NEWS
author webertj
Fri Oct 19 15:12:52 2012 +0200 (2012-10-19)
changeset 49962 a8cc904a6820
parent 49918 cf441f4a358b
child 49963 326f87427719
permissions -rw-r--r--
Renamed {left,right}_distrib to distrib_{right,left}.
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Prover IDE (PIDE) improvements:
    10     . parallel terminal proofs ('by');
    11     . improved output panel with tooltips, hyperlinks etc.;
    12     . improved tooltips with nested tooltips, hyperlinks etc.;
    13     . more efficient painting, improved reactivity;
    14     . more robust incremental parsing of outer syntax (partial
    15       comments, malformed symbols);
    16     . smarter handling of tracing messages (via tracing_limit);
    17     . more plugin options and preferences, based on Isabelle/Scala;
    18     . uniform Java 7 platform on Linux, Mac OS X, Windows;
    19 
    20 * Configuration option show_markup controls direct inlining of markup
    21 into the printed representation of formal entities --- notably type
    22 and sort constraints.  This enables Prover IDE users to retrieve that
    23 information via tooltips in the output window, for example.
    24 
    25 * Command 'ML_file' evaluates ML text from a file directly within the
    26 theory, without any predeclaration via 'uses' in the theory header.
    27 
    28 * Old command 'use' command and corresponding keyword 'uses' in the
    29 theory header are legacy features and will be discontinued soon.
    30 Tools that load their additional source files may imitate the
    31 'ML_file' implementation, such that the system can take care of
    32 dependencies properly.
    33 
    34 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    35 is called fastforce / fast_force_tac already since Isabelle2011-1.
    36 
    37 * Updated and extended "isar-ref" manual, reduced remaining material
    38 in old "ref" manual.
    39 
    40 * Improved support for auxiliary contexts indicate block structure for
    41 specifications: nesting of "context fixes ... context assumes ..."
    42 and "class ... context ...".
    43 
    44 * More informative error messages for Isar proof commands involving
    45 lazy enumerations (method applications etc.).
    46 
    47 
    48 *** Pure ***
    49 
    50 * Code generation for Haskell: restrict unqualified imports from
    51 Haskell Prelude to a small set of fundamental operations.
    52 
    53 * Command "export_code": relative file names are interpreted
    54 relatively to master directory of current theory rather than
    55 the rather arbitrary current working directory.
    56 INCOMPATIBILITY.
    57 
    58 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    59 use regular rule composition via "OF" / "THEN", or explicit proof
    60 structure instead.  Note that Isabelle/ML provides a variety of
    61 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    62 with some care where this is really required.
    63 
    64 * Command 'typ' supports an additional variant with explicit sort
    65 constraint, to infer and check the most general type conforming to a
    66 given given sort.  Example (in HOL):
    67 
    68   typ "_ * _ * bool * unit" :: finite
    69 
    70 
    71 *** HOL ***
    72 
    73 * Simplified 'typedef' specifications: historical options for implicit
    74 set definition and alternative name have been discontinued.  The
    75 former behavior of "typedef (open) t = A" is now the default, but
    76 written just "typedef t = A".  INCOMPATIBILITY, need to adapt theories
    77 accordingly.
    78 
    79 * Theory "Library/Multiset":
    80 
    81   - Renamed constants
    82       fold_mset ~> Multiset.fold  -- for coherence with other fold combinators
    83 
    84   - Renamed facts
    85       fold_mset_commute ~> fold_mset_comm  -- for coherence with fold_comm
    86 
    87 INCOMPATIBILITY.
    88 
    89 * Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
    90 
    91 * Class "comm_monoid_diff" formalises properties of bounded
    92 subtraction, with natural numbers and multisets as typical instances.
    93 
    94 * Theory "Library/Option_ord" provides instantiation of option type
    95 to lattice type classes.
    96 
    97 * New combinator "Option.these" with type "'a option set => 'a set".
    98 
    99 * Renamed theory Library/List_Prefix to Library/Sublist.
   100 INCOMPATIBILITY.  Related changes are:
   101 
   102   - Renamed constants:
   103 
   104       prefix ~> prefixeq
   105       strict_prefix ~> prefix
   106 
   107     Renamed lemmas accordingly, INCOMPATIBILITY.
   108 
   109   - Replaced constant "postfix" by "suffixeq" with swapped argument order
   110     (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix
   111     syntax "xs >>= ys"; use "suffixeq ys xs" instead.  Renamed lemmas
   112     accordingly.  INCOMPATIBILITY.
   113 
   114   - New constant "emb" for homeomorphic embedding on lists. New
   115     abbreviation "sub" for special case "emb (op =)".
   116 
   117   - Library/Sublist does no longer provide "order" and "bot" type class
   118     instances for the prefix order (merely corresponding locale
   119     interpretations). The type class instances are to be found in
   120     Library/Prefix_Order. INCOMPATIBILITY.
   121 
   122   - The sublist relation from Library/Sublist_Order is now based on
   123     "Sublist.sub". Replaced lemmas:
   124 
   125       le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff
   126       le_list_append_mono ~> Sublist.emb_append_mono
   127       le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2
   128       le_list_Cons_EX ~> Sublist.emb_ConsD
   129       le_list_drop_Cons2 ~> Sublist.sub_Cons2'
   130       le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq
   131       le_list_drop_Cons ~> Sublist.sub_Cons'
   132       le_list_drop_many ~> Sublist.sub_drop_many
   133       le_list_filter_left ~> Sublist.sub_filter_left
   134       le_list_rev_drop_many ~> Sublist.sub_rev_drop_many
   135       le_list_rev_take_iff ~> Sublist.sub_append
   136       le_list_same_length ~> Sublist.sub_same_length
   137       le_list_take_many_iff ~> Sublist.sub_append'
   138       less_eq_list.drop ~> less_eq_list_drop
   139       less_eq_list.induct ~> less_eq_list_induct
   140       not_le_list_length ~> Sublist.not_sub_length
   141 
   142     INCOMPATIBILITY.
   143 
   144 * HOL/Rings: renamed lemmas
   145 
   146 left_distrib ~> distrib_right
   147 right_distrib ~> distrib_left
   148 
   149 in class semiring.  INCOMPATIBILITY.
   150 
   151 * HOL/BNF: New (co)datatype package based on bounded natural
   152 functors with support for mixed, nested recursion and interesting
   153 non-free datatypes.
   154 
   155 * HOL/Cardinals: Theories of ordinals and cardinals
   156 (supersedes the AFP entry "Ordinals_and_Cardinals").
   157 
   158 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
   159 execution for code generated towards Isabelle/ML.
   160 
   161 * Library/FinFun.thy: theory of almost everywhere constant functions
   162 (supersedes the AFP entry "Code Generation for Functions as Data").
   163 
   164 * Library/Phantom.thy: generic phantom type to make a type parameter
   165 appear in a constant's type. This alternative to adding TYPE('a) as
   166 another parameter avoids unnecessary closures in generated code.
   167 
   168 * Library/RBT_Impl.thy: efficient construction of red-black trees 
   169 from sorted associative lists. Merging two trees with rbt_union may
   170 return a structurally different tree than before. MINOR INCOMPATIBILITY.
   171 
   172 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   173 expressions.
   174 
   175 * Quickcheck:
   176 
   177   - added an optimisation for equality premises.
   178     It is switched on by default, and can be switched off by setting
   179     the configuration quickcheck_optimise_equality to false.
   180 
   181 * The SMT solver Z3 has now by default a restricted set of directly
   182 supported features. For the full set of features (div/mod, nonlinear
   183 arithmetic, datatypes/records) with potential proof reconstruction
   184 failures, enable the configuration option "z3_with_extensions".
   185 Minor INCOMPATIBILITY.
   186 
   187 * Sledgehammer:
   188 
   189   - Added MaSh relevance filter based on machine-learning; see the
   190     Sledgehammer manual for details.
   191   - Rationalized type encodings ("type_enc" option).
   192   - Renamed "kill_provers" subcommand to "kill"
   193   - Renamed options:
   194       isar_proof ~> isar_proofs
   195       isar_shrink_factor ~> isar_shrinkage
   196       max_relevant ~> max_facts
   197       relevance_thresholds ~> fact_thresholds
   198 
   199 
   200 *** Document preparation ***
   201 
   202 * Default for \<euro> is now based on eurosym package, instead of
   203 slightly exotic babel/greek.
   204 
   205 * Document variant NAME may use different LaTeX entry point
   206 document/root_NAME.tex if that file exists, instead of the common
   207 document/root.tex.
   208 
   209 * Simplified custom document/build script, instead of old-style
   210 document/IsaMakefile.  Minor INCOMPATIBILITY.
   211 
   212 
   213 *** ML ***
   214 
   215 * Type Seq.results and related operations support embedded error
   216 messages within lazy enumerations, and thus allow to provide
   217 informative errors in the absence of any usable results.
   218 
   219 * Renamed Position.str_of to Position.here to emphasize that this is a
   220 formal device to inline positions into message text, but not
   221 necessarily printing visible text.
   222 
   223 
   224 *** System ***
   225 
   226 * The ML system is configured as regular component, and no longer
   227 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   228 for home-made configurations.
   229 
   230 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   231 Minor INCOMPATIBILITY in command-line options.
   232 
   233 * Advanced support for Isabelle sessions and build management, see
   234 "system" manual for the chapter of that name, especially the "isabelle
   235 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   236 mkdir / make are rendered obsolete.
   237 
   238 * Discontinued obsolete "isabelle makeall".
   239 
   240 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   241 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   242 provides some traditional targets that invoke "isabelle build".  Note
   243 that this is inefficient!  Applications of Isabelle/HOL involving
   244 "isabelle make" should be upgraded to use "isabelle build" directly.
   245 
   246 * Discontinued obsolete Isabelle/build script, it is superseded by the
   247 regular isabelle build tool.  For example:
   248 
   249   isabelle build -s -b HOLCF
   250 
   251 * The "isabelle options" tool prints Isabelle system options, as
   252 required for "isabelle build", for example.
   253 
   254 * The "isabelle mkroot" tool prepares session root directories for use
   255 with "isabelle build", similar to former "isabelle mkdir" for
   256 "isabelle usedir".
   257 
   258 * The "isabelle components" tool helps to resolve add-on components
   259 that are not bundled, or referenced from a bare-bones repository
   260 version of Isabelle.
   261 
   262 * Discontinued support for Poly/ML 5.2.1, which was the last version
   263 without exception positions and advanced ML compiler/toplevel
   264 configuration.
   265 
   266 * Discontinued special treatment of Proof General -- no longer guess
   267 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   268 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   269 settings manually, or use a Proof General version that has been
   270 bundled as Isabelle component.
   271 
   272 
   273 New in Isabelle2012 (May 2012)
   274 ------------------------------
   275 
   276 *** General ***
   277 
   278 * Prover IDE (PIDE) improvements:
   279 
   280   - more robust Sledgehammer integration (as before the sledgehammer
   281     command-line needs to be typed into the source buffer)
   282   - markup for bound variables
   283   - markup for types of term variables (displayed as tooltips)
   284   - support for user-defined Isar commands within the running session
   285   - improved support for Unicode outside original 16bit range
   286     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   287 
   288 * Forward declaration of outer syntax keywords within the theory
   289 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   290 commands to be used in the same theory where defined.
   291 
   292 * Auxiliary contexts indicate block structure for specifications with
   293 additional parameters and assumptions.  Such unnamed contexts may be
   294 nested within other targets, like 'theory', 'locale', 'class',
   295 'instantiation' etc.  Results from the local context are generalized
   296 accordingly and applied to the enclosing target context.  Example:
   297 
   298   context
   299     fixes x y z :: 'a
   300     assumes xy: "x = y" and yz: "y = z"
   301   begin
   302 
   303   lemma my_trans: "x = z" using xy yz by simp
   304 
   305   end
   306 
   307   thm my_trans
   308 
   309 The most basic application is to factor-out context elements of
   310 several fixes/assumes/shows theorem statements, e.g. see
   311 ~~/src/HOL/Isar_Examples/Group_Context.thy
   312 
   313 Any other local theory specification element works within the "context
   314 ... begin ... end" block as well.
   315 
   316 * Bundled declarations associate attributed fact expressions with a
   317 given name in the context.  These may be later included in other
   318 contexts.  This allows to manage context extensions casually, without
   319 the logical dependencies of locales and locale interpretation.  See
   320 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   321 
   322 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   323 declaration, and results are standardized before being stored.  Thus
   324 old-style "standard" after instantiation or composition of facts
   325 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   326 indices of schematic variables.
   327 
   328 * Rule attributes in local theory declarations (e.g. locale or class)
   329 are now statically evaluated: the resulting theorem is stored instead
   330 of the original expression.  INCOMPATIBILITY in rare situations, where
   331 the historic accident of dynamic re-evaluation in interpretations
   332 etc. was exploited.
   333 
   334 * New tutorial "Programming and Proving in Isabelle/HOL"
   335 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   336 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   337 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   338 Logic" as the recommended beginners tutorial, but does not cover all
   339 of the material of that old tutorial.
   340 
   341 * Updated and extended reference manuals: "isar-ref",
   342 "implementation", "system"; reduced remaining material in old "ref"
   343 manual.
   344 
   345 
   346 *** Pure ***
   347 
   348 * Command 'definition' no longer exports the foundational "raw_def"
   349 into the user context.  Minor INCOMPATIBILITY, may use the regular
   350 "def" result with attribute "abs_def" to imitate the old version.
   351 
   352 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   353 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   354 expand it.  This also works for object-logic equality.  (Formerly
   355 undocumented feature.)
   356 
   357 * Sort constraints are now propagated in simultaneous statements, just
   358 like type constraints.  INCOMPATIBILITY in rare situations, where
   359 distinct sorts used to be assigned accidentally.  For example:
   360 
   361   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   362 
   363   lemma "P (x::'a)" and "Q (y::'a::bar)"
   364     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   365 
   366 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   367 tolerant against multiple unifiers, as long as the final result is
   368 unique.  (As before, rules are composed in canonical right-to-left
   369 order to accommodate newly introduced premises.)
   370 
   371 * Renamed some inner syntax categories:
   372 
   373     num ~> num_token
   374     xnum ~> xnum_token
   375     xstr ~> str_token
   376 
   377 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   378 "num_position" etc. are mainly used instead (which also include
   379 position information via constraints).
   380 
   381 * Simplified configuration options for syntax ambiguity: see
   382 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   383 manual.  Minor INCOMPATIBILITY.
   384 
   385 * Discontinued configuration option "syntax_positions": atomic terms
   386 in parse trees are always annotated by position constraints.
   387 
   388 * Old code generator for SML and its commands 'code_module',
   389 'code_library', 'consts_code', 'types_code' have been discontinued.
   390 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   391 
   392 * Redundant attribute "code_inline" has been discontinued. Use
   393 "code_unfold" instead.  INCOMPATIBILITY.
   394 
   395 * Dropped attribute "code_unfold_post" in favor of the its dual
   396 "code_abbrev", which yields a common pattern in definitions like
   397 
   398   definition [code_abbrev]: "f = t"
   399 
   400 INCOMPATIBILITY.
   401 
   402 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   403 instead.  INCOMPATIBILITY.
   404 
   405 * Discontinued old "prems" fact, which used to refer to the accidental
   406 collection of foundational premises in the context (already marked as
   407 legacy since Isabelle2011).
   408 
   409 
   410 *** HOL ***
   411 
   412 * Type 'a set is now a proper type constructor (just as before
   413 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   414 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   415 sets separate, it is often sufficient to rephrase some set S that has
   416 been accidentally used as predicates by "%x. x : S", and some
   417 predicate P that has been accidentally used as set by "{x. P x}".
   418 Corresponding proofs in a first step should be pruned from any
   419 tinkering with former theorems mem_def and Collect_def as far as
   420 possible.
   421 
   422 For developments which deliberately mix predicates and sets, a
   423 planning step is necessary to determine what should become a predicate
   424 and what a set.  It can be helpful to carry out that step in
   425 Isabelle2011-1 before jumping right into the current release.
   426 
   427 * Code generation by default implements sets as container type rather
   428 than predicates.  INCOMPATIBILITY.
   429 
   430 * New type synonym 'a rel = ('a * 'a) set
   431 
   432 * The representation of numerals has changed.  Datatype "num"
   433 represents strictly positive binary numerals, along with functions
   434 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   435 positive and negated numeric literals, respectively.  See also
   436 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   437 user theories may require adaptations as follows:
   438 
   439   - Theorems with number_ring or number_semiring constraints: These
   440     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   441 
   442   - Theories defining numeric types: Remove number, number_semiring,
   443     and number_ring instances. Defer all theorems about numerals until
   444     after classes one and semigroup_add have been instantiated.
   445 
   446   - Numeral-only simp rules: Replace each rule having a "number_of v"
   447     pattern with two copies, one for numeral and one for neg_numeral.
   448 
   449   - Theorems about subclasses of semiring_1 or ring_1: These classes
   450     automatically support numerals now, so more simp rules and
   451     simprocs may now apply within the proof.
   452 
   453   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   454     Redefine using other integer operations.
   455 
   456 * Transfer: New package intended to generalize the existing
   457 "descending" method and related theorem attributes from the Quotient
   458 package.  (Not all functionality is implemented yet, but future
   459 development will focus on Transfer as an eventual replacement for the
   460 corresponding parts of the Quotient package.)
   461 
   462   - transfer_rule attribute: Maintains a collection of transfer rules,
   463     which relate constants at two different types. Transfer rules may
   464     relate different type instances of the same polymorphic constant,
   465     or they may relate an operation on a raw type to a corresponding
   466     operation on an abstract type (quotient or subtype). For example:
   467 
   468     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   469     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   470 
   471   - transfer method: Replaces a subgoal on abstract types with an
   472     equivalent subgoal on the corresponding raw types. Constants are
   473     replaced with corresponding ones according to the transfer rules.
   474     Goals are generalized over all free variables by default; this is
   475     necessary for variables whose types change, but can be overridden
   476     for specific variables with e.g. "transfer fixing: x y z".  The
   477     variant transfer' method allows replacing a subgoal with one that
   478     is logically stronger (rather than equivalent).
   479 
   480   - relator_eq attribute: Collects identity laws for relators of
   481     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   482     transfer method uses these lemmas to infer transfer rules for
   483     non-polymorphic constants on the fly.
   484 
   485   - transfer_prover method: Assists with proving a transfer rule for a
   486     new constant, provided the constant is defined in terms of other
   487     constants that already have transfer rules. It should be applied
   488     after unfolding the constant definitions.
   489 
   490   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   491     from type nat to type int.
   492 
   493 * Lifting: New package intended to generalize the quotient_definition
   494 facility of the Quotient package; designed to work with Transfer.
   495 
   496   - lift_definition command: Defines operations on an abstract type in
   497     terms of a corresponding operation on a representation
   498     type.  Example syntax:
   499 
   500     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   501       is List.insert
   502 
   503     Users must discharge a respectfulness proof obligation when each
   504     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   505     the proof is discharged automatically.) The obligation is
   506     presented in a user-friendly, readable form; a respectfulness
   507     theorem in the standard format and a transfer rule are generated
   508     by the package.
   509 
   510   - Integration with code_abstype: For typedefs (e.g. subtypes
   511     corresponding to a datatype invariant, such as dlist),
   512     lift_definition generates a code certificate theorem and sets up
   513     code generation for each constant.
   514 
   515   - setup_lifting command: Sets up the Lifting package to work with a
   516     user-defined type. The user must provide either a quotient theorem
   517     or a type_definition theorem.  The package configures transfer
   518     rules for equality and quantifiers on the type, and sets up the
   519     lift_definition command to work with the type.
   520 
   521   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   522     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   523     Word/Word.thy and Library/Float.thy.
   524 
   525 * Quotient package:
   526 
   527   - The 'quotient_type' command now supports a 'morphisms' option with
   528     rep and abs functions, similar to typedef.
   529 
   530   - 'quotient_type' sets up new types to work with the Lifting and
   531     Transfer packages, as with 'setup_lifting'.
   532 
   533   - The 'quotient_definition' command now requires the user to prove a
   534     respectfulness property at the point where the constant is
   535     defined, similar to lift_definition; INCOMPATIBILITY.
   536 
   537   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   538     accordingly, INCOMPATIBILITY.
   539 
   540 * New diagnostic command 'find_unused_assms' to find potentially
   541 superfluous assumptions in theorems using Quickcheck.
   542 
   543 * Quickcheck:
   544 
   545   - Quickcheck returns variable assignments as counterexamples, which
   546     allows to reveal the underspecification of functions under test.
   547     For example, refuting "hd xs = x", it presents the variable
   548     assignment xs = [] and x = a1 as a counterexample, assuming that
   549     any property is false whenever "hd []" occurs in it.
   550 
   551     These counterexample are marked as potentially spurious, as
   552     Quickcheck also returns "xs = []" as a counterexample to the
   553     obvious theorem "hd xs = hd xs".
   554 
   555     After finding a potentially spurious counterexample, Quickcheck
   556     continues searching for genuine ones.
   557 
   558     By default, Quickcheck shows potentially spurious and genuine
   559     counterexamples. The option "genuine_only" sets quickcheck to only
   560     show genuine counterexamples.
   561 
   562   - The command 'quickcheck_generator' creates random and exhaustive
   563     value generators for a given type and operations.
   564 
   565     It generates values by using the operations as if they were
   566     constructors of that type.
   567 
   568   - Support for multisets.
   569 
   570   - Added "use_subtype" options.
   571 
   572   - Added "quickcheck_locale" configuration to specify how to process
   573     conjectures in a locale context.
   574 
   575 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   576 and affecting 'rat' and 'real'.
   577 
   578 * Sledgehammer:
   579   - Integrated more tightly with SPASS, as described in the ITP 2012
   580     paper "More SPASS with Isabelle".
   581   - Made it try "smt" as a fallback if "metis" fails or times out.
   582   - Added support for the following provers: Alt-Ergo (via Why3 and
   583     TFF1), iProver, iProver-Eq.
   584   - Sped up the minimizer.
   585   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   586   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   587   - Renamed "sound" option to "strict".
   588 
   589 * Metis: Added possibility to specify lambda translations scheme as a
   590 parenthesized argument (e.g., "by (metis (lifting) ...)").
   591 
   592 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   593 
   594 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   595 
   596 * New "case_product" attribute to generate a case rule doing multiple
   597 case distinctions at the same time.  E.g.
   598 
   599   list.exhaust [case_product nat.exhaust]
   600 
   601 produces a rule which can be used to perform case distinction on both
   602 a list and a nat.
   603 
   604 * New "eventually_elim" method as a generalized variant of the
   605 eventually_elim* rules.  Supports structured proofs.
   606 
   607 * Typedef with implicit set definition is considered legacy.  Use
   608 "typedef (open)" form instead, which will eventually become the
   609 default.
   610 
   611 * Record: code generation can be switched off manually with
   612 
   613   declare [[record_coden = false]]  -- "default true"
   614 
   615 * Datatype: type parameters allow explicit sort constraints.
   616 
   617 * Concrete syntax for case expressions includes constraints for source
   618 positions, and thus produces Prover IDE markup for its bindings.
   619 INCOMPATIBILITY for old-style syntax translations that augment the
   620 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   621 one_case.
   622 
   623 * Clarified attribute "mono_set": pure declaration without modifying
   624 the result of the fact expression.
   625 
   626 * More default pred/set conversions on a couple of relation operations
   627 and predicates.  Added powers of predicate relations.  Consolidation
   628 of some relation theorems:
   629 
   630   converse_def ~> converse_unfold
   631   rel_comp_def ~> relcomp_unfold
   632   symp_def ~> (modified, use symp_def and sym_def instead)
   633   transp_def ~> transp_trans
   634   Domain_def ~> Domain_unfold
   635   Range_def ~> Domain_converse [symmetric]
   636 
   637 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   638 
   639 See theory "Relation" for examples for making use of pred/set
   640 conversions by means of attributes "to_set" and "to_pred".
   641 
   642 INCOMPATIBILITY.
   643 
   644 * Renamed facts about the power operation on relations, i.e., relpow
   645 to match the constant's name:
   646 
   647   rel_pow_1 ~> relpow_1
   648   rel_pow_0_I ~> relpow_0_I
   649   rel_pow_Suc_I ~> relpow_Suc_I
   650   rel_pow_Suc_I2 ~> relpow_Suc_I2
   651   rel_pow_0_E ~> relpow_0_E
   652   rel_pow_Suc_E ~> relpow_Suc_E
   653   rel_pow_E ~> relpow_E
   654   rel_pow_Suc_D2 ~> relpow_Suc_D2
   655   rel_pow_Suc_E2 ~> relpow_Suc_E2
   656   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   657   rel_pow_E2 ~> relpow_E2
   658   rel_pow_add ~> relpow_add
   659   rel_pow_commute ~> relpow
   660   rel_pow_empty ~> relpow_empty:
   661   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   662   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   663   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   664   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   665   rel_pow_fun_conv ~> relpow_fun_conv
   666   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   667   rel_pow_finite_bounded ~> relpow_finite_bounded
   668   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   669   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   670   single_valued_rel_pow ~> single_valued_relpow
   671 
   672 INCOMPATIBILITY.
   673 
   674 * Theory Relation: Consolidated constant name for relation composition
   675 and corresponding theorem names:
   676 
   677   - Renamed constant rel_comp to relcomp.
   678 
   679   - Dropped abbreviation pred_comp. Use relcompp instead.
   680 
   681   - Renamed theorems:
   682 
   683     rel_compI ~> relcompI
   684     rel_compEpair ~> relcompEpair
   685     rel_compE ~> relcompE
   686     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   687     rel_comp_empty1 ~> relcomp_empty1
   688     rel_comp_mono ~> relcomp_mono
   689     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   690     rel_comp_distrib ~> relcomp_distrib
   691     rel_comp_distrib2 ~> relcomp_distrib2
   692     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   693     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   694     single_valued_rel_comp ~> single_valued_relcomp
   695     rel_comp_def ~> relcomp_unfold
   696     converse_rel_comp ~> converse_relcomp
   697     pred_compI ~> relcomppI
   698     pred_compE ~> relcomppE
   699     pred_comp_bot1 ~> relcompp_bot1
   700     pred_comp_bot2 ~> relcompp_bot2
   701     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   702     pred_comp_mono ~> relcompp_mono
   703     pred_comp_distrib ~> relcompp_distrib
   704     pred_comp_distrib2 ~> relcompp_distrib2
   705     converse_pred_comp ~> converse_relcompp
   706 
   707     finite_rel_comp ~> finite_relcomp
   708 
   709     set_rel_comp ~> set_relcomp
   710 
   711 INCOMPATIBILITY.
   712 
   713 * Theory Divides: Discontinued redundant theorems about div and mod.
   714 INCOMPATIBILITY, use the corresponding generic theorems instead.
   715 
   716   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   717   zdiv_self ~> div_self
   718   zmod_self ~> mod_self
   719   zdiv_zero ~> div_0
   720   zmod_zero ~> mod_0
   721   zdiv_zmod_equality ~> div_mod_equality2
   722   zdiv_zmod_equality2 ~> div_mod_equality
   723   zmod_zdiv_trivial ~> mod_div_trivial
   724   zdiv_zminus_zminus ~> div_minus_minus
   725   zmod_zminus_zminus ~> mod_minus_minus
   726   zdiv_zminus2 ~> div_minus_right
   727   zmod_zminus2 ~> mod_minus_right
   728   zdiv_minus1_right ~> div_minus1_right
   729   zmod_minus1_right ~> mod_minus1_right
   730   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   731   zmod_zmult1_eq ~> mod_mult_right_eq
   732   zpower_zmod ~> power_mod
   733   zdvd_zmod ~> dvd_mod
   734   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   735   mod_mult_distrib ~> mult_mod_left
   736   mod_mult_distrib2 ~> mult_mod_right
   737 
   738 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   739 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   740 
   741 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   742 
   743 * Consolidated theorem names concerning fold combinators:
   744 
   745   inf_INFI_fold_inf ~> inf_INF_fold_inf
   746   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   747   INFI_fold_inf ~> INF_fold_inf
   748   SUPR_fold_sup ~> SUP_fold_sup
   749   union_set ~> union_set_fold
   750   minus_set ~> minus_set_fold
   751   INFI_set_fold ~> INF_set_fold
   752   SUPR_set_fold ~> SUP_set_fold
   753   INF_code ~> INF_set_foldr
   754   SUP_code ~> SUP_set_foldr
   755   foldr.simps ~> foldr.simps (in point-free formulation)
   756   foldr_fold_rev ~> foldr_conv_fold
   757   foldl_fold ~> foldl_conv_fold
   758   foldr_foldr ~> foldr_conv_foldl
   759   foldl_foldr ~> foldl_conv_foldr
   760   fold_set_remdups ~> fold_set_fold_remdups
   761   fold_set ~> fold_set_fold
   762   fold1_set ~> fold1_set_fold
   763 
   764 INCOMPATIBILITY.
   765 
   766 * Dropped rarely useful theorems concerning fold combinators:
   767 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   768 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   769 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   770 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   771 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   772 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   773 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   774 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   775 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   776 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   777 
   778 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   779 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   780 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   781 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   782 lemmas over fold rather than foldr, or make use of lemmas
   783 fold_conv_foldr and fold_rev.
   784 
   785 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   786 through option types.
   787 
   788 * "Transitive_Closure.ntrancl": bounded transitive closure on
   789 relations.
   790 
   791 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   792 
   793 * Theory Int: Discontinued many legacy theorems specific to type int.
   794 INCOMPATIBILITY, use the corresponding generic theorems instead.
   795 
   796   zminus_zminus ~> minus_minus
   797   zminus_0 ~> minus_zero
   798   zminus_zadd_distrib ~> minus_add_distrib
   799   zadd_commute ~> add_commute
   800   zadd_assoc ~> add_assoc
   801   zadd_left_commute ~> add_left_commute
   802   zadd_ac ~> add_ac
   803   zmult_ac ~> mult_ac
   804   zadd_0 ~> add_0_left
   805   zadd_0_right ~> add_0_right
   806   zadd_zminus_inverse2 ~> left_minus
   807   zmult_zminus ~> mult_minus_left
   808   zmult_commute ~> mult_commute
   809   zmult_assoc ~> mult_assoc
   810   zadd_zmult_distrib ~> left_distrib
   811   zadd_zmult_distrib2 ~> right_distrib
   812   zdiff_zmult_distrib ~> left_diff_distrib
   813   zdiff_zmult_distrib2 ~> right_diff_distrib
   814   zmult_1 ~> mult_1_left
   815   zmult_1_right ~> mult_1_right
   816   zle_refl ~> order_refl
   817   zle_trans ~> order_trans
   818   zle_antisym ~> order_antisym
   819   zle_linear ~> linorder_linear
   820   zless_linear ~> linorder_less_linear
   821   zadd_left_mono ~> add_left_mono
   822   zadd_strict_right_mono ~> add_strict_right_mono
   823   zadd_zless_mono ~> add_less_le_mono
   824   int_0_less_1 ~> zero_less_one
   825   int_0_neq_1 ~> zero_neq_one
   826   zless_le ~> less_le
   827   zpower_zadd_distrib ~> power_add
   828   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   829   zero_le_zpower_abs ~> zero_le_power_abs
   830 
   831 * Theory Deriv: Renamed
   832 
   833   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   834 
   835 * Theory Library/Multiset: Improved code generation of multisets.
   836 
   837 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   838 are expressed via type classes again. The special syntax
   839 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   840 setsum_set, which is now subsumed by Big_Operators.setsum.
   841 INCOMPATIBILITY.
   842 
   843 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   844 use theory HOL/Library/Nat_Bijection instead.
   845 
   846 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   847 trees is now inside a type class context.  Names of affected
   848 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   849 theories working directly with raw red-black trees, adapt the names as
   850 follows:
   851 
   852   Operations:
   853   bulkload -> rbt_bulkload
   854   del_from_left -> rbt_del_from_left
   855   del_from_right -> rbt_del_from_right
   856   del -> rbt_del
   857   delete -> rbt_delete
   858   ins -> rbt_ins
   859   insert -> rbt_insert
   860   insertw -> rbt_insert_with
   861   insert_with_key -> rbt_insert_with_key
   862   map_entry -> rbt_map_entry
   863   lookup -> rbt_lookup
   864   sorted -> rbt_sorted
   865   tree_greater -> rbt_greater
   866   tree_less -> rbt_less
   867   tree_less_symbol -> rbt_less_symbol
   868   union -> rbt_union
   869   union_with -> rbt_union_with
   870   union_with_key -> rbt_union_with_key
   871 
   872   Lemmas:
   873   balance_left_sorted -> balance_left_rbt_sorted
   874   balance_left_tree_greater -> balance_left_rbt_greater
   875   balance_left_tree_less -> balance_left_rbt_less
   876   balance_right_sorted -> balance_right_rbt_sorted
   877   balance_right_tree_greater -> balance_right_rbt_greater
   878   balance_right_tree_less -> balance_right_rbt_less
   879   balance_sorted -> balance_rbt_sorted
   880   balance_tree_greater -> balance_rbt_greater
   881   balance_tree_less -> balance_rbt_less
   882   bulkload_is_rbt -> rbt_bulkload_is_rbt
   883   combine_sorted -> combine_rbt_sorted
   884   combine_tree_greater -> combine_rbt_greater
   885   combine_tree_less -> combine_rbt_less
   886   delete_in_tree -> rbt_delete_in_tree
   887   delete_is_rbt -> rbt_delete_is_rbt
   888   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   889   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   890   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   891   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   892   del_in_tree -> rbt_del_in_tree
   893   del_inv1_inv2 -> rbt_del_inv1_inv2
   894   del_sorted -> rbt_del_rbt_sorted
   895   del_tree_greater -> rbt_del_rbt_greater
   896   del_tree_less -> rbt_del_rbt_less
   897   dom_lookup_Branch -> dom_rbt_lookup_Branch
   898   entries_lookup -> entries_rbt_lookup
   899   finite_dom_lookup -> finite_dom_rbt_lookup
   900   insert_sorted -> rbt_insert_rbt_sorted
   901   insertw_is_rbt -> rbt_insertw_is_rbt
   902   insertwk_is_rbt -> rbt_insertwk_is_rbt
   903   insertwk_sorted -> rbt_insertwk_rbt_sorted
   904   insertw_sorted -> rbt_insertw_rbt_sorted
   905   ins_sorted -> ins_rbt_sorted
   906   ins_tree_greater -> ins_rbt_greater
   907   ins_tree_less -> ins_rbt_less
   908   is_rbt_sorted -> is_rbt_rbt_sorted
   909   lookup_balance -> rbt_lookup_balance
   910   lookup_bulkload -> rbt_lookup_rbt_bulkload
   911   lookup_delete -> rbt_lookup_rbt_delete
   912   lookup_Empty -> rbt_lookup_Empty
   913   lookup_from_in_tree -> rbt_lookup_from_in_tree
   914   lookup_in_tree -> rbt_lookup_in_tree
   915   lookup_ins -> rbt_lookup_ins
   916   lookup_insert -> rbt_lookup_rbt_insert
   917   lookup_insertw -> rbt_lookup_rbt_insertw
   918   lookup_insertwk -> rbt_lookup_rbt_insertwk
   919   lookup_keys -> rbt_lookup_keys
   920   lookup_map -> rbt_lookup_map
   921   lookup_map_entry -> rbt_lookup_rbt_map_entry
   922   lookup_tree_greater -> rbt_lookup_rbt_greater
   923   lookup_tree_less -> rbt_lookup_rbt_less
   924   lookup_union -> rbt_lookup_rbt_union
   925   map_entry_color_of -> rbt_map_entry_color_of
   926   map_entry_inv1 -> rbt_map_entry_inv1
   927   map_entry_inv2 -> rbt_map_entry_inv2
   928   map_entry_is_rbt -> rbt_map_entry_is_rbt
   929   map_entry_sorted -> rbt_map_entry_rbt_sorted
   930   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   931   map_entry_tree_less -> rbt_map_entry_rbt_less
   932   map_tree_greater -> map_rbt_greater
   933   map_tree_less -> map_rbt_less
   934   map_sorted -> map_rbt_sorted
   935   paint_sorted -> paint_rbt_sorted
   936   paint_lookup -> paint_rbt_lookup
   937   paint_tree_greater -> paint_rbt_greater
   938   paint_tree_less -> paint_rbt_less
   939   sorted_entries -> rbt_sorted_entries
   940   tree_greater_eq_trans -> rbt_greater_eq_trans
   941   tree_greater_nit -> rbt_greater_nit
   942   tree_greater_prop -> rbt_greater_prop
   943   tree_greater_simps -> rbt_greater_simps
   944   tree_greater_trans -> rbt_greater_trans
   945   tree_less_eq_trans -> rbt_less_eq_trans
   946   tree_less_nit -> rbt_less_nit
   947   tree_less_prop -> rbt_less_prop
   948   tree_less_simps -> rbt_less_simps
   949   tree_less_trans -> rbt_less_trans
   950   tree_ord_props -> rbt_ord_props
   951   union_Branch -> rbt_union_Branch
   952   union_is_rbt -> rbt_union_is_rbt
   953   unionw_is_rbt -> rbt_unionw_is_rbt
   954   unionwk_is_rbt -> rbt_unionwk_is_rbt
   955   unionwk_sorted -> rbt_unionwk_rbt_sorted
   956 
   957 * Theory HOL/Library/Float: Floating point numbers are now defined as
   958 a subset of the real numbers.  All operations are defined using the
   959 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   960 
   961   Changed Operations:
   962   float_abs -> abs
   963   float_nprt -> nprt
   964   float_pprt -> pprt
   965   pow2 -> use powr
   966   round_down -> float_round_down
   967   round_up -> float_round_up
   968   scale -> exponent
   969 
   970   Removed Operations:
   971   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   972 
   973   Renamed Lemmas:
   974   abs_float_def -> Float.compute_float_abs
   975   bitlen_ge0 -> bitlen_nonneg
   976   bitlen.simps -> Float.compute_bitlen
   977   float_components -> Float_mantissa_exponent
   978   float_divl.simps -> Float.compute_float_divl
   979   float_divr.simps -> Float.compute_float_divr
   980   float_eq_odd -> mult_powr_eq_mult_powr_iff
   981   float_power -> real_of_float_power
   982   lapprox_posrat_def -> Float.compute_lapprox_posrat
   983   lapprox_rat.simps -> Float.compute_lapprox_rat
   984   le_float_def' -> Float.compute_float_le
   985   le_float_def -> less_eq_float.rep_eq
   986   less_float_def' -> Float.compute_float_less
   987   less_float_def -> less_float.rep_eq
   988   normfloat_def -> Float.compute_normfloat
   989   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   990   normfloat -> normfloat_def
   991   normfloat_unique -> use normfloat_def
   992   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   993   one_float_def -> Float.compute_float_one
   994   plus_float_def -> Float.compute_float_plus
   995   rapprox_posrat_def -> Float.compute_rapprox_posrat
   996   rapprox_rat.simps -> Float.compute_rapprox_rat
   997   real_of_float_0 -> zero_float.rep_eq
   998   real_of_float_1 -> one_float.rep_eq
   999   real_of_float_abs -> abs_float.rep_eq
  1000   real_of_float_add -> plus_float.rep_eq
  1001   real_of_float_minus -> uminus_float.rep_eq
  1002   real_of_float_mult -> times_float.rep_eq
  1003   real_of_float_simp -> Float.rep_eq
  1004   real_of_float_sub -> minus_float.rep_eq
  1005   round_down.simps -> Float.compute_float_round_down
  1006   round_up.simps -> Float.compute_float_round_up
  1007   times_float_def -> Float.compute_float_times
  1008   uminus_float_def -> Float.compute_float_uminus
  1009   zero_float_def -> Float.compute_float_zero
  1010 
  1011   Lemmas not necessary anymore, use the transfer method:
  1012   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
  1013   float_divr, float_le_simp, float_less1_mantissa_bound,
  1014   float_less_simp, float_less_zero, float_le_zero,
  1015   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
  1016   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
  1017   lapprox_rat_bottom, normalized_float, rapprox_posrat,
  1018   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
  1019   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
  1020   round_up, zero_le_float, zero_less_float
  1021 
  1022 * New theory HOL/Library/DAList provides an abstract type for
  1023 association lists with distinct keys.
  1024 
  1025 * Session HOL/IMP: Added new theory of abstract interpretation of
  1026 annotated commands.
  1027 
  1028 * Session HOL-Import: Re-implementation from scratch is faster,
  1029 simpler, and more scalable.  Requires a proof bundle, which is
  1030 available as an external component.  Discontinued old (and mostly
  1031 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
  1032 
  1033 * Session HOL-Word: Discontinued many redundant theorems specific to
  1034 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
  1035 instead.
  1036 
  1037   word_sub_alt ~> word_sub_wi
  1038   word_add_alt ~> word_add_def
  1039   word_mult_alt ~> word_mult_def
  1040   word_minus_alt ~> word_minus_def
  1041   word_0_alt ~> word_0_wi
  1042   word_1_alt ~> word_1_wi
  1043   word_add_0 ~> add_0_left
  1044   word_add_0_right ~> add_0_right
  1045   word_mult_1 ~> mult_1_left
  1046   word_mult_1_right ~> mult_1_right
  1047   word_add_commute ~> add_commute
  1048   word_add_assoc ~> add_assoc
  1049   word_add_left_commute ~> add_left_commute
  1050   word_mult_commute ~> mult_commute
  1051   word_mult_assoc ~> mult_assoc
  1052   word_mult_left_commute ~> mult_left_commute
  1053   word_left_distrib ~> left_distrib
  1054   word_right_distrib ~> right_distrib
  1055   word_left_minus ~> left_minus
  1056   word_diff_0_right ~> diff_0_right
  1057   word_diff_self ~> diff_self
  1058   word_sub_def ~> diff_minus
  1059   word_diff_minus ~> diff_minus
  1060   word_add_ac ~> add_ac
  1061   word_mult_ac ~> mult_ac
  1062   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1063   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1064   word_order_trans ~> order_trans
  1065   word_order_refl ~> order_refl
  1066   word_order_antisym ~> order_antisym
  1067   word_order_linear ~> linorder_linear
  1068   lenw1_zero_neq_one ~> zero_neq_one
  1069   word_number_of_eq ~> number_of_eq
  1070   word_of_int_add_hom ~> wi_hom_add
  1071   word_of_int_sub_hom ~> wi_hom_sub
  1072   word_of_int_mult_hom ~> wi_hom_mult
  1073   word_of_int_minus_hom ~> wi_hom_neg
  1074   word_of_int_succ_hom ~> wi_hom_succ
  1075   word_of_int_pred_hom ~> wi_hom_pred
  1076   word_of_int_0_hom ~> word_0_wi
  1077   word_of_int_1_hom ~> word_1_wi
  1078 
  1079 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1080 machine word equalities and inequalities into logical circuits,
  1081 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1082 multiplication, shifting by constants, bitwise operators and numeric
  1083 constants.  Requires fixed-length word types, not 'a word.  Solves
  1084 many standard word identities outright and converts more into first
  1085 order problems amenable to blast or similar.  See also examples in
  1086 HOL/Word/Examples/WordExamples.thy.
  1087 
  1088 * Session HOL-Probability: Introduced the type "'a measure" to
  1089 represent measures, this replaces the records 'a algebra and 'a
  1090 measure_space.  The locales based on subset_class now have two
  1091 locale-parameters the space \<Omega> and the set of measurable sets M.
  1092 The product of probability spaces uses now the same constant as the
  1093 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1094 measure".  Most constants are defined now outside of locales and gain
  1095 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1096 Measure space constructions for distributions and densities now got
  1097 their own constants distr and density.  Instead of using locales to
  1098 describe measure spaces with a finite space, the measure count_space
  1099 and point_measure is introduced.  INCOMPATIBILITY.
  1100 
  1101   Renamed constants:
  1102   measure -> emeasure
  1103   finite_measure.\<mu>' -> measure
  1104   product_algebra_generator -> prod_algebra
  1105   product_prob_space.emb -> prod_emb
  1106   product_prob_space.infprod_algebra -> PiM
  1107 
  1108   Removed locales:
  1109   completeable_measure_space
  1110   finite_measure_space
  1111   finite_prob_space
  1112   finite_product_finite_prob_space
  1113   finite_product_sigma_algebra
  1114   finite_sigma_algebra
  1115   measure_space
  1116   pair_finite_prob_space
  1117   pair_finite_sigma_algebra
  1118   pair_finite_space
  1119   pair_sigma_algebra
  1120   product_sigma_algebra
  1121 
  1122   Removed constants:
  1123   conditional_space
  1124   distribution -> use distr measure, or distributed predicate
  1125   image_space
  1126   joint_distribution -> use distr measure, or distributed predicate
  1127   pair_measure_generator
  1128   product_prob_space.infprod_algebra -> use PiM
  1129   subvimage
  1130 
  1131   Replacement theorems:
  1132   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1133   finite_measure.empty_measure -> measure_empty
  1134   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1135   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1136   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1137   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1138   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1139   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1140   finite_measure.positive_measure' -> measure_nonneg
  1141   finite_measure.real_measure -> finite_measure.emeasure_real
  1142   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1143   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1144   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1145   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1146   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1147   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1148   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1149   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1150   information_space.entropy_commute -> information_space.entropy_commute_simple
  1151   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1152   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1153   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1154   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1155   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1156   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1157   information_space.KL_ge_0 -> information_space.KL_nonneg
  1158   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1159   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1160   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1161   Int_stable_product_algebra_generator -> positive_integral
  1162   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1163   measure_space.additive -> emeasure_additive
  1164   measure_space.AE_iff_null_set -> AE_iff_null
  1165   measure_space.almost_everywhere_def -> eventually_ae_filter
  1166   measure_space.almost_everywhere_vimage -> AE_distrD
  1167   measure_space.continuity_from_above -> INF_emeasure_decseq
  1168   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1169   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1170   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1171   measure_space_density -> emeasure_density
  1172   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1173   measure_space.integrable_vimage -> integrable_distr
  1174   measure_space.integral_translated_density -> integral_density
  1175   measure_space.integral_vimage -> integral_distr
  1176   measure_space.measure_additive -> plus_emeasure
  1177   measure_space.measure_compl -> emeasure_compl
  1178   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1179   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1180   measure_space.measure_decseq -> decseq_emeasure
  1181   measure_space.measure_Diff -> emeasure_Diff
  1182   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1183   measure_space.measure_eq_0 -> emeasure_eq_0
  1184   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1185   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1186   measure_space.measure_incseq -> incseq_emeasure
  1187   measure_space.measure_insert -> emeasure_insert
  1188   measure_space.measure_mono -> emeasure_mono
  1189   measure_space.measure_not_negative -> emeasure_not_MInf
  1190   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1191   measure_space.measure_setsum -> setsum_emeasure
  1192   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1193   measure_space.measure_space_vimage -> emeasure_distr
  1194   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1195   measure_space.measure_subadditive -> subadditive
  1196   measure_space.measure_top -> emeasure_space
  1197   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1198   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1199   measure_space.positive_integral_translated_density -> positive_integral_density
  1200   measure_space.positive_integral_vimage -> positive_integral_distr
  1201   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1202   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1203   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1204   measure_space.real_measure_Diff -> measure_Diff
  1205   measure_space.real_measure_finite_Union -> measure_finite_Union
  1206   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1207   measure_space.real_measure_subadditive -> measure_subadditive
  1208   measure_space.real_measure_Union -> measure_Union
  1209   measure_space.real_measure_UNION -> measure_UNION
  1210   measure_space.simple_function_vimage -> simple_function_comp
  1211   measure_space.simple_integral_vimage -> simple_integral_distr
  1212   measure_space.simple_integral_vimage -> simple_integral_distr
  1213   measure_unique_Int_stable -> measure_eqI_generator_eq
  1214   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1215   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1216   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1217   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1218   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1219   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1220   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1221   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1222   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1223   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1224   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1225   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1226   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1227   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1228   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1229   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1230   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1231   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1232   prob_space.prob_space_vimage -> prob_space_distr
  1233   prob_space.random_variable_restrict -> measurable_restrict
  1234   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1235   product_algebraE -> prod_algebraE_all
  1236   product_algebra_generator_der -> prod_algebra_eq_finite
  1237   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1238   product_algebraI -> sets_PiM_I_finite
  1239   product_measure_exists -> product_sigma_finite.sigma_finite
  1240   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1241   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1242   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1243   product_prob_space.measurable_component -> measurable_component_singleton
  1244   product_prob_space.measurable_emb -> measurable_prod_emb
  1245   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1246   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1247   product_prob_space.measure_emb -> emeasure_prod_emb
  1248   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1249   product_sigma_algebra.product_algebra_into_space -> space_closed
  1250   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1251   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1252   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1253   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1254   sets_product_algebra -> sets_PiM
  1255   sigma_algebra.measurable_sigma -> measurable_measure_of
  1256   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1257   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1258   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1259   space_product_algebra -> space_PiM
  1260 
  1261 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1262 languages) into Isabelle/HOL.
  1263 
  1264 
  1265 *** FOL ***
  1266 
  1267 * New "case_product" attribute (see HOL).
  1268 
  1269 
  1270 *** ZF ***
  1271 
  1272 * Greater support for structured proofs involving induction or case
  1273 analysis.
  1274 
  1275 * Much greater use of mathematical symbols.
  1276 
  1277 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1278 
  1279 
  1280 *** ML ***
  1281 
  1282 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1283 from a minor keyword introduced via theory header declaration.
  1284 
  1285 * Antiquotation @{command_spec "name"} produces the
  1286 Outer_Syntax.command_spec from a major keyword introduced via theory
  1287 header declaration; it can be passed to Outer_Syntax.command etc.
  1288 
  1289 * Local_Theory.define no longer hard-wires default theorem name
  1290 "foo_def", but retains the binding as given.  If that is Binding.empty
  1291 / Attrib.empty_binding, the result is not registered as user-level
  1292 fact.  The Local_Theory.define_internal variant allows to specify a
  1293 non-empty name (used for the foundation in the background theory),
  1294 while omitting the fact binding in the user-context.  Potential
  1295 INCOMPATIBILITY for derived definitional packages: need to specify
  1296 naming policy for primitive definitions more explicitly.
  1297 
  1298 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1299 conformance with similar operations in structure Term and Logic.
  1300 
  1301 * Antiquotation @{attributes [...]} embeds attribute source
  1302 representation into the ML text, which is particularly useful with
  1303 declarations like Local_Theory.note.
  1304 
  1305 * Structure Proof_Context follows standard naming scheme.  Old
  1306 ProofContext has been discontinued.  INCOMPATIBILITY.
  1307 
  1308 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1309 change of semantics: update is applied to auxiliary local theory
  1310 context as well.
  1311 
  1312 * Modernized some old-style infix operations:
  1313 
  1314   addeqcongs    ~> Simplifier.add_eqcong
  1315   deleqcongs    ~> Simplifier.del_eqcong
  1316   addcongs      ~> Simplifier.add_cong
  1317   delcongs      ~> Simplifier.del_cong
  1318   setmksimps    ~> Simplifier.set_mksimps
  1319   setmkcong     ~> Simplifier.set_mkcong
  1320   setmksym      ~> Simplifier.set_mksym
  1321   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1322   settermless   ~> Simplifier.set_termless
  1323   setsubgoaler  ~> Simplifier.set_subgoaler
  1324   addsplits     ~> Splitter.add_split
  1325   delsplits     ~> Splitter.del_split
  1326 
  1327 
  1328 *** System ***
  1329 
  1330 * USER_HOME settings variable points to cross-platform user home
  1331 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1332 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1333 of former $HOME.  A different default for USER_HOME may be set
  1334 explicitly in shell environment, before Isabelle settings are
  1335 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1336 the generic user home was intended.
  1337 
  1338 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1339 notation, which is useful for the jEdit file browser, for example.
  1340 
  1341 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1342 (not just JRE).
  1343 
  1344 
  1345 
  1346 New in Isabelle2011-1 (October 2011)
  1347 ------------------------------------
  1348 
  1349 *** General ***
  1350 
  1351 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1352 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1353 
  1354   - Management of multiple theory files directly from the editor
  1355     buffer store -- bypassing the file-system (no requirement to save
  1356     files for checking).
  1357 
  1358   - Markup of formal entities within the text buffer, with semantic
  1359     highlighting, tooltips and hyperlinks to jump to defining source
  1360     positions.
  1361 
  1362   - Improved text rendering, with sub/superscripts in the source
  1363     buffer (including support for copy/paste wrt. output panel, HTML
  1364     theory output and other non-Isabelle text boxes).
  1365 
  1366   - Refined scheduling of proof checking and printing of results,
  1367     based on interactive editor view.  (Note: jEdit folding and
  1368     narrowing allows to restrict buffer perspectives explicitly.)
  1369 
  1370   - Reduced CPU performance requirements, usable on machines with few
  1371     cores.
  1372 
  1373   - Reduced memory requirements due to pruning of unused document
  1374     versions (garbage collection).
  1375 
  1376 See also ~~/src/Tools/jEdit/README.html for further information,
  1377 including some remaining limitations.
  1378 
  1379 * Theory loader: source files are exclusively located via the master
  1380 directory of each theory node (where the .thy file itself resides).
  1381 The global load path (such as src/HOL/Library) has been discontinued.
  1382 Note that the path element ~~ may be used to reference theories in the
  1383 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1384 INCOMPATIBILITY.
  1385 
  1386 * Theory loader: source files are identified by content via SHA1
  1387 digests.  Discontinued former path/modtime identification and optional
  1388 ISABELLE_FILE_IDENT plugin scripts.
  1389 
  1390 * Parallelization of nested Isar proofs is subject to
  1391 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1392 usedir option -Q.
  1393 
  1394 * Name space: former unsynchronized references are now proper
  1395 configuration options, with more conventional names:
  1396 
  1397   long_names   ~> names_long
  1398   short_names  ~> names_short
  1399   unique_names ~> names_unique
  1400 
  1401 Minor INCOMPATIBILITY, need to declare options in context like this:
  1402 
  1403   declare [[names_unique = false]]
  1404 
  1405 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1406 that the result needs to be unique, which means fact specifications
  1407 may have to be refined after enriching a proof context.
  1408 
  1409 * Attribute "case_names" has been refined: the assumptions in each case
  1410 can be named now by following the case name with [name1 name2 ...].
  1411 
  1412 * Isabelle/Isar reference manual has been updated and extended:
  1413   - "Synopsis" provides a catalog of main Isar language concepts.
  1414   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1415   - Updated material from classic "ref" manual, notably about
  1416     "Classical Reasoner".
  1417 
  1418 
  1419 *** HOL ***
  1420 
  1421 * Class bot and top require underlying partial order rather than
  1422 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1423 
  1424 * Class complete_lattice: generalized a couple of lemmas from sets;
  1425 generalized theorems INF_cong and SUP_cong.  New type classes for
  1426 complete boolean algebras and complete linear orders.  Lemmas
  1427 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1428 class complete_linorder.
  1429 
  1430 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1431 Sup_fun_def, Inf_apply, Sup_apply.
  1432 
  1433 Removed redundant lemmas (the right hand side gives hints how to
  1434 replace them for (metis ...), or (simp only: ...) proofs):
  1435 
  1436   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1437   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1438   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1439   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1440   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1441   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1442   Inter_def ~> INF_def, image_def
  1443   Union_def ~> SUP_def, image_def
  1444   INT_eq ~> INF_def, and image_def
  1445   UN_eq ~> SUP_def, and image_def
  1446   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1447 
  1448 More consistent and comprehensive names:
  1449 
  1450   INTER_eq_Inter_image ~> INF_def
  1451   UNION_eq_Union_image ~> SUP_def
  1452   INFI_def ~> INF_def
  1453   SUPR_def ~> SUP_def
  1454   INF_leI ~> INF_lower
  1455   INF_leI2 ~> INF_lower2
  1456   le_INFI ~> INF_greatest
  1457   le_SUPI ~> SUP_upper
  1458   le_SUPI2 ~> SUP_upper2
  1459   SUP_leI ~> SUP_least
  1460   INFI_bool_eq ~> INF_bool_eq
  1461   SUPR_bool_eq ~> SUP_bool_eq
  1462   INFI_apply ~> INF_apply
  1463   SUPR_apply ~> SUP_apply
  1464   INTER_def ~> INTER_eq
  1465   UNION_def ~> UNION_eq
  1466 
  1467 INCOMPATIBILITY.
  1468 
  1469 * Renamed theory Complete_Lattice to Complete_Lattices.
  1470 INCOMPATIBILITY.
  1471 
  1472 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1473 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1474 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1475 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1476 
  1477 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1478 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1479 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1480 INCOMPATIBILITY.
  1481 
  1482 * Added syntactic classes "inf" and "sup" for the respective
  1483 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1484 (mostly internal) locale predicates for some derived classes.
  1485 
  1486 * Theorem collections ball_simps and bex_simps do not contain theorems
  1487 referring to UNION any longer; these have been moved to collection
  1488 UN_ball_bex_simps.  INCOMPATIBILITY.
  1489 
  1490 * Theory Archimedean_Field: floor now is defined as parameter of a
  1491 separate type class floor_ceiling.
  1492 
  1493 * Theory Finite_Set: more coherent development of fold_set locales:
  1494 
  1495     locale fun_left_comm ~> locale comp_fun_commute
  1496     locale fun_left_comm_idem ~> locale comp_fun_idem
  1497 
  1498 Both use point-free characterization; interpretation proofs may need
  1499 adjustment.  INCOMPATIBILITY.
  1500 
  1501 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1502 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1503 
  1504 * Theory Complex_Main: The locale interpretations for the
  1505 bounded_linear and bounded_bilinear locales have been removed, in
  1506 order to reduce the number of duplicate lemmas. Users must use the
  1507 original names for distributivity theorems, potential INCOMPATIBILITY.
  1508 
  1509   divide.add ~> add_divide_distrib
  1510   divide.diff ~> diff_divide_distrib
  1511   divide.setsum ~> setsum_divide_distrib
  1512   mult.add_right ~> right_distrib
  1513   mult.diff_right ~> right_diff_distrib
  1514   mult_right.setsum ~> setsum_right_distrib
  1515   mult_left.diff ~> left_diff_distrib
  1516 
  1517 * Theory Complex_Main: Several redundant theorems have been removed or
  1518 replaced by more general versions. INCOMPATIBILITY.
  1519 
  1520   real_diff_def ~> minus_real_def
  1521   real_divide_def ~> divide_real_def
  1522   real_less_def ~> less_le
  1523   real_abs_def ~> abs_real_def
  1524   real_sgn_def ~> sgn_real_def
  1525   real_mult_commute ~> mult_commute
  1526   real_mult_assoc ~> mult_assoc
  1527   real_mult_1 ~> mult_1_left
  1528   real_add_mult_distrib ~> left_distrib
  1529   real_zero_not_eq_one ~> zero_neq_one
  1530   real_mult_inverse_left ~> left_inverse
  1531   INVERSE_ZERO ~> inverse_zero
  1532   real_le_refl ~> order_refl
  1533   real_le_antisym ~> order_antisym
  1534   real_le_trans ~> order_trans
  1535   real_le_linear ~> linear
  1536   real_le_eq_diff ~> le_iff_diff_le_0
  1537   real_add_left_mono ~> add_left_mono
  1538   real_mult_order ~> mult_pos_pos
  1539   real_mult_less_mono2 ~> mult_strict_left_mono
  1540   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1541   real_0_le_divide_iff ~> zero_le_divide_iff
  1542   realpow_two_disj ~> power2_eq_iff
  1543   real_squared_diff_one_factored ~> square_diff_one_factored
  1544   realpow_two_diff ~> square_diff_square_factored
  1545   reals_complete2 ~> complete_real
  1546   real_sum_squared_expand ~> power2_sum
  1547   exp_ln_eq ~> ln_unique
  1548   expi_add ~> exp_add
  1549   expi_zero ~> exp_zero
  1550   lemma_DERIV_subst ~> DERIV_cong
  1551   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1552   LIMSEQ_const ~> tendsto_const
  1553   LIMSEQ_norm ~> tendsto_norm
  1554   LIMSEQ_add ~> tendsto_add
  1555   LIMSEQ_minus ~> tendsto_minus
  1556   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1557   LIMSEQ_diff ~> tendsto_diff
  1558   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1559   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1560   LIMSEQ_mult ~> tendsto_mult
  1561   LIMSEQ_inverse ~> tendsto_inverse
  1562   LIMSEQ_divide ~> tendsto_divide
  1563   LIMSEQ_pow ~> tendsto_power
  1564   LIMSEQ_setsum ~> tendsto_setsum
  1565   LIMSEQ_setprod ~> tendsto_setprod
  1566   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1567   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1568   LIMSEQ_imp_rabs ~> tendsto_rabs
  1569   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1570   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1571   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1572   LIMSEQ_Complex ~> tendsto_Complex
  1573   LIM_ident ~> tendsto_ident_at
  1574   LIM_const ~> tendsto_const
  1575   LIM_add ~> tendsto_add
  1576   LIM_add_zero ~> tendsto_add_zero
  1577   LIM_minus ~> tendsto_minus
  1578   LIM_diff ~> tendsto_diff
  1579   LIM_norm ~> tendsto_norm
  1580   LIM_norm_zero ~> tendsto_norm_zero
  1581   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1582   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1583   LIM_rabs ~> tendsto_rabs
  1584   LIM_rabs_zero ~> tendsto_rabs_zero
  1585   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1586   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1587   LIM_compose ~> tendsto_compose
  1588   LIM_mult ~> tendsto_mult
  1589   LIM_scaleR ~> tendsto_scaleR
  1590   LIM_of_real ~> tendsto_of_real
  1591   LIM_power ~> tendsto_power
  1592   LIM_inverse ~> tendsto_inverse
  1593   LIM_sgn ~> tendsto_sgn
  1594   isCont_LIM_compose ~> isCont_tendsto_compose
  1595   bounded_linear.LIM ~> bounded_linear.tendsto
  1596   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1597   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1598   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1599   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1600   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1601   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1602 
  1603 * Theory Complex_Main: The definition of infinite series was
  1604 generalized.  Now it is defined on the type class {topological_space,
  1605 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1606 
  1607 * Theory Complex_Main: The complex exponential function "expi" is now
  1608 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1609 several polymorphic lemmas about "exp" are now applicable to "expi".
  1610 
  1611 * Code generation:
  1612 
  1613   - Theory Library/Code_Char_ord provides native ordering of
  1614     characters in the target language.
  1615 
  1616   - Commands code_module and code_library are legacy, use export_code
  1617     instead.
  1618 
  1619   - Method "evaluation" is legacy, use method "eval" instead.
  1620 
  1621   - Legacy evaluator "SML" is deactivated by default.  May be
  1622     reactivated by the following theory command:
  1623 
  1624       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1625 
  1626 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1627 
  1628 * New proof method "induction" that gives induction hypotheses the
  1629 name "IH", thus distinguishing them from further hypotheses that come
  1630 from rule induction.  The latter are still called "hyps".  Method
  1631 "induction" is a thin wrapper around "induct" and follows the same
  1632 syntax.
  1633 
  1634 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1635 still available as a legacy feature for some time.
  1636 
  1637 * Nitpick:
  1638   - Added "need" and "total_consts" options.
  1639   - Reintroduced "show_skolems" option by popular demand.
  1640   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1641     INCOMPATIBILITY.
  1642 
  1643 * Sledgehammer:
  1644   - Use quasi-sound (and efficient) translations by default.
  1645   - Added support for the following provers: E-ToFoF, LEO-II,
  1646     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1647   - Automatically preplay and minimize proofs before showing them if
  1648     this can be done within reasonable time.
  1649   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1650     INCOMPATIBILITY.
  1651   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1652     "max_mono_iters", and "max_new_mono_instances" options.
  1653   - Removed "explicit_apply" and "full_types" options as well as "Full
  1654     Types" Proof General menu item. INCOMPATIBILITY.
  1655 
  1656 * Metis:
  1657   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1658   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1659     INCOMPATIBILITY.
  1660 
  1661 * Command 'try':
  1662   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1663     "elim:" options. INCOMPATIBILITY.
  1664   - Introduced 'try' that not only runs 'try_methods' but also
  1665     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1666 
  1667 * Quickcheck:
  1668   - Added "eval" option to evaluate terms for the found counterexample
  1669     (currently only supported by the default (exhaustive) tester).
  1670   - Added post-processing of terms to obtain readable counterexamples
  1671     (currently only supported by the default (exhaustive) tester).
  1672   - New counterexample generator quickcheck[narrowing] enables
  1673     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1674     with its installation location defined in the Isabelle settings
  1675     environment as ISABELLE_GHC.
  1676   - Removed quickcheck tester "SML" based on the SML code generator
  1677     (formly in HOL/Library).
  1678 
  1679 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1680 use 'partial_function' instead.
  1681 
  1682 * Theory Library/Extended_Reals replaces now the positive extended
  1683 reals found in probability theory. This file is extended by
  1684 Multivariate_Analysis/Extended_Real_Limits.
  1685 
  1686 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1687 from where it must be imported explicitly if it is really required.
  1688 INCOMPATIBILITY.
  1689 
  1690 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1691 been moved here.  INCOMPATIBILITY.
  1692 
  1693 * Theory Library/Saturated provides type of numbers with saturated
  1694 arithmetic.
  1695 
  1696 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1697 product type 'a * 'b, and provides instance proofs for various order
  1698 and lattice type classes.
  1699 
  1700 * Theory Library/Countable now provides the "countable_datatype" proof
  1701 method for proving "countable" class instances for datatypes.
  1702 
  1703 * Theory Library/Cset_Monad allows do notation for computable sets
  1704 (cset) via the generic monad ad-hoc overloading facility.
  1705 
  1706 * Library: Theories of common data structures are split into theories
  1707 for implementation, an invariant-ensuring type, and connection to an
  1708 abstract type. INCOMPATIBILITY.
  1709 
  1710   - RBT is split into RBT and RBT_Mapping.
  1711   - AssocList is split and renamed into AList and AList_Mapping.
  1712   - DList is split into DList_Impl, DList, and DList_Cset.
  1713   - Cset is split into Cset and List_Cset.
  1714 
  1715 * Theory Library/Nat_Infinity has been renamed to
  1716 Library/Extended_Nat, with name changes of the following types and
  1717 constants:
  1718 
  1719   type inat   ~> type enat
  1720   Fin         ~> enat
  1721   Infty       ~> infinity (overloaded)
  1722   iSuc        ~> eSuc
  1723   the_Fin     ~> the_enat
  1724 
  1725 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1726 been renamed accordingly. INCOMPATIBILITY.
  1727 
  1728 * Session Multivariate_Analysis: The euclidean_space type class now
  1729 fixes a constant "Basis :: 'a set" consisting of the standard
  1730 orthonormal basis for the type. Users now have the option of
  1731 quantifying over this set instead of using the "basis" function, e.g.
  1732 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1733 
  1734 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1735 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1736 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1737 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1738 changed to match. Definition theorems for overloaded constants now use
  1739 the standard "foo_vec_def" naming scheme. A few other theorems have
  1740 been renamed as follows (INCOMPATIBILITY):
  1741 
  1742   Cart_eq          ~> vec_eq_iff
  1743   dist_nth_le_cart ~> dist_vec_nth_le
  1744   tendsto_vector   ~> vec_tendstoI
  1745   Cauchy_vector    ~> vec_CauchyI
  1746 
  1747 * Session Multivariate_Analysis: Several duplicate theorems have been
  1748 removed, and other theorems have been renamed or replaced with more
  1749 general versions. INCOMPATIBILITY.
  1750 
  1751   finite_choice ~> finite_set_choice
  1752   eventually_conjI ~> eventually_conj
  1753   eventually_and ~> eventually_conj_iff
  1754   eventually_false ~> eventually_False
  1755   setsum_norm ~> norm_setsum
  1756   Lim_sequentially ~> LIMSEQ_def
  1757   Lim_ident_at ~> LIM_ident
  1758   Lim_const ~> tendsto_const
  1759   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1760   Lim_neg ~> tendsto_minus
  1761   Lim_add ~> tendsto_add
  1762   Lim_sub ~> tendsto_diff
  1763   Lim_mul ~> tendsto_scaleR
  1764   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1765   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1766   Lim_linear ~> bounded_linear.tendsto
  1767   Lim_component ~> tendsto_euclidean_component
  1768   Lim_component_cart ~> tendsto_vec_nth
  1769   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1770   dot_lsum ~> inner_setsum_left
  1771   dot_rsum ~> inner_setsum_right
  1772   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1773   continuous_neg ~> continuous_minus
  1774   continuous_sub ~> continuous_diff
  1775   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1776   continuous_mul ~> continuous_scaleR
  1777   continuous_inv ~> continuous_inverse
  1778   continuous_at_within_inv ~> continuous_at_within_inverse
  1779   continuous_at_inv ~> continuous_at_inverse
  1780   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1781   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1782   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1783   continuous_on_neg ~> continuous_on_minus
  1784   continuous_on_sub ~> continuous_on_diff
  1785   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1786   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1787   continuous_on_mul ~> continuous_on_scaleR
  1788   continuous_on_mul_real ~> continuous_on_mult
  1789   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1790   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1791   continuous_on_inverse ~> continuous_on_inv
  1792   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1793   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1794   subset_interior ~> interior_mono
  1795   subset_closure ~> closure_mono
  1796   closure_univ ~> closure_UNIV
  1797   real_arch_lt ~> reals_Archimedean2
  1798   real_arch ~> reals_Archimedean3
  1799   real_abs_norm ~> abs_norm_cancel
  1800   real_abs_sub_norm ~> norm_triangle_ineq3
  1801   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1802 
  1803 * Session HOL-Probability:
  1804   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1805   - Infinite products of probability measures are now available.
  1806   - Sigma closure is independent, if the generator is independent
  1807   - Use extended reals instead of positive extended
  1808     reals. INCOMPATIBILITY.
  1809 
  1810 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1811 
  1812   expand_fun_below ~> fun_below_iff
  1813   below_fun_ext ~> fun_belowI
  1814   expand_cfun_eq ~> cfun_eq_iff
  1815   ext_cfun ~> cfun_eqI
  1816   expand_cfun_below ~> cfun_below_iff
  1817   below_cfun_ext ~> cfun_belowI
  1818   monofun_fun_fun ~> fun_belowD
  1819   monofun_fun_arg ~> monofunE
  1820   monofun_lub_fun ~> adm_monofun [THEN admD]
  1821   cont_lub_fun ~> adm_cont [THEN admD]
  1822   cont2cont_Rep_CFun ~> cont2cont_APP
  1823   cont_Rep_CFun_app ~> cont_APP_app
  1824   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1825   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1826   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1827   contlub_cfun ~> lub_APP [symmetric]
  1828   contlub_LAM ~> lub_LAM [symmetric]
  1829   thelubI ~> lub_eqI
  1830   UU_I ~> bottomI
  1831   lift_distinct1 ~> lift.distinct(1)
  1832   lift_distinct2 ~> lift.distinct(2)
  1833   Def_not_UU ~> lift.distinct(2)
  1834   Def_inject ~> lift.inject
  1835   below_UU_iff ~> below_bottom_iff
  1836   eq_UU_iff ~> eq_bottom_iff
  1837 
  1838 
  1839 *** Document preparation ***
  1840 
  1841 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1842 isar-ref manual, both for description and actual application of the
  1843 same.
  1844 
  1845 * Antiquotation @{value} evaluates the given term and presents its
  1846 result.
  1847 
  1848 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1849 variables x1, y23 into subscripted form x\<^isub>1,
  1850 y\<^isub>2\<^isub>3.
  1851 
  1852 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1853 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1854 
  1855 * Localized \isabellestyle switch can be used within blocks or groups
  1856 like this:
  1857 
  1858   \isabellestyle{it}  %preferred default
  1859   {\isabellestylett @{text "typewriter stuff"}}
  1860 
  1861 * Discontinued special treatment of hard tabulators.  Implicit
  1862 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1863 layouts.
  1864 
  1865 
  1866 *** ML ***
  1867 
  1868 * The inner syntax of sort/type/term/prop supports inlined YXML
  1869 representations within quoted string tokens.  By encoding logical
  1870 entities via Term_XML (in ML or Scala) concrete syntax can be
  1871 bypassed, which is particularly useful for producing bits of text
  1872 under external program control.
  1873 
  1874 * Antiquotations for ML and document preparation are managed as theory
  1875 data, which requires explicit setup.
  1876 
  1877 * Isabelle_Process.is_active allows tools to check if the official
  1878 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1879 (better known as Proof General).
  1880 
  1881 * Structure Proof_Context follows standard naming scheme.  Old
  1882 ProofContext is still available for some time as legacy alias.
  1883 
  1884 * Structure Timing provides various operations for timing; supersedes
  1885 former start_timing/end_timing etc.
  1886 
  1887 * Path.print is the official way to show file-system paths to users
  1888 (including quotes etc.).
  1889 
  1890 * Inner syntax: identifiers in parse trees of generic categories
  1891 "logic", "aprop", "idt" etc. carry position information (disguised as
  1892 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1893 translations that choke on unexpected type constraints.  Positions can
  1894 be stripped in ML translations via Syntax.strip_positions /
  1895 Syntax.strip_positions_ast, or via the syntax constant
  1896 "_strip_positions" within parse trees.  As last resort, positions can
  1897 be disabled via the configuration option Syntax.positions, which is
  1898 called "syntax_positions" in Isar attribute syntax.
  1899 
  1900 * Discontinued special status of various ML structures that contribute
  1901 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1902 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1903 refer directly to Ast.Constant, Lexicon.is_identifier,
  1904 Syntax_Trans.mk_binder_tr etc.
  1905 
  1906 * Typed print translation: discontinued show_sorts argument, which is
  1907 already available via context of "advanced" translation.
  1908 
  1909 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1910 goal states; body tactic needs to address all subgoals uniformly.
  1911 
  1912 * Slightly more special eq_list/eq_set, with shortcut involving
  1913 pointer equality (assumes that eq relation is reflexive).
  1914 
  1915 * Classical tactics use proper Proof.context instead of historic types
  1916 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1917 operate directly on Proof.context.  Raw type claset retains its use as
  1918 snapshot of the classical context, which can be recovered via
  1919 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1920 INCOMPATIBILITY, classical tactics and derived proof methods require
  1921 proper Proof.context.
  1922 
  1923 
  1924 *** System ***
  1925 
  1926 * Discontinued support for Poly/ML 5.2, which was the last version
  1927 without proper multithreading and TimeLimit implementation.
  1928 
  1929 * Discontinued old lib/scripts/polyml-platform, which has been
  1930 obsolete since Isabelle2009-2.
  1931 
  1932 * Various optional external tools are referenced more robustly and
  1933 uniformly by explicit Isabelle settings as follows:
  1934 
  1935   ISABELLE_CSDP   (formerly CSDP_EXE)
  1936   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1937   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1938   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1939   ISABELLE_YAP    (formerly EXEC_YAP)
  1940 
  1941 Note that automated detection from the file-system or search path has
  1942 been discontinued.  INCOMPATIBILITY.
  1943 
  1944 * Scala layer provides JVM method invocation service for static
  1945 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1946 example:
  1947 
  1948   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1949 
  1950 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1951 this allows to pass structured values between ML and Scala.
  1952 
  1953 * The IsabelleText fonts includes some further glyphs to support the
  1954 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1955 installed a local copy (which is normally *not* required) need to
  1956 delete or update it from ~~/lib/fonts/.
  1957 
  1958 
  1959 
  1960 New in Isabelle2011 (January 2011)
  1961 ----------------------------------
  1962 
  1963 *** General ***
  1964 
  1965 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1966 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1967 useful tooltips and hyperlinks produced from its static analysis.  The
  1968 bundled component provides an executable Isabelle tool that can be run
  1969 like this:
  1970 
  1971   Isabelle2011/bin/isabelle jedit
  1972 
  1973 * Significantly improved Isabelle/Isar implementation manual.
  1974 
  1975 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1976 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1977 while the default heap location within that directory lacks that extra
  1978 suffix.  This isolates multiple Isabelle installations from each
  1979 other, avoiding problems with old settings in new versions.
  1980 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1981 
  1982 * Source files are always encoded as UTF-8, instead of old-fashioned
  1983 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1984 the following package declarations:
  1985 
  1986   \usepackage[utf8]{inputenc}
  1987   \usepackage{textcomp}
  1988 
  1989 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1990 a Unicode character is treated as a single symbol, not a sequence of
  1991 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1992 contain symbols without further backslash escapes, Unicode can now be
  1993 used here as well.  Recall that Symbol.explode in ML provides a
  1994 consistent view on symbols, while raw explode (or String.explode)
  1995 merely give a byte-oriented representation.
  1996 
  1997 * Theory loader: source files are primarily located via the master
  1998 directory of each theory node (where the .thy file itself resides).
  1999 The global load path is still partially available as legacy feature.
  2000 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  2001 explicit paths, relatively to the theory.
  2002 
  2003 * Special treatment of ML file names has been discontinued.
  2004 Historically, optional extensions .ML or .sml were added on demand --
  2005 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  2006 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  2007 
  2008 * Various options that affect pretty printing etc. are now properly
  2009 handled within the context via configuration options, instead of
  2010 unsynchronized references or print modes.  There are both ML Config.T
  2011 entities and Isar declaration attributes to access these.
  2012 
  2013   ML (Config.T)                 Isar (attribute)
  2014 
  2015   eta_contract                  eta_contract
  2016   show_brackets                 show_brackets
  2017   show_sorts                    show_sorts
  2018   show_types                    show_types
  2019   show_question_marks           show_question_marks
  2020   show_consts                   show_consts
  2021   show_abbrevs                  show_abbrevs
  2022 
  2023   Syntax.ast_trace              syntax_ast_trace
  2024   Syntax.ast_stat               syntax_ast_stat
  2025   Syntax.ambiguity_level        syntax_ambiguity_level
  2026 
  2027   Goal_Display.goals_limit      goals_limit
  2028   Goal_Display.show_main_goal   show_main_goal
  2029 
  2030   Method.rule_trace             rule_trace
  2031 
  2032   Thy_Output.display            thy_output_display
  2033   Thy_Output.quotes             thy_output_quotes
  2034   Thy_Output.indent             thy_output_indent
  2035   Thy_Output.source             thy_output_source
  2036   Thy_Output.break              thy_output_break
  2037 
  2038 Note that corresponding "..._default" references in ML may only be
  2039 changed globally at the ROOT session setup, but *not* within a theory.
  2040 The option "show_abbrevs" supersedes the former print mode
  2041 "no_abbrevs" with inverted meaning.
  2042 
  2043 * More systematic naming of some configuration options.
  2044 INCOMPATIBILITY.
  2045 
  2046   trace_simp  ~>  simp_trace
  2047   debug_simp  ~>  simp_debug
  2048 
  2049 * Support for real valued configuration options, using simplistic
  2050 floating-point notation that coincides with the inner syntax for
  2051 float_token.
  2052 
  2053 * Support for real valued preferences (with approximative PGIP type):
  2054 front-ends need to accept "pgint" values in float notation.
  2055 INCOMPATIBILITY.
  2056 
  2057 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2058 DejaVu Sans.
  2059 
  2060 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2061 
  2062 
  2063 *** Pure ***
  2064 
  2065 * Command 'type_synonym' (with single argument) replaces somewhat
  2066 outdated 'types', which is still available as legacy feature for some
  2067 time.
  2068 
  2069 * Command 'nonterminal' (with 'and' separated list of arguments)
  2070 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2071 
  2072 * Command 'notepad' replaces former 'example_proof' for
  2073 experimentation in Isar without any result.  INCOMPATIBILITY.
  2074 
  2075 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2076 lists of equations to map definitions in a locale to appropriate
  2077 entities in the context of the interpretation.  The 'interpretation'
  2078 command already provided this functionality.
  2079 
  2080 * Diagnostic command 'print_dependencies' prints the locale instances
  2081 that would be activated if the specified expression was interpreted in
  2082 the current context.  Variant "print_dependencies!" assumes a context
  2083 without interpretations.
  2084 
  2085 * Diagnostic command 'print_interps' prints interpretations in proofs
  2086 in addition to interpretations in theories.
  2087 
  2088 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2089 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2090 Sign.root_path and Sign.local_path may be applied directly where this
  2091 feature is still required for historical reasons.
  2092 
  2093 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2094 'definition' instead.
  2095 
  2096 * The "prems" fact, which refers to the accidental collection of
  2097 foundational premises in the context, is now explicitly marked as
  2098 legacy feature and will be discontinued soon.  Consider using "assms"
  2099 of the head statement or reference facts by explicit names.
  2100 
  2101 * Document antiquotations @{class} and @{type} print classes and type
  2102 constructors.
  2103 
  2104 * Document antiquotation @{file} checks file/directory entries within
  2105 the local file system.
  2106 
  2107 
  2108 *** HOL ***
  2109 
  2110 * Coercive subtyping: functions can be declared as coercions and type
  2111 inference will add them as necessary upon input of a term.  Theory
  2112 Complex_Main declares real :: nat => real and real :: int => real as
  2113 coercions. A coercion function f is declared like this:
  2114 
  2115   declare [[coercion f]]
  2116 
  2117 To lift coercions through type constructors (e.g. from nat => real to
  2118 nat list => real list), map functions can be declared, e.g.
  2119 
  2120   declare [[coercion_map map]]
  2121 
  2122 Currently coercion inference is activated only in theories including
  2123 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2124 the configuration option "coercion_enabled", e.g. it can be enabled in
  2125 other theories like this:
  2126 
  2127   declare [[coercion_enabled]]
  2128 
  2129 * Command 'partial_function' provides basic support for recursive
  2130 function definitions over complete partial orders.  Concrete instances
  2131 are provided for i) the option type, ii) tail recursion on arbitrary
  2132 types, and iii) the heap monad of Imperative_HOL.  See
  2133 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2134 for examples.
  2135 
  2136 * Function package: f.psimps rules are no longer implicitly declared
  2137 as [simp].  INCOMPATIBILITY.
  2138 
  2139 * Datatype package: theorems generated for executable equality (class
  2140 "eq") carry proper names and are treated as default code equations.
  2141 
  2142 * Inductive package: now offers command 'inductive_simps' to
  2143 automatically derive instantiated and simplified equations for
  2144 inductive predicates, similar to 'inductive_cases'.
  2145 
  2146 * Command 'enriched_type' allows to register properties of the
  2147 functorial structure of types.
  2148 
  2149 * Improved infrastructure for term evaluation using code generator
  2150 techniques, in particular static evaluation conversions.
  2151 
  2152 * Code generator: Scala (2.8 or higher) has been added to the target
  2153 languages.
  2154 
  2155 * Code generator: globbing constant expressions "*" and "Theory.*"
  2156 have been replaced by the more idiomatic "_" and "Theory._".
  2157 INCOMPATIBILITY.
  2158 
  2159 * Code generator: export_code without explicit file declaration prints
  2160 to standard output.  INCOMPATIBILITY.
  2161 
  2162 * Code generator: do not print function definitions for case
  2163 combinators any longer.
  2164 
  2165 * Code generator: simplification with rules determined with
  2166 src/Tools/Code/code_simp.ML and method "code_simp".
  2167 
  2168 * Code generator for records: more idiomatic representation of record
  2169 types.  Warning: records are not covered by ancient SML code
  2170 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2171 rep_datatype declaration helps to succeed then:
  2172 
  2173   record 'a foo = ...
  2174   ...
  2175   rep_datatype foo_ext ...
  2176 
  2177 * Records: logical foundation type for records does not carry a
  2178 '_type' suffix any longer (obsolete due to authentic syntax).
  2179 INCOMPATIBILITY.
  2180 
  2181 * Quickcheck now by default uses exhaustive testing instead of random
  2182 testing.  Random testing can be invoked by "quickcheck [random]",
  2183 exhaustive testing by "quickcheck [exhaustive]".
  2184 
  2185 * Quickcheck instantiates polymorphic types with small finite
  2186 datatypes by default. This enables a simple execution mechanism to
  2187 handle quantifiers and function equality over the finite datatypes.
  2188 
  2189 * Quickcheck random generator has been renamed from "code" to
  2190 "random".  INCOMPATIBILITY.
  2191 
  2192 * Quickcheck now has a configurable time limit which is set to 30
  2193 seconds by default. This can be changed by adding [timeout = n] to the
  2194 quickcheck command. The time limit for Auto Quickcheck is still set
  2195 independently.
  2196 
  2197 * Quickcheck in locales considers interpretations of that locale for
  2198 counter example search.
  2199 
  2200 * Sledgehammer:
  2201   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2202     method. See the Sledgehammer manual for details ("isabelle doc
  2203     sledgehammer").
  2204   - Renamed commands:
  2205     sledgehammer atp_info ~> sledgehammer running_provers
  2206     sledgehammer atp_kill ~> sledgehammer kill_provers
  2207     sledgehammer available_atps ~> sledgehammer available_provers
  2208     INCOMPATIBILITY.
  2209   - Renamed options:
  2210     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2211     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2212     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2213     (and "ms" and "min" are no longer supported)
  2214     INCOMPATIBILITY.
  2215 
  2216 * Nitpick:
  2217   - Renamed options:
  2218     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2219     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2220     INCOMPATIBILITY.
  2221   - Added support for partial quotient types.
  2222   - Added local versions of the "Nitpick.register_xxx" functions.
  2223   - Added "whack" option.
  2224   - Allow registration of quotient types as codatatypes.
  2225   - Improved "merge_type_vars" option to merge more types.
  2226   - Removed unsound "fast_descrs" option.
  2227   - Added custom symmetry breaking for datatypes, making it possible to reach
  2228     higher cardinalities.
  2229   - Prevent the expansion of too large definitions.
  2230 
  2231 * Proof methods "metis" and "meson" now have configuration options
  2232 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2233 to diagnose these tools. E.g.
  2234 
  2235     using [[metis_trace = true]]
  2236 
  2237 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2238 manually as command 'solve_direct'.
  2239 
  2240 * The default SMT solver Z3 must be enabled explicitly (due to
  2241 licensing issues) by setting the environment variable
  2242 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2243 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2244 changing the SMT solver is done via the configuration option
  2245 "smt_solver".
  2246 
  2247 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2248 i.e. "remote_cvc3" and "remote_z3".
  2249 
  2250 * Added basic SMT support for datatypes, records, and typedefs using
  2251 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2252 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2253 support for a similar behavior).  Minor INCOMPATIBILITY.
  2254 
  2255 * Changed SMT configuration options:
  2256   - Renamed:
  2257     z3_proofs ~> smt_oracle (with inverted meaning)
  2258     z3_trace_assms ~> smt_trace_used_facts
  2259     INCOMPATIBILITY.
  2260   - Added:
  2261     smt_verbose
  2262     smt_random_seed
  2263     smt_datatypes
  2264     smt_infer_triggers
  2265     smt_monomorph_limit
  2266     cvc3_options
  2267     remote_cvc3_options
  2268     remote_z3_options
  2269     yices_options
  2270 
  2271 * Boogie output files (.b2i files) need to be declared in the theory
  2272 header.
  2273 
  2274 * Simplification procedure "list_to_set_comprehension" rewrites list
  2275 comprehensions applied to List.set to set comprehensions.  Occasional
  2276 INCOMPATIBILITY, may be deactivated like this:
  2277 
  2278   declare [[simproc del: list_to_set_comprehension]]
  2279 
  2280 * Removed old version of primrec package.  INCOMPATIBILITY.
  2281 
  2282 * Removed simplifier congruence rule of "prod_case", as has for long
  2283 been the case with "split".  INCOMPATIBILITY.
  2284 
  2285 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2286 
  2287 * Removed [split_format ... and ... and ...] version of
  2288 [split_format].  Potential INCOMPATIBILITY.
  2289 
  2290 * Predicate "sorted" now defined inductively, with nice induction
  2291 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2292 
  2293 * Constant "contents" renamed to "the_elem", to free the generic name
  2294 contents for other uses.  INCOMPATIBILITY.
  2295 
  2296 * Renamed class eq and constant eq (for code generation) to class
  2297 equal and constant equal, plus renaming of related facts and various
  2298 tuning.  INCOMPATIBILITY.
  2299 
  2300 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2301 
  2302 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2303 INCOMPATIBILITY.
  2304 
  2305 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2306 avoid confusion with finite sets.  INCOMPATIBILITY.
  2307 
  2308 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2309 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2310 for congruent(2)).
  2311 
  2312 * Some previously unqualified names have been qualified:
  2313 
  2314   types
  2315     bool ~> HOL.bool
  2316     nat ~> Nat.nat
  2317 
  2318   constants
  2319     Trueprop ~> HOL.Trueprop
  2320     True ~> HOL.True
  2321     False ~> HOL.False
  2322     op & ~> HOL.conj
  2323     op | ~> HOL.disj
  2324     op --> ~> HOL.implies
  2325     op = ~> HOL.eq
  2326     Not ~> HOL.Not
  2327     The ~> HOL.The
  2328     All ~> HOL.All
  2329     Ex ~> HOL.Ex
  2330     Ex1 ~> HOL.Ex1
  2331     Let ~> HOL.Let
  2332     If ~> HOL.If
  2333     Ball ~> Set.Ball
  2334     Bex ~> Set.Bex
  2335     Suc ~> Nat.Suc
  2336     Pair ~> Product_Type.Pair
  2337     fst ~> Product_Type.fst
  2338     snd ~> Product_Type.snd
  2339     curry ~> Product_Type.curry
  2340     op : ~> Set.member
  2341     Collect ~> Set.Collect
  2342 
  2343 INCOMPATIBILITY.
  2344 
  2345 * More canonical naming convention for some fundamental definitions:
  2346 
  2347     bot_bool_eq ~> bot_bool_def
  2348     top_bool_eq ~> top_bool_def
  2349     inf_bool_eq ~> inf_bool_def
  2350     sup_bool_eq ~> sup_bool_def
  2351     bot_fun_eq  ~> bot_fun_def
  2352     top_fun_eq  ~> top_fun_def
  2353     inf_fun_eq  ~> inf_fun_def
  2354     sup_fun_eq  ~> sup_fun_def
  2355 
  2356 INCOMPATIBILITY.
  2357 
  2358 * More stylized fact names:
  2359 
  2360   expand_fun_eq ~> fun_eq_iff
  2361   expand_set_eq ~> set_eq_iff
  2362   set_ext       ~> set_eqI
  2363   nat_number    ~> eval_nat_numeral
  2364 
  2365 INCOMPATIBILITY.
  2366 
  2367 * Refactoring of code-generation specific operations in theory List:
  2368 
  2369   constants
  2370     null ~> List.null
  2371 
  2372   facts
  2373     mem_iff ~> member_def
  2374     null_empty ~> null_def
  2375 
  2376 INCOMPATIBILITY.  Note that these were not supposed to be used
  2377 regularly unless for striking reasons; their main purpose was code
  2378 generation.
  2379 
  2380 Various operations from the Haskell prelude are used for generating
  2381 Haskell code.
  2382 
  2383 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2384 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2385 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2386 
  2387 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2388 "*" and "+" respectively.  INCOMPATIBILITY.
  2389 
  2390 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2391 "Sum_Type.Plus" instead.
  2392 
  2393 * Constant "split" has been merged with constant "prod_case"; names of
  2394 ML functions, facts etc. involving split have been retained so far,
  2395 though.  INCOMPATIBILITY.
  2396 
  2397 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2398 instead.  INCOMPATIBILITY.
  2399 
  2400 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2401 INCOMPATIBILITY.
  2402 
  2403 * Former theory Library/Enum is now part of the HOL-Main image.
  2404 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2405 referred to by its qualified name.
  2406 
  2407   enum    ~>  Enum.enum
  2408   nlists  ~>  Enum.nlists
  2409   product ~>  Enum.product
  2410 
  2411 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2412 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2413 INCOMPATIBILITY.
  2414 
  2415 * Theory Library/SetsAndFunctions has been split into
  2416 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2417 for instance definitions for functions; various improvements.
  2418 INCOMPATIBILITY.
  2419 
  2420 * Theory Library/Multiset provides stable quicksort implementation of
  2421 sort_key.
  2422 
  2423 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2424 INCOMPATIBILITY.
  2425 
  2426 * Session Multivariate_Analysis: introduced a type class for euclidean
  2427 space.  Most theorems are now stated in terms of euclidean spaces
  2428 instead of finite cartesian products.
  2429 
  2430   types
  2431     real ^ 'n ~>  'a::real_vector
  2432               ~>  'a::euclidean_space
  2433               ~>  'a::ordered_euclidean_space
  2434         (depends on your needs)
  2435 
  2436   constants
  2437      _ $ _        ~> _ $$ _
  2438      \<chi> x. _  ~> \<chi>\<chi> x. _
  2439      CARD('n)     ~> DIM('a)
  2440 
  2441 Also note that the indices are now natural numbers and not from some
  2442 finite type. Finite cartesian products of euclidean spaces, products
  2443 of euclidean spaces the real and complex numbers are instantiated to
  2444 be euclidean_spaces.  INCOMPATIBILITY.
  2445 
  2446 * Session Probability: introduced pextreal as positive extended real
  2447 numbers.  Use pextreal as value for measures.  Introduce the
  2448 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2449 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2450 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2451 
  2452 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2453 INCOMPATIBILITY.
  2454 
  2455 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2456 prove verification conditions generated by the SPARK Ada program
  2457 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2458 
  2459 
  2460 *** HOL-Algebra ***
  2461 
  2462 * Theorems for additive ring operations (locale abelian_monoid and
  2463 descendants) are generated by interpretation from their multiplicative
  2464 counterparts.  Names (in particular theorem names) have the mandatory
  2465 qualifier 'add'.  Previous theorem names are redeclared for
  2466 compatibility.
  2467 
  2468 * Structure "int_ring" is now an abbreviation (previously a
  2469 definition).  This fits more natural with advanced interpretations.
  2470 
  2471 
  2472 *** HOLCF ***
  2473 
  2474 * The domain package now runs in definitional mode by default: The
  2475 former command 'new_domain' is now called 'domain'.  To use the domain
  2476 package in its original axiomatic mode, use 'domain (unsafe)'.
  2477 INCOMPATIBILITY.
  2478 
  2479 * The new class "domain" is now the default sort.  Class "predomain"
  2480 is an unpointed version of "domain". Theories can be updated by
  2481 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2482 
  2483   'a::type ~> 'a::countable
  2484   'a::cpo  ~> 'a::predomain
  2485   'a::pcpo ~> 'a::domain
  2486 
  2487 * The old type class "rep" has been superseded by class "domain".
  2488 Accordingly, users of the definitional package must remove any
  2489 "default_sort rep" declarations.  INCOMPATIBILITY.
  2490 
  2491 * The domain package (definitional mode) now supports unpointed
  2492 predomain argument types, as long as they are marked 'lazy'. (Strict
  2493 arguments must be in class "domain".) For example, the following
  2494 domain definition now works:
  2495 
  2496   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2497 
  2498 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2499 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2500 'a option, and 'a list.  Additionally, it configures fixrec and the
  2501 domain package to work with these types.  For example:
  2502 
  2503   fixrec isInl :: "('a + 'b) u -> tr"
  2504     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2505 
  2506   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2507 
  2508 * The "(permissive)" option of fixrec has been replaced with a
  2509 per-equation "(unchecked)" option. See
  2510 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2511 
  2512 * The "bifinite" class no longer fixes a constant "approx"; the class
  2513 now just asserts that such a function exists.  INCOMPATIBILITY.
  2514 
  2515 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2516 defines an embedding of type 'a defl into udom by default; instances
  2517 of "bifinite" and "domain" classes are available in
  2518 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2519 
  2520 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2521 
  2522 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2523 
  2524 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2525 
  2526 * The function "cprod_map" has been renamed to "prod_map".
  2527 INCOMPATIBILITY.
  2528 
  2529 * The monadic bind operator on each powerdomain has new binder syntax
  2530 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2531 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2532 
  2533 * The infix syntax for binary union on each powerdomain has changed
  2534 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2535 syntax.  INCOMPATIBILITY.
  2536 
  2537 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2538 still supported as an input translation.
  2539 
  2540 * Renamed some theorems (the original names are also still available).
  2541 
  2542   expand_fun_below   ~> fun_below_iff
  2543   below_fun_ext      ~> fun_belowI
  2544   expand_cfun_eq     ~> cfun_eq_iff
  2545   ext_cfun           ~> cfun_eqI
  2546   expand_cfun_below  ~> cfun_below_iff
  2547   below_cfun_ext     ~> cfun_belowI
  2548   cont2cont_Rep_CFun ~> cont2cont_APP
  2549 
  2550 * The Abs and Rep functions for various types have changed names.
  2551 Related theorem names have also changed to match. INCOMPATIBILITY.
  2552 
  2553   Rep_CFun  ~> Rep_cfun
  2554   Abs_CFun  ~> Abs_cfun
  2555   Rep_Sprod ~> Rep_sprod
  2556   Abs_Sprod ~> Abs_sprod
  2557   Rep_Ssum  ~> Rep_ssum
  2558   Abs_Ssum  ~> Abs_ssum
  2559 
  2560 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2561 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2562 
  2563 * Various changes to bisimulation/coinduction with domain package:
  2564 
  2565   - Definitions of "bisim" constants no longer mention definedness.
  2566   - With mutual recursion, "bisim" predicate is now curried.
  2567   - With mutual recursion, each type gets a separate coind theorem.
  2568   - Variable names in bisim_def and coinduct rules have changed.
  2569 
  2570 INCOMPATIBILITY.
  2571 
  2572 * Case combinators generated by the domain package for type "foo" are
  2573 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2574 
  2575 * Several theorems have been renamed to more accurately reflect the
  2576 names of constants and types involved.  INCOMPATIBILITY.
  2577 
  2578   thelub_const    ~> lub_const
  2579   lub_const       ~> is_lub_const
  2580   thelubI         ~> lub_eqI
  2581   is_lub_lub      ~> is_lubD2
  2582   lubI            ~> is_lub_lub
  2583   unique_lub      ~> is_lub_unique
  2584   is_ub_lub       ~> is_lub_rangeD1
  2585   lub_bin_chain   ~> is_lub_bin_chain
  2586   lub_fun         ~> is_lub_fun
  2587   thelub_fun      ~> lub_fun
  2588   thelub_cfun     ~> lub_cfun
  2589   thelub_Pair     ~> lub_Pair
  2590   lub_cprod       ~> is_lub_prod
  2591   thelub_cprod    ~> lub_prod
  2592   minimal_cprod   ~> minimal_prod
  2593   inst_cprod_pcpo ~> inst_prod_pcpo
  2594   UU_I            ~> bottomI
  2595   compact_UU      ~> compact_bottom
  2596   deflation_UU    ~> deflation_bottom
  2597   finite_deflation_UU ~> finite_deflation_bottom
  2598 
  2599 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2600 
  2601   sq_ord_less_eq_trans ~> below_eq_trans
  2602   sq_ord_eq_less_trans ~> eq_below_trans
  2603   refl_less            ~> below_refl
  2604   trans_less           ~> below_trans
  2605   antisym_less         ~> below_antisym
  2606   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2607   box_less             ~> box_below
  2608   rev_trans_less       ~> rev_below_trans
  2609   not_less2not_eq      ~> not_below2not_eq
  2610   less_UU_iff          ~> below_UU_iff
  2611   flat_less_iff        ~> flat_below_iff
  2612   adm_less             ~> adm_below
  2613   adm_not_less         ~> adm_not_below
  2614   adm_compact_not_less ~> adm_compact_not_below
  2615   less_fun_def         ~> below_fun_def
  2616   expand_fun_less      ~> fun_below_iff
  2617   less_fun_ext         ~> fun_belowI
  2618   less_discr_def       ~> below_discr_def
  2619   discr_less_eq        ~> discr_below_eq
  2620   less_unit_def        ~> below_unit_def
  2621   less_cprod_def       ~> below_prod_def
  2622   prod_lessI           ~> prod_belowI
  2623   Pair_less_iff        ~> Pair_below_iff
  2624   fst_less_iff         ~> fst_below_iff
  2625   snd_less_iff         ~> snd_below_iff
  2626   expand_cfun_less     ~> cfun_below_iff
  2627   less_cfun_ext        ~> cfun_belowI
  2628   injection_less       ~> injection_below
  2629   less_up_def          ~> below_up_def
  2630   not_Iup_less         ~> not_Iup_below
  2631   Iup_less             ~> Iup_below
  2632   up_less              ~> up_below
  2633   Def_inject_less_eq   ~> Def_below_Def
  2634   Def_less_is_eq       ~> Def_below_iff
  2635   spair_less_iff       ~> spair_below_iff
  2636   less_sprod           ~> below_sprod
  2637   spair_less           ~> spair_below
  2638   sfst_less_iff        ~> sfst_below_iff
  2639   ssnd_less_iff        ~> ssnd_below_iff
  2640   fix_least_less       ~> fix_least_below
  2641   dist_less_one        ~> dist_below_one
  2642   less_ONE             ~> below_ONE
  2643   ONE_less_iff         ~> ONE_below_iff
  2644   less_sinlD           ~> below_sinlD
  2645   less_sinrD           ~> below_sinrD
  2646 
  2647 
  2648 *** FOL and ZF ***
  2649 
  2650 * All constant names are now qualified internally and use proper
  2651 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2652 
  2653 
  2654 *** ML ***
  2655 
  2656 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2657 Fail if the argument is false.  Due to inlining the source position of
  2658 failed assertions is included in the error output.
  2659 
  2660 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2661 text is in practice always evaluated with a stable theory checkpoint.
  2662 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2663 
  2664 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2665 the current context, not any accidental theory loader state as before.
  2666 Potential INCOMPATIBILITY, subtle change in semantics.
  2667 
  2668 * Syntax.pretty_priority (default 0) configures the required priority
  2669 of pretty-printed output and thus affects insertion of parentheses.
  2670 
  2671 * Syntax.default_root (default "any") configures the inner syntax
  2672 category (nonterminal symbol) for parsing of terms.
  2673 
  2674 * Former exception Library.UnequalLengths now coincides with
  2675 ListPair.UnequalLengths.
  2676 
  2677 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2678 main functionality is provided by structure Simplifier.
  2679 
  2680 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2681 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2682 almost all situations.
  2683 
  2684 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2685 See implementation manual for further details on exceptions in
  2686 Isabelle/ML.
  2687 
  2688 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2689 meaning.
  2690 
  2691 * Renamed structure PureThy to Pure_Thy and moved most of its
  2692 operations to structure Global_Theory, to emphasize that this is
  2693 rarely-used global-only stuff.
  2694 
  2695 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2696 instead (or tracing for high-volume output).
  2697 
  2698 * Configuration option show_question_marks only affects regular pretty
  2699 printing of types and terms, not raw Term.string_of_vname.
  2700 
  2701 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2702 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2703 @{thms} for most purposes.
  2704 
  2705 * ML structure Unsynchronized is never opened, not even in Isar
  2706 interaction mode as before.  Old Unsynchronized.set etc. have been
  2707 discontinued -- use plain := instead.  This should be *rare* anyway,
  2708 since modern tools always work via official context data, notably
  2709 configuration options.
  2710 
  2711 * Parallel and asynchronous execution requires special care concerning
  2712 interrupts.  Structure Exn provides some convenience functions that
  2713 avoid working directly with raw Interrupt.  User code must not absorb
  2714 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2715 followed by re-raising of the original exception.  Another common
  2716 source of mistakes are "handle _" patterns, which make the meaning of
  2717 the program subject to physical effects of the environment.
  2718 
  2719 
  2720 
  2721 New in Isabelle2009-2 (June 2010)
  2722 ---------------------------------
  2723 
  2724 *** General ***
  2725 
  2726 * Authentic syntax for *all* logical entities (type classes, type
  2727 constructors, term constants): provides simple and robust
  2728 correspondence between formal entities and concrete syntax.  Within
  2729 the parse tree / AST representations, "constants" are decorated by
  2730 their category (class, type, const) and spelled out explicitly with
  2731 their full internal name.
  2732 
  2733 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2734 and translations (translation rules and translation functions in ML).
  2735 Some hints on upgrading:
  2736 
  2737   - Many existing uses of 'syntax' and 'translations' can be replaced
  2738     by more modern 'type_notation', 'notation' and 'abbreviation',
  2739     which are independent of this issue.
  2740 
  2741   - 'translations' require markup within the AST; the term syntax
  2742     provides the following special forms:
  2743 
  2744       CONST c   -- produces syntax version of constant c from context
  2745       XCONST c  -- literally c, checked as constant from context
  2746       c         -- literally c, if declared by 'syntax'
  2747 
  2748     Plain identifiers are treated as AST variables -- occasionally the
  2749     system indicates accidental variables via the error "rhs contains
  2750     extra variables".
  2751 
  2752     Type classes and type constructors are marked according to their
  2753     concrete syntax.  Some old translations rules need to be written
  2754     for the "type" category, using type constructor application
  2755     instead of pseudo-term application of the default category
  2756     "logic".
  2757 
  2758   - 'parse_translation' etc. in ML may use the following
  2759     antiquotations:
  2760 
  2761       @{class_syntax c}   -- type class c within parse tree / AST
  2762       @{term_syntax c}    -- type constructor c within parse tree / AST
  2763       @{const_syntax c}   -- ML version of "CONST c" above
  2764       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2765 
  2766   - Literal types within 'typed_print_translations', i.e. those *not*
  2767     represented as pseudo-terms are represented verbatim.  Use @{class
  2768     c} or @{type_name c} here instead of the above syntax
  2769     antiquotations.
  2770 
  2771 Note that old non-authentic syntax was based on unqualified base
  2772 names, so all of the above "constant" names would coincide.  Recall
  2773 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2774 diagnose syntax problems.
  2775 
  2776 * Type constructors admit general mixfix syntax, not just infix.
  2777 
  2778 * Concrete syntax may be attached to local entities without a proof
  2779 body, too.  This works via regular mixfix annotations for 'fix',
  2780 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2781 similar to the 'notation' command in theory specifications.
  2782 
  2783 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2784 need to specify constant name and syntax separately.  Internal ML
  2785 datatype constructors have been renamed from InfixName to Infix etc.
  2786 Minor INCOMPATIBILITY.
  2787 
  2788 * Schematic theorem statements need to be explicitly markup as such,
  2789 via commands 'schematic_lemma', 'schematic_theorem',
  2790 'schematic_corollary'.  Thus the relevance of the proof is made
  2791 syntactically clear, which impacts performance in a parallel or
  2792 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2793 
  2794 * Use of cumulative prems via "!" in some proof methods has been
  2795 discontinued (old legacy feature).
  2796 
  2797 * References 'trace_simp' and 'debug_simp' have been replaced by
  2798 configuration options stored in the context. Enabling tracing (the
  2799 case of debugging is similar) in proofs works via
  2800 
  2801   using [[trace_simp = true]]
  2802 
  2803 Tracing is then active for all invocations of the simplifier in
  2804 subsequent goal refinement steps. Tracing may also still be enabled or
  2805 disabled via the ProofGeneral settings menu.
  2806 
  2807 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2808 'hide_fact' replace the former 'hide' KIND command.  Minor
  2809 INCOMPATIBILITY.
  2810 
  2811 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2812 more efficient than combinations with -q1 or -q2.
  2813 
  2814 
  2815 *** Pure ***
  2816 
  2817 * Proofterms record type-class reasoning explicitly, using the
  2818 "unconstrain" operation internally.  This eliminates all sort
  2819 constraints from a theorem and proof, introducing explicit
  2820 OFCLASS-premises.  On the proof term level, this operation is
  2821 automatically applied at theorem boundaries, such that closed proofs
  2822 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2823 inspect proof terms.
  2824 
  2825 * Local theory specifications may depend on extra type variables that
  2826 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2827 are added internally.  For example:
  2828 
  2829   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2830 
  2831 * Predicates of locales introduced by classes carry a mandatory
  2832 "class" prefix.  INCOMPATIBILITY.
  2833 
  2834 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2835 
  2836 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2837 'class' instead.
  2838 
  2839 * Command 'code_reflect' allows to incorporate generated ML code into
  2840 runtime environment; replaces immature code_datatype antiquotation.
  2841 INCOMPATIBILITY.
  2842 
  2843 * Code generator: simple concept for abstract datatypes obeying
  2844 invariants.
  2845 
  2846 * Code generator: details of internal data cache have no impact on the
  2847 user space functionality any longer.
  2848 
  2849 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2850 subgoals.  This is more appropriate for interpretations with 'where'.
  2851 INCOMPATIBILITY.
  2852 
  2853 * Command 'example_proof' opens an empty proof body.  This allows to
  2854 experiment with Isar, without producing any persistent result.
  2855 
  2856 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2857 within a local theory context, with explicit checking of the
  2858 constructors involved (in contrast to the raw 'syntax' versions).
  2859 
  2860 * Commands 'types' and 'typedecl' now work within a local theory
  2861 context -- without introducing dependencies on parameters or
  2862 assumptions, which is not possible in Isabelle/Pure.
  2863 
  2864 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2865 within a local theory context.  Minor INCOMPATIBILITY.
  2866 
  2867 
  2868 *** HOL ***
  2869 
  2870 * Command 'typedef' now works within a local theory context -- without
  2871 introducing dependencies on parameters or assumptions, which is not
  2872 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2873 contain multiple interpretations of local typedefs (with different
  2874 non-emptiness proofs), even in a global theory context.
  2875 
  2876 * New package for quotient types.  Commands 'quotient_type' and
  2877 'quotient_definition' may be used for defining types and constants by
  2878 quotient constructions.  An example is the type of integers created by
  2879 quotienting pairs of natural numbers:
  2880 
  2881   fun
  2882     intrel :: "(nat * nat) => (nat * nat) => bool"
  2883   where
  2884     "intrel (x, y) (u, v) = (x + v = u + y)"
  2885 
  2886   quotient_type int = "nat * nat" / intrel
  2887     by (auto simp add: equivp_def expand_fun_eq)
  2888 
  2889   quotient_definition
  2890     "0::int" is "(0::nat, 0::nat)"
  2891 
  2892 The method "lifting" can be used to lift of theorems from the
  2893 underlying "raw" type to the quotient type.  The example
  2894 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2895 construction and provides a reasoning infrastructure for finite sets.
  2896 
  2897 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2898 clash with new theory Quotient in Main HOL.
  2899 
  2900 * Moved the SMT binding into the main HOL session, eliminating
  2901 separate HOL-SMT session.
  2902 
  2903 * List membership infix mem operation is only an input abbreviation.
  2904 INCOMPATIBILITY.
  2905 
  2906 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2907 for future developements; former Library/Word.thy is still present in
  2908 the AFP entry RSAPPS.
  2909 
  2910 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2911 longer shadowed.  INCOMPATIBILITY.
  2912 
  2913 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2914 INCOMPATIBILITY.
  2915 
  2916 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2917 INCOMPATIBILITY.
  2918 
  2919 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2920 instead.  INCOMPATIBILITY.
  2921 
  2922 * Dropped several real-specific versions of lemmas about floor and
  2923 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2924 instead.  INCOMPATIBILITY.
  2925 
  2926   floor_number_of_eq         ~> floor_number_of
  2927   le_floor_eq_number_of      ~> number_of_le_floor
  2928   le_floor_eq_zero           ~> zero_le_floor
  2929   le_floor_eq_one            ~> one_le_floor
  2930   floor_less_eq_number_of    ~> floor_less_number_of
  2931   floor_less_eq_zero         ~> floor_less_zero
  2932   floor_less_eq_one          ~> floor_less_one
  2933   less_floor_eq_number_of    ~> number_of_less_floor
  2934   less_floor_eq_zero         ~> zero_less_floor
  2935   less_floor_eq_one          ~> one_less_floor
  2936   floor_le_eq_number_of      ~> floor_le_number_of
  2937   floor_le_eq_zero           ~> floor_le_zero
  2938   floor_le_eq_one            ~> floor_le_one
  2939   floor_subtract_number_of   ~> floor_diff_number_of
  2940   floor_subtract_one         ~> floor_diff_one
  2941   ceiling_number_of_eq       ~> ceiling_number_of
  2942   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2943   ceiling_le_zero_eq         ~> ceiling_le_zero
  2944   ceiling_le_eq_one          ~> ceiling_le_one
  2945   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2946   less_ceiling_eq_zero       ~> zero_less_ceiling
  2947   less_ceiling_eq_one        ~> one_less_ceiling
  2948   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2949   ceiling_less_eq_zero       ~> ceiling_less_zero
  2950   ceiling_less_eq_one        ~> ceiling_less_one
  2951   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2952   le_ceiling_eq_zero         ~> zero_le_ceiling
  2953   le_ceiling_eq_one          ~> one_le_ceiling
  2954   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2955   ceiling_subtract_one       ~> ceiling_diff_one
  2956 
  2957 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2958 application of the various fold combinators on finite sets.
  2959 
  2960 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2961 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2962 implementation.  INCOMPATIBILTY.
  2963 
  2964 * Theory Library/Coinductive_List has been removed -- superseded by
  2965 AFP/thys/Coinductive.
  2966 
  2967 * Theory PReal, including the type "preal" and related operations, has
  2968 been removed.  INCOMPATIBILITY.
  2969 
  2970 * Real: new development using Cauchy Sequences.
  2971 
  2972 * Split off theory "Big_Operators" containing setsum, setprod,
  2973 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2974 
  2975 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2976 "Int" etc.  INCOMPATIBILITY.
  2977 
  2978 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2979 
  2980 * New set of rules "ac_simps" provides combined assoc / commute
  2981 rewrites for all interpretations of the appropriate generic locales.
  2982 
  2983 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2984 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2985 appropriate and more consistent names suitable for name prefixes
  2986 within the HOL theories.  INCOMPATIBILITY.
  2987 
  2988 * Some generic constants have been put to appropriate theories:
  2989   - less_eq, less: Orderings
  2990   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2991   - inverse, divide: Rings
  2992 INCOMPATIBILITY.
  2993 
  2994 * More consistent naming of type classes involving orderings (and
  2995 lattices):
  2996 
  2997     lower_semilattice                   ~> semilattice_inf
  2998     upper_semilattice                   ~> semilattice_sup
  2999 
  3000     dense_linear_order                  ~> dense_linorder
  3001 
  3002     pordered_ab_group_add               ~> ordered_ab_group_add
  3003     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  3004     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  3005     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  3006     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  3007     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  3008     pordered_cancel_semiring            ~> ordered_cancel_semiring
  3009     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  3010     pordered_comm_ring                  ~> ordered_comm_ring
  3011     pordered_comm_semiring              ~> ordered_comm_semiring
  3012     pordered_ring                       ~> ordered_ring
  3013     pordered_ring_abs                   ~> ordered_ring_abs
  3014     pordered_semiring                   ~> ordered_semiring
  3015 
  3016     ordered_ab_group_add                ~> linordered_ab_group_add
  3017     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  3018     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  3019     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  3020     ordered_field                       ~> linordered_field
  3021     ordered_field_no_lb                 ~> linordered_field_no_lb
  3022     ordered_field_no_ub                 ~> linordered_field_no_ub
  3023     ordered_field_dense_linear_order    ~> dense_linordered_field
  3024     ordered_idom                        ~> linordered_idom
  3025     ordered_ring                        ~> linordered_ring
  3026     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  3027     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  3028     ordered_ring_strict                 ~> linordered_ring_strict
  3029     ordered_semidom                     ~> linordered_semidom
  3030     ordered_semiring                    ~> linordered_semiring
  3031     ordered_semiring_1                  ~> linordered_semiring_1
  3032     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  3033     ordered_semiring_strict             ~> linordered_semiring_strict
  3034 
  3035   The following slightly odd type classes have been moved to a
  3036   separate theory Library/Lattice_Algebras:
  3037 
  3038     lordered_ab_group_add               ~> lattice_ab_group_add
  3039     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  3040     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  3041     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  3042     lordered_ring                       ~> lattice_ring
  3043 
  3044 INCOMPATIBILITY.
  3045 
  3046 * Refined field classes:
  3047   - classes division_ring_inverse_zero, field_inverse_zero,
  3048     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3049     subsumes former division_by_zero class;
  3050   - numerous lemmas have been ported from field to division_ring.
  3051 INCOMPATIBILITY.
  3052 
  3053 * Refined algebra theorem collections:
  3054   - dropped theorem group group_simps, use algebra_simps instead;
  3055   - dropped theorem group ring_simps, use field_simps instead;
  3056   - proper theorem collection field_simps subsumes former theorem
  3057     groups field_eq_simps and field_simps;
  3058   - dropped lemma eq_minus_self_iff which is a duplicate for
  3059     equal_neg_zero.
  3060 INCOMPATIBILITY.
  3061 
  3062 * Theory Finite_Set and List: some lemmas have been generalized from
  3063 sets to lattices:
  3064 
  3065   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3066   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3067   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3068   union_Union_fold_union        ~> sup_Sup_fold_sup
  3069   Inter_fold_inter              ~> Inf_fold_inf
  3070   Union_fold_union              ~> Sup_fold_sup
  3071   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3072   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3073   INTER_fold_inter              ~> INFI_fold_inf
  3074   UNION_fold_union              ~> SUPR_fold_sup
  3075 
  3076 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3077 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3078 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3079 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3080 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3081 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3082 
  3083 * Reorganized theory Multiset: swapped notation of pointwise and
  3084 multiset order:
  3085 
  3086   - pointwise ordering is instance of class order with standard syntax
  3087     <= and <;
  3088   - multiset ordering has syntax <=# and <#; partial order properties
  3089     are provided by means of interpretation with prefix
  3090     multiset_order;
  3091   - less duplication, less historical organization of sections,
  3092     conversion from associations lists to multisets, rudimentary code
  3093     generation;
  3094   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3095     if needed.
  3096 
  3097 Renamed:
  3098 
  3099   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3100   multi_count_ext  ~>  multiset_ext
  3101   diff_union_inverse2  ~>  diff_union_cancelR
  3102 
  3103 INCOMPATIBILITY.
  3104 
  3105 * Theory Permutation: replaced local "remove" by List.remove1.
  3106 
  3107 * Code generation: ML and OCaml code is decorated with signatures.
  3108 
  3109 * Theory List: added transpose.
  3110 
  3111 * Library/Nat_Bijection.thy is a collection of bijective functions
  3112 between nat and other types, which supersedes the older libraries
  3113 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3114 
  3115   Constants:
  3116   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3117   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3118   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3119   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3120   Countable.pair_encode           ~> prod_encode
  3121   NatIso.prod2nat                 ~> prod_encode
  3122   NatIso.nat2prod                 ~> prod_decode
  3123   NatIso.sum2nat                  ~> sum_encode
  3124   NatIso.nat2sum                  ~> sum_decode
  3125   NatIso.list2nat                 ~> list_encode
  3126   NatIso.nat2list                 ~> list_decode
  3127   NatIso.set2nat                  ~> set_encode
  3128   NatIso.nat2set                  ~> set_decode
  3129 
  3130   Lemmas:
  3131   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3132   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3133   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3134   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3135   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3136   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3137   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3138   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3139   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3140   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3141   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3142   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3143   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3144 
  3145 * Sledgehammer:
  3146   - Renamed ATP commands:
  3147     atp_info     ~> sledgehammer running_atps
  3148     atp_kill     ~> sledgehammer kill_atps
  3149     atp_messages ~> sledgehammer messages
  3150     atp_minimize ~> sledgehammer minimize
  3151     print_atps   ~> sledgehammer available_atps
  3152     INCOMPATIBILITY.
  3153   - Added user's manual ("isabelle doc sledgehammer").
  3154   - Added option syntax and "sledgehammer_params" to customize
  3155     Sledgehammer's behavior.  See the manual for details.
  3156   - Modified the Isar proof reconstruction code so that it produces
  3157     direct proofs rather than proofs by contradiction.  (This feature
  3158     is still experimental.)
  3159   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3160     full-typed mode.
  3161   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3162 
  3163 * Nitpick:
  3164   - Added and implemented "binary_ints" and "bits" options.
  3165   - Added "std" option and implemented support for nonstandard models.
  3166   - Added and implemented "finitize" option to improve the precision
  3167     of infinite datatypes based on a monotonicity analysis.
  3168   - Added support for quotient types.
  3169   - Added support for "specification" and "ax_specification"
  3170     constructs.
  3171   - Added support for local definitions (for "function" and
  3172     "termination" proofs).
  3173   - Added support for term postprocessors.
  3174   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3175   - Improved efficiency of "destroy_constrs" optimization.
  3176   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3177     record getters.
  3178   - Fixed soundness bug related to higher-order constructors.
  3179   - Fixed soundness bug when "full_descrs" is enabled.
  3180   - Improved precision of set constructs.
  3181   - Added "atoms" option.
  3182   - Added cache to speed up repeated Kodkod invocations on the same
  3183     problems.
  3184   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3185     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3186     "SAT4J_Light".  INCOMPATIBILITY.
  3187   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3188     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3189   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3190 
  3191 * Method "induct" now takes instantiations of the form t, where t is not
  3192   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3193   If this is not intended, t has to be enclosed in parentheses.
  3194   By default, the equalities generated by definitional instantiations
  3195   are pre-simplified, which may cause parameters of inductive cases
  3196   to disappear, or may even delete some of the inductive cases.
  3197   Use "induct (no_simp)" instead of "induct" to restore the old
  3198   behaviour. The (no_simp) option is also understood by the "cases"
  3199   and "nominal_induct" methods, which now perform pre-simplification, too.
  3200   INCOMPATIBILITY.
  3201 
  3202 
  3203 *** HOLCF ***
  3204 
  3205 * Variable names in lemmas generated by the domain package have
  3206 changed; the naming scheme is now consistent with the HOL datatype
  3207 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3208 
  3209 * The domain package no longer defines the function "foo_copy" for
  3210 recursive domain "foo".  The reach lemma is now stated directly in
  3211 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3212 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3213 
  3214 * Most definedness lemmas generated by the domain package (previously
  3215 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3216 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3217 Proofs that used definedness lemmas as intro rules may break,
  3218 potential INCOMPATIBILITY.
  3219 
  3220 * Induction and casedist rules generated by the domain package now
  3221 declare proper case_names (one called "bottom", and one named for each
  3222 constructor).  INCOMPATIBILITY.
  3223 
  3224 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3225 rules are generated for each domain, INCOMPATIBILITY.
  3226 
  3227   foo_bar.reach       ~> foo.reach  bar.reach
  3228   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3229 
  3230 * Some lemmas generated by the domain package have been renamed for
  3231 consistency with the datatype package, INCOMPATIBILITY.
  3232 
  3233   foo.ind        ~> foo.induct
  3234   foo.finite_ind ~> foo.finite_induct
  3235   foo.coind      ~> foo.coinduct
  3236   foo.casedist   ~> foo.exhaust
  3237   foo.exhaust    ~> foo.nchotomy
  3238 
  3239 * For consistency with other definition packages, the fixrec package
  3240 now generates qualified theorem names, INCOMPATIBILITY.
  3241 
  3242   foo_simps  ~> foo.simps
  3243   foo_unfold ~> foo.unfold
  3244   foo_induct ~> foo.induct
  3245 
  3246 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3247 method and internal fixrec proofs now use the default simpset instead.
  3248 INCOMPATIBILITY.
  3249 
  3250 * The "contlub" predicate has been removed.  Proof scripts should use
  3251 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3252 
  3253 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3254 
  3255 * The constants cpair, cfst, and csnd have been removed in favor of
  3256 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3257 
  3258 
  3259 *** ML ***
  3260 
  3261 * Antiquotations for basic formal entities:
  3262 
  3263     @{class NAME}         -- type class
  3264     @{class_syntax NAME}  -- syntax representation of the above
  3265 
  3266     @{type_name NAME}     -- logical type
  3267     @{type_abbrev NAME}   -- type abbreviation
  3268     @{nonterminal NAME}   -- type of concrete syntactic category
  3269     @{type_syntax NAME}   -- syntax representation of any of the above
  3270 
  3271     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3272     @{const_abbrev NAME}  -- abbreviated constant
  3273     @{const_syntax NAME}  -- syntax representation of any of the above
  3274 
  3275 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3276 syntax constant (cf. 'syntax' command).
  3277 
  3278 * Antiquotation @{make_string} inlines a function to print arbitrary
  3279 values similar to the ML toplevel.  The result is compiler dependent
  3280 and may fall back on "?" in certain situations.
  3281 
  3282 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3283 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3284 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3285 not work with the asynchronous Isar document model.
  3286 
  3287 * Configuration options now admit dynamic default values, depending on
  3288 the context or even global references.
  3289 
  3290 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3291 uses an efficient external library if available (for Poly/ML).
  3292 
  3293 * Renamed some important ML structures, while keeping the old names
  3294 for some time as aliases within the structure Legacy:
  3295 
  3296   OuterKeyword  ~>  Keyword
  3297   OuterLex      ~>  Token
  3298   OuterParse    ~>  Parse
  3299   OuterSyntax   ~>  Outer_Syntax
  3300   PrintMode     ~>  Print_Mode
  3301   SpecParse     ~>  Parse_Spec
  3302   ThyInfo       ~>  Thy_Info
  3303   ThyLoad       ~>  Thy_Load
  3304   ThyOutput     ~>  Thy_Output
  3305   TypeInfer     ~>  Type_Infer
  3306 
  3307 Note that "open Legacy" simplifies porting of sources, but forgetting
  3308 to remove it again will complicate porting again in the future.
  3309 
  3310 * Most operations that refer to a global context are named
  3311 accordingly, e.g. Simplifier.global_context or
  3312 ProofContext.init_global.  There are some situations where a global
  3313 context actually works, but under normal circumstances one needs to
  3314 pass the proper local context through the code!
  3315 
  3316 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3317 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3318 Pretty.pp argument to merge, which is absent in the standard
  3319 Theory_Data version.
  3320 
  3321 * Sorts.certify_sort and derived "cert" operations for types and terms
  3322 no longer minimize sorts.  Thus certification at the boundary of the
  3323 inference kernel becomes invariant under addition of class relations,
  3324 which is an important monotonicity principle.  Sorts are now minimized
  3325 in the syntax layer only, at the boundary between the end-user and the
  3326 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3327 explicitly in rare situations.
  3328 
  3329 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3330 emphasize that this is in no way a standard operation.
  3331 INCOMPATIBILITY.
  3332 
  3333 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3334 usual for resolution.  Rare INCOMPATIBILITY.
  3335 
  3336 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3337 to emphasize that these only work in a global situation (which is
  3338 quite rare).
  3339 
  3340 * Curried take and drop in library.ML; negative length is interpreted
  3341 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3342 
  3343 * Proof terms: type substitutions on proof constants now use canonical
  3344 order of type variables.  INCOMPATIBILITY for tools working with proof
  3345 terms.
  3346 
  3347 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3348 may no longer carry premises.  User-level specifications are
  3349 transformed accordingly by Thm.add_axiom/add_def.
  3350 
  3351 
  3352 *** System ***
  3353 
  3354 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3355 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3356 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3357 
  3358 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3359 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3360 respectively.
  3361 
  3362 * Old lib/scripts/polyml-platform is superseded by the
  3363 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3364 variant, even on a 64 bit machine.  The following example setting
  3365 prefers 64 bit if available:
  3366 
  3367   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3368 
  3369 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3370 Isabelle/Scala layer for advanced prover interaction and integration.
  3371 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3372 component.
  3373 
  3374 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3375 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3376 similar to the default assignment of the document preparation system
  3377 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3378 provides some operations for direct access to the font without asking
  3379 the user for manual installation.
  3380 
  3381 
  3382 
  3383 New in Isabelle2009-1 (December 2009)
  3384 -------------------------------------
  3385 
  3386 *** General ***
  3387 
  3388 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3389 one backslash should be used, even in ML sources.
  3390 
  3391 
  3392 *** Pure ***
  3393 
  3394 * Locale interpretation propagates mixins along the locale hierarchy.
  3395 The currently only available mixins are the equations used to map
  3396 local definitions to terms of the target domain of an interpretation.
  3397 
  3398 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3399 loc" to print all interpretations of locale "loc" in the theory.
  3400 Interpretations in proofs are not shown.
  3401 
  3402 * Thoroughly revised locales tutorial.  New section on conditional
  3403 interpretation.
  3404 
  3405 * On instantiation of classes, remaining undefined class parameters
  3406 are formally declared.  INCOMPATIBILITY.
  3407 
  3408 
  3409 *** Document preparation ***
  3410 
  3411 * New generalized style concept for printing terms: @{foo (style) ...}
  3412 instead of @{foo_style style ...}  (old form is still retained for
  3413 backward compatibility).  Styles can be also applied for
  3414 antiquotations prop, term_type and typeof.
  3415 
  3416 
  3417 *** HOL ***
  3418 
  3419 * New proof method "smt" for a combination of first-order logic with
  3420 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3421 fixed-size bitvectors; there is also basic support for higher-order
  3422 features (esp. lambda abstractions).  It is an incomplete decision
  3423 procedure based on external SMT solvers using the oracle mechanism;
  3424 for the SMT solver Z3, this method is proof-producing.  Certificates
  3425 are provided to avoid calling the external solvers solely for
  3426 re-checking proofs.  Due to a remote SMT service there is no need for
  3427 installing SMT solvers locally.  See src/HOL/SMT.
  3428 
  3429 * New commands to load and prove verification conditions generated by
  3430 the Boogie program verifier or derived systems (e.g. the Verifying C
  3431 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3432 
  3433 * New counterexample generator tool 'nitpick' based on the Kodkod
  3434 relational model finder.  See src/HOL/Tools/Nitpick and
  3435 src/HOL/Nitpick_Examples.
  3436 
  3437 * New commands 'code_pred' and 'values' to invoke the predicate
  3438 compiler and to enumerate values of inductive predicates.
  3439 
  3440 * A tabled implementation of the reflexive transitive closure.
  3441 
  3442 * New implementation of quickcheck uses generic code generator;
  3443 default generators are provided for all suitable HOL types, records
  3444 and datatypes.  Old quickcheck can be re-activated importing theory
  3445 Library/SML_Quickcheck.
  3446 
  3447 * New testing tool Mirabelle for automated proof tools.  Applies
  3448 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3449 every proof step in a theory.  To be used in batch mode via the
  3450 "mirabelle" utility.
  3451 
  3452 * New proof method "sos" (sum of squares) for nonlinear real
  3453 arithmetic (originally due to John Harison). It requires theory
  3454 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3455 works well in practice on quantifier-free real arithmetic with +, -,
  3456 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3457 inequalities between polynomials.  It makes use of external
  3458 semidefinite programming solvers.  Method "sos" generates a
  3459 certificate that can be pasted into the proof thus avoiding the need
  3460 to call an external tool every time the proof is checked.  See
  3461 src/HOL/Library/Sum_Of_Squares.
  3462 
  3463 * New method "linarith" invokes existing linear arithmetic decision
  3464 procedure only.
  3465 
  3466 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3467 
  3468 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3469 Causes full type information to be output to the ATPs.  This slows
  3470 ATPs down considerably but eliminates a source of unsound "proofs"
  3471 that fail later.
  3472 
  3473 * New method "metisFT": A version of metis that uses full type
  3474 information in order to avoid failures of proof reconstruction.
  3475 
  3476 * New evaluator "approximate" approximates an real valued term using
  3477 the same method as the approximation method.
  3478 
  3479 * Method "approximate" now supports arithmetic expressions as
  3480 boundaries of intervals and implements interval splitting and Taylor
  3481 series expansion.
  3482 
  3483 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3484 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3485 
  3486 * New theory SupInf of the supremum and infimum operators for sets of
  3487 reals.
  3488 
  3489 * New theory Probability, which contains a development of measure
  3490 theory, eventually leading to Lebesgue integration and probability.
  3491 
  3492 * Extended Multivariate Analysis to include derivation and Brouwer's
  3493 fixpoint theorem.
  3494 
  3495 * Reorganization of number theory, INCOMPATIBILITY:
  3496   - new number theory development for nat and int, in theories Divides
  3497     and GCD as well as in new session Number_Theory
  3498   - some constants and facts now suffixed with _nat and _int
  3499     accordingly
  3500   - former session NumberTheory now named Old_Number_Theory, including
  3501     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3502   - moved theory Pocklington from src/HOL/Library to
  3503     src/HOL/Old_Number_Theory
  3504 
  3505 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3506 lcm of finite and infinite sets. It is shown that they form a complete
  3507 lattice.
  3508 
  3509 * Class semiring_div requires superclass no_zero_divisors and proof of
  3510 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3511 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3512 generalized to class semiring_div, subsuming former theorems
  3513 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3514 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3515 INCOMPATIBILITY.
  3516 
  3517 * Refinements to lattice classes and sets:
  3518   - less default intro/elim rules in locale variant, more default
  3519     intro/elim rules in class variant: more uniformity
  3520   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3521     le_inf_iff
  3522   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3523     sup_aci)
  3524   - renamed ACI to inf_sup_aci
  3525   - new class "boolean_algebra"
  3526   - class "complete_lattice" moved to separate theory
  3527     "Complete_Lattice"; corresponding constants (and abbreviations)
  3528     renamed and with authentic syntax:
  3529     Set.Inf ~>    Complete_Lattice.Inf
  3530     Set.Sup ~>    Complete_Lattice.Sup
  3531     Set.INFI ~>   Complete_Lattice.INFI
  3532     Set.SUPR ~>   Complete_Lattice.SUPR
  3533     Set.Inter ~>  Complete_Lattice.Inter
  3534     Set.Union ~>  Complete_Lattice.Union
  3535     Set.INTER ~>  Complete_Lattice.INTER
  3536     Set.UNION ~>  Complete_Lattice.UNION
  3537   - authentic syntax for
  3538     Set.Pow
  3539     Set.image
  3540   - mere abbreviations:
  3541     Set.empty               (for bot)
  3542     Set.UNIV                (for top)
  3543     Set.inter               (for inf, formerly Set.Int)
  3544     Set.union               (for sup, formerly Set.Un)
  3545     Complete_Lattice.Inter  (for Inf)
  3546     Complete_Lattice.Union  (for Sup)
  3547     Complete_Lattice.INTER  (for INFI)
  3548     Complete_Lattice.UNION  (for SUPR)
  3549   - object-logic definitions as far as appropriate
  3550 
  3551 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3552 Un_subset_iff are explicitly deleted as default simp rules; then also
  3553 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3554 deleted to achieve the desired effect.
  3555 
  3556 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3557 rules by default any longer; the same applies to min_max.inf_absorb1
  3558 etc.  INCOMPATIBILITY.
  3559 
  3560 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3561 pred_set_conv by default.  INCOMPATIBILITY.
  3562 
  3563 * Power operations on relations and functions are now one dedicated
  3564 constant "compow" with infix syntax "^^".  Power operation on
  3565 multiplicative monoids retains syntax "^" and is now defined generic
  3566 in class power.  INCOMPATIBILITY.
  3567 
  3568 * Relation composition "R O S" now has a more standard argument order:
  3569 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3570 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3571 break, since the O_assoc rule was not rewritten like this.  Fix using
  3572 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3573 
  3574 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3575 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3576 INCOMPATIBILITY.
  3577 
  3578 * Most rules produced by inductive and datatype package have mandatory
  3579 prefixes.  INCOMPATIBILITY.
  3580 
  3581 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3582 the attribute of the same name.  Each of the theorems in the list
  3583 DERIV_intros assumes composition with an additional function and
  3584 matches a variable to the derivative, which has to be solved by the
  3585 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3586 of most elementary terms.  Former Maclauren.DERIV_tac and
  3587 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3588 INCOMPATIBILITY.
  3589 
  3590 * Code generator attributes follow the usual underscore convention:
  3591     code_unfold     replaces    code unfold
  3592     code_post       replaces    code post
  3593     etc.
  3594   INCOMPATIBILITY.
  3595 
  3596 * Renamed methods:
  3597     sizechange -> size_change
  3598     induct_scheme -> induction_schema
  3599   INCOMPATIBILITY.
  3600 
  3601 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3602 INCOMPATIBILITY, use "undefined" directly.
  3603 
  3604 * Renamed theorems:
  3605     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3606     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3607     Suc_plus1 -> Suc_eq_plus1
  3608     *anti_sym -> *antisym*
  3609     vector_less_eq_def -> vector_le_def
  3610   INCOMPATIBILITY.
  3611 
  3612 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3613 INCOMPATIBILITY.
  3614 
  3615 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3616 INCOMPATIBILITY.
  3617 
  3618 
  3619 *** HOLCF ***
  3620 
  3621 * Theory Representable defines a class "rep" of domains that are
  3622 representable (via an ep-pair) in the universal domain type "udom".
  3623 Instances are provided for all type constructors defined in HOLCF.
  3624 
  3625 * The 'new_domain' command is a purely definitional version of the
  3626 domain package, for representable domains.  Syntax is identical to the
  3627 old domain package.  The 'new_domain' package also supports indirect
  3628 recursion using previously-defined type constructors.  See
  3629 src/HOLCF/ex/New_Domain.thy for examples.
  3630 
  3631 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3632 on the left-hand side of an equation, and then performs
  3633 simplification.  Rewriting is done using rules declared with the
  3634 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3635 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3636 
  3637 * The pattern-match compiler in 'fixrec' can now handle constructors
  3638 with HOL function types.  Pattern-match combinators for the Pair
  3639 constructor are pre-configured.
  3640 
  3641 * The 'fixrec' package now produces better fixed-point induction rules
  3642 for mutually-recursive definitions:  Induction rules have conclusions
  3643 of the form "P foo bar" instead of "P <foo, bar>".
  3644 
  3645 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3646 been renamed to "below".  The name "below" now replaces "less" in many
  3647 theorem names.  (Legacy theorem names using "less" are still supported
  3648 as well.)
  3649 
  3650 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3651 patterns can be used to generate strictness rules, or to make
  3652 functions more strict (much like the bang-patterns supported by the
  3653 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3654 examples.
  3655 
  3656 
  3657 *** ML ***
  3658 
  3659 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3660 errors and run-time exceptions, including detailed source positions.
  3661 
  3662 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3663 identified entries, with some additional information such as source
  3664 position, logical grouping etc.
  3665 
  3666 * Theory and context data is now introduced by the simplified and
  3667 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3668 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3669 copy operation) is still available for some time.
  3670 
  3671 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3672 provides a high-level programming interface to synchronized state
  3673 variables with atomic update.  This works via pure function
  3674 application within a critical section -- its runtime should be as
  3675 short as possible; beware of deadlocks if critical code is nested,
  3676 either directly or indirectly via other synchronized variables!
  3677 
  3678 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3679 wraps raw ML references, explicitly indicating their non-thread-safe
  3680 behaviour.  The Isar toplevel keeps this structure open, to
  3681 accommodate Proof General as well as quick and dirty interactive
  3682 experiments with references.
  3683 
  3684 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3685 parallel tactical reasoning.
  3686 
  3687 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3688 are similar to SUBPROOF, but are slightly more flexible: only the
  3689 specified parts of the subgoal are imported into the context, and the
  3690 body tactic may introduce new subgoals and schematic variables.
  3691 
  3692 * Old tactical METAHYPS, which does not observe the proof context, has
  3693 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3694 or Subgoal.FOCUS etc.
  3695 
  3696 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3697 functors have their own ML name space there is no point to mark them
  3698 separately.)  Minor INCOMPATIBILITY.
  3699 
  3700 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3701 
  3702 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3703 INCOMPATIBILITY.
  3704 
  3705 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3706 mode, which avoids critical setmp.
  3707 
  3708 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3709 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3710 Attrib/Method.setup introduced in Isabelle2009.
  3711 
  3712 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3713 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3714 last resort.  INCOMPATIBILITY.
  3715 
  3716 * Display.pretty_thm now requires a proper context (cf. former
  3717 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3718 or even Display.pretty_thm_without_context as last resort.
  3719 INCOMPATIBILITY.
  3720 
  3721 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3722 Syntax.pretty_typ/term directly, preferably with proper context
  3723 instead of global theory.
  3724 
  3725 
  3726 *** System ***
  3727 
  3728 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3729 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3730 usedir option -q.
  3731 
  3732 * Support for additional "Isabelle components" via etc/components, see
  3733 also the system manual.
  3734 
  3735 * The isabelle makeall tool now operates on all components with
  3736 IsaMakefile, not just hardwired "logics".
  3737 
  3738 * Removed "compress" option from isabelle-process and isabelle usedir;
  3739 this is always enabled.
  3740 
  3741 * Discontinued support for Poly/ML 4.x versions.
  3742 
  3743 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3744 on a given logic image.  This requires the lighttpd webserver and is
  3745 currently supported on Linux only.
  3746 
  3747 
  3748 
  3749 New in Isabelle2009 (April 2009)
  3750 --------------------------------
  3751 
  3752 *** General ***
  3753 
  3754 * Simplified main Isabelle executables, with less surprises on
  3755 case-insensitive file-systems (such as Mac OS).
  3756 
  3757   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3758     "isatool."
  3759 
  3760   - The former "isabelle" alias for "isabelle-process" has been
  3761     removed (should rarely occur to regular users).
  3762 
  3763   - The former "isabelle-interface" and its alias "Isabelle" have been
  3764     removed (interfaces are now regular Isabelle tools).
  3765 
  3766 Within scripts and make files, the Isabelle environment variables
  3767 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3768 respectively.  (The latter are still available as legacy feature.)
  3769 
  3770 The old isabelle-interface wrapper could react in confusing ways if
  3771 the interface was uninstalled or changed otherwise.  Individual
  3772 interface tool configuration is now more explicit, see also the
  3773 Isabelle system manual.  In particular, Proof General is now available
  3774 via "isabelle emacs".
  3775 
  3776 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3777 purge installed copies of Isabelle executables and re-run "isabelle
  3778 install -p ...", or use symlinks.
  3779 
  3780 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3781 old ~/isabelle, which was slightly non-standard and apt to cause
  3782 surprises on case-insensitive file-systems (such as Mac OS).
  3783 
  3784 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3785 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3786 care is required when using older releases of Isabelle.  Note that
  3787 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3788 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3789 
  3790 * Proofs of fully specified statements are run in parallel on
  3791 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3792 a regular 4-core machine, if the initial heap space is made reasonably
  3793 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3794 
  3795 * The main reference manuals ("isar-ref", "implementation", and
  3796 "system") have been updated and extended.  Formally checked references
  3797 as hyperlinks are now available uniformly.
  3798 
  3799 
  3800 *** Pure ***
  3801 
  3802 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3803 respects.  The most important changes are listed below.  See the
  3804 Tutorial on Locales ("locales" manual) for details.
  3805 
  3806 - In locale expressions, instantiation replaces renaming.  Parameters
  3807 must be declared in a for clause.  To aid compatibility with previous
  3808 parameter inheritance, in locale declarations, parameters that are not
  3809 'touched' (instantiation position "_" or omitted) are implicitly added
  3810 with their syntax at the beginning of the for clause.
  3811 
  3812 - Syntax from abbreviations and definitions in locales is available in
  3813 locale expressions and context elements.  The latter is particularly
  3814 useful in locale declarations.
  3815 
  3816 - More flexible mechanisms to qualify names generated by locale
  3817 expressions.  Qualifiers (prefixes) may be specified in locale
  3818 expressions, and can be marked as mandatory (syntax: "name!:") or
  3819 optional (syntax "name?:").  The default depends for plain "name:"
  3820 depends on the situation where a locale expression is used: in
  3821 commands 'locale' and 'sublocale' prefixes are optional, in
  3822 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3823 implicit qualifiers derived from the parameter names of a locale are
  3824 no longer generated.
  3825 
  3826 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3827 instantiation clause in "interpretation" and "interpret" (square
  3828 brackets) is no longer available.  Use locale expressions.
  3829 
  3830 - When converting proof scripts, mandatory qualifiers in
  3831 'interpretation' and 'interpret' should be retained by default, even
  3832 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3833 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3834 in locale expressions range over a single locale instance only.
  3835 
  3836 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3837 In existing theorem specifications replace the includes element by the
  3838 respective context elements of the included locale, omitting those
  3839 that are already present in the theorem specification.  Multiple
  3840 assume elements of a locale should be replaced by a single one
  3841 involving the locale predicate.  In the proof body, declarations (most
  3842 notably theorems) may be regained by interpreting the respective
  3843 locales in the proof context as required (command "interpret").
  3844 
  3845 If using "includes" in replacement of a target solely because the
  3846 parameter types in the theorem are not as general as in the target,
  3847 consider declaring a new locale with additional type constraints on
  3848 the parameters (context element "constrains").
  3849 
  3850 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3851 
  3852 - Locale interpretation commands no longer attempt to simplify goal.
  3853 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3854 methods intro_locales and unfold_locales to clarify.
  3855 
  3856 - Locale interpretation commands no longer accept interpretation
  3857 attributes.  INCOMPATIBILITY.
  3858 
  3859 * Class declaration: so-called "base sort" must not be given in import
  3860 list any longer, but is inferred from the specification.  Particularly
  3861 in HOL, write
  3862 
  3863     class foo = ...
  3864 
  3865 instead of
  3866 
  3867     class foo = type + ...
  3868 
  3869 * Class target: global versions of theorems stemming do not carry a
  3870 parameter prefix any longer.  INCOMPATIBILITY.
  3871 
  3872 * Class 'instance' command no longer accepts attached definitions.
  3873 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3874 
  3875 * Recovered hiding of consts, which was accidentally broken in
  3876 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3877 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3878 
  3879 * Slightly more coherent Pure syntax, with updated documentation in
  3880 isar-ref manual.  Removed locales meta_term_syntax and
  3881 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3882 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3883 directly in regular applications.
  3884 
  3885 * There is a new syntactic category "float_const" for signed decimal
  3886 fractions (e.g. 123.45 or -123.45).
  3887 
  3888 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3889 interface with 'setup' command instead.
  3890 
  3891 * Command 'local_setup' is similar to 'setup', but operates on a local
  3892 theory context.
  3893 
  3894 * The 'axiomatization' command now only works within a global theory
  3895 context.  INCOMPATIBILITY.
  3896 
  3897 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3898 hypotheses.  Sorts required in the course of reasoning need to be
  3899 covered by the constraints in the initial statement, completed by the
  3900 type instance information of the background theory.  Non-trivial sort
  3901 hypotheses, which rarely occur in practice, may be specified via
  3902 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3903 
  3904   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3905 
  3906 The result contains an implicit sort hypotheses as before --
  3907 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3908 normalization.
  3909 
  3910 * Generalized Isar history, with support for linear undo, direct state
  3911 addressing etc.
  3912 
  3913 * Changed defaults for unify configuration options:
  3914 
  3915   unify_trace_bound = 50 (formerly 25)
  3916   unify_search_bound = 60 (formerly 30)
  3917 
  3918 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3919 
  3920   a) On theory merge, the last set of code equations for a particular
  3921      constant is taken (in accordance with the policy applied by other
  3922      parts of the code generator framework).
  3923 
  3924   b) Code equations stemming from explicit declarations (e.g. code
  3925      attribute) gain priority over default code equations stemming
  3926      from definition, primrec, fun etc.
  3927 
  3928 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3929 
  3930 * Unified theorem tables for both code generators.  Thus [code
  3931 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3932 
  3933 * Command 'find_consts' searches for constants based on type and name
  3934 patterns, e.g.
  3935 
  3936     find_consts "_ => bool"
  3937 
  3938 By default, matching is against subtypes, but it may be restricted to
  3939 the whole type.  Searching by name is possible.  Multiple queries are
  3940 conjunctive and queries may be negated by prefixing them with a
  3941 hyphen:
  3942 
  3943     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3944 
  3945 * New 'find_theorems' criterion "solves" matches theorems that
  3946 directly solve the current goal (modulo higher-order unification).
  3947 
  3948 * Auto solve feature for main theorem statements: whenever a new goal
  3949 is stated, "find_theorems solves" is called; any theorems that could
  3950 solve the lemma directly are listed as part of the goal state.
  3951 Cf. associated options in Proof General Isabelle settings menu,
  3952 enabled by default, with reasonable timeout for pathological cases of
  3953 higher-order unification.
  3954 
  3955 
  3956 *** Document preparation ***
  3957 
  3958 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3959 demanding keyword 'by' and supporting the full method expression
  3960 syntax just like the Isar command 'by'.
  3961 
  3962 
  3963 *** HOL ***
  3964 
  3965 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3966 points Main and Complex_Main remain as before.
  3967 
  3968 * Logic image HOL-Plain provides a minimal HOL with the most important
  3969 tools available (inductive, datatype, primrec, ...).  This facilitates
  3970 experimentation and tool development.  Note that user applications
  3971 (and library theories) should never refer to anything below theory
  3972 Main, as before.
  3973 
  3974 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3975 experimentation due to shorter build times.
  3976 
  3977 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3978 were previously part of former HOL-Complex.  Entry point Hyperreal
  3979 remains valid, but theories formerly using Complex_Main should now use
  3980 new entry point Hypercomplex.
  3981 
  3982 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3983 Posix processes.  Avoids potentially expensive forking of the ML
  3984 process.  New thread-based implementation also works on non-Unix
  3985 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3986 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3987 commands are covered in the isar-ref manual.
  3988 
  3989 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3990 sledgehammer without local ATP installation (Vampire etc.). Other
  3991 provers may be included via suitable ML wrappers, see also
  3992 src/HOL/ATP_Linkup.thy.
  3993 
  3994 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3995 menu.
  3996 
  3997 * The metis method no longer fails because the theorem is too trivial
  3998 (contains the empty clause).
  3999 
  4000 * The metis method now fails in the usual manner, rather than raising
  4001 an exception, if it determines that it cannot prove the theorem.
  4002 
  4003 * Method "coherent" implements a prover for coherent logic (see also
  4004 src/Tools/coherent.ML).
  4005 
  4006 * Constants "undefined" and "default" replace "arbitrary".  Usually
  4007 "undefined" is the right choice to replace "arbitrary", though
  4008 logically there is no difference.  INCOMPATIBILITY.
  4009 
  4010 * Command "value" now integrates different evaluation mechanisms.  The
  4011 result of the first successful evaluation mechanism is printed.  In
  4012 square brackets a particular named evaluation mechanisms may be
  4013 specified (currently, [SML], [code] or [nbe]).  See further
  4014 src/HOL/ex/Eval_Examples.thy.
  4015 
  4016 * Normalization by evaluation now allows non-leftlinear equations.
  4017 Declare with attribute [code nbe].
  4018 
  4019 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  4020 as the structured Isar versions "cases" and "induct", cf. the
  4021 corresponding "cases" and "induct" attributes.  Mutual induction rules
  4022 are now presented as a list of individual projections
  4023 (e.g. foo_bar.inducts for types foo and bar); the old format with
  4024 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  4025 rare situations a different rule is selected --- notably nested tuple
  4026 elimination instead of former prod.exhaust: use explicit (case_tac t
  4027 rule: prod.exhaust) here.
  4028 
  4029 * Attributes "cases", "induct", "coinduct" support "del" option.
  4030 
  4031 * Removed fact "case_split_thm", which duplicates "case_split".
  4032 
  4033 * The option datatype has been moved to a new theory Option.  Renamed
  4034 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  4035 
  4036 * New predicate "strict_mono" classifies strict functions on partial
  4037 orders.  With strict functions on linear orders, reasoning about
  4038 (in)equalities is facilitated by theorems "strict_mono_eq",
  4039 "strict_mono_less_eq" and "strict_mono_less".
  4040 
  4041 * Some set operations are now proper qualified constants with
  4042 authentic syntax.  INCOMPATIBILITY:
  4043 
  4044     op Int ~>   Set.Int
  4045     op Un ~>    Set.Un
  4046     INTER ~>    Set.INTER
  4047     UNION ~>    Set.UNION
  4048     Inter ~>    Set.Inter
  4049     Union ~>    Set.Union
  4050     {} ~>       Set.empty
  4051     UNIV ~>     Set.UNIV
  4052 
  4053 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4054 theory Set.
  4055 
  4056 * Auxiliary class "itself" has disappeared -- classes without any
  4057 parameter are treated as expected by the 'class' command.
  4058 
  4059 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4060 
  4061 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4062 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4063 
  4064 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4065 method "approximation".  It proves formulas on real values by using
  4066 interval arithmetic.  In the formulas are also the transcendental
  4067 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4068 allowed. For examples see
  4069 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4070 
  4071 * Theory "Reflection" now resides in HOL/Library.
  4072 
  4073 * Entry point to Word library now simply named "Word".
  4074 INCOMPATIBILITY.
  4075 
  4076 * Made source layout more coherent with logical distribution
  4077 structure:
  4078 
  4079     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4080     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4081     src/HOL/Library/GCD.thy ~> src/HOL/
  4082     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4083     src/HOL/Library/Parity.thy ~> src/HOL/
  4084     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4085     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4086     src/HOL/Real/Lubs.thy ~> src/HOL/
  4087     src/HOL/Real/PReal.thy ~> src/HOL/
  4088     src/HOL/Real/Rational.thy ~> src/HOL/
  4089     src/HOL/Real/RComplete.thy ~> src/HOL/
  4090     src/HOL/Real/RealDef.thy ~> src/HOL/
  4091     src/HOL/Real/RealPow.thy ~> src/HOL/
  4092     src/HOL/Real/Real.thy ~> src/HOL/
  4093     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4094     src/HOL/Complex/Complex.thy ~> src/HOL/
  4095     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4096     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4097     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4098     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4099     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4100     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4101     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4102     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4103     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4104     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4105     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4106     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4107     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4108     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4109     src/HOL/Real/Float ~> src/HOL/Library/
  4110     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4111     src/HOL/Real/RealVector.thy ~> src/HOL/
  4112 
  4113     src/HOL/arith_data.ML ~> src/HOL/Tools
  4114     src/HOL/hologic.ML ~> src/HOL/Tools
  4115     src/HOL/simpdata.ML ~> src/HOL/Tools
  4116     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4117     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4118     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4119     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4120     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4121     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4122     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4123 
  4124     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4125     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4126     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4127     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4128     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4129     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4130 
  4131 * If methods "eval" and "evaluation" encounter a structured proof
  4132 state with !!/==>, only the conclusion is evaluated to True (if
  4133 possible), avoiding strange error messages.
  4134 
  4135 * Method "sizechange" automates termination proofs using (a
  4136 modification of) the size-change principle.  Requires SAT solver.  See
  4137 src/HOL/ex/Termination.thy for examples.
  4138 
  4139 * Simplifier: simproc for let expressions now unfolds if bound
  4140 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4141 
  4142 * Method "arith": Linear arithmetic now ignores all inequalities when
  4143 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4144 
  4145 * New attribute "arith" for facts that should always be used
  4146 automatically by arithmetic. It is intended to be used locally in
  4147 proofs, e.g.
  4148 
  4149   assumes [arith]: "x > 0"
  4150 
  4151 Global usage is discouraged because of possible performance impact.
  4152 
  4153 * New classes "top" and "bot" with corresponding operations "top" and
  4154 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4155 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4156 
  4157 * Changed definition lemma "less_fun_def" in order to provide an
  4158 instance for preorders on functions; use lemma "less_le" instead.
  4159 INCOMPATIBILITY.
  4160 
  4161 * Theory Orderings: class "wellorder" moved here, with explicit
  4162 induction rule "less_induct" as assumption.  For instantiation of
  4163 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4164 INCOMPATIBILITY.
  4165 
  4166 * Theory Orderings: added class "preorder" as superclass of "order".
  4167 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4168 etc. slightly changed.  Some theorems named order_class.* now named
  4169 preorder_class.*.
  4170 
  4171 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4172 "diag" to "Id_on".
  4173 
  4174 * Theory Finite_Set: added a new fold combinator of type
  4175 
  4176   ('a => 'b => 'b) => 'b => 'a set => 'b
  4177 
  4178 Occasionally this is more convenient than the old fold combinator
  4179 which is now defined in terms of the new one and renamed to
  4180 fold_image.
  4181 
  4182 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4183 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4184 extended with further lemmas!).  At the moment both still exist but
  4185 the former will disappear at some point.
  4186 
  4187 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4188 class recpower.  Type-specific simp rules for various recpower types
  4189 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4190 
  4191 rat_power_0    -> power_0
  4192 rat_power_Suc  -> power_Suc
  4193 realpow_0      -> power_0
  4194 realpow_Suc    -> power_Suc
  4195 complexpow_0   -> power_0
  4196 complexpow_Suc -> power_Suc
  4197 power_poly_0   -> power_0
  4198 power_poly_Suc -> power_Suc
  4199 
  4200 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4201 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4202 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4203 of lemmas from Divides has been generalized from nat to class
  4204 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4205 renames resulting from duplicate elimination:
  4206 
  4207     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4208     zero_dvd_iff ~>         dvd_0_left_iff
  4209     dvd_0 ~>                dvd_0_right
  4210     DIVISION_BY_ZERO_DIV ~> div_by_0
  4211     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4212     mult_div ~>             div_mult_self2_is_id
  4213     mult_mod ~>             mod_mult_self2_is_0
  4214 
  4215 * Theory IntDiv: removed many lemmas that are instances of class-based
  4216 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4217 rename old lemmas as follows:
  4218 
  4219 dvd_diff               -> nat_dvd_diff
  4220 dvd_zminus_iff         -> dvd_minus_iff
  4221 mod_add1_eq            -> mod_add_eq
  4222 mod_mult1_eq           -> mod_mult_right_eq
  4223 mod_mult1_eq'          -> mod_mult_left_eq
  4224 mod_mult_distrib_mod   -> mod_mult_eq
  4225 nat_mod_add_left_eq    -> mod_add_left_eq
  4226 nat_mod_add_right_eq   -> mod_add_right_eq
  4227 nat_mod_div_trivial    -> mod_div_trivial
  4228 nat_mod_mod_trivial    -> mod_mod_trivial
  4229 zdiv_zadd_self1        -> div_add_self1
  4230 zdiv_zadd_self2        -> div_add_self2
  4231 zdiv_zmult_self1       -> div_mult_self2_is_id
  4232 zdiv_zmult_self2       -> div_mult_self1_is_id
  4233 zdvd_triv_left         -> dvd_triv_left
  4234 zdvd_triv_right        -> dvd_triv_right
  4235 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4236 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4237 zmod_zadd_left_eq      -> mod_add_left_eq
  4238 zmod_zadd_right_eq     -> mod_add_right_eq
  4239 zmod_zadd_self1        -> mod_add_self1
  4240 zmod_zadd_self2        -> mod_add_self2
  4241 zmod_zadd1_eq          -> mod_add_eq
  4242 zmod_zdiff1_eq         -> mod_diff_eq
  4243 zmod_zdvd_zmod         -> mod_mod_cancel
  4244 zmod_zmod_cancel       -> mod_mod_cancel
  4245 zmod_zmult_self1       -> mod_mult_self2_is_0
  4246 zmod_zmult_self2       -> mod_mult_self1_is_0
  4247 zmod_1                 -> mod_by_1
  4248 zdiv_1                 -> div_by_1
  4249 zdvd_abs1              -> abs_dvd_iff
  4250 zdvd_abs2              -> dvd_abs_iff
  4251 zdvd_refl              -> dvd_refl
  4252 zdvd_trans             -> dvd_trans
  4253 zdvd_zadd              -> dvd_add
  4254 zdvd_zdiff             -> dvd_diff
  4255 zdvd_zminus_iff        -> dvd_minus_iff
  4256 zdvd_zminus2_iff       -> minus_dvd_iff
  4257 zdvd_zmultD            -> dvd_mult_right
  4258 zdvd_zmultD2           -> dvd_mult_left
  4259 zdvd_zmult_mono        -> mult_dvd_mono
  4260 zdvd_0_right           -> dvd_0_right
  4261 zdvd_0_left            -> dvd_0_left_iff
  4262 zdvd_1_left            -> one_dvd
  4263 zminus_dvd_iff         -> minus_dvd_iff
  4264 
  4265 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4266 
  4267 * The real numbers offer decimal input syntax: 12.34 is translated
  4268 into 1234/10^2. This translation is not reversed upon output.
  4269 
  4270 * Theory Library/Polynomial defines an abstract type 'a poly of
  4271 univariate polynomials with coefficients of type 'a.  In addition to
  4272 the standard ring operations, it also supports div and mod.  Code
  4273 generation is also supported, using list-style constructors.
  4274 
  4275 * Theory Library/Inner_Product defines a class of real_inner for real
  4276 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4277 => real.  Class real_inner is a subclass of real_normed_vector from
  4278 theory RealVector.
  4279 
  4280 * Theory Library/Product_Vector provides instances for the product
  4281 type 'a * 'b of several classes from RealVector and Inner_Product.
  4282 Definitions of addition, subtraction, scalar multiplication, norms,
  4283 and inner products are included.
  4284 
  4285 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4286 addition to the field operations, numerals and case syntax are also
  4287 supported.
  4288 
  4289 * Theory Library/Diagonalize provides constructive version of Cantor's
  4290 first diagonalization argument.
  4291 
  4292 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4293 zlcm (for int); carried together from various gcd/lcm developements in
  4294 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4295 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4296 may recover tupled syntax as follows:
  4297 
  4298     hide (open) const gcd
  4299     abbreviation gcd where
  4300       "gcd == (%(a, b). GCD.gcd a b)"
  4301     notation (output)
  4302       GCD.gcd ("gcd '(_, _')")
  4303 
  4304 The same works for lcm, zgcd, zlcm.
  4305 
  4306 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4307 instantiations for algebraic structures.  Removed some duplicate
  4308 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4309 
  4310 * ML antiquotation @{code} takes a constant as argument and generates
  4311 corresponding code in background and inserts name of the corresponding
  4312 resulting ML value/function/datatype constructor binding in place.
  4313 All occurrences of @{code} with a single ML block are generated
  4314 simultaneously.  Provides a generic and safe interface for
  4315 instrumentalizing code generation.  See
  4316 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4317 In future you ought to refrain from ad-hoc compiling generated SML
  4318 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4319 cannot refer to constants for which user-defined serializations are
  4320 set.  Refer to the corresponding ML counterpart directly in that
  4321 cases.
  4322 
  4323 * Command 'rep_datatype': instead of theorem names the command now
  4324 takes a list of terms denoting the constructors of the type to be
  4325 represented as datatype.  The characteristic theorems have to be
  4326 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4327 have disappeared in favour of existing ones:
  4328 
  4329     unit_induct                 ~> unit.induct
  4330     prod_induct                 ~> prod.induct
  4331     sum_induct                  ~> sum.induct
  4332     Suc_Suc_eq                  ~> nat.inject
  4333     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4334 
  4335 
  4336 *** HOL-Algebra ***
  4337 
  4338 * New locales for orders and lattices where the equivalence relation
  4339 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4340 locales use a record structure with field eq for the equivalence.
  4341 
  4342 * New theory of factorial domains.
  4343 
  4344 * Units_l_inv and Units_r_inv are now simp rules by default.
  4345 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4346 and/or r_inv will now also require deletion of these lemmas.
  4347 
  4348 * Renamed the following theorems, INCOMPATIBILITY:
  4349 
  4350 UpperD ~> Upper_memD
  4351 LowerD ~> Lower_memD
  4352 least_carrier ~> least_closed
  4353 greatest_carrier ~> greatest_closed
  4354 greatest_Lower_above ~> greatest_Lower_below
  4355 one_zero ~> carrier_one_zero
  4356 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4357 
  4358 
  4359 *** HOL-Nominal ***
  4360 
  4361 * Nominal datatypes can now contain type-variables.
  4362 
  4363 * Commands 'nominal_inductive' and 'equivariance' work with local
  4364 theory targets.
  4365 
  4366 * Nominal primrec can now works with local theory targets and its
  4367 specification syntax now conforms to the general format as seen in
  4368 'inductive' etc.
  4369 
  4370 * Method "perm_simp" honours the standard simplifier attributes
  4371 (no_asm), (no_asm_use) etc.
  4372 
  4373 * The new predicate #* is defined like freshness, except that on the
  4374 left hand side can be a set or list of atoms.
  4375 
  4376 * Experimental command 'nominal_inductive2' derives strong induction
  4377 principles for inductive definitions.  In contrast to
  4378 'nominal_inductive', which can only deal with a fixed number of
  4379 binders, it can deal with arbitrary expressions standing for sets of
  4380 atoms to be avoided.  The only inductive definition we have at the
  4381 moment that needs this generalisation is the typing rule for Lets in
  4382 the algorithm W:
  4383 
  4384  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4385  -----------------------------------------------------------------
  4386          Gamma |- Let x be t1 in t2 : T2
  4387 
  4388 In this rule one wants to avoid all the binders that are introduced by
  4389 "close Gamma T1".  We are looking for other examples where this
  4390 feature might be useful.  Please let us know.
  4391 
  4392 
  4393 *** HOLCF ***
  4394 
  4395 * Reimplemented the simplification procedure for proving continuity
  4396 subgoals.  The new simproc is extensible; users can declare additional
  4397 continuity introduction rules with the attribute [cont2cont].
  4398 
  4399 * The continuity simproc now uses a different introduction rule for
  4400 solving continuity subgoals on terms with lambda abstractions.  In
  4401 some rare cases the new simproc may fail to solve subgoals that the
  4402 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4403 Potential INCOMPATIBILITY.
  4404 
  4405 * Command 'fixrec': specification syntax now conforms to the general
  4406 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4407 examples.  INCOMPATIBILITY.
  4408 
  4409 
  4410 *** ZF ***
  4411 
  4412 * Proof of Zorn's Lemma for partial orders.
  4413 
  4414 
  4415 *** ML ***
  4416 
  4417 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4418 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4419 depends on multithreading, so timouts will not work before Poly/ML
  4420 5.2.1!
  4421 
  4422 * High-level support for concurrent ML programming, see
  4423 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4424 particularly convenient to organize independent functional
  4425 computations.  The concept of "synchronized variables" provides a
  4426 higher-order interface for components with shared state, avoiding the
  4427 delicate details of mutexes and condition variables.  (Requires
  4428 Poly/ML 5.2.1 or later.)
  4429 
  4430 * ML bindings produced via Isar commands are stored within the Isar
  4431 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4432 become thread-safe and work with undo as expected (concerning
  4433 top-level bindings, not side-effects on global references).
  4434 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4435 compiler at runtime; really global bindings need to be given outside a
  4436 theory.  (Requires Poly/ML 5.2 or later.)
  4437 
  4438 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4439 context.  Top-level ML bindings are stored within the proof context in
  4440 a purely sequential fashion, disregarding the nested proof structure.
  4441 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4442 proof.  (Requires Poly/ML 5.2 or later.)
  4443 
  4444 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4445 and Method.setup, as well as Isar commands 'attribute_setup' and
  4446 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4447 existing code accordingly, or use plain 'setup' together with old
  4448 Method.add_method.
  4449 
  4450 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4451 to 'a -> thm, while results are always tagged with an authentic oracle
  4452 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4453 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4454 accordingly.  Note that extra performance may be gained by producing
  4455 the cterm carefully, avoiding slow Thm.cterm_of.
  4456 
  4457 * Simplified interface for defining document antiquotations via
  4458 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4459 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4460 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4461 examples.
  4462 
  4463 * More systematic treatment of long names, abstract name bindings, and
  4464 name space operations.  Basic operations on qualified names have been
  4465 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4466 Long_Name.append.  Old type bstring has been mostly replaced by
  4467 abstract type binding (see structure Binding), which supports precise
  4468 qualification by packages and local theory targets, as well as proper
  4469 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4470 bstring values into Binding.name, or better pass through abstract
  4471 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4472 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4473 
  4474 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4475 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4476 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4477 
  4478 * Disposed old type and term read functions (Sign.read_def_typ,
  4479 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4480 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4481 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4482 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4483 resort for legacy applications.
  4484 
  4485 * Disposed old declarations, tactics, tactic combinators that refer to
  4486 the simpset or claset of an implicit theory (such as Addsimps,
  4487 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4488 embedded ML text, or local_simpset_of with a proper context passed as
  4489 explicit runtime argument.
  4490 
  4491 * Rules and tactics that read instantiations (read_instantiate,
  4492 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4493 context, which is required for parsing and type-checking.  Moreover,
  4494 the variables are specified as plain indexnames, not string encodings
  4495 thereof.  INCOMPATIBILITY.
  4496 
  4497 * Generic Toplevel.add_hook interface allows to analyze the result of
  4498 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4499 for theorem dependency output of transactions resulting in a new
  4500 theory state.
  4501 
  4502 * ML antiquotations: block-structured compilation context indicated by
  4503 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4504 
  4505   @{binding name}                         - basic name binding
  4506   @{let ?pat = term}                      - term abbreviation (HO matching)
  4507   @{note name = fact}                     - fact abbreviation
  4508   @{thm fact}                             - singleton fact (with attributes)
  4509   @{thms fact}                            - general fact (with attributes)
  4510   @{lemma prop by method}                 - singleton goal
  4511   @{lemma prop by meth1 meth2}            - singleton goal
  4512   @{lemma prop1 ... propN by method}      - general goal
  4513   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4514   @{lemma (open) ...}                     - open derivation
  4515 
  4516 
  4517 *** System ***
  4518 
  4519 * The Isabelle "emacs" tool provides a specific interface to invoke
  4520 Proof General / Emacs, with more explicit failure if that is not
  4521 installed (the old isabelle-interface script silently falls back on
  4522 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4523 installation location of the Proof General distribution.
  4524 
  4525 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4526 the Isabelle process into a JVM/Scala application.  See
  4527 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4528 process wrapper has been discontinued.)
  4529 
  4530 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4531 
  4532 * Various status messages (with exact source position information) are
  4533 emitted, if proper markup print mode is enabled.  This allows
  4534 user-interface components to provide detailed feedback on internal
  4535 prover operations.
  4536 
  4537 
  4538 
  4539 New in Isabelle2008 (June 2008)
  4540 -------------------------------
  4541 
  4542 *** General ***
  4543 
  4544 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4545 and updated, with formally checked references as hyperlinks.
  4546 
  4547 * Theory loader: use_thy (and similar operations) no longer set the
  4548 implicit ML context, which was occasionally hard to predict and in
  4549 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4550 provides a proper context already.
  4551 
  4552 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4553 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4554 'use' within a theory file will do the job.
  4555 
  4556 * Name space merge now observes canonical order, i.e. the second space
  4557 is inserted into the first one, while existing entries in the first
  4558 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4559 swap theory imports.
  4560 
  4561 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4562 INCOMPATIBILITY in identifier syntax etc.
  4563 
  4564 * Outer syntax: string tokens no longer admit escaped white space,
  4565 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4566 white space without escapes.
  4567 
  4568 * Outer syntax: string tokens may contain arbitrary character codes
  4569 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4570 "foo_bar".
  4571 
  4572 
  4573 *** Pure ***
  4574 
  4575 * Context-dependent token translations.  Default setup reverts locally
  4576 fixed variables, and adds hilite markup for undeclared frees.
  4577 
  4578 * Unused theorems can be found using the new command 'unused_thms'.
  4579 There are three ways of invoking it:
  4580 
  4581 (1) unused_thms
  4582      Only finds unused theorems in the current theory.
  4583 
  4584 (2) unused_thms thy_1 ... thy_n -
  4585      Finds unused theorems in the current theory and all of its ancestors,
  4586      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4587 
  4588 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4589      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4590      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4591      their ancestors.
  4592 
  4593 In order to increase the readability of the list produced by
  4594 unused_thms, theorems that have been created by a particular instance
  4595 of a theory command such as 'inductive' or 'function' are considered
  4596 to belong to the same "group", meaning that if at least one theorem in
  4597 this group is used, the other theorems in the same group are no longer
  4598 reported as unused.  Moreover, if all theorems in the group are
  4599 unused, only one theorem in the group is displayed.
  4600 
  4601 Note that proof objects have to be switched on in order for
  4602 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4603 usually the case when using Proof General with the default settings).
  4604 
  4605 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4606 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4607 duplicate fact bindings, or even accidental fact duplications.  Note
  4608 that tools may maintain dynamically scoped facts systematically, using
  4609 PureThy.add_thms_dynamic.
  4610 
  4611 * Command 'hide' now allows to hide from "fact" name space as well.
  4612 
  4613 * Eliminated destructive theorem database, simpset, claset, and
  4614 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4615 update of theories within ML code.
  4616 
  4617 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4618 INCOMPATIBILITY, object-logics depending on former Pure require
  4619 additional setup PureThy.old_appl_syntax_setup; object-logics
  4620 depending on former CPure need to refer to Pure.
  4621 
  4622 * Commands 'use' and 'ML' are now purely functional, operating on
  4623 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4624 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4625 INCOMPATIBILITY.
  4626 
  4627 * Command 'setup': discontinued implicit version with ML reference.
  4628 
  4629 * Instantiation target allows for simultaneous specification of class
  4630 instance operations together with an instantiation proof.
  4631 Type-checking phase allows to refer to class operations uniformly.
  4632 See src/HOL/Complex/Complex.thy for an Isar example and
  4633 src/HOL/Library/Eval.thy for an ML example.
  4634 
  4635 * Indexing of literal facts: be more serious about including only
  4636 facts from the visible specification/proof context, but not the
  4637 background context (locale etc.).  Affects `prop` notation and method
  4638 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4639 situations.
  4640 
  4641 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4642 "(open)" option, which used to expose internal bound variables to the
  4643 proof text.
  4644 
  4645 * Isar statements: removed obsolete case "rule_context".
  4646 INCOMPATIBILITY, better use explicit fixes/assumes.
  4647 
  4648 * Locale proofs: default proof step now includes 'unfold_locales';
  4649 hence 'proof' without argument may be used to unfold locale
  4650 predicates.
  4651 
  4652 
  4653 *** Document preparation ***
  4654 
  4655 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4656 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4657 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4658 spot); renamed link color from "darkblue" to "linkcolor" (default
  4659 value unchanged, can be redefined via \definecolor); no longer sets
  4660 "a4paper" option (unnecessary or even intrusive).
  4661 
  4662 * Antiquotation @{lemma A method} proves proposition A by the given
  4663 method (either a method name or a method name plus (optional) method
  4664 arguments in parentheses) and prints A just like @{prop A}.
  4665 
  4666 
  4667 *** HOL ***
  4668 
  4669 * New primrec package.  Specification syntax conforms in style to
  4670 definition/function/....  No separate induction rule is provided.  The
  4671 "primrec" command distinguishes old-style and new-style specifications
  4672 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4673 When adjusting theories, beware: constants stemming from new-style
  4674 primrec specifications have authentic syntax.
  4675 
  4676 * Metis prover is now an order of magnitude faster, and also works
  4677 with multithreading.
  4678 
  4679 * Metis: the maximum number of clauses that can be produced from a
  4680 theorem is now given by the attribute max_clauses.  Theorems that
  4681 exceed this number are ignored, with a warning printed.
  4682 
  4683 * Sledgehammer no longer produces structured proofs by default. To
  4684 enable, declare [[sledgehammer_full = true]].  Attributes
  4685 reconstruction_modulus, reconstruction_sorts renamed
  4686 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4687 
  4688 * Method "induct_scheme" derives user-specified induction rules
  4689 from well-founded induction and completeness of patterns. This factors
  4690 out some operations that are done internally by the function package
  4691 and makes them available separately.  See
  4692 src/HOL/ex/Induction_Scheme.thy for examples.
  4693 
  4694 * More flexible generation of measure functions for termination
  4695 proofs: Measure functions can be declared by proving a rule of the
  4696 form "is_measure f" and giving it the [measure_function] attribute.
  4697 The "is_measure" predicate is logically meaningless (always true), and
  4698 just guides the heuristic.  To find suitable measure functions, the
  4699 termination prover sets up the goal "is_measure ?f" of the appropriate
  4700 type and generates all solutions by prolog-style backwards proof using
  4701 the declared rules.
  4702 
  4703 This setup also deals with rules like
  4704 
  4705   "is_measure f ==> is_measure (list_size f)"
  4706 
  4707 which accommodates nested datatypes that recurse through lists.
  4708 Similar rules are predeclared for products and option types.
  4709 
  4710 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4711 
  4712   INCOMPATIBILITIES:
  4713 
  4714   - Definitions of overloaded constants on sets have to be replaced by
  4715     definitions on => and bool.
  4716 
  4717   - Some definitions of overloaded operators on sets can now be proved
  4718     using the definitions of the operators on => and bool.  Therefore,
  4719     the following theorems have been renamed:
  4720 
  4721       subset_def   -> subset_eq
  4722       psubset_def  -> psubset_eq
  4723       set_diff_def -> set_diff_eq
  4724       Compl_def    -> Compl_eq
  4725       Sup_set_def  -> Sup_set_eq
  4726       Inf_set_def  -> Inf_set_eq
  4727       sup_set_def  -> sup_set_eq
  4728       inf_set_def  -> inf_set_eq
  4729 
  4730   - Due to the incompleteness of the HO unification algorithm, some
  4731     rules such as subst may require manual instantiation, if some of
  4732     the unknowns in the rule is a set.
  4733 
  4734   - Higher order unification and forward proofs:
  4735     The proof pattern
  4736 
  4737       have "P (S::'a set)" <...>
  4738       then have "EX S. P S" ..
  4739 
  4740     no longer works (due to the incompleteness of the HO unification
  4741     algorithm) and must be replaced by the pattern
  4742 
  4743       have "EX S. P S"
  4744       proof
  4745         show "P S" <...>
  4746       qed
  4747 
  4748   - Calculational reasoning with subst (or similar rules):
  4749     The proof pattern
  4750 
  4751       have "P (S::'a set)" <...>
  4752       also have "S = T" <...>
  4753       finally have "P T" .
  4754 
  4755     no longer works (for similar reasons as the previous example) and
  4756     must be replaced by something like
  4757 
  4758       have "P (S::'a set)" <...>
  4759       moreover have "S = T" <...>
  4760       ultimately have "P T" by simp
  4761 
  4762   - Tactics or packages written in ML code:
  4763     Code performing pattern matching on types via
  4764 
  4765       Type ("set", [T]) => ...
  4766 
  4767     must be rewritten. Moreover, functions like strip_type or
  4768     binder_types no longer return the right value when applied to a
  4769     type of the form
  4770 
  4771       T1 => ... => Tn => U => bool
  4772 
  4773     rather than
  4774 
  4775       T1 => ... => Tn => U set
  4776 
  4777 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4778 Wellfounded_Relations to theory Wellfounded.
  4779 
  4780 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4781 
  4782 * Class finite no longer treats UNIV as class parameter.  Use class
  4783 enum from theory Library/Enum instead to achieve a similar effect.
  4784 INCOMPATIBILITY.
  4785 
  4786 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4787 and "Cons".  INCOMPATIBILITY.
  4788 
  4789 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4790 Potential INCOMPATIBILITY.
  4791 
  4792 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4793 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4794 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4795 
  4796 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4797 the same name in theory Orderings:
  4798 
  4799   less_trans
  4800   less_linear
  4801   le_imp_less_or_eq
  4802   le_less_trans
  4803   less_le_trans
  4804   less_not_sym
  4805   less_asym
  4806 
  4807 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4808 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4809 and different variable names.
  4810 
  4811 * Library/Option_ord.thy: Canonical order on option type.
  4812 
  4813 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4814 finite maps.
  4815 
  4816 * Library/Countable.thy: Type class for countable types.
  4817 
  4818 * Theory Int: The representation of numerals has changed.  The infix
  4819 operator BIT and the bit datatype with constructors B0 and B1 have
  4820 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4821 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4822 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4823 accordingly.
  4824 
  4825 * Theory Nat: definition of <= and < on natural numbers no longer
  4826 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4827 le_def and less_def have disappeared.  Consider lemmas not_less
  4828 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4829 
  4830 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4831 (whose purpose mainly is for various fold_set functionals) have been
  4832 abandoned in favor of the existing algebraic classes
  4833 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4834 lower_semilattice (resp. upper_semilattice) and linorder.
  4835 INCOMPATIBILITY.
  4836 
  4837 * Theory Transitive_Closure: induct and cases rules now declare proper
  4838 case_names ("base" and "step").  INCOMPATIBILITY.
  4839 
  4840 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4841 lattices.  The form set-specific version is available as
  4842 Inductive.lfp_ordinal_induct_set.
  4843 
  4844 * Renamed theorems "power.simps" to "power_int.simps".
  4845 INCOMPATIBILITY.
  4846 
  4847 * Class semiring_div provides basic abstract properties of semirings
  4848 with division and modulo operations.  Subsumes former class dvd_mod.
  4849 
  4850 * Merged theories IntDef, Numeral and IntArith into unified theory
  4851 Int.  INCOMPATIBILITY.
  4852 
  4853 * Theory Library/Code_Index: type "index" now represents natural
  4854 numbers rather than integers.  INCOMPATIBILITY.
  4855 
  4856 * New class "uminus" with operation "uminus" (split of from class
  4857 "minus" which now only has operation "minus", binary).
  4858 INCOMPATIBILITY.
  4859 
  4860 * Constants "card", "internal_split", "option_map" now with authentic
  4861 syntax.  INCOMPATIBILITY.
  4862 
  4863 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4864 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4865 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4866 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4867 Sup_set_def, le_def, less_def, option_map_def now with object
  4868 equality.  INCOMPATIBILITY.
  4869 
  4870 * Records. Removed K_record, and replaced it by pure lambda term
  4871 %x. c. The simplifier setup is now more robust against eta expansion.
  4872 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4873 
  4874 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4875 
  4876 * Library/ListVector: new theory of arithmetic vector operations.
  4877 
  4878 * Library/Order_Relation: new theory of various orderings as sets of
  4879 pairs.  Defines preorders, partial orders, linear orders and
  4880 well-orders on sets and on types.
  4881 
  4882 
  4883 *** ZF ***
  4884 
  4885 * Renamed some theories to allow to loading both ZF and HOL in the
  4886 same session:
  4887 
  4888   Datatype  -> Datatype_ZF
  4889   Inductive -> Inductive_ZF
  4890   Int       -> Int_ZF
  4891   IntDiv    -> IntDiv_ZF
  4892   Nat       -> Nat_ZF
  4893   List      -> List_ZF
  4894   Main      -> Main_ZF
  4895 
  4896 INCOMPATIBILITY: ZF theories that import individual theories below
  4897 Main might need to be adapted.  Regular theory Main is still
  4898 available, as trivial extension of Main_ZF.
  4899 
  4900 
  4901 *** ML ***
  4902 
  4903 * ML within Isar: antiquotation @{const name} or @{const
  4904 name(typargs)} produces statically-checked Const term.
  4905 
  4906 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4907 (of the same name).  Removed obsolete print command.
  4908 
  4909 * Removed obsolete "use_legacy_bindings" function.
  4910 
  4911 * The ``print mode'' is now a thread-local value derived from a global
  4912 template (the former print_mode reference), thus access becomes
  4913 non-critical.  The global print_mode reference is for session
  4914 management only; user-code should use print_mode_value,
  4915 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4916 
  4917 * Functions system/system_out provide a robust way to invoke external
  4918 shell commands, with propagation of interrupts (requires Poly/ML
  4919 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4920 
  4921 
  4922 *** System ***
  4923 
  4924 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4925 in accordance with Proof General 3.7, which prefers GNU emacs.
  4926 
  4927 * isatool tty runs Isabelle process with plain tty interaction;
  4928 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4929 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4930 
  4931 * isatool browser now works with Cygwin as well, using general
  4932 "javapath" function defined in Isabelle process environment.
  4933 
  4934 * YXML notation provides a simple and efficient alternative to
  4935 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4936 isatool yxml as described in the Isabelle system manual.
  4937 
  4938 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4939 provides general wrapper for managing an Isabelle process in a robust
  4940 fashion, with ``cooked'' output from stdin/stderr.
  4941 
  4942 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4943 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4944 
  4945 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4946 way of changing the user's settings is via
  4947 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4948 script.
  4949 
  4950 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4951 cores of the underlying machine, which is a good starting point for
  4952 optimal performance tuning.  The corresponding usedir option -M allows
  4953 "max" as an alias for "0".  WARNING: does not work on certain versions
  4954 of Mac OS (with Poly/ML 5.1).
  4955 
  4956 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4957 adverse effect of Isabelle flooding interactive front-ends (notably
  4958 ProofGeneral / XEmacs).
  4959 
  4960 
  4961 
  4962 New in Isabelle2007 (November 2007)
  4963 -----------------------------------
  4964 
  4965 *** General ***
  4966 
  4967 * More uniform information about legacy features, notably a
  4968 warning/error of "Legacy feature: ...", depending on the state of the
  4969 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4970 legacy features will disappear eventually.
  4971 
  4972 * Theory syntax: the header format ``theory A = B + C:'' has been
  4973 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4974 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4975 
  4976 * Theory syntax: the old non-Isar theory file format has been
  4977 discontinued altogether.  Note that ML proof scripts may still be used
  4978 with Isar theories; migration is usually quite simple with the ML
  4979 function use_legacy_bindings.  INCOMPATIBILITY.
  4980 
  4981 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4982 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4983 quotes.
  4984 
  4985 * Theory loader: be more serious about observing the static theory
  4986 header specifications (including optional directories), but not the
  4987 accidental file locations of previously successful loads.  The strict
  4988 update policy of former update_thy is now already performed by
  4989 use_thy, so the former has been removed; use_thys updates several
  4990 theories simultaneously, just as 'imports' within a theory header
  4991 specification, but without merging the results.  Potential
  4992 INCOMPATIBILITY: may need to refine theory headers and commands
  4993 ROOT.ML which depend on load order.
  4994 
  4995 * Theory loader: optional support for content-based file
  4996 identification, instead of the traditional scheme of full physical
  4997 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4998 (cf. the system manual).  The new scheme allows to work with
  4999 non-finished theories in persistent session images, such that source
  5000 files may be moved later on without requiring reloads.
  5001 
  5002 * Theory loader: old-style ML proof scripts being *attached* to a thy
  5003 file (with the same base name as the theory) are considered a legacy
  5004 feature, which will disappear eventually. Even now, the theory loader
  5005 no longer maintains dependencies on such files.
  5006 
  5007 * Syntax: the scope for resolving ambiguities via type-inference is
  5008 now limited to individual terms, instead of whole simultaneous
  5009 specifications as before. This greatly reduces the complexity of the
  5010 syntax module and improves flexibility by separating parsing and
  5011 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  5012 'fixes' etc.) are required in rare situations.
  5013 
  5014 * Syntax: constants introduced by new-style packages ('definition',
  5015 'abbreviation' etc.) are passed through the syntax module in
  5016 ``authentic mode''. This means that associated mixfix annotations
  5017 really stick to such constants, independently of potential name space
  5018 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  5019 trees are represented slightly differently, may need to adapt syntax
  5020 translations accordingly. Use CONST marker in 'translations' and
  5021 @{const_syntax} antiquotation in 'parse_translation' etc.
  5022 
  5023 * Legacy goal package: reduced interface to the bare minimum required
  5024 to keep existing proof scripts running.  Most other user-level
  5025 functions are now part of the OldGoals structure, which is *not* open
  5026 by default (consider isatool expandshort before open OldGoals).
  5027 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  5028 these tend to cause confusion about the actual goal (!) context being
  5029 used here, which is not necessarily the same as the_context().
  5030 
  5031 * Command 'find_theorems': supports "*" wild-card in "name:"
  5032 criterion; "with_dups" option.  Certain ProofGeneral versions might
  5033 support a specific search form (see ProofGeneral/CHANGES).
  5034 
  5035 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  5036 by default, which means that "prems" (and also "fixed variables") are
  5037 suppressed from proof state output.  Note that the ProofGeneral
  5038 settings mechanism allows to change and save options persistently, but
  5039 older versions of Isabelle will fail to start up if a negative prems
  5040 limit is imposed.
  5041 
  5042 * Local theory targets may be specified by non-nested blocks of
  5043 ``context/locale/class ... begin'' followed by ``end''.  The body may
  5044 contain definitions, theorems etc., including any derived mechanism
  5045 that has been implemented on top of these primitives.  This concept
  5046 generalizes the existing ``theorem (in ...)'' towards more versatility
  5047 and scalability.
  5048 
  5049 * Proof General interface: proper undo of final 'end' command;
  5050 discontinued Isabelle/classic mode (ML proof scripts).
  5051 
  5052 
  5053 *** Document preparation ***
  5054 
  5055 * Added antiquotation @{theory name} which prints the given name,
  5056 after checking that it refers to a valid ancestor theory in the
  5057 current context.
  5058 
  5059 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5060 check the given source text as ML type/structure, printing verbatim.
  5061 
  5062 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5063 "c args == rhs" given in the current context.  (Any number of
  5064 arguments may be given on the LHS.)
  5065 
  5066 
  5067 *** Pure ***
  5068 
  5069 * The 'class' package offers a combination of axclass and locale to
  5070 achieve Haskell-like type classes in Isabelle.  Definitions and
  5071 theorems within a class context produce both relative results (with
  5072 implicit parameters according to the locale context), and polymorphic
  5073 constants with qualified polymorphism (according to the class
  5074 context).  Within the body context of a 'class' target, a separate
  5075 syntax layer ("user space type system") takes care of converting
  5076 between global polymorphic consts and internal locale representation.
  5077 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5078 "isatool doc classes" provides a tutorial.
  5079 
  5080 * Generic code generator framework allows to generate executable
  5081 code for ML and Haskell (including Isabelle classes).  A short usage
  5082 sketch:
  5083 
  5084     internal compilation:
  5085         export_code <list of constants (term syntax)> in SML
  5086     writing SML code to a file:
  5087         export_code <list of constants (term syntax)> in SML <filename>
  5088     writing OCaml code to a file:
  5089         export_code <list of constants (term syntax)> in OCaml <filename>
  5090     writing Haskell code to a bunch of files:
  5091         export_code <list of constants (term syntax)> in Haskell <filename>
  5092 
  5093     evaluating closed propositions to True/False using code generation:
  5094         method ``eval''
  5095 
  5096 Reasonable default setup of framework in HOL.
  5097 
  5098 Theorem attributs for selecting and transforming function equations theorems:
  5099 
  5100     [code fun]:        select a theorem as function equation for a specific constant
  5101     [code fun del]:    deselect a theorem as function equation for a specific constant
  5102     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5103     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5104 
  5105 User-defined serializations (target in {SML, OCaml, Haskell}):
  5106 
  5107     code_const <and-list of constants (term syntax)>
  5108       {(target) <and-list of const target syntax>}+
  5109 
  5110     code_type <and-list of type constructors>
  5111       {(target) <and-list of type target syntax>}+
  5112 
  5113     code_instance <and-list of instances>
  5114       {(target)}+
  5115         where instance ::= <type constructor> :: <class>
  5116 
  5117     code_class <and_list of classes>
  5118       {(target) <and-list of class target syntax>}+
  5119         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5120 
  5121 code_instance and code_class only are effective to target Haskell.
  5122 
  5123 For example usage see src/HOL/ex/Codegenerator.thy and
  5124 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5125 generation from Isabelle/HOL theories is available via "isatool doc
  5126 codegen".
  5127 
  5128 * Code generator: consts in 'consts_code' Isar commands are now
  5129 referred to by usual term syntax (including optional type
  5130 annotations).
  5131 
  5132 * Command 'no_translations' removes translation rules from theory
  5133 syntax.
  5134 
  5135 * Overloaded definitions are now actually checked for acyclic
  5136 dependencies.  The overloading scheme is slightly more general than
  5137 that of Haskell98, although Isabelle does not demand an exact
  5138 correspondence to type class and instance declarations.
  5139 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5140 exotic versions of overloading -- at the discretion of the user!
  5141 
  5142 Polymorphic constants are represented via type arguments, i.e. the
  5143 instantiation that matches an instance against the most general
  5144 declaration given in the signature.  For example, with the declaration
  5145 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5146 as c(nat).  Overloading is essentially simultaneous structural
  5147 recursion over such type arguments.  Incomplete specification patterns
  5148 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5149 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5150 Command 'print_theory' outputs the normalized system of recursive
  5151 equations, see section "definitions".
  5152 
  5153 * Configuration options are maintained within the theory or proof
  5154 context (with name and type bool/int/string), providing a very simple
  5155 interface to a poor-man's version of general context data.  Tools may
  5156 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5157 these values using Config.get etc.  Users may change options via an
  5158 associated attribute of the same name.  This form of context
  5159 declaration works particularly well with commands 'declare' or
  5160 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5161 very easy to avoid global references, which would not observe Isar
  5162 toplevel undo/redo and fail to work with multithreading.
  5163 
  5164 Various global ML references of Pure and HOL have been turned into
  5165 configuration options:
  5166 
  5167   Unify.search_bound		unify_search_bound
  5168   Unify.trace_bound		unify_trace_bound
  5169   Unify.trace_simp		unify_trace_simp
  5170   Unify.trace_types		unify_trace_types
  5171   Simplifier.simp_depth_limit	simp_depth_limit
  5172   Blast.depth_limit		blast_depth_limit
  5173   DatatypeProp.dtK		datatype_distinctness_limit
  5174   fast_arith_neq_limit  	fast_arith_neq_limit
  5175   fast_arith_split_limit	fast_arith_split_limit
  5176 
  5177 * Named collections of theorems may be easily installed as context
  5178 data using the functor NamedThmsFun (see also
  5179 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5180 attributes; there is also a toplevel print command.  This facility is
  5181 just a common case of general context data, which is the preferred way
  5182 for anything more complex than just a list of facts in canonical
  5183 order.
  5184 
  5185 * Isar: command 'declaration' augments a local theory by generic
  5186 declaration functions written in ML.  This enables arbitrary content
  5187 being added to the context, depending on a morphism that tells the
  5188 difference of the original declaration context wrt. the application
  5189 context encountered later on.
  5190 
  5191 * Isar: proper interfaces for simplification procedures.  Command
  5192 'simproc_setup' declares named simprocs (with match patterns, and body
  5193 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5194 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5195 
  5196 * Isar: an extra pair of brackets around attribute declarations
  5197 abbreviates a theorem reference involving an internal dummy fact,
  5198 which will be ignored later --- only the effect of the attribute on
  5199 the background context will persist.  This form of in-place
  5200 declarations is particularly useful with commands like 'declare' and
  5201 'using', for example ``have A using [[simproc a]] by simp''.
  5202 
  5203 * Isar: method "assumption" (and implicit closing of subproofs) now
  5204 takes simple non-atomic goal assumptions into account: after applying
  5205 an assumption as a rule the resulting subgoals are solved by atomic
  5206 assumption steps.  This is particularly useful to finish 'obtain'
  5207 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5208 without referring to the original premise "!!x. P x ==> thesis" in the
  5209 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5210 more permissive.
  5211 
  5212 * Isar: implicit use of prems from the Isar proof context is
  5213 considered a legacy feature.  Common applications like ``have A .''
  5214 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5215 referencing facts explicitly here improves readability and
  5216 maintainability of proof texts.
  5217 
  5218 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5219 the obtained context from the course of reasoning!  For example:
  5220 
  5221   assume "EX x y. A x & B y"   -- "any previous fact"
  5222   then guess x and y by clarify
  5223 
  5224 This technique is potentially adventurous, depending on the facts and
  5225 proof tools being involved here.
  5226 
  5227 * Isar: known facts from the proof context may be specified as literal
  5228 propositions, using ASCII back-quote syntax.  This works wherever
  5229 named facts used to be allowed so far, in proof commands, proof
  5230 methods, attributes etc.  Literal facts are retrieved from the context
  5231 according to unification of type and term parameters.  For example,
  5232 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5233 theorems in the current context, then these are valid literal facts:
  5234 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5235 
  5236 There is also a proof method "fact" which does the same composition
  5237 for explicit goal states, e.g. the following proof texts coincide with
  5238 certain special cases of literal facts:
  5239 
  5240   have "A" by fact                 ==  note `A`
  5241   have "A ==> B" by fact           ==  note `A ==> B`
  5242   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5243   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5244 
  5245 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5246 outer syntax.  Thus symbolic identifiers may be used without
  5247 additional white space in declarations like this: ``assume *: A''.
  5248 
  5249 * Isar: 'print_facts' prints all local facts of the current context,
  5250 both named and unnamed ones.
  5251 
  5252 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5253 
  5254   def x == "t" and y == "u"
  5255 
  5256 * Isar: added command 'unfolding', which is structurally similar to
  5257 'using', but affects both the goal state and facts by unfolding given
  5258 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5259 'unfolded' attribute may be replaced by first-class proof text.
  5260 
  5261 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5262 and command 'unfolding' now all support object-level equalities
  5263 (potentially conditional).  The underlying notion of rewrite rule is
  5264 analogous to the 'rule_format' attribute, but *not* that of the
  5265 Simplifier (which is usually more generous).
  5266 
  5267 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5268 premises of a theorem by n. Useful in conjunction with drule.
  5269 
  5270 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5271 method expression within a sandbox consisting of the first N
  5272 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5273 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5274 simplifies all new goals that emerge from applying rule foo to the
  5275 originally first one.
  5276 
  5277 * Isar: schematic goals are no longer restricted to higher-order
  5278 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5279 expected.
  5280 
  5281 * Isar: the conclusion of a long theorem statement is now either
  5282 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5283 (essentially a disjunction of cases with local parameters and
  5284 assumptions).  The latter allows to express general elimination rules
  5285 adequately; in this notation common elimination rules look like this:
  5286 
  5287   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5288     assumes "EX x. P x"
  5289     obtains x where "P x"
  5290 
  5291   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5292     assumes "A & B"
  5293     obtains A and B
  5294 
  5295   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5296     assumes "A | B"
  5297     obtains
  5298       A
  5299     | B
  5300 
  5301 The subsequent classical rules even refer to the formal "thesis"
  5302 explicitly:
  5303 
  5304   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5305     obtains "~ thesis"
  5306 
  5307   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5308     obtains "thesis ==> something"
  5309 
  5310 The actual proof of an 'obtains' statement is analogous to that of the
  5311 Isar proof element 'obtain', only that there may be several cases.
  5312 Optional case names may be specified in parentheses; these will be
  5313 available both in the present proof and as annotations in the
  5314 resulting rule, for later use with the 'cases' method (cf. attribute
  5315 case_names).
  5316 
  5317 * Isar: the assumptions of a long theorem statement are available as
  5318 "assms" fact in the proof context.  This is more appropriate than the
  5319 (historical) "prems", which refers to all assumptions of the current
  5320 context, including those from the target locale, proof body etc.
  5321 
  5322 * Isar: 'print_statement' prints theorems from the current theory or
  5323 proof context in long statement form, according to the syntax of a
  5324 top-level lemma.
  5325 
  5326 * Isar: 'obtain' takes an optional case name for the local context
  5327 introduction rule (default "that").
  5328 
  5329 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5330 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5331 to occur.
  5332 
  5333 * Pure: syntax "CONST name" produces a fully internalized constant
  5334 according to the current context.  This is particularly useful for
  5335 syntax translations that should refer to internal constant
  5336 representations independently of name spaces.
  5337 
  5338 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5339 instead of "FOO ". This allows multiple binder declarations to coexist
  5340 in the same context.  INCOMPATIBILITY.
  5341 
  5342 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5343 primitive that also works in a locale context (both for constants and
  5344 fixed variables). Type declaration and internal syntactic representation
  5345 of given constants retrieved from the context. Likewise, the
  5346 'no_notation' command allows to remove given syntax annotations from the
  5347 current context.
  5348 
  5349 * Isar/locales: new derived specification elements 'axiomatization',
  5350 'definition', 'abbreviation', which support type-inference, admit
  5351 object-level specifications (equality, equivalence).  See also the
  5352 isar-ref manual.  Examples:
  5353 
  5354   axiomatization
  5355     eq  (infix "===" 50) where
  5356     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5357 
  5358   definition "f x y = x + y + 1"
  5359   definition g where "g x = f x x"
  5360 
  5361   abbreviation
  5362     neq  (infix "=!=" 50) where
  5363     "x =!= y == ~ (x === y)"
  5364 
  5365 These specifications may be also used in a locale context.  Then the
  5366 constants being introduced depend on certain fixed parameters, and the
  5367 constant name is qualified by the locale base name.  An internal
  5368 abbreviation takes care for convenient input and output, making the
  5369 parameters implicit and using the original short name.  See also
  5370 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5371 entities from a monomorphic theory.
  5372 
  5373 Presently, abbreviations are only available 'in' a target locale, but
  5374 not inherited by general import expressions.  Also note that
  5375 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5376 'translations' in common applications.  The "no_abbrevs" print mode
  5377 prevents folding of abbreviations in term output.
  5378 
  5379 Concrete syntax is attached to specified constants in internal form,
  5380 independently of name spaces.  The parse tree representation is
  5381 slightly different -- use 'notation' instead of raw 'syntax', and
  5382 'translations' with explicit "CONST" markup to accommodate this.
  5383 
  5384 * Pure/Isar: unified syntax for new-style specification mechanisms
  5385 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5386 full type inference and dummy patterns ("_").  For example:
  5387 
  5388   definition "K x _ = x"
  5389 
  5390   inductive conj for A B
  5391   where "A ==> B ==> conj A B"
  5392 
  5393 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5394 the current context.  Print mode "no_abbrevs" prevents inversion of
  5395 abbreviations on output.
  5396 
  5397 * Isar/locales: improved parameter handling: use of locales "var" and
  5398 "struct" no longer necessary; - parameter renamings are no longer
  5399 required to be injective.  For example, this allows to define
  5400 endomorphisms as locale endom = homom mult mult h.
  5401 
  5402 * Isar/locales: changed the way locales with predicates are defined.
  5403 Instead of accumulating the specification, the imported expression is
  5404 now an interpretation.  INCOMPATIBILITY: different normal form of
  5405 locale expressions.  In particular, in interpretations of locales with
  5406 predicates, goals repesenting already interpreted fragments are not
  5407 removed automatically.  Use methods `intro_locales' and
  5408 `unfold_locales'; see below.
  5409 
  5410 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5411 provide backward reasoning on locales predicates.  The methods are
  5412 aware of interpretations and discharge corresponding goals.
  5413 `intro_locales' is less aggressive then `unfold_locales' and does not
  5414 unfold predicates to assumptions.
  5415 
  5416 * Isar/locales: the order in which locale fragments are accumulated
  5417 has changed.  This enables to override declarations from fragments due
  5418 to interpretations -- for example, unwanted simp rules.
  5419 
  5420 * Isar/locales: interpretation in theories and proof contexts has been
  5421 extended.  One may now specify (and prove) equations, which are
  5422 unfolded in interpreted theorems.  This is useful for replacing
  5423 defined concepts (constants depending on locale parameters) by
  5424 concepts already existing in the target context.  Example:
  5425 
  5426   interpretation partial_order ["op <= :: [int, int] => bool"]
  5427     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5428 
  5429 Typically, the constant `partial_order.less' is created by a
  5430 definition specification element in the context of locale
  5431 partial_order.
  5432 
  5433 * Method "induct": improved internal context management to support
  5434 local fixes and defines on-the-fly. Thus explicit meta-level
  5435 connectives !!  and ==> are rarely required anymore in inductive goals
  5436 (using object-logic connectives for this purpose has been long
  5437 obsolete anyway). Common proof patterns are explained in
  5438 src/HOL/Induct/Common_Patterns.thy, see also
  5439 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5440 examples.
  5441 
  5442 * Method "induct": improved handling of simultaneous goals. Instead of
  5443 introducing object-level conjunction, the statement is now split into
  5444 several conclusions, while the corresponding symbolic cases are nested
  5445 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5446 see src/HOL/Induct/Common_Patterns.thy, for example.
  5447 
  5448 * Method "induct": mutual induction rules are now specified as a list
  5449 of rule sharing the same induction cases. HOL packages usually provide
  5450 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5451 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5452 mutual induction rules differently, i.e. like this:
  5453 
  5454   (induct rule: foo_bar.inducts)
  5455   (induct set: foo bar)
  5456   (induct pred: foo bar)
  5457   (induct type: foo bar)
  5458 
  5459 The ML function ProjectRule.projections turns old-style rules into the
  5460 new format.
  5461 
  5462 * Method "coinduct": dual of induction, see
  5463 src/HOL/Library/Coinductive_List.thy for various examples.
  5464 
  5465 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5466 considered a legacy feature.
  5467 
  5468 * Attribute "symmetric" produces result with standardized schematic
  5469 variables (index 0).  Potential INCOMPATIBILITY.
  5470 
  5471 * Simplifier: by default the simplifier trace only shows top level
  5472 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5473 default. Thus there is less danger of being flooded by the trace. The
  5474 trace indicates where parts have been suppressed.
  5475 
  5476 * Provers/classical: removed obsolete classical version of elim_format
  5477 attribute; classical elim/dest rules are now treated uniformly when
  5478 manipulating the claset.
  5479 
  5480 * Provers/classical: stricter checks to ensure that supplied intro,
  5481 dest and elim rules are well-formed; dest and elim rules must have at
  5482 least one premise.
  5483 
  5484 * Provers/classical: attributes dest/elim/intro take an optional
  5485 weight argument for the rule (just as the Pure versions).  Weights are
  5486 ignored by automated tools, but determine the search order of single
  5487 rule steps.
  5488 
  5489 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5490 where the body does not mention the bound variable.  Note that dummy
  5491 patterns implicitly depend on their context of bounds, which makes
  5492 "{_. _}" match any set comprehension as expected.  Potential
  5493 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5494 constant "_idtdummy" in the binding position.
  5495 
  5496 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5497 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5498 for example "A -> B" => "Pi A (%_. B)".
  5499 
  5500 * Pure: 'class_deps' command visualizes the subclass relation, using
  5501 the graph browser tool.
  5502 
  5503 * Pure: 'print_theory' now suppresses certain internal declarations by
  5504 default; use '!' option for full details.
  5505 
  5506 
  5507 *** HOL ***
  5508 
  5509 * Method "metis" proves goals by applying the Metis general-purpose
  5510 resolution prover (see also http://gilith.com/software/metis/).
  5511 Examples are in the directory MetisExamples.  WARNING: the
  5512 Isabelle/HOL-Metis integration does not yet work properly with
  5513 multi-threading.
  5514 
  5515 * Command 'sledgehammer' invokes external automatic theorem provers as
  5516 background processes.  It generates calls to the "metis" method if
  5517 successful. These can be pasted into the proof.  Users do not have to
  5518 wait for the automatic provers to return.  WARNING: does not really
  5519 work with multi-threading.
  5520 
  5521 * New "auto_quickcheck" feature tests outermost goal statements for
  5522 potential counter-examples.  Controlled by ML references
  5523 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5524 5000 milliseconds).  Fails silently if statements is outside of
  5525 executable fragment, or any other codgenerator problem occurs.
  5526 
  5527 * New constant "undefined" with axiom "undefined x = undefined".
  5528 
  5529 * Added class "HOL.eq", allowing for code generation with polymorphic
  5530 equality.
  5531 
  5532 * Some renaming of class constants due to canonical name prefixing in
  5533 the new 'class' package:
  5534 
  5535     HOL.abs ~> HOL.abs_class.abs
  5536     HOL.divide ~> HOL.divide_class.divide
  5537     0 ~> HOL.zero_class.zero
  5538     1 ~> HOL.one_class.one
  5539     op + ~> HOL.plus_class.plus
  5540     op - ~> HOL.minus_class.minus
  5541     uminus ~> HOL.minus_class.uminus
  5542     op * ~> HOL.times_class.times
  5543     op < ~> HOL.ord_class.less
  5544     op <= > HOL.ord_class.less_eq
  5545     Nat.power ~> Power.power_class.power
  5546     Nat.size ~> Nat.size_class.size
  5547     Numeral.number_of ~> Numeral.number_class.number_of
  5548     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5549     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5550     Orderings.min ~> Orderings.ord_class.min
  5551     Orderings.max ~> Orderings.ord_class.max
  5552     Divides.op div ~> Divides.div_class.div
  5553     Divides.op mod ~> Divides.div_class.mod
  5554     Divides.op dvd ~> Divides.div_class.dvd
  5555 
  5556 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5557 
  5558 a) User-defined constants using any of the names "plus", "minus",
  5559 "times", "less" or "less_eq". The standard syntax translations for
  5560 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5561 names.
  5562 
  5563 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5564 INCOMPATIBILITY: use more specific names.
  5565 
  5566 c) Permutative equations (e.g. "a + b = b + a")
  5567 Since the change of names also changes the order of terms, permutative
  5568 rewrite rules may get applied in a different order. Experience shows
  5569 that this is rarely the case (only two adaptions in the whole Isabelle
  5570 distribution).  INCOMPATIBILITY: rewrite proofs
  5571 
  5572 d) ML code directly refering to constant names
  5573 This in general only affects hand-written proof tactics, simprocs and
  5574 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5575 Consider using @{const_name} antiquotation.
  5576 
  5577 * New class "default" with associated constant "default".
  5578 
  5579 * Function "sgn" is now overloaded and available on int, real, complex
  5580 (and other numeric types), using class "sgn".  Two possible defs of
  5581 sgn are given as equational assumptions in the classes sgn_if and
  5582 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5583 INCOMPATIBILITY.
  5584 
  5585 * Locale "partial_order" now unified with class "order" (cf. theory
  5586 Orderings), added parameter "less".  INCOMPATIBILITY.
  5587 
  5588 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5589 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5590 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5591 
  5592 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5593 changed order of proof goals in instance proofs.
  5594 
  5595 * The transitivity reasoner for partial and linear orders is set up
  5596 for classes "order" and "linorder".  Instances of the reasoner are available
  5597 in all contexts importing or interpreting the corresponding locales.
  5598 Method "order" invokes the reasoner separately; the reasoner
  5599 is also integrated with the Simplifier as a solver.  Diagnostic
  5600 command 'print_orders' shows the available instances of the reasoner
  5601 in the current context.
  5602 
  5603 * Localized monotonicity predicate in theory "Orderings"; integrated
  5604 lemmas max_of_mono and min_of_mono with this predicate.
  5605 INCOMPATIBILITY.
  5606 
  5607 * Formulation of theorem "dense" changed slightly due to integration
  5608 with new class dense_linear_order.
  5609 
  5610 * Uniform lattice theory development in HOL.
  5611 
  5612     constants "meet" and "join" now named "inf" and "sup"
  5613     constant "Meet" now named "Inf"
  5614 
  5615     classes "meet_semilorder" and "join_semilorder" now named
  5616       "lower_semilattice" and "upper_semilattice"
  5617     class "lorder" now named "lattice"
  5618     class "comp_lat" now named "complete_lattice"
  5619 
  5620     Instantiation of lattice classes allows explicit definitions
  5621     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5622 
  5623   INCOMPATIBILITY.  Theorem renames:
  5624 
  5625     meet_left_le            ~> inf_le1
  5626     meet_right_le           ~> inf_le2
  5627     join_left_le            ~> sup_ge1
  5628     join_right_le           ~> sup_ge2
  5629     meet_join_le            ~> inf_sup_ord
  5630     le_meetI                ~> le_infI
  5631     join_leI                ~> le_supI
  5632     le_meet                 ~> le_inf_iff
  5633     le_join                 ~> ge_sup_conv
  5634     meet_idempotent         ~> inf_idem
  5635     join_idempotent         ~> sup_idem
  5636     meet_comm               ~> inf_commute
  5637     join_comm               ~> sup_commute
  5638     meet_leI1               ~> le_infI1
  5639     meet_leI2               ~> le_infI2
  5640     le_joinI1               ~> le_supI1
  5641     le_joinI2               ~> le_supI2
  5642     meet_assoc              ~> inf_assoc
  5643     join_assoc              ~> sup_assoc
  5644     meet_left_comm          ~> inf_left_commute
  5645     meet_left_idempotent    ~> inf_left_idem
  5646     join_left_comm          ~> sup_left_commute
  5647     join_left_idempotent    ~> sup_left_idem
  5648     meet_aci                ~> inf_aci
  5649     join_aci                ~> sup_aci
  5650     le_def_meet             ~> le_iff_inf
  5651     le_def_join             ~> le_iff_sup
  5652     join_absorp2            ~> sup_absorb2
  5653     join_absorp1            ~> sup_absorb1
  5654     meet_absorp1            ~> inf_absorb1
  5655     meet_absorp2            ~> inf_absorb2
  5656     meet_join_absorp        ~> inf_sup_absorb
  5657     join_meet_absorp        ~> sup_inf_absorb
  5658     distrib_join_le         ~> distrib_sup_le
  5659     distrib_meet_le         ~> distrib_inf_le
  5660 
  5661     add_meet_distrib_left   ~> add_inf_distrib_left
  5662     add_join_distrib_left   ~> add_sup_distrib_left
  5663     is_join_neg_meet        ~> is_join_neg_inf
  5664     is_meet_neg_join        ~> is_meet_neg_sup
  5665     add_meet_distrib_right  ~> add_inf_distrib_right
  5666     add_join_distrib_right  ~> add_sup_distrib_right
  5667     add_meet_join_distribs  ~> add_sup_inf_distribs
  5668     join_eq_neg_meet        ~> sup_eq_neg_inf
  5669     meet_eq_neg_join        ~> inf_eq_neg_sup
  5670     add_eq_meet_join        ~> add_eq_inf_sup
  5671     meet_0_imp_0            ~> inf_0_imp_0
  5672     join_0_imp_0            ~> sup_0_imp_0
  5673     meet_0_eq_0             ~> inf_0_eq_0
  5674     join_0_eq_0             ~> sup_0_eq_0
  5675     neg_meet_eq_join        ~> neg_inf_eq_sup
  5676     neg_join_eq_meet        ~> neg_sup_eq_inf
  5677     join_eq_if              ~> sup_eq_if
  5678 
  5679     mono_meet               ~> mono_inf
  5680     mono_join               ~> mono_sup
  5681     meet_bool_eq            ~> inf_bool_eq
  5682     join_bool_eq            ~> sup_bool_eq
  5683     meet_fun_eq             ~> inf_fun_eq
  5684     join_fun_eq             ~> sup_fun_eq
  5685     meet_set_eq             ~> inf_set_eq
  5686     join_set_eq             ~> sup_set_eq
  5687     meet1_iff               ~> inf1_iff
  5688     meet2_iff               ~> inf2_iff
  5689     meet1I                  ~> inf1I
  5690     meet2I                  ~> inf2I
  5691     meet1D1                 ~> inf1D1
  5692     meet2D1                 ~> inf2D1
  5693     meet1D2                 ~> inf1D2
  5694     meet2D2                 ~> inf2D2
  5695     meet1E                  ~> inf1E
  5696     meet2E                  ~> inf2E
  5697     join1_iff               ~> sup1_iff
  5698     join2_iff               ~> sup2_iff
  5699     join1I1                 ~> sup1I1
  5700     join2I1                 ~> sup2I1
  5701     join1I1                 ~> sup1I1
  5702     join2I2                 ~> sup1I2
  5703     join1CI                 ~> sup1CI
  5704     join2CI                 ~> sup2CI
  5705     join1E                  ~> sup1E
  5706     join2E                  ~> sup2E
  5707 
  5708     is_meet_Meet            ~> is_meet_Inf
  5709     Meet_bool_def           ~> Inf_bool_def
  5710     Meet_fun_def            ~> Inf_fun_def
  5711     Meet_greatest           ~> Inf_greatest
  5712     Meet_lower              ~> Inf_lower
  5713     Meet_set_def            ~> Inf_set_def
  5714 
  5715     Sup_def                 ~> Sup_Inf
  5716     Sup_bool_eq             ~> Sup_bool_def
  5717     Sup_fun_eq              ~> Sup_fun_def
  5718     Sup_set_eq              ~> Sup_set_def
  5719 
  5720     listsp_meetI            ~> listsp_infI
  5721     listsp_meet_eq          ~> listsp_inf_eq
  5722 
  5723     meet_min                ~> inf_min
  5724     join_max                ~> sup_max
  5725 
  5726 * Added syntactic class "size"; overloaded constant "size" now has
  5727 type "'a::size ==> bool"
  5728 
  5729 * Internal reorganisation of `size' of datatypes: size theorems
  5730 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5731 simplification rules by default!); theorems "prod.size" now named
  5732 "*.size".
  5733 
  5734 * Class "div" now inherits from class "times" rather than "type".
  5735 INCOMPATIBILITY.
  5736 
  5737 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5738 Linorder etc.  have disappeared; operations defined in terms of
  5739 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5740 
  5741 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5742 
  5743 * HOL-Word: New extensive library and type for generic, fixed size
  5744 machine words, with arithmetic, bit-wise, shifting and rotating
  5745 operations, reflection into int, nat, and bool lists, automation for
  5746 linear arithmetic (by automatic reflection into nat or int), including
  5747 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5748 arithmetic type classes, supporting automatic simplification of
  5749 numerals on all operations.
  5750 
  5751 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5752 
  5753 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5754 
  5755 * Code generator library theories:
  5756   - Code_Integer represents HOL integers by big integer literals in target
  5757     languages.
  5758   - Code_Char represents HOL characters by character literals in target
  5759     languages.
  5760   - Code_Char_chr like Code_Char, but also offers treatment of character
  5761     codes; includes Code_Integer.
  5762   - Executable_Set allows to generate code for finite sets using lists.
  5763   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5764     denominator).
  5765   - Executable_Real implements a subset of real numbers, namly those
  5766     representable by rational numbers.
  5767   - Efficient_Nat implements natural numbers by integers, which in general will
  5768     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5769     includes Code_Integer.
  5770   - Code_Index provides an additional datatype index which is mapped to
  5771     target-language built-in integers.
  5772   - Code_Message provides an additional datatype message_string which is isomorphic to
  5773     strings; messages are mapped to target-language strings.
  5774 
  5775 * New package for inductive predicates
  5776 
  5777   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5778 
  5779     inductive
  5780       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5781       for z_1 :: U_1 and ... and z_n :: U_m
  5782     where
  5783       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5784     | ...
  5785 
  5786   with full support for type-inference, rather than
  5787 
  5788     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5789 
  5790     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5791     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5792 
  5793     inductive "s z_1 ... z_m"
  5794     intros
  5795       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5796       ...
  5797 
  5798   For backward compatibility, there is a wrapper allowing inductive
  5799   sets to be defined with the new package via
  5800 
  5801     inductive_set
  5802       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5803       for z_1 :: U_1 and ... and z_n :: U_m
  5804     where
  5805       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5806     | ...
  5807 
  5808   or
  5809 
  5810     inductive_set
  5811       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5812       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5813       for z_1 :: U_1 and ... and z_n :: U_m
  5814     where
  5815       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5816     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5817     | ...
  5818 
  5819   if the additional syntax "p ..." is required.
  5820 
  5821   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5822   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5823 
  5824   INCOMPATIBILITIES:
  5825 
  5826   - Since declaration and definition of inductive sets or predicates
  5827     is no longer separated, abbreviations involving the newly
  5828     introduced sets or predicates must be specified together with the
  5829     introduction rules after the 'where' keyword (see above), rather
  5830     than before the actual inductive definition.
  5831 
  5832   - The variables in induction and elimination rules are now
  5833     quantified in the order of their occurrence in the introduction
  5834     rules, rather than in alphabetical order. Since this may break
  5835     some proofs, these proofs either have to be repaired, e.g. by
  5836     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5837     statements of the form
  5838 
  5839       case (rule_i a_i_1 ... a_i_{k_i})
  5840 
  5841     or the old order of quantification has to be restored by explicitly adding
  5842     meta-level quantifiers in the introduction rules, i.e.
  5843 
  5844       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5845 
  5846   - The format of the elimination rules is now
  5847 
  5848       p z_1 ... z_m x_1 ... x_n ==>
  5849         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5850         ==> ... ==> P
  5851 
  5852     for predicates and
  5853 
  5854       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5855         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5856         ==> ... ==> P
  5857 
  5858     for sets rather than
  5859 
  5860       x : s z_1 ... z_m ==>
  5861         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5862         ==> ... ==> P
  5863 
  5864     This may require terms in goals to be expanded to n-tuples
  5865     (e.g. using case_tac or simplification with the split_paired_all
  5866     rule) before the above elimination rule is applicable.
  5867 
  5868   - The elimination or case analysis rules for (mutually) inductive
  5869     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5870     list of rules "p_1_..._p_k.elims" is no longer available.
  5871 
  5872 * New package "function"/"fun" for general recursive functions,
  5873 supporting mutual and nested recursion, definitions in local contexts,
  5874 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5875 for small examples, and the separate tutorial on the function
  5876 package. The old recdef "package" is still available as before, but
  5877 users are encouraged to use the new package.
  5878 
  5879 * Method "lexicographic_order" automatically synthesizes termination
  5880 relations as lexicographic combinations of size measures.
  5881 
  5882 * Case-expressions allow arbitrary constructor-patterns (including
  5883 "_") and take their order into account, like in functional
  5884 programming.  Internally, this is translated into nested
  5885 case-expressions; missing cases are added and mapped to the predefined
  5886 constant "undefined". In complicated cases printing may no longer show
  5887 the original input but the internal form. Lambda-abstractions allow
  5888 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5889 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5890 variable.
  5891 
  5892 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5893 is an abbreviation for "of_nat :: nat => int". The simplification
  5894 rules for "of_nat" have been changed to work like "int" did
  5895 previously.  Potential INCOMPATIBILITY:
  5896   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5897   - of_nat_diff and of_nat_mult are no longer default simp rules
  5898 
  5899 * Method "algebra" solves polynomial equations over (semi)rings using
  5900 Groebner bases. The (semi)ring structure is defined by locales and the
  5901 tool setup depends on that generic context. Installing the method for
  5902 a specific type involves instantiating the locale and possibly adding
  5903 declarations for computation on the coefficients.  The method is
  5904 already instantiated for natural numbers and for the axiomatic class
  5905 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5906 CALCULEMUS 2007 for the general principles underlying this
  5907 architecture of context-aware proof-tools.
  5908 
  5909 * Method "ferrack" implements quantifier elimination over
  5910 special-purpose dense linear orders using locales (analogous to
  5911 "algebra"). The method is already installed for class
  5912 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5913 rat, etc.
  5914 
  5915 * Former constant "List.op @" now named "List.append".  Use ML
  5916 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5917 circumvent possible incompatibilities when working on ML level.
  5918 
  5919 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5920 
  5921 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5922 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5923 
  5924 * New syntax for Haskell-like list comprehension (input only), eg.
  5925 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5926 
  5927 * The special syntax for function "filter" has changed from [x :
  5928 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5929 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5930 
  5931 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5932 defined on nat only, as an abbreviation for [a..<Suc b]
  5933 INCOMPATIBILITY.
  5934 
  5935 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5936 
  5937 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5938 
  5939 * New lemma collection field_simps (an extension of ring_simps) for
  5940 manipulating (in)equations involving division. Multiplies with all
  5941 denominators that can be proved to be non-zero (in equations) or
  5942 positive/negative (in inequations).
  5943 
  5944 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5945 have been improved and renamed to ring_simps, group_simps and
  5946 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5947 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5948 
  5949 * Theory Library/Commutative_Ring: switched from recdef to function
  5950 package; constants add, mul, pow now curried.  Infix syntax for
  5951 algebraic operations.
  5952 
  5953 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5954 INCOMPATIBILITY.
  5955 
  5956 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5957 INCOMPATIBILITY.
  5958 
  5959 * HOL/records: generalised field-update to take a function on the
  5960 field rather than the new value: r(|A := x|) is translated to A_update
  5961 (K x) r The K-combinator that is internally used is called K_record.
  5962 INCOMPATIBILITY: Usage of the plain update functions has to be
  5963 adapted.
  5964 
  5965 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5966 * x = 0, which are required for a semiring.  Richer structures do not
  5967 inherit from semiring_0 anymore, because this property is a theorem
  5968 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5969 there is more to prove, but this is mostly trivial.
  5970 
  5971 * Class "recpower" is generalized to arbitrary monoids, not just
  5972 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5973 commutativity or semiring properties additionally.
  5974 
  5975 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5976 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5977 better use 'abbreviation'.
  5978 
  5979 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5980 
  5981 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5982 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5983 significant changes for setting up numeral syntax for types:
  5984   - New constants Numeral.pred and Numeral.succ instead
  5985       of former Numeral.bin_pred and Numeral.bin_succ.
  5986   - Use integer operations instead of bin_add, bin_mult and so on.
  5987   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5988   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5989 
  5990 See src/HOL/Integ/IntArith.thy for an example setup.
  5991 
  5992 * Command 'normal_form' computes the normal form of a term that may
  5993 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5994 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5995 heavy-duty computations because the functions are compiled to ML
  5996 first.  Correspondingly, a method "normalization" is provided.  See
  5997 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5998 
  5999 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  6000 25 like -->); output depends on the "iff" print_mode, the default is
  6001 "A = B" (with priority 50).
  6002 
  6003 * Relations less (<) and less_eq (<=) are also available on type bool.
  6004 Modified syntax to disallow nesting without explicit parentheses,
  6005 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  6006 INCOMPATIBILITY.
  6007 
  6008 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  6009 
  6010 * Relation composition operator "op O" now has precedence 75 and binds
  6011 stronger than union and intersection. INCOMPATIBILITY.
  6012 
  6013 * The old set interval syntax "{m..n(}" (and relatives) has been
  6014 removed.  Use "{m..<n}" (and relatives) instead.
  6015 
  6016 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  6017 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  6018 ``declare [[simproc del: neq]]''.
  6019 
  6020 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  6021 True/False.
  6022 
  6023 * Theorem Cons_eq_map_conv no longer declared as "simp".
  6024 
  6025 * Theorem setsum_mult renamed to setsum_right_distrib.
  6026 
  6027 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  6028 ``rule'' method.
  6029 
  6030 * Reimplemented methods "sat" and "satx", with several improvements:
  6031 goals no longer need to be stated as "<prems> ==> False", equivalences
  6032 (i.e. "=" on type bool) are handled, variable names of the form
  6033 "lit_<n>" are no longer reserved, significant speedup.
  6034 
  6035 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  6036 zChaff is still supported as well.
  6037 
  6038 * 'inductive' and 'datatype': provide projections of mutual rules,
  6039 bundled as foo_bar.inducts;
  6040 
  6041 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  6042 Library.
  6043 
  6044 * Library: moved theory Accessible_Part to main HOL.
  6045 
  6046 * Library: added theory Coinductive_List of potentially infinite lists
  6047 as greatest fixed-point.
  6048 
  6049 * Library: added theory AssocList which implements (finite) maps as
  6050 association lists.
  6051 
  6052 * Method "evaluation" solves goals (i.e. a boolean expression)
  6053 efficiently by compiling it to ML.  The goal is "proved" (via an
  6054 oracle) if it evaluates to True.
  6055 
  6056 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  6057 also when invoked by the simplifier.  This results in the Simplifier
  6058 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  6059 Configuration option fast_arith_split_limit=0 recovers the old
  6060 behavior.
  6061 
  6062 * Support for hex (0x20) and binary (0b1001) numerals.
  6063 
  6064 * New method: reify eqs (t), where eqs are equations for an
  6065 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  6066 parameter, computes a term s::'b and a list xs::'a list and proves the
  6067 theorem I xs s = t. This is also known as reification or quoting. The
  6068 resulting theorem is applied to the subgoal to substitute t with I xs
  6069 s.  If t is omitted, the subgoal itself is reified.
  6070 
  6071 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  6072 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  6073 where f is supposed to be a computable function (in the sense of code
  6074 generattion). The method uses reify to compute s and xs as above then
  6075 applies corr_thm and uses normalization by evaluation to "prove" f s =
  6076 r and finally gets the theorem t = r, which is again applied to the
  6077 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  6078 
  6079 * Reflection: Automatic reification now handels binding, an example is
  6080 available in src/HOL/ex/ReflectionEx.thy
  6081 
  6082 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  6083 command 'statespace' that is similar to 'record', but introduces an
  6084 abstract specification based on the locale infrastructure instead of
  6085 HOL types.  This leads to extra flexibility in composing state spaces,
  6086 in particular multiple inheritance and renaming of components.
  6087 
  6088 
  6089 *** HOL-Complex ***
  6090 
  6091 * Hyperreal: Functions root and sqrt are now defined on negative real
  6092 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  6093 Nonnegativity side conditions have been removed from many lemmas, so
  6094 that more subgoals may now be solved by simplification; potential
  6095 INCOMPATIBILITY.
  6096 
  6097 * Real: new type classes formalize real normed vector spaces and
  6098 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  6099 and norm :: 'a => real.
  6100 
  6101 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  6102 reals into other types. The overloaded constant Reals :: 'a set is now
  6103 defined as range of_real; potential INCOMPATIBILITY.
  6104 
  6105 * Real: proper support for ML code generation, including 'quickcheck'.
  6106 Reals are implemented as arbitrary precision rationals.
  6107 
  6108 * Hyperreal: Several constants that previously worked only for the
  6109 reals have been generalized, so they now work over arbitrary vector
  6110 spaces. Type annotations may need to be added in some cases; potential
  6111 INCOMPATIBILITY.
  6112 
  6113   Infinitesimal  :: ('a::real_normed_vector) star set
  6114   HFinite        :: ('a::real_normed_vector) star set
  6115   HInfinite      :: ('a::real_normed_vector) star set
  6116   approx         :: ('a::real_normed_vector) star => 'a star => bool
  6117   monad          :: ('a::real_normed_vector) star => 'a star set
  6118   galaxy         :: ('a::real_normed_vector) star => 'a star set
  6119   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  6120   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  6121   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  6122   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  6123   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  6124   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  6125   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6126   sgn            :: 'a::real_normed_vector => 'a
  6127   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6128 
  6129 * Complex: Some complex-specific constants are now abbreviations for
  6130 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6131 hnorm.  Other constants have been entirely removed in favor of the
  6132 polymorphic versions (INCOMPATIBILITY):
  6133 
  6134   approx        <-- capprox
  6135   HFinite       <-- CFinite
  6136   HInfinite     <-- CInfinite
  6137   Infinitesimal <-- CInfinitesimal
  6138   monad         <-- cmonad
  6139   galaxy        <-- cgalaxy
  6140   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  6141   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  6142   (ns)deriv     <-- (ns)cderiv
  6143 
  6144 
  6145 *** HOL-Algebra ***
  6146 
  6147 * Formalisation of ideals and the quotient construction over rings.
  6148 
  6149 * Order and lattice theory no longer based on records.
  6150 INCOMPATIBILITY.
  6151 
  6152 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  6153 greatest_closed.  INCOMPATIBILITY.
  6154 
  6155 * Method algebra is now set up via an attribute.  For examples see
  6156 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  6157 of algebraic structures.
  6158 
  6159 * Renamed theory CRing to Ring.
  6160 
  6161 
  6162 *** HOL-Nominal ***
  6163 
  6164 * Substantial, yet incomplete support for nominal datatypes (binding
  6165 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  6166 src/HOL/Nominal/Examples.  Prospective users should consult
  6167 http://isabelle.in.tum.de/nominal/
  6168 
  6169 
  6170 *** ML ***
  6171 
  6172 * ML basics: just one true type int, which coincides with IntInf.int
  6173 (even on SML/NJ).
  6174 
  6175 * ML within Isar: antiquotations allow to embed statically-checked
  6176 formal entities in the source, referring to the context available at
  6177 compile-time.  For example:
  6178 
  6179 ML {* @{sort "{zero,one}"} *}
  6180 ML {* @{typ "'a => 'b"} *}
  6181 ML {* @{term "%x. x"} *}
  6182 ML {* @{prop "x == y"} *}
  6183 ML {* @{ctyp "'a => 'b"} *}
  6184 ML {* @{cterm "%x. x"} *}
  6185 ML {* @{cprop "x == y"} *}
  6186 ML {* @{thm asm_rl} *}
  6187 ML {* @{thms asm_rl} *}
  6188 ML {* @{type_name c} *}
  6189 ML {* @{type_syntax c} *}
  6190 ML {* @{const_name c} *}
  6191 ML {* @{const_syntax c} *}
  6192 ML {* @{context} *}
  6193 ML {* @{theory} *}
  6194 ML {* @{theory Pure} *}
  6195 ML {* @{theory_ref} *}
  6196 ML {* @{theory_ref Pure} *}
  6197 ML {* @{simpset} *}
  6198 ML {* @{claset} *}
  6199 ML {* @{clasimpset} *}
  6200 
  6201 The same works for sources being ``used'' within an Isar context.
  6202 
  6203 * ML in Isar: improved error reporting; extra verbosity with
  6204 ML_Context.trace enabled.
  6205 
  6206 * Pure/General/table.ML: the join operations now works via exceptions
  6207 DUP/SAME instead of type option. This is simpler in simple cases, and
  6208 admits slightly more efficient complex applications.
  6209 
  6210 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6211 use Context.generic instead of just theory.
  6212 
  6213 * Pure: datatype Context.generic joins theory/Proof.context and
  6214 provides some facilities for code that works in either kind of
  6215 context, notably GenericDataFun for uniform theory and proof data.
  6216 
  6217 * Pure: simplified internal attribute type, which is now always
  6218 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6219 local (Proof.context) attributes have been discontinued, while
  6220 minimizing code duplication. Thm.rule_attribute and
  6221 Thm.declaration_attribute build canonical attributes; see also structure
  6222 Context for further operations on Context.generic, notably
  6223 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6224 declarations and definitions.
  6225 
  6226 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6227 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6228 init only required for impure data. INCOMPATIBILITY: empty really need
  6229 to be empty (no dependencies on theory content!)
  6230 
  6231 * Pure/kernel: consts certification ignores sort constraints given in
  6232 signature declarations. (This information is not relevant to the
  6233 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6234 potential INCOMPATIBILITY.
  6235 
  6236 * Pure: axiomatic type classes are now purely definitional, with
  6237 explicit proofs of class axioms and super class relations performed
  6238 internally. See Pure/axclass.ML for the main internal interfaces --
  6239 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6240 AxClass.axiomatize_class/classrel/arity supersede
  6241 Sign.add_classes/classrel/arities.