NEWS
author nipkow
Wed Jun 06 18:19:55 2018 +0200 (14 months ago)
changeset 68403 223172b97d0b
parent 68373 f254e383bfe9
child 68404 05605481935d
permissions -rw-r--r--
reorient -> split; documented split
wenzelm@57491
     1
Isabelle NEWS -- history of user-relevant changes
wenzelm@57491
     2
=================================================
wenzelm@2553
     3
wenzelm@62114
     4
(Note: Isabelle/jEdit shows a tree-view of the NEWS file in Sidekick.)
wenzelm@60006
     5
wenzelm@64603
     6
wenzelm@66651
     7
New in this Isabelle version
wenzelm@66651
     8
----------------------------
wenzelm@66651
     9
wenzelm@66712
    10
*** General ***
wenzelm@66712
    11
wenzelm@67446
    12
* Marginal comments need to be written exclusively in the new-style form
wenzelm@67446
    13
"\<comment> \<open>text\<close>", old ASCII variants like "-- {* ... *}" are no longer
wenzelm@67446
    14
supported. INCOMPATIBILITY, use the command-line tool "isabelle
wenzelm@67446
    15
update_comments" to update existing theory files.
wenzelm@67446
    16
wenzelm@67507
    17
* Old-style inner comments (* ... *) within the term language are legacy
wenzelm@67507
    18
and will be discontinued soon: use formal comments "\<comment> \<open>...\<close>" or "\<^cancel>\<open>...\<close>"
wenzelm@67507
    19
instead.
wenzelm@67507
    20
nipkow@67402
    21
* The "op <infix-op>" syntax for infix operators has been replaced by
nipkow@67400
    22
"(<infix-op>)". If <infix-op> begins or ends with a "*", there needs to
nipkow@67400
    23
be a space between the "*" and the corresponding parenthesis.
nipkow@67400
    24
INCOMPATIBILITY.
nipkow@67398
    25
There is an Isabelle tool "update_op" that converts theory and ML files
nipkow@67398
    26
to the new syntax. Because it is based on regular expression matching,
nipkow@67398
    27
the result may need a bit of manual postprocessing. Invoking "isabelle
nipkow@67398
    28
update_op" converts all files in the current directory (recursively).
nipkow@67398
    29
In case you want to exclude conversion of ML files (because the tool
nipkow@67398
    30
frequently also converts ML's "op" syntax), use option "-m".
nipkow@67398
    31
wenzelm@67119
    32
* The old 'def' command has been discontinued (legacy since
wenzelm@67119
    33
Isbelle2016-1). INCOMPATIBILITY, use 'define' instead -- usually with
wenzelm@67119
    34
object-logic equality or equivalence.
wenzelm@67119
    35
wenzelm@66712
    36
* Session-qualified theory names are mandatory: it is no longer possible
wenzelm@66712
    37
to refer to unqualified theories from the parent session.
wenzelm@66712
    38
INCOMPATIBILITY for old developments that have not been updated to
wenzelm@66712
    39
Isabelle2017 yet (using the "isabelle imports" tool).
wenzelm@66712
    40
wenzelm@67013
    41
* Theory header 'abbrevs' specifications need to be separated by 'and'.
wenzelm@67013
    42
INCOMPATIBILITY.
wenzelm@67013
    43
wenzelm@66993
    44
* Only the most fundamental theory names are global, usually the entry
wenzelm@66993
    45
points to major logic sessions: Pure, Main, Complex_Main, HOLCF, IFOL,
wenzelm@66993
    46
FOL, ZF, ZFC etc. INCOMPATIBILITY, need to use qualified names for
wenzelm@66993
    47
formerly global "HOL-Probability.Probability" and "HOL-SPARK.SPARK".
wenzelm@66993
    48
wenzelm@66757
    49
* Command 'external_file' declares the formal dependency on the given
wenzelm@66757
    50
file name, such that the Isabelle build process knows about it, but
wenzelm@66757
    51
without specific Prover IDE management.
wenzelm@66757
    52
wenzelm@66759
    53
* Session ROOT entries no longer allow specification of 'files'. Rare
wenzelm@66759
    54
INCOMPATIBILITY, use command 'external_file' within a proper theory
wenzelm@66759
    55
context.
wenzelm@66759
    56
wenzelm@66764
    57
* Session root directories may be specified multiple times: each
wenzelm@66764
    58
accessible ROOT file is processed only once. This facilitates
wenzelm@66764
    59
specification of $ISABELLE_HOME_USER/ROOTS or command-line options like
wenzelm@66764
    60
-d or -D for "isabelle build" and "isabelle jedit". Example:
wenzelm@66764
    61
wenzelm@66764
    62
  isabelle build -D '~~/src/ZF'
wenzelm@66764
    63
wenzelm@67263
    64
* The command 'display_drafts' has been discontinued. INCOMPATIBILITY,
wenzelm@67263
    65
use action "isabelle.draft" (or "print") in Isabelle/jEdit instead.
wenzelm@67263
    66
wenzelm@67305
    67
* Isabelle symbol "\<hyphen>" is rendered as explicit Unicode hyphen U+2010, to
wenzelm@67305
    68
avoid unclear meaning of the old "soft hyphen" U+00AD. Rare
wenzelm@67305
    69
INCOMPATIBILITY, e.g. copy-paste of historic Isabelle HTML output.
wenzelm@67305
    70
wenzelm@66712
    71
wenzelm@67261
    72
*** Isabelle/jEdit Prover IDE ***
wenzelm@66768
    73
wenzelm@68368
    74
* Slightly more parallel checking, notably for high priority print
wenzelm@68368
    75
functions (e.g. State output).
wenzelm@68368
    76
wenzelm@68080
    77
* The view title is set dynamically, according to the Isabelle
wenzelm@68080
    78
distribution and the logic session name. The user can override this via
wenzelm@68080
    79
set-view-title (stored persistently in $JEDIT_SETTINGS/perspective.xml).
wenzelm@68080
    80
wenzelm@67395
    81
* System options "spell_checker_include" and "spell_checker_exclude"
wenzelm@67395
    82
supersede former "spell_checker_elements" to determine regions of text
wenzelm@67395
    83
that are subject to spell-checking. Minor INCOMPATIBILITY.
wenzelm@67395
    84
wenzelm@67221
    85
* PIDE markup for session ROOT files: allows to complete session names,
wenzelm@67221
    86
follow links to theories and document files etc.
wenzelm@67215
    87
wenzelm@66993
    88
* Completion supports theory header imports, using theory base name.
wenzelm@66994
    89
E.g. "Prob" may be completed to "HOL-Probability.Probability".
wenzelm@66768
    90
wenzelm@66988
    91
* The command-line tool "isabelle jedit" provides more flexible options
wenzelm@68370
    92
for session management:
wenzelm@68370
    93
  - option -R builds an auxiliary logic image with all required theories
wenzelm@68370
    94
    from other sessions, relative to an ancestor session given by option
wenzelm@68370
    95
    -A (default: parent)
wenzelm@68370
    96
  - option -S is like -R, with a focus on the selected session and its
wenzelm@68370
    97
    descendants (this reduces startup time for big projects like AFP)
wenzelm@66973
    98
wenzelm@66990
    99
  Examples:
wenzelm@68370
   100
    isabelle jedit -R HOL-Number_Theory
wenzelm@68370
   101
    isabelle jedit -R HOL-Number_Theory -A HOL
wenzelm@66990
   102
    isabelle jedit -d '$AFP' -S Formal_SSA -A HOL
wenzelm@66990
   103
    isabelle jedit -d '$AFP' -S Formal_SSA -A HOL-Analysis
wenzelm@66988
   104
wenzelm@67133
   105
* Named control symbols (without special Unicode rendering) are shown as
wenzelm@67133
   106
bold-italic keyword. This is particularly useful for the short form of
wenzelm@67133
   107
antiquotations with control symbol: \<^name>\<open>argument\<close>. The action
wenzelm@67133
   108
"isabelle.antiquoted_cartouche" turns an antiquotation with 0 or 1
wenzelm@67133
   109
arguments into this format.
wenzelm@67133
   110
wenzelm@67311
   111
* Completion provides templates for named symbols with arguments,
wenzelm@67311
   112
e.g. "\<comment> \<open>ARGUMENT\<close>" or "\<^emph>\<open>ARGUMENT\<close>".
wenzelm@67311
   113
wenzelm@67295
   114
* Bibtex database files (.bib) are semantically checked.
wenzelm@67295
   115
wenzelm@67248
   116
* Action "isabelle.preview" is able to present more file formats,
wenzelm@67266
   117
notably bibtex database files and ML files.
wenzelm@67246
   118
wenzelm@67262
   119
* Action "isabelle.draft" is similar to "isabelle.preview", but shows a
wenzelm@68067
   120
plain-text document draft. Both are available via the menu "Plugins /
wenzelm@68067
   121
Isabelle".
wenzelm@67262
   122
wenzelm@67304
   123
* When loading text files, the Isabelle symbols encoding UTF-8-Isabelle
wenzelm@67304
   124
is only used if there is no conflict with existing Unicode sequences in
wenzelm@67304
   125
the file. Otherwise, the fallback encoding is plain UTF-8 and Isabelle
wenzelm@67304
   126
symbols remain in literal \<symbol> form. This avoids accidental loss of
wenzelm@67304
   127
Unicode content when saving the file.
wenzelm@67304
   128
wenzelm@67993
   129
* Update to jedit-5.5.0, the latest release.
wenzelm@67993
   130
wenzelm@67246
   131
wenzelm@67261
   132
*** Isabelle/VSCode Prover IDE ***
wenzelm@67261
   133
wenzelm@67261
   134
* HTML preview of theories and other file-formats similar to
wenzelm@67261
   135
Isabelle/jEdit.
wenzelm@67261
   136
wenzelm@66768
   137
wenzelm@67140
   138
*** Document preparation ***
wenzelm@67140
   139
wenzelm@67448
   140
* Formal comments work uniformly in outer syntax, inner syntax (term
wenzelm@67448
   141
language), Isabelle/ML and some other embedded languages of Isabelle.
wenzelm@67448
   142
See also "Document comments" in the isar-ref manual. The following forms
wenzelm@67448
   143
are supported:
wenzelm@67448
   144
wenzelm@67448
   145
  - marginal text comment: \<comment> \<open>\<dots>\<close>
wenzelm@67448
   146
  - canceled source: \<^cancel>\<open>\<dots>\<close>
wenzelm@67448
   147
  - raw LaTeX: \<^latex>\<open>\<dots>\<close>
wenzelm@67413
   148
wenzelm@67381
   149
* Outside of the inner theory body, the default presentation context is
wenzelm@67381
   150
theory Pure. Thus elementary antiquotations may be used in markup
wenzelm@67381
   151
commands (e.g. 'chapter', 'section', 'text') and formal comments.
wenzelm@67381
   152
wenzelm@67140
   153
* System option "document_tags" specifies a default for otherwise
wenzelm@67140
   154
untagged commands. This is occasionally useful to control the global
wenzelm@67140
   155
visibility of commands via session options (e.g. in ROOT).
wenzelm@67140
   156
wenzelm@67140
   157
* Document markup commands ('section', 'text' etc.) are implicitly
wenzelm@67140
   158
tagged as "document" and visible by default. This avoids the application
wenzelm@67140
   159
of option "document_tags" to these commands.
wenzelm@67140
   160
wenzelm@67145
   161
* Isabelle names are mangled into LaTeX macro names to allow the full
wenzelm@67145
   162
identifier syntax with underscore, prime, digits. This is relevant for
wenzelm@67145
   163
antiquotations in control symbol notation, e.g. \<^const_name> becomes
wenzelm@67145
   164
\isactrlconstUNDERSCOREname.
wenzelm@67145
   165
wenzelm@67297
   166
* Document antiquotation @{cite} now checks the given Bibtex entries
wenzelm@67297
   167
against the Bibtex database files -- only in batch-mode session builds.
wenzelm@67297
   168
wenzelm@67219
   169
* Document antiquotation @{session name} checks and prints the given
wenzelm@67219
   170
session name verbatim.
wenzelm@67219
   171
wenzelm@67157
   172
* Document preparation with skip_proofs option now preserves the content
wenzelm@67157
   173
more accurately: only terminal proof steps ('by' etc.) are skipped.
wenzelm@67157
   174
wenzelm@67176
   175
* Command-line tool "isabelle document" has been re-implemented in
wenzelm@67194
   176
Isabelle/Scala, with simplified arguments and explicit errors from the
wenzelm@67203
   177
latex and bibtex process. Minor INCOMPATIBILITY.
wenzelm@67173
   178
wenzelm@67140
   179
wenzelm@67702
   180
*** Isar ***
wenzelm@67702
   181
wenzelm@67702
   182
* Command 'interpret' no longer exposes resulting theorems as literal
wenzelm@67702
   183
facts, notably for the \<open>prop\<close> notation or the "fact" proof method. This
wenzelm@67702
   184
improves modularity of proofs and scalability of locale interpretation.
wenzelm@67702
   185
Rare INCOMPATIBILITY, need to refer to explicitly named facts instead
wenzelm@67702
   186
(e.g. use 'find_theorems' or 'try' to figure this out).
wenzelm@67702
   187
ballarin@67740
   188
* Rewrites clauses (keyword 'rewrites') were moved into the locale
ballarin@67764
   189
expression syntax, where they are part of locale instances.  In
ballarin@67764
   190
interpretation commands rewrites clauses now need to occur before
ballarin@67764
   191
'for' and 'defines'.  Minor INCOMPATIBILITY.
ballarin@67764
   192
ballarin@67764
   193
* For rewrites clauses, if activating a locale instance fails, fall
ballarin@67764
   194
back to reading the clause first.  This helps avoid qualification of
ballarin@67764
   195
locale instances where the qualifier's sole purpose is avoiding
ballarin@67764
   196
duplicate constant declarations.
ballarin@67741
   197
nipkow@68403
   198
* Proof method 'simp' now supports a new modifier 'flip:' followed by a list
nipkow@68403
   199
of theorems. Each of these theorems is removed from the simpset
nipkow@68403
   200
(without warning if it is not there) and the symmetric version of the theorem
nipkow@68403
   201
(i.e. lhs and rhs exchanged) is added to the simpset.
nipkow@68403
   202
For 'auto' and friends the modifier is "simp flip:".
nipkow@68403
   203
wenzelm@67702
   204
wenzelm@67718
   205
*** Pure ***
wenzelm@67718
   206
wenzelm@67718
   207
* The inner syntax category "sort" now includes notation "_" for the
wenzelm@67718
   208
dummy sort: it is effectively ignored in type-inference.
wenzelm@67718
   209
wenzelm@67718
   210
blanchet@66661
   211
*** HOL ***
blanchet@66661
   212
eberlm@68246
   213
* Landau_Symbols from the AFP was moved to HOL-Library
eberlm@68246
   214
haftmann@68028
   215
* Clarified relationship of characters, strings and code generation:
haftmann@68028
   216
haftmann@68028
   217
  * Type "char" is now a proper datatype of 8-bit values.
haftmann@68028
   218
haftmann@68028
   219
  * Conversions "nat_of_char" and "char_of_nat" are gone; use more
haftmann@68028
   220
    general conversions "of_char" and "char_of" with suitable
haftmann@68028
   221
    type constraints instead.
haftmann@68028
   222
haftmann@68028
   223
  * The zero character is just written "CHR 0x00", not
haftmann@68028
   224
    "0" any longer.
haftmann@68028
   225
haftmann@68028
   226
  * Type "String.literal" (for code generation) is now isomorphic
haftmann@68028
   227
    to lists of 7-bit (ASCII) values; concrete values can be written
haftmann@68028
   228
    as "STR ''...''" for sequences of printable characters and
haftmann@68033
   229
    "STR 0x..." for one single ASCII code point given
haftmann@68028
   230
    as hexadecimal numeral.
haftmann@68028
   231
haftmann@68028
   232
  * Type "String.literal" supports concatenation "... + ..."
haftmann@68028
   233
    for all standard target languages.
haftmann@68028
   234
haftmann@68028
   235
  * Theory Library/Code_Char is gone; study the explanations concerning
haftmann@68028
   236
    "String.literal" in the tutorial on code generation to get an idea
haftmann@68028
   237
    how target-language string literals can be converted to HOL string
haftmann@68028
   238
    values and vice versa.
haftmann@68028
   239
haftmann@68028
   240
  * Imperative-HOL: operation "raise" directly takes a value of type
haftmann@68028
   241
    "String.literal" as argument, not type "string".
haftmann@68028
   242
haftmann@68028
   243
INCOMPATIBILITY.
haftmann@68028
   244
haftmann@68027
   245
* Abstract bit operations as part of Main: push_bit, take_bit,
haftmann@68027
   246
drop_bit.
haftmann@67928
   247
lp15@68373
   248
* New, more general, axiomatization of complete_distrib_lattice.
eberlm@67831
   249
The former axioms:
eberlm@67831
   250
"sup x (Inf X) = Inf (sup x ` X)" and "inf x (Sup X) = Sup (inf x ` X)"
lp15@68373
   251
are replaced by
eberlm@67831
   252
"Inf (Sup ` A) <= Sup (Inf ` {f ` A | f . (! Y \<in> A . f Y \<in> Y)})"
lp15@68373
   253
The instantiations of sets and functions as complete_distrib_lattice
eberlm@67831
   254
are moved to Hilbert_Choice.thy because their proofs need the Hilbert
lp15@68373
   255
choice operator. The dual of this property is also proved in
eberlm@67831
   256
Hilbert_Choice.thy.
eberlm@67831
   257
lp15@67999
   258
* New syntax for the minimum/maximum of a function over a finite set:
lp15@67999
   259
MIN x\<in>A. B  and even  MIN x. B (only useful for finite types), also
lp15@67999
   260
MAX.
lp15@67999
   261
haftmann@67525
   262
* Clarifed theorem names:
haftmann@67525
   263
haftmann@67525
   264
  Min.antimono ~> Min.subset_imp
haftmann@67525
   265
  Max.antimono ~> Max.subset_imp
haftmann@67525
   266
haftmann@67525
   267
Minor INCOMPATIBILITY.
haftmann@67525
   268
traytel@67224
   269
* A new command parametric_constant for proving parametricity of
wenzelm@67303
   270
non-recursive definitions. For constants that are not fully parametric
wenzelm@67303
   271
the command will infer conditions on relations (e.g., bi_unique,
wenzelm@67303
   272
bi_total, or type class conditions such as "respects 0") sufficient for
wenzelm@67303
   273
parametricity. See ~~/src/HOL/ex/Conditional_Parametricity_Examples for
wenzelm@67303
   274
some examples.
traytel@67224
   275
Andreas@68200
   276
* A new preprocessor for the code generator to generate code for algebraic
Andreas@68200
   277
  types with lazy evaluation semantics even in call-by-value target languages.
Andreas@68200
   278
  See theory HOL-Library.Code_Lazy for the implementation and
Andreas@68200
   279
  HOL-Codegenerator_Test.Code_Lazy_Test for examples.
Andreas@68200
   280
blanchet@66661
   281
* SMT module:
blanchet@66661
   282
  - The 'smt_oracle' option is now necessary when using the 'smt' method
blanchet@66662
   283
    with a solver other than Z3. INCOMPATIBILITY.
wenzelm@66844
   284
  - The encoding to first-order logic is now more complete in the
wenzelm@66844
   285
    presence of higher-order quantifiers. An 'smt_explicit_application'
wenzelm@66844
   286
    option has been added to control this. INCOMPATIBILITY.
wenzelm@66844
   287
blanchet@67335
   288
* Datatypes:
blanchet@67335
   289
  - The legacy command 'old_datatype', provided by
blanchet@67335
   290
    '~~/src/HOL/Library/Old_Datatype.thy', has been removed. INCOMPATIBILITY.
blanchet@67335
   291
wenzelm@66844
   292
* Theory HOL-Computational_Algebra.Polynomial_Factorial does not provide
wenzelm@66844
   293
instances of rat, real, complex as factorial rings etc. Import
wenzelm@66844
   294
HOL-Computational_Algebra.Field_as_Ring explicitly in case of need.
wenzelm@66844
   295
INCOMPATIBILITY.
wenzelm@66844
   296
haftmann@66804
   297
* Facts sum.commute(_restrict) and prod.commute(_restrict) renamed to
wenzelm@66844
   298
sum.swap(_restrict) and prod.swap(_restrict), to avoid name clashes on
wenzelm@66844
   299
interpretation of abstract locales. INCOMPATIBILITY.
haftmann@66804
   300
haftmann@66803
   301
* Predicate pairwise_coprime abolished, use "pairwise coprime" instead.
haftmann@66803
   302
INCOMPATIBILITY.
haftmann@66803
   303
lp15@68373
   304
* The relator rel_filter on filters has been strengthened to its
Andreas@67616
   305
canonical categorical definition with better properties. INCOMPATIBILITY.
Andreas@67616
   306
immler@68072
   307
* Generalized linear algebra involving linear, span, dependent, dim
immler@68072
   308
from type class real_vector to locales module and vector_space.
immler@68072
   309
Renamed:
immler@68072
   310
  - span_inc ~> span_superset
immler@68072
   311
  - span_superset ~> span_base
immler@68072
   312
  - span_eq ~> span_eq_iff
immler@68072
   313
INCOMPATIBILITY.
immler@68072
   314
nipkow@67341
   315
* HOL-Algebra: renamed (^) to [^]
nipkow@67341
   316
lp15@68373
   317
* Session HOL-Analysis: infinite products, Moebius functions, the Riemann mapping
lp15@67999
   318
theorem, the Vitali covering theorem, change-of-variables results for
lp15@67999
   319
integration and measures.
wenzelm@66844
   320
haftmann@66937
   321
* Class linordered_semiring_1 covers zero_less_one also, ruling out
haftmann@66937
   322
pathologic instances. Minor INCOMPATIBILITY.
haftmann@66937
   323
nipkow@68125
   324
* Theory List: functions "sorted_wrt" and "sorted" now compare every
nipkow@68125
   325
  element in a list to all following elements, not just the next one.
nipkow@68125
   326
nipkow@68364
   327
* Theory List: Synatx:
nipkow@68364
   328
  - filter-syntax "[x <- xs. P]" is no longer output syntax
nipkow@68364
   329
    but only input syntax.
nipkow@68364
   330
  - list comprehension syntax now supports tuple patterns in "pat <- xs".
nipkow@68249
   331
haftmann@67051
   332
* Removed nat-int transfer machinery. Rare INCOMPATIBILITY.
haftmann@67051
   333
haftmann@67051
   334
* Predicate coprime is now a real definition, not a mere
haftmann@67051
   335
abbreviation. INCOMPATIBILITY.
haftmann@66954
   336
haftmann@67207
   337
* Code generation: Code generation takes an explicit option
haftmann@67207
   338
"case_insensitive" to accomodate case-insensitive file systems.
haftmann@67207
   339
haftmann@68100
   340
* Fact mod_mult_self4 (on nat) renamed to Suc_mod_mult_self3, to avoid
haftmann@68100
   341
clash with fact mod_mult_self4 (on more generic semirings).
haftmann@68100
   342
INCOMPATIBILITY.
haftmann@68100
   343
haftmann@68100
   344
* Eliminated some theorem aliasses:
haftmann@68100
   345
haftmann@68100
   346
  even_times_iff ~> even_mult_iff
haftmann@68100
   347
  mod_2_not_eq_zero_eq_one_nat ~> not_mod_2_eq_0_eq_1
haftmann@68100
   348
  even_of_nat ~> even_int_iff
haftmann@68100
   349
haftmann@68100
   350
INCOMPATIBILITY.
haftmann@68100
   351
haftmann@68157
   352
* Eliminated some theorem duplicate variations:
haftmann@68157
   353
  * dvd_eq_mod_eq_0_numeral can be replaced by dvd_eq_mod_eq_0.
haftmann@68157
   354
  * mod_Suc_eq_Suc_mod can be replaced by mod_Suc.
haftmann@68157
   355
  * mod_Suc_eq_Suc_mod [symmetrict] can be replaced by mod_simps.
haftmann@68157
   356
  * mod_eq_0_iff can be replaced by mod_eq_0_iff_dvd and dvd_def.
haftmann@68157
   357
  * The witness of mod_eqD can be given directly as "_ div _".
haftmann@68157
   358
haftmann@68157
   359
INCOMPATIBILITY.
haftmann@68157
   360
haftmann@68260
   361
* Classical setup: Assumption "m mod d = 0" (for m d :: nat) is no
haftmann@68260
   362
longer aggresively destroyed to "?q. m = d * q".  INCOMPATIBILITY,
haftmann@68260
   363
adding "elim!: dvd" to classical proof methods in most situations
haftmann@68260
   364
restores broken proofs.
haftmann@68260
   365
wenzelm@66651
   366
wenzelm@68116
   367
*** ML ***
wenzelm@68116
   368
wenzelm@68116
   369
* Operation Export.export emits theory exports (arbitrary blobs), which
wenzelm@68116
   370
are stored persistently in the session build database.
wenzelm@68116
   371
wenzelm@68276
   372
* Command 'ML_export' exports ML toplevel bindings to the global
wenzelm@68276
   373
bootstrap environment of the ML process. This allows ML evaluation
wenzelm@68276
   374
without a formal theory context, e.g. in command-line tools like
wenzelm@68276
   375
"isabelle process".
wenzelm@68276
   376
wenzelm@68116
   377
wenzelm@66729
   378
*** System ***
wenzelm@66729
   379
wenzelm@68219
   380
* Settings ISABELLE_PATH and ISABELLE_OUTPUT have been discontinued:
wenzelm@68219
   381
heap images and session databases are always stored in
wenzelm@68219
   382
$ISABELLE_HOME_USER/heaps/$ML_IDENTIFIER (command-line default) or
wenzelm@68219
   383
$ISABELLE_HOME/heaps/$ML_IDENTIFIER (main Isabelle application or
wenzelm@68219
   384
"isabelle jedit -s" or "isabelle build -s").
wenzelm@68219
   385
wenzelm@68292
   386
* The command-line tool "export" and 'export_files' in session ROOT
wenzelm@68292
   387
entries retrieve theory exports from the session build database.
wenzelm@68116
   388
wenzelm@67904
   389
* The command-line tools "isabelle server" and "isabelle client" provide
wenzelm@67904
   390
access to the Isabelle Server: it supports responsive session management
wenzelm@67904
   391
and concurrent use of theories, based on Isabelle/PIDE infrastructure.
wenzelm@67904
   392
See also the "system" manual.
wenzelm@67904
   393
wenzelm@68348
   394
* The command-line tool "dump" dumps information from the cumulative
wenzelm@68348
   395
PIDE session database: many sessions may be loaded into a given logic
wenzelm@68348
   396
image, results from all loaded theories are written to the output
wenzelm@68348
   397
directory.
wenzelm@68348
   398
wenzelm@67433
   399
* The command-line tool "isabelle update_comments" normalizes formal
wenzelm@67433
   400
comments in outer syntax as follows: \<comment> \<open>text\<close> (whith a single space to
wenzelm@67433
   401
approximate the appearance in document output). This is more specific
wenzelm@67433
   402
than former "isabelle update_cartouches -c": the latter tool option has
wenzelm@67433
   403
been discontinued.
wenzelm@67433
   404
wenzelm@66949
   405
* Session ROOT entry: empty 'document_files' means there is no document
wenzelm@66949
   406
for this session. There is no need to specify options [document = false]
wenzelm@66949
   407
anymore.
wenzelm@66949
   408
wenzelm@67043
   409
* The command-line tool "isabelle mkroot" now always produces a document
wenzelm@67043
   410
outline: its options have been adapted accordingly. INCOMPATIBILITY.
wenzelm@67043
   411
wenzelm@67069
   412
* The command-line tool "isabelle mkroot -I" initializes a Mercurial
wenzelm@67069
   413
repository for the generated session files.
wenzelm@67069
   414
wenzelm@66909
   415
* Linux and Windows/Cygwin is for x86_64 only. Old 32bit platform
wenzelm@66909
   416
support has been discontinued.
wenzelm@66729
   417
wenzelm@67088
   418
* Mac OS X 10.10 Yosemite is now the baseline version; Mavericks is no
wenzelm@67088
   419
longer supported.
wenzelm@67088
   420
wenzelm@66906
   421
* Java runtime is for x86_64 only. Corresponding Isabelle settings have
wenzelm@66906
   422
been renamed to ISABELLE_TOOL_JAVA_OPTIONS and JEDIT_JAVA_OPTIONS,
wenzelm@66906
   423
instead of former 32/64 variants. INCOMPATIBILITY.
wenzelm@66906
   424
wenzelm@68003
   425
* Old settings ISABELLE_PLATFORM and ISABELLE_WINDOWS_PLATFORM should be
wenzelm@68003
   426
phased out due to unclear preference of 32bit vs. 64bit architecture.
wenzelm@68003
   427
Explicit GNU bash expressions are now preferred, for example (with
wenzelm@68003
   428
quotes):
wenzelm@68003
   429
wenzelm@68003
   430
  #Posix executables (Unix or Cygwin), with preference for 64bit
wenzelm@68003
   431
  "${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM32}"
wenzelm@68003
   432
wenzelm@68003
   433
  #native Windows or Unix executables, with preference for 64bit
wenzelm@68003
   434
  "${ISABELLE_WINDOWS_PLATFORM64:-${ISABELLE_WINDOWS_PLATFORM32:-${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM32}}}"
wenzelm@68003
   435
wenzelm@68003
   436
  #native Windows (32bit) or Unix executables (preference for 64bit)
wenzelm@68003
   437
  "${ISABELLE_WINDOWS_PLATFORM32:-${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM32}}"
wenzelm@68003
   438
wenzelm@66745
   439
* Command-line tool "isabelle build" supports new options:
wenzelm@66745
   440
  - option -B NAME: include session NAME and all descendants
wenzelm@66745
   441
  - option -S: only observe changes of sources, not heap images
wenzelm@66841
   442
  - option -f: forces a fresh build
wenzelm@66737
   443
wenzelm@66843
   444
* Command-line tool "isabelle build" takes "condition" options with the
wenzelm@66843
   445
corresponding environment values into account, when determining the
wenzelm@66843
   446
up-to-date status of a session.
wenzelm@66843
   447
wenzelm@66851
   448
* Command-line tool "isabelle imports -I" also reports actual session
wenzelm@66851
   449
imports. This helps to minimize the session dependency graph.
wenzelm@66851
   450
wenzelm@67099
   451
* Update to current Poly/ML 5.7.1 with slightly improved performance and
wenzelm@67099
   452
PIDE markup for identifier bindings.
wenzelm@67099
   453
wenzelm@67199
   454
* ISABELLE_LATEX and ISABELLE_PDFLATEX now include platform-specific
wenzelm@67199
   455
options for improved error reporting. Potential INCOMPATIBILITY with
wenzelm@67199
   456
unusual LaTeX installations, may have to adapt these settings.
wenzelm@67199
   457
wenzelm@67591
   458
* The bundled Poly/ML 5.7.1 now uses The GNU Multiple Precision
wenzelm@67591
   459
Arithmetic Library (libgmp) on all platforms, notably Mac OS X with
wenzelm@67591
   460
32/64 bit.
wenzelm@67591
   461
wenzelm@67099
   462
wenzelm@66729
   463
wenzelm@66472
   464
New in Isabelle2017 (October 2017)
wenzelm@66472
   465
----------------------------------
wenzelm@64439
   466
wenzelm@64986
   467
*** General ***
wenzelm@64986
   468
wenzelm@66238
   469
* Experimental support for Visual Studio Code (VSCode) as alternative
wenzelm@66238
   470
Isabelle/PIDE front-end, see also
wenzelm@66599
   471
https://marketplace.visualstudio.com/items?itemName=makarius.Isabelle2017
wenzelm@66238
   472
wenzelm@66238
   473
VSCode is a new type of application that continues the concepts of
wenzelm@66238
   474
"programmer's editor" and "integrated development environment" towards
wenzelm@66238
   475
fully semantic editing and debugging -- in a relatively light-weight
wenzelm@66238
   476
manner. Thus it fits nicely on top of the Isabelle/PIDE infrastructure.
wenzelm@66238
   477
Technically, VSCode is based on the Electron application framework
wenzelm@66238
   478
(Node.js + Chromium browser + V8), which is implemented in JavaScript
wenzelm@66238
   479
and TypeScript, while Isabelle/VSCode mainly consists of Isabelle/Scala
wenzelm@66238
   480
modules around a Language Server implementation.
wenzelm@66238
   481
wenzelm@65504
   482
* Theory names are qualified by the session name that they belong to.
wenzelm@66454
   483
This affects imports, but not the theory name space prefix (which is
wenzelm@66454
   484
just the theory base name as before).
wenzelm@66454
   485
wenzelm@66454
   486
In order to import theories from other sessions, the ROOT file format
wenzelm@66454
   487
provides a new 'sessions' keyword. In contrast, a theory that is
wenzelm@66454
   488
imported in the old-fashioned manner via an explicit file-system path
wenzelm@66472
   489
belongs to the current session, and might cause theory name conflicts
wenzelm@66454
   490
later on. Theories that are imported from other sessions are excluded
wenzelm@66454
   491
from the current session document. The command-line tool "isabelle
wenzelm@66454
   492
imports" helps to update theory imports.
wenzelm@66454
   493
wenzelm@65451
   494
* The main theory entry points for some non-HOL sessions have changed,
wenzelm@65451
   495
to avoid confusion with the global name "Main" of the session HOL. This
wenzelm@65451
   496
leads to the follow renamings:
wenzelm@65451
   497
wenzelm@65451
   498
  CTT/Main.thy    ~>  CTT/CTT.thy
wenzelm@65451
   499
  ZF/Main.thy     ~>  ZF/ZF.thy
wenzelm@65451
   500
  ZF/Main_ZF.thy  ~>  ZF/ZF.thy
wenzelm@65451
   501
  ZF/Main_ZFC.thy ~>  ZF/ZFC.thy
wenzelm@65451
   502
  ZF/ZF.thy       ~>  ZF/ZF_Base.thy
wenzelm@65451
   503
wenzelm@65451
   504
INCOMPATIBILITY.
wenzelm@65451
   505
wenzelm@66472
   506
* Commands 'alias' and 'type_alias' introduce aliases for constants and
wenzelm@66472
   507
type constructors, respectively. This allows adhoc changes to name-space
wenzelm@66472
   508
accesses within global or local theory contexts, e.g. within a 'bundle'.
wenzelm@66472
   509
wenzelm@64986
   510
* Document antiquotations @{prf} and @{full_prf} output proof terms
wenzelm@64986
   511
(again) in the same way as commands 'prf' and 'full_prf'.
wenzelm@64986
   512
wenzelm@65055
   513
* Computations generated by the code generator can be embedded directly
wenzelm@65055
   514
into ML, alongside with @{code} antiquotations, using the following
wenzelm@65055
   515
antiquotations:
wenzelm@65055
   516
wenzelm@65055
   517
  @{computation ... terms: ... datatypes: ...} :
wenzelm@65055
   518
    ((term -> term) -> 'ml option -> 'a) -> Proof.context -> term -> 'a
wenzelm@65055
   519
  @{computation_conv ... terms: ... datatypes: ...} :
wenzelm@65055
   520
    (Proof.context -> 'ml -> conv) -> Proof.context -> conv
wenzelm@65045
   521
  @{computation_check terms: ... datatypes: ...} : Proof.context -> conv
haftmann@65041
   522
wenzelm@65055
   523
See src/HOL/ex/Computations.thy,
wenzelm@65055
   524
src/HOL/Decision_Procs/Commutative_Ring.thy and
wenzelm@65055
   525
src/HOL/Decision_Procs/Reflective_Field.thy for examples and the
wenzelm@65055
   526
tutorial on code generation.
haftmann@65041
   527
wenzelm@64986
   528
wenzelm@64602
   529
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@64602
   530
wenzelm@66472
   531
* Session-qualified theory imports allow the Prover IDE to process
wenzelm@66472
   532
arbitrary theory hierarchies independently of the underlying logic
wenzelm@66472
   533
session image (e.g. option "isabelle jedit -l"), but the directory
wenzelm@66472
   534
structure needs to be known in advance (e.g. option "isabelle jedit -d"
wenzelm@66472
   535
or a line in the file $ISABELLE_HOME_USER/ROOTS).
wenzelm@64602
   536
wenzelm@64842
   537
* The PIDE document model maintains file content independently of the
wenzelm@64842
   538
status of jEdit editor buffers. Reloading jEdit buffers no longer causes
wenzelm@64842
   539
changes of formal document content. Theory dependencies are always
wenzelm@64842
   540
resolved internally, without the need for corresponding editor buffers.
wenzelm@64842
   541
The system option "jedit_auto_load" has been discontinued: it is
wenzelm@64842
   542
effectively always enabled.
wenzelm@64842
   543
wenzelm@64867
   544
* The Theories dockable provides a "Purge" button, in order to restrict
wenzelm@64867
   545
the document model to theories that are required for open editor
wenzelm@64867
   546
buffers.
wenzelm@64867
   547
wenzelm@66424
   548
* The Theories dockable indicates the overall status of checking of each
wenzelm@66424
   549
entry. When all forked tasks of a theory are finished, the border is
wenzelm@66424
   550
painted with thick lines; remaining errors in this situation are
wenzelm@66424
   551
represented by a different border color.
wenzelm@66424
   552
wenzelm@66472
   553
* Automatic indentation is more careful to avoid redundant spaces in
wenzelm@66472
   554
intermediate situations. Keywords are indented after input (via typed
wenzelm@66472
   555
characters or completion); see also option "jedit_indent_input".
wenzelm@66472
   556
wenzelm@66472
   557
* Action "isabelle.preview" opens an HTML preview of the current theory
wenzelm@66472
   558
document in the default web browser.
wenzelm@66472
   559
wenzelm@66574
   560
* Command-line invocation "isabelle jedit -R -l LOGIC" opens the ROOT
wenzelm@66574
   561
entry of the specified logic session in the editor, while its parent is
wenzelm@66574
   562
used for formal checking.
wenzelm@66472
   563
wenzelm@66462
   564
* The main Isabelle/jEdit plugin may be restarted manually (using the
wenzelm@66462
   565
jEdit Plugin Manager), as long as the "Isabelle Base" plugin remains
wenzelm@66462
   566
enabled at all times.
wenzelm@66462
   567
wenzelm@66472
   568
* Update to current jedit-5.4.0.
wenzelm@65329
   569
wenzelm@64602
   570
haftmann@66149
   571
*** Pure ***
haftmann@66149
   572
haftmann@66149
   573
* Deleting the last code equations for a particular function using
haftmann@66149
   574
[code del] results in function with no equations (runtime abort) rather
wenzelm@66472
   575
than an unimplemented function (generation time abort). Use explicit
haftmann@66665
   576
[[code drop:]] to enforce the latter. Minor INCOMPATIBILITY.
haftmann@66149
   577
haftmann@66310
   578
* Proper concept of code declarations in code.ML:
wenzelm@66472
   579
  - Regular code declarations act only on the global theory level, being
wenzelm@66472
   580
    ignored with warnings if syntactically malformed.
wenzelm@66472
   581
  - Explicitly global code declarations yield errors if syntactically
wenzelm@66472
   582
    malformed.
wenzelm@66472
   583
  - Default code declarations are silently ignored if syntactically
wenzelm@66472
   584
    malformed.
haftmann@66310
   585
Minor INCOMPATIBILITY.
haftmann@66310
   586
wenzelm@66472
   587
* Clarified and standardized internal data bookkeeping of code
wenzelm@66472
   588
declarations: history of serials allows to track potentially
wenzelm@66472
   589
non-monotonous declarations appropriately. Minor INCOMPATIBILITY.
haftmann@66310
   590
haftmann@66149
   591
haftmann@64593
   592
*** HOL ***
haftmann@64593
   593
blanchet@66614
   594
* The Nunchaku model finder is now part of "Main".
blanchet@66614
   595
wenzelm@66472
   596
* SMT module:
wenzelm@66472
   597
  - A new option, 'smt_nat_as_int', has been added to translate 'nat' to
wenzelm@66472
   598
    'int' and benefit from the SMT solver's theory reasoning. It is
wenzelm@66472
   599
    disabled by default.
wenzelm@66472
   600
  - The legacy module "src/HOL/Library/Old_SMT.thy" has been removed.
wenzelm@66472
   601
  - Several small issues have been rectified in the 'smt' command.
wenzelm@66472
   602
wenzelm@66472
   603
* (Co)datatype package: The 'size_gen_o_map' lemma is no longer
wenzelm@66472
   604
generated for datatypes with type class annotations. As a result, the
wenzelm@66472
   605
tactic that derives it no longer fails on nested datatypes. Slight
wenzelm@66472
   606
INCOMPATIBILITY.
eberlm@66450
   607
haftmann@66345
   608
* Command and antiquotation "value" with modified default strategy:
haftmann@66345
   609
terms without free variables are always evaluated using plain evaluation
wenzelm@66472
   610
only, with no fallback on normalization by evaluation. Minor
wenzelm@66472
   611
INCOMPATIBILITY.
eberlm@65956
   612
wenzelm@65552
   613
* Theories "GCD" and "Binomial" are already included in "Main" (instead
wenzelm@65575
   614
of "Complex_Main").
wenzelm@65552
   615
haftmann@65170
   616
* Constant "surj" is a full input/output abbreviation (again).
haftmann@65170
   617
Minor INCOMPATIBILITY.
haftmann@65170
   618
haftmann@64632
   619
* Dropped aliasses RangeP, DomainP for Rangep, Domainp respectively.
haftmann@64632
   620
INCOMPATIBILITY.
haftmann@64632
   621
wenzelm@66472
   622
* Renamed ii to imaginary_unit in order to free up ii as a variable
wenzelm@66472
   623
name. The syntax \<i> remains available. INCOMPATIBILITY.
wenzelm@66472
   624
wenzelm@66472
   625
* Dropped abbreviations transP, antisymP, single_valuedP; use constants
wenzelm@66472
   626
transp, antisymp, single_valuedp instead. INCOMPATIBILITY.
wenzelm@66472
   627
wenzelm@66472
   628
* Constant "subseq" in Topological_Spaces has been removed -- it is
wenzelm@66472
   629
subsumed by "strict_mono". Some basic lemmas specific to "subseq" have
wenzelm@66472
   630
been renamed accordingly, e.g. "subseq_o" -> "strict_mono_o" etc.
wenzelm@66472
   631
wenzelm@66472
   632
* Theory List: "sublist" renamed to "nths" in analogy with "nth", and
wenzelm@66472
   633
"sublisteq" renamed to "subseq". Minor INCOMPATIBILITY.
wenzelm@66472
   634
wenzelm@66472
   635
* Theory List: new generic function "sorted_wrt".
wenzelm@66472
   636
wenzelm@66472
   637
* Named theorems mod_simps covers various congruence rules concerning
wenzelm@66472
   638
mod, replacing former zmod_simps. INCOMPATIBILITY.
haftmann@64787
   639
haftmann@64593
   640
* Swapped orientation of congruence rules mod_add_left_eq,
haftmann@64593
   641
mod_add_right_eq, mod_add_eq, mod_mult_left_eq, mod_mult_right_eq,
haftmann@64593
   642
mod_mult_eq, mod_minus_eq, mod_diff_left_eq, mod_diff_right_eq,
wenzelm@66472
   643
mod_diff_eq. INCOMPATIBILITY.
haftmann@64593
   644
haftmann@64593
   645
* Generalized some facts:
blanchet@64876
   646
    measure_induct_rule
blanchet@64876
   647
    measure_induct
haftmann@64593
   648
    zminus_zmod ~> mod_minus_eq
haftmann@64593
   649
    zdiff_zmod_left ~> mod_diff_left_eq
haftmann@64593
   650
    zdiff_zmod_right ~> mod_diff_right_eq
haftmann@64593
   651
    zmod_eq_dvd_iff ~> mod_eq_dvd_iff
haftmann@64593
   652
INCOMPATIBILITY.
haftmann@64593
   653
wenzelm@66472
   654
* Algebraic type class hierarchy of euclidean (semi)rings in HOL:
wenzelm@66472
   655
euclidean_(semi)ring, euclidean_(semi)ring_cancel,
wenzelm@66472
   656
unique_euclidean_(semi)ring; instantiation requires provision of a
wenzelm@66472
   657
euclidean size.
wenzelm@66472
   658
wenzelm@66472
   659
* Theory "HOL-Number_Theory.Euclidean_Algorithm" has been reworked:
wenzelm@66472
   660
  - Euclidean induction is available as rule eucl_induct.
wenzelm@66472
   661
  - Constants Euclidean_Algorithm.gcd, Euclidean_Algorithm.lcm,
wenzelm@66472
   662
    Euclidean_Algorithm.Gcd and Euclidean_Algorithm.Lcm allow
wenzelm@66472
   663
    easy instantiation of euclidean (semi)rings as GCD (semi)rings.
wenzelm@66472
   664
  - Coefficients obtained by extended euclidean algorithm are
wenzelm@66472
   665
    available as "bezout_coefficients".
wenzelm@66472
   666
INCOMPATIBILITY.
wenzelm@66472
   667
wenzelm@66472
   668
* Theory "Number_Theory.Totient" introduces basic notions about Euler's
wenzelm@66472
   669
totient function previously hidden as solitary example in theory
wenzelm@66472
   670
Residues. Definition changed so that "totient 1 = 1" in agreement with
wenzelm@66472
   671
the literature. Minor INCOMPATIBILITY.
wenzelm@66472
   672
wenzelm@66542
   673
* New styles in theory "HOL-Library.LaTeXsugar":
nipkow@66541
   674
  - "dummy_pats" for printing equations with "_" on the lhs;
nipkow@66541
   675
  - "eta_expand" for printing eta-expanded terms.
nipkow@66541
   676
wenzelm@66472
   677
* Theory "HOL-Library.Permutations": theorem bij_swap_ompose_bij has
wenzelm@66472
   678
been renamed to bij_swap_compose_bij. INCOMPATIBILITY.
wenzelm@66472
   679
lp15@66643
   680
* New theory "HOL-Library.Going_To_Filter" providing the "f going_to F"
eberlm@66488
   681
filter for describing points x such that f(x) is in the filter F.
eberlm@66488
   682
eberlm@66480
   683
* Theory "HOL-Library.Formal_Power_Series": constants X/E/L/F have been
eberlm@66480
   684
renamed to fps_X/fps_exp/fps_ln/fps_hypergeo to avoid polluting the name
wenzelm@66472
   685
space. INCOMPATIBILITY.
wenzelm@66472
   686
wenzelm@66472
   687
* Theory "HOL-Library.FinFun" has been moved to AFP (again).
wenzelm@66472
   688
INCOMPATIBILITY.
wenzelm@66472
   689
wenzelm@66472
   690
* Theory "HOL-Library.FuncSet": some old and rarely used ASCII
wenzelm@66472
   691
replacement syntax has been removed. INCOMPATIBILITY, standard syntax
wenzelm@66472
   692
with symbols should be used instead. The subsequent commands help to
wenzelm@66472
   693
reproduce the old forms, e.g. to simplify porting old theories:
wenzelm@66472
   694
wenzelm@66472
   695
syntax (ASCII)
wenzelm@66472
   696
  "_PiE" :: "pttrn \<Rightarrow> 'a set \<Rightarrow> 'b set \<Rightarrow> ('a \<Rightarrow> 'b) set"  ("(3PIE _:_./ _)" 10)
wenzelm@66472
   697
  "_Pi"  :: "pttrn \<Rightarrow> 'a set \<Rightarrow> 'b set \<Rightarrow> ('a \<Rightarrow> 'b) set"  ("(3PI _:_./ _)" 10)
wenzelm@66472
   698
  "_lam" :: "pttrn \<Rightarrow> 'a set \<Rightarrow> 'a \<Rightarrow> 'b \<Rightarrow> ('a \<Rightarrow> 'b)"  ("(3%_:_./ _)" [0,0,3] 3)
wenzelm@66472
   699
wenzelm@66472
   700
* Theory "HOL-Library.Multiset": the simprocs on subsets operators of
wenzelm@66472
   701
multisets have been renamed:
wenzelm@66472
   702
wenzelm@66472
   703
  msetless_cancel_numerals ~> msetsubset_cancel
wenzelm@66472
   704
  msetle_cancel_numerals ~> msetsubset_eq_cancel
wenzelm@66472
   705
wenzelm@66472
   706
INCOMPATIBILITY.
wenzelm@66472
   707
lars@66481
   708
* Theory "HOL-Library.Pattern_Aliases" provides input and output syntax
lars@66481
   709
for pattern aliases as known from Haskell, Scala and ML.
blanchet@65515
   710
Andreas@66563
   711
* Theory "HOL-Library.Uprod" formalizes the type of unordered pairs.
Andreas@66563
   712
wenzelm@64898
   713
* Session HOL-Analysis: more material involving arcs, paths, covering
wenzelm@66650
   714
spaces, innessential maps, retracts, infinite products, simplicial
wenzelm@66650
   715
complexes. Baire Category theorem. Major results include the Jordan
wenzelm@66650
   716
Curve Theorem and the Great Picard Theorem.
wenzelm@66472
   717
wenzelm@66472
   718
* Session HOL-Algebra has been extended by additional lattice theory:
wenzelm@66472
   719
the Knaster-Tarski fixed point theorem and Galois Connections.
wenzelm@66472
   720
wenzelm@66472
   721
* Sessions HOL-Computational_Algebra and HOL-Number_Theory: new notions
wenzelm@66472
   722
of squarefreeness, n-th powers, and prime powers.
wenzelm@66472
   723
wenzelm@66472
   724
* Session "HOL-Computional_Algebra" covers many previously scattered
wenzelm@66472
   725
theories, notably Euclidean_Algorithm, Factorial_Ring,
wenzelm@66472
   726
Formal_Power_Series, Fraction_Field, Fundamental_Theorem_Algebra,
wenzelm@66472
   727
Normalized_Fraction, Polynomial_FPS, Polynomial, Primes. Minor
wenzelm@66472
   728
INCOMPATIBILITY.
Mathias@65027
   729
wenzelm@60331
   730
wenzelm@64844
   731
*** System ***
wenzelm@64844
   732
wenzelm@66474
   733
* Isabelle/Scala: the SQL module supports access to relational
wenzelm@66474
   734
databases, either as plain file (SQLite) or full-scale server
wenzelm@66474
   735
(PostgreSQL via local port or remote ssh connection).
wenzelm@66474
   736
wenzelm@66474
   737
* Results of "isabelle build" are recorded as SQLite database (i.e.
wenzelm@66474
   738
"Application File Format" in the sense of
wenzelm@66474
   739
https://www.sqlite.org/appfileformat.html). This allows systematic
wenzelm@66474
   740
access via operations from module Sessions.Store in Isabelle/Scala.
wenzelm@66474
   741
wenzelm@66472
   742
* System option "parallel_proofs" is 1 by default (instead of more
wenzelm@66472
   743
aggressive 2). This requires less heap space and avoids burning parallel
wenzelm@66472
   744
CPU cycles, while full subproof parallelization is enabled for repeated
wenzelm@66472
   745
builds (according to parallel_subproofs_threshold).
wenzelm@65072
   746
wenzelm@65448
   747
* System option "record_proofs" allows to change the global
wenzelm@65448
   748
Proofterm.proofs variable for a session. Regular values are are 0, 1, 2;
wenzelm@65448
   749
a negative value means the current state in the ML heap image remains
wenzelm@65448
   750
unchanged.
wenzelm@65448
   751
wenzelm@66472
   752
* Isabelle settings variable ISABELLE_SCALA_BUILD_OPTIONS has been
wenzelm@66472
   753
renamed to ISABELLE_SCALAC_OPTIONS. Rare INCOMPATIBILITY.
wenzelm@66472
   754
wenzelm@66472
   755
* Isabelle settings variables ISABELLE_WINDOWS_PLATFORM,
wenzelm@66472
   756
ISABELLE_WINDOWS_PLATFORM32, ISABELLE_WINDOWS_PLATFORM64 indicate the
wenzelm@66472
   757
native Windows platform (independently of the Cygwin installation). This
wenzelm@66472
   758
is analogous to ISABELLE_PLATFORM, ISABELLE_PLATFORM32,
wenzelm@66472
   759
ISABELLE_PLATFORM64.
wenzelm@65557
   760
wenzelm@66786
   761
* Command-line tool "isabelle build_docker" builds a Docker image from
wenzelm@66786
   762
the Isabelle application bundle for Linux. See also
wenzelm@66786
   763
https://hub.docker.com/r/makarius/isabelle
wenzelm@66786
   764
wenzelm@66473
   765
* Command-line tool "isabelle vscode_server" provides a Language Server
wenzelm@66473
   766
Protocol implementation, e.g. for the Visual Studio Code editor. It
wenzelm@66473
   767
serves as example for alternative PIDE front-ends.
wenzelm@66473
   768
wenzelm@66472
   769
* Command-line tool "isabelle imports" helps to maintain theory imports
wenzelm@66671
   770
wrt. session structure. Examples for the main Isabelle distribution:
wenzelm@66472
   771
wenzelm@66472
   772
  isabelle imports -I -a
wenzelm@66472
   773
  isabelle imports -U -a
wenzelm@66472
   774
  isabelle imports -U -i -a
wenzelm@66472
   775
  isabelle imports -M -a -d '~~/src/Benchmarks'
wenzelm@66472
   776
wenzelm@64844
   777
wenzelm@67099
   778
wenzelm@64072
   779
New in Isabelle2016-1 (December 2016)
wenzelm@64072
   780
-------------------------------------
wenzelm@62216
   781
wenzelm@62440
   782
*** General ***
wenzelm@62440
   783
wenzelm@64390
   784
* Splitter in proof methods "simp", "auto" and friends:
wenzelm@64390
   785
  - The syntax "split add" has been discontinued, use plain "split",
wenzelm@64390
   786
    INCOMPATIBILITY.
wenzelm@64390
   787
  - For situations with many conditional or case expressions, there is
wenzelm@64390
   788
    an alternative splitting strategy that can be much faster. It is
wenzelm@64390
   789
    selected by writing "split!" instead of "split". It applies safe
wenzelm@64390
   790
    introduction and elimination rules after each split rule. As a
wenzelm@64390
   791
    result the subgoal may be split into several subgoals.
wenzelm@64390
   792
wenzelm@63273
   793
* Command 'bundle' provides a local theory target to define a bundle
wenzelm@63273
   794
from the body of specification commands (such as 'declare',
wenzelm@63273
   795
'declaration', 'notation', 'lemmas', 'lemma'). For example:
wenzelm@63273
   796
wenzelm@63273
   797
bundle foo
wenzelm@63273
   798
begin
wenzelm@63273
   799
  declare a [simp]
wenzelm@63273
   800
  declare b [intro]
wenzelm@63273
   801
end
wenzelm@63272
   802
wenzelm@63282
   803
* Command 'unbundle' is like 'include', but works within a local theory
wenzelm@63282
   804
context. Unlike "context includes ... begin", the effect of 'unbundle'
wenzelm@63282
   805
on the target context persists, until different declarations are given.
wenzelm@63282
   806
wenzelm@63977
   807
* Simplified outer syntax: uniform category "name" includes long
wenzelm@63977
   808
identifiers. Former "xname" / "nameref" / "name reference" has been
wenzelm@63977
   809
discontinued.
wenzelm@63977
   810
wenzelm@63977
   811
* Embedded content (e.g. the inner syntax of types, terms, props) may be
wenzelm@63977
   812
delimited uniformly via cartouches. This works better than old-fashioned
wenzelm@63977
   813
quotes when sub-languages are nested.
wenzelm@63977
   814
wenzelm@63977
   815
* Mixfix annotations support general block properties, with syntax
wenzelm@63977
   816
"(\<open>x=a y=b z \<dots>\<close>". Notable property names are "indent", "consistent",
wenzelm@63977
   817
"unbreakable", "markup". The existing notation "(DIGITS" is equivalent
wenzelm@63977
   818
to "(\<open>indent=DIGITS\<close>". The former notation "(00" for unbreakable blocks
wenzelm@63977
   819
is superseded by "(\<open>unbreabable\<close>" --- rare INCOMPATIBILITY.
nipkow@63650
   820
wenzelm@63383
   821
* Proof method "blast" is more robust wrt. corner cases of Pure
wenzelm@63383
   822
statements without object-logic judgment.
wenzelm@63383
   823
wenzelm@63624
   824
* Commands 'prf' and 'full_prf' are somewhat more informative (again):
wenzelm@63977
   825
proof terms are reconstructed and cleaned from administrative thm nodes.
wenzelm@63977
   826
wenzelm@63977
   827
* Code generator: config option "code_timing" triggers measurements of
wenzelm@63977
   828
different phases of code generation. See src/HOL/ex/Code_Timing.thy for
wenzelm@63977
   829
examples.
wenzelm@63977
   830
wenzelm@63977
   831
* Code generator: implicits in Scala (stemming from type class
wenzelm@63977
   832
instances) are generated into companion object of corresponding type
wenzelm@63977
   833
class, to resolve some situations where ambiguities may occur.
wenzelm@63977
   834
wenzelm@64390
   835
* Solve direct: option "solve_direct_strict_warnings" gives explicit
wenzelm@64390
   836
warnings for lemma statements with trivial proofs.
haftmann@64013
   837
wenzelm@64602
   838
wenzelm@62904
   839
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@62904
   840
wenzelm@64527
   841
* More aggressive flushing of machine-generated input, according to
wenzelm@64527
   842
system option editor_generated_input_delay (in addition to existing
wenzelm@64527
   843
editor_input_delay for regular user edits). This may affect overall PIDE
wenzelm@64527
   844
reactivity and CPU usage.
wenzelm@64527
   845
wenzelm@64390
   846
* Syntactic indentation according to Isabelle outer syntax. Action
wenzelm@64390
   847
"indent-lines" (shortcut C+i) indents the current line according to
wenzelm@64390
   848
command keywords and some command substructure. Action
wenzelm@64390
   849
"isabelle.newline" (shortcut ENTER) indents the old and the new line
wenzelm@64390
   850
according to command keywords only; see also option
wenzelm@64390
   851
"jedit_indent_newline".
wenzelm@64390
   852
wenzelm@64390
   853
* Semantic indentation for unstructured proof scripts ('apply' etc.) via
wenzelm@64390
   854
number of subgoals. This requires information of ongoing document
wenzelm@64390
   855
processing and may thus lag behind, when the user is editing too
wenzelm@64390
   856
quickly; see also option "jedit_script_indent" and
wenzelm@64390
   857
"jedit_script_indent_limit".
wenzelm@64390
   858
wenzelm@64390
   859
* Refined folding mode "isabelle" based on Isar syntax: 'next' and 'qed'
wenzelm@64390
   860
are treated as delimiters for fold structure; 'begin' and 'end'
wenzelm@64390
   861
structure of theory specifications is treated as well.
wenzelm@64390
   862
wenzelm@64390
   863
* Command 'proof' provides information about proof outline with cases,
wenzelm@64390
   864
e.g. for proof methods "cases", "induct", "goal_cases".
wenzelm@64390
   865
wenzelm@64390
   866
* Completion templates for commands involving "begin ... end" blocks,
wenzelm@64390
   867
e.g. 'context', 'notepad'.
wenzelm@64390
   868
wenzelm@64390
   869
* Sidekick parser "isabelle-context" shows nesting of context blocks
wenzelm@64390
   870
according to 'begin' and 'end' structure.
wenzelm@64390
   871
wenzelm@63977
   872
* Highlighting of entity def/ref positions wrt. cursor.
wenzelm@63977
   873
wenzelm@63977
   874
* Action "isabelle.select-entity" (shortcut CS+ENTER) selects all
wenzelm@64514
   875
occurrences of the formal entity at the caret position. This facilitates
wenzelm@63977
   876
systematic renaming.
wenzelm@63977
   877
wenzelm@63977
   878
* PIDE document markup works across multiple Isar commands, e.g. the
wenzelm@63977
   879
results established at the end of a proof are properly identified in the
wenzelm@63977
   880
theorem statement.
wenzelm@63977
   881
wenzelm@63977
   882
* Cartouche abbreviations work both for " and ` to accomodate typical
wenzelm@63977
   883
situations where old ASCII notation may be updated.
wenzelm@63977
   884
wenzelm@63875
   885
* Dockable window "Symbols" also provides access to 'abbrevs' from the
wenzelm@63875
   886
outer syntax of the current theory buffer. This provides clickable
wenzelm@63875
   887
syntax templates, including entries with empty abbrevs name (which are
wenzelm@63875
   888
inaccessible via keyboard completion).
wenzelm@63875
   889
wenzelm@63022
   890
* IDE support for the Isabelle/Pure bootstrap process, with the
wenzelm@63022
   891
following independent stages:
wenzelm@63022
   892
wenzelm@63022
   893
  src/Pure/ROOT0.ML
wenzelm@63022
   894
  src/Pure/ROOT.ML
wenzelm@63022
   895
  src/Pure/Pure.thy
wenzelm@63022
   896
  src/Pure/ML_Bootstrap.thy
wenzelm@63022
   897
wenzelm@63022
   898
The ML ROOT files act like quasi-theories in the context of theory
wenzelm@63022
   899
ML_Bootstrap: this allows continuous checking of all loaded ML files.
wenzelm@63022
   900
The theory files are presented with a modified header to import Pure
wenzelm@63022
   901
from the running Isabelle instance. Results from changed versions of
wenzelm@63022
   902
each stage are *not* propagated to the next stage, and isolated from the
wenzelm@63022
   903
actual Isabelle/Pure that runs the IDE itself. The sequential
wenzelm@63307
   904
dependencies of the above files are only observed for batch build.
wenzelm@62904
   905
wenzelm@63977
   906
* Isabelle/ML and Standard ML files are presented in Sidekick with the
wenzelm@63977
   907
tree structure of section headings: this special comment format is
wenzelm@63977
   908
described in "implementation" chapter 0, e.g. (*** section ***).
wenzelm@63461
   909
wenzelm@63581
   910
* Additional abbreviations for syntactic completion may be specified
wenzelm@63871
   911
within the theory header as 'abbrevs'. The theory syntax for 'keywords'
wenzelm@63871
   912
has been simplified accordingly: optional abbrevs need to go into the
wenzelm@63871
   913
new 'abbrevs' section.
wenzelm@63871
   914
wenzelm@63871
   915
* Global abbreviations via $ISABELLE_HOME/etc/abbrevs and
wenzelm@63871
   916
$ISABELLE_HOME_USER/etc/abbrevs are no longer supported. Minor
wenzelm@63871
   917
INCOMPATIBILITY, use 'abbrevs' within theory header instead.
wenzelm@63579
   918
wenzelm@64390
   919
* Action "isabelle.keymap-merge" asks the user to resolve pending
wenzelm@64390
   920
Isabelle keymap changes that are in conflict with the current jEdit
wenzelm@64390
   921
keymap; non-conflicting changes are always applied implicitly. This
wenzelm@64390
   922
action is automatically invoked on Isabelle/jEdit startup and thus
wenzelm@64390
   923
increases chances that users see new keyboard shortcuts when re-using
wenzelm@64390
   924
old keymaps.
wenzelm@64390
   925
wenzelm@63675
   926
* ML and document antiquotations for file-systems paths are more uniform
wenzelm@63675
   927
and diverse:
wenzelm@63675
   928
wenzelm@63675
   929
  @{path NAME}   -- no file-system check
wenzelm@63675
   930
  @{file NAME}   -- check for plain file
wenzelm@63675
   931
  @{dir NAME}    -- check for directory
wenzelm@63675
   932
wenzelm@63675
   933
Minor INCOMPATIBILITY, former uses of @{file} and @{file_unchecked} may
wenzelm@63675
   934
have to be changed.
wenzelm@63669
   935
wenzelm@63669
   936
wenzelm@63977
   937
*** Document preparation ***
wenzelm@63977
   938
wenzelm@63977
   939
* New symbol \<circle>, e.g. for temporal operator.
wenzelm@63977
   940
wenzelm@64073
   941
* New document and ML antiquotation @{locale} for locales, similar to
wenzelm@64073
   942
existing antiquotation @{class}.
wenzelm@64073
   943
wenzelm@63977
   944
* Mixfix annotations support delimiters like \<^control>\<open>cartouche\<close> --
wenzelm@63977
   945
this allows special forms of document output.
wenzelm@63977
   946
wenzelm@63977
   947
* Raw LaTeX output now works via \<^latex>\<open>...\<close> instead of raw control
wenzelm@63977
   948
symbol \<^raw:...>. INCOMPATIBILITY, notably for LaTeXsugar.thy and its
wenzelm@63977
   949
derivatives.
wenzelm@63977
   950
wenzelm@63977
   951
* \<^raw:...> symbols are no longer supported.
wenzelm@63977
   952
wenzelm@63977
   953
* Old 'header' command is no longer supported (legacy since
wenzelm@63977
   954
Isabelle2015).
wenzelm@63977
   955
wenzelm@63977
   956
wenzelm@62312
   957
*** Isar ***
wenzelm@62312
   958
wenzelm@63180
   959
* Many specification elements support structured statements with 'if' /
wenzelm@63180
   960
'for' eigen-context, e.g. 'axiomatization', 'abbreviation',
wenzelm@63180
   961
'definition', 'inductive', 'function'.
wenzelm@63180
   962
wenzelm@63094
   963
* Toplevel theorem statements support eigen-context notation with 'if' /
wenzelm@63284
   964
'for' (in postfix), which corresponds to 'assumes' / 'fixes' in the
wenzelm@63094
   965
traditional long statement form (in prefix). Local premises are called
wenzelm@63094
   966
"that" or "assms", respectively. Empty premises are *not* bound in the
wenzelm@63094
   967
context: INCOMPATIBILITY.
wenzelm@63094
   968
wenzelm@63039
   969
* Command 'define' introduces a local (non-polymorphic) definition, with
wenzelm@63039
   970
optional abstraction over local parameters. The syntax resembles
wenzelm@63043
   971
'definition' and 'obtain'. It fits better into the Isar language than
wenzelm@63043
   972
old 'def', which is now a legacy feature.
wenzelm@63039
   973
wenzelm@63059
   974
* Command 'obtain' supports structured statements with 'if' / 'for'
wenzelm@63059
   975
context.
wenzelm@63059
   976
wenzelm@62312
   977
* Command '\<proof>' is an alias for 'sorry', with different
wenzelm@62312
   978
typesetting. E.g. to produce proof holes in examples and documentation.
wenzelm@62216
   979
wenzelm@63977
   980
* The defining position of a literal fact \<open>prop\<close> is maintained more
wenzelm@63977
   981
carefully, and made accessible as hyperlink in the Prover IDE.
wenzelm@63977
   982
wenzelm@63977
   983
* Commands 'finally' and 'ultimately' used to expose the result as
wenzelm@63977
   984
literal fact: this accidental behaviour has been discontinued. Rare
wenzelm@63977
   985
INCOMPATIBILITY, use more explicit means to refer to facts in Isar.
wenzelm@63977
   986
wenzelm@63977
   987
* Command 'axiomatization' has become more restrictive to correspond
wenzelm@63977
   988
better to internal axioms as singleton facts with mandatory name. Minor
wenzelm@63977
   989
INCOMPATIBILITY.
wenzelm@62939
   990
wenzelm@63259
   991
* Proof methods may refer to the main facts via the dynamic fact
wenzelm@63259
   992
"method_facts". This is particularly useful for Eisbach method
wenzelm@63259
   993
definitions.
wenzelm@63259
   994
wenzelm@63527
   995
* Proof method "use" allows to modify the main facts of a given method
wenzelm@63527
   996
expression, e.g.
wenzelm@63259
   997
wenzelm@63259
   998
  (use facts in simp)
wenzelm@63259
   999
  (use facts in \<open>simp add: ...\<close>)
wenzelm@63259
  1000
wenzelm@63977
  1001
* The old proof method "default" has been removed (legacy since
wenzelm@63977
  1002
Isabelle2016). INCOMPATIBILITY, use "standard" instead.
wenzelm@63977
  1003
wenzelm@62216
  1004
haftmann@63165
  1005
*** Pure ***
haftmann@63165
  1006
wenzelm@63977
  1007
* Pure provides basic versions of proof methods "simp" and "simp_all"
wenzelm@63977
  1008
that only know about meta-equality (==). Potential INCOMPATIBILITY in
wenzelm@63977
  1009
theory imports that merge Pure with e.g. Main of Isabelle/HOL: the order
wenzelm@63977
  1010
is relevant to avoid confusion of Pure.simp vs. HOL.simp.
wenzelm@63977
  1011
wenzelm@63977
  1012
* The command 'unfolding' and proof method "unfold" include a second
wenzelm@63977
  1013
stage where given equations are passed through the attribute "abs_def"
wenzelm@63977
  1014
before rewriting. This ensures that definitions are fully expanded,
wenzelm@63977
  1015
regardless of the actual parameters that are provided. Rare
wenzelm@63977
  1016
INCOMPATIBILITY in some corner cases: use proof method (simp only:)
wenzelm@63977
  1017
instead, or declare [[unfold_abs_def = false]] in the proof context.
wenzelm@63977
  1018
wenzelm@63977
  1019
* Type-inference improves sorts of newly introduced type variables for
wenzelm@63977
  1020
the object-logic, using its base sort (i.e. HOL.type for Isabelle/HOL).
wenzelm@63977
  1021
Thus terms like "f x" or "\<And>x. P x" without any further syntactic context
wenzelm@63977
  1022
produce x::'a::type in HOL instead of x::'a::{} in Pure. Rare
wenzelm@63977
  1023
INCOMPATIBILITY, need to provide explicit type constraints for Pure
wenzelm@63977
  1024
types where this is really intended.
haftmann@63350
  1025
haftmann@63165
  1026
blanchet@62327
  1027
*** HOL ***
blanchet@62327
  1028
wenzelm@63977
  1029
* New proof method "argo" using the built-in Argo solver based on SMT
wenzelm@63977
  1030
technology. The method can be used to prove goals of quantifier-free
wenzelm@63977
  1031
propositional logic, goals based on a combination of quantifier-free
wenzelm@63977
  1032
propositional logic with equality, and goals based on a combination of
wenzelm@63977
  1033
quantifier-free propositional logic with linear real arithmetic
wenzelm@63977
  1034
including min/max/abs. See HOL/ex/Argo_Examples.thy for examples.
wenzelm@63977
  1035
blanchet@66614
  1036
* The new "nunchaku" command integrates the Nunchaku model finder. The
wenzelm@64390
  1037
tool is experimental. See ~~/src/HOL/Nunchaku/Nunchaku.thy for details.
wenzelm@64390
  1038
wenzelm@63977
  1039
* Metis: The problem encoding has changed very slightly. This might
blanchet@63785
  1040
break existing proofs. INCOMPATIBILITY.
blanchet@63785
  1041
blanchet@63116
  1042
* Sledgehammer:
lp15@63967
  1043
  - The MaSh relevance filter is now faster than before.
blanchet@63116
  1044
  - Produce syntactically correct Vampire 4.0 problem files.
blanchet@63116
  1045
blanchet@62327
  1046
* (Co)datatype package:
blanchet@62693
  1047
  - New commands for defining corecursive functions and reasoning about
blanchet@62693
  1048
    them in "~~/src/HOL/Library/BNF_Corec.thy": 'corec', 'corecursive',
blanchet@62693
  1049
    'friend_of_corec', and 'corecursion_upto'; and 'corec_unique' proof
blanchet@62842
  1050
    method. See 'isabelle doc corec'.
wenzelm@63977
  1051
  - The predicator :: ('a \<Rightarrow> bool) \<Rightarrow> 'a F \<Rightarrow> bool is now a first-class
blanchet@63855
  1052
    citizen in bounded natural functors.
blanchet@62693
  1053
  - 'primrec' now allows nested calls through the predicator in addition
blanchet@62327
  1054
    to the map function.
blanchet@63855
  1055
  - 'bnf' automatically discharges reflexive proof obligations.
blanchet@62693
  1056
  - 'bnf' outputs a slightly modified proof obligation expressing rel in
traytel@62332
  1057
       terms of map and set
blanchet@63855
  1058
       (not giving a specification for rel makes this one reflexive).
blanchet@62693
  1059
  - 'bnf' outputs a new proof obligation expressing pred in terms of set
blanchet@63855
  1060
       (not giving a specification for pred makes this one reflexive).
blanchet@63855
  1061
    INCOMPATIBILITY: manual 'bnf' declarations may need adjustment.
blanchet@62335
  1062
  - Renamed lemmas:
blanchet@62335
  1063
      rel_prod_apply ~> rel_prod_inject
blanchet@62335
  1064
      pred_prod_apply ~> pred_prod_inject
blanchet@62335
  1065
    INCOMPATIBILITY.
blanchet@62536
  1066
  - The "size" plugin has been made compatible again with locales.
blanchet@63855
  1067
  - The theorems about "rel" and "set" may have a slightly different (but
blanchet@63855
  1068
    equivalent) form.
blanchet@63855
  1069
    INCOMPATIBILITY.
blanchet@62327
  1070
wenzelm@63977
  1071
* The 'coinductive' command produces a proper coinduction rule for
wenzelm@63977
  1072
mutual coinductive predicates. This new rule replaces the old rule,
wenzelm@63977
  1073
which exposed details of the internal fixpoint construction and was
wenzelm@63977
  1074
hard to use. INCOMPATIBILITY.
wenzelm@63977
  1075
wenzelm@63977
  1076
* New abbreviations for negated existence (but not bounded existence):
wenzelm@63977
  1077
wenzelm@63977
  1078
  \<nexists>x. P x \<equiv> \<not> (\<exists>x. P x)
wenzelm@63977
  1079
  \<nexists>!x. P x \<equiv> \<not> (\<exists>!x. P x)
wenzelm@63977
  1080
wenzelm@63977
  1081
* The print mode "HOL" for ASCII syntax of binders "!", "?", "?!", "@"
wenzelm@63977
  1082
has been removed for output. It is retained for input only, until it is
wenzelm@63977
  1083
eliminated altogether.
wenzelm@63977
  1084
wenzelm@63977
  1085
* The unique existence quantifier no longer provides 'binder' syntax,
wenzelm@63977
  1086
but uses syntax translations (as for bounded unique existence). Thus
wenzelm@63977
  1087
iterated quantification \<exists>!x y. P x y with its slightly confusing
wenzelm@63977
  1088
sequential meaning \<exists>!x. \<exists>!y. P x y is no longer possible. Instead,
wenzelm@63977
  1089
pattern abstraction admits simultaneous unique existence \<exists>!(x, y). P x y
wenzelm@63977
  1090
(analogous to existing notation \<exists>!(x, y)\<in>A. P x y). Potential
wenzelm@63977
  1091
INCOMPATIBILITY in rare situations.
wenzelm@63977
  1092
wenzelm@63977
  1093
* Conventional syntax "%(). t" for unit abstractions. Slight syntactic
wenzelm@63977
  1094
INCOMPATIBILITY.
wenzelm@63977
  1095
wenzelm@64390
  1096
* Renamed constants and corresponding theorems:
wenzelm@64390
  1097
wenzelm@64390
  1098
    setsum ~> sum
wenzelm@64390
  1099
    setprod ~> prod
wenzelm@64390
  1100
    listsum ~> sum_list
wenzelm@64390
  1101
    listprod ~> prod_list
wenzelm@64390
  1102
wenzelm@64390
  1103
INCOMPATIBILITY.
wenzelm@64390
  1104
wenzelm@64390
  1105
* Sligthly more standardized theorem names:
wenzelm@64390
  1106
    sgn_times ~> sgn_mult
wenzelm@64390
  1107
    sgn_mult' ~> Real_Vector_Spaces.sgn_mult
wenzelm@64390
  1108
    divide_zero_left ~> div_0
wenzelm@64390
  1109
    zero_mod_left ~> mod_0
wenzelm@64390
  1110
    divide_zero ~> div_by_0
wenzelm@64390
  1111
    divide_1 ~> div_by_1
wenzelm@64390
  1112
    nonzero_mult_divide_cancel_left ~> nonzero_mult_div_cancel_left
wenzelm@64390
  1113
    div_mult_self1_is_id ~> nonzero_mult_div_cancel_left
wenzelm@64390
  1114
    nonzero_mult_divide_cancel_right ~> nonzero_mult_div_cancel_right
wenzelm@64390
  1115
    div_mult_self2_is_id ~> nonzero_mult_div_cancel_right
wenzelm@64390
  1116
    is_unit_divide_mult_cancel_left ~> is_unit_div_mult_cancel_left
wenzelm@64390
  1117
    is_unit_divide_mult_cancel_right ~> is_unit_div_mult_cancel_right
wenzelm@64390
  1118
    mod_div_equality ~> div_mult_mod_eq
wenzelm@64390
  1119
    mod_div_equality2 ~> mult_div_mod_eq
wenzelm@64390
  1120
    mod_div_equality3 ~> mod_div_mult_eq
wenzelm@64390
  1121
    mod_div_equality4 ~> mod_mult_div_eq
wenzelm@64390
  1122
    minus_div_eq_mod ~> minus_div_mult_eq_mod
wenzelm@64390
  1123
    minus_div_eq_mod2 ~> minus_mult_div_eq_mod
wenzelm@64390
  1124
    minus_mod_eq_div ~> minus_mod_eq_div_mult
wenzelm@64390
  1125
    minus_mod_eq_div2 ~> minus_mod_eq_mult_div
wenzelm@64390
  1126
    div_mod_equality' ~> minus_mod_eq_div_mult [symmetric]
wenzelm@64390
  1127
    mod_div_equality' ~> minus_div_mult_eq_mod [symmetric]
wenzelm@64390
  1128
    zmod_zdiv_equality ~> mult_div_mod_eq [symmetric]
wenzelm@64390
  1129
    zmod_zdiv_equality' ~> minus_div_mult_eq_mod [symmetric]
wenzelm@64390
  1130
    Divides.mult_div_cancel ~> minus_mod_eq_mult_div [symmetric]
wenzelm@64390
  1131
    mult_div_cancel ~> minus_mod_eq_mult_div [symmetric]
wenzelm@64390
  1132
    zmult_div_cancel ~> minus_mod_eq_mult_div [symmetric]
wenzelm@64390
  1133
    div_1 ~> div_by_Suc_0
wenzelm@64390
  1134
    mod_1 ~> mod_by_Suc_0
wenzelm@64390
  1135
INCOMPATIBILITY.
wenzelm@64390
  1136
wenzelm@64390
  1137
* New type class "idom_abs_sgn" specifies algebraic properties
wenzelm@64390
  1138
of sign and absolute value functions.  Type class "sgn_if" has
wenzelm@64390
  1139
disappeared.  Slight INCOMPATIBILITY.
wenzelm@64390
  1140
wenzelm@64390
  1141
* Dedicated syntax LENGTH('a) for length of types.
wenzelm@64390
  1142
wenzelm@63977
  1143
* Characters (type char) are modelled as finite algebraic type
wenzelm@63977
  1144
corresponding to {0..255}.
wenzelm@63977
  1145
wenzelm@63977
  1146
  - Logical representation:
wenzelm@63977
  1147
    * 0 is instantiated to the ASCII zero character.
wenzelm@63977
  1148
    * All other characters are represented as "Char n"
wenzelm@63977
  1149
      with n being a raw numeral expression less than 256.
wenzelm@63977
  1150
    * Expressions of the form "Char n" with n greater than 255
wenzelm@63977
  1151
      are non-canonical.
wenzelm@63977
  1152
  - Printing and parsing:
wenzelm@63977
  1153
    * Printable characters are printed and parsed as "CHR ''\<dots>''"
wenzelm@63977
  1154
      (as before).
wenzelm@63977
  1155
    * The ASCII zero character is printed and parsed as "0".
wenzelm@63977
  1156
    * All other canonical characters are printed as "CHR 0xXX"
wenzelm@63977
  1157
      with XX being the hexadecimal character code.  "CHR n"
wenzelm@63977
  1158
      is parsable for every numeral expression n.
wenzelm@63977
  1159
    * Non-canonical characters have no special syntax and are
wenzelm@63977
  1160
      printed as their logical representation.
wenzelm@63977
  1161
  - Explicit conversions from and to the natural numbers are
wenzelm@63977
  1162
    provided as char_of_nat, nat_of_char (as before).
wenzelm@63977
  1163
  - The auxiliary nibble type has been discontinued.
wenzelm@63977
  1164
wenzelm@63977
  1165
INCOMPATIBILITY.
wenzelm@63977
  1166
wenzelm@63977
  1167
* Type class "div" with operation "mod" renamed to type class "modulo"
wenzelm@63977
  1168
with operation "modulo", analogously to type class "divide". This
wenzelm@63977
  1169
eliminates the need to qualify any of those names in the presence of
wenzelm@63977
  1170
infix "mod" syntax. INCOMPATIBILITY.
wenzelm@63977
  1171
wenzelm@63979
  1172
* Statements and proofs of Knaster-Tarski fixpoint combinators lfp/gfp
wenzelm@63979
  1173
have been clarified. The fixpoint properties are lfp_fixpoint, its
wenzelm@63979
  1174
symmetric lfp_unfold (as before), and the duals for gfp. Auxiliary items
wenzelm@63979
  1175
for the proof (lfp_lemma2 etc.) are no longer exported, but can be
wenzelm@63979
  1176
easily recovered by composition with eq_refl. Minor INCOMPATIBILITY.
wenzelm@63979
  1177
wenzelm@63977
  1178
* Constant "surj" is a mere input abbreviation, to avoid hiding an
wenzelm@63977
  1179
equation in term output. Minor INCOMPATIBILITY.
wenzelm@63977
  1180
wenzelm@63977
  1181
* Command 'code_reflect' accepts empty constructor lists for datatypes,
wenzelm@63977
  1182
which renders those abstract effectively.
wenzelm@63977
  1183
wenzelm@63977
  1184
* Command 'export_code' checks given constants for abstraction
wenzelm@63977
  1185
violations: a small guarantee that given constants specify a safe
wenzelm@63977
  1186
interface for the generated code.
wenzelm@63977
  1187
wenzelm@63977
  1188
* Code generation for Scala: ambiguous implicts in class diagrams are
wenzelm@63977
  1189
spelt out explicitly.
wenzelm@63977
  1190
wenzelm@63977
  1191
* Static evaluators (Code_Evaluation.static_* in Isabelle/ML) rely on
wenzelm@63977
  1192
explicitly provided auxiliary definitions for required type class
wenzelm@64390
  1193
dictionaries rather than half-working magic. INCOMPATIBILITY, see the
wenzelm@64390
  1194
tutorial on code generation for details.
wenzelm@64390
  1195
wenzelm@64390
  1196
* Theory Set_Interval: substantial new theorems on indexed sums and
wenzelm@64390
  1197
products.
wenzelm@63977
  1198
wenzelm@63977
  1199
* Locale bijection establishes convenient default simp rules such as
wenzelm@63977
  1200
"inv f (f a) = a" for total bijections.
wenzelm@63977
  1201
wenzelm@63977
  1202
* Abstract locales semigroup, abel_semigroup, semilattice,
wenzelm@63977
  1203
semilattice_neutr, ordering, ordering_top, semilattice_order,
wenzelm@63977
  1204
semilattice_neutr_order, comm_monoid_set, semilattice_set,
wenzelm@63977
  1205
semilattice_neutr_set, semilattice_order_set,
wenzelm@63977
  1206
semilattice_order_neutr_set monoid_list, comm_monoid_list,
wenzelm@63977
  1207
comm_monoid_list_set, comm_monoid_mset, comm_monoid_fun use boldified
wenzelm@63977
  1208
syntax uniformly that does not clash with corresponding global syntax.
wenzelm@63977
  1209
INCOMPATIBILITY.
wenzelm@63977
  1210
wenzelm@63977
  1211
* Former locale lifting_syntax is now a bundle, which is easier to
wenzelm@63977
  1212
include in a local context or theorem statement, e.g. "context includes
wenzelm@63977
  1213
lifting_syntax begin ... end". Minor INCOMPATIBILITY.
wenzelm@63977
  1214
wenzelm@63807
  1215
* Some old / obsolete theorems have been renamed / removed, potential
wenzelm@63807
  1216
INCOMPATIBILITY.
wenzelm@63807
  1217
wenzelm@63807
  1218
  nat_less_cases  --  removed, use linorder_cases instead
wenzelm@63807
  1219
  inv_image_comp  --  removed, use image_inv_f_f instead
wenzelm@63807
  1220
  image_surj_f_inv_f  ~>  image_f_inv_f
wenzelm@63113
  1221
Mathias@63456
  1222
* Some theorems about groups and orders have been generalised from
Mathias@63456
  1223
  groups to semi-groups that are also monoids:
Mathias@63456
  1224
    le_add_same_cancel1
Mathias@63456
  1225
    le_add_same_cancel2
Mathias@63456
  1226
    less_add_same_cancel1
Mathias@63456
  1227
    less_add_same_cancel2
Mathias@63456
  1228
    add_le_same_cancel1
Mathias@63456
  1229
    add_le_same_cancel2
Mathias@63456
  1230
    add_less_same_cancel1
Mathias@63456
  1231
    add_less_same_cancel2
Mathias@63456
  1232
Mathias@63456
  1233
* Some simplifications theorems about rings have been removed, since
Mathias@63456
  1234
  superseeded by a more general version:
Mathias@63456
  1235
    less_add_cancel_left_greater_zero ~> less_add_same_cancel1
Mathias@63456
  1236
    less_add_cancel_right_greater_zero ~> less_add_same_cancel2
Mathias@63456
  1237
    less_eq_add_cancel_left_greater_eq_zero ~> le_add_same_cancel1
Mathias@63456
  1238
    less_eq_add_cancel_right_greater_eq_zero ~> le_add_same_cancel2
Mathias@63456
  1239
    less_eq_add_cancel_left_less_eq_zero ~> add_le_same_cancel1
Mathias@63456
  1240
    less_eq_add_cancel_right_less_eq_zero ~> add_le_same_cancel2
Mathias@63456
  1241
    less_add_cancel_left_less_zero ~> add_less_same_cancel1
Mathias@63456
  1242
    less_add_cancel_right_less_zero ~> add_less_same_cancel2
Mathias@63456
  1243
INCOMPATIBILITY.
Mathias@63456
  1244
wenzelm@62407
  1245
* Renamed split_if -> if_split and split_if_asm -> if_split_asm to
wenzelm@62407
  1246
resemble the f.split naming convention, INCOMPATIBILITY.
nipkow@62396
  1247
wenzelm@63977
  1248
* Added class topological_monoid.
wenzelm@63977
  1249
wenzelm@64391
  1250
* The following theorems have been renamed:
wenzelm@64391
  1251
nipkow@64457
  1252
  setsum_left_distrib ~> sum_distrib_right
nipkow@64457
  1253
  setsum_right_distrib ~> sum_distrib_left
wenzelm@64391
  1254
wenzelm@64391
  1255
INCOMPATIBILITY.
wenzelm@64391
  1256
wenzelm@64391
  1257
* Compound constants INFIMUM and SUPREMUM are mere abbreviations now.
wenzelm@64391
  1258
INCOMPATIBILITY.
wenzelm@64391
  1259
wenzelm@64391
  1260
* "Gcd (f ` A)" and "Lcm (f ` A)" are printed with optional
wenzelm@64391
  1261
comprehension-like syntax analogously to "Inf (f ` A)" and "Sup (f `
wenzelm@64391
  1262
A)".
wenzelm@64391
  1263
wenzelm@64391
  1264
* Class semiring_Lcd merged into semiring_Gcd. INCOMPATIBILITY.
wenzelm@64391
  1265
wenzelm@64391
  1266
* The type class ordered_comm_monoid_add is now called
wenzelm@64391
  1267
ordered_cancel_comm_monoid_add. A new type class ordered_comm_monoid_add
wenzelm@64391
  1268
is introduced as the combination of ordered_ab_semigroup_add +
wenzelm@64391
  1269
comm_monoid_add. INCOMPATIBILITY.
wenzelm@64391
  1270
wenzelm@64391
  1271
* Introduced the type classes canonically_ordered_comm_monoid_add and
wenzelm@64391
  1272
dioid.
wenzelm@64391
  1273
wenzelm@64391
  1274
* Introduced the type class ordered_ab_semigroup_monoid_add_imp_le. When
wenzelm@64391
  1275
instantiating linordered_semiring_strict and ordered_ab_group_add, an
wenzelm@64391
  1276
explicit instantiation of ordered_ab_semigroup_monoid_add_imp_le might
wenzelm@64391
  1277
be required. INCOMPATIBILITY.
nipkow@63117
  1278
haftmann@62348
  1279
* Dropped various legacy fact bindings, whose replacements are often
haftmann@62348
  1280
of a more general type also:
haftmann@62348
  1281
  lcm_left_commute_nat ~> lcm.left_commute
haftmann@62348
  1282
  lcm_left_commute_int ~> lcm.left_commute
haftmann@62348
  1283
  gcd_left_commute_nat ~> gcd.left_commute
haftmann@62348
  1284
  gcd_left_commute_int ~> gcd.left_commute
haftmann@62348
  1285
  gcd_greatest_iff_nat ~> gcd_greatest_iff
haftmann@62348
  1286
  gcd_greatest_iff_int ~> gcd_greatest_iff
haftmann@62348
  1287
  coprime_dvd_mult_nat ~> coprime_dvd_mult
haftmann@62348
  1288
  coprime_dvd_mult_int ~> coprime_dvd_mult
haftmann@62348
  1289
  zpower_numeral_even ~> power_numeral_even
haftmann@62348
  1290
  gcd_mult_cancel_nat ~> gcd_mult_cancel
haftmann@62348
  1291
  gcd_mult_cancel_int ~> gcd_mult_cancel
haftmann@62348
  1292
  div_gcd_coprime_nat ~> div_gcd_coprime
haftmann@62348
  1293
  div_gcd_coprime_int ~> div_gcd_coprime
haftmann@62348
  1294
  zpower_numeral_odd ~> power_numeral_odd
haftmann@62348
  1295
  zero_less_int_conv ~> of_nat_0_less_iff
haftmann@62348
  1296
  gcd_greatest_nat ~> gcd_greatest
haftmann@62348
  1297
  gcd_greatest_int ~> gcd_greatest
haftmann@62348
  1298
  coprime_mult_nat ~> coprime_mult
haftmann@62348
  1299
  coprime_mult_int ~> coprime_mult
haftmann@62348
  1300
  lcm_commute_nat ~> lcm.commute
haftmann@62348
  1301
  lcm_commute_int ~> lcm.commute
haftmann@62348
  1302
  int_less_0_conv ~> of_nat_less_0_iff
haftmann@62348
  1303
  gcd_commute_nat ~> gcd.commute
haftmann@62348
  1304
  gcd_commute_int ~> gcd.commute
haftmann@62348
  1305
  Gcd_insert_nat ~> Gcd_insert
haftmann@62348
  1306
  Gcd_insert_int ~> Gcd_insert
haftmann@62348
  1307
  of_int_int_eq ~> of_int_of_nat_eq
haftmann@62348
  1308
  lcm_least_nat ~> lcm_least
haftmann@62348
  1309
  lcm_least_int ~> lcm_least
haftmann@62348
  1310
  lcm_assoc_nat ~> lcm.assoc
haftmann@62348
  1311
  lcm_assoc_int ~> lcm.assoc
haftmann@62348
  1312
  int_le_0_conv ~> of_nat_le_0_iff
haftmann@62348
  1313
  int_eq_0_conv ~> of_nat_eq_0_iff
haftmann@62348
  1314
  Gcd_empty_nat ~> Gcd_empty
haftmann@62348
  1315
  Gcd_empty_int ~> Gcd_empty
haftmann@62348
  1316
  gcd_assoc_nat ~> gcd.assoc
haftmann@62348
  1317
  gcd_assoc_int ~> gcd.assoc
haftmann@62348
  1318
  zero_zle_int ~> of_nat_0_le_iff
haftmann@62348
  1319
  lcm_dvd2_nat ~> dvd_lcm2
haftmann@62348
  1320
  lcm_dvd2_int ~> dvd_lcm2
haftmann@62348
  1321
  lcm_dvd1_nat ~> dvd_lcm1
haftmann@62348
  1322
  lcm_dvd1_int ~> dvd_lcm1
haftmann@62348
  1323
  gcd_zero_nat ~> gcd_eq_0_iff
haftmann@62348
  1324
  gcd_zero_int ~> gcd_eq_0_iff
haftmann@62348
  1325
  gcd_dvd2_nat ~> gcd_dvd2
haftmann@62348
  1326
  gcd_dvd2_int ~> gcd_dvd2
haftmann@62348
  1327
  gcd_dvd1_nat ~> gcd_dvd1
haftmann@62348
  1328
  gcd_dvd1_int ~> gcd_dvd1
haftmann@62348
  1329
  int_numeral ~> of_nat_numeral
haftmann@62348
  1330
  lcm_ac_nat ~> ac_simps
haftmann@62348
  1331
  lcm_ac_int ~> ac_simps
haftmann@62348
  1332
  gcd_ac_nat ~> ac_simps
haftmann@62348
  1333
  gcd_ac_int ~> ac_simps
haftmann@62348
  1334
  abs_int_eq ~> abs_of_nat
haftmann@62348
  1335
  zless_int ~> of_nat_less_iff
haftmann@62348
  1336
  zdiff_int ~> of_nat_diff
haftmann@62348
  1337
  zadd_int ~> of_nat_add
haftmann@62348
  1338
  int_mult ~> of_nat_mult
haftmann@62348
  1339
  int_Suc ~> of_nat_Suc
haftmann@62348
  1340
  inj_int ~> inj_of_nat
haftmann@62348
  1341
  int_1 ~> of_nat_1
haftmann@62348
  1342
  int_0 ~> of_nat_0
haftmann@62353
  1343
  Lcm_empty_nat ~> Lcm_empty
haftmann@62353
  1344
  Lcm_empty_int ~> Lcm_empty
haftmann@62353
  1345
  Lcm_insert_nat ~> Lcm_insert
haftmann@62353
  1346
  Lcm_insert_int ~> Lcm_insert
haftmann@62353
  1347
  comp_fun_idem_gcd_nat ~> comp_fun_idem_gcd
haftmann@62353
  1348
  comp_fun_idem_gcd_int ~> comp_fun_idem_gcd
haftmann@62353
  1349
  comp_fun_idem_lcm_nat ~> comp_fun_idem_lcm
haftmann@62353
  1350
  comp_fun_idem_lcm_int ~> comp_fun_idem_lcm
haftmann@62353
  1351
  Lcm_eq_0 ~> Lcm_eq_0_I
haftmann@62353
  1352
  Lcm0_iff ~> Lcm_0_iff
haftmann@62353
  1353
  Lcm_dvd_int ~> Lcm_least
haftmann@62353
  1354
  divides_mult_nat ~> divides_mult
haftmann@62353
  1355
  divides_mult_int ~> divides_mult
haftmann@62353
  1356
  lcm_0_nat ~> lcm_0_right
haftmann@62353
  1357
  lcm_0_int ~> lcm_0_right
haftmann@62353
  1358
  lcm_0_left_nat ~> lcm_0_left
haftmann@62353
  1359
  lcm_0_left_int ~> lcm_0_left
haftmann@62353
  1360
  dvd_gcd_D1_nat ~> dvd_gcdD1
haftmann@62353
  1361
  dvd_gcd_D1_int ~> dvd_gcdD1
haftmann@62353
  1362
  dvd_gcd_D2_nat ~> dvd_gcdD2
haftmann@62353
  1363
  dvd_gcd_D2_int ~> dvd_gcdD2
haftmann@62353
  1364
  coprime_dvd_mult_iff_nat ~> coprime_dvd_mult_iff
haftmann@62353
  1365
  coprime_dvd_mult_iff_int ~> coprime_dvd_mult_iff
haftmann@62348
  1366
  realpow_minus_mult ~> power_minus_mult
haftmann@62348
  1367
  realpow_Suc_le_self ~> power_Suc_le_self
haftmann@62353
  1368
  dvd_Gcd, dvd_Gcd_nat, dvd_Gcd_int removed in favour of Gcd_greatest
haftmann@62347
  1369
INCOMPATIBILITY.
haftmann@62347
  1370
lp15@63967
  1371
* Renamed HOL/Quotient_Examples/FSet.thy to
wenzelm@63977
  1372
HOL/Quotient_Examples/Quotient_FSet.thy INCOMPATIBILITY.
wenzelm@63977
  1373
wenzelm@64390
  1374
* Session HOL-Library: theory FinFun bundles "finfun_syntax" and
wenzelm@64390
  1375
"no_finfun_syntax" allow to control optional syntax in local contexts;
wenzelm@64390
  1376
this supersedes former theory FinFun_Syntax. INCOMPATIBILITY, e.g. use
wenzelm@64390
  1377
"unbundle finfun_syntax" to imitate import of
wenzelm@64390
  1378
"~~/src/HOL/Library/FinFun_Syntax".
wenzelm@64390
  1379
wenzelm@64390
  1380
* Session HOL-Library: theory Multiset_Permutations (executably) defines
wenzelm@64390
  1381
the set of permutations of a given set or multiset, i.e. the set of all
wenzelm@64390
  1382
lists that contain every element of the carrier (multi-)set exactly
wenzelm@64390
  1383
once.
wenzelm@64390
  1384
wenzelm@64390
  1385
* Session HOL-Library: multiset membership is now expressed using
wenzelm@64390
  1386
set_mset rather than count.
wenzelm@64390
  1387
wenzelm@64390
  1388
  - Expressions "count M a > 0" and similar simplify to membership
wenzelm@64390
  1389
    by default.
wenzelm@64390
  1390
wenzelm@64390
  1391
  - Converting between "count M a = 0" and non-membership happens using
wenzelm@64390
  1392
    equations count_eq_zero_iff and not_in_iff.
wenzelm@64390
  1393
wenzelm@64390
  1394
  - Rules count_inI and in_countE obtain facts of the form
wenzelm@64390
  1395
    "count M a = n" from membership.
wenzelm@64390
  1396
wenzelm@64390
  1397
  - Rules count_in_diffI and in_diff_countE obtain facts of the form
wenzelm@64390
  1398
    "count M a = n + count N a" from membership on difference sets.
wenzelm@64390
  1399
wenzelm@64390
  1400
INCOMPATIBILITY.
wenzelm@64390
  1401
wenzelm@64390
  1402
* Session HOL-Library: theory LaTeXsugar uses new-style "dummy_pats" for
wenzelm@64390
  1403
displaying equations in functional programming style --- variables
wenzelm@64390
  1404
present on the left-hand but not on the righ-hand side are replaced by
wenzelm@64390
  1405
underscores.
wenzelm@64390
  1406
wenzelm@64390
  1407
* Session HOL-Library: theory Combinator_PER provides combinator to
wenzelm@64390
  1408
build partial equivalence relations from a predicate and an equivalence
wenzelm@64390
  1409
relation.
wenzelm@64390
  1410
wenzelm@64390
  1411
* Session HOL-Library: theory Perm provides basic facts about almost
wenzelm@64390
  1412
everywhere fix bijections.
wenzelm@64390
  1413
wenzelm@64390
  1414
* Session HOL-Library: theory Normalized_Fraction allows viewing an
wenzelm@64390
  1415
element of a field of fractions as a normalized fraction (i.e. a pair of
wenzelm@64390
  1416
numerator and denominator such that the two are coprime and the
wenzelm@64390
  1417
denominator is normalized wrt. unit factors).
wenzelm@64390
  1418
wenzelm@64390
  1419
* Session HOL-NSA has been renamed to HOL-Nonstandard_Analysis.
wenzelm@64390
  1420
wenzelm@64390
  1421
* Session HOL-Multivariate_Analysis has been renamed to HOL-Analysis.
wenzelm@64390
  1422
wenzelm@64390
  1423
* Session HOL-Analysis: measure theory has been moved here from
wenzelm@64390
  1424
HOL-Probability. When importing HOL-Analysis some theorems need
wenzelm@64390
  1425
additional name spaces prefixes due to name clashes. INCOMPATIBILITY.
wenzelm@64390
  1426
wenzelm@64390
  1427
* Session HOL-Analysis: more complex analysis including Cauchy's
wenzelm@64390
  1428
inequality, Liouville theorem, open mapping theorem, maximum modulus
wenzelm@64390
  1429
principle, Residue theorem, Schwarz Lemma.
wenzelm@64390
  1430
wenzelm@64390
  1431
* Session HOL-Analysis: Theory of polyhedra: faces, extreme points,
wenzelm@64390
  1432
polytopes, and the Krein–Milman Minkowski theorem.
wenzelm@64390
  1433
wenzelm@64390
  1434
* Session HOL-Analysis: Numerous results ported from the HOL Light
wenzelm@64390
  1435
libraries: homeomorphisms, continuous function extensions, invariance of
wenzelm@64390
  1436
domain.
wenzelm@64390
  1437
wenzelm@64390
  1438
* Session HOL-Probability: the type of emeasure and nn_integral was
wenzelm@64390
  1439
changed from ereal to ennreal, INCOMPATIBILITY.
wenzelm@64390
  1440
wenzelm@64390
  1441
  emeasure :: 'a measure \<Rightarrow> 'a set \<Rightarrow> ennreal
wenzelm@64390
  1442
  nn_integral :: 'a measure \<Rightarrow> ('a \<Rightarrow> ennreal) \<Rightarrow> ennreal
wenzelm@64390
  1443
wenzelm@64390
  1444
* Session HOL-Probability: Code generation and QuickCheck for
wenzelm@64390
  1445
Probability Mass Functions.
wenzelm@64390
  1446
wenzelm@64390
  1447
* Session HOL-Probability: theory Random_Permutations contains some
wenzelm@64390
  1448
theory about choosing a permutation of a set uniformly at random and
wenzelm@64390
  1449
folding over a list in random order.
wenzelm@64390
  1450
wenzelm@64390
  1451
* Session HOL-Probability: theory SPMF formalises discrete
wenzelm@64390
  1452
subprobability distributions.
wenzelm@64390
  1453
wenzelm@64390
  1454
* Session HOL-Library: the names of multiset theorems have been
wenzelm@64390
  1455
normalised to distinguish which ordering the theorems are about
wenzelm@64390
  1456
wenzelm@64390
  1457
    mset_less_eqI ~> mset_subset_eqI
wenzelm@64390
  1458
    mset_less_insertD ~> mset_subset_insertD
wenzelm@64390
  1459
    mset_less_eq_count ~> mset_subset_eq_count
wenzelm@64390
  1460
    mset_less_diff_self ~> mset_subset_diff_self
wenzelm@64390
  1461
    mset_le_exists_conv ~> mset_subset_eq_exists_conv
wenzelm@64390
  1462
    mset_le_mono_add_right_cancel ~> mset_subset_eq_mono_add_right_cancel
wenzelm@64390
  1463
    mset_le_mono_add_left_cancel ~> mset_subset_eq_mono_add_left_cancel
wenzelm@64390
  1464
    mset_le_mono_add ~> mset_subset_eq_mono_add
wenzelm@64390
  1465
    mset_le_add_left ~> mset_subset_eq_add_left
wenzelm@64390
  1466
    mset_le_add_right ~> mset_subset_eq_add_right
wenzelm@64390
  1467
    mset_le_single ~> mset_subset_eq_single
wenzelm@64390
  1468
    mset_le_multiset_union_diff_commute ~> mset_subset_eq_multiset_union_diff_commute
wenzelm@64390
  1469
    diff_le_self ~> diff_subset_eq_self
wenzelm@64390
  1470
    mset_leD ~> mset_subset_eqD
wenzelm@64390
  1471
    mset_lessD ~> mset_subsetD
wenzelm@64390
  1472
    mset_le_insertD ~> mset_subset_eq_insertD
wenzelm@64390
  1473
    mset_less_of_empty ~> mset_subset_of_empty
wenzelm@64390
  1474
    mset_less_size ~> mset_subset_size
wenzelm@64390
  1475
    wf_less_mset_rel ~> wf_subset_mset_rel
wenzelm@64390
  1476
    count_le_replicate_mset_le ~> count_le_replicate_mset_subset_eq
wenzelm@64390
  1477
    mset_remdups_le ~> mset_remdups_subset_eq
wenzelm@64390
  1478
    ms_lesseq_impl ~> subset_eq_mset_impl
wenzelm@64390
  1479
wenzelm@64390
  1480
Some functions have been renamed:
wenzelm@64390
  1481
    ms_lesseq_impl -> subset_eq_mset_impl
wenzelm@64390
  1482
wenzelm@64390
  1483
* HOL-Library: multisets are now ordered with the multiset ordering
wenzelm@64390
  1484
    #\<subseteq># ~> \<le>
wenzelm@64390
  1485
    #\<subset># ~> <
wenzelm@64390
  1486
    le_multiset ~> less_eq_multiset
wenzelm@64390
  1487
    less_multiset ~> le_multiset
wenzelm@64390
  1488
INCOMPATIBILITY.
wenzelm@64390
  1489
wenzelm@64390
  1490
* Session HOL-Library: the prefix multiset_order has been discontinued:
wenzelm@64390
  1491
the theorems can be directly accessed. As a consequence, the lemmas
wenzelm@64390
  1492
"order_multiset" and "linorder_multiset" have been discontinued, and the
wenzelm@64390
  1493
interpretations "multiset_linorder" and "multiset_wellorder" have been
wenzelm@64390
  1494
replaced by instantiations. INCOMPATIBILITY.
wenzelm@64390
  1495
wenzelm@64390
  1496
* Session HOL-Library: some theorems about the multiset ordering have
wenzelm@64390
  1497
been renamed:
wenzelm@64390
  1498
wenzelm@64390
  1499
    le_multiset_def ~> less_eq_multiset_def
wenzelm@64390
  1500
    less_multiset_def ~> le_multiset_def
wenzelm@64390
  1501
    less_eq_imp_le_multiset ~> subset_eq_imp_le_multiset
wenzelm@64390
  1502
    mult_less_not_refl ~> mset_le_not_refl
wenzelm@64390
  1503
    mult_less_trans ~> mset_le_trans
wenzelm@64390
  1504
    mult_less_not_sym ~> mset_le_not_sym
wenzelm@64390
  1505
    mult_less_asym ~> mset_le_asym
wenzelm@64390
  1506
    mult_less_irrefl ~> mset_le_irrefl
wenzelm@64390
  1507
    union_less_mono2{,1,2} ~> union_le_mono2{,1,2}
wenzelm@64390
  1508
wenzelm@64390
  1509
    le_multiset\<^sub>H\<^sub>O ~> less_eq_multiset\<^sub>H\<^sub>O
wenzelm@64390
  1510
    le_multiset_total ~> less_eq_multiset_total
wenzelm@64390
  1511
    less_multiset_right_total ~> subset_eq_imp_le_multiset
wenzelm@64390
  1512
    le_multiset_empty_left ~> less_eq_multiset_empty_left
wenzelm@64390
  1513
    le_multiset_empty_right ~> less_eq_multiset_empty_right
wenzelm@64390
  1514
    less_multiset_empty_right ~> le_multiset_empty_left
wenzelm@64390
  1515
    less_multiset_empty_left ~> le_multiset_empty_right
wenzelm@64390
  1516
    union_less_diff_plus ~> union_le_diff_plus
wenzelm@64390
  1517
    ex_gt_count_imp_less_multiset ~> ex_gt_count_imp_le_multiset
wenzelm@64390
  1518
    less_multiset_plus_left_nonempty ~> le_multiset_plus_left_nonempty
wenzelm@64390
  1519
    le_multiset_plus_right_nonempty ~> le_multiset_plus_right_nonempty
wenzelm@64390
  1520
INCOMPATIBILITY.
wenzelm@64390
  1521
wenzelm@64390
  1522
* Session HOL-Library: the lemma mset_map has now the attribute [simp].
wenzelm@64390
  1523
INCOMPATIBILITY.
wenzelm@64390
  1524
wenzelm@64390
  1525
* Session HOL-Library: some theorems about multisets have been removed.
wenzelm@64390
  1526
INCOMPATIBILITY, use the following replacements:
wenzelm@64390
  1527
wenzelm@64390
  1528
    le_multiset_plus_plus_left_iff ~> add_less_cancel_right
wenzelm@64390
  1529
    less_multiset_plus_plus_left_iff ~> add_less_cancel_right
wenzelm@64390
  1530
    le_multiset_plus_plus_right_iff ~> add_less_cancel_left
wenzelm@64390
  1531
    less_multiset_plus_plus_right_iff ~> add_less_cancel_left
wenzelm@64390
  1532
    add_eq_self_empty_iff ~> add_cancel_left_right
wenzelm@64390
  1533
    mset_subset_add_bothsides ~> subset_mset.add_less_cancel_right
wenzelm@64390
  1534
    mset_less_add_bothsides ~> subset_mset.add_less_cancel_right
wenzelm@64390
  1535
    mset_le_add_bothsides ~> subset_mset.add_less_cancel_right
wenzelm@64390
  1536
    empty_inter ~> subset_mset.inf_bot_left
wenzelm@64390
  1537
    inter_empty ~> subset_mset.inf_bot_right
wenzelm@64390
  1538
    empty_sup ~> subset_mset.sup_bot_left
wenzelm@64390
  1539
    sup_empty ~> subset_mset.sup_bot_right
wenzelm@64390
  1540
    bdd_below_multiset ~> subset_mset.bdd_above_bot
wenzelm@64390
  1541
    subset_eq_empty ~> subset_mset.le_zero_eq
wenzelm@64390
  1542
    le_empty ~> subset_mset.le_zero_eq
wenzelm@64390
  1543
    mset_subset_empty_nonempty ~> subset_mset.zero_less_iff_neq_zero
wenzelm@64390
  1544
    mset_less_empty_nonempty ~> subset_mset.zero_less_iff_neq_zero
wenzelm@64390
  1545
wenzelm@64390
  1546
* Session HOL-Library: some typeclass constraints about multisets have
wenzelm@64390
  1547
been reduced from ordered or linordered to preorder. Multisets have the
wenzelm@64390
  1548
additional typeclasses order_bot, no_top,
wenzelm@64390
  1549
ordered_ab_semigroup_add_imp_le, ordered_cancel_comm_monoid_add,
wenzelm@64390
  1550
linordered_cancel_ab_semigroup_add, and
wenzelm@64390
  1551
ordered_ab_semigroup_monoid_add_imp_le. INCOMPATIBILITY.
wenzelm@64390
  1552
wenzelm@64390
  1553
* Session HOL-Library: there are some new simplification rules about
wenzelm@64390
  1554
multisets, the multiset ordering, and the subset ordering on multisets.
wenzelm@64390
  1555
INCOMPATIBILITY.
wenzelm@64390
  1556
wenzelm@64390
  1557
* Session HOL-Library: the subset ordering on multisets has now the
wenzelm@64390
  1558
interpretations ordered_ab_semigroup_monoid_add_imp_le and
wenzelm@64390
  1559
bounded_lattice_bot. INCOMPATIBILITY.
wenzelm@64390
  1560
wenzelm@64390
  1561
* Session HOL-Library, theory Multiset: single has been removed in favor
wenzelm@64390
  1562
of add_mset that roughly corresponds to Set.insert. Some theorems have
wenzelm@64390
  1563
removed or changed:
wenzelm@64390
  1564
wenzelm@64390
  1565
  single_not_empty ~> add_mset_not_empty or empty_not_add_mset
wenzelm@64390
  1566
  fold_mset_insert ~> fold_mset_add_mset
wenzelm@64390
  1567
  image_mset_insert ~> image_mset_add_mset
wenzelm@64390
  1568
  union_single_eq_diff
wenzelm@64390
  1569
  multi_self_add_other_not_self
wenzelm@64390
  1570
  diff_single_eq_union
wenzelm@64390
  1571
INCOMPATIBILITY.
wenzelm@64390
  1572
wenzelm@64390
  1573
* Session HOL-Library, theory Multiset: some theorems have been changed
wenzelm@64390
  1574
to use add_mset instead of single:
wenzelm@64390
  1575
wenzelm@64390
  1576
  mset_add
wenzelm@64390
  1577
  multi_self_add_other_not_self
wenzelm@64390
  1578
  diff_single_eq_union
wenzelm@64390
  1579
  union_single_eq_diff
wenzelm@64390
  1580
  union_single_eq_member
wenzelm@64390
  1581
  add_eq_conv_diff
wenzelm@64390
  1582
  insert_noteq_member
wenzelm@64390
  1583
  add_eq_conv_ex
wenzelm@64390
  1584
  multi_member_split
wenzelm@64390
  1585
  multiset_add_sub_el_shuffle
wenzelm@64390
  1586
  mset_subset_eq_insertD
wenzelm@64390
  1587
  mset_subset_insertD
wenzelm@64390
  1588
  insert_subset_eq_iff
wenzelm@64390
  1589
  insert_union_subset_iff
wenzelm@64390
  1590
  multi_psub_of_add_self
wenzelm@64390
  1591
  inter_add_left1
wenzelm@64390
  1592
  inter_add_left2
wenzelm@64390
  1593
  inter_add_right1
wenzelm@64390
  1594
  inter_add_right2
wenzelm@64390
  1595
  sup_union_left1
wenzelm@64390
  1596
  sup_union_left2
wenzelm@64390
  1597
  sup_union_right1
wenzelm@64390
  1598
  sup_union_right2
wenzelm@64390
  1599
  size_eq_Suc_imp_eq_union
wenzelm@64390
  1600
  multi_nonempty_split
wenzelm@64390
  1601
  mset_insort
wenzelm@64390
  1602
  mset_update
wenzelm@64390
  1603
  mult1I
wenzelm@64390
  1604
  less_add
wenzelm@64390
  1605
  mset_zip_take_Cons_drop_twice
wenzelm@64390
  1606
  rel_mset_Zero
wenzelm@64390
  1607
  msed_map_invL
wenzelm@64390
  1608
  msed_map_invR
wenzelm@64390
  1609
  msed_rel_invL
wenzelm@64390
  1610
  msed_rel_invR
wenzelm@64390
  1611
  le_multiset_right_total
wenzelm@64390
  1612
  multiset_induct
wenzelm@64390
  1613
  multiset_induct2_size
wenzelm@64390
  1614
  multiset_induct2
wenzelm@64390
  1615
INCOMPATIBILITY.
wenzelm@64390
  1616
wenzelm@64390
  1617
* Session HOL-Library, theory Multiset: the definitions of some
wenzelm@64390
  1618
constants have changed to use add_mset instead of adding a single
wenzelm@64390
  1619
element:
wenzelm@64390
  1620
wenzelm@64390
  1621
  image_mset
wenzelm@64390
  1622
  mset
wenzelm@64390
  1623
  replicate_mset
wenzelm@64390
  1624
  mult1
wenzelm@64390
  1625
  pred_mset
wenzelm@64390
  1626
  rel_mset'
wenzelm@64390
  1627
  mset_insort
wenzelm@64390
  1628
wenzelm@64390
  1629
INCOMPATIBILITY.
wenzelm@64390
  1630
wenzelm@64390
  1631
* Session HOL-Library, theory Multiset: due to the above changes, the
wenzelm@64390
  1632
attributes of some multiset theorems have been changed:
wenzelm@64390
  1633
wenzelm@64390
  1634
  insert_DiffM  [] ~> [simp]
wenzelm@64390
  1635
  insert_DiffM2 [simp] ~> []
wenzelm@64390
  1636
  diff_add_mset_swap [simp]
wenzelm@64390
  1637
  fold_mset_add_mset [simp]
wenzelm@64390
  1638
  diff_diff_add [simp] (for multisets only)
wenzelm@64390
  1639
  diff_cancel [simp] ~> []
wenzelm@64390
  1640
  count_single [simp] ~> []
wenzelm@64390
  1641
  set_mset_single [simp] ~> []
wenzelm@64390
  1642
  size_multiset_single [simp] ~> []
wenzelm@64390
  1643
  size_single [simp] ~> []
wenzelm@64390
  1644
  image_mset_single [simp] ~> []
wenzelm@64390
  1645
  mset_subset_eq_mono_add_right_cancel [simp] ~> []
wenzelm@64390
  1646
  mset_subset_eq_mono_add_left_cancel [simp] ~> []
wenzelm@64390
  1647
  fold_mset_single [simp] ~> []
wenzelm@64390
  1648
  subset_eq_empty [simp] ~> []
wenzelm@64390
  1649
  empty_sup [simp] ~> []
wenzelm@64390
  1650
  sup_empty [simp] ~> []
wenzelm@64390
  1651
  inter_empty [simp] ~> []
wenzelm@64390
  1652
  empty_inter [simp] ~> []
wenzelm@64390
  1653
INCOMPATIBILITY.
wenzelm@64390
  1654
wenzelm@64391
  1655
* Session HOL-Library, theory Multiset: the order of the variables in
wenzelm@64390
  1656
the second cases of multiset_induct, multiset_induct2_size,
wenzelm@64390
  1657
multiset_induct2 has been changed (e.g. Add A a ~> Add a A).
wenzelm@64390
  1658
INCOMPATIBILITY.
wenzelm@64390
  1659
wenzelm@64390
  1660
* Session HOL-Library, theory Multiset: there is now a simplification
wenzelm@64390
  1661
procedure on multisets. It mimics the behavior of the procedure on
wenzelm@64390
  1662
natural numbers. INCOMPATIBILITY.
wenzelm@64390
  1663
wenzelm@64390
  1664
* Session HOL-Library, theory Multiset: renamed sums and products of
wenzelm@64390
  1665
multisets:
wenzelm@64390
  1666
wenzelm@64390
  1667
  msetsum ~> sum_mset
wenzelm@64390
  1668
  msetprod ~> prod_mset
wenzelm@64390
  1669
wenzelm@64390
  1670
* Session HOL-Library, theory Multiset: the notation for intersection
wenzelm@64390
  1671
and union of multisets have been changed:
wenzelm@64390
  1672
wenzelm@64390
  1673
  #\<inter> ~> \<inter>#
wenzelm@64390
  1674
  #\<union> ~> \<union>#
wenzelm@64390
  1675
wenzelm@64390
  1676
INCOMPATIBILITY.
wenzelm@64390
  1677
wenzelm@64390
  1678
* Session HOL-Library, theory Multiset: the lemma
wenzelm@64390
  1679
one_step_implies_mult_aux on multisets has been removed, use
wenzelm@64390
  1680
one_step_implies_mult instead. INCOMPATIBILITY.
wenzelm@64390
  1681
wenzelm@64390
  1682
* Session HOL-Library: theory Complete_Partial_Order2 provides reasoning
wenzelm@64390
  1683
support for monotonicity and continuity in chain-complete partial orders
wenzelm@64390
  1684
and about admissibility conditions for fixpoint inductions.
wenzelm@64390
  1685
haftmann@64523
  1686
* Session HOL-Library: theory Library/Polynomial contains also
haftmann@64523
  1687
derivation of polynomials (formerly in Library/Poly_Deriv) but not
haftmann@64523
  1688
gcd/lcm on polynomials over fields. This has been moved to a separate
haftmann@64523
  1689
theory Library/Polynomial_GCD_euclidean.thy, to pave way for a possible
haftmann@64523
  1690
future different type class instantiation for polynomials over factorial
haftmann@64523
  1691
rings. INCOMPATIBILITY.
wenzelm@64390
  1692
wenzelm@64390
  1693
* Session HOL-Library: theory Sublist provides function "prefixes" with
wenzelm@64390
  1694
the following renaming
wenzelm@64390
  1695
wenzelm@64390
  1696
  prefixeq -> prefix
wenzelm@64390
  1697
  prefix -> strict_prefix
wenzelm@64390
  1698
  suffixeq -> suffix
wenzelm@64390
  1699
  suffix -> strict_suffix
wenzelm@64390
  1700
wenzelm@64390
  1701
Added theory of longest common prefixes.
blanchet@64389
  1702
wenzelm@64391
  1703
* Session HOL-Number_Theory: algebraic foundation for primes:
wenzelm@64391
  1704
Generalisation of predicate "prime" and introduction of predicates
wenzelm@64391
  1705
"prime_elem", "irreducible", a "prime_factorization" function, and the
wenzelm@64391
  1706
"factorial_ring" typeclass with instance proofs for nat, int, poly. Some
wenzelm@64391
  1707
theorems now have different names, most notably "prime_def" is now
wenzelm@64391
  1708
"prime_nat_iff". INCOMPATIBILITY.
wenzelm@64391
  1709
wenzelm@64391
  1710
* Session Old_Number_Theory has been removed, after porting remaining
wenzelm@64391
  1711
theories.
wenzelm@64391
  1712
wenzelm@64551
  1713
* Session HOL-Types_To_Sets provides an experimental extension of
wenzelm@64551
  1714
Higher-Order Logic to allow translation of types to sets.
wenzelm@64551
  1715
wenzelm@63198
  1716
wenzelm@62498
  1717
*** ML ***
wenzelm@62498
  1718
wenzelm@63227
  1719
* Integer.gcd and Integer.lcm use efficient operations from the Poly/ML
wenzelm@63228
  1720
library (notably for big integers). Subtle change of semantics:
wenzelm@63228
  1721
Integer.gcd and Integer.lcm both normalize the sign, results are never
wenzelm@63228
  1722
negative. This coincides with the definitions in HOL/GCD.thy.
wenzelm@63228
  1723
INCOMPATIBILITY.
wenzelm@63227
  1724
wenzelm@63212
  1725
* Structure Rat for rational numbers is now an integral part of
wenzelm@63215
  1726
Isabelle/ML, with special notation @int/nat or @int for numerals (an
wenzelm@63215
  1727
abbreviation for antiquotation @{Pure.rat argument}) and ML pretty
wenzelm@63212
  1728
printing. Standard operations on type Rat.rat are provided via ad-hoc
wenzelm@63215
  1729
overloading of + - * / < <= > >= ~ abs. INCOMPATIBILITY, need to
wenzelm@63212
  1730
use + instead of +/ etc. Moreover, exception Rat.DIVZERO has been
wenzelm@63212
  1731
superseded by General.Div.
wenzelm@63198
  1732
wenzelm@64390
  1733
* ML antiquotation @{path} is superseded by @{file}, which ensures that
wenzelm@64390
  1734
the argument is a plain file. Minor INCOMPATIBILITY.
wenzelm@64390
  1735
wenzelm@64390
  1736
* Antiquotation @{make_string} is available during Pure bootstrap --
wenzelm@64390
  1737
with approximative output quality.
wenzelm@64390
  1738
wenzelm@64390
  1739
* Low-level ML system structures (like PolyML and RunCall) are no longer
wenzelm@64390
  1740
exposed to Isabelle/ML user-space. Potential INCOMPATIBILITY.
wenzelm@64390
  1741
wenzelm@62861
  1742
* The ML function "ML" provides easy access to run-time compilation.
wenzelm@62861
  1743
This is particularly useful for conditional compilation, without
wenzelm@62861
  1744
requiring separate files.
wenzelm@62861
  1745
wenzelm@62498
  1746
* Option ML_exception_debugger controls detailed exception trace via the
wenzelm@62498
  1747
Poly/ML debugger. Relevant ML modules need to be compiled beforehand
wenzelm@62498
  1748
with ML_file_debug, or with ML_file and option ML_debugger enabled. Note
wenzelm@62498
  1749
debugger information requires consirable time and space: main
wenzelm@62498
  1750
Isabelle/HOL with full debugger support may need ML_system_64.
wenzelm@62498
  1751
wenzelm@62514
  1752
* Local_Theory.restore has been renamed to Local_Theory.reset to
wenzelm@62514
  1753
emphasize its disruptive impact on the cumulative context, notably the
wenzelm@62514
  1754
scope of 'private' or 'qualified' names. Note that Local_Theory.reset is
wenzelm@62514
  1755
only appropriate when targets are managed, e.g. starting from a global
wenzelm@62514
  1756
theory and returning to it. Regular definitional packages should use
wenzelm@62514
  1757
balanced blocks of Local_Theory.open_target versus
wenzelm@62514
  1758
Local_Theory.close_target instead. Rare INCOMPATIBILITY.
wenzelm@62514
  1759
wenzelm@62519
  1760
* Structure TimeLimit (originally from the SML/NJ library) has been
wenzelm@62519
  1761
replaced by structure Timeout, with slightly different signature.
wenzelm@62519
  1762
INCOMPATIBILITY.
wenzelm@62519
  1763
wenzelm@62551
  1764
* Discontinued cd and pwd operations, which are not well-defined in a
wenzelm@62551
  1765
multi-threaded environment. Note that files are usually located
wenzelm@62551
  1766
relatively to the master directory of a theory (see also
wenzelm@62551
  1767
File.full_path). Potential INCOMPATIBILITY.
wenzelm@62551
  1768
wenzelm@63352
  1769
* Binding.empty_atts supersedes Thm.empty_binding and
wenzelm@63352
  1770
Attrib.empty_binding. Minor INCOMPATIBILITY.
wenzelm@63352
  1771
wenzelm@62498
  1772
wenzelm@62354
  1773
*** System ***
wenzelm@62354
  1774
wenzelm@64390
  1775
* SML/NJ and old versions of Poly/ML are no longer supported.
wenzelm@64390
  1776
wenzelm@64390
  1777
* Poly/ML heaps now follow the hierarchy of sessions, and thus require
wenzelm@64390
  1778
much less disk space.
wenzelm@63226
  1779
wenzelm@62591
  1780
* The Isabelle ML process is now managed directly by Isabelle/Scala, and
wenzelm@62591
  1781
shell scripts merely provide optional command-line access. In
wenzelm@62591
  1782
particular:
wenzelm@62591
  1783
wenzelm@62591
  1784
  . Scala module ML_Process to connect to the raw ML process,
wenzelm@62591
  1785
    with interaction via stdin/stdout/stderr or in batch mode;
wenzelm@62591
  1786
  . command-line tool "isabelle console" as interactive wrapper;
wenzelm@62591
  1787
  . command-line tool "isabelle process" as batch mode wrapper.
wenzelm@62588
  1788
wenzelm@62588
  1789
* The executable "isabelle_process" has been discontinued. Tools and
wenzelm@62588
  1790
prover front-ends should use ML_Process or Isabelle_Process in
wenzelm@62591
  1791
Isabelle/Scala. INCOMPATIBILITY.
wenzelm@62588
  1792
wenzelm@62588
  1793
* New command-line tool "isabelle process" supports ML evaluation of
wenzelm@62588
  1794
literal expressions (option -e) or files (option -f) in the context of a
wenzelm@62588
  1795
given heap image. Errors lead to premature exit of the ML process with
wenzelm@62588
  1796
return code 1.
wenzelm@62588
  1797
wenzelm@64390
  1798
* The command-line tool "isabelle build" supports option -N for cyclic
wenzelm@64390
  1799
shuffling of NUMA CPU nodes. This may help performance tuning on Linux
wenzelm@64390
  1800
servers with separate CPU/memory modules.
wenzelm@64390
  1801
wenzelm@64390
  1802
* System option "threads" (for the size of the Isabelle/ML thread farm)
wenzelm@64390
  1803
is also passed to the underlying ML runtime system as --gcthreads,
wenzelm@64274
  1804
unless there is already a default provided via ML_OPTIONS settings.
wenzelm@64274
  1805
wenzelm@63827
  1806
* System option "checkpoint" helps to fine-tune the global heap space
wenzelm@63827
  1807
management of isabelle build. This is relevant for big sessions that may
wenzelm@63827
  1808
exhaust the small 32-bit address space of the ML process (which is used
wenzelm@63827
  1809
by default).
wenzelm@63827
  1810
wenzelm@64308
  1811
* System option "profiling" specifies the mode for global ML profiling
wenzelm@64342
  1812
in "isabelle build". Possible values are "time", "allocations". The
wenzelm@64342
  1813
command-line tool "isabelle profiling_report" helps to digest the
wenzelm@64342
  1814
resulting log files.
wenzelm@64308
  1815
wenzelm@63986
  1816
* System option "ML_process_policy" specifies an optional command prefix
wenzelm@63986
  1817
for the underlying ML process, e.g. to control CPU affinity on
wenzelm@63987
  1818
multiprocessor systems. The "isabelle jedit" tool allows to override the
wenzelm@63987
  1819
implicit default via option -p.
wenzelm@63986
  1820
wenzelm@64390
  1821
* Command-line tool "isabelle console" provides option -r to help to
wenzelm@64390
  1822
bootstrapping Isabelle/Pure interactively.
wenzelm@64390
  1823
wenzelm@64390
  1824
* Command-line tool "isabelle yxml" has been discontinued.
wenzelm@64390
  1825
INCOMPATIBILITY, use operations from the modules "XML" and "YXML" in
wenzelm@64390
  1826
Isabelle/ML or Isabelle/Scala.
wenzelm@64390
  1827
wenzelm@64390
  1828
* Many Isabelle tools that require a Java runtime system refer to the
wenzelm@64390
  1829
settings ISABELLE_TOOL_JAVA_OPTIONS32 / ISABELLE_TOOL_JAVA_OPTIONS64,
wenzelm@64390
  1830
depending on the underlying platform. The settings for "isabelle build"
wenzelm@64390
  1831
ISABELLE_BUILD_JAVA_OPTIONS32 / ISABELLE_BUILD_JAVA_OPTIONS64 have been
wenzelm@64390
  1832
discontinued. Potential INCOMPATIBILITY.
wenzelm@64390
  1833
wenzelm@64390
  1834
* The Isabelle system environment always ensures that the main
wenzelm@64390
  1835
executables are found within the shell search $PATH: "isabelle" and
wenzelm@64390
  1836
"isabelle_scala_script".
wenzelm@64390
  1837
wenzelm@64390
  1838
* Isabelle tools may consist of .scala files: the Scala compiler is
wenzelm@64390
  1839
invoked on the spot. The source needs to define some object that extends
wenzelm@64390
  1840
Isabelle_Tool.Body.
wenzelm@64390
  1841
wenzelm@64390
  1842
* File.bash_string, File.bash_path etc. represent Isabelle/ML and
wenzelm@64390
  1843
Isabelle/Scala strings authentically within GNU bash. This is useful to
wenzelm@64390
  1844
produce robust shell scripts under program control, without worrying
wenzelm@64390
  1845
about spaces or special characters. Note that user output works via
wenzelm@64390
  1846
Path.print (ML) or Path.toString (Scala). INCOMPATIBILITY, the old (and
wenzelm@64390
  1847
less versatile) operations File.shell_quote, File.shell_path etc. have
wenzelm@64390
  1848
been discontinued.
wenzelm@64390
  1849
wenzelm@63995
  1850
* The isabelle_java executable allows to run a Java process within the
wenzelm@63995
  1851
name space of Java and Scala components that are bundled with Isabelle,
wenzelm@63995
  1852
but without the Isabelle settings environment.
wenzelm@63995
  1853
wenzelm@64390
  1854
* Isabelle/Scala: the SSH module supports ssh and sftp connections, for
wenzelm@64390
  1855
remote command-execution and file-system access. This resembles
wenzelm@64390
  1856
operations from module File and Isabelle_System to some extent. Note
wenzelm@64390
  1857
that Path specifications need to be resolved remotely via
wenzelm@64390
  1858
ssh.remote_path instead of File.standard_path: the implicit process
wenzelm@64390
  1859
environment is different, Isabelle settings are not available remotely.
wenzelm@64390
  1860
wenzelm@64390
  1861
* Isabelle/Scala: the Mercurial module supports repositories via the
wenzelm@64390
  1862
regular hg command-line interface. The repositroy clone and working
wenzelm@64390
  1863
directory may reside on a local or remote file-system (via ssh
wenzelm@64390
  1864
connection).
wenzelm@64265
  1865
wenzelm@62354
  1866
wenzelm@62354
  1867
wenzelm@62031
  1868
New in Isabelle2016 (February 2016)
wenzelm@62016
  1869
-----------------------------------
wenzelm@60138
  1870
wenzelm@61337
  1871
*** General ***
wenzelm@61337
  1872
wenzelm@62168
  1873
* Eisbach is now based on Pure instead of HOL. Objects-logics may import
wenzelm@62168
  1874
either the theory ~~/src/HOL/Eisbach/Eisbach (for HOL etc.) or
wenzelm@62168
  1875
~~/src/HOL/Eisbach/Eisbach_Old_Appl_Syntax (for FOL, ZF etc.). Note that
wenzelm@62168
  1876
the HOL-Eisbach session located in ~~/src/HOL/Eisbach/ contains further
wenzelm@62168
  1877
examples that do require HOL.
wenzelm@62168
  1878
wenzelm@62157
  1879
* Better resource usage on all platforms (Linux, Windows, Mac OS X) for
wenzelm@62157
  1880
both Isabelle/ML and Isabelle/Scala.  Slightly reduced heap space usage.
wenzelm@62157
  1881
wenzelm@62017
  1882
* Former "xsymbols" syntax with Isabelle symbols is used by default,
wenzelm@62017
  1883
without any special print mode. Important ASCII replacement syntax
wenzelm@62017
  1884
remains available under print mode "ASCII", but less important syntax
wenzelm@62017
  1885
has been removed (see below).
wenzelm@62017
  1886
wenzelm@62109
  1887
* Support for more arrow symbols, with rendering in LaTeX and Isabelle
wenzelm@62109
  1888
fonts: \<Lleftarrow> \<Rrightarrow> \<longlongleftarrow> \<longlongrightarrow> \<longlonglongleftarrow> \<longlonglongrightarrow>.
wenzelm@62017
  1889
wenzelm@62108
  1890
* Special notation \<struct> for the first implicit 'structure' in the
wenzelm@62108
  1891
context has been discontinued. Rare INCOMPATIBILITY, use explicit
wenzelm@62108
  1892
structure name instead, notably in indexed notation with block-subscript
wenzelm@62108
  1893
(e.g. \<odot>\<^bsub>A\<^esub>).
wenzelm@62108
  1894
wenzelm@62108
  1895
* The glyph for \<diamond> in the IsabelleText font now corresponds better to its
wenzelm@62108
  1896
counterpart \<box> as quantifier-like symbol. A small diamond is available as
wenzelm@62108
  1897
\<diamondop>; the old symbol \<struct> loses this rendering and any special
wenzelm@62108
  1898
meaning.
wenzelm@62108
  1899
wenzelm@62017
  1900
* Syntax for formal comments "-- text" now also supports the symbolic
wenzelm@62017
  1901
form "\<comment> text". Command-line tool "isabelle update_cartouches -c" helps
wenzelm@62017
  1902
to update old sources.
wenzelm@62017
  1903
wenzelm@61337
  1904
* Toplevel theorem statements have been simplified as follows:
wenzelm@61337
  1905
wenzelm@61337
  1906
  theorems             ~>  lemmas
wenzelm@61337
  1907
  schematic_lemma      ~>  schematic_goal
wenzelm@61337
  1908
  schematic_theorem    ~>  schematic_goal
wenzelm@61337
  1909
  schematic_corollary  ~>  schematic_goal
wenzelm@61337
  1910
wenzelm@61337
  1911
Command-line tool "isabelle update_theorems" updates theory sources
wenzelm@61337
  1912
accordingly.
wenzelm@61337
  1913
wenzelm@61338
  1914
* Toplevel theorem statement 'proposition' is another alias for
wenzelm@61338
  1915
'theorem'.
wenzelm@61338
  1916
wenzelm@62169
  1917
* The old 'defs' command has been removed (legacy since Isabelle2014).
wenzelm@62169
  1918
INCOMPATIBILITY, use regular 'definition' instead. Overloaded and/or
wenzelm@62169
  1919
deferred definitions require a surrounding 'overloading' block.
wenzelm@62169
  1920
wenzelm@61337
  1921
wenzelm@60610
  1922
*** Prover IDE -- Isabelle/Scala/jEdit ***
wenzelm@60610
  1923
wenzelm@60986
  1924
* IDE support for the source-level debugger of Poly/ML, to work with
wenzelm@62253
  1925
Isabelle/ML and official Standard ML. Option "ML_debugger" and commands
wenzelm@62253
  1926
'ML_file_debug', 'ML_file_no_debug', 'SML_file_debug',
wenzelm@62253
  1927
'SML_file_no_debug' control compilation of sources with or without
wenzelm@62253
  1928
debugging information. The Debugger panel allows to set breakpoints (via
wenzelm@62253
  1929
context menu), step through stopped threads, evaluate local ML
wenzelm@62253
  1930
expressions etc. At least one Debugger view needs to be active to have
wenzelm@62253
  1931
any effect on the running ML program.
wenzelm@60984
  1932
wenzelm@61803
  1933
* The State panel manages explicit proof state output, with dynamic
wenzelm@61803
  1934
auto-update according to cursor movement. Alternatively, the jEdit
wenzelm@61803
  1935
action "isabelle.update-state" (shortcut S+ENTER) triggers manual
wenzelm@61803
  1936
update.
wenzelm@61729
  1937
wenzelm@61729
  1938
* The Output panel no longer shows proof state output by default, to
wenzelm@61729
  1939
avoid GUI overcrowding. INCOMPATIBILITY, use the State panel instead or
wenzelm@61729
  1940
enable option "editor_output_state".
wenzelm@61215
  1941
wenzelm@61803
  1942
* The text overview column (status of errors, warnings etc.) is updated
wenzelm@61803
  1943
asynchronously, leading to much better editor reactivity. Moreover, the
wenzelm@61803
  1944
full document node content is taken into account. The width of the
wenzelm@61803
  1945
column is scaled according to the main text area font, for improved
wenzelm@61803
  1946
visibility.
wenzelm@61803
  1947
wenzelm@61803
  1948
* The main text area no longer changes its color hue in outdated
wenzelm@61803
  1949
situations. The text overview column takes over the role to indicate
wenzelm@61803
  1950
unfinished edits in the PIDE pipeline. This avoids flashing text display
wenzelm@61803
  1951
due to ad-hoc updates by auxiliary GUI components, such as the State
wenzelm@61803
  1952
panel.
wenzelm@61803
  1953
wenzelm@62254
  1954
* Slightly improved scheduling for urgent print tasks (e.g. command
wenzelm@62254
  1955
state output, interactive queries) wrt. long-running background tasks.
wenzelm@62017
  1956
wenzelm@62017
  1957
* Completion of symbols via prefix of \<name> or \<^name> or \name is
wenzelm@62017
  1958
always possible, independently of the language context. It is never
wenzelm@62017
  1959
implicit: a popup will show up unconditionally.
wenzelm@62017
  1960
wenzelm@62017
  1961
* Additional abbreviations for syntactic completion may be specified in
wenzelm@62017
  1962
$ISABELLE_HOME/etc/abbrevs and $ISABELLE_HOME_USER/etc/abbrevs, with
wenzelm@62017
  1963
support for simple templates using ASCII 007 (bell) as placeholder.
wenzelm@62017
  1964
wenzelm@62234
  1965
* Symbols \<oplus>, \<Oplus>, \<otimes>, \<Otimes>, \<odot>, \<Odot>, \<ominus>, \<oslash> no longer provide abbreviations for
wenzelm@62234
  1966
completion like "+o", "*o", ".o" etc. -- due to conflicts with other
wenzelm@62234
  1967
ASCII syntax. INCOMPATIBILITY, use plain backslash-completion or define
wenzelm@62234
  1968
suitable abbreviations in $ISABELLE_HOME_USER/etc/abbrevs.
wenzelm@62234
  1969
wenzelm@61483
  1970
* Action "isabelle-emph" (with keyboard shortcut C+e LEFT) controls
wenzelm@61483
  1971
emphasized text style; the effect is visible in document output, not in
wenzelm@61483
  1972
the editor.
wenzelm@61483
  1973
wenzelm@61483
  1974
* Action "isabelle-reset" now uses keyboard shortcut C+e BACK_SPACE,
wenzelm@61483
  1975
instead of former C+e LEFT.
wenzelm@61483
  1976
wenzelm@61512
  1977
* The command-line tool "isabelle jedit" and the isabelle.Main
wenzelm@62027
  1978
application wrapper treat the default $USER_HOME/Scratch.thy more
wenzelm@61512
  1979
uniformly, and allow the dummy file argument ":" to open an empty buffer
wenzelm@61512
  1980
instead.
wenzelm@61512
  1981
wenzelm@62017
  1982
* New command-line tool "isabelle jedit_client" allows to connect to an
wenzelm@62017
  1983
already running Isabelle/jEdit process. This achieves the effect of
wenzelm@62017
  1984
single-instance applications seen on common GUI desktops.
wenzelm@62017
  1985
wenzelm@61529
  1986
* The default look-and-feel for Linux is the traditional "Metal", which
wenzelm@61529
  1987
works better with GUI scaling for very high-resolution displays (e.g.
wenzelm@61529
  1988
4K). Moreover, it is generally more robust than "Nimbus".
wenzelm@61529
  1989
wenzelm@62163
  1990
* Update to jedit-5.3.0, with improved GUI scaling and support of
wenzelm@62163
  1991
high-resolution displays (e.g. 4K).
wenzelm@62163
  1992
wenzelm@62034
  1993
* The main Isabelle executable is managed as single-instance Desktop
wenzelm@62034
  1994
application uniformly on all platforms: Linux, Windows, Mac OS X.
wenzelm@62034
  1995
wenzelm@60610
  1996
wenzelm@61405
  1997
*** Document preparation ***
wenzelm@61405
  1998
wenzelm@62017
  1999
* Commands 'paragraph' and 'subparagraph' provide additional section
wenzelm@62017
  2000
headings. Thus there are 6 levels of standard headings, as in HTML.
wenzelm@62017
  2001
wenzelm@62017
  2002
* Command 'text_raw' has been clarified: input text is processed as in
wenzelm@62017
  2003
'text' (with antiquotations and control symbols). The key difference is
wenzelm@62017
  2004
the lack of the surrounding isabelle markup environment in output.
wenzelm@62017
  2005
wenzelm@62017
  2006
* Text is structured in paragraphs and nested lists, using notation that
wenzelm@62017
  2007
is similar to Markdown. The control symbols for list items are as
wenzelm@62017
  2008
follows:
wenzelm@62017
  2009
wenzelm@62017
  2010
  \<^item>  itemize
wenzelm@62017
  2011
  \<^enum>  enumerate
wenzelm@62017
  2012
  \<^descr>  description
wenzelm@62017
  2013
wenzelm@61491
  2014
* There is a new short form for antiquotations with a single argument
wenzelm@61491
  2015
that is a cartouche: \<^name>\<open>...\<close> is equivalent to @{name \<open>...\<close>} and
wenzelm@61595
  2016
\<open>...\<close> without control symbol is equivalent to @{cartouche \<open>...\<close>}.
wenzelm@61595
  2017
\<^name> without following cartouche is equivalent to @{name}. The
wenzelm@61501
  2018
standard Isabelle fonts provide glyphs to render important control
wenzelm@61501
  2019
symbols, e.g. "\<^verbatim>", "\<^emph>", "\<^bold>".
wenzelm@61491
  2020
wenzelm@61595
  2021
* Antiquotations @{noindent}, @{smallskip}, @{medskip}, @{bigskip} with
wenzelm@61595
  2022
corresponding control symbols \<^noindent>, \<^smallskip>, \<^medskip>, \<^bigskip> specify spacing formally, using
wenzelm@61595
  2023
standard LaTeX macros of the same names.
wenzelm@61595
  2024
wenzelm@61491
  2025
* Antiquotation @{cartouche} in Isabelle/Pure is the same as @{text}.
wenzelm@61491
  2026
Consequently, \<open>...\<close> without any decoration prints literal quasi-formal
wenzelm@61492
  2027
text. Command-line tool "isabelle update_cartouches -t" helps to update
wenzelm@61492
  2028
old sources, by approximative patching of the content of string and
wenzelm@61492
  2029
cartouche tokens seen in theory sources.
wenzelm@61491
  2030
wenzelm@61491
  2031
* The @{text} antiquotation now ignores the antiquotation option
wenzelm@61491
  2032
"source". The given text content is output unconditionally, without any
wenzelm@61491
  2033
surrounding quotes etc. Subtle INCOMPATIBILITY, put quotes into the
wenzelm@61494
  2034
argument where they are really intended, e.g. @{text \<open>"foo"\<close>}. Initial
wenzelm@61494
  2035
or terminal spaces are ignored.
wenzelm@61491
  2036
wenzelm@62017
  2037
* Antiquotations @{emph} and @{bold} output LaTeX source recursively,
wenzelm@62017
  2038
adding appropriate text style markup. These may be used in the short
wenzelm@62017
  2039
form \<^emph>\<open>...\<close> and \<^bold>\<open>...\<close>.
wenzelm@62017
  2040
wenzelm@62017
  2041
* Document antiquotation @{footnote} outputs LaTeX source recursively,
wenzelm@62017
  2042
marked as \footnote{}. This may be used in the short form \<^footnote>\<open>...\<close>.
wenzelm@62017
  2043
wenzelm@62017
  2044
* Antiquotation @{verbatim [display]} supports option "indent".
wenzelm@62017
  2045
wenzelm@62017
  2046
* Antiquotation @{theory_text} prints uninterpreted theory source text
wenzelm@62231
  2047
(Isar outer syntax with command keywords etc.). This may be used in the
wenzelm@62231
  2048
short form \<^theory_text>\<open>...\<close>. @{theory_text [display]} supports option "indent".
wenzelm@62017
  2049
wenzelm@62017
  2050
* Antiquotation @{doc ENTRY} provides a reference to the given
wenzelm@62017
  2051
documentation, with a hyperlink in the Prover IDE.
wenzelm@62017
  2052
wenzelm@62017
  2053
* Antiquotations @{command}, @{method}, @{attribute} print checked
wenzelm@62017
  2054
entities of the Isar language.
wenzelm@62017
  2055
wenzelm@61471
  2056
* HTML presentation uses the standard IsabelleText font and Unicode
wenzelm@61471
  2057
rendering of Isabelle symbols like Isabelle/Scala/jEdit.  The former
wenzelm@61488
  2058
print mode "HTML" loses its special meaning.
wenzelm@61471
  2059
wenzelm@61405
  2060
wenzelm@60406
  2061
*** Isar ***
wenzelm@60406
  2062
wenzelm@62205
  2063
* Local goals ('have', 'show', 'hence', 'thus') allow structured rule
wenzelm@62205
  2064
statements like fixes/assumes/shows in theorem specifications, but the
wenzelm@62205
  2065
notation is postfix with keywords 'if' (or 'when') and 'for'. For
wenzelm@60555
  2066
example:
wenzelm@60414
  2067
wenzelm@60414
  2068
  have result: "C x y"
wenzelm@60414
  2069
    if "A x" and "B y"
wenzelm@60414
  2070
    for x :: 'a and y :: 'a
wenzelm@60414
  2071
    <proof>
wenzelm@60414
  2072
wenzelm@60449
  2073
The local assumptions are bound to the name "that". The result is
wenzelm@60449
  2074
exported from context of the statement as usual. The above roughly
wenzelm@60414
  2075
corresponds to a raw proof block like this:
wenzelm@60414
  2076
wenzelm@60414
  2077
  {
wenzelm@60414
  2078
    fix x :: 'a and y :: 'a
wenzelm@60449
  2079
    assume that: "A x" "B y"
wenzelm@60414
  2080
    have "C x y" <proof>
wenzelm@60414
  2081
  }
wenzelm@60414
  2082
  note result = this
wenzelm@60406
  2083
wenzelm@60555
  2084
The keyword 'when' may be used instead of 'if', to indicate 'presume'
wenzelm@60555
  2085
instead of 'assume' above.
wenzelm@60555
  2086
wenzelm@61733
  2087
* Assumptions ('assume', 'presume') allow structured rule statements
wenzelm@61733
  2088
using 'if' and 'for', similar to 'have' etc. above. For example:
wenzelm@61658
  2089
wenzelm@61658
  2090
  assume result: "C x y"
wenzelm@61658
  2091
    if "A x" and "B y"
wenzelm@61658
  2092
    for x :: 'a and y :: 'a
wenzelm@61658
  2093
wenzelm@61658
  2094
This assumes "\<And>x y::'a. A x \<Longrightarrow> B y \<Longrightarrow> C x y" and produces a general
wenzelm@61658
  2095
result as usual: "A ?x \<Longrightarrow> B ?y \<Longrightarrow> C ?x ?y".
wenzelm@61658
  2096
wenzelm@61658
  2097
Vacuous quantification in assumptions is omitted, i.e. a for-context
wenzelm@61658
  2098
only effects propositions according to actual use of variables. For
wenzelm@61658
  2099
example:
wenzelm@61658
  2100
wenzelm@61658
  2101
  assume "A x" and "B y" for x and y
wenzelm@61658
  2102
wenzelm@61658
  2103
is equivalent to:
wenzelm@61658
  2104
wenzelm@61658
  2105
  assume "\<And>x. A x" and "\<And>y. B y"
wenzelm@61658
  2106
wenzelm@60595
  2107
* The meaning of 'show' with Pure rule statements has changed: premises
wenzelm@60595
  2108
are treated in the sense of 'assume', instead of 'presume'. This means,
wenzelm@62205
  2109
a goal like "\<And>x. A x \<Longrightarrow> B x \<Longrightarrow> C x" can be solved completely as
wenzelm@62205
  2110
follows:
wenzelm@60595
  2111
wenzelm@60595
  2112
  show "\<And>x. A x \<Longrightarrow> B x \<Longrightarrow> C x"
wenzelm@60595
  2113
wenzelm@60595
  2114
or:
wenzelm@60595
  2115
wenzelm@60595
  2116
  show "C x" if "A x" "B x" for x
wenzelm@60595
  2117
wenzelm@60595
  2118
Rare INCOMPATIBILITY, the old behaviour may be recovered as follows:
wenzelm@60595
  2119
wenzelm@60595
  2120
  show "C x" when "A x" "B x" for x
wenzelm@60595
  2121
wenzelm@60459
  2122
* New command 'consider' states rules for generalized elimination and
wenzelm@60459
  2123
case splitting. This is like a toplevel statement "theorem obtains" used
wenzelm@60459
  2124
within a proof body; or like a multi-branch 'obtain' without activation
wenzelm@60459
  2125
of the local context elements yet.
wenzelm@60459
  2126
wenzelm@60455
  2127
* Proof method "cases" allows to specify the rule as first entry of
wenzelm@60455
  2128
chained facts.  This is particularly useful with 'consider':
wenzelm@60455
  2129
wenzelm@60455
  2130
  consider (a) A | (b) B | (c) C <proof>
wenzelm@60455
  2131
  then have something
wenzelm@60455
  2132
  proof cases
wenzelm@60455
  2133
    case a
wenzelm@60455
  2134
    then show ?thesis <proof>
wenzelm@60455
  2135
  next
wenzelm@60455
  2136
    case b
wenzelm@60455
  2137
    then show ?thesis <proof>
wenzelm@60455
  2138
  next
wenzelm@60455
  2139
    case c
wenzelm@60455
  2140
    then show ?thesis <proof>
wenzelm@60455
  2141
  qed
wenzelm@60455
  2142
wenzelm@60565
  2143
* Command 'case' allows fact name and attribute specification like this:
wenzelm@60565
  2144
wenzelm@60565
  2145
  case a: (c xs)
wenzelm@60565
  2146
  case a [attributes]: (c xs)
wenzelm@60565
  2147
wenzelm@60565
  2148
Facts that are introduced by invoking the case context are uniformly
wenzelm@60565
  2149
qualified by "a"; the same name is used for the cumulative fact. The old
wenzelm@60565
  2150
form "case (c xs) [attributes]" is no longer supported. Rare
wenzelm@60565
  2151
INCOMPATIBILITY, need to adapt uses of case facts in exotic situations,
wenzelm@60565
  2152
and always put attributes in front.
wenzelm@60565
  2153
wenzelm@60618
  2154
* The standard proof method of commands 'proof' and '..' is now called
wenzelm@60618
  2155
"standard" to make semantically clear what it is; the old name "default"
wenzelm@60618
  2156
is still available as legacy for some time. Documentation now explains
wenzelm@60618
  2157
'..' more accurately as "by standard" instead of "by rule".
wenzelm@60618
  2158
wenzelm@62017
  2159
* Nesting of Isar goal structure has been clarified: the context after
wenzelm@62017
  2160
the initial backwards refinement is retained for the whole proof, within
wenzelm@62017
  2161
all its context sections (as indicated via 'next'). This is e.g.
wenzelm@62017
  2162
relevant for 'using', 'including', 'supply':
wenzelm@62017
  2163
wenzelm@62017
  2164
  have "A \<and> A" if a: A for A
wenzelm@62017
  2165
    supply [simp] = a
wenzelm@62017
  2166
  proof
wenzelm@62017
  2167
    show A by simp
wenzelm@62017
  2168
  next
wenzelm@62017
  2169
    show A by simp
wenzelm@62017
  2170
  qed
wenzelm@62017
  2171
wenzelm@62017
  2172
* Command 'obtain' binds term abbreviations (via 'is' patterns) in the
wenzelm@62017
  2173
proof body as well, abstracted over relevant parameters.
wenzelm@62017
  2174
wenzelm@62017
  2175
* Improved type-inference for theorem statement 'obtains': separate
wenzelm@62017
  2176
parameter scope for of each clause.
wenzelm@62017
  2177
wenzelm@62017
  2178
* Term abbreviations via 'is' patterns also work for schematic
wenzelm@62017
  2179
statements: result is abstracted over unknowns.
wenzelm@62017
  2180
wenzelm@60631
  2181
* Command 'subgoal' allows to impose some structure on backward
wenzelm@60631
  2182
refinements, to avoid proof scripts degenerating into long of 'apply'
wenzelm@60631
  2183
sequences. Further explanations and examples are given in the isar-ref
wenzelm@60631
  2184
manual.
wenzelm@60631
  2185
wenzelm@62017
  2186
* Command 'supply' supports fact definitions during goal refinement
wenzelm@62017
  2187
('apply' scripts).
wenzelm@62017
  2188
wenzelm@61166
  2189
* Proof method "goal_cases" turns the current subgoals into cases within
wenzelm@61166
  2190
the context; the conclusion is bound to variable ?case in each case. For
wenzelm@61166
  2191
example:
wenzelm@60617
  2192
wenzelm@60617
  2193
lemma "\<And>x. A x \<Longrightarrow> B x \<Longrightarrow> C x"
wenzelm@60622
  2194
  and "\<And>y z. U y \<Longrightarrow> V z \<Longrightarrow> W y z"
wenzelm@61166
  2195
proof goal_cases
wenzelm@60622
  2196
  case (1 x)
wenzelm@60622
  2197
  then show ?case using \<open>A x\<close> \<open>B x\<close> sorry
wenzelm@60622
  2198
next
wenzelm@60622
  2199
  case (2 y z)
wenzelm@60622
  2200
  then show ?case using \<open>U y\<close> \<open>V z\<close> sorry
wenzelm@60622
  2201
qed
wenzelm@60622
  2202
wenzelm@60622
  2203
lemma "\<And>x. A x \<Longrightarrow> B x \<Longrightarrow> C x"
wenzelm@60622
  2204
  and "\<And>y z. U y \<Longrightarrow> V z \<Longrightarrow> W y z"
wenzelm@61166
  2205
proof goal_cases
wenzelm@60617
  2206
  case prems: 1
wenzelm@60617
  2207
  then show ?case using prems sorry
wenzelm@60617
  2208
next
wenzelm@60617
  2209
  case prems: 2
wenzelm@60617
  2210
  then show ?case using prems sorry
wenzelm@60617
  2211
qed
wenzelm@60578
  2212
wenzelm@60581
  2213
* The undocumented feature of implicit cases goal1, goal2, goal3, etc.
wenzelm@60617
  2214
is marked as legacy, and will be removed eventually. The proof method
wenzelm@60617
  2215
"goals" achieves a similar effect within regular Isar; often it can be
wenzelm@60617
  2216
done more adequately by other means (e.g. 'consider').
wenzelm@60581
  2217
wenzelm@62017
  2218
* The vacuous fact "TERM x" may be established "by fact" or as `TERM x`
wenzelm@62017
  2219
as well, not just "by this" or "." as before.
wenzelm@60551
  2220
wenzelm@60554
  2221
* Method "sleep" succeeds after a real-time delay (in seconds). This is
wenzelm@60554
  2222
occasionally useful for demonstration and testing purposes.
wenzelm@60554
  2223
wenzelm@60406
  2224
wenzelm@60331
  2225
*** Pure ***
wenzelm@60331
  2226
wenzelm@61606
  2227
* Qualifiers in locale expressions default to mandatory ('!') regardless
wenzelm@61606
  2228
of the command. Previously, for 'locale' and 'sublocale' the default was
wenzelm@61606
  2229
optional ('?'). The old synatx '!' has been discontinued.
wenzelm@61606
  2230
INCOMPATIBILITY, remove '!' and add '?' as required.
ballarin@61565
  2231
ballarin@61566
  2232
* Keyword 'rewrites' identifies rewrite morphisms in interpretation
wenzelm@62017
  2233
commands. Previously, the keyword was 'where'. INCOMPATIBILITY.
ballarin@61566
  2234
ballarin@61701
  2235
* More gentle suppression of syntax along locale morphisms while
wenzelm@62017
  2236
printing terms. Previously 'abbreviation' and 'notation' declarations
wenzelm@62017
  2237
would be suppressed for morphisms except term identity. Now
ballarin@61701
  2238
'abbreviation' is also kept for morphims that only change the involved
wenzelm@62017
  2239
parameters, and only 'notation' is suppressed. This can be of great help
wenzelm@62017
  2240
when working with complex locale hierarchies, because proof states are
wenzelm@62017
  2241
displayed much more succinctly. It also means that only notation needs
wenzelm@62017
  2242
to be redeclared if desired, as illustrated by this example:
ballarin@61701
  2243
ballarin@61701
  2244
  locale struct = fixes composition :: "'a => 'a => 'a" (infixl "\<cdot>" 65)
ballarin@61701
  2245
  begin
ballarin@61701
  2246
    definition derived (infixl "\<odot>" 65) where ...
ballarin@61701
  2247
  end
ballarin@61701
  2248
ballarin@61701
  2249
  locale morphism =
ballarin@61701
  2250
    left: struct composition + right: struct composition'
ballarin@61701
  2251
    for composition (infix "\<cdot>" 65) and composition' (infix "\<cdot>''" 65)
ballarin@61701
  2252
  begin
ballarin@61701
  2253
    notation right.derived ("\<odot>''")
ballarin@61701
  2254
  end
ballarin@61701
  2255
wenzelm@61895
  2256
* Command 'global_interpretation' issues interpretations into global
wenzelm@61895
  2257
theories, with optional rewrite definitions following keyword 'defines'.
wenzelm@61895
  2258
wenzelm@61895
  2259
* Command 'sublocale' accepts optional rewrite definitions after keyword
haftmann@61675
  2260
'defines'.
haftmann@61675
  2261
wenzelm@61895
  2262
* Command 'permanent_interpretation' has been discontinued. Use
wenzelm@61895
  2263
'global_interpretation' or 'sublocale' instead. INCOMPATIBILITY.
haftmann@61670
  2264
wenzelm@61252
  2265
* Command 'print_definitions' prints dependencies of definitional
wenzelm@61252
  2266
specifications. This functionality used to be part of 'print_theory'.
wenzelm@61252
  2267
wenzelm@60331
  2268
* Configuration option rule_insts_schematic has been discontinued
wenzelm@62017
  2269
(intermediate legacy feature in Isabelle2015). INCOMPATIBILITY.
wenzelm@60331
  2270
wenzelm@62205
  2271
* Abbreviations in type classes now carry proper sort constraint. Rare
wenzelm@62205
  2272
INCOMPATIBILITY in situations where the previous misbehaviour has been
wenzelm@62205
  2273
exploited.
haftmann@60347
  2274
haftmann@60347
  2275
* Refinement of user-space type system in type classes: pseudo-local
wenzelm@62205
  2276
operations behave more similar to abbreviations. Potential
haftmann@60347
  2277
INCOMPATIBILITY in exotic situations.
haftmann@60347
  2278
haftmann@60347
  2279
nipkow@60171
  2280
*** HOL ***
nipkow@60171
  2281
wenzelm@62017
  2282
* The 'typedef' command has been upgraded from a partially checked
wenzelm@62017
  2283
"axiomatization", to a full definitional specification that takes the
wenzelm@62017
  2284
global collection of overloaded constant / type definitions into
wenzelm@62017
  2285
account. Type definitions with open dependencies on overloaded
wenzelm@62017
  2286
definitions need to be specified as "typedef (overloaded)". This
wenzelm@62017
  2287
provides extra robustness in theory construction. Rare INCOMPATIBILITY.
wenzelm@62017
  2288
wenzelm@62017
  2289
* Qualification of various formal entities in the libraries is done more
wenzelm@62017
  2290
uniformly via "context begin qualified definition ... end" instead of
wenzelm@62017
  2291
old-style "hide_const (open) ...". Consequently, both the defined
wenzelm@62017
  2292
constant and its defining fact become qualified, e.g. Option.is_none and
wenzelm@62017
  2293
Option.is_none_def. Occasional INCOMPATIBILITY in applications.
wenzelm@62017
  2294
wenzelm@62017
  2295
* Some old and rarely used ASCII replacement syntax has been removed.
wenzelm@62017
  2296
INCOMPATIBILITY, standard syntax with symbols should be used instead.
wenzelm@62017
  2297
The subsequent commands help to reproduce the old forms, e.g. to
wenzelm@62017
  2298
simplify porting old theories:
wenzelm@62017
  2299
wenzelm@62017
  2300
  notation iff  (infixr "<->" 25)
wenzelm@62017
  2301
wenzelm@62017
  2302
  notation Times  (infixr "<*>" 80)
wenzelm@62017
  2303
wenzelm@62017
  2304
  type_notation Map.map  (infixr "~=>" 0)
wenzelm@62017
  2305
  notation Map.map_comp  (infixl "o'_m" 55)
wenzelm@62017
  2306
wenzelm@62017
  2307
  type_notation FinFun.finfun ("(_ =>f /_)" [22, 21] 21)
wenzelm@62017
  2308
wenzelm@62017
  2309
  notation FuncSet.funcset  (infixr "->" 60)
wenzelm@62017
  2310
  notation FuncSet.extensional_funcset  (infixr "->\<^sub>E" 60)
wenzelm@62017
  2311
wenzelm@62017
  2312
  notation Omega_Words_Fun.conc (infixr "conc" 65)
wenzelm@62017
  2313
wenzelm@62017
  2314
  notation Preorder.equiv ("op ~~")
wenzelm@62017
  2315
    and Preorder.equiv ("(_/ ~~ _)" [51, 51] 50)
wenzelm@62017
  2316
wenzelm@62017
  2317
  notation (in topological_space) tendsto (infixr "--->" 55)
wenzelm@62017
  2318
  notation (in topological_space) LIMSEQ ("((_)/ ----> (_))" [60, 60] 60)
wenzelm@62017
  2319
  notation LIM ("((_)/ -- (_)/ --> (_))" [60, 0, 60] 60)
wenzelm@62017
  2320
wenzelm@62017
  2321
  notation NSA.approx (infixl "@=" 50)
wenzelm@62017
  2322
  notation NSLIMSEQ ("((_)/ ----NS> (_))" [60, 60] 60)
wenzelm@62017
  2323
  notation NSLIM ("((_)/ -- (_)/ --NS> (_))" [60, 0, 60] 60)
wenzelm@62017
  2324
wenzelm@62017
  2325
* The alternative notation "\<Colon>" for type and sort constraints has been
wenzelm@62017
  2326
removed: in LaTeX document output it looks the same as "::".
wenzelm@62017
  2327
INCOMPATIBILITY, use plain "::" instead.
wenzelm@62017
  2328
wenzelm@62017
  2329
* Commands 'inductive' and 'inductive_set' work better when names for
wenzelm@62017
  2330
intro rules are omitted: the "cases" and "induct" rules no longer
wenzelm@62017
  2331
declare empty case_names, but no case_names at all. This allows to use
wenzelm@62017
  2332
numbered cases in proofs, without requiring method "goal_cases".
wenzelm@62017
  2333
wenzelm@62017
  2334
* Inductive definitions ('inductive', 'coinductive', etc.) expose
wenzelm@62017
  2335
low-level facts of the internal construction only if the option
wenzelm@62093
  2336
"inductive_internals" is enabled. This refers to the internal predicate
wenzelm@62017
  2337
definition and its monotonicity result. Rare INCOMPATIBILITY.
wenzelm@62017
  2338
wenzelm@62017
  2339
* Recursive function definitions ('fun', 'function', 'partial_function')
wenzelm@62017
  2340
expose low-level facts of the internal construction only if the option
wenzelm@62205
  2341
"function_internals" is enabled. Its internal inductive definition is
wenzelm@62205
  2342
also subject to "inductive_internals". Rare INCOMPATIBILITY.
wenzelm@62093
  2343
wenzelm@62093
  2344
* BNF datatypes ('datatype', 'codatatype', etc.) expose low-level facts
wenzelm@62093
  2345
of the internal construction only if the option "bnf_internals" is
wenzelm@62093
  2346
enabled. This supersedes the former option "bnf_note_all". Rare
wenzelm@62093
  2347
INCOMPATIBILITY.
wenzelm@62017
  2348
wenzelm@62017
  2349
* Combinator to represent case distinction on products is named
wenzelm@62017
  2350
"case_prod", uniformly, discontinuing any input aliasses. Very popular
wenzelm@62017
  2351
theorem aliasses have been retained.
wenzelm@62017
  2352
haftmann@61424
  2353
Consolidated facts:
haftmann@61424
  2354
  PairE ~> prod.exhaust
haftmann@61424
  2355
  Pair_eq ~> prod.inject
haftmann@61424
  2356
  pair_collapse ~> prod.collapse
haftmann@61424
  2357
  Pair_fst_snd_eq ~> prod_eq_iff
haftmann@61424
  2358
  split_twice ~> prod.case_distrib
haftmann@61424
  2359
  split_weak_cong ~> prod.case_cong_weak
haftmann@61424
  2360
  split_split ~> prod.split
haftmann@61424
  2361
  split_split_asm ~> prod.split_asm
haftmann@61424
  2362
  splitI ~> case_prodI
haftmann@61424
  2363
  splitD ~> case_prodD
haftmann@61424
  2364
  splitI2 ~> case_prodI2
haftmann@61424
  2365
  splitI2' ~> case_prodI2'
haftmann@61424
  2366
  splitE ~> case_prodE
haftmann@61424
  2367
  splitE' ~> case_prodE'
haftmann@61424
  2368
  split_pair ~> case_prod_Pair
haftmann@61424
  2369
  split_eta ~> case_prod_eta
haftmann@61424
  2370
  split_comp ~> case_prod_comp
haftmann@61424
  2371
  mem_splitI ~> mem_case_prodI
haftmann@61424
  2372
  mem_splitI2 ~> mem_case_prodI2
haftmann@61424
  2373
  mem_splitE ~> mem_case_prodE
haftmann@61424
  2374
  The_split ~> The_case_prod
haftmann@61424
  2375
  cond_split_eta ~> cond_case_prod_eta
haftmann@61424
  2376
  Collect_split_in_rel_leE ~> Collect_case_prod_in_rel_leE
haftmann@61424
  2377
  Collect_split_in_rel_leI ~> Collect_case_prod_in_rel_leI
haftmann@61424
  2378
  in_rel_Collect_split_eq ~> in_rel_Collect_case_prod_eq
haftmann@61424
  2379
  Collect_split_Grp_eqD ~> Collect_case_prod_Grp_eqD
haftmann@61424
  2380
  Collect_split_Grp_inD ~> Collect_case_prod_Grp_in
haftmann@61424
  2381
  Domain_Collect_split ~> Domain_Collect_case_prod
haftmann@61424
  2382
  Image_Collect_split ~> Image_Collect_case_prod
haftmann@61424
  2383
  Range_Collect_split ~> Range_Collect_case_prod
haftmann@61424
  2384
  Eps_split ~> Eps_case_prod
haftmann@61424
  2385
  Eps_split_eq ~> Eps_case_prod_eq
haftmann@61424
  2386
  split_rsp ~> case_prod_rsp
haftmann@61424
  2387
  curry_split ~> curry_case_prod
haftmann@61424
  2388
  split_curry ~> case_prod_curry
wenzelm@62017
  2389
haftmann@61424
  2390
Changes in structure HOLogic:
haftmann@61424
  2391
  split_const ~> case_prod_const
haftmann@61424
  2392
  mk_split ~> mk_case_prod
haftmann@61424
  2393
  mk_psplits ~> mk_ptupleabs
haftmann@61424
  2394
  strip_psplits ~> strip_ptupleabs
wenzelm@62017
  2395
wenzelm@62017
  2396
INCOMPATIBILITY.
wenzelm@62017
  2397
wenzelm@62017
  2398
* The coercions to type 'real' have been reorganised. The function
wenzelm@62017
  2399
'real' is no longer overloaded, but has type 'nat => real' and
wenzelm@62017
  2400
abbreviates of_nat for that type. Also 'real_of_int :: int => real'
wenzelm@62017
  2401
abbreviates of_int for that type. Other overloaded instances of 'real'
wenzelm@62017
  2402
have been replaced by 'real_of_ereal' and 'real_of_float'.
wenzelm@62017
  2403
lp15@61694
  2404
Consolidated facts (among others):
lp15@61694
  2405
  real_of_nat_le_iff -> of_nat_le_iff
lp15@61694
  2406
  real_of_nat_numeral of_nat_numeral
lp15@61694
  2407
  real_of_int_zero of_int_0
lp15@61694
  2408
  real_of_nat_zero of_nat_0
lp15@61694
  2409
  real_of_one of_int_1
lp15@61694
  2410
  real_of_int_add of_int_add
lp15@61694
  2411
  real_of_nat_add of_nat_add
lp15@61694
  2412
  real_of_int_diff of_int_diff
lp15@61694
  2413
  real_of_nat_diff of_nat_diff
lp15@61694
  2414
  floor_subtract floor_diff_of_int
lp15@61694
  2415
  real_of_int_inject of_int_eq_iff
lp15@61694
  2416
  real_of_int_gt_zero_cancel_iff of_int_0_less_iff
lp15@61694
  2417
  real_of_int_ge_zero_cancel_iff of_int_0_le_iff
lp15@61694
  2418
  real_of_nat_ge_zero of_nat_0_le_iff
lp15@61694
  2419
  real_of_int_ceiling_ge le_of_int_ceiling
lp15@61694
  2420
  ceiling_less_eq ceiling_less_iff
lp15@61694
  2421
  ceiling_le_eq ceiling_le_iff
lp15@61694
  2422
  less_floor_eq less_floor_iff
lp15@61694
  2423
  floor_less_eq floor_less_iff
lp15@61694
  2424
  floor_divide_eq_div floor_divide_of_int_eq
lp15@61694
  2425
  real_of_int_zero_cancel of_nat_eq_0_iff
lp15@61694
  2426
  ceiling_real_of_int ceiling_of_int
wenzelm@62017
  2427
wenzelm@62017
  2428
INCOMPATIBILITY.
wenzelm@61143
  2429
wenzelm@60841
  2430
* Theory Map: lemma map_of_is_SomeD was a clone of map_of_SomeD and has
wenzelm@60841
  2431
been removed. INCOMPATIBILITY.
wenzelm@60841
  2432
lars@60712
  2433
* Quickcheck setup for finite sets.
lars@60712
  2434
nipkow@60171
  2435
* Discontinued simp_legacy_precond. Potential INCOMPATIBILITY.
wenzelm@60138
  2436
blanchet@60306
  2437
* Sledgehammer:
blanchet@61318
  2438
  - The MaSh relevance filter has been sped up.
blanchet@60306
  2439
  - Proof reconstruction has been improved, to minimize the incidence of
blanchet@60306
  2440
    cases where Sledgehammer gives a proof that does not work.
blanchet@60306
  2441
  - Auto Sledgehammer now minimizes and preplays the results.
blanchet@61030
  2442
  - Handle Vampire 4.0 proof output without raising exception.
blanchet@61043
  2443
  - Eliminated "MASH" environment variable. Use the "MaSh" option in
blanchet@61043
  2444
    Isabelle/jEdit instead. INCOMPATIBILITY.
blanchet@61317
  2445
  - Eliminated obsolete "blocking" option and related subcommands.
blanchet@60306
  2446
blanchet@60310
  2447
* Nitpick:
blanchet@61325
  2448
  - Fixed soundness bug in translation of "finite" predicate.
blanchet@61324
  2449
  - Fixed soundness bug in "destroy_constrs" optimization.
blanchet@62080
  2450
  - Fixed soundness bug in translation of "rat" type.
blanchet@60310
  2451
  - Removed "check_potential" and "check_genuine" options.
blanchet@61317
  2452
  - Eliminated obsolete "blocking" option.
blanchet@60310
  2453
wenzelm@62027
  2454
* (Co)datatype package:
blanchet@61345
  2455
  - New commands "lift_bnf" and "copy_bnf" for lifting (copying) a BNF
blanchet@61345
  2456
    structure on the raw type to an abstract type defined using typedef.
blanchet@61345
  2457
  - Always generate "case_transfer" theorem.
wenzelm@62235
  2458
  - For mutual types, generate slightly stronger "rel_induct",
haftmann@63946
  2459
    "rel_coinduct", and "coinduct" theorems. INCOMPATIBILITY.
blanchet@61551
  2460
  - Allow discriminators and selectors with the same name as the type
blanchet@61551
  2461
    being defined.
blanchet@61551
  2462
  - Avoid various internal name clashes (e.g., 'datatype f = f').
traytel@60920
  2463
wenzelm@62098
  2464
* Transfer: new methods for interactive debugging of 'transfer' and
wenzelm@62098
  2465
'transfer_prover': 'transfer_start', 'transfer_step', 'transfer_end',
wenzelm@62098
  2466
'transfer_prover_start' and 'transfer_prover_end'.
kuncar@61370
  2467
kleing@62118
  2468
* New diagnostic command print_record for displaying record definitions.
kleing@62118
  2469
haftmann@60868
  2470
* Division on integers is bootstrapped directly from division on
wenzelm@62017
  2471
naturals and uses generic numeral algorithm for computations. Slight
wenzelm@62017
  2472
INCOMPATIBILITY, simproc numeral_divmod replaces and generalizes former
wenzelm@62017
  2473
simprocs binary_int_div and binary_int_mod
wenzelm@62017
  2474
wenzelm@62017
  2475
* Tightened specification of class semiring_no_zero_divisors. Minor
haftmann@60516
  2476
INCOMPATIBILITY.
haftmann@60516
  2477
haftmann@60688
  2478
* Class algebraic_semidom introduces common algebraic notions of
wenzelm@62017
  2479
integral (semi)domains, particularly units. Although logically subsumed
wenzelm@62017
  2480
by fields, is is not a super class of these in order not to burden
wenzelm@62017
  2481
fields with notions that are trivial there.
wenzelm@62017
  2482
wenzelm@62017
  2483
* Class normalization_semidom specifies canonical representants for
wenzelm@62017
  2484
equivalence classes of associated elements in an integral (semi)domain.
wenzelm@62017
  2485
This formalizes associated elements as well.
haftmann@60688
  2486
haftmann@60688
  2487
* Abstract specification of gcd/lcm operations in classes semiring_gcd,
wenzelm@62017
  2488
semiring_Gcd, semiring_Lcd. Minor INCOMPATIBILITY: facts gcd_nat.commute
wenzelm@62017
  2489
and gcd_int.commute are subsumed by gcd.commute, as well as
wenzelm@62017
  2490
gcd_nat.assoc and gcd_int.assoc by gcd.assoc.
wenzelm@62017
  2491
wenzelm@62017
  2492
* Former constants Fields.divide (_ / _) and Divides.div (_ div _) are
wenzelm@62017
  2493
logically unified to Rings.divide in syntactic type class Rings.divide,
wenzelm@62017
  2494
with infix syntax (_ div _). Infix syntax (_ / _) for field division is
wenzelm@62017
  2495
added later as abbreviation in class Fields.inverse. INCOMPATIBILITY,
wenzelm@62017
  2496
instantiations must refer to Rings.divide rather than the former
wenzelm@62017
  2497
separate constants, hence infix syntax (_ / _) is usually not available
wenzelm@62017
  2498
during instantiation.
wenzelm@62017
  2499
wenzelm@62017
  2500
* New cancellation simprocs for boolean algebras to cancel complementary
wenzelm@62017
  2501
terms for sup and inf. For example, "sup x (sup y (- x))" simplifies to
wenzelm@62017
  2502
"top". INCOMPATIBILITY.
Andreas@61629
  2503
hoelzl@62101
  2504
* Class uniform_space introduces uniform spaces btw topological spaces
hoelzl@62101
  2505
and metric spaces. Minor INCOMPATIBILITY: open_<type>_def needs to be
wenzelm@62205
  2506
introduced in the form of an uniformity. Some constants are more general
wenzelm@62205
  2507
now, it may be necessary to add type class constraints.
hoelzl@62101
  2508
hoelzl@62101
  2509
  open_real_def \<leadsto> open_dist
hoelzl@62101
  2510
  open_complex_def \<leadsto> open_dist
hoelzl@62101
  2511
wenzelm@62026
  2512
* Library/Monad_Syntax: notation uses symbols \<bind> and \<then>. INCOMPATIBILITY.
wenzelm@62026
  2513
Mathias@60397
  2514
* Library/Multiset:
Mathias@60397
  2515
  - Renamed multiset inclusion operators:
Mathias@60397
  2516
      < ~> <#
blanchet@62208
  2517
      > ~> >#
Mathias@60397
  2518
      <= ~> <=#
blanchet@62208
  2519
      >= ~> >=#
Mathias@60397
  2520
      \<le> ~> \<le>#
blanchet@62208
  2521
      \<ge> ~> \<ge>#
Mathias@60397
  2522
    INCOMPATIBILITY.
blanchet@62209
  2523
  - Added multiset inclusion operator syntax:
blanchet@62209
  2524
      \<subset>#
blanchet@62209
  2525
      \<subseteq>#
blanchet@62209
  2526
      \<supset>#
blanchet@62209
  2527
      \<supseteq>#
Mathias@60397
  2528
  - "'a multiset" is no longer an instance of the "order",
Mathias@60397
  2529
    "ordered_ab_semigroup_add_imp_le", "ordered_cancel_comm_monoid_diff",
Mathias@60397
  2530
    "semilattice_inf", and "semilattice_sup" type classes. The theorems
Mathias@60397
  2531
    previously provided by these type classes (directly or indirectly)
Mathias@60397
  2532
    are now available through the "subset_mset" interpretation
Mathias@60397
  2533
    (e.g. add_mono ~> subset_mset.add_mono).
Mathias@60397
  2534
    INCOMPATIBILITY.
nipkow@60497
  2535
  - Renamed conversions:
nipkow@60515
  2536
      multiset_of ~> mset
nipkow@60515
  2537
      multiset_of_set ~> mset_set
nipkow@60497
  2538
      set_of ~> set_mset
nipkow@60497
  2539
    INCOMPATIBILITY
Mathias@60398
  2540
  - Renamed lemmas:
Mathias@60398
  2541
      mset_le_def ~> subseteq_mset_def
Mathias@60398
  2542
      mset_less_def ~> subset_mset_def
Mathias@60400
  2543
      less_eq_multiset.rep_eq ~> subseteq_mset_def
Mathias@60400
  2544
    INCOMPATIBILITY
Mathias@60400
  2545
  - Removed lemmas generated by lift_definition:
wenzelm@62235
  2546
    less_eq_multiset.abs_eq, less_eq_multiset.rsp,
wenzelm@62235
  2547
    less_eq_multiset.transfer, less_eq_multiset_def
Mathias@60400
  2548
    INCOMPATIBILITY
wenzelm@60006
  2549
wenzelm@62017
  2550
* Library/Omega_Words_Fun: Infinite words modeled as functions nat \<Rightarrow> 'a.
wenzelm@62017
  2551
wenzelm@62017
  2552
* Library/Bourbaki_Witt_Fixpoint: Added formalisation of the
wenzelm@62017
  2553
Bourbaki-Witt fixpoint theorem for increasing functions in
wenzelm@62017
  2554
chain-complete partial orders.
wenzelm@62017
  2555
wenzelm@62017
  2556
* Library/Old_Recdef: discontinued obsolete 'defer_recdef' command.
wenzelm@62017
  2557
Minor INCOMPATIBILITY, use 'function' instead.
wenzelm@62017
  2558
wenzelm@62064
  2559
* Library/Periodic_Fun: a locale that provides convenient lemmas for
wenzelm@62064
  2560
periodic functions.
eberlm@62060
  2561
wenzelm@62098
  2562
* Library/Formal_Power_Series: proper definition of division (with
wenzelm@62098
  2563
remainder) for formal power series; instances for Euclidean Ring and
wenzelm@62098
  2564
GCD.
eberlm@62086
  2565
wenzelm@62084
  2566
* HOL-Imperative_HOL: obsolete theory Legacy_Mrec has been removed.
wenzelm@62084
  2567
wenzelm@62084
  2568
* HOL-Statespace: command 'statespace' uses mandatory qualifier for
wenzelm@62084
  2569
import of parent, as for general 'locale' expressions. INCOMPATIBILITY,
wenzelm@62084
  2570
remove '!' and add '?' as required.
wenzelm@62084
  2571
wenzelm@62237
  2572
* HOL-Decision_Procs: The "approximation" method works with "powr"
wenzelm@62237
  2573
(exponentiation on real numbers) again.
wenzelm@62237
  2574
wenzelm@62084
  2575
* HOL-Multivariate_Analysis: theory Cauchy_Integral_Thm with Contour
wenzelm@62084
  2576
integrals (= complex path integrals), Cauchy's integral theorem, winding
wenzelm@62084
  2577
numbers and Cauchy's integral formula, Liouville theorem, Fundamental
wenzelm@62084
  2578
Theorem of Algebra. Ported from HOL Light.
wenzelm@62084
  2579
wenzelm@62084
  2580
* HOL-Multivariate_Analysis: topological concepts such as connected
wenzelm@62017
  2581
components, homotopic paths and the inside or outside of a set.
wenzelm@61121
  2582
wenzelm@62084
  2583
* HOL-Multivariate_Analysis: radius of convergence of power series and
wenzelm@62064
  2584
various summability tests; Harmonic numbers and the Euler–Mascheroni
wenzelm@62064
  2585
constant; the Generalised Binomial Theorem; the complex and real
wenzelm@62064
  2586
Gamma/log-Gamma/Digamma/ Polygamma functions and their most important
wenzelm@62064
  2587
properties.
eberlm@62060
  2588
wenzelm@62084
  2589
* HOL-Probability: The central limit theorem based on Levy's uniqueness
wenzelm@62084
  2590
and continuity theorems, weak convergence, and characterisitc functions.
wenzelm@62084
  2591
wenzelm@62084
  2592
* HOL-Data_Structures: new and growing session of standard data
wenzelm@62084
  2593
structures.
lammich@61178
  2594
wenzelm@60479
  2595
wenzelm@60793
  2596
*** ML ***
wenzelm@60793
  2597
wenzelm@62017
  2598
* The following combinators for low-level profiling of the ML runtime
wenzelm@62017
  2599
system are available:
wenzelm@62017
  2600
wenzelm@62017
  2601
  profile_time          (*CPU time*)
wenzelm@62017
  2602
  profile_time_thread   (*CPU time on this thread*)
wenzelm@62017
  2603
  profile_allocations   (*overall heap allocations*)
wenzelm@62017
  2604
wenzelm@62017
  2605
* Antiquotation @{undefined} or \<^undefined> inlines (raise Match).
wenzelm@62017
  2606
wenzelm@62075
  2607
* Antiquotation @{method NAME} inlines the (checked) name of the given
wenzelm@62075
  2608
Isar proof method.
wenzelm@62075
  2609
wenzelm@61922
  2610
* Pretty printing of Poly/ML compiler output in Isabelle has been
wenzelm@61922
  2611
improved: proper treatment of break offsets and blocks with consistent
wenzelm@61922
  2612
breaks.
wenzelm@61922
  2613
wenzelm@61268
  2614
* The auxiliary module Pure/display.ML has been eliminated. Its
wenzelm@61268
  2615
elementary thm print operations are now in Pure/more_thm.ML and thus
wenzelm@61268
  2616
called Thm.pretty_thm, Thm.string_of_thm etc. INCOMPATIBILITY.
wenzelm@61268
  2617
wenzelm@61144
  2618
* Simproc programming interfaces have been simplified:
wenzelm@61144
  2619
Simplifier.make_simproc and Simplifier.define_simproc supersede various
wenzelm@61144
  2620
forms of Simplifier.mk_simproc, Simplifier.simproc_global etc. Note that
wenzelm@61144
  2621
term patterns for the left-hand sides are specified with implicitly
wenzelm@61144
  2622
fixed variables, like top-level theorem statements. INCOMPATIBILITY.
wenzelm@61144
  2623
wenzelm@60802
  2624
* Instantiation rules have been re-organized as follows:
wenzelm@60802
  2625
wenzelm@60802
  2626
  Thm.instantiate  (*low-level instantiation with named arguments*)
wenzelm@60802
  2627
  Thm.instantiate' (*version with positional arguments*)
wenzelm@60802
  2628
wenzelm@60802
  2629
  Drule.infer_instantiate  (*instantiation with type inference*)
wenzelm@60802
  2630
  Drule.infer_instantiate'  (*version with positional arguments*)
wenzelm@60802
  2631
wenzelm@60802
  2632
The LHS only requires variable specifications, instead of full terms.
wenzelm@60802
  2633
Old cterm_instantiate is superseded by infer_instantiate.
wenzelm@60802
  2634
INCOMPATIBILITY, need to re-adjust some ML names and types accordingly.
wenzelm@60802
  2635
wenzelm@60793
  2636
* Old tactic shorthands atac, rtac, etac, dtac, ftac have been
wenzelm@60793
  2637
discontinued. INCOMPATIBILITY, use regular assume_tac, resolve_tac etc.
wenzelm@60793
  2638
instead (with proper context).
wenzelm@60642
  2639
wenzelm@60642
  2640
* Thm.instantiate (and derivatives) no longer require the LHS of the
wenzelm@60642
  2641
instantiation to be certified: plain variables are given directly.
wenzelm@60642
  2642
wenzelm@60707
  2643
* Subgoal.SUBPROOF and Subgoal.FOCUS combinators use anonymous
wenzelm@60707
  2644
quasi-bound variables (like the Simplifier), instead of accidentally
wenzelm@60707
  2645
named local fixes. This has the potential to improve stability of proof
wenzelm@60707
  2646
tools, but can also cause INCOMPATIBILITY for tools that don't observe
wenzelm@60707
  2647
the proof context discipline.
wenzelm@60707
  2648
wenzelm@62017
  2649
* Isar proof methods are based on a slightly more general type
wenzelm@62017
  2650
context_tactic, which allows to change the proof context dynamically
wenzelm@62017
  2651
(e.g. to update cases) and indicate explicit Seq.Error results. Former
wenzelm@62017
  2652
METHOD_CASES is superseded by CONTEXT_METHOD; further combinators are
wenzelm@62017
  2653
provided in src/Pure/Isar/method.ML for convenience. INCOMPATIBILITY.
wenzelm@61885
  2654
wenzelm@60642
  2655
wenzelm@60983
  2656
*** System ***
wenzelm@60983
  2657
wenzelm@62525
  2658
* Command-line tool "isabelle console" enables print mode "ASCII".
wenzelm@61958
  2659
wenzelm@62017
  2660
* Command-line tool "isabelle update_then" expands old Isar command
wenzelm@62017
  2661
conflations:
wenzelm@62017
  2662
wenzelm@62017
  2663
    hence  ~>  then have
wenzelm@62017
  2664
    thus   ~>  then show
wenzelm@62017
  2665
wenzelm@62017
  2666
This syntax is more orthogonal and improves readability and
wenzelm@62017
  2667
maintainability of proofs.
wenzelm@62017
  2668
wenzelm@61602
  2669
* Global session timeout is multiplied by timeout_scale factor. This
wenzelm@61602
  2670
allows to adjust large-scale tests (e.g. AFP) to overall hardware
wenzelm@61602
  2671
performance.
wenzelm@61602
  2672
wenzelm@61174
  2673
* Property values in etc/symbols may contain spaces, if written with the
wenzelm@62671
  2674
replacement character "␣" (Unicode point 0x2324). For example:
wenzelm@62671
  2675
wenzelm@62671
  2676
    \<star>  code: 0x0022c6  group: operator  font: Deja␣Vu␣Sans␣Mono
wenzelm@61174
  2677
wenzelm@60995
  2678
* Java runtime environment for x86_64-windows allows to use larger heap
wenzelm@60995
  2679
space.
wenzelm@60995
  2680
wenzelm@61135
  2681
* Java runtime options are determined separately for 32bit vs. 64bit
wenzelm@61135
  2682
platforms as follows.
wenzelm@61135
  2683
wenzelm@61135
  2684
  - Isabelle desktop application: platform-specific files that are
wenzelm@61135
  2685
    associated with the main app bundle
wenzelm@61135
  2686
wenzelm@61135
  2687
  - isabelle jedit: settings
wenzelm@61135
  2688
    JEDIT_JAVA_SYSTEM_OPTIONS
wenzelm@61135
  2689
    JEDIT_JAVA_OPTIONS32 vs. JEDIT_JAVA_OPTIONS64
wenzelm@61135
  2690
wenzelm@61135
  2691
  - isabelle build: settings
wenzelm@61135
  2692
    ISABELLE_BUILD_JAVA_OPTIONS32 vs. ISABELLE_BUILD_JAVA_OPTIONS64
wenzelm@61135
  2693
wenzelm@61294
  2694
* Bash shell function "jvmpath" has been renamed to "platform_path": it
wenzelm@61294
  2695
is relevant both for Poly/ML and JVM processes.
wenzelm@61294
  2696
wenzelm@62017
  2697
* Poly/ML default platform architecture may be changed from 32bit to
wenzelm@62205
  2698
64bit via system option ML_system_64. A system restart (and rebuild) is
wenzelm@62205
  2699
required after change.
wenzelm@62017
  2700
wenzelm@62017
  2701
* Poly/ML 5.6 runs natively on x86-windows and x86_64-windows, which
wenzelm@62017
  2702
both allow larger heap space than former x86-cygwin.
wenzelm@62017
  2703
wenzelm@62157
  2704
* Heap images are 10-15% smaller due to less wasteful persistent theory
wenzelm@62157
  2705
content (using ML type theory_id instead of theory);
wenzelm@62157
  2706
wenzelm@60983
  2707
wenzelm@60479
  2708
wenzelm@60009
  2709
New in Isabelle2015 (May 2015)
wenzelm@60009
  2710
------------------------------
wenzelm@57695
  2711
wenzelm@57941
  2712
*** General ***
wenzelm@57941
  2713
wenzelm@59939
  2714
* Local theory specification commands may have a 'private' or
wenzelm@59990
  2715
'qualified' modifier to restrict name space accesses to the local scope,
wenzelm@59939
  2716
as provided by some "context begin ... end" block. For example:
wenzelm@59926
  2717
wenzelm@59926
  2718
  context
wenzelm@59926
  2719
  begin
wenzelm@59926
  2720
wenzelm@59926
  2721
  private definition ...
wenzelm@59926
  2722
  private lemma ...
wenzelm@59926
  2723
wenzelm@59990
  2724
  qualified definition ...
wenzelm@59990
  2725
  qualified lemma ...
wenzelm@59990
  2726
wenzelm@59926
  2727
  lemma ...
wenzelm@59926
  2728
  theorem ...
wenzelm@59926
  2729
wenzelm@59926
  2730
  end
wenzelm@59926
  2731
wenzelm@59901
  2732
* Command 'experiment' opens an anonymous locale context with private
wenzelm@59901
  2733
naming policy.
wenzelm@59901
  2734
wenzelm@59951
  2735
* Command 'notepad' requires proper nesting of begin/end and its proof
wenzelm@59951
  2736
structure in the body: 'oops' is no longer supported here. Minor
wenzelm@59951
  2737
INCOMPATIBILITY, use 'sorry' instead.
wenzelm@59951
  2738
wenzelm@59951