NEWS
author haftmann
Fri Feb 15 08:31:31 2013 +0100 (2013-02-15)
changeset 51143 0a2371e7ced3
parent 51137 077456580eca
child 51168 35d00ce58626
permissions -rw-r--r--
two target language numeral types: integer and natural, as replacement for code_numeral;
former theory HOL/Library/Code_Numeral_Types replaces HOL/Code_Numeral;
refined stack of theories implementing int and/or nat by target language numerals;
reduced number of target language numeral types to exactly one
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** HOL ***
     8 
     9 * Theory "RealVector" and "Limits": Introduce type class
    10 (lin)order_topology. Allows to generalize theorems about limits and
    11 order. Instances are reals and extended reals.
    12 
    13 *** HOL ***
    14 
    15 * Consolidation of library theories on product orders:
    16 
    17     Product_Lattice ~> Product_Order -- pointwise order on products
    18     Product_ord ~> Product_Lexorder -- lexicographic order on products
    19 
    20 INCOMPATIBILITY.
    21 
    22 * Sledgehammer:
    23 
    24   - Renamed option:
    25       isar_shrink ~> isar_compress
    26 
    27 
    28 New in Isabelle2013 (February 2013)
    29 -----------------------------------
    30 
    31 *** General ***
    32 
    33 * Theorem status about oracles and unfinished/failed future proofs is
    34 no longer printed by default, since it is incompatible with
    35 incremental / parallel checking of the persistent document model.  ML
    36 function Thm.peek_status may be used to inspect a snapshot of the
    37 ongoing evaluation process.  Note that in batch mode --- notably
    38 isabelle build --- the system ensures that future proofs of all
    39 accessible theorems in the theory context are finished (as before).
    40 
    41 * Configuration option show_markup controls direct inlining of markup
    42 into the printed representation of formal entities --- notably type
    43 and sort constraints.  This enables Prover IDE users to retrieve that
    44 information via tooltips in the output window, for example.
    45 
    46 * Command 'ML_file' evaluates ML text from a file directly within the
    47 theory, without any predeclaration via 'uses' in the theory header.
    48 
    49 * Old command 'use' command and corresponding keyword 'uses' in the
    50 theory header are legacy features and will be discontinued soon.
    51 Tools that load their additional source files may imitate the
    52 'ML_file' implementation, such that the system can take care of
    53 dependencies properly.
    54 
    55 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    56 is called fastforce / fast_force_tac already since Isabelle2011-1.
    57 
    58 * Updated and extended "isar-ref" and "implementation" manual, reduced
    59 remaining material in old "ref" manual.
    60 
    61 * Improved support for auxiliary contexts that indicate block structure
    62 for specifications.  Nesting of "context fixes ... context assumes ..."
    63 and "class ... context ...".
    64 
    65 * Attribute "consumes" allows a negative value as well, which is
    66 interpreted relatively to the total number of premises of the rule in
    67 the target context.  This form of declaration is stable when exported
    68 from a nested 'context' with additional assumptions.  It is the
    69 preferred form for definitional packages, notably cases/rules produced
    70 in HOL/inductive and HOL/function.
    71 
    72 * More informative error messages for Isar proof commands involving
    73 lazy enumerations (method applications etc.).
    74 
    75 * Refined 'help' command to retrieve outer syntax commands according
    76 to name patterns (with clickable results).
    77 
    78 
    79 *** Prover IDE -- Isabelle/Scala/jEdit ***
    80 
    81 * Parallel terminal proofs ('by') are enabled by default, likewise
    82 proofs that are built into packages like 'datatype', 'function'.  This
    83 allows to "run ahead" checking the theory specifications on the
    84 surface, while the prover is still crunching on internal
    85 justifications.  Unfinished / cancelled proofs are restarted as
    86 required to complete full proof checking eventually.
    87 
    88 * Improved output panel with tooltips, hyperlinks etc. based on the
    89 same Rich_Text_Area as regular Isabelle/jEdit buffers.  Activation of
    90 tooltips leads to some window that supports the same recursively,
    91 which can lead to stacks of tooltips as the semantic document content
    92 is explored.  ESCAPE closes the whole stack, individual windows may be
    93 closed separately, or detached to become independent jEdit dockables.
    94 
    95 * Improved support for commands that produce graph output: the text
    96 message contains a clickable area to open a new instance of the graph
    97 browser on demand.
    98 
    99 * More robust incremental parsing of outer syntax (partial comments,
   100 malformed symbols).  Changing the balance of open/close quotes and
   101 comment delimiters works more conveniently with unfinished situations
   102 that frequently occur in user interaction.
   103 
   104 * More efficient painting and improved reactivity when editing large
   105 files.  More scalable management of formal document content.
   106 
   107 * Smarter handling of tracing messages: prover process pauses after
   108 certain number of messages per command transaction, with some user
   109 dialog to stop or continue.  This avoids swamping the front-end with
   110 potentially infinite message streams.
   111 
   112 * More plugin options and preferences, based on Isabelle/Scala.  The
   113 jEdit plugin option panel provides access to some Isabelle/Scala
   114 options, including tuning parameters for editor reactivity and color
   115 schemes.
   116 
   117 * Dockable window "Symbols" provides some editing support for Isabelle
   118 symbols.
   119 
   120 * Dockable window "Monitor" shows ML runtime statistics.  Note that
   121 continuous display of the chart slows down the system.
   122 
   123 * Improved editing support for control styles: subscript, superscript,
   124 bold, reset of style -- operating on single symbols or text
   125 selections.  Cf. keyboard shortcuts C+e DOWN/UP/RIGHT/LEFT.
   126 
   127 * Actions isabelle.increase-font-size and isabelle.decrease-font-size
   128 adjust the main text area font size, and its derivatives for output,
   129 tooltips etc.  Cf. keyboard shortcuts C-PLUS and C-MINUS, which often
   130 need to be adapted to local keyboard layouts.
   131 
   132 * More reactive completion popup by default: use \t (TAB) instead of
   133 \n (NEWLINE) to minimize intrusion into regular flow of editing.  See
   134 also "Plugin Options / SideKick / General / Code Completion Options".
   135 
   136 * Implicit check and build dialog of the specified logic session
   137 image.  For example, HOL, HOLCF, HOL-Nominal can be produced on
   138 demand, without bundling big platform-dependent heap images in the
   139 Isabelle distribution.
   140 
   141 * Uniform Java 7 platform on Linux, Mac OS X, Windows: recent updates
   142 from Oracle provide better multi-platform experience.  This version is
   143 now bundled exclusively with Isabelle.
   144 
   145 
   146 *** Pure ***
   147 
   148 * Code generation for Haskell: restrict unqualified imports from
   149 Haskell Prelude to a small set of fundamental operations.
   150 
   151 * Command 'export_code': relative file names are interpreted
   152 relatively to master directory of current theory rather than the
   153 rather arbitrary current working directory.  INCOMPATIBILITY.
   154 
   155 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
   156 use regular rule composition via "OF" / "THEN", or explicit proof
   157 structure instead.  Note that Isabelle/ML provides a variety of
   158 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
   159 with some care where this is really required.
   160 
   161 * Command 'typ' supports an additional variant with explicit sort
   162 constraint, to infer and check the most general type conforming to a
   163 given sort.  Example (in HOL):
   164 
   165   typ "_ * _ * bool * unit" :: finite
   166 
   167 * Command 'locale_deps' visualizes all locales and their relations as
   168 a Hasse diagram.
   169 
   170 
   171 *** HOL ***
   172 
   173 * Numeric types mapped by default to target language numerals:
   174 natural (replaces former code_numeral) and integer (replaces
   175 former code_int).  Conversions are available as integer_of_natural /
   176 natural_of_integer / integer_of_nat / nat_of_integer (in HOL) and
   177 Code_Numeral.integer_of_natural / Code_Numeral.natural_of_integer (in ML).
   178 INCOMPATIBILITY.
   179 
   180 * Discontinued theories Code_Integer and Efficient_Nat by a more
   181 fine-grain stack of theories Code_Target_Int, Code_Binary_Nat,
   182 Code_Target_Nat and Code_Target_Numeral.  See the tutorial on
   183 code generation for details.  INCOMPATIBILITY.
   184 
   185 * Sledgehammer:
   186 
   187   - Added MaSh relevance filter based on machine-learning; see the
   188     Sledgehammer manual for details.
   189   - Polished Isar proofs generated with "isar_proofs" option.
   190   - Rationalized type encodings ("type_enc" option).
   191   - Renamed "kill_provers" subcommand to "kill_all".
   192   - Renamed options:
   193       isar_proof ~> isar_proofs
   194       isar_shrink_factor ~> isar_shrink
   195       max_relevant ~> max_facts
   196       relevance_thresholds ~> fact_thresholds
   197 
   198 * Quickcheck: added an optimisation for equality premises.  It is
   199 switched on by default, and can be switched off by setting the
   200 configuration quickcheck_optimise_equality to false.
   201 
   202 * Quotient: only one quotient can be defined by quotient_type
   203 INCOMPATIBILITY.
   204 
   205 * Lifting:
   206   - generation of an abstraction function equation in lift_definition
   207   - quot_del attribute
   208   - renamed no_abs_code -> no_code (INCOMPATIBILITY.)
   209 
   210 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   211 expressions.
   212 
   213 * Preprocessing of the code generator rewrites set comprehensions into
   214 pointfree expressions.
   215 
   216 * The SMT solver Z3 has now by default a restricted set of directly
   217 supported features. For the full set of features (div/mod, nonlinear
   218 arithmetic, datatypes/records) with potential proof reconstruction
   219 failures, enable the configuration option "z3_with_extensions".  Minor
   220 INCOMPATIBILITY.
   221 
   222 * Simplified 'typedef' specifications: historical options for implicit
   223 set definition and alternative name have been discontinued.  The
   224 former behavior of "typedef (open) t = A" is now the default, but
   225 written just "typedef t = A".  INCOMPATIBILITY, need to adapt theories
   226 accordingly.
   227 
   228 * Removed constant "chars"; prefer "Enum.enum" on type "char"
   229 directly.  INCOMPATIBILITY.
   230 
   231 * Moved operation product, sublists and n_lists from theory Enum to
   232 List.  INCOMPATIBILITY.
   233 
   234 * Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
   235 
   236 * Class "comm_monoid_diff" formalises properties of bounded
   237 subtraction, with natural numbers and multisets as typical instances.
   238 
   239 * Added combinator "Option.these" with type "'a option set => 'a set".
   240 
   241 * Theory "Transitive_Closure": renamed lemmas
   242 
   243   reflcl_tranclp -> reflclp_tranclp
   244   rtranclp_reflcl -> rtranclp_reflclp
   245 
   246 INCOMPATIBILITY.
   247 
   248 * Theory "Rings": renamed lemmas (in class semiring)
   249 
   250   left_distrib ~> distrib_right
   251   right_distrib ~> distrib_left
   252 
   253 INCOMPATIBILITY.
   254 
   255 * Generalized the definition of limits:
   256 
   257   - Introduced the predicate filterlim (LIM x F. f x :> G) which
   258     expresses that when the input values x converge to F then the
   259     output f x converges to G.
   260 
   261   - Added filters for convergence to positive (at_top) and negative
   262     infinity (at_bot).
   263 
   264   - Moved infinity in the norm (at_infinity) from
   265     Multivariate_Analysis to Complex_Main.
   266 
   267   - Removed real_tendsto_inf, it is superseded by "LIM x F. f x :>
   268     at_top".
   269 
   270 INCOMPATIBILITY.
   271 
   272 * Theory "Library/Option_ord" provides instantiation of option type to
   273 lattice type classes.
   274 
   275 * Theory "Library/Multiset": renamed
   276 
   277     constant fold_mset ~> Multiset.fold
   278     fact fold_mset_commute ~> fold_mset_comm
   279 
   280 INCOMPATIBILITY.
   281 
   282 * Renamed theory Library/List_Prefix to Library/Sublist, with related
   283 changes as follows.
   284 
   285   - Renamed constants (and related lemmas)
   286 
   287       prefix ~> prefixeq
   288       strict_prefix ~> prefix
   289 
   290   - Replaced constant "postfix" by "suffixeq" with swapped argument
   291     order (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped
   292     old infix syntax "xs >>= ys"; use "suffixeq ys xs" instead.
   293     Renamed lemmas accordingly.
   294 
   295   - Added constant "list_hembeq" for homeomorphic embedding on
   296     lists. Added abbreviation "sublisteq" for special case
   297     "list_hembeq (op =)".
   298 
   299   - Theory Library/Sublist no longer provides "order" and "bot" type
   300     class instances for the prefix order (merely corresponding locale
   301     interpretations). The type class instances are now in theory
   302     Library/Prefix_Order.
   303 
   304   - The sublist relation of theory Library/Sublist_Order is now based
   305     on "Sublist.sublisteq".  Renamed lemmas accordingly:
   306 
   307       le_list_append_le_same_iff ~> Sublist.sublisteq_append_le_same_iff
   308       le_list_append_mono ~> Sublist.list_hembeq_append_mono
   309       le_list_below_empty ~> Sublist.list_hembeq_Nil, Sublist.list_hembeq_Nil2
   310       le_list_Cons_EX ~> Sublist.list_hembeq_ConsD
   311       le_list_drop_Cons2 ~> Sublist.sublisteq_Cons2'
   312       le_list_drop_Cons_neq ~> Sublist.sublisteq_Cons2_neq
   313       le_list_drop_Cons ~> Sublist.sublisteq_Cons'
   314       le_list_drop_many ~> Sublist.sublisteq_drop_many
   315       le_list_filter_left ~> Sublist.sublisteq_filter_left
   316       le_list_rev_drop_many ~> Sublist.sublisteq_rev_drop_many
   317       le_list_rev_take_iff ~> Sublist.sublisteq_append
   318       le_list_same_length ~> Sublist.sublisteq_same_length
   319       le_list_take_many_iff ~> Sublist.sublisteq_append'
   320       less_eq_list.drop ~> less_eq_list_drop
   321       less_eq_list.induct ~> less_eq_list_induct
   322       not_le_list_length ~> Sublist.not_sublisteq_length
   323 
   324 INCOMPATIBILITY.
   325 
   326 * New theory Library/Countable_Set.
   327 
   328 * Theory Library/Debug and Library/Parallel provide debugging and
   329 parallel execution for code generated towards Isabelle/ML.
   330 
   331 * Theory Library/FuncSet: Extended support for Pi and extensional and
   332 introduce the extensional dependent function space "PiE". Replaced
   333 extensional_funcset by an abbreviation, and renamed lemmas from
   334 extensional_funcset to PiE as follows:
   335 
   336   extensional_empty  ~>  PiE_empty
   337   extensional_funcset_empty_domain  ~>  PiE_empty_domain
   338   extensional_funcset_empty_range  ~>  PiE_empty_range
   339   extensional_funcset_arb  ~>  PiE_arb
   340   extensional_funcset_mem  ~>  PiE_mem
   341   extensional_funcset_extend_domainI  ~>  PiE_fun_upd
   342   extensional_funcset_restrict_domain  ~>  fun_upd_in_PiE
   343   extensional_funcset_extend_domain_eq  ~>  PiE_insert_eq
   344   card_extensional_funcset  ~>  card_PiE
   345   finite_extensional_funcset  ~>  finite_PiE
   346 
   347 INCOMPATIBILITY.
   348 
   349 * Theory Library/FinFun: theory of almost everywhere constant
   350 functions (supersedes the AFP entry "Code Generation for Functions as
   351 Data").
   352 
   353 * Theory Library/Phantom: generic phantom type to make a type
   354 parameter appear in a constant's type.  This alternative to adding
   355 TYPE('a) as another parameter avoids unnecessary closures in generated
   356 code.
   357 
   358 * Theory Library/RBT_Impl: efficient construction of red-black trees
   359 from sorted associative lists. Merging two trees with rbt_union may
   360 return a structurally different tree than before.  Potential
   361 INCOMPATIBILITY.
   362 
   363 * Theory Library/IArray: immutable arrays with code generation.
   364 
   365 * Theory Library/Finite_Lattice: theory of finite lattices.
   366 
   367 * HOL/Multivariate_Analysis: replaced
   368 
   369   "basis :: 'a::euclidean_space => nat => real"
   370   "\<Chi>\<Chi> :: (nat => real) => 'a::euclidean_space"
   371 
   372 on euclidean spaces by using the inner product "_ \<bullet> _" with
   373 vectors from the Basis set: "\<Chi>\<Chi> i. f i" is superseded by
   374 "SUM i : Basis. f i * r i".
   375 
   376   With this change the following constants are also changed or removed:
   377 
   378     DIM('a) :: nat  ~>  card (Basis :: 'a set)   (is an abbreviation)
   379     a $$ i  ~>  inner a i  (where i : Basis)
   380     cart_base i  removed
   381     \<pi>, \<pi>'  removed
   382 
   383   Theorems about these constants where removed.
   384 
   385   Renamed lemmas:
   386 
   387     component_le_norm  ~>  Basis_le_norm
   388     euclidean_eq  ~>  euclidean_eq_iff
   389     differential_zero_maxmin_component  ~>  differential_zero_maxmin_cart
   390     euclidean_simps  ~>  inner_simps
   391     independent_basis  ~>  independent_Basis
   392     span_basis  ~>  span_Basis
   393     in_span_basis  ~>  in_span_Basis
   394     norm_bound_component_le  ~>  norm_boound_Basis_le
   395     norm_bound_component_lt  ~>  norm_boound_Basis_lt
   396     component_le_infnorm  ~>  Basis_le_infnorm
   397 
   398 INCOMPATIBILITY.
   399 
   400 * HOL/Probability:
   401 
   402   - Added simproc "measurable" to automatically prove measurability.
   403 
   404   - Added induction rules for sigma sets with disjoint union
   405     (sigma_sets_induct_disjoint) and for Borel-measurable functions
   406     (borel_measurable_induct).
   407 
   408   - Added the Daniell-Kolmogorov theorem (the existence the limit of a
   409     projective family).
   410 
   411 * HOL/Cardinals: Theories of ordinals and cardinals (supersedes the
   412 AFP entry "Ordinals_and_Cardinals").
   413 
   414 * HOL/BNF: New (co)datatype package based on bounded natural functors
   415 with support for mixed, nested recursion and interesting non-free
   416 datatypes.
   417 
   418 * HOL/Finite_Set and Relation: added new set and relation operations
   419 expressed by Finite_Set.fold.
   420 
   421 * New theory HOL/Library/RBT_Set: implementation of sets by red-black
   422 trees for the code generator.
   423 
   424 * HOL/Library/RBT and HOL/Library/Mapping have been converted to
   425 Lifting/Transfer.
   426 possible INCOMPATIBILITY.
   427 
   428 * HOL/Set: renamed Set.project -> Set.filter
   429 INCOMPATIBILITY.
   430 
   431 
   432 *** Document preparation ***
   433 
   434 * Dropped legacy antiquotations "term_style" and "thm_style", since
   435 styles may be given as arguments to "term" and "thm" already.
   436 Discontinued legacy styles "prem1" .. "prem19".
   437 
   438 * Default LaTeX rendering for \<euro> is now based on eurosym package,
   439 instead of slightly exotic babel/greek.
   440 
   441 * Document variant NAME may use different LaTeX entry point
   442 document/root_NAME.tex if that file exists, instead of the common
   443 document/root.tex.
   444 
   445 * Simplified custom document/build script, instead of old-style
   446 document/IsaMakefile.  Minor INCOMPATIBILITY.
   447 
   448 
   449 *** ML ***
   450 
   451 * The default limit for maximum number of worker threads is now 8,
   452 instead of 4, in correspondence to capabilities of contemporary
   453 hardware and Poly/ML runtime system.
   454 
   455 * Type Seq.results and related operations support embedded error
   456 messages within lazy enumerations, and thus allow to provide
   457 informative errors in the absence of any usable results.
   458 
   459 * Renamed Position.str_of to Position.here to emphasize that this is a
   460 formal device to inline positions into message text, but not
   461 necessarily printing visible text.
   462 
   463 
   464 *** System ***
   465 
   466 * Advanced support for Isabelle sessions and build management, see
   467 "system" manual for the chapter of that name, especially the "isabelle
   468 build" tool and its examples.  The "isabelle mkroot" tool prepares
   469 session root directories for use with "isabelle build", similar to
   470 former "isabelle mkdir" for "isabelle usedir".  Note that this affects
   471 document preparation as well.  INCOMPATIBILITY, isabelle usedir /
   472 mkdir / make are rendered obsolete.
   473 
   474 * Discontinued obsolete Isabelle/build script, it is superseded by the
   475 regular isabelle build tool.  For example:
   476 
   477   isabelle build -s -b HOL
   478 
   479 * Discontinued obsolete "isabelle makeall".
   480 
   481 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   482 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   483 provides some traditional targets that invoke "isabelle build".  Note
   484 that this is inefficient!  Applications of Isabelle/HOL involving
   485 "isabelle make" should be upgraded to use "isabelle build" directly.
   486 
   487 * The "isabelle options" tool prints Isabelle system options, as
   488 required for "isabelle build", for example.
   489 
   490 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   491 Minor INCOMPATIBILITY in command-line options.
   492 
   493 * The "isabelle install" tool has now a simpler command-line.  Minor
   494 INCOMPATIBILITY.
   495 
   496 * The "isabelle components" tool helps to resolve add-on components
   497 that are not bundled, or referenced from a bare-bones repository
   498 version of Isabelle.
   499 
   500 * Settings variable ISABELLE_PLATFORM_FAMILY refers to the general
   501 platform family: "linux", "macos", "windows".
   502 
   503 * The ML system is configured as regular component, and no longer
   504 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   505 for home-made settings.
   506 
   507 * Improved ML runtime statistics (heap, threads, future tasks etc.).
   508 
   509 * Discontinued support for Poly/ML 5.2.1, which was the last version
   510 without exception positions and advanced ML compiler/toplevel
   511 configuration.
   512 
   513 * Discontinued special treatment of Proof General -- no longer guess
   514 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   515 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   516 settings manually, or use a Proof General version that has been
   517 bundled as Isabelle component.
   518 
   519 
   520 
   521 New in Isabelle2012 (May 2012)
   522 ------------------------------
   523 
   524 *** General ***
   525 
   526 * Prover IDE (PIDE) improvements:
   527 
   528   - more robust Sledgehammer integration (as before the sledgehammer
   529     command-line needs to be typed into the source buffer)
   530   - markup for bound variables
   531   - markup for types of term variables (displayed as tooltips)
   532   - support for user-defined Isar commands within the running session
   533   - improved support for Unicode outside original 16bit range
   534     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   535 
   536 * Forward declaration of outer syntax keywords within the theory
   537 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   538 commands to be used in the same theory where defined.
   539 
   540 * Auxiliary contexts indicate block structure for specifications with
   541 additional parameters and assumptions.  Such unnamed contexts may be
   542 nested within other targets, like 'theory', 'locale', 'class',
   543 'instantiation' etc.  Results from the local context are generalized
   544 accordingly and applied to the enclosing target context.  Example:
   545 
   546   context
   547     fixes x y z :: 'a
   548     assumes xy: "x = y" and yz: "y = z"
   549   begin
   550 
   551   lemma my_trans: "x = z" using xy yz by simp
   552 
   553   end
   554 
   555   thm my_trans
   556 
   557 The most basic application is to factor-out context elements of
   558 several fixes/assumes/shows theorem statements, e.g. see
   559 ~~/src/HOL/Isar_Examples/Group_Context.thy
   560 
   561 Any other local theory specification element works within the "context
   562 ... begin ... end" block as well.
   563 
   564 * Bundled declarations associate attributed fact expressions with a
   565 given name in the context.  These may be later included in other
   566 contexts.  This allows to manage context extensions casually, without
   567 the logical dependencies of locales and locale interpretation.  See
   568 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   569 
   570 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   571 declaration, and results are standardized before being stored.  Thus
   572 old-style "standard" after instantiation or composition of facts
   573 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   574 indices of schematic variables.
   575 
   576 * Rule attributes in local theory declarations (e.g. locale or class)
   577 are now statically evaluated: the resulting theorem is stored instead
   578 of the original expression.  INCOMPATIBILITY in rare situations, where
   579 the historic accident of dynamic re-evaluation in interpretations
   580 etc. was exploited.
   581 
   582 * New tutorial "Programming and Proving in Isabelle/HOL"
   583 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   584 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   585 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   586 Logic" as the recommended beginners tutorial, but does not cover all
   587 of the material of that old tutorial.
   588 
   589 * Updated and extended reference manuals: "isar-ref",
   590 "implementation", "system"; reduced remaining material in old "ref"
   591 manual.
   592 
   593 
   594 *** Pure ***
   595 
   596 * Command 'definition' no longer exports the foundational "raw_def"
   597 into the user context.  Minor INCOMPATIBILITY, may use the regular
   598 "def" result with attribute "abs_def" to imitate the old version.
   599 
   600 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   601 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   602 expand it.  This also works for object-logic equality.  (Formerly
   603 undocumented feature.)
   604 
   605 * Sort constraints are now propagated in simultaneous statements, just
   606 like type constraints.  INCOMPATIBILITY in rare situations, where
   607 distinct sorts used to be assigned accidentally.  For example:
   608 
   609   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   610 
   611   lemma "P (x::'a)" and "Q (y::'a::bar)"
   612     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   613 
   614 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   615 tolerant against multiple unifiers, as long as the final result is
   616 unique.  (As before, rules are composed in canonical right-to-left
   617 order to accommodate newly introduced premises.)
   618 
   619 * Renamed some inner syntax categories:
   620 
   621     num ~> num_token
   622     xnum ~> xnum_token
   623     xstr ~> str_token
   624 
   625 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   626 "num_position" etc. are mainly used instead (which also include
   627 position information via constraints).
   628 
   629 * Simplified configuration options for syntax ambiguity: see
   630 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   631 manual.  Minor INCOMPATIBILITY.
   632 
   633 * Discontinued configuration option "syntax_positions": atomic terms
   634 in parse trees are always annotated by position constraints.
   635 
   636 * Old code generator for SML and its commands 'code_module',
   637 'code_library', 'consts_code', 'types_code' have been discontinued.
   638 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   639 
   640 * Redundant attribute "code_inline" has been discontinued. Use
   641 "code_unfold" instead.  INCOMPATIBILITY.
   642 
   643 * Dropped attribute "code_unfold_post" in favor of the its dual
   644 "code_abbrev", which yields a common pattern in definitions like
   645 
   646   definition [code_abbrev]: "f = t"
   647 
   648 INCOMPATIBILITY.
   649 
   650 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   651 instead.  INCOMPATIBILITY.
   652 
   653 * Discontinued old "prems" fact, which used to refer to the accidental
   654 collection of foundational premises in the context (already marked as
   655 legacy since Isabelle2011).
   656 
   657 
   658 *** HOL ***
   659 
   660 * Type 'a set is now a proper type constructor (just as before
   661 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   662 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   663 sets separate, it is often sufficient to rephrase some set S that has
   664 been accidentally used as predicates by "%x. x : S", and some
   665 predicate P that has been accidentally used as set by "{x. P x}".
   666 Corresponding proofs in a first step should be pruned from any
   667 tinkering with former theorems mem_def and Collect_def as far as
   668 possible.
   669 
   670 For developments which deliberately mix predicates and sets, a
   671 planning step is necessary to determine what should become a predicate
   672 and what a set.  It can be helpful to carry out that step in
   673 Isabelle2011-1 before jumping right into the current release.
   674 
   675 * Code generation by default implements sets as container type rather
   676 than predicates.  INCOMPATIBILITY.
   677 
   678 * New type synonym 'a rel = ('a * 'a) set
   679 
   680 * The representation of numerals has changed.  Datatype "num"
   681 represents strictly positive binary numerals, along with functions
   682 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   683 positive and negated numeric literals, respectively.  See also
   684 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   685 user theories may require adaptations as follows:
   686 
   687   - Theorems with number_ring or number_semiring constraints: These
   688     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   689 
   690   - Theories defining numeric types: Remove number, number_semiring,
   691     and number_ring instances. Defer all theorems about numerals until
   692     after classes one and semigroup_add have been instantiated.
   693 
   694   - Numeral-only simp rules: Replace each rule having a "number_of v"
   695     pattern with two copies, one for numeral and one for neg_numeral.
   696 
   697   - Theorems about subclasses of semiring_1 or ring_1: These classes
   698     automatically support numerals now, so more simp rules and
   699     simprocs may now apply within the proof.
   700 
   701   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   702     Redefine using other integer operations.
   703 
   704 * Transfer: New package intended to generalize the existing
   705 "descending" method and related theorem attributes from the Quotient
   706 package.  (Not all functionality is implemented yet, but future
   707 development will focus on Transfer as an eventual replacement for the
   708 corresponding parts of the Quotient package.)
   709 
   710   - transfer_rule attribute: Maintains a collection of transfer rules,
   711     which relate constants at two different types. Transfer rules may
   712     relate different type instances of the same polymorphic constant,
   713     or they may relate an operation on a raw type to a corresponding
   714     operation on an abstract type (quotient or subtype). For example:
   715 
   716     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   717     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   718 
   719   - transfer method: Replaces a subgoal on abstract types with an
   720     equivalent subgoal on the corresponding raw types. Constants are
   721     replaced with corresponding ones according to the transfer rules.
   722     Goals are generalized over all free variables by default; this is
   723     necessary for variables whose types change, but can be overridden
   724     for specific variables with e.g. "transfer fixing: x y z".  The
   725     variant transfer' method allows replacing a subgoal with one that
   726     is logically stronger (rather than equivalent).
   727 
   728   - relator_eq attribute: Collects identity laws for relators of
   729     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   730     transfer method uses these lemmas to infer transfer rules for
   731     non-polymorphic constants on the fly.
   732 
   733   - transfer_prover method: Assists with proving a transfer rule for a
   734     new constant, provided the constant is defined in terms of other
   735     constants that already have transfer rules. It should be applied
   736     after unfolding the constant definitions.
   737 
   738   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   739     from type nat to type int.
   740 
   741 * Lifting: New package intended to generalize the quotient_definition
   742 facility of the Quotient package; designed to work with Transfer.
   743 
   744   - lift_definition command: Defines operations on an abstract type in
   745     terms of a corresponding operation on a representation
   746     type.  Example syntax:
   747 
   748     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   749       is List.insert
   750 
   751     Users must discharge a respectfulness proof obligation when each
   752     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   753     the proof is discharged automatically.) The obligation is
   754     presented in a user-friendly, readable form; a respectfulness
   755     theorem in the standard format and a transfer rule are generated
   756     by the package.
   757 
   758   - Integration with code_abstype: For typedefs (e.g. subtypes
   759     corresponding to a datatype invariant, such as dlist),
   760     lift_definition generates a code certificate theorem and sets up
   761     code generation for each constant.
   762 
   763   - setup_lifting command: Sets up the Lifting package to work with a
   764     user-defined type. The user must provide either a quotient theorem
   765     or a type_definition theorem.  The package configures transfer
   766     rules for equality and quantifiers on the type, and sets up the
   767     lift_definition command to work with the type.
   768 
   769   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   770     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   771     Word/Word.thy and Library/Float.thy.
   772 
   773 * Quotient package:
   774 
   775   - The 'quotient_type' command now supports a 'morphisms' option with
   776     rep and abs functions, similar to typedef.
   777 
   778   - 'quotient_type' sets up new types to work with the Lifting and
   779     Transfer packages, as with 'setup_lifting'.
   780 
   781   - The 'quotient_definition' command now requires the user to prove a
   782     respectfulness property at the point where the constant is
   783     defined, similar to lift_definition; INCOMPATIBILITY.
   784 
   785   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   786     accordingly, INCOMPATIBILITY.
   787 
   788 * New diagnostic command 'find_unused_assms' to find potentially
   789 superfluous assumptions in theorems using Quickcheck.
   790 
   791 * Quickcheck:
   792 
   793   - Quickcheck returns variable assignments as counterexamples, which
   794     allows to reveal the underspecification of functions under test.
   795     For example, refuting "hd xs = x", it presents the variable
   796     assignment xs = [] and x = a1 as a counterexample, assuming that
   797     any property is false whenever "hd []" occurs in it.
   798 
   799     These counterexample are marked as potentially spurious, as
   800     Quickcheck also returns "xs = []" as a counterexample to the
   801     obvious theorem "hd xs = hd xs".
   802 
   803     After finding a potentially spurious counterexample, Quickcheck
   804     continues searching for genuine ones.
   805 
   806     By default, Quickcheck shows potentially spurious and genuine
   807     counterexamples. The option "genuine_only" sets quickcheck to only
   808     show genuine counterexamples.
   809 
   810   - The command 'quickcheck_generator' creates random and exhaustive
   811     value generators for a given type and operations.
   812 
   813     It generates values by using the operations as if they were
   814     constructors of that type.
   815 
   816   - Support for multisets.
   817 
   818   - Added "use_subtype" options.
   819 
   820   - Added "quickcheck_locale" configuration to specify how to process
   821     conjectures in a locale context.
   822 
   823 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   824 and affecting 'rat' and 'real'.
   825 
   826 * Sledgehammer:
   827   - Integrated more tightly with SPASS, as described in the ITP 2012
   828     paper "More SPASS with Isabelle".
   829   - Made it try "smt" as a fallback if "metis" fails or times out.
   830   - Added support for the following provers: Alt-Ergo (via Why3 and
   831     TFF1), iProver, iProver-Eq.
   832   - Sped up the minimizer.
   833   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   834   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   835   - Renamed "sound" option to "strict".
   836 
   837 * Metis: Added possibility to specify lambda translations scheme as a
   838 parenthesized argument (e.g., "by (metis (lifting) ...)").
   839 
   840 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   841 
   842 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   843 
   844 * New "case_product" attribute to generate a case rule doing multiple
   845 case distinctions at the same time.  E.g.
   846 
   847   list.exhaust [case_product nat.exhaust]
   848 
   849 produces a rule which can be used to perform case distinction on both
   850 a list and a nat.
   851 
   852 * New "eventually_elim" method as a generalized variant of the
   853 eventually_elim* rules.  Supports structured proofs.
   854 
   855 * Typedef with implicit set definition is considered legacy.  Use
   856 "typedef (open)" form instead, which will eventually become the
   857 default.
   858 
   859 * Record: code generation can be switched off manually with
   860 
   861   declare [[record_coden = false]]  -- "default true"
   862 
   863 * Datatype: type parameters allow explicit sort constraints.
   864 
   865 * Concrete syntax for case expressions includes constraints for source
   866 positions, and thus produces Prover IDE markup for its bindings.
   867 INCOMPATIBILITY for old-style syntax translations that augment the
   868 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   869 one_case.
   870 
   871 * Clarified attribute "mono_set": pure declaration without modifying
   872 the result of the fact expression.
   873 
   874 * More default pred/set conversions on a couple of relation operations
   875 and predicates.  Added powers of predicate relations.  Consolidation
   876 of some relation theorems:
   877 
   878   converse_def ~> converse_unfold
   879   rel_comp_def ~> relcomp_unfold
   880   symp_def ~> (modified, use symp_def and sym_def instead)
   881   transp_def ~> transp_trans
   882   Domain_def ~> Domain_unfold
   883   Range_def ~> Domain_converse [symmetric]
   884 
   885 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   886 
   887 See theory "Relation" for examples for making use of pred/set
   888 conversions by means of attributes "to_set" and "to_pred".
   889 
   890 INCOMPATIBILITY.
   891 
   892 * Renamed facts about the power operation on relations, i.e., relpow
   893 to match the constant's name:
   894 
   895   rel_pow_1 ~> relpow_1
   896   rel_pow_0_I ~> relpow_0_I
   897   rel_pow_Suc_I ~> relpow_Suc_I
   898   rel_pow_Suc_I2 ~> relpow_Suc_I2
   899   rel_pow_0_E ~> relpow_0_E
   900   rel_pow_Suc_E ~> relpow_Suc_E
   901   rel_pow_E ~> relpow_E
   902   rel_pow_Suc_D2 ~> relpow_Suc_D2
   903   rel_pow_Suc_E2 ~> relpow_Suc_E2
   904   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   905   rel_pow_E2 ~> relpow_E2
   906   rel_pow_add ~> relpow_add
   907   rel_pow_commute ~> relpow
   908   rel_pow_empty ~> relpow_empty:
   909   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   910   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   911   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   912   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   913   rel_pow_fun_conv ~> relpow_fun_conv
   914   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   915   rel_pow_finite_bounded ~> relpow_finite_bounded
   916   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   917   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   918   single_valued_rel_pow ~> single_valued_relpow
   919 
   920 INCOMPATIBILITY.
   921 
   922 * Theory Relation: Consolidated constant name for relation composition
   923 and corresponding theorem names:
   924 
   925   - Renamed constant rel_comp to relcomp.
   926 
   927   - Dropped abbreviation pred_comp. Use relcompp instead.
   928 
   929   - Renamed theorems:
   930 
   931     rel_compI ~> relcompI
   932     rel_compEpair ~> relcompEpair
   933     rel_compE ~> relcompE
   934     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   935     rel_comp_empty1 ~> relcomp_empty1
   936     rel_comp_mono ~> relcomp_mono
   937     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   938     rel_comp_distrib ~> relcomp_distrib
   939     rel_comp_distrib2 ~> relcomp_distrib2
   940     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   941     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   942     single_valued_rel_comp ~> single_valued_relcomp
   943     rel_comp_def ~> relcomp_unfold
   944     converse_rel_comp ~> converse_relcomp
   945     pred_compI ~> relcomppI
   946     pred_compE ~> relcomppE
   947     pred_comp_bot1 ~> relcompp_bot1
   948     pred_comp_bot2 ~> relcompp_bot2
   949     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   950     pred_comp_mono ~> relcompp_mono
   951     pred_comp_distrib ~> relcompp_distrib
   952     pred_comp_distrib2 ~> relcompp_distrib2
   953     converse_pred_comp ~> converse_relcompp
   954 
   955     finite_rel_comp ~> finite_relcomp
   956 
   957     set_rel_comp ~> set_relcomp
   958 
   959 INCOMPATIBILITY.
   960 
   961 * Theory Divides: Discontinued redundant theorems about div and mod.
   962 INCOMPATIBILITY, use the corresponding generic theorems instead.
   963 
   964   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   965   zdiv_self ~> div_self
   966   zmod_self ~> mod_self
   967   zdiv_zero ~> div_0
   968   zmod_zero ~> mod_0
   969   zdiv_zmod_equality ~> div_mod_equality2
   970   zdiv_zmod_equality2 ~> div_mod_equality
   971   zmod_zdiv_trivial ~> mod_div_trivial
   972   zdiv_zminus_zminus ~> div_minus_minus
   973   zmod_zminus_zminus ~> mod_minus_minus
   974   zdiv_zminus2 ~> div_minus_right
   975   zmod_zminus2 ~> mod_minus_right
   976   zdiv_minus1_right ~> div_minus1_right
   977   zmod_minus1_right ~> mod_minus1_right
   978   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   979   zmod_zmult1_eq ~> mod_mult_right_eq
   980   zpower_zmod ~> power_mod
   981   zdvd_zmod ~> dvd_mod
   982   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   983   mod_mult_distrib ~> mult_mod_left
   984   mod_mult_distrib2 ~> mult_mod_right
   985 
   986 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   987 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   988 
   989 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   990 
   991 * Consolidated theorem names concerning fold combinators:
   992 
   993   inf_INFI_fold_inf ~> inf_INF_fold_inf
   994   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   995   INFI_fold_inf ~> INF_fold_inf
   996   SUPR_fold_sup ~> SUP_fold_sup
   997   union_set ~> union_set_fold
   998   minus_set ~> minus_set_fold
   999   INFI_set_fold ~> INF_set_fold
  1000   SUPR_set_fold ~> SUP_set_fold
  1001   INF_code ~> INF_set_foldr
  1002   SUP_code ~> SUP_set_foldr
  1003   foldr.simps ~> foldr.simps (in point-free formulation)
  1004   foldr_fold_rev ~> foldr_conv_fold
  1005   foldl_fold ~> foldl_conv_fold
  1006   foldr_foldr ~> foldr_conv_foldl
  1007   foldl_foldr ~> foldl_conv_foldr
  1008   fold_set_remdups ~> fold_set_fold_remdups
  1009   fold_set ~> fold_set_fold
  1010   fold1_set ~> fold1_set_fold
  1011 
  1012 INCOMPATIBILITY.
  1013 
  1014 * Dropped rarely useful theorems concerning fold combinators:
  1015 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
  1016 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
  1017 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
  1018 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
  1019 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
  1020 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
  1021 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
  1022 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
  1023 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
  1024 unfolding "foldr_conv_fold" and "foldl_conv_fold".
  1025 
  1026 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
  1027 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
  1028 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
  1029 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
  1030 lemmas over fold rather than foldr, or make use of lemmas
  1031 fold_conv_foldr and fold_rev.
  1032 
  1033 * Congruence rules Option.map_cong and Option.bind_cong for recursion
  1034 through option types.
  1035 
  1036 * "Transitive_Closure.ntrancl": bounded transitive closure on
  1037 relations.
  1038 
  1039 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
  1040 
  1041 * Theory Int: Discontinued many legacy theorems specific to type int.
  1042 INCOMPATIBILITY, use the corresponding generic theorems instead.
  1043 
  1044   zminus_zminus ~> minus_minus
  1045   zminus_0 ~> minus_zero
  1046   zminus_zadd_distrib ~> minus_add_distrib
  1047   zadd_commute ~> add_commute
  1048   zadd_assoc ~> add_assoc
  1049   zadd_left_commute ~> add_left_commute
  1050   zadd_ac ~> add_ac
  1051   zmult_ac ~> mult_ac
  1052   zadd_0 ~> add_0_left
  1053   zadd_0_right ~> add_0_right
  1054   zadd_zminus_inverse2 ~> left_minus
  1055   zmult_zminus ~> mult_minus_left
  1056   zmult_commute ~> mult_commute
  1057   zmult_assoc ~> mult_assoc
  1058   zadd_zmult_distrib ~> left_distrib
  1059   zadd_zmult_distrib2 ~> right_distrib
  1060   zdiff_zmult_distrib ~> left_diff_distrib
  1061   zdiff_zmult_distrib2 ~> right_diff_distrib
  1062   zmult_1 ~> mult_1_left
  1063   zmult_1_right ~> mult_1_right
  1064   zle_refl ~> order_refl
  1065   zle_trans ~> order_trans
  1066   zle_antisym ~> order_antisym
  1067   zle_linear ~> linorder_linear
  1068   zless_linear ~> linorder_less_linear
  1069   zadd_left_mono ~> add_left_mono
  1070   zadd_strict_right_mono ~> add_strict_right_mono
  1071   zadd_zless_mono ~> add_less_le_mono
  1072   int_0_less_1 ~> zero_less_one
  1073   int_0_neq_1 ~> zero_neq_one
  1074   zless_le ~> less_le
  1075   zpower_zadd_distrib ~> power_add
  1076   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
  1077   zero_le_zpower_abs ~> zero_le_power_abs
  1078 
  1079 * Theory Deriv: Renamed
  1080 
  1081   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
  1082 
  1083 * Theory Library/Multiset: Improved code generation of multisets.
  1084 
  1085 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
  1086 are expressed via type classes again. The special syntax
  1087 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
  1088 setsum_set, which is now subsumed by Big_Operators.setsum.
  1089 INCOMPATIBILITY.
  1090 
  1091 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
  1092 use theory HOL/Library/Nat_Bijection instead.
  1093 
  1094 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
  1095 trees is now inside a type class context.  Names of affected
  1096 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
  1097 theories working directly with raw red-black trees, adapt the names as
  1098 follows:
  1099 
  1100   Operations:
  1101   bulkload -> rbt_bulkload
  1102   del_from_left -> rbt_del_from_left
  1103   del_from_right -> rbt_del_from_right
  1104   del -> rbt_del
  1105   delete -> rbt_delete
  1106   ins -> rbt_ins
  1107   insert -> rbt_insert
  1108   insertw -> rbt_insert_with
  1109   insert_with_key -> rbt_insert_with_key
  1110   map_entry -> rbt_map_entry
  1111   lookup -> rbt_lookup
  1112   sorted -> rbt_sorted
  1113   tree_greater -> rbt_greater
  1114   tree_less -> rbt_less
  1115   tree_less_symbol -> rbt_less_symbol
  1116   union -> rbt_union
  1117   union_with -> rbt_union_with
  1118   union_with_key -> rbt_union_with_key
  1119 
  1120   Lemmas:
  1121   balance_left_sorted -> balance_left_rbt_sorted
  1122   balance_left_tree_greater -> balance_left_rbt_greater
  1123   balance_left_tree_less -> balance_left_rbt_less
  1124   balance_right_sorted -> balance_right_rbt_sorted
  1125   balance_right_tree_greater -> balance_right_rbt_greater
  1126   balance_right_tree_less -> balance_right_rbt_less
  1127   balance_sorted -> balance_rbt_sorted
  1128   balance_tree_greater -> balance_rbt_greater
  1129   balance_tree_less -> balance_rbt_less
  1130   bulkload_is_rbt -> rbt_bulkload_is_rbt
  1131   combine_sorted -> combine_rbt_sorted
  1132   combine_tree_greater -> combine_rbt_greater
  1133   combine_tree_less -> combine_rbt_less
  1134   delete_in_tree -> rbt_delete_in_tree
  1135   delete_is_rbt -> rbt_delete_is_rbt
  1136   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
  1137   del_from_left_tree_less -> rbt_del_from_left_rbt_less
  1138   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
  1139   del_from_right_tree_less -> rbt_del_from_right_rbt_less
  1140   del_in_tree -> rbt_del_in_tree
  1141   del_inv1_inv2 -> rbt_del_inv1_inv2
  1142   del_sorted -> rbt_del_rbt_sorted
  1143   del_tree_greater -> rbt_del_rbt_greater
  1144   del_tree_less -> rbt_del_rbt_less
  1145   dom_lookup_Branch -> dom_rbt_lookup_Branch
  1146   entries_lookup -> entries_rbt_lookup
  1147   finite_dom_lookup -> finite_dom_rbt_lookup
  1148   insert_sorted -> rbt_insert_rbt_sorted
  1149   insertw_is_rbt -> rbt_insertw_is_rbt
  1150   insertwk_is_rbt -> rbt_insertwk_is_rbt
  1151   insertwk_sorted -> rbt_insertwk_rbt_sorted
  1152   insertw_sorted -> rbt_insertw_rbt_sorted
  1153   ins_sorted -> ins_rbt_sorted
  1154   ins_tree_greater -> ins_rbt_greater
  1155   ins_tree_less -> ins_rbt_less
  1156   is_rbt_sorted -> is_rbt_rbt_sorted
  1157   lookup_balance -> rbt_lookup_balance
  1158   lookup_bulkload -> rbt_lookup_rbt_bulkload
  1159   lookup_delete -> rbt_lookup_rbt_delete
  1160   lookup_Empty -> rbt_lookup_Empty
  1161   lookup_from_in_tree -> rbt_lookup_from_in_tree
  1162   lookup_in_tree -> rbt_lookup_in_tree
  1163   lookup_ins -> rbt_lookup_ins
  1164   lookup_insert -> rbt_lookup_rbt_insert
  1165   lookup_insertw -> rbt_lookup_rbt_insertw
  1166   lookup_insertwk -> rbt_lookup_rbt_insertwk
  1167   lookup_keys -> rbt_lookup_keys
  1168   lookup_map -> rbt_lookup_map
  1169   lookup_map_entry -> rbt_lookup_rbt_map_entry
  1170   lookup_tree_greater -> rbt_lookup_rbt_greater
  1171   lookup_tree_less -> rbt_lookup_rbt_less
  1172   lookup_union -> rbt_lookup_rbt_union
  1173   map_entry_color_of -> rbt_map_entry_color_of
  1174   map_entry_inv1 -> rbt_map_entry_inv1
  1175   map_entry_inv2 -> rbt_map_entry_inv2
  1176   map_entry_is_rbt -> rbt_map_entry_is_rbt
  1177   map_entry_sorted -> rbt_map_entry_rbt_sorted
  1178   map_entry_tree_greater -> rbt_map_entry_rbt_greater
  1179   map_entry_tree_less -> rbt_map_entry_rbt_less
  1180   map_tree_greater -> map_rbt_greater
  1181   map_tree_less -> map_rbt_less
  1182   map_sorted -> map_rbt_sorted
  1183   paint_sorted -> paint_rbt_sorted
  1184   paint_lookup -> paint_rbt_lookup
  1185   paint_tree_greater -> paint_rbt_greater
  1186   paint_tree_less -> paint_rbt_less
  1187   sorted_entries -> rbt_sorted_entries
  1188   tree_greater_eq_trans -> rbt_greater_eq_trans
  1189   tree_greater_nit -> rbt_greater_nit
  1190   tree_greater_prop -> rbt_greater_prop
  1191   tree_greater_simps -> rbt_greater_simps
  1192   tree_greater_trans -> rbt_greater_trans
  1193   tree_less_eq_trans -> rbt_less_eq_trans
  1194   tree_less_nit -> rbt_less_nit
  1195   tree_less_prop -> rbt_less_prop
  1196   tree_less_simps -> rbt_less_simps
  1197   tree_less_trans -> rbt_less_trans
  1198   tree_ord_props -> rbt_ord_props
  1199   union_Branch -> rbt_union_Branch
  1200   union_is_rbt -> rbt_union_is_rbt
  1201   unionw_is_rbt -> rbt_unionw_is_rbt
  1202   unionwk_is_rbt -> rbt_unionwk_is_rbt
  1203   unionwk_sorted -> rbt_unionwk_rbt_sorted
  1204 
  1205 * Theory HOL/Library/Float: Floating point numbers are now defined as
  1206 a subset of the real numbers.  All operations are defined using the
  1207 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
  1208 
  1209   Changed Operations:
  1210   float_abs -> abs
  1211   float_nprt -> nprt
  1212   float_pprt -> pprt
  1213   pow2 -> use powr
  1214   round_down -> float_round_down
  1215   round_up -> float_round_up
  1216   scale -> exponent
  1217 
  1218   Removed Operations:
  1219   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
  1220 
  1221   Renamed Lemmas:
  1222   abs_float_def -> Float.compute_float_abs
  1223   bitlen_ge0 -> bitlen_nonneg
  1224   bitlen.simps -> Float.compute_bitlen
  1225   float_components -> Float_mantissa_exponent
  1226   float_divl.simps -> Float.compute_float_divl
  1227   float_divr.simps -> Float.compute_float_divr
  1228   float_eq_odd -> mult_powr_eq_mult_powr_iff
  1229   float_power -> real_of_float_power
  1230   lapprox_posrat_def -> Float.compute_lapprox_posrat
  1231   lapprox_rat.simps -> Float.compute_lapprox_rat
  1232   le_float_def' -> Float.compute_float_le
  1233   le_float_def -> less_eq_float.rep_eq
  1234   less_float_def' -> Float.compute_float_less
  1235   less_float_def -> less_float.rep_eq
  1236   normfloat_def -> Float.compute_normfloat
  1237   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
  1238   normfloat -> normfloat_def
  1239   normfloat_unique -> use normfloat_def
  1240   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
  1241   one_float_def -> Float.compute_float_one
  1242   plus_float_def -> Float.compute_float_plus
  1243   rapprox_posrat_def -> Float.compute_rapprox_posrat
  1244   rapprox_rat.simps -> Float.compute_rapprox_rat
  1245   real_of_float_0 -> zero_float.rep_eq
  1246   real_of_float_1 -> one_float.rep_eq
  1247   real_of_float_abs -> abs_float.rep_eq
  1248   real_of_float_add -> plus_float.rep_eq
  1249   real_of_float_minus -> uminus_float.rep_eq
  1250   real_of_float_mult -> times_float.rep_eq
  1251   real_of_float_simp -> Float.rep_eq
  1252   real_of_float_sub -> minus_float.rep_eq
  1253   round_down.simps -> Float.compute_float_round_down
  1254   round_up.simps -> Float.compute_float_round_up
  1255   times_float_def -> Float.compute_float_times
  1256   uminus_float_def -> Float.compute_float_uminus
  1257   zero_float_def -> Float.compute_float_zero
  1258 
  1259   Lemmas not necessary anymore, use the transfer method:
  1260   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
  1261   float_divr, float_le_simp, float_less1_mantissa_bound,
  1262   float_less_simp, float_less_zero, float_le_zero,
  1263   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
  1264   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
  1265   lapprox_rat_bottom, normalized_float, rapprox_posrat,
  1266   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
  1267   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
  1268   round_up, zero_le_float, zero_less_float
  1269 
  1270 * New theory HOL/Library/DAList provides an abstract type for
  1271 association lists with distinct keys.
  1272 
  1273 * Session HOL/IMP: Added new theory of abstract interpretation of
  1274 annotated commands.
  1275 
  1276 * Session HOL-Import: Re-implementation from scratch is faster,
  1277 simpler, and more scalable.  Requires a proof bundle, which is
  1278 available as an external component.  Discontinued old (and mostly
  1279 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
  1280 
  1281 * Session HOL-Word: Discontinued many redundant theorems specific to
  1282 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
  1283 instead.
  1284 
  1285   word_sub_alt ~> word_sub_wi
  1286   word_add_alt ~> word_add_def
  1287   word_mult_alt ~> word_mult_def
  1288   word_minus_alt ~> word_minus_def
  1289   word_0_alt ~> word_0_wi
  1290   word_1_alt ~> word_1_wi
  1291   word_add_0 ~> add_0_left
  1292   word_add_0_right ~> add_0_right
  1293   word_mult_1 ~> mult_1_left
  1294   word_mult_1_right ~> mult_1_right
  1295   word_add_commute ~> add_commute
  1296   word_add_assoc ~> add_assoc
  1297   word_add_left_commute ~> add_left_commute
  1298   word_mult_commute ~> mult_commute
  1299   word_mult_assoc ~> mult_assoc
  1300   word_mult_left_commute ~> mult_left_commute
  1301   word_left_distrib ~> left_distrib
  1302   word_right_distrib ~> right_distrib
  1303   word_left_minus ~> left_minus
  1304   word_diff_0_right ~> diff_0_right
  1305   word_diff_self ~> diff_self
  1306   word_sub_def ~> diff_minus
  1307   word_diff_minus ~> diff_minus
  1308   word_add_ac ~> add_ac
  1309   word_mult_ac ~> mult_ac
  1310   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1311   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1312   word_order_trans ~> order_trans
  1313   word_order_refl ~> order_refl
  1314   word_order_antisym ~> order_antisym
  1315   word_order_linear ~> linorder_linear
  1316   lenw1_zero_neq_one ~> zero_neq_one
  1317   word_number_of_eq ~> number_of_eq
  1318   word_of_int_add_hom ~> wi_hom_add
  1319   word_of_int_sub_hom ~> wi_hom_sub
  1320   word_of_int_mult_hom ~> wi_hom_mult
  1321   word_of_int_minus_hom ~> wi_hom_neg
  1322   word_of_int_succ_hom ~> wi_hom_succ
  1323   word_of_int_pred_hom ~> wi_hom_pred
  1324   word_of_int_0_hom ~> word_0_wi
  1325   word_of_int_1_hom ~> word_1_wi
  1326 
  1327 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1328 machine word equalities and inequalities into logical circuits,
  1329 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1330 multiplication, shifting by constants, bitwise operators and numeric
  1331 constants.  Requires fixed-length word types, not 'a word.  Solves
  1332 many standard word identities outright and converts more into first
  1333 order problems amenable to blast or similar.  See also examples in
  1334 HOL/Word/Examples/WordExamples.thy.
  1335 
  1336 * Session HOL-Probability: Introduced the type "'a measure" to
  1337 represent measures, this replaces the records 'a algebra and 'a
  1338 measure_space.  The locales based on subset_class now have two
  1339 locale-parameters the space \<Omega> and the set of measurable sets M.
  1340 The product of probability spaces uses now the same constant as the
  1341 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1342 measure".  Most constants are defined now outside of locales and gain
  1343 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1344 Measure space constructions for distributions and densities now got
  1345 their own constants distr and density.  Instead of using locales to
  1346 describe measure spaces with a finite space, the measure count_space
  1347 and point_measure is introduced.  INCOMPATIBILITY.
  1348 
  1349   Renamed constants:
  1350   measure -> emeasure
  1351   finite_measure.\<mu>' -> measure
  1352   product_algebra_generator -> prod_algebra
  1353   product_prob_space.emb -> prod_emb
  1354   product_prob_space.infprod_algebra -> PiM
  1355 
  1356   Removed locales:
  1357   completeable_measure_space
  1358   finite_measure_space
  1359   finite_prob_space
  1360   finite_product_finite_prob_space
  1361   finite_product_sigma_algebra
  1362   finite_sigma_algebra
  1363   measure_space
  1364   pair_finite_prob_space
  1365   pair_finite_sigma_algebra
  1366   pair_finite_space
  1367   pair_sigma_algebra
  1368   product_sigma_algebra
  1369 
  1370   Removed constants:
  1371   conditional_space
  1372   distribution -> use distr measure, or distributed predicate
  1373   image_space
  1374   joint_distribution -> use distr measure, or distributed predicate
  1375   pair_measure_generator
  1376   product_prob_space.infprod_algebra -> use PiM
  1377   subvimage
  1378 
  1379   Replacement theorems:
  1380   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1381   finite_measure.empty_measure -> measure_empty
  1382   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1383   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1384   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1385   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1386   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1387   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1388   finite_measure.positive_measure' -> measure_nonneg
  1389   finite_measure.real_measure -> finite_measure.emeasure_real
  1390   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1391   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1392   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1393   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1394   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1395   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1396   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1397   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1398   information_space.entropy_commute -> information_space.entropy_commute_simple
  1399   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1400   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1401   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1402   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1403   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1404   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1405   information_space.KL_ge_0 -> information_space.KL_nonneg
  1406   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1407   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1408   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1409   Int_stable_product_algebra_generator -> positive_integral
  1410   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1411   measure_space.additive -> emeasure_additive
  1412   measure_space.AE_iff_null_set -> AE_iff_null
  1413   measure_space.almost_everywhere_def -> eventually_ae_filter
  1414   measure_space.almost_everywhere_vimage -> AE_distrD
  1415   measure_space.continuity_from_above -> INF_emeasure_decseq
  1416   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1417   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1418   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1419   measure_space_density -> emeasure_density
  1420   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1421   measure_space.integrable_vimage -> integrable_distr
  1422   measure_space.integral_translated_density -> integral_density
  1423   measure_space.integral_vimage -> integral_distr
  1424   measure_space.measure_additive -> plus_emeasure
  1425   measure_space.measure_compl -> emeasure_compl
  1426   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1427   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1428   measure_space.measure_decseq -> decseq_emeasure
  1429   measure_space.measure_Diff -> emeasure_Diff
  1430   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1431   measure_space.measure_eq_0 -> emeasure_eq_0
  1432   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1433   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1434   measure_space.measure_incseq -> incseq_emeasure
  1435   measure_space.measure_insert -> emeasure_insert
  1436   measure_space.measure_mono -> emeasure_mono
  1437   measure_space.measure_not_negative -> emeasure_not_MInf
  1438   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1439   measure_space.measure_setsum -> setsum_emeasure
  1440   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1441   measure_space.measure_space_vimage -> emeasure_distr
  1442   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1443   measure_space.measure_subadditive -> subadditive
  1444   measure_space.measure_top -> emeasure_space
  1445   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1446   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1447   measure_space.positive_integral_translated_density -> positive_integral_density
  1448   measure_space.positive_integral_vimage -> positive_integral_distr
  1449   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1450   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1451   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1452   measure_space.real_measure_Diff -> measure_Diff
  1453   measure_space.real_measure_finite_Union -> measure_finite_Union
  1454   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1455   measure_space.real_measure_subadditive -> measure_subadditive
  1456   measure_space.real_measure_Union -> measure_Union
  1457   measure_space.real_measure_UNION -> measure_UNION
  1458   measure_space.simple_function_vimage -> simple_function_comp
  1459   measure_space.simple_integral_vimage -> simple_integral_distr
  1460   measure_space.simple_integral_vimage -> simple_integral_distr
  1461   measure_unique_Int_stable -> measure_eqI_generator_eq
  1462   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1463   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1464   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1465   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1466   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1467   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1468   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1469   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1470   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1471   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1472   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1473   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1474   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1475   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1476   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1477   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1478   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1479   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1480   prob_space.prob_space_vimage -> prob_space_distr
  1481   prob_space.random_variable_restrict -> measurable_restrict
  1482   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1483   product_algebraE -> prod_algebraE_all
  1484   product_algebra_generator_der -> prod_algebra_eq_finite
  1485   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1486   product_algebraI -> sets_PiM_I_finite
  1487   product_measure_exists -> product_sigma_finite.sigma_finite
  1488   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1489   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1490   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1491   product_prob_space.measurable_component -> measurable_component_singleton
  1492   product_prob_space.measurable_emb -> measurable_prod_emb
  1493   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1494   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1495   product_prob_space.measure_emb -> emeasure_prod_emb
  1496   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1497   product_sigma_algebra.product_algebra_into_space -> space_closed
  1498   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1499   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1500   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1501   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1502   sets_product_algebra -> sets_PiM
  1503   sigma_algebra.measurable_sigma -> measurable_measure_of
  1504   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1505   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1506   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1507   space_product_algebra -> space_PiM
  1508 
  1509 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1510 languages) into Isabelle/HOL.
  1511 
  1512 
  1513 *** FOL ***
  1514 
  1515 * New "case_product" attribute (see HOL).
  1516 
  1517 
  1518 *** ZF ***
  1519 
  1520 * Greater support for structured proofs involving induction or case
  1521 analysis.
  1522 
  1523 * Much greater use of mathematical symbols.
  1524 
  1525 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1526 
  1527 
  1528 *** ML ***
  1529 
  1530 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1531 from a minor keyword introduced via theory header declaration.
  1532 
  1533 * Antiquotation @{command_spec "name"} produces the
  1534 Outer_Syntax.command_spec from a major keyword introduced via theory
  1535 header declaration; it can be passed to Outer_Syntax.command etc.
  1536 
  1537 * Local_Theory.define no longer hard-wires default theorem name
  1538 "foo_def", but retains the binding as given.  If that is Binding.empty
  1539 / Attrib.empty_binding, the result is not registered as user-level
  1540 fact.  The Local_Theory.define_internal variant allows to specify a
  1541 non-empty name (used for the foundation in the background theory),
  1542 while omitting the fact binding in the user-context.  Potential
  1543 INCOMPATIBILITY for derived definitional packages: need to specify
  1544 naming policy for primitive definitions more explicitly.
  1545 
  1546 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1547 conformance with similar operations in structure Term and Logic.
  1548 
  1549 * Antiquotation @{attributes [...]} embeds attribute source
  1550 representation into the ML text, which is particularly useful with
  1551 declarations like Local_Theory.note.
  1552 
  1553 * Structure Proof_Context follows standard naming scheme.  Old
  1554 ProofContext has been discontinued.  INCOMPATIBILITY.
  1555 
  1556 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1557 change of semantics: update is applied to auxiliary local theory
  1558 context as well.
  1559 
  1560 * Modernized some old-style infix operations:
  1561 
  1562   addeqcongs    ~> Simplifier.add_eqcong
  1563   deleqcongs    ~> Simplifier.del_eqcong
  1564   addcongs      ~> Simplifier.add_cong
  1565   delcongs      ~> Simplifier.del_cong
  1566   setmksimps    ~> Simplifier.set_mksimps
  1567   setmkcong     ~> Simplifier.set_mkcong
  1568   setmksym      ~> Simplifier.set_mksym
  1569   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1570   settermless   ~> Simplifier.set_termless
  1571   setsubgoaler  ~> Simplifier.set_subgoaler
  1572   addsplits     ~> Splitter.add_split
  1573   delsplits     ~> Splitter.del_split
  1574 
  1575 
  1576 *** System ***
  1577 
  1578 * USER_HOME settings variable points to cross-platform user home
  1579 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1580 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1581 of former $HOME.  A different default for USER_HOME may be set
  1582 explicitly in shell environment, before Isabelle settings are
  1583 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1584 the generic user home was intended.
  1585 
  1586 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1587 notation, which is useful for the jEdit file browser, for example.
  1588 
  1589 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1590 (not just JRE).
  1591 
  1592 
  1593 
  1594 New in Isabelle2011-1 (October 2011)
  1595 ------------------------------------
  1596 
  1597 *** General ***
  1598 
  1599 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1600 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1601 
  1602   - Management of multiple theory files directly from the editor
  1603     buffer store -- bypassing the file-system (no requirement to save
  1604     files for checking).
  1605 
  1606   - Markup of formal entities within the text buffer, with semantic
  1607     highlighting, tooltips and hyperlinks to jump to defining source
  1608     positions.
  1609 
  1610   - Improved text rendering, with sub/superscripts in the source
  1611     buffer (including support for copy/paste wrt. output panel, HTML
  1612     theory output and other non-Isabelle text boxes).
  1613 
  1614   - Refined scheduling of proof checking and printing of results,
  1615     based on interactive editor view.  (Note: jEdit folding and
  1616     narrowing allows to restrict buffer perspectives explicitly.)
  1617 
  1618   - Reduced CPU performance requirements, usable on machines with few
  1619     cores.
  1620 
  1621   - Reduced memory requirements due to pruning of unused document
  1622     versions (garbage collection).
  1623 
  1624 See also ~~/src/Tools/jEdit/README.html for further information,
  1625 including some remaining limitations.
  1626 
  1627 * Theory loader: source files are exclusively located via the master
  1628 directory of each theory node (where the .thy file itself resides).
  1629 The global load path (such as src/HOL/Library) has been discontinued.
  1630 Note that the path element ~~ may be used to reference theories in the
  1631 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1632 INCOMPATIBILITY.
  1633 
  1634 * Theory loader: source files are identified by content via SHA1
  1635 digests.  Discontinued former path/modtime identification and optional
  1636 ISABELLE_FILE_IDENT plugin scripts.
  1637 
  1638 * Parallelization of nested Isar proofs is subject to
  1639 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1640 usedir option -Q.
  1641 
  1642 * Name space: former unsynchronized references are now proper
  1643 configuration options, with more conventional names:
  1644 
  1645   long_names   ~> names_long
  1646   short_names  ~> names_short
  1647   unique_names ~> names_unique
  1648 
  1649 Minor INCOMPATIBILITY, need to declare options in context like this:
  1650 
  1651   declare [[names_unique = false]]
  1652 
  1653 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1654 that the result needs to be unique, which means fact specifications
  1655 may have to be refined after enriching a proof context.
  1656 
  1657 * Attribute "case_names" has been refined: the assumptions in each case
  1658 can be named now by following the case name with [name1 name2 ...].
  1659 
  1660 * Isabelle/Isar reference manual has been updated and extended:
  1661   - "Synopsis" provides a catalog of main Isar language concepts.
  1662   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1663   - Updated material from classic "ref" manual, notably about
  1664     "Classical Reasoner".
  1665 
  1666 
  1667 *** HOL ***
  1668 
  1669 * Class bot and top require underlying partial order rather than
  1670 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1671 
  1672 * Class complete_lattice: generalized a couple of lemmas from sets;
  1673 generalized theorems INF_cong and SUP_cong.  New type classes for
  1674 complete boolean algebras and complete linear orders.  Lemmas
  1675 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1676 class complete_linorder.
  1677 
  1678 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1679 Sup_fun_def, Inf_apply, Sup_apply.
  1680 
  1681 Removed redundant lemmas (the right hand side gives hints how to
  1682 replace them for (metis ...), or (simp only: ...) proofs):
  1683 
  1684   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1685   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1686   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1687   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1688   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1689   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1690   Inter_def ~> INF_def, image_def
  1691   Union_def ~> SUP_def, image_def
  1692   INT_eq ~> INF_def, and image_def
  1693   UN_eq ~> SUP_def, and image_def
  1694   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1695 
  1696 More consistent and comprehensive names:
  1697 
  1698   INTER_eq_Inter_image ~> INF_def
  1699   UNION_eq_Union_image ~> SUP_def
  1700   INFI_def ~> INF_def
  1701   SUPR_def ~> SUP_def
  1702   INF_leI ~> INF_lower
  1703   INF_leI2 ~> INF_lower2
  1704   le_INFI ~> INF_greatest
  1705   le_SUPI ~> SUP_upper
  1706   le_SUPI2 ~> SUP_upper2
  1707   SUP_leI ~> SUP_least
  1708   INFI_bool_eq ~> INF_bool_eq
  1709   SUPR_bool_eq ~> SUP_bool_eq
  1710   INFI_apply ~> INF_apply
  1711   SUPR_apply ~> SUP_apply
  1712   INTER_def ~> INTER_eq
  1713   UNION_def ~> UNION_eq
  1714 
  1715 INCOMPATIBILITY.
  1716 
  1717 * Renamed theory Complete_Lattice to Complete_Lattices.
  1718 INCOMPATIBILITY.
  1719 
  1720 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1721 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1722 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1723 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1724 
  1725 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1726 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1727 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1728 INCOMPATIBILITY.
  1729 
  1730 * Added syntactic classes "inf" and "sup" for the respective
  1731 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1732 (mostly internal) locale predicates for some derived classes.
  1733 
  1734 * Theorem collections ball_simps and bex_simps do not contain theorems
  1735 referring to UNION any longer; these have been moved to collection
  1736 UN_ball_bex_simps.  INCOMPATIBILITY.
  1737 
  1738 * Theory Archimedean_Field: floor now is defined as parameter of a
  1739 separate type class floor_ceiling.
  1740 
  1741 * Theory Finite_Set: more coherent development of fold_set locales:
  1742 
  1743     locale fun_left_comm ~> locale comp_fun_commute
  1744     locale fun_left_comm_idem ~> locale comp_fun_idem
  1745 
  1746 Both use point-free characterization; interpretation proofs may need
  1747 adjustment.  INCOMPATIBILITY.
  1748 
  1749 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1750 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1751 
  1752 * Theory Complex_Main: The locale interpretations for the
  1753 bounded_linear and bounded_bilinear locales have been removed, in
  1754 order to reduce the number of duplicate lemmas. Users must use the
  1755 original names for distributivity theorems, potential INCOMPATIBILITY.
  1756 
  1757   divide.add ~> add_divide_distrib
  1758   divide.diff ~> diff_divide_distrib
  1759   divide.setsum ~> setsum_divide_distrib
  1760   mult.add_right ~> right_distrib
  1761   mult.diff_right ~> right_diff_distrib
  1762   mult_right.setsum ~> setsum_right_distrib
  1763   mult_left.diff ~> left_diff_distrib
  1764 
  1765 * Theory Complex_Main: Several redundant theorems have been removed or
  1766 replaced by more general versions. INCOMPATIBILITY.
  1767 
  1768   real_diff_def ~> minus_real_def
  1769   real_divide_def ~> divide_real_def
  1770   real_less_def ~> less_le
  1771   real_abs_def ~> abs_real_def
  1772   real_sgn_def ~> sgn_real_def
  1773   real_mult_commute ~> mult_commute
  1774   real_mult_assoc ~> mult_assoc
  1775   real_mult_1 ~> mult_1_left
  1776   real_add_mult_distrib ~> left_distrib
  1777   real_zero_not_eq_one ~> zero_neq_one
  1778   real_mult_inverse_left ~> left_inverse
  1779   INVERSE_ZERO ~> inverse_zero
  1780   real_le_refl ~> order_refl
  1781   real_le_antisym ~> order_antisym
  1782   real_le_trans ~> order_trans
  1783   real_le_linear ~> linear
  1784   real_le_eq_diff ~> le_iff_diff_le_0
  1785   real_add_left_mono ~> add_left_mono
  1786   real_mult_order ~> mult_pos_pos
  1787   real_mult_less_mono2 ~> mult_strict_left_mono
  1788   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1789   real_0_le_divide_iff ~> zero_le_divide_iff
  1790   realpow_two_disj ~> power2_eq_iff
  1791   real_squared_diff_one_factored ~> square_diff_one_factored
  1792   realpow_two_diff ~> square_diff_square_factored
  1793   reals_complete2 ~> complete_real
  1794   real_sum_squared_expand ~> power2_sum
  1795   exp_ln_eq ~> ln_unique
  1796   expi_add ~> exp_add
  1797   expi_zero ~> exp_zero
  1798   lemma_DERIV_subst ~> DERIV_cong
  1799   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1800   LIMSEQ_const ~> tendsto_const
  1801   LIMSEQ_norm ~> tendsto_norm
  1802   LIMSEQ_add ~> tendsto_add
  1803   LIMSEQ_minus ~> tendsto_minus
  1804   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1805   LIMSEQ_diff ~> tendsto_diff
  1806   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1807   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1808   LIMSEQ_mult ~> tendsto_mult
  1809   LIMSEQ_inverse ~> tendsto_inverse
  1810   LIMSEQ_divide ~> tendsto_divide
  1811   LIMSEQ_pow ~> tendsto_power
  1812   LIMSEQ_setsum ~> tendsto_setsum
  1813   LIMSEQ_setprod ~> tendsto_setprod
  1814   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1815   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1816   LIMSEQ_imp_rabs ~> tendsto_rabs
  1817   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1818   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1819   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1820   LIMSEQ_Complex ~> tendsto_Complex
  1821   LIM_ident ~> tendsto_ident_at
  1822   LIM_const ~> tendsto_const
  1823   LIM_add ~> tendsto_add
  1824   LIM_add_zero ~> tendsto_add_zero
  1825   LIM_minus ~> tendsto_minus
  1826   LIM_diff ~> tendsto_diff
  1827   LIM_norm ~> tendsto_norm
  1828   LIM_norm_zero ~> tendsto_norm_zero
  1829   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1830   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1831   LIM_rabs ~> tendsto_rabs
  1832   LIM_rabs_zero ~> tendsto_rabs_zero
  1833   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1834   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1835   LIM_compose ~> tendsto_compose
  1836   LIM_mult ~> tendsto_mult
  1837   LIM_scaleR ~> tendsto_scaleR
  1838   LIM_of_real ~> tendsto_of_real
  1839   LIM_power ~> tendsto_power
  1840   LIM_inverse ~> tendsto_inverse
  1841   LIM_sgn ~> tendsto_sgn
  1842   isCont_LIM_compose ~> isCont_tendsto_compose
  1843   bounded_linear.LIM ~> bounded_linear.tendsto
  1844   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1845   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1846   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1847   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1848   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1849   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1850 
  1851 * Theory Complex_Main: The definition of infinite series was
  1852 generalized.  Now it is defined on the type class {topological_space,
  1853 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1854 
  1855 * Theory Complex_Main: The complex exponential function "expi" is now
  1856 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1857 several polymorphic lemmas about "exp" are now applicable to "expi".
  1858 
  1859 * Code generation:
  1860 
  1861   - Theory Library/Code_Char_ord provides native ordering of
  1862     characters in the target language.
  1863 
  1864   - Commands code_module and code_library are legacy, use export_code
  1865     instead.
  1866 
  1867   - Method "evaluation" is legacy, use method "eval" instead.
  1868 
  1869   - Legacy evaluator "SML" is deactivated by default.  May be
  1870     reactivated by the following theory command:
  1871 
  1872       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1873 
  1874 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1875 
  1876 * New proof method "induction" that gives induction hypotheses the
  1877 name "IH", thus distinguishing them from further hypotheses that come
  1878 from rule induction.  The latter are still called "hyps".  Method
  1879 "induction" is a thin wrapper around "induct" and follows the same
  1880 syntax.
  1881 
  1882 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1883 still available as a legacy feature for some time.
  1884 
  1885 * Nitpick:
  1886   - Added "need" and "total_consts" options.
  1887   - Reintroduced "show_skolems" option by popular demand.
  1888   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1889     INCOMPATIBILITY.
  1890 
  1891 * Sledgehammer:
  1892   - Use quasi-sound (and efficient) translations by default.
  1893   - Added support for the following provers: E-ToFoF, LEO-II,
  1894     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1895   - Automatically preplay and minimize proofs before showing them if
  1896     this can be done within reasonable time.
  1897   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1898     INCOMPATIBILITY.
  1899   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1900     "max_mono_iters", and "max_new_mono_instances" options.
  1901   - Removed "explicit_apply" and "full_types" options as well as "Full
  1902     Types" Proof General menu item. INCOMPATIBILITY.
  1903 
  1904 * Metis:
  1905   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1906   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1907     INCOMPATIBILITY.
  1908 
  1909 * Command 'try':
  1910   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1911     "elim:" options. INCOMPATIBILITY.
  1912   - Introduced 'try' that not only runs 'try_methods' but also
  1913     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1914 
  1915 * Quickcheck:
  1916   - Added "eval" option to evaluate terms for the found counterexample
  1917     (currently only supported by the default (exhaustive) tester).
  1918   - Added post-processing of terms to obtain readable counterexamples
  1919     (currently only supported by the default (exhaustive) tester).
  1920   - New counterexample generator quickcheck[narrowing] enables
  1921     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1922     with its installation location defined in the Isabelle settings
  1923     environment as ISABELLE_GHC.
  1924   - Removed quickcheck tester "SML" based on the SML code generator
  1925     (formly in HOL/Library).
  1926 
  1927 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1928 use 'partial_function' instead.
  1929 
  1930 * Theory Library/Extended_Reals replaces now the positive extended
  1931 reals found in probability theory. This file is extended by
  1932 Multivariate_Analysis/Extended_Real_Limits.
  1933 
  1934 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1935 from where it must be imported explicitly if it is really required.
  1936 INCOMPATIBILITY.
  1937 
  1938 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1939 been moved here.  INCOMPATIBILITY.
  1940 
  1941 * Theory Library/Saturated provides type of numbers with saturated
  1942 arithmetic.
  1943 
  1944 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1945 product type 'a * 'b, and provides instance proofs for various order
  1946 and lattice type classes.
  1947 
  1948 * Theory Library/Countable now provides the "countable_datatype" proof
  1949 method for proving "countable" class instances for datatypes.
  1950 
  1951 * Theory Library/Cset_Monad allows do notation for computable sets
  1952 (cset) via the generic monad ad-hoc overloading facility.
  1953 
  1954 * Library: Theories of common data structures are split into theories
  1955 for implementation, an invariant-ensuring type, and connection to an
  1956 abstract type. INCOMPATIBILITY.
  1957 
  1958   - RBT is split into RBT and RBT_Mapping.
  1959   - AssocList is split and renamed into AList and AList_Mapping.
  1960   - DList is split into DList_Impl, DList, and DList_Cset.
  1961   - Cset is split into Cset and List_Cset.
  1962 
  1963 * Theory Library/Nat_Infinity has been renamed to
  1964 Library/Extended_Nat, with name changes of the following types and
  1965 constants:
  1966 
  1967   type inat   ~> type enat
  1968   Fin         ~> enat
  1969   Infty       ~> infinity (overloaded)
  1970   iSuc        ~> eSuc
  1971   the_Fin     ~> the_enat
  1972 
  1973 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1974 been renamed accordingly. INCOMPATIBILITY.
  1975 
  1976 * Session Multivariate_Analysis: The euclidean_space type class now
  1977 fixes a constant "Basis :: 'a set" consisting of the standard
  1978 orthonormal basis for the type. Users now have the option of
  1979 quantifying over this set instead of using the "basis" function, e.g.
  1980 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1981 
  1982 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1983 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1984 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1985 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1986 changed to match. Definition theorems for overloaded constants now use
  1987 the standard "foo_vec_def" naming scheme. A few other theorems have
  1988 been renamed as follows (INCOMPATIBILITY):
  1989 
  1990   Cart_eq          ~> vec_eq_iff
  1991   dist_nth_le_cart ~> dist_vec_nth_le
  1992   tendsto_vector   ~> vec_tendstoI
  1993   Cauchy_vector    ~> vec_CauchyI
  1994 
  1995 * Session Multivariate_Analysis: Several duplicate theorems have been
  1996 removed, and other theorems have been renamed or replaced with more
  1997 general versions. INCOMPATIBILITY.
  1998 
  1999   finite_choice ~> finite_set_choice
  2000   eventually_conjI ~> eventually_conj
  2001   eventually_and ~> eventually_conj_iff
  2002   eventually_false ~> eventually_False
  2003   setsum_norm ~> norm_setsum
  2004   Lim_sequentially ~> LIMSEQ_def
  2005   Lim_ident_at ~> LIM_ident
  2006   Lim_const ~> tendsto_const
  2007   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  2008   Lim_neg ~> tendsto_minus
  2009   Lim_add ~> tendsto_add
  2010   Lim_sub ~> tendsto_diff
  2011   Lim_mul ~> tendsto_scaleR
  2012   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  2013   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  2014   Lim_linear ~> bounded_linear.tendsto
  2015   Lim_component ~> tendsto_euclidean_component
  2016   Lim_component_cart ~> tendsto_vec_nth
  2017   Lim_inner ~> tendsto_inner [OF tendsto_const]
  2018   dot_lsum ~> inner_setsum_left
  2019   dot_rsum ~> inner_setsum_right
  2020   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  2021   continuous_neg ~> continuous_minus
  2022   continuous_sub ~> continuous_diff
  2023   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  2024   continuous_mul ~> continuous_scaleR
  2025   continuous_inv ~> continuous_inverse
  2026   continuous_at_within_inv ~> continuous_at_within_inverse
  2027   continuous_at_inv ~> continuous_at_inverse
  2028   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  2029   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  2030   continuous_at_component ~> continuous_component [OF continuous_at_id]
  2031   continuous_on_neg ~> continuous_on_minus
  2032   continuous_on_sub ~> continuous_on_diff
  2033   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  2034   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  2035   continuous_on_mul ~> continuous_on_scaleR
  2036   continuous_on_mul_real ~> continuous_on_mult
  2037   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  2038   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  2039   continuous_on_inverse ~> continuous_on_inv
  2040   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  2041   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  2042   subset_interior ~> interior_mono
  2043   subset_closure ~> closure_mono
  2044   closure_univ ~> closure_UNIV
  2045   real_arch_lt ~> reals_Archimedean2
  2046   real_arch ~> reals_Archimedean3
  2047   real_abs_norm ~> abs_norm_cancel
  2048   real_abs_sub_norm ~> norm_triangle_ineq3
  2049   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  2050 
  2051 * Session HOL-Probability:
  2052   - Caratheodory's extension lemma is now proved for ring_of_sets.
  2053   - Infinite products of probability measures are now available.
  2054   - Sigma closure is independent, if the generator is independent
  2055   - Use extended reals instead of positive extended
  2056     reals. INCOMPATIBILITY.
  2057 
  2058 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  2059 
  2060   expand_fun_below ~> fun_below_iff
  2061   below_fun_ext ~> fun_belowI
  2062   expand_cfun_eq ~> cfun_eq_iff
  2063   ext_cfun ~> cfun_eqI
  2064   expand_cfun_below ~> cfun_below_iff
  2065   below_cfun_ext ~> cfun_belowI
  2066   monofun_fun_fun ~> fun_belowD
  2067   monofun_fun_arg ~> monofunE
  2068   monofun_lub_fun ~> adm_monofun [THEN admD]
  2069   cont_lub_fun ~> adm_cont [THEN admD]
  2070   cont2cont_Rep_CFun ~> cont2cont_APP
  2071   cont_Rep_CFun_app ~> cont_APP_app
  2072   cont_Rep_CFun_app_app ~> cont_APP_app_app
  2073   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  2074   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  2075   contlub_cfun ~> lub_APP [symmetric]
  2076   contlub_LAM ~> lub_LAM [symmetric]
  2077   thelubI ~> lub_eqI
  2078   UU_I ~> bottomI
  2079   lift_distinct1 ~> lift.distinct(1)
  2080   lift_distinct2 ~> lift.distinct(2)
  2081   Def_not_UU ~> lift.distinct(2)
  2082   Def_inject ~> lift.inject
  2083   below_UU_iff ~> below_bottom_iff
  2084   eq_UU_iff ~> eq_bottom_iff
  2085 
  2086 
  2087 *** Document preparation ***
  2088 
  2089 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  2090 isar-ref manual, both for description and actual application of the
  2091 same.
  2092 
  2093 * Antiquotation @{value} evaluates the given term and presents its
  2094 result.
  2095 
  2096 * Antiquotations: term style "isub" provides ad-hoc conversion of
  2097 variables x1, y23 into subscripted form x\<^isub>1,
  2098 y\<^isub>2\<^isub>3.
  2099 
  2100 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  2101 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  2102 
  2103 * Localized \isabellestyle switch can be used within blocks or groups
  2104 like this:
  2105 
  2106   \isabellestyle{it}  %preferred default
  2107   {\isabellestylett @{text "typewriter stuff"}}
  2108 
  2109 * Discontinued special treatment of hard tabulators.  Implicit
  2110 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  2111 layouts.
  2112 
  2113 
  2114 *** ML ***
  2115 
  2116 * The inner syntax of sort/type/term/prop supports inlined YXML
  2117 representations within quoted string tokens.  By encoding logical
  2118 entities via Term_XML (in ML or Scala) concrete syntax can be
  2119 bypassed, which is particularly useful for producing bits of text
  2120 under external program control.
  2121 
  2122 * Antiquotations for ML and document preparation are managed as theory
  2123 data, which requires explicit setup.
  2124 
  2125 * Isabelle_Process.is_active allows tools to check if the official
  2126 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  2127 (better known as Proof General).
  2128 
  2129 * Structure Proof_Context follows standard naming scheme.  Old
  2130 ProofContext is still available for some time as legacy alias.
  2131 
  2132 * Structure Timing provides various operations for timing; supersedes
  2133 former start_timing/end_timing etc.
  2134 
  2135 * Path.print is the official way to show file-system paths to users
  2136 (including quotes etc.).
  2137 
  2138 * Inner syntax: identifiers in parse trees of generic categories
  2139 "logic", "aprop", "idt" etc. carry position information (disguised as
  2140 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  2141 translations that choke on unexpected type constraints.  Positions can
  2142 be stripped in ML translations via Syntax.strip_positions /
  2143 Syntax.strip_positions_ast, or via the syntax constant
  2144 "_strip_positions" within parse trees.  As last resort, positions can
  2145 be disabled via the configuration option Syntax.positions, which is
  2146 called "syntax_positions" in Isar attribute syntax.
  2147 
  2148 * Discontinued special status of various ML structures that contribute
  2149 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  2150 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  2151 refer directly to Ast.Constant, Lexicon.is_identifier,
  2152 Syntax_Trans.mk_binder_tr etc.
  2153 
  2154 * Typed print translation: discontinued show_sorts argument, which is
  2155 already available via context of "advanced" translation.
  2156 
  2157 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  2158 goal states; body tactic needs to address all subgoals uniformly.
  2159 
  2160 * Slightly more special eq_list/eq_set, with shortcut involving
  2161 pointer equality (assumes that eq relation is reflexive).
  2162 
  2163 * Classical tactics use proper Proof.context instead of historic types
  2164 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  2165 operate directly on Proof.context.  Raw type claset retains its use as
  2166 snapshot of the classical context, which can be recovered via
  2167 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  2168 INCOMPATIBILITY, classical tactics and derived proof methods require
  2169 proper Proof.context.
  2170 
  2171 
  2172 *** System ***
  2173 
  2174 * Discontinued support for Poly/ML 5.2, which was the last version
  2175 without proper multithreading and TimeLimit implementation.
  2176 
  2177 * Discontinued old lib/scripts/polyml-platform, which has been
  2178 obsolete since Isabelle2009-2.
  2179 
  2180 * Various optional external tools are referenced more robustly and
  2181 uniformly by explicit Isabelle settings as follows:
  2182 
  2183   ISABELLE_CSDP   (formerly CSDP_EXE)
  2184   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  2185   ISABELLE_OCAML  (formerly EXEC_OCAML)
  2186   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  2187   ISABELLE_YAP    (formerly EXEC_YAP)
  2188 
  2189 Note that automated detection from the file-system or search path has
  2190 been discontinued.  INCOMPATIBILITY.
  2191 
  2192 * Scala layer provides JVM method invocation service for static
  2193 methods of type (String)String, see Invoke_Scala.method in ML.  For
  2194 example:
  2195 
  2196   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  2197 
  2198 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  2199 this allows to pass structured values between ML and Scala.
  2200 
  2201 * The IsabelleText fonts includes some further glyphs to support the
  2202 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  2203 installed a local copy (which is normally *not* required) need to
  2204 delete or update it from ~~/lib/fonts/.
  2205 
  2206 
  2207 
  2208 New in Isabelle2011 (January 2011)
  2209 ----------------------------------
  2210 
  2211 *** General ***
  2212 
  2213 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  2214 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  2215 useful tooltips and hyperlinks produced from its static analysis.  The
  2216 bundled component provides an executable Isabelle tool that can be run
  2217 like this:
  2218 
  2219   Isabelle2011/bin/isabelle jedit
  2220 
  2221 * Significantly improved Isabelle/Isar implementation manual.
  2222 
  2223 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  2224 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  2225 while the default heap location within that directory lacks that extra
  2226 suffix.  This isolates multiple Isabelle installations from each
  2227 other, avoiding problems with old settings in new versions.
  2228 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  2229 
  2230 * Source files are always encoded as UTF-8, instead of old-fashioned
  2231 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  2232 the following package declarations:
  2233 
  2234   \usepackage[utf8]{inputenc}
  2235   \usepackage{textcomp}
  2236 
  2237 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  2238 a Unicode character is treated as a single symbol, not a sequence of
  2239 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  2240 contain symbols without further backslash escapes, Unicode can now be
  2241 used here as well.  Recall that Symbol.explode in ML provides a
  2242 consistent view on symbols, while raw explode (or String.explode)
  2243 merely give a byte-oriented representation.
  2244 
  2245 * Theory loader: source files are primarily located via the master
  2246 directory of each theory node (where the .thy file itself resides).
  2247 The global load path is still partially available as legacy feature.
  2248 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  2249 explicit paths, relatively to the theory.
  2250 
  2251 * Special treatment of ML file names has been discontinued.
  2252 Historically, optional extensions .ML or .sml were added on demand --
  2253 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  2254 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  2255 
  2256 * Various options that affect pretty printing etc. are now properly
  2257 handled within the context via configuration options, instead of
  2258 unsynchronized references or print modes.  There are both ML Config.T
  2259 entities and Isar declaration attributes to access these.
  2260 
  2261   ML (Config.T)                 Isar (attribute)
  2262 
  2263   eta_contract                  eta_contract
  2264   show_brackets                 show_brackets
  2265   show_sorts                    show_sorts
  2266   show_types                    show_types
  2267   show_question_marks           show_question_marks
  2268   show_consts                   show_consts
  2269   show_abbrevs                  show_abbrevs
  2270 
  2271   Syntax.ast_trace              syntax_ast_trace
  2272   Syntax.ast_stat               syntax_ast_stat
  2273   Syntax.ambiguity_level        syntax_ambiguity_level
  2274 
  2275   Goal_Display.goals_limit      goals_limit
  2276   Goal_Display.show_main_goal   show_main_goal
  2277 
  2278   Method.rule_trace             rule_trace
  2279 
  2280   Thy_Output.display            thy_output_display
  2281   Thy_Output.quotes             thy_output_quotes
  2282   Thy_Output.indent             thy_output_indent
  2283   Thy_Output.source             thy_output_source
  2284   Thy_Output.break              thy_output_break
  2285 
  2286 Note that corresponding "..._default" references in ML may only be
  2287 changed globally at the ROOT session setup, but *not* within a theory.
  2288 The option "show_abbrevs" supersedes the former print mode
  2289 "no_abbrevs" with inverted meaning.
  2290 
  2291 * More systematic naming of some configuration options.
  2292 INCOMPATIBILITY.
  2293 
  2294   trace_simp  ~>  simp_trace
  2295   debug_simp  ~>  simp_debug
  2296 
  2297 * Support for real valued configuration options, using simplistic
  2298 floating-point notation that coincides with the inner syntax for
  2299 float_token.
  2300 
  2301 * Support for real valued preferences (with approximative PGIP type):
  2302 front-ends need to accept "pgint" values in float notation.
  2303 INCOMPATIBILITY.
  2304 
  2305 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2306 DejaVu Sans.
  2307 
  2308 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2309 
  2310 
  2311 *** Pure ***
  2312 
  2313 * Command 'type_synonym' (with single argument) replaces somewhat
  2314 outdated 'types', which is still available as legacy feature for some
  2315 time.
  2316 
  2317 * Command 'nonterminal' (with 'and' separated list of arguments)
  2318 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2319 
  2320 * Command 'notepad' replaces former 'example_proof' for
  2321 experimentation in Isar without any result.  INCOMPATIBILITY.
  2322 
  2323 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2324 lists of equations to map definitions in a locale to appropriate
  2325 entities in the context of the interpretation.  The 'interpretation'
  2326 command already provided this functionality.
  2327 
  2328 * Diagnostic command 'print_dependencies' prints the locale instances
  2329 that would be activated if the specified expression was interpreted in
  2330 the current context.  Variant "print_dependencies!" assumes a context
  2331 without interpretations.
  2332 
  2333 * Diagnostic command 'print_interps' prints interpretations in proofs
  2334 in addition to interpretations in theories.
  2335 
  2336 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2337 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2338 Sign.root_path and Sign.local_path may be applied directly where this
  2339 feature is still required for historical reasons.
  2340 
  2341 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2342 'definition' instead.
  2343 
  2344 * The "prems" fact, which refers to the accidental collection of
  2345 foundational premises in the context, is now explicitly marked as
  2346 legacy feature and will be discontinued soon.  Consider using "assms"
  2347 of the head statement or reference facts by explicit names.
  2348 
  2349 * Document antiquotations @{class} and @{type} print classes and type
  2350 constructors.
  2351 
  2352 * Document antiquotation @{file} checks file/directory entries within
  2353 the local file system.
  2354 
  2355 
  2356 *** HOL ***
  2357 
  2358 * Coercive subtyping: functions can be declared as coercions and type
  2359 inference will add them as necessary upon input of a term.  Theory
  2360 Complex_Main declares real :: nat => real and real :: int => real as
  2361 coercions. A coercion function f is declared like this:
  2362 
  2363   declare [[coercion f]]
  2364 
  2365 To lift coercions through type constructors (e.g. from nat => real to
  2366 nat list => real list), map functions can be declared, e.g.
  2367 
  2368   declare [[coercion_map map]]
  2369 
  2370 Currently coercion inference is activated only in theories including
  2371 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2372 the configuration option "coercion_enabled", e.g. it can be enabled in
  2373 other theories like this:
  2374 
  2375   declare [[coercion_enabled]]
  2376 
  2377 * Command 'partial_function' provides basic support for recursive
  2378 function definitions over complete partial orders.  Concrete instances
  2379 are provided for i) the option type, ii) tail recursion on arbitrary
  2380 types, and iii) the heap monad of Imperative_HOL.  See
  2381 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2382 for examples.
  2383 
  2384 * Function package: f.psimps rules are no longer implicitly declared
  2385 as [simp].  INCOMPATIBILITY.
  2386 
  2387 * Datatype package: theorems generated for executable equality (class
  2388 "eq") carry proper names and are treated as default code equations.
  2389 
  2390 * Inductive package: now offers command 'inductive_simps' to
  2391 automatically derive instantiated and simplified equations for
  2392 inductive predicates, similar to 'inductive_cases'.
  2393 
  2394 * Command 'enriched_type' allows to register properties of the
  2395 functorial structure of types.
  2396 
  2397 * Improved infrastructure for term evaluation using code generator
  2398 techniques, in particular static evaluation conversions.
  2399 
  2400 * Code generator: Scala (2.8 or higher) has been added to the target
  2401 languages.
  2402 
  2403 * Code generator: globbing constant expressions "*" and "Theory.*"
  2404 have been replaced by the more idiomatic "_" and "Theory._".
  2405 INCOMPATIBILITY.
  2406 
  2407 * Code generator: export_code without explicit file declaration prints
  2408 to standard output.  INCOMPATIBILITY.
  2409 
  2410 * Code generator: do not print function definitions for case
  2411 combinators any longer.
  2412 
  2413 * Code generator: simplification with rules determined with
  2414 src/Tools/Code/code_simp.ML and method "code_simp".
  2415 
  2416 * Code generator for records: more idiomatic representation of record
  2417 types.  Warning: records are not covered by ancient SML code
  2418 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2419 rep_datatype declaration helps to succeed then:
  2420 
  2421   record 'a foo = ...
  2422   ...
  2423   rep_datatype foo_ext ...
  2424 
  2425 * Records: logical foundation type for records does not carry a
  2426 '_type' suffix any longer (obsolete due to authentic syntax).
  2427 INCOMPATIBILITY.
  2428 
  2429 * Quickcheck now by default uses exhaustive testing instead of random
  2430 testing.  Random testing can be invoked by "quickcheck [random]",
  2431 exhaustive testing by "quickcheck [exhaustive]".
  2432 
  2433 * Quickcheck instantiates polymorphic types with small finite
  2434 datatypes by default. This enables a simple execution mechanism to
  2435 handle quantifiers and function equality over the finite datatypes.
  2436 
  2437 * Quickcheck random generator has been renamed from "code" to
  2438 "random".  INCOMPATIBILITY.
  2439 
  2440 * Quickcheck now has a configurable time limit which is set to 30
  2441 seconds by default. This can be changed by adding [timeout = n] to the
  2442 quickcheck command. The time limit for Auto Quickcheck is still set
  2443 independently.
  2444 
  2445 * Quickcheck in locales considers interpretations of that locale for
  2446 counter example search.
  2447 
  2448 * Sledgehammer:
  2449   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2450     method. See the Sledgehammer manual for details ("isabelle doc
  2451     sledgehammer").
  2452   - Renamed commands:
  2453     sledgehammer atp_info ~> sledgehammer running_provers
  2454     sledgehammer atp_kill ~> sledgehammer kill_provers
  2455     sledgehammer available_atps ~> sledgehammer available_provers
  2456     INCOMPATIBILITY.
  2457   - Renamed options:
  2458     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2459     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2460     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2461     (and "ms" and "min" are no longer supported)
  2462     INCOMPATIBILITY.
  2463 
  2464 * Nitpick:
  2465   - Renamed options:
  2466     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2467     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2468     INCOMPATIBILITY.
  2469   - Added support for partial quotient types.
  2470   - Added local versions of the "Nitpick.register_xxx" functions.
  2471   - Added "whack" option.
  2472   - Allow registration of quotient types as codatatypes.
  2473   - Improved "merge_type_vars" option to merge more types.
  2474   - Removed unsound "fast_descrs" option.
  2475   - Added custom symmetry breaking for datatypes, making it possible to reach
  2476     higher cardinalities.
  2477   - Prevent the expansion of too large definitions.
  2478 
  2479 * Proof methods "metis" and "meson" now have configuration options
  2480 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2481 to diagnose these tools. E.g.
  2482 
  2483     using [[metis_trace = true]]
  2484 
  2485 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2486 manually as command 'solve_direct'.
  2487 
  2488 * The default SMT solver Z3 must be enabled explicitly (due to
  2489 licensing issues) by setting the environment variable
  2490 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2491 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2492 changing the SMT solver is done via the configuration option
  2493 "smt_solver".
  2494 
  2495 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2496 i.e. "remote_cvc3" and "remote_z3".
  2497 
  2498 * Added basic SMT support for datatypes, records, and typedefs using
  2499 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2500 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2501 support for a similar behavior).  Minor INCOMPATIBILITY.
  2502 
  2503 * Changed SMT configuration options:
  2504   - Renamed:
  2505     z3_proofs ~> smt_oracle (with inverted meaning)
  2506     z3_trace_assms ~> smt_trace_used_facts
  2507     INCOMPATIBILITY.
  2508   - Added:
  2509     smt_verbose
  2510     smt_random_seed
  2511     smt_datatypes
  2512     smt_infer_triggers
  2513     smt_monomorph_limit
  2514     cvc3_options
  2515     remote_cvc3_options
  2516     remote_z3_options
  2517     yices_options
  2518 
  2519 * Boogie output files (.b2i files) need to be declared in the theory
  2520 header.
  2521 
  2522 * Simplification procedure "list_to_set_comprehension" rewrites list
  2523 comprehensions applied to List.set to set comprehensions.  Occasional
  2524 INCOMPATIBILITY, may be deactivated like this:
  2525 
  2526   declare [[simproc del: list_to_set_comprehension]]
  2527 
  2528 * Removed old version of primrec package.  INCOMPATIBILITY.
  2529 
  2530 * Removed simplifier congruence rule of "prod_case", as has for long
  2531 been the case with "split".  INCOMPATIBILITY.
  2532 
  2533 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2534 
  2535 * Removed [split_format ... and ... and ...] version of
  2536 [split_format].  Potential INCOMPATIBILITY.
  2537 
  2538 * Predicate "sorted" now defined inductively, with nice induction
  2539 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2540 
  2541 * Constant "contents" renamed to "the_elem", to free the generic name
  2542 contents for other uses.  INCOMPATIBILITY.
  2543 
  2544 * Renamed class eq and constant eq (for code generation) to class
  2545 equal and constant equal, plus renaming of related facts and various
  2546 tuning.  INCOMPATIBILITY.
  2547 
  2548 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2549 
  2550 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2551 INCOMPATIBILITY.
  2552 
  2553 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2554 avoid confusion with finite sets.  INCOMPATIBILITY.
  2555 
  2556 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2557 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2558 for congruent(2)).
  2559 
  2560 * Some previously unqualified names have been qualified:
  2561 
  2562   types
  2563     bool ~> HOL.bool
  2564     nat ~> Nat.nat
  2565 
  2566   constants
  2567     Trueprop ~> HOL.Trueprop
  2568     True ~> HOL.True
  2569     False ~> HOL.False
  2570     op & ~> HOL.conj
  2571     op | ~> HOL.disj
  2572     op --> ~> HOL.implies
  2573     op = ~> HOL.eq
  2574     Not ~> HOL.Not
  2575     The ~> HOL.The
  2576     All ~> HOL.All
  2577     Ex ~> HOL.Ex
  2578     Ex1 ~> HOL.Ex1
  2579     Let ~> HOL.Let
  2580     If ~> HOL.If
  2581     Ball ~> Set.Ball
  2582     Bex ~> Set.Bex
  2583     Suc ~> Nat.Suc
  2584     Pair ~> Product_Type.Pair
  2585     fst ~> Product_Type.fst
  2586     snd ~> Product_Type.snd
  2587     curry ~> Product_Type.curry
  2588     op : ~> Set.member
  2589     Collect ~> Set.Collect
  2590 
  2591 INCOMPATIBILITY.
  2592 
  2593 * More canonical naming convention for some fundamental definitions:
  2594 
  2595     bot_bool_eq ~> bot_bool_def
  2596     top_bool_eq ~> top_bool_def
  2597     inf_bool_eq ~> inf_bool_def
  2598     sup_bool_eq ~> sup_bool_def
  2599     bot_fun_eq  ~> bot_fun_def
  2600     top_fun_eq  ~> top_fun_def
  2601     inf_fun_eq  ~> inf_fun_def
  2602     sup_fun_eq  ~> sup_fun_def
  2603 
  2604 INCOMPATIBILITY.
  2605 
  2606 * More stylized fact names:
  2607 
  2608   expand_fun_eq ~> fun_eq_iff
  2609   expand_set_eq ~> set_eq_iff
  2610   set_ext       ~> set_eqI
  2611   nat_number    ~> eval_nat_numeral
  2612 
  2613 INCOMPATIBILITY.
  2614 
  2615 * Refactoring of code-generation specific operations in theory List:
  2616 
  2617   constants
  2618     null ~> List.null
  2619 
  2620   facts
  2621     mem_iff ~> member_def
  2622     null_empty ~> null_def
  2623 
  2624 INCOMPATIBILITY.  Note that these were not supposed to be used
  2625 regularly unless for striking reasons; their main purpose was code
  2626 generation.
  2627 
  2628 Various operations from the Haskell prelude are used for generating
  2629 Haskell code.
  2630 
  2631 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2632 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2633 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2634 
  2635 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2636 "*" and "+" respectively.  INCOMPATIBILITY.
  2637 
  2638 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2639 "Sum_Type.Plus" instead.
  2640 
  2641 * Constant "split" has been merged with constant "prod_case"; names of
  2642 ML functions, facts etc. involving split have been retained so far,
  2643 though.  INCOMPATIBILITY.
  2644 
  2645 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2646 instead.  INCOMPATIBILITY.
  2647 
  2648 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2649 INCOMPATIBILITY.
  2650 
  2651 * Former theory Library/Enum is now part of the HOL-Main image.
  2652 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2653 referred to by its qualified name.
  2654 
  2655   enum    ~>  Enum.enum
  2656   nlists  ~>  Enum.nlists
  2657   product ~>  Enum.product
  2658 
  2659 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2660 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2661 INCOMPATIBILITY.
  2662 
  2663 * Theory Library/SetsAndFunctions has been split into
  2664 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2665 for instance definitions for functions; various improvements.
  2666 INCOMPATIBILITY.
  2667 
  2668 * Theory Library/Multiset provides stable quicksort implementation of
  2669 sort_key.
  2670 
  2671 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2672 INCOMPATIBILITY.
  2673 
  2674 * Session Multivariate_Analysis: introduced a type class for euclidean
  2675 space.  Most theorems are now stated in terms of euclidean spaces
  2676 instead of finite cartesian products.
  2677 
  2678   types
  2679     real ^ 'n ~>  'a::real_vector
  2680               ~>  'a::euclidean_space
  2681               ~>  'a::ordered_euclidean_space
  2682         (depends on your needs)
  2683 
  2684   constants
  2685      _ $ _        ~> _ $$ _
  2686      \<chi> x. _  ~> \<chi>\<chi> x. _
  2687      CARD('n)     ~> DIM('a)
  2688 
  2689 Also note that the indices are now natural numbers and not from some
  2690 finite type. Finite cartesian products of euclidean spaces, products
  2691 of euclidean spaces the real and complex numbers are instantiated to
  2692 be euclidean_spaces.  INCOMPATIBILITY.
  2693 
  2694 * Session Probability: introduced pextreal as positive extended real
  2695 numbers.  Use pextreal as value for measures.  Introduce the
  2696 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2697 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2698 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2699 
  2700 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2701 INCOMPATIBILITY.
  2702 
  2703 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2704 prove verification conditions generated by the SPARK Ada program
  2705 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2706 
  2707 
  2708 *** HOL-Algebra ***
  2709 
  2710 * Theorems for additive ring operations (locale abelian_monoid and
  2711 descendants) are generated by interpretation from their multiplicative
  2712 counterparts.  Names (in particular theorem names) have the mandatory
  2713 qualifier 'add'.  Previous theorem names are redeclared for
  2714 compatibility.
  2715 
  2716 * Structure "int_ring" is now an abbreviation (previously a
  2717 definition).  This fits more natural with advanced interpretations.
  2718 
  2719 
  2720 *** HOLCF ***
  2721 
  2722 * The domain package now runs in definitional mode by default: The
  2723 former command 'new_domain' is now called 'domain'.  To use the domain
  2724 package in its original axiomatic mode, use 'domain (unsafe)'.
  2725 INCOMPATIBILITY.
  2726 
  2727 * The new class "domain" is now the default sort.  Class "predomain"
  2728 is an unpointed version of "domain". Theories can be updated by
  2729 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2730 
  2731   'a::type ~> 'a::countable
  2732   'a::cpo  ~> 'a::predomain
  2733   'a::pcpo ~> 'a::domain
  2734 
  2735 * The old type class "rep" has been superseded by class "domain".
  2736 Accordingly, users of the definitional package must remove any
  2737 "default_sort rep" declarations.  INCOMPATIBILITY.
  2738 
  2739 * The domain package (definitional mode) now supports unpointed
  2740 predomain argument types, as long as they are marked 'lazy'. (Strict
  2741 arguments must be in class "domain".) For example, the following
  2742 domain definition now works:
  2743 
  2744   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2745 
  2746 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2747 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2748 'a option, and 'a list.  Additionally, it configures fixrec and the
  2749 domain package to work with these types.  For example:
  2750 
  2751   fixrec isInl :: "('a + 'b) u -> tr"
  2752     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2753 
  2754   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2755 
  2756 * The "(permissive)" option of fixrec has been replaced with a
  2757 per-equation "(unchecked)" option. See
  2758 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2759 
  2760 * The "bifinite" class no longer fixes a constant "approx"; the class
  2761 now just asserts that such a function exists.  INCOMPATIBILITY.
  2762 
  2763 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2764 defines an embedding of type 'a defl into udom by default; instances
  2765 of "bifinite" and "domain" classes are available in
  2766 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2767 
  2768 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2769 
  2770 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2771 
  2772 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2773 
  2774 * The function "cprod_map" has been renamed to "prod_map".
  2775 INCOMPATIBILITY.
  2776 
  2777 * The monadic bind operator on each powerdomain has new binder syntax
  2778 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2779 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2780 
  2781 * The infix syntax for binary union on each powerdomain has changed
  2782 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2783 syntax.  INCOMPATIBILITY.
  2784 
  2785 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2786 still supported as an input translation.
  2787 
  2788 * Renamed some theorems (the original names are also still available).
  2789 
  2790   expand_fun_below   ~> fun_below_iff
  2791   below_fun_ext      ~> fun_belowI
  2792   expand_cfun_eq     ~> cfun_eq_iff
  2793   ext_cfun           ~> cfun_eqI
  2794   expand_cfun_below  ~> cfun_below_iff
  2795   below_cfun_ext     ~> cfun_belowI
  2796   cont2cont_Rep_CFun ~> cont2cont_APP
  2797 
  2798 * The Abs and Rep functions for various types have changed names.
  2799 Related theorem names have also changed to match. INCOMPATIBILITY.
  2800 
  2801   Rep_CFun  ~> Rep_cfun
  2802   Abs_CFun  ~> Abs_cfun
  2803   Rep_Sprod ~> Rep_sprod
  2804   Abs_Sprod ~> Abs_sprod
  2805   Rep_Ssum  ~> Rep_ssum
  2806   Abs_Ssum  ~> Abs_ssum
  2807 
  2808 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2809 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2810 
  2811 * Various changes to bisimulation/coinduction with domain package:
  2812 
  2813   - Definitions of "bisim" constants no longer mention definedness.
  2814   - With mutual recursion, "bisim" predicate is now curried.
  2815   - With mutual recursion, each type gets a separate coind theorem.
  2816   - Variable names in bisim_def and coinduct rules have changed.
  2817 
  2818 INCOMPATIBILITY.
  2819 
  2820 * Case combinators generated by the domain package for type "foo" are
  2821 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2822 
  2823 * Several theorems have been renamed to more accurately reflect the
  2824 names of constants and types involved.  INCOMPATIBILITY.
  2825 
  2826   thelub_const    ~> lub_const
  2827   lub_const       ~> is_lub_const
  2828   thelubI         ~> lub_eqI
  2829   is_lub_lub      ~> is_lubD2
  2830   lubI            ~> is_lub_lub
  2831   unique_lub      ~> is_lub_unique
  2832   is_ub_lub       ~> is_lub_rangeD1
  2833   lub_bin_chain   ~> is_lub_bin_chain
  2834   lub_fun         ~> is_lub_fun
  2835   thelub_fun      ~> lub_fun
  2836   thelub_cfun     ~> lub_cfun
  2837   thelub_Pair     ~> lub_Pair
  2838   lub_cprod       ~> is_lub_prod
  2839   thelub_cprod    ~> lub_prod
  2840   minimal_cprod   ~> minimal_prod
  2841   inst_cprod_pcpo ~> inst_prod_pcpo
  2842   UU_I            ~> bottomI
  2843   compact_UU      ~> compact_bottom
  2844   deflation_UU    ~> deflation_bottom
  2845   finite_deflation_UU ~> finite_deflation_bottom
  2846 
  2847 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2848 
  2849   sq_ord_less_eq_trans ~> below_eq_trans
  2850   sq_ord_eq_less_trans ~> eq_below_trans
  2851   refl_less            ~> below_refl
  2852   trans_less           ~> below_trans
  2853   antisym_less         ~> below_antisym
  2854   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2855   box_less             ~> box_below
  2856   rev_trans_less       ~> rev_below_trans
  2857   not_less2not_eq      ~> not_below2not_eq
  2858   less_UU_iff          ~> below_UU_iff
  2859   flat_less_iff        ~> flat_below_iff
  2860   adm_less             ~> adm_below
  2861   adm_not_less         ~> adm_not_below
  2862   adm_compact_not_less ~> adm_compact_not_below
  2863   less_fun_def         ~> below_fun_def
  2864   expand_fun_less      ~> fun_below_iff
  2865   less_fun_ext         ~> fun_belowI
  2866   less_discr_def       ~> below_discr_def
  2867   discr_less_eq        ~> discr_below_eq
  2868   less_unit_def        ~> below_unit_def
  2869   less_cprod_def       ~> below_prod_def
  2870   prod_lessI           ~> prod_belowI
  2871   Pair_less_iff        ~> Pair_below_iff
  2872   fst_less_iff         ~> fst_below_iff
  2873   snd_less_iff         ~> snd_below_iff
  2874   expand_cfun_less     ~> cfun_below_iff
  2875   less_cfun_ext        ~> cfun_belowI
  2876   injection_less       ~> injection_below
  2877   less_up_def          ~> below_up_def
  2878   not_Iup_less         ~> not_Iup_below
  2879   Iup_less             ~> Iup_below
  2880   up_less              ~> up_below
  2881   Def_inject_less_eq   ~> Def_below_Def
  2882   Def_less_is_eq       ~> Def_below_iff
  2883   spair_less_iff       ~> spair_below_iff
  2884   less_sprod           ~> below_sprod
  2885   spair_less           ~> spair_below
  2886   sfst_less_iff        ~> sfst_below_iff
  2887   ssnd_less_iff        ~> ssnd_below_iff
  2888   fix_least_less       ~> fix_least_below
  2889   dist_less_one        ~> dist_below_one
  2890   less_ONE             ~> below_ONE
  2891   ONE_less_iff         ~> ONE_below_iff
  2892   less_sinlD           ~> below_sinlD
  2893   less_sinrD           ~> below_sinrD
  2894 
  2895 
  2896 *** FOL and ZF ***
  2897 
  2898 * All constant names are now qualified internally and use proper
  2899 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2900 
  2901 
  2902 *** ML ***
  2903 
  2904 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2905 Fail if the argument is false.  Due to inlining the source position of
  2906 failed assertions is included in the error output.
  2907 
  2908 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2909 text is in practice always evaluated with a stable theory checkpoint.
  2910 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2911 
  2912 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2913 the current context, not any accidental theory loader state as before.
  2914 Potential INCOMPATIBILITY, subtle change in semantics.
  2915 
  2916 * Syntax.pretty_priority (default 0) configures the required priority
  2917 of pretty-printed output and thus affects insertion of parentheses.
  2918 
  2919 * Syntax.default_root (default "any") configures the inner syntax
  2920 category (nonterminal symbol) for parsing of terms.
  2921 
  2922 * Former exception Library.UnequalLengths now coincides with
  2923 ListPair.UnequalLengths.
  2924 
  2925 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2926 main functionality is provided by structure Simplifier.
  2927 
  2928 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2929 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2930 almost all situations.
  2931 
  2932 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2933 See implementation manual for further details on exceptions in
  2934 Isabelle/ML.
  2935 
  2936 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2937 meaning.
  2938 
  2939 * Renamed structure PureThy to Pure_Thy and moved most of its
  2940 operations to structure Global_Theory, to emphasize that this is
  2941 rarely-used global-only stuff.
  2942 
  2943 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2944 instead (or tracing for high-volume output).
  2945 
  2946 * Configuration option show_question_marks only affects regular pretty
  2947 printing of types and terms, not raw Term.string_of_vname.
  2948 
  2949 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2950 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2951 @{thms} for most purposes.
  2952 
  2953 * ML structure Unsynchronized is never opened, not even in Isar
  2954 interaction mode as before.  Old Unsynchronized.set etc. have been
  2955 discontinued -- use plain := instead.  This should be *rare* anyway,
  2956 since modern tools always work via official context data, notably
  2957 configuration options.
  2958 
  2959 * Parallel and asynchronous execution requires special care concerning
  2960 interrupts.  Structure Exn provides some convenience functions that
  2961 avoid working directly with raw Interrupt.  User code must not absorb
  2962 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2963 followed by re-raising of the original exception.  Another common
  2964 source of mistakes are "handle _" patterns, which make the meaning of
  2965 the program subject to physical effects of the environment.
  2966 
  2967 
  2968 
  2969 New in Isabelle2009-2 (June 2010)
  2970 ---------------------------------
  2971 
  2972 *** General ***
  2973 
  2974 * Authentic syntax for *all* logical entities (type classes, type
  2975 constructors, term constants): provides simple and robust
  2976 correspondence between formal entities and concrete syntax.  Within
  2977 the parse tree / AST representations, "constants" are decorated by
  2978 their category (class, type, const) and spelled out explicitly with
  2979 their full internal name.
  2980 
  2981 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2982 and translations (translation rules and translation functions in ML).
  2983 Some hints on upgrading:
  2984 
  2985   - Many existing uses of 'syntax' and 'translations' can be replaced
  2986     by more modern 'type_notation', 'notation' and 'abbreviation',
  2987     which are independent of this issue.
  2988 
  2989   - 'translations' require markup within the AST; the term syntax
  2990     provides the following special forms:
  2991 
  2992       CONST c   -- produces syntax version of constant c from context
  2993       XCONST c  -- literally c, checked as constant from context
  2994       c         -- literally c, if declared by 'syntax'
  2995 
  2996     Plain identifiers are treated as AST variables -- occasionally the
  2997     system indicates accidental variables via the error "rhs contains
  2998     extra variables".
  2999 
  3000     Type classes and type constructors are marked according to their
  3001     concrete syntax.  Some old translations rules need to be written
  3002     for the "type" category, using type constructor application
  3003     instead of pseudo-term application of the default category
  3004     "logic".
  3005 
  3006   - 'parse_translation' etc. in ML may use the following
  3007     antiquotations:
  3008 
  3009       @{class_syntax c}   -- type class c within parse tree / AST
  3010       @{term_syntax c}    -- type constructor c within parse tree / AST
  3011       @{const_syntax c}   -- ML version of "CONST c" above
  3012       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  3013 
  3014   - Literal types within 'typed_print_translations', i.e. those *not*
  3015     represented as pseudo-terms are represented verbatim.  Use @{class
  3016     c} or @{type_name c} here instead of the above syntax
  3017     antiquotations.
  3018 
  3019 Note that old non-authentic syntax was based on unqualified base
  3020 names, so all of the above "constant" names would coincide.  Recall
  3021 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  3022 diagnose syntax problems.
  3023 
  3024 * Type constructors admit general mixfix syntax, not just infix.
  3025 
  3026 * Concrete syntax may be attached to local entities without a proof
  3027 body, too.  This works via regular mixfix annotations for 'fix',
  3028 'def', 'obtain' etc. or via the explicit 'write' command, which is
  3029 similar to the 'notation' command in theory specifications.
  3030 
  3031 * Discontinued unnamed infix syntax (legacy feature for many years) --
  3032 need to specify constant name and syntax separately.  Internal ML
  3033 datatype constructors have been renamed from InfixName to Infix etc.
  3034 Minor INCOMPATIBILITY.
  3035 
  3036 * Schematic theorem statements need to be explicitly markup as such,
  3037 via commands 'schematic_lemma', 'schematic_theorem',
  3038 'schematic_corollary'.  Thus the relevance of the proof is made
  3039 syntactically clear, which impacts performance in a parallel or
  3040 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  3041 
  3042 * Use of cumulative prems via "!" in some proof methods has been
  3043 discontinued (old legacy feature).
  3044 
  3045 * References 'trace_simp' and 'debug_simp' have been replaced by
  3046 configuration options stored in the context. Enabling tracing (the
  3047 case of debugging is similar) in proofs works via
  3048 
  3049   using [[trace_simp = true]]
  3050 
  3051 Tracing is then active for all invocations of the simplifier in
  3052 subsequent goal refinement steps. Tracing may also still be enabled or
  3053 disabled via the ProofGeneral settings menu.
  3054 
  3055 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  3056 'hide_fact' replace the former 'hide' KIND command.  Minor
  3057 INCOMPATIBILITY.
  3058 
  3059 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  3060 more efficient than combinations with -q1 or -q2.
  3061 
  3062 
  3063 *** Pure ***
  3064 
  3065 * Proofterms record type-class reasoning explicitly, using the
  3066 "unconstrain" operation internally.  This eliminates all sort
  3067 constraints from a theorem and proof, introducing explicit
  3068 OFCLASS-premises.  On the proof term level, this operation is
  3069 automatically applied at theorem boundaries, such that closed proofs
  3070 are always free of sort constraints.  INCOMPATIBILITY for tools that
  3071 inspect proof terms.
  3072 
  3073 * Local theory specifications may depend on extra type variables that
  3074 are not present in the result type -- arguments TYPE('a) :: 'a itself
  3075 are added internally.  For example:
  3076 
  3077   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  3078 
  3079 * Predicates of locales introduced by classes carry a mandatory
  3080 "class" prefix.  INCOMPATIBILITY.
  3081 
  3082 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  3083 
  3084 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  3085 'class' instead.
  3086 
  3087 * Command 'code_reflect' allows to incorporate generated ML code into
  3088 runtime environment; replaces immature code_datatype antiquotation.
  3089 INCOMPATIBILITY.
  3090 
  3091 * Code generator: simple concept for abstract datatypes obeying
  3092 invariants.
  3093 
  3094 * Code generator: details of internal data cache have no impact on the
  3095 user space functionality any longer.
  3096 
  3097 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  3098 subgoals.  This is more appropriate for interpretations with 'where'.
  3099 INCOMPATIBILITY.
  3100 
  3101 * Command 'example_proof' opens an empty proof body.  This allows to
  3102 experiment with Isar, without producing any persistent result.
  3103 
  3104 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  3105 within a local theory context, with explicit checking of the
  3106 constructors involved (in contrast to the raw 'syntax' versions).
  3107 
  3108 * Commands 'types' and 'typedecl' now work within a local theory
  3109 context -- without introducing dependencies on parameters or
  3110 assumptions, which is not possible in Isabelle/Pure.
  3111 
  3112 * Command 'defaultsort' has been renamed to 'default_sort', it works
  3113 within a local theory context.  Minor INCOMPATIBILITY.
  3114 
  3115 
  3116 *** HOL ***
  3117 
  3118 * Command 'typedef' now works within a local theory context -- without
  3119 introducing dependencies on parameters or assumptions, which is not
  3120 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  3121 contain multiple interpretations of local typedefs (with different
  3122 non-emptiness proofs), even in a global theory context.
  3123 
  3124 * New package for quotient types.  Commands 'quotient_type' and
  3125 'quotient_definition' may be used for defining types and constants by
  3126 quotient constructions.  An example is the type of integers created by
  3127 quotienting pairs of natural numbers:
  3128 
  3129   fun
  3130     intrel :: "(nat * nat) => (nat * nat) => bool"
  3131   where
  3132     "intrel (x, y) (u, v) = (x + v = u + y)"
  3133 
  3134   quotient_type int = "nat * nat" / intrel
  3135     by (auto simp add: equivp_def expand_fun_eq)
  3136 
  3137   quotient_definition
  3138     "0::int" is "(0::nat, 0::nat)"
  3139 
  3140 The method "lifting" can be used to lift of theorems from the
  3141 underlying "raw" type to the quotient type.  The example
  3142 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  3143 construction and provides a reasoning infrastructure for finite sets.
  3144 
  3145 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  3146 clash with new theory Quotient in Main HOL.
  3147 
  3148 * Moved the SMT binding into the main HOL session, eliminating
  3149 separate HOL-SMT session.
  3150 
  3151 * List membership infix mem operation is only an input abbreviation.
  3152 INCOMPATIBILITY.
  3153 
  3154 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  3155 for future developements; former Library/Word.thy is still present in
  3156 the AFP entry RSAPPS.
  3157 
  3158 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  3159 longer shadowed.  INCOMPATIBILITY.
  3160 
  3161 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  3162 INCOMPATIBILITY.
  3163 
  3164 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  3165 INCOMPATIBILITY.
  3166 
  3167 * Dropped normalizing_semiring etc; use the facts in semiring classes
  3168 instead.  INCOMPATIBILITY.
  3169 
  3170 * Dropped several real-specific versions of lemmas about floor and
  3171 ceiling; use the generic lemmas from theory "Archimedean_Field"
  3172 instead.  INCOMPATIBILITY.
  3173 
  3174   floor_number_of_eq         ~> floor_number_of
  3175   le_floor_eq_number_of      ~> number_of_le_floor
  3176   le_floor_eq_zero           ~> zero_le_floor
  3177   le_floor_eq_one            ~> one_le_floor
  3178   floor_less_eq_number_of    ~> floor_less_number_of
  3179   floor_less_eq_zero         ~> floor_less_zero
  3180   floor_less_eq_one          ~> floor_less_one
  3181   less_floor_eq_number_of    ~> number_of_less_floor
  3182   less_floor_eq_zero         ~> zero_less_floor
  3183   less_floor_eq_one          ~> one_less_floor
  3184   floor_le_eq_number_of      ~> floor_le_number_of
  3185   floor_le_eq_zero           ~> floor_le_zero
  3186   floor_le_eq_one            ~> floor_le_one
  3187   floor_subtract_number_of   ~> floor_diff_number_of
  3188   floor_subtract_one         ~> floor_diff_one
  3189   ceiling_number_of_eq       ~> ceiling_number_of
  3190   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  3191   ceiling_le_zero_eq         ~> ceiling_le_zero
  3192   ceiling_le_eq_one          ~> ceiling_le_one
  3193   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  3194   less_ceiling_eq_zero       ~> zero_less_ceiling
  3195   less_ceiling_eq_one        ~> one_less_ceiling
  3196   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  3197   ceiling_less_eq_zero       ~> ceiling_less_zero
  3198   ceiling_less_eq_one        ~> ceiling_less_one
  3199   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  3200   le_ceiling_eq_zero         ~> zero_le_ceiling
  3201   le_ceiling_eq_one          ~> one_le_ceiling
  3202   ceiling_subtract_number_of ~> ceiling_diff_number_of
  3203   ceiling_subtract_one       ~> ceiling_diff_one
  3204 
  3205 * Theory "Finite_Set": various folding_XXX locales facilitate the
  3206 application of the various fold combinators on finite sets.
  3207 
  3208 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  3209 provides abstract red-black tree type which is backed by "RBT_Impl" as
  3210 implementation.  INCOMPATIBILTY.
  3211 
  3212 * Theory Library/Coinductive_List has been removed -- superseded by
  3213 AFP/thys/Coinductive.
  3214 
  3215 * Theory PReal, including the type "preal" and related operations, has
  3216 been removed.  INCOMPATIBILITY.
  3217 
  3218 * Real: new development using Cauchy Sequences.
  3219 
  3220 * Split off theory "Big_Operators" containing setsum, setprod,
  3221 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  3222 
  3223 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  3224 "Int" etc.  INCOMPATIBILITY.
  3225 
  3226 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  3227 
  3228 * New set of rules "ac_simps" provides combined assoc / commute
  3229 rewrites for all interpretations of the appropriate generic locales.
  3230 
  3231 * Renamed theory "OrderedGroup" to "Groups" and split theory
  3232 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  3233 appropriate and more consistent names suitable for name prefixes
  3234 within the HOL theories.  INCOMPATIBILITY.
  3235 
  3236 * Some generic constants have been put to appropriate theories:
  3237   - less_eq, less: Orderings
  3238   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  3239   - inverse, divide: Rings
  3240 INCOMPATIBILITY.
  3241 
  3242 * More consistent naming of type classes involving orderings (and
  3243 lattices):
  3244 
  3245     lower_semilattice                   ~> semilattice_inf
  3246     upper_semilattice                   ~> semilattice_sup
  3247 
  3248     dense_linear_order                  ~> dense_linorder
  3249 
  3250     pordered_ab_group_add               ~> ordered_ab_group_add
  3251     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  3252     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  3253     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  3254     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  3255     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  3256     pordered_cancel_semiring            ~> ordered_cancel_semiring
  3257     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  3258     pordered_comm_ring                  ~> ordered_comm_ring
  3259     pordered_comm_semiring              ~> ordered_comm_semiring
  3260     pordered_ring                       ~> ordered_ring
  3261     pordered_ring_abs                   ~> ordered_ring_abs
  3262     pordered_semiring                   ~> ordered_semiring
  3263 
  3264     ordered_ab_group_add                ~> linordered_ab_group_add
  3265     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  3266     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  3267     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  3268     ordered_field                       ~> linordered_field
  3269     ordered_field_no_lb                 ~> linordered_field_no_lb
  3270     ordered_field_no_ub                 ~> linordered_field_no_ub
  3271     ordered_field_dense_linear_order    ~> dense_linordered_field
  3272     ordered_idom                        ~> linordered_idom
  3273     ordered_ring                        ~> linordered_ring
  3274     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  3275     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  3276     ordered_ring_strict                 ~> linordered_ring_strict
  3277     ordered_semidom                     ~> linordered_semidom
  3278     ordered_semiring                    ~> linordered_semiring
  3279     ordered_semiring_1                  ~> linordered_semiring_1
  3280     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  3281     ordered_semiring_strict             ~> linordered_semiring_strict
  3282 
  3283   The following slightly odd type classes have been moved to a
  3284   separate theory Library/Lattice_Algebras:
  3285 
  3286     lordered_ab_group_add               ~> lattice_ab_group_add
  3287     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  3288     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  3289     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  3290     lordered_ring                       ~> lattice_ring
  3291 
  3292 INCOMPATIBILITY.
  3293 
  3294 * Refined field classes:
  3295   - classes division_ring_inverse_zero, field_inverse_zero,
  3296     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3297     subsumes former division_by_zero class;
  3298   - numerous lemmas have been ported from field to division_ring.
  3299 INCOMPATIBILITY.
  3300 
  3301 * Refined algebra theorem collections:
  3302   - dropped theorem group group_simps, use algebra_simps instead;
  3303   - dropped theorem group ring_simps, use field_simps instead;
  3304   - proper theorem collection field_simps subsumes former theorem
  3305     groups field_eq_simps and field_simps;
  3306   - dropped lemma eq_minus_self_iff which is a duplicate for
  3307     equal_neg_zero.
  3308 INCOMPATIBILITY.
  3309 
  3310 * Theory Finite_Set and List: some lemmas have been generalized from
  3311 sets to lattices:
  3312 
  3313   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3314   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3315   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3316   union_Union_fold_union        ~> sup_Sup_fold_sup
  3317   Inter_fold_inter              ~> Inf_fold_inf
  3318   Union_fold_union              ~> Sup_fold_sup
  3319   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3320   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3321   INTER_fold_inter              ~> INFI_fold_inf
  3322   UNION_fold_union              ~> SUPR_fold_sup
  3323 
  3324 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3325 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3326 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3327 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3328 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3329 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3330 
  3331 * Reorganized theory Multiset: swapped notation of pointwise and
  3332 multiset order:
  3333 
  3334   - pointwise ordering is instance of class order with standard syntax
  3335     <= and <;
  3336   - multiset ordering has syntax <=# and <#; partial order properties
  3337     are provided by means of interpretation with prefix
  3338     multiset_order;
  3339   - less duplication, less historical organization of sections,
  3340     conversion from associations lists to multisets, rudimentary code
  3341     generation;
  3342   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3343     if needed.
  3344 
  3345 Renamed:
  3346 
  3347   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3348   multi_count_ext  ~>  multiset_ext
  3349   diff_union_inverse2  ~>  diff_union_cancelR
  3350 
  3351 INCOMPATIBILITY.
  3352 
  3353 * Theory Permutation: replaced local "remove" by List.remove1.
  3354 
  3355 * Code generation: ML and OCaml code is decorated with signatures.
  3356 
  3357 * Theory List: added transpose.
  3358 
  3359 * Library/Nat_Bijection.thy is a collection of bijective functions
  3360 between nat and other types, which supersedes the older libraries
  3361 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3362 
  3363   Constants:
  3364   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3365   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3366   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3367   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3368   Countable.pair_encode           ~> prod_encode
  3369   NatIso.prod2nat                 ~> prod_encode
  3370   NatIso.nat2prod                 ~> prod_decode
  3371   NatIso.sum2nat                  ~> sum_encode
  3372   NatIso.nat2sum                  ~> sum_decode
  3373   NatIso.list2nat                 ~> list_encode
  3374   NatIso.nat2list                 ~> list_decode
  3375   NatIso.set2nat                  ~> set_encode
  3376   NatIso.nat2set                  ~> set_decode
  3377 
  3378   Lemmas:
  3379   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3380   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3381   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3382   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3383   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3384   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3385   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3386   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3387   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3388   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3389   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3390   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3391   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3392 
  3393 * Sledgehammer:
  3394   - Renamed ATP commands:
  3395     atp_info     ~> sledgehammer running_atps
  3396     atp_kill     ~> sledgehammer kill_atps
  3397     atp_messages ~> sledgehammer messages
  3398     atp_minimize ~> sledgehammer minimize
  3399     print_atps   ~> sledgehammer available_atps
  3400     INCOMPATIBILITY.
  3401   - Added user's manual ("isabelle doc sledgehammer").
  3402   - Added option syntax and "sledgehammer_params" to customize
  3403     Sledgehammer's behavior.  See the manual for details.
  3404   - Modified the Isar proof reconstruction code so that it produces
  3405     direct proofs rather than proofs by contradiction.  (This feature
  3406     is still experimental.)
  3407   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3408     full-typed mode.
  3409   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3410 
  3411 * Nitpick:
  3412   - Added and implemented "binary_ints" and "bits" options.
  3413   - Added "std" option and implemented support for nonstandard models.
  3414   - Added and implemented "finitize" option to improve the precision
  3415     of infinite datatypes based on a monotonicity analysis.
  3416   - Added support for quotient types.
  3417   - Added support for "specification" and "ax_specification"
  3418     constructs.
  3419   - Added support for local definitions (for "function" and
  3420     "termination" proofs).
  3421   - Added support for term postprocessors.
  3422   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3423   - Improved efficiency of "destroy_constrs" optimization.
  3424   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3425     record getters.
  3426   - Fixed soundness bug related to higher-order constructors.
  3427   - Fixed soundness bug when "full_descrs" is enabled.
  3428   - Improved precision of set constructs.
  3429   - Added "atoms" option.
  3430   - Added cache to speed up repeated Kodkod invocations on the same
  3431     problems.
  3432   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3433     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3434     "SAT4J_Light".  INCOMPATIBILITY.
  3435   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3436     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3437   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3438 
  3439 * Method "induct" now takes instantiations of the form t, where t is not
  3440   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3441   If this is not intended, t has to be enclosed in parentheses.
  3442   By default, the equalities generated by definitional instantiations
  3443   are pre-simplified, which may cause parameters of inductive cases
  3444   to disappear, or may even delete some of the inductive cases.
  3445   Use "induct (no_simp)" instead of "induct" to restore the old
  3446   behaviour. The (no_simp) option is also understood by the "cases"
  3447   and "nominal_induct" methods, which now perform pre-simplification, too.
  3448   INCOMPATIBILITY.
  3449 
  3450 
  3451 *** HOLCF ***
  3452 
  3453 * Variable names in lemmas generated by the domain package have
  3454 changed; the naming scheme is now consistent with the HOL datatype
  3455 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3456 
  3457 * The domain package no longer defines the function "foo_copy" for
  3458 recursive domain "foo".  The reach lemma is now stated directly in
  3459 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3460 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3461 
  3462 * Most definedness lemmas generated by the domain package (previously
  3463 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3464 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3465 Proofs that used definedness lemmas as intro rules may break,
  3466 potential INCOMPATIBILITY.
  3467 
  3468 * Induction and casedist rules generated by the domain package now
  3469 declare proper case_names (one called "bottom", and one named for each
  3470 constructor).  INCOMPATIBILITY.
  3471 
  3472 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3473 rules are generated for each domain, INCOMPATIBILITY.
  3474 
  3475   foo_bar.reach       ~> foo.reach  bar.reach
  3476   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3477 
  3478 * Some lemmas generated by the domain package have been renamed for
  3479 consistency with the datatype package, INCOMPATIBILITY.
  3480 
  3481   foo.ind        ~> foo.induct
  3482   foo.finite_ind ~> foo.finite_induct
  3483   foo.coind      ~> foo.coinduct
  3484   foo.casedist   ~> foo.exhaust
  3485   foo.exhaust    ~> foo.nchotomy
  3486 
  3487 * For consistency with other definition packages, the fixrec package
  3488 now generates qualified theorem names, INCOMPATIBILITY.
  3489 
  3490   foo_simps  ~> foo.simps
  3491   foo_unfold ~> foo.unfold
  3492   foo_induct ~> foo.induct
  3493 
  3494 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3495 method and internal fixrec proofs now use the default simpset instead.
  3496 INCOMPATIBILITY.
  3497 
  3498 * The "contlub" predicate has been removed.  Proof scripts should use
  3499 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3500 
  3501 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3502 
  3503 * The constants cpair, cfst, and csnd have been removed in favor of
  3504 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3505 
  3506 
  3507 *** ML ***
  3508 
  3509 * Antiquotations for basic formal entities:
  3510 
  3511     @{class NAME}         -- type class
  3512     @{class_syntax NAME}  -- syntax representation of the above
  3513 
  3514     @{type_name NAME}     -- logical type
  3515     @{type_abbrev NAME}   -- type abbreviation
  3516     @{nonterminal NAME}   -- type of concrete syntactic category
  3517     @{type_syntax NAME}   -- syntax representation of any of the above
  3518 
  3519     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3520     @{const_abbrev NAME}  -- abbreviated constant
  3521     @{const_syntax NAME}  -- syntax representation of any of the above
  3522 
  3523 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3524 syntax constant (cf. 'syntax' command).
  3525 
  3526 * Antiquotation @{make_string} inlines a function to print arbitrary
  3527 values similar to the ML toplevel.  The result is compiler dependent
  3528 and may fall back on "?" in certain situations.
  3529 
  3530 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3531 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3532 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3533 not work with the asynchronous Isar document model.
  3534 
  3535 * Configuration options now admit dynamic default values, depending on
  3536 the context or even global references.
  3537 
  3538 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3539 uses an efficient external library if available (for Poly/ML).
  3540 
  3541 * Renamed some important ML structures, while keeping the old names
  3542 for some time as aliases within the structure Legacy:
  3543 
  3544   OuterKeyword  ~>  Keyword
  3545   OuterLex      ~>  Token
  3546   OuterParse    ~>  Parse
  3547   OuterSyntax   ~>  Outer_Syntax
  3548   PrintMode     ~>  Print_Mode
  3549   SpecParse     ~>  Parse_Spec
  3550   ThyInfo       ~>  Thy_Info
  3551   ThyLoad       ~>  Thy_Load
  3552   ThyOutput     ~>  Thy_Output
  3553   TypeInfer     ~>  Type_Infer
  3554 
  3555 Note that "open Legacy" simplifies porting of sources, but forgetting
  3556 to remove it again will complicate porting again in the future.
  3557 
  3558 * Most operations that refer to a global context are named
  3559 accordingly, e.g. Simplifier.global_context or
  3560 ProofContext.init_global.  There are some situations where a global
  3561 context actually works, but under normal circumstances one needs to
  3562 pass the proper local context through the code!
  3563 
  3564 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3565 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3566 Pretty.pp argument to merge, which is absent in the standard
  3567 Theory_Data version.
  3568 
  3569 * Sorts.certify_sort and derived "cert" operations for types and terms
  3570 no longer minimize sorts.  Thus certification at the boundary of the
  3571 inference kernel becomes invariant under addition of class relations,
  3572 which is an important monotonicity principle.  Sorts are now minimized
  3573 in the syntax layer only, at the boundary between the end-user and the
  3574 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3575 explicitly in rare situations.
  3576 
  3577 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3578 emphasize that this is in no way a standard operation.
  3579 INCOMPATIBILITY.
  3580 
  3581 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3582 usual for resolution.  Rare INCOMPATIBILITY.
  3583 
  3584 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3585 to emphasize that these only work in a global situation (which is
  3586 quite rare).
  3587 
  3588 * Curried take and drop in library.ML; negative length is interpreted
  3589 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3590 
  3591 * Proof terms: type substitutions on proof constants now use canonical
  3592 order of type variables.  INCOMPATIBILITY for tools working with proof
  3593 terms.
  3594 
  3595 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3596 may no longer carry premises.  User-level specifications are
  3597 transformed accordingly by Thm.add_axiom/add_def.
  3598 
  3599 
  3600 *** System ***
  3601 
  3602 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3603 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3604 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3605 
  3606 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3607 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3608 respectively.
  3609 
  3610 * Old lib/scripts/polyml-platform is superseded by the
  3611 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3612 variant, even on a 64 bit machine.  The following example setting
  3613 prefers 64 bit if available:
  3614 
  3615   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3616 
  3617 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3618 Isabelle/Scala layer for advanced prover interaction and integration.
  3619 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3620 component.
  3621 
  3622 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3623 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3624 similar to the default assignment of the document preparation system
  3625 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3626 provides some operations for direct access to the font without asking
  3627 the user for manual installation.
  3628 
  3629 
  3630 
  3631 New in Isabelle2009-1 (December 2009)
  3632 -------------------------------------
  3633 
  3634 *** General ***
  3635 
  3636 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3637 one backslash should be used, even in ML sources.
  3638 
  3639 
  3640 *** Pure ***
  3641 
  3642 * Locale interpretation propagates mixins along the locale hierarchy.
  3643 The currently only available mixins are the equations used to map
  3644 local definitions to terms of the target domain of an interpretation.
  3645 
  3646 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3647 loc" to print all interpretations of locale "loc" in the theory.
  3648 Interpretations in proofs are not shown.
  3649 
  3650 * Thoroughly revised locales tutorial.  New section on conditional
  3651 interpretation.
  3652 
  3653 * On instantiation of classes, remaining undefined class parameters
  3654 are formally declared.  INCOMPATIBILITY.
  3655 
  3656 
  3657 *** Document preparation ***
  3658 
  3659 * New generalized style concept for printing terms: @{foo (style) ...}
  3660 instead of @{foo_style style ...}  (old form is still retained for
  3661 backward compatibility).  Styles can be also applied for
  3662 antiquotations prop, term_type and typeof.
  3663 
  3664 
  3665 *** HOL ***
  3666 
  3667 * New proof method "smt" for a combination of first-order logic with
  3668 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3669 fixed-size bitvectors; there is also basic support for higher-order
  3670 features (esp. lambda abstractions).  It is an incomplete decision
  3671 procedure based on external SMT solvers using the oracle mechanism;
  3672 for the SMT solver Z3, this method is proof-producing.  Certificates
  3673 are provided to avoid calling the external solvers solely for
  3674 re-checking proofs.  Due to a remote SMT service there is no need for
  3675 installing SMT solvers locally.  See src/HOL/SMT.
  3676 
  3677 * New commands to load and prove verification conditions generated by
  3678 the Boogie program verifier or derived systems (e.g. the Verifying C
  3679 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3680 
  3681 * New counterexample generator tool 'nitpick' based on the Kodkod
  3682 relational model finder.  See src/HOL/Tools/Nitpick and
  3683 src/HOL/Nitpick_Examples.
  3684 
  3685 * New commands 'code_pred' and 'values' to invoke the predicate
  3686 compiler and to enumerate values of inductive predicates.
  3687 
  3688 * A tabled implementation of the reflexive transitive closure.
  3689 
  3690 * New implementation of quickcheck uses generic code generator;
  3691 default generators are provided for all suitable HOL types, records
  3692 and datatypes.  Old quickcheck can be re-activated importing theory
  3693 Library/SML_Quickcheck.
  3694 
  3695 * New testing tool Mirabelle for automated proof tools.  Applies
  3696 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3697 every proof step in a theory.  To be used in batch mode via the
  3698 "mirabelle" utility.
  3699 
  3700 * New proof method "sos" (sum of squares) for nonlinear real
  3701 arithmetic (originally due to John Harison). It requires theory
  3702 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3703 works well in practice on quantifier-free real arithmetic with +, -,
  3704 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3705 inequalities between polynomials.  It makes use of external
  3706 semidefinite programming solvers.  Method "sos" generates a
  3707 certificate that can be pasted into the proof thus avoiding the need
  3708 to call an external tool every time the proof is checked.  See
  3709 src/HOL/Library/Sum_Of_Squares.
  3710 
  3711 * New method "linarith" invokes existing linear arithmetic decision
  3712 procedure only.
  3713 
  3714 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3715 
  3716 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3717 Causes full type information to be output to the ATPs.  This slows
  3718 ATPs down considerably but eliminates a source of unsound "proofs"
  3719 that fail later.
  3720 
  3721 * New method "metisFT": A version of metis that uses full type
  3722 information in order to avoid failures of proof reconstruction.
  3723 
  3724 * New evaluator "approximate" approximates an real valued term using
  3725 the same method as the approximation method.
  3726 
  3727 * Method "approximate" now supports arithmetic expressions as
  3728 boundaries of intervals and implements interval splitting and Taylor
  3729 series expansion.
  3730 
  3731 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3732 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3733 
  3734 * New theory SupInf of the supremum and infimum operators for sets of
  3735 reals.
  3736 
  3737 * New theory Probability, which contains a development of measure
  3738 theory, eventually leading to Lebesgue integration and probability.
  3739 
  3740 * Extended Multivariate Analysis to include derivation and Brouwer's
  3741 fixpoint theorem.
  3742 
  3743 * Reorganization of number theory, INCOMPATIBILITY:
  3744   - new number theory development for nat and int, in theories Divides
  3745     and GCD as well as in new session Number_Theory
  3746   - some constants and facts now suffixed with _nat and _int
  3747     accordingly
  3748   - former session NumberTheory now named Old_Number_Theory, including
  3749     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3750   - moved theory Pocklington from src/HOL/Library to
  3751     src/HOL/Old_Number_Theory
  3752 
  3753 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3754 lcm of finite and infinite sets. It is shown that they form a complete
  3755 lattice.
  3756 
  3757 * Class semiring_div requires superclass no_zero_divisors and proof of
  3758 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3759 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3760 generalized to class semiring_div, subsuming former theorems
  3761 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3762 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3763 INCOMPATIBILITY.
  3764 
  3765 * Refinements to lattice classes and sets:
  3766   - less default intro/elim rules in locale variant, more default
  3767     intro/elim rules in class variant: more uniformity
  3768   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3769     le_inf_iff
  3770   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3771     sup_aci)
  3772   - renamed ACI to inf_sup_aci
  3773   - new class "boolean_algebra"
  3774   - class "complete_lattice" moved to separate theory
  3775     "Complete_Lattice"; corresponding constants (and abbreviations)
  3776     renamed and with authentic syntax:
  3777     Set.Inf ~>    Complete_Lattice.Inf
  3778     Set.Sup ~>    Complete_Lattice.Sup
  3779     Set.INFI ~>   Complete_Lattice.INFI
  3780     Set.SUPR ~>   Complete_Lattice.SUPR
  3781     Set.Inter ~>  Complete_Lattice.Inter
  3782     Set.Union ~>  Complete_Lattice.Union
  3783     Set.INTER ~>  Complete_Lattice.INTER
  3784     Set.UNION ~>  Complete_Lattice.UNION
  3785   - authentic syntax for
  3786     Set.Pow
  3787     Set.image
  3788   - mere abbreviations:
  3789     Set.empty               (for bot)
  3790     Set.UNIV                (for top)
  3791     Set.inter               (for inf, formerly Set.Int)
  3792     Set.union               (for sup, formerly Set.Un)
  3793     Complete_Lattice.Inter  (for Inf)
  3794     Complete_Lattice.Union  (for Sup)
  3795     Complete_Lattice.INTER  (for INFI)
  3796     Complete_Lattice.UNION  (for SUPR)
  3797   - object-logic definitions as far as appropriate
  3798 
  3799 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3800 Un_subset_iff are explicitly deleted as default simp rules; then also
  3801 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3802 deleted to achieve the desired effect.
  3803 
  3804 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3805 rules by default any longer; the same applies to min_max.inf_absorb1
  3806 etc.  INCOMPATIBILITY.
  3807 
  3808 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3809 pred_set_conv by default.  INCOMPATIBILITY.
  3810 
  3811 * Power operations on relations and functions are now one dedicated
  3812 constant "compow" with infix syntax "^^".  Power operation on
  3813 multiplicative monoids retains syntax "^" and is now defined generic
  3814 in class power.  INCOMPATIBILITY.
  3815 
  3816 * Relation composition "R O S" now has a more standard argument order:
  3817 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3818 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3819 break, since the O_assoc rule was not rewritten like this.  Fix using
  3820 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3821 
  3822 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3823 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3824 INCOMPATIBILITY.
  3825 
  3826 * Most rules produced by inductive and datatype package have mandatory
  3827 prefixes.  INCOMPATIBILITY.
  3828 
  3829 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3830 the attribute of the same name.  Each of the theorems in the list
  3831 DERIV_intros assumes composition with an additional function and
  3832 matches a variable to the derivative, which has to be solved by the
  3833 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3834 of most elementary terms.  Former Maclauren.DERIV_tac and
  3835 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3836 INCOMPATIBILITY.
  3837 
  3838 * Code generator attributes follow the usual underscore convention:
  3839     code_unfold     replaces    code unfold
  3840     code_post       replaces    code post
  3841     etc.
  3842   INCOMPATIBILITY.
  3843 
  3844 * Renamed methods:
  3845     sizechange -> size_change
  3846     induct_scheme -> induction_schema
  3847   INCOMPATIBILITY.
  3848 
  3849 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3850 INCOMPATIBILITY, use "undefined" directly.
  3851 
  3852 * Renamed theorems:
  3853     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3854     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3855     Suc_plus1 -> Suc_eq_plus1
  3856     *anti_sym -> *antisym*
  3857     vector_less_eq_def -> vector_le_def
  3858   INCOMPATIBILITY.
  3859 
  3860 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3861 INCOMPATIBILITY.
  3862 
  3863 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3864 INCOMPATIBILITY.
  3865 
  3866 
  3867 *** HOLCF ***
  3868 
  3869 * Theory Representable defines a class "rep" of domains that are
  3870 representable (via an ep-pair) in the universal domain type "udom".
  3871 Instances are provided for all type constructors defined in HOLCF.
  3872 
  3873 * The 'new_domain' command is a purely definitional version of the
  3874 domain package, for representable domains.  Syntax is identical to the
  3875 old domain package.  The 'new_domain' package also supports indirect
  3876 recursion using previously-defined type constructors.  See
  3877 src/HOLCF/ex/New_Domain.thy for examples.
  3878 
  3879 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3880 on the left-hand side of an equation, and then performs
  3881 simplification.  Rewriting is done using rules declared with the
  3882 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3883 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3884 
  3885 * The pattern-match compiler in 'fixrec' can now handle constructors
  3886 with HOL function types.  Pattern-match combinators for the Pair
  3887 constructor are pre-configured.
  3888 
  3889 * The 'fixrec' package now produces better fixed-point induction rules
  3890 for mutually-recursive definitions:  Induction rules have conclusions
  3891 of the form "P foo bar" instead of "P <foo, bar>".
  3892 
  3893 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3894 been renamed to "below".  The name "below" now replaces "less" in many
  3895 theorem names.  (Legacy theorem names using "less" are still supported
  3896 as well.)
  3897 
  3898 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3899 patterns can be used to generate strictness rules, or to make
  3900 functions more strict (much like the bang-patterns supported by the
  3901 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3902 examples.
  3903 
  3904 
  3905 *** ML ***
  3906 
  3907 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3908 errors and run-time exceptions, including detailed source positions.
  3909 
  3910 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3911 identified entries, with some additional information such as source
  3912 position, logical grouping etc.
  3913 
  3914 * Theory and context data is now introduced by the simplified and
  3915 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3916 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3917 copy operation) is still available for some time.
  3918 
  3919 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3920 provides a high-level programming interface to synchronized state
  3921 variables with atomic update.  This works via pure function
  3922 application within a critical section -- its runtime should be as
  3923 short as possible; beware of deadlocks if critical code is nested,
  3924 either directly or indirectly via other synchronized variables!
  3925 
  3926 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3927 wraps raw ML references, explicitly indicating their non-thread-safe
  3928 behaviour.  The Isar toplevel keeps this structure open, to
  3929 accommodate Proof General as well as quick and dirty interactive
  3930 experiments with references.
  3931 
  3932 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3933 parallel tactical reasoning.
  3934 
  3935 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3936 are similar to SUBPROOF, but are slightly more flexible: only the
  3937 specified parts of the subgoal are imported into the context, and the
  3938 body tactic may introduce new subgoals and schematic variables.
  3939 
  3940 * Old tactical METAHYPS, which does not observe the proof context, has
  3941 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3942 or Subgoal.FOCUS etc.
  3943 
  3944 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3945 functors have their own ML name space there is no point to mark them
  3946 separately.)  Minor INCOMPATIBILITY.
  3947 
  3948 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3949 
  3950 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3951 INCOMPATIBILITY.
  3952 
  3953 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3954 mode, which avoids critical setmp.
  3955 
  3956 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3957 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3958 Attrib/Method.setup introduced in Isabelle2009.
  3959 
  3960 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3961 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3962 last resort.  INCOMPATIBILITY.
  3963 
  3964 * Display.pretty_thm now requires a proper context (cf. former
  3965 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3966 or even Display.pretty_thm_without_context as last resort.
  3967 INCOMPATIBILITY.
  3968 
  3969 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3970 Syntax.pretty_typ/term directly, preferably with proper context
  3971 instead of global theory.
  3972 
  3973 
  3974 *** System ***
  3975 
  3976 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3977 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3978 usedir option -q.
  3979 
  3980 * Support for additional "Isabelle components" via etc/components, see
  3981 also the system manual.
  3982 
  3983 * The isabelle makeall tool now operates on all components with
  3984 IsaMakefile, not just hardwired "logics".
  3985 
  3986 * Removed "compress" option from isabelle-process and isabelle usedir;
  3987 this is always enabled.
  3988 
  3989 * Discontinued support for Poly/ML 4.x versions.
  3990 
  3991 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3992 on a given logic image.  This requires the lighttpd webserver and is
  3993 currently supported on Linux only.
  3994 
  3995 
  3996 
  3997 New in Isabelle2009 (April 2009)
  3998 --------------------------------
  3999 
  4000 *** General ***
  4001 
  4002 * Simplified main Isabelle executables, with less surprises on
  4003 case-insensitive file-systems (such as Mac OS).
  4004 
  4005   - The main Isabelle tool wrapper is now called "isabelle" instead of
  4006     "isatool."
  4007 
  4008   - The former "isabelle" alias for "isabelle-process" has been
  4009     removed (should rarely occur to regular users).
  4010 
  4011   - The former "isabelle-interface" and its alias "Isabelle" have been
  4012     removed (interfaces are now regular Isabelle tools).
  4013 
  4014 Within scripts and make files, the Isabelle environment variables
  4015 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  4016 respectively.  (The latter are still available as legacy feature.)
  4017 
  4018 The old isabelle-interface wrapper could react in confusing ways if
  4019 the interface was uninstalled or changed otherwise.  Individual
  4020 interface tool configuration is now more explicit, see also the
  4021 Isabelle system manual.  In particular, Proof General is now available
  4022 via "isabelle emacs".
  4023 
  4024 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  4025 purge installed copies of Isabelle executables and re-run "isabelle
  4026 install -p ...", or use symlinks.
  4027 
  4028 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  4029 old ~/isabelle, which was slightly non-standard and apt to cause
  4030 surprises on case-insensitive file-systems (such as Mac OS).
  4031 
  4032 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  4033 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  4034 care is required when using older releases of Isabelle.  Note that
  4035 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  4036 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  4037 
  4038 * Proofs of fully specified statements are run in parallel on
  4039 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  4040 a regular 4-core machine, if the initial heap space is made reasonably
  4041 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  4042 
  4043 * The main reference manuals ("isar-ref", "implementation", and
  4044 "system") have been updated and extended.  Formally checked references
  4045 as hyperlinks are now available uniformly.
  4046 
  4047 
  4048 *** Pure ***
  4049 
  4050 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  4051 respects.  The most important changes are listed below.  See the
  4052 Tutorial on Locales ("locales" manual) for details.
  4053 
  4054 - In locale expressions, instantiation replaces renaming.  Parameters
  4055 must be declared in a for clause.  To aid compatibility with previous
  4056 parameter inheritance, in locale declarations, parameters that are not
  4057 'touched' (instantiation position "_" or omitted) are implicitly added
  4058 with their syntax at the beginning of the for clause.
  4059 
  4060 - Syntax from abbreviations and definitions in locales is available in
  4061 locale expressions and context elements.  The latter is particularly
  4062 useful in locale declarations.
  4063 
  4064 - More flexible mechanisms to qualify names generated by locale
  4065 expressions.  Qualifiers (prefixes) may be specified in locale
  4066 expressions, and can be marked as mandatory (syntax: "name!:") or
  4067 optional (syntax "name?:").  The default depends for plain "name:"
  4068 depends on the situation where a locale expression is used: in
  4069 commands 'locale' and 'sublocale' prefixes are optional, in
  4070 'interpretation' and 'interpret' prefixes are mandatory.  The old
  4071 implicit qualifiers derived from the parameter names of a locale are
  4072 no longer generated.
  4073 
  4074 - Command "sublocale l < e" replaces "interpretation l < e".  The
  4075 instantiation clause in "interpretation" and "interpret" (square
  4076 brackets) is no longer available.  Use locale expressions.
  4077 
  4078 - When converting proof scripts, mandatory qualifiers in
  4079 'interpretation' and 'interpret' should be retained by default, even
  4080 if this is an INCOMPATIBILITY compared to former behavior.  In the
  4081 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  4082 in locale expressions range over a single locale instance only.
  4083 
  4084 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  4085 In existing theorem specifications replace the includes element by the
  4086 respective context elements of the included locale, omitting those
  4087 that are already present in the theorem specification.  Multiple
  4088 assume elements of a locale should be replaced by a single one
  4089 involving the locale predicate.  In the proof body, declarations (most
  4090 notably theorems) may be regained by interpreting the respective
  4091 locales in the proof context as required (command "interpret").
  4092 
  4093 If using "includes" in replacement of a target solely because the
  4094 parameter types in the theorem are not as general as in the target,
  4095 consider declaring a new locale with additional type constraints on
  4096 the parameters (context element "constrains").
  4097 
  4098 - Discontinued "locale (open)".  INCOMPATIBILITY.
  4099 
  4100 - Locale interpretation commands no longer attempt to simplify goal.
  4101 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  4102 methods intro_locales and unfold_locales to clarify.
  4103 
  4104 - Locale interpretation commands no longer accept interpretation
  4105 attributes.  INCOMPATIBILITY.
  4106 
  4107 * Class declaration: so-called "base sort" must not be given in import
  4108 list any longer, but is inferred from the specification.  Particularly
  4109 in HOL, write
  4110 
  4111     class foo = ...
  4112 
  4113 instead of
  4114 
  4115     class foo = type + ...
  4116 
  4117 * Class target: global versions of theorems stemming do not carry a
  4118 parameter prefix any longer.  INCOMPATIBILITY.
  4119 
  4120 * Class 'instance' command no longer accepts attached definitions.
  4121 INCOMPATIBILITY, use proper 'instantiation' target instead.
  4122 
  4123 * Recovered hiding of consts, which was accidentally broken in
  4124 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  4125 makes c inaccessible; consider using ``hide (open) const c'' instead.
  4126 
  4127 * Slightly more coherent Pure syntax, with updated documentation in
  4128 isar-ref manual.  Removed locales meta_term_syntax and
  4129 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  4130 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  4131 directly in regular applications.
  4132 
  4133 * There is a new syntactic category "float_const" for signed decimal
  4134 fractions (e.g. 123.45 or -123.45).
  4135 
  4136 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  4137 interface with 'setup' command instead.
  4138 
  4139 * Command 'local_setup' is similar to 'setup', but operates on a local
  4140 theory context.
  4141 
  4142 * The 'axiomatization' command now only works within a global theory
  4143 context.  INCOMPATIBILITY.
  4144 
  4145 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  4146 hypotheses.  Sorts required in the course of reasoning need to be
  4147 covered by the constraints in the initial statement, completed by the
  4148 type instance information of the background theory.  Non-trivial sort
  4149 hypotheses, which rarely occur in practice, may be specified via
  4150 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  4151 
  4152   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  4153 
  4154 The result contains an implicit sort hypotheses as before --
  4155 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  4156 normalization.
  4157 
  4158 * Generalized Isar history, with support for linear undo, direct state
  4159 addressing etc.
  4160 
  4161 * Changed defaults for unify configuration options:
  4162 
  4163   unify_trace_bound = 50 (formerly 25)
  4164   unify_search_bound = 60 (formerly 30)
  4165 
  4166 * Different bookkeeping for code equations (INCOMPATIBILITY):
  4167 
  4168   a) On theory merge, the last set of code equations for a particular
  4169      constant is taken (in accordance with the policy applied by other
  4170      parts of the code generator framework).
  4171 
  4172   b) Code equations stemming from explicit declarations (e.g. code
  4173      attribute) gain priority over default code equations stemming
  4174      from definition, primrec, fun etc.
  4175 
  4176 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  4177 
  4178 * Unified theorem tables for both code generators.  Thus [code
  4179 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  4180 
  4181 * Command 'find_consts' searches for constants based on type and name
  4182 patterns, e.g.
  4183 
  4184     find_consts "_ => bool"
  4185 
  4186 By default, matching is against subtypes, but it may be restricted to
  4187 the whole type.  Searching by name is possible.  Multiple queries are
  4188 conjunctive and queries may be negated by prefixing them with a
  4189 hyphen:
  4190 
  4191     find_consts strict: "_ => bool" name: "Int" -"int => int"
  4192 
  4193 * New 'find_theorems' criterion "solves" matches theorems that
  4194 directly solve the current goal (modulo higher-order unification).
  4195 
  4196 * Auto solve feature for main theorem statements: whenever a new goal
  4197 is stated, "find_theorems solves" is called; any theorems that could
  4198 solve the lemma directly are listed as part of the goal state.
  4199 Cf. associated options in Proof General Isabelle settings menu,
  4200 enabled by default, with reasonable timeout for pathological cases of
  4201 higher-order unification.
  4202 
  4203 
  4204 *** Document preparation ***
  4205 
  4206 * Antiquotation @{lemma} now imitates a regular terminal proof,
  4207 demanding keyword 'by' and supporting the full method expression
  4208 syntax just like the Isar command 'by'.
  4209 
  4210 
  4211 *** HOL ***
  4212 
  4213 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  4214 points Main and Complex_Main remain as before.
  4215 
  4216 * Logic image HOL-Plain provides a minimal HOL with the most important
  4217 tools available (inductive, datatype, primrec, ...).  This facilitates
  4218 experimentation and tool development.  Note that user applications
  4219 (and library theories) should never refer to anything below theory
  4220 Main, as before.
  4221 
  4222 * Logic image HOL-Main stops at theory Main, and thus facilitates
  4223 experimentation due to shorter build times.
  4224 
  4225 * Logic image HOL-NSA contains theories of nonstandard analysis which
  4226 were previously part of former HOL-Complex.  Entry point Hyperreal
  4227 remains valid, but theories formerly using Complex_Main should now use
  4228 new entry point Hypercomplex.
  4229 
  4230 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  4231 Posix processes.  Avoids potentially expensive forking of the ML
  4232 process.  New thread-based implementation also works on non-Unix
  4233 platforms (Cygwin).  Provers are no longer hardwired, but defined
  4234 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  4235 commands are covered in the isar-ref manual.
  4236 
  4237 * Wrapper scripts for remote SystemOnTPTP service allows to use
  4238 sledgehammer without local ATP installation (Vampire etc.). Other
  4239 provers may be included via suitable ML wrappers, see also
  4240 src/HOL/ATP_Linkup.thy.
  4241 
  4242 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  4243 menu.
  4244 
  4245 * The metis method no longer fails because the theorem is too trivial
  4246 (contains the empty clause).
  4247 
  4248 * The metis method now fails in the usual manner, rather than raising
  4249 an exception, if it determines that it cannot prove the theorem.
  4250 
  4251 * Method "coherent" implements a prover for coherent logic (see also
  4252 src/Tools/coherent.ML).
  4253 
  4254 * Constants "undefined" and "default" replace "arbitrary".  Usually
  4255 "undefined" is the right choice to replace "arbitrary", though
  4256 logically there is no difference.  INCOMPATIBILITY.
  4257 
  4258 * Command "value" now integrates different evaluation mechanisms.  The
  4259 result of the first successful evaluation mechanism is printed.  In
  4260 square brackets a particular named evaluation mechanisms may be
  4261 specified (currently, [SML], [code] or [nbe]).  See further
  4262 src/HOL/ex/Eval_Examples.thy.
  4263 
  4264 * Normalization by evaluation now allows non-leftlinear equations.
  4265 Declare with attribute [code nbe].
  4266 
  4267 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  4268 as the structured Isar versions "cases" and "induct", cf. the
  4269 corresponding "cases" and "induct" attributes.  Mutual induction rules
  4270 are now presented as a list of individual projections
  4271 (e.g. foo_bar.inducts for types foo and bar); the old format with
  4272 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  4273 rare situations a different rule is selected --- notably nested tuple
  4274 elimination instead of former prod.exhaust: use explicit (case_tac t
  4275 rule: prod.exhaust) here.
  4276 
  4277 * Attributes "cases", "induct", "coinduct" support "del" option.
  4278 
  4279 * Removed fact "case_split_thm", which duplicates "case_split".
  4280 
  4281 * The option datatype has been moved to a new theory Option.  Renamed
  4282 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  4283 
  4284 * New predicate "strict_mono" classifies strict functions on partial
  4285 orders.  With strict functions on linear orders, reasoning about
  4286 (in)equalities is facilitated by theorems "strict_mono_eq",
  4287 "strict_mono_less_eq" and "strict_mono_less".
  4288 
  4289 * Some set operations are now proper qualified constants with
  4290 authentic syntax.  INCOMPATIBILITY:
  4291 
  4292     op Int ~>   Set.Int
  4293     op Un ~>    Set.Un
  4294     INTER ~>    Set.INTER
  4295     UNION ~>    Set.UNION
  4296     Inter ~>    Set.Inter
  4297     Union ~>    Set.Union
  4298     {} ~>       Set.empty
  4299     UNIV ~>     Set.UNIV
  4300 
  4301 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4302 theory Set.
  4303 
  4304 * Auxiliary class "itself" has disappeared -- classes without any
  4305 parameter are treated as expected by the 'class' command.
  4306 
  4307 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4308 
  4309 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4310 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4311 
  4312 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4313 method "approximation".  It proves formulas on real values by using
  4314 interval arithmetic.  In the formulas are also the transcendental
  4315 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4316 allowed. For examples see
  4317 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4318 
  4319 * Theory "Reflection" now resides in HOL/Library.
  4320 
  4321 * Entry point to Word library now simply named "Word".
  4322 INCOMPATIBILITY.
  4323 
  4324 * Made source layout more coherent with logical distribution
  4325 structure:
  4326 
  4327     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4328     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4329     src/HOL/Library/GCD.thy ~> src/HOL/
  4330     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4331     src/HOL/Library/Parity.thy ~> src/HOL/
  4332     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4333     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4334     src/HOL/Real/Lubs.thy ~> src/HOL/
  4335     src/HOL/Real/PReal.thy ~> src/HOL/
  4336     src/HOL/Real/Rational.thy ~> src/HOL/
  4337     src/HOL/Real/RComplete.thy ~> src/HOL/
  4338     src/HOL/Real/RealDef.thy ~> src/HOL/
  4339     src/HOL/Real/RealPow.thy ~> src/HOL/
  4340     src/HOL/Real/Real.thy ~> src/HOL/
  4341     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4342     src/HOL/Complex/Complex.thy ~> src/HOL/
  4343     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4344     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4345     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4346     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4347     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4348     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4349     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4350     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4351     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4352     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4353     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4354     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4355     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4356     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4357     src/HOL/Real/Float ~> src/HOL/Library/
  4358     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4359     src/HOL/Real/RealVector.thy ~> src/HOL/
  4360 
  4361     src/HOL/arith_data.ML ~> src/HOL/Tools
  4362     src/HOL/hologic.ML ~> src/HOL/Tools
  4363     src/HOL/simpdata.ML ~> src/HOL/Tools
  4364     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4365     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4366     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4367     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4368     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4369     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4370     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4371 
  4372     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4373     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4374     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4375     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4376     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4377     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4378 
  4379 * If methods "eval" and "evaluation" encounter a structured proof
  4380 state with !!/==>, only the conclusion is evaluated to True (if
  4381 possible), avoiding strange error messages.
  4382 
  4383 * Method "sizechange" automates termination proofs using (a
  4384 modification of) the size-change principle.  Requires SAT solver.  See
  4385 src/HOL/ex/Termination.thy for examples.
  4386 
  4387 * Simplifier: simproc for let expressions now unfolds if bound
  4388 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4389 
  4390 * Method "arith": Linear arithmetic now ignores all inequalities when
  4391 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4392 
  4393 * New attribute "arith" for facts that should always be used
  4394 automatically by arithmetic. It is intended to be used locally in
  4395 proofs, e.g.
  4396 
  4397   assumes [arith]: "x > 0"
  4398 
  4399 Global usage is discouraged because of possible performance impact.
  4400 
  4401 * New classes "top" and "bot" with corresponding operations "top" and
  4402 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4403 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4404 
  4405 * Changed definition lemma "less_fun_def" in order to provide an
  4406 instance for preorders on functions; use lemma "less_le" instead.
  4407 INCOMPATIBILITY.
  4408 
  4409 * Theory Orderings: class "wellorder" moved here, with explicit
  4410 induction rule "less_induct" as assumption.  For instantiation of
  4411 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4412 INCOMPATIBILITY.
  4413 
  4414 * Theory Orderings: added class "preorder" as superclass of "order".
  4415 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4416 etc. slightly changed.  Some theorems named order_class.* now named
  4417 preorder_class.*.
  4418 
  4419 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4420 "diag" to "Id_on".
  4421 
  4422 * Theory Finite_Set: added a new fold combinator of type
  4423 
  4424   ('a => 'b => 'b) => 'b => 'a set => 'b
  4425 
  4426 Occasionally this is more convenient than the old fold combinator
  4427 which is now defined in terms of the new one and renamed to
  4428 fold_image.
  4429 
  4430 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4431 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4432 extended with further lemmas!).  At the moment both still exist but
  4433 the former will disappear at some point.
  4434 
  4435 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4436 class recpower.  Type-specific simp rules for various recpower types
  4437 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4438 
  4439 rat_power_0    -> power_0
  4440 rat_power_Suc  -> power_Suc
  4441 realpow_0      -> power_0
  4442 realpow_Suc    -> power_Suc
  4443 complexpow_0   -> power_0
  4444 complexpow_Suc -> power_Suc
  4445 power_poly_0   -> power_0
  4446 power_poly_Suc -> power_Suc
  4447 
  4448 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4449 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4450 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4451 of lemmas from Divides has been generalized from nat to class
  4452 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4453 renames resulting from duplicate elimination:
  4454 
  4455     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4456     zero_dvd_iff ~>         dvd_0_left_iff
  4457     dvd_0 ~>                dvd_0_right
  4458     DIVISION_BY_ZERO_DIV ~> div_by_0
  4459     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4460     mult_div ~>             div_mult_self2_is_id
  4461     mult_mod ~>             mod_mult_self2_is_0
  4462 
  4463 * Theory IntDiv: removed many lemmas that are instances of class-based
  4464 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4465 rename old lemmas as follows:
  4466 
  4467 dvd_diff               -> nat_dvd_diff
  4468 dvd_zminus_iff         -> dvd_minus_iff
  4469 mod_add1_eq            -> mod_add_eq
  4470 mod_mult1_eq           -> mod_mult_right_eq
  4471 mod_mult1_eq'          -> mod_mult_left_eq
  4472 mod_mult_distrib_mod   -> mod_mult_eq
  4473 nat_mod_add_left_eq    -> mod_add_left_eq
  4474 nat_mod_add_right_eq   -> mod_add_right_eq
  4475 nat_mod_div_trivial    -> mod_div_trivial
  4476 nat_mod_mod_trivial    -> mod_mod_trivial
  4477 zdiv_zadd_self1        -> div_add_self1
  4478 zdiv_zadd_self2        -> div_add_self2
  4479 zdiv_zmult_self1       -> div_mult_self2_is_id
  4480 zdiv_zmult_self2       -> div_mult_self1_is_id
  4481 zdvd_triv_left         -> dvd_triv_left
  4482 zdvd_triv_right        -> dvd_triv_right
  4483 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4484 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4485 zmod_zadd_left_eq      -> mod_add_left_eq
  4486 zmod_zadd_right_eq     -> mod_add_right_eq
  4487 zmod_zadd_self1        -> mod_add_self1
  4488 zmod_zadd_self2        -> mod_add_self2
  4489 zmod_zadd1_eq          -> mod_add_eq
  4490 zmod_zdiff1_eq         -> mod_diff_eq
  4491 zmod_zdvd_zmod         -> mod_mod_cancel
  4492 zmod_zmod_cancel       -> mod_mod_cancel
  4493 zmod_zmult_self1       -> mod_mult_self2_is_0
  4494 zmod_zmult_self2       -> mod_mult_self1_is_0
  4495 zmod_1                 -> mod_by_1
  4496 zdiv_1                 -> div_by_1
  4497 zdvd_abs1              -> abs_dvd_iff
  4498 zdvd_abs2              -> dvd_abs_iff
  4499 zdvd_refl              -> dvd_refl
  4500 zdvd_trans             -> dvd_trans
  4501 zdvd_zadd              -> dvd_add
  4502 zdvd_zdiff             -> dvd_diff
  4503 zdvd_zminus_iff        -> dvd_minus_iff
  4504 zdvd_zminus2_iff       -> minus_dvd_iff
  4505 zdvd_zmultD            -> dvd_mult_right
  4506 zdvd_zmultD2           -> dvd_mult_left
  4507 zdvd_zmult_mono        -> mult_dvd_mono
  4508 zdvd_0_right           -> dvd_0_right
  4509 zdvd_0_left            -> dvd_0_left_iff
  4510 zdvd_1_left            -> one_dvd
  4511 zminus_dvd_iff         -> minus_dvd_iff
  4512 
  4513 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4514 
  4515 * The real numbers offer decimal input syntax: 12.34 is translated
  4516 into 1234/10^2. This translation is not reversed upon output.
  4517 
  4518 * Theory Library/Polynomial defines an abstract type 'a poly of
  4519 univariate polynomials with coefficients of type 'a.  In addition to
  4520 the standard ring operations, it also supports div and mod.  Code
  4521 generation is also supported, using list-style constructors.
  4522 
  4523 * Theory Library/Inner_Product defines a class of real_inner for real
  4524 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4525 => real.  Class real_inner is a subclass of real_normed_vector from
  4526 theory RealVector.
  4527 
  4528 * Theory Library/Product_Vector provides instances for the product
  4529 type 'a * 'b of several classes from RealVector and Inner_Product.
  4530 Definitions of addition, subtraction, scalar multiplication, norms,
  4531 and inner products are included.
  4532 
  4533 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4534 addition to the field operations, numerals and case syntax are also
  4535 supported.
  4536 
  4537 * Theory Library/Diagonalize provides constructive version of Cantor's
  4538 first diagonalization argument.
  4539 
  4540 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4541 zlcm (for int); carried together from various gcd/lcm developements in
  4542 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4543 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4544 may recover tupled syntax as follows:
  4545 
  4546     hide (open) const gcd
  4547     abbreviation gcd where
  4548       "gcd == (%(a, b). GCD.gcd a b)"
  4549     notation (output)
  4550       GCD.gcd ("gcd '(_, _')")
  4551 
  4552 The same works for lcm, zgcd, zlcm.
  4553 
  4554 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4555 instantiations for algebraic structures.  Removed some duplicate
  4556 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4557 
  4558 * ML antiquotation @{code} takes a constant as argument and generates
  4559 corresponding code in background and inserts name of the corresponding
  4560 resulting ML value/function/datatype constructor binding in place.
  4561 All occurrences of @{code} with a single ML block are generated
  4562 simultaneously.  Provides a generic and safe interface for
  4563 instrumentalizing code generation.  See
  4564 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4565 In future you ought to refrain from ad-hoc compiling generated SML
  4566 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4567 cannot refer to constants for which user-defined serializations are
  4568 set.  Refer to the corresponding ML counterpart directly in that
  4569 cases.
  4570 
  4571 * Command 'rep_datatype': instead of theorem names the command now
  4572 takes a list of terms denoting the constructors of the type to be
  4573 represented as datatype.  The characteristic theorems have to be
  4574 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4575 have disappeared in favour of existing ones:
  4576 
  4577     unit_induct                 ~> unit.induct
  4578     prod_induct                 ~> prod.induct
  4579     sum_induct                  ~> sum.induct
  4580     Suc_Suc_eq                  ~> nat.inject
  4581     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4582 
  4583 
  4584 *** HOL-Algebra ***
  4585 
  4586 * New locales for orders and lattices where the equivalence relation
  4587 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4588 locales use a record structure with field eq for the equivalence.
  4589 
  4590 * New theory of factorial domains.
  4591 
  4592 * Units_l_inv and Units_r_inv are now simp rules by default.
  4593 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4594 and/or r_inv will now also require deletion of these lemmas.
  4595 
  4596 * Renamed the following theorems, INCOMPATIBILITY:
  4597 
  4598 UpperD ~> Upper_memD
  4599 LowerD ~> Lower_memD
  4600 least_carrier ~> least_closed
  4601 greatest_carrier ~> greatest_closed
  4602 greatest_Lower_above ~> greatest_Lower_below
  4603 one_zero ~> carrier_one_zero
  4604 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4605 
  4606 
  4607 *** HOL-Nominal ***
  4608 
  4609 * Nominal datatypes can now contain type-variables.
  4610 
  4611 * Commands 'nominal_inductive' and 'equivariance' work with local
  4612 theory targets.
  4613 
  4614 * Nominal primrec can now works with local theory targets and its
  4615 specification syntax now conforms to the general format as seen in
  4616 'inductive' etc.
  4617 
  4618 * Method "perm_simp" honours the standard simplifier attributes
  4619 (no_asm), (no_asm_use) etc.
  4620 
  4621 * The new predicate #* is defined like freshness, except that on the
  4622 left hand side can be a set or list of atoms.
  4623 
  4624 * Experimental command 'nominal_inductive2' derives strong induction
  4625 principles for inductive definitions.  In contrast to
  4626 'nominal_inductive', which can only deal with a fixed number of
  4627 binders, it can deal with arbitrary expressions standing for sets of
  4628 atoms to be avoided.  The only inductive definition we have at the
  4629 moment that needs this generalisation is the typing rule for Lets in
  4630 the algorithm W:
  4631 
  4632  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4633  -----------------------------------------------------------------
  4634          Gamma |- Let x be t1 in t2 : T2
  4635 
  4636 In this rule one wants to avoid all the binders that are introduced by
  4637 "close Gamma T1".  We are looking for other examples where this
  4638 feature might be useful.  Please let us know.
  4639 
  4640 
  4641 *** HOLCF ***
  4642 
  4643 * Reimplemented the simplification procedure for proving continuity
  4644 subgoals.  The new simproc is extensible; users can declare additional
  4645 continuity introduction rules with the attribute [cont2cont].
  4646 
  4647 * The continuity simproc now uses a different introduction rule for
  4648 solving continuity subgoals on terms with lambda abstractions.  In
  4649 some rare cases the new simproc may fail to solve subgoals that the
  4650 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4651 Potential INCOMPATIBILITY.
  4652 
  4653 * Command 'fixrec': specification syntax now conforms to the general
  4654 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4655 examples.  INCOMPATIBILITY.
  4656 
  4657 
  4658 *** ZF ***
  4659 
  4660 * Proof of Zorn's Lemma for partial orders.
  4661 
  4662 
  4663 *** ML ***
  4664 
  4665 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4666 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4667 depends on multithreading, so timouts will not work before Poly/ML
  4668 5.2.1!
  4669 
  4670 * High-level support for concurrent ML programming, see
  4671 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4672 particularly convenient to organize independent functional
  4673 computations.  The concept of "synchronized variables" provides a
  4674 higher-order interface for components with shared state, avoiding the
  4675 delicate details of mutexes and condition variables.  (Requires
  4676 Poly/ML 5.2.1 or later.)
  4677 
  4678 * ML bindings produced via Isar commands are stored within the Isar
  4679 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4680 become thread-safe and work with undo as expected (concerning
  4681 top-level bindings, not side-effects on global references).
  4682 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4683 compiler at runtime; really global bindings need to be given outside a
  4684 theory.  (Requires Poly/ML 5.2 or later.)
  4685 
  4686 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4687 context.  Top-level ML bindings are stored within the proof context in
  4688 a purely sequential fashion, disregarding the nested proof structure.
  4689 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4690 proof.  (Requires Poly/ML 5.2 or later.)
  4691 
  4692 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4693 and Method.setup, as well as Isar commands 'attribute_setup' and
  4694 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4695 existing code accordingly, or use plain 'setup' together with old
  4696 Method.add_method.
  4697 
  4698 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4699 to 'a -> thm, while results are always tagged with an authentic oracle
  4700 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4701 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4702 accordingly.  Note that extra performance may be gained by producing
  4703 the cterm carefully, avoiding slow Thm.cterm_of.
  4704 
  4705 * Simplified interface for defining document antiquotations via
  4706 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4707 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4708 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4709 examples.
  4710 
  4711 * More systematic treatment of long names, abstract name bindings, and
  4712 name space operations.  Basic operations on qualified names have been
  4713 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4714 Long_Name.append.  Old type bstring has been mostly replaced by
  4715 abstract type binding (see structure Binding), which supports precise
  4716 qualification by packages and local theory targets, as well as proper
  4717 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4718 bstring values into Binding.name, or better pass through abstract
  4719 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4720 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4721 
  4722 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4723 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4724 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4725 
  4726 * Disposed old type and term read functions (Sign.read_def_typ,
  4727 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4728 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4729 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4730 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4731 resort for legacy applications.
  4732 
  4733 * Disposed old declarations, tactics, tactic combinators that refer to
  4734 the simpset or claset of an implicit theory (such as Addsimps,
  4735 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4736 embedded ML text, or local_simpset_of with a proper context passed as
  4737 explicit runtime argument.
  4738 
  4739 * Rules and tactics that read instantiations (read_instantiate,
  4740 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4741 context, which is required for parsing and type-checking.  Moreover,
  4742 the variables are specified as plain indexnames, not string encodings
  4743 thereof.  INCOMPATIBILITY.
  4744 
  4745 * Generic Toplevel.add_hook interface allows to analyze the result of
  4746 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4747 for theorem dependency output of transactions resulting in a new
  4748 theory state.
  4749 
  4750 * ML antiquotations: block-structured compilation context indicated by
  4751 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4752 
  4753   @{binding name}                         - basic name binding
  4754   @{let ?pat = term}                      - term abbreviation (HO matching)
  4755   @{note name = fact}                     - fact abbreviation
  4756   @{thm fact}                             - singleton fact (with attributes)
  4757   @{thms fact}                            - general fact (with attributes)
  4758   @{lemma prop by method}                 - singleton goal
  4759   @{lemma prop by meth1 meth2}            - singleton goal
  4760   @{lemma prop1 ... propN by method}      - general goal
  4761   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4762   @{lemma (open) ...}                     - open derivation
  4763 
  4764 
  4765 *** System ***
  4766 
  4767 * The Isabelle "emacs" tool provides a specific interface to invoke
  4768 Proof General / Emacs, with more explicit failure if that is not
  4769 installed (the old isabelle-interface script silently falls back on
  4770 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4771 installation location of the Proof General distribution.
  4772 
  4773 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4774 the Isabelle process into a JVM/Scala application.  See
  4775 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4776 process wrapper has been discontinued.)
  4777 
  4778 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4779 
  4780 * Various status messages (with exact source position information) are
  4781 emitted, if proper markup print mode is enabled.  This allows
  4782 user-interface components to provide detailed feedback on internal
  4783 prover operations.
  4784 
  4785 
  4786 
  4787 New in Isabelle2008 (June 2008)
  4788 -------------------------------
  4789 
  4790 *** General ***
  4791 
  4792 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4793 and updated, with formally checked references as hyperlinks.
  4794 
  4795 * Theory loader: use_thy (and similar operations) no longer set the
  4796 implicit ML context, which was occasionally hard to predict and in
  4797 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4798 provides a proper context already.
  4799 
  4800 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4801 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4802 'use' within a theory file will do the job.
  4803 
  4804 * Name space merge now observes canonical order, i.e. the second space
  4805 is inserted into the first one, while existing entries in the first
  4806 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4807 swap theory imports.
  4808 
  4809 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4810 INCOMPATIBILITY in identifier syntax etc.
  4811 
  4812 * Outer syntax: string tokens no longer admit escaped white space,
  4813 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4814 white space without escapes.
  4815 
  4816 * Outer syntax: string tokens may contain arbitrary character codes
  4817 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4818 "foo_bar".
  4819 
  4820 
  4821 *** Pure ***
  4822 
  4823 * Context-dependent token translations.  Default setup reverts locally
  4824 fixed variables, and adds hilite markup for undeclared frees.
  4825 
  4826 * Unused theorems can be found using the new command 'unused_thms'.
  4827 There are three ways of invoking it:
  4828 
  4829 (1) unused_thms
  4830      Only finds unused theorems in the current theory.
  4831 
  4832 (2) unused_thms thy_1 ... thy_n -
  4833      Finds unused theorems in the current theory and all of its ancestors,
  4834      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4835 
  4836 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4837      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4838      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4839      their ancestors.
  4840 
  4841 In order to increase the readability of the list produced by
  4842 unused_thms, theorems that have been created by a particular instance
  4843 of a theory command such as 'inductive' or 'function' are considered
  4844 to belong to the same "group", meaning that if at least one theorem in
  4845 this group is used, the other theorems in the same group are no longer
  4846 reported as unused.  Moreover, if all theorems in the group are
  4847 unused, only one theorem in the group is displayed.
  4848 
  4849 Note that proof objects have to be switched on in order for
  4850 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4851 usually the case when using Proof General with the default settings).
  4852 
  4853 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4854 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4855 duplicate fact bindings, or even accidental fact duplications.  Note
  4856 that tools may maintain dynamically scoped facts systematically, using
  4857 PureThy.add_thms_dynamic.
  4858 
  4859 * Command 'hide' now allows to hide from "fact" name space as well.
  4860 
  4861 * Eliminated destructive theorem database, simpset, claset, and
  4862 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4863 update of theories within ML code.
  4864 
  4865 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4866 INCOMPATIBILITY, object-logics depending on former Pure require
  4867 additional setup PureThy.old_appl_syntax_setup; object-logics
  4868 depending on former CPure need to refer to Pure.
  4869 
  4870 * Commands 'use' and 'ML' are now purely functional, operating on
  4871 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4872 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4873 INCOMPATIBILITY.
  4874 
  4875 * Command 'setup': discontinued implicit version with ML reference.
  4876 
  4877 * Instantiation target allows for simultaneous specification of class
  4878 instance operations together with an instantiation proof.
  4879 Type-checking phase allows to refer to class operations uniformly.
  4880 See src/HOL/Complex/Complex.thy for an Isar example and
  4881 src/HOL/Library/Eval.thy for an ML example.
  4882 
  4883 * Indexing of literal facts: be more serious about including only
  4884 facts from the visible specification/proof context, but not the
  4885 background context (locale etc.).  Affects `prop` notation and method
  4886 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4887 situations.
  4888 
  4889 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4890 "(open)" option, which used to expose internal bound variables to the
  4891 proof text.
  4892 
  4893 * Isar statements: removed obsolete case "rule_context".
  4894 INCOMPATIBILITY, better use explicit fixes/assumes.
  4895 
  4896 * Locale proofs: default proof step now includes 'unfold_locales';
  4897 hence 'proof' without argument may be used to unfold locale
  4898 predicates.
  4899 
  4900 
  4901 *** Document preparation ***
  4902 
  4903 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4904 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4905 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4906 spot); renamed link color from "darkblue" to "linkcolor" (default
  4907 value unchanged, can be redefined via \definecolor); no longer sets
  4908 "a4paper" option (unnecessary or even intrusive).
  4909 
  4910 * Antiquotation @{lemma A method} proves proposition A by the given
  4911 method (either a method name or a method name plus (optional) method
  4912 arguments in parentheses) and prints A just like @{prop A}.
  4913 
  4914 
  4915 *** HOL ***
  4916 
  4917 * New primrec package.  Specification syntax conforms in style to
  4918 definition/function/....  No separate induction rule is provided.  The
  4919 "primrec" command distinguishes old-style and new-style specifications
  4920 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4921 When adjusting theories, beware: constants stemming from new-style
  4922 primrec specifications have authentic syntax.
  4923 
  4924 * Metis prover is now an order of magnitude faster, and also works
  4925 with multithreading.
  4926 
  4927 * Metis: the maximum number of clauses that can be produced from a
  4928 theorem is now given by the attribute max_clauses.  Theorems that
  4929 exceed this number are ignored, with a warning printed.
  4930 
  4931 * Sledgehammer no longer produces structured proofs by default. To
  4932 enable, declare [[sledgehammer_full = true]].  Attributes
  4933 reconstruction_modulus, reconstruction_sorts renamed
  4934 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4935 
  4936 * Method "induct_scheme" derives user-specified induction rules
  4937 from well-founded induction and completeness of patterns. This factors
  4938 out some operations that are done internally by the function package
  4939 and makes them available separately.  See
  4940 src/HOL/ex/Induction_Scheme.thy for examples.
  4941 
  4942 * More flexible generation of measure functions for termination
  4943 proofs: Measure functions can be declared by proving a rule of the
  4944 form "is_measure f" and giving it the [measure_function] attribute.
  4945 The "is_measure" predicate is logically meaningless (always true), and
  4946 just guides the heuristic.  To find suitable measure functions, the
  4947 termination prover sets up the goal "is_measure ?f" of the appropriate
  4948 type and generates all solutions by prolog-style backwards proof using
  4949 the declared rules.
  4950 
  4951 This setup also deals with rules like 
  4952 
  4953   "is_measure f ==> is_measure (list_size f)"
  4954 
  4955 which accommodates nested datatypes that recurse through lists.
  4956 Similar rules are predeclared for products and option types.
  4957 
  4958 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4959 
  4960   INCOMPATIBILITIES:
  4961 
  4962   - Definitions of overloaded constants on sets have to be replaced by
  4963     definitions on => and bool.
  4964 
  4965   - Some definitions of overloaded operators on sets can now be proved
  4966     using the definitions of the operators on => and bool.  Therefore,
  4967     the following theorems have been renamed:
  4968 
  4969       subset_def   -> subset_eq
  4970       psubset_def  -> psubset_eq
  4971       set_diff_def -> set_diff_eq
  4972       Compl_def    -> Compl_eq
  4973       Sup_set_def  -> Sup_set_eq
  4974       Inf_set_def  -> Inf_set_eq
  4975       sup_set_def  -> sup_set_eq
  4976       inf_set_def  -> inf_set_eq
  4977 
  4978   - Due to the incompleteness of the HO unification algorithm, some
  4979     rules such as subst may require manual instantiation, if some of
  4980     the unknowns in the rule is a set.
  4981 
  4982   - Higher order unification and forward proofs:
  4983     The proof pattern
  4984 
  4985       have "P (S::'a set)" <...>
  4986       then have "EX S. P S" ..
  4987 
  4988     no longer works (due to the incompleteness of the HO unification
  4989     algorithm) and must be replaced by the pattern
  4990 
  4991       have "EX S. P S"
  4992       proof
  4993         show "P S" <...>
  4994       qed
  4995 
  4996   - Calculational reasoning with subst (or similar rules):
  4997     The proof pattern
  4998 
  4999       have "P (S::'a set)" <...>
  5000       also have "S = T" <...>
  5001       finally have "P T" .
  5002 
  5003     no longer works (for similar reasons as the previous example) and
  5004     must be replaced by something like
  5005 
  5006       have "P (S::'a set)" <...>
  5007       moreover have "S = T" <...>
  5008       ultimately have "P T" by simp
  5009 
  5010   - Tactics or packages written in ML code:
  5011     Code performing pattern matching on types via
  5012 
  5013       Type ("set", [T]) => ...
  5014 
  5015     must be rewritten. Moreover, functions like strip_type or
  5016     binder_types no longer return the right value when applied to a
  5017     type of the form
  5018 
  5019       T1 => ... => Tn => U => bool
  5020 
  5021     rather than
  5022 
  5023       T1 => ... => Tn => U set
  5024 
  5025 * Merged theories Wellfounded_Recursion, Accessible_Part and
  5026 Wellfounded_Relations to theory Wellfounded.
  5027 
  5028 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  5029 
  5030 * Class finite no longer treats UNIV as class parameter.  Use class
  5031 enum from theory Library/Enum instead to achieve a similar effect.
  5032 INCOMPATIBILITY.
  5033 
  5034 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  5035 and "Cons".  INCOMPATIBILITY.
  5036 
  5037 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  5038 Potential INCOMPATIBILITY.
  5039 
  5040 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  5041 injective_fst_snd removed, use split_eta and prod_eqI instead.
  5042 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  5043 
  5044 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  5045 the same name in theory Orderings:
  5046 
  5047   less_trans
  5048   less_linear
  5049   le_imp_less_or_eq
  5050   le_less_trans
  5051   less_le_trans
  5052   less_not_sym
  5053   less_asym
  5054 
  5055 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  5056 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  5057 and different variable names.
  5058 
  5059 * Library/Option_ord.thy: Canonical order on option type.
  5060 
  5061 * Library/RBT.thy: Red-black trees, an efficient implementation of
  5062 finite maps.
  5063 
  5064 * Library/Countable.thy: Type class for countable types.
  5065 
  5066 * Theory Int: The representation of numerals has changed.  The infix
  5067 operator BIT and the bit datatype with constructors B0 and B1 have
  5068 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  5069 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  5070 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  5071 accordingly.
  5072 
  5073 * Theory Nat: definition of <= and < on natural numbers no longer
  5074 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  5075 le_def and less_def have disappeared.  Consider lemmas not_less
  5076 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  5077 
  5078 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  5079 (whose purpose mainly is for various fold_set functionals) have been
  5080 abandoned in favor of the existing algebraic classes
  5081 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  5082 lower_semilattice (resp. upper_semilattice) and linorder.
  5083 INCOMPATIBILITY.
  5084 
  5085 * Theory Transitive_Closure: induct and cases rules now declare proper
  5086 case_names ("base" and "step").  INCOMPATIBILITY.
  5087 
  5088 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  5089 lattices.  The form set-specific version is available as
  5090 Inductive.lfp_ordinal_induct_set.
  5091 
  5092 * Renamed theorems "power.simps" to "power_int.simps".
  5093 INCOMPATIBILITY.
  5094 
  5095 * Class semiring_div provides basic abstract properties of semirings
  5096 with division and modulo operations.  Subsumes former class dvd_mod.
  5097 
  5098 * Merged theories IntDef, Numeral and IntArith into unified theory
  5099 Int.  INCOMPATIBILITY.
  5100 
  5101 * Theory Library/Code_Index: type "index" now represents natural
  5102 numbers rather than integers.  INCOMPATIBILITY.
  5103 
  5104 * New class "uminus" with operation "uminus" (split of from class
  5105 "minus" which now only has operation "minus", binary).
  5106 INCOMPATIBILITY.
  5107 
  5108 * Constants "card", "internal_split", "option_map" now with authentic
  5109 syntax.  INCOMPATIBILITY.
  5110 
  5111 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  5112 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  5113 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  5114 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  5115 Sup_set_def, le_def, less_def, option_map_def now with object
  5116 equality.  INCOMPATIBILITY.
  5117 
  5118 * Records. Removed K_record, and replaced it by pure lambda term
  5119 %x. c. The simplifier setup is now more robust against eta expansion.
  5120 INCOMPATIBILITY: in cases explicitly referring to K_record.
  5121 
  5122 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  5123 
  5124 * Library/ListVector: new theory of arithmetic vector operations.
  5125 
  5126 * Library/Order_Relation: new theory of various orderings as sets of
  5127 pairs.  Defines preorders, partial orders, linear orders and
  5128 well-orders on sets and on types.
  5129 
  5130 
  5131 *** ZF ***
  5132 
  5133 * Renamed some theories to allow to loading both ZF and HOL in the
  5134 same session:
  5135 
  5136   Datatype  -> Datatype_ZF
  5137   Inductive -> Inductive_ZF
  5138   Int       -> Int_ZF
  5139   IntDiv    -> IntDiv_ZF
  5140   Nat       -> Nat_ZF
  5141   List      -> List_ZF
  5142   Main      -> Main_ZF
  5143 
  5144 INCOMPATIBILITY: ZF theories that import individual theories below
  5145 Main might need to be adapted.  Regular theory Main is still
  5146 available, as trivial extension of Main_ZF.
  5147 
  5148 
  5149 *** ML ***
  5150 
  5151 * ML within Isar: antiquotation @{const name} or @{const
  5152 name(typargs)} produces statically-checked Const term.
  5153 
  5154 * Functor NamedThmsFun: data is available to the user as dynamic fact
  5155 (of the same name).  Removed obsolete print command.
  5156 
  5157 * Removed obsolete "use_legacy_bindings" function.
  5158 
  5159 * The ``print mode'' is now a thread-local value derived from a global
  5160 template (the former print_mode reference), thus access becomes
  5161 non-critical.  The global print_mode reference is for session
  5162 management only; user-code should use print_mode_value,
  5163 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  5164 
  5165 * Functions system/system_out provide a robust way to invoke external
  5166 shell commands, with propagation of interrupts (requires Poly/ML
  5167 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  5168 
  5169 
  5170 *** System ***
  5171 
  5172 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  5173 in accordance with Proof General 3.7, which prefers GNU emacs.
  5174 
  5175 * isatool tty runs Isabelle process with plain tty interaction;
  5176 optional line editor may be specified via ISABELLE_LINE_EDITOR
  5177 setting, the default settings attempt to locate "ledit" and "rlwrap".
  5178 
  5179 * isatool browser now works with Cygwin as well, using general
  5180 "javapath" function defined in Isabelle process environment.
  5181 
  5182 * YXML notation provides a simple and efficient alternative to
  5183 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  5184 isatool yxml as described in the Isabelle system manual.
  5185 
  5186 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  5187 provides general wrapper for managing an Isabelle process in a robust
  5188 fashion, with ``cooked'' output from stdin/stderr.
  5189 
  5190 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  5191 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  5192 
  5193 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  5194 way of changing the user's settings is via
  5195 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  5196 script.
  5197 
  5198 * Multithreading.max_threads := 0 refers to the number of actual CPU
  5199 cores of the underlying machine, which is a good starting point for
  5200 optimal performance tuning.  The corresponding usedir option -M allows
  5201 "max" as an alias for "0".  WARNING: does not work on certain versions
  5202 of Mac OS (with Poly/ML 5.1).
  5203 
  5204 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  5205 adverse effect of Isabelle flooding interactive front-ends (notably
  5206 ProofGeneral / XEmacs).
  5207 
  5208 
  5209 
  5210 New in Isabelle2007 (November 2007)
  5211 -----------------------------------
  5212 
  5213 *** General ***
  5214 
  5215 * More uniform information about legacy features, notably a
  5216 warning/error of "Legacy feature: ...", depending on the state of the
  5217 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  5218 legacy features will disappear eventually.
  5219 
  5220 * Theory syntax: the header format ``theory A = B + C:'' has been
  5221 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  5222 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  5223 
  5224 * Theory syntax: the old non-Isar theory file format has been
  5225 discontinued altogether.  Note that ML proof scripts may still be used
  5226 with Isar theories; migration is usually quite simple with the ML
  5227 function use_legacy_bindings.  INCOMPATIBILITY.
  5228 
  5229 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  5230 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  5231 quotes.
  5232 
  5233 * Theory loader: be more serious about observing the static theory
  5234 header specifications (including optional directories), but not the
  5235 accidental file locations of previously successful loads.  The strict
  5236 update policy of former update_thy is now already performed by
  5237 use_thy, so the former has been removed; use_thys updates several
  5238 theories simultaneously, just as 'imports' within a theory header
  5239 specification, but without merging the results.  Potential
  5240 INCOMPATIBILITY: may need to refine theory headers and commands
  5241 ROOT.ML which depend on load order.
  5242 
  5243 * Theory loader: optional support for content-based file
  5244 identification, instead of the traditional scheme of full physical
  5245 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  5246 (cf. the system manual).  The new scheme allows to work with
  5247 non-finished theories in persistent session images, such that source
  5248 files may be moved later on without requiring reloads.
  5249 
  5250 * Theory loader: old-style ML proof scripts being *attached* to a thy
  5251 file (with the same base name as the theory) are considered a legacy
  5252 feature, which will disappear eventually. Even now, the theory loader
  5253 no longer maintains dependencies on such files.
  5254 
  5255 * Syntax: the scope for resolving ambiguities via type-inference is
  5256 now limited to individual terms, instead of whole simultaneous
  5257 specifications as before. This greatly reduces the complexity of the
  5258 syntax module and improves flexibility by separating parsing and
  5259 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  5260 'fixes' etc.) are required in rare situations.
  5261 
  5262 * Syntax: constants introduced by new-style packages ('definition',
  5263 'abbreviation' etc.) are passed through the syntax module in
  5264 ``authentic mode''. This means that associated mixfix annotations
  5265 really stick to such constants, independently of potential name space
  5266 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  5267 trees are represented slightly differently, may need to adapt syntax
  5268 translations accordingly. Use CONST marker in 'translations' and
  5269 @{const_syntax} antiquotation in 'parse_translation' etc.
  5270 
  5271 * Legacy goal package: reduced interface to the bare minimum required
  5272 to keep existing proof scripts running.  Most other user-level
  5273 functions are now part of the OldGoals structure, which is *not* open
  5274 by default (consider isatool expandshort before open OldGoals).
  5275 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  5276 these tend to cause confusion about the actual goal (!) context being
  5277 used here, which is not necessarily the same as the_context().
  5278 
  5279 * Command 'find_theorems': supports "*" wild-card in "name:"
  5280 criterion; "with_dups" option.  Certain ProofGeneral versions might
  5281 support a specific search form (see ProofGeneral/CHANGES).
  5282 
  5283 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  5284 by default, which means that "prems" (and also "fixed variables") are
  5285 suppressed from proof state output.  Note that the ProofGeneral
  5286 settings mechanism allows to change and save options persistently, but
  5287 older versions of Isabelle will fail to start up if a negative prems
  5288 limit is imposed.
  5289 
  5290 * Local theory targets may be specified by non-nested blocks of
  5291 ``context/locale/class ... begin'' followed by ``end''.  The body may
  5292 contain definitions, theorems etc., including any derived mechanism
  5293 that has been implemented on top of these primitives.  This concept
  5294 generalizes the existing ``theorem (in ...)'' towards more versatility
  5295 and scalability.
  5296 
  5297 * Proof General interface: proper undo of final 'end' command;
  5298 discontinued Isabelle/classic mode (ML proof scripts).
  5299 
  5300 
  5301 *** Document preparation ***
  5302 
  5303 * Added antiquotation @{theory name} which prints the given name,
  5304 after checking that it refers to a valid ancestor theory in the
  5305 current context.
  5306 
  5307 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5308 check the given source text as ML type/structure, printing verbatim.
  5309 
  5310 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5311 "c args == rhs" given in the current context.  (Any number of
  5312 arguments may be given on the LHS.)
  5313 
  5314 
  5315 *** Pure ***
  5316 
  5317 * The 'class' package offers a combination of axclass and locale to
  5318 achieve Haskell-like type classes in Isabelle.  Definitions and
  5319 theorems within a class context produce both relative results (with
  5320 implicit parameters according to the locale context), and polymorphic
  5321 constants with qualified polymorphism (according to the class
  5322 context).  Within the body context of a 'class' target, a separate
  5323 syntax layer ("user space type system") takes care of converting
  5324 between global polymorphic consts and internal locale representation.
  5325 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5326 "isatool doc classes" provides a tutorial.
  5327 
  5328 * Generic code generator framework allows to generate executable
  5329 code for ML and Haskell (including Isabelle classes).  A short usage
  5330 sketch:
  5331 
  5332     internal compilation:
  5333         export_code <list of constants (term syntax)> in SML
  5334     writing SML code to a file:
  5335         export_code <list of constants (term syntax)> in SML <filename>
  5336     writing OCaml code to a file:
  5337         export_code <list of constants (term syntax)> in OCaml <filename>
  5338     writing Haskell code to a bunch of files:
  5339         export_code <list of constants (term syntax)> in Haskell <filename>
  5340 
  5341     evaluating closed propositions to True/False using code generation:
  5342         method ``eval''
  5343 
  5344 Reasonable default setup of framework in HOL.
  5345 
  5346 Theorem attributs for selecting and transforming function equations theorems:
  5347 
  5348     [code fun]:        select a theorem as function equation for a specific constant
  5349     [code fun del]:    deselect a theorem as function equation for a specific constant
  5350     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5351     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5352 
  5353 User-defined serializations (target in {SML, OCaml, Haskell}):
  5354 
  5355     code_const <and-list of constants (term syntax)>
  5356       {(target) <and-list of const target syntax>}+
  5357 
  5358     code_type <and-list of type constructors>
  5359       {(target) <and-list of type target syntax>}+
  5360 
  5361     code_instance <and-list of instances>
  5362       {(target)}+
  5363         where instance ::= <type constructor> :: <class>
  5364 
  5365     code_class <and_list of classes>
  5366       {(target) <and-list of class target syntax>}+
  5367         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5368 
  5369 code_instance and code_class only are effective to target Haskell.
  5370 
  5371 For example usage see src/HOL/ex/Codegenerator.thy and
  5372 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5373 generation from Isabelle/HOL theories is available via "isatool doc
  5374 codegen".
  5375 
  5376 * Code generator: consts in 'consts_code' Isar commands are now
  5377 referred to by usual term syntax (including optional type
  5378 annotations).
  5379 
  5380 * Command 'no_translations' removes translation rules from theory
  5381 syntax.
  5382 
  5383 * Overloaded definitions are now actually checked for acyclic
  5384 dependencies.  The overloading scheme is slightly more general than
  5385 that of Haskell98, although Isabelle does not demand an exact
  5386 correspondence to type class and instance declarations.
  5387 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5388 exotic versions of overloading -- at the discretion of the user!
  5389 
  5390 Polymorphic constants are represented via type arguments, i.e. the
  5391 instantiation that matches an instance against the most general
  5392 declaration given in the signature.  For example, with the declaration
  5393 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5394 as c(nat).  Overloading is essentially simultaneous structural
  5395 recursion over such type arguments.  Incomplete specification patterns
  5396 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5397 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5398 Command 'print_theory' outputs the normalized system of recursive
  5399 equations, see section "definitions".
  5400 
  5401 * Configuration options are maintained within the theory or proof
  5402 context (with name and type bool/int/string), providing a very simple
  5403 interface to a poor-man's version of general context data.  Tools may
  5404 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5405 these values using Config.get etc.  Users may change options via an
  5406 associated attribute of the same name.  This form of context
  5407 declaration works particularly well with commands 'declare' or
  5408 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5409 very easy to avoid global references, which would not observe Isar
  5410 toplevel undo/redo and fail to work with multithreading.
  5411 
  5412 Various global ML references of Pure and HOL have been turned into
  5413 configuration options:
  5414 
  5415   Unify.search_bound		unify_search_bound
  5416   Unify.trace_bound		unify_trace_bound
  5417   Unify.trace_simp		unify_trace_simp
  5418   Unify.trace_types		unify_trace_types
  5419   Simplifier.simp_depth_limit	simp_depth_limit
  5420   Blast.depth_limit		blast_depth_limit
  5421   DatatypeProp.dtK		datatype_distinctness_limit
  5422   fast_arith_neq_limit  	fast_arith_neq_limit
  5423   fast_arith_split_limit	fast_arith_split_limit
  5424 
  5425 * Named collections of theorems may be easily installed as context
  5426 data using the functor NamedThmsFun (see also
  5427 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5428 attributes; there is also a toplevel print command.  This facility is
  5429 just a common case of general context data, which is the preferred way
  5430 for anything more complex than just a list of facts in canonical
  5431 order.
  5432 
  5433 * Isar: command 'declaration' augments a local theory by generic
  5434 declaration functions written in ML.  This enables arbitrary content
  5435 being added to the context, depending on a morphism that tells the
  5436 difference of the original declaration context wrt. the application
  5437 context encountered later on.
  5438 
  5439 * Isar: proper interfaces for simplification procedures.  Command
  5440 'simproc_setup' declares named simprocs (with match patterns, and body
  5441 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5442 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5443 
  5444 * Isar: an extra pair of brackets around attribute declarations
  5445 abbreviates a theorem reference involving an internal dummy fact,
  5446 which will be ignored later --- only the effect of the attribute on
  5447 the background context will persist.  This form of in-place
  5448 declarations is particularly useful with commands like 'declare' and
  5449 'using', for example ``have A using [[simproc a]] by simp''.
  5450 
  5451 * Isar: method "assumption" (and implicit closing of subproofs) now
  5452 takes simple non-atomic goal assumptions into account: after applying
  5453 an assumption as a rule the resulting subgoals are solved by atomic
  5454 assumption steps.  This is particularly useful to finish 'obtain'
  5455 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5456 without referring to the original premise "!!x. P x ==> thesis" in the
  5457 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5458 more permissive.
  5459 
  5460 * Isar: implicit use of prems from the Isar proof context is
  5461 considered a legacy feature.  Common applications like ``have A .''
  5462 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5463 referencing facts explicitly here improves readability and
  5464 maintainability of proof texts.
  5465 
  5466 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5467 the obtained context from the course of reasoning!  For example:
  5468 
  5469   assume "EX x y. A x & B y"   -- "any previous fact"
  5470   then guess x and y by clarify
  5471 
  5472 This technique is potentially adventurous, depending on the facts and
  5473 proof tools being involved here.
  5474 
  5475 * Isar: known facts from the proof context may be specified as literal
  5476 propositions, using ASCII back-quote syntax.  This works wherever
  5477 named facts used to be allowed so far, in proof commands, proof
  5478 methods, attributes etc.  Literal facts are retrieved from the context
  5479 according to unification of type and term parameters.  For example,
  5480 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5481 theorems in the current context, then these are valid literal facts:
  5482 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5483 
  5484 There is also a proof method "fact" which does the same composition
  5485 for explicit goal states, e.g. the following proof texts coincide with
  5486 certain special cases of literal facts:
  5487 
  5488   have "A" by fact                 ==  note `A`
  5489   have "A ==> B" by fact           ==  note `A ==> B`
  5490   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5491   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5492 
  5493 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5494 outer syntax.  Thus symbolic identifiers may be used without
  5495 additional white space in declarations like this: ``assume *: A''.
  5496 
  5497 * Isar: 'print_facts' prints all local facts of the current context,
  5498 both named and unnamed ones.
  5499 
  5500 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5501 
  5502   def x == "t" and y == "u"
  5503 
  5504 * Isar: added command 'unfolding', which is structurally similar to
  5505 'using', but affects both the goal state and facts by unfolding given
  5506 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5507 'unfolded' attribute may be replaced by first-class proof text.
  5508 
  5509 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5510 and command 'unfolding' now all support object-level equalities
  5511 (potentially conditional).  The underlying notion of rewrite rule is
  5512 analogous to the 'rule_format' attribute, but *not* that of the
  5513 Simplifier (which is usually more generous).
  5514 
  5515 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5516 premises of a theorem by n. Useful in conjunction with drule.
  5517 
  5518 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5519 method expression within a sandbox consisting of the first N
  5520 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5521 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5522 simplifies all new goals that emerge from applying rule foo to the
  5523 originally first one.
  5524 
  5525 * Isar: schematic goals are no longer restricted to higher-order
  5526 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5527 expected.
  5528 
  5529 * Isar: the conclusion of a long theorem statement is now either
  5530 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5531 (essentially a disjunction of cases with local parameters and
  5532 assumptions).  The latter allows to express general elimination rules
  5533 adequately; in this notation common elimination rules look like this:
  5534 
  5535   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5536     assumes "EX x. P x"
  5537     obtains x where "P x"
  5538 
  5539   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5540     assumes "A & B"
  5541     obtains A and B
  5542 
  5543   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5544     assumes "A | B"
  5545     obtains
  5546       A
  5547     | B
  5548 
  5549 The subsequent classical rules even refer to the formal "thesis"
  5550 explicitly:
  5551 
  5552   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5553     obtains "~ thesis"
  5554 
  5555   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5556     obtains "thesis ==> something"
  5557 
  5558 The actual proof of an 'obtains' statement is analogous to that of the
  5559 Isar proof element 'obtain', only that there may be several cases.
  5560 Optional case names may be specified in parentheses; these will be
  5561 available both in the present proof and as annotations in the
  5562 resulting rule, for later use with the 'cases' method (cf. attribute
  5563 case_names).
  5564 
  5565 * Isar: the assumptions of a long theorem statement are available as
  5566 "assms" fact in the proof context.  This is more appropriate than the
  5567 (historical) "prems", which refers to all assumptions of the current
  5568 context, including those from the target locale, proof body etc.
  5569 
  5570 * Isar: 'print_statement' prints theorems from the current theory or
  5571 proof context in long statement form, according to the syntax of a
  5572 top-level lemma.
  5573 
  5574 * Isar: 'obtain' takes an optional case name for the local context
  5575 introduction rule (default "that").
  5576 
  5577 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5578 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5579 to occur.
  5580 
  5581 * Pure: syntax "CONST name" produces a fully internalized constant
  5582 according to the current context.  This is particularly useful for
  5583 syntax translations that should refer to internal constant
  5584 representations independently of name spaces.
  5585 
  5586 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5587 instead of "FOO ". This allows multiple binder declarations to coexist
  5588 in the same context.  INCOMPATIBILITY.
  5589 
  5590 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5591 primitive that also works in a locale context (both for constants and
  5592 fixed variables). Type declaration and internal syntactic representation
  5593 of given constants retrieved from the context. Likewise, the
  5594 'no_notation' command allows to remove given syntax annotations from the
  5595 current context.
  5596 
  5597 * Isar/locales: new derived specification elements 'axiomatization',
  5598 'definition', 'abbreviation', which support type-inference, admit
  5599 object-level specifications (equality, equivalence).  See also the
  5600 isar-ref manual.  Examples:
  5601 
  5602   axiomatization
  5603     eq  (infix "===" 50) where
  5604     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5605 
  5606   definition "f x y = x + y + 1"
  5607   definition g where "g x = f x x"
  5608 
  5609   abbreviation
  5610     neq  (infix "=!=" 50) where
  5611     "x =!= y == ~ (x === y)"
  5612 
  5613 These specifications may be also used in a locale context.  Then the
  5614 constants being introduced depend on certain fixed parameters, and the
  5615 constant name is qualified by the locale base name.  An internal
  5616 abbreviation takes care for convenient input and output, making the
  5617 parameters implicit and using the original short name.  See also
  5618 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5619 entities from a monomorphic theory.
  5620 
  5621 Presently, abbreviations are only available 'in' a target locale, but
  5622 not inherited by general import expressions.  Also note that
  5623 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5624 'translations' in common applications.  The "no_abbrevs" print mode
  5625 prevents folding of abbreviations in term output.
  5626 
  5627 Concrete syntax is attached to specified constants in internal form,
  5628 independently of name spaces.  The parse tree representation is
  5629 slightly different -- use 'notation' instead of raw 'syntax', and
  5630 'translations' with explicit "CONST" markup to accommodate this.
  5631 
  5632 * Pure/Isar: unified syntax for new-style specification mechanisms
  5633 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5634 full type inference and dummy patterns ("_").  For example:
  5635 
  5636   definition "K x _ = x"
  5637 
  5638   inductive conj for A B
  5639   where "A ==> B ==> conj A B"
  5640 
  5641 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5642 the current context.  Print mode "no_abbrevs" prevents inversion of
  5643 abbreviations on output.
  5644 
  5645 * Isar/locales: improved parameter handling: use of locales "var" and
  5646 "struct" no longer necessary; - parameter renamings are no longer
  5647 required to be injective.  For example, this allows to define
  5648 endomorphisms as locale endom = homom mult mult h.
  5649 
  5650 * Isar/locales: changed the way locales with predicates are defined.
  5651 Instead of accumulating the specification, the imported expression is
  5652 now an interpretation.  INCOMPATIBILITY: different normal form of
  5653 locale expressions.  In particular, in interpretations of locales with
  5654 predicates, goals repesenting already interpreted fragments are not
  5655 removed automatically.  Use methods `intro_locales' and
  5656 `unfold_locales'; see below.
  5657 
  5658 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5659 provide backward reasoning on locales predicates.  The methods are
  5660 aware of interpretations and discharge corresponding goals.
  5661 `intro_locales' is less aggressive then `unfold_locales' and does not
  5662 unfold predicates to assumptions.
  5663 
  5664 * Isar/locales: the order in which locale fragments are accumulated
  5665 has changed.  This enables to override declarations from fragments due
  5666 to interpretations -- for example, unwanted simp rules.
  5667 
  5668 * Isar/locales: interpretation in theories and proof contexts has been
  5669 extended.  One may now specify (and prove) equations, which are
  5670 unfolded in interpreted theorems.  This is useful for replacing
  5671 defined concepts (constants depending on locale parameters) by
  5672 concepts already existing in the target context.  Example:
  5673 
  5674   interpretation partial_order ["op <= :: [int, int] => bool"]
  5675     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5676 
  5677 Typically, the constant `partial_order.less' is created by a
  5678 definition specification element in the context of locale
  5679 partial_order.
  5680 
  5681 * Method "induct": improved internal context management to support
  5682 local fixes and defines on-the-fly. Thus explicit meta-level
  5683 connectives !!  and ==> are rarely required anymore in inductive goals
  5684 (using object-logic connectives for this purpose has been long
  5685 obsolete anyway). Common proof patterns are explained in
  5686 src/HOL/Induct/Common_Patterns.thy, see also
  5687 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5688 examples.
  5689 
  5690 * Method "induct": improved handling of simultaneous goals. Instead of
  5691 introducing object-level conjunction, the statement is now split into
  5692 several conclusions, while the corresponding symbolic cases are nested
  5693 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5694 see src/HOL/Induct/Common_Patterns.thy, for example.
  5695 
  5696 * Method "induct": mutual induction rules are now specified as a list
  5697 of rule sharing the same induction cases. HOL packages usually provide
  5698 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5699 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5700 mutual induction rules differently, i.e. like this:
  5701 
  5702   (induct rule: foo_bar.inducts)
  5703   (induct set: foo bar)
  5704   (induct pred: foo bar)
  5705   (induct type: foo bar)
  5706 
  5707 The ML function ProjectRule.projections turns old-style rules into the
  5708 new format.
  5709 
  5710 * Method "coinduct": dual of induction, see
  5711 src/HOL/Library/Coinductive_List.thy for various examples.
  5712 
  5713 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5714 considered a legacy feature.
  5715 
  5716 * Attribute "symmetric" produces result with standardized schematic
  5717 variables (index 0).  Potential INCOMPATIBILITY.
  5718 
  5719 * Simplifier: by default the simplifier trace only shows top level
  5720 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5721 default. Thus there is less danger of being flooded by the trace. The
  5722 trace indicates where parts have been suppressed.
  5723   
  5724 * Provers/classical: removed obsolete classical version of elim_format
  5725 attribute; classical elim/dest rules are now treated uniformly when
  5726 manipulating the claset.
  5727 
  5728 * Provers/classical: stricter checks to ensure that supplied intro,
  5729 dest and elim rules are well-formed; dest and elim rules must have at
  5730 least one premise.
  5731 
  5732 * Provers/classical: attributes dest/elim/intro take an optional
  5733 weight argument for the rule (just as the Pure versions).  Weights are
  5734 ignored by automated tools, but determine the search order of single
  5735 rule steps.
  5736 
  5737 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5738 where the body does not mention the bound variable.  Note that dummy
  5739 patterns implicitly depend on their context of bounds, which makes
  5740 "{_. _}" match any set comprehension as expected.  Potential
  5741 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5742 constant "_idtdummy" in the binding position.
  5743 
  5744 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5745 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5746 for example "A -> B" => "Pi A (%_. B)".
  5747 
  5748 * Pure: 'class_deps' command visualizes the subclass relation, using
  5749 the graph browser tool.
  5750 
  5751 * Pure: 'print_theory' now suppresses certain internal declarations by
  5752 default; use '!' option for full details.
  5753 
  5754 
  5755 *** HOL ***
  5756 
  5757 * Method "metis" proves goals by applying the Metis general-purpose
  5758 resolution prover (see also http://gilith.com/software/metis/).
  5759 Examples are in the directory MetisExamples.  WARNING: the
  5760 Isabelle/HOL-Metis integration does not yet work properly with
  5761 multi-threading.
  5762   
  5763 * Command 'sledgehammer' invokes external automatic theorem provers as
  5764 background processes.  It generates calls to the "metis" method if
  5765 successful. These can be pasted into the proof.  Users do not have to
  5766 wait for the automatic provers to return.  WARNING: does not really
  5767 work with multi-threading.
  5768 
  5769 * New "auto_quickcheck" feature tests outermost goal statements for
  5770 potential counter-examples.  Controlled by ML references
  5771 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5772 5000 milliseconds).  Fails silently if statements is outside of
  5773 executable fragment, or any other codgenerator problem occurs.
  5774 
  5775 * New constant "undefined" with axiom "undefined x = undefined".
  5776 
  5777 * Added class "HOL.eq", allowing for code generation with polymorphic
  5778 equality.
  5779 
  5780 * Some renaming of class constants due to canonical name prefixing in
  5781 the new 'class' package:
  5782 
  5783     HOL.abs ~> HOL.abs_class.abs
  5784     HOL.divide ~> HOL.divide_class.divide
  5785     0 ~> HOL.zero_class.zero
  5786     1 ~> HOL.one_class.one
  5787     op + ~> HOL.plus_class.plus
  5788     op - ~> HOL.minus_class.minus
  5789     uminus ~> HOL.minus_class.uminus
  5790     op * ~> HOL.times_class.times
  5791     op < ~> HOL.ord_class.less
  5792     op <= > HOL.ord_class.less_eq
  5793     Nat.power ~> Power.power_class.power
  5794     Nat.size ~> Nat.size_class.size
  5795     Numeral.number_of ~> Numeral.number_class.number_of
  5796     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5797     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5798     Orderings.min ~> Orderings.ord_class.min
  5799     Orderings.max ~> Orderings.ord_class.max
  5800     Divides.op div ~> Divides.div_class.div
  5801     Divides.op mod ~> Divides.div_class.mod
  5802     Divides.op dvd ~> Divides.div_class.dvd
  5803 
  5804 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5805 
  5806 a) User-defined constants using any of the names "plus", "minus",
  5807 "times", "less" or "less_eq". The standard syntax translations for
  5808 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5809 names.
  5810 
  5811 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5812 INCOMPATIBILITY: use more specific names.
  5813 
  5814 c) Permutative equations (e.g. "a + b = b + a")
  5815 Since the change of names also changes the order of terms, permutative
  5816 rewrite rules may get applied in a different order. Experience shows
  5817 that this is rarely the case (only two adaptions in the whole Isabelle
  5818 distribution).  INCOMPATIBILITY: rewrite proofs
  5819 
  5820 d) ML code directly refering to constant names
  5821 This in general only affects hand-written proof tactics, simprocs and
  5822 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5823 Consider using @{const_name} antiquotation.
  5824 
  5825 * New class "default" with associated constant "default".
  5826 
  5827 * Function "sgn" is now overloaded and available on int, real, complex
  5828 (and other numeric types), using class "sgn".  Two possible defs of
  5829 sgn are given as equational assumptions in the classes sgn_if and
  5830 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5831 INCOMPATIBILITY.
  5832 
  5833 * Locale "partial_order" now unified with class "order" (cf. theory
  5834 Orderings), added parameter "less".  INCOMPATIBILITY.
  5835 
  5836 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5837 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5838 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5839 
  5840 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5841 changed order of proof goals in instance proofs.
  5842 
  5843 * The transitivity reasoner for partial and linear orders is set up
  5844 for classes "order" and "linorder".  Instances of the reasoner are available
  5845 in all contexts importing or interpreting the corresponding locales.
  5846 Method "order" invokes the reasoner separately; the reasoner
  5847 is also integrated with the Simplifier as a solver.  Diagnostic
  5848 command 'print_orders' shows the available instances of the reasoner
  5849 in the current context.
  5850 
  5851 * Localized monotonicity predicate in theory "Orderings"; integrated
  5852 lemmas max_of_mono and min_of_mono with this predicate.
  5853 INCOMPATIBILITY.
  5854 
  5855 * Formulation of theorem "dense" changed slightly due to integration
  5856 with new class dense_linear_order.
  5857 
  5858 * Uniform lattice theory development in HOL.
  5859 
  5860     constants "meet" and "join" now named "inf" and "sup"
  5861     constant "Meet" now named "Inf"
  5862 
  5863     classes "meet_semilorder" and "join_semilorder" now named
  5864       "lower_semilattice" and "upper_semilattice"
  5865     class "lorder" now named "lattice"
  5866     class "comp_lat" now named "complete_lattice"
  5867 
  5868     Instantiation of lattice classes allows explicit definitions
  5869     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5870 
  5871   INCOMPATIBILITY.  Theorem renames:
  5872 
  5873     meet_left_le            ~> inf_le1
  5874     meet_right_le           ~> inf_le2
  5875     join_left_le            ~> sup_ge1
  5876     join_right_le           ~> sup_ge2
  5877     meet_join_le            ~> inf_sup_ord
  5878     le_meetI                ~> le_infI
  5879     join_leI                ~> le_supI
  5880     le_meet                 ~> le_inf_iff
  5881     le_join                 ~> ge_sup_conv
  5882     meet_idempotent         ~> inf_idem
  5883     join_idempotent         ~> sup_idem
  5884     meet_comm               ~> inf_commute
  5885     join_comm               ~> sup_commute
  5886     meet_leI1               ~> le_infI1
  5887     meet_leI2               ~> le_infI2
  5888     le_joinI1               ~> le_supI1
  5889     le_joinI2               ~> le_supI2
  5890     meet_assoc              ~> inf_assoc
  5891     join_assoc              ~> sup_assoc
  5892     meet_left_comm          ~> inf_left_commute
  5893     meet_left_idempotent    ~> inf_left_idem
  5894     join_left_comm          ~> sup_left_commute
  5895     join_left_idempotent    ~> sup_left_idem
  5896     meet_aci                ~> inf_aci
  5897     join_aci                ~> sup_aci
  5898     le_def_meet             ~> le_iff_inf
  5899     le_def_join             ~> le_iff_sup
  5900     join_absorp2            ~> sup_absorb2
  5901     join_absorp1            ~> sup_absorb1
  5902     meet_absorp1            ~> inf_absorb1
  5903     meet_absorp2            ~> inf_absorb2
  5904     meet_join_absorp        ~> inf_sup_absorb
  5905     join_meet_absorp        ~> sup_inf_absorb
  5906     distrib_join_le         ~> distrib_sup_le
  5907     distrib_meet_le         ~> distrib_inf_le
  5908 
  5909     add_meet_distrib_left   ~> add_inf_distrib_left
  5910     add_join_distrib_left   ~> add_sup_distrib_left
  5911     is_join_neg_meet        ~> is_join_neg_inf
  5912     is_meet_neg_join        ~> is_meet_neg_sup
  5913     add_meet_distrib_right  ~> add_inf_distrib_right
  5914     add_join_distrib_right  ~> add_sup_distrib_right
  5915     add_meet_join_distribs  ~> add_sup_inf_distribs
  5916     join_eq_neg_meet        ~> sup_eq_neg_inf
  5917     meet_eq_neg_join        ~> inf_eq_neg_sup
  5918     add_eq_meet_join        ~> add_eq_inf_sup
  5919     meet_0_imp_0            ~> inf_0_imp_0
  5920     join_0_imp_0            ~> sup_0_imp_0
  5921     meet_0_eq_0             ~> inf_0_eq_0
  5922     join_0_eq_0             ~> sup_0_eq_0
  5923     neg_meet_eq_join        ~> neg_inf_eq_sup
  5924     neg_join_eq_meet        ~> neg_sup_eq_inf
  5925     join_eq_if              ~> sup_eq_if
  5926 
  5927     mono_meet               ~> mono_inf
  5928     mono_join               ~> mono_sup
  5929     meet_bool_eq            ~> inf_bool_eq
  5930     join_bool_eq            ~> sup_bool_eq
  5931     meet_fun_eq             ~> inf_fun_eq
  5932     join_fun_eq             ~> sup_fun_eq
  5933     meet_set_eq             ~> inf_set_eq
  5934     join_set_eq             ~> sup_set_eq
  5935     meet1_iff               ~> inf1_iff
  5936     meet2_iff               ~> inf2_iff
  5937     meet1I                  ~> inf1I
  5938     meet2I                  ~> inf2I
  5939     meet1D1                 ~> inf1D1
  5940     meet2D1                 ~> inf2D1
  5941     meet1D2                 ~> inf1D2
  5942     meet2D2                 ~> inf2D2
  5943     meet1E                  ~> inf1E
  5944     meet2E                  ~> inf2E
  5945     join1_iff               ~> sup1_iff
  5946     join2_iff               ~> sup2_iff
  5947     join1I1                 ~> sup1I1
  5948     join2I1                 ~> sup2I1
  5949     join1I1                 ~> sup1I1
  5950     join2I2                 ~> sup1I2
  5951     join1CI                 ~> sup1CI
  5952     join2CI                 ~> sup2CI
  5953     join1E                  ~> sup1E
  5954     join2E                  ~> sup2E
  5955 
  5956     is_meet_Meet            ~> is_meet_Inf
  5957     Meet_bool_def           ~> Inf_bool_def
  5958     Meet_fun_def            ~> Inf_fun_def
  5959     Meet_greatest           ~> Inf_greatest
  5960     Meet_lower              ~> Inf_lower
  5961     Meet_set_def            ~> Inf_set_def
  5962 
  5963     Sup_def                 ~> Sup_Inf
  5964     Sup_bool_eq             ~> Sup_bool_def
  5965     Sup_fun_eq              ~> Sup_fun_def
  5966     Sup_set_eq              ~> Sup_set_def
  5967 
  5968     listsp_meetI            ~> listsp_infI
  5969     listsp_meet_eq          ~> listsp_inf_eq
  5970 
  5971     meet_min                ~> inf_min
  5972     join_max                ~> sup_max
  5973 
  5974 * Added syntactic class "size"; overloaded constant "size" now has
  5975 type "'a::size ==> bool"
  5976 
  5977 * Internal reorganisation of `size' of datatypes: size theorems
  5978 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5979 simplification rules by default!); theorems "prod.size" now named
  5980 "*.size".
  5981 
  5982 * Class "div" now inherits from class "times" rather than "type".
  5983 INCOMPATIBILITY.
  5984 
  5985 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5986 Linorder etc.  have disappeared; operations defined in terms of
  5987 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5988 
  5989 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5990 
  5991 * HOL-Word: New extensive library and type for generic, fixed size
  5992 machine words, with arithmetic, bit-wise, shifting and rotating
  5993 operations, reflection into int, nat, and bool lists, automation for
  5994 linear arithmetic (by automatic reflection into nat or int), including
  5995 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5996 arithmetic type classes, supporting automatic simplification of
  5997 numerals on all operations.
  5998 
  5999 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  6000 
  6001 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  6002 
  6003 * Code generator library theories:
  6004   - Code_Integer represents HOL integers by big integer literals in target
  6005     languages.
  6006   - Code_Char represents HOL characters by character literals in target
  6007     languages.
  6008   - Code_Char_chr like Code_Char, but also offers treatment of character
  6009     codes; includes Code_Integer.
  6010   - Executable_Set allows to generate code for finite sets using lists.
  6011   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  6012     denominator).
  6013   - Executable_Real implements a subset of real numbers, namly those
  6014     representable by rational numbers.
  6015   - Efficient_Nat implements natural numbers by integers, which in general will
  6016     result in higher efficency; pattern matching with 0/Suc is eliminated;
  6017     includes Code_Integer.
  6018   - Code_Index provides an additional datatype index which is mapped to
  6019     target-language built-in integers.
  6020   - Code_Message provides an additional datatype message_string which is isomorphic to
  6021     strings; messages are mapped to target-language strings.
  6022 
  6023 * New package for inductive predicates
  6024 
  6025   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  6026 
  6027     inductive
  6028       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  6029       for z_1 :: U_1 and ... and z_n :: U_m
  6030     where
  6031       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  6032     | ...
  6033 
  6034   with full support for type-inference, rather than
  6035 
  6036     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  6037 
  6038     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  6039     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  6040 
  6041     inductive "s z_1 ... z_m"
  6042     intros
  6043       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  6044       ...
  6045 
  6046   For backward compatibility, there is a wrapper allowing inductive
  6047   sets to be defined with the new package via
  6048 
  6049     inductive_set
  6050       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  6051       for z_1 :: U_1 and ... and z_n :: U_m
  6052     where
  6053       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  6054     | ...
  6055 
  6056   or
  6057 
  6058     inductive_set
  6059       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  6060       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  6061       for z_1 :: U_1 and ... and z_n :: U_m
  6062     where
  6063       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  6064     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  6065     | ...
  6066 
  6067   if the additional syntax "p ..." is required.
  6068 
  6069   Numerous examples can be found in the subdirectories src/HOL/Auth,
  6070   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  6071 
  6072   INCOMPATIBILITIES:
  6073 
  6074   - Since declaration and definition of inductive sets or predicates
  6075     is no longer separated, abbreviations involving the newly
  6076     introduced sets or predicates must be specified together with the
  6077     introduction rules after the 'where' keyword (see above), rather
  6078     than before the actual inductive definition.
  6079 
  6080   - The variables in induction and elimination rules are now
  6081     quantified in the order of their occurrence in the introduction
  6082     rules, rather than in alphabetical order. Since this may break
  6083     some proofs, these proofs either have to be repaired, e.g. by
  6084     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  6085     statements of the form
  6086 
  6087       case (rule_i a_i_1 ... a_i_{k_i})
  6088 
  6089     or the old order of quantification has to be restored by explicitly adding
  6090     meta-level quantifiers in the introduction rules, i.e.
  6091 
  6092       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  6093 
  6094   - The format of the elimination rules is now
  6095 
  6096       p z_1 ... z_m x_1 ... x_n ==>
  6097         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  6098         ==> ... ==> P
  6099 
  6100     for predicates and
  6101 
  6102       (x_1, ..., x_n) : s z_1 ... z_m ==>
  6103         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  6104         ==> ... ==> P
  6105 
  6106     for sets rather than
  6107 
  6108       x : s z_1 ... z_m ==>
  6109         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  6110         ==> ... ==> P
  6111 
  6112     This may require terms in goals to be expanded to n-tuples
  6113     (e.g. using case_tac or simplification with the split_paired_all
  6114     rule) before the above elimination rule is applicable.
  6115 
  6116   - The elimination or case analysis rules for (mutually) inductive
  6117     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  6118     list of rules "p_1_..._p_k.elims" is no longer available.
  6119 
  6120 * New package "function"/"fun" for general recursive functions,
  6121 supporting mutual and nested recursion, definitions in local contexts,
  6122 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  6123 for small examples, and the separate tutorial on the function
  6124 package. The old recdef "package" is still available as before, but
  6125 users are encouraged to use the new package.
  6126 
  6127 * Method "lexicographic_order" automatically synthesizes termination
  6128 relations as lexicographic combinations of size measures. 
  6129 
  6130 * Case-expressions allow arbitrary constructor-patterns (including
  6131 "_") and take their order into account, like in functional
  6132 programming.  Internally, this is translated into nested
  6133 case-expressions; missing cases are added and mapped to the predefined
  6134 constant "undefined". In complicated cases printing may no longer show
  6135 the original input but the internal form. Lambda-abstractions allow
  6136 the same form of pattern matching: "% pat1 => e1 | ..." is an
  6137 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  6138 variable.
  6139 
  6140 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  6141 is an abbreviation for "of_nat :: nat => int". The simplification
  6142 rules for "of_nat" have been changed to work like "int" did
  6143 previously.  Potential INCOMPATIBILITY:
  6144   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  6145   - of_nat_diff and of_nat_mult are no longer default simp rules
  6146 
  6147 * Method "algebra" solves polynomial equations over (semi)rings using
  6148 Groebner bases. The (semi)ring structure is defined by locales and the
  6149 tool setup depends on that generic context. Installing the method for
  6150 a specific type involves instantiating the locale and possibly adding
  6151 declarations for computation on the coefficients.  The method is
  6152 already instantiated for natural numbers and for the axiomatic class
  6153 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  6154 CALCULEMUS 2007 for the general principles underlying this
  6155 architecture of context-aware proof-tools.
  6156 
  6157 * Method "ferrack" implements quantifier elimination over
  6158 special-purpose dense linear orders using locales (analogous to
  6159 "algebra"). The method is already installed for class
  6160 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  6161 rat, etc.
  6162 
  6163 * Former constant "List.op @" now named "List.append".  Use ML
  6164 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  6165 circumvent possible incompatibilities when working on ML level.
  6166 
  6167 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  6168 
  6169 * New function listsum :: 'a list => 'a for arbitrary monoids.
  6170 Special syntax: "SUM x <- xs. f x" (and latex variants)
  6171 
  6172 * New syntax for Haskell-like list comprehension (input only), eg.
  6173 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  6174 
  6175 * The special syntax for function "filter" has changed from [x :
  6176 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  6177 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  6178 
  6179 * [a..b] is now defined for arbitrary linear orders.  It used to be
  6180 defined on nat only, as an abbreviation for [a..<Suc b]
  6181 INCOMPATIBILITY.
  6182 
  6183 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  6184 
  6185 * New functions "sorted" and "sort" in src/HOL/List.thy.
  6186 
  6187 * New lemma collection field_simps (an extension of ring_simps) for
  6188 manipulating (in)equations involving division. Multiplies with all
  6189 denominators that can be proved to be non-zero (in equations) or
  6190 positive/negative (in inequations).
  6191 
  6192 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  6193 have been improved and renamed to ring_simps, group_simps and
  6194 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  6195 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  6196 
  6197 * Theory Library/Commutative_Ring: switched from recdef to function
  6198 package; constants add, mul, pow now curried.  Infix syntax for
  6199 algebraic operations.
  6200 
  6201 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  6202 INCOMPATIBILITY.
  6203 
  6204 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  6205 INCOMPATIBILITY.
  6206 
  6207 * HOL/records: generalised field-update to take a function on the
  6208 field rather than the new value: r(|A := x|) is translated to A_update
  6209 (K x) r The K-combinator that is internally used is called K_record.
  6210 INCOMPATIBILITY: Usage of the plain update functions has to be
  6211 adapted.
  6212  
  6213 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  6214 * x = 0, which are required for a semiring.  Richer structures do not
  6215 inherit from semiring_0 anymore, because this property is a theorem
  6216 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  6217 there is more to prove, but this is mostly trivial.
  6218 
  6219 * Class "recpower" is generalized to arbitrary monoids, not just
  6220 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  6221 commutativity or semiring properties additionally.
  6222 
  6223 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  6224 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  6225 better use 'abbreviation'.
  6226 
  6227 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  6228 
  6229 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  6230 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  6231 significant changes for setting up numeral syntax for types:
  6232   - New constants Numeral.pred and Numeral.succ instead
  6233       of former Numeral.bin_pred and Numeral.bin_succ.
  6234   - Use integer operations instead of bin_add, bin_mult and so on.
  6235   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  6236   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  6237 
  6238 See src/HOL/Integ/IntArith.thy for an example setup.
  6239 
  6240 * Command 'normal_form' computes the normal form of a term that may
  6241 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  6242 produces ``[b, c, a]'' (without proof).  This command is suitable for
  6243 heavy-duty computations because the functions are compiled to ML