NEWS
author wenzelm
Sat Nov 24 14:50:19 2012 +0100 (2012-11-24)
changeset 50183 2b3e24e1c9e7
parent 50182 30177ec0be36
child 50184 5a16f42a9b44
permissions -rw-r--r--
improved editing support for control styles;
separate module for Isabelle actions;
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Theorem status about oracles and unfinished/failed future proofs is
    10 no longer printed by default, since it is incompatible with
    11 incremental / parallel checking of the persistent document model.  ML
    12 function Thm.peek_status may be used to inspect a snapshot of the
    13 ongoing evaluation process.  Note that in batch mode --- notably
    14 isabelle build --- the system ensures that future proofs of all
    15 accessible theorems in the theory context are finished (as before).
    16 
    17 * Configuration option show_markup controls direct inlining of markup
    18 into the printed representation of formal entities --- notably type
    19 and sort constraints.  This enables Prover IDE users to retrieve that
    20 information via tooltips in the output window, for example.
    21 
    22 * Command 'ML_file' evaluates ML text from a file directly within the
    23 theory, without any predeclaration via 'uses' in the theory header.
    24 
    25 * Old command 'use' command and corresponding keyword 'uses' in the
    26 theory header are legacy features and will be discontinued soon.
    27 Tools that load their additional source files may imitate the
    28 'ML_file' implementation, such that the system can take care of
    29 dependencies properly.
    30 
    31 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    32 is called fastforce / fast_force_tac already since Isabelle2011-1.
    33 
    34 * Updated and extended "isar-ref" and "implementation" manual, reduced
    35 remaining material in old "ref" manual.
    36 
    37 * Improved support for auxiliary contexts indicate block structure for
    38 specifications: nesting of "context fixes ... context assumes ..."
    39 and "class ... context ...".
    40 
    41 * More informative error messages for Isar proof commands involving
    42 lazy enumerations (method applications etc.).
    43 
    44 
    45 *** Prover IDE -- Isabelle/Scala/jEdit ***
    46 
    47 * Parallel terminal proofs ('by') are enabled by default, likewise
    48 proofs that are built into packages like 'datatype', 'function'.  This
    49 allows to "run ahead" checking the theory specifications on the
    50 surface, while the prover is still crunching on internal
    51 justifications.  Unfinished / cancelled proofs are restarted as
    52 required to complete full proof checking eventually.
    53 
    54 * Improved output panel with tooltips, hyperlinks etc. based on the
    55 same Rich_Text_Area as regular Isabelle/jEdit buffers.  Activation of
    56 tooltips leads to some window that supports the same recursively,
    57 which can lead to stacks of tooltips as the semantic document content
    58 is explored.  ESCAPE closes the whole stack, individual windows may be
    59 closed separately, or detached to become independent jEdit dockables.
    60 
    61 * More robust incremental parsing of outer syntax (partial comments,
    62 malformed symbols).  Changing the balance of open/close quotes and
    63 comment delimiters works more conveniently with unfinished situations
    64 that frequently occur in user interaction.
    65 
    66 * More efficient painting and improved reactivity when editing large
    67 files.  More scalable management of formal document content.
    68 
    69 * Smarter handling of tracing messages: output window informs about
    70 accumulated messages; prover transactions are limited to emit maximum
    71 amount of output, before being canceled (cf. system option
    72 "editor_tracing_limit").  This avoids swamping the front-end with
    73 potentially infinite message streams.
    74 
    75 * More plugin options and preferences, based on Isabelle/Scala.  The
    76 jEdit plugin option panel provides access to some Isabelle/Scala
    77 options, including tuning parameters for editor reactivity and color
    78 schemes.
    79 
    80 * Improved editing support for control styles: subscript, superscript,
    81 bold, reset of style -- operating on single symbols or text
    82 selections.  Cf. keyboard short-cuts C+e DOWN/UP/RIGHT/LEFT.
    83 
    84 * Uniform Java 7 platform on Linux, Mac OS X, Windows: recent updates
    85 from Oracle provide better multi-platform experience.  This version is
    86 now bundled exclusively with Isabelle.
    87 
    88 
    89 *** Pure ***
    90 
    91 * Code generation for Haskell: restrict unqualified imports from
    92 Haskell Prelude to a small set of fundamental operations.
    93 
    94 * Command "export_code": relative file names are interpreted
    95 relatively to master directory of current theory rather than
    96 the rather arbitrary current working directory.
    97 INCOMPATIBILITY.
    98 
    99 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
   100 use regular rule composition via "OF" / "THEN", or explicit proof
   101 structure instead.  Note that Isabelle/ML provides a variety of
   102 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
   103 with some care where this is really required.
   104 
   105 * Command 'typ' supports an additional variant with explicit sort
   106 constraint, to infer and check the most general type conforming to a
   107 given given sort.  Example (in HOL):
   108 
   109   typ "_ * _ * bool * unit" :: finite
   110 
   111 
   112 *** HOL ***
   113 
   114 * Removed constant "chars".  Prefer "Enum.enum" on type "char"
   115 directly.  INCOMPATIBILITY.
   116 
   117 * Moved operation product, sublists and n_lists from Enum.thy
   118 to List.thy.  INCOMPATIBILITY.
   119 
   120 * Simplified 'typedef' specifications: historical options for implicit
   121 set definition and alternative name have been discontinued.  The
   122 former behavior of "typedef (open) t = A" is now the default, but
   123 written just "typedef t = A".  INCOMPATIBILITY, need to adapt theories
   124 accordingly.
   125 
   126 * Theory "Library/Multiset":
   127 
   128   - Renamed constants
   129       fold_mset ~> Multiset.fold  -- for coherence with other fold combinators
   130 
   131   - Renamed facts
   132       fold_mset_commute ~> fold_mset_comm  -- for coherence with fold_comm
   133 
   134 INCOMPATIBILITY.
   135 
   136 * Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
   137 
   138 * Class "comm_monoid_diff" formalises properties of bounded
   139 subtraction, with natural numbers and multisets as typical instances.
   140 
   141 * Theory "Library/Option_ord" provides instantiation of option type
   142 to lattice type classes.
   143 
   144 * New combinator "Option.these" with type "'a option set => 'a set".
   145 
   146 * Renamed theory Library/List_Prefix to Library/Sublist.
   147 INCOMPATIBILITY.  Related changes are:
   148 
   149   - Renamed constants:
   150 
   151       prefix ~> prefixeq
   152       strict_prefix ~> prefix
   153 
   154     Renamed lemmas accordingly, INCOMPATIBILITY.
   155 
   156   - Replaced constant "postfix" by "suffixeq" with swapped argument order
   157     (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix
   158     syntax "xs >>= ys"; use "suffixeq ys xs" instead.  Renamed lemmas
   159     accordingly.  INCOMPATIBILITY.
   160 
   161   - New constant "emb" for homeomorphic embedding on lists. New
   162     abbreviation "sub" for special case "emb (op =)".
   163 
   164   - Library/Sublist does no longer provide "order" and "bot" type class
   165     instances for the prefix order (merely corresponding locale
   166     interpretations). The type class instances are to be found in
   167     Library/Prefix_Order. INCOMPATIBILITY.
   168 
   169   - The sublist relation from Library/Sublist_Order is now based on
   170     "Sublist.sub". Replaced lemmas:
   171 
   172       le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff
   173       le_list_append_mono ~> Sublist.emb_append_mono
   174       le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2
   175       le_list_Cons_EX ~> Sublist.emb_ConsD
   176       le_list_drop_Cons2 ~> Sublist.sub_Cons2'
   177       le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq
   178       le_list_drop_Cons ~> Sublist.sub_Cons'
   179       le_list_drop_many ~> Sublist.sub_drop_many
   180       le_list_filter_left ~> Sublist.sub_filter_left
   181       le_list_rev_drop_many ~> Sublist.sub_rev_drop_many
   182       le_list_rev_take_iff ~> Sublist.sub_append
   183       le_list_same_length ~> Sublist.sub_same_length
   184       le_list_take_many_iff ~> Sublist.sub_append'
   185       less_eq_list.drop ~> less_eq_list_drop
   186       less_eq_list.induct ~> less_eq_list_induct
   187       not_le_list_length ~> Sublist.not_sub_length
   188 
   189     INCOMPATIBILITY.
   190 
   191 * HOL/Rings: renamed lemmas
   192 
   193 left_distrib ~> distrib_right
   194 right_distrib ~> distrib_left
   195 
   196 in class semiring.  INCOMPATIBILITY.
   197 
   198 * HOL/BNF: New (co)datatype package based on bounded natural
   199 functors with support for mixed, nested recursion and interesting
   200 non-free datatypes.
   201 
   202 * HOL/Cardinals: Theories of ordinals and cardinals
   203 (supersedes the AFP entry "Ordinals_and_Cardinals").
   204 
   205 * HOL/Probability:
   206   - Add simproc "measurable" to automatically prove measurability
   207 
   208   - Add induction rules for sigma sets with disjoint union (sigma_sets_induct_disjoint)
   209     and for Borel-measurable functions (borel_measurable_induct).
   210 
   211   - The Daniell-Kolmogorov theorem (the existence the limit of a projective family)
   212 
   213 * Library/Countable_Set.thy: Theory of countable sets.
   214 
   215 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
   216 execution for code generated towards Isabelle/ML.
   217 
   218 * Library/FuncSet.thy: Extended support for Pi and extensional and introduce the
   219 extensional dependent function space "PiE". Replaces extensional_funcset by an
   220 abbreviation, rename a couple of lemmas from extensional_funcset to PiE:
   221 
   222       extensional_empty ~> PiE_empty
   223       extensional_funcset_empty_domain ~> PiE_empty_domain
   224       extensional_funcset_empty_range ~> PiE_empty_range
   225       extensional_funcset_arb ~> PiE_arb
   226       extensional_funcset_mem > PiE_mem
   227       extensional_funcset_extend_domainI ~> PiE_fun_upd
   228       extensional_funcset_restrict_domain ~> fun_upd_in_PiE
   229       extensional_funcset_extend_domain_eq ~> PiE_insert_eq
   230       card_extensional_funcset ~> card_PiE
   231       finite_extensional_funcset ~> finite_PiE
   232 
   233   INCOMPATIBUILITY.
   234 
   235 * Library/FinFun.thy: theory of almost everywhere constant functions
   236 (supersedes the AFP entry "Code Generation for Functions as Data").
   237 
   238 * Library/Phantom.thy: generic phantom type to make a type parameter
   239 appear in a constant's type. This alternative to adding TYPE('a) as
   240 another parameter avoids unnecessary closures in generated code.
   241 
   242 * Library/RBT_Impl.thy: efficient construction of red-black trees 
   243 from sorted associative lists. Merging two trees with rbt_union may
   244 return a structurally different tree than before. MINOR INCOMPATIBILITY.
   245 
   246 * Library/IArray.thy: immutable arrays with code generation.
   247 
   248 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   249 expressions.
   250 
   251 * Preprocessing of the code generator rewrites set comprehensions into
   252 pointfree expressions.
   253 
   254 * Quickcheck:
   255 
   256   - added an optimisation for equality premises.
   257     It is switched on by default, and can be switched off by setting
   258     the configuration quickcheck_optimise_equality to false.
   259 
   260 * The SMT solver Z3 has now by default a restricted set of directly
   261 supported features. For the full set of features (div/mod, nonlinear
   262 arithmetic, datatypes/records) with potential proof reconstruction
   263 failures, enable the configuration option "z3_with_extensions".
   264 Minor INCOMPATIBILITY.
   265 
   266 * Sledgehammer:
   267 
   268   - Added MaSh relevance filter based on machine-learning; see the
   269     Sledgehammer manual for details.
   270   - Rationalized type encodings ("type_enc" option).
   271   - Renamed "kill_provers" subcommand to "kill"
   272   - Renamed options:
   273       isar_proof ~> isar_proofs
   274       isar_shrink_factor ~> isar_shrink
   275       max_relevant ~> max_facts
   276       relevance_thresholds ~> fact_thresholds
   277 
   278 
   279 *** Document preparation ***
   280 
   281 * Default for \<euro> is now based on eurosym package, instead of
   282 slightly exotic babel/greek.
   283 
   284 * Document variant NAME may use different LaTeX entry point
   285 document/root_NAME.tex if that file exists, instead of the common
   286 document/root.tex.
   287 
   288 * Simplified custom document/build script, instead of old-style
   289 document/IsaMakefile.  Minor INCOMPATIBILITY.
   290 
   291 
   292 *** ML ***
   293 
   294 * Type Seq.results and related operations support embedded error
   295 messages within lazy enumerations, and thus allow to provide
   296 informative errors in the absence of any usable results.
   297 
   298 * Renamed Position.str_of to Position.here to emphasize that this is a
   299 formal device to inline positions into message text, but not
   300 necessarily printing visible text.
   301 
   302 
   303 *** System ***
   304 
   305 * The default limit for maximum number of worker threads is now 8,
   306 instead of 4.
   307 
   308 * The ML system is configured as regular component, and no longer
   309 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   310 for home-made configurations.
   311 
   312 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   313 Minor INCOMPATIBILITY in command-line options.
   314 
   315 * Advanced support for Isabelle sessions and build management, see
   316 "system" manual for the chapter of that name, especially the "isabelle
   317 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   318 mkdir / make are rendered obsolete.
   319 
   320 * Discontinued obsolete "isabelle makeall".
   321 
   322 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   323 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   324 provides some traditional targets that invoke "isabelle build".  Note
   325 that this is inefficient!  Applications of Isabelle/HOL involving
   326 "isabelle make" should be upgraded to use "isabelle build" directly.
   327 
   328 * Discontinued obsolete Isabelle/build script, it is superseded by the
   329 regular isabelle build tool.  For example:
   330 
   331   isabelle build -s -b HOLCF
   332 
   333 * The "isabelle options" tool prints Isabelle system options, as
   334 required for "isabelle build", for example.
   335 
   336 * The "isabelle mkroot" tool prepares session root directories for use
   337 with "isabelle build", similar to former "isabelle mkdir" for
   338 "isabelle usedir".
   339 
   340 * The "isabelle components" tool helps to resolve add-on components
   341 that are not bundled, or referenced from a bare-bones repository
   342 version of Isabelle.
   343 
   344 * The "isabelle install" tool has now a simpler command-line.  Minor
   345 INCOMPATIBILITY.
   346 
   347 * Discontinued support for Poly/ML 5.2.1, which was the last version
   348 without exception positions and advanced ML compiler/toplevel
   349 configuration.
   350 
   351 * Discontinued special treatment of Proof General -- no longer guess
   352 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   353 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   354 settings manually, or use a Proof General version that has been
   355 bundled as Isabelle component.
   356 
   357 * Settings variable ISABELLE_PLATFORM_FAMILY refers to the general
   358 platform family: "linux", "macos", "windows".
   359 
   360 
   361 New in Isabelle2012 (May 2012)
   362 ------------------------------
   363 
   364 *** General ***
   365 
   366 * Prover IDE (PIDE) improvements:
   367 
   368   - more robust Sledgehammer integration (as before the sledgehammer
   369     command-line needs to be typed into the source buffer)
   370   - markup for bound variables
   371   - markup for types of term variables (displayed as tooltips)
   372   - support for user-defined Isar commands within the running session
   373   - improved support for Unicode outside original 16bit range
   374     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   375 
   376 * Forward declaration of outer syntax keywords within the theory
   377 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   378 commands to be used in the same theory where defined.
   379 
   380 * Auxiliary contexts indicate block structure for specifications with
   381 additional parameters and assumptions.  Such unnamed contexts may be
   382 nested within other targets, like 'theory', 'locale', 'class',
   383 'instantiation' etc.  Results from the local context are generalized
   384 accordingly and applied to the enclosing target context.  Example:
   385 
   386   context
   387     fixes x y z :: 'a
   388     assumes xy: "x = y" and yz: "y = z"
   389   begin
   390 
   391   lemma my_trans: "x = z" using xy yz by simp
   392 
   393   end
   394 
   395   thm my_trans
   396 
   397 The most basic application is to factor-out context elements of
   398 several fixes/assumes/shows theorem statements, e.g. see
   399 ~~/src/HOL/Isar_Examples/Group_Context.thy
   400 
   401 Any other local theory specification element works within the "context
   402 ... begin ... end" block as well.
   403 
   404 * Bundled declarations associate attributed fact expressions with a
   405 given name in the context.  These may be later included in other
   406 contexts.  This allows to manage context extensions casually, without
   407 the logical dependencies of locales and locale interpretation.  See
   408 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   409 
   410 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   411 declaration, and results are standardized before being stored.  Thus
   412 old-style "standard" after instantiation or composition of facts
   413 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   414 indices of schematic variables.
   415 
   416 * Rule attributes in local theory declarations (e.g. locale or class)
   417 are now statically evaluated: the resulting theorem is stored instead
   418 of the original expression.  INCOMPATIBILITY in rare situations, where
   419 the historic accident of dynamic re-evaluation in interpretations
   420 etc. was exploited.
   421 
   422 * New tutorial "Programming and Proving in Isabelle/HOL"
   423 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   424 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   425 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   426 Logic" as the recommended beginners tutorial, but does not cover all
   427 of the material of that old tutorial.
   428 
   429 * Updated and extended reference manuals: "isar-ref",
   430 "implementation", "system"; reduced remaining material in old "ref"
   431 manual.
   432 
   433 
   434 *** Pure ***
   435 
   436 * Command 'definition' no longer exports the foundational "raw_def"
   437 into the user context.  Minor INCOMPATIBILITY, may use the regular
   438 "def" result with attribute "abs_def" to imitate the old version.
   439 
   440 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   441 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   442 expand it.  This also works for object-logic equality.  (Formerly
   443 undocumented feature.)
   444 
   445 * Sort constraints are now propagated in simultaneous statements, just
   446 like type constraints.  INCOMPATIBILITY in rare situations, where
   447 distinct sorts used to be assigned accidentally.  For example:
   448 
   449   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   450 
   451   lemma "P (x::'a)" and "Q (y::'a::bar)"
   452     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   453 
   454 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   455 tolerant against multiple unifiers, as long as the final result is
   456 unique.  (As before, rules are composed in canonical right-to-left
   457 order to accommodate newly introduced premises.)
   458 
   459 * Renamed some inner syntax categories:
   460 
   461     num ~> num_token
   462     xnum ~> xnum_token
   463     xstr ~> str_token
   464 
   465 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   466 "num_position" etc. are mainly used instead (which also include
   467 position information via constraints).
   468 
   469 * Simplified configuration options for syntax ambiguity: see
   470 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   471 manual.  Minor INCOMPATIBILITY.
   472 
   473 * Discontinued configuration option "syntax_positions": atomic terms
   474 in parse trees are always annotated by position constraints.
   475 
   476 * Old code generator for SML and its commands 'code_module',
   477 'code_library', 'consts_code', 'types_code' have been discontinued.
   478 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   479 
   480 * Redundant attribute "code_inline" has been discontinued. Use
   481 "code_unfold" instead.  INCOMPATIBILITY.
   482 
   483 * Dropped attribute "code_unfold_post" in favor of the its dual
   484 "code_abbrev", which yields a common pattern in definitions like
   485 
   486   definition [code_abbrev]: "f = t"
   487 
   488 INCOMPATIBILITY.
   489 
   490 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   491 instead.  INCOMPATIBILITY.
   492 
   493 * Discontinued old "prems" fact, which used to refer to the accidental
   494 collection of foundational premises in the context (already marked as
   495 legacy since Isabelle2011).
   496 
   497 
   498 *** HOL ***
   499 
   500 * Type 'a set is now a proper type constructor (just as before
   501 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   502 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   503 sets separate, it is often sufficient to rephrase some set S that has
   504 been accidentally used as predicates by "%x. x : S", and some
   505 predicate P that has been accidentally used as set by "{x. P x}".
   506 Corresponding proofs in a first step should be pruned from any
   507 tinkering with former theorems mem_def and Collect_def as far as
   508 possible.
   509 
   510 For developments which deliberately mix predicates and sets, a
   511 planning step is necessary to determine what should become a predicate
   512 and what a set.  It can be helpful to carry out that step in
   513 Isabelle2011-1 before jumping right into the current release.
   514 
   515 * Code generation by default implements sets as container type rather
   516 than predicates.  INCOMPATIBILITY.
   517 
   518 * New type synonym 'a rel = ('a * 'a) set
   519 
   520 * The representation of numerals has changed.  Datatype "num"
   521 represents strictly positive binary numerals, along with functions
   522 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   523 positive and negated numeric literals, respectively.  See also
   524 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   525 user theories may require adaptations as follows:
   526 
   527   - Theorems with number_ring or number_semiring constraints: These
   528     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   529 
   530   - Theories defining numeric types: Remove number, number_semiring,
   531     and number_ring instances. Defer all theorems about numerals until
   532     after classes one and semigroup_add have been instantiated.
   533 
   534   - Numeral-only simp rules: Replace each rule having a "number_of v"
   535     pattern with two copies, one for numeral and one for neg_numeral.
   536 
   537   - Theorems about subclasses of semiring_1 or ring_1: These classes
   538     automatically support numerals now, so more simp rules and
   539     simprocs may now apply within the proof.
   540 
   541   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   542     Redefine using other integer operations.
   543 
   544 * Transfer: New package intended to generalize the existing
   545 "descending" method and related theorem attributes from the Quotient
   546 package.  (Not all functionality is implemented yet, but future
   547 development will focus on Transfer as an eventual replacement for the
   548 corresponding parts of the Quotient package.)
   549 
   550   - transfer_rule attribute: Maintains a collection of transfer rules,
   551     which relate constants at two different types. Transfer rules may
   552     relate different type instances of the same polymorphic constant,
   553     or they may relate an operation on a raw type to a corresponding
   554     operation on an abstract type (quotient or subtype). For example:
   555 
   556     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   557     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   558 
   559   - transfer method: Replaces a subgoal on abstract types with an
   560     equivalent subgoal on the corresponding raw types. Constants are
   561     replaced with corresponding ones according to the transfer rules.
   562     Goals are generalized over all free variables by default; this is
   563     necessary for variables whose types change, but can be overridden
   564     for specific variables with e.g. "transfer fixing: x y z".  The
   565     variant transfer' method allows replacing a subgoal with one that
   566     is logically stronger (rather than equivalent).
   567 
   568   - relator_eq attribute: Collects identity laws for relators of
   569     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   570     transfer method uses these lemmas to infer transfer rules for
   571     non-polymorphic constants on the fly.
   572 
   573   - transfer_prover method: Assists with proving a transfer rule for a
   574     new constant, provided the constant is defined in terms of other
   575     constants that already have transfer rules. It should be applied
   576     after unfolding the constant definitions.
   577 
   578   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   579     from type nat to type int.
   580 
   581 * Lifting: New package intended to generalize the quotient_definition
   582 facility of the Quotient package; designed to work with Transfer.
   583 
   584   - lift_definition command: Defines operations on an abstract type in
   585     terms of a corresponding operation on a representation
   586     type.  Example syntax:
   587 
   588     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   589       is List.insert
   590 
   591     Users must discharge a respectfulness proof obligation when each
   592     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   593     the proof is discharged automatically.) The obligation is
   594     presented in a user-friendly, readable form; a respectfulness
   595     theorem in the standard format and a transfer rule are generated
   596     by the package.
   597 
   598   - Integration with code_abstype: For typedefs (e.g. subtypes
   599     corresponding to a datatype invariant, such as dlist),
   600     lift_definition generates a code certificate theorem and sets up
   601     code generation for each constant.
   602 
   603   - setup_lifting command: Sets up the Lifting package to work with a
   604     user-defined type. The user must provide either a quotient theorem
   605     or a type_definition theorem.  The package configures transfer
   606     rules for equality and quantifiers on the type, and sets up the
   607     lift_definition command to work with the type.
   608 
   609   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   610     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   611     Word/Word.thy and Library/Float.thy.
   612 
   613 * Quotient package:
   614 
   615   - The 'quotient_type' command now supports a 'morphisms' option with
   616     rep and abs functions, similar to typedef.
   617 
   618   - 'quotient_type' sets up new types to work with the Lifting and
   619     Transfer packages, as with 'setup_lifting'.
   620 
   621   - The 'quotient_definition' command now requires the user to prove a
   622     respectfulness property at the point where the constant is
   623     defined, similar to lift_definition; INCOMPATIBILITY.
   624 
   625   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   626     accordingly, INCOMPATIBILITY.
   627 
   628 * New diagnostic command 'find_unused_assms' to find potentially
   629 superfluous assumptions in theorems using Quickcheck.
   630 
   631 * Quickcheck:
   632 
   633   - Quickcheck returns variable assignments as counterexamples, which
   634     allows to reveal the underspecification of functions under test.
   635     For example, refuting "hd xs = x", it presents the variable
   636     assignment xs = [] and x = a1 as a counterexample, assuming that
   637     any property is false whenever "hd []" occurs in it.
   638 
   639     These counterexample are marked as potentially spurious, as
   640     Quickcheck also returns "xs = []" as a counterexample to the
   641     obvious theorem "hd xs = hd xs".
   642 
   643     After finding a potentially spurious counterexample, Quickcheck
   644     continues searching for genuine ones.
   645 
   646     By default, Quickcheck shows potentially spurious and genuine
   647     counterexamples. The option "genuine_only" sets quickcheck to only
   648     show genuine counterexamples.
   649 
   650   - The command 'quickcheck_generator' creates random and exhaustive
   651     value generators for a given type and operations.
   652 
   653     It generates values by using the operations as if they were
   654     constructors of that type.
   655 
   656   - Support for multisets.
   657 
   658   - Added "use_subtype" options.
   659 
   660   - Added "quickcheck_locale" configuration to specify how to process
   661     conjectures in a locale context.
   662 
   663 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   664 and affecting 'rat' and 'real'.
   665 
   666 * Sledgehammer:
   667   - Integrated more tightly with SPASS, as described in the ITP 2012
   668     paper "More SPASS with Isabelle".
   669   - Made it try "smt" as a fallback if "metis" fails or times out.
   670   - Added support for the following provers: Alt-Ergo (via Why3 and
   671     TFF1), iProver, iProver-Eq.
   672   - Sped up the minimizer.
   673   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   674   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   675   - Renamed "sound" option to "strict".
   676 
   677 * Metis: Added possibility to specify lambda translations scheme as a
   678 parenthesized argument (e.g., "by (metis (lifting) ...)").
   679 
   680 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   681 
   682 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   683 
   684 * New "case_product" attribute to generate a case rule doing multiple
   685 case distinctions at the same time.  E.g.
   686 
   687   list.exhaust [case_product nat.exhaust]
   688 
   689 produces a rule which can be used to perform case distinction on both
   690 a list and a nat.
   691 
   692 * New "eventually_elim" method as a generalized variant of the
   693 eventually_elim* rules.  Supports structured proofs.
   694 
   695 * Typedef with implicit set definition is considered legacy.  Use
   696 "typedef (open)" form instead, which will eventually become the
   697 default.
   698 
   699 * Record: code generation can be switched off manually with
   700 
   701   declare [[record_coden = false]]  -- "default true"
   702 
   703 * Datatype: type parameters allow explicit sort constraints.
   704 
   705 * Concrete syntax for case expressions includes constraints for source
   706 positions, and thus produces Prover IDE markup for its bindings.
   707 INCOMPATIBILITY for old-style syntax translations that augment the
   708 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   709 one_case.
   710 
   711 * Clarified attribute "mono_set": pure declaration without modifying
   712 the result of the fact expression.
   713 
   714 * More default pred/set conversions on a couple of relation operations
   715 and predicates.  Added powers of predicate relations.  Consolidation
   716 of some relation theorems:
   717 
   718   converse_def ~> converse_unfold
   719   rel_comp_def ~> relcomp_unfold
   720   symp_def ~> (modified, use symp_def and sym_def instead)
   721   transp_def ~> transp_trans
   722   Domain_def ~> Domain_unfold
   723   Range_def ~> Domain_converse [symmetric]
   724 
   725 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   726 
   727 See theory "Relation" for examples for making use of pred/set
   728 conversions by means of attributes "to_set" and "to_pred".
   729 
   730 INCOMPATIBILITY.
   731 
   732 * Renamed facts about the power operation on relations, i.e., relpow
   733 to match the constant's name:
   734 
   735   rel_pow_1 ~> relpow_1
   736   rel_pow_0_I ~> relpow_0_I
   737   rel_pow_Suc_I ~> relpow_Suc_I
   738   rel_pow_Suc_I2 ~> relpow_Suc_I2
   739   rel_pow_0_E ~> relpow_0_E
   740   rel_pow_Suc_E ~> relpow_Suc_E
   741   rel_pow_E ~> relpow_E
   742   rel_pow_Suc_D2 ~> relpow_Suc_D2
   743   rel_pow_Suc_E2 ~> relpow_Suc_E2
   744   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   745   rel_pow_E2 ~> relpow_E2
   746   rel_pow_add ~> relpow_add
   747   rel_pow_commute ~> relpow
   748   rel_pow_empty ~> relpow_empty:
   749   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   750   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   751   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   752   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   753   rel_pow_fun_conv ~> relpow_fun_conv
   754   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   755   rel_pow_finite_bounded ~> relpow_finite_bounded
   756   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   757   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   758   single_valued_rel_pow ~> single_valued_relpow
   759 
   760 INCOMPATIBILITY.
   761 
   762 * Theory Relation: Consolidated constant name for relation composition
   763 and corresponding theorem names:
   764 
   765   - Renamed constant rel_comp to relcomp.
   766 
   767   - Dropped abbreviation pred_comp. Use relcompp instead.
   768 
   769   - Renamed theorems:
   770 
   771     rel_compI ~> relcompI
   772     rel_compEpair ~> relcompEpair
   773     rel_compE ~> relcompE
   774     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   775     rel_comp_empty1 ~> relcomp_empty1
   776     rel_comp_mono ~> relcomp_mono
   777     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   778     rel_comp_distrib ~> relcomp_distrib
   779     rel_comp_distrib2 ~> relcomp_distrib2
   780     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   781     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   782     single_valued_rel_comp ~> single_valued_relcomp
   783     rel_comp_def ~> relcomp_unfold
   784     converse_rel_comp ~> converse_relcomp
   785     pred_compI ~> relcomppI
   786     pred_compE ~> relcomppE
   787     pred_comp_bot1 ~> relcompp_bot1
   788     pred_comp_bot2 ~> relcompp_bot2
   789     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   790     pred_comp_mono ~> relcompp_mono
   791     pred_comp_distrib ~> relcompp_distrib
   792     pred_comp_distrib2 ~> relcompp_distrib2
   793     converse_pred_comp ~> converse_relcompp
   794 
   795     finite_rel_comp ~> finite_relcomp
   796 
   797     set_rel_comp ~> set_relcomp
   798 
   799 INCOMPATIBILITY.
   800 
   801 * Theory Divides: Discontinued redundant theorems about div and mod.
   802 INCOMPATIBILITY, use the corresponding generic theorems instead.
   803 
   804   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   805   zdiv_self ~> div_self
   806   zmod_self ~> mod_self
   807   zdiv_zero ~> div_0
   808   zmod_zero ~> mod_0
   809   zdiv_zmod_equality ~> div_mod_equality2
   810   zdiv_zmod_equality2 ~> div_mod_equality
   811   zmod_zdiv_trivial ~> mod_div_trivial
   812   zdiv_zminus_zminus ~> div_minus_minus
   813   zmod_zminus_zminus ~> mod_minus_minus
   814   zdiv_zminus2 ~> div_minus_right
   815   zmod_zminus2 ~> mod_minus_right
   816   zdiv_minus1_right ~> div_minus1_right
   817   zmod_minus1_right ~> mod_minus1_right
   818   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   819   zmod_zmult1_eq ~> mod_mult_right_eq
   820   zpower_zmod ~> power_mod
   821   zdvd_zmod ~> dvd_mod
   822   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   823   mod_mult_distrib ~> mult_mod_left
   824   mod_mult_distrib2 ~> mult_mod_right
   825 
   826 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   827 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   828 
   829 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   830 
   831 * Consolidated theorem names concerning fold combinators:
   832 
   833   inf_INFI_fold_inf ~> inf_INF_fold_inf
   834   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   835   INFI_fold_inf ~> INF_fold_inf
   836   SUPR_fold_sup ~> SUP_fold_sup
   837   union_set ~> union_set_fold
   838   minus_set ~> minus_set_fold
   839   INFI_set_fold ~> INF_set_fold
   840   SUPR_set_fold ~> SUP_set_fold
   841   INF_code ~> INF_set_foldr
   842   SUP_code ~> SUP_set_foldr
   843   foldr.simps ~> foldr.simps (in point-free formulation)
   844   foldr_fold_rev ~> foldr_conv_fold
   845   foldl_fold ~> foldl_conv_fold
   846   foldr_foldr ~> foldr_conv_foldl
   847   foldl_foldr ~> foldl_conv_foldr
   848   fold_set_remdups ~> fold_set_fold_remdups
   849   fold_set ~> fold_set_fold
   850   fold1_set ~> fold1_set_fold
   851 
   852 INCOMPATIBILITY.
   853 
   854 * Dropped rarely useful theorems concerning fold combinators:
   855 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   856 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   857 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   858 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   859 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   860 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   861 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   862 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   863 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   864 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   865 
   866 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   867 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   868 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   869 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   870 lemmas over fold rather than foldr, or make use of lemmas
   871 fold_conv_foldr and fold_rev.
   872 
   873 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   874 through option types.
   875 
   876 * "Transitive_Closure.ntrancl": bounded transitive closure on
   877 relations.
   878 
   879 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   880 
   881 * Theory Int: Discontinued many legacy theorems specific to type int.
   882 INCOMPATIBILITY, use the corresponding generic theorems instead.
   883 
   884   zminus_zminus ~> minus_minus
   885   zminus_0 ~> minus_zero
   886   zminus_zadd_distrib ~> minus_add_distrib
   887   zadd_commute ~> add_commute
   888   zadd_assoc ~> add_assoc
   889   zadd_left_commute ~> add_left_commute
   890   zadd_ac ~> add_ac
   891   zmult_ac ~> mult_ac
   892   zadd_0 ~> add_0_left
   893   zadd_0_right ~> add_0_right
   894   zadd_zminus_inverse2 ~> left_minus
   895   zmult_zminus ~> mult_minus_left
   896   zmult_commute ~> mult_commute
   897   zmult_assoc ~> mult_assoc
   898   zadd_zmult_distrib ~> left_distrib
   899   zadd_zmult_distrib2 ~> right_distrib
   900   zdiff_zmult_distrib ~> left_diff_distrib
   901   zdiff_zmult_distrib2 ~> right_diff_distrib
   902   zmult_1 ~> mult_1_left
   903   zmult_1_right ~> mult_1_right
   904   zle_refl ~> order_refl
   905   zle_trans ~> order_trans
   906   zle_antisym ~> order_antisym
   907   zle_linear ~> linorder_linear
   908   zless_linear ~> linorder_less_linear
   909   zadd_left_mono ~> add_left_mono
   910   zadd_strict_right_mono ~> add_strict_right_mono
   911   zadd_zless_mono ~> add_less_le_mono
   912   int_0_less_1 ~> zero_less_one
   913   int_0_neq_1 ~> zero_neq_one
   914   zless_le ~> less_le
   915   zpower_zadd_distrib ~> power_add
   916   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   917   zero_le_zpower_abs ~> zero_le_power_abs
   918 
   919 * Theory Deriv: Renamed
   920 
   921   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   922 
   923 * Theory Library/Multiset: Improved code generation of multisets.
   924 
   925 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   926 are expressed via type classes again. The special syntax
   927 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   928 setsum_set, which is now subsumed by Big_Operators.setsum.
   929 INCOMPATIBILITY.
   930 
   931 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   932 use theory HOL/Library/Nat_Bijection instead.
   933 
   934 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   935 trees is now inside a type class context.  Names of affected
   936 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   937 theories working directly with raw red-black trees, adapt the names as
   938 follows:
   939 
   940   Operations:
   941   bulkload -> rbt_bulkload
   942   del_from_left -> rbt_del_from_left
   943   del_from_right -> rbt_del_from_right
   944   del -> rbt_del
   945   delete -> rbt_delete
   946   ins -> rbt_ins
   947   insert -> rbt_insert
   948   insertw -> rbt_insert_with
   949   insert_with_key -> rbt_insert_with_key
   950   map_entry -> rbt_map_entry
   951   lookup -> rbt_lookup
   952   sorted -> rbt_sorted
   953   tree_greater -> rbt_greater
   954   tree_less -> rbt_less
   955   tree_less_symbol -> rbt_less_symbol
   956   union -> rbt_union
   957   union_with -> rbt_union_with
   958   union_with_key -> rbt_union_with_key
   959 
   960   Lemmas:
   961   balance_left_sorted -> balance_left_rbt_sorted
   962   balance_left_tree_greater -> balance_left_rbt_greater
   963   balance_left_tree_less -> balance_left_rbt_less
   964   balance_right_sorted -> balance_right_rbt_sorted
   965   balance_right_tree_greater -> balance_right_rbt_greater
   966   balance_right_tree_less -> balance_right_rbt_less
   967   balance_sorted -> balance_rbt_sorted
   968   balance_tree_greater -> balance_rbt_greater
   969   balance_tree_less -> balance_rbt_less
   970   bulkload_is_rbt -> rbt_bulkload_is_rbt
   971   combine_sorted -> combine_rbt_sorted
   972   combine_tree_greater -> combine_rbt_greater
   973   combine_tree_less -> combine_rbt_less
   974   delete_in_tree -> rbt_delete_in_tree
   975   delete_is_rbt -> rbt_delete_is_rbt
   976   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   977   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   978   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   979   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   980   del_in_tree -> rbt_del_in_tree
   981   del_inv1_inv2 -> rbt_del_inv1_inv2
   982   del_sorted -> rbt_del_rbt_sorted
   983   del_tree_greater -> rbt_del_rbt_greater
   984   del_tree_less -> rbt_del_rbt_less
   985   dom_lookup_Branch -> dom_rbt_lookup_Branch
   986   entries_lookup -> entries_rbt_lookup
   987   finite_dom_lookup -> finite_dom_rbt_lookup
   988   insert_sorted -> rbt_insert_rbt_sorted
   989   insertw_is_rbt -> rbt_insertw_is_rbt
   990   insertwk_is_rbt -> rbt_insertwk_is_rbt
   991   insertwk_sorted -> rbt_insertwk_rbt_sorted
   992   insertw_sorted -> rbt_insertw_rbt_sorted
   993   ins_sorted -> ins_rbt_sorted
   994   ins_tree_greater -> ins_rbt_greater
   995   ins_tree_less -> ins_rbt_less
   996   is_rbt_sorted -> is_rbt_rbt_sorted
   997   lookup_balance -> rbt_lookup_balance
   998   lookup_bulkload -> rbt_lookup_rbt_bulkload
   999   lookup_delete -> rbt_lookup_rbt_delete
  1000   lookup_Empty -> rbt_lookup_Empty
  1001   lookup_from_in_tree -> rbt_lookup_from_in_tree
  1002   lookup_in_tree -> rbt_lookup_in_tree
  1003   lookup_ins -> rbt_lookup_ins
  1004   lookup_insert -> rbt_lookup_rbt_insert
  1005   lookup_insertw -> rbt_lookup_rbt_insertw
  1006   lookup_insertwk -> rbt_lookup_rbt_insertwk
  1007   lookup_keys -> rbt_lookup_keys
  1008   lookup_map -> rbt_lookup_map
  1009   lookup_map_entry -> rbt_lookup_rbt_map_entry
  1010   lookup_tree_greater -> rbt_lookup_rbt_greater
  1011   lookup_tree_less -> rbt_lookup_rbt_less
  1012   lookup_union -> rbt_lookup_rbt_union
  1013   map_entry_color_of -> rbt_map_entry_color_of
  1014   map_entry_inv1 -> rbt_map_entry_inv1
  1015   map_entry_inv2 -> rbt_map_entry_inv2
  1016   map_entry_is_rbt -> rbt_map_entry_is_rbt
  1017   map_entry_sorted -> rbt_map_entry_rbt_sorted
  1018   map_entry_tree_greater -> rbt_map_entry_rbt_greater
  1019   map_entry_tree_less -> rbt_map_entry_rbt_less
  1020   map_tree_greater -> map_rbt_greater
  1021   map_tree_less -> map_rbt_less
  1022   map_sorted -> map_rbt_sorted
  1023   paint_sorted -> paint_rbt_sorted
  1024   paint_lookup -> paint_rbt_lookup
  1025   paint_tree_greater -> paint_rbt_greater
  1026   paint_tree_less -> paint_rbt_less
  1027   sorted_entries -> rbt_sorted_entries
  1028   tree_greater_eq_trans -> rbt_greater_eq_trans
  1029   tree_greater_nit -> rbt_greater_nit
  1030   tree_greater_prop -> rbt_greater_prop
  1031   tree_greater_simps -> rbt_greater_simps
  1032   tree_greater_trans -> rbt_greater_trans
  1033   tree_less_eq_trans -> rbt_less_eq_trans
  1034   tree_less_nit -> rbt_less_nit
  1035   tree_less_prop -> rbt_less_prop
  1036   tree_less_simps -> rbt_less_simps
  1037   tree_less_trans -> rbt_less_trans
  1038   tree_ord_props -> rbt_ord_props
  1039   union_Branch -> rbt_union_Branch
  1040   union_is_rbt -> rbt_union_is_rbt
  1041   unionw_is_rbt -> rbt_unionw_is_rbt
  1042   unionwk_is_rbt -> rbt_unionwk_is_rbt
  1043   unionwk_sorted -> rbt_unionwk_rbt_sorted
  1044 
  1045 * Theory HOL/Library/Float: Floating point numbers are now defined as
  1046 a subset of the real numbers.  All operations are defined using the
  1047 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
  1048 
  1049   Changed Operations:
  1050   float_abs -> abs
  1051   float_nprt -> nprt
  1052   float_pprt -> pprt
  1053   pow2 -> use powr
  1054   round_down -> float_round_down
  1055   round_up -> float_round_up
  1056   scale -> exponent
  1057 
  1058   Removed Operations:
  1059   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
  1060 
  1061   Renamed Lemmas:
  1062   abs_float_def -> Float.compute_float_abs
  1063   bitlen_ge0 -> bitlen_nonneg
  1064   bitlen.simps -> Float.compute_bitlen
  1065   float_components -> Float_mantissa_exponent
  1066   float_divl.simps -> Float.compute_float_divl
  1067   float_divr.simps -> Float.compute_float_divr
  1068   float_eq_odd -> mult_powr_eq_mult_powr_iff
  1069   float_power -> real_of_float_power
  1070   lapprox_posrat_def -> Float.compute_lapprox_posrat
  1071   lapprox_rat.simps -> Float.compute_lapprox_rat
  1072   le_float_def' -> Float.compute_float_le
  1073   le_float_def -> less_eq_float.rep_eq
  1074   less_float_def' -> Float.compute_float_less
  1075   less_float_def -> less_float.rep_eq
  1076   normfloat_def -> Float.compute_normfloat
  1077   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
  1078   normfloat -> normfloat_def
  1079   normfloat_unique -> use normfloat_def
  1080   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
  1081   one_float_def -> Float.compute_float_one
  1082   plus_float_def -> Float.compute_float_plus
  1083   rapprox_posrat_def -> Float.compute_rapprox_posrat
  1084   rapprox_rat.simps -> Float.compute_rapprox_rat
  1085   real_of_float_0 -> zero_float.rep_eq
  1086   real_of_float_1 -> one_float.rep_eq
  1087   real_of_float_abs -> abs_float.rep_eq
  1088   real_of_float_add -> plus_float.rep_eq
  1089   real_of_float_minus -> uminus_float.rep_eq
  1090   real_of_float_mult -> times_float.rep_eq
  1091   real_of_float_simp -> Float.rep_eq
  1092   real_of_float_sub -> minus_float.rep_eq
  1093   round_down.simps -> Float.compute_float_round_down
  1094   round_up.simps -> Float.compute_float_round_up
  1095   times_float_def -> Float.compute_float_times
  1096   uminus_float_def -> Float.compute_float_uminus
  1097   zero_float_def -> Float.compute_float_zero
  1098 
  1099   Lemmas not necessary anymore, use the transfer method:
  1100   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
  1101   float_divr, float_le_simp, float_less1_mantissa_bound,
  1102   float_less_simp, float_less_zero, float_le_zero,
  1103   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
  1104   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
  1105   lapprox_rat_bottom, normalized_float, rapprox_posrat,
  1106   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
  1107   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
  1108   round_up, zero_le_float, zero_less_float
  1109 
  1110 * New theory HOL/Library/DAList provides an abstract type for
  1111 association lists with distinct keys.
  1112 
  1113 * Session HOL/IMP: Added new theory of abstract interpretation of
  1114 annotated commands.
  1115 
  1116 * Session HOL-Import: Re-implementation from scratch is faster,
  1117 simpler, and more scalable.  Requires a proof bundle, which is
  1118 available as an external component.  Discontinued old (and mostly
  1119 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
  1120 
  1121 * Session HOL-Word: Discontinued many redundant theorems specific to
  1122 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
  1123 instead.
  1124 
  1125   word_sub_alt ~> word_sub_wi
  1126   word_add_alt ~> word_add_def
  1127   word_mult_alt ~> word_mult_def
  1128   word_minus_alt ~> word_minus_def
  1129   word_0_alt ~> word_0_wi
  1130   word_1_alt ~> word_1_wi
  1131   word_add_0 ~> add_0_left
  1132   word_add_0_right ~> add_0_right
  1133   word_mult_1 ~> mult_1_left
  1134   word_mult_1_right ~> mult_1_right
  1135   word_add_commute ~> add_commute
  1136   word_add_assoc ~> add_assoc
  1137   word_add_left_commute ~> add_left_commute
  1138   word_mult_commute ~> mult_commute
  1139   word_mult_assoc ~> mult_assoc
  1140   word_mult_left_commute ~> mult_left_commute
  1141   word_left_distrib ~> left_distrib
  1142   word_right_distrib ~> right_distrib
  1143   word_left_minus ~> left_minus
  1144   word_diff_0_right ~> diff_0_right
  1145   word_diff_self ~> diff_self
  1146   word_sub_def ~> diff_minus
  1147   word_diff_minus ~> diff_minus
  1148   word_add_ac ~> add_ac
  1149   word_mult_ac ~> mult_ac
  1150   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1151   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1152   word_order_trans ~> order_trans
  1153   word_order_refl ~> order_refl
  1154   word_order_antisym ~> order_antisym
  1155   word_order_linear ~> linorder_linear
  1156   lenw1_zero_neq_one ~> zero_neq_one
  1157   word_number_of_eq ~> number_of_eq
  1158   word_of_int_add_hom ~> wi_hom_add
  1159   word_of_int_sub_hom ~> wi_hom_sub
  1160   word_of_int_mult_hom ~> wi_hom_mult
  1161   word_of_int_minus_hom ~> wi_hom_neg
  1162   word_of_int_succ_hom ~> wi_hom_succ
  1163   word_of_int_pred_hom ~> wi_hom_pred
  1164   word_of_int_0_hom ~> word_0_wi
  1165   word_of_int_1_hom ~> word_1_wi
  1166 
  1167 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1168 machine word equalities and inequalities into logical circuits,
  1169 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1170 multiplication, shifting by constants, bitwise operators and numeric
  1171 constants.  Requires fixed-length word types, not 'a word.  Solves
  1172 many standard word identities outright and converts more into first
  1173 order problems amenable to blast or similar.  See also examples in
  1174 HOL/Word/Examples/WordExamples.thy.
  1175 
  1176 * Session HOL-Probability: Introduced the type "'a measure" to
  1177 represent measures, this replaces the records 'a algebra and 'a
  1178 measure_space.  The locales based on subset_class now have two
  1179 locale-parameters the space \<Omega> and the set of measurable sets M.
  1180 The product of probability spaces uses now the same constant as the
  1181 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1182 measure".  Most constants are defined now outside of locales and gain
  1183 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1184 Measure space constructions for distributions and densities now got
  1185 their own constants distr and density.  Instead of using locales to
  1186 describe measure spaces with a finite space, the measure count_space
  1187 and point_measure is introduced.  INCOMPATIBILITY.
  1188 
  1189   Renamed constants:
  1190   measure -> emeasure
  1191   finite_measure.\<mu>' -> measure
  1192   product_algebra_generator -> prod_algebra
  1193   product_prob_space.emb -> prod_emb
  1194   product_prob_space.infprod_algebra -> PiM
  1195 
  1196   Removed locales:
  1197   completeable_measure_space
  1198   finite_measure_space
  1199   finite_prob_space
  1200   finite_product_finite_prob_space
  1201   finite_product_sigma_algebra
  1202   finite_sigma_algebra
  1203   measure_space
  1204   pair_finite_prob_space
  1205   pair_finite_sigma_algebra
  1206   pair_finite_space
  1207   pair_sigma_algebra
  1208   product_sigma_algebra
  1209 
  1210   Removed constants:
  1211   conditional_space
  1212   distribution -> use distr measure, or distributed predicate
  1213   image_space
  1214   joint_distribution -> use distr measure, or distributed predicate
  1215   pair_measure_generator
  1216   product_prob_space.infprod_algebra -> use PiM
  1217   subvimage
  1218 
  1219   Replacement theorems:
  1220   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1221   finite_measure.empty_measure -> measure_empty
  1222   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1223   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1224   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1225   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1226   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1227   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1228   finite_measure.positive_measure' -> measure_nonneg
  1229   finite_measure.real_measure -> finite_measure.emeasure_real
  1230   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1231   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1232   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1233   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1234   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1235   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1236   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1237   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1238   information_space.entropy_commute -> information_space.entropy_commute_simple
  1239   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1240   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1241   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1242   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1243   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1244   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1245   information_space.KL_ge_0 -> information_space.KL_nonneg
  1246   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1247   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1248   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1249   Int_stable_product_algebra_generator -> positive_integral
  1250   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1251   measure_space.additive -> emeasure_additive
  1252   measure_space.AE_iff_null_set -> AE_iff_null
  1253   measure_space.almost_everywhere_def -> eventually_ae_filter
  1254   measure_space.almost_everywhere_vimage -> AE_distrD
  1255   measure_space.continuity_from_above -> INF_emeasure_decseq
  1256   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1257   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1258   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1259   measure_space_density -> emeasure_density
  1260   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1261   measure_space.integrable_vimage -> integrable_distr
  1262   measure_space.integral_translated_density -> integral_density
  1263   measure_space.integral_vimage -> integral_distr
  1264   measure_space.measure_additive -> plus_emeasure
  1265   measure_space.measure_compl -> emeasure_compl
  1266   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1267   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1268   measure_space.measure_decseq -> decseq_emeasure
  1269   measure_space.measure_Diff -> emeasure_Diff
  1270   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1271   measure_space.measure_eq_0 -> emeasure_eq_0
  1272   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1273   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1274   measure_space.measure_incseq -> incseq_emeasure
  1275   measure_space.measure_insert -> emeasure_insert
  1276   measure_space.measure_mono -> emeasure_mono
  1277   measure_space.measure_not_negative -> emeasure_not_MInf
  1278   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1279   measure_space.measure_setsum -> setsum_emeasure
  1280   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1281   measure_space.measure_space_vimage -> emeasure_distr
  1282   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1283   measure_space.measure_subadditive -> subadditive
  1284   measure_space.measure_top -> emeasure_space
  1285   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1286   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1287   measure_space.positive_integral_translated_density -> positive_integral_density
  1288   measure_space.positive_integral_vimage -> positive_integral_distr
  1289   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1290   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1291   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1292   measure_space.real_measure_Diff -> measure_Diff
  1293   measure_space.real_measure_finite_Union -> measure_finite_Union
  1294   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1295   measure_space.real_measure_subadditive -> measure_subadditive
  1296   measure_space.real_measure_Union -> measure_Union
  1297   measure_space.real_measure_UNION -> measure_UNION
  1298   measure_space.simple_function_vimage -> simple_function_comp
  1299   measure_space.simple_integral_vimage -> simple_integral_distr
  1300   measure_space.simple_integral_vimage -> simple_integral_distr
  1301   measure_unique_Int_stable -> measure_eqI_generator_eq
  1302   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1303   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1304   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1305   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1306   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1307   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1308   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1309   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1310   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1311   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1312   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1313   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1314   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1315   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1316   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1317   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1318   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1319   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1320   prob_space.prob_space_vimage -> prob_space_distr
  1321   prob_space.random_variable_restrict -> measurable_restrict
  1322   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1323   product_algebraE -> prod_algebraE_all
  1324   product_algebra_generator_der -> prod_algebra_eq_finite
  1325   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1326   product_algebraI -> sets_PiM_I_finite
  1327   product_measure_exists -> product_sigma_finite.sigma_finite
  1328   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1329   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1330   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1331   product_prob_space.measurable_component -> measurable_component_singleton
  1332   product_prob_space.measurable_emb -> measurable_prod_emb
  1333   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1334   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1335   product_prob_space.measure_emb -> emeasure_prod_emb
  1336   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1337   product_sigma_algebra.product_algebra_into_space -> space_closed
  1338   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1339   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1340   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1341   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1342   sets_product_algebra -> sets_PiM
  1343   sigma_algebra.measurable_sigma -> measurable_measure_of
  1344   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1345   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1346   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1347   space_product_algebra -> space_PiM
  1348 
  1349 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1350 languages) into Isabelle/HOL.
  1351 
  1352 
  1353 *** FOL ***
  1354 
  1355 * New "case_product" attribute (see HOL).
  1356 
  1357 
  1358 *** ZF ***
  1359 
  1360 * Greater support for structured proofs involving induction or case
  1361 analysis.
  1362 
  1363 * Much greater use of mathematical symbols.
  1364 
  1365 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1366 
  1367 
  1368 *** ML ***
  1369 
  1370 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1371 from a minor keyword introduced via theory header declaration.
  1372 
  1373 * Antiquotation @{command_spec "name"} produces the
  1374 Outer_Syntax.command_spec from a major keyword introduced via theory
  1375 header declaration; it can be passed to Outer_Syntax.command etc.
  1376 
  1377 * Local_Theory.define no longer hard-wires default theorem name
  1378 "foo_def", but retains the binding as given.  If that is Binding.empty
  1379 / Attrib.empty_binding, the result is not registered as user-level
  1380 fact.  The Local_Theory.define_internal variant allows to specify a
  1381 non-empty name (used for the foundation in the background theory),
  1382 while omitting the fact binding in the user-context.  Potential
  1383 INCOMPATIBILITY for derived definitional packages: need to specify
  1384 naming policy for primitive definitions more explicitly.
  1385 
  1386 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1387 conformance with similar operations in structure Term and Logic.
  1388 
  1389 * Antiquotation @{attributes [...]} embeds attribute source
  1390 representation into the ML text, which is particularly useful with
  1391 declarations like Local_Theory.note.
  1392 
  1393 * Structure Proof_Context follows standard naming scheme.  Old
  1394 ProofContext has been discontinued.  INCOMPATIBILITY.
  1395 
  1396 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1397 change of semantics: update is applied to auxiliary local theory
  1398 context as well.
  1399 
  1400 * Modernized some old-style infix operations:
  1401 
  1402   addeqcongs    ~> Simplifier.add_eqcong
  1403   deleqcongs    ~> Simplifier.del_eqcong
  1404   addcongs      ~> Simplifier.add_cong
  1405   delcongs      ~> Simplifier.del_cong
  1406   setmksimps    ~> Simplifier.set_mksimps
  1407   setmkcong     ~> Simplifier.set_mkcong
  1408   setmksym      ~> Simplifier.set_mksym
  1409   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1410   settermless   ~> Simplifier.set_termless
  1411   setsubgoaler  ~> Simplifier.set_subgoaler
  1412   addsplits     ~> Splitter.add_split
  1413   delsplits     ~> Splitter.del_split
  1414 
  1415 
  1416 *** System ***
  1417 
  1418 * USER_HOME settings variable points to cross-platform user home
  1419 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1420 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1421 of former $HOME.  A different default for USER_HOME may be set
  1422 explicitly in shell environment, before Isabelle settings are
  1423 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1424 the generic user home was intended.
  1425 
  1426 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1427 notation, which is useful for the jEdit file browser, for example.
  1428 
  1429 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1430 (not just JRE).
  1431 
  1432 
  1433 
  1434 New in Isabelle2011-1 (October 2011)
  1435 ------------------------------------
  1436 
  1437 *** General ***
  1438 
  1439 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1440 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1441 
  1442   - Management of multiple theory files directly from the editor
  1443     buffer store -- bypassing the file-system (no requirement to save
  1444     files for checking).
  1445 
  1446   - Markup of formal entities within the text buffer, with semantic
  1447     highlighting, tooltips and hyperlinks to jump to defining source
  1448     positions.
  1449 
  1450   - Improved text rendering, with sub/superscripts in the source
  1451     buffer (including support for copy/paste wrt. output panel, HTML
  1452     theory output and other non-Isabelle text boxes).
  1453 
  1454   - Refined scheduling of proof checking and printing of results,
  1455     based on interactive editor view.  (Note: jEdit folding and
  1456     narrowing allows to restrict buffer perspectives explicitly.)
  1457 
  1458   - Reduced CPU performance requirements, usable on machines with few
  1459     cores.
  1460 
  1461   - Reduced memory requirements due to pruning of unused document
  1462     versions (garbage collection).
  1463 
  1464 See also ~~/src/Tools/jEdit/README.html for further information,
  1465 including some remaining limitations.
  1466 
  1467 * Theory loader: source files are exclusively located via the master
  1468 directory of each theory node (where the .thy file itself resides).
  1469 The global load path (such as src/HOL/Library) has been discontinued.
  1470 Note that the path element ~~ may be used to reference theories in the
  1471 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1472 INCOMPATIBILITY.
  1473 
  1474 * Theory loader: source files are identified by content via SHA1
  1475 digests.  Discontinued former path/modtime identification and optional
  1476 ISABELLE_FILE_IDENT plugin scripts.
  1477 
  1478 * Parallelization of nested Isar proofs is subject to
  1479 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1480 usedir option -Q.
  1481 
  1482 * Name space: former unsynchronized references are now proper
  1483 configuration options, with more conventional names:
  1484 
  1485   long_names   ~> names_long
  1486   short_names  ~> names_short
  1487   unique_names ~> names_unique
  1488 
  1489 Minor INCOMPATIBILITY, need to declare options in context like this:
  1490 
  1491   declare [[names_unique = false]]
  1492 
  1493 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1494 that the result needs to be unique, which means fact specifications
  1495 may have to be refined after enriching a proof context.
  1496 
  1497 * Attribute "case_names" has been refined: the assumptions in each case
  1498 can be named now by following the case name with [name1 name2 ...].
  1499 
  1500 * Isabelle/Isar reference manual has been updated and extended:
  1501   - "Synopsis" provides a catalog of main Isar language concepts.
  1502   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1503   - Updated material from classic "ref" manual, notably about
  1504     "Classical Reasoner".
  1505 
  1506 
  1507 *** HOL ***
  1508 
  1509 * Class bot and top require underlying partial order rather than
  1510 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1511 
  1512 * Class complete_lattice: generalized a couple of lemmas from sets;
  1513 generalized theorems INF_cong and SUP_cong.  New type classes for
  1514 complete boolean algebras and complete linear orders.  Lemmas
  1515 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1516 class complete_linorder.
  1517 
  1518 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1519 Sup_fun_def, Inf_apply, Sup_apply.
  1520 
  1521 Removed redundant lemmas (the right hand side gives hints how to
  1522 replace them for (metis ...), or (simp only: ...) proofs):
  1523 
  1524   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1525   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1526   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1527   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1528   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1529   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1530   Inter_def ~> INF_def, image_def
  1531   Union_def ~> SUP_def, image_def
  1532   INT_eq ~> INF_def, and image_def
  1533   UN_eq ~> SUP_def, and image_def
  1534   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1535 
  1536 More consistent and comprehensive names:
  1537 
  1538   INTER_eq_Inter_image ~> INF_def
  1539   UNION_eq_Union_image ~> SUP_def
  1540   INFI_def ~> INF_def
  1541   SUPR_def ~> SUP_def
  1542   INF_leI ~> INF_lower
  1543   INF_leI2 ~> INF_lower2
  1544   le_INFI ~> INF_greatest
  1545   le_SUPI ~> SUP_upper
  1546   le_SUPI2 ~> SUP_upper2
  1547   SUP_leI ~> SUP_least
  1548   INFI_bool_eq ~> INF_bool_eq
  1549   SUPR_bool_eq ~> SUP_bool_eq
  1550   INFI_apply ~> INF_apply
  1551   SUPR_apply ~> SUP_apply
  1552   INTER_def ~> INTER_eq
  1553   UNION_def ~> UNION_eq
  1554 
  1555 INCOMPATIBILITY.
  1556 
  1557 * Renamed theory Complete_Lattice to Complete_Lattices.
  1558 INCOMPATIBILITY.
  1559 
  1560 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1561 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1562 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1563 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1564 
  1565 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1566 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1567 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1568 INCOMPATIBILITY.
  1569 
  1570 * Added syntactic classes "inf" and "sup" for the respective
  1571 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1572 (mostly internal) locale predicates for some derived classes.
  1573 
  1574 * Theorem collections ball_simps and bex_simps do not contain theorems
  1575 referring to UNION any longer; these have been moved to collection
  1576 UN_ball_bex_simps.  INCOMPATIBILITY.
  1577 
  1578 * Theory Archimedean_Field: floor now is defined as parameter of a
  1579 separate type class floor_ceiling.
  1580 
  1581 * Theory Finite_Set: more coherent development of fold_set locales:
  1582 
  1583     locale fun_left_comm ~> locale comp_fun_commute
  1584     locale fun_left_comm_idem ~> locale comp_fun_idem
  1585 
  1586 Both use point-free characterization; interpretation proofs may need
  1587 adjustment.  INCOMPATIBILITY.
  1588 
  1589 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1590 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1591 
  1592 * Theory Complex_Main: The locale interpretations for the
  1593 bounded_linear and bounded_bilinear locales have been removed, in
  1594 order to reduce the number of duplicate lemmas. Users must use the
  1595 original names for distributivity theorems, potential INCOMPATIBILITY.
  1596 
  1597   divide.add ~> add_divide_distrib
  1598   divide.diff ~> diff_divide_distrib
  1599   divide.setsum ~> setsum_divide_distrib
  1600   mult.add_right ~> right_distrib
  1601   mult.diff_right ~> right_diff_distrib
  1602   mult_right.setsum ~> setsum_right_distrib
  1603   mult_left.diff ~> left_diff_distrib
  1604 
  1605 * Theory Complex_Main: Several redundant theorems have been removed or
  1606 replaced by more general versions. INCOMPATIBILITY.
  1607 
  1608   real_diff_def ~> minus_real_def
  1609   real_divide_def ~> divide_real_def
  1610   real_less_def ~> less_le
  1611   real_abs_def ~> abs_real_def
  1612   real_sgn_def ~> sgn_real_def
  1613   real_mult_commute ~> mult_commute
  1614   real_mult_assoc ~> mult_assoc
  1615   real_mult_1 ~> mult_1_left
  1616   real_add_mult_distrib ~> left_distrib
  1617   real_zero_not_eq_one ~> zero_neq_one
  1618   real_mult_inverse_left ~> left_inverse
  1619   INVERSE_ZERO ~> inverse_zero
  1620   real_le_refl ~> order_refl
  1621   real_le_antisym ~> order_antisym
  1622   real_le_trans ~> order_trans
  1623   real_le_linear ~> linear
  1624   real_le_eq_diff ~> le_iff_diff_le_0
  1625   real_add_left_mono ~> add_left_mono
  1626   real_mult_order ~> mult_pos_pos
  1627   real_mult_less_mono2 ~> mult_strict_left_mono
  1628   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1629   real_0_le_divide_iff ~> zero_le_divide_iff
  1630   realpow_two_disj ~> power2_eq_iff
  1631   real_squared_diff_one_factored ~> square_diff_one_factored
  1632   realpow_two_diff ~> square_diff_square_factored
  1633   reals_complete2 ~> complete_real
  1634   real_sum_squared_expand ~> power2_sum
  1635   exp_ln_eq ~> ln_unique
  1636   expi_add ~> exp_add
  1637   expi_zero ~> exp_zero
  1638   lemma_DERIV_subst ~> DERIV_cong
  1639   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1640   LIMSEQ_const ~> tendsto_const
  1641   LIMSEQ_norm ~> tendsto_norm
  1642   LIMSEQ_add ~> tendsto_add
  1643   LIMSEQ_minus ~> tendsto_minus
  1644   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1645   LIMSEQ_diff ~> tendsto_diff
  1646   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1647   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1648   LIMSEQ_mult ~> tendsto_mult
  1649   LIMSEQ_inverse ~> tendsto_inverse
  1650   LIMSEQ_divide ~> tendsto_divide
  1651   LIMSEQ_pow ~> tendsto_power
  1652   LIMSEQ_setsum ~> tendsto_setsum
  1653   LIMSEQ_setprod ~> tendsto_setprod
  1654   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1655   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1656   LIMSEQ_imp_rabs ~> tendsto_rabs
  1657   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1658   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1659   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1660   LIMSEQ_Complex ~> tendsto_Complex
  1661   LIM_ident ~> tendsto_ident_at
  1662   LIM_const ~> tendsto_const
  1663   LIM_add ~> tendsto_add
  1664   LIM_add_zero ~> tendsto_add_zero
  1665   LIM_minus ~> tendsto_minus
  1666   LIM_diff ~> tendsto_diff
  1667   LIM_norm ~> tendsto_norm
  1668   LIM_norm_zero ~> tendsto_norm_zero
  1669   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1670   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1671   LIM_rabs ~> tendsto_rabs
  1672   LIM_rabs_zero ~> tendsto_rabs_zero
  1673   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1674   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1675   LIM_compose ~> tendsto_compose
  1676   LIM_mult ~> tendsto_mult
  1677   LIM_scaleR ~> tendsto_scaleR
  1678   LIM_of_real ~> tendsto_of_real
  1679   LIM_power ~> tendsto_power
  1680   LIM_inverse ~> tendsto_inverse
  1681   LIM_sgn ~> tendsto_sgn
  1682   isCont_LIM_compose ~> isCont_tendsto_compose
  1683   bounded_linear.LIM ~> bounded_linear.tendsto
  1684   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1685   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1686   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1687   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1688   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1689   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1690 
  1691 * Theory Complex_Main: The definition of infinite series was
  1692 generalized.  Now it is defined on the type class {topological_space,
  1693 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1694 
  1695 * Theory Complex_Main: The complex exponential function "expi" is now
  1696 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1697 several polymorphic lemmas about "exp" are now applicable to "expi".
  1698 
  1699 * Code generation:
  1700 
  1701   - Theory Library/Code_Char_ord provides native ordering of
  1702     characters in the target language.
  1703 
  1704   - Commands code_module and code_library are legacy, use export_code
  1705     instead.
  1706 
  1707   - Method "evaluation" is legacy, use method "eval" instead.
  1708 
  1709   - Legacy evaluator "SML" is deactivated by default.  May be
  1710     reactivated by the following theory command:
  1711 
  1712       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1713 
  1714 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1715 
  1716 * New proof method "induction" that gives induction hypotheses the
  1717 name "IH", thus distinguishing them from further hypotheses that come
  1718 from rule induction.  The latter are still called "hyps".  Method
  1719 "induction" is a thin wrapper around "induct" and follows the same
  1720 syntax.
  1721 
  1722 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1723 still available as a legacy feature for some time.
  1724 
  1725 * Nitpick:
  1726   - Added "need" and "total_consts" options.
  1727   - Reintroduced "show_skolems" option by popular demand.
  1728   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1729     INCOMPATIBILITY.
  1730 
  1731 * Sledgehammer:
  1732   - Use quasi-sound (and efficient) translations by default.
  1733   - Added support for the following provers: E-ToFoF, LEO-II,
  1734     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1735   - Automatically preplay and minimize proofs before showing them if
  1736     this can be done within reasonable time.
  1737   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1738     INCOMPATIBILITY.
  1739   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1740     "max_mono_iters", and "max_new_mono_instances" options.
  1741   - Removed "explicit_apply" and "full_types" options as well as "Full
  1742     Types" Proof General menu item. INCOMPATIBILITY.
  1743 
  1744 * Metis:
  1745   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1746   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1747     INCOMPATIBILITY.
  1748 
  1749 * Command 'try':
  1750   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1751     "elim:" options. INCOMPATIBILITY.
  1752   - Introduced 'try' that not only runs 'try_methods' but also
  1753     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1754 
  1755 * Quickcheck:
  1756   - Added "eval" option to evaluate terms for the found counterexample
  1757     (currently only supported by the default (exhaustive) tester).
  1758   - Added post-processing of terms to obtain readable counterexamples
  1759     (currently only supported by the default (exhaustive) tester).
  1760   - New counterexample generator quickcheck[narrowing] enables
  1761     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1762     with its installation location defined in the Isabelle settings
  1763     environment as ISABELLE_GHC.
  1764   - Removed quickcheck tester "SML" based on the SML code generator
  1765     (formly in HOL/Library).
  1766 
  1767 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1768 use 'partial_function' instead.
  1769 
  1770 * Theory Library/Extended_Reals replaces now the positive extended
  1771 reals found in probability theory. This file is extended by
  1772 Multivariate_Analysis/Extended_Real_Limits.
  1773 
  1774 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1775 from where it must be imported explicitly if it is really required.
  1776 INCOMPATIBILITY.
  1777 
  1778 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1779 been moved here.  INCOMPATIBILITY.
  1780 
  1781 * Theory Library/Saturated provides type of numbers with saturated
  1782 arithmetic.
  1783 
  1784 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1785 product type 'a * 'b, and provides instance proofs for various order
  1786 and lattice type classes.
  1787 
  1788 * Theory Library/Countable now provides the "countable_datatype" proof
  1789 method for proving "countable" class instances for datatypes.
  1790 
  1791 * Theory Library/Cset_Monad allows do notation for computable sets
  1792 (cset) via the generic monad ad-hoc overloading facility.
  1793 
  1794 * Library: Theories of common data structures are split into theories
  1795 for implementation, an invariant-ensuring type, and connection to an
  1796 abstract type. INCOMPATIBILITY.
  1797 
  1798   - RBT is split into RBT and RBT_Mapping.
  1799   - AssocList is split and renamed into AList and AList_Mapping.
  1800   - DList is split into DList_Impl, DList, and DList_Cset.
  1801   - Cset is split into Cset and List_Cset.
  1802 
  1803 * Theory Library/Nat_Infinity has been renamed to
  1804 Library/Extended_Nat, with name changes of the following types and
  1805 constants:
  1806 
  1807   type inat   ~> type enat
  1808   Fin         ~> enat
  1809   Infty       ~> infinity (overloaded)
  1810   iSuc        ~> eSuc
  1811   the_Fin     ~> the_enat
  1812 
  1813 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1814 been renamed accordingly. INCOMPATIBILITY.
  1815 
  1816 * Session Multivariate_Analysis: The euclidean_space type class now
  1817 fixes a constant "Basis :: 'a set" consisting of the standard
  1818 orthonormal basis for the type. Users now have the option of
  1819 quantifying over this set instead of using the "basis" function, e.g.
  1820 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1821 
  1822 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1823 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1824 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1825 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1826 changed to match. Definition theorems for overloaded constants now use
  1827 the standard "foo_vec_def" naming scheme. A few other theorems have
  1828 been renamed as follows (INCOMPATIBILITY):
  1829 
  1830   Cart_eq          ~> vec_eq_iff
  1831   dist_nth_le_cart ~> dist_vec_nth_le
  1832   tendsto_vector   ~> vec_tendstoI
  1833   Cauchy_vector    ~> vec_CauchyI
  1834 
  1835 * Session Multivariate_Analysis: Several duplicate theorems have been
  1836 removed, and other theorems have been renamed or replaced with more
  1837 general versions. INCOMPATIBILITY.
  1838 
  1839   finite_choice ~> finite_set_choice
  1840   eventually_conjI ~> eventually_conj
  1841   eventually_and ~> eventually_conj_iff
  1842   eventually_false ~> eventually_False
  1843   setsum_norm ~> norm_setsum
  1844   Lim_sequentially ~> LIMSEQ_def
  1845   Lim_ident_at ~> LIM_ident
  1846   Lim_const ~> tendsto_const
  1847   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1848   Lim_neg ~> tendsto_minus
  1849   Lim_add ~> tendsto_add
  1850   Lim_sub ~> tendsto_diff
  1851   Lim_mul ~> tendsto_scaleR
  1852   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1853   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1854   Lim_linear ~> bounded_linear.tendsto
  1855   Lim_component ~> tendsto_euclidean_component
  1856   Lim_component_cart ~> tendsto_vec_nth
  1857   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1858   dot_lsum ~> inner_setsum_left
  1859   dot_rsum ~> inner_setsum_right
  1860   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1861   continuous_neg ~> continuous_minus
  1862   continuous_sub ~> continuous_diff
  1863   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1864   continuous_mul ~> continuous_scaleR
  1865   continuous_inv ~> continuous_inverse
  1866   continuous_at_within_inv ~> continuous_at_within_inverse
  1867   continuous_at_inv ~> continuous_at_inverse
  1868   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1869   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1870   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1871   continuous_on_neg ~> continuous_on_minus
  1872   continuous_on_sub ~> continuous_on_diff
  1873   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1874   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1875   continuous_on_mul ~> continuous_on_scaleR
  1876   continuous_on_mul_real ~> continuous_on_mult
  1877   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1878   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1879   continuous_on_inverse ~> continuous_on_inv
  1880   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1881   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1882   subset_interior ~> interior_mono
  1883   subset_closure ~> closure_mono
  1884   closure_univ ~> closure_UNIV
  1885   real_arch_lt ~> reals_Archimedean2
  1886   real_arch ~> reals_Archimedean3
  1887   real_abs_norm ~> abs_norm_cancel
  1888   real_abs_sub_norm ~> norm_triangle_ineq3
  1889   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1890 
  1891 * Session HOL-Probability:
  1892   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1893   - Infinite products of probability measures are now available.
  1894   - Sigma closure is independent, if the generator is independent
  1895   - Use extended reals instead of positive extended
  1896     reals. INCOMPATIBILITY.
  1897 
  1898 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1899 
  1900   expand_fun_below ~> fun_below_iff
  1901   below_fun_ext ~> fun_belowI
  1902   expand_cfun_eq ~> cfun_eq_iff
  1903   ext_cfun ~> cfun_eqI
  1904   expand_cfun_below ~> cfun_below_iff
  1905   below_cfun_ext ~> cfun_belowI
  1906   monofun_fun_fun ~> fun_belowD
  1907   monofun_fun_arg ~> monofunE
  1908   monofun_lub_fun ~> adm_monofun [THEN admD]
  1909   cont_lub_fun ~> adm_cont [THEN admD]
  1910   cont2cont_Rep_CFun ~> cont2cont_APP
  1911   cont_Rep_CFun_app ~> cont_APP_app
  1912   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1913   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1914   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1915   contlub_cfun ~> lub_APP [symmetric]
  1916   contlub_LAM ~> lub_LAM [symmetric]
  1917   thelubI ~> lub_eqI
  1918   UU_I ~> bottomI
  1919   lift_distinct1 ~> lift.distinct(1)
  1920   lift_distinct2 ~> lift.distinct(2)
  1921   Def_not_UU ~> lift.distinct(2)
  1922   Def_inject ~> lift.inject
  1923   below_UU_iff ~> below_bottom_iff
  1924   eq_UU_iff ~> eq_bottom_iff
  1925 
  1926 
  1927 *** Document preparation ***
  1928 
  1929 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1930 isar-ref manual, both for description and actual application of the
  1931 same.
  1932 
  1933 * Antiquotation @{value} evaluates the given term and presents its
  1934 result.
  1935 
  1936 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1937 variables x1, y23 into subscripted form x\<^isub>1,
  1938 y\<^isub>2\<^isub>3.
  1939 
  1940 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1941 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1942 
  1943 * Localized \isabellestyle switch can be used within blocks or groups
  1944 like this:
  1945 
  1946   \isabellestyle{it}  %preferred default
  1947   {\isabellestylett @{text "typewriter stuff"}}
  1948 
  1949 * Discontinued special treatment of hard tabulators.  Implicit
  1950 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1951 layouts.
  1952 
  1953 
  1954 *** ML ***
  1955 
  1956 * The inner syntax of sort/type/term/prop supports inlined YXML
  1957 representations within quoted string tokens.  By encoding logical
  1958 entities via Term_XML (in ML or Scala) concrete syntax can be
  1959 bypassed, which is particularly useful for producing bits of text
  1960 under external program control.
  1961 
  1962 * Antiquotations for ML and document preparation are managed as theory
  1963 data, which requires explicit setup.
  1964 
  1965 * Isabelle_Process.is_active allows tools to check if the official
  1966 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1967 (better known as Proof General).
  1968 
  1969 * Structure Proof_Context follows standard naming scheme.  Old
  1970 ProofContext is still available for some time as legacy alias.
  1971 
  1972 * Structure Timing provides various operations for timing; supersedes
  1973 former start_timing/end_timing etc.
  1974 
  1975 * Path.print is the official way to show file-system paths to users
  1976 (including quotes etc.).
  1977 
  1978 * Inner syntax: identifiers in parse trees of generic categories
  1979 "logic", "aprop", "idt" etc. carry position information (disguised as
  1980 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1981 translations that choke on unexpected type constraints.  Positions can
  1982 be stripped in ML translations via Syntax.strip_positions /
  1983 Syntax.strip_positions_ast, or via the syntax constant
  1984 "_strip_positions" within parse trees.  As last resort, positions can
  1985 be disabled via the configuration option Syntax.positions, which is
  1986 called "syntax_positions" in Isar attribute syntax.
  1987 
  1988 * Discontinued special status of various ML structures that contribute
  1989 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1990 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1991 refer directly to Ast.Constant, Lexicon.is_identifier,
  1992 Syntax_Trans.mk_binder_tr etc.
  1993 
  1994 * Typed print translation: discontinued show_sorts argument, which is
  1995 already available via context of "advanced" translation.
  1996 
  1997 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1998 goal states; body tactic needs to address all subgoals uniformly.
  1999 
  2000 * Slightly more special eq_list/eq_set, with shortcut involving
  2001 pointer equality (assumes that eq relation is reflexive).
  2002 
  2003 * Classical tactics use proper Proof.context instead of historic types
  2004 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  2005 operate directly on Proof.context.  Raw type claset retains its use as
  2006 snapshot of the classical context, which can be recovered via
  2007 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  2008 INCOMPATIBILITY, classical tactics and derived proof methods require
  2009 proper Proof.context.
  2010 
  2011 
  2012 *** System ***
  2013 
  2014 * Discontinued support for Poly/ML 5.2, which was the last version
  2015 without proper multithreading and TimeLimit implementation.
  2016 
  2017 * Discontinued old lib/scripts/polyml-platform, which has been
  2018 obsolete since Isabelle2009-2.
  2019 
  2020 * Various optional external tools are referenced more robustly and
  2021 uniformly by explicit Isabelle settings as follows:
  2022 
  2023   ISABELLE_CSDP   (formerly CSDP_EXE)
  2024   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  2025   ISABELLE_OCAML  (formerly EXEC_OCAML)
  2026   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  2027   ISABELLE_YAP    (formerly EXEC_YAP)
  2028 
  2029 Note that automated detection from the file-system or search path has
  2030 been discontinued.  INCOMPATIBILITY.
  2031 
  2032 * Scala layer provides JVM method invocation service for static
  2033 methods of type (String)String, see Invoke_Scala.method in ML.  For
  2034 example:
  2035 
  2036   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  2037 
  2038 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  2039 this allows to pass structured values between ML and Scala.
  2040 
  2041 * The IsabelleText fonts includes some further glyphs to support the
  2042 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  2043 installed a local copy (which is normally *not* required) need to
  2044 delete or update it from ~~/lib/fonts/.
  2045 
  2046 
  2047 
  2048 New in Isabelle2011 (January 2011)
  2049 ----------------------------------
  2050 
  2051 *** General ***
  2052 
  2053 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  2054 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  2055 useful tooltips and hyperlinks produced from its static analysis.  The
  2056 bundled component provides an executable Isabelle tool that can be run
  2057 like this:
  2058 
  2059   Isabelle2011/bin/isabelle jedit
  2060 
  2061 * Significantly improved Isabelle/Isar implementation manual.
  2062 
  2063 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  2064 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  2065 while the default heap location within that directory lacks that extra
  2066 suffix.  This isolates multiple Isabelle installations from each
  2067 other, avoiding problems with old settings in new versions.
  2068 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  2069 
  2070 * Source files are always encoded as UTF-8, instead of old-fashioned
  2071 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  2072 the following package declarations:
  2073 
  2074   \usepackage[utf8]{inputenc}
  2075   \usepackage{textcomp}
  2076 
  2077 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  2078 a Unicode character is treated as a single symbol, not a sequence of
  2079 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  2080 contain symbols without further backslash escapes, Unicode can now be
  2081 used here as well.  Recall that Symbol.explode in ML provides a
  2082 consistent view on symbols, while raw explode (or String.explode)
  2083 merely give a byte-oriented representation.
  2084 
  2085 * Theory loader: source files are primarily located via the master
  2086 directory of each theory node (where the .thy file itself resides).
  2087 The global load path is still partially available as legacy feature.
  2088 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  2089 explicit paths, relatively to the theory.
  2090 
  2091 * Special treatment of ML file names has been discontinued.
  2092 Historically, optional extensions .ML or .sml were added on demand --
  2093 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  2094 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  2095 
  2096 * Various options that affect pretty printing etc. are now properly
  2097 handled within the context via configuration options, instead of
  2098 unsynchronized references or print modes.  There are both ML Config.T
  2099 entities and Isar declaration attributes to access these.
  2100 
  2101   ML (Config.T)                 Isar (attribute)
  2102 
  2103   eta_contract                  eta_contract
  2104   show_brackets                 show_brackets
  2105   show_sorts                    show_sorts
  2106   show_types                    show_types
  2107   show_question_marks           show_question_marks
  2108   show_consts                   show_consts
  2109   show_abbrevs                  show_abbrevs
  2110 
  2111   Syntax.ast_trace              syntax_ast_trace
  2112   Syntax.ast_stat               syntax_ast_stat
  2113   Syntax.ambiguity_level        syntax_ambiguity_level
  2114 
  2115   Goal_Display.goals_limit      goals_limit
  2116   Goal_Display.show_main_goal   show_main_goal
  2117 
  2118   Method.rule_trace             rule_trace
  2119 
  2120   Thy_Output.display            thy_output_display
  2121   Thy_Output.quotes             thy_output_quotes
  2122   Thy_Output.indent             thy_output_indent
  2123   Thy_Output.source             thy_output_source
  2124   Thy_Output.break              thy_output_break
  2125 
  2126 Note that corresponding "..._default" references in ML may only be
  2127 changed globally at the ROOT session setup, but *not* within a theory.
  2128 The option "show_abbrevs" supersedes the former print mode
  2129 "no_abbrevs" with inverted meaning.
  2130 
  2131 * More systematic naming of some configuration options.
  2132 INCOMPATIBILITY.
  2133 
  2134   trace_simp  ~>  simp_trace
  2135   debug_simp  ~>  simp_debug
  2136 
  2137 * Support for real valued configuration options, using simplistic
  2138 floating-point notation that coincides with the inner syntax for
  2139 float_token.
  2140 
  2141 * Support for real valued preferences (with approximative PGIP type):
  2142 front-ends need to accept "pgint" values in float notation.
  2143 INCOMPATIBILITY.
  2144 
  2145 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2146 DejaVu Sans.
  2147 
  2148 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2149 
  2150 
  2151 *** Pure ***
  2152 
  2153 * Command 'type_synonym' (with single argument) replaces somewhat
  2154 outdated 'types', which is still available as legacy feature for some
  2155 time.
  2156 
  2157 * Command 'nonterminal' (with 'and' separated list of arguments)
  2158 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2159 
  2160 * Command 'notepad' replaces former 'example_proof' for
  2161 experimentation in Isar without any result.  INCOMPATIBILITY.
  2162 
  2163 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2164 lists of equations to map definitions in a locale to appropriate
  2165 entities in the context of the interpretation.  The 'interpretation'
  2166 command already provided this functionality.
  2167 
  2168 * Diagnostic command 'print_dependencies' prints the locale instances
  2169 that would be activated if the specified expression was interpreted in
  2170 the current context.  Variant "print_dependencies!" assumes a context
  2171 without interpretations.
  2172 
  2173 * Diagnostic command 'print_interps' prints interpretations in proofs
  2174 in addition to interpretations in theories.
  2175 
  2176 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2177 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2178 Sign.root_path and Sign.local_path may be applied directly where this
  2179 feature is still required for historical reasons.
  2180 
  2181 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2182 'definition' instead.
  2183 
  2184 * The "prems" fact, which refers to the accidental collection of
  2185 foundational premises in the context, is now explicitly marked as
  2186 legacy feature and will be discontinued soon.  Consider using "assms"
  2187 of the head statement or reference facts by explicit names.
  2188 
  2189 * Document antiquotations @{class} and @{type} print classes and type
  2190 constructors.
  2191 
  2192 * Document antiquotation @{file} checks file/directory entries within
  2193 the local file system.
  2194 
  2195 
  2196 *** HOL ***
  2197 
  2198 * Coercive subtyping: functions can be declared as coercions and type
  2199 inference will add them as necessary upon input of a term.  Theory
  2200 Complex_Main declares real :: nat => real and real :: int => real as
  2201 coercions. A coercion function f is declared like this:
  2202 
  2203   declare [[coercion f]]
  2204 
  2205 To lift coercions through type constructors (e.g. from nat => real to
  2206 nat list => real list), map functions can be declared, e.g.
  2207 
  2208   declare [[coercion_map map]]
  2209 
  2210 Currently coercion inference is activated only in theories including
  2211 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2212 the configuration option "coercion_enabled", e.g. it can be enabled in
  2213 other theories like this:
  2214 
  2215   declare [[coercion_enabled]]
  2216 
  2217 * Command 'partial_function' provides basic support for recursive
  2218 function definitions over complete partial orders.  Concrete instances
  2219 are provided for i) the option type, ii) tail recursion on arbitrary
  2220 types, and iii) the heap monad of Imperative_HOL.  See
  2221 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2222 for examples.
  2223 
  2224 * Function package: f.psimps rules are no longer implicitly declared
  2225 as [simp].  INCOMPATIBILITY.
  2226 
  2227 * Datatype package: theorems generated for executable equality (class
  2228 "eq") carry proper names and are treated as default code equations.
  2229 
  2230 * Inductive package: now offers command 'inductive_simps' to
  2231 automatically derive instantiated and simplified equations for
  2232 inductive predicates, similar to 'inductive_cases'.
  2233 
  2234 * Command 'enriched_type' allows to register properties of the
  2235 functorial structure of types.
  2236 
  2237 * Improved infrastructure for term evaluation using code generator
  2238 techniques, in particular static evaluation conversions.
  2239 
  2240 * Code generator: Scala (2.8 or higher) has been added to the target
  2241 languages.
  2242 
  2243 * Code generator: globbing constant expressions "*" and "Theory.*"
  2244 have been replaced by the more idiomatic "_" and "Theory._".
  2245 INCOMPATIBILITY.
  2246 
  2247 * Code generator: export_code without explicit file declaration prints
  2248 to standard output.  INCOMPATIBILITY.
  2249 
  2250 * Code generator: do not print function definitions for case
  2251 combinators any longer.
  2252 
  2253 * Code generator: simplification with rules determined with
  2254 src/Tools/Code/code_simp.ML and method "code_simp".
  2255 
  2256 * Code generator for records: more idiomatic representation of record
  2257 types.  Warning: records are not covered by ancient SML code
  2258 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2259 rep_datatype declaration helps to succeed then:
  2260 
  2261   record 'a foo = ...
  2262   ...
  2263   rep_datatype foo_ext ...
  2264 
  2265 * Records: logical foundation type for records does not carry a
  2266 '_type' suffix any longer (obsolete due to authentic syntax).
  2267 INCOMPATIBILITY.
  2268 
  2269 * Quickcheck now by default uses exhaustive testing instead of random
  2270 testing.  Random testing can be invoked by "quickcheck [random]",
  2271 exhaustive testing by "quickcheck [exhaustive]".
  2272 
  2273 * Quickcheck instantiates polymorphic types with small finite
  2274 datatypes by default. This enables a simple execution mechanism to
  2275 handle quantifiers and function equality over the finite datatypes.
  2276 
  2277 * Quickcheck random generator has been renamed from "code" to
  2278 "random".  INCOMPATIBILITY.
  2279 
  2280 * Quickcheck now has a configurable time limit which is set to 30
  2281 seconds by default. This can be changed by adding [timeout = n] to the
  2282 quickcheck command. The time limit for Auto Quickcheck is still set
  2283 independently.
  2284 
  2285 * Quickcheck in locales considers interpretations of that locale for
  2286 counter example search.
  2287 
  2288 * Sledgehammer:
  2289   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2290     method. See the Sledgehammer manual for details ("isabelle doc
  2291     sledgehammer").
  2292   - Renamed commands:
  2293     sledgehammer atp_info ~> sledgehammer running_provers
  2294     sledgehammer atp_kill ~> sledgehammer kill_provers
  2295     sledgehammer available_atps ~> sledgehammer available_provers
  2296     INCOMPATIBILITY.
  2297   - Renamed options:
  2298     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2299     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2300     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2301     (and "ms" and "min" are no longer supported)
  2302     INCOMPATIBILITY.
  2303 
  2304 * Nitpick:
  2305   - Renamed options:
  2306     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2307     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2308     INCOMPATIBILITY.
  2309   - Added support for partial quotient types.
  2310   - Added local versions of the "Nitpick.register_xxx" functions.
  2311   - Added "whack" option.
  2312   - Allow registration of quotient types as codatatypes.
  2313   - Improved "merge_type_vars" option to merge more types.
  2314   - Removed unsound "fast_descrs" option.
  2315   - Added custom symmetry breaking for datatypes, making it possible to reach
  2316     higher cardinalities.
  2317   - Prevent the expansion of too large definitions.
  2318 
  2319 * Proof methods "metis" and "meson" now have configuration options
  2320 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2321 to diagnose these tools. E.g.
  2322 
  2323     using [[metis_trace = true]]
  2324 
  2325 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2326 manually as command 'solve_direct'.
  2327 
  2328 * The default SMT solver Z3 must be enabled explicitly (due to
  2329 licensing issues) by setting the environment variable
  2330 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2331 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2332 changing the SMT solver is done via the configuration option
  2333 "smt_solver".
  2334 
  2335 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2336 i.e. "remote_cvc3" and "remote_z3".
  2337 
  2338 * Added basic SMT support for datatypes, records, and typedefs using
  2339 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2340 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2341 support for a similar behavior).  Minor INCOMPATIBILITY.
  2342 
  2343 * Changed SMT configuration options:
  2344   - Renamed:
  2345     z3_proofs ~> smt_oracle (with inverted meaning)
  2346     z3_trace_assms ~> smt_trace_used_facts
  2347     INCOMPATIBILITY.
  2348   - Added:
  2349     smt_verbose
  2350     smt_random_seed
  2351     smt_datatypes
  2352     smt_infer_triggers
  2353     smt_monomorph_limit
  2354     cvc3_options
  2355     remote_cvc3_options
  2356     remote_z3_options
  2357     yices_options
  2358 
  2359 * Boogie output files (.b2i files) need to be declared in the theory
  2360 header.
  2361 
  2362 * Simplification procedure "list_to_set_comprehension" rewrites list
  2363 comprehensions applied to List.set to set comprehensions.  Occasional
  2364 INCOMPATIBILITY, may be deactivated like this:
  2365 
  2366   declare [[simproc del: list_to_set_comprehension]]
  2367 
  2368 * Removed old version of primrec package.  INCOMPATIBILITY.
  2369 
  2370 * Removed simplifier congruence rule of "prod_case", as has for long
  2371 been the case with "split".  INCOMPATIBILITY.
  2372 
  2373 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2374 
  2375 * Removed [split_format ... and ... and ...] version of
  2376 [split_format].  Potential INCOMPATIBILITY.
  2377 
  2378 * Predicate "sorted" now defined inductively, with nice induction
  2379 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2380 
  2381 * Constant "contents" renamed to "the_elem", to free the generic name
  2382 contents for other uses.  INCOMPATIBILITY.
  2383 
  2384 * Renamed class eq and constant eq (for code generation) to class
  2385 equal and constant equal, plus renaming of related facts and various
  2386 tuning.  INCOMPATIBILITY.
  2387 
  2388 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2389 
  2390 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2391 INCOMPATIBILITY.
  2392 
  2393 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2394 avoid confusion with finite sets.  INCOMPATIBILITY.
  2395 
  2396 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2397 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2398 for congruent(2)).
  2399 
  2400 * Some previously unqualified names have been qualified:
  2401 
  2402   types
  2403     bool ~> HOL.bool
  2404     nat ~> Nat.nat
  2405 
  2406   constants
  2407     Trueprop ~> HOL.Trueprop
  2408     True ~> HOL.True
  2409     False ~> HOL.False
  2410     op & ~> HOL.conj
  2411     op | ~> HOL.disj
  2412     op --> ~> HOL.implies
  2413     op = ~> HOL.eq
  2414     Not ~> HOL.Not
  2415     The ~> HOL.The
  2416     All ~> HOL.All
  2417     Ex ~> HOL.Ex
  2418     Ex1 ~> HOL.Ex1
  2419     Let ~> HOL.Let
  2420     If ~> HOL.If
  2421     Ball ~> Set.Ball
  2422     Bex ~> Set.Bex
  2423     Suc ~> Nat.Suc
  2424     Pair ~> Product_Type.Pair
  2425     fst ~> Product_Type.fst
  2426     snd ~> Product_Type.snd
  2427     curry ~> Product_Type.curry
  2428     op : ~> Set.member
  2429     Collect ~> Set.Collect
  2430 
  2431 INCOMPATIBILITY.
  2432 
  2433 * More canonical naming convention for some fundamental definitions:
  2434 
  2435     bot_bool_eq ~> bot_bool_def
  2436     top_bool_eq ~> top_bool_def
  2437     inf_bool_eq ~> inf_bool_def
  2438     sup_bool_eq ~> sup_bool_def
  2439     bot_fun_eq  ~> bot_fun_def
  2440     top_fun_eq  ~> top_fun_def
  2441     inf_fun_eq  ~> inf_fun_def
  2442     sup_fun_eq  ~> sup_fun_def
  2443 
  2444 INCOMPATIBILITY.
  2445 
  2446 * More stylized fact names:
  2447 
  2448   expand_fun_eq ~> fun_eq_iff
  2449   expand_set_eq ~> set_eq_iff
  2450   set_ext       ~> set_eqI
  2451   nat_number    ~> eval_nat_numeral
  2452 
  2453 INCOMPATIBILITY.
  2454 
  2455 * Refactoring of code-generation specific operations in theory List:
  2456 
  2457   constants
  2458     null ~> List.null
  2459 
  2460   facts
  2461     mem_iff ~> member_def
  2462     null_empty ~> null_def
  2463 
  2464 INCOMPATIBILITY.  Note that these were not supposed to be used
  2465 regularly unless for striking reasons; their main purpose was code
  2466 generation.
  2467 
  2468 Various operations from the Haskell prelude are used for generating
  2469 Haskell code.
  2470 
  2471 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2472 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2473 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2474 
  2475 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2476 "*" and "+" respectively.  INCOMPATIBILITY.
  2477 
  2478 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2479 "Sum_Type.Plus" instead.
  2480 
  2481 * Constant "split" has been merged with constant "prod_case"; names of
  2482 ML functions, facts etc. involving split have been retained so far,
  2483 though.  INCOMPATIBILITY.
  2484 
  2485 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2486 instead.  INCOMPATIBILITY.
  2487 
  2488 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2489 INCOMPATIBILITY.
  2490 
  2491 * Former theory Library/Enum is now part of the HOL-Main image.
  2492 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2493 referred to by its qualified name.
  2494 
  2495   enum    ~>  Enum.enum
  2496   nlists  ~>  Enum.nlists
  2497   product ~>  Enum.product
  2498 
  2499 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2500 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2501 INCOMPATIBILITY.
  2502 
  2503 * Theory Library/SetsAndFunctions has been split into
  2504 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2505 for instance definitions for functions; various improvements.
  2506 INCOMPATIBILITY.
  2507 
  2508 * Theory Library/Multiset provides stable quicksort implementation of
  2509 sort_key.
  2510 
  2511 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2512 INCOMPATIBILITY.
  2513 
  2514 * Session Multivariate_Analysis: introduced a type class for euclidean
  2515 space.  Most theorems are now stated in terms of euclidean spaces
  2516 instead of finite cartesian products.
  2517 
  2518   types
  2519     real ^ 'n ~>  'a::real_vector
  2520               ~>  'a::euclidean_space
  2521               ~>  'a::ordered_euclidean_space
  2522         (depends on your needs)
  2523 
  2524   constants
  2525      _ $ _        ~> _ $$ _
  2526      \<chi> x. _  ~> \<chi>\<chi> x. _
  2527      CARD('n)     ~> DIM('a)
  2528 
  2529 Also note that the indices are now natural numbers and not from some
  2530 finite type. Finite cartesian products of euclidean spaces, products
  2531 of euclidean spaces the real and complex numbers are instantiated to
  2532 be euclidean_spaces.  INCOMPATIBILITY.
  2533 
  2534 * Session Probability: introduced pextreal as positive extended real
  2535 numbers.  Use pextreal as value for measures.  Introduce the
  2536 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2537 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2538 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2539 
  2540 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2541 INCOMPATIBILITY.
  2542 
  2543 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2544 prove verification conditions generated by the SPARK Ada program
  2545 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2546 
  2547 
  2548 *** HOL-Algebra ***
  2549 
  2550 * Theorems for additive ring operations (locale abelian_monoid and
  2551 descendants) are generated by interpretation from their multiplicative
  2552 counterparts.  Names (in particular theorem names) have the mandatory
  2553 qualifier 'add'.  Previous theorem names are redeclared for
  2554 compatibility.
  2555 
  2556 * Structure "int_ring" is now an abbreviation (previously a
  2557 definition).  This fits more natural with advanced interpretations.
  2558 
  2559 
  2560 *** HOLCF ***
  2561 
  2562 * The domain package now runs in definitional mode by default: The
  2563 former command 'new_domain' is now called 'domain'.  To use the domain
  2564 package in its original axiomatic mode, use 'domain (unsafe)'.
  2565 INCOMPATIBILITY.
  2566 
  2567 * The new class "domain" is now the default sort.  Class "predomain"
  2568 is an unpointed version of "domain". Theories can be updated by
  2569 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2570 
  2571   'a::type ~> 'a::countable
  2572   'a::cpo  ~> 'a::predomain
  2573   'a::pcpo ~> 'a::domain
  2574 
  2575 * The old type class "rep" has been superseded by class "domain".
  2576 Accordingly, users of the definitional package must remove any
  2577 "default_sort rep" declarations.  INCOMPATIBILITY.
  2578 
  2579 * The domain package (definitional mode) now supports unpointed
  2580 predomain argument types, as long as they are marked 'lazy'. (Strict
  2581 arguments must be in class "domain".) For example, the following
  2582 domain definition now works:
  2583 
  2584   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2585 
  2586 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2587 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2588 'a option, and 'a list.  Additionally, it configures fixrec and the
  2589 domain package to work with these types.  For example:
  2590 
  2591   fixrec isInl :: "('a + 'b) u -> tr"
  2592     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2593 
  2594   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2595 
  2596 * The "(permissive)" option of fixrec has been replaced with a
  2597 per-equation "(unchecked)" option. See
  2598 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2599 
  2600 * The "bifinite" class no longer fixes a constant "approx"; the class
  2601 now just asserts that such a function exists.  INCOMPATIBILITY.
  2602 
  2603 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2604 defines an embedding of type 'a defl into udom by default; instances
  2605 of "bifinite" and "domain" classes are available in
  2606 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2607 
  2608 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2609 
  2610 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2611 
  2612 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2613 
  2614 * The function "cprod_map" has been renamed to "prod_map".
  2615 INCOMPATIBILITY.
  2616 
  2617 * The monadic bind operator on each powerdomain has new binder syntax
  2618 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2619 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2620 
  2621 * The infix syntax for binary union on each powerdomain has changed
  2622 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2623 syntax.  INCOMPATIBILITY.
  2624 
  2625 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2626 still supported as an input translation.
  2627 
  2628 * Renamed some theorems (the original names are also still available).
  2629 
  2630   expand_fun_below   ~> fun_below_iff
  2631   below_fun_ext      ~> fun_belowI
  2632   expand_cfun_eq     ~> cfun_eq_iff
  2633   ext_cfun           ~> cfun_eqI
  2634   expand_cfun_below  ~> cfun_below_iff
  2635   below_cfun_ext     ~> cfun_belowI
  2636   cont2cont_Rep_CFun ~> cont2cont_APP
  2637 
  2638 * The Abs and Rep functions for various types have changed names.
  2639 Related theorem names have also changed to match. INCOMPATIBILITY.
  2640 
  2641   Rep_CFun  ~> Rep_cfun
  2642   Abs_CFun  ~> Abs_cfun
  2643   Rep_Sprod ~> Rep_sprod
  2644   Abs_Sprod ~> Abs_sprod
  2645   Rep_Ssum  ~> Rep_ssum
  2646   Abs_Ssum  ~> Abs_ssum
  2647 
  2648 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2649 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2650 
  2651 * Various changes to bisimulation/coinduction with domain package:
  2652 
  2653   - Definitions of "bisim" constants no longer mention definedness.
  2654   - With mutual recursion, "bisim" predicate is now curried.
  2655   - With mutual recursion, each type gets a separate coind theorem.
  2656   - Variable names in bisim_def and coinduct rules have changed.
  2657 
  2658 INCOMPATIBILITY.
  2659 
  2660 * Case combinators generated by the domain package for type "foo" are
  2661 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2662 
  2663 * Several theorems have been renamed to more accurately reflect the
  2664 names of constants and types involved.  INCOMPATIBILITY.
  2665 
  2666   thelub_const    ~> lub_const
  2667   lub_const       ~> is_lub_const
  2668   thelubI         ~> lub_eqI
  2669   is_lub_lub      ~> is_lubD2
  2670   lubI            ~> is_lub_lub
  2671   unique_lub      ~> is_lub_unique
  2672   is_ub_lub       ~> is_lub_rangeD1
  2673   lub_bin_chain   ~> is_lub_bin_chain
  2674   lub_fun         ~> is_lub_fun
  2675   thelub_fun      ~> lub_fun
  2676   thelub_cfun     ~> lub_cfun
  2677   thelub_Pair     ~> lub_Pair
  2678   lub_cprod       ~> is_lub_prod
  2679   thelub_cprod    ~> lub_prod
  2680   minimal_cprod   ~> minimal_prod
  2681   inst_cprod_pcpo ~> inst_prod_pcpo
  2682   UU_I            ~> bottomI
  2683   compact_UU      ~> compact_bottom
  2684   deflation_UU    ~> deflation_bottom
  2685   finite_deflation_UU ~> finite_deflation_bottom
  2686 
  2687 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2688 
  2689   sq_ord_less_eq_trans ~> below_eq_trans
  2690   sq_ord_eq_less_trans ~> eq_below_trans
  2691   refl_less            ~> below_refl
  2692   trans_less           ~> below_trans
  2693   antisym_less         ~> below_antisym
  2694   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2695   box_less             ~> box_below
  2696   rev_trans_less       ~> rev_below_trans
  2697   not_less2not_eq      ~> not_below2not_eq
  2698   less_UU_iff          ~> below_UU_iff
  2699   flat_less_iff        ~> flat_below_iff
  2700   adm_less             ~> adm_below
  2701   adm_not_less         ~> adm_not_below
  2702   adm_compact_not_less ~> adm_compact_not_below
  2703   less_fun_def         ~> below_fun_def
  2704   expand_fun_less      ~> fun_below_iff
  2705   less_fun_ext         ~> fun_belowI
  2706   less_discr_def       ~> below_discr_def
  2707   discr_less_eq        ~> discr_below_eq
  2708   less_unit_def        ~> below_unit_def
  2709   less_cprod_def       ~> below_prod_def
  2710   prod_lessI           ~> prod_belowI
  2711   Pair_less_iff        ~> Pair_below_iff
  2712   fst_less_iff         ~> fst_below_iff
  2713   snd_less_iff         ~> snd_below_iff
  2714   expand_cfun_less     ~> cfun_below_iff
  2715   less_cfun_ext        ~> cfun_belowI
  2716   injection_less       ~> injection_below
  2717   less_up_def          ~> below_up_def
  2718   not_Iup_less         ~> not_Iup_below
  2719   Iup_less             ~> Iup_below
  2720   up_less              ~> up_below
  2721   Def_inject_less_eq   ~> Def_below_Def
  2722   Def_less_is_eq       ~> Def_below_iff
  2723   spair_less_iff       ~> spair_below_iff
  2724   less_sprod           ~> below_sprod
  2725   spair_less           ~> spair_below
  2726   sfst_less_iff        ~> sfst_below_iff
  2727   ssnd_less_iff        ~> ssnd_below_iff
  2728   fix_least_less       ~> fix_least_below
  2729   dist_less_one        ~> dist_below_one
  2730   less_ONE             ~> below_ONE
  2731   ONE_less_iff         ~> ONE_below_iff
  2732   less_sinlD           ~> below_sinlD
  2733   less_sinrD           ~> below_sinrD
  2734 
  2735 
  2736 *** FOL and ZF ***
  2737 
  2738 * All constant names are now qualified internally and use proper
  2739 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2740 
  2741 
  2742 *** ML ***
  2743 
  2744 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2745 Fail if the argument is false.  Due to inlining the source position of
  2746 failed assertions is included in the error output.
  2747 
  2748 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2749 text is in practice always evaluated with a stable theory checkpoint.
  2750 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2751 
  2752 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2753 the current context, not any accidental theory loader state as before.
  2754 Potential INCOMPATIBILITY, subtle change in semantics.
  2755 
  2756 * Syntax.pretty_priority (default 0) configures the required priority
  2757 of pretty-printed output and thus affects insertion of parentheses.
  2758 
  2759 * Syntax.default_root (default "any") configures the inner syntax
  2760 category (nonterminal symbol) for parsing of terms.
  2761 
  2762 * Former exception Library.UnequalLengths now coincides with
  2763 ListPair.UnequalLengths.
  2764 
  2765 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2766 main functionality is provided by structure Simplifier.
  2767 
  2768 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2769 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2770 almost all situations.
  2771 
  2772 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2773 See implementation manual for further details on exceptions in
  2774 Isabelle/ML.
  2775 
  2776 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2777 meaning.
  2778 
  2779 * Renamed structure PureThy to Pure_Thy and moved most of its
  2780 operations to structure Global_Theory, to emphasize that this is
  2781 rarely-used global-only stuff.
  2782 
  2783 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2784 instead (or tracing for high-volume output).
  2785 
  2786 * Configuration option show_question_marks only affects regular pretty
  2787 printing of types and terms, not raw Term.string_of_vname.
  2788 
  2789 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2790 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2791 @{thms} for most purposes.
  2792 
  2793 * ML structure Unsynchronized is never opened, not even in Isar
  2794 interaction mode as before.  Old Unsynchronized.set etc. have been
  2795 discontinued -- use plain := instead.  This should be *rare* anyway,
  2796 since modern tools always work via official context data, notably
  2797 configuration options.
  2798 
  2799 * Parallel and asynchronous execution requires special care concerning
  2800 interrupts.  Structure Exn provides some convenience functions that
  2801 avoid working directly with raw Interrupt.  User code must not absorb
  2802 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2803 followed by re-raising of the original exception.  Another common
  2804 source of mistakes are "handle _" patterns, which make the meaning of
  2805 the program subject to physical effects of the environment.
  2806 
  2807 
  2808 
  2809 New in Isabelle2009-2 (June 2010)
  2810 ---------------------------------
  2811 
  2812 *** General ***
  2813 
  2814 * Authentic syntax for *all* logical entities (type classes, type
  2815 constructors, term constants): provides simple and robust
  2816 correspondence between formal entities and concrete syntax.  Within
  2817 the parse tree / AST representations, "constants" are decorated by
  2818 their category (class, type, const) and spelled out explicitly with
  2819 their full internal name.
  2820 
  2821 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2822 and translations (translation rules and translation functions in ML).
  2823 Some hints on upgrading:
  2824 
  2825   - Many existing uses of 'syntax' and 'translations' can be replaced
  2826     by more modern 'type_notation', 'notation' and 'abbreviation',
  2827     which are independent of this issue.
  2828 
  2829   - 'translations' require markup within the AST; the term syntax
  2830     provides the following special forms:
  2831 
  2832       CONST c   -- produces syntax version of constant c from context
  2833       XCONST c  -- literally c, checked as constant from context
  2834       c         -- literally c, if declared by 'syntax'
  2835 
  2836     Plain identifiers are treated as AST variables -- occasionally the
  2837     system indicates accidental variables via the error "rhs contains
  2838     extra variables".
  2839 
  2840     Type classes and type constructors are marked according to their
  2841     concrete syntax.  Some old translations rules need to be written
  2842     for the "type" category, using type constructor application
  2843     instead of pseudo-term application of the default category
  2844     "logic".
  2845 
  2846   - 'parse_translation' etc. in ML may use the following
  2847     antiquotations:
  2848 
  2849       @{class_syntax c}   -- type class c within parse tree / AST
  2850       @{term_syntax c}    -- type constructor c within parse tree / AST
  2851       @{const_syntax c}   -- ML version of "CONST c" above
  2852       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2853 
  2854   - Literal types within 'typed_print_translations', i.e. those *not*
  2855     represented as pseudo-terms are represented verbatim.  Use @{class
  2856     c} or @{type_name c} here instead of the above syntax
  2857     antiquotations.
  2858 
  2859 Note that old non-authentic syntax was based on unqualified base
  2860 names, so all of the above "constant" names would coincide.  Recall
  2861 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2862 diagnose syntax problems.
  2863 
  2864 * Type constructors admit general mixfix syntax, not just infix.
  2865 
  2866 * Concrete syntax may be attached to local entities without a proof
  2867 body, too.  This works via regular mixfix annotations for 'fix',
  2868 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2869 similar to the 'notation' command in theory specifications.
  2870 
  2871 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2872 need to specify constant name and syntax separately.  Internal ML
  2873 datatype constructors have been renamed from InfixName to Infix etc.
  2874 Minor INCOMPATIBILITY.
  2875 
  2876 * Schematic theorem statements need to be explicitly markup as such,
  2877 via commands 'schematic_lemma', 'schematic_theorem',
  2878 'schematic_corollary'.  Thus the relevance of the proof is made
  2879 syntactically clear, which impacts performance in a parallel or
  2880 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2881 
  2882 * Use of cumulative prems via "!" in some proof methods has been
  2883 discontinued (old legacy feature).
  2884 
  2885 * References 'trace_simp' and 'debug_simp' have been replaced by
  2886 configuration options stored in the context. Enabling tracing (the
  2887 case of debugging is similar) in proofs works via
  2888 
  2889   using [[trace_simp = true]]
  2890 
  2891 Tracing is then active for all invocations of the simplifier in
  2892 subsequent goal refinement steps. Tracing may also still be enabled or
  2893 disabled via the ProofGeneral settings menu.
  2894 
  2895 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2896 'hide_fact' replace the former 'hide' KIND command.  Minor
  2897 INCOMPATIBILITY.
  2898 
  2899 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2900 more efficient than combinations with -q1 or -q2.
  2901 
  2902 
  2903 *** Pure ***
  2904 
  2905 * Proofterms record type-class reasoning explicitly, using the
  2906 "unconstrain" operation internally.  This eliminates all sort
  2907 constraints from a theorem and proof, introducing explicit
  2908 OFCLASS-premises.  On the proof term level, this operation is
  2909 automatically applied at theorem boundaries, such that closed proofs
  2910 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2911 inspect proof terms.
  2912 
  2913 * Local theory specifications may depend on extra type variables that
  2914 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2915 are added internally.  For example:
  2916 
  2917   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2918 
  2919 * Predicates of locales introduced by classes carry a mandatory
  2920 "class" prefix.  INCOMPATIBILITY.
  2921 
  2922 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2923 
  2924 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2925 'class' instead.
  2926 
  2927 * Command 'code_reflect' allows to incorporate generated ML code into
  2928 runtime environment; replaces immature code_datatype antiquotation.
  2929 INCOMPATIBILITY.
  2930 
  2931 * Code generator: simple concept for abstract datatypes obeying
  2932 invariants.
  2933 
  2934 * Code generator: details of internal data cache have no impact on the
  2935 user space functionality any longer.
  2936 
  2937 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2938 subgoals.  This is more appropriate for interpretations with 'where'.
  2939 INCOMPATIBILITY.
  2940 
  2941 * Command 'example_proof' opens an empty proof body.  This allows to
  2942 experiment with Isar, without producing any persistent result.
  2943 
  2944 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2945 within a local theory context, with explicit checking of the
  2946 constructors involved (in contrast to the raw 'syntax' versions).
  2947 
  2948 * Commands 'types' and 'typedecl' now work within a local theory
  2949 context -- without introducing dependencies on parameters or
  2950 assumptions, which is not possible in Isabelle/Pure.
  2951 
  2952 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2953 within a local theory context.  Minor INCOMPATIBILITY.
  2954 
  2955 
  2956 *** HOL ***
  2957 
  2958 * Command 'typedef' now works within a local theory context -- without
  2959 introducing dependencies on parameters or assumptions, which is not
  2960 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2961 contain multiple interpretations of local typedefs (with different
  2962 non-emptiness proofs), even in a global theory context.
  2963 
  2964 * New package for quotient types.  Commands 'quotient_type' and
  2965 'quotient_definition' may be used for defining types and constants by
  2966 quotient constructions.  An example is the type of integers created by
  2967 quotienting pairs of natural numbers:
  2968 
  2969   fun
  2970     intrel :: "(nat * nat) => (nat * nat) => bool"
  2971   where
  2972     "intrel (x, y) (u, v) = (x + v = u + y)"
  2973 
  2974   quotient_type int = "nat * nat" / intrel
  2975     by (auto simp add: equivp_def expand_fun_eq)
  2976 
  2977   quotient_definition
  2978     "0::int" is "(0::nat, 0::nat)"
  2979 
  2980 The method "lifting" can be used to lift of theorems from the
  2981 underlying "raw" type to the quotient type.  The example
  2982 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2983 construction and provides a reasoning infrastructure for finite sets.
  2984 
  2985 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2986 clash with new theory Quotient in Main HOL.
  2987 
  2988 * Moved the SMT binding into the main HOL session, eliminating
  2989 separate HOL-SMT session.
  2990 
  2991 * List membership infix mem operation is only an input abbreviation.
  2992 INCOMPATIBILITY.
  2993 
  2994 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2995 for future developements; former Library/Word.thy is still present in
  2996 the AFP entry RSAPPS.
  2997 
  2998 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2999 longer shadowed.  INCOMPATIBILITY.
  3000 
  3001 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  3002 INCOMPATIBILITY.
  3003 
  3004 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  3005 INCOMPATIBILITY.
  3006 
  3007 * Dropped normalizing_semiring etc; use the facts in semiring classes
  3008 instead.  INCOMPATIBILITY.
  3009 
  3010 * Dropped several real-specific versions of lemmas about floor and
  3011 ceiling; use the generic lemmas from theory "Archimedean_Field"
  3012 instead.  INCOMPATIBILITY.
  3013 
  3014   floor_number_of_eq         ~> floor_number_of
  3015   le_floor_eq_number_of      ~> number_of_le_floor
  3016   le_floor_eq_zero           ~> zero_le_floor
  3017   le_floor_eq_one            ~> one_le_floor
  3018   floor_less_eq_number_of    ~> floor_less_number_of
  3019   floor_less_eq_zero         ~> floor_less_zero
  3020   floor_less_eq_one          ~> floor_less_one
  3021   less_floor_eq_number_of    ~> number_of_less_floor
  3022   less_floor_eq_zero         ~> zero_less_floor
  3023   less_floor_eq_one          ~> one_less_floor
  3024   floor_le_eq_number_of      ~> floor_le_number_of
  3025   floor_le_eq_zero           ~> floor_le_zero
  3026   floor_le_eq_one            ~> floor_le_one
  3027   floor_subtract_number_of   ~> floor_diff_number_of
  3028   floor_subtract_one         ~> floor_diff_one
  3029   ceiling_number_of_eq       ~> ceiling_number_of
  3030   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  3031   ceiling_le_zero_eq         ~> ceiling_le_zero
  3032   ceiling_le_eq_one          ~> ceiling_le_one
  3033   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  3034   less_ceiling_eq_zero       ~> zero_less_ceiling
  3035   less_ceiling_eq_one        ~> one_less_ceiling
  3036   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  3037   ceiling_less_eq_zero       ~> ceiling_less_zero
  3038   ceiling_less_eq_one        ~> ceiling_less_one
  3039   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  3040   le_ceiling_eq_zero         ~> zero_le_ceiling
  3041   le_ceiling_eq_one          ~> one_le_ceiling
  3042   ceiling_subtract_number_of ~> ceiling_diff_number_of
  3043   ceiling_subtract_one       ~> ceiling_diff_one
  3044 
  3045 * Theory "Finite_Set": various folding_XXX locales facilitate the
  3046 application of the various fold combinators on finite sets.
  3047 
  3048 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  3049 provides abstract red-black tree type which is backed by "RBT_Impl" as
  3050 implementation.  INCOMPATIBILTY.
  3051 
  3052 * Theory Library/Coinductive_List has been removed -- superseded by
  3053 AFP/thys/Coinductive.
  3054 
  3055 * Theory PReal, including the type "preal" and related operations, has
  3056 been removed.  INCOMPATIBILITY.
  3057 
  3058 * Real: new development using Cauchy Sequences.
  3059 
  3060 * Split off theory "Big_Operators" containing setsum, setprod,
  3061 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  3062 
  3063 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  3064 "Int" etc.  INCOMPATIBILITY.
  3065 
  3066 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  3067 
  3068 * New set of rules "ac_simps" provides combined assoc / commute
  3069 rewrites for all interpretations of the appropriate generic locales.
  3070 
  3071 * Renamed theory "OrderedGroup" to "Groups" and split theory
  3072 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  3073 appropriate and more consistent names suitable for name prefixes
  3074 within the HOL theories.  INCOMPATIBILITY.
  3075 
  3076 * Some generic constants have been put to appropriate theories:
  3077   - less_eq, less: Orderings
  3078   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  3079   - inverse, divide: Rings
  3080 INCOMPATIBILITY.
  3081 
  3082 * More consistent naming of type classes involving orderings (and
  3083 lattices):
  3084 
  3085     lower_semilattice                   ~> semilattice_inf
  3086     upper_semilattice                   ~> semilattice_sup
  3087 
  3088     dense_linear_order                  ~> dense_linorder
  3089 
  3090     pordered_ab_group_add               ~> ordered_ab_group_add
  3091     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  3092     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  3093     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  3094     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  3095     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  3096     pordered_cancel_semiring            ~> ordered_cancel_semiring
  3097     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  3098     pordered_comm_ring                  ~> ordered_comm_ring
  3099     pordered_comm_semiring              ~> ordered_comm_semiring
  3100     pordered_ring                       ~> ordered_ring
  3101     pordered_ring_abs                   ~> ordered_ring_abs
  3102     pordered_semiring                   ~> ordered_semiring
  3103 
  3104     ordered_ab_group_add                ~> linordered_ab_group_add
  3105     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  3106     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  3107     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  3108     ordered_field                       ~> linordered_field
  3109     ordered_field_no_lb                 ~> linordered_field_no_lb
  3110     ordered_field_no_ub                 ~> linordered_field_no_ub
  3111     ordered_field_dense_linear_order    ~> dense_linordered_field
  3112     ordered_idom                        ~> linordered_idom
  3113     ordered_ring                        ~> linordered_ring
  3114     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  3115     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  3116     ordered_ring_strict                 ~> linordered_ring_strict
  3117     ordered_semidom                     ~> linordered_semidom
  3118     ordered_semiring                    ~> linordered_semiring
  3119     ordered_semiring_1                  ~> linordered_semiring_1
  3120     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  3121     ordered_semiring_strict             ~> linordered_semiring_strict
  3122 
  3123   The following slightly odd type classes have been moved to a
  3124   separate theory Library/Lattice_Algebras:
  3125 
  3126     lordered_ab_group_add               ~> lattice_ab_group_add
  3127     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  3128     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  3129     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  3130     lordered_ring                       ~> lattice_ring
  3131 
  3132 INCOMPATIBILITY.
  3133 
  3134 * Refined field classes:
  3135   - classes division_ring_inverse_zero, field_inverse_zero,
  3136     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3137     subsumes former division_by_zero class;
  3138   - numerous lemmas have been ported from field to division_ring.
  3139 INCOMPATIBILITY.
  3140 
  3141 * Refined algebra theorem collections:
  3142   - dropped theorem group group_simps, use algebra_simps instead;
  3143   - dropped theorem group ring_simps, use field_simps instead;
  3144   - proper theorem collection field_simps subsumes former theorem
  3145     groups field_eq_simps and field_simps;
  3146   - dropped lemma eq_minus_self_iff which is a duplicate for
  3147     equal_neg_zero.
  3148 INCOMPATIBILITY.
  3149 
  3150 * Theory Finite_Set and List: some lemmas have been generalized from
  3151 sets to lattices:
  3152 
  3153   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3154   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3155   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3156   union_Union_fold_union        ~> sup_Sup_fold_sup
  3157   Inter_fold_inter              ~> Inf_fold_inf
  3158   Union_fold_union              ~> Sup_fold_sup
  3159   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3160   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3161   INTER_fold_inter              ~> INFI_fold_inf
  3162   UNION_fold_union              ~> SUPR_fold_sup
  3163 
  3164 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3165 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3166 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3167 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3168 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3169 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3170 
  3171 * Reorganized theory Multiset: swapped notation of pointwise and
  3172 multiset order:
  3173 
  3174   - pointwise ordering is instance of class order with standard syntax
  3175     <= and <;
  3176   - multiset ordering has syntax <=# and <#; partial order properties
  3177     are provided by means of interpretation with prefix
  3178     multiset_order;
  3179   - less duplication, less historical organization of sections,
  3180     conversion from associations lists to multisets, rudimentary code
  3181     generation;
  3182   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3183     if needed.
  3184 
  3185 Renamed:
  3186 
  3187   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3188   multi_count_ext  ~>  multiset_ext
  3189   diff_union_inverse2  ~>  diff_union_cancelR
  3190 
  3191 INCOMPATIBILITY.
  3192 
  3193 * Theory Permutation: replaced local "remove" by List.remove1.
  3194 
  3195 * Code generation: ML and OCaml code is decorated with signatures.
  3196 
  3197 * Theory List: added transpose.
  3198 
  3199 * Library/Nat_Bijection.thy is a collection of bijective functions
  3200 between nat and other types, which supersedes the older libraries
  3201 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3202 
  3203   Constants:
  3204   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3205   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3206   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3207   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3208   Countable.pair_encode           ~> prod_encode
  3209   NatIso.prod2nat                 ~> prod_encode
  3210   NatIso.nat2prod                 ~> prod_decode
  3211   NatIso.sum2nat                  ~> sum_encode
  3212   NatIso.nat2sum                  ~> sum_decode
  3213   NatIso.list2nat                 ~> list_encode
  3214   NatIso.nat2list                 ~> list_decode
  3215   NatIso.set2nat                  ~> set_encode
  3216   NatIso.nat2set                  ~> set_decode
  3217 
  3218   Lemmas:
  3219   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3220   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3221   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3222   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3223   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3224   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3225   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3226   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3227   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3228   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3229   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3230   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3231   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3232 
  3233 * Sledgehammer:
  3234   - Renamed ATP commands:
  3235     atp_info     ~> sledgehammer running_atps
  3236     atp_kill     ~> sledgehammer kill_atps
  3237     atp_messages ~> sledgehammer messages
  3238     atp_minimize ~> sledgehammer minimize
  3239     print_atps   ~> sledgehammer available_atps
  3240     INCOMPATIBILITY.
  3241   - Added user's manual ("isabelle doc sledgehammer").
  3242   - Added option syntax and "sledgehammer_params" to customize
  3243     Sledgehammer's behavior.  See the manual for details.
  3244   - Modified the Isar proof reconstruction code so that it produces
  3245     direct proofs rather than proofs by contradiction.  (This feature
  3246     is still experimental.)
  3247   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3248     full-typed mode.
  3249   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3250 
  3251 * Nitpick:
  3252   - Added and implemented "binary_ints" and "bits" options.
  3253   - Added "std" option and implemented support for nonstandard models.
  3254   - Added and implemented "finitize" option to improve the precision
  3255     of infinite datatypes based on a monotonicity analysis.
  3256   - Added support for quotient types.
  3257   - Added support for "specification" and "ax_specification"
  3258     constructs.
  3259   - Added support for local definitions (for "function" and
  3260     "termination" proofs).
  3261   - Added support for term postprocessors.
  3262   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3263   - Improved efficiency of "destroy_constrs" optimization.
  3264   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3265     record getters.
  3266   - Fixed soundness bug related to higher-order constructors.
  3267   - Fixed soundness bug when "full_descrs" is enabled.
  3268   - Improved precision of set constructs.
  3269   - Added "atoms" option.
  3270   - Added cache to speed up repeated Kodkod invocations on the same
  3271     problems.
  3272   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3273     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3274     "SAT4J_Light".  INCOMPATIBILITY.
  3275   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3276     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3277   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3278 
  3279 * Method "induct" now takes instantiations of the form t, where t is not
  3280   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3281   If this is not intended, t has to be enclosed in parentheses.
  3282   By default, the equalities generated by definitional instantiations
  3283   are pre-simplified, which may cause parameters of inductive cases
  3284   to disappear, or may even delete some of the inductive cases.
  3285   Use "induct (no_simp)" instead of "induct" to restore the old
  3286   behaviour. The (no_simp) option is also understood by the "cases"
  3287   and "nominal_induct" methods, which now perform pre-simplification, too.
  3288   INCOMPATIBILITY.
  3289 
  3290 
  3291 *** HOLCF ***
  3292 
  3293 * Variable names in lemmas generated by the domain package have
  3294 changed; the naming scheme is now consistent with the HOL datatype
  3295 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3296 
  3297 * The domain package no longer defines the function "foo_copy" for
  3298 recursive domain "foo".  The reach lemma is now stated directly in
  3299 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3300 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3301 
  3302 * Most definedness lemmas generated by the domain package (previously
  3303 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3304 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3305 Proofs that used definedness lemmas as intro rules may break,
  3306 potential INCOMPATIBILITY.
  3307 
  3308 * Induction and casedist rules generated by the domain package now
  3309 declare proper case_names (one called "bottom", and one named for each
  3310 constructor).  INCOMPATIBILITY.
  3311 
  3312 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3313 rules are generated for each domain, INCOMPATIBILITY.
  3314 
  3315   foo_bar.reach       ~> foo.reach  bar.reach
  3316   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3317 
  3318 * Some lemmas generated by the domain package have been renamed for
  3319 consistency with the datatype package, INCOMPATIBILITY.
  3320 
  3321   foo.ind        ~> foo.induct
  3322   foo.finite_ind ~> foo.finite_induct
  3323   foo.coind      ~> foo.coinduct
  3324   foo.casedist   ~> foo.exhaust
  3325   foo.exhaust    ~> foo.nchotomy
  3326 
  3327 * For consistency with other definition packages, the fixrec package
  3328 now generates qualified theorem names, INCOMPATIBILITY.
  3329 
  3330   foo_simps  ~> foo.simps
  3331   foo_unfold ~> foo.unfold
  3332   foo_induct ~> foo.induct
  3333 
  3334 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3335 method and internal fixrec proofs now use the default simpset instead.
  3336 INCOMPATIBILITY.
  3337 
  3338 * The "contlub" predicate has been removed.  Proof scripts should use
  3339 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3340 
  3341 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3342 
  3343 * The constants cpair, cfst, and csnd have been removed in favor of
  3344 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3345 
  3346 
  3347 *** ML ***
  3348 
  3349 * Antiquotations for basic formal entities:
  3350 
  3351     @{class NAME}         -- type class
  3352     @{class_syntax NAME}  -- syntax representation of the above
  3353 
  3354     @{type_name NAME}     -- logical type
  3355     @{type_abbrev NAME}   -- type abbreviation
  3356     @{nonterminal NAME}   -- type of concrete syntactic category
  3357     @{type_syntax NAME}   -- syntax representation of any of the above
  3358 
  3359     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3360     @{const_abbrev NAME}  -- abbreviated constant
  3361     @{const_syntax NAME}  -- syntax representation of any of the above
  3362 
  3363 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3364 syntax constant (cf. 'syntax' command).
  3365 
  3366 * Antiquotation @{make_string} inlines a function to print arbitrary
  3367 values similar to the ML toplevel.  The result is compiler dependent
  3368 and may fall back on "?" in certain situations.
  3369 
  3370 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3371 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3372 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3373 not work with the asynchronous Isar document model.
  3374 
  3375 * Configuration options now admit dynamic default values, depending on
  3376 the context or even global references.
  3377 
  3378 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3379 uses an efficient external library if available (for Poly/ML).
  3380 
  3381 * Renamed some important ML structures, while keeping the old names
  3382 for some time as aliases within the structure Legacy:
  3383 
  3384   OuterKeyword  ~>  Keyword
  3385   OuterLex      ~>  Token
  3386   OuterParse    ~>  Parse
  3387   OuterSyntax   ~>  Outer_Syntax
  3388   PrintMode     ~>  Print_Mode
  3389   SpecParse     ~>  Parse_Spec
  3390   ThyInfo       ~>  Thy_Info
  3391   ThyLoad       ~>  Thy_Load
  3392   ThyOutput     ~>  Thy_Output
  3393   TypeInfer     ~>  Type_Infer
  3394 
  3395 Note that "open Legacy" simplifies porting of sources, but forgetting
  3396 to remove it again will complicate porting again in the future.
  3397 
  3398 * Most operations that refer to a global context are named
  3399 accordingly, e.g. Simplifier.global_context or
  3400 ProofContext.init_global.  There are some situations where a global
  3401 context actually works, but under normal circumstances one needs to
  3402 pass the proper local context through the code!
  3403 
  3404 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3405 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3406 Pretty.pp argument to merge, which is absent in the standard
  3407 Theory_Data version.
  3408 
  3409 * Sorts.certify_sort and derived "cert" operations for types and terms
  3410 no longer minimize sorts.  Thus certification at the boundary of the
  3411 inference kernel becomes invariant under addition of class relations,
  3412 which is an important monotonicity principle.  Sorts are now minimized
  3413 in the syntax layer only, at the boundary between the end-user and the
  3414 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3415 explicitly in rare situations.
  3416 
  3417 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3418 emphasize that this is in no way a standard operation.
  3419 INCOMPATIBILITY.
  3420 
  3421 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3422 usual for resolution.  Rare INCOMPATIBILITY.
  3423 
  3424 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3425 to emphasize that these only work in a global situation (which is
  3426 quite rare).
  3427 
  3428 * Curried take and drop in library.ML; negative length is interpreted
  3429 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3430 
  3431 * Proof terms: type substitutions on proof constants now use canonical
  3432 order of type variables.  INCOMPATIBILITY for tools working with proof
  3433 terms.
  3434 
  3435 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3436 may no longer carry premises.  User-level specifications are
  3437 transformed accordingly by Thm.add_axiom/add_def.
  3438 
  3439 
  3440 *** System ***
  3441 
  3442 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3443 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3444 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3445 
  3446 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3447 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3448 respectively.
  3449 
  3450 * Old lib/scripts/polyml-platform is superseded by the
  3451 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3452 variant, even on a 64 bit machine.  The following example setting
  3453 prefers 64 bit if available:
  3454 
  3455   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3456 
  3457 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3458 Isabelle/Scala layer for advanced prover interaction and integration.
  3459 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3460 component.
  3461 
  3462 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3463 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3464 similar to the default assignment of the document preparation system
  3465 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3466 provides some operations for direct access to the font without asking
  3467 the user for manual installation.
  3468 
  3469 
  3470 
  3471 New in Isabelle2009-1 (December 2009)
  3472 -------------------------------------
  3473 
  3474 *** General ***
  3475 
  3476 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3477 one backslash should be used, even in ML sources.
  3478 
  3479 
  3480 *** Pure ***
  3481 
  3482 * Locale interpretation propagates mixins along the locale hierarchy.
  3483 The currently only available mixins are the equations used to map
  3484 local definitions to terms of the target domain of an interpretation.
  3485 
  3486 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3487 loc" to print all interpretations of locale "loc" in the theory.
  3488 Interpretations in proofs are not shown.
  3489 
  3490 * Thoroughly revised locales tutorial.  New section on conditional
  3491 interpretation.
  3492 
  3493 * On instantiation of classes, remaining undefined class parameters
  3494 are formally declared.  INCOMPATIBILITY.
  3495 
  3496 
  3497 *** Document preparation ***
  3498 
  3499 * New generalized style concept for printing terms: @{foo (style) ...}
  3500 instead of @{foo_style style ...}  (old form is still retained for
  3501 backward compatibility).  Styles can be also applied for
  3502 antiquotations prop, term_type and typeof.
  3503 
  3504 
  3505 *** HOL ***
  3506 
  3507 * New proof method "smt" for a combination of first-order logic with
  3508 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3509 fixed-size bitvectors; there is also basic support for higher-order
  3510 features (esp. lambda abstractions).  It is an incomplete decision
  3511 procedure based on external SMT solvers using the oracle mechanism;
  3512 for the SMT solver Z3, this method is proof-producing.  Certificates
  3513 are provided to avoid calling the external solvers solely for
  3514 re-checking proofs.  Due to a remote SMT service there is no need for
  3515 installing SMT solvers locally.  See src/HOL/SMT.
  3516 
  3517 * New commands to load and prove verification conditions generated by
  3518 the Boogie program verifier or derived systems (e.g. the Verifying C
  3519 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3520 
  3521 * New counterexample generator tool 'nitpick' based on the Kodkod
  3522 relational model finder.  See src/HOL/Tools/Nitpick and
  3523 src/HOL/Nitpick_Examples.
  3524 
  3525 * New commands 'code_pred' and 'values' to invoke the predicate
  3526 compiler and to enumerate values of inductive predicates.
  3527 
  3528 * A tabled implementation of the reflexive transitive closure.
  3529 
  3530 * New implementation of quickcheck uses generic code generator;
  3531 default generators are provided for all suitable HOL types, records
  3532 and datatypes.  Old quickcheck can be re-activated importing theory
  3533 Library/SML_Quickcheck.
  3534 
  3535 * New testing tool Mirabelle for automated proof tools.  Applies
  3536 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3537 every proof step in a theory.  To be used in batch mode via the
  3538 "mirabelle" utility.
  3539 
  3540 * New proof method "sos" (sum of squares) for nonlinear real
  3541 arithmetic (originally due to John Harison). It requires theory
  3542 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3543 works well in practice on quantifier-free real arithmetic with +, -,
  3544 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3545 inequalities between polynomials.  It makes use of external
  3546 semidefinite programming solvers.  Method "sos" generates a
  3547 certificate that can be pasted into the proof thus avoiding the need
  3548 to call an external tool every time the proof is checked.  See
  3549 src/HOL/Library/Sum_Of_Squares.
  3550 
  3551 * New method "linarith" invokes existing linear arithmetic decision
  3552 procedure only.
  3553 
  3554 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3555 
  3556 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3557 Causes full type information to be output to the ATPs.  This slows
  3558 ATPs down considerably but eliminates a source of unsound "proofs"
  3559 that fail later.
  3560 
  3561 * New method "metisFT": A version of metis that uses full type
  3562 information in order to avoid failures of proof reconstruction.
  3563 
  3564 * New evaluator "approximate" approximates an real valued term using
  3565 the same method as the approximation method.
  3566 
  3567 * Method "approximate" now supports arithmetic expressions as
  3568 boundaries of intervals and implements interval splitting and Taylor
  3569 series expansion.
  3570 
  3571 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3572 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3573 
  3574 * New theory SupInf of the supremum and infimum operators for sets of
  3575 reals.
  3576 
  3577 * New theory Probability, which contains a development of measure
  3578 theory, eventually leading to Lebesgue integration and probability.
  3579 
  3580 * Extended Multivariate Analysis to include derivation and Brouwer's
  3581 fixpoint theorem.
  3582 
  3583 * Reorganization of number theory, INCOMPATIBILITY:
  3584   - new number theory development for nat and int, in theories Divides
  3585     and GCD as well as in new session Number_Theory
  3586   - some constants and facts now suffixed with _nat and _int
  3587     accordingly
  3588   - former session NumberTheory now named Old_Number_Theory, including
  3589     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3590   - moved theory Pocklington from src/HOL/Library to
  3591     src/HOL/Old_Number_Theory
  3592 
  3593 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3594 lcm of finite and infinite sets. It is shown that they form a complete
  3595 lattice.
  3596 
  3597 * Class semiring_div requires superclass no_zero_divisors and proof of
  3598 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3599 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3600 generalized to class semiring_div, subsuming former theorems
  3601 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3602 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3603 INCOMPATIBILITY.
  3604 
  3605 * Refinements to lattice classes and sets:
  3606   - less default intro/elim rules in locale variant, more default
  3607     intro/elim rules in class variant: more uniformity
  3608   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3609     le_inf_iff
  3610   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3611     sup_aci)
  3612   - renamed ACI to inf_sup_aci
  3613   - new class "boolean_algebra"
  3614   - class "complete_lattice" moved to separate theory
  3615     "Complete_Lattice"; corresponding constants (and abbreviations)
  3616     renamed and with authentic syntax:
  3617     Set.Inf ~>    Complete_Lattice.Inf
  3618     Set.Sup ~>    Complete_Lattice.Sup
  3619     Set.INFI ~>   Complete_Lattice.INFI
  3620     Set.SUPR ~>   Complete_Lattice.SUPR
  3621     Set.Inter ~>  Complete_Lattice.Inter
  3622     Set.Union ~>  Complete_Lattice.Union
  3623     Set.INTER ~>  Complete_Lattice.INTER
  3624     Set.UNION ~>  Complete_Lattice.UNION
  3625   - authentic syntax for
  3626     Set.Pow
  3627     Set.image
  3628   - mere abbreviations:
  3629     Set.empty               (for bot)
  3630     Set.UNIV                (for top)
  3631     Set.inter               (for inf, formerly Set.Int)
  3632     Set.union               (for sup, formerly Set.Un)
  3633     Complete_Lattice.Inter  (for Inf)
  3634     Complete_Lattice.Union  (for Sup)
  3635     Complete_Lattice.INTER  (for INFI)
  3636     Complete_Lattice.UNION  (for SUPR)
  3637   - object-logic definitions as far as appropriate
  3638 
  3639 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3640 Un_subset_iff are explicitly deleted as default simp rules; then also
  3641 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3642 deleted to achieve the desired effect.
  3643 
  3644 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3645 rules by default any longer; the same applies to min_max.inf_absorb1
  3646 etc.  INCOMPATIBILITY.
  3647 
  3648 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3649 pred_set_conv by default.  INCOMPATIBILITY.
  3650 
  3651 * Power operations on relations and functions are now one dedicated
  3652 constant "compow" with infix syntax "^^".  Power operation on
  3653 multiplicative monoids retains syntax "^" and is now defined generic
  3654 in class power.  INCOMPATIBILITY.
  3655 
  3656 * Relation composition "R O S" now has a more standard argument order:
  3657 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3658 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3659 break, since the O_assoc rule was not rewritten like this.  Fix using
  3660 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3661 
  3662 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3663 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3664 INCOMPATIBILITY.
  3665 
  3666 * Most rules produced by inductive and datatype package have mandatory
  3667 prefixes.  INCOMPATIBILITY.
  3668 
  3669 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3670 the attribute of the same name.  Each of the theorems in the list
  3671 DERIV_intros assumes composition with an additional function and
  3672 matches a variable to the derivative, which has to be solved by the
  3673 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3674 of most elementary terms.  Former Maclauren.DERIV_tac and
  3675 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3676 INCOMPATIBILITY.
  3677 
  3678 * Code generator attributes follow the usual underscore convention:
  3679     code_unfold     replaces    code unfold
  3680     code_post       replaces    code post
  3681     etc.
  3682   INCOMPATIBILITY.
  3683 
  3684 * Renamed methods:
  3685     sizechange -> size_change
  3686     induct_scheme -> induction_schema
  3687   INCOMPATIBILITY.
  3688 
  3689 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3690 INCOMPATIBILITY, use "undefined" directly.
  3691 
  3692 * Renamed theorems:
  3693     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3694     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3695     Suc_plus1 -> Suc_eq_plus1
  3696     *anti_sym -> *antisym*
  3697     vector_less_eq_def -> vector_le_def
  3698   INCOMPATIBILITY.
  3699 
  3700 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3701 INCOMPATIBILITY.
  3702 
  3703 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3704 INCOMPATIBILITY.
  3705 
  3706 
  3707 *** HOLCF ***
  3708 
  3709 * Theory Representable defines a class "rep" of domains that are
  3710 representable (via an ep-pair) in the universal domain type "udom".
  3711 Instances are provided for all type constructors defined in HOLCF.
  3712 
  3713 * The 'new_domain' command is a purely definitional version of the
  3714 domain package, for representable domains.  Syntax is identical to the
  3715 old domain package.  The 'new_domain' package also supports indirect
  3716 recursion using previously-defined type constructors.  See
  3717 src/HOLCF/ex/New_Domain.thy for examples.
  3718 
  3719 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3720 on the left-hand side of an equation, and then performs
  3721 simplification.  Rewriting is done using rules declared with the
  3722 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3723 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3724 
  3725 * The pattern-match compiler in 'fixrec' can now handle constructors
  3726 with HOL function types.  Pattern-match combinators for the Pair
  3727 constructor are pre-configured.
  3728 
  3729 * The 'fixrec' package now produces better fixed-point induction rules
  3730 for mutually-recursive definitions:  Induction rules have conclusions
  3731 of the form "P foo bar" instead of "P <foo, bar>".
  3732 
  3733 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3734 been renamed to "below".  The name "below" now replaces "less" in many
  3735 theorem names.  (Legacy theorem names using "less" are still supported
  3736 as well.)
  3737 
  3738 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3739 patterns can be used to generate strictness rules, or to make
  3740 functions more strict (much like the bang-patterns supported by the
  3741 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3742 examples.
  3743 
  3744 
  3745 *** ML ***
  3746 
  3747 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3748 errors and run-time exceptions, including detailed source positions.
  3749 
  3750 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3751 identified entries, with some additional information such as source
  3752 position, logical grouping etc.
  3753 
  3754 * Theory and context data is now introduced by the simplified and
  3755 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3756 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3757 copy operation) is still available for some time.
  3758 
  3759 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3760 provides a high-level programming interface to synchronized state
  3761 variables with atomic update.  This works via pure function
  3762 application within a critical section -- its runtime should be as
  3763 short as possible; beware of deadlocks if critical code is nested,
  3764 either directly or indirectly via other synchronized variables!
  3765 
  3766 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3767 wraps raw ML references, explicitly indicating their non-thread-safe
  3768 behaviour.  The Isar toplevel keeps this structure open, to
  3769 accommodate Proof General as well as quick and dirty interactive
  3770 experiments with references.
  3771 
  3772 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3773 parallel tactical reasoning.
  3774 
  3775 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3776 are similar to SUBPROOF, but are slightly more flexible: only the
  3777 specified parts of the subgoal are imported into the context, and the
  3778 body tactic may introduce new subgoals and schematic variables.
  3779 
  3780 * Old tactical METAHYPS, which does not observe the proof context, has
  3781 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3782 or Subgoal.FOCUS etc.
  3783 
  3784 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3785 functors have their own ML name space there is no point to mark them
  3786 separately.)  Minor INCOMPATIBILITY.
  3787 
  3788 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3789 
  3790 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3791 INCOMPATIBILITY.
  3792 
  3793 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3794 mode, which avoids critical setmp.
  3795 
  3796 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3797 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3798 Attrib/Method.setup introduced in Isabelle2009.
  3799 
  3800 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3801 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3802 last resort.  INCOMPATIBILITY.
  3803 
  3804 * Display.pretty_thm now requires a proper context (cf. former
  3805 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3806 or even Display.pretty_thm_without_context as last resort.
  3807 INCOMPATIBILITY.
  3808 
  3809 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3810 Syntax.pretty_typ/term directly, preferably with proper context
  3811 instead of global theory.
  3812 
  3813 
  3814 *** System ***
  3815 
  3816 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3817 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3818 usedir option -q.
  3819 
  3820 * Support for additional "Isabelle components" via etc/components, see
  3821 also the system manual.
  3822 
  3823 * The isabelle makeall tool now operates on all components with
  3824 IsaMakefile, not just hardwired "logics".
  3825 
  3826 * Removed "compress" option from isabelle-process and isabelle usedir;
  3827 this is always enabled.
  3828 
  3829 * Discontinued support for Poly/ML 4.x versions.
  3830 
  3831 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3832 on a given logic image.  This requires the lighttpd webserver and is
  3833 currently supported on Linux only.
  3834 
  3835 
  3836 
  3837 New in Isabelle2009 (April 2009)
  3838 --------------------------------
  3839 
  3840 *** General ***
  3841 
  3842 * Simplified main Isabelle executables, with less surprises on
  3843 case-insensitive file-systems (such as Mac OS).
  3844 
  3845   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3846     "isatool."
  3847 
  3848   - The former "isabelle" alias for "isabelle-process" has been
  3849     removed (should rarely occur to regular users).
  3850 
  3851   - The former "isabelle-interface" and its alias "Isabelle" have been
  3852     removed (interfaces are now regular Isabelle tools).
  3853 
  3854 Within scripts and make files, the Isabelle environment variables
  3855 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3856 respectively.  (The latter are still available as legacy feature.)
  3857 
  3858 The old isabelle-interface wrapper could react in confusing ways if
  3859 the interface was uninstalled or changed otherwise.  Individual
  3860 interface tool configuration is now more explicit, see also the
  3861 Isabelle system manual.  In particular, Proof General is now available
  3862 via "isabelle emacs".
  3863 
  3864 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3865 purge installed copies of Isabelle executables and re-run "isabelle
  3866 install -p ...", or use symlinks.
  3867 
  3868 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3869 old ~/isabelle, which was slightly non-standard and apt to cause
  3870 surprises on case-insensitive file-systems (such as Mac OS).
  3871 
  3872 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3873 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3874 care is required when using older releases of Isabelle.  Note that
  3875 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3876 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3877 
  3878 * Proofs of fully specified statements are run in parallel on
  3879 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3880 a regular 4-core machine, if the initial heap space is made reasonably
  3881 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3882 
  3883 * The main reference manuals ("isar-ref", "implementation", and
  3884 "system") have been updated and extended.  Formally checked references
  3885 as hyperlinks are now available uniformly.
  3886 
  3887 
  3888 *** Pure ***
  3889 
  3890 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3891 respects.  The most important changes are listed below.  See the
  3892 Tutorial on Locales ("locales" manual) for details.
  3893 
  3894 - In locale expressions, instantiation replaces renaming.  Parameters
  3895 must be declared in a for clause.  To aid compatibility with previous
  3896 parameter inheritance, in locale declarations, parameters that are not
  3897 'touched' (instantiation position "_" or omitted) are implicitly added
  3898 with their syntax at the beginning of the for clause.
  3899 
  3900 - Syntax from abbreviations and definitions in locales is available in
  3901 locale expressions and context elements.  The latter is particularly
  3902 useful in locale declarations.
  3903 
  3904 - More flexible mechanisms to qualify names generated by locale
  3905 expressions.  Qualifiers (prefixes) may be specified in locale
  3906 expressions, and can be marked as mandatory (syntax: "name!:") or
  3907 optional (syntax "name?:").  The default depends for plain "name:"
  3908 depends on the situation where a locale expression is used: in
  3909 commands 'locale' and 'sublocale' prefixes are optional, in
  3910 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3911 implicit qualifiers derived from the parameter names of a locale are
  3912 no longer generated.
  3913 
  3914 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3915 instantiation clause in "interpretation" and "interpret" (square
  3916 brackets) is no longer available.  Use locale expressions.
  3917 
  3918 - When converting proof scripts, mandatory qualifiers in
  3919 'interpretation' and 'interpret' should be retained by default, even
  3920 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3921 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3922 in locale expressions range over a single locale instance only.
  3923 
  3924 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3925 In existing theorem specifications replace the includes element by the
  3926 respective context elements of the included locale, omitting those
  3927 that are already present in the theorem specification.  Multiple
  3928 assume elements of a locale should be replaced by a single one
  3929 involving the locale predicate.  In the proof body, declarations (most
  3930 notably theorems) may be regained by interpreting the respective
  3931 locales in the proof context as required (command "interpret").
  3932 
  3933 If using "includes" in replacement of a target solely because the
  3934 parameter types in the theorem are not as general as in the target,
  3935 consider declaring a new locale with additional type constraints on
  3936 the parameters (context element "constrains").
  3937 
  3938 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3939 
  3940 - Locale interpretation commands no longer attempt to simplify goal.
  3941 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3942 methods intro_locales and unfold_locales to clarify.
  3943 
  3944 - Locale interpretation commands no longer accept interpretation
  3945 attributes.  INCOMPATIBILITY.
  3946 
  3947 * Class declaration: so-called "base sort" must not be given in import
  3948 list any longer, but is inferred from the specification.  Particularly
  3949 in HOL, write
  3950 
  3951     class foo = ...
  3952 
  3953 instead of
  3954 
  3955     class foo = type + ...
  3956 
  3957 * Class target: global versions of theorems stemming do not carry a
  3958 parameter prefix any longer.  INCOMPATIBILITY.
  3959 
  3960 * Class 'instance' command no longer accepts attached definitions.
  3961 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3962 
  3963 * Recovered hiding of consts, which was accidentally broken in
  3964 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3965 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3966 
  3967 * Slightly more coherent Pure syntax, with updated documentation in
  3968 isar-ref manual.  Removed locales meta_term_syntax and
  3969 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3970 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3971 directly in regular applications.
  3972 
  3973 * There is a new syntactic category "float_const" for signed decimal
  3974 fractions (e.g. 123.45 or -123.45).
  3975 
  3976 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3977 interface with 'setup' command instead.
  3978 
  3979 * Command 'local_setup' is similar to 'setup', but operates on a local
  3980 theory context.
  3981 
  3982 * The 'axiomatization' command now only works within a global theory
  3983 context.  INCOMPATIBILITY.
  3984 
  3985 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3986 hypotheses.  Sorts required in the course of reasoning need to be
  3987 covered by the constraints in the initial statement, completed by the
  3988 type instance information of the background theory.  Non-trivial sort
  3989 hypotheses, which rarely occur in practice, may be specified via
  3990 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3991 
  3992   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3993 
  3994 The result contains an implicit sort hypotheses as before --
  3995 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3996 normalization.
  3997 
  3998 * Generalized Isar history, with support for linear undo, direct state
  3999 addressing etc.
  4000 
  4001 * Changed defaults for unify configuration options:
  4002 
  4003   unify_trace_bound = 50 (formerly 25)
  4004   unify_search_bound = 60 (formerly 30)
  4005 
  4006 * Different bookkeeping for code equations (INCOMPATIBILITY):
  4007 
  4008   a) On theory merge, the last set of code equations for a particular
  4009      constant is taken (in accordance with the policy applied by other
  4010      parts of the code generator framework).
  4011 
  4012   b) Code equations stemming from explicit declarations (e.g. code
  4013      attribute) gain priority over default code equations stemming
  4014      from definition, primrec, fun etc.
  4015 
  4016 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  4017 
  4018 * Unified theorem tables for both code generators.  Thus [code
  4019 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  4020 
  4021 * Command 'find_consts' searches for constants based on type and name
  4022 patterns, e.g.
  4023 
  4024     find_consts "_ => bool"
  4025 
  4026 By default, matching is against subtypes, but it may be restricted to
  4027 the whole type.  Searching by name is possible.  Multiple queries are
  4028 conjunctive and queries may be negated by prefixing them with a
  4029 hyphen:
  4030 
  4031     find_consts strict: "_ => bool" name: "Int" -"int => int"
  4032 
  4033 * New 'find_theorems' criterion "solves" matches theorems that
  4034 directly solve the current goal (modulo higher-order unification).
  4035 
  4036 * Auto solve feature for main theorem statements: whenever a new goal
  4037 is stated, "find_theorems solves" is called; any theorems that could
  4038 solve the lemma directly are listed as part of the goal state.
  4039 Cf. associated options in Proof General Isabelle settings menu,
  4040 enabled by default, with reasonable timeout for pathological cases of
  4041 higher-order unification.
  4042 
  4043 
  4044 *** Document preparation ***
  4045 
  4046 * Antiquotation @{lemma} now imitates a regular terminal proof,
  4047 demanding keyword 'by' and supporting the full method expression
  4048 syntax just like the Isar command 'by'.
  4049 
  4050 
  4051 *** HOL ***
  4052 
  4053 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  4054 points Main and Complex_Main remain as before.
  4055 
  4056 * Logic image HOL-Plain provides a minimal HOL with the most important
  4057 tools available (inductive, datatype, primrec, ...).  This facilitates
  4058 experimentation and tool development.  Note that user applications
  4059 (and library theories) should never refer to anything below theory
  4060 Main, as before.
  4061 
  4062 * Logic image HOL-Main stops at theory Main, and thus facilitates
  4063 experimentation due to shorter build times.
  4064 
  4065 * Logic image HOL-NSA contains theories of nonstandard analysis which
  4066 were previously part of former HOL-Complex.  Entry point Hyperreal
  4067 remains valid, but theories formerly using Complex_Main should now use
  4068 new entry point Hypercomplex.
  4069 
  4070 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  4071 Posix processes.  Avoids potentially expensive forking of the ML
  4072 process.  New thread-based implementation also works on non-Unix
  4073 platforms (Cygwin).  Provers are no longer hardwired, but defined
  4074 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  4075 commands are covered in the isar-ref manual.
  4076 
  4077 * Wrapper scripts for remote SystemOnTPTP service allows to use
  4078 sledgehammer without local ATP installation (Vampire etc.). Other
  4079 provers may be included via suitable ML wrappers, see also
  4080 src/HOL/ATP_Linkup.thy.
  4081 
  4082 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  4083 menu.
  4084 
  4085 * The metis method no longer fails because the theorem is too trivial
  4086 (contains the empty clause).
  4087 
  4088 * The metis method now fails in the usual manner, rather than raising
  4089 an exception, if it determines that it cannot prove the theorem.
  4090 
  4091 * Method "coherent" implements a prover for coherent logic (see also
  4092 src/Tools/coherent.ML).
  4093 
  4094 * Constants "undefined" and "default" replace "arbitrary".  Usually
  4095 "undefined" is the right choice to replace "arbitrary", though
  4096 logically there is no difference.  INCOMPATIBILITY.
  4097 
  4098 * Command "value" now integrates different evaluation mechanisms.  The
  4099 result of the first successful evaluation mechanism is printed.  In
  4100 square brackets a particular named evaluation mechanisms may be
  4101 specified (currently, [SML], [code] or [nbe]).  See further
  4102 src/HOL/ex/Eval_Examples.thy.
  4103 
  4104 * Normalization by evaluation now allows non-leftlinear equations.
  4105 Declare with attribute [code nbe].
  4106 
  4107 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  4108 as the structured Isar versions "cases" and "induct", cf. the
  4109 corresponding "cases" and "induct" attributes.  Mutual induction rules
  4110 are now presented as a list of individual projections
  4111 (e.g. foo_bar.inducts for types foo and bar); the old format with
  4112 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  4113 rare situations a different rule is selected --- notably nested tuple
  4114 elimination instead of former prod.exhaust: use explicit (case_tac t
  4115 rule: prod.exhaust) here.
  4116 
  4117 * Attributes "cases", "induct", "coinduct" support "del" option.
  4118 
  4119 * Removed fact "case_split_thm", which duplicates "case_split".
  4120 
  4121 * The option datatype has been moved to a new theory Option.  Renamed
  4122 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  4123 
  4124 * New predicate "strict_mono" classifies strict functions on partial
  4125 orders.  With strict functions on linear orders, reasoning about
  4126 (in)equalities is facilitated by theorems "strict_mono_eq",
  4127 "strict_mono_less_eq" and "strict_mono_less".
  4128 
  4129 * Some set operations are now proper qualified constants with
  4130 authentic syntax.  INCOMPATIBILITY:
  4131 
  4132     op Int ~>   Set.Int
  4133     op Un ~>    Set.Un
  4134     INTER ~>    Set.INTER
  4135     UNION ~>    Set.UNION
  4136     Inter ~>    Set.Inter
  4137     Union ~>    Set.Union
  4138     {} ~>       Set.empty
  4139     UNIV ~>     Set.UNIV
  4140 
  4141 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4142 theory Set.
  4143 
  4144 * Auxiliary class "itself" has disappeared -- classes without any
  4145 parameter are treated as expected by the 'class' command.
  4146 
  4147 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4148 
  4149 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4150 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4151 
  4152 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4153 method "approximation".  It proves formulas on real values by using
  4154 interval arithmetic.  In the formulas are also the transcendental
  4155 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4156 allowed. For examples see
  4157 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4158 
  4159 * Theory "Reflection" now resides in HOL/Library.
  4160 
  4161 * Entry point to Word library now simply named "Word".
  4162 INCOMPATIBILITY.
  4163 
  4164 * Made source layout more coherent with logical distribution
  4165 structure:
  4166 
  4167     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4168     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4169     src/HOL/Library/GCD.thy ~> src/HOL/
  4170     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4171     src/HOL/Library/Parity.thy ~> src/HOL/
  4172     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4173     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4174     src/HOL/Real/Lubs.thy ~> src/HOL/
  4175     src/HOL/Real/PReal.thy ~> src/HOL/
  4176     src/HOL/Real/Rational.thy ~> src/HOL/
  4177     src/HOL/Real/RComplete.thy ~> src/HOL/
  4178     src/HOL/Real/RealDef.thy ~> src/HOL/
  4179     src/HOL/Real/RealPow.thy ~> src/HOL/
  4180     src/HOL/Real/Real.thy ~> src/HOL/
  4181     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4182     src/HOL/Complex/Complex.thy ~> src/HOL/
  4183     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4184     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4185     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4186     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4187     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4188     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4189     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4190     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4191     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4192     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4193     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4194     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4195     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4196     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4197     src/HOL/Real/Float ~> src/HOL/Library/
  4198     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4199     src/HOL/Real/RealVector.thy ~> src/HOL/
  4200 
  4201     src/HOL/arith_data.ML ~> src/HOL/Tools
  4202     src/HOL/hologic.ML ~> src/HOL/Tools
  4203     src/HOL/simpdata.ML ~> src/HOL/Tools
  4204     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4205     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4206     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4207     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4208     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4209     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4210     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4211 
  4212     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4213     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4214     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4215     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4216     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4217     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4218 
  4219 * If methods "eval" and "evaluation" encounter a structured proof
  4220 state with !!/==>, only the conclusion is evaluated to True (if
  4221 possible), avoiding strange error messages.
  4222 
  4223 * Method "sizechange" automates termination proofs using (a
  4224 modification of) the size-change principle.  Requires SAT solver.  See
  4225 src/HOL/ex/Termination.thy for examples.
  4226 
  4227 * Simplifier: simproc for let expressions now unfolds if bound
  4228 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4229 
  4230 * Method "arith": Linear arithmetic now ignores all inequalities when
  4231 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4232 
  4233 * New attribute "arith" for facts that should always be used
  4234 automatically by arithmetic. It is intended to be used locally in
  4235 proofs, e.g.
  4236 
  4237   assumes [arith]: "x > 0"
  4238 
  4239 Global usage is discouraged because of possible performance impact.
  4240 
  4241 * New classes "top" and "bot" with corresponding operations "top" and
  4242 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4243 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4244 
  4245 * Changed definition lemma "less_fun_def" in order to provide an
  4246 instance for preorders on functions; use lemma "less_le" instead.
  4247 INCOMPATIBILITY.
  4248 
  4249 * Theory Orderings: class "wellorder" moved here, with explicit
  4250 induction rule "less_induct" as assumption.  For instantiation of
  4251 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4252 INCOMPATIBILITY.
  4253 
  4254 * Theory Orderings: added class "preorder" as superclass of "order".
  4255 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4256 etc. slightly changed.  Some theorems named order_class.* now named
  4257 preorder_class.*.
  4258 
  4259 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4260 "diag" to "Id_on".
  4261 
  4262 * Theory Finite_Set: added a new fold combinator of type
  4263 
  4264   ('a => 'b => 'b) => 'b => 'a set => 'b
  4265 
  4266 Occasionally this is more convenient than the old fold combinator
  4267 which is now defined in terms of the new one and renamed to
  4268 fold_image.
  4269 
  4270 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4271 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4272 extended with further lemmas!).  At the moment both still exist but
  4273 the former will disappear at some point.
  4274 
  4275 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4276 class recpower.  Type-specific simp rules for various recpower types
  4277 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4278 
  4279 rat_power_0    -> power_0
  4280 rat_power_Suc  -> power_Suc
  4281 realpow_0      -> power_0
  4282 realpow_Suc    -> power_Suc
  4283 complexpow_0   -> power_0
  4284 complexpow_Suc -> power_Suc
  4285 power_poly_0   -> power_0
  4286 power_poly_Suc -> power_Suc
  4287 
  4288 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4289 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4290 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4291 of lemmas from Divides has been generalized from nat to class
  4292 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4293 renames resulting from duplicate elimination:
  4294 
  4295     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4296     zero_dvd_iff ~>         dvd_0_left_iff
  4297     dvd_0 ~>                dvd_0_right
  4298     DIVISION_BY_ZERO_DIV ~> div_by_0
  4299     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4300     mult_div ~>             div_mult_self2_is_id
  4301     mult_mod ~>             mod_mult_self2_is_0
  4302 
  4303 * Theory IntDiv: removed many lemmas that are instances of class-based
  4304 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4305 rename old lemmas as follows:
  4306 
  4307 dvd_diff               -> nat_dvd_diff
  4308 dvd_zminus_iff         -> dvd_minus_iff
  4309 mod_add1_eq            -> mod_add_eq
  4310 mod_mult1_eq           -> mod_mult_right_eq
  4311 mod_mult1_eq'          -> mod_mult_left_eq
  4312 mod_mult_distrib_mod   -> mod_mult_eq
  4313 nat_mod_add_left_eq    -> mod_add_left_eq
  4314 nat_mod_add_right_eq   -> mod_add_right_eq
  4315 nat_mod_div_trivial    -> mod_div_trivial
  4316 nat_mod_mod_trivial    -> mod_mod_trivial
  4317 zdiv_zadd_self1        -> div_add_self1
  4318 zdiv_zadd_self2        -> div_add_self2
  4319 zdiv_zmult_self1       -> div_mult_self2_is_id
  4320 zdiv_zmult_self2       -> div_mult_self1_is_id
  4321 zdvd_triv_left         -> dvd_triv_left
  4322 zdvd_triv_right        -> dvd_triv_right
  4323 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4324 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4325 zmod_zadd_left_eq      -> mod_add_left_eq
  4326 zmod_zadd_right_eq     -> mod_add_right_eq
  4327 zmod_zadd_self1        -> mod_add_self1
  4328 zmod_zadd_self2        -> mod_add_self2
  4329 zmod_zadd1_eq          -> mod_add_eq
  4330 zmod_zdiff1_eq         -> mod_diff_eq
  4331 zmod_zdvd_zmod         -> mod_mod_cancel
  4332 zmod_zmod_cancel       -> mod_mod_cancel
  4333 zmod_zmult_self1       -> mod_mult_self2_is_0
  4334 zmod_zmult_self2       -> mod_mult_self1_is_0
  4335 zmod_1                 -> mod_by_1
  4336 zdiv_1                 -> div_by_1
  4337 zdvd_abs1              -> abs_dvd_iff
  4338 zdvd_abs2              -> dvd_abs_iff
  4339 zdvd_refl              -> dvd_refl
  4340 zdvd_trans             -> dvd_trans
  4341 zdvd_zadd              -> dvd_add
  4342 zdvd_zdiff             -> dvd_diff
  4343 zdvd_zminus_iff        -> dvd_minus_iff
  4344 zdvd_zminus2_iff       -> minus_dvd_iff
  4345 zdvd_zmultD            -> dvd_mult_right
  4346 zdvd_zmultD2           -> dvd_mult_left
  4347 zdvd_zmult_mono        -> mult_dvd_mono
  4348 zdvd_0_right           -> dvd_0_right
  4349 zdvd_0_left            -> dvd_0_left_iff
  4350 zdvd_1_left            -> one_dvd
  4351 zminus_dvd_iff         -> minus_dvd_iff
  4352 
  4353 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4354 
  4355 * The real numbers offer decimal input syntax: 12.34 is translated
  4356 into 1234/10^2. This translation is not reversed upon output.
  4357 
  4358 * Theory Library/Polynomial defines an abstract type 'a poly of
  4359 univariate polynomials with coefficients of type 'a.  In addition to
  4360 the standard ring operations, it also supports div and mod.  Code
  4361 generation is also supported, using list-style constructors.
  4362 
  4363 * Theory Library/Inner_Product defines a class of real_inner for real
  4364 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4365 => real.  Class real_inner is a subclass of real_normed_vector from
  4366 theory RealVector.
  4367 
  4368 * Theory Library/Product_Vector provides instances for the product
  4369 type 'a * 'b of several classes from RealVector and Inner_Product.
  4370 Definitions of addition, subtraction, scalar multiplication, norms,
  4371 and inner products are included.
  4372 
  4373 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4374 addition to the field operations, numerals and case syntax are also
  4375 supported.
  4376 
  4377 * Theory Library/Diagonalize provides constructive version of Cantor's
  4378 first diagonalization argument.
  4379 
  4380 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4381 zlcm (for int); carried together from various gcd/lcm developements in
  4382 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4383 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4384 may recover tupled syntax as follows:
  4385 
  4386     hide (open) const gcd
  4387     abbreviation gcd where
  4388       "gcd == (%(a, b). GCD.gcd a b)"
  4389     notation (output)
  4390       GCD.gcd ("gcd '(_, _')")
  4391 
  4392 The same works for lcm, zgcd, zlcm.
  4393 
  4394 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4395 instantiations for algebraic structures.  Removed some duplicate
  4396 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4397 
  4398 * ML antiquotation @{code} takes a constant as argument and generates
  4399 corresponding code in background and inserts name of the corresponding
  4400 resulting ML value/function/datatype constructor binding in place.
  4401 All occurrences of @{code} with a single ML block are generated
  4402 simultaneously.  Provides a generic and safe interface for
  4403 instrumentalizing code generation.  See
  4404 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4405 In future you ought to refrain from ad-hoc compiling generated SML
  4406 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4407 cannot refer to constants for which user-defined serializations are
  4408 set.  Refer to the corresponding ML counterpart directly in that
  4409 cases.
  4410 
  4411 * Command 'rep_datatype': instead of theorem names the command now
  4412 takes a list of terms denoting the constructors of the type to be
  4413 represented as datatype.  The characteristic theorems have to be
  4414 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4415 have disappeared in favour of existing ones:
  4416 
  4417     unit_induct                 ~> unit.induct
  4418     prod_induct                 ~> prod.induct
  4419     sum_induct                  ~> sum.induct
  4420     Suc_Suc_eq                  ~> nat.inject
  4421     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4422 
  4423 
  4424 *** HOL-Algebra ***
  4425 
  4426 * New locales for orders and lattices where the equivalence relation
  4427 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4428 locales use a record structure with field eq for the equivalence.
  4429 
  4430 * New theory of factorial domains.
  4431 
  4432 * Units_l_inv and Units_r_inv are now simp rules by default.
  4433 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4434 and/or r_inv will now also require deletion of these lemmas.
  4435 
  4436 * Renamed the following theorems, INCOMPATIBILITY:
  4437 
  4438 UpperD ~> Upper_memD
  4439 LowerD ~> Lower_memD
  4440 least_carrier ~> least_closed
  4441 greatest_carrier ~> greatest_closed
  4442 greatest_Lower_above ~> greatest_Lower_below
  4443 one_zero ~> carrier_one_zero
  4444 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4445 
  4446 
  4447 *** HOL-Nominal ***
  4448 
  4449 * Nominal datatypes can now contain type-variables.
  4450 
  4451 * Commands 'nominal_inductive' and 'equivariance' work with local
  4452 theory targets.
  4453 
  4454 * Nominal primrec can now works with local theory targets and its
  4455 specification syntax now conforms to the general format as seen in
  4456 'inductive' etc.
  4457 
  4458 * Method "perm_simp" honours the standard simplifier attributes
  4459 (no_asm), (no_asm_use) etc.
  4460 
  4461 * The new predicate #* is defined like freshness, except that on the
  4462 left hand side can be a set or list of atoms.
  4463 
  4464 * Experimental command 'nominal_inductive2' derives strong induction
  4465 principles for inductive definitions.  In contrast to
  4466 'nominal_inductive', which can only deal with a fixed number of
  4467 binders, it can deal with arbitrary expressions standing for sets of
  4468 atoms to be avoided.  The only inductive definition we have at the
  4469 moment that needs this generalisation is the typing rule for Lets in
  4470 the algorithm W:
  4471 
  4472  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4473  -----------------------------------------------------------------
  4474          Gamma |- Let x be t1 in t2 : T2
  4475 
  4476 In this rule one wants to avoid all the binders that are introduced by
  4477 "close Gamma T1".  We are looking for other examples where this
  4478 feature might be useful.  Please let us know.
  4479 
  4480 
  4481 *** HOLCF ***
  4482 
  4483 * Reimplemented the simplification procedure for proving continuity
  4484 subgoals.  The new simproc is extensible; users can declare additional
  4485 continuity introduction rules with the attribute [cont2cont].
  4486 
  4487 * The continuity simproc now uses a different introduction rule for
  4488 solving continuity subgoals on terms with lambda abstractions.  In
  4489 some rare cases the new simproc may fail to solve subgoals that the
  4490 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4491 Potential INCOMPATIBILITY.
  4492 
  4493 * Command 'fixrec': specification syntax now conforms to the general
  4494 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4495 examples.  INCOMPATIBILITY.
  4496 
  4497 
  4498 *** ZF ***
  4499 
  4500 * Proof of Zorn's Lemma for partial orders.
  4501 
  4502 
  4503 *** ML ***
  4504 
  4505 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4506 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4507 depends on multithreading, so timouts will not work before Poly/ML
  4508 5.2.1!
  4509 
  4510 * High-level support for concurrent ML programming, see
  4511 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4512 particularly convenient to organize independent functional
  4513 computations.  The concept of "synchronized variables" provides a
  4514 higher-order interface for components with shared state, avoiding the
  4515 delicate details of mutexes and condition variables.  (Requires
  4516 Poly/ML 5.2.1 or later.)
  4517 
  4518 * ML bindings produced via Isar commands are stored within the Isar
  4519 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4520 become thread-safe and work with undo as expected (concerning
  4521 top-level bindings, not side-effects on global references).
  4522 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4523 compiler at runtime; really global bindings need to be given outside a
  4524 theory.  (Requires Poly/ML 5.2 or later.)
  4525 
  4526 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4527 context.  Top-level ML bindings are stored within the proof context in
  4528 a purely sequential fashion, disregarding the nested proof structure.
  4529 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4530 proof.  (Requires Poly/ML 5.2 or later.)
  4531 
  4532 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4533 and Method.setup, as well as Isar commands 'attribute_setup' and
  4534 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4535 existing code accordingly, or use plain 'setup' together with old
  4536 Method.add_method.
  4537 
  4538 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4539 to 'a -> thm, while results are always tagged with an authentic oracle
  4540 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4541 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4542 accordingly.  Note that extra performance may be gained by producing
  4543 the cterm carefully, avoiding slow Thm.cterm_of.
  4544 
  4545 * Simplified interface for defining document antiquotations via
  4546 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4547 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4548 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4549 examples.
  4550 
  4551 * More systematic treatment of long names, abstract name bindings, and
  4552 name space operations.  Basic operations on qualified names have been
  4553 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4554 Long_Name.append.  Old type bstring has been mostly replaced by
  4555 abstract type binding (see structure Binding), which supports precise
  4556 qualification by packages and local theory targets, as well as proper
  4557 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4558 bstring values into Binding.name, or better pass through abstract
  4559 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4560 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4561 
  4562 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4563 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4564 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4565 
  4566 * Disposed old type and term read functions (Sign.read_def_typ,
  4567 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4568 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4569 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4570 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4571 resort for legacy applications.
  4572 
  4573 * Disposed old declarations, tactics, tactic combinators that refer to
  4574 the simpset or claset of an implicit theory (such as Addsimps,
  4575 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4576 embedded ML text, or local_simpset_of with a proper context passed as
  4577 explicit runtime argument.
  4578 
  4579 * Rules and tactics that read instantiations (read_instantiate,
  4580 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4581 context, which is required for parsing and type-checking.  Moreover,
  4582 the variables are specified as plain indexnames, not string encodings
  4583 thereof.  INCOMPATIBILITY.
  4584 
  4585 * Generic Toplevel.add_hook interface allows to analyze the result of
  4586 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4587 for theorem dependency output of transactions resulting in a new
  4588 theory state.
  4589 
  4590 * ML antiquotations: block-structured compilation context indicated by
  4591 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4592 
  4593   @{binding name}                         - basic name binding
  4594   @{let ?pat = term}                      - term abbreviation (HO matching)
  4595   @{note name = fact}                     - fact abbreviation
  4596   @{thm fact}                             - singleton fact (with attributes)
  4597   @{thms fact}                            - general fact (with attributes)
  4598   @{lemma prop by method}                 - singleton goal
  4599   @{lemma prop by meth1 meth2}            - singleton goal
  4600   @{lemma prop1 ... propN by method}      - general goal
  4601   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4602   @{lemma (open) ...}                     - open derivation
  4603 
  4604 
  4605 *** System ***
  4606 
  4607 * The Isabelle "emacs" tool provides a specific interface to invoke
  4608 Proof General / Emacs, with more explicit failure if that is not
  4609 installed (the old isabelle-interface script silently falls back on
  4610 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4611 installation location of the Proof General distribution.
  4612 
  4613 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4614 the Isabelle process into a JVM/Scala application.  See
  4615 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4616 process wrapper has been discontinued.)
  4617 
  4618 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4619 
  4620 * Various status messages (with exact source position information) are
  4621 emitted, if proper markup print mode is enabled.  This allows
  4622 user-interface components to provide detailed feedback on internal
  4623 prover operations.
  4624 
  4625 
  4626 
  4627 New in Isabelle2008 (June 2008)
  4628 -------------------------------
  4629 
  4630 *** General ***
  4631 
  4632 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4633 and updated, with formally checked references as hyperlinks.
  4634 
  4635 * Theory loader: use_thy (and similar operations) no longer set the
  4636 implicit ML context, which was occasionally hard to predict and in
  4637 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4638 provides a proper context already.
  4639 
  4640 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4641 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4642 'use' within a theory file will do the job.
  4643 
  4644 * Name space merge now observes canonical order, i.e. the second space
  4645 is inserted into the first one, while existing entries in the first
  4646 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4647 swap theory imports.
  4648 
  4649 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4650 INCOMPATIBILITY in identifier syntax etc.
  4651 
  4652 * Outer syntax: string tokens no longer admit escaped white space,
  4653 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4654 white space without escapes.
  4655 
  4656 * Outer syntax: string tokens may contain arbitrary character codes
  4657 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4658 "foo_bar".
  4659 
  4660 
  4661 *** Pure ***
  4662 
  4663 * Context-dependent token translations.  Default setup reverts locally
  4664 fixed variables, and adds hilite markup for undeclared frees.
  4665 
  4666 * Unused theorems can be found using the new command 'unused_thms'.
  4667 There are three ways of invoking it:
  4668 
  4669 (1) unused_thms
  4670      Only finds unused theorems in the current theory.
  4671 
  4672 (2) unused_thms thy_1 ... thy_n -
  4673      Finds unused theorems in the current theory and all of its ancestors,
  4674      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4675 
  4676 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4677      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4678      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4679      their ancestors.
  4680 
  4681 In order to increase the readability of the list produced by
  4682 unused_thms, theorems that have been created by a particular instance
  4683 of a theory command such as 'inductive' or 'function' are considered
  4684 to belong to the same "group", meaning that if at least one theorem in
  4685 this group is used, the other theorems in the same group are no longer
  4686 reported as unused.  Moreover, if all theorems in the group are
  4687 unused, only one theorem in the group is displayed.
  4688 
  4689 Note that proof objects have to be switched on in order for
  4690 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4691 usually the case when using Proof General with the default settings).
  4692 
  4693 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4694 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4695 duplicate fact bindings, or even accidental fact duplications.  Note
  4696 that tools may maintain dynamically scoped facts systematically, using
  4697 PureThy.add_thms_dynamic.
  4698 
  4699 * Command 'hide' now allows to hide from "fact" name space as well.
  4700 
  4701 * Eliminated destructive theorem database, simpset, claset, and
  4702 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4703 update of theories within ML code.
  4704 
  4705 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4706 INCOMPATIBILITY, object-logics depending on former Pure require
  4707 additional setup PureThy.old_appl_syntax_setup; object-logics
  4708 depending on former CPure need to refer to Pure.
  4709 
  4710 * Commands 'use' and 'ML' are now purely functional, operating on
  4711 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4712 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4713 INCOMPATIBILITY.
  4714 
  4715 * Command 'setup': discontinued implicit version with ML reference.
  4716 
  4717 * Instantiation target allows for simultaneous specification of class
  4718 instance operations together with an instantiation proof.
  4719 Type-checking phase allows to refer to class operations uniformly.
  4720 See src/HOL/Complex/Complex.thy for an Isar example and
  4721 src/HOL/Library/Eval.thy for an ML example.
  4722 
  4723 * Indexing of literal facts: be more serious about including only
  4724 facts from the visible specification/proof context, but not the
  4725 background context (locale etc.).  Affects `prop` notation and method
  4726 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4727 situations.
  4728 
  4729 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4730 "(open)" option, which used to expose internal bound variables to the
  4731 proof text.
  4732 
  4733 * Isar statements: removed obsolete case "rule_context".
  4734 INCOMPATIBILITY, better use explicit fixes/assumes.
  4735 
  4736 * Locale proofs: default proof step now includes 'unfold_locales';
  4737 hence 'proof' without argument may be used to unfold locale
  4738 predicates.
  4739 
  4740 
  4741 *** Document preparation ***
  4742 
  4743 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4744 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4745 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4746 spot); renamed link color from "darkblue" to "linkcolor" (default
  4747 value unchanged, can be redefined via \definecolor); no longer sets
  4748 "a4paper" option (unnecessary or even intrusive).
  4749 
  4750 * Antiquotation @{lemma A method} proves proposition A by the given
  4751 method (either a method name or a method name plus (optional) method
  4752 arguments in parentheses) and prints A just like @{prop A}.
  4753 
  4754 
  4755 *** HOL ***
  4756 
  4757 * New primrec package.  Specification syntax conforms in style to
  4758 definition/function/....  No separate induction rule is provided.  The
  4759 "primrec" command distinguishes old-style and new-style specifications
  4760 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4761 When adjusting theories, beware: constants stemming from new-style
  4762 primrec specifications have authentic syntax.
  4763 
  4764 * Metis prover is now an order of magnitude faster, and also works
  4765 with multithreading.
  4766 
  4767 * Metis: the maximum number of clauses that can be produced from a
  4768 theorem is now given by the attribute max_clauses.  Theorems that
  4769 exceed this number are ignored, with a warning printed.
  4770 
  4771 * Sledgehammer no longer produces structured proofs by default. To
  4772 enable, declare [[sledgehammer_full = true]].  Attributes
  4773 reconstruction_modulus, reconstruction_sorts renamed
  4774 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4775 
  4776 * Method "induct_scheme" derives user-specified induction rules
  4777 from well-founded induction and completeness of patterns. This factors
  4778 out some operations that are done internally by the function package
  4779 and makes them available separately.  See
  4780 src/HOL/ex/Induction_Scheme.thy for examples.
  4781 
  4782 * More flexible generation of measure functions for termination
  4783 proofs: Measure functions can be declared by proving a rule of the
  4784 form "is_measure f" and giving it the [measure_function] attribute.
  4785 The "is_measure" predicate is logically meaningless (always true), and
  4786 just guides the heuristic.  To find suitable measure functions, the
  4787 termination prover sets up the goal "is_measure ?f" of the appropriate
  4788 type and generates all solutions by prolog-style backwards proof using
  4789 the declared rules.
  4790 
  4791 This setup also deals with rules like
  4792 
  4793   "is_measure f ==> is_measure (list_size f)"
  4794 
  4795 which accommodates nested datatypes that recurse through lists.
  4796 Similar rules are predeclared for products and option types.
  4797 
  4798 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4799 
  4800   INCOMPATIBILITIES:
  4801 
  4802   - Definitions of overloaded constants on sets have to be replaced by
  4803     definitions on => and bool.
  4804 
  4805   - Some definitions of overloaded operators on sets can now be proved
  4806     using the definitions of the operators on => and bool.  Therefore,
  4807     the following theorems have been renamed:
  4808 
  4809       subset_def   -> subset_eq
  4810       psubset_def  -> psubset_eq
  4811       set_diff_def -> set_diff_eq
  4812       Compl_def    -> Compl_eq
  4813       Sup_set_def  -> Sup_set_eq
  4814       Inf_set_def  -> Inf_set_eq
  4815       sup_set_def  -> sup_set_eq
  4816       inf_set_def  -> inf_set_eq
  4817 
  4818   - Due to the incompleteness of the HO unification algorithm, some
  4819     rules such as subst may require manual instantiation, if some of
  4820     the unknowns in the rule is a set.
  4821 
  4822   - Higher order unification and forward proofs:
  4823     The proof pattern
  4824 
  4825       have "P (S::'a set)" <...>
  4826       then have "EX S. P S" ..
  4827 
  4828     no longer works (due to the incompleteness of the HO unification
  4829     algorithm) and must be replaced by the pattern
  4830 
  4831       have "EX S. P S"
  4832       proof
  4833         show "P S" <...>
  4834       qed
  4835 
  4836   - Calculational reasoning with subst (or similar rules):
  4837     The proof pattern
  4838 
  4839       have "P (S::'a set)" <...>
  4840       also have "S = T" <...>
  4841       finally have "P T" .
  4842 
  4843     no longer works (for similar reasons as the previous example) and
  4844     must be replaced by something like
  4845 
  4846       have "P (S::'a set)" <...>
  4847       moreover have "S = T" <...>
  4848       ultimately have "P T" by simp
  4849 
  4850   - Tactics or packages written in ML code:
  4851     Code performing pattern matching on types via
  4852 
  4853       Type ("set", [T]) => ...
  4854 
  4855     must be rewritten. Moreover, functions like strip_type or
  4856     binder_types no longer return the right value when applied to a
  4857     type of the form
  4858 
  4859       T1 => ... => Tn => U => bool
  4860 
  4861     rather than
  4862 
  4863       T1 => ... => Tn => U set
  4864 
  4865 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4866 Wellfounded_Relations to theory Wellfounded.
  4867 
  4868 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4869 
  4870 * Class finite no longer treats UNIV as class parameter.  Use class
  4871 enum from theory Library/Enum instead to achieve a similar effect.
  4872 INCOMPATIBILITY.
  4873 
  4874 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4875 and "Cons".  INCOMPATIBILITY.
  4876 
  4877 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4878 Potential INCOMPATIBILITY.
  4879 
  4880 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4881 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4882 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4883 
  4884 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4885 the same name in theory Orderings:
  4886 
  4887   less_trans
  4888   less_linear
  4889   le_imp_less_or_eq
  4890   le_less_trans
  4891   less_le_trans
  4892   less_not_sym
  4893   less_asym
  4894 
  4895 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4896 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4897 and different variable names.
  4898 
  4899 * Library/Option_ord.thy: Canonical order on option type.
  4900 
  4901 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4902 finite maps.
  4903 
  4904 * Library/Countable.thy: Type class for countable types.
  4905 
  4906 * Theory Int: The representation of numerals has changed.  The infix
  4907 operator BIT and the bit datatype with constructors B0 and B1 have
  4908 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4909 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4910 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4911 accordingly.
  4912 
  4913 * Theory Nat: definition of <= and < on natural numbers no longer
  4914 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4915 le_def and less_def have disappeared.  Consider lemmas not_less
  4916 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4917 
  4918 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4919 (whose purpose mainly is for various fold_set functionals) have been
  4920 abandoned in favor of the existing algebraic classes
  4921 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4922 lower_semilattice (resp. upper_semilattice) and linorder.
  4923 INCOMPATIBILITY.
  4924 
  4925 * Theory Transitive_Closure: induct and cases rules now declare proper
  4926 case_names ("base" and "step").  INCOMPATIBILITY.
  4927 
  4928 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4929 lattices.  The form set-specific version is available as
  4930 Inductive.lfp_ordinal_induct_set.
  4931 
  4932 * Renamed theorems "power.simps" to "power_int.simps".
  4933 INCOMPATIBILITY.
  4934 
  4935 * Class semiring_div provides basic abstract properties of semirings
  4936 with division and modulo operations.  Subsumes former class dvd_mod.
  4937 
  4938 * Merged theories IntDef, Numeral and IntArith into unified theory
  4939 Int.  INCOMPATIBILITY.
  4940 
  4941 * Theory Library/Code_Index: type "index" now represents natural
  4942 numbers rather than integers.  INCOMPATIBILITY.
  4943 
  4944 * New class "uminus" with operation "uminus" (split of from class
  4945 "minus" which now only has operation "minus", binary).
  4946 INCOMPATIBILITY.
  4947 
  4948 * Constants "card", "internal_split", "option_map" now with authentic
  4949 syntax.  INCOMPATIBILITY.
  4950 
  4951 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4952 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4953 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4954 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4955 Sup_set_def, le_def, less_def, option_map_def now with object
  4956 equality.  INCOMPATIBILITY.
  4957 
  4958 * Records. Removed K_record, and replaced it by pure lambda term
  4959 %x. c. The simplifier setup is now more robust against eta expansion.
  4960 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4961 
  4962 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4963 
  4964 * Library/ListVector: new theory of arithmetic vector operations.
  4965 
  4966 * Library/Order_Relation: new theory of various orderings as sets of
  4967 pairs.  Defines preorders, partial orders, linear orders and
  4968 well-orders on sets and on types.
  4969 
  4970 
  4971 *** ZF ***
  4972 
  4973 * Renamed some theories to allow to loading both ZF and HOL in the
  4974 same session:
  4975 
  4976   Datatype  -> Datatype_ZF
  4977   Inductive -> Inductive_ZF
  4978   Int       -> Int_ZF
  4979   IntDiv    -> IntDiv_ZF
  4980   Nat       -> Nat_ZF
  4981   List      -> List_ZF
  4982   Main      -> Main_ZF
  4983 
  4984 INCOMPATIBILITY: ZF theories that import individual theories below
  4985 Main might need to be adapted.  Regular theory Main is still
  4986 available, as trivial extension of Main_ZF.
  4987 
  4988 
  4989 *** ML ***
  4990 
  4991 * ML within Isar: antiquotation @{const name} or @{const
  4992 name(typargs)} produces statically-checked Const term.
  4993 
  4994 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4995 (of the same name).  Removed obsolete print command.
  4996 
  4997 * Removed obsolete "use_legacy_bindings" function.
  4998 
  4999 * The ``print mode'' is now a thread-local value derived from a global
  5000 template (the former print_mode reference), thus access becomes
  5001 non-critical.  The global print_mode reference is for session
  5002 management only; user-code should use print_mode_value,
  5003 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  5004 
  5005 * Functions system/system_out provide a robust way to invoke external
  5006 shell commands, with propagation of interrupts (requires Poly/ML
  5007 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  5008 
  5009 
  5010 *** System ***
  5011 
  5012 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  5013 in accordance with Proof General 3.7, which prefers GNU emacs.
  5014 
  5015 * isatool tty runs Isabelle process with plain tty interaction;
  5016 optional line editor may be specified via ISABELLE_LINE_EDITOR
  5017 setting, the default settings attempt to locate "ledit" and "rlwrap".
  5018 
  5019 * isatool browser now works with Cygwin as well, using general
  5020 "javapath" function defined in Isabelle process environment.
  5021 
  5022 * YXML notation provides a simple and efficient alternative to
  5023 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  5024 isatool yxml as described in the Isabelle system manual.
  5025 
  5026 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  5027 provides general wrapper for managing an Isabelle process in a robust
  5028 fashion, with ``cooked'' output from stdin/stderr.
  5029 
  5030 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  5031 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  5032 
  5033 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  5034 way of changing the user's settings is via
  5035 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  5036 script.
  5037 
  5038 * Multithreading.max_threads := 0 refers to the number of actual CPU
  5039 cores of the underlying machine, which is a good starting point for
  5040 optimal performance tuning.  The corresponding usedir option -M allows
  5041 "max" as an alias for "0".  WARNING: does not work on certain versions
  5042 of Mac OS (with Poly/ML 5.1).
  5043 
  5044 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  5045 adverse effect of Isabelle flooding interactive front-ends (notably
  5046 ProofGeneral / XEmacs).
  5047 
  5048 
  5049 
  5050 New in Isabelle2007 (November 2007)
  5051 -----------------------------------
  5052 
  5053 *** General ***
  5054 
  5055 * More uniform information about legacy features, notably a
  5056 warning/error of "Legacy feature: ...", depending on the state of the
  5057 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  5058 legacy features will disappear eventually.
  5059 
  5060 * Theory syntax: the header format ``theory A = B + C:'' has been
  5061 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  5062 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  5063 
  5064 * Theory syntax: the old non-Isar theory file format has been
  5065 discontinued altogether.  Note that ML proof scripts may still be used
  5066 with Isar theories; migration is usually quite simple with the ML
  5067 function use_legacy_bindings.  INCOMPATIBILITY.
  5068 
  5069 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  5070 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  5071 quotes.
  5072 
  5073 * Theory loader: be more serious about observing the static theory
  5074 header specifications (including optional directories), but not the
  5075 accidental file locations of previously successful loads.  The strict
  5076 update policy of former update_thy is now already performed by
  5077 use_thy, so the former has been removed; use_thys updates several
  5078 theories simultaneously, just as 'imports' within a theory header
  5079 specification, but without merging the results.  Potential
  5080 INCOMPATIBILITY: may need to refine theory headers and commands
  5081 ROOT.ML which depend on load order.
  5082 
  5083 * Theory loader: optional support for content-based file
  5084 identification, instead of the traditional scheme of full physical
  5085 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  5086 (cf. the system manual).  The new scheme allows to work with
  5087 non-finished theories in persistent session images, such that source
  5088 files may be moved later on without requiring reloads.
  5089 
  5090 * Theory loader: old-style ML proof scripts being *attached* to a thy
  5091 file (with the same base name as the theory) are considered a legacy
  5092 feature, which will disappear eventually. Even now, the theory loader
  5093 no longer maintains dependencies on such files.
  5094 
  5095 * Syntax: the scope for resolving ambiguities via type-inference is
  5096 now limited to individual terms, instead of whole simultaneous
  5097 specifications as before. This greatly reduces the complexity of the
  5098 syntax module and improves flexibility by separating parsing and
  5099 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  5100 'fixes' etc.) are required in rare situations.
  5101 
  5102 * Syntax: constants introduced by new-style packages ('definition',
  5103 'abbreviation' etc.) are passed through the syntax module in
  5104 ``authentic mode''. This means that associated mixfix annotations
  5105 really stick to such constants, independently of potential name space
  5106 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  5107 trees are represented slightly differently, may need to adapt syntax
  5108 translations accordingly. Use CONST marker in 'translations' and
  5109 @{const_syntax} antiquotation in 'parse_translation' etc.
  5110 
  5111 * Legacy goal package: reduced interface to the bare minimum required
  5112 to keep existing proof scripts running.  Most other user-level
  5113 functions are now part of the OldGoals structure, which is *not* open
  5114 by default (consider isatool expandshort before open OldGoals).
  5115 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  5116 these tend to cause confusion about the actual goal (!) context being
  5117 used here, which is not necessarily the same as the_context().
  5118 
  5119 * Command 'find_theorems': supports "*" wild-card in "name:"
  5120 criterion; "with_dups" option.  Certain ProofGeneral versions might
  5121 support a specific search form (see ProofGeneral/CHANGES).
  5122 
  5123 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  5124 by default, which means that "prems" (and also "fixed variables") are
  5125 suppressed from proof state output.  Note that the ProofGeneral
  5126 settings mechanism allows to change and save options persistently, but
  5127 older versions of Isabelle will fail to start up if a negative prems
  5128 limit is imposed.
  5129 
  5130 * Local theory targets may be specified by non-nested blocks of
  5131 ``context/locale/class ... begin'' followed by ``end''.  The body may
  5132 contain definitions, theorems etc., including any derived mechanism
  5133 that has been implemented on top of these primitives.  This concept
  5134 generalizes the existing ``theorem (in ...)'' towards more versatility
  5135 and scalability.
  5136 
  5137 * Proof General interface: proper undo of final 'end' command;
  5138 discontinued Isabelle/classic mode (ML proof scripts).
  5139 
  5140 
  5141 *** Document preparation ***
  5142 
  5143 * Added antiquotation @{theory name} which prints the given name,
  5144 after checking that it refers to a valid ancestor theory in the
  5145 current context.
  5146 
  5147 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5148 check the given source text as ML type/structure, printing verbatim.
  5149 
  5150 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5151 "c args == rhs" given in the current context.  (Any number of
  5152 arguments may be given on the LHS.)
  5153 
  5154 
  5155 *** Pure ***
  5156 
  5157 * The 'class' package offers a combination of axclass and locale to
  5158 achieve Haskell-like type classes in Isabelle.  Definitions and
  5159 theorems within a class context produce both relative results (with
  5160 implicit parameters according to the locale context), and polymorphic
  5161 constants with qualified polymorphism (according to the class
  5162 context).  Within the body context of a 'class' target, a separate
  5163 syntax layer ("user space type system") takes care of converting
  5164 between global polymorphic consts and internal locale representation.
  5165 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5166 "isatool doc classes" provides a tutorial.
  5167 
  5168 * Generic code generator framework allows to generate executable
  5169 code for ML and Haskell (including Isabelle classes).  A short usage
  5170 sketch:
  5171 
  5172     internal compilation:
  5173         export_code <list of constants (term syntax)> in SML
  5174     writing SML code to a file:
  5175         export_code <list of constants (term syntax)> in SML <filename>
  5176     writing OCaml code to a file:
  5177         export_code <list of constants (term syntax)> in OCaml <filename>
  5178     writing Haskell code to a bunch of files:
  5179         export_code <list of constants (term syntax)> in Haskell <filename>
  5180 
  5181     evaluating closed propositions to True/False using code generation:
  5182         method ``eval''
  5183 
  5184 Reasonable default setup of framework in HOL.
  5185 
  5186 Theorem attributs for selecting and transforming function equations theorems:
  5187 
  5188     [code fun]:        select a theorem as function equation for a specific constant
  5189     [code fun del]:    deselect a theorem as function equation for a specific constant
  5190     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5191     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5192 
  5193 User-defined serializations (target in {SML, OCaml, Haskell}):
  5194 
  5195     code_const <and-list of constants (term syntax)>
  5196       {(target) <and-list of const target syntax>}+
  5197 
  5198     code_type <and-list of type constructors>
  5199       {(target) <and-list of type target syntax>}+
  5200 
  5201     code_instance <and-list of instances>
  5202       {(target)}+
  5203         where instance ::= <type constructor> :: <class>
  5204 
  5205     code_class <and_list of classes>
  5206       {(target) <and-list of class target syntax>}+
  5207         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5208 
  5209 code_instance and code_class only are effective to target Haskell.
  5210 
  5211 For example usage see src/HOL/ex/Codegenerator.thy and
  5212 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5213 generation from Isabelle/HOL theories is available via "isatool doc
  5214 codegen".
  5215 
  5216 * Code generator: consts in 'consts_code' Isar commands are now
  5217 referred to by usual term syntax (including optional type
  5218 annotations).
  5219 
  5220 * Command 'no_translations' removes translation rules from theory
  5221 syntax.
  5222 
  5223 * Overloaded definitions are now actually checked for acyclic
  5224 dependencies.  The overloading scheme is slightly more general than
  5225 that of Haskell98, although Isabelle does not demand an exact
  5226 correspondence to type class and instance declarations.
  5227 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5228 exotic versions of overloading -- at the discretion of the user!
  5229 
  5230 Polymorphic constants are represented via type arguments, i.e. the
  5231 instantiation that matches an instance against the most general
  5232 declaration given in the signature.  For example, with the declaration
  5233 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5234 as c(nat).  Overloading is essentially simultaneous structural
  5235 recursion over such type arguments.  Incomplete specification patterns
  5236 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5237 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5238 Command 'print_theory' outputs the normalized system of recursive
  5239 equations, see section "definitions".
  5240 
  5241 * Configuration options are maintained within the theory or proof
  5242 context (with name and type bool/int/string), providing a very simple
  5243 interface to a poor-man's version of general context data.  Tools may
  5244 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5245 these values using Config.get etc.  Users may change options via an
  5246 associated attribute of the same name.  This form of context
  5247 declaration works particularly well with commands 'declare' or
  5248 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5249 very easy to avoid global references, which would not observe Isar
  5250 toplevel undo/redo and fail to work with multithreading.
  5251 
  5252 Various global ML references of Pure and HOL have been turned into
  5253 configuration options:
  5254 
  5255   Unify.search_bound		unify_search_bound
  5256   Unify.trace_bound		unify_trace_bound
  5257   Unify.trace_simp		unify_trace_simp
  5258   Unify.trace_types		unify_trace_types
  5259   Simplifier.simp_depth_limit	simp_depth_limit
  5260   Blast.depth_limit		blast_depth_limit
  5261   DatatypeProp.dtK		datatype_distinctness_limit
  5262   fast_arith_neq_limit  	fast_arith_neq_limit
  5263   fast_arith_split_limit	fast_arith_split_limit
  5264 
  5265 * Named collections of theorems may be easily installed as context
  5266 data using the functor NamedThmsFun (see also
  5267 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5268 attributes; there is also a toplevel print command.  This facility is
  5269 just a common case of general context data, which is the preferred way
  5270 for anything more complex than just a list of facts in canonical
  5271 order.
  5272 
  5273 * Isar: command 'declaration' augments a local theory by generic
  5274 declaration functions written in ML.  This enables arbitrary content
  5275 being added to the context, depending on a morphism that tells the
  5276 difference of the original declaration context wrt. the application
  5277 context encountered later on.
  5278 
  5279 * Isar: proper interfaces for simplification procedures.  Command
  5280 'simproc_setup' declares named simprocs (with match patterns, and body
  5281 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5282 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5283 
  5284 * Isar: an extra pair of brackets around attribute declarations
  5285 abbreviates a theorem reference involving an internal dummy fact,
  5286 which will be ignored later --- only the effect of the attribute on
  5287 the background context will persist.  This form of in-place
  5288 declarations is particularly useful with commands like 'declare' and
  5289 'using', for example ``have A using [[simproc a]] by simp''.
  5290 
  5291 * Isar: method "assumption" (and implicit closing of subproofs) now
  5292 takes simple non-atomic goal assumptions into account: after applying
  5293 an assumption as a rule the resulting subgoals are solved by atomic
  5294 assumption steps.  This is particularly useful to finish 'obtain'
  5295 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5296 without referring to the original premise "!!x. P x ==> thesis" in the
  5297 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5298 more permissive.
  5299 
  5300 * Isar: implicit use of prems from the Isar proof context is
  5301 considered a legacy feature.  Common applications like ``have A .''
  5302 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5303 referencing facts explicitly here improves readability and
  5304 maintainability of proof texts.
  5305 
  5306 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5307 the obtained context from the course of reasoning!  For example:
  5308 
  5309   assume "EX x y. A x & B y"   -- "any previous fact"
  5310   then guess x and y by clarify
  5311 
  5312 This technique is potentially adventurous, depending on the facts and
  5313 proof tools being involved here.
  5314 
  5315 * Isar: known facts from the proof context may be specified as literal
  5316 propositions, using ASCII back-quote syntax.  This works wherever
  5317 named facts used to be allowed so far, in proof commands, proof
  5318 methods, attributes etc.  Literal facts are retrieved from the context
  5319 according to unification of type and term parameters.  For example,
  5320 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5321 theorems in the current context, then these are valid literal facts:
  5322 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5323 
  5324 There is also a proof method "fact" which does the same composition
  5325 for explicit goal states, e.g. the following proof texts coincide with
  5326 certain special cases of literal facts:
  5327 
  5328   have "A" by fact                 ==  note `A`
  5329   have "A ==> B" by fact           ==  note `A ==> B`
  5330   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5331   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5332 
  5333 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5334 outer syntax.  Thus symbolic identifiers may be used without
  5335 additional white space in declarations like this: ``assume *: A''.
  5336 
  5337 * Isar: 'print_facts' prints all local facts of the current context,
  5338 both named and unnamed ones.
  5339 
  5340 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5341 
  5342   def x == "t" and y == "u"
  5343 
  5344 * Isar: added command 'unfolding', which is structurally similar to
  5345 'using', but affects both the goal state and facts by unfolding given
  5346 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5347 'unfolded' attribute may be replaced by first-class proof text.
  5348 
  5349 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5350 and command 'unfolding' now all support object-level equalities
  5351 (potentially conditional).  The underlying notion of rewrite rule is
  5352 analogous to the 'rule_format' attribute, but *not* that of the
  5353 Simplifier (which is usually more generous).
  5354 
  5355 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5356 premises of a theorem by n. Useful in conjunction with drule.
  5357 
  5358 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5359 method expression within a sandbox consisting of the first N
  5360 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5361 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5362 simplifies all new goals that emerge from applying rule foo to the
  5363 originally first one.
  5364 
  5365 * Isar: schematic goals are no longer restricted to higher-order
  5366 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5367 expected.
  5368 
  5369 * Isar: the conclusion of a long theorem statement is now either
  5370 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5371 (essentially a disjunction of cases with local parameters and
  5372 assumptions).  The latter allows to express general elimination rules
  5373 adequately; in this notation common elimination rules look like this:
  5374 
  5375   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5376     assumes "EX x. P x"
  5377     obtains x where "P x"
  5378 
  5379   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5380     assumes "A & B"
  5381     obtains A and B
  5382 
  5383   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5384     assumes "A | B"
  5385     obtains
  5386       A
  5387     | B
  5388 
  5389 The subsequent classical rules even refer to the formal "thesis"
  5390 explicitly:
  5391 
  5392   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5393     obtains "~ thesis"
  5394 
  5395   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5396     obtains "thesis ==> something"
  5397 
  5398 The actual proof of an 'obtains' statement is analogous to that of the
  5399 Isar proof element 'obtain', only that there may be several cases.
  5400 Optional case names may be specified in parentheses; these will be
  5401 available both in the present proof and as annotations in the
  5402 resulting rule, for later use with the 'cases' method (cf. attribute
  5403 case_names).
  5404 
  5405 * Isar: the assumptions of a long theorem statement are available as
  5406 "assms" fact in the proof context.  This is more appropriate than the
  5407 (historical) "prems", which refers to all assumptions of the current
  5408 context, including those from the target locale, proof body etc.
  5409 
  5410 * Isar: 'print_statement' prints theorems from the current theory or
  5411 proof context in long statement form, according to the syntax of a
  5412 top-level lemma.
  5413 
  5414 * Isar: 'obtain' takes an optional case name for the local context
  5415 introduction rule (default "that").
  5416 
  5417 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5418 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5419 to occur.
  5420 
  5421 * Pure: syntax "CONST name" produces a fully internalized constant
  5422 according to the current context.  This is particularly useful for
  5423 syntax translations that should refer to internal constant
  5424 representations independently of name spaces.
  5425 
  5426 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5427 instead of "FOO ". This allows multiple binder declarations to coexist
  5428 in the same context.  INCOMPATIBILITY.
  5429 
  5430 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5431 primitive that also works in a locale context (both for constants and
  5432 fixed variables). Type declaration and internal syntactic representation
  5433 of given constants retrieved from the context. Likewise, the
  5434 'no_notation' command allows to remove given syntax annotations from the
  5435 current context.
  5436 
  5437 * Isar/locales: new derived specification elements 'axiomatization',
  5438 'definition', 'abbreviation', which support type-inference, admit
  5439 object-level specifications (equality, equivalence).  See also the
  5440 isar-ref manual.  Examples:
  5441 
  5442   axiomatization
  5443     eq  (infix "===" 50) where
  5444     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5445 
  5446   definition "f x y = x + y + 1"
  5447   definition g where "g x = f x x"
  5448 
  5449   abbreviation
  5450     neq  (infix "=!=" 50) where
  5451     "x =!= y == ~ (x === y)"
  5452 
  5453 These specifications may be also used in a locale context.  Then the
  5454 constants being introduced depend on certain fixed parameters, and the
  5455 constant name is qualified by the locale base name.  An internal
  5456 abbreviation takes care for convenient input and output, making the
  5457 parameters implicit and using the original short name.  See also
  5458 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5459 entities from a monomorphic theory.
  5460 
  5461 Presently, abbreviations are only available 'in' a target locale, but
  5462 not inherited by general import expressions.  Also note that
  5463 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5464 'translations' in common applications.  The "no_abbrevs" print mode
  5465 prevents folding of abbreviations in term output.
  5466 
  5467 Concrete syntax is attached to specified constants in internal form,
  5468 independently of name spaces.  The parse tree representation is
  5469 slightly different -- use 'notation' instead of raw 'syntax', and
  5470 'translations' with explicit "CONST" markup to accommodate this.
  5471 
  5472 * Pure/Isar: unified syntax for new-style specification mechanisms
  5473 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5474 full type inference and dummy patterns ("_").  For example:
  5475 
  5476   definition "K x _ = x"
  5477 
  5478   inductive conj for A B
  5479   where "A ==> B ==> conj A B"
  5480 
  5481 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5482 the current context.  Print mode "no_abbrevs" prevents inversion of
  5483 abbreviations on output.
  5484 
  5485 * Isar/locales: improved parameter handling: use of locales "var" and
  5486 "struct" no longer necessary; - parameter renamings are no longer
  5487 required to be injective.  For example, this allows to define
  5488 endomorphisms as locale endom = homom mult mult h.
  5489 
  5490 * Isar/locales: changed the way locales with predicates are defined.
  5491 Instead of accumulating the specification, the imported expression is
  5492 now an interpretation.  INCOMPATIBILITY: different normal form of
  5493 locale expressions.  In particular, in interpretations of locales with
  5494 predicates, goals repesenting already interpreted fragments are not
  5495 removed automatically.  Use methods `intro_locales' and
  5496 `unfold_locales'; see below.
  5497 
  5498 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5499 provide backward reasoning on locales predicates.  The methods are
  5500 aware of interpretations and discharge corresponding goals.
  5501 `intro_locales' is less aggressive then `unfold_locales' and does not
  5502 unfold predicates to assumptions.
  5503 
  5504 * Isar/locales: the order in which locale fragments are accumulated
  5505 has changed.  This enables to override declarations from fragments due
  5506 to interpretations -- for example, unwanted simp rules.
  5507 
  5508 * Isar/locales: interpretation in theories and proof contexts has been
  5509 extended.  One may now specify (and prove) equations, which are
  5510 unfolded in interpreted theorems.  This is useful for replacing
  5511 defined concepts (constants depending on locale parameters) by
  5512 concepts already existing in the target context.  Example:
  5513 
  5514   interpretation partial_order ["op <= :: [int, int] => bool"]
  5515     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5516 
  5517 Typically, the constant `partial_order.less' is created by a
  5518 definition specification element in the context of locale
  5519 partial_order.
  5520 
  5521 * Method "induct": improved internal context management to support
  5522 local fixes and defines on-the-fly. Thus explicit meta-level
  5523 connectives !!  and ==> are rarely required anymore in inductive goals
  5524 (using object-logic connectives for this purpose has been long
  5525 obsolete anyway). Common proof patterns are explained in
  5526 src/HOL/Induct/Common_Patterns.thy, see also
  5527 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5528 examples.
  5529 
  5530 * Method "induct": improved handling of simultaneous goals. Instead of
  5531 introducing object-level conjunction, the statement is now split into
  5532 several conclusions, while the corresponding symbolic cases are nested
  5533 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5534 see src/HOL/Induct/Common_Patterns.thy, for example.
  5535 
  5536 * Method "induct": mutual induction rules are now specified as a list
  5537 of rule sharing the same induction cases. HOL packages usually provide
  5538 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5539 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5540 mutual induction rules differently, i.e. like this:
  5541 
  5542   (induct rule: foo_bar.inducts)
  5543   (induct set: foo bar)
  5544   (induct pred: foo bar)
  5545   (induct type: foo bar)
  5546 
  5547 The ML function ProjectRule.projections turns old-style rules into the
  5548 new format.
  5549 
  5550 * Method "coinduct": dual of induction, see
  5551 src/HOL/Library/Coinductive_List.thy for various examples.
  5552 
  5553 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5554 considered a legacy feature.
  5555 
  5556 * Attribute "symmetric" produces result with standardized schematic
  5557 variables (index 0).  Potential INCOMPATIBILITY.
  5558 
  5559 * Simplifier: by default the simplifier trace only shows top level
  5560 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5561 default. Thus there is less danger of being flooded by the trace. The
  5562 trace indicates where parts have been suppressed.
  5563 
  5564 * Provers/classical: removed obsolete classical version of elim_format
  5565 attribute; classical elim/dest rules are now treated uniformly when
  5566 manipulating the claset.
  5567 
  5568 * Provers/classical: stricter checks to ensure that supplied intro,
  5569 dest and elim rules are well-formed; dest and elim rules must have at
  5570 least one premise.
  5571 
  5572 * Provers/classical: attributes dest/elim/intro take an optional
  5573 weight argument for the rule (just as the Pure versions).  Weights are
  5574 ignored by automated tools, but determine the search order of single
  5575 rule steps.
  5576 
  5577 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5578 where the body does not mention the bound variable.  Note that dummy
  5579 patterns implicitly depend on their context of bounds, which makes
  5580 "{_. _}" match any set comprehension as expected.  Potential
  5581 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5582 constant "_idtdummy" in the binding position.
  5583 
  5584 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5585 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5586 for example "A -> B" => "Pi A (%_. B)".
  5587 
  5588 * Pure: 'class_deps' command visualizes the subclass relation, using
  5589 the graph browser tool.
  5590 
  5591 * Pure: 'print_theory' now suppresses certain internal declarations by
  5592 default; use '!' option for full details.
  5593 
  5594 
  5595 *** HOL ***
  5596 
  5597 * Method "metis" proves goals by applying the Metis general-purpose
  5598 resolution prover (see also http://gilith.com/software/metis/).
  5599 Examples are in the directory MetisExamples.  WARNING: the
  5600 Isabelle/HOL-Metis integration does not yet work properly with
  5601 multi-threading.
  5602 
  5603 * Command 'sledgehammer' invokes external automatic theorem provers as
  5604 background processes.  It generates calls to the "metis" method if
  5605 successful. These can be pasted into the proof.  Users do not have to
  5606 wait for the automatic provers to return.  WARNING: does not really
  5607 work with multi-threading.
  5608 
  5609 * New "auto_quickcheck" feature tests outermost goal statements for
  5610 potential counter-examples.  Controlled by ML references
  5611 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5612 5000 milliseconds).  Fails silently if statements is outside of
  5613 executable fragment, or any other codgenerator problem occurs.
  5614 
  5615 * New constant "undefined" with axiom "undefined x = undefined".
  5616 
  5617 * Added class "HOL.eq", allowing for code generation with polymorphic
  5618 equality.
  5619 
  5620 * Some renaming of class constants due to canonical name prefixing in
  5621 the new 'class' package:
  5622 
  5623     HOL.abs ~> HOL.abs_class.abs
  5624     HOL.divide ~> HOL.divide_class.divide
  5625     0 ~> HOL.zero_class.zero
  5626     1 ~> HOL.one_class.one
  5627     op + ~> HOL.plus_class.plus
  5628     op - ~> HOL.minus_class.minus
  5629     uminus ~> HOL.minus_class.uminus
  5630     op * ~> HOL.times_class.times
  5631     op < ~> HOL.ord_class.less
  5632     op <= > HOL.ord_class.less_eq
  5633     Nat.power ~> Power.power_class.power
  5634     Nat.size ~> Nat.size_class.size
  5635     Numeral.number_of ~> Numeral.number_class.number_of
  5636     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5637     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5638     Orderings.min ~> Orderings.ord_class.min
  5639     Orderings.max ~> Orderings.ord_class.max
  5640     Divides.op div ~> Divides.div_class.div
  5641     Divides.op mod ~> Divides.div_class.mod
  5642     Divides.op dvd ~> Divides.div_class.dvd
  5643 
  5644 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5645 
  5646 a) User-defined constants using any of the names "plus", "minus",
  5647 "times", "less" or "less_eq". The standard syntax translations for
  5648 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5649 names.
  5650 
  5651 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5652 INCOMPATIBILITY: use more specific names.
  5653 
  5654 c) Permutative equations (e.g. "a + b = b + a")
  5655 Since the change of names also changes the order of terms, permutative
  5656 rewrite rules may get applied in a different order. Experience shows
  5657 that this is rarely the case (only two adaptions in the whole Isabelle
  5658 distribution).  INCOMPATIBILITY: rewrite proofs
  5659 
  5660 d) ML code directly refering to constant names
  5661 This in general only affects hand-written proof tactics, simprocs and
  5662 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5663 Consider using @{const_name} antiquotation.
  5664 
  5665 * New class "default" with associated constant "default".
  5666 
  5667 * Function "sgn" is now overloaded and available on int, real, complex
  5668 (and other numeric types), using class "sgn".  Two possible defs of
  5669 sgn are given as equational assumptions in the classes sgn_if and
  5670 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5671 INCOMPATIBILITY.
  5672 
  5673 * Locale "partial_order" now unified with class "order" (cf. theory
  5674 Orderings), added parameter "less".  INCOMPATIBILITY.
  5675 
  5676 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5677 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5678 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5679 
  5680 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5681 changed order of proof goals in instance proofs.
  5682 
  5683 * The transitivity reasoner for partial and linear orders is set up
  5684 for classes "order" and "linorder".  Instances of the reasoner are available
  5685 in all contexts importing or interpreting the corresponding locales.
  5686 Method "order" invokes the reasoner separately; the reasoner
  5687 is also integrated with the Simplifier as a solver.  Diagnostic
  5688 command 'print_orders' shows the available instances of the reasoner
  5689 in the current context.
  5690 
  5691 * Localized monotonicity predicate in theory "Orderings"; integrated
  5692 lemmas max_of_mono and min_of_mono with this predicate.
  5693 INCOMPATIBILITY.
  5694 
  5695 * Formulation of theorem "dense" changed slightly due to integration
  5696 with new class dense_linear_order.
  5697 
  5698 * Uniform lattice theory development in HOL.
  5699 
  5700     constants "meet" and "join" now named "inf" and "sup"
  5701     constant "Meet" now named "Inf"
  5702 
  5703     classes "meet_semilorder" and "join_semilorder" now named
  5704       "lower_semilattice" and "upper_semilattice"
  5705     class "lorder" now named "lattice"
  5706     class "comp_lat" now named "complete_lattice"
  5707 
  5708     Instantiation of lattice classes allows explicit definitions
  5709     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5710 
  5711   INCOMPATIBILITY.  Theorem renames:
  5712 
  5713     meet_left_le            ~> inf_le1
  5714     meet_right_le           ~> inf_le2
  5715     join_left_le            ~> sup_ge1
  5716     join_right_le           ~> sup_ge2
  5717     meet_join_le            ~> inf_sup_ord
  5718     le_meetI                ~> le_infI
  5719     join_leI                ~> le_supI
  5720     le_meet                 ~> le_inf_iff
  5721     le_join                 ~> ge_sup_conv
  5722     meet_idempotent         ~> inf_idem
  5723     join_idempotent         ~> sup_idem
  5724     meet_comm               ~> inf_commute
  5725     join_comm               ~> sup_commute
  5726     meet_leI1               ~> le_infI1
  5727     meet_leI2               ~> le_infI2
  5728     le_joinI1               ~> le_supI1
  5729     le_joinI2               ~> le_supI2
  5730     meet_assoc              ~> inf_assoc
  5731     join_assoc              ~> sup_assoc
  5732     meet_left_comm          ~> inf_left_commute
  5733     meet_left_idempotent    ~> inf_left_idem
  5734     join_left_comm          ~> sup_left_commute
  5735     join_left_idempotent    ~> sup_left_idem
  5736     meet_aci                ~> inf_aci
  5737     join_aci                ~> sup_aci
  5738     le_def_meet             ~> le_iff_inf
  5739     le_def_join             ~> le_iff_sup
  5740     join_absorp2            ~> sup_absorb2
  5741     join_absorp1            ~> sup_absorb1
  5742     meet_absorp1            ~> inf_absorb1
  5743     meet_absorp2            ~> inf_absorb2
  5744     meet_join_absorp        ~> inf_sup_absorb
  5745     join_meet_absorp        ~> sup_inf_absorb
  5746     distrib_join_le         ~> distrib_sup_le
  5747     distrib_meet_le         ~> distrib_inf_le
  5748 
  5749     add_meet_distrib_left   ~> add_inf_distrib_left
  5750     add_join_distrib_left   ~> add_sup_distrib_left
  5751     is_join_neg_meet        ~> is_join_neg_inf
  5752     is_meet_neg_join        ~> is_meet_neg_sup
  5753     add_meet_distrib_right  ~> add_inf_distrib_right
  5754     add_join_distrib_right  ~> add_sup_distrib_right
  5755     add_meet_join_distribs  ~> add_sup_inf_distribs
  5756     join_eq_neg_meet        ~> sup_eq_neg_inf
  5757     meet_eq_neg_join        ~> inf_eq_neg_sup
  5758     add_eq_meet_join        ~> add_eq_inf_sup
  5759     meet_0_imp_0            ~> inf_0_imp_0
  5760     join_0_imp_0            ~> sup_0_imp_0
  5761     meet_0_eq_0             ~> inf_0_eq_0
  5762     join_0_eq_0             ~> sup_0_eq_0
  5763     neg_meet_eq_join        ~> neg_inf_eq_sup
  5764     neg_join_eq_meet        ~> neg_sup_eq_inf
  5765     join_eq_if              ~> sup_eq_if
  5766 
  5767     mono_meet               ~> mono_inf
  5768     mono_join               ~> mono_sup
  5769     meet_bool_eq            ~> inf_bool_eq
  5770     join_bool_eq            ~> sup_bool_eq
  5771     meet_fun_eq             ~> inf_fun_eq
  5772     join_fun_eq             ~> sup_fun_eq
  5773     meet_set_eq             ~> inf_set_eq
  5774     join_set_eq             ~> sup_set_eq
  5775     meet1_iff               ~> inf1_iff
  5776     meet2_iff               ~> inf2_iff
  5777     meet1I                  ~> inf1I
  5778     meet2I                  ~> inf2I
  5779     meet1D1                 ~> inf1D1
  5780     meet2D1                 ~> inf2D1
  5781     meet1D2                 ~> inf1D2
  5782     meet2D2                 ~> inf2D2
  5783     meet1E                  ~> inf1E
  5784     meet2E                  ~> inf2E
  5785     join1_iff               ~> sup1_iff
  5786     join2_iff               ~> sup2_iff
  5787     join1I1                 ~> sup1I1
  5788     join2I1                 ~> sup2I1
  5789     join1I1                 ~> sup1I1
  5790     join2I2                 ~> sup1I2
  5791     join1CI                 ~> sup1CI
  5792     join2CI                 ~> sup2CI
  5793     join1E                  ~> sup1E
  5794     join2E                  ~> sup2E
  5795 
  5796     is_meet_Meet            ~> is_meet_Inf
  5797     Meet_bool_def           ~> Inf_bool_def
  5798     Meet_fun_def            ~> Inf_fun_def
  5799     Meet_greatest           ~> Inf_greatest
  5800     Meet_lower              ~> Inf_lower
  5801     Meet_set_def            ~> Inf_set_def
  5802 
  5803     Sup_def                 ~> Sup_Inf
  5804     Sup_bool_eq             ~> Sup_bool_def
  5805     Sup_fun_eq              ~> Sup_fun_def
  5806     Sup_set_eq              ~> Sup_set_def
  5807 
  5808     listsp_meetI            ~> listsp_infI
  5809     listsp_meet_eq          ~> listsp_inf_eq
  5810 
  5811     meet_min                ~> inf_min
  5812     join_max                ~> sup_max
  5813 
  5814 * Added syntactic class "size"; overloaded constant "size" now has
  5815 type "'a::size ==> bool"
  5816 
  5817 * Internal reorganisation of `size' of datatypes: size theorems
  5818 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5819 simplification rules by default!); theorems "prod.size" now named
  5820 "*.size".
  5821 
  5822 * Class "div" now inherits from class "times" rather than "type".
  5823 INCOMPATIBILITY.
  5824 
  5825 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5826 Linorder etc.  have disappeared; operations defined in terms of
  5827 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5828 
  5829 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5830 
  5831 * HOL-Word: New extensive library and type for generic, fixed size
  5832 machine words, with arithmetic, bit-wise, shifting and rotating
  5833 operations, reflection into int, nat, and bool lists, automation for
  5834 linear arithmetic (by automatic reflection into nat or int), including
  5835 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5836 arithmetic type classes, supporting automatic simplification of
  5837 numerals on all operations.
  5838 
  5839 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5840 
  5841 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5842 
  5843 * Code generator library theories:
  5844   - Code_Integer represents HOL integers by big integer literals in target
  5845     languages.
  5846   - Code_Char represents HOL characters by character literals in target
  5847     languages.
  5848   - Code_Char_chr like Code_Char, but also offers treatment of character
  5849     codes; includes Code_Integer.
  5850   - Executable_Set allows to generate code for finite sets using lists.
  5851   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5852     denominator).
  5853   - Executable_Real implements a subset of real numbers, namly those
  5854     representable by rational numbers.
  5855   - Efficient_Nat implements natural numbers by integers, which in general will
  5856     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5857     includes Code_Integer.
  5858   - Code_Index provides an additional datatype index which is mapped to
  5859     target-language built-in integers.
  5860   - Code_Message provides an additional datatype message_string which is isomorphic to
  5861     strings; messages are mapped to target-language strings.
  5862 
  5863 * New package for inductive predicates
  5864 
  5865   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5866 
  5867     inductive
  5868       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5869       for z_1 :: U_1 and ... and z_n :: U_m
  5870     where
  5871       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5872     | ...
  5873 
  5874   with full support for type-inference, rather than
  5875 
  5876     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5877 
  5878     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5879     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5880 
  5881     inductive "s z_1 ... z_m"
  5882     intros
  5883       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5884       ...
  5885 
  5886   For backward compatibility, there is a wrapper allowing inductive
  5887   sets to be defined with the new package via
  5888 
  5889     inductive_set
  5890       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5891       for z_1 :: U_1 and ... and z_n :: U_m
  5892     where
  5893       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5894     | ...
  5895 
  5896   or
  5897 
  5898     inductive_set
  5899       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5900       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5901       for z_1 :: U_1 and ... and z_n :: U_m
  5902     where
  5903       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5904     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5905     | ...
  5906 
  5907   if the additional syntax "p ..." is required.
  5908 
  5909   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5910   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5911 
  5912   INCOMPATIBILITIES:
  5913 
  5914   - Since declaration and definition of inductive sets or predicates
  5915     is no longer separated, abbreviations involving the newly
  5916     introduced sets or predicates must be specified together with the
  5917     introduction rules after the 'where' keyword (see above), rather
  5918     than before the actual inductive definition.
  5919 
  5920   - The variables in induction and elimination rules are now
  5921     quantified in the order of their occurrence in the introduction
  5922     rules, rather than in alphabetical order. Since this may break
  5923     some proofs, these proofs either have to be repaired, e.g. by
  5924     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5925     statements of the form
  5926 
  5927       case (rule_i a_i_1 ... a_i_{k_i})
  5928 
  5929     or the old order of quantification has to be restored by explicitly adding
  5930     meta-level quantifiers in the introduction rules, i.e.
  5931 
  5932       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5933 
  5934   - The format of the elimination rules is now
  5935 
  5936       p z_1 ... z_m x_1 ... x_n ==>
  5937         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5938         ==> ... ==> P
  5939 
  5940     for predicates and
  5941 
  5942       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5943         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5944         ==> ... ==> P
  5945 
  5946     for sets rather than
  5947 
  5948       x : s z_1 ... z_m ==>
  5949         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5950         ==> ... ==> P
  5951 
  5952     This may require terms in goals to be expanded to n-tuples
  5953     (e.g. using case_tac or simplification with the split_paired_all
  5954     rule) before the above elimination rule is applicable.
  5955 
  5956   - The elimination or case analysis rules for (mutually) inductive
  5957     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5958     list of rules "p_1_..._p_k.elims" is no longer available.
  5959 
  5960 * New package "function"/"fun" for general recursive functions,
  5961 supporting mutual and nested recursion, definitions in local contexts,
  5962 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5963 for small examples, and the separate tutorial on the function
  5964 package. The old recdef "package" is still available as before, but
  5965 users are encouraged to use the new package.
  5966 
  5967 * Method "lexicographic_order" automatically synthesizes termination
  5968 relations as lexicographic combinations of size measures.
  5969 
  5970 * Case-expressions allow arbitrary constructor-patterns (including
  5971 "_") and take their order into account, like in functional
  5972 programming.  Internally, this is translated into nested
  5973 case-expressions; missing cases are added and mapped to the predefined
  5974 constant "undefined". In complicated cases printing may no longer show
  5975 the original input but the internal form. Lambda-abstractions allow
  5976 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5977 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5978 variable.
  5979 
  5980 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5981 is an abbreviation for "of_nat :: nat => int". The simplification
  5982 rules for "of_nat" have been changed to work like "int" did
  5983 previously.  Potential INCOMPATIBILITY:
  5984   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5985   - of_nat_diff and of_nat_mult are no longer default simp rules
  5986 
  5987 * Method "algebra" solves polynomial equations over (semi)rings using
  5988 Groebner bases. The (semi)ring structure is defined by locales and the
  5989 tool setup depends on that generic context. Installing the method for
  5990 a specific type involves instantiating the locale and possibly adding
  5991 declarations for computation on the coefficients.  The method is
  5992 already instantiated for natural numbers and for the axiomatic class
  5993 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5994 CALCULEMUS 2007 for the general principles underlying this
  5995 architecture of context-aware proof-tools.
  5996 
  5997 * Method "ferrack" implements quantifier elimination over
  5998 special-purpose dense linear orders using locales (analogous to
  5999 "algebra"). The method is already installed for class
  6000 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  6001 rat, etc.
  6002 
  6003 * Former constant "List.op @" now named "List.append".  Use ML
  6004 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  6005 circumvent possible incompatibilities when working on ML level.
  6006 
  6007 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  6008 
  6009 * New function listsum :: 'a list => 'a for arbitrary monoids.
  6010 Special syntax: "SUM x <- xs. f x" (and latex variants)
  6011 
  6012 * New syntax for Haskell-like list comprehension (input only), eg.
  6013 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  6014 
  6015 * The special syntax for function "filter" has changed from [x :
  6016 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  6017 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  6018 
  6019 * [a..b] is now defined for arbitrary linear orders.  It used to be
  6020 defined on nat only, as an abbreviation for [a..<Suc b]
  6021 INCOMPATIBILITY.
  6022 
  6023 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  6024 
  6025 * New functions "sorted" and "sort" in src/HOL/List.thy.
  6026 
  6027 * New lemma collection field_simps (an extension of ring_simps) for
  6028 manipulating (in)equations involving division. Multiplies with all
  6029 denominators that can be proved to be non-zero (in equations) or
  6030 positive/negative (in inequations).
  6031 
  6032 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  6033 have been improved and renamed to ring_simps, group_simps and
  6034 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  6035 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  6036 
  6037 * Theory Library/Commutative_Ring: switched from recdef to function
  6038 package; constants add, mul, pow now curried.  Infix syntax for
  6039 algebraic operations.
  6040 
  6041 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  6042 INCOMPATIBILITY.
  6043 
  6044 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  6045 INCOMPATIBILITY.
  6046 
  6047 * HOL/records: generalised field-update to take a function on the
  6048 field rather than the new value: r(|A := x|) is translated to A_update
  6049 (K x) r The K-combinator that is internally used is called K_record.
  6050 INCOMPATIBILITY: Usage of the plain update functions has to be
  6051 adapted.
  6052 
  6053 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  6054 * x = 0, which are required for a semiring.  Richer structures do not
  6055 inherit from semiring_0 anymore, because this property is a theorem
  6056 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  6057 there is more to prove, but this is mostly trivial.
  6058 
  6059 * Class "recpower" is generalized to arbitrary monoids, not just
  6060 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  6061 commutativity or semiring properties additionally.
  6062 
  6063 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  6064 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  6065 better use 'abbreviation'.
  6066 
  6067 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  6068 
  6069 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  6070 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  6071 significant changes for setting up numeral syntax for types:
  6072   - New constants Numeral.pred and Numeral.succ instead
  6073       of former Numeral.bin_pred and Numeral.bin_succ.
  6074   - Use integer operations instead of bin_add, bin_mult and so on.
  6075   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  6076   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  6077 
  6078 See src/HOL/Integ/IntArith.thy for an example setup.
  6079 
  6080 * Command 'normal_form' computes the normal form of a term that may
  6081 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  6082 produces ``[b, c, a]'' (without proof).  This command is suitable for
  6083 heavy-duty computations because the functions are compiled to ML
  6084 first.  Correspondingly, a method "normalization" is provided.  See
  6085 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  6086 
  6087 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  6088 25 like -->); output depends on the "iff" print_mode, the default is
  6089 "A = B" (with priority 50).
  6090 
  6091 * Relations less (<) and less_eq (<=) are also available on type bool.
  6092 Modified syntax to disallow nesting without explicit parentheses,
  6093 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  6094 INCOMPATIBILITY.
  6095 
  6096 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  6097 
  6098 * Relation composition operator "op O" now has precedence 75 and binds
  6099 stronger than union and intersection. INCOMPATIBILITY.
  6100 
  6101 * The old set interval syntax "{m..n(}" (and relatives) has been
  6102 removed.  Use "{m..<n}" (and relatives) instead.
  6103 
  6104 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  6105 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  6106 ``declare [[simproc del: neq]]''.
  6107 
  6108 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  6109 True/False.
  6110 
  6111 * Theorem Cons_eq_map_conv no longer declared as "simp".
  6112 
  6113 * Theorem setsum_mult renamed to setsum_right_distrib.
  6114 
  6115 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  6116 ``rule'' method.
  6117 
  6118 * Reimplemented methods "sat" and "satx", with several improvements:
  6119 goals no longer need to be stated as "<prems> ==> False", equivalences
  6120 (i.e. "=" on type bool) are handled, variable names of the form
  6121 "lit_<n>" are no longer reserved, significant speedup.
  6122 
  6123 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  6124 zChaff is still supported as well.
  6125 
  6126 * 'inductive' and 'datatype': provide projections of mutual rules,
  6127 bundled as foo_bar.inducts;
  6128 
  6129 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  6130 Library.
  6131 
  6132 * Library: moved theory Accessible_Part to main HOL.
  6133 
  6134 * Library: added theory Coinductive_List of potentially infinite lists
  6135 as greatest fixed-point.
  6136 
  6137 * Library: added theory AssocList which implements (finite) maps as
  6138 association lists.
  6139 
  6140 * Method "evaluation" solves goals (i.e. a boolean expression)
  6141 efficiently by compiling it to ML.  The goal is "proved" (via an
  6142 oracle) if it evaluates to True.
  6143 
  6144 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  6145 also when invoked by the simplifier.  This results in the Simplifier
  6146 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  6147 Configuration option fast_arith_split_limit=0 recovers the old
  6148 behavior.
  6149 
  6150 * Support for hex (0x20) and binary (0b1001) numerals.
  6151 
  6152 * New method: reify eqs (t), where eqs are equations for an
  6153 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  6154 parameter, computes a term s::'b and a list xs::'a list and proves the
  6155 theorem I xs s = t. This is also known as reification or quoting. The
  6156 resulting theorem is applied to the subgoal to substitute t with I xs
  6157 s.  If t is omitted, the subgoal itself is reified.
  6158 
  6159 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  6160 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  6161 where f is supposed to be a computable function (in the sense of code
  6162 generattion). The method uses reify to compute s and xs as above then
  6163 applies corr_thm and uses normalization by evaluation to "prove" f s =
  6164 r and finally gets the theorem t = r, which is again applied to the
  6165 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  6166 
  6167 * Reflection: Automatic reification now handels binding, an example is
  6168 available in src/HOL/ex/ReflectionEx.thy
  6169 
  6170 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  6171 command 'statespace' that is similar to 'record', but introduces an
  6172 abstract specification based on the locale infrastructure instead of
  6173 HOL types.  This leads to extra flexibility in composing state spaces,
  6174 in particular multiple inheritance and renaming of components.
  6175 
  6176 
  6177 *** HOL-Complex ***
  6178 
  6179 * Hyperreal: Functions root and sqrt are now defined on negative real
  6180 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  6181 Nonnegativity side conditions have been removed from many lemmas, so
  6182 that more subgoals may now be solved by simplification; potential
  6183 INCOMPATIBILITY.
  6184 
  6185 * Real: new type classes formalize real normed vector spaces and
  6186 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  6187 and norm :: 'a => real.
  6188 
  6189 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  6190 reals into other types. The overloaded constant Reals :: 'a set is now
  6191 defined as range of_real; potential INCOMPATIBILITY.
  6192 
  6193 * Real: proper support for ML code generation, including 'quickcheck'.
  6194 Reals are implemented as arbitrary precision rationals.
  6195 
  6196 * Hyperreal: Several constants that previously worked only for the
  6197 reals have been generalized, so they now work over arbitrary vector
  6198 spaces. Type annotations may need to be added in some cases; potential
  6199 INCOMPATIBILITY.
  6200 
  6201   Infinitesimal  :: ('a::real_normed_vector) star set
  6202   HFinite        :: ('a::real_normed_vector) star set
  6203   HInfinite      :: ('a::real_normed_vector) star set
  6204   approx         :: ('a::real_normed_vector) star => 'a star => bool
  6205   monad          :: ('a::real_normed_vector) star => 'a star set
  6206   galaxy         :: ('a::real_normed_vector) star => 'a star set
  6207   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  6208   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  6209   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  6210   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  6211   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  6212   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  6213   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6214   sgn            :: 'a::real_normed_vector => 'a
  6215   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6216 
  6217 * Complex: Some complex-specific constants are now abbreviations for
  6218 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6219 hnorm.  Other constants have been entirely removed in favor of the
  6220 polymorphic versions (INCOMPATIBILITY):
  6221 
  6222   approx        <-- capprox
  6223   HFinite       <-- CFinite
  6224   HInfinite     <-- CInfinite
  6225   Infinitesimal <-- CInfinitesimal
  6226   monad         <-- cmonad
  6227   galaxy        <-- cgalaxy
  6228   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  6229   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  6230   (ns)deriv     <-- (ns)cderiv
  6231 
  6232 
  6233 *** HOL-Algebra ***
  6234 
  6235 * Formalisation of ideals and the quotient construction over rings.
  6236 
  6237 * Order and lattice theory no longer based on records.