_xxx", "rec_xxx", and "size_xxx" (e.g.,
wenzelm@57452
   648
    "prod_case" ~> "case_prod").  INCOMPATIBILITY.
wenzelm@57452
   649
wenzelm@57452
   650
* The types "'a list" and "'a option", their set and map functions,
wenzelm@57452
   651
  their relators, and their selectors are now produced using the new
wenzelm@57452
   652
  BNF-based datatype package.
wenzelm@57452
   653
blanchet@55519
   654
  Renamed constants:
blanchet@55519
   655
    Option.set ~> set_option
blanchet@55519
   656
    Option.map ~> map_option
blanchet@55525
   657
    option_rel ~> rel_option
wenzelm@57452
   658
blanchet@55519
   659
  Renamed theorems:
blanchet@55585
   660
    set_def ~> set_rec[abs_def]
blanchet@55519
   661
    map_def ~> map_rec[abs_def]
blanchet@55519
   662
    Option.map_def ~> map_option_case[abs_def] (with "case_option" instead of "rec_option")
blanchet@56652
   663
    option.recs ~> option.rec
blanchet@55524
   664
    list_all2_def ~> list_all2_iff
blanchet@55585
   665
    set.simps ~> set_simps (or the slightly different "list.set")
blanchet@55519
   666
    map.simps ~> list.map
blanchet@55519
   667
    hd.simps ~> list.sel(1)
blanchet@55519
   668
    tl.simps ~> list.sel(2-3)
blanchet@55519
   669
    the.simps ~> option.sel
wenzelm@57452
   670
wenzelm@57452
   671
INCOMPATIBILITY.
blanchet@55519
   672
blanchet@55933
   673
* The following map functions and relators have been renamed:
blanchet@55939
   674
    sum_map ~> map_sum
blanchet@55939
   675
    map_pair ~> map_prod
blanchet@55944
   676
    prod_rel ~> rel_prod
blanchet@55943
   677
    sum_rel ~> rel_sum
blanchet@55945
   678
    fun_rel ~> rel_fun
blanchet@55942
   679
    set_rel ~> rel_set
blanchet@55942
   680
    filter_rel ~> rel_filter
wenzelm@57452
   681
    fset_rel ~> rel_fset (in "src/HOL/Library/FSet.thy")
wenzelm@57452
   682
    cset_rel ~> rel_cset (in "src/HOL/Library/Countable_Set_Type.thy")
wenzelm@57452
   683
    vset ~> rel_vset (in "src/HOL/Library/Quotient_Set.thy")
wenzelm@57452
   684
wenzelm@57452
   685
INCOMPATIBILITY.
wenzelm@57452
   686
kuncar@57826
   687
* Lifting and Transfer:
kuncar@57826
   688
  - a type variable as a raw type is supported
kuncar@57826
   689
  - stronger reflexivity prover
kuncar@57826
   690
  - rep_eq is always generated by lift_definition
wenzelm@57856
   691
  - setup for Lifting/Transfer is now automated for BNFs
kuncar@57826
   692
    + holds for BNFs that do not contain a dead variable
wenzelm@57856
   693
    + relator_eq, relator_mono, relator_distr, relator_domain,
kuncar@57826
   694
      relator_eq_onp, quot_map, transfer rules for bi_unique, bi_total,
kuncar@57826
   695
      right_unique, right_total, left_unique, left_total are proved
kuncar@57826
   696
      automatically
kuncar@57826
   697
    + definition of a predicator is generated automatically
kuncar@57826
   698
    + simplification rules for a predicator definition are proved
kuncar@57826
   699
      automatically for datatypes
kuncar@57826
   700
  - consolidation of the setup of Lifting/Transfer
wenzelm@57856
   701
    + property that a relator preservers reflexivity is not needed any
kuncar@57826
   702
      more
kuncar@57826
   703
      Minor INCOMPATIBILITY.
wenzelm@57856
   704
    + left_total and left_unique rules are now transfer rules
kuncar@57826
   705
      (reflexivity_rule attribute not needed anymore)
kuncar@57826
   706
      INCOMPATIBILITY.
wenzelm@57856
   707
    + Domainp does not have to be a separate assumption in
kuncar@57826
   708
      relator_domain theorems (=> more natural statement)
kuncar@57826
   709
      INCOMPATIBILITY.
kuncar@57826
   710
  - registration of code equations is more robust
kuncar@57826
   711
    Potential INCOMPATIBILITY.
kuncar@57826
   712
  - respectfulness proof obligation is preprocessed to a more readable
kuncar@57826
   713
    form
kuncar@57826
   714
    Potential INCOMPATIBILITY.
kuncar@57826
   715
  - eq_onp is always unfolded in respectfulness proof obligation
kuncar@57826
   716
    Potential INCOMPATIBILITY.
wenzelm@57856
   717
  - unregister lifting setup for Code_Numeral.integer and
kuncar@57826
   718
    Code_Numeral.natural
kuncar@57826
   719
    Potential INCOMPATIBILITY.
kuncar@57826
   720
  - Lifting.invariant -> eq_onp
kuncar@57826
   721
    INCOMPATIBILITY.
wenzelm@57856
   722
wenzelm@57508
   723
* New internal SAT solver "cdclite" that produces models and proof
wenzelm@57508
   724
traces.  This solver replaces the internal SAT solvers "enumerate" and
wenzelm@57508
   725
"dpll".  Applications that explicitly used one of these two SAT
wenzelm@57508
   726
solvers should use "cdclite" instead. In addition, "cdclite" is now
wenzelm@57508
   727
the default SAT solver for the "sat" and "satx" proof methods and
wenzelm@57508
   728
corresponding tactics; the old default can be restored using "declare
wenzelm@57508
   729
[[sat_solver = zchaff_with_proofs]]".  Minor INCOMPATIBILITY.
wenzelm@57508
   730
wenzelm@57508
   731
* SMT module: A new version of the SMT module, temporarily called
wenzelm@57508
   732
"SMT2", uses SMT-LIB 2 and supports recent versions of Z3 (e.g.,
wenzelm@57508
   733
4.3). The new proof method is called "smt2". CVC3 and CVC4 are also
wenzelm@57508
   734
supported as oracles. Yices is no longer supported, because no version
wenzelm@57508
   735
of the solver can handle both SMT-LIB 2 and quantifiers.
wenzelm@57508
   736
wenzelm@57508
   737
* Activation of Z3 now works via "z3_non_commercial" system option
wenzelm@57508
   738
(without requiring restart), instead of former settings variable
wenzelm@57508
   739
"Z3_NON_COMMERCIAL".  The option can be edited in Isabelle/jEdit menu
wenzelm@57508
   740
Plugin Options / Isabelle / General.
wenzelm@57508
   741
wenzelm@57508
   742
* Sledgehammer:
wenzelm@57508
   743
  - Z3 can now produce Isar proofs.
wenzelm@57508
   744
  - MaSh overhaul:
blanchet@57532
   745
    . New SML-based learning algorithms eliminate the dependency on
wenzelm@57508
   746
      Python and increase performance and reliability.
wenzelm@57508
   747
    . MaSh and MeSh are now used by default together with the
wenzelm@57508
   748
      traditional MePo (Meng-Paulson) relevance filter. To disable
wenzelm@57508
   749
      MaSh, set the "MaSh" system option in Isabelle/jEdit Plugin
wenzelm@57508
   750
      Options / Isabelle / General to "none".
wenzelm@57508
   751
  - New option:
wenzelm@57508
   752
      smt_proofs
wenzelm@57508
   753
  - Renamed options:
wenzelm@57508
   754
      isar_compress ~> compress
wenzelm@57508
   755
      isar_try0 ~> try0
wenzelm@57508
   756
wenzelm@57508
   757
INCOMPATIBILITY.
wenzelm@57508
   758
wenzelm@57508
   759
* Removed solvers remote_cvc3 and remote_z3. Use cvc3 and z3 instead.
wenzelm@57508
   760
wenzelm@57508
   761
* Nitpick:
wenzelm@57508
   762
  - Fixed soundness bug whereby mutually recursive datatypes could
wenzelm@57508
   763
    take infinite values.
wenzelm@57508
   764
  - Fixed soundness bug with low-level number functions such as
wenzelm@57508
   765
    "Abs_Integ" and "Rep_Integ".
wenzelm@57508
   766
  - Removed "std" option.
wenzelm@57508
   767
  - Renamed "show_datatypes" to "show_types" and "hide_datatypes" to
wenzelm@57508
   768
    "hide_types".
wenzelm@57508
   769
wenzelm@57508
   770
* Metis: Removed legacy proof method 'metisFT'. Use 'metis
wenzelm@57508
   771
(full_types)' instead. INCOMPATIBILITY.
wenzelm@57508
   772
wenzelm@57508
   773
* Try0: Added 'algebra' and 'meson' to the set of proof methods.
wenzelm@57508
   774
wenzelm@57508
   775
* Adjustion of INF and SUP operations:
wenzelm@57508
   776
  - Elongated constants INFI and SUPR to INFIMUM and SUPREMUM.
wenzelm@57508
   777
  - Consolidated theorem names containing INFI and SUPR: have INF and
wenzelm@57508
   778
    SUP instead uniformly.
wenzelm@57508
   779
  - More aggressive normalization of expressions involving INF and Inf
wenzelm@57508
   780
    or SUP and Sup.
wenzelm@57508
   781
  - INF_image and SUP_image do not unfold composition.
wenzelm@57508
   782
  - Dropped facts INF_comp, SUP_comp.
wenzelm@57508
   783
  - Default congruence rules strong_INF_cong and strong_SUP_cong, with
wenzelm@57508
   784
    simplifier implication in premises.  Generalize and replace former
wenzelm@57508
   785
    INT_cong, SUP_cong
wenzelm@57508
   786
wenzelm@57508
   787
INCOMPATIBILITY.
wenzelm@57508
   788
wenzelm@57508
   789
* SUP and INF generalized to conditionally_complete_lattice.
wenzelm@57508
   790
wenzelm@57508
   791
* Swapped orientation of facts image_comp and vimage_comp:
wenzelm@57508
   792
wenzelm@57508
   793
  image_compose ~> image_comp [symmetric]
wenzelm@57508
   794
  image_comp ~> image_comp [symmetric]
wenzelm@57508
   795
  vimage_compose ~> vimage_comp [symmetric]
wenzelm@57508
   796
  vimage_comp ~> vimage_comp [symmetric]
wenzelm@57508
   797
wenzelm@57508
   798
INCOMPATIBILITY.
wenzelm@57508
   799
wenzelm@57504
   800
* Theory reorganization: split of Big_Operators.thy into
wenzelm@57504
   801
Groups_Big.thy and Lattices_Big.thy.
blanchet@55098
   802
haftmann@57418
   803
* Consolidated some facts about big group operators:
haftmann@57418
   804
haftmann@57418
   805
    setsum_0' ~> setsum.neutral
haftmann@57418
   806
    setsum_0 ~> setsum.neutral_const
haftmann@57418
   807
    setsum_addf ~> setsum.distrib
haftmann@57418
   808
    setsum_cartesian_product ~> setsum.cartesian_product
haftmann@57418
   809
    setsum_cases ~> setsum.If_cases
haftmann@57418
   810
    setsum_commute ~> setsum.commute
haftmann@57418
   811
    setsum_cong ~> setsum.cong
haftmann@57418
   812
    setsum_delta ~> setsum.delta
haftmann@57418
   813
    setsum_delta' ~> setsum.delta'
haftmann@57418
   814
    setsum_diff1' ~> setsum.remove
haftmann@57418
   815
    setsum_empty ~> setsum.empty
haftmann@57418
   816
    setsum_infinite ~> setsum.infinite
haftmann@57418
   817
    setsum_insert ~> setsum.insert
haftmann@57418
   818
    setsum_inter_restrict'' ~> setsum.inter_filter
haftmann@57418
   819
    setsum_mono_zero_cong_left ~> setsum.mono_neutral_cong_left
haftmann@57418
   820
    setsum_mono_zero_cong_right ~> setsum.mono_neutral_cong_right
haftmann@57418
   821
    setsum_mono_zero_left ~> setsum.mono_neutral_left
haftmann@57418
   822
    setsum_mono_zero_right ~> setsum.mono_neutral_right
haftmann@57418
   823
    setsum_reindex ~> setsum.reindex
haftmann@57418
   824
    setsum_reindex_cong ~> setsum.reindex_cong
haftmann@57418
   825
    setsum_reindex_nonzero ~> setsum.reindex_nontrivial
haftmann@57418
   826
    setsum_restrict_set ~> setsum.inter_restrict
haftmann@57418
   827
    setsum_Plus ~> setsum.Plus
haftmann@57418
   828
    setsum_setsum_restrict ~> setsum.commute_restrict
haftmann@57418
   829
    setsum_Sigma ~> setsum.Sigma
haftmann@57418
   830
    setsum_subset_diff ~> setsum.subset_diff
haftmann@57418
   831
    setsum_Un_disjoint ~> setsum.union_disjoint
haftmann@57418
   832
    setsum_UN_disjoint ~> setsum.UNION_disjoint
haftmann@57418
   833
    setsum_Un_Int ~> setsum.union_inter
haftmann@57418
   834
    setsum_Union_disjoint ~> setsum.Union_disjoint
haftmann@57418
   835
    setsum_UNION_zero ~> setsum.Union_comp
haftmann@57418
   836
    setsum_Un_zero ~> setsum.union_inter_neutral
haftmann@57418
   837
    strong_setprod_cong ~> setprod.strong_cong
haftmann@57418
   838
    strong_setsum_cong ~> setsum.strong_cong
haftmann@57418
   839
    setprod_1' ~> setprod.neutral
haftmann@57418
   840
    setprod_1 ~> setprod.neutral_const
haftmann@57418
   841
    setprod_cartesian_product ~> setprod.cartesian_product
haftmann@57418
   842
    setprod_cong ~> setprod.cong
haftmann@57418
   843
    setprod_delta ~> setprod.delta
haftmann@57418
   844
    setprod_delta' ~> setprod.delta'
haftmann@57418
   845
    setprod_empty ~> setprod.empty
haftmann@57418
   846
    setprod_infinite ~> setprod.infinite
haftmann@57418
   847
    setprod_insert ~> setprod.insert
haftmann@57418
   848
    setprod_mono_one_cong_left ~> setprod.mono_neutral_cong_left
haftmann@57418
   849
    setprod_mono_one_cong_right ~> setprod.mono_neutral_cong_right
haftmann@57418
   850
    setprod_mono_one_left ~> setprod.mono_neutral_left
haftmann@57418
   851
    setprod_mono_one_right ~> setprod.mono_neutral_right
haftmann@57418
   852
    setprod_reindex ~> setprod.reindex
haftmann@57418
   853
    setprod_reindex_cong ~> setprod.reindex_cong
haftmann@57418
   854
    setprod_reindex_nonzero ~> setprod.reindex_nontrivial
haftmann@57418
   855
    setprod_Sigma ~> setprod.Sigma
haftmann@57418
   856
    setprod_subset_diff ~> setprod.subset_diff
haftmann@57418
   857
    setprod_timesf ~> setprod.distrib
haftmann@57418
   858
    setprod_Un2 ~> setprod.union_diff2
haftmann@57418
   859
    setprod_Un_disjoint ~> setprod.union_disjoint
haftmann@57418
   860
    setprod_UN_disjoint ~> setprod.UNION_disjoint
haftmann@57418
   861
    setprod_Un_Int ~> setprod.union_inter
haftmann@57418
   862
    setprod_Union_disjoint ~> setprod.Union_disjoint
haftmann@57418
   863
    setprod_Un_one ~> setprod.union_inter_neutral
haftmann@57418
   864
haftmann@57418
   865
  Dropped setsum_cong2 (simple variant of setsum.cong).
haftmann@57418
   866
  Dropped setsum_inter_restrict' (simple variant of setsum.inter_restrict)
haftmann@57418
   867
  Dropped setsum_reindex_id, setprod_reindex_id
haftmann@57418
   868
    (simple variants of setsum.reindex [symmetric], setprod.reindex [symmetric]).
haftmann@57418
   869
wenzelm@57452
   870
INCOMPATIBILITY.
wenzelm@57452
   871
haftmann@54864
   872
* Abolished slightly odd global lattice interpretation for min/max.
haftmann@54864
   873
wenzelm@57452
   874
  Fact consolidations:
haftmann@54864
   875
    min_max.inf_assoc ~> min.assoc
haftmann@54864
   876
    min_max.inf_commute ~> min.commute
haftmann@54864
   877
    min_max.inf_left_commute ~> min.left_commute
haftmann@54864
   878
    min_max.inf_idem ~> min.idem
haftmann@54864
   879
    min_max.inf_left_idem ~> min.left_idem
haftmann@54864
   880
    min_max.inf_right_idem ~> min.right_idem
haftmann@54864
   881
    min_max.sup_assoc ~> max.assoc
haftmann@54864
   882
    min_max.sup_commute ~> max.commute
haftmann@54864
   883
    min_max.sup_left_commute ~> max.left_commute
haftmann@54864
   884
    min_max.sup_idem ~> max.idem
haftmann@54864
   885
    min_max.sup_left_idem ~> max.left_idem
haftmann@54864
   886
    min_max.sup_inf_distrib1 ~> max_min_distrib2
haftmann@54864
   887
    min_max.sup_inf_distrib2 ~> max_min_distrib1
haftmann@54864
   888
    min_max.inf_sup_distrib1 ~> min_max_distrib2
haftmann@54864
   889
    min_max.inf_sup_distrib2 ~> min_max_distrib1
haftmann@54864
   890
    min_max.distrib ~> min_max_distribs
haftmann@54864
   891
    min_max.inf_absorb1 ~> min.absorb1
haftmann@54864
   892
    min_max.inf_absorb2 ~> min.absorb2
haftmann@54864
   893
    min_max.sup_absorb1 ~> max.absorb1
haftmann@54864
   894
    min_max.sup_absorb2 ~> max.absorb2
haftmann@54864
   895
    min_max.le_iff_inf ~> min.absorb_iff1
haftmann@54864
   896
    min_max.le_iff_sup ~> max.absorb_iff2
haftmann@54864
   897
    min_max.inf_le1 ~> min.cobounded1
haftmann@54864
   898
    min_max.inf_le2 ~> min.cobounded2
haftmann@54864
   899
    le_maxI1, min_max.sup_ge1 ~> max.cobounded1
haftmann@54864
   900
    le_maxI2, min_max.sup_ge2 ~> max.cobounded2
haftmann@54864
   901
    min_max.le_infI1 ~> min.coboundedI1
haftmann@54864
   902
    min_max.le_infI2 ~> min.coboundedI2
haftmann@54864
   903
    min_max.le_supI1 ~> max.coboundedI1
haftmann@54864
   904
    min_max.le_supI2 ~> max.coboundedI2
haftmann@54864
   905
    min_max.less_infI1 ~> min.strict_coboundedI1
haftmann@54864
   906
    min_max.less_infI2 ~> min.strict_coboundedI2
haftmann@54864
   907
    min_max.less_supI1 ~> max.strict_coboundedI1
haftmann@54864
   908
    min_max.less_supI2 ~> max.strict_coboundedI2
haftmann@54864
   909
    min_max.inf_mono ~> min.mono
haftmann@54864
   910
    min_max.sup_mono ~> max.mono
haftmann@54864
   911
    min_max.le_infI, min_max.inf_greatest ~> min.boundedI
haftmann@54864
   912
    min_max.le_supI, min_max.sup_least ~> max.boundedI
haftmann@54864
   913
    min_max.le_inf_iff ~> min.bounded_iff
haftmann@54864
   914
    min_max.le_sup_iff ~> max.bounded_iff
haftmann@54864
   915
haftmann@54864
   916
For min_max.inf_sup_aci, prefer (one of) min.commute, min.assoc,
haftmann@54864
   917
min.left_commute, min.left_idem, max.commute, max.assoc,
haftmann@54864
   918
max.left_commute, max.left_idem directly.
haftmann@54864
   919
wenzelm@57452
   920
For min_max.inf_sup_ord, prefer (one of) min.cobounded1,
wenzelm@57452
   921
min.cobounded2, max.cobounded1m max.cobounded2 directly.
haftmann@54864
   922
haftmann@56807
   923
For min_ac or max_ac, prefer more general collection ac_simps.
haftmann@54864
   924
wenzelm@58604
   925
INCOMPATIBILITY.
haftmann@54864
   926
wenzelm@57452
   927
* Theorem disambiguation Inf_le_Sup (on finite sets) ~>
wenzelm@57452
   928
Inf_fin_le_Sup_fin.  INCOMPATIBILITY.
haftmann@54745
   929
haftmann@54295
   930
* Qualified constant names Wellfounded.acc, Wellfounded.accp.
haftmann@54295
   931
INCOMPATIBILITY.
haftmann@54295
   932
haftmann@54228
   933
* Fact generalization and consolidation:
haftmann@54228
   934
    neq_one_mod_two, mod_2_not_eq_zero_eq_one_int ~> not_mod_2_eq_0_eq_1
wenzelm@57452
   935
wenzelm@57452
   936
INCOMPATIBILITY.
wenzelm@57452
   937
wenzelm@57452
   938
* Purely algebraic definition of even.  Fact generalization and
wenzelm@57452
   939
  consolidation:
haftmann@54228
   940
    nat_even_iff_2_dvd, int_even_iff_2_dvd ~> even_iff_2_dvd
haftmann@54228
   941
    even_zero_(nat|int) ~> even_zero
wenzelm@57452
   942
haftmann@54228
   943
INCOMPATIBILITY.
wenzelm@54055
   944
haftmann@54489
   945
* Abolished neg_numeral.
wenzelm@57452
   946
  - Canonical representation for minus one is "- 1".
wenzelm@57452
   947
  - Canonical representation for other negative numbers is "- (numeral _)".
wenzelm@57452
   948
  - When devising rule sets for number calculation, consider the
haftmann@54587
   949
    following canonical cases: 0, 1, numeral _, - 1, - numeral _.
wenzelm@57452
   950
  - HOLogic.dest_number also recognizes numerals in non-canonical forms
wenzelm@54893
   951
    like "numeral One", "- numeral One", "- 0" and even "- ... - _".
wenzelm@57452
   952
  - Syntax for negative numerals is mere input syntax.
wenzelm@57452
   953
haftmann@56964
   954
INCOMPATIBILITY.
haftmann@54489
   955
wenzelm@57517
   956
* Reduced name variants for rules on associativity and commutativity:
wenzelm@57517
   957
wenzelm@57517
   958
    add_assoc ~> add.assoc
wenzelm@57517
   959
    add_commute ~> add.commute
wenzelm@57517
   960
    add_left_commute ~> add.left_commute
wenzelm@57517
   961
    mult_assoc ~> mult.assoc
wenzelm@57517
   962
    mult_commute ~> mult.commute
wenzelm@57517
   963
    mult_left_commute ~> mult.left_commute
wenzelm@57517
   964
    nat_add_assoc ~> add.assoc
wenzelm@57517
   965
    nat_add_commute ~> add.commute
wenzelm@57517
   966
    nat_add_left_commute ~> add.left_commute
wenzelm@57517
   967
    nat_mult_assoc ~> mult.assoc
wenzelm@57517
   968
    nat_mult_commute ~> mult.commute
wenzelm@57517
   969
    eq_assoc ~> iff_assoc
wenzelm@57517
   970
    eq_left_commute ~> iff_left_commute
wenzelm@57517
   971
wenzelm@57517
   972
INCOMPATIBILITY.
wenzelm@57517
   973
wenzelm@57650
   974
* Fact collections add_ac and mult_ac are considered old-fashioned.
wenzelm@57637
   975
Prefer ac_simps instead, or specify rules
wenzelm@57637
   976
(add|mult).(assoc|commute|left_commute) individually.
wenzelm@57517
   977
haftmann@54230
   978
* Elimination of fact duplicates:
haftmann@54230
   979
    equals_zero_I ~> minus_unique
haftmann@54230
   980
    diff_eq_0_iff_eq ~> right_minus_eq
traytel@54588
   981
    nat_infinite ~> infinite_UNIV_nat
traytel@54588
   982
    int_infinite ~> infinite_UNIV_int
wenzelm@57452
   983
haftmann@54230
   984
INCOMPATIBILITY.
haftmann@54230
   985
haftmann@54230
   986
* Fact name consolidation:
haftmann@54230
   987
    diff_def, diff_minus, ab_diff_minus ~> diff_conv_add_uminus
haftmann@54250
   988
    minus_le_self_iff ~> neg_less_eq_nonneg
haftmann@54250
   989
    le_minus_self_iff ~> less_eq_neg_nonpos
haftmann@54250
   990
    neg_less_nonneg ~> neg_less_pos
haftmann@54250
   991
    less_minus_self_iff ~> less_neg_neg [simp]
wenzelm@57452
   992
haftmann@54230
   993
INCOMPATIBILITY.
haftmann@54230
   994
haftmann@54230
   995
* More simplification rules on unary and binary minus:
haftmann@54230
   996
add_diff_cancel, add_diff_cancel_left, add_le_same_cancel1,
haftmann@54230
   997
add_le_same_cancel2, add_less_same_cancel1, add_less_same_cancel2,
haftmann@54230
   998
add_minus_cancel, diff_add_cancel, le_add_same_cancel1,
haftmann@54230
   999
le_add_same_cancel2, less_add_same_cancel1, less_add_same_cancel2,
wenzelm@57452
  1000
minus_add_cancel, uminus_add_conv_diff.  These correspondingly have
wenzelm@57452
  1001
been taken away from fact collections algebra_simps and field_simps.
wenzelm@57452
  1002
INCOMPATIBILITY.
haftmann@54230
  1003
haftmann@54230
  1004
To restore proofs, the following patterns are helpful:
haftmann@54230
  1005
haftmann@54230
  1006
a) Arbitrary failing proof not involving "diff_def":
haftmann@54230
  1007
Consider simplification with algebra_simps or field_simps.
haftmann@54230
  1008
haftmann@54230
  1009
b) Lifting rules from addition to subtraction:
wenzelm@54893
  1010
Try with "using <rule for addition> of [... "- _" ...]" by simp".
haftmann@54230
  1011
haftmann@54230
  1012
c) Simplification with "diff_def": just drop "diff_def".
haftmann@54230
  1013
Consider simplification with algebra_simps or field_simps;
haftmann@54230
  1014
or the brute way with
haftmann@54230
  1015
"simp add: diff_conv_add_uminus del: add_uminus_conv_diff".
haftmann@54230
  1016
wenzelm@57452
  1017
* Introduce bdd_above and bdd_below in theory
wenzelm@57452
  1018
Conditionally_Complete_Lattices, use them instead of explicitly
wenzelm@57452
  1019
stating boundedness of sets.
wenzelm@57452
  1020
wenzelm@57452
  1021
* ccpo.admissible quantifies only over non-empty chains to allow more
wenzelm@57452
  1022
syntax-directed proof rules; the case of the empty chain shows up as
wenzelm@57452
  1023
additional case in fixpoint induction proofs.  INCOMPATIBILITY.
hoelzl@54264
  1024
hoelzl@56214
  1025
* Removed and renamed theorems in Series:
hoelzl@56214
  1026
  summable_le         ~>  suminf_le
hoelzl@56214
  1027
  suminf_le           ~>  suminf_le_const
hoelzl@56214
  1028
  series_pos_le       ~>  setsum_le_suminf
hoelzl@56214
  1029
  series_pos_less     ~>  setsum_less_suminf
hoelzl@56214
  1030
  suminf_ge_zero      ~>  suminf_nonneg
hoelzl@56214
  1031
  suminf_gt_zero      ~>  suminf_pos
hoelzl@56214
  1032
  suminf_gt_zero_iff  ~>  suminf_pos_iff
hoelzl@56214
  1033
  summable_sumr_LIMSEQ_suminf  ~>  summable_LIMSEQ
hoelzl@56214
  1034
  suminf_0_le         ~>  suminf_nonneg [rotate]
hoelzl@56214
  1035
  pos_summable        ~>  summableI_nonneg_bounded
hoelzl@56214
  1036
  ratio_test          ~>  summable_ratio_test
hoelzl@56214
  1037
hoelzl@56214
  1038
  removed series_zero, replaced by sums_finite
hoelzl@56214
  1039
hoelzl@56214
  1040
  removed auxiliary lemmas:
wenzelm@57452
  1041
hoelzl@56214
  1042
    sumr_offset, sumr_offset2, sumr_offset3, sumr_offset4, sumr_group,
wenzelm@57452
  1043
    half, le_Suc_ex_iff, lemma_realpow_diff_sumr,
wenzelm@57452
  1044
    real_setsum_nat_ivl_bounded, summable_le2, ratio_test_lemma2,
wenzelm@57452
  1045
    sumr_minus_one_realpow_zerom, sumr_one_lb_realpow_zero,
wenzelm@57452
  1046
    summable_convergent_sumr_iff, sumr_diff_mult_const
wenzelm@57452
  1047
hoelzl@56214
  1048
INCOMPATIBILITY.
hoelzl@56214
  1049
hoelzl@56214
  1050
* Replace (F)DERIV syntax by has_derivative:
hoelzl@56214
  1051
  - "(f has_derivative f') (at x within s)" replaces "FDERIV f x : s : f'"
hoelzl@56214
  1052
hoelzl@56214
  1053
  - "(f has_field_derivative f') (at x within s)" replaces "DERIV f x : s : f'"
hoelzl@56214
  1054
hoelzl@56214
  1055
  - "f differentiable at x within s" replaces "_ differentiable _ in _" syntax
hoelzl@56214
  1056
hoelzl@56214
  1057
  - removed constant isDiff
hoelzl@56214
  1058
wenzelm@57452
  1059
  - "DERIV f x : f'" and "FDERIV f x : f'" syntax is only available as
wenzelm@57452
  1060
    input syntax.
wenzelm@57452
  1061
wenzelm@57452
  1062
  - "DERIV f x : s : f'" and "FDERIV f x : s : f'" syntax removed.
hoelzl@56214
  1063
hoelzl@56214
  1064
  - Renamed FDERIV_... lemmas to has_derivative_...
hoelzl@56214
  1065
hoelzl@56381
  1066
  - renamed deriv (the syntax constant used for "DERIV _ _ :> _") to DERIV
hoelzl@56381
  1067
hoelzl@56381
  1068
  - removed DERIV_intros, has_derivative_eq_intros
hoelzl@56381
  1069
wenzelm@57452
  1070
  - introduced derivative_intros and deriative_eq_intros which
wenzelm@57452
  1071
    includes now rules for DERIV, has_derivative and
wenzelm@57452
  1072
    has_vector_derivative.
hoelzl@56381
  1073
hoelzl@56214
  1074
  - Other renamings:
hoelzl@56214
  1075
    differentiable_def        ~>  real_differentiable_def
hoelzl@56214
  1076
    differentiableE           ~>  real_differentiableE
hoelzl@56214
  1077
    fderiv_def                ~>  has_derivative_at
hoelzl@56214
  1078
    field_fderiv_def          ~>  field_has_derivative_at
hoelzl@56214
  1079
    isDiff_der                ~>  differentiable_def
hoelzl@56214
  1080
    deriv_fderiv              ~>  has_field_derivative_def
hoelzl@56381
  1081
    deriv_def                 ~>  DERIV_def
wenzelm@57452
  1082
wenzelm@57452
  1083
INCOMPATIBILITY.
wenzelm@57452
  1084
wenzelm@57452
  1085
* Include more theorems in continuous_intros. Remove the
wenzelm@57452
  1086
continuous_on_intros, isCont_intros collections, these facts are now
wenzelm@57452
  1087
in continuous_intros.
wenzelm@57452
  1088
wenzelm@57452
  1089
* Theorems about complex numbers are now stated only using Re and Im,
wenzelm@57452
  1090
the Complex constructor is not used anymore. It is possible to use
wenzelm@57452
  1091
primcorec to defined the behaviour of a complex-valued function.
wenzelm@57452
  1092
wenzelm@57452
  1093
Removed theorems about the Complex constructor from the simpset, they
wenzelm@57452
  1094
are available as the lemma collection legacy_Complex_simps. This
wenzelm@57452
  1095
especially removes
wenzelm@57452
  1096
hoelzl@56889
  1097
    i_complex_of_real: "ii * complex_of_real r = Complex 0 r".
hoelzl@56889
  1098
wenzelm@57452
  1099
Instead the reverse direction is supported with
hoelzl@56889
  1100
    Complex_eq: "Complex a b = a + \<i> * b"
hoelzl@56889
  1101
wenzelm@57452
  1102
Moved csqrt from Fundamental_Algebra_Theorem to Complex.
hoelzl@56889
  1103
hoelzl@56889
  1104
  Renamings:
hoelzl@56889
  1105
    Re/Im                  ~>  complex.sel
hoelzl@56889
  1106
    complex_Re/Im_zero     ~>  zero_complex.sel
hoelzl@56889
  1107
    complex_Re/Im_add      ~>  plus_complex.sel
hoelzl@56889
  1108
    complex_Re/Im_minus    ~>  uminus_complex.sel
hoelzl@56889
  1109
    complex_Re/Im_diff     ~>  minus_complex.sel
hoelzl@56889
  1110
    complex_Re/Im_one      ~>  one_complex.sel
hoelzl@56889
  1111
    complex_Re/Im_mult     ~>  times_complex.sel
hoelzl@56889
  1112
    complex_Re/Im_inverse  ~>  inverse_complex.sel
hoelzl@56889
  1113
    complex_Re/Im_scaleR   ~>  scaleR_complex.sel
hoelzl@56889
  1114
    complex_Re/Im_i        ~>  ii.sel
hoelzl@56889
  1115
    complex_Re/Im_cnj      ~>  cnj.sel
hoelzl@56889
  1116
    Re/Im_cis              ~>  cis.sel
hoelzl@56889
  1117
hoelzl@56889
  1118
    complex_divide_def   ~>  divide_complex_def
hoelzl@56889
  1119
    complex_norm_def     ~>  norm_complex_def
hoelzl@56889
  1120
    cmod_def             ~>  norm_complex_de
hoelzl@56889
  1121
hoelzl@56889
  1122
  Removed theorems:
hoelzl@56889
  1123
    complex_zero_def
hoelzl@56889
  1124
    complex_add_def
hoelzl@56889
  1125
    complex_minus_def
hoelzl@56889
  1126
    complex_diff_def
hoelzl@56889
  1127
    complex_one_def
hoelzl@56889
  1128
    complex_mult_def
hoelzl@56889
  1129
    complex_inverse_def
hoelzl@56889
  1130
    complex_scaleR_def
hoelzl@56889
  1131
wenzelm@57452
  1132
INCOMPATIBILITY.
wenzelm@57452
  1133
wenzelm@57504
  1134
* Theory Lubs moved HOL image to HOL-Library. It is replaced by
wenzelm@57504
  1135
Conditionally_Complete_Lattices.  INCOMPATIBILITY.
wenzelm@57504
  1136
wenzelm@57504
  1137
* HOL-Library: new theory src/HOL/Library/Tree.thy.
wenzelm@57504
  1138
wenzelm@57504
  1139
* HOL-Library: removed theory src/HOL/Library/Kleene_Algebra.thy; it
wenzelm@57504
  1140
is subsumed by session Kleene_Algebra in AFP.
wenzelm@57504
  1141
wenzelm@57856
  1142
* HOL-Library / theory RBT: various constants and facts are hidden;
wenzelm@57856
  1143
lifting setup is unregistered.  INCOMPATIBILITY.
kuncar@57826
  1144
wenzelm@57504
  1145
* HOL-Cardinals: new theory src/HOL/Cardinals/Ordinal_Arithmetic.thy.
wenzelm@57504
  1146
wenzelm@57504
  1147
* HOL-Word: bit representations prefer type bool over type bit.
wenzelm@57504
  1148
INCOMPATIBILITY.
wenzelm@57504
  1149
wenzelm@57504
  1150
* HOL-Word:
wenzelm@57504
  1151
  - Abandoned fact collection "word_arith_alts", which is a duplicate
wenzelm@57504
  1152
    of "word_arith_wis".
wenzelm@57504
  1153
  - Dropped first (duplicated) element in fact collections
wenzelm@57504
  1154
    "sint_word_ariths", "word_arith_alts", "uint_word_ariths",
wenzelm@57504
  1155
    "uint_word_arith_bintrs".
wenzelm@57504
  1156
wenzelm@57504
  1157
* HOL-Number_Theory:
wenzelm@57504
  1158
  - consolidated the proofs of the binomial theorem
wenzelm@57504
  1159
  - the function fib is again of type nat => nat and not overloaded
wenzelm@57504
  1160
  - no more references to Old_Number_Theory in the HOL libraries
wenzelm@57504
  1161
    (except the AFP)
wenzelm@57504
  1162
wenzelm@57504
  1163
INCOMPATIBILITY.
wenzelm@57504
  1164
immler@54787
  1165
* HOL-Multivariate_Analysis:
wenzelm@57452
  1166
  - Type class ordered_real_vector for ordered vector spaces.
wenzelm@57452
  1167
  - New theory Complex_Basic_Analysis defining complex derivatives,
lp15@57253
  1168
    holomorphic functions, etc., ported from HOL Light's canal.ml.
wenzelm@57452
  1169
  - Changed order of ordered_euclidean_space to be compatible with
immler@54787
  1170
    pointwise ordering on products. Therefore instance of
immler@54787
  1171
    conditionally_complete_lattice and ordered_real_vector.
immler@54787
  1172
    INCOMPATIBILITY: use box instead of greaterThanLessThan or
wenzelm@57452
  1173
    explicit set-comprehensions with eucl_less for other (half-)open
immler@54787
  1174
    intervals.
immler@57476
  1175
  - removed dependencies on type class ordered_euclidean_space with
immler@57476
  1176
    introduction of "cbox" on euclidean_space
immler@57476
  1177
    - renamed theorems:
immler@57476
  1178
        interval ~> box
immler@57476
  1179
        mem_interval ~> mem_box
immler@57476
  1180
        interval_eq_empty ~> box_eq_empty
immler@57476
  1181
        interval_ne_empty ~> box_ne_empty
immler@57476
  1182
        interval_sing(1) ~> cbox_sing
immler@57476
  1183
        interval_sing(2) ~> box_sing
immler@57476
  1184
        subset_interval_imp ~> subset_box_imp
immler@57476
  1185
        subset_interval ~> subset_box
immler@57476
  1186
        open_interval ~> open_box
immler@57476
  1187
        closed_interval ~> closed_cbox
immler@57476
  1188
        interior_closed_interval ~> interior_cbox
immler@57476
  1189
        bounded_closed_interval ~> bounded_cbox
immler@57476
  1190
        compact_interval ~> compact_cbox
immler@57476
  1191
        bounded_subset_closed_interval_symmetric ~> bounded_subset_cbox_symmetric
immler@57476
  1192
        bounded_subset_closed_interval ~> bounded_subset_cbox
immler@57476
  1193
        mem_interval_componentwiseI ~> mem_box_componentwiseI
immler@57476
  1194
        convex_box ~> convex_prod
immler@57476
  1195
        rel_interior_real_interval ~> rel_interior_real_box
immler@57476
  1196
        convex_interval ~> convex_box
immler@57476
  1197
        convex_hull_eq_real_interval ~> convex_hull_eq_real_cbox
immler@57476
  1198
        frechet_derivative_within_closed_interval ~> frechet_derivative_within_cbox
immler@57476
  1199
        content_closed_interval' ~> content_cbox'
immler@57476
  1200
        elementary_subset_interval ~> elementary_subset_box
immler@57476
  1201
        diameter_closed_interval ~> diameter_cbox
immler@57476
  1202
        frontier_closed_interval ~> frontier_cbox
immler@57476
  1203
        frontier_open_interval ~> frontier_box
immler@57476
  1204
        bounded_subset_open_interval_symmetric ~> bounded_subset_box_symmetric
immler@57476
  1205
        closure_open_interval ~> closure_box
immler@57476
  1206
        open_closed_interval_convex ~> open_cbox_convex
immler@57476
  1207
        open_interval_midpoint ~> box_midpoint
immler@57476
  1208
        content_image_affinity_interval ~> content_image_affinity_cbox
immler@57476
  1209
        is_interval_interval ~> is_interval_cbox + is_interval_box + is_interval_closed_interval
immler@57476
  1210
        bounded_interval ~> bounded_closed_interval + bounded_boxes
immler@57476
  1211
immler@57476
  1212
    - respective theorems for intervals over the reals:
immler@57476
  1213
        content_closed_interval + content_cbox
immler@57476
  1214
        has_integral + has_integral_real
immler@57476
  1215
        fine_division_exists + fine_division_exists_real
immler@57476
  1216
        has_integral_null + has_integral_null_real
immler@57476
  1217
        tagged_division_union_interval + tagged_division_union_interval_real
immler@57476
  1218
        has_integral_const + has_integral_const_real
immler@57476
  1219
        integral_const + integral_const_real
immler@57476
  1220
        has_integral_bound + has_integral_bound_real
immler@57476
  1221
        integrable_continuous + integrable_continuous_real
immler@57476
  1222
        integrable_subinterval + integrable_subinterval_real
immler@57476
  1223
        has_integral_reflect_lemma + has_integral_reflect_lemma_real
immler@57476
  1224
        integrable_reflect + integrable_reflect_real
immler@57476
  1225
        integral_reflect + integral_reflect_real
immler@57476
  1226
        image_affinity_interval + image_affinity_cbox
immler@57476
  1227
        image_smult_interval + image_smult_cbox
immler@57476
  1228
        integrable_const + integrable_const_ivl
immler@57476
  1229
        integrable_on_subinterval + integrable_on_subcbox
immler@57476
  1230
hoelzl@56369
  1231
  - renamed theorems:
hoelzl@56369
  1232
    derivative_linear         ~>  has_derivative_bounded_linear
hoelzl@56369
  1233
    derivative_is_linear      ~>  has_derivative_linear
hoelzl@56369
  1234
    bounded_linear_imp_linear ~>  bounded_linear.linear
hoelzl@56369
  1235
hoelzl@56993
  1236
* HOL-Probability:
hoelzl@57825
  1237
  - Renamed positive_integral to nn_integral:
hoelzl@57825
  1238
hoelzl@57825
  1239
    . Renamed all lemmas "*positive_integral*" to *nn_integral*"
hoelzl@57825
  1240
      positive_integral_positive ~> nn_integral_nonneg
hoelzl@57825
  1241
hoelzl@57825
  1242
    . Renamed abbreviation integral\<^sup>P to integral\<^sup>N.
hoelzl@57825
  1243
wenzelm@57452
  1244
  - replaced the Lebesgue integral on real numbers by the more general
wenzelm@57452
  1245
    Bochner integral for functions into a real-normed vector space.
hoelzl@56993
  1246
hoelzl@56993
  1247
    integral_zero               ~>  integral_zero / integrable_zero
hoelzl@56993
  1248
    integral_minus              ~>  integral_minus / integrable_minus
hoelzl@56993
  1249
    integral_add                ~>  integral_add / integrable_add
hoelzl@56993
  1250
    integral_diff               ~>  integral_diff / integrable_diff
hoelzl@56993
  1251
    integral_setsum             ~>  integral_setsum / integrable_setsum
hoelzl@56993
  1252
    integral_multc              ~>  integral_mult_left / integrable_mult_left
hoelzl@56993
  1253
    integral_cmult              ~>  integral_mult_right / integrable_mult_right
hoelzl@56993
  1254
    integral_triangle_inequality~>  integral_norm_bound
hoelzl@56993
  1255
    integrable_nonneg           ~>  integrableI_nonneg
hoelzl@56993
  1256
    integral_positive           ~>  integral_nonneg_AE
hoelzl@56993
  1257
    integrable_abs_iff          ~>  integrable_abs_cancel
hoelzl@57825
  1258
    positive_integral_lim_INF   ~>  nn_integral_liminf
hoelzl@56993
  1259
    lebesgue_real_affine        ~>  lborel_real_affine
hoelzl@56993
  1260
    borel_integral_has_integral ~>  has_integral_lebesgue_integral
hoelzl@56993
  1261
    integral_indicator          ~>
hoelzl@56993
  1262
         integral_real_indicator / integrable_real_indicator
hoelzl@57825
  1263
    positive_integral_fst       ~>  nn_integral_fst'
hoelzl@57825
  1264
    positive_integral_fst_measurable ~> nn_integral_fst
hoelzl@57825
  1265
    positive_integral_snd_measurable ~> nn_integral_snd
hoelzl@56993
  1266
hoelzl@56993
  1267
    integrable_fst_measurable   ~>
hoelzl@56993
  1268
         integral_fst / integrable_fst / AE_integrable_fst
hoelzl@56993
  1269
hoelzl@56993
  1270
    integrable_snd_measurable   ~>
hoelzl@56993
  1271
         integral_snd / integrable_snd / AE_integrable_snd
hoelzl@56993
  1272
hoelzl@56993
  1273
    integral_monotone_convergence  ~>
hoelzl@56993
  1274
         integral_monotone_convergence / integrable_monotone_convergence
hoelzl@56993
  1275
hoelzl@56993
  1276
    integral_monotone_convergence_at_top  ~>
hoelzl@56993
  1277
         integral_monotone_convergence_at_top /
hoelzl@56993
  1278
         integrable_monotone_convergence_at_top
hoelzl@56993
  1279
hoelzl@56993
  1280
    has_integral_iff_positive_integral_lebesgue  ~>
hoelzl@56993
  1281
         has_integral_iff_has_bochner_integral_lebesgue_nonneg
hoelzl@56993
  1282
hoelzl@56993
  1283
    lebesgue_integral_has_integral  ~>
hoelzl@56993
  1284
         has_integral_integrable_lebesgue_nonneg
hoelzl@56993
  1285
hoelzl@56993
  1286
    positive_integral_lebesgue_has_integral  ~>
hoelzl@56993
  1287
         integral_has_integral_lebesgue_nonneg /
hoelzl@56993
  1288
         integrable_has_integral_lebesgue_nonneg
hoelzl@56993
  1289
hoelzl@56993
  1290
    lebesgue_integral_real_affine  ~>
hoelzl@57825
  1291
         nn_integral_real_affine
hoelzl@56993
  1292
hoelzl@56993
  1293
    has_integral_iff_positive_integral_lborel  ~>
hoelzl@56993
  1294
         integral_has_integral_nonneg / integrable_has_integral_nonneg
hoelzl@56993
  1295
hoelzl@56993
  1296
    The following theorems where removed:
hoelzl@56993
  1297
hoelzl@56993
  1298
    lebesgue_integral_nonneg
hoelzl@56993
  1299
    lebesgue_integral_uminus
hoelzl@56993
  1300
    lebesgue_integral_cmult
hoelzl@56993
  1301
    lebesgue_integral_multc
hoelzl@56993
  1302
    lebesgue_integral_cmult_nonneg
hoelzl@56993
  1303
    integral_cmul_indicator
hoelzl@56993
  1304
    integral_real
wenzelm@54672
  1305
wenzelm@57452
  1306
  - Formalized properties about exponentially, Erlang, and normal
wenzelm@57452
  1307
    distributed random variables.
wenzelm@57452
  1308
wenzelm@57504
  1309
* HOL-Decision_Procs: Separate command 'approximate' for approximative
wenzelm@57504
  1310
computation in src/HOL/Decision_Procs/Approximation.  Minor
wenzelm@57504
  1311
INCOMPATIBILITY.
wenzelm@57452
  1312
nipkow@57112
  1313
wenzelm@55622
  1314
*** Scala ***
wenzelm@55622
  1315
wenzelm@55622
  1316
* The signature and semantics of Document.Snapshot.cumulate_markup /
wenzelm@55622
  1317
select_markup have been clarified.  Markup is now traversed in the
wenzelm@55622
  1318
order of reports given by the prover: later markup is usually more
wenzelm@55622
  1319
specific and may override results accumulated so far.  The elements
wenzelm@55622
  1320
guard is mandatory and checked precisely.  Subtle INCOMPATIBILITY.
wenzelm@55622
  1321
wenzelm@57452
  1322
* Substantial reworking of internal PIDE protocol communication
wenzelm@57452
  1323
channels.  INCOMPATIBILITY.
wenzelm@57452
  1324
wenzelm@55622
  1325
wenzelm@54449
  1326
*** ML ***
wenzelm@54449
  1327
wenzelm@57504
  1328
* Subtle change of semantics of Thm.eq_thm: theory stamps are not
wenzelm@57504
  1329
compared (according to Thm.thm_ord), but assumed to be covered by the
wenzelm@57504
  1330
current background theory.  Thus equivalent data produced in different
wenzelm@57504
  1331
branches of the theory graph usually coincides (e.g. relevant for
wenzelm@57504
  1332
theory merge).  Note that the softer Thm.eq_thm_prop is often more
wenzelm@57504
  1333
appropriate than Thm.eq_thm.
wenzelm@57504
  1334
wenzelm@57504
  1335
* Proper context for basic Simplifier operations: rewrite_rule,
wenzelm@57504
  1336
rewrite_goals_rule, rewrite_goals_tac etc. INCOMPATIBILITY, need to
wenzelm@57504
  1337
pass runtime Proof.context (and ensure that the simplified entity
wenzelm@57504
  1338
actually belongs to it).
wenzelm@57504
  1339
wenzelm@57504
  1340
* Proper context discipline for read_instantiate and instantiate_tac:
wenzelm@57504
  1341
variables that are meant to become schematic need to be given as
wenzelm@57504
  1342
fixed, and are generalized by the explicit context of local variables.
wenzelm@57504
  1343
This corresponds to Isar attributes "where" and "of" with 'for'
wenzelm@57504
  1344
declaration.  INCOMPATIBILITY, also due to potential change of indices
wenzelm@57504
  1345
of schematic variables.
wenzelm@57504
  1346
wenzelm@56303
  1347
* Moved ML_Compiler.exn_trace and other operations on exceptions to
wenzelm@56303
  1348
structure Runtime.  Minor INCOMPATIBILITY.
wenzelm@56303
  1349
wenzelm@56279
  1350
* Discontinued old Toplevel.debug in favour of system option
wenzelm@57452
  1351
"ML_exception_trace", which may be also declared within the context
wenzelm@57452
  1352
via "declare [[ML_exception_trace = true]]".  Minor INCOMPATIBILITY.
wenzelm@56279
  1353
wenzelm@56281
  1354
* Renamed configuration option "ML_trace" to "ML_source_trace". Minor
wenzelm@56281
  1355
INCOMPATIBILITY.
wenzelm@56281
  1356
wenzelm@56281
  1357
* Configuration option "ML_print_depth" controls the pretty-printing
wenzelm@56281
  1358
depth of the ML compiler within the context.  The old print_depth in
wenzelm@56285
  1359
ML is still available as default_print_depth, but rarely used.  Minor
wenzelm@56285
  1360
INCOMPATIBILITY.
wenzelm@56279
  1361
wenzelm@54449
  1362
* Toplevel function "use" refers to raw ML bootstrap environment,
wenzelm@54449
  1363
without Isar context nor antiquotations.  Potential INCOMPATIBILITY.
wenzelm@54449
  1364
Note that 'ML_file' is the canonical command to load ML files into the
wenzelm@54449
  1365
formal context.
wenzelm@54449
  1366
wenzelm@56205
  1367
* Simplified programming interface to define ML antiquotations, see
wenzelm@56205
  1368
structure ML_Antiquotation.  Minor INCOMPATIBILITY.
wenzelm@56069
  1369
wenzelm@56071
  1370
* ML antiquotation @{here} refers to its source position, which is
wenzelm@56071
  1371
occasionally useful for experimentation and diagnostic purposes.
wenzelm@56071
  1372
wenzelm@56135
  1373
* ML antiquotation @{path} produces a Path.T value, similarly to
wenzelm@56135
  1374
Path.explode, but with compile-time check against the file-system and
wenzelm@56135
  1375
some PIDE markup.  Note that unlike theory source, ML does not have a
wenzelm@56135
  1376
well-defined master directory, so an absolute symbolic path
wenzelm@56135
  1377
specification is usually required, e.g. "~~/src/HOL".
wenzelm@56135
  1378
wenzelm@56399
  1379
* ML antiquotation @{print} inlines a function to print an arbitrary
wenzelm@56399
  1380
ML value, which is occasionally useful for diagnostic or demonstration
wenzelm@56399
  1381
purposes.
wenzelm@56399
  1382
wenzelm@54449
  1383
wenzelm@54683
  1384
*** System ***
wenzelm@54683
  1385
wenzelm@57443
  1386
* Proof General with its traditional helper scripts is now an optional
wenzelm@57504
  1387
Isabelle component, e.g. see ProofGeneral-4.2-2 from the Isabelle
wenzelm@57504
  1388
component repository http://isabelle.in.tum.de/components/.  Note that
wenzelm@57504
  1389
the "system" manual provides general explanations about add-on
wenzelm@57504
  1390
components, especially those that are not bundled with the release.
wenzelm@57443
  1391
wenzelm@56439
  1392
* The raw Isabelle process executable has been renamed from
wenzelm@56439
  1393
"isabelle-process" to "isabelle_process", which conforms to common
wenzelm@56439
  1394
shell naming conventions, and allows to define a shell function within
wenzelm@56439
  1395
the Isabelle environment to avoid dynamic path lookup.  Rare
wenzelm@57504
  1396
incompatibility for old tools that do not use the ISABELLE_PROCESS
wenzelm@57504
  1397
settings variable.
wenzelm@56439
  1398
wenzelm@57439
  1399
* Former "isabelle tty" has been superseded by "isabelle console",
wenzelm@57439
  1400
with implicit build like "isabelle jedit", and without the mostly
wenzelm@57439
  1401
obsolete Isar TTY loop.
wenzelm@57439
  1402
wenzelm@57452
  1403
* Simplified "isabelle display" tool.  Settings variables DVI_VIEWER
wenzelm@57452
  1404
and PDF_VIEWER now refer to the actual programs, not shell
wenzelm@57452
  1405
command-lines.  Discontinued option -c: invocation may be asynchronous
wenzelm@57452
  1406
via desktop environment, without any special precautions.  Potential
wenzelm@57452
  1407
INCOMPATIBILITY with ambitious private settings.
wenzelm@57452
  1408
wenzelm@57413
  1409
* Removed obsolete "isabelle unsymbolize".  Note that the usual format
wenzelm@57413
  1410
for email communication is the Unicode rendering of Isabelle symbols,
wenzelm@57413
  1411
as produced by Isabelle/jEdit, for example.
wenzelm@57413
  1412
wenzelm@57452
  1413
* Removed obsolete tool "wwwfind". Similar functionality may be
wenzelm@57452
  1414
integrated into Isabelle/jEdit eventually.
wenzelm@57452
  1415
wenzelm@57452
  1416
* Improved 'display_drafts' concerning desktop integration and
wenzelm@57452
  1417
repeated invocation in PIDE front-end: re-use single file
wenzelm@57452
  1418
$ISABELLE_HOME_USER/tmp/drafts.pdf and corresponding views.
wenzelm@57084
  1419
wenzelm@57452
  1420
* Session ROOT specifications require explicit 'document_files' for
wenzelm@57452
  1421
robust dependencies on LaTeX sources.  Only these explicitly given
wenzelm@57452
  1422
files are copied to the document output directory, before document
wenzelm@57452
  1423
processing is started.
wenzelm@57452
  1424
wenzelm@57504
  1425
* Windows: support for regular TeX installation (e.g. MiKTeX) instead
wenzelm@57504
  1426
of TeX Live from Cygwin.
wenzelm@57504
  1427
wenzelm@57084
  1428
wenzelm@57693
  1429
wenzelm@54639
  1430
New in Isabelle2013-2 (December 2013)
wenzelm@54639
  1431
-------------------------------------
wenzelm@54639
  1432
wenzelm@54639
  1433
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@54639
  1434
wenzelm@54639
  1435
* More robust editing of running commands with internal forks,
wenzelm@54639
  1436
e.g. non-terminating 'by' steps.
wenzelm@54639
  1437
wenzelm@54641
  1438
* More relaxed Sledgehammer panel: avoid repeated application of query
wenzelm@54641
  1439
after edits surrounding the command location.
wenzelm@54641
  1440
wenzelm@54648
  1441
* More status information about commands that are interrupted
wenzelm@54648
  1442
accidentally (via physical event or Poly/ML runtime system signal,
wenzelm@54648
  1443
e.g. out-of-memory).
wenzelm@54648
  1444
wenzelm@54653
  1445
wenzelm@54653
  1446
*** System ***
wenzelm@54653
  1447
wenzelm@54653
  1448
* More robust termination of external processes managed by
wenzelm@54664
  1449
Isabelle/ML: support cancellation of tasks within the range of
wenzelm@54664
  1450
milliseconds, as required for PIDE document editing with automatically
wenzelm@54664
  1451
tried tools (e.g. Sledgehammer).
wenzelm@54653
  1452
wenzelm@54648
  1453
* Reactivated Isabelle/Scala kill command for external processes on
wenzelm@54648
  1454
Mac OS X, which was accidentally broken in Isabelle2013-1 due to a
wenzelm@54648
  1455
workaround for some Debian/Ubuntu Linux versions from 2013.
wenzelm@54648
  1456
wenzelm@54639
  1457
wenzelm@54639
  1458
wenzelm@53971
  1459
New in Isabelle2013-1 (November 2013)
wenzelm@53971
  1460
-------------------------------------
wenzelm@50994
  1461
wenzelm@51293
  1462
*** General ***
wenzelm@51293
  1463
wenzelm@53971
  1464
* Discontinued obsolete 'uses' within theory header.  Note that
wenzelm@53971
  1465
commands like 'ML_file' work without separate declaration of file
wenzelm@53971
  1466
dependencies.  Minor INCOMPATIBILITY.
wenzelm@53971
  1467
wenzelm@53971
  1468
* Discontinued redundant 'use' command, which was superseded by
wenzelm@53971
  1469
'ML_file' in Isabelle2013.  Minor INCOMPATIBILITY.
wenzelm@53971
  1470
wenzelm@53016
  1471
* Simplified subscripts within identifiers, using plain \<^sub>
wenzelm@53016
  1472
instead of the second copy \<^isub> and \<^isup>.  Superscripts are
wenzelm@53016
  1473
only for literal tokens within notation; explicit mixfix annotations
wenzelm@53016
  1474
for consts or fixed variables may be used as fall-back for unusual
wenzelm@53016
  1475
names.  Obsolete \<twosuperior> has been expanded to \<^sup>2 in
wenzelm@53016
  1476
Isabelle/HOL.  INCOMPATIBILITY, use "isabelle update_sub_sup" to
wenzelm@53016
  1477
standardize symbols as a starting point for further manual cleanup.
wenzelm@53016
  1478
The ML reference variable "legacy_isub_isup" may be set as temporary
wenzelm@53016
  1479
workaround, to make the prover accept a subset of the old identifier
wenzelm@53016
  1480
syntax.
wenzelm@53016
  1481
wenzelm@53021
  1482
* Document antiquotations: term style "isub" has been renamed to
wenzelm@53021
  1483
"sub".  Minor INCOMPATIBILITY.
wenzelm@53021
  1484
wenzelm@52487
  1485
* Uniform management of "quick_and_dirty" as system option (see also
wenzelm@52487
  1486
"isabelle options"), configuration option within the context (see also
wenzelm@52487
  1487
Config.get in Isabelle/ML), and attribute in Isabelle/Isar.  Minor
wenzelm@52487
  1488
INCOMPATIBILITY, need to use more official Isabelle means to access
wenzelm@52487
  1489
quick_and_dirty, instead of historical poking into mutable reference.
wenzelm@52059
  1490
wenzelm@52060
  1491
* Renamed command 'print_configs' to 'print_options'.  Minor
wenzelm@52060
  1492
INCOMPATIBILITY.
wenzelm@52060
  1493
wenzelm@52430
  1494
* Proper diagnostic command 'print_state'.  Old 'pr' (with its
wenzelm@52430
  1495
implicit change of some global references) is retained for now as
wenzelm@52430
  1496
control command, e.g. for ProofGeneral 3.7.x.
wenzelm@52430
  1497
wenzelm@52549
  1498
* Discontinued 'print_drafts' command with its old-fashioned PS output
wenzelm@52549
  1499
and Unix command-line print spooling.  Minor INCOMPATIBILITY: use
wenzelm@52549
  1500
'display_drafts' instead and print via the regular document viewer.
wenzelm@52549
  1501
wenzelm@53971
  1502
* Updated and extended "isar-ref" and "implementation" manual,
wenzelm@53971
  1503
eliminated old "ref" manual.
wenzelm@53971
  1504
wenzelm@51293
  1505
wenzelm@51533
  1506
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@51533
  1507
wenzelm@53971
  1508
* New manual "jedit" for Isabelle/jEdit, see isabelle doc or
wenzelm@53852
  1509
Documentation panel.
wenzelm@53852
  1510
wenzelm@53971
  1511
* Dockable window "Documentation" provides access to Isabelle
wenzelm@53971
  1512
documentation.
wenzelm@52646
  1513
wenzelm@52949
  1514
* Dockable window "Find" provides query operations for formal entities
wenzelm@52949
  1515
(GUI front-end to 'find_theorems' command).
wenzelm@52949
  1516
wenzelm@53050
  1517
* Dockable window "Sledgehammer" manages asynchronous / parallel
wenzelm@53050
  1518
sledgehammer runs over existing document sources, independently of
wenzelm@53050
  1519
normal editing and checking process.
wenzelm@53050
  1520
wenzelm@51533
  1521
* Dockable window "Timing" provides an overview of relevant command
wenzelm@54332
  1522
timing information, depending on option jedit_timing_threshold.  The
wenzelm@54332
  1523
same timing information is shown in the extended tooltip of the
wenzelm@54332
  1524
command keyword, when hovering the mouse over it while the CONTROL or
wenzelm@54332
  1525
COMMAND modifier is pressed.
wenzelm@51533
  1526
wenzelm@53971
  1527
* Improved dockable window "Theories": Continuous checking of proof
wenzelm@53971
  1528
document (visible and required parts) may be controlled explicitly,
wenzelm@53971
  1529
using check box or shortcut "C+e ENTER".  Individual theory nodes may
wenzelm@53971
  1530
be marked explicitly as required and checked in full, using check box
wenzelm@53971
  1531
or shortcut "C+e SPACE".
wenzelm@53971
  1532
wenzelm@54305
  1533
* Improved completion mechanism, which is now managed by the
wenzelm@54305
  1534
Isabelle/jEdit plugin instead of SideKick.  Refined table of Isabelle
wenzelm@54305
  1535
symbol abbreviations (see $ISABELLE_HOME/etc/symbols).
wenzelm@54305
  1536
wenzelm@54319
  1537
* Standard jEdit keyboard shortcut C+b complete-word is remapped to
wenzelm@54319
  1538
isabelle.complete for explicit completion in Isabelle sources.
wenzelm@54319
  1539
INCOMPATIBILITY wrt. jEdit defaults, may have to invent new shortcuts
wenzelm@54319
  1540
to resolve conflict.
wenzelm@54319
  1541
wenzelm@54305
  1542
* Improved support of various "minor modes" for Isabelle NEWS,
wenzelm@54305
  1543
options, session ROOT etc., with completion and SideKick tree view.
wenzelm@54305
  1544
wenzelm@53971
  1545
* Strictly monotonic document update, without premature cancellation of
wenzelm@53971
  1546
running transactions that are still needed: avoid reset/restart of
wenzelm@53971
  1547
such command executions while editing.
wenzelm@53271
  1548
wenzelm@53971
  1549
* Support for asynchronous print functions, as overlay to existing
wenzelm@53971
  1550
document content.
wenzelm@53971
  1551
wenzelm@53971
  1552
* Support for automatic tools in HOL, which try to prove or disprove
wenzelm@53971
  1553
toplevel theorem statements.
wenzelm@53971
  1554
wenzelm@53971
  1555
* Action isabelle.reset-font-size resets main text area font size
wenzelm@54365
  1556
according to Isabelle/Scala plugin option "jedit_font_reset_size" (see
wenzelm@54365
  1557
also "Plugin Options / Isabelle / General").  It can be bound to some
wenzelm@54365
  1558
keyboard shortcut by the user (e.g. C+0 and/or C+NUMPAD0).
wenzelm@53971
  1559
wenzelm@53971
  1560
* File specifications in jEdit (e.g. file browser) may refer to
wenzelm@54351
  1561
$ISABELLE_HOME and $ISABELLE_HOME_USER on all platforms.  Discontinued
wenzelm@54351
  1562
obsolete $ISABELLE_HOME_WINDOWS variable.
wenzelm@53971
  1563
wenzelm@53971
  1564
* Improved support for Linux look-and-feel "GTK+", see also "Utilities
wenzelm@53971
  1565
/ Global Options / Appearance".
wenzelm@53971
  1566
wenzelm@53971
  1567
* Improved support of native Mac OS X functionality via "MacOSX"
wenzelm@53971
  1568
plugin, which is now enabled by default.
wenzelm@53971
  1569
wenzelm@51533
  1570
wenzelm@51313
  1571
*** Pure ***
wenzelm@51313
  1572
ballarin@54049
  1573
* Commands 'interpretation' and 'sublocale' are now target-sensitive.
ballarin@54049
  1574
In particular, 'interpretation' allows for non-persistent
ballarin@54049
  1575
interpretation within "context ... begin ... end" blocks offering a
ballarin@54049
  1576
light-weight alternative to 'sublocale'.  See "isar-ref" manual for
ballarin@54049
  1577
details.
haftmann@51747
  1578
ballarin@51565
  1579
* Improved locales diagnostic command 'print_dependencies'.
ballarin@51565
  1580
wenzelm@51313
  1581
* Discontinued obsolete 'axioms' command, which has been marked as
wenzelm@51313
  1582
legacy since Isabelle2009-2.  INCOMPATIBILITY, use 'axiomatization'
wenzelm@51313
  1583
instead, while observing its uniform scope for polymorphism.
wenzelm@51313
  1584
wenzelm@51316
  1585
* Discontinued empty name bindings in 'axiomatization'.
wenzelm@51316
  1586
INCOMPATIBILITY.
wenzelm@51316
  1587
wenzelm@53971
  1588
* System option "proofs" has been discontinued.  Instead the global
wenzelm@53971
  1589
state of Proofterm.proofs is persistently compiled into logic images
wenzelm@53971
  1590
as required, notably HOL-Proofs.  Users no longer need to change
wenzelm@53971
  1591
Proofterm.proofs dynamically.  Minor INCOMPATIBILITY.
wenzelm@53971
  1592
wenzelm@53971
  1593
* Syntax translation functions (print_translation etc.) always depend
wenzelm@53971
  1594
on Proof.context.  Discontinued former "(advanced)" option -- this is
wenzelm@53971
  1595
now the default.  Minor INCOMPATIBILITY.
wenzelm@53971
  1596
wenzelm@53971
  1597
* Former global reference trace_unify_fail is now available as
wenzelm@53971
  1598
configuration option "unify_trace_failure" (global context only).
wenzelm@53971
  1599
wenzelm@52463
  1600
* SELECT_GOAL now retains the syntactic context of the overall goal
wenzelm@52463
  1601
state (schematic variables etc.).  Potential INCOMPATIBILITY in rare
wenzelm@52463
  1602
situations.
wenzelm@52463
  1603
wenzelm@51313
  1604
hoelzl@51002
  1605
*** HOL ***
hoelzl@51002
  1606
wenzelm@54032
  1607
* Stronger precedence of syntax for big intersection and union on
wenzelm@54032
  1608
sets, in accordance with corresponding lattice operations.
wenzelm@54032
  1609
INCOMPATIBILITY.
wenzelm@54032
  1610
wenzelm@54032
  1611
* Notation "{p:A. P}" now allows tuple patterns as well.
wenzelm@54032
  1612
wenzelm@54032
  1613
* Nested case expressions are now translated in a separate check phase
wenzelm@54032
  1614
rather than during parsing. The data for case combinators is separated
wenzelm@54032
  1615
from the datatype package. The declaration attribute
wenzelm@54032
  1616
"case_translation" can be used to register new case combinators:
wenzelm@54032
  1617
wenzelm@54032
  1618
  declare [[case_translation case_combinator constructor1 ... constructorN]]
haftmann@52637
  1619
haftmann@52435
  1620
* Code generator:
wenzelm@53160
  1621
  - 'code_printing' unifies 'code_const' / 'code_type' / 'code_class' /
wenzelm@53160
  1622
    'code_instance'.
wenzelm@53160
  1623
  - 'code_identifier' declares name hints for arbitrary identifiers in
wenzelm@53160
  1624
    generated code, subsuming 'code_modulename'.
wenzelm@53983
  1625
wenzelm@53983
  1626
See the isar-ref manual for syntax diagrams, and the HOL theories for
wenzelm@53983
  1627
examples.
haftmann@52435
  1628
wenzelm@54032
  1629
* Attibute 'code': 'code' now declares concrete and abstract code
wenzelm@54032
  1630
equations uniformly.  Use explicit 'code equation' and 'code abstract'
wenzelm@54032
  1631
to distinguish both when desired.
wenzelm@54032
  1632
wenzelm@54032
  1633
* Discontinued theories Code_Integer and Efficient_Nat by a more
wenzelm@54032
  1634
fine-grain stack of theories Code_Target_Int, Code_Binary_Nat,
wenzelm@54032
  1635
Code_Target_Nat and Code_Target_Numeral.  See the tutorial on code
wenzelm@54032
  1636
generation for details.  INCOMPATIBILITY.
wenzelm@54032
  1637
wenzelm@54032
  1638
* Numeric types are mapped by default to target language numerals:
wenzelm@54032
  1639
natural (replaces former code_numeral) and integer (replaces former
wenzelm@54032
  1640
code_int).  Conversions are available as integer_of_natural /
wenzelm@54032
  1641
natural_of_integer / integer_of_nat / nat_of_integer (in HOL) and
wenzelm@54032
  1642
Code_Numeral.integer_of_natural / Code_Numeral.natural_of_integer (in
wenzelm@54032
  1643
ML).  INCOMPATIBILITY.
wenzelm@54032
  1644
wenzelm@54032
  1645
* Function package: For mutually recursive functions f and g, separate
wenzelm@54032
  1646
cases rules f.cases and g.cases are generated instead of unusable
wenzelm@54032
  1647
f_g.cases which exposed internal sum types. Potential INCOMPATIBILITY,
wenzelm@54032
  1648
in the case that the unusable rule was used nevertheless.
wenzelm@54032
  1649
wenzelm@54032
  1650
* Function package: For each function f, new rules f.elims are
wenzelm@54032
  1651
generated, which eliminate equalities of the form "f x = t".
wenzelm@54032
  1652
wenzelm@54032
  1653
* New command 'fun_cases' derives ad-hoc elimination rules for
wenzelm@54032
  1654
function equations as simplified instances of f.elims, analogous to
wenzelm@54032
  1655
inductive_cases.  See ~~/src/HOL/ex/Fundefs.thy for some examples.
blanchet@53307
  1656
kuncar@54021
  1657
* Lifting:
kuncar@54021
  1658
  - parametrized correspondence relations are now supported:
wenzelm@54378
  1659
    + parametricity theorems for the raw term can be specified in
kuncar@54021
  1660
      the command lift_definition, which allow us to generate stronger
kuncar@54021
  1661
      transfer rules
kuncar@54021
  1662
    + setup_lifting generates stronger transfer rules if parametric
kuncar@54021
  1663
      correspondence relation can be generated
kuncar@54021
  1664
    + various new properties of the relator must be specified to support
kuncar@54021
  1665
      parametricity
kuncar@54021
  1666
    + parametricity theorem for the Quotient relation can be specified
kuncar@54021
  1667
  - setup_lifting generates domain rules for the Transfer package
kuncar@54021
  1668
  - stronger reflexivity prover of respectfulness theorems for type
kuncar@54021
  1669
    copies
kuncar@54021
  1670
  - ===> and --> are now local. The symbols can be introduced
kuncar@54021
  1671
    by interpreting the locale lifting_syntax (typically in an
kuncar@54021
  1672
    anonymous context)
wenzelm@54378
  1673
  - Lifting/Transfer relevant parts of Library/Quotient_* are now in
kuncar@54021
  1674
    Main. Potential INCOMPATIBILITY
kuncar@54021
  1675
  - new commands for restoring and deleting Lifting/Transfer context:
kuncar@54021
  1676
    lifting_forget, lifting_update
wenzelm@54378
  1677
  - the command print_quotmaps was renamed to print_quot_maps.
kuncar@54021
  1678
    INCOMPATIBILITY
kuncar@54021
  1679
kuncar@54021
  1680
* Transfer:
wenzelm@54378
  1681
  - better support for domains in Transfer: replace Domainp T
kuncar@54021
  1682
    by the actual invariant in a transferred goal
kuncar@54021
  1683
  - transfer rules can have as assumptions other transfer rules
kuncar@54021
  1684
  - Experimental support for transferring from the raw level to the
kuncar@54021
  1685
    abstract level: Transfer.transferred attribute
kuncar@54021
  1686
  - Attribute version of the transfer method: untransferred attribute
kuncar@54021
  1687
haftmann@52286
  1688
* Reification and reflection:
wenzelm@53160
  1689
  - Reification is now directly available in HOL-Main in structure
wenzelm@53160
  1690
    "Reification".
wenzelm@53160
  1691
  - Reflection now handles multiple lists with variables also.
wenzelm@53160
  1692
  - The whole reflection stack has been decomposed into conversions.
haftmann@52286
  1693
INCOMPATIBILITY.
haftmann@52286
  1694
haftmann@51489
  1695
* Revised devices for recursive definitions over finite sets:
haftmann@51489
  1696
  - Only one fundamental fold combinator on finite set remains:
haftmann@51489
  1697
    Finite_Set.fold :: ('a => 'b => 'b) => 'b => 'a set => 'b
haftmann@51489
  1698
    This is now identity on infinite sets.
wenzelm@52745
  1699
  - Locales ("mini packages") for fundamental definitions with
haftmann@51489
  1700
    Finite_Set.fold: folding, folding_idem.
haftmann@51489
  1701
  - Locales comm_monoid_set, semilattice_order_set and
haftmann@51489
  1702
    semilattice_neutr_order_set for big operators on sets.
haftmann@51489
  1703
    See theory Big_Operators for canonical examples.
haftmann@51489
  1704
    Note that foundational constants comm_monoid_set.F and
haftmann@51489
  1705
    semilattice_set.F correspond to former combinators fold_image
haftmann@51489
  1706
    and fold1 respectively.  These are now gone.  You may use
haftmann@51490
  1707
    those foundational constants as substitutes, but it is
wenzelm@53983
  1708
    preferable to interpret the above locales accordingly.
haftmann@51489
  1709
  - Dropped class ab_semigroup_idem_mult (special case of lattice,
haftmann@51489
  1710
    no longer needed in connection with Finite_Set.fold etc.)
haftmann@51489
  1711
  - Fact renames:
haftmann@51489
  1712
      card.union_inter ~> card_Un_Int [symmetric]
haftmann@51489
  1713
      card.union_disjoint ~> card_Un_disjoint
haftmann@51489
  1714
INCOMPATIBILITY.
haftmann@51489
  1715
haftmann@51487
  1716
* Locale hierarchy for abstract orderings and (semi)lattices.
haftmann@51487
  1717
wenzelm@53526
  1718
* Complete_Partial_Order.admissible is defined outside the type class
wenzelm@53526
  1719
ccpo, but with mandatory prefix ccpo. Admissibility theorems lose the
wenzelm@53526
  1720
class predicate assumption or sort constraint when possible.
Andreas@53362
  1721
INCOMPATIBILITY.
Andreas@53362
  1722
wenzelm@53160
  1723
* Introduce type class "conditionally_complete_lattice": Like a
wenzelm@53160
  1724
complete lattice but does not assume the existence of the top and
wenzelm@53160
  1725
bottom elements.  Allows to generalize some lemmas about reals and
wenzelm@53160
  1726
extended reals.  Removed SupInf and replaced it by the instantiation
wenzelm@53160
  1727
of conditionally_complete_lattice for real. Renamed lemmas about
wenzelm@53160
  1728
conditionally-complete lattice from Sup_... to cSup_... and from
wenzelm@53160
  1729
Inf_...  to cInf_... to avoid hidding of similar complete lattice
wenzelm@53160
  1730
lemmas.
wenzelm@53160
  1731
wenzelm@53160
  1732
* Introduce type class linear_continuum as combination of
wenzelm@53160
  1733
conditionally-complete lattices and inner dense linorders which have
wenzelm@53160
  1734
more than one element.  INCOMPATIBILITY.
wenzelm@53160
  1735
wenzelm@53983
  1736
* Introduced type classes order_top and order_bot. The old classes top
wenzelm@53983
  1737
and bot only contain the syntax without assumptions.  INCOMPATIBILITY:
wenzelm@53983
  1738
Rename bot -> order_bot, top -> order_top
lammich@53683
  1739
wenzelm@53160
  1740
* Introduce type classes "no_top" and "no_bot" for orderings without
wenzelm@53160
  1741
top and bottom elements.
hoelzl@51732
  1742
hoelzl@51732
  1743
* Split dense_linorder into inner_dense_order and no_top, no_bot.
hoelzl@51732
  1744
hoelzl@51732
  1745
* Complex_Main: Unify and move various concepts from
wenzelm@53160
  1746
HOL-Multivariate_Analysis to HOL-Complex_Main.
hoelzl@51732
  1747
wenzelm@53983
  1748
 - Introduce type class (lin)order_topology and
wenzelm@53983
  1749
   linear_continuum_topology.  Allows to generalize theorems about
wenzelm@53983
  1750
   limits and order.  Instances are reals and extended reals.
hoelzl@51732
  1751
hoelzl@51732
  1752
 - continuous and continuos_on from Multivariate_Analysis:
wenzelm@53983
  1753
   "continuous" is the continuity of a function at a filter.  "isCont"
wenzelm@53983
  1754
   is now an abbrevitation: "isCont x f == continuous (at _) f".
wenzelm@53983
  1755
wenzelm@53983
  1756
   Generalized continuity lemmas from isCont to continuous on an
wenzelm@53983
  1757
   arbitrary filter.
wenzelm@53983
  1758
wenzelm@53983
  1759
 - compact from Multivariate_Analysis. Use Bolzano's lemma to prove
wenzelm@53983
  1760
   compactness of closed intervals on reals. Continuous functions
wenzelm@53983
  1761
   attain infimum and supremum on compact sets. The inverse of a
wenzelm@53983
  1762
   continuous function is continuous, when the function is continuous
wenzelm@53983
  1763
   on a compact set.
hoelzl@51732
  1764
hoelzl@51732
  1765
 - connected from Multivariate_Analysis. Use it to prove the
hoelzl@51775
  1766
   intermediate value theorem. Show connectedness of intervals on
hoelzl@51775
  1767
   linear_continuum_topology).
hoelzl@51732
  1768
hoelzl@51732
  1769
 - first_countable_topology from Multivariate_Analysis. Is used to
wenzelm@53983
  1770
   show equivalence of properties on the neighbourhood filter of x and
wenzelm@53983
  1771
   on all sequences converging to x.
wenzelm@53983
  1772
wenzelm@53983
  1773
 - FDERIV: Definition of has_derivative moved to Deriv.thy. Moved
wenzelm@53983
  1774
   theorems from Library/FDERIV.thy to Deriv.thy and base the
wenzelm@53983
  1775
   definition of DERIV on FDERIV. Add variants of DERIV and FDERIV
wenzelm@53983
  1776
   which are restricted to sets, i.e. to represent derivatives from
wenzelm@53983
  1777
   left or right.
hoelzl@51732
  1778
hoelzl@51732
  1779
 - Removed the within-filter. It is replaced by the principal filter:
hoelzl@51732
  1780
hoelzl@51732
  1781
     F within X = inf F (principal X)
hoelzl@51732
  1782
hoelzl@51732
  1783
 - Introduce "at x within U" as a single constant, "at x" is now an
hoelzl@51732
  1784
   abbreviation for "at x within UNIV"
hoelzl@51732
  1785
wenzelm@53983
  1786
 - Introduce named theorem collections tendsto_intros,
wenzelm@53983
  1787
   continuous_intros, continuous_on_intros and FDERIV_intros. Theorems
wenzelm@53983
  1788
   in tendsto_intros (or FDERIV_intros) are also available as
wenzelm@53983
  1789
   tendsto_eq_intros (or FDERIV_eq_intros) where the right-hand side
wenzelm@53983
  1790
   is replaced by a congruence rule. This allows to apply them as
wenzelm@53983
  1791
   intro rules and then proving equivalence by the simplifier.
hoelzl@51732
  1792
hoelzl@51732
  1793
 - Restructured theories in HOL-Complex_Main:
hoelzl@51732
  1794
hoelzl@51732
  1795
   + Moved RealDef and RComplete into Real
hoelzl@51732
  1796
hoelzl@51732
  1797
   + Introduced Topological_Spaces and moved theorems about
hoelzl@51732
  1798
     topological spaces, filters, limits and continuity to it
hoelzl@51732
  1799
hoelzl@51732
  1800
   + Renamed RealVector to Real_Vector_Spaces
hoelzl@51732
  1801
wenzelm@53983
  1802
   + Split Lim, SEQ, Series into Topological_Spaces,
wenzelm@53983
  1803
     Real_Vector_Spaces, and Limits
hoelzl@51732
  1804
hoelzl@51732
  1805
   + Moved Ln and Log to Transcendental
hoelzl@51732
  1806
hoelzl@51732
  1807
   + Moved theorems about continuity from Deriv to Topological_Spaces
hoelzl@51732
  1808
hoelzl@51732
  1809
 - Remove various auxiliary lemmas.
hoelzl@51732
  1810
hoelzl@51732
  1811
INCOMPATIBILITY.
hoelzl@51002
  1812
blanchet@53738
  1813
* Nitpick:
blanchet@55889
  1814
  - Added option "spy".
blanchet@55889
  1815
  - Reduce incidence of "too high arity" errors.
blanchet@53738
  1816
blanchet@51137
  1817
* Sledgehammer:
blanchet@51137
  1818
  - Renamed option:
blanchet@51137
  1819
      isar_shrink ~> isar_compress
blanchet@53738
  1820
    INCOMPATIBILITY.
blanchet@55889
  1821
  - Added options "isar_try0", "spy".
blanchet@55889
  1822
  - Better support for "isar_proofs".
blanchet@55889
  1823
  - MaSh has been fined-tuned and now runs as a local server.
blanchet@51137
  1824
wenzelm@54032
  1825
* Improved support for ad hoc overloading of constants (see also
wenzelm@54032
  1826
isar-ref manual and ~~/src/HOL/ex/Adhoc_Overloading_Examples.thy).
wenzelm@54032
  1827
wenzelm@54032
  1828
* Library/Polynomial.thy:
wenzelm@54032
  1829
  - Use lifting for primitive definitions.
wenzelm@54032
  1830
  - Explicit conversions from and to lists of coefficients, used for
wenzelm@54032
  1831
    generated code.
wenzelm@54032
  1832
  - Replaced recursion operator poly_rec by fold_coeffs.
wenzelm@54032
  1833
  - Prefer pre-existing gcd operation for gcd.
wenzelm@54032
  1834
  - Fact renames:
wenzelm@54032
  1835
    poly_eq_iff ~> poly_eq_poly_eq_iff
wenzelm@54032
  1836
    poly_ext ~> poly_eqI
wenzelm@54032
  1837
    expand_poly_eq ~> poly_eq_iff
wenzelm@54032
  1838
IMCOMPATIBILITY.
wenzelm@54032
  1839
wenzelm@54032
  1840
* New Library/Simps_Case_Conv.thy: Provides commands simps_of_case and
wenzelm@54032
  1841
case_of_simps to convert function definitions between a list of
wenzelm@54032
  1842
equations with patterns on the lhs and a single equation with case
wenzelm@54032
  1843
expressions on the rhs. See also Ex/Simps_Case_Conv_Examples.thy.
wenzelm@54032
  1844
wenzelm@54032
  1845
* New Library/FSet.thy: type of finite sets defined as a subtype of
wenzelm@54032
  1846
sets defined by Lifting/Transfer.
wenzelm@54032
  1847
wenzelm@54032
  1848
* Discontinued theory src/HOL/Library/Eval_Witness.  INCOMPATIBILITY.
wenzelm@54032
  1849
wenzelm@54032
  1850
* Consolidation of library theories on product orders:
wenzelm@54032
  1851
wenzelm@54032
  1852
    Product_Lattice ~> Product_Order -- pointwise order on products
wenzelm@54032
  1853
    Product_ord ~> Product_Lexorder -- lexicographic order on products
wenzelm@54032
  1854
wenzelm@54032
  1855
INCOMPATIBILITY.
wenzelm@54032
  1856
wenzelm@53160
  1857
* Imperative-HOL: The MREC combinator is considered legacy and no
wenzelm@53160
  1858
longer included by default. INCOMPATIBILITY, use partial_function
wenzelm@53160
  1859
instead, or import theory Legacy_Mrec as a fallback.
wenzelm@53160
  1860
wenzelm@53983
  1861
* HOL-Algebra: Discontinued theories ~~/src/HOL/Algebra/abstract and
wenzelm@53983
  1862
~~/src/HOL/Algebra/poly.  Existing theories should be based on
wenzelm@53983
  1863
~~/src/HOL/Library/Polynomial instead.  The latter provides
wenzelm@53983
  1864
integration with HOL's type classes for rings.  INCOMPATIBILITY.
ballarin@51517
  1865
wenzelm@54033
  1866
* HOL-BNF:
wenzelm@54032
  1867
  - Various improvements to BNF-based (co)datatype package, including
wenzelm@54032
  1868
    new commands "primrec_new", "primcorec", and
wenzelm@54032
  1869
    "datatype_new_compat", as well as documentation. See
wenzelm@54032
  1870
    "datatypes.pdf" for details.
wenzelm@54032
  1871
  - New "coinduction" method to avoid some boilerplate (compared to
wenzelm@54032
  1872
    coinduct).
wenzelm@54032
  1873
  - Renamed keywords:
wenzelm@54032
  1874
    data ~> datatype_new
wenzelm@54032
  1875
    codata ~> codatatype
wenzelm@54032
  1876
    bnf_def ~> bnf
wenzelm@54032
  1877
  - Renamed many generated theorems, including
wenzelm@54032
  1878
    discs ~> disc
wenzelm@54032
  1879
    map_comp' ~> map_comp
wenzelm@54032
  1880
    map_id' ~> map_id
wenzelm@54032
  1881
    sels ~> sel
wenzelm@54032
  1882
    set_map' ~> set_map
wenzelm@54032
  1883
    sets ~> set
wenzelm@54032
  1884
IMCOMPATIBILITY.
wenzelm@54032
  1885
ballarin@51517
  1886
wenzelm@51551
  1887
*** ML ***
wenzelm@51551
  1888
wenzelm@53971
  1889
* Spec_Check is a Quickcheck tool for Isabelle/ML.  The ML function
wenzelm@53971
  1890
"check_property" allows to check specifications of the form "ALL x y
wenzelm@53971
  1891
z. prop x y z".  See also ~~/src/Tools/Spec_Check/ with its
wenzelm@53971
  1892
Examples.thy in particular.
wenzelm@53971
  1893
wenzelm@53709
  1894
* Improved printing of exception trace in Poly/ML 5.5.1, with regular
wenzelm@53709
  1895
tracing output in the command transaction context instead of physical
wenzelm@53709
  1896
stdout.  See also Toplevel.debug, Toplevel.debugging and
wenzelm@53709
  1897
ML_Compiler.exn_trace.
wenzelm@53709
  1898
wenzelm@53971
  1899
* ML type "theory" is now immutable, without any special treatment of
wenzelm@53971
  1900
drafts or linear updates (which could lead to "stale theory" errors in
wenzelm@53971
  1901
the past).  Discontinued obsolete operations like Theory.copy,
wenzelm@53971
  1902
Theory.checkpoint, and the auxiliary type theory_ref.  Minor
wenzelm@53971
  1903
INCOMPATIBILITY.
wenzelm@53164
  1904
wenzelm@51551
  1905
* More uniform naming of goal functions for skipped proofs:
wenzelm@51551
  1906
wenzelm@51551
  1907
    Skip_Proof.prove  ~>  Goal.prove_sorry
wenzelm@51551
  1908
    Skip_Proof.prove_global  ~>  Goal.prove_sorry_global
wenzelm@51551
  1909
wenzelm@53971
  1910
Minor INCOMPATIBILITY.
wenzelm@51703
  1911
wenzelm@51717
  1912
* Simplifier tactics and tools use proper Proof.context instead of
wenzelm@51717
  1913
historic type simpset.  Old-style declarations like addsimps,
wenzelm@51717
  1914
addsimprocs etc. operate directly on Proof.context.  Raw type simpset
wenzelm@51717
  1915
retains its use as snapshot of the main Simplifier context, using
wenzelm@51717
  1916
simpset_of and put_simpset on Proof.context.  INCOMPATIBILITY -- port
wenzelm@51717
  1917
old tools by making them depend on (ctxt : Proof.context) instead of
wenzelm@51717
  1918
(ss : simpset), then turn (simpset_of ctxt) into ctxt.
wenzelm@51717
  1919
wenzelm@53971
  1920
* Modifiers for classical wrappers (e.g. addWrapper, delWrapper)
wenzelm@53971
  1921
operate on Proof.context instead of claset, for uniformity with addIs,
wenzelm@53971
  1922
addEs, addDs etc. Note that claset_of and put_claset allow to manage
wenzelm@53971
  1923
clasets separately from the context.
wenzelm@53971
  1924
wenzelm@51717
  1925
* Discontinued obsolete ML antiquotations @{claset} and @{simpset}.
wenzelm@51717
  1926
INCOMPATIBILITY, use @{context} instead.
wenzelm@51717
  1927
wenzelm@53971
  1928
* Antiquotation @{theory_context A} is similar to @{theory A}, but
wenzelm@53971
  1929
presents the result as initial Proof.context.
wenzelm@53971
  1930
wenzelm@51551
  1931
wenzelm@51398
  1932
*** System ***
wenzelm@51398
  1933
wenzelm@52052
  1934
* Discontinued obsolete isabelle usedir, mkdir, make -- superseded by
wenzelm@52052
  1935
"isabelle build" in Isabelle2013.  INCOMPATIBILITY.
wenzelm@51398
  1936
wenzelm@52054
  1937
* Discontinued obsolete isabelle-process options -f and -u (former
wenzelm@52054
  1938
administrative aliases of option -e).  Minor INCOMPATIBILITY.
wenzelm@52054
  1939
wenzelm@52550
  1940
* Discontinued obsolete isabelle print tool, and PRINT_COMMAND
wenzelm@52550
  1941
settings variable.
wenzelm@52550
  1942
wenzelm@52746
  1943
* Discontinued ISABELLE_DOC_FORMAT settings variable and historic
wenzelm@52746
  1944
document formats: dvi.gz, ps, ps.gz -- the default document format is
wenzelm@52746
  1945
always pdf.
wenzelm@52743
  1946
wenzelm@52053
  1947
* Isabelle settings variable ISABELLE_BUILD_JAVA_OPTIONS allows to
wenzelm@52053
  1948
specify global resources of the JVM process run by isabelle build.
wenzelm@52053
  1949
wenzelm@52116
  1950
* Toplevel executable $ISABELLE_HOME/bin/isabelle_scala_script allows
wenzelm@52116
  1951
to run Isabelle/Scala source files as standalone programs.
wenzelm@52116
  1952
wenzelm@52439
  1953
* Improved "isabelle keywords" tool (for old-style ProofGeneral
wenzelm@52439
  1954
keyword tables): use Isabelle/Scala operations, which inspect outer
wenzelm@52439
  1955
syntax without requiring to build sessions first.
wenzelm@52439
  1956
wenzelm@53971
  1957
* Sessions may be organized via 'chapter' specifications in the ROOT
wenzelm@53971
  1958
file, which determines a two-level hierarchy of browser info.  The old
wenzelm@53971
  1959
tree-like organization via implicit sub-session relation (with its
wenzelm@53971
  1960
tendency towards erratic fluctuation of URLs) has been discontinued.
wenzelm@53971
  1961
The default chapter is called "Unsorted".  Potential INCOMPATIBILITY
wenzelm@53971
  1962
for HTML presentation of theories.
wenzelm@53971
  1963
wenzelm@51398
  1964
wenzelm@51398
  1965
wenzelm@50993
  1966
New in Isabelle2013 (February 2013)
wenzelm@50993
  1967
-----------------------------------
wenzelm@47887
  1968
wenzelm@47967
  1969
*** General ***
wenzelm@47967
  1970
wenzelm@50126
  1971
* Theorem status about oracles and unfinished/failed future proofs is
wenzelm@50126
  1972
no longer printed by default, since it is incompatible with
wenzelm@50126
  1973
incremental / parallel checking of the persistent document model.  ML
wenzelm@50126
  1974
function Thm.peek_status may be used to inspect a snapshot of the
wenzelm@50126
  1975
ongoing evaluation process.  Note that in batch mode --- notably
wenzelm@50126
  1976
isabelle build --- the system ensures that future proofs of all
wenzelm@50126
  1977
accessible theorems in the theory context are finished (as before).
wenzelm@50126
  1978
wenzelm@49699
  1979
* Configuration option show_markup controls direct inlining of markup
wenzelm@49699
  1980
into the printed representation of formal entities --- notably type
wenzelm@49699
  1981
and sort constraints.  This enables Prover IDE users to retrieve that
wenzelm@49699
  1982
information via tooltips in the output window, for example.
wenzelm@49699
  1983
wenzelm@48890
  1984
* Command 'ML_file' evaluates ML text from a file directly within the
wenzelm@48890
  1985
theory, without any predeclaration via 'uses' in the theory header.
wenzelm@48890
  1986
wenzelm@49243
  1987
* Old command 'use' command and corresponding keyword 'uses' in the
wenzelm@49243
  1988
theory header are legacy features and will be discontinued soon.
wenzelm@49243
  1989
Tools that load their additional source files may imitate the
wenzelm@49243
  1990
'ML_file' implementation, such that the system can take care of
wenzelm@49243
  1991
dependencies properly.
wenzelm@49243
  1992
wenzelm@47967
  1993
* Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
wenzelm@47967
  1994
is called fastforce / fast_force_tac already since Isabelle2011-1.
wenzelm@47967
  1995
wenzelm@50110
  1996
* Updated and extended "isar-ref" and "implementation" manual, reduced
wenzelm@50110
  1997
remaining material in old "ref" manual.
wenzelm@48120
  1998
wenzelm@51050
  1999
* Improved support for auxiliary contexts that indicate block structure
wenzelm@51050
  2000
for specifications.  Nesting of "context fixes ... context assumes ..."
wenzelm@49841
  2001
and "class ... context ...".
wenzelm@49841
  2002
wenzelm@50772
  2003
* Attribute "consumes" allows a negative value as well, which is
wenzelm@50778
  2004
interpreted relatively to the total number of premises of the rule in
wenzelm@50772
  2005
the target context.  This form of declaration is stable when exported
wenzelm@50772
  2006
from a nested 'context' with additional assumptions.  It is the
wenzelm@50772
  2007
preferred form for definitional packages, notably cases/rules produced
wenzelm@50772
  2008
in HOL/inductive and HOL/function.
wenzelm@50772
  2009
wenzelm@49869
  2010
* More informative error messages for Isar proof commands involving
wenzelm@49869
  2011
lazy enumerations (method applications etc.).
wenzelm@49869
  2012
wenzelm@50213
  2013
* Refined 'help' command to retrieve outer syntax commands according
wenzelm@50213
  2014
to name patterns (with clickable results).
wenzelm@50213
  2015
wenzelm@47967
  2016
wenzelm@49968
  2017
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@49968
  2018
wenzelm@49968
  2019
* Parallel terminal proofs ('by') are enabled by default, likewise
wenzelm@49968
  2020
proofs that are built into packages like 'datatype', 'function'.  This
wenzelm@49968
  2021
allows to "run ahead" checking the theory specifications on the
wenzelm@49968
  2022
surface, while the prover is still crunching on internal
wenzelm@49968
  2023
justifications.  Unfinished / cancelled proofs are restarted as
wenzelm@49968
  2024
required to complete full proof checking eventually.
wenzelm@49968
  2025
wenzelm@49968
  2026
* Improved output panel with tooltips, hyperlinks etc. based on the
wenzelm@49968
  2027
same Rich_Text_Area as regular Isabelle/jEdit buffers.  Activation of
wenzelm@49968
  2028
tooltips leads to some window that supports the same recursively,
wenzelm@49968
  2029
which can lead to stacks of tooltips as the semantic document content
wenzelm@49968
  2030
is explored.  ESCAPE closes the whole stack, individual windows may be
wenzelm@49968
  2031
closed separately, or detached to become independent jEdit dockables.
wenzelm@49968
  2032
wenzelm@50717
  2033
* Improved support for commands that produce graph output: the text
wenzelm@50717
  2034
message contains a clickable area to open a new instance of the graph
wenzelm@50717
  2035
browser on demand.
wenzelm@50717
  2036
wenzelm@49968
  2037
* More robust incremental parsing of outer syntax (partial comments,
wenzelm@49968
  2038
malformed symbols).  Changing the balance of open/close quotes and
wenzelm@49968
  2039
comment delimiters works more conveniently with unfinished situations
wenzelm@49968
  2040
that frequently occur in user interaction.
wenzelm@49968
  2041
wenzelm@49968
  2042
* More efficient painting and improved reactivity when editing large
wenzelm@49968
  2043
files.  More scalable management of formal document content.
wenzelm@49968
  2044
wenzelm@50505
  2045
* Smarter handling of tracing messages: prover process pauses after
wenzelm@50505
  2046
certain number of messages per command transaction, with some user
wenzelm@50505
  2047
dialog to stop or continue.  This avoids swamping the front-end with
wenzelm@50119
  2048
potentially infinite message streams.
wenzelm@49968
  2049
wenzelm@49968
  2050
* More plugin options and preferences, based on Isabelle/Scala.  The
wenzelm@49968
  2051
jEdit plugin option panel provides access to some Isabelle/Scala
wenzelm@49968
  2052
options, including tuning parameters for editor reactivity and color
wenzelm@49968
  2053
schemes.
wenzelm@49968
  2054
wenzelm@50184
  2055
* Dockable window "Symbols" provides some editing support for Isabelle
wenzelm@50184
  2056
symbols.
wenzelm@50184
  2057
wenzelm@51082
  2058
* Dockable window "Monitor" shows ML runtime statistics.  Note that
wenzelm@51082
  2059
continuous display of the chart slows down the system.
wenzelm@50701
  2060
wenzelm@50183
  2061
* Improved editing support for control styles: subscript, superscript,
wenzelm@50183
  2062
bold, reset of style -- operating on single symbols or text
wenzelm@50198
  2063
selections.  Cf. keyboard shortcuts C+e DOWN/UP/RIGHT/LEFT.
wenzelm@50198
  2064
wenzelm@50198
  2065
* Actions isabelle.increase-font-size and isabelle.decrease-font-size
wenzelm@50198
  2066
adjust the main text area font size, and its derivatives for output,
wenzelm@50836
  2067
tooltips etc.  Cf. keyboard shortcuts C-PLUS and C-MINUS, which often
wenzelm@50836
  2068
need to be adapted to local keyboard layouts.
wenzelm@50183
  2069
wenzelm@50730
  2070
* More reactive completion popup by default: use \t (TAB) instead of
wenzelm@50730
  2071
\n (NEWLINE) to minimize intrusion into regular flow of editing.  See
wenzelm@50730
  2072
also "Plugin Options / SideKick / General / Code Completion Options".
wenzelm@50730
  2073
wenzelm@50406
  2074
* Implicit check and build dialog of the specified logic session
wenzelm@50406
  2075
image.  For example, HOL, HOLCF, HOL-Nominal can be produced on
wenzelm@50406
  2076
demand, without bundling big platform-dependent heap images in the
wenzelm@50406
  2077
Isabelle distribution.
wenzelm@50406
  2078
wenzelm@49968
  2079
* Uniform Java 7 platform on Linux, Mac OS X, Windows: recent updates
wenzelm@49968
  2080
from Oracle provide better multi-platform experience.  This version is
wenzelm@49968
  2081
now bundled exclusively with Isabelle.
wenzelm@49968
  2082
wenzelm@49968
  2083
wenzelm@48205
  2084
*** Pure ***
wenzelm@48205
  2085
haftmann@48431
  2086
* Code generation for Haskell: restrict unqualified imports from
haftmann@48431
  2087
Haskell Prelude to a small set of fundamental operations.
haftmann@48431
  2088
wenzelm@50646
  2089
* Command 'export_code': relative file names are interpreted
wenzelm@50646
  2090
relatively to master directory of current theory rather than the
wenzelm@50646
  2091
rather arbitrary current working directory.  INCOMPATIBILITY.
haftmann@48371
  2092
wenzelm@48205
  2093
* Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
wenzelm@48205
  2094
use regular rule composition via "OF" / "THEN", or explicit proof
wenzelm@48205
  2095
structure instead.  Note that Isabelle/ML provides a variety of
wenzelm@48205
  2096
operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
wenzelm@48205
  2097
with some care where this is really required.
wenzelm@48205
  2098
wenzelm@48792
  2099
* Command 'typ' supports an additional variant with explicit sort
wenzelm@48792
  2100
constraint, to infer and check the most general type conforming to a
wenzelm@51063
  2101
given sort.  Example (in HOL):
wenzelm@48792
  2102
wenzelm@48792
  2103
  typ "_ * _ * bool * unit" :: finite
wenzelm@48792
  2104
wenzelm@50716
  2105
* Command 'locale_deps' visualizes all locales and their relations as
wenzelm@50716
  2106
a Hasse diagram.
wenzelm@50716
  2107
wenzelm@48205
  2108
bulwahn@48013
  2109
*** HOL ***
bulwahn@48013
  2110
wenzelm@50646
  2111
* Sledgehammer:
wenzelm@50646
  2112
wenzelm@50646
  2113
  - Added MaSh relevance filter based on machine-learning; see the
wenzelm@50646
  2114
    Sledgehammer manual for details.
wenzelm@50646
  2115
  - Polished Isar proofs generated with "isar_proofs" option.
wenzelm@50646
  2116
  - Rationalized type encodings ("type_enc" option).
blanchet@50720
  2117
  - Renamed "kill_provers" subcommand to "kill_all".
wenzelm@50646
  2118
  - Renamed options:
wenzelm@50646
  2119
      isar_proof ~> isar_proofs
wenzelm@50646
  2120
      isar_shrink_factor ~> isar_shrink
wenzelm@50646
  2121
      max_relevant ~> max_facts
wenzelm@50646
  2122
      relevance_thresholds ~> fact_thresholds
wenzelm@50646
  2123
wenzelm@50646
  2124
* Quickcheck: added an optimisation for equality premises.  It is
wenzelm@50646
  2125
switched on by default, and can be switched off by setting the
wenzelm@50646
  2126
configuration quickcheck_optimise_equality to false.
wenzelm@50646
  2127
kuncar@50878
  2128
* Quotient: only one quotient can be defined by quotient_type
kuncar@50878
  2129
INCOMPATIBILITY.
kuncar@50878
  2130
kuncar@50878
  2131
* Lifting:
kuncar@50878
  2132
  - generation of an abstraction function equation in lift_definition
kuncar@50878
  2133
  - quot_del attribute
kuncar@50878
  2134
  - renamed no_abs_code -> no_code (INCOMPATIBILITY.)
kuncar@50878
  2135
wenzelm@50646
  2136
* Simproc "finite_Collect" rewrites set comprehensions into pointfree
wenzelm@50646
  2137
expressions.
wenzelm@50646
  2138
wenzelm@50646
  2139
* Preprocessing of the code generator rewrites set comprehensions into
wenzelm@50646
  2140
pointfree expressions.
wenzelm@50646
  2141
wenzelm@50646
  2142
* The SMT solver Z3 has now by default a restricted set of directly
wenzelm@50646
  2143
supported features. For the full set of features (div/mod, nonlinear
wenzelm@50646
  2144
arithmetic, datatypes/records) with potential proof reconstruction
wenzelm@50646
  2145
failures, enable the configuration option "z3_with_extensions".  Minor
wenzelm@50646
  2146
INCOMPATIBILITY.
haftmann@49948
  2147
wenzelm@49836
  2148
* Simplified 'typedef' specifications: historical options for implicit
wenzelm@49836
  2149
set definition and alternative name have been discontinued.  The
wenzelm@49836
  2150
former behavior of "typedef (open) t = A" is now the default, but
wenzelm@49836
  2151
written just "typedef t = A".  INCOMPATIBILITY, need to adapt theories
wenzelm@49836
  2152
accordingly.
wenzelm@49836
  2153
wenzelm@50646
  2154
* Removed constant "chars"; prefer "Enum.enum" on type "char"
wenzelm@50646
  2155
directly.  INCOMPATIBILITY.
wenzelm@50646
  2156
wenzelm@50646
  2157
* Moved operation product, sublists and n_lists from theory Enum to
wenzelm@50646
  2158
List.  INCOMPATIBILITY.
haftmann@49822
  2159
haftmann@49739
  2160
* Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
haftmann@49739
  2161
haftmann@49738
  2162
* Class "comm_monoid_diff" formalises properties of bounded
haftmann@49388
  2163
subtraction, with natural numbers and multisets as typical instances.
haftmann@49388
  2164
wenzelm@50646
  2165
* Added combinator "Option.these" with type "'a option set => 'a set".
wenzelm@50646
  2166
wenzelm@50646
  2167
* Theory "Transitive_Closure": renamed lemmas
wenzelm@50646
  2168
wenzelm@50646
  2169
  reflcl_tranclp -> reflclp_tranclp
wenzelm@50646
  2170
  rtranclp_reflcl -> rtranclp_reflclp
wenzelm@50646
  2171
wenzelm@50646
  2172
INCOMPATIBILITY.
wenzelm@50646
  2173
wenzelm@50646
  2174
* Theory "Rings": renamed lemmas (in class semiring)
wenzelm@50646
  2175
wenzelm@50646
  2176
  left_distrib ~> distrib_right
wenzelm@50646
  2177
  right_distrib ~> distrib_left
wenzelm@50646
  2178
wenzelm@50646
  2179
INCOMPATIBILITY.
wenzelm@50646
  2180
wenzelm@50646
  2181
* Generalized the definition of limits:
wenzelm@50646
  2182
wenzelm@50646
  2183
  - Introduced the predicate filterlim (LIM x F. f x :> G) which
wenzelm@50646
  2184
    expresses that when the input values x converge to F then the
wenzelm@50646
  2185
    output f x converges to G.
wenzelm@50646
  2186
wenzelm@50646
  2187
  - Added filters for convergence to positive (at_top) and negative
wenzelm@50646
  2188
    infinity (at_bot).
wenzelm@50646
  2189
wenzelm@50646
  2190
  - Moved infinity in the norm (at_infinity) from
wenzelm@50646
  2191
    Multivariate_Analysis to Complex_Main.
wenzelm@50646
  2192
wenzelm@50646
  2193
  - Removed real_tendsto_inf, it is superseded by "LIM x F. f x :>
wenzelm@50646
  2194
    at_top".
wenzelm@50646
  2195
wenzelm@50646
  2196
INCOMPATIBILITY.
wenzelm@50646
  2197
wenzelm@50646
  2198
* Theory "Library/Option_ord" provides instantiation of option type to
wenzelm@50646
  2199
lattice type classes.
wenzelm@50646
  2200
wenzelm@50646
  2201
* Theory "Library/Multiset": renamed
wenzelm@50646
  2202
wenzelm@50646
  2203
    constant fold_mset ~> Multiset.fold
wenzelm@50646
  2204
    fact fold_mset_commute ~> fold_mset_comm
wenzelm@50646
  2205
wenzelm@50646
  2206
INCOMPATIBILITY.
wenzelm@50646
  2207
wenzelm@50646
  2208
* Renamed theory Library/List_Prefix to Library/Sublist, with related
wenzelm@50646
  2209
changes as follows.
wenzelm@50646
  2210
wenzelm@50646
  2211
  - Renamed constants (and related lemmas)
Christian@49145
  2212
Christian@49145
  2213
      prefix ~> prefixeq
Christian@49145
  2214
      strict_prefix ~> prefix
Christian@49145
  2215
wenzelm@50646
  2216
  - Replaced constant "postfix" by "suffixeq" with swapped argument
wenzelm@50646
  2217
    order (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped
wenzelm@50646
  2218
    old infix syntax "xs >>= ys"; use "suffixeq ys xs" instead.
wenzelm@50646
  2219
    Renamed lemmas accordingly.
wenzelm@50646
  2220
wenzelm@50646
  2221
  - Added constant "list_hembeq" for homeomorphic embedding on
wenzelm@50646
  2222
    lists. Added abbreviation "sublisteq" for special case
wenzelm@50646
  2223
    "list_hembeq (op =)".
wenzelm@50646
  2224
wenzelm@50646
  2225
  - Theory Library/Sublist no longer provides "order" and "bot" type
wenzelm@50646
  2226
    class instances for the prefix order (merely corresponding locale
wenzelm@50646
  2227
    interpretations). The type class instances are now in theory
wenzelm@50646
  2228
    Library/Prefix_Order.
wenzelm@50646
  2229
wenzelm@50646
  2230
  - The sublist relation of theory Library/Sublist_Order is now based
wenzelm@50646
  2231
    on "Sublist.sublisteq".  Renamed lemmas accordingly:
Christian@50516
  2232
Christian@50516
  2233
      le_list_append_le_same_iff ~> Sublist.sublisteq_append_le_same_iff
Christian@50516
  2234
      le_list_append_mono ~> Sublist.list_hembeq_append_mono
Christian@50516
  2235
      le_list_below_empty ~> Sublist.list_hembeq_Nil, Sublist.list_hembeq_Nil2
Christian@50516
  2236
      le_list_Cons_EX ~> Sublist.list_hembeq_ConsD
Christian@50516
  2237
      le_list_drop_Cons2 ~> Sublist.sublisteq_Cons2'
Christian@50516
  2238
      le_list_drop_Cons_neq ~> Sublist.sublisteq_Cons2_neq
Christian@50516
  2239
      le_list_drop_Cons ~> Sublist.sublisteq_Cons'
Christian@50516
  2240
      le_list_drop_many ~> Sublist.sublisteq_drop_many
Christian@50516
  2241
      le_list_filter_left ~> Sublist.sublisteq_filter_left
Christian@50516
  2242
      le_list_rev_drop_many ~> Sublist.sublisteq_rev_drop_many
Christian@50516
  2243
      le_list_rev_take_iff ~> Sublist.sublisteq_append
Christian@50516
  2244
      le_list_same_length ~> Sublist.sublisteq_same_length
Christian@50516
  2245
      le_list_take_many_iff ~> Sublist.sublisteq_append'
Christian@49145
  2246
      less_eq_list.drop ~> less_eq_list_drop
Christian@49145
  2247
      less_eq_list.induct ~> less_eq_list_induct
Christian@50516
  2248
      not_le_list_length ~> Sublist.not_sublisteq_length
Christian@49145
  2249
wenzelm@50646
  2250
INCOMPATIBILITY.
wenzelm@50646
  2251
wenzelm@50646
  2252
* New theory Library/Countable_Set.
wenzelm@50646
  2253
wenzelm@50646
  2254
* Theory Library/Debug and Library/Parallel provide debugging and
wenzelm@50646
  2255
parallel execution for code generated towards Isabelle/ML.
wenzelm@50646
  2256
wenzelm@50646
  2257
* Theory Library/FuncSet: Extended support for Pi and extensional and
wenzelm@50646
  2258
introduce the extensional dependent function space "PiE". Replaced
wenzelm@50646
  2259
extensional_funcset by an abbreviation, and renamed lemmas from
wenzelm@50646
  2260
extensional_funcset to PiE as follows:
wenzelm@50646
  2261
wenzelm@50646
  2262
  extensional_empty  ~>  PiE_empty
wenzelm@50646
  2263
  extensional_funcset_empty_domain  ~>  PiE_empty_domain
wenzelm@50646
  2264
  extensional_funcset_empty_range  ~>  PiE_empty_range
wenzelm@50646
  2265
  extensional_funcset_arb  ~>  PiE_arb
wenzelm@50646
  2266
  extensional_funcset_mem  ~>  PiE_mem
wenzelm@50646
  2267
  extensional_funcset_extend_domainI  ~>  PiE_fun_upd
wenzelm@50646
  2268
  extensional_funcset_restrict_domain  ~>  fun_upd_in_PiE
wenzelm@50646
  2269
  extensional_funcset_extend_domain_eq  ~>  PiE_insert_eq
wenzelm@50646
  2270
  card_extensional_funcset  ~>  card_PiE
wenzelm@50646
  2271
  finite_extensional_funcset  ~>  finite_PiE
wenzelm@50646
  2272
wenzelm@50646
  2273
INCOMPATIBILITY.
wenzelm@50646
  2274
wenzelm@50646
  2275
* Theory Library/FinFun: theory of almost everywhere constant
wenzelm@50646
  2276
functions (supersedes the AFP entry "Code Generation for Functions as
wenzelm@50646
  2277
Data").
wenzelm@50646
  2278
wenzelm@50646
  2279
* Theory Library/Phantom: generic phantom type to make a type
wenzelm@50646
  2280
parameter appear in a constant's type.  This alternative to adding
wenzelm@50646
  2281
TYPE('a) as another parameter avoids unnecessary closures in generated
wenzelm@50646
  2282
code.
wenzelm@50646
  2283
wenzelm@50646
  2284
* Theory Library/RBT_Impl: efficient construction of red-black trees
wenzelm@50646
  2285
from sorted associative lists. Merging two trees with rbt_union may
wenzelm@50646
  2286
return a structurally different tree than before.  Potential
wenzelm@50646
  2287
INCOMPATIBILITY.
wenzelm@50646
  2288
wenzelm@50646
  2289
* Theory Library/IArray: immutable arrays with code generation.
wenzelm@50646
  2290
wenzelm@50646
  2291
* Theory Library/Finite_Lattice: theory of finite lattices.
wenzelm@50646
  2292
wenzelm@50646
  2293
* HOL/Multivariate_Analysis: replaced
wenzelm@50646
  2294
wenzelm@50646
  2295
  "basis :: 'a::euclidean_space => nat => real"
wenzelm@50646
  2296
  "\<Chi>\<Chi> :: (nat => real) => 'a::euclidean_space"
wenzelm@50646
  2297
wenzelm@50646
  2298
on euclidean spaces by using the inner product "_ \<bullet> _" with
wenzelm@50646
  2299
vectors from the Basis set: "\<Chi>\<Chi> i. f i" is superseded by
wenzelm@50646
  2300
"SUM i : Basis. f i * r i".
wenzelm@50646
  2301
wenzelm@50646
  2302
  With this change the following constants are also changed or removed:
wenzelm@50646
  2303
wenzelm@50646
  2304
    DIM('a) :: nat  ~>  card (Basis :: 'a set)   (is an abbreviation)
wenzelm@50646
  2305
    a $$ i  ~>  inner a i  (where i : Basis)
wenzelm@50646
  2306
    cart_base i  removed
wenzelm@50646
  2307
    \<pi>, \<pi>'  removed
hoelzl@50526
  2308
hoelzl@50526
  2309
  Theorems about these constants where removed.
hoelzl@50526
  2310
hoelzl@50526
  2311
  Renamed lemmas:
hoelzl@50526
  2312
wenzelm@50646
  2313
    component_le_norm  ~>  Basis_le_norm
wenzelm@50646
  2314
    euclidean_eq  ~>  euclidean_eq_iff
wenzelm@50646
  2315
    differential_zero_maxmin_component  ~>  differential_zero_maxmin_cart
wenzelm@50646
  2316
    euclidean_simps  ~>  inner_simps
wenzelm@50646
  2317
    independent_basis  ~>  independent_Basis
wenzelm@50646
  2318
    span_basis  ~>  span_Basis
wenzelm@50646
  2319
    in_span_basis  ~>  in_span_Basis
wenzelm@50646
  2320
    norm_bound_component_le  ~>  norm_boound_Basis_le
wenzelm@50646
  2321
    norm_bound_component_lt  ~>  norm_boound_Basis_lt
wenzelm@50646
  2322
    component_le_infnorm  ~>  Basis_le_infnorm
wenzelm@50646
  2323
wenzelm@50646
  2324
INCOMPATIBILITY.
hoelzl@50526
  2325
hoelzl@50141
  2326
* HOL/Probability:
wenzelm@50646
  2327
wenzelm@50646
  2328
  - Added simproc "measurable" to automatically prove measurability.
wenzelm@50646
  2329
wenzelm@50646
  2330
  - Added induction rules for sigma sets with disjoint union
wenzelm@50646
  2331
    (sigma_sets_induct_disjoint) and for Borel-measurable functions
wenzelm@50646
  2332
    (borel_measurable_induct).
wenzelm@50646
  2333
wenzelm@50646
  2334
  - Added the Daniell-Kolmogorov theorem (the existence the limit of a
wenzelm@50646
  2335
    projective family).
wenzelm@50646
  2336
wenzelm@50646
  2337
* HOL/Cardinals: Theories of ordinals and cardinals (supersedes the
wenzelm@50646
  2338
AFP entry "Ordinals_and_Cardinals").
wenzelm@50646
  2339
wenzelm@50646
  2340
* HOL/BNF: New (co)datatype package based on bounded natural functors
wenzelm@50646
  2341
with support for mixed, nested recursion and interesting non-free
wenzelm@50646
  2342
datatypes.
blanchet@48094
  2343
wenzelm@50991
  2344
* HOL/Finite_Set and Relation: added new set and relation operations
kuncar@50878
  2345
expressed by Finite_Set.fold.
kuncar@50878
  2346
kuncar@50878
  2347
* New theory HOL/Library/RBT_Set: implementation of sets by red-black
kuncar@50878
  2348
trees for the code generator.
kuncar@50878
  2349
kuncar@50878
  2350
* HOL/Library/RBT and HOL/Library/Mapping have been converted to
kuncar@50878
  2351
Lifting/Transfer.
kuncar@50878
  2352
possible INCOMPATIBILITY.
kuncar@50878
  2353
kuncar@50878
  2354
* HOL/Set: renamed Set.project -> Set.filter
kuncar@50878
  2355
INCOMPATIBILITY.
kuncar@50878
  2356
wenzelm@48120
  2357
wenzelm@48206
  2358
*** Document preparation ***
wenzelm@48206
  2359
wenzelm@50646
  2360
* Dropped legacy antiquotations "term_style" and "thm_style", since
wenzelm@50646
  2361
styles may be given as arguments to "term" and "thm" already.
wenzelm@50646
  2362
Discontinued legacy styles "prem1" .. "prem19".
wenzelm@50646
  2363
wenzelm@50646
  2364
* Default LaTeX rendering for \<euro> is now based on eurosym package,
wenzelm@50646
  2365
instead of slightly exotic babel/greek.
wenzelm@48206
  2366
wenzelm@48616
  2367
* Document variant NAME may use different LaTeX entry point
wenzelm@48616
  2368
document/root_NAME.tex if that file exists, instead of the common
wenzelm@48616
  2369
document/root.tex.
wenzelm@48616
  2370
wenzelm@48657
  2371
* Simplified custom document/build script, instead of old-style
wenzelm@48657
  2372
document/IsaMakefile.  Minor INCOMPATIBILITY.
wenzelm@48657
  2373
wenzelm@48206
  2374
wenzelm@48992
  2375
*** ML ***
wenzelm@48992
  2376
wenzelm@50646
  2377
* The default limit for maximum number of worker threads is now 8,
wenzelm@50646
  2378
instead of 4, in correspondence to capabilities of contemporary
wenzelm@50646
  2379
hardware and Poly/ML runtime system.
wenzelm@50646
  2380
wenzelm@49869
  2381
* Type Seq.results and related operations support embedded error
wenzelm@49869
  2382
messages within lazy enumerations, and thus allow to provide
wenzelm@49869
  2383
informative errors in the absence of any usable results.
wenzelm@49869
  2384
wenzelm@48992
  2385
* Renamed Position.str_of to Position.here to emphasize that this is a
wenzelm@48992
  2386
formal device to inline positions into message text, but not
wenzelm@48992
  2387
necessarily printing visible text.
wenzelm@48992
  2388
wenzelm@48992
  2389
wenzelm@48206
  2390
*** System ***
wenzelm@48206
  2391
wenzelm@48585
  2392
* Advanced support for Isabelle sessions and build management, see
wenzelm@48585
  2393
"system" manual for the chapter of that name, especially the "isabelle
wenzelm@51056
  2394
build" tool and its examples.  The "isabelle mkroot" tool prepares
wenzelm@51056
  2395
session root directories for use with "isabelle build", similar to
wenzelm@51056
  2396
former "isabelle mkdir" for "isabelle usedir".  Note that this affects
wenzelm@51056
  2397
document preparation as well.  INCOMPATIBILITY, isabelle usedir /
wenzelm@48736
  2398
mkdir / make are rendered obsolete.
wenzelm@48736
  2399
wenzelm@51056
  2400
* Discontinued obsolete Isabelle/build script, it is superseded by the
wenzelm@51056
  2401
regular isabelle build tool.  For example:
wenzelm@51056
  2402
wenzelm@51056
  2403
  isabelle build -s -b HOL
wenzelm@51056
  2404
wenzelm@48736
  2405
* Discontinued obsolete "isabelle makeall".
wenzelm@48585
  2406
wenzelm@48722
  2407
* Discontinued obsolete IsaMakefile and ROOT.ML files from the
wenzelm@48722
  2408
Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
wenzelm@48722
  2409
provides some traditional targets that invoke "isabelle build".  Note
wenzelm@48722
  2410
that this is inefficient!  Applications of Isabelle/HOL involving
wenzelm@48722
  2411
"isabelle make" should be upgraded to use "isabelle build" directly.
wenzelm@48722
  2412
wenzelm@48693
  2413
* The "isabelle options" tool prints Isabelle system options, as
wenzelm@48693
  2414
required for "isabelle build", for example.
wenzelm@48693
  2415
wenzelm@50646
  2416
* The "isabelle logo" tool produces EPS and PDF format simultaneously.
wenzelm@50646
  2417
Minor INCOMPATIBILITY in command-line options.
wenzelm@50646
  2418
wenzelm@50646
  2419
* The "isabelle install" tool has now a simpler command-line.  Minor
wenzelm@50646
  2420
INCOMPATIBILITY.
wenzelm@50646
  2421
wenzelm@48844
  2422
* The "isabelle components" tool helps to resolve add-on components
wenzelm@48844
  2423
that are not bundled, or referenced from a bare-bones repository
wenzelm@48844
  2424
version of Isabelle.
wenzelm@48844
  2425
wenzelm@50646
  2426
* Settings variable ISABELLE_PLATFORM_FAMILY refers to the general
wenzelm@50646
  2427
platform family: "linux", "macos", "windows".
wenzelm@50646
  2428
wenzelm@50646
  2429
* The ML system is configured as regular component, and no longer
wenzelm@50646
  2430
picked up from some surrounding directory.  Potential INCOMPATIBILITY
wenzelm@50646
  2431
for home-made settings.
wenzelm@50132
  2432
wenzelm@50701
  2433
* Improved ML runtime statistics (heap, threads, future tasks etc.).
wenzelm@50701
  2434
wenzelm@48206
  2435
* Discontinued support for Poly/ML 5.2.1, which was the last version
wenzelm@48206
  2436
without exception positions and advanced ML compiler/toplevel
wenzelm@48206
  2437
configuration.
wenzelm@48206
  2438
wenzelm@48574
  2439
* Discontinued special treatment of Proof General -- no longer guess
wenzelm@48574
  2440
PROOFGENERAL_HOME based on accidental file-system layout.  Minor
wenzelm@48574
  2441
INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
wenzelm@48574
  2442
settings manually, or use a Proof General version that has been
wenzelm@48574
  2443
bundled as Isabelle component.
wenzelm@48206
  2444
wenzelm@50182
  2445
wenzelm@48120
  2446
wenzelm@47462
  2447
New in Isabelle2012 (May 2012)
wenzelm@47462
  2448
------------------------------
wenzelm@45109
  2449
wenzelm@45593
  2450
*** General ***
wenzelm@45593
  2451
wenzelm@45614
  2452
* Prover IDE (PIDE) improvements:
wenzelm@45614
  2453
wenzelm@47585
  2454
  - more robust Sledgehammer integration (as before the sledgehammer
wenzelm@47806
  2455
    command-line needs to be typed into the source buffer)
wenzelm@45614
  2456
  - markup for bound variables
wenzelm@47806
  2457
  - markup for types of term variables (displayed as tooltips)
wenzelm@46956
  2458
  - support for user-defined Isar commands within the running session
wenzelm@47158
  2459
  - improved support for Unicode outside original 16bit range
wenzelm@47158
  2460
    e.g. glyph for \<A> (thanks to jEdit 4.5.1)
wenzelm@45614
  2461
wenzelm@47806
  2462
* Forward declaration of outer syntax keywords within the theory
wenzelm@47806
  2463
header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
wenzelm@47806
  2464
commands to be used in the same theory where defined.
wenzelm@46485
  2465
wenzelm@47482
  2466
* Auxiliary contexts indicate block structure for specifications with
wenzelm@47482
  2467
additional parameters and assumptions.  Such unnamed contexts may be
wenzelm@47482
  2468
nested within other targets, like 'theory', 'locale', 'class',
wenzelm@47482
  2469
'instantiation' etc.  Results from the local context are generalized
wenzelm@47482
  2470
accordingly and applied to the enclosing target context.  Example:
wenzelm@47482
  2471
wenzelm@47482
  2472
  context
wenzelm@47482
  2473
    fixes x y z :: 'a
wenzelm@47482
  2474
    assumes xy: "x = y" and yz: "y = z"
wenzelm@47482
  2475
  begin
wenzelm@47482
  2476
wenzelm@47482
  2477
  lemma my_trans: "x = z" using xy yz by simp
wenzelm@47482
  2478
wenzelm@47482
  2479
  end
wenzelm@47482
  2480
wenzelm@47482
  2481
  thm my_trans
wenzelm@47482
  2482
wenzelm@47482
  2483
The most basic application is to factor-out context elements of
wenzelm@47482
  2484
several fixes/assumes/shows theorem statements, e.g. see
wenzelm@47482
  2485
~~/src/HOL/Isar_Examples/Group_Context.thy
wenzelm@47482
  2486
wenzelm@47482
  2487
Any other local theory specification element works within the "context
wenzelm@47482
  2488
... begin ... end" block as well.
wenzelm@47482
  2489
wenzelm@47484
  2490
* Bundled declarations associate attributed fact expressions with a
wenzelm@47484
  2491
given name in the context.  These may be later included in other
wenzelm@47484
  2492
contexts.  This allows to manage context extensions casually, without
wenzelm@47855
  2493
the logical dependencies of locales and locale interpretation.  See
wenzelm@47855
  2494
commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
wenzelm@47484
  2495
wenzelm@47829
  2496
* Commands 'lemmas' and 'theorems' allow local variables using 'for'
wenzelm@47829
  2497
declaration, and results are standardized before being stored.  Thus
wenzelm@47829
  2498
old-style "standard" after instantiation or composition of facts
wenzelm@47829
  2499
becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
wenzelm@47829
  2500
indices of schematic variables.
wenzelm@47829
  2501
wenzelm@47829
  2502
* Rule attributes in local theory declarations (e.g. locale or class)
wenzelm@47829
  2503
are now statically evaluated: the resulting theorem is stored instead
wenzelm@47829
  2504
of the original expression.  INCOMPATIBILITY in rare situations, where
wenzelm@47829
  2505
the historic accident of dynamic re-evaluation in interpretations
wenzelm@47829
  2506
etc. was exploited.
wenzelm@47829
  2507
wenzelm@47829
  2508
* New tutorial "Programming and Proving in Isabelle/HOL"
wenzelm@47829
  2509
("prog-prove").  It completely supersedes "A Tutorial Introduction to
wenzelm@47829
  2510
Structured Isar Proofs" ("isar-overview"), which has been removed.  It
wenzelm@47829
  2511
also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
wenzelm@47829
  2512
Logic" as the recommended beginners tutorial, but does not cover all
wenzelm@47829
  2513
of the material of that old tutorial.
wenzelm@47829
  2514
wenzelm@47829
  2515
* Updated and extended reference manuals: "isar-ref",
wenzelm@47829
  2516
"implementation", "system"; reduced remaining material in old "ref"
wenzelm@47829
  2517
manual.
wenzelm@47829
  2518
wenzelm@47829
  2519
wenzelm@47829
  2520
*** Pure ***
wenzelm@47829
  2521
wenzelm@46976
  2522
* Command 'definition' no longer exports the foundational "raw_def"
wenzelm@46976
  2523
into the user context.  Minor INCOMPATIBILITY, may use the regular
wenzelm@46976
  2524
"def" result with attribute "abs_def" to imitate the old version.
wenzelm@46976
  2525
wenzelm@47855
  2526
* Attribute "abs_def" turns an equation of the form "f x y == t" into
wenzelm@47855
  2527
"f == %x y. t", which ensures that "simp" or "unfold" steps always
wenzelm@47855
  2528
expand it.  This also works for object-logic equality.  (Formerly
wenzelm@47855
  2529
undocumented feature.)
wenzelm@47855
  2530
wenzelm@47856
  2531
* Sort constraints are now propagated in simultaneous statements, just
wenzelm@47856
  2532
like type constraints.  INCOMPATIBILITY in rare situations, where
wenzelm@47856
  2533
distinct sorts used to be assigned accidentally.  For example:
wenzelm@47856
  2534
wenzelm@47856
  2535
  lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
wenzelm@47856
  2536
wenzelm@47856
  2537
  lemma "P (x::'a)" and "Q (y::'a::bar)"
wenzelm@47856
  2538
    -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
wenzelm@47856
  2539
wenzelm@47856
  2540
* Rule composition via attribute "OF" (or ML functions OF/MRS) is more
wenzelm@47856
  2541
tolerant against multiple unifiers, as long as the final result is
wenzelm@47856
  2542
unique.  (As before, rules are composed in canonical right-to-left
wenzelm@47856
  2543
order to accommodate newly introduced premises.)
wenzelm@47856
  2544
wenzelm@47806
  2545
* Renamed some inner syntax categories:
wenzelm@47806
  2546
wenzelm@47806
  2547
    num ~> num_token
wenzelm@47806
  2548
    xnum ~> xnum_token
wenzelm@47806
  2549
    xstr ~> str_token
wenzelm@47806
  2550
wenzelm@47806
  2551
Minor INCOMPATIBILITY.  Note that in practice "num_const" or
wenzelm@47806
  2552
"num_position" etc. are mainly used instead (which also include
wenzelm@47806
  2553
position information via constraints).
wenzelm@47806
  2554
wenzelm@47829
  2555
* Simplified configuration options for syntax ambiguity: see
wenzelm@47829
  2556
"syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
wenzelm@47829
  2557
manual.  Minor INCOMPATIBILITY.
wenzelm@47829
  2558
wenzelm@47856
  2559
* Discontinued configuration option "syntax_positions": atomic terms
wenzelm@47856
  2560
in parse trees are always annotated by position constraints.
wenzelm@45134
  2561
wenzelm@47464
  2562
* Old code generator for SML and its commands 'code_module',
wenzelm@45383
  2563
'code_library', 'consts_code', 'types_code' have been discontinued.
haftmann@46028
  2564
Use commands of the generic code generator instead.  INCOMPATIBILITY.
wenzelm@45383
  2565
wenzelm@47464
  2566
* Redundant attribute "code_inline" has been discontinued. Use
wenzelm@47464
  2567
"code_unfold" instead.  INCOMPATIBILITY.
wenzelm@47464
  2568
wenzelm@47464
  2569
* Dropped attribute "code_unfold_post" in favor of the its dual
wenzelm@47464
  2570
"code_abbrev", which yields a common pattern in definitions like
haftmann@46028
  2571
haftmann@46028
  2572
  definition [code_abbrev]: "f = t"
haftmann@46028
  2573
haftmann@46028
  2574
INCOMPATIBILITY.
wenzelm@45383
  2575
wenzelm@47856
  2576
* Obsolete 'types' command has been discontinued.  Use 'type_synonym'
wenzelm@47856
  2577
instead.  INCOMPATIBILITY.
wenzelm@47856
  2578
wenzelm@47856
  2579
* Discontinued old "prems" fact, which used to refer to the accidental
wenzelm@47856
  2580
collection of foundational premises in the context (already marked as
wenzelm@47856
  2581
legacy since Isabelle2011).
wenzelm@47855
  2582
wenzelm@45427
  2583
huffman@45122
  2584
*** HOL ***
huffman@45122
  2585
wenzelm@47464
  2586
* Type 'a set is now a proper type constructor (just as before
wenzelm@47464
  2587
Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
wenzelm@47464
  2588
Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
wenzelm@47855
  2589
sets separate, it is often sufficient to rephrase some set S that has
wenzelm@47855
  2590
been accidentally used as predicates by "%x. x : S", and some
wenzelm@47855
  2591
predicate P that has been accidentally used as set by "{x. P x}".
wenzelm@47855
  2592
Corresponding proofs in a first step should be pruned from any
wenzelm@47855
  2593
tinkering with former theorems mem_def and Collect_def as far as
wenzelm@47855
  2594
possible.
wenzelm@47855
  2595
wenzelm@47855
  2596
For developments which deliberately mix predicates and sets, a
wenzelm@47464
  2597
planning step is necessary to determine what should become a predicate
wenzelm@47464
  2598
and what a set.  It can be helpful to carry out that step in
wenzelm@47464
  2599
Isabelle2011-1 before jumping right into the current release.
wenzelm@47464
  2600
wenzelm@47855
  2601
* Code generation by default implements sets as container type rather
wenzelm@47855
  2602
than predicates.  INCOMPATIBILITY.
wenzelm@47855
  2603
wenzelm@47855
  2604
* New type synonym 'a rel = ('a * 'a) set
wenzelm@47855
  2605
wenzelm@47464
  2606
* The representation of numerals has changed.  Datatype "num"
wenzelm@47464
  2607
represents strictly positive binary numerals, along with functions
wenzelm@47464
  2608
"numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
wenzelm@47855
  2609
positive and negated numeric literals, respectively.  See also
wenzelm@47855
  2610
definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
wenzelm@47855
  2611
user theories may require adaptations as follows:
huffman@47108
  2612
huffman@47108
  2613
  - Theorems with number_ring or number_semiring constraints: These
huffman@47108
  2614
    classes are gone; use comm_ring_1 or comm_semiring_1 instead.
huffman@47108
  2615
huffman@47108
  2616
  - Theories defining numeric types: Remove number, number_semiring,
huffman@47108
  2617
    and number_ring instances. Defer all theorems about numerals until
huffman@47108
  2618
    after classes one and semigroup_add have been instantiated.
huffman@47108
  2619
huffman@47108
  2620
  - Numeral-only simp rules: Replace each rule having a "number_of v"
huffman@47108
  2621
    pattern with two copies, one for numeral and one for neg_numeral.
huffman@47108
  2622
huffman@47108
  2623
  - Theorems about subclasses of semiring_1 or ring_1: These classes
huffman@47108
  2624
    automatically support numerals now, so more simp rules and
huffman@47108
  2625
    simprocs may now apply within the proof.
huffman@47108
  2626
huffman@47108
  2627
  - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
huffman@47108
  2628
    Redefine using other integer operations.
huffman@47108
  2629
wenzelm@47855
  2630
* Transfer: New package intended to generalize the existing
wenzelm@47855
  2631
"descending" method and related theorem attributes from the Quotient
wenzelm@47855
  2632
package.  (Not all functionality is implemented yet, but future
wenzelm@47855
  2633
development will focus on Transfer as an eventual replacement for the
wenzelm@47855
  2634
corresponding parts of the Quotient package.)
wenzelm@47809
  2635
wenzelm@47809
  2636
  - transfer_rule attribute: Maintains a collection of transfer rules,
wenzelm@47809
  2637
    which relate constants at two different types. Transfer rules may
wenzelm@47809
  2638
    relate different type instances of the same polymorphic constant,
wenzelm@47809
  2639
    or they may relate an operation on a raw type to a corresponding
wenzelm@47809
  2640
    operation on an abstract type (quotient or subtype). For example:
wenzelm@47809
  2641
wenzelm@47809
  2642
    ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
wenzelm@47809
  2643
    (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
wenzelm@47809
  2644
wenzelm@47809
  2645
  - transfer method: Replaces a subgoal on abstract types with an
wenzelm@47809
  2646
    equivalent subgoal on the corresponding raw types. Constants are
wenzelm@47809
  2647
    replaced with corresponding ones according to the transfer rules.
wenzelm@47809
  2648
    Goals are generalized over all free variables by default; this is
huffman@47851
  2649
    necessary for variables whose types change, but can be overridden
wenzelm@47855
  2650
    for specific variables with e.g. "transfer fixing: x y z".  The
wenzelm@47809
  2651
    variant transfer' method allows replacing a subgoal with one that
wenzelm@47809
  2652
    is logically stronger (rather than equivalent).
wenzelm@47809
  2653
wenzelm@47809
  2654
  - relator_eq attribute: Collects identity laws for relators of
wenzelm@47809
  2655
    various type constructors, e.g. "list_all2 (op =) = (op =)".  The
wenzelm@47809
  2656
    transfer method uses these lemmas to infer transfer rules for
wenzelm@47809
  2657
    non-polymorphic constants on the fly.
wenzelm@47809
  2658
wenzelm@47809
  2659
  - transfer_prover method: Assists with proving a transfer rule for a
wenzelm@47809
  2660
    new constant, provided the constant is defined in terms of other
wenzelm@47809
  2661
    constants that already have transfer rules. It should be applied
wenzelm@47809
  2662
    after unfolding the constant definitions.
wenzelm@47809
  2663
wenzelm@47809
  2664
  - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
wenzelm@47809
  2665
    from type nat to type int.
wenzelm@47809
  2666
huffman@47851
  2667
* Lifting: New package intended to generalize the quotient_definition
huffman@47851
  2668
facility of the Quotient package; designed to work with Transfer.
wenzelm@47809
  2669
wenzelm@47809
  2670
  - lift_definition command: Defines operations on an abstract type in
wenzelm@47809
  2671
    terms of a corresponding operation on a representation
wenzelm@47809
  2672
    type.  Example syntax:
wenzelm@47809
  2673
wenzelm@47809
  2674
    lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
wenzelm@47809
  2675
      is List.insert
wenzelm@47809
  2676
wenzelm@47809
  2677
    Users must discharge a respectfulness proof obligation when each
wenzelm@47809
  2678
    constant is defined. (For a type copy, i.e. a typedef with UNIV,
wenzelm@47809
  2679
    the proof is discharged automatically.) The obligation is
wenzelm@47809
  2680
    presented in a user-friendly, readable form; a respectfulness
wenzelm@47809
  2681
    theorem in the standard format and a transfer rule are generated
wenzelm@47809
  2682
    by the package.
wenzelm@47809
  2683
wenzelm@47809
  2684
  - Integration with code_abstype: For typedefs (e.g. subtypes
wenzelm@47809
  2685
    corresponding to a datatype invariant, such as dlist),
wenzelm@47809
  2686
    lift_definition generates a code certificate theorem and sets up
wenzelm@47809
  2687
    code generation for each constant.
wenzelm@47809
  2688
wenzelm@47809
  2689
  - setup_lifting command: Sets up the Lifting package to work with a
wenzelm@47809
  2690
    user-defined type. The user must provide either a quotient theorem
wenzelm@47809
  2691
    or a type_definition theorem.  The package configures transfer
wenzelm@47809
  2692
    rules for equality and quantifiers on the type, and sets up the
wenzelm@47809
  2693
    lift_definition command to work with the type.
wenzelm@47809
  2694
wenzelm@47809
  2695
  - Usage examples: See Quotient_Examples/Lift_DList.thy,
huffman@47851
  2696
    Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
huffman@47851
  2697
    Word/Word.thy and Library/Float.thy.
wenzelm@47809
  2698
wenzelm@47809
  2699
* Quotient package:
wenzelm@47809
  2700
wenzelm@47809
  2701
  - The 'quotient_type' command now supports a 'morphisms' option with
wenzelm@47809
  2702
    rep and abs functions, similar to typedef.
wenzelm@47809
  2703
wenzelm@47809
  2704
  - 'quotient_type' sets up new types to work with the Lifting and
wenzelm@47809
  2705
    Transfer packages, as with 'setup_lifting'.
wenzelm@47809
  2706
wenzelm@47809
  2707
  - The 'quotient_definition' command now requires the user to prove a
wenzelm@47809
  2708
    respectfulness property at the point where the constant is
wenzelm@47809
  2709
    defined, similar to lift_definition; INCOMPATIBILITY.
wenzelm@478