NEWS
author wenzelm
Mon Nov 26 19:53:43 2012 +0100 (2012-11-26)
changeset 50232 289a34f9c383
parent 50231 81a067b188b8
child 50406 c28753665b8e
permissions -rw-r--r--
tuned;
     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 * Refined 'help' command to retrieve outer syntax commands according
    45 to name patterns (with clickable results).
    46 
    47 
    48 *** Prover IDE -- Isabelle/Scala/jEdit ***
    49 
    50 * Parallel terminal proofs ('by') are enabled by default, likewise
    51 proofs that are built into packages like 'datatype', 'function'.  This
    52 allows to "run ahead" checking the theory specifications on the
    53 surface, while the prover is still crunching on internal
    54 justifications.  Unfinished / cancelled proofs are restarted as
    55 required to complete full proof checking eventually.
    56 
    57 * Improved output panel with tooltips, hyperlinks etc. based on the
    58 same Rich_Text_Area as regular Isabelle/jEdit buffers.  Activation of
    59 tooltips leads to some window that supports the same recursively,
    60 which can lead to stacks of tooltips as the semantic document content
    61 is explored.  ESCAPE closes the whole stack, individual windows may be
    62 closed separately, or detached to become independent jEdit dockables.
    63 
    64 * More robust incremental parsing of outer syntax (partial comments,
    65 malformed symbols).  Changing the balance of open/close quotes and
    66 comment delimiters works more conveniently with unfinished situations
    67 that frequently occur in user interaction.
    68 
    69 * More efficient painting and improved reactivity when editing large
    70 files.  More scalable management of formal document content.
    71 
    72 * Smarter handling of tracing messages: output window informs about
    73 accumulated messages; prover transactions are limited to emit maximum
    74 amount of output, before being canceled (cf. system option
    75 "editor_tracing_limit").  This avoids swamping the front-end with
    76 potentially infinite message streams.
    77 
    78 * More plugin options and preferences, based on Isabelle/Scala.  The
    79 jEdit plugin option panel provides access to some Isabelle/Scala
    80 options, including tuning parameters for editor reactivity and color
    81 schemes.
    82 
    83 * Dockable window "Symbols" provides some editing support for Isabelle
    84 symbols.
    85 
    86 * Improved editing support for control styles: subscript, superscript,
    87 bold, reset of style -- operating on single symbols or text
    88 selections.  Cf. keyboard shortcuts C+e DOWN/UP/RIGHT/LEFT.
    89 
    90 * Actions isabelle.increase-font-size and isabelle.decrease-font-size
    91 adjust the main text area font size, and its derivatives for output,
    92 tooltips etc.  Cf. keyboard shortcuts C-PLUS and C-MINUS.
    93 
    94 * Uniform Java 7 platform on Linux, Mac OS X, Windows: recent updates
    95 from Oracle provide better multi-platform experience.  This version is
    96 now bundled exclusively with Isabelle.
    97 
    98 
    99 *** Pure ***
   100 
   101 * Code generation for Haskell: restrict unqualified imports from
   102 Haskell Prelude to a small set of fundamental operations.
   103 
   104 * Command "export_code": relative file names are interpreted
   105 relatively to master directory of current theory rather than
   106 the rather arbitrary current working directory.
   107 INCOMPATIBILITY.
   108 
   109 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
   110 use regular rule composition via "OF" / "THEN", or explicit proof
   111 structure instead.  Note that Isabelle/ML provides a variety of
   112 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
   113 with some care where this is really required.
   114 
   115 * Command 'typ' supports an additional variant with explicit sort
   116 constraint, to infer and check the most general type conforming to a
   117 given given sort.  Example (in HOL):
   118 
   119   typ "_ * _ * bool * unit" :: finite
   120 
   121 
   122 *** HOL ***
   123 
   124 * Removed constant "chars".  Prefer "Enum.enum" on type "char"
   125 directly.  INCOMPATIBILITY.
   126 
   127 * Moved operation product, sublists and n_lists from Enum.thy
   128 to List.thy.  INCOMPATIBILITY.
   129 
   130 * Simplified 'typedef' specifications: historical options for implicit
   131 set definition and alternative name have been discontinued.  The
   132 former behavior of "typedef (open) t = A" is now the default, but
   133 written just "typedef t = A".  INCOMPATIBILITY, need to adapt theories
   134 accordingly.
   135 
   136 * Theory "Library/Multiset":
   137 
   138   - Renamed constants
   139       fold_mset ~> Multiset.fold  -- for coherence with other fold combinators
   140 
   141   - Renamed facts
   142       fold_mset_commute ~> fold_mset_comm  -- for coherence with fold_comm
   143 
   144 INCOMPATIBILITY.
   145 
   146 * Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
   147 
   148 * Class "comm_monoid_diff" formalises properties of bounded
   149 subtraction, with natural numbers and multisets as typical instances.
   150 
   151 * Theory "Library/Option_ord" provides instantiation of option type
   152 to lattice type classes.
   153 
   154 * New combinator "Option.these" with type "'a option set => 'a set".
   155 
   156 * Renamed theory Library/List_Prefix to Library/Sublist.
   157 INCOMPATIBILITY.  Related changes are:
   158 
   159   - Renamed constants:
   160 
   161       prefix ~> prefixeq
   162       strict_prefix ~> prefix
   163 
   164     Renamed lemmas accordingly, INCOMPATIBILITY.
   165 
   166   - Replaced constant "postfix" by "suffixeq" with swapped argument order
   167     (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix
   168     syntax "xs >>= ys"; use "suffixeq ys xs" instead.  Renamed lemmas
   169     accordingly.  INCOMPATIBILITY.
   170 
   171   - New constant "emb" for homeomorphic embedding on lists. New
   172     abbreviation "sub" for special case "emb (op =)".
   173 
   174   - Library/Sublist does no longer provide "order" and "bot" type class
   175     instances for the prefix order (merely corresponding locale
   176     interpretations). The type class instances are to be found in
   177     Library/Prefix_Order. INCOMPATIBILITY.
   178 
   179   - The sublist relation from Library/Sublist_Order is now based on
   180     "Sublist.sub". Replaced lemmas:
   181 
   182       le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff
   183       le_list_append_mono ~> Sublist.emb_append_mono
   184       le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2
   185       le_list_Cons_EX ~> Sublist.emb_ConsD
   186       le_list_drop_Cons2 ~> Sublist.sub_Cons2'
   187       le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq
   188       le_list_drop_Cons ~> Sublist.sub_Cons'
   189       le_list_drop_many ~> Sublist.sub_drop_many
   190       le_list_filter_left ~> Sublist.sub_filter_left
   191       le_list_rev_drop_many ~> Sublist.sub_rev_drop_many
   192       le_list_rev_take_iff ~> Sublist.sub_append
   193       le_list_same_length ~> Sublist.sub_same_length
   194       le_list_take_many_iff ~> Sublist.sub_append'
   195       less_eq_list.drop ~> less_eq_list_drop
   196       less_eq_list.induct ~> less_eq_list_induct
   197       not_le_list_length ~> Sublist.not_sub_length
   198 
   199     INCOMPATIBILITY.
   200 
   201 * HOL/Rings: renamed lemmas
   202 
   203 left_distrib ~> distrib_right
   204 right_distrib ~> distrib_left
   205 
   206 in class semiring.  INCOMPATIBILITY.
   207 
   208 * HOL/BNF: New (co)datatype package based on bounded natural
   209 functors with support for mixed, nested recursion and interesting
   210 non-free datatypes.
   211 
   212 * HOL/Cardinals: Theories of ordinals and cardinals
   213 (supersedes the AFP entry "Ordinals_and_Cardinals").
   214 
   215 * HOL/Probability:
   216   - Add simproc "measurable" to automatically prove measurability
   217 
   218   - Add induction rules for sigma sets with disjoint union (sigma_sets_induct_disjoint)
   219     and for Borel-measurable functions (borel_measurable_induct).
   220 
   221   - The Daniell-Kolmogorov theorem (the existence the limit of a projective family)
   222 
   223 * Library/Countable_Set.thy: Theory of countable sets.
   224 
   225 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
   226 execution for code generated towards Isabelle/ML.
   227 
   228 * Library/FuncSet.thy: Extended support for Pi and extensional and introduce the
   229 extensional dependent function space "PiE". Replaces extensional_funcset by an
   230 abbreviation, rename a couple of lemmas from extensional_funcset to PiE:
   231 
   232       extensional_empty ~> PiE_empty
   233       extensional_funcset_empty_domain ~> PiE_empty_domain
   234       extensional_funcset_empty_range ~> PiE_empty_range
   235       extensional_funcset_arb ~> PiE_arb
   236       extensional_funcset_mem > PiE_mem
   237       extensional_funcset_extend_domainI ~> PiE_fun_upd
   238       extensional_funcset_restrict_domain ~> fun_upd_in_PiE
   239       extensional_funcset_extend_domain_eq ~> PiE_insert_eq
   240       card_extensional_funcset ~> card_PiE
   241       finite_extensional_funcset ~> finite_PiE
   242 
   243   INCOMPATIBILITY.
   244 
   245 * Library/FinFun.thy: theory of almost everywhere constant functions
   246 (supersedes the AFP entry "Code Generation for Functions as Data").
   247 
   248 * Library/Phantom.thy: generic phantom type to make a type parameter
   249 appear in a constant's type. This alternative to adding TYPE('a) as
   250 another parameter avoids unnecessary closures in generated code.
   251 
   252 * Library/RBT_Impl.thy: efficient construction of red-black trees 
   253 from sorted associative lists. Merging two trees with rbt_union may
   254 return a structurally different tree than before. MINOR INCOMPATIBILITY.
   255 
   256 * Library/IArray.thy: immutable arrays with code generation.
   257 
   258 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   259 expressions.
   260 
   261 * Preprocessing of the code generator rewrites set comprehensions into
   262 pointfree expressions.
   263 
   264 * Quickcheck:
   265 
   266   - added an optimisation for equality premises.
   267     It is switched on by default, and can be switched off by setting
   268     the configuration quickcheck_optimise_equality to false.
   269 
   270 * The SMT solver Z3 has now by default a restricted set of directly
   271 supported features. For the full set of features (div/mod, nonlinear
   272 arithmetic, datatypes/records) with potential proof reconstruction
   273 failures, enable the configuration option "z3_with_extensions".
   274 Minor INCOMPATIBILITY.
   275 
   276 * Sledgehammer:
   277 
   278   - Added MaSh relevance filter based on machine-learning; see the
   279     Sledgehammer manual for details.
   280   - Polished Isar proofs generated with "isar_proofs" option.
   281   - Rationalized type encodings ("type_enc" option).
   282   - Renamed "kill_provers" subcommand to "kill".
   283   - Renamed options:
   284       isar_proof ~> isar_proofs
   285       isar_shrink_factor ~> isar_shrink
   286       max_relevant ~> max_facts
   287       relevance_thresholds ~> fact_thresholds
   288 
   289 
   290 *** Document preparation ***
   291 
   292 * Default for \<euro> is now based on eurosym package, instead of
   293 slightly exotic babel/greek.
   294 
   295 * Document variant NAME may use different LaTeX entry point
   296 document/root_NAME.tex if that file exists, instead of the common
   297 document/root.tex.
   298 
   299 * Simplified custom document/build script, instead of old-style
   300 document/IsaMakefile.  Minor INCOMPATIBILITY.
   301 
   302 
   303 *** ML ***
   304 
   305 * Type Seq.results and related operations support embedded error
   306 messages within lazy enumerations, and thus allow to provide
   307 informative errors in the absence of any usable results.
   308 
   309 * Renamed Position.str_of to Position.here to emphasize that this is a
   310 formal device to inline positions into message text, but not
   311 necessarily printing visible text.
   312 
   313 
   314 *** System ***
   315 
   316 * The default limit for maximum number of worker threads is now 8,
   317 instead of 4.
   318 
   319 * The ML system is configured as regular component, and no longer
   320 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   321 for home-made configurations.
   322 
   323 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   324 Minor INCOMPATIBILITY in command-line options.
   325 
   326 * Advanced support for Isabelle sessions and build management, see
   327 "system" manual for the chapter of that name, especially the "isabelle
   328 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   329 mkdir / make are rendered obsolete.
   330 
   331 * Discontinued obsolete "isabelle makeall".
   332 
   333 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   334 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   335 provides some traditional targets that invoke "isabelle build".  Note
   336 that this is inefficient!  Applications of Isabelle/HOL involving
   337 "isabelle make" should be upgraded to use "isabelle build" directly.
   338 
   339 * Discontinued obsolete Isabelle/build script, it is superseded by the
   340 regular isabelle build tool.  For example:
   341 
   342   isabelle build -s -b HOLCF
   343 
   344 * The "isabelle options" tool prints Isabelle system options, as
   345 required for "isabelle build", for example.
   346 
   347 * The "isabelle mkroot" tool prepares session root directories for use
   348 with "isabelle build", similar to former "isabelle mkdir" for
   349 "isabelle usedir".
   350 
   351 * The "isabelle components" tool helps to resolve add-on components
   352 that are not bundled, or referenced from a bare-bones repository
   353 version of Isabelle.
   354 
   355 * The "isabelle install" tool has now a simpler command-line.  Minor
   356 INCOMPATIBILITY.
   357 
   358 * Discontinued support for Poly/ML 5.2.1, which was the last version
   359 without exception positions and advanced ML compiler/toplevel
   360 configuration.
   361 
   362 * Discontinued special treatment of Proof General -- no longer guess
   363 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   364 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   365 settings manually, or use a Proof General version that has been
   366 bundled as Isabelle component.
   367 
   368 * Settings variable ISABELLE_PLATFORM_FAMILY refers to the general
   369 platform family: "linux", "macos", "windows".
   370 
   371 
   372 New in Isabelle2012 (May 2012)
   373 ------------------------------
   374 
   375 *** General ***
   376 
   377 * Prover IDE (PIDE) improvements:
   378 
   379   - more robust Sledgehammer integration (as before the sledgehammer
   380     command-line needs to be typed into the source buffer)
   381   - markup for bound variables
   382   - markup for types of term variables (displayed as tooltips)
   383   - support for user-defined Isar commands within the running session
   384   - improved support for Unicode outside original 16bit range
   385     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   386 
   387 * Forward declaration of outer syntax keywords within the theory
   388 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   389 commands to be used in the same theory where defined.
   390 
   391 * Auxiliary contexts indicate block structure for specifications with
   392 additional parameters and assumptions.  Such unnamed contexts may be
   393 nested within other targets, like 'theory', 'locale', 'class',
   394 'instantiation' etc.  Results from the local context are generalized
   395 accordingly and applied to the enclosing target context.  Example:
   396 
   397   context
   398     fixes x y z :: 'a
   399     assumes xy: "x = y" and yz: "y = z"
   400   begin
   401 
   402   lemma my_trans: "x = z" using xy yz by simp
   403 
   404   end
   405 
   406   thm my_trans
   407 
   408 The most basic application is to factor-out context elements of
   409 several fixes/assumes/shows theorem statements, e.g. see
   410 ~~/src/HOL/Isar_Examples/Group_Context.thy
   411 
   412 Any other local theory specification element works within the "context
   413 ... begin ... end" block as well.
   414 
   415 * Bundled declarations associate attributed fact expressions with a
   416 given name in the context.  These may be later included in other
   417 contexts.  This allows to manage context extensions casually, without
   418 the logical dependencies of locales and locale interpretation.  See
   419 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   420 
   421 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   422 declaration, and results are standardized before being stored.  Thus
   423 old-style "standard" after instantiation or composition of facts
   424 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   425 indices of schematic variables.
   426 
   427 * Rule attributes in local theory declarations (e.g. locale or class)
   428 are now statically evaluated: the resulting theorem is stored instead
   429 of the original expression.  INCOMPATIBILITY in rare situations, where
   430 the historic accident of dynamic re-evaluation in interpretations
   431 etc. was exploited.
   432 
   433 * New tutorial "Programming and Proving in Isabelle/HOL"
   434 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   435 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   436 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   437 Logic" as the recommended beginners tutorial, but does not cover all
   438 of the material of that old tutorial.
   439 
   440 * Updated and extended reference manuals: "isar-ref",
   441 "implementation", "system"; reduced remaining material in old "ref"
   442 manual.
   443 
   444 
   445 *** Pure ***
   446 
   447 * Command 'definition' no longer exports the foundational "raw_def"
   448 into the user context.  Minor INCOMPATIBILITY, may use the regular
   449 "def" result with attribute "abs_def" to imitate the old version.
   450 
   451 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   452 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   453 expand it.  This also works for object-logic equality.  (Formerly
   454 undocumented feature.)
   455 
   456 * Sort constraints are now propagated in simultaneous statements, just
   457 like type constraints.  INCOMPATIBILITY in rare situations, where
   458 distinct sorts used to be assigned accidentally.  For example:
   459 
   460   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   461 
   462   lemma "P (x::'a)" and "Q (y::'a::bar)"
   463     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   464 
   465 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   466 tolerant against multiple unifiers, as long as the final result is
   467 unique.  (As before, rules are composed in canonical right-to-left
   468 order to accommodate newly introduced premises.)
   469 
   470 * Renamed some inner syntax categories:
   471 
   472     num ~> num_token
   473     xnum ~> xnum_token
   474     xstr ~> str_token
   475 
   476 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   477 "num_position" etc. are mainly used instead (which also include
   478 position information via constraints).
   479 
   480 * Simplified configuration options for syntax ambiguity: see
   481 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   482 manual.  Minor INCOMPATIBILITY.
   483 
   484 * Discontinued configuration option "syntax_positions": atomic terms
   485 in parse trees are always annotated by position constraints.
   486 
   487 * Old code generator for SML and its commands 'code_module',
   488 'code_library', 'consts_code', 'types_code' have been discontinued.
   489 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   490 
   491 * Redundant attribute "code_inline" has been discontinued. Use
   492 "code_unfold" instead.  INCOMPATIBILITY.
   493 
   494 * Dropped attribute "code_unfold_post" in favor of the its dual
   495 "code_abbrev", which yields a common pattern in definitions like
   496 
   497   definition [code_abbrev]: "f = t"
   498 
   499 INCOMPATIBILITY.
   500 
   501 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   502 instead.  INCOMPATIBILITY.
   503 
   504 * Discontinued old "prems" fact, which used to refer to the accidental
   505 collection of foundational premises in the context (already marked as
   506 legacy since Isabelle2011).
   507 
   508 
   509 *** HOL ***
   510 
   511 * Type 'a set is now a proper type constructor (just as before
   512 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   513 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   514 sets separate, it is often sufficient to rephrase some set S that has
   515 been accidentally used as predicates by "%x. x : S", and some
   516 predicate P that has been accidentally used as set by "{x. P x}".
   517 Corresponding proofs in a first step should be pruned from any
   518 tinkering with former theorems mem_def and Collect_def as far as
   519 possible.
   520 
   521 For developments which deliberately mix predicates and sets, a
   522 planning step is necessary to determine what should become a predicate
   523 and what a set.  It can be helpful to carry out that step in
   524 Isabelle2011-1 before jumping right into the current release.
   525 
   526 * Code generation by default implements sets as container type rather
   527 than predicates.  INCOMPATIBILITY.
   528 
   529 * New type synonym 'a rel = ('a * 'a) set
   530 
   531 * The representation of numerals has changed.  Datatype "num"
   532 represents strictly positive binary numerals, along with functions
   533 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   534 positive and negated numeric literals, respectively.  See also
   535 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   536 user theories may require adaptations as follows:
   537 
   538   - Theorems with number_ring or number_semiring constraints: These
   539     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   540 
   541   - Theories defining numeric types: Remove number, number_semiring,
   542     and number_ring instances. Defer all theorems about numerals until
   543     after classes one and semigroup_add have been instantiated.
   544 
   545   - Numeral-only simp rules: Replace each rule having a "number_of v"
   546     pattern with two copies, one for numeral and one for neg_numeral.
   547 
   548   - Theorems about subclasses of semiring_1 or ring_1: These classes
   549     automatically support numerals now, so more simp rules and
   550     simprocs may now apply within the proof.
   551 
   552   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   553     Redefine using other integer operations.
   554 
   555 * Transfer: New package intended to generalize the existing
   556 "descending" method and related theorem attributes from the Quotient
   557 package.  (Not all functionality is implemented yet, but future
   558 development will focus on Transfer as an eventual replacement for the
   559 corresponding parts of the Quotient package.)
   560 
   561   - transfer_rule attribute: Maintains a collection of transfer rules,
   562     which relate constants at two different types. Transfer rules may
   563     relate different type instances of the same polymorphic constant,
   564     or they may relate an operation on a raw type to a corresponding
   565     operation on an abstract type (quotient or subtype). For example:
   566 
   567     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   568     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   569 
   570   - transfer method: Replaces a subgoal on abstract types with an
   571     equivalent subgoal on the corresponding raw types. Constants are
   572     replaced with corresponding ones according to the transfer rules.
   573     Goals are generalized over all free variables by default; this is
   574     necessary for variables whose types change, but can be overridden
   575     for specific variables with e.g. "transfer fixing: x y z".  The
   576     variant transfer' method allows replacing a subgoal with one that
   577     is logically stronger (rather than equivalent).
   578 
   579   - relator_eq attribute: Collects identity laws for relators of
   580     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   581     transfer method uses these lemmas to infer transfer rules for
   582     non-polymorphic constants on the fly.
   583 
   584   - transfer_prover method: Assists with proving a transfer rule for a
   585     new constant, provided the constant is defined in terms of other
   586     constants that already have transfer rules. It should be applied
   587     after unfolding the constant definitions.
   588 
   589   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   590     from type nat to type int.
   591 
   592 * Lifting: New package intended to generalize the quotient_definition
   593 facility of the Quotient package; designed to work with Transfer.
   594 
   595   - lift_definition command: Defines operations on an abstract type in
   596     terms of a corresponding operation on a representation
   597     type.  Example syntax:
   598 
   599     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   600       is List.insert
   601 
   602     Users must discharge a respectfulness proof obligation when each
   603     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   604     the proof is discharged automatically.) The obligation is
   605     presented in a user-friendly, readable form; a respectfulness
   606     theorem in the standard format and a transfer rule are generated
   607     by the package.
   608 
   609   - Integration with code_abstype: For typedefs (e.g. subtypes
   610     corresponding to a datatype invariant, such as dlist),
   611     lift_definition generates a code certificate theorem and sets up
   612     code generation for each constant.
   613 
   614   - setup_lifting command: Sets up the Lifting package to work with a
   615     user-defined type. The user must provide either a quotient theorem
   616     or a type_definition theorem.  The package configures transfer
   617     rules for equality and quantifiers on the type, and sets up the
   618     lift_definition command to work with the type.
   619 
   620   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   621     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   622     Word/Word.thy and Library/Float.thy.
   623 
   624 * Quotient package:
   625 
   626   - The 'quotient_type' command now supports a 'morphisms' option with
   627     rep and abs functions, similar to typedef.
   628 
   629   - 'quotient_type' sets up new types to work with the Lifting and
   630     Transfer packages, as with 'setup_lifting'.
   631 
   632   - The 'quotient_definition' command now requires the user to prove a
   633     respectfulness property at the point where the constant is
   634     defined, similar to lift_definition; INCOMPATIBILITY.
   635 
   636   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   637     accordingly, INCOMPATIBILITY.
   638 
   639 * New diagnostic command 'find_unused_assms' to find potentially
   640 superfluous assumptions in theorems using Quickcheck.
   641 
   642 * Quickcheck:
   643 
   644   - Quickcheck returns variable assignments as counterexamples, which
   645     allows to reveal the underspecification of functions under test.
   646     For example, refuting "hd xs = x", it presents the variable
   647     assignment xs = [] and x = a1 as a counterexample, assuming that
   648     any property is false whenever "hd []" occurs in it.
   649 
   650     These counterexample are marked as potentially spurious, as
   651     Quickcheck also returns "xs = []" as a counterexample to the
   652     obvious theorem "hd xs = hd xs".
   653 
   654     After finding a potentially spurious counterexample, Quickcheck
   655     continues searching for genuine ones.
   656 
   657     By default, Quickcheck shows potentially spurious and genuine
   658     counterexamples. The option "genuine_only" sets quickcheck to only
   659     show genuine counterexamples.
   660 
   661   - The command 'quickcheck_generator' creates random and exhaustive
   662     value generators for a given type and operations.
   663 
   664     It generates values by using the operations as if they were
   665     constructors of that type.
   666 
   667   - Support for multisets.
   668 
   669   - Added "use_subtype" options.
   670 
   671   - Added "quickcheck_locale" configuration to specify how to process
   672     conjectures in a locale context.
   673 
   674 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   675 and affecting 'rat' and 'real'.
   676 
   677 * Sledgehammer:
   678   - Integrated more tightly with SPASS, as described in the ITP 2012
   679     paper "More SPASS with Isabelle".
   680   - Made it try "smt" as a fallback if "metis" fails or times out.
   681   - Added support for the following provers: Alt-Ergo (via Why3 and
   682     TFF1), iProver, iProver-Eq.
   683   - Sped up the minimizer.
   684   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   685   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   686   - Renamed "sound" option to "strict".
   687 
   688 * Metis: Added possibility to specify lambda translations scheme as a
   689 parenthesized argument (e.g., "by (metis (lifting) ...)").
   690 
   691 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   692 
   693 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   694 
   695 * New "case_product" attribute to generate a case rule doing multiple
   696 case distinctions at the same time.  E.g.
   697 
   698   list.exhaust [case_product nat.exhaust]
   699 
   700 produces a rule which can be used to perform case distinction on both
   701 a list and a nat.
   702 
   703 * New "eventually_elim" method as a generalized variant of the
   704 eventually_elim* rules.  Supports structured proofs.
   705 
   706 * Typedef with implicit set definition is considered legacy.  Use
   707 "typedef (open)" form instead, which will eventually become the
   708 default.
   709 
   710 * Record: code generation can be switched off manually with
   711 
   712   declare [[record_coden = false]]  -- "default true"
   713 
   714 * Datatype: type parameters allow explicit sort constraints.
   715 
   716 * Concrete syntax for case expressions includes constraints for source
   717 positions, and thus produces Prover IDE markup for its bindings.
   718 INCOMPATIBILITY for old-style syntax translations that augment the
   719 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   720 one_case.
   721 
   722 * Clarified attribute "mono_set": pure declaration without modifying
   723 the result of the fact expression.
   724 
   725 * More default pred/set conversions on a couple of relation operations
   726 and predicates.  Added powers of predicate relations.  Consolidation
   727 of some relation theorems:
   728 
   729   converse_def ~> converse_unfold
   730   rel_comp_def ~> relcomp_unfold
   731   symp_def ~> (modified, use symp_def and sym_def instead)
   732   transp_def ~> transp_trans
   733   Domain_def ~> Domain_unfold
   734   Range_def ~> Domain_converse [symmetric]
   735 
   736 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   737 
   738 See theory "Relation" for examples for making use of pred/set
   739 conversions by means of attributes "to_set" and "to_pred".
   740 
   741 INCOMPATIBILITY.
   742 
   743 * Renamed facts about the power operation on relations, i.e., relpow
   744 to match the constant's name:
   745 
   746   rel_pow_1 ~> relpow_1
   747   rel_pow_0_I ~> relpow_0_I
   748   rel_pow_Suc_I ~> relpow_Suc_I
   749   rel_pow_Suc_I2 ~> relpow_Suc_I2
   750   rel_pow_0_E ~> relpow_0_E
   751   rel_pow_Suc_E ~> relpow_Suc_E
   752   rel_pow_E ~> relpow_E
   753   rel_pow_Suc_D2 ~> relpow_Suc_D2
   754   rel_pow_Suc_E2 ~> relpow_Suc_E2
   755   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   756   rel_pow_E2 ~> relpow_E2
   757   rel_pow_add ~> relpow_add
   758   rel_pow_commute ~> relpow
   759   rel_pow_empty ~> relpow_empty:
   760   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   761   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   762   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   763   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   764   rel_pow_fun_conv ~> relpow_fun_conv
   765   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   766   rel_pow_finite_bounded ~> relpow_finite_bounded
   767   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   768   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   769   single_valued_rel_pow ~> single_valued_relpow
   770 
   771 INCOMPATIBILITY.
   772 
   773 * Theory Relation: Consolidated constant name for relation composition
   774 and corresponding theorem names:
   775 
   776   - Renamed constant rel_comp to relcomp.
   777 
   778   - Dropped abbreviation pred_comp. Use relcompp instead.
   779 
   780   - Renamed theorems:
   781 
   782     rel_compI ~> relcompI
   783     rel_compEpair ~> relcompEpair
   784     rel_compE ~> relcompE
   785     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   786     rel_comp_empty1 ~> relcomp_empty1
   787     rel_comp_mono ~> relcomp_mono
   788     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   789     rel_comp_distrib ~> relcomp_distrib
   790     rel_comp_distrib2 ~> relcomp_distrib2
   791     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   792     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   793     single_valued_rel_comp ~> single_valued_relcomp
   794     rel_comp_def ~> relcomp_unfold
   795     converse_rel_comp ~> converse_relcomp
   796     pred_compI ~> relcomppI
   797     pred_compE ~> relcomppE
   798     pred_comp_bot1 ~> relcompp_bot1
   799     pred_comp_bot2 ~> relcompp_bot2
   800     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   801     pred_comp_mono ~> relcompp_mono
   802     pred_comp_distrib ~> relcompp_distrib
   803     pred_comp_distrib2 ~> relcompp_distrib2
   804     converse_pred_comp ~> converse_relcompp
   805 
   806     finite_rel_comp ~> finite_relcomp
   807 
   808     set_rel_comp ~> set_relcomp
   809 
   810 INCOMPATIBILITY.
   811 
   812 * Theory Divides: Discontinued redundant theorems about div and mod.
   813 INCOMPATIBILITY, use the corresponding generic theorems instead.
   814 
   815   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   816   zdiv_self ~> div_self
   817   zmod_self ~> mod_self
   818   zdiv_zero ~> div_0
   819   zmod_zero ~> mod_0
   820   zdiv_zmod_equality ~> div_mod_equality2
   821   zdiv_zmod_equality2 ~> div_mod_equality
   822   zmod_zdiv_trivial ~> mod_div_trivial
   823   zdiv_zminus_zminus ~> div_minus_minus
   824   zmod_zminus_zminus ~> mod_minus_minus
   825   zdiv_zminus2 ~> div_minus_right
   826   zmod_zminus2 ~> mod_minus_right
   827   zdiv_minus1_right ~> div_minus1_right
   828   zmod_minus1_right ~> mod_minus1_right
   829   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   830   zmod_zmult1_eq ~> mod_mult_right_eq
   831   zpower_zmod ~> power_mod
   832   zdvd_zmod ~> dvd_mod
   833   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   834   mod_mult_distrib ~> mult_mod_left
   835   mod_mult_distrib2 ~> mult_mod_right
   836 
   837 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   838 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   839 
   840 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   841 
   842 * Consolidated theorem names concerning fold combinators:
   843 
   844   inf_INFI_fold_inf ~> inf_INF_fold_inf
   845   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   846   INFI_fold_inf ~> INF_fold_inf
   847   SUPR_fold_sup ~> SUP_fold_sup
   848   union_set ~> union_set_fold
   849   minus_set ~> minus_set_fold
   850   INFI_set_fold ~> INF_set_fold
   851   SUPR_set_fold ~> SUP_set_fold
   852   INF_code ~> INF_set_foldr
   853   SUP_code ~> SUP_set_foldr
   854   foldr.simps ~> foldr.simps (in point-free formulation)
   855   foldr_fold_rev ~> foldr_conv_fold
   856   foldl_fold ~> foldl_conv_fold
   857   foldr_foldr ~> foldr_conv_foldl
   858   foldl_foldr ~> foldl_conv_foldr
   859   fold_set_remdups ~> fold_set_fold_remdups
   860   fold_set ~> fold_set_fold
   861   fold1_set ~> fold1_set_fold
   862 
   863 INCOMPATIBILITY.
   864 
   865 * Dropped rarely useful theorems concerning fold combinators:
   866 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   867 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   868 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   869 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   870 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   871 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   872 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   873 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   874 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   875 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   876 
   877 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   878 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   879 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   880 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   881 lemmas over fold rather than foldr, or make use of lemmas
   882 fold_conv_foldr and fold_rev.
   883 
   884 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   885 through option types.
   886 
   887 * "Transitive_Closure.ntrancl": bounded transitive closure on
   888 relations.
   889 
   890 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   891 
   892 * Theory Int: Discontinued many legacy theorems specific to type int.
   893 INCOMPATIBILITY, use the corresponding generic theorems instead.
   894 
   895   zminus_zminus ~> minus_minus
   896   zminus_0 ~> minus_zero
   897   zminus_zadd_distrib ~> minus_add_distrib
   898   zadd_commute ~> add_commute
   899   zadd_assoc ~> add_assoc
   900   zadd_left_commute ~> add_left_commute
   901   zadd_ac ~> add_ac
   902   zmult_ac ~> mult_ac
   903   zadd_0 ~> add_0_left
   904   zadd_0_right ~> add_0_right
   905   zadd_zminus_inverse2 ~> left_minus
   906   zmult_zminus ~> mult_minus_left
   907   zmult_commute ~> mult_commute
   908   zmult_assoc ~> mult_assoc
   909   zadd_zmult_distrib ~> left_distrib
   910   zadd_zmult_distrib2 ~> right_distrib
   911   zdiff_zmult_distrib ~> left_diff_distrib
   912   zdiff_zmult_distrib2 ~> right_diff_distrib
   913   zmult_1 ~> mult_1_left
   914   zmult_1_right ~> mult_1_right
   915   zle_refl ~> order_refl
   916   zle_trans ~> order_trans
   917   zle_antisym ~> order_antisym
   918   zle_linear ~> linorder_linear
   919   zless_linear ~> linorder_less_linear
   920   zadd_left_mono ~> add_left_mono
   921   zadd_strict_right_mono ~> add_strict_right_mono
   922   zadd_zless_mono ~> add_less_le_mono
   923   int_0_less_1 ~> zero_less_one
   924   int_0_neq_1 ~> zero_neq_one
   925   zless_le ~> less_le
   926   zpower_zadd_distrib ~> power_add
   927   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   928   zero_le_zpower_abs ~> zero_le_power_abs
   929 
   930 * Theory Deriv: Renamed
   931 
   932   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   933 
   934 * Theory Library/Multiset: Improved code generation of multisets.
   935 
   936 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   937 are expressed via type classes again. The special syntax
   938 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   939 setsum_set, which is now subsumed by Big_Operators.setsum.
   940 INCOMPATIBILITY.
   941 
   942 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   943 use theory HOL/Library/Nat_Bijection instead.
   944 
   945 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   946 trees is now inside a type class context.  Names of affected
   947 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   948 theories working directly with raw red-black trees, adapt the names as
   949 follows:
   950 
   951   Operations:
   952   bulkload -> rbt_bulkload
   953   del_from_left -> rbt_del_from_left
   954   del_from_right -> rbt_del_from_right
   955   del -> rbt_del
   956   delete -> rbt_delete
   957   ins -> rbt_ins
   958   insert -> rbt_insert
   959   insertw -> rbt_insert_with
   960   insert_with_key -> rbt_insert_with_key
   961   map_entry -> rbt_map_entry
   962   lookup -> rbt_lookup
   963   sorted -> rbt_sorted
   964   tree_greater -> rbt_greater
   965   tree_less -> rbt_less
   966   tree_less_symbol -> rbt_less_symbol
   967   union -> rbt_union
   968   union_with -> rbt_union_with
   969   union_with_key -> rbt_union_with_key
   970 
   971   Lemmas:
   972   balance_left_sorted -> balance_left_rbt_sorted
   973   balance_left_tree_greater -> balance_left_rbt_greater
   974   balance_left_tree_less -> balance_left_rbt_less
   975   balance_right_sorted -> balance_right_rbt_sorted
   976   balance_right_tree_greater -> balance_right_rbt_greater
   977   balance_right_tree_less -> balance_right_rbt_less
   978   balance_sorted -> balance_rbt_sorted
   979   balance_tree_greater -> balance_rbt_greater
   980   balance_tree_less -> balance_rbt_less
   981   bulkload_is_rbt -> rbt_bulkload_is_rbt
   982   combine_sorted -> combine_rbt_sorted
   983   combine_tree_greater -> combine_rbt_greater
   984   combine_tree_less -> combine_rbt_less
   985   delete_in_tree -> rbt_delete_in_tree
   986   delete_is_rbt -> rbt_delete_is_rbt
   987   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   988   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   989   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   990   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   991   del_in_tree -> rbt_del_in_tree
   992   del_inv1_inv2 -> rbt_del_inv1_inv2
   993   del_sorted -> rbt_del_rbt_sorted
   994   del_tree_greater -> rbt_del_rbt_greater
   995   del_tree_less -> rbt_del_rbt_less
   996   dom_lookup_Branch -> dom_rbt_lookup_Branch
   997   entries_lookup -> entries_rbt_lookup
   998   finite_dom_lookup -> finite_dom_rbt_lookup
   999   insert_sorted -> rbt_insert_rbt_sorted
  1000   insertw_is_rbt -> rbt_insertw_is_rbt
  1001   insertwk_is_rbt -> rbt_insertwk_is_rbt
  1002   insertwk_sorted -> rbt_insertwk_rbt_sorted
  1003   insertw_sorted -> rbt_insertw_rbt_sorted
  1004   ins_sorted -> ins_rbt_sorted
  1005   ins_tree_greater -> ins_rbt_greater
  1006   ins_tree_less -> ins_rbt_less
  1007   is_rbt_sorted -> is_rbt_rbt_sorted
  1008   lookup_balance -> rbt_lookup_balance
  1009   lookup_bulkload -> rbt_lookup_rbt_bulkload
  1010   lookup_delete -> rbt_lookup_rbt_delete
  1011   lookup_Empty -> rbt_lookup_Empty
  1012   lookup_from_in_tree -> rbt_lookup_from_in_tree
  1013   lookup_in_tree -> rbt_lookup_in_tree
  1014   lookup_ins -> rbt_lookup_ins
  1015   lookup_insert -> rbt_lookup_rbt_insert
  1016   lookup_insertw -> rbt_lookup_rbt_insertw
  1017   lookup_insertwk -> rbt_lookup_rbt_insertwk
  1018   lookup_keys -> rbt_lookup_keys
  1019   lookup_map -> rbt_lookup_map
  1020   lookup_map_entry -> rbt_lookup_rbt_map_entry
  1021   lookup_tree_greater -> rbt_lookup_rbt_greater
  1022   lookup_tree_less -> rbt_lookup_rbt_less
  1023   lookup_union -> rbt_lookup_rbt_union
  1024   map_entry_color_of -> rbt_map_entry_color_of
  1025   map_entry_inv1 -> rbt_map_entry_inv1
  1026   map_entry_inv2 -> rbt_map_entry_inv2
  1027   map_entry_is_rbt -> rbt_map_entry_is_rbt
  1028   map_entry_sorted -> rbt_map_entry_rbt_sorted
  1029   map_entry_tree_greater -> rbt_map_entry_rbt_greater
  1030   map_entry_tree_less -> rbt_map_entry_rbt_less
  1031   map_tree_greater -> map_rbt_greater
  1032   map_tree_less -> map_rbt_less
  1033   map_sorted -> map_rbt_sorted
  1034   paint_sorted -> paint_rbt_sorted
  1035   paint_lookup -> paint_rbt_lookup
  1036   paint_tree_greater -> paint_rbt_greater
  1037   paint_tree_less -> paint_rbt_less
  1038   sorted_entries -> rbt_sorted_entries
  1039   tree_greater_eq_trans -> rbt_greater_eq_trans
  1040   tree_greater_nit -> rbt_greater_nit
  1041   tree_greater_prop -> rbt_greater_prop
  1042   tree_greater_simps -> rbt_greater_simps
  1043   tree_greater_trans -> rbt_greater_trans
  1044   tree_less_eq_trans -> rbt_less_eq_trans
  1045   tree_less_nit -> rbt_less_nit
  1046   tree_less_prop -> rbt_less_prop
  1047   tree_less_simps -> rbt_less_simps
  1048   tree_less_trans -> rbt_less_trans
  1049   tree_ord_props -> rbt_ord_props
  1050   union_Branch -> rbt_union_Branch
  1051   union_is_rbt -> rbt_union_is_rbt
  1052   unionw_is_rbt -> rbt_unionw_is_rbt
  1053   unionwk_is_rbt -> rbt_unionwk_is_rbt
  1054   unionwk_sorted -> rbt_unionwk_rbt_sorted
  1055 
  1056 * Theory HOL/Library/Float: Floating point numbers are now defined as
  1057 a subset of the real numbers.  All operations are defined using the
  1058 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
  1059 
  1060   Changed Operations:
  1061   float_abs -> abs
  1062   float_nprt -> nprt
  1063   float_pprt -> pprt
  1064   pow2 -> use powr
  1065   round_down -> float_round_down
  1066   round_up -> float_round_up
  1067   scale -> exponent
  1068 
  1069   Removed Operations:
  1070   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
  1071 
  1072   Renamed Lemmas:
  1073   abs_float_def -> Float.compute_float_abs
  1074   bitlen_ge0 -> bitlen_nonneg
  1075   bitlen.simps -> Float.compute_bitlen
  1076   float_components -> Float_mantissa_exponent
  1077   float_divl.simps -> Float.compute_float_divl
  1078   float_divr.simps -> Float.compute_float_divr
  1079   float_eq_odd -> mult_powr_eq_mult_powr_iff
  1080   float_power -> real_of_float_power
  1081   lapprox_posrat_def -> Float.compute_lapprox_posrat
  1082   lapprox_rat.simps -> Float.compute_lapprox_rat
  1083   le_float_def' -> Float.compute_float_le
  1084   le_float_def -> less_eq_float.rep_eq
  1085   less_float_def' -> Float.compute_float_less
  1086   less_float_def -> less_float.rep_eq
  1087   normfloat_def -> Float.compute_normfloat
  1088   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
  1089   normfloat -> normfloat_def
  1090   normfloat_unique -> use normfloat_def
  1091   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
  1092   one_float_def -> Float.compute_float_one
  1093   plus_float_def -> Float.compute_float_plus
  1094   rapprox_posrat_def -> Float.compute_rapprox_posrat
  1095   rapprox_rat.simps -> Float.compute_rapprox_rat
  1096   real_of_float_0 -> zero_float.rep_eq
  1097   real_of_float_1 -> one_float.rep_eq
  1098   real_of_float_abs -> abs_float.rep_eq
  1099   real_of_float_add -> plus_float.rep_eq
  1100   real_of_float_minus -> uminus_float.rep_eq
  1101   real_of_float_mult -> times_float.rep_eq
  1102   real_of_float_simp -> Float.rep_eq
  1103   real_of_float_sub -> minus_float.rep_eq
  1104   round_down.simps -> Float.compute_float_round_down
  1105   round_up.simps -> Float.compute_float_round_up
  1106   times_float_def -> Float.compute_float_times
  1107   uminus_float_def -> Float.compute_float_uminus
  1108   zero_float_def -> Float.compute_float_zero
  1109 
  1110   Lemmas not necessary anymore, use the transfer method:
  1111   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
  1112   float_divr, float_le_simp, float_less1_mantissa_bound,
  1113   float_less_simp, float_less_zero, float_le_zero,
  1114   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
  1115   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
  1116   lapprox_rat_bottom, normalized_float, rapprox_posrat,
  1117   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
  1118   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
  1119   round_up, zero_le_float, zero_less_float
  1120 
  1121 * New theory HOL/Library/DAList provides an abstract type for
  1122 association lists with distinct keys.
  1123 
  1124 * Session HOL/IMP: Added new theory of abstract interpretation of
  1125 annotated commands.
  1126 
  1127 * Session HOL-Import: Re-implementation from scratch is faster,
  1128 simpler, and more scalable.  Requires a proof bundle, which is
  1129 available as an external component.  Discontinued old (and mostly
  1130 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
  1131 
  1132 * Session HOL-Word: Discontinued many redundant theorems specific to
  1133 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
  1134 instead.
  1135 
  1136   word_sub_alt ~> word_sub_wi
  1137   word_add_alt ~> word_add_def
  1138   word_mult_alt ~> word_mult_def
  1139   word_minus_alt ~> word_minus_def
  1140   word_0_alt ~> word_0_wi
  1141   word_1_alt ~> word_1_wi
  1142   word_add_0 ~> add_0_left
  1143   word_add_0_right ~> add_0_right
  1144   word_mult_1 ~> mult_1_left
  1145   word_mult_1_right ~> mult_1_right
  1146   word_add_commute ~> add_commute
  1147   word_add_assoc ~> add_assoc
  1148   word_add_left_commute ~> add_left_commute
  1149   word_mult_commute ~> mult_commute
  1150   word_mult_assoc ~> mult_assoc
  1151   word_mult_left_commute ~> mult_left_commute
  1152   word_left_distrib ~> left_distrib
  1153   word_right_distrib ~> right_distrib
  1154   word_left_minus ~> left_minus
  1155   word_diff_0_right ~> diff_0_right
  1156   word_diff_self ~> diff_self
  1157   word_sub_def ~> diff_minus
  1158   word_diff_minus ~> diff_minus
  1159   word_add_ac ~> add_ac
  1160   word_mult_ac ~> mult_ac
  1161   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1162   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1163   word_order_trans ~> order_trans
  1164   word_order_refl ~> order_refl
  1165   word_order_antisym ~> order_antisym
  1166   word_order_linear ~> linorder_linear
  1167   lenw1_zero_neq_one ~> zero_neq_one
  1168   word_number_of_eq ~> number_of_eq
  1169   word_of_int_add_hom ~> wi_hom_add
  1170   word_of_int_sub_hom ~> wi_hom_sub
  1171   word_of_int_mult_hom ~> wi_hom_mult
  1172   word_of_int_minus_hom ~> wi_hom_neg
  1173   word_of_int_succ_hom ~> wi_hom_succ
  1174   word_of_int_pred_hom ~> wi_hom_pred
  1175   word_of_int_0_hom ~> word_0_wi
  1176   word_of_int_1_hom ~> word_1_wi
  1177 
  1178 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1179 machine word equalities and inequalities into logical circuits,
  1180 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1181 multiplication, shifting by constants, bitwise operators and numeric
  1182 constants.  Requires fixed-length word types, not 'a word.  Solves
  1183 many standard word identities outright and converts more into first
  1184 order problems amenable to blast or similar.  See also examples in
  1185 HOL/Word/Examples/WordExamples.thy.
  1186 
  1187 * Session HOL-Probability: Introduced the type "'a measure" to
  1188 represent measures, this replaces the records 'a algebra and 'a
  1189 measure_space.  The locales based on subset_class now have two
  1190 locale-parameters the space \<Omega> and the set of measurable sets M.
  1191 The product of probability spaces uses now the same constant as the
  1192 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1193 measure".  Most constants are defined now outside of locales and gain
  1194 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1195 Measure space constructions for distributions and densities now got
  1196 their own constants distr and density.  Instead of using locales to
  1197 describe measure spaces with a finite space, the measure count_space
  1198 and point_measure is introduced.  INCOMPATIBILITY.
  1199 
  1200   Renamed constants:
  1201   measure -> emeasure
  1202   finite_measure.\<mu>' -> measure
  1203   product_algebra_generator -> prod_algebra
  1204   product_prob_space.emb -> prod_emb
  1205   product_prob_space.infprod_algebra -> PiM
  1206 
  1207   Removed locales:
  1208   completeable_measure_space
  1209   finite_measure_space
  1210   finite_prob_space
  1211   finite_product_finite_prob_space
  1212   finite_product_sigma_algebra
  1213   finite_sigma_algebra
  1214   measure_space
  1215   pair_finite_prob_space
  1216   pair_finite_sigma_algebra
  1217   pair_finite_space
  1218   pair_sigma_algebra
  1219   product_sigma_algebra
  1220 
  1221   Removed constants:
  1222   conditional_space
  1223   distribution -> use distr measure, or distributed predicate
  1224   image_space
  1225   joint_distribution -> use distr measure, or distributed predicate
  1226   pair_measure_generator
  1227   product_prob_space.infprod_algebra -> use PiM
  1228   subvimage
  1229 
  1230   Replacement theorems:
  1231   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1232   finite_measure.empty_measure -> measure_empty
  1233   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1234   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1235   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1236   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1237   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1238   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1239   finite_measure.positive_measure' -> measure_nonneg
  1240   finite_measure.real_measure -> finite_measure.emeasure_real
  1241   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1242   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1243   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1244   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1245   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1246   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1247   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1248   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1249   information_space.entropy_commute -> information_space.entropy_commute_simple
  1250   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1251   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1252   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1253   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1254   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1255   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1256   information_space.KL_ge_0 -> information_space.KL_nonneg
  1257   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1258   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1259   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1260   Int_stable_product_algebra_generator -> positive_integral
  1261   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1262   measure_space.additive -> emeasure_additive
  1263   measure_space.AE_iff_null_set -> AE_iff_null
  1264   measure_space.almost_everywhere_def -> eventually_ae_filter
  1265   measure_space.almost_everywhere_vimage -> AE_distrD
  1266   measure_space.continuity_from_above -> INF_emeasure_decseq
  1267   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1268   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1269   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1270   measure_space_density -> emeasure_density
  1271   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1272   measure_space.integrable_vimage -> integrable_distr
  1273   measure_space.integral_translated_density -> integral_density
  1274   measure_space.integral_vimage -> integral_distr
  1275   measure_space.measure_additive -> plus_emeasure
  1276   measure_space.measure_compl -> emeasure_compl
  1277   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1278   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1279   measure_space.measure_decseq -> decseq_emeasure
  1280   measure_space.measure_Diff -> emeasure_Diff
  1281   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1282   measure_space.measure_eq_0 -> emeasure_eq_0
  1283   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1284   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1285   measure_space.measure_incseq -> incseq_emeasure
  1286   measure_space.measure_insert -> emeasure_insert
  1287   measure_space.measure_mono -> emeasure_mono
  1288   measure_space.measure_not_negative -> emeasure_not_MInf
  1289   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1290   measure_space.measure_setsum -> setsum_emeasure
  1291   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1292   measure_space.measure_space_vimage -> emeasure_distr
  1293   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1294   measure_space.measure_subadditive -> subadditive
  1295   measure_space.measure_top -> emeasure_space
  1296   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1297   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1298   measure_space.positive_integral_translated_density -> positive_integral_density
  1299   measure_space.positive_integral_vimage -> positive_integral_distr
  1300   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1301   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1302   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1303   measure_space.real_measure_Diff -> measure_Diff
  1304   measure_space.real_measure_finite_Union -> measure_finite_Union
  1305   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1306   measure_space.real_measure_subadditive -> measure_subadditive
  1307   measure_space.real_measure_Union -> measure_Union
  1308   measure_space.real_measure_UNION -> measure_UNION
  1309   measure_space.simple_function_vimage -> simple_function_comp
  1310   measure_space.simple_integral_vimage -> simple_integral_distr
  1311   measure_space.simple_integral_vimage -> simple_integral_distr
  1312   measure_unique_Int_stable -> measure_eqI_generator_eq
  1313   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1314   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1315   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1316   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1317   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1318   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1319   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1320   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1321   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1322   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1323   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1324   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1325   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1326   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1327   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1328   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1329   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1330   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1331   prob_space.prob_space_vimage -> prob_space_distr
  1332   prob_space.random_variable_restrict -> measurable_restrict
  1333   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1334   product_algebraE -> prod_algebraE_all
  1335   product_algebra_generator_der -> prod_algebra_eq_finite
  1336   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1337   product_algebraI -> sets_PiM_I_finite
  1338   product_measure_exists -> product_sigma_finite.sigma_finite
  1339   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1340   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1341   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1342   product_prob_space.measurable_component -> measurable_component_singleton
  1343   product_prob_space.measurable_emb -> measurable_prod_emb
  1344   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1345   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1346   product_prob_space.measure_emb -> emeasure_prod_emb
  1347   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1348   product_sigma_algebra.product_algebra_into_space -> space_closed
  1349   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1350   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1351   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1352   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1353   sets_product_algebra -> sets_PiM
  1354   sigma_algebra.measurable_sigma -> measurable_measure_of
  1355   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1356   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1357   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1358   space_product_algebra -> space_PiM
  1359 
  1360 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1361 languages) into Isabelle/HOL.
  1362 
  1363 
  1364 *** FOL ***
  1365 
  1366 * New "case_product" attribute (see HOL).
  1367 
  1368 
  1369 *** ZF ***
  1370 
  1371 * Greater support for structured proofs involving induction or case
  1372 analysis.
  1373 
  1374 * Much greater use of mathematical symbols.
  1375 
  1376 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1377 
  1378 
  1379 *** ML ***
  1380 
  1381 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1382 from a minor keyword introduced via theory header declaration.
  1383 
  1384 * Antiquotation @{command_spec "name"} produces the
  1385 Outer_Syntax.command_spec from a major keyword introduced via theory
  1386 header declaration; it can be passed to Outer_Syntax.command etc.
  1387 
  1388 * Local_Theory.define no longer hard-wires default theorem name
  1389 "foo_def", but retains the binding as given.  If that is Binding.empty
  1390 / Attrib.empty_binding, the result is not registered as user-level
  1391 fact.  The Local_Theory.define_internal variant allows to specify a
  1392 non-empty name (used for the foundation in the background theory),
  1393 while omitting the fact binding in the user-context.  Potential
  1394 INCOMPATIBILITY for derived definitional packages: need to specify
  1395 naming policy for primitive definitions more explicitly.
  1396 
  1397 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1398 conformance with similar operations in structure Term and Logic.
  1399 
  1400 * Antiquotation @{attributes [...]} embeds attribute source
  1401 representation into the ML text, which is particularly useful with
  1402 declarations like Local_Theory.note.
  1403 
  1404 * Structure Proof_Context follows standard naming scheme.  Old
  1405 ProofContext has been discontinued.  INCOMPATIBILITY.
  1406 
  1407 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1408 change of semantics: update is applied to auxiliary local theory
  1409 context as well.
  1410 
  1411 * Modernized some old-style infix operations:
  1412 
  1413   addeqcongs    ~> Simplifier.add_eqcong
  1414   deleqcongs    ~> Simplifier.del_eqcong
  1415   addcongs      ~> Simplifier.add_cong
  1416   delcongs      ~> Simplifier.del_cong
  1417   setmksimps    ~> Simplifier.set_mksimps
  1418   setmkcong     ~> Simplifier.set_mkcong
  1419   setmksym      ~> Simplifier.set_mksym
  1420   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1421   settermless   ~> Simplifier.set_termless
  1422   setsubgoaler  ~> Simplifier.set_subgoaler
  1423   addsplits     ~> Splitter.add_split
  1424   delsplits     ~> Splitter.del_split
  1425 
  1426 
  1427 *** System ***
  1428 
  1429 * USER_HOME settings variable points to cross-platform user home
  1430 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1431 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1432 of former $HOME.  A different default for USER_HOME may be set
  1433 explicitly in shell environment, before Isabelle settings are
  1434 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1435 the generic user home was intended.
  1436 
  1437 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1438 notation, which is useful for the jEdit file browser, for example.
  1439 
  1440 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1441 (not just JRE).
  1442 
  1443 
  1444 
  1445 New in Isabelle2011-1 (October 2011)
  1446 ------------------------------------
  1447 
  1448 *** General ***
  1449 
  1450 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1451 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1452 
  1453   - Management of multiple theory files directly from the editor
  1454     buffer store -- bypassing the file-system (no requirement to save
  1455     files for checking).
  1456 
  1457   - Markup of formal entities within the text buffer, with semantic
  1458     highlighting, tooltips and hyperlinks to jump to defining source
  1459     positions.
  1460 
  1461   - Improved text rendering, with sub/superscripts in the source
  1462     buffer (including support for copy/paste wrt. output panel, HTML
  1463     theory output and other non-Isabelle text boxes).
  1464 
  1465   - Refined scheduling of proof checking and printing of results,
  1466     based on interactive editor view.  (Note: jEdit folding and
  1467     narrowing allows to restrict buffer perspectives explicitly.)
  1468 
  1469   - Reduced CPU performance requirements, usable on machines with few
  1470     cores.
  1471 
  1472   - Reduced memory requirements due to pruning of unused document
  1473     versions (garbage collection).
  1474 
  1475 See also ~~/src/Tools/jEdit/README.html for further information,
  1476 including some remaining limitations.
  1477 
  1478 * Theory loader: source files are exclusively located via the master
  1479 directory of each theory node (where the .thy file itself resides).
  1480 The global load path (such as src/HOL/Library) has been discontinued.
  1481 Note that the path element ~~ may be used to reference theories in the
  1482 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1483 INCOMPATIBILITY.
  1484 
  1485 * Theory loader: source files are identified by content via SHA1
  1486 digests.  Discontinued former path/modtime identification and optional
  1487 ISABELLE_FILE_IDENT plugin scripts.
  1488 
  1489 * Parallelization of nested Isar proofs is subject to
  1490 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1491 usedir option -Q.
  1492 
  1493 * Name space: former unsynchronized references are now proper
  1494 configuration options, with more conventional names:
  1495 
  1496   long_names   ~> names_long
  1497   short_names  ~> names_short
  1498   unique_names ~> names_unique
  1499 
  1500 Minor INCOMPATIBILITY, need to declare options in context like this:
  1501 
  1502   declare [[names_unique = false]]
  1503 
  1504 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1505 that the result needs to be unique, which means fact specifications
  1506 may have to be refined after enriching a proof context.
  1507 
  1508 * Attribute "case_names" has been refined: the assumptions in each case
  1509 can be named now by following the case name with [name1 name2 ...].
  1510 
  1511 * Isabelle/Isar reference manual has been updated and extended:
  1512   - "Synopsis" provides a catalog of main Isar language concepts.
  1513   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1514   - Updated material from classic "ref" manual, notably about
  1515     "Classical Reasoner".
  1516 
  1517 
  1518 *** HOL ***
  1519 
  1520 * Class bot and top require underlying partial order rather than
  1521 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1522 
  1523 * Class complete_lattice: generalized a couple of lemmas from sets;
  1524 generalized theorems INF_cong and SUP_cong.  New type classes for
  1525 complete boolean algebras and complete linear orders.  Lemmas
  1526 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1527 class complete_linorder.
  1528 
  1529 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1530 Sup_fun_def, Inf_apply, Sup_apply.
  1531 
  1532 Removed redundant lemmas (the right hand side gives hints how to
  1533 replace them for (metis ...), or (simp only: ...) proofs):
  1534 
  1535   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1536   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1537   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1538   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1539   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1540   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1541   Inter_def ~> INF_def, image_def
  1542   Union_def ~> SUP_def, image_def
  1543   INT_eq ~> INF_def, and image_def
  1544   UN_eq ~> SUP_def, and image_def
  1545   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1546 
  1547 More consistent and comprehensive names:
  1548 
  1549   INTER_eq_Inter_image ~> INF_def
  1550   UNION_eq_Union_image ~> SUP_def
  1551   INFI_def ~> INF_def
  1552   SUPR_def ~> SUP_def
  1553   INF_leI ~> INF_lower
  1554   INF_leI2 ~> INF_lower2
  1555   le_INFI ~> INF_greatest
  1556   le_SUPI ~> SUP_upper
  1557   le_SUPI2 ~> SUP_upper2
  1558   SUP_leI ~> SUP_least
  1559   INFI_bool_eq ~> INF_bool_eq
  1560   SUPR_bool_eq ~> SUP_bool_eq
  1561   INFI_apply ~> INF_apply
  1562   SUPR_apply ~> SUP_apply
  1563   INTER_def ~> INTER_eq
  1564   UNION_def ~> UNION_eq
  1565 
  1566 INCOMPATIBILITY.
  1567 
  1568 * Renamed theory Complete_Lattice to Complete_Lattices.
  1569 INCOMPATIBILITY.
  1570 
  1571 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1572 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1573 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1574 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1575 
  1576 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1577 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1578 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1579 INCOMPATIBILITY.
  1580 
  1581 * Added syntactic classes "inf" and "sup" for the respective
  1582 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1583 (mostly internal) locale predicates for some derived classes.
  1584 
  1585 * Theorem collections ball_simps and bex_simps do not contain theorems
  1586 referring to UNION any longer; these have been moved to collection
  1587 UN_ball_bex_simps.  INCOMPATIBILITY.
  1588 
  1589 * Theory Archimedean_Field: floor now is defined as parameter of a
  1590 separate type class floor_ceiling.
  1591 
  1592 * Theory Finite_Set: more coherent development of fold_set locales:
  1593 
  1594     locale fun_left_comm ~> locale comp_fun_commute
  1595     locale fun_left_comm_idem ~> locale comp_fun_idem
  1596 
  1597 Both use point-free characterization; interpretation proofs may need
  1598 adjustment.  INCOMPATIBILITY.
  1599 
  1600 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1601 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1602 
  1603 * Theory Complex_Main: The locale interpretations for the
  1604 bounded_linear and bounded_bilinear locales have been removed, in
  1605 order to reduce the number of duplicate lemmas. Users must use the
  1606 original names for distributivity theorems, potential INCOMPATIBILITY.
  1607 
  1608   divide.add ~> add_divide_distrib
  1609   divide.diff ~> diff_divide_distrib
  1610   divide.setsum ~> setsum_divide_distrib
  1611   mult.add_right ~> right_distrib
  1612   mult.diff_right ~> right_diff_distrib
  1613   mult_right.setsum ~> setsum_right_distrib
  1614   mult_left.diff ~> left_diff_distrib
  1615 
  1616 * Theory Complex_Main: Several redundant theorems have been removed or
  1617 replaced by more general versions. INCOMPATIBILITY.
  1618 
  1619   real_diff_def ~> minus_real_def
  1620   real_divide_def ~> divide_real_def
  1621   real_less_def ~> less_le
  1622   real_abs_def ~> abs_real_def
  1623   real_sgn_def ~> sgn_real_def
  1624   real_mult_commute ~> mult_commute
  1625   real_mult_assoc ~> mult_assoc
  1626   real_mult_1 ~> mult_1_left
  1627   real_add_mult_distrib ~> left_distrib
  1628   real_zero_not_eq_one ~> zero_neq_one
  1629   real_mult_inverse_left ~> left_inverse
  1630   INVERSE_ZERO ~> inverse_zero
  1631   real_le_refl ~> order_refl
  1632   real_le_antisym ~> order_antisym
  1633   real_le_trans ~> order_trans
  1634   real_le_linear ~> linear
  1635   real_le_eq_diff ~> le_iff_diff_le_0
  1636   real_add_left_mono ~> add_left_mono
  1637   real_mult_order ~> mult_pos_pos
  1638   real_mult_less_mono2 ~> mult_strict_left_mono
  1639   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1640   real_0_le_divide_iff ~> zero_le_divide_iff
  1641   realpow_two_disj ~> power2_eq_iff
  1642   real_squared_diff_one_factored ~> square_diff_one_factored
  1643   realpow_two_diff ~> square_diff_square_factored
  1644   reals_complete2 ~> complete_real
  1645   real_sum_squared_expand ~> power2_sum
  1646   exp_ln_eq ~> ln_unique
  1647   expi_add ~> exp_add
  1648   expi_zero ~> exp_zero
  1649   lemma_DERIV_subst ~> DERIV_cong
  1650   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1651   LIMSEQ_const ~> tendsto_const
  1652   LIMSEQ_norm ~> tendsto_norm
  1653   LIMSEQ_add ~> tendsto_add
  1654   LIMSEQ_minus ~> tendsto_minus
  1655   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1656   LIMSEQ_diff ~> tendsto_diff
  1657   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1658   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1659   LIMSEQ_mult ~> tendsto_mult
  1660   LIMSEQ_inverse ~> tendsto_inverse
  1661   LIMSEQ_divide ~> tendsto_divide
  1662   LIMSEQ_pow ~> tendsto_power
  1663   LIMSEQ_setsum ~> tendsto_setsum
  1664   LIMSEQ_setprod ~> tendsto_setprod
  1665   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1666   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1667   LIMSEQ_imp_rabs ~> tendsto_rabs
  1668   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1669   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1670   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1671   LIMSEQ_Complex ~> tendsto_Complex
  1672   LIM_ident ~> tendsto_ident_at
  1673   LIM_const ~> tendsto_const
  1674   LIM_add ~> tendsto_add
  1675   LIM_add_zero ~> tendsto_add_zero
  1676   LIM_minus ~> tendsto_minus
  1677   LIM_diff ~> tendsto_diff
  1678   LIM_norm ~> tendsto_norm
  1679   LIM_norm_zero ~> tendsto_norm_zero
  1680   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1681   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1682   LIM_rabs ~> tendsto_rabs
  1683   LIM_rabs_zero ~> tendsto_rabs_zero
  1684   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1685   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1686   LIM_compose ~> tendsto_compose
  1687   LIM_mult ~> tendsto_mult
  1688   LIM_scaleR ~> tendsto_scaleR
  1689   LIM_of_real ~> tendsto_of_real
  1690   LIM_power ~> tendsto_power
  1691   LIM_inverse ~> tendsto_inverse
  1692   LIM_sgn ~> tendsto_sgn
  1693   isCont_LIM_compose ~> isCont_tendsto_compose
  1694   bounded_linear.LIM ~> bounded_linear.tendsto
  1695   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1696   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1697   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1698   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1699   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1700   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1701 
  1702 * Theory Complex_Main: The definition of infinite series was
  1703 generalized.  Now it is defined on the type class {topological_space,
  1704 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1705 
  1706 * Theory Complex_Main: The complex exponential function "expi" is now
  1707 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1708 several polymorphic lemmas about "exp" are now applicable to "expi".
  1709 
  1710 * Code generation:
  1711 
  1712   - Theory Library/Code_Char_ord provides native ordering of
  1713     characters in the target language.
  1714 
  1715   - Commands code_module and code_library are legacy, use export_code
  1716     instead.
  1717 
  1718   - Method "evaluation" is legacy, use method "eval" instead.
  1719 
  1720   - Legacy evaluator "SML" is deactivated by default.  May be
  1721     reactivated by the following theory command:
  1722 
  1723       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1724 
  1725 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1726 
  1727 * New proof method "induction" that gives induction hypotheses the
  1728 name "IH", thus distinguishing them from further hypotheses that come
  1729 from rule induction.  The latter are still called "hyps".  Method
  1730 "induction" is a thin wrapper around "induct" and follows the same
  1731 syntax.
  1732 
  1733 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1734 still available as a legacy feature for some time.
  1735 
  1736 * Nitpick:
  1737   - Added "need" and "total_consts" options.
  1738   - Reintroduced "show_skolems" option by popular demand.
  1739   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1740     INCOMPATIBILITY.
  1741 
  1742 * Sledgehammer:
  1743   - Use quasi-sound (and efficient) translations by default.
  1744   - Added support for the following provers: E-ToFoF, LEO-II,
  1745     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1746   - Automatically preplay and minimize proofs before showing them if
  1747     this can be done within reasonable time.
  1748   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1749     INCOMPATIBILITY.
  1750   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1751     "max_mono_iters", and "max_new_mono_instances" options.
  1752   - Removed "explicit_apply" and "full_types" options as well as "Full
  1753     Types" Proof General menu item. INCOMPATIBILITY.
  1754 
  1755 * Metis:
  1756   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1757   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1758     INCOMPATIBILITY.
  1759 
  1760 * Command 'try':
  1761   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1762     "elim:" options. INCOMPATIBILITY.
  1763   - Introduced 'try' that not only runs 'try_methods' but also
  1764     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1765 
  1766 * Quickcheck:
  1767   - Added "eval" option to evaluate terms for the found counterexample
  1768     (currently only supported by the default (exhaustive) tester).
  1769   - Added post-processing of terms to obtain readable counterexamples
  1770     (currently only supported by the default (exhaustive) tester).
  1771   - New counterexample generator quickcheck[narrowing] enables
  1772     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1773     with its installation location defined in the Isabelle settings
  1774     environment as ISABELLE_GHC.
  1775   - Removed quickcheck tester "SML" based on the SML code generator
  1776     (formly in HOL/Library).
  1777 
  1778 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1779 use 'partial_function' instead.
  1780 
  1781 * Theory Library/Extended_Reals replaces now the positive extended
  1782 reals found in probability theory. This file is extended by
  1783 Multivariate_Analysis/Extended_Real_Limits.
  1784 
  1785 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1786 from where it must be imported explicitly if it is really required.
  1787 INCOMPATIBILITY.
  1788 
  1789 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1790 been moved here.  INCOMPATIBILITY.
  1791 
  1792 * Theory Library/Saturated provides type of numbers with saturated
  1793 arithmetic.
  1794 
  1795 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1796 product type 'a * 'b, and provides instance proofs for various order
  1797 and lattice type classes.
  1798 
  1799 * Theory Library/Countable now provides the "countable_datatype" proof
  1800 method for proving "countable" class instances for datatypes.
  1801 
  1802 * Theory Library/Cset_Monad allows do notation for computable sets
  1803 (cset) via the generic monad ad-hoc overloading facility.
  1804 
  1805 * Library: Theories of common data structures are split into theories
  1806 for implementation, an invariant-ensuring type, and connection to an
  1807 abstract type. INCOMPATIBILITY.
  1808 
  1809   - RBT is split into RBT and RBT_Mapping.
  1810   - AssocList is split and renamed into AList and AList_Mapping.
  1811   - DList is split into DList_Impl, DList, and DList_Cset.
  1812   - Cset is split into Cset and List_Cset.
  1813 
  1814 * Theory Library/Nat_Infinity has been renamed to
  1815 Library/Extended_Nat, with name changes of the following types and
  1816 constants:
  1817 
  1818   type inat   ~> type enat
  1819   Fin         ~> enat
  1820   Infty       ~> infinity (overloaded)
  1821   iSuc        ~> eSuc
  1822   the_Fin     ~> the_enat
  1823 
  1824 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1825 been renamed accordingly. INCOMPATIBILITY.
  1826 
  1827 * Session Multivariate_Analysis: The euclidean_space type class now
  1828 fixes a constant "Basis :: 'a set" consisting of the standard
  1829 orthonormal basis for the type. Users now have the option of
  1830 quantifying over this set instead of using the "basis" function, e.g.
  1831 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1832 
  1833 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1834 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1835 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1836 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1837 changed to match. Definition theorems for overloaded constants now use
  1838 the standard "foo_vec_def" naming scheme. A few other theorems have
  1839 been renamed as follows (INCOMPATIBILITY):
  1840 
  1841   Cart_eq          ~> vec_eq_iff
  1842   dist_nth_le_cart ~> dist_vec_nth_le
  1843   tendsto_vector   ~> vec_tendstoI
  1844   Cauchy_vector    ~> vec_CauchyI
  1845 
  1846 * Session Multivariate_Analysis: Several duplicate theorems have been
  1847 removed, and other theorems have been renamed or replaced with more
  1848 general versions. INCOMPATIBILITY.
  1849 
  1850   finite_choice ~> finite_set_choice
  1851   eventually_conjI ~> eventually_conj
  1852   eventually_and ~> eventually_conj_iff
  1853   eventually_false ~> eventually_False
  1854   setsum_norm ~> norm_setsum
  1855   Lim_sequentially ~> LIMSEQ_def
  1856   Lim_ident_at ~> LIM_ident
  1857   Lim_const ~> tendsto_const
  1858   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1859   Lim_neg ~> tendsto_minus
  1860   Lim_add ~> tendsto_add
  1861   Lim_sub ~> tendsto_diff
  1862   Lim_mul ~> tendsto_scaleR
  1863   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1864   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1865   Lim_linear ~> bounded_linear.tendsto
  1866   Lim_component ~> tendsto_euclidean_component
  1867   Lim_component_cart ~> tendsto_vec_nth
  1868   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1869   dot_lsum ~> inner_setsum_left
  1870   dot_rsum ~> inner_setsum_right
  1871   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1872   continuous_neg ~> continuous_minus
  1873   continuous_sub ~> continuous_diff
  1874   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1875   continuous_mul ~> continuous_scaleR
  1876   continuous_inv ~> continuous_inverse
  1877   continuous_at_within_inv ~> continuous_at_within_inverse
  1878   continuous_at_inv ~> continuous_at_inverse
  1879   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1880   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1881   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1882   continuous_on_neg ~> continuous_on_minus
  1883   continuous_on_sub ~> continuous_on_diff
  1884   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1885   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1886   continuous_on_mul ~> continuous_on_scaleR
  1887   continuous_on_mul_real ~> continuous_on_mult
  1888   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1889   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1890   continuous_on_inverse ~> continuous_on_inv
  1891   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1892   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1893   subset_interior ~> interior_mono
  1894   subset_closure ~> closure_mono
  1895   closure_univ ~> closure_UNIV
  1896   real_arch_lt ~> reals_Archimedean2
  1897   real_arch ~> reals_Archimedean3
  1898   real_abs_norm ~> abs_norm_cancel
  1899   real_abs_sub_norm ~> norm_triangle_ineq3
  1900   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1901 
  1902 * Session HOL-Probability:
  1903   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1904   - Infinite products of probability measures are now available.
  1905   - Sigma closure is independent, if the generator is independent
  1906   - Use extended reals instead of positive extended
  1907     reals. INCOMPATIBILITY.
  1908 
  1909 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1910 
  1911   expand_fun_below ~> fun_below_iff
  1912   below_fun_ext ~> fun_belowI
  1913   expand_cfun_eq ~> cfun_eq_iff
  1914   ext_cfun ~> cfun_eqI
  1915   expand_cfun_below ~> cfun_below_iff
  1916   below_cfun_ext ~> cfun_belowI
  1917   monofun_fun_fun ~> fun_belowD
  1918   monofun_fun_arg ~> monofunE
  1919   monofun_lub_fun ~> adm_monofun [THEN admD]
  1920   cont_lub_fun ~> adm_cont [THEN admD]
  1921   cont2cont_Rep_CFun ~> cont2cont_APP
  1922   cont_Rep_CFun_app ~> cont_APP_app
  1923   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1924   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1925   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1926   contlub_cfun ~> lub_APP [symmetric]
  1927   contlub_LAM ~> lub_LAM [symmetric]
  1928   thelubI ~> lub_eqI
  1929   UU_I ~> bottomI
  1930   lift_distinct1 ~> lift.distinct(1)
  1931   lift_distinct2 ~> lift.distinct(2)
  1932   Def_not_UU ~> lift.distinct(2)
  1933   Def_inject ~> lift.inject
  1934   below_UU_iff ~> below_bottom_iff
  1935   eq_UU_iff ~> eq_bottom_iff
  1936 
  1937 
  1938 *** Document preparation ***
  1939 
  1940 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1941 isar-ref manual, both for description and actual application of the
  1942 same.
  1943 
  1944 * Antiquotation @{value} evaluates the given term and presents its
  1945 result.
  1946 
  1947 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1948 variables x1, y23 into subscripted form x\<^isub>1,
  1949 y\<^isub>2\<^isub>3.
  1950 
  1951 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1952 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1953 
  1954 * Localized \isabellestyle switch can be used within blocks or groups
  1955 like this:
  1956 
  1957   \isabellestyle{it}  %preferred default
  1958   {\isabellestylett @{text "typewriter stuff"}}
  1959 
  1960 * Discontinued special treatment of hard tabulators.  Implicit
  1961 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1962 layouts.
  1963 
  1964 
  1965 *** ML ***
  1966 
  1967 * The inner syntax of sort/type/term/prop supports inlined YXML
  1968 representations within quoted string tokens.  By encoding logical
  1969 entities via Term_XML (in ML or Scala) concrete syntax can be
  1970 bypassed, which is particularly useful for producing bits of text
  1971 under external program control.
  1972 
  1973 * Antiquotations for ML and document preparation are managed as theory
  1974 data, which requires explicit setup.
  1975 
  1976 * Isabelle_Process.is_active allows tools to check if the official
  1977 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1978 (better known as Proof General).
  1979 
  1980 * Structure Proof_Context follows standard naming scheme.  Old
  1981 ProofContext is still available for some time as legacy alias.
  1982 
  1983 * Structure Timing provides various operations for timing; supersedes
  1984 former start_timing/end_timing etc.
  1985 
  1986 * Path.print is the official way to show file-system paths to users
  1987 (including quotes etc.).
  1988 
  1989 * Inner syntax: identifiers in parse trees of generic categories
  1990 "logic", "aprop", "idt" etc. carry position information (disguised as
  1991 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1992 translations that choke on unexpected type constraints.  Positions can
  1993 be stripped in ML translations via Syntax.strip_positions /
  1994 Syntax.strip_positions_ast, or via the syntax constant
  1995 "_strip_positions" within parse trees.  As last resort, positions can
  1996 be disabled via the configuration option Syntax.positions, which is
  1997 called "syntax_positions" in Isar attribute syntax.
  1998 
  1999 * Discontinued special status of various ML structures that contribute
  2000 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  2001 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  2002 refer directly to Ast.Constant, Lexicon.is_identifier,
  2003 Syntax_Trans.mk_binder_tr etc.
  2004 
  2005 * Typed print translation: discontinued show_sorts argument, which is
  2006 already available via context of "advanced" translation.
  2007 
  2008 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  2009 goal states; body tactic needs to address all subgoals uniformly.
  2010 
  2011 * Slightly more special eq_list/eq_set, with shortcut involving
  2012 pointer equality (assumes that eq relation is reflexive).
  2013 
  2014 * Classical tactics use proper Proof.context instead of historic types
  2015 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  2016 operate directly on Proof.context.  Raw type claset retains its use as
  2017 snapshot of the classical context, which can be recovered via
  2018 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  2019 INCOMPATIBILITY, classical tactics and derived proof methods require
  2020 proper Proof.context.
  2021 
  2022 
  2023 *** System ***
  2024 
  2025 * Discontinued support for Poly/ML 5.2, which was the last version
  2026 without proper multithreading and TimeLimit implementation.
  2027 
  2028 * Discontinued old lib/scripts/polyml-platform, which has been
  2029 obsolete since Isabelle2009-2.
  2030 
  2031 * Various optional external tools are referenced more robustly and
  2032 uniformly by explicit Isabelle settings as follows:
  2033 
  2034   ISABELLE_CSDP   (formerly CSDP_EXE)
  2035   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  2036   ISABELLE_OCAML  (formerly EXEC_OCAML)
  2037   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  2038   ISABELLE_YAP    (formerly EXEC_YAP)
  2039 
  2040 Note that automated detection from the file-system or search path has
  2041 been discontinued.  INCOMPATIBILITY.
  2042 
  2043 * Scala layer provides JVM method invocation service for static
  2044 methods of type (String)String, see Invoke_Scala.method in ML.  For
  2045 example:
  2046 
  2047   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  2048 
  2049 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  2050 this allows to pass structured values between ML and Scala.
  2051 
  2052 * The IsabelleText fonts includes some further glyphs to support the
  2053 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  2054 installed a local copy (which is normally *not* required) need to
  2055 delete or update it from ~~/lib/fonts/.
  2056 
  2057 
  2058 
  2059 New in Isabelle2011 (January 2011)
  2060 ----------------------------------
  2061 
  2062 *** General ***
  2063 
  2064 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  2065 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  2066 useful tooltips and hyperlinks produced from its static analysis.  The
  2067 bundled component provides an executable Isabelle tool that can be run
  2068 like this:
  2069 
  2070   Isabelle2011/bin/isabelle jedit
  2071 
  2072 * Significantly improved Isabelle/Isar implementation manual.
  2073 
  2074 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  2075 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  2076 while the default heap location within that directory lacks that extra
  2077 suffix.  This isolates multiple Isabelle installations from each
  2078 other, avoiding problems with old settings in new versions.
  2079 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  2080 
  2081 * Source files are always encoded as UTF-8, instead of old-fashioned
  2082 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  2083 the following package declarations:
  2084 
  2085   \usepackage[utf8]{inputenc}
  2086   \usepackage{textcomp}
  2087 
  2088 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  2089 a Unicode character is treated as a single symbol, not a sequence of
  2090 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  2091 contain symbols without further backslash escapes, Unicode can now be
  2092 used here as well.  Recall that Symbol.explode in ML provides a
  2093 consistent view on symbols, while raw explode (or String.explode)
  2094 merely give a byte-oriented representation.
  2095 
  2096 * Theory loader: source files are primarily located via the master
  2097 directory of each theory node (where the .thy file itself resides).
  2098 The global load path is still partially available as legacy feature.
  2099 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  2100 explicit paths, relatively to the theory.
  2101 
  2102 * Special treatment of ML file names has been discontinued.
  2103 Historically, optional extensions .ML or .sml were added on demand --
  2104 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  2105 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  2106 
  2107 * Various options that affect pretty printing etc. are now properly
  2108 handled within the context via configuration options, instead of
  2109 unsynchronized references or print modes.  There are both ML Config.T
  2110 entities and Isar declaration attributes to access these.
  2111 
  2112   ML (Config.T)                 Isar (attribute)
  2113 
  2114   eta_contract                  eta_contract
  2115   show_brackets                 show_brackets
  2116   show_sorts                    show_sorts
  2117   show_types                    show_types
  2118   show_question_marks           show_question_marks
  2119   show_consts                   show_consts
  2120   show_abbrevs                  show_abbrevs
  2121 
  2122   Syntax.ast_trace              syntax_ast_trace
  2123   Syntax.ast_stat               syntax_ast_stat
  2124   Syntax.ambiguity_level        syntax_ambiguity_level
  2125 
  2126   Goal_Display.goals_limit      goals_limit
  2127   Goal_Display.show_main_goal   show_main_goal
  2128 
  2129   Method.rule_trace             rule_trace
  2130 
  2131   Thy_Output.display            thy_output_display
  2132   Thy_Output.quotes             thy_output_quotes
  2133   Thy_Output.indent             thy_output_indent
  2134   Thy_Output.source             thy_output_source
  2135   Thy_Output.break              thy_output_break
  2136 
  2137 Note that corresponding "..._default" references in ML may only be
  2138 changed globally at the ROOT session setup, but *not* within a theory.
  2139 The option "show_abbrevs" supersedes the former print mode
  2140 "no_abbrevs" with inverted meaning.
  2141 
  2142 * More systematic naming of some configuration options.
  2143 INCOMPATIBILITY.
  2144 
  2145   trace_simp  ~>  simp_trace
  2146   debug_simp  ~>  simp_debug
  2147 
  2148 * Support for real valued configuration options, using simplistic
  2149 floating-point notation that coincides with the inner syntax for
  2150 float_token.
  2151 
  2152 * Support for real valued preferences (with approximative PGIP type):
  2153 front-ends need to accept "pgint" values in float notation.
  2154 INCOMPATIBILITY.
  2155 
  2156 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2157 DejaVu Sans.
  2158 
  2159 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2160 
  2161 
  2162 *** Pure ***
  2163 
  2164 * Command 'type_synonym' (with single argument) replaces somewhat
  2165 outdated 'types', which is still available as legacy feature for some
  2166 time.
  2167 
  2168 * Command 'nonterminal' (with 'and' separated list of arguments)
  2169 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2170 
  2171 * Command 'notepad' replaces former 'example_proof' for
  2172 experimentation in Isar without any result.  INCOMPATIBILITY.
  2173 
  2174 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2175 lists of equations to map definitions in a locale to appropriate
  2176 entities in the context of the interpretation.  The 'interpretation'
  2177 command already provided this functionality.
  2178 
  2179 * Diagnostic command 'print_dependencies' prints the locale instances
  2180 that would be activated if the specified expression was interpreted in
  2181 the current context.  Variant "print_dependencies!" assumes a context
  2182 without interpretations.
  2183 
  2184 * Diagnostic command 'print_interps' prints interpretations in proofs
  2185 in addition to interpretations in theories.
  2186 
  2187 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2188 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2189 Sign.root_path and Sign.local_path may be applied directly where this
  2190 feature is still required for historical reasons.
  2191 
  2192 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2193 'definition' instead.
  2194 
  2195 * The "prems" fact, which refers to the accidental collection of
  2196 foundational premises in the context, is now explicitly marked as
  2197 legacy feature and will be discontinued soon.  Consider using "assms"
  2198 of the head statement or reference facts by explicit names.
  2199 
  2200 * Document antiquotations @{class} and @{type} print classes and type
  2201 constructors.
  2202 
  2203 * Document antiquotation @{file} checks file/directory entries within
  2204 the local file system.
  2205 
  2206 
  2207 *** HOL ***
  2208 
  2209 * Coercive subtyping: functions can be declared as coercions and type
  2210 inference will add them as necessary upon input of a term.  Theory
  2211 Complex_Main declares real :: nat => real and real :: int => real as
  2212 coercions. A coercion function f is declared like this:
  2213 
  2214   declare [[coercion f]]
  2215 
  2216 To lift coercions through type constructors (e.g. from nat => real to
  2217 nat list => real list), map functions can be declared, e.g.
  2218 
  2219   declare [[coercion_map map]]
  2220 
  2221 Currently coercion inference is activated only in theories including
  2222 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2223 the configuration option "coercion_enabled", e.g. it can be enabled in
  2224 other theories like this:
  2225 
  2226   declare [[coercion_enabled]]
  2227 
  2228 * Command 'partial_function' provides basic support for recursive
  2229 function definitions over complete partial orders.  Concrete instances
  2230 are provided for i) the option type, ii) tail recursion on arbitrary
  2231 types, and iii) the heap monad of Imperative_HOL.  See
  2232 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2233 for examples.
  2234 
  2235 * Function package: f.psimps rules are no longer implicitly declared
  2236 as [simp].  INCOMPATIBILITY.
  2237 
  2238 * Datatype package: theorems generated for executable equality (class
  2239 "eq") carry proper names and are treated as default code equations.
  2240 
  2241 * Inductive package: now offers command 'inductive_simps' to
  2242 automatically derive instantiated and simplified equations for
  2243 inductive predicates, similar to 'inductive_cases'.
  2244 
  2245 * Command 'enriched_type' allows to register properties of the
  2246 functorial structure of types.
  2247 
  2248 * Improved infrastructure for term evaluation using code generator
  2249 techniques, in particular static evaluation conversions.
  2250 
  2251 * Code generator: Scala (2.8 or higher) has been added to the target
  2252 languages.
  2253 
  2254 * Code generator: globbing constant expressions "*" and "Theory.*"
  2255 have been replaced by the more idiomatic "_" and "Theory._".
  2256 INCOMPATIBILITY.
  2257 
  2258 * Code generator: export_code without explicit file declaration prints
  2259 to standard output.  INCOMPATIBILITY.
  2260 
  2261 * Code generator: do not print function definitions for case
  2262 combinators any longer.
  2263 
  2264 * Code generator: simplification with rules determined with
  2265 src/Tools/Code/code_simp.ML and method "code_simp".
  2266 
  2267 * Code generator for records: more idiomatic representation of record
  2268 types.  Warning: records are not covered by ancient SML code
  2269 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2270 rep_datatype declaration helps to succeed then:
  2271 
  2272   record 'a foo = ...
  2273   ...
  2274   rep_datatype foo_ext ...
  2275 
  2276 * Records: logical foundation type for records does not carry a
  2277 '_type' suffix any longer (obsolete due to authentic syntax).
  2278 INCOMPATIBILITY.
  2279 
  2280 * Quickcheck now by default uses exhaustive testing instead of random
  2281 testing.  Random testing can be invoked by "quickcheck [random]",
  2282 exhaustive testing by "quickcheck [exhaustive]".
  2283 
  2284 * Quickcheck instantiates polymorphic types with small finite
  2285 datatypes by default. This enables a simple execution mechanism to
  2286 handle quantifiers and function equality over the finite datatypes.
  2287 
  2288 * Quickcheck random generator has been renamed from "code" to
  2289 "random".  INCOMPATIBILITY.
  2290 
  2291 * Quickcheck now has a configurable time limit which is set to 30
  2292 seconds by default. This can be changed by adding [timeout = n] to the
  2293 quickcheck command. The time limit for Auto Quickcheck is still set
  2294 independently.
  2295 
  2296 * Quickcheck in locales considers interpretations of that locale for
  2297 counter example search.
  2298 
  2299 * Sledgehammer:
  2300   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2301     method. See the Sledgehammer manual for details ("isabelle doc
  2302     sledgehammer").
  2303   - Renamed commands:
  2304     sledgehammer atp_info ~> sledgehammer running_provers
  2305     sledgehammer atp_kill ~> sledgehammer kill_provers
  2306     sledgehammer available_atps ~> sledgehammer available_provers
  2307     INCOMPATIBILITY.
  2308   - Renamed options:
  2309     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2310     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2311     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2312     (and "ms" and "min" are no longer supported)
  2313     INCOMPATIBILITY.
  2314 
  2315 * Nitpick:
  2316   - Renamed options:
  2317     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2318     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2319     INCOMPATIBILITY.
  2320   - Added support for partial quotient types.
  2321   - Added local versions of the "Nitpick.register_xxx" functions.
  2322   - Added "whack" option.
  2323   - Allow registration of quotient types as codatatypes.
  2324   - Improved "merge_type_vars" option to merge more types.
  2325   - Removed unsound "fast_descrs" option.
  2326   - Added custom symmetry breaking for datatypes, making it possible to reach
  2327     higher cardinalities.
  2328   - Prevent the expansion of too large definitions.
  2329 
  2330 * Proof methods "metis" and "meson" now have configuration options
  2331 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2332 to diagnose these tools. E.g.
  2333 
  2334     using [[metis_trace = true]]
  2335 
  2336 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2337 manually as command 'solve_direct'.
  2338 
  2339 * The default SMT solver Z3 must be enabled explicitly (due to
  2340 licensing issues) by setting the environment variable
  2341 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2342 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2343 changing the SMT solver is done via the configuration option
  2344 "smt_solver".
  2345 
  2346 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2347 i.e. "remote_cvc3" and "remote_z3".
  2348 
  2349 * Added basic SMT support for datatypes, records, and typedefs using
  2350 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2351 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2352 support for a similar behavior).  Minor INCOMPATIBILITY.
  2353 
  2354 * Changed SMT configuration options:
  2355   - Renamed:
  2356     z3_proofs ~> smt_oracle (with inverted meaning)
  2357     z3_trace_assms ~> smt_trace_used_facts
  2358     INCOMPATIBILITY.
  2359   - Added:
  2360     smt_verbose
  2361     smt_random_seed
  2362     smt_datatypes
  2363     smt_infer_triggers
  2364     smt_monomorph_limit
  2365     cvc3_options
  2366     remote_cvc3_options
  2367     remote_z3_options
  2368     yices_options
  2369 
  2370 * Boogie output files (.b2i files) need to be declared in the theory
  2371 header.
  2372 
  2373 * Simplification procedure "list_to_set_comprehension" rewrites list
  2374 comprehensions applied to List.set to set comprehensions.  Occasional
  2375 INCOMPATIBILITY, may be deactivated like this:
  2376 
  2377   declare [[simproc del: list_to_set_comprehension]]
  2378 
  2379 * Removed old version of primrec package.  INCOMPATIBILITY.
  2380 
  2381 * Removed simplifier congruence rule of "prod_case", as has for long
  2382 been the case with "split".  INCOMPATIBILITY.
  2383 
  2384 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2385 
  2386 * Removed [split_format ... and ... and ...] version of
  2387 [split_format].  Potential INCOMPATIBILITY.
  2388 
  2389 * Predicate "sorted" now defined inductively, with nice induction
  2390 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2391 
  2392 * Constant "contents" renamed to "the_elem", to free the generic name
  2393 contents for other uses.  INCOMPATIBILITY.
  2394 
  2395 * Renamed class eq and constant eq (for code generation) to class
  2396 equal and constant equal, plus renaming of related facts and various
  2397 tuning.  INCOMPATIBILITY.
  2398 
  2399 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2400 
  2401 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2402 INCOMPATIBILITY.
  2403 
  2404 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2405 avoid confusion with finite sets.  INCOMPATIBILITY.
  2406 
  2407 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2408 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2409 for congruent(2)).
  2410 
  2411 * Some previously unqualified names have been qualified:
  2412 
  2413   types
  2414     bool ~> HOL.bool
  2415     nat ~> Nat.nat
  2416 
  2417   constants
  2418     Trueprop ~> HOL.Trueprop
  2419     True ~> HOL.True
  2420     False ~> HOL.False
  2421     op & ~> HOL.conj
  2422     op | ~> HOL.disj
  2423     op --> ~> HOL.implies
  2424     op = ~> HOL.eq
  2425     Not ~> HOL.Not
  2426     The ~> HOL.The
  2427     All ~> HOL.All
  2428     Ex ~> HOL.Ex
  2429     Ex1 ~> HOL.Ex1
  2430     Let ~> HOL.Let
  2431     If ~> HOL.If
  2432     Ball ~> Set.Ball
  2433     Bex ~> Set.Bex
  2434     Suc ~> Nat.Suc
  2435     Pair ~> Product_Type.Pair
  2436     fst ~> Product_Type.fst
  2437     snd ~> Product_Type.snd
  2438     curry ~> Product_Type.curry
  2439     op : ~> Set.member
  2440     Collect ~> Set.Collect
  2441 
  2442 INCOMPATIBILITY.
  2443 
  2444 * More canonical naming convention for some fundamental definitions:
  2445 
  2446     bot_bool_eq ~> bot_bool_def
  2447     top_bool_eq ~> top_bool_def
  2448     inf_bool_eq ~> inf_bool_def
  2449     sup_bool_eq ~> sup_bool_def
  2450     bot_fun_eq  ~> bot_fun_def
  2451     top_fun_eq  ~> top_fun_def
  2452     inf_fun_eq  ~> inf_fun_def
  2453     sup_fun_eq  ~> sup_fun_def
  2454 
  2455 INCOMPATIBILITY.
  2456 
  2457 * More stylized fact names:
  2458 
  2459   expand_fun_eq ~> fun_eq_iff
  2460   expand_set_eq ~> set_eq_iff
  2461   set_ext       ~> set_eqI
  2462   nat_number    ~> eval_nat_numeral
  2463 
  2464 INCOMPATIBILITY.
  2465 
  2466 * Refactoring of code-generation specific operations in theory List:
  2467 
  2468   constants
  2469     null ~> List.null
  2470 
  2471   facts
  2472     mem_iff ~> member_def
  2473     null_empty ~> null_def
  2474 
  2475 INCOMPATIBILITY.  Note that these were not supposed to be used
  2476 regularly unless for striking reasons; their main purpose was code
  2477 generation.
  2478 
  2479 Various operations from the Haskell prelude are used for generating
  2480 Haskell code.
  2481 
  2482 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2483 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2484 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2485 
  2486 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2487 "*" and "+" respectively.  INCOMPATIBILITY.
  2488 
  2489 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2490 "Sum_Type.Plus" instead.
  2491 
  2492 * Constant "split" has been merged with constant "prod_case"; names of
  2493 ML functions, facts etc. involving split have been retained so far,
  2494 though.  INCOMPATIBILITY.
  2495 
  2496 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2497 instead.  INCOMPATIBILITY.
  2498 
  2499 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2500 INCOMPATIBILITY.
  2501 
  2502 * Former theory Library/Enum is now part of the HOL-Main image.
  2503 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2504 referred to by its qualified name.
  2505 
  2506   enum    ~>  Enum.enum
  2507   nlists  ~>  Enum.nlists
  2508   product ~>  Enum.product
  2509 
  2510 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2511 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2512 INCOMPATIBILITY.
  2513 
  2514 * Theory Library/SetsAndFunctions has been split into
  2515 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2516 for instance definitions for functions; various improvements.
  2517 INCOMPATIBILITY.
  2518 
  2519 * Theory Library/Multiset provides stable quicksort implementation of
  2520 sort_key.
  2521 
  2522 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2523 INCOMPATIBILITY.
  2524 
  2525 * Session Multivariate_Analysis: introduced a type class for euclidean
  2526 space.  Most theorems are now stated in terms of euclidean spaces
  2527 instead of finite cartesian products.
  2528 
  2529   types
  2530     real ^ 'n ~>  'a::real_vector
  2531               ~>  'a::euclidean_space
  2532               ~>  'a::ordered_euclidean_space
  2533         (depends on your needs)
  2534 
  2535   constants
  2536      _ $ _        ~> _ $$ _
  2537      \<chi> x. _  ~> \<chi>\<chi> x. _
  2538      CARD('n)     ~> DIM('a)
  2539 
  2540 Also note that the indices are now natural numbers and not from some
  2541 finite type. Finite cartesian products of euclidean spaces, products
  2542 of euclidean spaces the real and complex numbers are instantiated to
  2543 be euclidean_spaces.  INCOMPATIBILITY.
  2544 
  2545 * Session Probability: introduced pextreal as positive extended real
  2546 numbers.  Use pextreal as value for measures.  Introduce the
  2547 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2548 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2549 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2550 
  2551 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2552 INCOMPATIBILITY.
  2553 
  2554 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2555 prove verification conditions generated by the SPARK Ada program
  2556 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2557 
  2558 
  2559 *** HOL-Algebra ***
  2560 
  2561 * Theorems for additive ring operations (locale abelian_monoid and
  2562 descendants) are generated by interpretation from their multiplicative
  2563 counterparts.  Names (in particular theorem names) have the mandatory
  2564 qualifier 'add'.  Previous theorem names are redeclared for
  2565 compatibility.
  2566 
  2567 * Structure "int_ring" is now an abbreviation (previously a
  2568 definition).  This fits more natural with advanced interpretations.
  2569 
  2570 
  2571 *** HOLCF ***
  2572 
  2573 * The domain package now runs in definitional mode by default: The
  2574 former command 'new_domain' is now called 'domain'.  To use the domain
  2575 package in its original axiomatic mode, use 'domain (unsafe)'.
  2576 INCOMPATIBILITY.
  2577 
  2578 * The new class "domain" is now the default sort.  Class "predomain"
  2579 is an unpointed version of "domain". Theories can be updated by
  2580 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2581 
  2582   'a::type ~> 'a::countable
  2583   'a::cpo  ~> 'a::predomain
  2584   'a::pcpo ~> 'a::domain
  2585 
  2586 * The old type class "rep" has been superseded by class "domain".
  2587 Accordingly, users of the definitional package must remove any
  2588 "default_sort rep" declarations.  INCOMPATIBILITY.
  2589 
  2590 * The domain package (definitional mode) now supports unpointed
  2591 predomain argument types, as long as they are marked 'lazy'. (Strict
  2592 arguments must be in class "domain".) For example, the following
  2593 domain definition now works:
  2594 
  2595   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2596 
  2597 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2598 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2599 'a option, and 'a list.  Additionally, it configures fixrec and the
  2600 domain package to work with these types.  For example:
  2601 
  2602   fixrec isInl :: "('a + 'b) u -> tr"
  2603     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2604 
  2605   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2606 
  2607 * The "(permissive)" option of fixrec has been replaced with a
  2608 per-equation "(unchecked)" option. See
  2609 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2610 
  2611 * The "bifinite" class no longer fixes a constant "approx"; the class
  2612 now just asserts that such a function exists.  INCOMPATIBILITY.
  2613 
  2614 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2615 defines an embedding of type 'a defl into udom by default; instances
  2616 of "bifinite" and "domain" classes are available in
  2617 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2618 
  2619 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2620 
  2621 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2622 
  2623 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2624 
  2625 * The function "cprod_map" has been renamed to "prod_map".
  2626 INCOMPATIBILITY.
  2627 
  2628 * The monadic bind operator on each powerdomain has new binder syntax
  2629 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2630 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2631 
  2632 * The infix syntax for binary union on each powerdomain has changed
  2633 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2634 syntax.  INCOMPATIBILITY.
  2635 
  2636 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2637 still supported as an input translation.
  2638 
  2639 * Renamed some theorems (the original names are also still available).
  2640 
  2641   expand_fun_below   ~> fun_below_iff
  2642   below_fun_ext      ~> fun_belowI
  2643   expand_cfun_eq     ~> cfun_eq_iff
  2644   ext_cfun           ~> cfun_eqI
  2645   expand_cfun_below  ~> cfun_below_iff
  2646   below_cfun_ext     ~> cfun_belowI
  2647   cont2cont_Rep_CFun ~> cont2cont_APP
  2648 
  2649 * The Abs and Rep functions for various types have changed names.
  2650 Related theorem names have also changed to match. INCOMPATIBILITY.
  2651 
  2652   Rep_CFun  ~> Rep_cfun
  2653   Abs_CFun  ~> Abs_cfun
  2654   Rep_Sprod ~> Rep_sprod
  2655   Abs_Sprod ~> Abs_sprod
  2656   Rep_Ssum  ~> Rep_ssum
  2657   Abs_Ssum  ~> Abs_ssum
  2658 
  2659 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2660 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2661 
  2662 * Various changes to bisimulation/coinduction with domain package:
  2663 
  2664   - Definitions of "bisim" constants no longer mention definedness.
  2665   - With mutual recursion, "bisim" predicate is now curried.
  2666   - With mutual recursion, each type gets a separate coind theorem.
  2667   - Variable names in bisim_def and coinduct rules have changed.
  2668 
  2669 INCOMPATIBILITY.
  2670 
  2671 * Case combinators generated by the domain package for type "foo" are
  2672 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2673 
  2674 * Several theorems have been renamed to more accurately reflect the
  2675 names of constants and types involved.  INCOMPATIBILITY.
  2676 
  2677   thelub_const    ~> lub_const
  2678   lub_const       ~> is_lub_const
  2679   thelubI         ~> lub_eqI
  2680   is_lub_lub      ~> is_lubD2
  2681   lubI            ~> is_lub_lub
  2682   unique_lub      ~> is_lub_unique
  2683   is_ub_lub       ~> is_lub_rangeD1
  2684   lub_bin_chain   ~> is_lub_bin_chain
  2685   lub_fun         ~> is_lub_fun
  2686   thelub_fun      ~> lub_fun
  2687   thelub_cfun     ~> lub_cfun
  2688   thelub_Pair     ~> lub_Pair
  2689   lub_cprod       ~> is_lub_prod
  2690   thelub_cprod    ~> lub_prod
  2691   minimal_cprod   ~> minimal_prod
  2692   inst_cprod_pcpo ~> inst_prod_pcpo
  2693   UU_I            ~> bottomI
  2694   compact_UU      ~> compact_bottom
  2695   deflation_UU    ~> deflation_bottom
  2696   finite_deflation_UU ~> finite_deflation_bottom
  2697 
  2698 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2699 
  2700   sq_ord_less_eq_trans ~> below_eq_trans
  2701   sq_ord_eq_less_trans ~> eq_below_trans
  2702   refl_less            ~> below_refl
  2703   trans_less           ~> below_trans
  2704   antisym_less         ~> below_antisym
  2705   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2706   box_less             ~> box_below
  2707   rev_trans_less       ~> rev_below_trans
  2708   not_less2not_eq      ~> not_below2not_eq
  2709   less_UU_iff          ~> below_UU_iff
  2710   flat_less_iff        ~> flat_below_iff
  2711   adm_less             ~> adm_below
  2712   adm_not_less         ~> adm_not_below
  2713   adm_compact_not_less ~> adm_compact_not_below
  2714   less_fun_def         ~> below_fun_def
  2715   expand_fun_less      ~> fun_below_iff
  2716   less_fun_ext         ~> fun_belowI
  2717   less_discr_def       ~> below_discr_def
  2718   discr_less_eq        ~> discr_below_eq
  2719   less_unit_def        ~> below_unit_def
  2720   less_cprod_def       ~> below_prod_def
  2721   prod_lessI           ~> prod_belowI
  2722   Pair_less_iff        ~> Pair_below_iff
  2723   fst_less_iff         ~> fst_below_iff
  2724   snd_less_iff         ~> snd_below_iff
  2725   expand_cfun_less     ~> cfun_below_iff
  2726   less_cfun_ext        ~> cfun_belowI
  2727   injection_less       ~> injection_below
  2728   less_up_def          ~> below_up_def
  2729   not_Iup_less         ~> not_Iup_below
  2730   Iup_less             ~> Iup_below
  2731   up_less              ~> up_below
  2732   Def_inject_less_eq   ~> Def_below_Def
  2733   Def_less_is_eq       ~> Def_below_iff
  2734   spair_less_iff       ~> spair_below_iff
  2735   less_sprod           ~> below_sprod
  2736   spair_less           ~> spair_below
  2737   sfst_less_iff        ~> sfst_below_iff
  2738   ssnd_less_iff        ~> ssnd_below_iff
  2739   fix_least_less       ~> fix_least_below
  2740   dist_less_one        ~> dist_below_one
  2741   less_ONE             ~> below_ONE
  2742   ONE_less_iff         ~> ONE_below_iff
  2743   less_sinlD           ~> below_sinlD
  2744   less_sinrD           ~> below_sinrD
  2745 
  2746 
  2747 *** FOL and ZF ***
  2748 
  2749 * All constant names are now qualified internally and use proper
  2750 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2751 
  2752 
  2753 *** ML ***
  2754 
  2755 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2756 Fail if the argument is false.  Due to inlining the source position of
  2757 failed assertions is included in the error output.
  2758 
  2759 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2760 text is in practice always evaluated with a stable theory checkpoint.
  2761 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2762 
  2763 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2764 the current context, not any accidental theory loader state as before.
  2765 Potential INCOMPATIBILITY, subtle change in semantics.
  2766 
  2767 * Syntax.pretty_priority (default 0) configures the required priority
  2768 of pretty-printed output and thus affects insertion of parentheses.
  2769 
  2770 * Syntax.default_root (default "any") configures the inner syntax
  2771 category (nonterminal symbol) for parsing of terms.
  2772 
  2773 * Former exception Library.UnequalLengths now coincides with
  2774 ListPair.UnequalLengths.
  2775 
  2776 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2777 main functionality is provided by structure Simplifier.
  2778 
  2779 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2780 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2781 almost all situations.
  2782 
  2783 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2784 See implementation manual for further details on exceptions in
  2785 Isabelle/ML.
  2786 
  2787 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2788 meaning.
  2789 
  2790 * Renamed structure PureThy to Pure_Thy and moved most of its
  2791 operations to structure Global_Theory, to emphasize that this is
  2792 rarely-used global-only stuff.
  2793 
  2794 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2795 instead (or tracing for high-volume output).
  2796 
  2797 * Configuration option show_question_marks only affects regular pretty
  2798 printing of types and terms, not raw Term.string_of_vname.
  2799 
  2800 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2801 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2802 @{thms} for most purposes.
  2803 
  2804 * ML structure Unsynchronized is never opened, not even in Isar
  2805 interaction mode as before.  Old Unsynchronized.set etc. have been
  2806 discontinued -- use plain := instead.  This should be *rare* anyway,
  2807 since modern tools always work via official context data, notably
  2808 configuration options.
  2809 
  2810 * Parallel and asynchronous execution requires special care concerning
  2811 interrupts.  Structure Exn provides some convenience functions that
  2812 avoid working directly with raw Interrupt.  User code must not absorb
  2813 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2814 followed by re-raising of the original exception.  Another common
  2815 source of mistakes are "handle _" patterns, which make the meaning of
  2816 the program subject to physical effects of the environment.
  2817 
  2818 
  2819 
  2820 New in Isabelle2009-2 (June 2010)
  2821 ---------------------------------
  2822 
  2823 *** General ***
  2824 
  2825 * Authentic syntax for *all* logical entities (type classes, type
  2826 constructors, term constants): provides simple and robust
  2827 correspondence between formal entities and concrete syntax.  Within
  2828 the parse tree / AST representations, "constants" are decorated by
  2829 their category (class, type, const) and spelled out explicitly with
  2830 their full internal name.
  2831 
  2832 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2833 and translations (translation rules and translation functions in ML).
  2834 Some hints on upgrading:
  2835 
  2836   - Many existing uses of 'syntax' and 'translations' can be replaced
  2837     by more modern 'type_notation', 'notation' and 'abbreviation',
  2838     which are independent of this issue.
  2839 
  2840   - 'translations' require markup within the AST; the term syntax
  2841     provides the following special forms:
  2842 
  2843       CONST c   -- produces syntax version of constant c from context
  2844       XCONST c  -- literally c, checked as constant from context
  2845       c         -- literally c, if declared by 'syntax'
  2846 
  2847     Plain identifiers are treated as AST variables -- occasionally the
  2848     system indicates accidental variables via the error "rhs contains
  2849     extra variables".
  2850 
  2851     Type classes and type constructors are marked according to their
  2852     concrete syntax.  Some old translations rules need to be written
  2853     for the "type" category, using type constructor application
  2854     instead of pseudo-term application of the default category
  2855     "logic".
  2856 
  2857   - 'parse_translation' etc. in ML may use the following
  2858     antiquotations:
  2859 
  2860       @{class_syntax c}   -- type class c within parse tree / AST
  2861       @{term_syntax c}    -- type constructor c within parse tree / AST
  2862       @{const_syntax c}   -- ML version of "CONST c" above
  2863       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2864 
  2865   - Literal types within 'typed_print_translations', i.e. those *not*
  2866     represented as pseudo-terms are represented verbatim.  Use @{class
  2867     c} or @{type_name c} here instead of the above syntax
  2868     antiquotations.
  2869 
  2870 Note that old non-authentic syntax was based on unqualified base
  2871 names, so all of the above "constant" names would coincide.  Recall
  2872 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2873 diagnose syntax problems.
  2874 
  2875 * Type constructors admit general mixfix syntax, not just infix.
  2876 
  2877 * Concrete syntax may be attached to local entities without a proof
  2878 body, too.  This works via regular mixfix annotations for 'fix',
  2879 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2880 similar to the 'notation' command in theory specifications.
  2881 
  2882 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2883 need to specify constant name and syntax separately.  Internal ML
  2884 datatype constructors have been renamed from InfixName to Infix etc.
  2885 Minor INCOMPATIBILITY.
  2886 
  2887 * Schematic theorem statements need to be explicitly markup as such,
  2888 via commands 'schematic_lemma', 'schematic_theorem',
  2889 'schematic_corollary'.  Thus the relevance of the proof is made
  2890 syntactically clear, which impacts performance in a parallel or
  2891 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2892 
  2893 * Use of cumulative prems via "!" in some proof methods has been
  2894 discontinued (old legacy feature).
  2895 
  2896 * References 'trace_simp' and 'debug_simp' have been replaced by
  2897 configuration options stored in the context. Enabling tracing (the
  2898 case of debugging is similar) in proofs works via
  2899 
  2900   using [[trace_simp = true]]
  2901 
  2902 Tracing is then active for all invocations of the simplifier in
  2903 subsequent goal refinement steps. Tracing may also still be enabled or
  2904 disabled via the ProofGeneral settings menu.
  2905 
  2906 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2907 'hide_fact' replace the former 'hide' KIND command.  Minor
  2908 INCOMPATIBILITY.
  2909 
  2910 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2911 more efficient than combinations with -q1 or -q2.
  2912 
  2913 
  2914 *** Pure ***
  2915 
  2916 * Proofterms record type-class reasoning explicitly, using the
  2917 "unconstrain" operation internally.  This eliminates all sort
  2918 constraints from a theorem and proof, introducing explicit
  2919 OFCLASS-premises.  On the proof term level, this operation is
  2920 automatically applied at theorem boundaries, such that closed proofs
  2921 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2922 inspect proof terms.
  2923 
  2924 * Local theory specifications may depend on extra type variables that
  2925 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2926 are added internally.  For example:
  2927 
  2928   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2929 
  2930 * Predicates of locales introduced by classes carry a mandatory
  2931 "class" prefix.  INCOMPATIBILITY.
  2932 
  2933 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2934 
  2935 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2936 'class' instead.
  2937 
  2938 * Command 'code_reflect' allows to incorporate generated ML code into
  2939 runtime environment; replaces immature code_datatype antiquotation.
  2940 INCOMPATIBILITY.
  2941 
  2942 * Code generator: simple concept for abstract datatypes obeying
  2943 invariants.
  2944 
  2945 * Code generator: details of internal data cache have no impact on the
  2946 user space functionality any longer.
  2947 
  2948 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2949 subgoals.  This is more appropriate for interpretations with 'where'.
  2950 INCOMPATIBILITY.
  2951 
  2952 * Command 'example_proof' opens an empty proof body.  This allows to
  2953 experiment with Isar, without producing any persistent result.
  2954 
  2955 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2956 within a local theory context, with explicit checking of the
  2957 constructors involved (in contrast to the raw 'syntax' versions).
  2958 
  2959 * Commands 'types' and 'typedecl' now work within a local theory
  2960 context -- without introducing dependencies on parameters or
  2961 assumptions, which is not possible in Isabelle/Pure.
  2962 
  2963 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2964 within a local theory context.  Minor INCOMPATIBILITY.
  2965 
  2966 
  2967 *** HOL ***
  2968 
  2969 * Command 'typedef' now works within a local theory context -- without
  2970 introducing dependencies on parameters or assumptions, which is not
  2971 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2972 contain multiple interpretations of local typedefs (with different
  2973 non-emptiness proofs), even in a global theory context.
  2974 
  2975 * New package for quotient types.  Commands 'quotient_type' and
  2976 'quotient_definition' may be used for defining types and constants by
  2977 quotient constructions.  An example is the type of integers created by
  2978 quotienting pairs of natural numbers:
  2979 
  2980   fun
  2981     intrel :: "(nat * nat) => (nat * nat) => bool"
  2982   where
  2983     "intrel (x, y) (u, v) = (x + v = u + y)"
  2984 
  2985   quotient_type int = "nat * nat" / intrel
  2986     by (auto simp add: equivp_def expand_fun_eq)
  2987 
  2988   quotient_definition
  2989     "0::int" is "(0::nat, 0::nat)"
  2990 
  2991 The method "lifting" can be used to lift of theorems from the
  2992 underlying "raw" type to the quotient type.  The example
  2993 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2994 construction and provides a reasoning infrastructure for finite sets.
  2995 
  2996 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2997 clash with new theory Quotient in Main HOL.
  2998 
  2999 * Moved the SMT binding into the main HOL session, eliminating
  3000 separate HOL-SMT session.
  3001 
  3002 * List membership infix mem operation is only an input abbreviation.
  3003 INCOMPATIBILITY.
  3004 
  3005 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  3006 for future developements; former Library/Word.thy is still present in
  3007 the AFP entry RSAPPS.
  3008 
  3009 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  3010 longer shadowed.  INCOMPATIBILITY.
  3011 
  3012 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  3013 INCOMPATIBILITY.
  3014 
  3015 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  3016 INCOMPATIBILITY.
  3017 
  3018 * Dropped normalizing_semiring etc; use the facts in semiring classes
  3019 instead.  INCOMPATIBILITY.
  3020 
  3021 * Dropped several real-specific versions of lemmas about floor and
  3022 ceiling; use the generic lemmas from theory "Archimedean_Field"
  3023 instead.  INCOMPATIBILITY.
  3024 
  3025   floor_number_of_eq         ~> floor_number_of
  3026   le_floor_eq_number_of      ~> number_of_le_floor
  3027   le_floor_eq_zero           ~> zero_le_floor
  3028   le_floor_eq_one            ~> one_le_floor
  3029   floor_less_eq_number_of    ~> floor_less_number_of
  3030   floor_less_eq_zero         ~> floor_less_zero
  3031   floor_less_eq_one          ~> floor_less_one
  3032   less_floor_eq_number_of    ~> number_of_less_floor
  3033   less_floor_eq_zero         ~> zero_less_floor
  3034   less_floor_eq_one          ~> one_less_floor
  3035   floor_le_eq_number_of      ~> floor_le_number_of
  3036   floor_le_eq_zero           ~> floor_le_zero
  3037   floor_le_eq_one            ~> floor_le_one
  3038   floor_subtract_number_of   ~> floor_diff_number_of
  3039   floor_subtract_one         ~> floor_diff_one
  3040   ceiling_number_of_eq       ~> ceiling_number_of
  3041   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  3042   ceiling_le_zero_eq         ~> ceiling_le_zero
  3043   ceiling_le_eq_one          ~> ceiling_le_one
  3044   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  3045   less_ceiling_eq_zero       ~> zero_less_ceiling
  3046   less_ceiling_eq_one        ~> one_less_ceiling
  3047   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  3048   ceiling_less_eq_zero       ~> ceiling_less_zero
  3049   ceiling_less_eq_one        ~> ceiling_less_one
  3050   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  3051   le_ceiling_eq_zero         ~> zero_le_ceiling
  3052   le_ceiling_eq_one          ~> one_le_ceiling
  3053   ceiling_subtract_number_of ~> ceiling_diff_number_of
  3054   ceiling_subtract_one       ~> ceiling_diff_one
  3055 
  3056 * Theory "Finite_Set": various folding_XXX locales facilitate the
  3057 application of the various fold combinators on finite sets.
  3058 
  3059 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  3060 provides abstract red-black tree type which is backed by "RBT_Impl" as
  3061 implementation.  INCOMPATIBILTY.
  3062 
  3063 * Theory Library/Coinductive_List has been removed -- superseded by
  3064 AFP/thys/Coinductive.
  3065 
  3066 * Theory PReal, including the type "preal" and related operations, has
  3067 been removed.  INCOMPATIBILITY.
  3068 
  3069 * Real: new development using Cauchy Sequences.
  3070 
  3071 * Split off theory "Big_Operators" containing setsum, setprod,
  3072 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  3073 
  3074 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  3075 "Int" etc.  INCOMPATIBILITY.
  3076 
  3077 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  3078 
  3079 * New set of rules "ac_simps" provides combined assoc / commute
  3080 rewrites for all interpretations of the appropriate generic locales.
  3081 
  3082 * Renamed theory "OrderedGroup" to "Groups" and split theory
  3083 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  3084 appropriate and more consistent names suitable for name prefixes
  3085 within the HOL theories.  INCOMPATIBILITY.
  3086 
  3087 * Some generic constants have been put to appropriate theories:
  3088   - less_eq, less: Orderings
  3089   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  3090   - inverse, divide: Rings
  3091 INCOMPATIBILITY.
  3092 
  3093 * More consistent naming of type classes involving orderings (and
  3094 lattices):
  3095 
  3096     lower_semilattice                   ~> semilattice_inf
  3097     upper_semilattice                   ~> semilattice_sup
  3098 
  3099     dense_linear_order                  ~> dense_linorder
  3100 
  3101     pordered_ab_group_add               ~> ordered_ab_group_add
  3102     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  3103     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  3104     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  3105     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  3106     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  3107     pordered_cancel_semiring            ~> ordered_cancel_semiring
  3108     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  3109     pordered_comm_ring                  ~> ordered_comm_ring
  3110     pordered_comm_semiring              ~> ordered_comm_semiring
  3111     pordered_ring                       ~> ordered_ring
  3112     pordered_ring_abs                   ~> ordered_ring_abs
  3113     pordered_semiring                   ~> ordered_semiring
  3114 
  3115     ordered_ab_group_add                ~> linordered_ab_group_add
  3116     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  3117     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  3118     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  3119     ordered_field                       ~> linordered_field
  3120     ordered_field_no_lb                 ~> linordered_field_no_lb
  3121     ordered_field_no_ub                 ~> linordered_field_no_ub
  3122     ordered_field_dense_linear_order    ~> dense_linordered_field
  3123     ordered_idom                        ~> linordered_idom
  3124     ordered_ring                        ~> linordered_ring
  3125     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  3126     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  3127     ordered_ring_strict                 ~> linordered_ring_strict
  3128     ordered_semidom                     ~> linordered_semidom
  3129     ordered_semiring                    ~> linordered_semiring
  3130     ordered_semiring_1                  ~> linordered_semiring_1
  3131     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  3132     ordered_semiring_strict             ~> linordered_semiring_strict
  3133 
  3134   The following slightly odd type classes have been moved to a
  3135   separate theory Library/Lattice_Algebras:
  3136 
  3137     lordered_ab_group_add               ~> lattice_ab_group_add
  3138     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  3139     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  3140     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  3141     lordered_ring                       ~> lattice_ring
  3142 
  3143 INCOMPATIBILITY.
  3144 
  3145 * Refined field classes:
  3146   - classes division_ring_inverse_zero, field_inverse_zero,
  3147     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3148     subsumes former division_by_zero class;
  3149   - numerous lemmas have been ported from field to division_ring.
  3150 INCOMPATIBILITY.
  3151 
  3152 * Refined algebra theorem collections:
  3153   - dropped theorem group group_simps, use algebra_simps instead;
  3154   - dropped theorem group ring_simps, use field_simps instead;
  3155   - proper theorem collection field_simps subsumes former theorem
  3156     groups field_eq_simps and field_simps;
  3157   - dropped lemma eq_minus_self_iff which is a duplicate for
  3158     equal_neg_zero.
  3159 INCOMPATIBILITY.
  3160 
  3161 * Theory Finite_Set and List: some lemmas have been generalized from
  3162 sets to lattices:
  3163 
  3164   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3165   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3166   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3167   union_Union_fold_union        ~> sup_Sup_fold_sup
  3168   Inter_fold_inter              ~> Inf_fold_inf
  3169   Union_fold_union              ~> Sup_fold_sup
  3170   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3171   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3172   INTER_fold_inter              ~> INFI_fold_inf
  3173   UNION_fold_union              ~> SUPR_fold_sup
  3174 
  3175 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3176 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3177 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3178 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3179 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3180 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3181 
  3182 * Reorganized theory Multiset: swapped notation of pointwise and
  3183 multiset order:
  3184 
  3185   - pointwise ordering is instance of class order with standard syntax
  3186     <= and <;
  3187   - multiset ordering has syntax <=# and <#; partial order properties
  3188     are provided by means of interpretation with prefix
  3189     multiset_order;
  3190   - less duplication, less historical organization of sections,
  3191     conversion from associations lists to multisets, rudimentary code
  3192     generation;
  3193   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3194     if needed.
  3195 
  3196 Renamed:
  3197 
  3198   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3199   multi_count_ext  ~>  multiset_ext
  3200   diff_union_inverse2  ~>  diff_union_cancelR
  3201 
  3202 INCOMPATIBILITY.
  3203 
  3204 * Theory Permutation: replaced local "remove" by List.remove1.
  3205 
  3206 * Code generation: ML and OCaml code is decorated with signatures.
  3207 
  3208 * Theory List: added transpose.
  3209 
  3210 * Library/Nat_Bijection.thy is a collection of bijective functions
  3211 between nat and other types, which supersedes the older libraries
  3212 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3213 
  3214   Constants:
  3215   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3216   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3217   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3218   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3219   Countable.pair_encode           ~> prod_encode
  3220   NatIso.prod2nat                 ~> prod_encode
  3221   NatIso.nat2prod                 ~> prod_decode
  3222   NatIso.sum2nat                  ~> sum_encode
  3223   NatIso.nat2sum                  ~> sum_decode
  3224   NatIso.list2nat                 ~> list_encode
  3225   NatIso.nat2list                 ~> list_decode
  3226   NatIso.set2nat                  ~> set_encode
  3227   NatIso.nat2set                  ~> set_decode
  3228 
  3229   Lemmas:
  3230   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3231   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3232   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3233   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3234   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3235   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3236   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3237   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3238   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3239   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3240   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3241   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3242   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3243 
  3244 * Sledgehammer:
  3245   - Renamed ATP commands:
  3246     atp_info     ~> sledgehammer running_atps
  3247     atp_kill     ~> sledgehammer kill_atps
  3248     atp_messages ~> sledgehammer messages
  3249     atp_minimize ~> sledgehammer minimize
  3250     print_atps   ~> sledgehammer available_atps
  3251     INCOMPATIBILITY.
  3252   - Added user's manual ("isabelle doc sledgehammer").
  3253   - Added option syntax and "sledgehammer_params" to customize
  3254     Sledgehammer's behavior.  See the manual for details.
  3255   - Modified the Isar proof reconstruction code so that it produces
  3256     direct proofs rather than proofs by contradiction.  (This feature
  3257     is still experimental.)
  3258   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3259     full-typed mode.
  3260   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3261 
  3262 * Nitpick:
  3263   - Added and implemented "binary_ints" and "bits" options.
  3264   - Added "std" option and implemented support for nonstandard models.
  3265   - Added and implemented "finitize" option to improve the precision
  3266     of infinite datatypes based on a monotonicity analysis.
  3267   - Added support for quotient types.
  3268   - Added support for "specification" and "ax_specification"
  3269     constructs.
  3270   - Added support for local definitions (for "function" and
  3271     "termination" proofs).
  3272   - Added support for term postprocessors.
  3273   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3274   - Improved efficiency of "destroy_constrs" optimization.
  3275   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3276     record getters.
  3277   - Fixed soundness bug related to higher-order constructors.
  3278   - Fixed soundness bug when "full_descrs" is enabled.
  3279   - Improved precision of set constructs.
  3280   - Added "atoms" option.
  3281   - Added cache to speed up repeated Kodkod invocations on the same
  3282     problems.
  3283   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3284     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3285     "SAT4J_Light".  INCOMPATIBILITY.
  3286   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3287     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3288   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3289 
  3290 * Method "induct" now takes instantiations of the form t, where t is not
  3291   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3292   If this is not intended, t has to be enclosed in parentheses.
  3293   By default, the equalities generated by definitional instantiations
  3294   are pre-simplified, which may cause parameters of inductive cases
  3295   to disappear, or may even delete some of the inductive cases.
  3296   Use "induct (no_simp)" instead of "induct" to restore the old
  3297   behaviour. The (no_simp) option is also understood by the "cases"
  3298   and "nominal_induct" methods, which now perform pre-simplification, too.
  3299   INCOMPATIBILITY.
  3300 
  3301 
  3302 *** HOLCF ***
  3303 
  3304 * Variable names in lemmas generated by the domain package have
  3305 changed; the naming scheme is now consistent with the HOL datatype
  3306 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3307 
  3308 * The domain package no longer defines the function "foo_copy" for
  3309 recursive domain "foo".  The reach lemma is now stated directly in
  3310 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3311 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3312 
  3313 * Most definedness lemmas generated by the domain package (previously
  3314 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3315 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3316 Proofs that used definedness lemmas as intro rules may break,
  3317 potential INCOMPATIBILITY.
  3318 
  3319 * Induction and casedist rules generated by the domain package now
  3320 declare proper case_names (one called "bottom", and one named for each
  3321 constructor).  INCOMPATIBILITY.
  3322 
  3323 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3324 rules are generated for each domain, INCOMPATIBILITY.
  3325 
  3326   foo_bar.reach       ~> foo.reach  bar.reach
  3327   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3328 
  3329 * Some lemmas generated by the domain package have been renamed for
  3330 consistency with the datatype package, INCOMPATIBILITY.
  3331 
  3332   foo.ind        ~> foo.induct
  3333   foo.finite_ind ~> foo.finite_induct
  3334   foo.coind      ~> foo.coinduct
  3335   foo.casedist   ~> foo.exhaust
  3336   foo.exhaust    ~> foo.nchotomy
  3337 
  3338 * For consistency with other definition packages, the fixrec package
  3339 now generates qualified theorem names, INCOMPATIBILITY.
  3340 
  3341   foo_simps  ~> foo.simps
  3342   foo_unfold ~> foo.unfold
  3343   foo_induct ~> foo.induct
  3344 
  3345 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3346 method and internal fixrec proofs now use the default simpset instead.
  3347 INCOMPATIBILITY.
  3348 
  3349 * The "contlub" predicate has been removed.  Proof scripts should use
  3350 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3351 
  3352 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3353 
  3354 * The constants cpair, cfst, and csnd have been removed in favor of
  3355 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3356 
  3357 
  3358 *** ML ***
  3359 
  3360 * Antiquotations for basic formal entities:
  3361 
  3362     @{class NAME}         -- type class
  3363     @{class_syntax NAME}  -- syntax representation of the above
  3364 
  3365     @{type_name NAME}     -- logical type
  3366     @{type_abbrev NAME}   -- type abbreviation
  3367     @{nonterminal NAME}   -- type of concrete syntactic category
  3368     @{type_syntax NAME}   -- syntax representation of any of the above
  3369 
  3370     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3371     @{const_abbrev NAME}  -- abbreviated constant
  3372     @{const_syntax NAME}  -- syntax representation of any of the above
  3373 
  3374 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3375 syntax constant (cf. 'syntax' command).
  3376 
  3377 * Antiquotation @{make_string} inlines a function to print arbitrary
  3378 values similar to the ML toplevel.  The result is compiler dependent
  3379 and may fall back on "?" in certain situations.
  3380 
  3381 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3382 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3383 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3384 not work with the asynchronous Isar document model.
  3385 
  3386 * Configuration options now admit dynamic default values, depending on
  3387 the context or even global references.
  3388 
  3389 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3390 uses an efficient external library if available (for Poly/ML).
  3391 
  3392 * Renamed some important ML structures, while keeping the old names
  3393 for some time as aliases within the structure Legacy:
  3394 
  3395   OuterKeyword  ~>  Keyword
  3396   OuterLex      ~>  Token
  3397   OuterParse    ~>  Parse
  3398   OuterSyntax   ~>  Outer_Syntax
  3399   PrintMode     ~>  Print_Mode
  3400   SpecParse     ~>  Parse_Spec
  3401   ThyInfo       ~>  Thy_Info
  3402   ThyLoad       ~>  Thy_Load
  3403   ThyOutput     ~>  Thy_Output
  3404   TypeInfer     ~>  Type_Infer
  3405 
  3406 Note that "open Legacy" simplifies porting of sources, but forgetting
  3407 to remove it again will complicate porting again in the future.
  3408 
  3409 * Most operations that refer to a global context are named
  3410 accordingly, e.g. Simplifier.global_context or
  3411 ProofContext.init_global.  There are some situations where a global
  3412 context actually works, but under normal circumstances one needs to
  3413 pass the proper local context through the code!
  3414 
  3415 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3416 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3417 Pretty.pp argument to merge, which is absent in the standard
  3418 Theory_Data version.
  3419 
  3420 * Sorts.certify_sort and derived "cert" operations for types and terms
  3421 no longer minimize sorts.  Thus certification at the boundary of the
  3422 inference kernel becomes invariant under addition of class relations,
  3423 which is an important monotonicity principle.  Sorts are now minimized
  3424 in the syntax layer only, at the boundary between the end-user and the
  3425 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3426 explicitly in rare situations.
  3427 
  3428 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3429 emphasize that this is in no way a standard operation.
  3430 INCOMPATIBILITY.
  3431 
  3432 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3433 usual for resolution.  Rare INCOMPATIBILITY.
  3434 
  3435 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3436 to emphasize that these only work in a global situation (which is
  3437 quite rare).
  3438 
  3439 * Curried take and drop in library.ML; negative length is interpreted
  3440 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3441 
  3442 * Proof terms: type substitutions on proof constants now use canonical
  3443 order of type variables.  INCOMPATIBILITY for tools working with proof
  3444 terms.
  3445 
  3446 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3447 may no longer carry premises.  User-level specifications are
  3448 transformed accordingly by Thm.add_axiom/add_def.
  3449 
  3450 
  3451 *** System ***
  3452 
  3453 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3454 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3455 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3456 
  3457 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3458 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3459 respectively.
  3460 
  3461 * Old lib/scripts/polyml-platform is superseded by the
  3462 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3463 variant, even on a 64 bit machine.  The following example setting
  3464 prefers 64 bit if available:
  3465 
  3466   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3467 
  3468 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3469 Isabelle/Scala layer for advanced prover interaction and integration.
  3470 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3471 component.
  3472 
  3473 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3474 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3475 similar to the default assignment of the document preparation system
  3476 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3477 provides some operations for direct access to the font without asking
  3478 the user for manual installation.
  3479 
  3480 
  3481 
  3482 New in Isabelle2009-1 (December 2009)
  3483 -------------------------------------
  3484 
  3485 *** General ***
  3486 
  3487 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3488 one backslash should be used, even in ML sources.
  3489 
  3490 
  3491 *** Pure ***
  3492 
  3493 * Locale interpretation propagates mixins along the locale hierarchy.
  3494 The currently only available mixins are the equations used to map
  3495 local definitions to terms of the target domain of an interpretation.
  3496 
  3497 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3498 loc" to print all interpretations of locale "loc" in the theory.
  3499 Interpretations in proofs are not shown.
  3500 
  3501 * Thoroughly revised locales tutorial.  New section on conditional
  3502 interpretation.
  3503 
  3504 * On instantiation of classes, remaining undefined class parameters
  3505 are formally declared.  INCOMPATIBILITY.
  3506 
  3507 
  3508 *** Document preparation ***
  3509 
  3510 * New generalized style concept for printing terms: @{foo (style) ...}
  3511 instead of @{foo_style style ...}  (old form is still retained for
  3512 backward compatibility).  Styles can be also applied for
  3513 antiquotations prop, term_type and typeof.
  3514 
  3515 
  3516 *** HOL ***
  3517 
  3518 * New proof method "smt" for a combination of first-order logic with
  3519 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3520 fixed-size bitvectors; there is also basic support for higher-order
  3521 features (esp. lambda abstractions).  It is an incomplete decision
  3522 procedure based on external SMT solvers using the oracle mechanism;
  3523 for the SMT solver Z3, this method is proof-producing.  Certificates
  3524 are provided to avoid calling the external solvers solely for
  3525 re-checking proofs.  Due to a remote SMT service there is no need for
  3526 installing SMT solvers locally.  See src/HOL/SMT.
  3527 
  3528 * New commands to load and prove verification conditions generated by
  3529 the Boogie program verifier or derived systems (e.g. the Verifying C
  3530 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3531 
  3532 * New counterexample generator tool 'nitpick' based on the Kodkod
  3533 relational model finder.  See src/HOL/Tools/Nitpick and
  3534 src/HOL/Nitpick_Examples.
  3535 
  3536 * New commands 'code_pred' and 'values' to invoke the predicate
  3537 compiler and to enumerate values of inductive predicates.
  3538 
  3539 * A tabled implementation of the reflexive transitive closure.
  3540 
  3541 * New implementation of quickcheck uses generic code generator;
  3542 default generators are provided for all suitable HOL types, records
  3543 and datatypes.  Old quickcheck can be re-activated importing theory
  3544 Library/SML_Quickcheck.
  3545 
  3546 * New testing tool Mirabelle for automated proof tools.  Applies
  3547 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3548 every proof step in a theory.  To be used in batch mode via the
  3549 "mirabelle" utility.
  3550 
  3551 * New proof method "sos" (sum of squares) for nonlinear real
  3552 arithmetic (originally due to John Harison). It requires theory
  3553 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3554 works well in practice on quantifier-free real arithmetic with +, -,
  3555 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3556 inequalities between polynomials.  It makes use of external
  3557 semidefinite programming solvers.  Method "sos" generates a
  3558 certificate that can be pasted into the proof thus avoiding the need
  3559 to call an external tool every time the proof is checked.  See
  3560 src/HOL/Library/Sum_Of_Squares.
  3561 
  3562 * New method "linarith" invokes existing linear arithmetic decision
  3563 procedure only.
  3564 
  3565 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3566 
  3567 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3568 Causes full type information to be output to the ATPs.  This slows
  3569 ATPs down considerably but eliminates a source of unsound "proofs"
  3570 that fail later.
  3571 
  3572 * New method "metisFT": A version of metis that uses full type
  3573 information in order to avoid failures of proof reconstruction.
  3574 
  3575 * New evaluator "approximate" approximates an real valued term using
  3576 the same method as the approximation method.
  3577 
  3578 * Method "approximate" now supports arithmetic expressions as
  3579 boundaries of intervals and implements interval splitting and Taylor
  3580 series expansion.
  3581 
  3582 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3583 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3584 
  3585 * New theory SupInf of the supremum and infimum operators for sets of
  3586 reals.
  3587 
  3588 * New theory Probability, which contains a development of measure
  3589 theory, eventually leading to Lebesgue integration and probability.
  3590 
  3591 * Extended Multivariate Analysis to include derivation and Brouwer's
  3592 fixpoint theorem.
  3593 
  3594 * Reorganization of number theory, INCOMPATIBILITY:
  3595   - new number theory development for nat and int, in theories Divides
  3596     and GCD as well as in new session Number_Theory
  3597   - some constants and facts now suffixed with _nat and _int
  3598     accordingly
  3599   - former session NumberTheory now named Old_Number_Theory, including
  3600     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3601   - moved theory Pocklington from src/HOL/Library to
  3602     src/HOL/Old_Number_Theory
  3603 
  3604 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3605 lcm of finite and infinite sets. It is shown that they form a complete
  3606 lattice.
  3607 
  3608 * Class semiring_div requires superclass no_zero_divisors and proof of
  3609 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3610 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3611 generalized to class semiring_div, subsuming former theorems
  3612 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3613 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3614 INCOMPATIBILITY.
  3615 
  3616 * Refinements to lattice classes and sets:
  3617   - less default intro/elim rules in locale variant, more default
  3618     intro/elim rules in class variant: more uniformity
  3619   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3620     le_inf_iff
  3621   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3622     sup_aci)
  3623   - renamed ACI to inf_sup_aci
  3624   - new class "boolean_algebra"
  3625   - class "complete_lattice" moved to separate theory
  3626     "Complete_Lattice"; corresponding constants (and abbreviations)
  3627     renamed and with authentic syntax:
  3628     Set.Inf ~>    Complete_Lattice.Inf
  3629     Set.Sup ~>    Complete_Lattice.Sup
  3630     Set.INFI ~>   Complete_Lattice.INFI
  3631     Set.SUPR ~>   Complete_Lattice.SUPR
  3632     Set.Inter ~>  Complete_Lattice.Inter
  3633     Set.Union ~>  Complete_Lattice.Union
  3634     Set.INTER ~>  Complete_Lattice.INTER
  3635     Set.UNION ~>  Complete_Lattice.UNION
  3636   - authentic syntax for
  3637     Set.Pow
  3638     Set.image
  3639   - mere abbreviations:
  3640     Set.empty               (for bot)
  3641     Set.UNIV                (for top)
  3642     Set.inter               (for inf, formerly Set.Int)
  3643     Set.union               (for sup, formerly Set.Un)
  3644     Complete_Lattice.Inter  (for Inf)
  3645     Complete_Lattice.Union  (for Sup)
  3646     Complete_Lattice.INTER  (for INFI)
  3647     Complete_Lattice.UNION  (for SUPR)
  3648   - object-logic definitions as far as appropriate
  3649 
  3650 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3651 Un_subset_iff are explicitly deleted as default simp rules; then also
  3652 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3653 deleted to achieve the desired effect.
  3654 
  3655 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3656 rules by default any longer; the same applies to min_max.inf_absorb1
  3657 etc.  INCOMPATIBILITY.
  3658 
  3659 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3660 pred_set_conv by default.  INCOMPATIBILITY.
  3661 
  3662 * Power operations on relations and functions are now one dedicated
  3663 constant "compow" with infix syntax "^^".  Power operation on
  3664 multiplicative monoids retains syntax "^" and is now defined generic
  3665 in class power.  INCOMPATIBILITY.
  3666 
  3667 * Relation composition "R O S" now has a more standard argument order:
  3668 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3669 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3670 break, since the O_assoc rule was not rewritten like this.  Fix using
  3671 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3672 
  3673 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3674 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3675 INCOMPATIBILITY.
  3676 
  3677 * Most rules produced by inductive and datatype package have mandatory
  3678 prefixes.  INCOMPATIBILITY.
  3679 
  3680 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3681 the attribute of the same name.  Each of the theorems in the list
  3682 DERIV_intros assumes composition with an additional function and
  3683 matches a variable to the derivative, which has to be solved by the
  3684 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3685 of most elementary terms.  Former Maclauren.DERIV_tac and
  3686 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3687 INCOMPATIBILITY.
  3688 
  3689 * Code generator attributes follow the usual underscore convention:
  3690     code_unfold     replaces    code unfold
  3691     code_post       replaces    code post
  3692     etc.
  3693   INCOMPATIBILITY.
  3694 
  3695 * Renamed methods:
  3696     sizechange -> size_change
  3697     induct_scheme -> induction_schema
  3698   INCOMPATIBILITY.
  3699 
  3700 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3701 INCOMPATIBILITY, use "undefined" directly.
  3702 
  3703 * Renamed theorems:
  3704     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3705     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3706     Suc_plus1 -> Suc_eq_plus1
  3707     *anti_sym -> *antisym*
  3708     vector_less_eq_def -> vector_le_def
  3709   INCOMPATIBILITY.
  3710 
  3711 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3712 INCOMPATIBILITY.
  3713 
  3714 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3715 INCOMPATIBILITY.
  3716 
  3717 
  3718 *** HOLCF ***
  3719 
  3720 * Theory Representable defines a class "rep" of domains that are
  3721 representable (via an ep-pair) in the universal domain type "udom".
  3722 Instances are provided for all type constructors defined in HOLCF.
  3723 
  3724 * The 'new_domain' command is a purely definitional version of the
  3725 domain package, for representable domains.  Syntax is identical to the
  3726 old domain package.  The 'new_domain' package also supports indirect
  3727 recursion using previously-defined type constructors.  See
  3728 src/HOLCF/ex/New_Domain.thy for examples.
  3729 
  3730 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3731 on the left-hand side of an equation, and then performs
  3732 simplification.  Rewriting is done using rules declared with the
  3733 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3734 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3735 
  3736 * The pattern-match compiler in 'fixrec' can now handle constructors
  3737 with HOL function types.  Pattern-match combinators for the Pair
  3738 constructor are pre-configured.
  3739 
  3740 * The 'fixrec' package now produces better fixed-point induction rules
  3741 for mutually-recursive definitions:  Induction rules have conclusions
  3742 of the form "P foo bar" instead of "P <foo, bar>".
  3743 
  3744 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3745 been renamed to "below".  The name "below" now replaces "less" in many
  3746 theorem names.  (Legacy theorem names using "less" are still supported
  3747 as well.)
  3748 
  3749 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3750 patterns can be used to generate strictness rules, or to make
  3751 functions more strict (much like the bang-patterns supported by the
  3752 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3753 examples.
  3754 
  3755 
  3756 *** ML ***
  3757 
  3758 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3759 errors and run-time exceptions, including detailed source positions.
  3760 
  3761 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3762 identified entries, with some additional information such as source
  3763 position, logical grouping etc.
  3764 
  3765 * Theory and context data is now introduced by the simplified and
  3766 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3767 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3768 copy operation) is still available for some time.
  3769 
  3770 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3771 provides a high-level programming interface to synchronized state
  3772 variables with atomic update.  This works via pure function
  3773 application within a critical section -- its runtime should be as
  3774 short as possible; beware of deadlocks if critical code is nested,
  3775 either directly or indirectly via other synchronized variables!
  3776 
  3777 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3778 wraps raw ML references, explicitly indicating their non-thread-safe
  3779 behaviour.  The Isar toplevel keeps this structure open, to
  3780 accommodate Proof General as well as quick and dirty interactive
  3781 experiments with references.
  3782 
  3783 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3784 parallel tactical reasoning.
  3785 
  3786 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3787 are similar to SUBPROOF, but are slightly more flexible: only the
  3788 specified parts of the subgoal are imported into the context, and the
  3789 body tactic may introduce new subgoals and schematic variables.
  3790 
  3791 * Old tactical METAHYPS, which does not observe the proof context, has
  3792 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3793 or Subgoal.FOCUS etc.
  3794 
  3795 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3796 functors have their own ML name space there is no point to mark them
  3797 separately.)  Minor INCOMPATIBILITY.
  3798 
  3799 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3800 
  3801 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3802 INCOMPATIBILITY.
  3803 
  3804 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3805 mode, which avoids critical setmp.
  3806 
  3807 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3808 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3809 Attrib/Method.setup introduced in Isabelle2009.
  3810 
  3811 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3812 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3813 last resort.  INCOMPATIBILITY.
  3814 
  3815 * Display.pretty_thm now requires a proper context (cf. former
  3816 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3817 or even Display.pretty_thm_without_context as last resort.
  3818 INCOMPATIBILITY.
  3819 
  3820 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3821 Syntax.pretty_typ/term directly, preferably with proper context
  3822 instead of global theory.
  3823 
  3824 
  3825 *** System ***
  3826 
  3827 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3828 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3829 usedir option -q.
  3830 
  3831 * Support for additional "Isabelle components" via etc/components, see
  3832 also the system manual.
  3833 
  3834 * The isabelle makeall tool now operates on all components with
  3835 IsaMakefile, not just hardwired "logics".
  3836 
  3837 * Removed "compress" option from isabelle-process and isabelle usedir;
  3838 this is always enabled.
  3839 
  3840 * Discontinued support for Poly/ML 4.x versions.
  3841 
  3842 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3843 on a given logic image.  This requires the lighttpd webserver and is
  3844 currently supported on Linux only.
  3845 
  3846 
  3847 
  3848 New in Isabelle2009 (April 2009)
  3849 --------------------------------
  3850 
  3851 *** General ***
  3852 
  3853 * Simplified main Isabelle executables, with less surprises on
  3854 case-insensitive file-systems (such as Mac OS).
  3855 
  3856   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3857     "isatool."
  3858 
  3859   - The former "isabelle" alias for "isabelle-process" has been
  3860     removed (should rarely occur to regular users).
  3861 
  3862   - The former "isabelle-interface" and its alias "Isabelle" have been
  3863     removed (interfaces are now regular Isabelle tools).
  3864 
  3865 Within scripts and make files, the Isabelle environment variables
  3866 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3867 respectively.  (The latter are still available as legacy feature.)
  3868 
  3869 The old isabelle-interface wrapper could react in confusing ways if
  3870 the interface was uninstalled or changed otherwise.  Individual
  3871 interface tool configuration is now more explicit, see also the
  3872 Isabelle system manual.  In particular, Proof General is now available
  3873 via "isabelle emacs".
  3874 
  3875 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3876 purge installed copies of Isabelle executables and re-run "isabelle
  3877 install -p ...", or use symlinks.
  3878 
  3879 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3880 old ~/isabelle, which was slightly non-standard and apt to cause
  3881 surprises on case-insensitive file-systems (such as Mac OS).
  3882 
  3883 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3884 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3885 care is required when using older releases of Isabelle.  Note that
  3886 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3887 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3888 
  3889 * Proofs of fully specified statements are run in parallel on
  3890 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3891 a regular 4-core machine, if the initial heap space is made reasonably
  3892 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3893 
  3894 * The main reference manuals ("isar-ref", "implementation", and
  3895 "system") have been updated and extended.  Formally checked references
  3896 as hyperlinks are now available uniformly.
  3897 
  3898 
  3899 *** Pure ***
  3900 
  3901 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3902 respects.  The most important changes are listed below.  See the
  3903 Tutorial on Locales ("locales" manual) for details.
  3904 
  3905 - In locale expressions, instantiation replaces renaming.  Parameters
  3906 must be declared in a for clause.  To aid compatibility with previous
  3907 parameter inheritance, in locale declarations, parameters that are not
  3908 'touched' (instantiation position "_" or omitted) are implicitly added
  3909 with their syntax at the beginning of the for clause.
  3910 
  3911 - Syntax from abbreviations and definitions in locales is available in
  3912 locale expressions and context elements.  The latter is particularly
  3913 useful in locale declarations.
  3914 
  3915 - More flexible mechanisms to qualify names generated by locale
  3916 expressions.  Qualifiers (prefixes) may be specified in locale
  3917 expressions, and can be marked as mandatory (syntax: "name!:") or
  3918 optional (syntax "name?:").  The default depends for plain "name:"
  3919 depends on the situation where a locale expression is used: in
  3920 commands 'locale' and 'sublocale' prefixes are optional, in
  3921 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3922 implicit qualifiers derived from the parameter names of a locale are
  3923 no longer generated.
  3924 
  3925 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3926 instantiation clause in "interpretation" and "interpret" (square
  3927 brackets) is no longer available.  Use locale expressions.
  3928 
  3929 - When converting proof scripts, mandatory qualifiers in
  3930 'interpretation' and 'interpret' should be retained by default, even
  3931 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3932 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3933 in locale expressions range over a single locale instance only.
  3934 
  3935 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3936 In existing theorem specifications replace the includes element by the
  3937 respective context elements of the included locale, omitting those
  3938 that are already present in the theorem specification.  Multiple
  3939 assume elements of a locale should be replaced by a single one
  3940 involving the locale predicate.  In the proof body, declarations (most
  3941 notably theorems) may be regained by interpreting the respective
  3942 locales in the proof context as required (command "interpret").
  3943 
  3944 If using "includes" in replacement of a target solely because the
  3945 parameter types in the theorem are not as general as in the target,
  3946 consider declaring a new locale with additional type constraints on
  3947 the parameters (context element "constrains").
  3948 
  3949 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3950 
  3951 - Locale interpretation commands no longer attempt to simplify goal.
  3952 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3953 methods intro_locales and unfold_locales to clarify.
  3954 
  3955 - Locale interpretation commands no longer accept interpretation
  3956 attributes.  INCOMPATIBILITY.
  3957 
  3958 * Class declaration: so-called "base sort" must not be given in import
  3959 list any longer, but is inferred from the specification.  Particularly
  3960 in HOL, write
  3961 
  3962     class foo = ...
  3963 
  3964 instead of
  3965 
  3966     class foo = type + ...
  3967 
  3968 * Class target: global versions of theorems stemming do not carry a
  3969 parameter prefix any longer.  INCOMPATIBILITY.
  3970 
  3971 * Class 'instance' command no longer accepts attached definitions.
  3972 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3973 
  3974 * Recovered hiding of consts, which was accidentally broken in
  3975 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3976 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3977 
  3978 * Slightly more coherent Pure syntax, with updated documentation in
  3979 isar-ref manual.  Removed locales meta_term_syntax and
  3980 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3981 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3982 directly in regular applications.
  3983 
  3984 * There is a new syntactic category "float_const" for signed decimal
  3985 fractions (e.g. 123.45 or -123.45).
  3986 
  3987 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3988 interface with 'setup' command instead.
  3989 
  3990 * Command 'local_setup' is similar to 'setup', but operates on a local
  3991 theory context.
  3992 
  3993 * The 'axiomatization' command now only works within a global theory
  3994 context.  INCOMPATIBILITY.
  3995 
  3996 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3997 hypotheses.  Sorts required in the course of reasoning need to be
  3998 covered by the constraints in the initial statement, completed by the
  3999 type instance information of the background theory.  Non-trivial sort
  4000 hypotheses, which rarely occur in practice, may be specified via
  4001 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  4002 
  4003   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  4004 
  4005 The result contains an implicit sort hypotheses as before --
  4006 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  4007 normalization.
  4008 
  4009 * Generalized Isar history, with support for linear undo, direct state
  4010 addressing etc.
  4011 
  4012 * Changed defaults for unify configuration options:
  4013 
  4014   unify_trace_bound = 50 (formerly 25)
  4015   unify_search_bound = 60 (formerly 30)
  4016 
  4017 * Different bookkeeping for code equations (INCOMPATIBILITY):
  4018 
  4019   a) On theory merge, the last set of code equations for a particular
  4020      constant is taken (in accordance with the policy applied by other
  4021      parts of the code generator framework).
  4022 
  4023   b) Code equations stemming from explicit declarations (e.g. code
  4024      attribute) gain priority over default code equations stemming
  4025      from definition, primrec, fun etc.
  4026 
  4027 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  4028 
  4029 * Unified theorem tables for both code generators.  Thus [code
  4030 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  4031 
  4032 * Command 'find_consts' searches for constants based on type and name
  4033 patterns, e.g.
  4034 
  4035     find_consts "_ => bool"
  4036 
  4037 By default, matching is against subtypes, but it may be restricted to
  4038 the whole type.  Searching by name is possible.  Multiple queries are
  4039 conjunctive and queries may be negated by prefixing them with a
  4040 hyphen:
  4041 
  4042     find_consts strict: "_ => bool" name: "Int" -"int => int"
  4043 
  4044 * New 'find_theorems' criterion "solves" matches theorems that
  4045 directly solve the current goal (modulo higher-order unification).
  4046 
  4047 * Auto solve feature for main theorem statements: whenever a new goal
  4048 is stated, "find_theorems solves" is called; any theorems that could
  4049 solve the lemma directly are listed as part of the goal state.
  4050 Cf. associated options in Proof General Isabelle settings menu,
  4051 enabled by default, with reasonable timeout for pathological cases of
  4052 higher-order unification.
  4053 
  4054 
  4055 *** Document preparation ***
  4056 
  4057 * Antiquotation @{lemma} now imitates a regular terminal proof,
  4058 demanding keyword 'by' and supporting the full method expression
  4059 syntax just like the Isar command 'by'.
  4060 
  4061 
  4062 *** HOL ***
  4063 
  4064 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  4065 points Main and Complex_Main remain as before.
  4066 
  4067 * Logic image HOL-Plain provides a minimal HOL with the most important
  4068 tools available (inductive, datatype, primrec, ...).  This facilitates
  4069 experimentation and tool development.  Note that user applications
  4070 (and library theories) should never refer to anything below theory
  4071 Main, as before.
  4072 
  4073 * Logic image HOL-Main stops at theory Main, and thus facilitates
  4074 experimentation due to shorter build times.
  4075 
  4076 * Logic image HOL-NSA contains theories of nonstandard analysis which
  4077 were previously part of former HOL-Complex.  Entry point Hyperreal
  4078 remains valid, but theories formerly using Complex_Main should now use
  4079 new entry point Hypercomplex.
  4080 
  4081 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  4082 Posix processes.  Avoids potentially expensive forking of the ML
  4083 process.  New thread-based implementation also works on non-Unix
  4084 platforms (Cygwin).  Provers are no longer hardwired, but defined
  4085 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  4086 commands are covered in the isar-ref manual.
  4087 
  4088 * Wrapper scripts for remote SystemOnTPTP service allows to use
  4089 sledgehammer without local ATP installation (Vampire etc.). Other
  4090 provers may be included via suitable ML wrappers, see also
  4091 src/HOL/ATP_Linkup.thy.
  4092 
  4093 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  4094 menu.
  4095 
  4096 * The metis method no longer fails because the theorem is too trivial
  4097 (contains the empty clause).
  4098 
  4099 * The metis method now fails in the usual manner, rather than raising
  4100 an exception, if it determines that it cannot prove the theorem.
  4101 
  4102 * Method "coherent" implements a prover for coherent logic (see also
  4103 src/Tools/coherent.ML).
  4104 
  4105 * Constants "undefined" and "default" replace "arbitrary".  Usually
  4106 "undefined" is the right choice to replace "arbitrary", though
  4107 logically there is no difference.  INCOMPATIBILITY.
  4108 
  4109 * Command "value" now integrates different evaluation mechanisms.  The
  4110 result of the first successful evaluation mechanism is printed.  In
  4111 square brackets a particular named evaluation mechanisms may be
  4112 specified (currently, [SML], [code] or [nbe]).  See further
  4113 src/HOL/ex/Eval_Examples.thy.
  4114 
  4115 * Normalization by evaluation now allows non-leftlinear equations.
  4116 Declare with attribute [code nbe].
  4117 
  4118 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  4119 as the structured Isar versions "cases" and "induct", cf. the
  4120 corresponding "cases" and "induct" attributes.  Mutual induction rules
  4121 are now presented as a list of individual projections
  4122 (e.g. foo_bar.inducts for types foo and bar); the old format with
  4123 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  4124 rare situations a different rule is selected --- notably nested tuple
  4125 elimination instead of former prod.exhaust: use explicit (case_tac t
  4126 rule: prod.exhaust) here.
  4127 
  4128 * Attributes "cases", "induct", "coinduct" support "del" option.
  4129 
  4130 * Removed fact "case_split_thm", which duplicates "case_split".
  4131 
  4132 * The option datatype has been moved to a new theory Option.  Renamed
  4133 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  4134 
  4135 * New predicate "strict_mono" classifies strict functions on partial
  4136 orders.  With strict functions on linear orders, reasoning about
  4137 (in)equalities is facilitated by theorems "strict_mono_eq",
  4138 "strict_mono_less_eq" and "strict_mono_less".
  4139 
  4140 * Some set operations are now proper qualified constants with
  4141 authentic syntax.  INCOMPATIBILITY:
  4142 
  4143     op Int ~>   Set.Int
  4144     op Un ~>    Set.Un
  4145     INTER ~>    Set.INTER
  4146     UNION ~>    Set.UNION
  4147     Inter ~>    Set.Inter
  4148     Union ~>    Set.Union
  4149     {} ~>       Set.empty
  4150     UNIV ~>     Set.UNIV
  4151 
  4152 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4153 theory Set.
  4154 
  4155 * Auxiliary class "itself" has disappeared -- classes without any
  4156 parameter are treated as expected by the 'class' command.
  4157 
  4158 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4159 
  4160 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4161 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4162 
  4163 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4164 method "approximation".  It proves formulas on real values by using
  4165 interval arithmetic.  In the formulas are also the transcendental
  4166 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4167 allowed. For examples see
  4168 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4169 
  4170 * Theory "Reflection" now resides in HOL/Library.
  4171 
  4172 * Entry point to Word library now simply named "Word".
  4173 INCOMPATIBILITY.
  4174 
  4175 * Made source layout more coherent with logical distribution
  4176 structure:
  4177 
  4178     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4179     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4180     src/HOL/Library/GCD.thy ~> src/HOL/
  4181     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4182     src/HOL/Library/Parity.thy ~> src/HOL/
  4183     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4184     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4185     src/HOL/Real/Lubs.thy ~> src/HOL/
  4186     src/HOL/Real/PReal.thy ~> src/HOL/
  4187     src/HOL/Real/Rational.thy ~> src/HOL/
  4188     src/HOL/Real/RComplete.thy ~> src/HOL/
  4189     src/HOL/Real/RealDef.thy ~> src/HOL/
  4190     src/HOL/Real/RealPow.thy ~> src/HOL/
  4191     src/HOL/Real/Real.thy ~> src/HOL/
  4192     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4193     src/HOL/Complex/Complex.thy ~> src/HOL/
  4194     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4195     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4196     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4197     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4198     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4199     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4200     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4201     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4202     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4203     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4204     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4205     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4206     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4207     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4208     src/HOL/Real/Float ~> src/HOL/Library/
  4209     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4210     src/HOL/Real/RealVector.thy ~> src/HOL/
  4211 
  4212     src/HOL/arith_data.ML ~> src/HOL/Tools
  4213     src/HOL/hologic.ML ~> src/HOL/Tools
  4214     src/HOL/simpdata.ML ~> src/HOL/Tools
  4215     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4216     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4217     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4218     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4219     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4220     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4221     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4222 
  4223     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4224     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4225     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4226     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4227     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4228     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4229 
  4230 * If methods "eval" and "evaluation" encounter a structured proof
  4231 state with !!/==>, only the conclusion is evaluated to True (if
  4232 possible), avoiding strange error messages.
  4233 
  4234 * Method "sizechange" automates termination proofs using (a
  4235 modification of) the size-change principle.  Requires SAT solver.  See
  4236 src/HOL/ex/Termination.thy for examples.
  4237 
  4238 * Simplifier: simproc for let expressions now unfolds if bound
  4239 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4240 
  4241 * Method "arith": Linear arithmetic now ignores all inequalities when
  4242 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4243 
  4244 * New attribute "arith" for facts that should always be used
  4245 automatically by arithmetic. It is intended to be used locally in
  4246 proofs, e.g.
  4247 
  4248   assumes [arith]: "x > 0"
  4249 
  4250 Global usage is discouraged because of possible performance impact.
  4251 
  4252 * New classes "top" and "bot" with corresponding operations "top" and
  4253 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4254 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4255 
  4256 * Changed definition lemma "less_fun_def" in order to provide an
  4257 instance for preorders on functions; use lemma "less_le" instead.
  4258 INCOMPATIBILITY.
  4259 
  4260 * Theory Orderings: class "wellorder" moved here, with explicit
  4261 induction rule "less_induct" as assumption.  For instantiation of
  4262 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4263 INCOMPATIBILITY.
  4264 
  4265 * Theory Orderings: added class "preorder" as superclass of "order".
  4266 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4267 etc. slightly changed.  Some theorems named order_class.* now named
  4268 preorder_class.*.
  4269 
  4270 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4271 "diag" to "Id_on".
  4272 
  4273 * Theory Finite_Set: added a new fold combinator of type
  4274 
  4275   ('a => 'b => 'b) => 'b => 'a set => 'b
  4276 
  4277 Occasionally this is more convenient than the old fold combinator
  4278 which is now defined in terms of the new one and renamed to
  4279 fold_image.
  4280 
  4281 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4282 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4283 extended with further lemmas!).  At the moment both still exist but
  4284 the former will disappear at some point.
  4285 
  4286 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4287 class recpower.  Type-specific simp rules for various recpower types
  4288 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4289 
  4290 rat_power_0    -> power_0
  4291 rat_power_Suc  -> power_Suc
  4292 realpow_0      -> power_0
  4293 realpow_Suc    -> power_Suc
  4294 complexpow_0   -> power_0
  4295 complexpow_Suc -> power_Suc
  4296 power_poly_0   -> power_0
  4297 power_poly_Suc -> power_Suc
  4298 
  4299 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4300 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4301 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4302 of lemmas from Divides has been generalized from nat to class
  4303 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4304 renames resulting from duplicate elimination:
  4305 
  4306     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4307     zero_dvd_iff ~>         dvd_0_left_iff
  4308     dvd_0 ~>                dvd_0_right
  4309     DIVISION_BY_ZERO_DIV ~> div_by_0
  4310     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4311     mult_div ~>             div_mult_self2_is_id
  4312     mult_mod ~>             mod_mult_self2_is_0
  4313 
  4314 * Theory IntDiv: removed many lemmas that are instances of class-based
  4315 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4316 rename old lemmas as follows:
  4317 
  4318 dvd_diff               -> nat_dvd_diff
  4319 dvd_zminus_iff         -> dvd_minus_iff
  4320 mod_add1_eq            -> mod_add_eq
  4321 mod_mult1_eq           -> mod_mult_right_eq
  4322 mod_mult1_eq'          -> mod_mult_left_eq
  4323 mod_mult_distrib_mod   -> mod_mult_eq
  4324 nat_mod_add_left_eq    -> mod_add_left_eq
  4325 nat_mod_add_right_eq   -> mod_add_right_eq
  4326 nat_mod_div_trivial    -> mod_div_trivial
  4327 nat_mod_mod_trivial    -> mod_mod_trivial
  4328 zdiv_zadd_self1        -> div_add_self1
  4329 zdiv_zadd_self2        -> div_add_self2
  4330 zdiv_zmult_self1       -> div_mult_self2_is_id
  4331 zdiv_zmult_self2       -> div_mult_self1_is_id
  4332 zdvd_triv_left         -> dvd_triv_left
  4333 zdvd_triv_right        -> dvd_triv_right
  4334 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4335 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4336 zmod_zadd_left_eq      -> mod_add_left_eq
  4337 zmod_zadd_right_eq     -> mod_add_right_eq
  4338 zmod_zadd_self1        -> mod_add_self1
  4339 zmod_zadd_self2        -> mod_add_self2
  4340 zmod_zadd1_eq          -> mod_add_eq
  4341 zmod_zdiff1_eq         -> mod_diff_eq
  4342 zmod_zdvd_zmod         -> mod_mod_cancel
  4343 zmod_zmod_cancel       -> mod_mod_cancel
  4344 zmod_zmult_self1       -> mod_mult_self2_is_0
  4345 zmod_zmult_self2       -> mod_mult_self1_is_0
  4346 zmod_1                 -> mod_by_1
  4347 zdiv_1                 -> div_by_1
  4348 zdvd_abs1              -> abs_dvd_iff
  4349 zdvd_abs2              -> dvd_abs_iff
  4350 zdvd_refl              -> dvd_refl
  4351 zdvd_trans             -> dvd_trans
  4352 zdvd_zadd              -> dvd_add
  4353 zdvd_zdiff             -> dvd_diff
  4354 zdvd_zminus_iff        -> dvd_minus_iff
  4355 zdvd_zminus2_iff       -> minus_dvd_iff
  4356 zdvd_zmultD            -> dvd_mult_right
  4357 zdvd_zmultD2           -> dvd_mult_left
  4358 zdvd_zmult_mono        -> mult_dvd_mono
  4359 zdvd_0_right           -> dvd_0_right
  4360 zdvd_0_left            -> dvd_0_left_iff
  4361 zdvd_1_left            -> one_dvd
  4362 zminus_dvd_iff         -> minus_dvd_iff
  4363 
  4364 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4365 
  4366 * The real numbers offer decimal input syntax: 12.34 is translated
  4367 into 1234/10^2. This translation is not reversed upon output.
  4368 
  4369 * Theory Library/Polynomial defines an abstract type 'a poly of
  4370 univariate polynomials with coefficients of type 'a.  In addition to
  4371 the standard ring operations, it also supports div and mod.  Code
  4372 generation is also supported, using list-style constructors.
  4373 
  4374 * Theory Library/Inner_Product defines a class of real_inner for real
  4375 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4376 => real.  Class real_inner is a subclass of real_normed_vector from
  4377 theory RealVector.
  4378 
  4379 * Theory Library/Product_Vector provides instances for the product
  4380 type 'a * 'b of several classes from RealVector and Inner_Product.
  4381 Definitions of addition, subtraction, scalar multiplication, norms,
  4382 and inner products are included.
  4383 
  4384 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4385 addition to the field operations, numerals and case syntax are also
  4386 supported.
  4387 
  4388 * Theory Library/Diagonalize provides constructive version of Cantor's
  4389 first diagonalization argument.
  4390 
  4391 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4392 zlcm (for int); carried together from various gcd/lcm developements in
  4393 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4394 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4395 may recover tupled syntax as follows:
  4396 
  4397     hide (open) const gcd
  4398     abbreviation gcd where
  4399       "gcd == (%(a, b). GCD.gcd a b)"
  4400     notation (output)
  4401       GCD.gcd ("gcd '(_, _')")
  4402 
  4403 The same works for lcm, zgcd, zlcm.
  4404 
  4405 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4406 instantiations for algebraic structures.  Removed some duplicate
  4407 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4408 
  4409 * ML antiquotation @{code} takes a constant as argument and generates
  4410 corresponding code in background and inserts name of the corresponding
  4411 resulting ML value/function/datatype constructor binding in place.
  4412 All occurrences of @{code} with a single ML block are generated
  4413 simultaneously.  Provides a generic and safe interface for
  4414 instrumentalizing code generation.  See
  4415 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4416 In future you ought to refrain from ad-hoc compiling generated SML
  4417 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4418 cannot refer to constants for which user-defined serializations are
  4419 set.  Refer to the corresponding ML counterpart directly in that
  4420 cases.
  4421 
  4422 * Command 'rep_datatype': instead of theorem names the command now
  4423 takes a list of terms denoting the constructors of the type to be
  4424 represented as datatype.  The characteristic theorems have to be
  4425 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4426 have disappeared in favour of existing ones:
  4427 
  4428     unit_induct                 ~> unit.induct
  4429     prod_induct                 ~> prod.induct
  4430     sum_induct                  ~> sum.induct
  4431     Suc_Suc_eq                  ~> nat.inject
  4432     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4433 
  4434 
  4435 *** HOL-Algebra ***
  4436 
  4437 * New locales for orders and lattices where the equivalence relation
  4438 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4439 locales use a record structure with field eq for the equivalence.
  4440 
  4441 * New theory of factorial domains.
  4442 
  4443 * Units_l_inv and Units_r_inv are now simp rules by default.
  4444 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4445 and/or r_inv will now also require deletion of these lemmas.
  4446 
  4447 * Renamed the following theorems, INCOMPATIBILITY:
  4448 
  4449 UpperD ~> Upper_memD
  4450 LowerD ~> Lower_memD
  4451 least_carrier ~> least_closed
  4452 greatest_carrier ~> greatest_closed
  4453 greatest_Lower_above ~> greatest_Lower_below
  4454 one_zero ~> carrier_one_zero
  4455 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4456 
  4457 
  4458 *** HOL-Nominal ***
  4459 
  4460 * Nominal datatypes can now contain type-variables.
  4461 
  4462 * Commands 'nominal_inductive' and 'equivariance' work with local
  4463 theory targets.
  4464 
  4465 * Nominal primrec can now works with local theory targets and its
  4466 specification syntax now conforms to the general format as seen in
  4467 'inductive' etc.
  4468 
  4469 * Method "perm_simp" honours the standard simplifier attributes
  4470 (no_asm), (no_asm_use) etc.
  4471 
  4472 * The new predicate #* is defined like freshness, except that on the
  4473 left hand side can be a set or list of atoms.
  4474 
  4475 * Experimental command 'nominal_inductive2' derives strong induction
  4476 principles for inductive definitions.  In contrast to
  4477 'nominal_inductive', which can only deal with a fixed number of
  4478 binders, it can deal with arbitrary expressions standing for sets of
  4479 atoms to be avoided.  The only inductive definition we have at the
  4480 moment that needs this generalisation is the typing rule for Lets in
  4481 the algorithm W:
  4482 
  4483  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4484  -----------------------------------------------------------------
  4485          Gamma |- Let x be t1 in t2 : T2
  4486 
  4487 In this rule one wants to avoid all the binders that are introduced by
  4488 "close Gamma T1".  We are looking for other examples where this
  4489 feature might be useful.  Please let us know.
  4490 
  4491 
  4492 *** HOLCF ***
  4493 
  4494 * Reimplemented the simplification procedure for proving continuity
  4495 subgoals.  The new simproc is extensible; users can declare additional
  4496 continuity introduction rules with the attribute [cont2cont].
  4497 
  4498 * The continuity simproc now uses a different introduction rule for
  4499 solving continuity subgoals on terms with lambda abstractions.  In
  4500 some rare cases the new simproc may fail to solve subgoals that the
  4501 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4502 Potential INCOMPATIBILITY.
  4503 
  4504 * Command 'fixrec': specification syntax now conforms to the general
  4505 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4506 examples.  INCOMPATIBILITY.
  4507 
  4508 
  4509 *** ZF ***
  4510 
  4511 * Proof of Zorn's Lemma for partial orders.
  4512 
  4513 
  4514 *** ML ***
  4515 
  4516 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4517 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4518 depends on multithreading, so timouts will not work before Poly/ML
  4519 5.2.1!
  4520 
  4521 * High-level support for concurrent ML programming, see
  4522 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4523 particularly convenient to organize independent functional
  4524 computations.  The concept of "synchronized variables" provides a
  4525 higher-order interface for components with shared state, avoiding the
  4526 delicate details of mutexes and condition variables.  (Requires
  4527 Poly/ML 5.2.1 or later.)
  4528 
  4529 * ML bindings produced via Isar commands are stored within the Isar
  4530 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4531 become thread-safe and work with undo as expected (concerning
  4532 top-level bindings, not side-effects on global references).
  4533 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4534 compiler at runtime; really global bindings need to be given outside a
  4535 theory.  (Requires Poly/ML 5.2 or later.)
  4536 
  4537 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4538 context.  Top-level ML bindings are stored within the proof context in
  4539 a purely sequential fashion, disregarding the nested proof structure.
  4540 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4541 proof.  (Requires Poly/ML 5.2 or later.)
  4542 
  4543 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4544 and Method.setup, as well as Isar commands 'attribute_setup' and
  4545 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4546 existing code accordingly, or use plain 'setup' together with old
  4547 Method.add_method.
  4548 
  4549 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4550 to 'a -> thm, while results are always tagged with an authentic oracle
  4551 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4552 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4553 accordingly.  Note that extra performance may be gained by producing
  4554 the cterm carefully, avoiding slow Thm.cterm_of.
  4555 
  4556 * Simplified interface for defining document antiquotations via
  4557 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4558 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4559 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4560 examples.
  4561 
  4562 * More systematic treatment of long names, abstract name bindings, and
  4563 name space operations.  Basic operations on qualified names have been
  4564 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4565 Long_Name.append.  Old type bstring has been mostly replaced by
  4566 abstract type binding (see structure Binding), which supports precise
  4567 qualification by packages and local theory targets, as well as proper
  4568 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4569 bstring values into Binding.name, or better pass through abstract
  4570 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4571 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4572 
  4573 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4574 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4575 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4576 
  4577 * Disposed old type and term read functions (Sign.read_def_typ,
  4578 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4579 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4580 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4581 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4582 resort for legacy applications.
  4583 
  4584 * Disposed old declarations, tactics, tactic combinators that refer to
  4585 the simpset or claset of an implicit theory (such as Addsimps,
  4586 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4587 embedded ML text, or local_simpset_of with a proper context passed as
  4588 explicit runtime argument.
  4589 
  4590 * Rules and tactics that read instantiations (read_instantiate,
  4591 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4592 context, which is required for parsing and type-checking.  Moreover,
  4593 the variables are specified as plain indexnames, not string encodings
  4594 thereof.  INCOMPATIBILITY.
  4595 
  4596 * Generic Toplevel.add_hook interface allows to analyze the result of
  4597 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4598 for theorem dependency output of transactions resulting in a new
  4599 theory state.
  4600 
  4601 * ML antiquotations: block-structured compilation context indicated by
  4602 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4603 
  4604   @{binding name}                         - basic name binding
  4605   @{let ?pat = term}                      - term abbreviation (HO matching)
  4606   @{note name = fact}                     - fact abbreviation
  4607   @{thm fact}                             - singleton fact (with attributes)
  4608   @{thms fact}                            - general fact (with attributes)
  4609   @{lemma prop by method}                 - singleton goal
  4610   @{lemma prop by meth1 meth2}            - singleton goal
  4611   @{lemma prop1 ... propN by method}      - general goal
  4612   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4613   @{lemma (open) ...}                     - open derivation
  4614 
  4615 
  4616 *** System ***
  4617 
  4618 * The Isabelle "emacs" tool provides a specific interface to invoke
  4619 Proof General / Emacs, with more explicit failure if that is not
  4620 installed (the old isabelle-interface script silently falls back on
  4621 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4622 installation location of the Proof General distribution.
  4623 
  4624 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4625 the Isabelle process into a JVM/Scala application.  See
  4626 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4627 process wrapper has been discontinued.)
  4628 
  4629 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4630 
  4631 * Various status messages (with exact source position information) are
  4632 emitted, if proper markup print mode is enabled.  This allows
  4633 user-interface components to provide detailed feedback on internal
  4634 prover operations.
  4635 
  4636 
  4637 
  4638 New in Isabelle2008 (June 2008)
  4639 -------------------------------
  4640 
  4641 *** General ***
  4642 
  4643 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4644 and updated, with formally checked references as hyperlinks.
  4645 
  4646 * Theory loader: use_thy (and similar operations) no longer set the
  4647 implicit ML context, which was occasionally hard to predict and in
  4648 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4649 provides a proper context already.
  4650 
  4651 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4652 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4653 'use' within a theory file will do the job.
  4654 
  4655 * Name space merge now observes canonical order, i.e. the second space
  4656 is inserted into the first one, while existing entries in the first
  4657 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4658 swap theory imports.
  4659 
  4660 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4661 INCOMPATIBILITY in identifier syntax etc.
  4662 
  4663 * Outer syntax: string tokens no longer admit escaped white space,
  4664 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4665 white space without escapes.
  4666 
  4667 * Outer syntax: string tokens may contain arbitrary character codes
  4668 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4669 "foo_bar".
  4670 
  4671 
  4672 *** Pure ***
  4673 
  4674 * Context-dependent token translations.  Default setup reverts locally
  4675 fixed variables, and adds hilite markup for undeclared frees.
  4676 
  4677 * Unused theorems can be found using the new command 'unused_thms'.
  4678 There are three ways of invoking it:
  4679 
  4680 (1) unused_thms
  4681      Only finds unused theorems in the current theory.
  4682 
  4683 (2) unused_thms thy_1 ... thy_n -
  4684      Finds unused theorems in the current theory and all of its ancestors,
  4685      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4686 
  4687 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4688      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4689      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4690      their ancestors.
  4691 
  4692 In order to increase the readability of the list produced by
  4693 unused_thms, theorems that have been created by a particular instance
  4694 of a theory command such as 'inductive' or 'function' are considered
  4695 to belong to the same "group", meaning that if at least one theorem in
  4696 this group is used, the other theorems in the same group are no longer
  4697 reported as unused.  Moreover, if all theorems in the group are
  4698 unused, only one theorem in the group is displayed.
  4699 
  4700 Note that proof objects have to be switched on in order for
  4701 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4702 usually the case when using Proof General with the default settings).
  4703 
  4704 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4705 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4706 duplicate fact bindings, or even accidental fact duplications.  Note
  4707 that tools may maintain dynamically scoped facts systematically, using
  4708 PureThy.add_thms_dynamic.
  4709 
  4710 * Command 'hide' now allows to hide from "fact" name space as well.
  4711 
  4712 * Eliminated destructive theorem database, simpset, claset, and
  4713 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4714 update of theories within ML code.
  4715 
  4716 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4717 INCOMPATIBILITY, object-logics depending on former Pure require
  4718 additional setup PureThy.old_appl_syntax_setup; object-logics
  4719 depending on former CPure need to refer to Pure.
  4720 
  4721 * Commands 'use' and 'ML' are now purely functional, operating on
  4722 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4723 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4724 INCOMPATIBILITY.
  4725 
  4726 * Command 'setup': discontinued implicit version with ML reference.
  4727 
  4728 * Instantiation target allows for simultaneous specification of class
  4729 instance operations together with an instantiation proof.
  4730 Type-checking phase allows to refer to class operations uniformly.
  4731 See src/HOL/Complex/Complex.thy for an Isar example and
  4732 src/HOL/Library/Eval.thy for an ML example.
  4733 
  4734 * Indexing of literal facts: be more serious about including only
  4735 facts from the visible specification/proof context, but not the
  4736 background context (locale etc.).  Affects `prop` notation and method
  4737 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4738 situations.
  4739 
  4740 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4741 "(open)" option, which used to expose internal bound variables to the
  4742 proof text.
  4743 
  4744 * Isar statements: removed obsolete case "rule_context".
  4745 INCOMPATIBILITY, better use explicit fixes/assumes.
  4746 
  4747 * Locale proofs: default proof step now includes 'unfold_locales';
  4748 hence 'proof' without argument may be used to unfold locale
  4749 predicates.
  4750 
  4751 
  4752 *** Document preparation ***
  4753 
  4754 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4755 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4756 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4757 spot); renamed link color from "darkblue" to "linkcolor" (default
  4758 value unchanged, can be redefined via \definecolor); no longer sets
  4759 "a4paper" option (unnecessary or even intrusive).
  4760 
  4761 * Antiquotation @{lemma A method} proves proposition A by the given
  4762 method (either a method name or a method name plus (optional) method
  4763 arguments in parentheses) and prints A just like @{prop A}.
  4764 
  4765 
  4766 *** HOL ***
  4767 
  4768 * New primrec package.  Specification syntax conforms in style to
  4769 definition/function/....  No separate induction rule is provided.  The
  4770 "primrec" command distinguishes old-style and new-style specifications
  4771 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4772 When adjusting theories, beware: constants stemming from new-style
  4773 primrec specifications have authentic syntax.
  4774 
  4775 * Metis prover is now an order of magnitude faster, and also works
  4776 with multithreading.
  4777 
  4778 * Metis: the maximum number of clauses that can be produced from a
  4779 theorem is now given by the attribute max_clauses.  Theorems that
  4780 exceed this number are ignored, with a warning printed.
  4781 
  4782 * Sledgehammer no longer produces structured proofs by default. To
  4783 enable, declare [[sledgehammer_full = true]].  Attributes
  4784 reconstruction_modulus, reconstruction_sorts renamed
  4785 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4786 
  4787 * Method "induct_scheme" derives user-specified induction rules
  4788 from well-founded induction and completeness of patterns. This factors
  4789 out some operations that are done internally by the function package
  4790 and makes them available separately.  See
  4791 src/HOL/ex/Induction_Scheme.thy for examples.
  4792 
  4793 * More flexible generation of measure functions for termination
  4794 proofs: Measure functions can be declared by proving a rule of the
  4795 form "is_measure f" and giving it the [measure_function] attribute.
  4796 The "is_measure" predicate is logically meaningless (always true), and
  4797 just guides the heuristic.  To find suitable measure functions, the
  4798 termination prover sets up the goal "is_measure ?f" of the appropriate
  4799 type and generates all solutions by prolog-style backwards proof using
  4800 the declared rules.
  4801 
  4802 This setup also deals with rules like
  4803 
  4804   "is_measure f ==> is_measure (list_size f)"
  4805 
  4806 which accommodates nested datatypes that recurse through lists.
  4807 Similar rules are predeclared for products and option types.
  4808 
  4809 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4810 
  4811   INCOMPATIBILITIES:
  4812 
  4813   - Definitions of overloaded constants on sets have to be replaced by
  4814     definitions on => and bool.
  4815 
  4816   - Some definitions of overloaded operators on sets can now be proved
  4817     using the definitions of the operators on => and bool.  Therefore,
  4818     the following theorems have been renamed:
  4819 
  4820       subset_def   -> subset_eq
  4821       psubset_def  -> psubset_eq
  4822       set_diff_def -> set_diff_eq
  4823       Compl_def    -> Compl_eq
  4824       Sup_set_def  -> Sup_set_eq
  4825       Inf_set_def  -> Inf_set_eq
  4826       sup_set_def  -> sup_set_eq
  4827       inf_set_def  -> inf_set_eq
  4828 
  4829   - Due to the incompleteness of the HO unification algorithm, some
  4830     rules such as subst may require manual instantiation, if some of
  4831     the unknowns in the rule is a set.
  4832 
  4833   - Higher order unification and forward proofs:
  4834     The proof pattern
  4835 
  4836       have "P (S::'a set)" <...>
  4837       then have "EX S. P S" ..
  4838 
  4839     no longer works (due to the incompleteness of the HO unification
  4840     algorithm) and must be replaced by the pattern
  4841 
  4842       have "EX S. P S"
  4843       proof
  4844         show "P S" <...>
  4845       qed
  4846 
  4847   - Calculational reasoning with subst (or similar rules):
  4848     The proof pattern
  4849 
  4850       have "P (S::'a set)" <...>
  4851       also have "S = T" <...>
  4852       finally have "P T" .
  4853 
  4854     no longer works (for similar reasons as the previous example) and
  4855     must be replaced by something like
  4856 
  4857       have "P (S::'a set)" <...>
  4858       moreover have "S = T" <...>
  4859       ultimately have "P T" by simp
  4860 
  4861   - Tactics or packages written in ML code:
  4862     Code performing pattern matching on types via
  4863 
  4864       Type ("set", [T]) => ...
  4865 
  4866     must be rewritten. Moreover, functions like strip_type or
  4867     binder_types no longer return the right value when applied to a
  4868     type of the form
  4869 
  4870       T1 => ... => Tn => U => bool
  4871 
  4872     rather than
  4873 
  4874       T1 => ... => Tn => U set
  4875 
  4876 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4877 Wellfounded_Relations to theory Wellfounded.
  4878 
  4879 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4880 
  4881 * Class finite no longer treats UNIV as class parameter.  Use class
  4882 enum from theory Library/Enum instead to achieve a similar effect.
  4883 INCOMPATIBILITY.
  4884 
  4885 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4886 and "Cons".  INCOMPATIBILITY.
  4887 
  4888 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4889 Potential INCOMPATIBILITY.
  4890 
  4891 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4892 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4893 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4894 
  4895 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4896 the same name in theory Orderings:
  4897 
  4898   less_trans
  4899   less_linear
  4900   le_imp_less_or_eq
  4901   le_less_trans
  4902   less_le_trans
  4903   less_not_sym
  4904   less_asym
  4905 
  4906 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4907 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4908 and different variable names.
  4909 
  4910 * Library/Option_ord.thy: Canonical order on option type.
  4911 
  4912 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4913 finite maps.
  4914 
  4915 * Library/Countable.thy: Type class for countable types.
  4916 
  4917 * Theory Int: The representation of numerals has changed.  The infix
  4918 operator BIT and the bit datatype with constructors B0 and B1 have
  4919 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4920 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4921 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4922 accordingly.
  4923 
  4924 * Theory Nat: definition of <= and < on natural numbers no longer
  4925 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4926 le_def and less_def have disappeared.  Consider lemmas not_less
  4927 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4928 
  4929 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4930 (whose purpose mainly is for various fold_set functionals) have been
  4931 abandoned in favor of the existing algebraic classes
  4932 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4933 lower_semilattice (resp. upper_semilattice) and linorder.
  4934 INCOMPATIBILITY.
  4935 
  4936 * Theory Transitive_Closure: induct and cases rules now declare proper
  4937 case_names ("base" and "step").  INCOMPATIBILITY.
  4938 
  4939 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4940 lattices.  The form set-specific version is available as
  4941 Inductive.lfp_ordinal_induct_set.
  4942 
  4943 * Renamed theorems "power.simps" to "power_int.simps".
  4944 INCOMPATIBILITY.
  4945 
  4946 * Class semiring_div provides basic abstract properties of semirings
  4947 with division and modulo operations.  Subsumes former class dvd_mod.
  4948 
  4949 * Merged theories IntDef, Numeral and IntArith into unified theory
  4950 Int.  INCOMPATIBILITY.
  4951 
  4952 * Theory Library/Code_Index: type "index" now represents natural
  4953 numbers rather than integers.  INCOMPATIBILITY.
  4954 
  4955 * New class "uminus" with operation "uminus" (split of from class
  4956 "minus" which now only has operation "minus", binary).
  4957 INCOMPATIBILITY.
  4958 
  4959 * Constants "card", "internal_split", "option_map" now with authentic
  4960 syntax.  INCOMPATIBILITY.
  4961 
  4962 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4963 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4964 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4965 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4966 Sup_set_def, le_def, less_def, option_map_def now with object
  4967 equality.  INCOMPATIBILITY.
  4968 
  4969 * Records. Removed K_record, and replaced it by pure lambda term
  4970 %x. c. The simplifier setup is now more robust against eta expansion.
  4971 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4972 
  4973 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4974 
  4975 * Library/ListVector: new theory of arithmetic vector operations.
  4976 
  4977 * Library/Order_Relation: new theory of various orderings as sets of
  4978 pairs.  Defines preorders, partial orders, linear orders and
  4979 well-orders on sets and on types.
  4980 
  4981 
  4982 *** ZF ***
  4983 
  4984 * Renamed some theories to allow to loading both ZF and HOL in the
  4985 same session:
  4986 
  4987   Datatype  -> Datatype_ZF
  4988   Inductive -> Inductive_ZF
  4989   Int       -> Int_ZF
  4990   IntDiv    -> IntDiv_ZF
  4991   Nat       -> Nat_ZF
  4992   List      -> List_ZF
  4993   Main      -> Main_ZF
  4994 
  4995 INCOMPATIBILITY: ZF theories that import individual theories below
  4996 Main might need to be adapted.  Regular theory Main is still
  4997 available, as trivial extension of Main_ZF.
  4998 
  4999 
  5000 *** ML ***
  5001 
  5002 * ML within Isar: antiquotation @{const name} or @{const
  5003 name(typargs)} produces statically-checked Const term.
  5004 
  5005 * Functor NamedThmsFun: data is available to the user as dynamic fact
  5006 (of the same name).  Removed obsolete print command.
  5007 
  5008 * Removed obsolete "use_legacy_bindings" function.
  5009 
  5010 * The ``print mode'' is now a thread-local value derived from a global
  5011 template (the former print_mode reference), thus access becomes
  5012 non-critical.  The global print_mode reference is for session
  5013 management only; user-code should use print_mode_value,
  5014 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  5015 
  5016 * Functions system/system_out provide a robust way to invoke external
  5017 shell commands, with propagation of interrupts (requires Poly/ML
  5018 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  5019 
  5020 
  5021 *** System ***
  5022 
  5023 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  5024 in accordance with Proof General 3.7, which prefers GNU emacs.
  5025 
  5026 * isatool tty runs Isabelle process with plain tty interaction;
  5027 optional line editor may be specified via ISABELLE_LINE_EDITOR
  5028 setting, the default settings attempt to locate "ledit" and "rlwrap".
  5029 
  5030 * isatool browser now works with Cygwin as well, using general
  5031 "javapath" function defined in Isabelle process environment.
  5032 
  5033 * YXML notation provides a simple and efficient alternative to
  5034 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  5035 isatool yxml as described in the Isabelle system manual.
  5036 
  5037 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  5038 provides general wrapper for managing an Isabelle process in a robust
  5039 fashion, with ``cooked'' output from stdin/stderr.
  5040 
  5041 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  5042 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  5043 
  5044 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  5045 way of changing the user's settings is via
  5046 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  5047 script.
  5048 
  5049 * Multithreading.max_threads := 0 refers to the number of actual CPU
  5050 cores of the underlying machine, which is a good starting point for
  5051 optimal performance tuning.  The corresponding usedir option -M allows
  5052 "max" as an alias for "0".  WARNING: does not work on certain versions
  5053 of Mac OS (with Poly/ML 5.1).
  5054 
  5055 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  5056 adverse effect of Isabelle flooding interactive front-ends (notably
  5057 ProofGeneral / XEmacs).
  5058 
  5059 
  5060 
  5061 New in Isabelle2007 (November 2007)
  5062 -----------------------------------
  5063 
  5064 *** General ***
  5065 
  5066 * More uniform information about legacy features, notably a
  5067 warning/error of "Legacy feature: ...", depending on the state of the
  5068 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  5069 legacy features will disappear eventually.
  5070 
  5071 * Theory syntax: the header format ``theory A = B + C:'' has been
  5072 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  5073 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  5074 
  5075 * Theory syntax: the old non-Isar theory file format has been
  5076 discontinued altogether.  Note that ML proof scripts may still be used
  5077 with Isar theories; migration is usually quite simple with the ML
  5078 function use_legacy_bindings.  INCOMPATIBILITY.
  5079 
  5080 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  5081 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  5082 quotes.
  5083 
  5084 * Theory loader: be more serious about observing the static theory
  5085 header specifications (including optional directories), but not the
  5086 accidental file locations of previously successful loads.  The strict
  5087 update policy of former update_thy is now already performed by
  5088 use_thy, so the former has been removed; use_thys updates several
  5089 theories simultaneously, just as 'imports' within a theory header
  5090 specification, but without merging the results.  Potential
  5091 INCOMPATIBILITY: may need to refine theory headers and commands
  5092 ROOT.ML which depend on load order.
  5093 
  5094 * Theory loader: optional support for content-based file
  5095 identification, instead of the traditional scheme of full physical
  5096 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  5097 (cf. the system manual).  The new scheme allows to work with
  5098 non-finished theories in persistent session images, such that source
  5099 files may be moved later on without requiring reloads.
  5100 
  5101 * Theory loader: old-style ML proof scripts being *attached* to a thy
  5102 file (with the same base name as the theory) are considered a legacy
  5103 feature, which will disappear eventually. Even now, the theory loader
  5104 no longer maintains dependencies on such files.
  5105 
  5106 * Syntax: the scope for resolving ambiguities via type-inference is
  5107 now limited to individual terms, instead of whole simultaneous
  5108 specifications as before. This greatly reduces the complexity of the
  5109 syntax module and improves flexibility by separating parsing and
  5110 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  5111 'fixes' etc.) are required in rare situations.
  5112 
  5113 * Syntax: constants introduced by new-style packages ('definition',
  5114 'abbreviation' etc.) are passed through the syntax module in
  5115 ``authentic mode''. This means that associated mixfix annotations
  5116 really stick to such constants, independently of potential name space
  5117 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  5118 trees are represented slightly differently, may need to adapt syntax
  5119 translations accordingly. Use CONST marker in 'translations' and
  5120 @{const_syntax} antiquotation in 'parse_translation' etc.
  5121 
  5122 * Legacy goal package: reduced interface to the bare minimum required
  5123 to keep existing proof scripts running.  Most other user-level
  5124 functions are now part of the OldGoals structure, which is *not* open
  5125 by default (consider isatool expandshort before open OldGoals).
  5126 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  5127 these tend to cause confusion about the actual goal (!) context being
  5128 used here, which is not necessarily the same as the_context().
  5129 
  5130 * Command 'find_theorems': supports "*" wild-card in "name:"
  5131 criterion; "with_dups" option.  Certain ProofGeneral versions might
  5132 support a specific search form (see ProofGeneral/CHANGES).
  5133 
  5134 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  5135 by default, which means that "prems" (and also "fixed variables") are
  5136 suppressed from proof state output.  Note that the ProofGeneral
  5137 settings mechanism allows to change and save options persistently, but
  5138 older versions of Isabelle will fail to start up if a negative prems
  5139 limit is imposed.
  5140 
  5141 * Local theory targets may be specified by non-nested blocks of
  5142 ``context/locale/class ... begin'' followed by ``end''.  The body may
  5143 contain definitions, theorems etc., including any derived mechanism
  5144 that has been implemented on top of these primitives.  This concept
  5145 generalizes the existing ``theorem (in ...)'' towards more versatility
  5146 and scalability.
  5147 
  5148 * Proof General interface: proper undo of final 'end' command;
  5149 discontinued Isabelle/classic mode (ML proof scripts).
  5150 
  5151 
  5152 *** Document preparation ***
  5153 
  5154 * Added antiquotation @{theory name} which prints the given name,
  5155 after checking that it refers to a valid ancestor theory in the
  5156 current context.
  5157 
  5158 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5159 check the given source text as ML type/structure, printing verbatim.
  5160 
  5161 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5162 "c args == rhs" given in the current context.  (Any number of
  5163 arguments may be given on the LHS.)
  5164 
  5165 
  5166 *** Pure ***
  5167 
  5168 * The 'class' package offers a combination of axclass and locale to
  5169 achieve Haskell-like type classes in Isabelle.  Definitions and
  5170 theorems within a class context produce both relative results (with
  5171 implicit parameters according to the locale context), and polymorphic
  5172 constants with qualified polymorphism (according to the class
  5173 context).  Within the body context of a 'class' target, a separate
  5174 syntax layer ("user space type system") takes care of converting
  5175 between global polymorphic consts and internal locale representation.
  5176 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5177 "isatool doc classes" provides a tutorial.
  5178 
  5179 * Generic code generator framework allows to generate executable
  5180 code for ML and Haskell (including Isabelle classes).  A short usage
  5181 sketch:
  5182 
  5183     internal compilation:
  5184         export_code <list of constants (term syntax)> in SML
  5185     writing SML code to a file:
  5186         export_code <list of constants (term syntax)> in SML <filename>
  5187     writing OCaml code to a file:
  5188         export_code <list of constants (term syntax)> in OCaml <filename>
  5189     writing Haskell code to a bunch of files:
  5190         export_code <list of constants (term syntax)> in Haskell <filename>
  5191 
  5192     evaluating closed propositions to True/False using code generation:
  5193         method ``eval''
  5194 
  5195 Reasonable default setup of framework in HOL.
  5196 
  5197 Theorem attributs for selecting and transforming function equations theorems:
  5198 
  5199     [code fun]:        select a theorem as function equation for a specific constant
  5200     [code fun del]:    deselect a theorem as function equation for a specific constant
  5201     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5202     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5203 
  5204 User-defined serializations (target in {SML, OCaml, Haskell}):
  5205 
  5206     code_const <and-list of constants (term syntax)>
  5207       {(target) <and-list of const target syntax>}+
  5208 
  5209     code_type <and-list of type constructors>
  5210       {(target) <and-list of type target syntax>}+
  5211 
  5212     code_instance <and-list of instances>
  5213       {(target)}+
  5214         where instance ::= <type constructor> :: <class>
  5215 
  5216     code_class <and_list of classes>
  5217       {(target) <and-list of class target syntax>}+
  5218         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5219 
  5220 code_instance and code_class only are effective to target Haskell.
  5221 
  5222 For example usage see src/HOL/ex/Codegenerator.thy and
  5223 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5224 generation from Isabelle/HOL theories is available via "isatool doc
  5225 codegen".
  5226 
  5227 * Code generator: consts in 'consts_code' Isar commands are now
  5228 referred to by usual term syntax (including optional type
  5229 annotations).
  5230 
  5231 * Command 'no_translations' removes translation rules from theory
  5232 syntax.
  5233 
  5234 * Overloaded definitions are now actually checked for acyclic
  5235 dependencies.  The overloading scheme is slightly more general than
  5236 that of Haskell98, although Isabelle does not demand an exact
  5237 correspondence to type class and instance declarations.
  5238 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5239 exotic versions of overloading -- at the discretion of the user!
  5240 
  5241 Polymorphic constants are represented via type arguments, i.e. the
  5242 instantiation that matches an instance against the most general
  5243 declaration given in the signature.  For example, with the declaration
  5244 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5245 as c(nat).  Overloading is essentially simultaneous structural
  5246 recursion over such type arguments.  Incomplete specification patterns
  5247 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5248 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5249 Command 'print_theory' outputs the normalized system of recursive
  5250 equations, see section "definitions".
  5251 
  5252 * Configuration options are maintained within the theory or proof
  5253 context (with name and type bool/int/string), providing a very simple
  5254 interface to a poor-man's version of general context data.  Tools may
  5255 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5256 these values using Config.get etc.  Users may change options via an
  5257 associated attribute of the same name.  This form of context
  5258 declaration works particularly well with commands 'declare' or
  5259 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5260 very easy to avoid global references, which would not observe Isar
  5261 toplevel undo/redo and fail to work with multithreading.
  5262 
  5263 Various global ML references of Pure and HOL have been turned into
  5264 configuration options:
  5265 
  5266   Unify.search_bound		unify_search_bound
  5267   Unify.trace_bound		unify_trace_bound
  5268   Unify.trace_simp		unify_trace_simp
  5269   Unify.trace_types		unify_trace_types
  5270   Simplifier.simp_depth_limit	simp_depth_limit
  5271   Blast.depth_limit		blast_depth_limit
  5272   DatatypeProp.dtK		datatype_distinctness_limit
  5273   fast_arith_neq_limit  	fast_arith_neq_limit
  5274   fast_arith_split_limit	fast_arith_split_limit
  5275 
  5276 * Named collections of theorems may be easily installed as context
  5277 data using the functor NamedThmsFun (see also
  5278 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5279 attributes; there is also a toplevel print command.  This facility is
  5280 just a common case of general context data, which is the preferred way
  5281 for anything more complex than just a list of facts in canonical
  5282 order.
  5283 
  5284 * Isar: command 'declaration' augments a local theory by generic
  5285 declaration functions written in ML.  This enables arbitrary content
  5286 being added to the context, depending on a morphism that tells the
  5287 difference of the original declaration context wrt. the application
  5288 context encountered later on.
  5289 
  5290 * Isar: proper interfaces for simplification procedures.  Command
  5291 'simproc_setup' declares named simprocs (with match patterns, and body
  5292 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5293 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5294 
  5295 * Isar: an extra pair of brackets around attribute declarations
  5296 abbreviates a theorem reference involving an internal dummy fact,
  5297 which will be ignored later --- only the effect of the attribute on
  5298 the background context will persist.  This form of in-place
  5299 declarations is particularly useful with commands like 'declare' and
  5300 'using', for example ``have A using [[simproc a]] by simp''.
  5301 
  5302 * Isar: method "assumption" (and implicit closing of subproofs) now
  5303 takes simple non-atomic goal assumptions into account: after applying
  5304 an assumption as a rule the resulting subgoals are solved by atomic
  5305 assumption steps.  This is particularly useful to finish 'obtain'
  5306 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5307 without referring to the original premise "!!x. P x ==> thesis" in the
  5308 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5309 more permissive.
  5310 
  5311 * Isar: implicit use of prems from the Isar proof context is
  5312 considered a legacy feature.  Common applications like ``have A .''
  5313 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5314 referencing facts explicitly here improves readability and
  5315 maintainability of proof texts.
  5316 
  5317 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5318 the obtained context from the course of reasoning!  For example:
  5319 
  5320   assume "EX x y. A x & B y"   -- "any previous fact"
  5321   then guess x and y by clarify
  5322 
  5323 This technique is potentially adventurous, depending on the facts and
  5324 proof tools being involved here.
  5325 
  5326 * Isar: known facts from the proof context may be specified as literal
  5327 propositions, using ASCII back-quote syntax.  This works wherever
  5328 named facts used to be allowed so far, in proof commands, proof
  5329 methods, attributes etc.  Literal facts are retrieved from the context
  5330 according to unification of type and term parameters.  For example,
  5331 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5332 theorems in the current context, then these are valid literal facts:
  5333 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5334 
  5335 There is also a proof method "fact" which does the same composition
  5336 for explicit goal states, e.g. the following proof texts coincide with
  5337 certain special cases of literal facts:
  5338 
  5339   have "A" by fact                 ==  note `A`
  5340   have "A ==> B" by fact           ==  note `A ==> B`
  5341   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5342   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5343 
  5344 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5345 outer syntax.  Thus symbolic identifiers may be used without
  5346 additional white space in declarations like this: ``assume *: A''.
  5347 
  5348 * Isar: 'print_facts' prints all local facts of the current context,
  5349 both named and unnamed ones.
  5350 
  5351 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5352 
  5353   def x == "t" and y == "u"
  5354 
  5355 * Isar: added command 'unfolding', which is structurally similar to
  5356 'using', but affects both the goal state and facts by unfolding given
  5357 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5358 'unfolded' attribute may be replaced by first-class proof text.
  5359 
  5360 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5361 and command 'unfolding' now all support object-level equalities
  5362 (potentially conditional).  The underlying notion of rewrite rule is
  5363 analogous to the 'rule_format' attribute, but *not* that of the
  5364 Simplifier (which is usually more generous).
  5365 
  5366 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5367 premises of a theorem by n. Useful in conjunction with drule.
  5368 
  5369 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5370 method expression within a sandbox consisting of the first N
  5371 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5372 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5373 simplifies all new goals that emerge from applying rule foo to the
  5374 originally first one.
  5375 
  5376 * Isar: schematic goals are no longer restricted to higher-order
  5377 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5378 expected.
  5379 
  5380 * Isar: the conclusion of a long theorem statement is now either
  5381 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5382 (essentially a disjunction of cases with local parameters and
  5383 assumptions).  The latter allows to express general elimination rules
  5384 adequately; in this notation common elimination rules look like this:
  5385 
  5386   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5387     assumes "EX x. P x"
  5388     obtains x where "P x"
  5389 
  5390   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5391     assumes "A & B"
  5392     obtains A and B
  5393 
  5394   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5395     assumes "A | B"
  5396     obtains
  5397       A
  5398     | B
  5399 
  5400 The subsequent classical rules even refer to the formal "thesis"
  5401 explicitly:
  5402 
  5403   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5404     obtains "~ thesis"
  5405 
  5406   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5407     obtains "thesis ==> something"
  5408 
  5409 The actual proof of an 'obtains' statement is analogous to that of the
  5410 Isar proof element 'obtain', only that there may be several cases.
  5411 Optional case names may be specified in parentheses; these will be
  5412 available both in the present proof and as annotations in the
  5413 resulting rule, for later use with the 'cases' method (cf. attribute
  5414 case_names).
  5415 
  5416 * Isar: the assumptions of a long theorem statement are available as
  5417 "assms" fact in the proof context.  This is more appropriate than the
  5418 (historical) "prems", which refers to all assumptions of the current
  5419 context, including those from the target locale, proof body etc.
  5420 
  5421 * Isar: 'print_statement' prints theorems from the current theory or
  5422 proof context in long statement form, according to the syntax of a
  5423 top-level lemma.
  5424 
  5425 * Isar: 'obtain' takes an optional case name for the local context
  5426 introduction rule (default "that").
  5427 
  5428 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5429 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5430 to occur.
  5431 
  5432 * Pure: syntax "CONST name" produces a fully internalized constant
  5433 according to the current context.  This is particularly useful for
  5434 syntax translations that should refer to internal constant
  5435 representations independently of name spaces.
  5436 
  5437 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5438 instead of "FOO ". This allows multiple binder declarations to coexist
  5439 in the same context.  INCOMPATIBILITY.
  5440 
  5441 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5442 primitive that also works in a locale context (both for constants and
  5443 fixed variables). Type declaration and internal syntactic representation
  5444 of given constants retrieved from the context. Likewise, the
  5445 'no_notation' command allows to remove given syntax annotations from the
  5446 current context.
  5447 
  5448 * Isar/locales: new derived specification elements 'axiomatization',
  5449 'definition', 'abbreviation', which support type-inference, admit
  5450 object-level specifications (equality, equivalence).  See also the
  5451 isar-ref manual.  Examples:
  5452 
  5453   axiomatization
  5454     eq  (infix "===" 50) where
  5455     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5456 
  5457   definition "f x y = x + y + 1"
  5458   definition g where "g x = f x x"
  5459 
  5460   abbreviation
  5461     neq  (infix "=!=" 50) where
  5462     "x =!= y == ~ (x === y)"
  5463 
  5464 These specifications may be also used in a locale context.  Then the
  5465 constants being introduced depend on certain fixed parameters, and the
  5466 constant name is qualified by the locale base name.  An internal
  5467 abbreviation takes care for convenient input and output, making the
  5468 parameters implicit and using the original short name.  See also
  5469 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5470 entities from a monomorphic theory.
  5471 
  5472 Presently, abbreviations are only available 'in' a target locale, but
  5473 not inherited by general import expressions.  Also note that
  5474 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5475 'translations' in common applications.  The "no_abbrevs" print mode
  5476 prevents folding of abbreviations in term output.
  5477 
  5478 Concrete syntax is attached to specified constants in internal form,
  5479 independently of name spaces.  The parse tree representation is
  5480 slightly different -- use 'notation' instead of raw 'syntax', and
  5481 'translations' with explicit "CONST" markup to accommodate this.
  5482 
  5483 * Pure/Isar: unified syntax for new-style specification mechanisms
  5484 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5485 full type inference and dummy patterns ("_").  For example:
  5486 
  5487   definition "K x _ = x"
  5488 
  5489   inductive conj for A B
  5490   where "A ==> B ==> conj A B"
  5491 
  5492 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5493 the current context.  Print mode "no_abbrevs" prevents inversion of
  5494 abbreviations on output.
  5495 
  5496 * Isar/locales: improved parameter handling: use of locales "var" and
  5497 "struct" no longer necessary; - parameter renamings are no longer
  5498 required to be injective.  For example, this allows to define
  5499 endomorphisms as locale endom = homom mult mult h.
  5500 
  5501 * Isar/locales: changed the way locales with predicates are defined.
  5502 Instead of accumulating the specification, the imported expression is
  5503 now an interpretation.  INCOMPATIBILITY: different normal form of
  5504 locale expressions.  In particular, in interpretations of locales with
  5505 predicates, goals repesenting already interpreted fragments are not
  5506 removed automatically.  Use methods `intro_locales' and
  5507 `unfold_locales'; see below.
  5508 
  5509 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5510 provide backward reasoning on locales predicates.  The methods are
  5511 aware of interpretations and discharge corresponding goals.
  5512 `intro_locales' is less aggressive then `unfold_locales' and does not
  5513 unfold predicates to assumptions.
  5514 
  5515 * Isar/locales: the order in which locale fragments are accumulated
  5516 has changed.  This enables to override declarations from fragments due
  5517 to interpretations -- for example, unwanted simp rules.
  5518 
  5519 * Isar/locales: interpretation in theories and proof contexts has been
  5520 extended.  One may now specify (and prove) equations, which are
  5521 unfolded in interpreted theorems.  This is useful for replacing
  5522 defined concepts (constants depending on locale parameters) by
  5523 concepts already existing in the target context.  Example:
  5524 
  5525   interpretation partial_order ["op <= :: [int, int] => bool"]
  5526     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5527 
  5528 Typically, the constant `partial_order.less' is created by a
  5529 definition specification element in the context of locale
  5530 partial_order.
  5531 
  5532 * Method "induct": improved internal context management to support
  5533 local fixes and defines on-the-fly. Thus explicit meta-level
  5534 connectives !!  and ==> are rarely required anymore in inductive goals
  5535 (using object-logic connectives for this purpose has been long
  5536 obsolete anyway). Common proof patterns are explained in
  5537 src/HOL/Induct/Common_Patterns.thy, see also
  5538 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5539 examples.
  5540 
  5541 * Method "induct": improved handling of simultaneous goals. Instead of
  5542 introducing object-level conjunction, the statement is now split into
  5543 several conclusions, while the corresponding symbolic cases are nested
  5544 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5545 see src/HOL/Induct/Common_Patterns.thy, for example.
  5546 
  5547 * Method "induct": mutual induction rules are now specified as a list
  5548 of rule sharing the same induction cases. HOL packages usually provide
  5549 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5550 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5551 mutual induction rules differently, i.e. like this:
  5552 
  5553   (induct rule: foo_bar.inducts)
  5554   (induct set: foo bar)
  5555   (induct pred: foo bar)
  5556   (induct type: foo bar)
  5557 
  5558 The ML function ProjectRule.projections turns old-style rules into the
  5559 new format.
  5560 
  5561 * Method "coinduct": dual of induction, see
  5562 src/HOL/Library/Coinductive_List.thy for various examples.
  5563 
  5564 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5565 considered a legacy feature.
  5566 
  5567 * Attribute "symmetric" produces result with standardized schematic
  5568 variables (index 0).  Potential INCOMPATIBILITY.
  5569 
  5570 * Simplifier: by default the simplifier trace only shows top level
  5571 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5572 default. Thus there is less danger of being flooded by the trace. The
  5573 trace indicates where parts have been suppressed.
  5574 
  5575 * Provers/classical: removed obsolete classical version of elim_format
  5576 attribute; classical elim/dest rules are now treated uniformly when
  5577 manipulating the claset.
  5578 
  5579 * Provers/classical: stricter checks to ensure that supplied intro,
  5580 dest and elim rules are well-formed; dest and elim rules must have at
  5581 least one premise.
  5582 
  5583 * Provers/classical: attributes dest/elim/intro take an optional
  5584 weight argument for the rule (just as the Pure versions).  Weights are
  5585 ignored by automated tools, but determine the search order of single
  5586 rule steps.
  5587 
  5588 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5589 where the body does not mention the bound variable.  Note that dummy
  5590 patterns implicitly depend on their context of bounds, which makes
  5591 "{_. _}" match any set comprehension as expected.  Potential
  5592 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5593 constant "_idtdummy" in the binding position.
  5594 
  5595 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5596 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5597 for example "A -> B" => "Pi A (%_. B)".
  5598 
  5599 * Pure: 'class_deps' command visualizes the subclass relation, using
  5600 the graph browser tool.
  5601 
  5602 * Pure: 'print_theory' now suppresses certain internal declarations by
  5603 default; use '!' option for full details.
  5604 
  5605 
  5606 *** HOL ***
  5607 
  5608 * Method "metis" proves goals by applying the Metis general-purpose
  5609 resolution prover (see also http://gilith.com/software/metis/).
  5610 Examples are in the directory MetisExamples.  WARNING: the
  5611 Isabelle/HOL-Metis integration does not yet work properly with
  5612 multi-threading.
  5613 
  5614 * Command 'sledgehammer' invokes external automatic theorem provers as
  5615 background processes.  It generates calls to the "metis" method if
  5616 successful. These can be pasted into the proof.  Users do not have to
  5617 wait for the automatic provers to return.  WARNING: does not really
  5618 work with multi-threading.
  5619 
  5620 * New "auto_quickcheck" feature tests outermost goal statements for
  5621 potential counter-examples.  Controlled by ML references
  5622 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5623 5000 milliseconds).  Fails silently if statements is outside of
  5624 executable fragment, or any other codgenerator problem occurs.
  5625 
  5626 * New constant "undefined" with axiom "undefined x = undefined".
  5627 
  5628 * Added class "HOL.eq", allowing for code generation with polymorphic
  5629 equality.
  5630 
  5631 * Some renaming of class constants due to canonical name prefixing in
  5632 the new 'class' package:
  5633 
  5634     HOL.abs ~> HOL.abs_class.abs
  5635     HOL.divide ~> HOL.divide_class.divide
  5636     0 ~> HOL.zero_class.zero
  5637     1 ~> HOL.one_class.one
  5638     op + ~> HOL.plus_class.plus
  5639     op - ~> HOL.minus_class.minus
  5640     uminus ~> HOL.minus_class.uminus
  5641     op * ~> HOL.times_class.times
  5642     op < ~> HOL.ord_class.less
  5643     op <= > HOL.ord_class.less_eq
  5644     Nat.power ~> Power.power_class.power
  5645     Nat.size ~> Nat.size_class.size
  5646     Numeral.number_of ~> Numeral.number_class.number_of
  5647     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5648     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5649     Orderings.min ~> Orderings.ord_class.min
  5650     Orderings.max ~> Orderings.ord_class.max
  5651     Divides.op div ~> Divides.div_class.div
  5652     Divides.op mod ~> Divides.div_class.mod
  5653     Divides.op dvd ~> Divides.div_class.dvd
  5654 
  5655 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5656 
  5657 a) User-defined constants using any of the names "plus", "minus",
  5658 "times", "less" or "less_eq". The standard syntax translations for
  5659 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5660 names.
  5661 
  5662 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5663 INCOMPATIBILITY: use more specific names.
  5664 
  5665 c) Permutative equations (e.g. "a + b = b + a")
  5666 Since the change of names also changes the order of terms, permutative
  5667 rewrite rules may get applied in a different order. Experience shows
  5668 that this is rarely the case (only two adaptions in the whole Isabelle
  5669 distribution).  INCOMPATIBILITY: rewrite proofs
  5670 
  5671 d) ML code directly refering to constant names
  5672 This in general only affects hand-written proof tactics, simprocs and
  5673 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5674 Consider using @{const_name} antiquotation.
  5675 
  5676 * New class "default" with associated constant "default".
  5677 
  5678 * Function "sgn" is now overloaded and available on int, real, complex
  5679 (and other numeric types), using class "sgn".  Two possible defs of
  5680 sgn are given as equational assumptions in the classes sgn_if and
  5681 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5682 INCOMPATIBILITY.
  5683 
  5684 * Locale "partial_order" now unified with class "order" (cf. theory
  5685 Orderings), added parameter "less".  INCOMPATIBILITY.
  5686 
  5687 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5688 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5689 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5690 
  5691 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5692 changed order of proof goals in instance proofs.
  5693 
  5694 * The transitivity reasoner for partial and linear orders is set up
  5695 for classes "order" and "linorder".  Instances of the reasoner are available
  5696 in all contexts importing or interpreting the corresponding locales.
  5697 Method "order" invokes the reasoner separately; the reasoner
  5698 is also integrated with the Simplifier as a solver.  Diagnostic
  5699 command 'print_orders' shows the available instances of the reasoner
  5700 in the current context.
  5701 
  5702 * Localized monotonicity predicate in theory "Orderings"; integrated
  5703 lemmas max_of_mono and min_of_mono with this predicate.
  5704 INCOMPATIBILITY.
  5705 
  5706 * Formulation of theorem "dense" changed slightly due to integration
  5707 with new class dense_linear_order.
  5708 
  5709 * Uniform lattice theory development in HOL.
  5710 
  5711     constants "meet" and "join" now named "inf" and "sup"
  5712     constant "Meet" now named "Inf"
  5713 
  5714     classes "meet_semilorder" and "join_semilorder" now named
  5715       "lower_semilattice" and "upper_semilattice"
  5716     class "lorder" now named "lattice"
  5717     class "comp_lat" now named "complete_lattice"
  5718 
  5719     Instantiation of lattice classes allows explicit definitions
  5720     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5721 
  5722   INCOMPATIBILITY.  Theorem renames:
  5723 
  5724     meet_left_le            ~> inf_le1
  5725     meet_right_le           ~> inf_le2
  5726     join_left_le            ~> sup_ge1
  5727     join_right_le           ~> sup_ge2
  5728     meet_join_le            ~> inf_sup_ord
  5729     le_meetI                ~> le_infI
  5730     join_leI                ~> le_supI
  5731     le_meet                 ~> le_inf_iff
  5732     le_join                 ~> ge_sup_conv
  5733     meet_idempotent         ~> inf_idem
  5734     join_idempotent         ~> sup_idem
  5735     meet_comm               ~> inf_commute
  5736     join_comm               ~> sup_commute
  5737     meet_leI1               ~> le_infI1
  5738     meet_leI2               ~> le_infI2
  5739     le_joinI1               ~> le_supI1
  5740     le_joinI2               ~> le_supI2
  5741     meet_assoc              ~> inf_assoc
  5742     join_assoc              ~> sup_assoc
  5743     meet_left_comm          ~> inf_left_commute
  5744     meet_left_idempotent    ~> inf_left_idem
  5745     join_left_comm          ~> sup_left_commute
  5746     join_left_idempotent    ~> sup_left_idem
  5747     meet_aci                ~> inf_aci
  5748     join_aci                ~> sup_aci
  5749     le_def_meet             ~> le_iff_inf
  5750     le_def_join             ~> le_iff_sup
  5751     join_absorp2            ~> sup_absorb2
  5752     join_absorp1            ~> sup_absorb1
  5753     meet_absorp1            ~> inf_absorb1
  5754     meet_absorp2            ~> inf_absorb2
  5755     meet_join_absorp        ~> inf_sup_absorb
  5756     join_meet_absorp        ~> sup_inf_absorb
  5757     distrib_join_le         ~> distrib_sup_le
  5758     distrib_meet_le         ~> distrib_inf_le
  5759 
  5760     add_meet_distrib_left   ~> add_inf_distrib_left
  5761     add_join_distrib_left   ~> add_sup_distrib_left
  5762     is_join_neg_meet        ~> is_join_neg_inf
  5763     is_meet_neg_join        ~> is_meet_neg_sup
  5764     add_meet_distrib_right  ~> add_inf_distrib_right
  5765     add_join_distrib_right  ~> add_sup_distrib_right
  5766     add_meet_join_distribs  ~> add_sup_inf_distribs
  5767     join_eq_neg_meet        ~> sup_eq_neg_inf
  5768     meet_eq_neg_join        ~> inf_eq_neg_sup
  5769     add_eq_meet_join        ~> add_eq_inf_sup
  5770     meet_0_imp_0            ~> inf_0_imp_0
  5771     join_0_imp_0            ~> sup_0_imp_0
  5772     meet_0_eq_0             ~> inf_0_eq_0
  5773     join_0_eq_0             ~> sup_0_eq_0
  5774     neg_meet_eq_join        ~> neg_inf_eq_sup
  5775     neg_join_eq_meet        ~> neg_sup_eq_inf
  5776     join_eq_if              ~> sup_eq_if
  5777 
  5778     mono_meet               ~> mono_inf
  5779     mono_join               ~> mono_sup
  5780     meet_bool_eq            ~> inf_bool_eq
  5781     join_bool_eq            ~> sup_bool_eq
  5782     meet_fun_eq             ~> inf_fun_eq
  5783     join_fun_eq             ~> sup_fun_eq
  5784     meet_set_eq             ~> inf_set_eq
  5785     join_set_eq             ~> sup_set_eq
  5786     meet1_iff               ~> inf1_iff
  5787     meet2_iff               ~> inf2_iff
  5788     meet1I                  ~> inf1I
  5789     meet2I                  ~> inf2I
  5790     meet1D1                 ~> inf1D1
  5791     meet2D1                 ~> inf2D1
  5792     meet1D2                 ~> inf1D2
  5793     meet2D2                 ~> inf2D2
  5794     meet1E                  ~> inf1E
  5795     meet2E                  ~> inf2E
  5796     join1_iff               ~> sup1_iff
  5797     join2_iff               ~> sup2_iff
  5798     join1I1                 ~> sup1I1
  5799     join2I1                 ~> sup2I1
  5800     join1I1                 ~> sup1I1
  5801     join2I2                 ~> sup1I2
  5802     join1CI                 ~> sup1CI
  5803     join2CI                 ~> sup2CI
  5804     join1E                  ~> sup1E
  5805     join2E                  ~> sup2E
  5806 
  5807     is_meet_Meet            ~> is_meet_Inf
  5808     Meet_bool_def           ~> Inf_bool_def
  5809     Meet_fun_def            ~> Inf_fun_def
  5810     Meet_greatest           ~> Inf_greatest
  5811     Meet_lower              ~> Inf_lower
  5812     Meet_set_def            ~> Inf_set_def
  5813 
  5814     Sup_def                 ~> Sup_Inf
  5815     Sup_bool_eq             ~> Sup_bool_def
  5816     Sup_fun_eq              ~> Sup_fun_def
  5817     Sup_set_eq              ~> Sup_set_def
  5818 
  5819     listsp_meetI            ~> listsp_infI
  5820     listsp_meet_eq          ~> listsp_inf_eq
  5821 
  5822     meet_min                ~> inf_min
  5823     join_max                ~> sup_max
  5824 
  5825 * Added syntactic class "size"; overloaded constant "size" now has
  5826 type "'a::size ==> bool"
  5827 
  5828 * Internal reorganisation of `size' of datatypes: size theorems
  5829 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5830 simplification rules by default!); theorems "prod.size" now named
  5831 "*.size".
  5832 
  5833 * Class "div" now inherits from class "times" rather than "type".
  5834 INCOMPATIBILITY.
  5835 
  5836 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5837 Linorder etc.  have disappeared; operations defined in terms of
  5838 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5839 
  5840 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5841 
  5842 * HOL-Word: New extensive library and type for generic, fixed size
  5843 machine words, with arithmetic, bit-wise, shifting and rotating
  5844 operations, reflection into int, nat, and bool lists, automation for
  5845 linear arithmetic (by automatic reflection into nat or int), including
  5846 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5847 arithmetic type classes, supporting automatic simplification of
  5848 numerals on all operations.
  5849 
  5850 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5851 
  5852 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5853 
  5854 * Code generator library theories:
  5855   - Code_Integer represents HOL integers by big integer literals in target
  5856     languages.
  5857   - Code_Char represents HOL characters by character literals in target
  5858     languages.
  5859   - Code_Char_chr like Code_Char, but also offers treatment of character
  5860     codes; includes Code_Integer.
  5861   - Executable_Set allows to generate code for finite sets using lists.
  5862   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5863     denominator).
  5864   - Executable_Real implements a subset of real numbers, namly those
  5865     representable by rational numbers.
  5866   - Efficient_Nat implements natural numbers by integers, which in general will
  5867     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5868     includes Code_Integer.
  5869   - Code_Index provides an additional datatype index which is mapped to
  5870     target-language built-in integers.
  5871   - Code_Message provides an additional datatype message_string which is isomorphic to
  5872     strings; messages are mapped to target-language strings.
  5873 
  5874 * New package for inductive predicates
  5875 
  5876   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5877 
  5878     inductive
  5879       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5880       for z_1 :: U_1 and ... and z_n :: U_m
  5881     where
  5882       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5883     | ...
  5884 
  5885   with full support for type-inference, rather than
  5886 
  5887     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5888 
  5889     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5890     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5891 
  5892     inductive "s z_1 ... z_m"
  5893     intros
  5894       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5895       ...
  5896 
  5897   For backward compatibility, there is a wrapper allowing inductive
  5898   sets to be defined with the new package via
  5899 
  5900     inductive_set
  5901       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5902       for z_1 :: U_1 and ... and z_n :: U_m
  5903     where
  5904       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5905     | ...
  5906 
  5907   or
  5908 
  5909     inductive_set
  5910       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5911       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5912       for z_1 :: U_1 and ... and z_n :: U_m
  5913     where
  5914       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5915     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5916     | ...
  5917 
  5918   if the additional syntax "p ..." is required.
  5919 
  5920   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5921   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5922 
  5923   INCOMPATIBILITIES:
  5924 
  5925   - Since declaration and definition of inductive sets or predicates
  5926     is no longer separated, abbreviations involving the newly
  5927     introduced sets or predicates must be specified together with the
  5928     introduction rules after the 'where' keyword (see above), rather
  5929     than before the actual inductive definition.
  5930 
  5931   - The variables in induction and elimination rules are now
  5932     quantified in the order of their occurrence in the introduction
  5933     rules, rather than in alphabetical order. Since this may break
  5934     some proofs, these proofs either have to be repaired, e.g. by
  5935     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5936     statements of the form
  5937 
  5938       case (rule_i a_i_1 ... a_i_{k_i})
  5939 
  5940     or the old order of quantification has to be restored by explicitly adding
  5941     meta-level quantifiers in the introduction rules, i.e.
  5942 
  5943       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5944 
  5945   - The format of the elimination rules is now
  5946 
  5947       p z_1 ... z_m x_1 ... x_n ==>
  5948         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5949         ==> ... ==> P
  5950 
  5951     for predicates and
  5952 
  5953       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5954         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5955         ==> ... ==> P
  5956 
  5957     for sets rather than
  5958 
  5959       x : s z_1 ... z_m ==>
  5960         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5961         ==> ... ==> P
  5962 
  5963     This may require terms in goals to be expanded to n-tuples
  5964     (e.g. using case_tac or simplification with the split_paired_all
  5965     rule) before the above elimination rule is applicable.
  5966 
  5967   - The elimination or case analysis rules for (mutually) inductive
  5968     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5969     list of rules "p_1_..._p_k.elims" is no longer available.
  5970 
  5971 * New package "function"/"fun" for general recursive functions,
  5972 supporting mutual and nested recursion, definitions in local contexts,
  5973 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5974 for small examples, and the separate tutorial on the function
  5975 package. The old recdef "package" is still available as before, but
  5976 users are encouraged to use the new package.
  5977 
  5978 * Method "lexicographic_order" automatically synthesizes termination
  5979 relations as lexicographic combinations of size measures.
  5980 
  5981 * Case-expressions allow arbitrary constructor-patterns (including
  5982 "_") and take their order into account, like in functional
  5983 programming.  Internally, this is translated into nested
  5984 case-expressions; missing cases are added and mapped to the predefined
  5985 constant "undefined". In complicated cases printing may no longer show
  5986 the original input but the internal form. Lambda-abstractions allow
  5987 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5988 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5989 variable.
  5990 
  5991 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5992 is an abbreviation for "of_nat :: nat => int". The simplification
  5993 rules for "of_nat" have been changed to work like "int" did
  5994 previously.  Potential INCOMPATIBILITY:
  5995   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5996   - of_nat_diff and of_nat_mult are no longer default simp rules
  5997 
  5998 * Method "algebra" solves polynomial equations over (semi)rings using
  5999 Groebner bases. The (semi)ring structure is defined by locales and the
  6000 tool setup depends on that generic context. Installing the method for
  6001 a specific type involves instantiating the locale and possibly adding
  6002 declarations for computation on the coefficients.  The method is
  6003 already instantiated for natural numbers and for the axiomatic class
  6004 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  6005 CALCULEMUS 2007 for the general principles underlying this
  6006 architecture of context-aware proof-tools.
  6007 
  6008 * Method "ferrack" implements quantifier elimination over
  6009 special-purpose dense linear orders using locales (analogous to
  6010 "algebra"). The method is already installed for class
  6011 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  6012 rat, etc.
  6013 
  6014 * Former constant "List.op @" now named "List.append".  Use ML
  6015 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  6016 circumvent possible incompatibilities when working on ML level.
  6017 
  6018 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  6019 
  6020 * New function listsum :: 'a list => 'a for arbitrary monoids.
  6021 Special syntax: "SUM x <- xs. f x" (and latex variants)
  6022 
  6023 * New syntax for Haskell-like list comprehension (input only), eg.
  6024 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  6025 
  6026 * The special syntax for function "filter" has changed from [x :
  6027 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  6028 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  6029 
  6030 * [a..b] is now defined for arbitrary linear orders.  It used to be
  6031 defined on nat only, as an abbreviation for [a..<Suc b]
  6032 INCOMPATIBILITY.
  6033 
  6034 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  6035 
  6036 * New functions "sorted" and "sort" in src/HOL/List.thy.
  6037 
  6038 * New lemma collection field_simps (an extension of ring_simps) for
  6039 manipulating (in)equations involving division. Multiplies with all
  6040 denominators that can be proved to be non-zero (in equations) or
  6041 positive/negative (in inequations).
  6042 
  6043 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  6044 have been improved and renamed to ring_simps, group_simps and
  6045 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  6046 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  6047 
  6048 * Theory Library/Commutative_Ring: switched from recdef to function
  6049 package; constants add, mul, pow now curried.  Infix syntax for
  6050 algebraic operations.
  6051 
  6052 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  6053 INCOMPATIBILITY.
  6054 
  6055 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  6056 INCOMPATIBILITY.
  6057 
  6058 * HOL/records: generalised field-update to take a function on the
  6059 field rather than the new value: r(|A := x|) is translated to A_update
  6060 (K x) r The K-combinator that is internally used is called K_record.
  6061 INCOMPATIBILITY: Usage of the plain update functions has to be
  6062 adapted.
  6063 
  6064 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  6065 * x = 0, which are required for a semiring.  Richer structures do not
  6066 inherit from semiring_0 anymore, because this property is a theorem
  6067 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  6068 there is more to prove, but this is mostly trivial.
  6069 
  6070 * Class "recpower" is generalized to arbitrary monoids, not just
  6071 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  6072 commutativity or semiring properties additionally.
  6073 
  6074 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  6075 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  6076 better use 'abbreviation'.
  6077 
  6078 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  6079 
  6080 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  6081 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  6082 significant changes for setting up numeral syntax for types:
  6083   - New constants Numeral.pred and Numeral.succ instead
  6084       of former Numeral.bin_pred and Numeral.bin_succ.
  6085   - Use integer operations instead of bin_add, bin_mult and so on.
  6086   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  6087   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  6088 
  6089 See src/HOL/Integ/IntArith.thy for an example setup.
  6090 
  6091 * Command 'normal_form' computes the normal form of a term that may
  6092 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  6093 produces ``[b, c, a]'' (without proof).  This command is suitable for
  6094 heavy-duty computations because the functions are compiled to ML
  6095 first.  Correspondingly, a method "normalization" is provided.  See
  6096 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  6097 
  6098 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  6099 25 like -->); output depends on the "iff" print_mode, the default is
  6100 "A = B" (with priority 50).
  6101 
  6102 * Relations less (<) and less_eq (<=) are also available on type bool.
  6103 Modified syntax to disallow nesting without explicit parentheses,
  6104 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  6105 INCOMPATIBILITY.
  6106 
  6107 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  6108 
  6109 * Relation composition operator "op O" now has precedence 75 and binds
  6110 stronger than union and intersection. INCOMPATIBILITY.
  6111 
  6112 * The old set interval syntax "{m..n(}" (and relatives) has been
  6113 removed.  Use "{m..<n}" (and relatives) instead.
  6114 
  6115 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  6116 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  6117 ``declare [[simproc del: neq]]''.
  6118 
  6119 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  6120 True/False.
  6121 
  6122 * Theorem Cons_eq_map_conv no longer declared as "simp".
  6123 
  6124 * Theorem setsum_mult renamed to setsum_right_distrib.
  6125 
  6126 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  6127 ``rule'' method.
  6128 
  6129 * Reimplemented methods "sat" and "satx", with several improvements:
  6130 goals no longer need to be stated as "<prems> ==> False", equivalences
  6131 (i.e. "=" on type bool) are handled, variable names of the form
  6132 "lit_<n>" are no longer reserved, significant speedup.
  6133 
  6134 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  6135 zChaff is still supported as well.
  6136 
  6137 * 'inductive' and 'datatype': provide projections of mutual rules,
  6138 bundled as foo_bar.inducts;
  6139 
  6140 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  6141 Library.
  6142 
  6143 * Library: moved theory Accessible_Part to main HOL.
  6144 
  6145 * Library: added theory Coinductive_List of potentially infinite lists
  6146 as greatest fixed-point.
  6147 
  6148 * Library: added theory AssocList which implements (finite) maps as
  6149 association lists.
  6150 
  6151 * Method "evaluation" solves goals (i.e. a boolean expression)
  6152 efficiently by compiling it to ML.  The goal is "proved" (via an
  6153 oracle) if it evaluates to True.
  6154 
  6155 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  6156 also when invoked by the simplifier.  This results in the Simplifier
  6157 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  6158 Configuration option fast_arith_split_limit=0 recovers the old
  6159 behavior.
  6160 
  6161 * Support for hex (0x20) and binary (0b1001) numerals.
  6162 
  6163 * New method: reify eqs (t), where eqs are equations for an
  6164 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  6165 parameter, computes a term s::'b and a list xs::'a list and proves the
  6166 theorem I xs s = t. This is also known as reification or quoting. The
  6167 resulting theorem is applied to the subgoal to substitute t with I xs
  6168 s.  If t is omitted, the subgoal itself is reified.
  6169 
  6170 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  6171 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  6172 where f is supposed to be a computable function (in the sense of code
  6173 generattion). The method uses reify to compute s and xs as above then
  6174 applies corr_thm and uses normalization by evaluation to "prove" f s =
  6175 r and finally gets the theorem t = r, which is again applied to the
  6176 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  6177 
  6178 * Reflection: Automatic reification now handels binding, an example is
  6179 available in src/HOL/ex/ReflectionEx.thy
  6180 
  6181 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  6182 command 'statespace' that is similar to 'record', but introduces an
  6183 abstract specification based on the locale infrastructure instead of
  6184 HOL types.  This leads to extra flexibility in composing state spaces,
  6185 in particular multiple inheritance and renaming of components.
  6186 
  6187 
  6188 *** HOL-Complex ***
  6189 
  6190 * Hyperreal: Functions root and sqrt are now defined on negative real
  6191 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  6192 Nonnegativity side conditions have been removed from many lemmas, so
  6193 that more subgoals may now be solved by simplification; potential
  6194 INCOMPATIBILITY.
  6195 
  6196 * Real: new type classes formalize real normed vector spaces and
  6197 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  6198 and norm :: 'a => real.
  6199 
  6200 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  6201 reals into other types. The overloaded constant Reals :: 'a set is now
  6202 defined as range of_real; potential INCOMPATIBILITY.
  6203 
  6204 * Real: proper support for ML code generation, including 'quickcheck'.
  6205 Reals are implemented as arbitrary precision rationals.
  6206 
  6207 * Hyperreal: Several constants that previously worked only for the
  6208 reals have been generalized, so they now work over arbitrary vector
  6209 spaces. Type annotations may need to be added in some cases; potential
  6210 INCOMPATIBILITY.
  6211 
  6212   Infinitesimal  :: ('a::real_normed_vector) star set
  6213   HFinite        :: ('a::real_normed_vector) star set
  6214   HInfinite      :: ('a::real_normed_vector) star set
  6215   approx         :: ('a::real_normed_vector) star => 'a star => bool
  6216   monad          :: ('a::real_normed_vector) star => 'a star set
  6217   galaxy         :: ('a::real_normed_vector) star => 'a star set
  6218   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  6219   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  6220   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  6221   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  6222   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  6223   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  6224   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6225   sgn            :: 'a::real_normed_vector => 'a
  6226   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6227 
  6228 * Complex: Some complex-specific constants are now abbreviations for
  6229 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6230 hnorm.  Other constants have been entirely removed in favor of the
  6231 polymorphic versions (INCOMPATIBILITY):
  6232 
  6233   approx        <-- capprox
  6234   HFinite       <-- CFinite
  6235   HInfinite     <-- CInfinite
  6236   Infinitesimal <-- CInfinitesimal