NEWS
author Andreas Lochbihler
Wed Oct 10 13:03:50 2012 +0200 (2012-10-10)
changeset 49770 cf6a78acf445
parent 49739 13aa6d8268ec
child 49822 0cfc1651be25
permissions -rw-r--r--
efficient construction of red black trees from sorted associative lists
     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     . more efficient painting, improved reactivity;
    13     . more robust incremental parsing of outer syntax (partial
    14       comments, malformed symbols);
    15     . smarter handling of tracing messages (via tracing_limit);
    16     . more plugin options and preferences, based on Isabelle/Scala;
    17     . uniform Java 7 platform on Linux, Mac OS X, Windows;
    18 
    19 * Configuration option show_markup controls direct inlining of markup
    20 into the printed representation of formal entities --- notably type
    21 and sort constraints.  This enables Prover IDE users to retrieve that
    22 information via tooltips in the output window, for example.
    23 
    24 * Command 'ML_file' evaluates ML text from a file directly within the
    25 theory, without any predeclaration via 'uses' in the theory header.
    26 
    27 * Old command 'use' command and corresponding keyword 'uses' in the
    28 theory header are legacy features and will be discontinued soon.
    29 Tools that load their additional source files may imitate the
    30 'ML_file' implementation, such that the system can take care of
    31 dependencies properly.
    32 
    33 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    34 is called fastforce / fast_force_tac already since Isabelle2011-1.
    35 
    36 * Updated and extended "isar-ref" manual, reduced remaining material
    37 in old "ref" manual.
    38 
    39 
    40 *** Pure ***
    41 
    42 * Code generation for Haskell: restrict unqualified imports from
    43 Haskell Prelude to a small set of fundamental operations.
    44 
    45 * Command "export_code": relative file names are interpreted
    46 relatively to master directory of current theory rather than
    47 the rather arbitrary current working directory.
    48 INCOMPATIBILITY.
    49 
    50 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    51 use regular rule composition via "OF" / "THEN", or explicit proof
    52 structure instead.  Note that Isabelle/ML provides a variety of
    53 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    54 with some care where this is really required.
    55 
    56 * Command 'typ' supports an additional variant with explicit sort
    57 constraint, to infer and check the most general type conforming to a
    58 given given sort.  Example (in HOL):
    59 
    60   typ "_ * _ * bool * unit" :: finite
    61 
    62 
    63 *** HOL ***
    64 
    65 * Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
    66 
    67 * Class "comm_monoid_diff" formalises properties of bounded
    68 subtraction, with natural numbers and multisets as typical instances.
    69 
    70 * Theory "Library/Option_ord" provides instantiation of option type
    71 to lattice type classes.
    72 
    73 * New combinator "Option.these" with type "'a option set => 'a set".
    74 
    75 * Renamed theory Library/List_Prefix to Library/Sublist.
    76 INCOMPATIBILITY.  Related changes are:
    77 
    78   - Renamed constants:
    79 
    80       prefix ~> prefixeq
    81       strict_prefix ~> prefix
    82 
    83     Renamed lemmas accordingly, INCOMPATIBILITY.
    84 
    85   - Replaced constant "postfix" by "suffixeq" with swapped argument order
    86     (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix
    87     syntax "xs >>= ys"; use "suffixeq ys xs" instead.  Renamed lemmas
    88     accordingly.  INCOMPATIBILITY.
    89 
    90   - New constant "emb" for homeomorphic embedding on lists. New
    91     abbreviation "sub" for special case "emb (op =)".
    92 
    93   - Library/Sublist does no longer provide "order" and "bot" type class
    94     instances for the prefix order (merely corresponding locale
    95     interpretations). The type class instances are to be found in
    96     Library/Prefix_Order. INCOMPATIBILITY.
    97 
    98   - The sublist relation from Library/Sublist_Order is now based on
    99     "Sublist.sub". Replaced lemmas:
   100 
   101       le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff
   102       le_list_append_mono ~> Sublist.emb_append_mono
   103       le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2
   104       le_list_Cons_EX ~> Sublist.emb_ConsD
   105       le_list_drop_Cons2 ~> Sublist.sub_Cons2'
   106       le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq
   107       le_list_drop_Cons ~> Sublist.sub_Cons'
   108       le_list_drop_many ~> Sublist.sub_drop_many
   109       le_list_filter_left ~> Sublist.sub_filter_left
   110       le_list_rev_drop_many ~> Sublist.sub_rev_drop_many
   111       le_list_rev_take_iff ~> Sublist.sub_append
   112       le_list_same_length ~> Sublist.sub_same_length
   113       le_list_take_many_iff ~> Sublist.sub_append'
   114       less_eq_list.drop ~> less_eq_list_drop
   115       less_eq_list.induct ~> less_eq_list_induct
   116       not_le_list_length ~> Sublist.not_sub_length
   117 
   118     INCOMPATIBILITY.
   119 
   120 * HOL/BNF: New (co)datatype package based on bounded natural
   121 functors with support for mixed, nested recursion and interesting
   122 non-free datatypes.
   123 
   124 * HOL/Cardinals: Theories of ordinals and cardinals
   125 (supersedes the AFP entry "Ordinals_and_Cardinals").
   126 
   127 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
   128 execution for code generated towards Isabelle/ML.
   129 
   130 * Library/FinFun.thy: theory of almost everywhere constant functions
   131 (supersedes the AFP entry "Code Generation for Functions as Data").
   132 
   133 * Library/Phantom.thy: generic phantom type to make a type parameter
   134 appear in a constant's type. This alternative to adding TYPE('a) as
   135 another parameter avoids unnecessary closures in generated code.
   136 
   137 * Library/RBT_Impl.thy: efficient construction of red-black trees 
   138 from sorted associative lists. Merging two trees with rbt_union may
   139 return a structurally different tree than before. MINOR INCOMPATIBILITY.
   140 
   141 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   142 expressions.
   143 
   144 * Quickcheck:
   145 
   146   - added an optimisation for equality premises.
   147     It is switched on by default, and can be switched off by setting
   148     the configuration quickcheck_optimise_equality to false.
   149 
   150 * The SMT solver Z3 has now by default a restricted set of directly
   151 supported features. For the full set of features (div/mod, nonlinear
   152 arithmetic, datatypes/records) with potential proof reconstruction
   153 failures, enable the configuration option "z3_with_extensions".
   154 Minor INCOMPATIBILITY.
   155 
   156 * Sledgehammer:
   157 
   158   - Added MaSh relevance filter based on machine-learning; see the
   159     Sledgehammer manual for details.
   160   - Rationalized type encodings ("type_enc" option).
   161   - Renamed "kill_provers" subcommand to "kill"
   162   - Renamed options:
   163       max_relevant ~> max_facts
   164       relevance_thresholds ~> fact_thresholds
   165 
   166 
   167 *** Document preparation ***
   168 
   169 * Default for \<euro> is now based on eurosym package, instead of
   170 slightly exotic babel/greek.
   171 
   172 * Document variant NAME may use different LaTeX entry point
   173 document/root_NAME.tex if that file exists, instead of the common
   174 document/root.tex.
   175 
   176 * Simplified custom document/build script, instead of old-style
   177 document/IsaMakefile.  Minor INCOMPATIBILITY.
   178 
   179 
   180 *** ML ***
   181 
   182 * Renamed Position.str_of to Position.here to emphasize that this is a
   183 formal device to inline positions into message text, but not
   184 necessarily printing visible text.
   185 
   186 
   187 *** System ***
   188 
   189 * The ML system is configured as regular component, and no longer
   190 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   191 for home-made configurations.
   192 
   193 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   194 Minor INCOMPATIBILITY in command-line options.
   195 
   196 * Advanced support for Isabelle sessions and build management, see
   197 "system" manual for the chapter of that name, especially the "isabelle
   198 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   199 mkdir / make are rendered obsolete.
   200 
   201 * Discontinued obsolete "isabelle makeall".
   202 
   203 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   204 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   205 provides some traditional targets that invoke "isabelle build".  Note
   206 that this is inefficient!  Applications of Isabelle/HOL involving
   207 "isabelle make" should be upgraded to use "isabelle build" directly.
   208 
   209 * Discontinued obsolete Isabelle/build script, it is superseded by the
   210 regular isabelle build tool.  For example:
   211 
   212   isabelle build -s -b HOLCF
   213 
   214 * The "isabelle options" tool prints Isabelle system options, as
   215 required for "isabelle build", for example.
   216 
   217 * The "isabelle mkroot" tool prepares session root directories for use
   218 with "isabelle build", similar to former "isabelle mkdir" for
   219 "isabelle usedir".
   220 
   221 * The "isabelle components" tool helps to resolve add-on components
   222 that are not bundled, or referenced from a bare-bones repository
   223 version of Isabelle.
   224 
   225 * Discontinued support for Poly/ML 5.2.1, which was the last version
   226 without exception positions and advanced ML compiler/toplevel
   227 configuration.
   228 
   229 * Discontinued special treatment of Proof General -- no longer guess
   230 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   231 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   232 settings manually, or use a Proof General version that has been
   233 bundled as Isabelle component.
   234 
   235 
   236 New in Isabelle2012 (May 2012)
   237 ------------------------------
   238 
   239 *** General ***
   240 
   241 * Prover IDE (PIDE) improvements:
   242 
   243   - more robust Sledgehammer integration (as before the sledgehammer
   244     command-line needs to be typed into the source buffer)
   245   - markup for bound variables
   246   - markup for types of term variables (displayed as tooltips)
   247   - support for user-defined Isar commands within the running session
   248   - improved support for Unicode outside original 16bit range
   249     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   250 
   251 * Forward declaration of outer syntax keywords within the theory
   252 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   253 commands to be used in the same theory where defined.
   254 
   255 * Auxiliary contexts indicate block structure for specifications with
   256 additional parameters and assumptions.  Such unnamed contexts may be
   257 nested within other targets, like 'theory', 'locale', 'class',
   258 'instantiation' etc.  Results from the local context are generalized
   259 accordingly and applied to the enclosing target context.  Example:
   260 
   261   context
   262     fixes x y z :: 'a
   263     assumes xy: "x = y" and yz: "y = z"
   264   begin
   265 
   266   lemma my_trans: "x = z" using xy yz by simp
   267 
   268   end
   269 
   270   thm my_trans
   271 
   272 The most basic application is to factor-out context elements of
   273 several fixes/assumes/shows theorem statements, e.g. see
   274 ~~/src/HOL/Isar_Examples/Group_Context.thy
   275 
   276 Any other local theory specification element works within the "context
   277 ... begin ... end" block as well.
   278 
   279 * Bundled declarations associate attributed fact expressions with a
   280 given name in the context.  These may be later included in other
   281 contexts.  This allows to manage context extensions casually, without
   282 the logical dependencies of locales and locale interpretation.  See
   283 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   284 
   285 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   286 declaration, and results are standardized before being stored.  Thus
   287 old-style "standard" after instantiation or composition of facts
   288 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   289 indices of schematic variables.
   290 
   291 * Rule attributes in local theory declarations (e.g. locale or class)
   292 are now statically evaluated: the resulting theorem is stored instead
   293 of the original expression.  INCOMPATIBILITY in rare situations, where
   294 the historic accident of dynamic re-evaluation in interpretations
   295 etc. was exploited.
   296 
   297 * New tutorial "Programming and Proving in Isabelle/HOL"
   298 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   299 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   300 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   301 Logic" as the recommended beginners tutorial, but does not cover all
   302 of the material of that old tutorial.
   303 
   304 * Updated and extended reference manuals: "isar-ref",
   305 "implementation", "system"; reduced remaining material in old "ref"
   306 manual.
   307 
   308 
   309 *** Pure ***
   310 
   311 * Command 'definition' no longer exports the foundational "raw_def"
   312 into the user context.  Minor INCOMPATIBILITY, may use the regular
   313 "def" result with attribute "abs_def" to imitate the old version.
   314 
   315 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   316 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   317 expand it.  This also works for object-logic equality.  (Formerly
   318 undocumented feature.)
   319 
   320 * Sort constraints are now propagated in simultaneous statements, just
   321 like type constraints.  INCOMPATIBILITY in rare situations, where
   322 distinct sorts used to be assigned accidentally.  For example:
   323 
   324   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   325 
   326   lemma "P (x::'a)" and "Q (y::'a::bar)"
   327     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   328 
   329 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   330 tolerant against multiple unifiers, as long as the final result is
   331 unique.  (As before, rules are composed in canonical right-to-left
   332 order to accommodate newly introduced premises.)
   333 
   334 * Renamed some inner syntax categories:
   335 
   336     num ~> num_token
   337     xnum ~> xnum_token
   338     xstr ~> str_token
   339 
   340 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   341 "num_position" etc. are mainly used instead (which also include
   342 position information via constraints).
   343 
   344 * Simplified configuration options for syntax ambiguity: see
   345 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   346 manual.  Minor INCOMPATIBILITY.
   347 
   348 * Discontinued configuration option "syntax_positions": atomic terms
   349 in parse trees are always annotated by position constraints.
   350 
   351 * Old code generator for SML and its commands 'code_module',
   352 'code_library', 'consts_code', 'types_code' have been discontinued.
   353 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   354 
   355 * Redundant attribute "code_inline" has been discontinued. Use
   356 "code_unfold" instead.  INCOMPATIBILITY.
   357 
   358 * Dropped attribute "code_unfold_post" in favor of the its dual
   359 "code_abbrev", which yields a common pattern in definitions like
   360 
   361   definition [code_abbrev]: "f = t"
   362 
   363 INCOMPATIBILITY.
   364 
   365 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   366 instead.  INCOMPATIBILITY.
   367 
   368 * Discontinued old "prems" fact, which used to refer to the accidental
   369 collection of foundational premises in the context (already marked as
   370 legacy since Isabelle2011).
   371 
   372 
   373 *** HOL ***
   374 
   375 * Type 'a set is now a proper type constructor (just as before
   376 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   377 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   378 sets separate, it is often sufficient to rephrase some set S that has
   379 been accidentally used as predicates by "%x. x : S", and some
   380 predicate P that has been accidentally used as set by "{x. P x}".
   381 Corresponding proofs in a first step should be pruned from any
   382 tinkering with former theorems mem_def and Collect_def as far as
   383 possible.
   384 
   385 For developments which deliberately mix predicates and sets, a
   386 planning step is necessary to determine what should become a predicate
   387 and what a set.  It can be helpful to carry out that step in
   388 Isabelle2011-1 before jumping right into the current release.
   389 
   390 * Code generation by default implements sets as container type rather
   391 than predicates.  INCOMPATIBILITY.
   392 
   393 * New type synonym 'a rel = ('a * 'a) set
   394 
   395 * The representation of numerals has changed.  Datatype "num"
   396 represents strictly positive binary numerals, along with functions
   397 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   398 positive and negated numeric literals, respectively.  See also
   399 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   400 user theories may require adaptations as follows:
   401 
   402   - Theorems with number_ring or number_semiring constraints: These
   403     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   404 
   405   - Theories defining numeric types: Remove number, number_semiring,
   406     and number_ring instances. Defer all theorems about numerals until
   407     after classes one and semigroup_add have been instantiated.
   408 
   409   - Numeral-only simp rules: Replace each rule having a "number_of v"
   410     pattern with two copies, one for numeral and one for neg_numeral.
   411 
   412   - Theorems about subclasses of semiring_1 or ring_1: These classes
   413     automatically support numerals now, so more simp rules and
   414     simprocs may now apply within the proof.
   415 
   416   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   417     Redefine using other integer operations.
   418 
   419 * Transfer: New package intended to generalize the existing
   420 "descending" method and related theorem attributes from the Quotient
   421 package.  (Not all functionality is implemented yet, but future
   422 development will focus on Transfer as an eventual replacement for the
   423 corresponding parts of the Quotient package.)
   424 
   425   - transfer_rule attribute: Maintains a collection of transfer rules,
   426     which relate constants at two different types. Transfer rules may
   427     relate different type instances of the same polymorphic constant,
   428     or they may relate an operation on a raw type to a corresponding
   429     operation on an abstract type (quotient or subtype). For example:
   430 
   431     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   432     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   433 
   434   - transfer method: Replaces a subgoal on abstract types with an
   435     equivalent subgoal on the corresponding raw types. Constants are
   436     replaced with corresponding ones according to the transfer rules.
   437     Goals are generalized over all free variables by default; this is
   438     necessary for variables whose types change, but can be overridden
   439     for specific variables with e.g. "transfer fixing: x y z".  The
   440     variant transfer' method allows replacing a subgoal with one that
   441     is logically stronger (rather than equivalent).
   442 
   443   - relator_eq attribute: Collects identity laws for relators of
   444     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   445     transfer method uses these lemmas to infer transfer rules for
   446     non-polymorphic constants on the fly.
   447 
   448   - transfer_prover method: Assists with proving a transfer rule for a
   449     new constant, provided the constant is defined in terms of other
   450     constants that already have transfer rules. It should be applied
   451     after unfolding the constant definitions.
   452 
   453   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   454     from type nat to type int.
   455 
   456 * Lifting: New package intended to generalize the quotient_definition
   457 facility of the Quotient package; designed to work with Transfer.
   458 
   459   - lift_definition command: Defines operations on an abstract type in
   460     terms of a corresponding operation on a representation
   461     type.  Example syntax:
   462 
   463     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   464       is List.insert
   465 
   466     Users must discharge a respectfulness proof obligation when each
   467     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   468     the proof is discharged automatically.) The obligation is
   469     presented in a user-friendly, readable form; a respectfulness
   470     theorem in the standard format and a transfer rule are generated
   471     by the package.
   472 
   473   - Integration with code_abstype: For typedefs (e.g. subtypes
   474     corresponding to a datatype invariant, such as dlist),
   475     lift_definition generates a code certificate theorem and sets up
   476     code generation for each constant.
   477 
   478   - setup_lifting command: Sets up the Lifting package to work with a
   479     user-defined type. The user must provide either a quotient theorem
   480     or a type_definition theorem.  The package configures transfer
   481     rules for equality and quantifiers on the type, and sets up the
   482     lift_definition command to work with the type.
   483 
   484   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   485     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   486     Word/Word.thy and Library/Float.thy.
   487 
   488 * Quotient package:
   489 
   490   - The 'quotient_type' command now supports a 'morphisms' option with
   491     rep and abs functions, similar to typedef.
   492 
   493   - 'quotient_type' sets up new types to work with the Lifting and
   494     Transfer packages, as with 'setup_lifting'.
   495 
   496   - The 'quotient_definition' command now requires the user to prove a
   497     respectfulness property at the point where the constant is
   498     defined, similar to lift_definition; INCOMPATIBILITY.
   499 
   500   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   501     accordingly, INCOMPATIBILITY.
   502 
   503 * New diagnostic command 'find_unused_assms' to find potentially
   504 superfluous assumptions in theorems using Quickcheck.
   505 
   506 * Quickcheck:
   507 
   508   - Quickcheck returns variable assignments as counterexamples, which
   509     allows to reveal the underspecification of functions under test.
   510     For example, refuting "hd xs = x", it presents the variable
   511     assignment xs = [] and x = a1 as a counterexample, assuming that
   512     any property is false whenever "hd []" occurs in it.
   513 
   514     These counterexample are marked as potentially spurious, as
   515     Quickcheck also returns "xs = []" as a counterexample to the
   516     obvious theorem "hd xs = hd xs".
   517 
   518     After finding a potentially spurious counterexample, Quickcheck
   519     continues searching for genuine ones.
   520 
   521     By default, Quickcheck shows potentially spurious and genuine
   522     counterexamples. The option "genuine_only" sets quickcheck to only
   523     show genuine counterexamples.
   524 
   525   - The command 'quickcheck_generator' creates random and exhaustive
   526     value generators for a given type and operations.
   527 
   528     It generates values by using the operations as if they were
   529     constructors of that type.
   530 
   531   - Support for multisets.
   532 
   533   - Added "use_subtype" options.
   534 
   535   - Added "quickcheck_locale" configuration to specify how to process
   536     conjectures in a locale context.
   537 
   538 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   539 and affecting 'rat' and 'real'.
   540 
   541 * Sledgehammer:
   542   - Integrated more tightly with SPASS, as described in the ITP 2012
   543     paper "More SPASS with Isabelle".
   544   - Made it try "smt" as a fallback if "metis" fails or times out.
   545   - Added support for the following provers: Alt-Ergo (via Why3 and
   546     TFF1), iProver, iProver-Eq.
   547   - Sped up the minimizer.
   548   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   549   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   550   - Renamed "sound" option to "strict".
   551 
   552 * Metis: Added possibility to specify lambda translations scheme as a
   553 parenthesized argument (e.g., "by (metis (lifting) ...)").
   554 
   555 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   556 
   557 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   558 
   559 * New "case_product" attribute to generate a case rule doing multiple
   560 case distinctions at the same time.  E.g.
   561 
   562   list.exhaust [case_product nat.exhaust]
   563 
   564 produces a rule which can be used to perform case distinction on both
   565 a list and a nat.
   566 
   567 * New "eventually_elim" method as a generalized variant of the
   568 eventually_elim* rules.  Supports structured proofs.
   569 
   570 * Typedef with implicit set definition is considered legacy.  Use
   571 "typedef (open)" form instead, which will eventually become the
   572 default.
   573 
   574 * Record: code generation can be switched off manually with
   575 
   576   declare [[record_coden = false]]  -- "default true"
   577 
   578 * Datatype: type parameters allow explicit sort constraints.
   579 
   580 * Concrete syntax for case expressions includes constraints for source
   581 positions, and thus produces Prover IDE markup for its bindings.
   582 INCOMPATIBILITY for old-style syntax translations that augment the
   583 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   584 one_case.
   585 
   586 * Clarified attribute "mono_set": pure declaration without modifying
   587 the result of the fact expression.
   588 
   589 * More default pred/set conversions on a couple of relation operations
   590 and predicates.  Added powers of predicate relations.  Consolidation
   591 of some relation theorems:
   592 
   593   converse_def ~> converse_unfold
   594   rel_comp_def ~> relcomp_unfold
   595   symp_def ~> (modified, use symp_def and sym_def instead)
   596   transp_def ~> transp_trans
   597   Domain_def ~> Domain_unfold
   598   Range_def ~> Domain_converse [symmetric]
   599 
   600 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   601 
   602 See theory "Relation" for examples for making use of pred/set
   603 conversions by means of attributes "to_set" and "to_pred".
   604 
   605 INCOMPATIBILITY.
   606 
   607 * Renamed facts about the power operation on relations, i.e., relpow
   608 to match the constant's name:
   609 
   610   rel_pow_1 ~> relpow_1
   611   rel_pow_0_I ~> relpow_0_I
   612   rel_pow_Suc_I ~> relpow_Suc_I
   613   rel_pow_Suc_I2 ~> relpow_Suc_I2
   614   rel_pow_0_E ~> relpow_0_E
   615   rel_pow_Suc_E ~> relpow_Suc_E
   616   rel_pow_E ~> relpow_E
   617   rel_pow_Suc_D2 ~> relpow_Suc_D2
   618   rel_pow_Suc_E2 ~> relpow_Suc_E2
   619   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   620   rel_pow_E2 ~> relpow_E2
   621   rel_pow_add ~> relpow_add
   622   rel_pow_commute ~> relpow
   623   rel_pow_empty ~> relpow_empty:
   624   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   625   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   626   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   627   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   628   rel_pow_fun_conv ~> relpow_fun_conv
   629   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   630   rel_pow_finite_bounded ~> relpow_finite_bounded
   631   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   632   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   633   single_valued_rel_pow ~> single_valued_relpow
   634 
   635 INCOMPATIBILITY.
   636 
   637 * Theory Relation: Consolidated constant name for relation composition
   638 and corresponding theorem names:
   639 
   640   - Renamed constant rel_comp to relcomp.
   641 
   642   - Dropped abbreviation pred_comp. Use relcompp instead.
   643 
   644   - Renamed theorems:
   645 
   646     rel_compI ~> relcompI
   647     rel_compEpair ~> relcompEpair
   648     rel_compE ~> relcompE
   649     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   650     rel_comp_empty1 ~> relcomp_empty1
   651     rel_comp_mono ~> relcomp_mono
   652     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   653     rel_comp_distrib ~> relcomp_distrib
   654     rel_comp_distrib2 ~> relcomp_distrib2
   655     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   656     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   657     single_valued_rel_comp ~> single_valued_relcomp
   658     rel_comp_def ~> relcomp_unfold
   659     converse_rel_comp ~> converse_relcomp
   660     pred_compI ~> relcomppI
   661     pred_compE ~> relcomppE
   662     pred_comp_bot1 ~> relcompp_bot1
   663     pred_comp_bot2 ~> relcompp_bot2
   664     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   665     pred_comp_mono ~> relcompp_mono
   666     pred_comp_distrib ~> relcompp_distrib
   667     pred_comp_distrib2 ~> relcompp_distrib2
   668     converse_pred_comp ~> converse_relcompp
   669 
   670     finite_rel_comp ~> finite_relcomp
   671 
   672     set_rel_comp ~> set_relcomp
   673 
   674 INCOMPATIBILITY.
   675 
   676 * Theory Divides: Discontinued redundant theorems about div and mod.
   677 INCOMPATIBILITY, use the corresponding generic theorems instead.
   678 
   679   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   680   zdiv_self ~> div_self
   681   zmod_self ~> mod_self
   682   zdiv_zero ~> div_0
   683   zmod_zero ~> mod_0
   684   zdiv_zmod_equality ~> div_mod_equality2
   685   zdiv_zmod_equality2 ~> div_mod_equality
   686   zmod_zdiv_trivial ~> mod_div_trivial
   687   zdiv_zminus_zminus ~> div_minus_minus
   688   zmod_zminus_zminus ~> mod_minus_minus
   689   zdiv_zminus2 ~> div_minus_right
   690   zmod_zminus2 ~> mod_minus_right
   691   zdiv_minus1_right ~> div_minus1_right
   692   zmod_minus1_right ~> mod_minus1_right
   693   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   694   zmod_zmult1_eq ~> mod_mult_right_eq
   695   zpower_zmod ~> power_mod
   696   zdvd_zmod ~> dvd_mod
   697   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   698   mod_mult_distrib ~> mult_mod_left
   699   mod_mult_distrib2 ~> mult_mod_right
   700 
   701 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   702 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   703 
   704 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   705 
   706 * Consolidated theorem names concerning fold combinators:
   707 
   708   inf_INFI_fold_inf ~> inf_INF_fold_inf
   709   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   710   INFI_fold_inf ~> INF_fold_inf
   711   SUPR_fold_sup ~> SUP_fold_sup
   712   union_set ~> union_set_fold
   713   minus_set ~> minus_set_fold
   714   INFI_set_fold ~> INF_set_fold
   715   SUPR_set_fold ~> SUP_set_fold
   716   INF_code ~> INF_set_foldr
   717   SUP_code ~> SUP_set_foldr
   718   foldr.simps ~> foldr.simps (in point-free formulation)
   719   foldr_fold_rev ~> foldr_conv_fold
   720   foldl_fold ~> foldl_conv_fold
   721   foldr_foldr ~> foldr_conv_foldl
   722   foldl_foldr ~> foldl_conv_foldr
   723   fold_set_remdups ~> fold_set_fold_remdups
   724   fold_set ~> fold_set_fold
   725   fold1_set ~> fold1_set_fold
   726 
   727 INCOMPATIBILITY.
   728 
   729 * Dropped rarely useful theorems concerning fold combinators:
   730 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   731 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   732 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   733 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   734 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   735 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   736 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   737 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   738 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   739 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   740 
   741 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   742 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   743 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   744 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   745 lemmas over fold rather than foldr, or make use of lemmas
   746 fold_conv_foldr and fold_rev.
   747 
   748 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   749 through option types.
   750 
   751 * "Transitive_Closure.ntrancl": bounded transitive closure on
   752 relations.
   753 
   754 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   755 
   756 * Theory Int: Discontinued many legacy theorems specific to type int.
   757 INCOMPATIBILITY, use the corresponding generic theorems instead.
   758 
   759   zminus_zminus ~> minus_minus
   760   zminus_0 ~> minus_zero
   761   zminus_zadd_distrib ~> minus_add_distrib
   762   zadd_commute ~> add_commute
   763   zadd_assoc ~> add_assoc
   764   zadd_left_commute ~> add_left_commute
   765   zadd_ac ~> add_ac
   766   zmult_ac ~> mult_ac
   767   zadd_0 ~> add_0_left
   768   zadd_0_right ~> add_0_right
   769   zadd_zminus_inverse2 ~> left_minus
   770   zmult_zminus ~> mult_minus_left
   771   zmult_commute ~> mult_commute
   772   zmult_assoc ~> mult_assoc
   773   zadd_zmult_distrib ~> left_distrib
   774   zadd_zmult_distrib2 ~> right_distrib
   775   zdiff_zmult_distrib ~> left_diff_distrib
   776   zdiff_zmult_distrib2 ~> right_diff_distrib
   777   zmult_1 ~> mult_1_left
   778   zmult_1_right ~> mult_1_right
   779   zle_refl ~> order_refl
   780   zle_trans ~> order_trans
   781   zle_antisym ~> order_antisym
   782   zle_linear ~> linorder_linear
   783   zless_linear ~> linorder_less_linear
   784   zadd_left_mono ~> add_left_mono
   785   zadd_strict_right_mono ~> add_strict_right_mono
   786   zadd_zless_mono ~> add_less_le_mono
   787   int_0_less_1 ~> zero_less_one
   788   int_0_neq_1 ~> zero_neq_one
   789   zless_le ~> less_le
   790   zpower_zadd_distrib ~> power_add
   791   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   792   zero_le_zpower_abs ~> zero_le_power_abs
   793 
   794 * Theory Deriv: Renamed
   795 
   796   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   797 
   798 * Theory Library/Multiset: Improved code generation of multisets.
   799 
   800 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   801 are expressed via type classes again. The special syntax
   802 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   803 setsum_set, which is now subsumed by Big_Operators.setsum.
   804 INCOMPATIBILITY.
   805 
   806 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   807 use theory HOL/Library/Nat_Bijection instead.
   808 
   809 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   810 trees is now inside a type class context.  Names of affected
   811 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   812 theories working directly with raw red-black trees, adapt the names as
   813 follows:
   814 
   815   Operations:
   816   bulkload -> rbt_bulkload
   817   del_from_left -> rbt_del_from_left
   818   del_from_right -> rbt_del_from_right
   819   del -> rbt_del
   820   delete -> rbt_delete
   821   ins -> rbt_ins
   822   insert -> rbt_insert
   823   insertw -> rbt_insert_with
   824   insert_with_key -> rbt_insert_with_key
   825   map_entry -> rbt_map_entry
   826   lookup -> rbt_lookup
   827   sorted -> rbt_sorted
   828   tree_greater -> rbt_greater
   829   tree_less -> rbt_less
   830   tree_less_symbol -> rbt_less_symbol
   831   union -> rbt_union
   832   union_with -> rbt_union_with
   833   union_with_key -> rbt_union_with_key
   834 
   835   Lemmas:
   836   balance_left_sorted -> balance_left_rbt_sorted
   837   balance_left_tree_greater -> balance_left_rbt_greater
   838   balance_left_tree_less -> balance_left_rbt_less
   839   balance_right_sorted -> balance_right_rbt_sorted
   840   balance_right_tree_greater -> balance_right_rbt_greater
   841   balance_right_tree_less -> balance_right_rbt_less
   842   balance_sorted -> balance_rbt_sorted
   843   balance_tree_greater -> balance_rbt_greater
   844   balance_tree_less -> balance_rbt_less
   845   bulkload_is_rbt -> rbt_bulkload_is_rbt
   846   combine_sorted -> combine_rbt_sorted
   847   combine_tree_greater -> combine_rbt_greater
   848   combine_tree_less -> combine_rbt_less
   849   delete_in_tree -> rbt_delete_in_tree
   850   delete_is_rbt -> rbt_delete_is_rbt
   851   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   852   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   853   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   854   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   855   del_in_tree -> rbt_del_in_tree
   856   del_inv1_inv2 -> rbt_del_inv1_inv2
   857   del_sorted -> rbt_del_rbt_sorted
   858   del_tree_greater -> rbt_del_rbt_greater
   859   del_tree_less -> rbt_del_rbt_less
   860   dom_lookup_Branch -> dom_rbt_lookup_Branch
   861   entries_lookup -> entries_rbt_lookup
   862   finite_dom_lookup -> finite_dom_rbt_lookup
   863   insert_sorted -> rbt_insert_rbt_sorted
   864   insertw_is_rbt -> rbt_insertw_is_rbt
   865   insertwk_is_rbt -> rbt_insertwk_is_rbt
   866   insertwk_sorted -> rbt_insertwk_rbt_sorted
   867   insertw_sorted -> rbt_insertw_rbt_sorted
   868   ins_sorted -> ins_rbt_sorted
   869   ins_tree_greater -> ins_rbt_greater
   870   ins_tree_less -> ins_rbt_less
   871   is_rbt_sorted -> is_rbt_rbt_sorted
   872   lookup_balance -> rbt_lookup_balance
   873   lookup_bulkload -> rbt_lookup_rbt_bulkload
   874   lookup_delete -> rbt_lookup_rbt_delete
   875   lookup_Empty -> rbt_lookup_Empty
   876   lookup_from_in_tree -> rbt_lookup_from_in_tree
   877   lookup_in_tree -> rbt_lookup_in_tree
   878   lookup_ins -> rbt_lookup_ins
   879   lookup_insert -> rbt_lookup_rbt_insert
   880   lookup_insertw -> rbt_lookup_rbt_insertw
   881   lookup_insertwk -> rbt_lookup_rbt_insertwk
   882   lookup_keys -> rbt_lookup_keys
   883   lookup_map -> rbt_lookup_map
   884   lookup_map_entry -> rbt_lookup_rbt_map_entry
   885   lookup_tree_greater -> rbt_lookup_rbt_greater
   886   lookup_tree_less -> rbt_lookup_rbt_less
   887   lookup_union -> rbt_lookup_rbt_union
   888   map_entry_color_of -> rbt_map_entry_color_of
   889   map_entry_inv1 -> rbt_map_entry_inv1
   890   map_entry_inv2 -> rbt_map_entry_inv2
   891   map_entry_is_rbt -> rbt_map_entry_is_rbt
   892   map_entry_sorted -> rbt_map_entry_rbt_sorted
   893   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   894   map_entry_tree_less -> rbt_map_entry_rbt_less
   895   map_tree_greater -> map_rbt_greater
   896   map_tree_less -> map_rbt_less
   897   map_sorted -> map_rbt_sorted
   898   paint_sorted -> paint_rbt_sorted
   899   paint_lookup -> paint_rbt_lookup
   900   paint_tree_greater -> paint_rbt_greater
   901   paint_tree_less -> paint_rbt_less
   902   sorted_entries -> rbt_sorted_entries
   903   tree_greater_eq_trans -> rbt_greater_eq_trans
   904   tree_greater_nit -> rbt_greater_nit
   905   tree_greater_prop -> rbt_greater_prop
   906   tree_greater_simps -> rbt_greater_simps
   907   tree_greater_trans -> rbt_greater_trans
   908   tree_less_eq_trans -> rbt_less_eq_trans
   909   tree_less_nit -> rbt_less_nit
   910   tree_less_prop -> rbt_less_prop
   911   tree_less_simps -> rbt_less_simps
   912   tree_less_trans -> rbt_less_trans
   913   tree_ord_props -> rbt_ord_props
   914   union_Branch -> rbt_union_Branch
   915   union_is_rbt -> rbt_union_is_rbt
   916   unionw_is_rbt -> rbt_unionw_is_rbt
   917   unionwk_is_rbt -> rbt_unionwk_is_rbt
   918   unionwk_sorted -> rbt_unionwk_rbt_sorted
   919 
   920 * Theory HOL/Library/Float: Floating point numbers are now defined as
   921 a subset of the real numbers.  All operations are defined using the
   922 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   923 
   924   Changed Operations:
   925   float_abs -> abs
   926   float_nprt -> nprt
   927   float_pprt -> pprt
   928   pow2 -> use powr
   929   round_down -> float_round_down
   930   round_up -> float_round_up
   931   scale -> exponent
   932 
   933   Removed Operations:
   934   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   935 
   936   Renamed Lemmas:
   937   abs_float_def -> Float.compute_float_abs
   938   bitlen_ge0 -> bitlen_nonneg
   939   bitlen.simps -> Float.compute_bitlen
   940   float_components -> Float_mantissa_exponent
   941   float_divl.simps -> Float.compute_float_divl
   942   float_divr.simps -> Float.compute_float_divr
   943   float_eq_odd -> mult_powr_eq_mult_powr_iff
   944   float_power -> real_of_float_power
   945   lapprox_posrat_def -> Float.compute_lapprox_posrat
   946   lapprox_rat.simps -> Float.compute_lapprox_rat
   947   le_float_def' -> Float.compute_float_le
   948   le_float_def -> less_eq_float.rep_eq
   949   less_float_def' -> Float.compute_float_less
   950   less_float_def -> less_float.rep_eq
   951   normfloat_def -> Float.compute_normfloat
   952   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   953   normfloat -> normfloat_def
   954   normfloat_unique -> use normfloat_def
   955   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   956   one_float_def -> Float.compute_float_one
   957   plus_float_def -> Float.compute_float_plus
   958   rapprox_posrat_def -> Float.compute_rapprox_posrat
   959   rapprox_rat.simps -> Float.compute_rapprox_rat
   960   real_of_float_0 -> zero_float.rep_eq
   961   real_of_float_1 -> one_float.rep_eq
   962   real_of_float_abs -> abs_float.rep_eq
   963   real_of_float_add -> plus_float.rep_eq
   964   real_of_float_minus -> uminus_float.rep_eq
   965   real_of_float_mult -> times_float.rep_eq
   966   real_of_float_simp -> Float.rep_eq
   967   real_of_float_sub -> minus_float.rep_eq
   968   round_down.simps -> Float.compute_float_round_down
   969   round_up.simps -> Float.compute_float_round_up
   970   times_float_def -> Float.compute_float_times
   971   uminus_float_def -> Float.compute_float_uminus
   972   zero_float_def -> Float.compute_float_zero
   973 
   974   Lemmas not necessary anymore, use the transfer method:
   975   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   976   float_divr, float_le_simp, float_less1_mantissa_bound,
   977   float_less_simp, float_less_zero, float_le_zero,
   978   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   979   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   980   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   981   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   982   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   983   round_up, zero_le_float, zero_less_float
   984 
   985 * New theory HOL/Library/DAList provides an abstract type for
   986 association lists with distinct keys.
   987 
   988 * Session HOL/IMP: Added new theory of abstract interpretation of
   989 annotated commands.
   990 
   991 * Session HOL-Import: Re-implementation from scratch is faster,
   992 simpler, and more scalable.  Requires a proof bundle, which is
   993 available as an external component.  Discontinued old (and mostly
   994 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   995 
   996 * Session HOL-Word: Discontinued many redundant theorems specific to
   997 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   998 instead.
   999 
  1000   word_sub_alt ~> word_sub_wi
  1001   word_add_alt ~> word_add_def
  1002   word_mult_alt ~> word_mult_def
  1003   word_minus_alt ~> word_minus_def
  1004   word_0_alt ~> word_0_wi
  1005   word_1_alt ~> word_1_wi
  1006   word_add_0 ~> add_0_left
  1007   word_add_0_right ~> add_0_right
  1008   word_mult_1 ~> mult_1_left
  1009   word_mult_1_right ~> mult_1_right
  1010   word_add_commute ~> add_commute
  1011   word_add_assoc ~> add_assoc
  1012   word_add_left_commute ~> add_left_commute
  1013   word_mult_commute ~> mult_commute
  1014   word_mult_assoc ~> mult_assoc
  1015   word_mult_left_commute ~> mult_left_commute
  1016   word_left_distrib ~> left_distrib
  1017   word_right_distrib ~> right_distrib
  1018   word_left_minus ~> left_minus
  1019   word_diff_0_right ~> diff_0_right
  1020   word_diff_self ~> diff_self
  1021   word_sub_def ~> diff_minus
  1022   word_diff_minus ~> diff_minus
  1023   word_add_ac ~> add_ac
  1024   word_mult_ac ~> mult_ac
  1025   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1026   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1027   word_order_trans ~> order_trans
  1028   word_order_refl ~> order_refl
  1029   word_order_antisym ~> order_antisym
  1030   word_order_linear ~> linorder_linear
  1031   lenw1_zero_neq_one ~> zero_neq_one
  1032   word_number_of_eq ~> number_of_eq
  1033   word_of_int_add_hom ~> wi_hom_add
  1034   word_of_int_sub_hom ~> wi_hom_sub
  1035   word_of_int_mult_hom ~> wi_hom_mult
  1036   word_of_int_minus_hom ~> wi_hom_neg
  1037   word_of_int_succ_hom ~> wi_hom_succ
  1038   word_of_int_pred_hom ~> wi_hom_pred
  1039   word_of_int_0_hom ~> word_0_wi
  1040   word_of_int_1_hom ~> word_1_wi
  1041 
  1042 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1043 machine word equalities and inequalities into logical circuits,
  1044 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1045 multiplication, shifting by constants, bitwise operators and numeric
  1046 constants.  Requires fixed-length word types, not 'a word.  Solves
  1047 many standard word identities outright and converts more into first
  1048 order problems amenable to blast or similar.  See also examples in
  1049 HOL/Word/Examples/WordExamples.thy.
  1050 
  1051 * Session HOL-Probability: Introduced the type "'a measure" to
  1052 represent measures, this replaces the records 'a algebra and 'a
  1053 measure_space.  The locales based on subset_class now have two
  1054 locale-parameters the space \<Omega> and the set of measurable sets M.
  1055 The product of probability spaces uses now the same constant as the
  1056 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1057 measure".  Most constants are defined now outside of locales and gain
  1058 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1059 Measure space constructions for distributions and densities now got
  1060 their own constants distr and density.  Instead of using locales to
  1061 describe measure spaces with a finite space, the measure count_space
  1062 and point_measure is introduced.  INCOMPATIBILITY.
  1063 
  1064   Renamed constants:
  1065   measure -> emeasure
  1066   finite_measure.\<mu>' -> measure
  1067   product_algebra_generator -> prod_algebra
  1068   product_prob_space.emb -> prod_emb
  1069   product_prob_space.infprod_algebra -> PiM
  1070 
  1071   Removed locales:
  1072   completeable_measure_space
  1073   finite_measure_space
  1074   finite_prob_space
  1075   finite_product_finite_prob_space
  1076   finite_product_sigma_algebra
  1077   finite_sigma_algebra
  1078   measure_space
  1079   pair_finite_prob_space
  1080   pair_finite_sigma_algebra
  1081   pair_finite_space
  1082   pair_sigma_algebra
  1083   product_sigma_algebra
  1084 
  1085   Removed constants:
  1086   conditional_space
  1087   distribution -> use distr measure, or distributed predicate
  1088   image_space
  1089   joint_distribution -> use distr measure, or distributed predicate
  1090   pair_measure_generator
  1091   product_prob_space.infprod_algebra -> use PiM
  1092   subvimage
  1093 
  1094   Replacement theorems:
  1095   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1096   finite_measure.empty_measure -> measure_empty
  1097   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1098   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1099   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1100   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1101   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1102   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1103   finite_measure.positive_measure' -> measure_nonneg
  1104   finite_measure.real_measure -> finite_measure.emeasure_real
  1105   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1106   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1107   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1108   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1109   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1110   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1111   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1112   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1113   information_space.entropy_commute -> information_space.entropy_commute_simple
  1114   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1115   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1116   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1117   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1118   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1119   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1120   information_space.KL_ge_0 -> information_space.KL_nonneg
  1121   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1122   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1123   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1124   Int_stable_product_algebra_generator -> positive_integral
  1125   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1126   measure_space.additive -> emeasure_additive
  1127   measure_space.AE_iff_null_set -> AE_iff_null
  1128   measure_space.almost_everywhere_def -> eventually_ae_filter
  1129   measure_space.almost_everywhere_vimage -> AE_distrD
  1130   measure_space.continuity_from_above -> INF_emeasure_decseq
  1131   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1132   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1133   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1134   measure_space_density -> emeasure_density
  1135   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1136   measure_space.integrable_vimage -> integrable_distr
  1137   measure_space.integral_translated_density -> integral_density
  1138   measure_space.integral_vimage -> integral_distr
  1139   measure_space.measure_additive -> plus_emeasure
  1140   measure_space.measure_compl -> emeasure_compl
  1141   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1142   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1143   measure_space.measure_decseq -> decseq_emeasure
  1144   measure_space.measure_Diff -> emeasure_Diff
  1145   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1146   measure_space.measure_eq_0 -> emeasure_eq_0
  1147   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1148   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1149   measure_space.measure_incseq -> incseq_emeasure
  1150   measure_space.measure_insert -> emeasure_insert
  1151   measure_space.measure_mono -> emeasure_mono
  1152   measure_space.measure_not_negative -> emeasure_not_MInf
  1153   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1154   measure_space.measure_setsum -> setsum_emeasure
  1155   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1156   measure_space.measure_space_vimage -> emeasure_distr
  1157   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1158   measure_space.measure_subadditive -> subadditive
  1159   measure_space.measure_top -> emeasure_space
  1160   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1161   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1162   measure_space.positive_integral_translated_density -> positive_integral_density
  1163   measure_space.positive_integral_vimage -> positive_integral_distr
  1164   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1165   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1166   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1167   measure_space.real_measure_Diff -> measure_Diff
  1168   measure_space.real_measure_finite_Union -> measure_finite_Union
  1169   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1170   measure_space.real_measure_subadditive -> measure_subadditive
  1171   measure_space.real_measure_Union -> measure_Union
  1172   measure_space.real_measure_UNION -> measure_UNION
  1173   measure_space.simple_function_vimage -> simple_function_comp
  1174   measure_space.simple_integral_vimage -> simple_integral_distr
  1175   measure_space.simple_integral_vimage -> simple_integral_distr
  1176   measure_unique_Int_stable -> measure_eqI_generator_eq
  1177   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1178   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1179   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1180   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1181   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1182   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1183   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1184   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1185   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1186   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1187   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1188   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1189   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1190   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1191   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1192   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1193   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1194   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1195   prob_space.prob_space_vimage -> prob_space_distr
  1196   prob_space.random_variable_restrict -> measurable_restrict
  1197   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1198   product_algebraE -> prod_algebraE_all
  1199   product_algebra_generator_der -> prod_algebra_eq_finite
  1200   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1201   product_algebraI -> sets_PiM_I_finite
  1202   product_measure_exists -> product_sigma_finite.sigma_finite
  1203   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1204   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1205   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1206   product_prob_space.measurable_component -> measurable_component_singleton
  1207   product_prob_space.measurable_emb -> measurable_prod_emb
  1208   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1209   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1210   product_prob_space.measure_emb -> emeasure_prod_emb
  1211   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1212   product_sigma_algebra.product_algebra_into_space -> space_closed
  1213   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1214   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1215   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1216   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1217   sets_product_algebra -> sets_PiM
  1218   sigma_algebra.measurable_sigma -> measurable_measure_of
  1219   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1220   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1221   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1222   space_product_algebra -> space_PiM
  1223 
  1224 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1225 languages) into Isabelle/HOL.
  1226 
  1227 
  1228 *** FOL ***
  1229 
  1230 * New "case_product" attribute (see HOL).
  1231 
  1232 
  1233 *** ZF ***
  1234 
  1235 * Greater support for structured proofs involving induction or case
  1236 analysis.
  1237 
  1238 * Much greater use of mathematical symbols.
  1239 
  1240 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1241 
  1242 
  1243 *** ML ***
  1244 
  1245 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1246 from a minor keyword introduced via theory header declaration.
  1247 
  1248 * Antiquotation @{command_spec "name"} produces the
  1249 Outer_Syntax.command_spec from a major keyword introduced via theory
  1250 header declaration; it can be passed to Outer_Syntax.command etc.
  1251 
  1252 * Local_Theory.define no longer hard-wires default theorem name
  1253 "foo_def", but retains the binding as given.  If that is Binding.empty
  1254 / Attrib.empty_binding, the result is not registered as user-level
  1255 fact.  The Local_Theory.define_internal variant allows to specify a
  1256 non-empty name (used for the foundation in the background theory),
  1257 while omitting the fact binding in the user-context.  Potential
  1258 INCOMPATIBILITY for derived definitional packages: need to specify
  1259 naming policy for primitive definitions more explicitly.
  1260 
  1261 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1262 conformance with similar operations in structure Term and Logic.
  1263 
  1264 * Antiquotation @{attributes [...]} embeds attribute source
  1265 representation into the ML text, which is particularly useful with
  1266 declarations like Local_Theory.note.
  1267 
  1268 * Structure Proof_Context follows standard naming scheme.  Old
  1269 ProofContext has been discontinued.  INCOMPATIBILITY.
  1270 
  1271 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1272 change of semantics: update is applied to auxiliary local theory
  1273 context as well.
  1274 
  1275 * Modernized some old-style infix operations:
  1276 
  1277   addeqcongs    ~> Simplifier.add_eqcong
  1278   deleqcongs    ~> Simplifier.del_eqcong
  1279   addcongs      ~> Simplifier.add_cong
  1280   delcongs      ~> Simplifier.del_cong
  1281   setmksimps    ~> Simplifier.set_mksimps
  1282   setmkcong     ~> Simplifier.set_mkcong
  1283   setmksym      ~> Simplifier.set_mksym
  1284   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1285   settermless   ~> Simplifier.set_termless
  1286   setsubgoaler  ~> Simplifier.set_subgoaler
  1287   addsplits     ~> Splitter.add_split
  1288   delsplits     ~> Splitter.del_split
  1289 
  1290 
  1291 *** System ***
  1292 
  1293 * USER_HOME settings variable points to cross-platform user home
  1294 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1295 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1296 of former $HOME.  A different default for USER_HOME may be set
  1297 explicitly in shell environment, before Isabelle settings are
  1298 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1299 the generic user home was intended.
  1300 
  1301 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1302 notation, which is useful for the jEdit file browser, for example.
  1303 
  1304 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1305 (not just JRE).
  1306 
  1307 
  1308 
  1309 New in Isabelle2011-1 (October 2011)
  1310 ------------------------------------
  1311 
  1312 *** General ***
  1313 
  1314 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1315 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1316 
  1317   - Management of multiple theory files directly from the editor
  1318     buffer store -- bypassing the file-system (no requirement to save
  1319     files for checking).
  1320 
  1321   - Markup of formal entities within the text buffer, with semantic
  1322     highlighting, tooltips and hyperlinks to jump to defining source
  1323     positions.
  1324 
  1325   - Improved text rendering, with sub/superscripts in the source
  1326     buffer (including support for copy/paste wrt. output panel, HTML
  1327     theory output and other non-Isabelle text boxes).
  1328 
  1329   - Refined scheduling of proof checking and printing of results,
  1330     based on interactive editor view.  (Note: jEdit folding and
  1331     narrowing allows to restrict buffer perspectives explicitly.)
  1332 
  1333   - Reduced CPU performance requirements, usable on machines with few
  1334     cores.
  1335 
  1336   - Reduced memory requirements due to pruning of unused document
  1337     versions (garbage collection).
  1338 
  1339 See also ~~/src/Tools/jEdit/README.html for further information,
  1340 including some remaining limitations.
  1341 
  1342 * Theory loader: source files are exclusively located via the master
  1343 directory of each theory node (where the .thy file itself resides).
  1344 The global load path (such as src/HOL/Library) has been discontinued.
  1345 Note that the path element ~~ may be used to reference theories in the
  1346 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1347 INCOMPATIBILITY.
  1348 
  1349 * Theory loader: source files are identified by content via SHA1
  1350 digests.  Discontinued former path/modtime identification and optional
  1351 ISABELLE_FILE_IDENT plugin scripts.
  1352 
  1353 * Parallelization of nested Isar proofs is subject to
  1354 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1355 usedir option -Q.
  1356 
  1357 * Name space: former unsynchronized references are now proper
  1358 configuration options, with more conventional names:
  1359 
  1360   long_names   ~> names_long
  1361   short_names  ~> names_short
  1362   unique_names ~> names_unique
  1363 
  1364 Minor INCOMPATIBILITY, need to declare options in context like this:
  1365 
  1366   declare [[names_unique = false]]
  1367 
  1368 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1369 that the result needs to be unique, which means fact specifications
  1370 may have to be refined after enriching a proof context.
  1371 
  1372 * Attribute "case_names" has been refined: the assumptions in each case
  1373 can be named now by following the case name with [name1 name2 ...].
  1374 
  1375 * Isabelle/Isar reference manual has been updated and extended:
  1376   - "Synopsis" provides a catalog of main Isar language concepts.
  1377   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1378   - Updated material from classic "ref" manual, notably about
  1379     "Classical Reasoner".
  1380 
  1381 
  1382 *** HOL ***
  1383 
  1384 * Class bot and top require underlying partial order rather than
  1385 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1386 
  1387 * Class complete_lattice: generalized a couple of lemmas from sets;
  1388 generalized theorems INF_cong and SUP_cong.  New type classes for
  1389 complete boolean algebras and complete linear orders.  Lemmas
  1390 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1391 class complete_linorder.
  1392 
  1393 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1394 Sup_fun_def, Inf_apply, Sup_apply.
  1395 
  1396 Removed redundant lemmas (the right hand side gives hints how to
  1397 replace them for (metis ...), or (simp only: ...) proofs):
  1398 
  1399   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1400   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1401   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1402   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1403   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1404   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1405   Inter_def ~> INF_def, image_def
  1406   Union_def ~> SUP_def, image_def
  1407   INT_eq ~> INF_def, and image_def
  1408   UN_eq ~> SUP_def, and image_def
  1409   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1410 
  1411 More consistent and comprehensive names:
  1412 
  1413   INTER_eq_Inter_image ~> INF_def
  1414   UNION_eq_Union_image ~> SUP_def
  1415   INFI_def ~> INF_def
  1416   SUPR_def ~> SUP_def
  1417   INF_leI ~> INF_lower
  1418   INF_leI2 ~> INF_lower2
  1419   le_INFI ~> INF_greatest
  1420   le_SUPI ~> SUP_upper
  1421   le_SUPI2 ~> SUP_upper2
  1422   SUP_leI ~> SUP_least
  1423   INFI_bool_eq ~> INF_bool_eq
  1424   SUPR_bool_eq ~> SUP_bool_eq
  1425   INFI_apply ~> INF_apply
  1426   SUPR_apply ~> SUP_apply
  1427   INTER_def ~> INTER_eq
  1428   UNION_def ~> UNION_eq
  1429 
  1430 INCOMPATIBILITY.
  1431 
  1432 * Renamed theory Complete_Lattice to Complete_Lattices.
  1433 INCOMPATIBILITY.
  1434 
  1435 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1436 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1437 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1438 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1439 
  1440 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1441 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1442 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1443 INCOMPATIBILITY.
  1444 
  1445 * Added syntactic classes "inf" and "sup" for the respective
  1446 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1447 (mostly internal) locale predicates for some derived classes.
  1448 
  1449 * Theorem collections ball_simps and bex_simps do not contain theorems
  1450 referring to UNION any longer; these have been moved to collection
  1451 UN_ball_bex_simps.  INCOMPATIBILITY.
  1452 
  1453 * Theory Archimedean_Field: floor now is defined as parameter of a
  1454 separate type class floor_ceiling.
  1455 
  1456 * Theory Finite_Set: more coherent development of fold_set locales:
  1457 
  1458     locale fun_left_comm ~> locale comp_fun_commute
  1459     locale fun_left_comm_idem ~> locale comp_fun_idem
  1460 
  1461 Both use point-free characterization; interpretation proofs may need
  1462 adjustment.  INCOMPATIBILITY.
  1463 
  1464 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1465 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1466 
  1467 * Theory Complex_Main: The locale interpretations for the
  1468 bounded_linear and bounded_bilinear locales have been removed, in
  1469 order to reduce the number of duplicate lemmas. Users must use the
  1470 original names for distributivity theorems, potential INCOMPATIBILITY.
  1471 
  1472   divide.add ~> add_divide_distrib
  1473   divide.diff ~> diff_divide_distrib
  1474   divide.setsum ~> setsum_divide_distrib
  1475   mult.add_right ~> right_distrib
  1476   mult.diff_right ~> right_diff_distrib
  1477   mult_right.setsum ~> setsum_right_distrib
  1478   mult_left.diff ~> left_diff_distrib
  1479 
  1480 * Theory Complex_Main: Several redundant theorems have been removed or
  1481 replaced by more general versions. INCOMPATIBILITY.
  1482 
  1483   real_diff_def ~> minus_real_def
  1484   real_divide_def ~> divide_real_def
  1485   real_less_def ~> less_le
  1486   real_abs_def ~> abs_real_def
  1487   real_sgn_def ~> sgn_real_def
  1488   real_mult_commute ~> mult_commute
  1489   real_mult_assoc ~> mult_assoc
  1490   real_mult_1 ~> mult_1_left
  1491   real_add_mult_distrib ~> left_distrib
  1492   real_zero_not_eq_one ~> zero_neq_one
  1493   real_mult_inverse_left ~> left_inverse
  1494   INVERSE_ZERO ~> inverse_zero
  1495   real_le_refl ~> order_refl
  1496   real_le_antisym ~> order_antisym
  1497   real_le_trans ~> order_trans
  1498   real_le_linear ~> linear
  1499   real_le_eq_diff ~> le_iff_diff_le_0
  1500   real_add_left_mono ~> add_left_mono
  1501   real_mult_order ~> mult_pos_pos
  1502   real_mult_less_mono2 ~> mult_strict_left_mono
  1503   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1504   real_0_le_divide_iff ~> zero_le_divide_iff
  1505   realpow_two_disj ~> power2_eq_iff
  1506   real_squared_diff_one_factored ~> square_diff_one_factored
  1507   realpow_two_diff ~> square_diff_square_factored
  1508   reals_complete2 ~> complete_real
  1509   real_sum_squared_expand ~> power2_sum
  1510   exp_ln_eq ~> ln_unique
  1511   expi_add ~> exp_add
  1512   expi_zero ~> exp_zero
  1513   lemma_DERIV_subst ~> DERIV_cong
  1514   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1515   LIMSEQ_const ~> tendsto_const
  1516   LIMSEQ_norm ~> tendsto_norm
  1517   LIMSEQ_add ~> tendsto_add
  1518   LIMSEQ_minus ~> tendsto_minus
  1519   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1520   LIMSEQ_diff ~> tendsto_diff
  1521   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1522   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1523   LIMSEQ_mult ~> tendsto_mult
  1524   LIMSEQ_inverse ~> tendsto_inverse
  1525   LIMSEQ_divide ~> tendsto_divide
  1526   LIMSEQ_pow ~> tendsto_power
  1527   LIMSEQ_setsum ~> tendsto_setsum
  1528   LIMSEQ_setprod ~> tendsto_setprod
  1529   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1530   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1531   LIMSEQ_imp_rabs ~> tendsto_rabs
  1532   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1533   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1534   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1535   LIMSEQ_Complex ~> tendsto_Complex
  1536   LIM_ident ~> tendsto_ident_at
  1537   LIM_const ~> tendsto_const
  1538   LIM_add ~> tendsto_add
  1539   LIM_add_zero ~> tendsto_add_zero
  1540   LIM_minus ~> tendsto_minus
  1541   LIM_diff ~> tendsto_diff
  1542   LIM_norm ~> tendsto_norm
  1543   LIM_norm_zero ~> tendsto_norm_zero
  1544   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1545   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1546   LIM_rabs ~> tendsto_rabs
  1547   LIM_rabs_zero ~> tendsto_rabs_zero
  1548   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1549   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1550   LIM_compose ~> tendsto_compose
  1551   LIM_mult ~> tendsto_mult
  1552   LIM_scaleR ~> tendsto_scaleR
  1553   LIM_of_real ~> tendsto_of_real
  1554   LIM_power ~> tendsto_power
  1555   LIM_inverse ~> tendsto_inverse
  1556   LIM_sgn ~> tendsto_sgn
  1557   isCont_LIM_compose ~> isCont_tendsto_compose
  1558   bounded_linear.LIM ~> bounded_linear.tendsto
  1559   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1560   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1561   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1562   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1563   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1564   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1565 
  1566 * Theory Complex_Main: The definition of infinite series was
  1567 generalized.  Now it is defined on the type class {topological_space,
  1568 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1569 
  1570 * Theory Complex_Main: The complex exponential function "expi" is now
  1571 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1572 several polymorphic lemmas about "exp" are now applicable to "expi".
  1573 
  1574 * Code generation:
  1575 
  1576   - Theory Library/Code_Char_ord provides native ordering of
  1577     characters in the target language.
  1578 
  1579   - Commands code_module and code_library are legacy, use export_code
  1580     instead.
  1581 
  1582   - Method "evaluation" is legacy, use method "eval" instead.
  1583 
  1584   - Legacy evaluator "SML" is deactivated by default.  May be
  1585     reactivated by the following theory command:
  1586 
  1587       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1588 
  1589 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1590 
  1591 * New proof method "induction" that gives induction hypotheses the
  1592 name "IH", thus distinguishing them from further hypotheses that come
  1593 from rule induction.  The latter are still called "hyps".  Method
  1594 "induction" is a thin wrapper around "induct" and follows the same
  1595 syntax.
  1596 
  1597 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1598 still available as a legacy feature for some time.
  1599 
  1600 * Nitpick:
  1601   - Added "need" and "total_consts" options.
  1602   - Reintroduced "show_skolems" option by popular demand.
  1603   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1604     INCOMPATIBILITY.
  1605 
  1606 * Sledgehammer:
  1607   - Use quasi-sound (and efficient) translations by default.
  1608   - Added support for the following provers: E-ToFoF, LEO-II,
  1609     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1610   - Automatically preplay and minimize proofs before showing them if
  1611     this can be done within reasonable time.
  1612   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1613     INCOMPATIBILITY.
  1614   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1615     "max_mono_iters", and "max_new_mono_instances" options.
  1616   - Removed "explicit_apply" and "full_types" options as well as "Full
  1617     Types" Proof General menu item. INCOMPATIBILITY.
  1618 
  1619 * Metis:
  1620   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1621   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1622     INCOMPATIBILITY.
  1623 
  1624 * Command 'try':
  1625   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1626     "elim:" options. INCOMPATIBILITY.
  1627   - Introduced 'try' that not only runs 'try_methods' but also
  1628     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1629 
  1630 * Quickcheck:
  1631   - Added "eval" option to evaluate terms for the found counterexample
  1632     (currently only supported by the default (exhaustive) tester).
  1633   - Added post-processing of terms to obtain readable counterexamples
  1634     (currently only supported by the default (exhaustive) tester).
  1635   - New counterexample generator quickcheck[narrowing] enables
  1636     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1637     with its installation location defined in the Isabelle settings
  1638     environment as ISABELLE_GHC.
  1639   - Removed quickcheck tester "SML" based on the SML code generator
  1640     (formly in HOL/Library).
  1641 
  1642 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1643 use 'partial_function' instead.
  1644 
  1645 * Theory Library/Extended_Reals replaces now the positive extended
  1646 reals found in probability theory. This file is extended by
  1647 Multivariate_Analysis/Extended_Real_Limits.
  1648 
  1649 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1650 from where it must be imported explicitly if it is really required.
  1651 INCOMPATIBILITY.
  1652 
  1653 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1654 been moved here.  INCOMPATIBILITY.
  1655 
  1656 * Theory Library/Saturated provides type of numbers with saturated
  1657 arithmetic.
  1658 
  1659 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1660 product type 'a * 'b, and provides instance proofs for various order
  1661 and lattice type classes.
  1662 
  1663 * Theory Library/Countable now provides the "countable_datatype" proof
  1664 method for proving "countable" class instances for datatypes.
  1665 
  1666 * Theory Library/Cset_Monad allows do notation for computable sets
  1667 (cset) via the generic monad ad-hoc overloading facility.
  1668 
  1669 * Library: Theories of common data structures are split into theories
  1670 for implementation, an invariant-ensuring type, and connection to an
  1671 abstract type. INCOMPATIBILITY.
  1672 
  1673   - RBT is split into RBT and RBT_Mapping.
  1674   - AssocList is split and renamed into AList and AList_Mapping.
  1675   - DList is split into DList_Impl, DList, and DList_Cset.
  1676   - Cset is split into Cset and List_Cset.
  1677 
  1678 * Theory Library/Nat_Infinity has been renamed to
  1679 Library/Extended_Nat, with name changes of the following types and
  1680 constants:
  1681 
  1682   type inat   ~> type enat
  1683   Fin         ~> enat
  1684   Infty       ~> infinity (overloaded)
  1685   iSuc        ~> eSuc
  1686   the_Fin     ~> the_enat
  1687 
  1688 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1689 been renamed accordingly. INCOMPATIBILITY.
  1690 
  1691 * Session Multivariate_Analysis: The euclidean_space type class now
  1692 fixes a constant "Basis :: 'a set" consisting of the standard
  1693 orthonormal basis for the type. Users now have the option of
  1694 quantifying over this set instead of using the "basis" function, e.g.
  1695 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1696 
  1697 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1698 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1699 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1700 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1701 changed to match. Definition theorems for overloaded constants now use
  1702 the standard "foo_vec_def" naming scheme. A few other theorems have
  1703 been renamed as follows (INCOMPATIBILITY):
  1704 
  1705   Cart_eq          ~> vec_eq_iff
  1706   dist_nth_le_cart ~> dist_vec_nth_le
  1707   tendsto_vector   ~> vec_tendstoI
  1708   Cauchy_vector    ~> vec_CauchyI
  1709 
  1710 * Session Multivariate_Analysis: Several duplicate theorems have been
  1711 removed, and other theorems have been renamed or replaced with more
  1712 general versions. INCOMPATIBILITY.
  1713 
  1714   finite_choice ~> finite_set_choice
  1715   eventually_conjI ~> eventually_conj
  1716   eventually_and ~> eventually_conj_iff
  1717   eventually_false ~> eventually_False
  1718   setsum_norm ~> norm_setsum
  1719   Lim_sequentially ~> LIMSEQ_def
  1720   Lim_ident_at ~> LIM_ident
  1721   Lim_const ~> tendsto_const
  1722   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1723   Lim_neg ~> tendsto_minus
  1724   Lim_add ~> tendsto_add
  1725   Lim_sub ~> tendsto_diff
  1726   Lim_mul ~> tendsto_scaleR
  1727   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1728   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1729   Lim_linear ~> bounded_linear.tendsto
  1730   Lim_component ~> tendsto_euclidean_component
  1731   Lim_component_cart ~> tendsto_vec_nth
  1732   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1733   dot_lsum ~> inner_setsum_left
  1734   dot_rsum ~> inner_setsum_right
  1735   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1736   continuous_neg ~> continuous_minus
  1737   continuous_sub ~> continuous_diff
  1738   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1739   continuous_mul ~> continuous_scaleR
  1740   continuous_inv ~> continuous_inverse
  1741   continuous_at_within_inv ~> continuous_at_within_inverse
  1742   continuous_at_inv ~> continuous_at_inverse
  1743   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1744   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1745   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1746   continuous_on_neg ~> continuous_on_minus
  1747   continuous_on_sub ~> continuous_on_diff
  1748   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1749   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1750   continuous_on_mul ~> continuous_on_scaleR
  1751   continuous_on_mul_real ~> continuous_on_mult
  1752   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1753   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1754   continuous_on_inverse ~> continuous_on_inv
  1755   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1756   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1757   subset_interior ~> interior_mono
  1758   subset_closure ~> closure_mono
  1759   closure_univ ~> closure_UNIV
  1760   real_arch_lt ~> reals_Archimedean2
  1761   real_arch ~> reals_Archimedean3
  1762   real_abs_norm ~> abs_norm_cancel
  1763   real_abs_sub_norm ~> norm_triangle_ineq3
  1764   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1765 
  1766 * Session HOL-Probability:
  1767   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1768   - Infinite products of probability measures are now available.
  1769   - Sigma closure is independent, if the generator is independent
  1770   - Use extended reals instead of positive extended
  1771     reals. INCOMPATIBILITY.
  1772 
  1773 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1774 
  1775   expand_fun_below ~> fun_below_iff
  1776   below_fun_ext ~> fun_belowI
  1777   expand_cfun_eq ~> cfun_eq_iff
  1778   ext_cfun ~> cfun_eqI
  1779   expand_cfun_below ~> cfun_below_iff
  1780   below_cfun_ext ~> cfun_belowI
  1781   monofun_fun_fun ~> fun_belowD
  1782   monofun_fun_arg ~> monofunE
  1783   monofun_lub_fun ~> adm_monofun [THEN admD]
  1784   cont_lub_fun ~> adm_cont [THEN admD]
  1785   cont2cont_Rep_CFun ~> cont2cont_APP
  1786   cont_Rep_CFun_app ~> cont_APP_app
  1787   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1788   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1789   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1790   contlub_cfun ~> lub_APP [symmetric]
  1791   contlub_LAM ~> lub_LAM [symmetric]
  1792   thelubI ~> lub_eqI
  1793   UU_I ~> bottomI
  1794   lift_distinct1 ~> lift.distinct(1)
  1795   lift_distinct2 ~> lift.distinct(2)
  1796   Def_not_UU ~> lift.distinct(2)
  1797   Def_inject ~> lift.inject
  1798   below_UU_iff ~> below_bottom_iff
  1799   eq_UU_iff ~> eq_bottom_iff
  1800 
  1801 
  1802 *** Document preparation ***
  1803 
  1804 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1805 isar-ref manual, both for description and actual application of the
  1806 same.
  1807 
  1808 * Antiquotation @{value} evaluates the given term and presents its
  1809 result.
  1810 
  1811 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1812 variables x1, y23 into subscripted form x\<^isub>1,
  1813 y\<^isub>2\<^isub>3.
  1814 
  1815 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1816 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1817 
  1818 * Localized \isabellestyle switch can be used within blocks or groups
  1819 like this:
  1820 
  1821   \isabellestyle{it}  %preferred default
  1822   {\isabellestylett @{text "typewriter stuff"}}
  1823 
  1824 * Discontinued special treatment of hard tabulators.  Implicit
  1825 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1826 layouts.
  1827 
  1828 
  1829 *** ML ***
  1830 
  1831 * The inner syntax of sort/type/term/prop supports inlined YXML
  1832 representations within quoted string tokens.  By encoding logical
  1833 entities via Term_XML (in ML or Scala) concrete syntax can be
  1834 bypassed, which is particularly useful for producing bits of text
  1835 under external program control.
  1836 
  1837 * Antiquotations for ML and document preparation are managed as theory
  1838 data, which requires explicit setup.
  1839 
  1840 * Isabelle_Process.is_active allows tools to check if the official
  1841 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1842 (better known as Proof General).
  1843 
  1844 * Structure Proof_Context follows standard naming scheme.  Old
  1845 ProofContext is still available for some time as legacy alias.
  1846 
  1847 * Structure Timing provides various operations for timing; supersedes
  1848 former start_timing/end_timing etc.
  1849 
  1850 * Path.print is the official way to show file-system paths to users
  1851 (including quotes etc.).
  1852 
  1853 * Inner syntax: identifiers in parse trees of generic categories
  1854 "logic", "aprop", "idt" etc. carry position information (disguised as
  1855 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1856 translations that choke on unexpected type constraints.  Positions can
  1857 be stripped in ML translations via Syntax.strip_positions /
  1858 Syntax.strip_positions_ast, or via the syntax constant
  1859 "_strip_positions" within parse trees.  As last resort, positions can
  1860 be disabled via the configuration option Syntax.positions, which is
  1861 called "syntax_positions" in Isar attribute syntax.
  1862 
  1863 * Discontinued special status of various ML structures that contribute
  1864 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1865 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1866 refer directly to Ast.Constant, Lexicon.is_identifier,
  1867 Syntax_Trans.mk_binder_tr etc.
  1868 
  1869 * Typed print translation: discontinued show_sorts argument, which is
  1870 already available via context of "advanced" translation.
  1871 
  1872 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1873 goal states; body tactic needs to address all subgoals uniformly.
  1874 
  1875 * Slightly more special eq_list/eq_set, with shortcut involving
  1876 pointer equality (assumes that eq relation is reflexive).
  1877 
  1878 * Classical tactics use proper Proof.context instead of historic types
  1879 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1880 operate directly on Proof.context.  Raw type claset retains its use as
  1881 snapshot of the classical context, which can be recovered via
  1882 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1883 INCOMPATIBILITY, classical tactics and derived proof methods require
  1884 proper Proof.context.
  1885 
  1886 
  1887 *** System ***
  1888 
  1889 * Discontinued support for Poly/ML 5.2, which was the last version
  1890 without proper multithreading and TimeLimit implementation.
  1891 
  1892 * Discontinued old lib/scripts/polyml-platform, which has been
  1893 obsolete since Isabelle2009-2.
  1894 
  1895 * Various optional external tools are referenced more robustly and
  1896 uniformly by explicit Isabelle settings as follows:
  1897 
  1898   ISABELLE_CSDP   (formerly CSDP_EXE)
  1899   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1900   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1901   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1902   ISABELLE_YAP    (formerly EXEC_YAP)
  1903 
  1904 Note that automated detection from the file-system or search path has
  1905 been discontinued.  INCOMPATIBILITY.
  1906 
  1907 * Scala layer provides JVM method invocation service for static
  1908 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1909 example:
  1910 
  1911   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1912 
  1913 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1914 this allows to pass structured values between ML and Scala.
  1915 
  1916 * The IsabelleText fonts includes some further glyphs to support the
  1917 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1918 installed a local copy (which is normally *not* required) need to
  1919 delete or update it from ~~/lib/fonts/.
  1920 
  1921 
  1922 
  1923 New in Isabelle2011 (January 2011)
  1924 ----------------------------------
  1925 
  1926 *** General ***
  1927 
  1928 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1929 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1930 useful tooltips and hyperlinks produced from its static analysis.  The
  1931 bundled component provides an executable Isabelle tool that can be run
  1932 like this:
  1933 
  1934   Isabelle2011/bin/isabelle jedit
  1935 
  1936 * Significantly improved Isabelle/Isar implementation manual.
  1937 
  1938 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1939 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1940 while the default heap location within that directory lacks that extra
  1941 suffix.  This isolates multiple Isabelle installations from each
  1942 other, avoiding problems with old settings in new versions.
  1943 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1944 
  1945 * Source files are always encoded as UTF-8, instead of old-fashioned
  1946 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1947 the following package declarations:
  1948 
  1949   \usepackage[utf8]{inputenc}
  1950   \usepackage{textcomp}
  1951 
  1952 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1953 a Unicode character is treated as a single symbol, not a sequence of
  1954 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1955 contain symbols without further backslash escapes, Unicode can now be
  1956 used here as well.  Recall that Symbol.explode in ML provides a
  1957 consistent view on symbols, while raw explode (or String.explode)
  1958 merely give a byte-oriented representation.
  1959 
  1960 * Theory loader: source files are primarily located via the master
  1961 directory of each theory node (where the .thy file itself resides).
  1962 The global load path is still partially available as legacy feature.
  1963 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1964 explicit paths, relatively to the theory.
  1965 
  1966 * Special treatment of ML file names has been discontinued.
  1967 Historically, optional extensions .ML or .sml were added on demand --
  1968 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1969 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1970 
  1971 * Various options that affect pretty printing etc. are now properly
  1972 handled within the context via configuration options, instead of
  1973 unsynchronized references or print modes.  There are both ML Config.T
  1974 entities and Isar declaration attributes to access these.
  1975 
  1976   ML (Config.T)                 Isar (attribute)
  1977 
  1978   eta_contract                  eta_contract
  1979   show_brackets                 show_brackets
  1980   show_sorts                    show_sorts
  1981   show_types                    show_types
  1982   show_question_marks           show_question_marks
  1983   show_consts                   show_consts
  1984   show_abbrevs                  show_abbrevs
  1985 
  1986   Syntax.ast_trace              syntax_ast_trace
  1987   Syntax.ast_stat               syntax_ast_stat
  1988   Syntax.ambiguity_level        syntax_ambiguity_level
  1989 
  1990   Goal_Display.goals_limit      goals_limit
  1991   Goal_Display.show_main_goal   show_main_goal
  1992 
  1993   Method.rule_trace             rule_trace
  1994 
  1995   Thy_Output.display            thy_output_display
  1996   Thy_Output.quotes             thy_output_quotes
  1997   Thy_Output.indent             thy_output_indent
  1998   Thy_Output.source             thy_output_source
  1999   Thy_Output.break              thy_output_break
  2000 
  2001 Note that corresponding "..._default" references in ML may only be
  2002 changed globally at the ROOT session setup, but *not* within a theory.
  2003 The option "show_abbrevs" supersedes the former print mode
  2004 "no_abbrevs" with inverted meaning.
  2005 
  2006 * More systematic naming of some configuration options.
  2007 INCOMPATIBILITY.
  2008 
  2009   trace_simp  ~>  simp_trace
  2010   debug_simp  ~>  simp_debug
  2011 
  2012 * Support for real valued configuration options, using simplistic
  2013 floating-point notation that coincides with the inner syntax for
  2014 float_token.
  2015 
  2016 * Support for real valued preferences (with approximative PGIP type):
  2017 front-ends need to accept "pgint" values in float notation.
  2018 INCOMPATIBILITY.
  2019 
  2020 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2021 DejaVu Sans.
  2022 
  2023 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2024 
  2025 
  2026 *** Pure ***
  2027 
  2028 * Command 'type_synonym' (with single argument) replaces somewhat
  2029 outdated 'types', which is still available as legacy feature for some
  2030 time.
  2031 
  2032 * Command 'nonterminal' (with 'and' separated list of arguments)
  2033 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2034 
  2035 * Command 'notepad' replaces former 'example_proof' for
  2036 experimentation in Isar without any result.  INCOMPATIBILITY.
  2037 
  2038 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2039 lists of equations to map definitions in a locale to appropriate
  2040 entities in the context of the interpretation.  The 'interpretation'
  2041 command already provided this functionality.
  2042 
  2043 * Diagnostic command 'print_dependencies' prints the locale instances
  2044 that would be activated if the specified expression was interpreted in
  2045 the current context.  Variant "print_dependencies!" assumes a context
  2046 without interpretations.
  2047 
  2048 * Diagnostic command 'print_interps' prints interpretations in proofs
  2049 in addition to interpretations in theories.
  2050 
  2051 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2052 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2053 Sign.root_path and Sign.local_path may be applied directly where this
  2054 feature is still required for historical reasons.
  2055 
  2056 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2057 'definition' instead.
  2058 
  2059 * The "prems" fact, which refers to the accidental collection of
  2060 foundational premises in the context, is now explicitly marked as
  2061 legacy feature and will be discontinued soon.  Consider using "assms"
  2062 of the head statement or reference facts by explicit names.
  2063 
  2064 * Document antiquotations @{class} and @{type} print classes and type
  2065 constructors.
  2066 
  2067 * Document antiquotation @{file} checks file/directory entries within
  2068 the local file system.
  2069 
  2070 
  2071 *** HOL ***
  2072 
  2073 * Coercive subtyping: functions can be declared as coercions and type
  2074 inference will add them as necessary upon input of a term.  Theory
  2075 Complex_Main declares real :: nat => real and real :: int => real as
  2076 coercions. A coercion function f is declared like this:
  2077 
  2078   declare [[coercion f]]
  2079 
  2080 To lift coercions through type constructors (e.g. from nat => real to
  2081 nat list => real list), map functions can be declared, e.g.
  2082 
  2083   declare [[coercion_map map]]
  2084 
  2085 Currently coercion inference is activated only in theories including
  2086 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2087 the configuration option "coercion_enabled", e.g. it can be enabled in
  2088 other theories like this:
  2089 
  2090   declare [[coercion_enabled]]
  2091 
  2092 * Command 'partial_function' provides basic support for recursive
  2093 function definitions over complete partial orders.  Concrete instances
  2094 are provided for i) the option type, ii) tail recursion on arbitrary
  2095 types, and iii) the heap monad of Imperative_HOL.  See
  2096 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2097 for examples.
  2098 
  2099 * Function package: f.psimps rules are no longer implicitly declared
  2100 as [simp].  INCOMPATIBILITY.
  2101 
  2102 * Datatype package: theorems generated for executable equality (class
  2103 "eq") carry proper names and are treated as default code equations.
  2104 
  2105 * Inductive package: now offers command 'inductive_simps' to
  2106 automatically derive instantiated and simplified equations for
  2107 inductive predicates, similar to 'inductive_cases'.
  2108 
  2109 * Command 'enriched_type' allows to register properties of the
  2110 functorial structure of types.
  2111 
  2112 * Improved infrastructure for term evaluation using code generator
  2113 techniques, in particular static evaluation conversions.
  2114 
  2115 * Code generator: Scala (2.8 or higher) has been added to the target
  2116 languages.
  2117 
  2118 * Code generator: globbing constant expressions "*" and "Theory.*"
  2119 have been replaced by the more idiomatic "_" and "Theory._".
  2120 INCOMPATIBILITY.
  2121 
  2122 * Code generator: export_code without explicit file declaration prints
  2123 to standard output.  INCOMPATIBILITY.
  2124 
  2125 * Code generator: do not print function definitions for case
  2126 combinators any longer.
  2127 
  2128 * Code generator: simplification with rules determined with
  2129 src/Tools/Code/code_simp.ML and method "code_simp".
  2130 
  2131 * Code generator for records: more idiomatic representation of record
  2132 types.  Warning: records are not covered by ancient SML code
  2133 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2134 rep_datatype declaration helps to succeed then:
  2135 
  2136   record 'a foo = ...
  2137   ...
  2138   rep_datatype foo_ext ...
  2139 
  2140 * Records: logical foundation type for records does not carry a
  2141 '_type' suffix any longer (obsolete due to authentic syntax).
  2142 INCOMPATIBILITY.
  2143 
  2144 * Quickcheck now by default uses exhaustive testing instead of random
  2145 testing.  Random testing can be invoked by "quickcheck [random]",
  2146 exhaustive testing by "quickcheck [exhaustive]".
  2147 
  2148 * Quickcheck instantiates polymorphic types with small finite
  2149 datatypes by default. This enables a simple execution mechanism to
  2150 handle quantifiers and function equality over the finite datatypes.
  2151 
  2152 * Quickcheck random generator has been renamed from "code" to
  2153 "random".  INCOMPATIBILITY.
  2154 
  2155 * Quickcheck now has a configurable time limit which is set to 30
  2156 seconds by default. This can be changed by adding [timeout = n] to the
  2157 quickcheck command. The time limit for Auto Quickcheck is still set
  2158 independently.
  2159 
  2160 * Quickcheck in locales considers interpretations of that locale for
  2161 counter example search.
  2162 
  2163 * Sledgehammer:
  2164   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2165     method. See the Sledgehammer manual for details ("isabelle doc
  2166     sledgehammer").
  2167   - Renamed commands:
  2168     sledgehammer atp_info ~> sledgehammer running_provers
  2169     sledgehammer atp_kill ~> sledgehammer kill_provers
  2170     sledgehammer available_atps ~> sledgehammer available_provers
  2171     INCOMPATIBILITY.
  2172   - Renamed options:
  2173     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2174     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2175     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2176     (and "ms" and "min" are no longer supported)
  2177     INCOMPATIBILITY.
  2178 
  2179 * Nitpick:
  2180   - Renamed options:
  2181     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2182     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2183     INCOMPATIBILITY.
  2184   - Added support for partial quotient types.
  2185   - Added local versions of the "Nitpick.register_xxx" functions.
  2186   - Added "whack" option.
  2187   - Allow registration of quotient types as codatatypes.
  2188   - Improved "merge_type_vars" option to merge more types.
  2189   - Removed unsound "fast_descrs" option.
  2190   - Added custom symmetry breaking for datatypes, making it possible to reach
  2191     higher cardinalities.
  2192   - Prevent the expansion of too large definitions.
  2193 
  2194 * Proof methods "metis" and "meson" now have configuration options
  2195 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2196 to diagnose these tools. E.g.
  2197 
  2198     using [[metis_trace = true]]
  2199 
  2200 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2201 manually as command 'solve_direct'.
  2202 
  2203 * The default SMT solver Z3 must be enabled explicitly (due to
  2204 licensing issues) by setting the environment variable
  2205 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2206 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2207 changing the SMT solver is done via the configuration option
  2208 "smt_solver".
  2209 
  2210 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2211 i.e. "remote_cvc3" and "remote_z3".
  2212 
  2213 * Added basic SMT support for datatypes, records, and typedefs using
  2214 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2215 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2216 support for a similar behavior).  Minor INCOMPATIBILITY.
  2217 
  2218 * Changed SMT configuration options:
  2219   - Renamed:
  2220     z3_proofs ~> smt_oracle (with inverted meaning)
  2221     z3_trace_assms ~> smt_trace_used_facts
  2222     INCOMPATIBILITY.
  2223   - Added:
  2224     smt_verbose
  2225     smt_random_seed
  2226     smt_datatypes
  2227     smt_infer_triggers
  2228     smt_monomorph_limit
  2229     cvc3_options
  2230     remote_cvc3_options
  2231     remote_z3_options
  2232     yices_options
  2233 
  2234 * Boogie output files (.b2i files) need to be declared in the theory
  2235 header.
  2236 
  2237 * Simplification procedure "list_to_set_comprehension" rewrites list
  2238 comprehensions applied to List.set to set comprehensions.  Occasional
  2239 INCOMPATIBILITY, may be deactivated like this:
  2240 
  2241   declare [[simproc del: list_to_set_comprehension]]
  2242 
  2243 * Removed old version of primrec package.  INCOMPATIBILITY.
  2244 
  2245 * Removed simplifier congruence rule of "prod_case", as has for long
  2246 been the case with "split".  INCOMPATIBILITY.
  2247 
  2248 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2249 
  2250 * Removed [split_format ... and ... and ...] version of
  2251 [split_format].  Potential INCOMPATIBILITY.
  2252 
  2253 * Predicate "sorted" now defined inductively, with nice induction
  2254 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2255 
  2256 * Constant "contents" renamed to "the_elem", to free the generic name
  2257 contents for other uses.  INCOMPATIBILITY.
  2258 
  2259 * Renamed class eq and constant eq (for code generation) to class
  2260 equal and constant equal, plus renaming of related facts and various
  2261 tuning.  INCOMPATIBILITY.
  2262 
  2263 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2264 
  2265 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2266 INCOMPATIBILITY.
  2267 
  2268 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2269 avoid confusion with finite sets.  INCOMPATIBILITY.
  2270 
  2271 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2272 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2273 for congruent(2)).
  2274 
  2275 * Some previously unqualified names have been qualified:
  2276 
  2277   types
  2278     bool ~> HOL.bool
  2279     nat ~> Nat.nat
  2280 
  2281   constants
  2282     Trueprop ~> HOL.Trueprop
  2283     True ~> HOL.True
  2284     False ~> HOL.False
  2285     op & ~> HOL.conj
  2286     op | ~> HOL.disj
  2287     op --> ~> HOL.implies
  2288     op = ~> HOL.eq
  2289     Not ~> HOL.Not
  2290     The ~> HOL.The
  2291     All ~> HOL.All
  2292     Ex ~> HOL.Ex
  2293     Ex1 ~> HOL.Ex1
  2294     Let ~> HOL.Let
  2295     If ~> HOL.If
  2296     Ball ~> Set.Ball
  2297     Bex ~> Set.Bex
  2298     Suc ~> Nat.Suc
  2299     Pair ~> Product_Type.Pair
  2300     fst ~> Product_Type.fst
  2301     snd ~> Product_Type.snd
  2302     curry ~> Product_Type.curry
  2303     op : ~> Set.member
  2304     Collect ~> Set.Collect
  2305 
  2306 INCOMPATIBILITY.
  2307 
  2308 * More canonical naming convention for some fundamental definitions:
  2309 
  2310     bot_bool_eq ~> bot_bool_def
  2311     top_bool_eq ~> top_bool_def
  2312     inf_bool_eq ~> inf_bool_def
  2313     sup_bool_eq ~> sup_bool_def
  2314     bot_fun_eq  ~> bot_fun_def
  2315     top_fun_eq  ~> top_fun_def
  2316     inf_fun_eq  ~> inf_fun_def
  2317     sup_fun_eq  ~> sup_fun_def
  2318 
  2319 INCOMPATIBILITY.
  2320 
  2321 * More stylized fact names:
  2322 
  2323   expand_fun_eq ~> fun_eq_iff
  2324   expand_set_eq ~> set_eq_iff
  2325   set_ext       ~> set_eqI
  2326   nat_number    ~> eval_nat_numeral
  2327 
  2328 INCOMPATIBILITY.
  2329 
  2330 * Refactoring of code-generation specific operations in theory List:
  2331 
  2332   constants
  2333     null ~> List.null
  2334 
  2335   facts
  2336     mem_iff ~> member_def
  2337     null_empty ~> null_def
  2338 
  2339 INCOMPATIBILITY.  Note that these were not supposed to be used
  2340 regularly unless for striking reasons; their main purpose was code
  2341 generation.
  2342 
  2343 Various operations from the Haskell prelude are used for generating
  2344 Haskell code.
  2345 
  2346 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2347 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2348 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2349 
  2350 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2351 "*" and "+" respectively.  INCOMPATIBILITY.
  2352 
  2353 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2354 "Sum_Type.Plus" instead.
  2355 
  2356 * Constant "split" has been merged with constant "prod_case"; names of
  2357 ML functions, facts etc. involving split have been retained so far,
  2358 though.  INCOMPATIBILITY.
  2359 
  2360 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2361 instead.  INCOMPATIBILITY.
  2362 
  2363 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2364 INCOMPATIBILITY.
  2365 
  2366 * Former theory Library/Enum is now part of the HOL-Main image.
  2367 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2368 referred to by its qualified name.
  2369 
  2370   enum    ~>  Enum.enum
  2371   nlists  ~>  Enum.nlists
  2372   product ~>  Enum.product
  2373 
  2374 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2375 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2376 INCOMPATIBILITY.
  2377 
  2378 * Theory Library/SetsAndFunctions has been split into
  2379 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2380 for instance definitions for functions; various improvements.
  2381 INCOMPATIBILITY.
  2382 
  2383 * Theory Library/Multiset provides stable quicksort implementation of
  2384 sort_key.
  2385 
  2386 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2387 INCOMPATIBILITY.
  2388 
  2389 * Session Multivariate_Analysis: introduced a type class for euclidean
  2390 space.  Most theorems are now stated in terms of euclidean spaces
  2391 instead of finite cartesian products.
  2392 
  2393   types
  2394     real ^ 'n ~>  'a::real_vector
  2395               ~>  'a::euclidean_space
  2396               ~>  'a::ordered_euclidean_space
  2397         (depends on your needs)
  2398 
  2399   constants
  2400      _ $ _        ~> _ $$ _
  2401      \<chi> x. _  ~> \<chi>\<chi> x. _
  2402      CARD('n)     ~> DIM('a)
  2403 
  2404 Also note that the indices are now natural numbers and not from some
  2405 finite type. Finite cartesian products of euclidean spaces, products
  2406 of euclidean spaces the real and complex numbers are instantiated to
  2407 be euclidean_spaces.  INCOMPATIBILITY.
  2408 
  2409 * Session Probability: introduced pextreal as positive extended real
  2410 numbers.  Use pextreal as value for measures.  Introduce the
  2411 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2412 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2413 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2414 
  2415 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2416 INCOMPATIBILITY.
  2417 
  2418 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2419 prove verification conditions generated by the SPARK Ada program
  2420 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2421 
  2422 
  2423 *** HOL-Algebra ***
  2424 
  2425 * Theorems for additive ring operations (locale abelian_monoid and
  2426 descendants) are generated by interpretation from their multiplicative
  2427 counterparts.  Names (in particular theorem names) have the mandatory
  2428 qualifier 'add'.  Previous theorem names are redeclared for
  2429 compatibility.
  2430 
  2431 * Structure "int_ring" is now an abbreviation (previously a
  2432 definition).  This fits more natural with advanced interpretations.
  2433 
  2434 
  2435 *** HOLCF ***
  2436 
  2437 * The domain package now runs in definitional mode by default: The
  2438 former command 'new_domain' is now called 'domain'.  To use the domain
  2439 package in its original axiomatic mode, use 'domain (unsafe)'.
  2440 INCOMPATIBILITY.
  2441 
  2442 * The new class "domain" is now the default sort.  Class "predomain"
  2443 is an unpointed version of "domain". Theories can be updated by
  2444 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2445 
  2446   'a::type ~> 'a::countable
  2447   'a::cpo  ~> 'a::predomain
  2448   'a::pcpo ~> 'a::domain
  2449 
  2450 * The old type class "rep" has been superseded by class "domain".
  2451 Accordingly, users of the definitional package must remove any
  2452 "default_sort rep" declarations.  INCOMPATIBILITY.
  2453 
  2454 * The domain package (definitional mode) now supports unpointed
  2455 predomain argument types, as long as they are marked 'lazy'. (Strict
  2456 arguments must be in class "domain".) For example, the following
  2457 domain definition now works:
  2458 
  2459   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2460 
  2461 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2462 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2463 'a option, and 'a list.  Additionally, it configures fixrec and the
  2464 domain package to work with these types.  For example:
  2465 
  2466   fixrec isInl :: "('a + 'b) u -> tr"
  2467     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2468 
  2469   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2470 
  2471 * The "(permissive)" option of fixrec has been replaced with a
  2472 per-equation "(unchecked)" option. See
  2473 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2474 
  2475 * The "bifinite" class no longer fixes a constant "approx"; the class
  2476 now just asserts that such a function exists.  INCOMPATIBILITY.
  2477 
  2478 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2479 defines an embedding of type 'a defl into udom by default; instances
  2480 of "bifinite" and "domain" classes are available in
  2481 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2482 
  2483 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2484 
  2485 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2486 
  2487 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2488 
  2489 * The function "cprod_map" has been renamed to "prod_map".
  2490 INCOMPATIBILITY.
  2491 
  2492 * The monadic bind operator on each powerdomain has new binder syntax
  2493 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2494 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2495 
  2496 * The infix syntax for binary union on each powerdomain has changed
  2497 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2498 syntax.  INCOMPATIBILITY.
  2499 
  2500 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2501 still supported as an input translation.
  2502 
  2503 * Renamed some theorems (the original names are also still available).
  2504 
  2505   expand_fun_below   ~> fun_below_iff
  2506   below_fun_ext      ~> fun_belowI
  2507   expand_cfun_eq     ~> cfun_eq_iff
  2508   ext_cfun           ~> cfun_eqI
  2509   expand_cfun_below  ~> cfun_below_iff
  2510   below_cfun_ext     ~> cfun_belowI
  2511   cont2cont_Rep_CFun ~> cont2cont_APP
  2512 
  2513 * The Abs and Rep functions for various types have changed names.
  2514 Related theorem names have also changed to match. INCOMPATIBILITY.
  2515 
  2516   Rep_CFun  ~> Rep_cfun
  2517   Abs_CFun  ~> Abs_cfun
  2518   Rep_Sprod ~> Rep_sprod
  2519   Abs_Sprod ~> Abs_sprod
  2520   Rep_Ssum  ~> Rep_ssum
  2521   Abs_Ssum  ~> Abs_ssum
  2522 
  2523 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2524 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2525 
  2526 * Various changes to bisimulation/coinduction with domain package:
  2527 
  2528   - Definitions of "bisim" constants no longer mention definedness.
  2529   - With mutual recursion, "bisim" predicate is now curried.
  2530   - With mutual recursion, each type gets a separate coind theorem.
  2531   - Variable names in bisim_def and coinduct rules have changed.
  2532 
  2533 INCOMPATIBILITY.
  2534 
  2535 * Case combinators generated by the domain package for type "foo" are
  2536 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2537 
  2538 * Several theorems have been renamed to more accurately reflect the
  2539 names of constants and types involved.  INCOMPATIBILITY.
  2540 
  2541   thelub_const    ~> lub_const
  2542   lub_const       ~> is_lub_const
  2543   thelubI         ~> lub_eqI
  2544   is_lub_lub      ~> is_lubD2
  2545   lubI            ~> is_lub_lub
  2546   unique_lub      ~> is_lub_unique
  2547   is_ub_lub       ~> is_lub_rangeD1
  2548   lub_bin_chain   ~> is_lub_bin_chain
  2549   lub_fun         ~> is_lub_fun
  2550   thelub_fun      ~> lub_fun
  2551   thelub_cfun     ~> lub_cfun
  2552   thelub_Pair     ~> lub_Pair
  2553   lub_cprod       ~> is_lub_prod
  2554   thelub_cprod    ~> lub_prod
  2555   minimal_cprod   ~> minimal_prod
  2556   inst_cprod_pcpo ~> inst_prod_pcpo
  2557   UU_I            ~> bottomI
  2558   compact_UU      ~> compact_bottom
  2559   deflation_UU    ~> deflation_bottom
  2560   finite_deflation_UU ~> finite_deflation_bottom
  2561 
  2562 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2563 
  2564   sq_ord_less_eq_trans ~> below_eq_trans
  2565   sq_ord_eq_less_trans ~> eq_below_trans
  2566   refl_less            ~> below_refl
  2567   trans_less           ~> below_trans
  2568   antisym_less         ~> below_antisym
  2569   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2570   box_less             ~> box_below
  2571   rev_trans_less       ~> rev_below_trans
  2572   not_less2not_eq      ~> not_below2not_eq
  2573   less_UU_iff          ~> below_UU_iff
  2574   flat_less_iff        ~> flat_below_iff
  2575   adm_less             ~> adm_below
  2576   adm_not_less         ~> adm_not_below
  2577   adm_compact_not_less ~> adm_compact_not_below
  2578   less_fun_def         ~> below_fun_def
  2579   expand_fun_less      ~> fun_below_iff
  2580   less_fun_ext         ~> fun_belowI
  2581   less_discr_def       ~> below_discr_def
  2582   discr_less_eq        ~> discr_below_eq
  2583   less_unit_def        ~> below_unit_def
  2584   less_cprod_def       ~> below_prod_def
  2585   prod_lessI           ~> prod_belowI
  2586   Pair_less_iff        ~> Pair_below_iff
  2587   fst_less_iff         ~> fst_below_iff
  2588   snd_less_iff         ~> snd_below_iff
  2589   expand_cfun_less     ~> cfun_below_iff
  2590   less_cfun_ext        ~> cfun_belowI
  2591   injection_less       ~> injection_below
  2592   less_up_def          ~> below_up_def
  2593   not_Iup_less         ~> not_Iup_below
  2594   Iup_less             ~> Iup_below
  2595   up_less              ~> up_below
  2596   Def_inject_less_eq   ~> Def_below_Def
  2597   Def_less_is_eq       ~> Def_below_iff
  2598   spair_less_iff       ~> spair_below_iff
  2599   less_sprod           ~> below_sprod
  2600   spair_less           ~> spair_below
  2601   sfst_less_iff        ~> sfst_below_iff
  2602   ssnd_less_iff        ~> ssnd_below_iff
  2603   fix_least_less       ~> fix_least_below
  2604   dist_less_one        ~> dist_below_one
  2605   less_ONE             ~> below_ONE
  2606   ONE_less_iff         ~> ONE_below_iff
  2607   less_sinlD           ~> below_sinlD
  2608   less_sinrD           ~> below_sinrD
  2609 
  2610 
  2611 *** FOL and ZF ***
  2612 
  2613 * All constant names are now qualified internally and use proper
  2614 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2615 
  2616 
  2617 *** ML ***
  2618 
  2619 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2620 Fail if the argument is false.  Due to inlining the source position of
  2621 failed assertions is included in the error output.
  2622 
  2623 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2624 text is in practice always evaluated with a stable theory checkpoint.
  2625 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2626 
  2627 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2628 the current context, not any accidental theory loader state as before.
  2629 Potential INCOMPATIBILITY, subtle change in semantics.
  2630 
  2631 * Syntax.pretty_priority (default 0) configures the required priority
  2632 of pretty-printed output and thus affects insertion of parentheses.
  2633 
  2634 * Syntax.default_root (default "any") configures the inner syntax
  2635 category (nonterminal symbol) for parsing of terms.
  2636 
  2637 * Former exception Library.UnequalLengths now coincides with
  2638 ListPair.UnequalLengths.
  2639 
  2640 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2641 main functionality is provided by structure Simplifier.
  2642 
  2643 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2644 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2645 almost all situations.
  2646 
  2647 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2648 See implementation manual for further details on exceptions in
  2649 Isabelle/ML.
  2650 
  2651 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2652 meaning.
  2653 
  2654 * Renamed structure PureThy to Pure_Thy and moved most of its
  2655 operations to structure Global_Theory, to emphasize that this is
  2656 rarely-used global-only stuff.
  2657 
  2658 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2659 instead (or tracing for high-volume output).
  2660 
  2661 * Configuration option show_question_marks only affects regular pretty
  2662 printing of types and terms, not raw Term.string_of_vname.
  2663 
  2664 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2665 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2666 @{thms} for most purposes.
  2667 
  2668 * ML structure Unsynchronized is never opened, not even in Isar
  2669 interaction mode as before.  Old Unsynchronized.set etc. have been
  2670 discontinued -- use plain := instead.  This should be *rare* anyway,
  2671 since modern tools always work via official context data, notably
  2672 configuration options.
  2673 
  2674 * Parallel and asynchronous execution requires special care concerning
  2675 interrupts.  Structure Exn provides some convenience functions that
  2676 avoid working directly with raw Interrupt.  User code must not absorb
  2677 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2678 followed by re-raising of the original exception.  Another common
  2679 source of mistakes are "handle _" patterns, which make the meaning of
  2680 the program subject to physical effects of the environment.
  2681 
  2682 
  2683 
  2684 New in Isabelle2009-2 (June 2010)
  2685 ---------------------------------
  2686 
  2687 *** General ***
  2688 
  2689 * Authentic syntax for *all* logical entities (type classes, type
  2690 constructors, term constants): provides simple and robust
  2691 correspondence between formal entities and concrete syntax.  Within
  2692 the parse tree / AST representations, "constants" are decorated by
  2693 their category (class, type, const) and spelled out explicitly with
  2694 their full internal name.
  2695 
  2696 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2697 and translations (translation rules and translation functions in ML).
  2698 Some hints on upgrading:
  2699 
  2700   - Many existing uses of 'syntax' and 'translations' can be replaced
  2701     by more modern 'type_notation', 'notation' and 'abbreviation',
  2702     which are independent of this issue.
  2703 
  2704   - 'translations' require markup within the AST; the term syntax
  2705     provides the following special forms:
  2706 
  2707       CONST c   -- produces syntax version of constant c from context
  2708       XCONST c  -- literally c, checked as constant from context
  2709       c         -- literally c, if declared by 'syntax'
  2710 
  2711     Plain identifiers are treated as AST variables -- occasionally the
  2712     system indicates accidental variables via the error "rhs contains
  2713     extra variables".
  2714 
  2715     Type classes and type constructors are marked according to their
  2716     concrete syntax.  Some old translations rules need to be written
  2717     for the "type" category, using type constructor application
  2718     instead of pseudo-term application of the default category
  2719     "logic".
  2720 
  2721   - 'parse_translation' etc. in ML may use the following
  2722     antiquotations:
  2723 
  2724       @{class_syntax c}   -- type class c within parse tree / AST
  2725       @{term_syntax c}    -- type constructor c within parse tree / AST
  2726       @{const_syntax c}   -- ML version of "CONST c" above
  2727       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2728 
  2729   - Literal types within 'typed_print_translations', i.e. those *not*
  2730     represented as pseudo-terms are represented verbatim.  Use @{class
  2731     c} or @{type_name c} here instead of the above syntax
  2732     antiquotations.
  2733 
  2734 Note that old non-authentic syntax was based on unqualified base
  2735 names, so all of the above "constant" names would coincide.  Recall
  2736 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2737 diagnose syntax problems.
  2738 
  2739 * Type constructors admit general mixfix syntax, not just infix.
  2740 
  2741 * Concrete syntax may be attached to local entities without a proof
  2742 body, too.  This works via regular mixfix annotations for 'fix',
  2743 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2744 similar to the 'notation' command in theory specifications.
  2745 
  2746 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2747 need to specify constant name and syntax separately.  Internal ML
  2748 datatype constructors have been renamed from InfixName to Infix etc.
  2749 Minor INCOMPATIBILITY.
  2750 
  2751 * Schematic theorem statements need to be explicitly markup as such,
  2752 via commands 'schematic_lemma', 'schematic_theorem',
  2753 'schematic_corollary'.  Thus the relevance of the proof is made
  2754 syntactically clear, which impacts performance in a parallel or
  2755 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2756 
  2757 * Use of cumulative prems via "!" in some proof methods has been
  2758 discontinued (old legacy feature).
  2759 
  2760 * References 'trace_simp' and 'debug_simp' have been replaced by
  2761 configuration options stored in the context. Enabling tracing (the
  2762 case of debugging is similar) in proofs works via
  2763 
  2764   using [[trace_simp = true]]
  2765 
  2766 Tracing is then active for all invocations of the simplifier in
  2767 subsequent goal refinement steps. Tracing may also still be enabled or
  2768 disabled via the ProofGeneral settings menu.
  2769 
  2770 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2771 'hide_fact' replace the former 'hide' KIND command.  Minor
  2772 INCOMPATIBILITY.
  2773 
  2774 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2775 more efficient than combinations with -q1 or -q2.
  2776 
  2777 
  2778 *** Pure ***
  2779 
  2780 * Proofterms record type-class reasoning explicitly, using the
  2781 "unconstrain" operation internally.  This eliminates all sort
  2782 constraints from a theorem and proof, introducing explicit
  2783 OFCLASS-premises.  On the proof term level, this operation is
  2784 automatically applied at theorem boundaries, such that closed proofs
  2785 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2786 inspect proof terms.
  2787 
  2788 * Local theory specifications may depend on extra type variables that
  2789 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2790 are added internally.  For example:
  2791 
  2792   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2793 
  2794 * Predicates of locales introduced by classes carry a mandatory
  2795 "class" prefix.  INCOMPATIBILITY.
  2796 
  2797 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2798 
  2799 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2800 'class' instead.
  2801 
  2802 * Command 'code_reflect' allows to incorporate generated ML code into
  2803 runtime environment; replaces immature code_datatype antiquotation.
  2804 INCOMPATIBILITY.
  2805 
  2806 * Code generator: simple concept for abstract datatypes obeying
  2807 invariants.
  2808 
  2809 * Code generator: details of internal data cache have no impact on the
  2810 user space functionality any longer.
  2811 
  2812 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2813 subgoals.  This is more appropriate for interpretations with 'where'.
  2814 INCOMPATIBILITY.
  2815 
  2816 * Command 'example_proof' opens an empty proof body.  This allows to
  2817 experiment with Isar, without producing any persistent result.
  2818 
  2819 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2820 within a local theory context, with explicit checking of the
  2821 constructors involved (in contrast to the raw 'syntax' versions).
  2822 
  2823 * Commands 'types' and 'typedecl' now work within a local theory
  2824 context -- without introducing dependencies on parameters or
  2825 assumptions, which is not possible in Isabelle/Pure.
  2826 
  2827 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2828 within a local theory context.  Minor INCOMPATIBILITY.
  2829 
  2830 
  2831 *** HOL ***
  2832 
  2833 * Command 'typedef' now works within a local theory context -- without
  2834 introducing dependencies on parameters or assumptions, which is not
  2835 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2836 contain multiple interpretations of local typedefs (with different
  2837 non-emptiness proofs), even in a global theory context.
  2838 
  2839 * New package for quotient types.  Commands 'quotient_type' and
  2840 'quotient_definition' may be used for defining types and constants by
  2841 quotient constructions.  An example is the type of integers created by
  2842 quotienting pairs of natural numbers:
  2843 
  2844   fun
  2845     intrel :: "(nat * nat) => (nat * nat) => bool"
  2846   where
  2847     "intrel (x, y) (u, v) = (x + v = u + y)"
  2848 
  2849   quotient_type int = "nat * nat" / intrel
  2850     by (auto simp add: equivp_def expand_fun_eq)
  2851 
  2852   quotient_definition
  2853     "0::int" is "(0::nat, 0::nat)"
  2854 
  2855 The method "lifting" can be used to lift of theorems from the
  2856 underlying "raw" type to the quotient type.  The example
  2857 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2858 construction and provides a reasoning infrastructure for finite sets.
  2859 
  2860 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2861 clash with new theory Quotient in Main HOL.
  2862 
  2863 * Moved the SMT binding into the main HOL session, eliminating
  2864 separate HOL-SMT session.
  2865 
  2866 * List membership infix mem operation is only an input abbreviation.
  2867 INCOMPATIBILITY.
  2868 
  2869 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2870 for future developements; former Library/Word.thy is still present in
  2871 the AFP entry RSAPPS.
  2872 
  2873 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2874 longer shadowed.  INCOMPATIBILITY.
  2875 
  2876 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2877 INCOMPATIBILITY.
  2878 
  2879 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2880 INCOMPATIBILITY.
  2881 
  2882 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2883 instead.  INCOMPATIBILITY.
  2884 
  2885 * Dropped several real-specific versions of lemmas about floor and
  2886 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2887 instead.  INCOMPATIBILITY.
  2888 
  2889   floor_number_of_eq         ~> floor_number_of
  2890   le_floor_eq_number_of      ~> number_of_le_floor
  2891   le_floor_eq_zero           ~> zero_le_floor
  2892   le_floor_eq_one            ~> one_le_floor
  2893   floor_less_eq_number_of    ~> floor_less_number_of
  2894   floor_less_eq_zero         ~> floor_less_zero
  2895   floor_less_eq_one          ~> floor_less_one
  2896   less_floor_eq_number_of    ~> number_of_less_floor
  2897   less_floor_eq_zero         ~> zero_less_floor
  2898   less_floor_eq_one          ~> one_less_floor
  2899   floor_le_eq_number_of      ~> floor_le_number_of
  2900   floor_le_eq_zero           ~> floor_le_zero
  2901   floor_le_eq_one            ~> floor_le_one
  2902   floor_subtract_number_of   ~> floor_diff_number_of
  2903   floor_subtract_one         ~> floor_diff_one
  2904   ceiling_number_of_eq       ~> ceiling_number_of
  2905   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2906   ceiling_le_zero_eq         ~> ceiling_le_zero
  2907   ceiling_le_eq_one          ~> ceiling_le_one
  2908   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2909   less_ceiling_eq_zero       ~> zero_less_ceiling
  2910   less_ceiling_eq_one        ~> one_less_ceiling
  2911   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2912   ceiling_less_eq_zero       ~> ceiling_less_zero
  2913   ceiling_less_eq_one        ~> ceiling_less_one
  2914   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2915   le_ceiling_eq_zero         ~> zero_le_ceiling
  2916   le_ceiling_eq_one          ~> one_le_ceiling
  2917   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2918   ceiling_subtract_one       ~> ceiling_diff_one
  2919 
  2920 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2921 application of the various fold combinators on finite sets.
  2922 
  2923 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2924 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2925 implementation.  INCOMPATIBILTY.
  2926 
  2927 * Theory Library/Coinductive_List has been removed -- superseded by
  2928 AFP/thys/Coinductive.
  2929 
  2930 * Theory PReal, including the type "preal" and related operations, has
  2931 been removed.  INCOMPATIBILITY.
  2932 
  2933 * Real: new development using Cauchy Sequences.
  2934 
  2935 * Split off theory "Big_Operators" containing setsum, setprod,
  2936 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2937 
  2938 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2939 "Int" etc.  INCOMPATIBILITY.
  2940 
  2941 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2942 
  2943 * New set of rules "ac_simps" provides combined assoc / commute
  2944 rewrites for all interpretations of the appropriate generic locales.
  2945 
  2946 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2947 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2948 appropriate and more consistent names suitable for name prefixes
  2949 within the HOL theories.  INCOMPATIBILITY.
  2950 
  2951 * Some generic constants have been put to appropriate theories:
  2952   - less_eq, less: Orderings
  2953   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2954   - inverse, divide: Rings
  2955 INCOMPATIBILITY.
  2956 
  2957 * More consistent naming of type classes involving orderings (and
  2958 lattices):
  2959 
  2960     lower_semilattice                   ~> semilattice_inf
  2961     upper_semilattice                   ~> semilattice_sup
  2962 
  2963     dense_linear_order                  ~> dense_linorder
  2964 
  2965     pordered_ab_group_add               ~> ordered_ab_group_add
  2966     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2967     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2968     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2969     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2970     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2971     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2972     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2973     pordered_comm_ring                  ~> ordered_comm_ring
  2974     pordered_comm_semiring              ~> ordered_comm_semiring
  2975     pordered_ring                       ~> ordered_ring
  2976     pordered_ring_abs                   ~> ordered_ring_abs
  2977     pordered_semiring                   ~> ordered_semiring
  2978 
  2979     ordered_ab_group_add                ~> linordered_ab_group_add
  2980     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2981     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2982     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2983     ordered_field                       ~> linordered_field
  2984     ordered_field_no_lb                 ~> linordered_field_no_lb
  2985     ordered_field_no_ub                 ~> linordered_field_no_ub
  2986     ordered_field_dense_linear_order    ~> dense_linordered_field
  2987     ordered_idom                        ~> linordered_idom
  2988     ordered_ring                        ~> linordered_ring
  2989     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2990     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2991     ordered_ring_strict                 ~> linordered_ring_strict
  2992     ordered_semidom                     ~> linordered_semidom
  2993     ordered_semiring                    ~> linordered_semiring
  2994     ordered_semiring_1                  ~> linordered_semiring_1
  2995     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2996     ordered_semiring_strict             ~> linordered_semiring_strict
  2997 
  2998   The following slightly odd type classes have been moved to a
  2999   separate theory Library/Lattice_Algebras:
  3000 
  3001     lordered_ab_group_add               ~> lattice_ab_group_add
  3002     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  3003     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  3004     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  3005     lordered_ring                       ~> lattice_ring
  3006 
  3007 INCOMPATIBILITY.
  3008 
  3009 * Refined field classes:
  3010   - classes division_ring_inverse_zero, field_inverse_zero,
  3011     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3012     subsumes former division_by_zero class;
  3013   - numerous lemmas have been ported from field to division_ring.
  3014 INCOMPATIBILITY.
  3015 
  3016 * Refined algebra theorem collections:
  3017   - dropped theorem group group_simps, use algebra_simps instead;
  3018   - dropped theorem group ring_simps, use field_simps instead;
  3019   - proper theorem collection field_simps subsumes former theorem
  3020     groups field_eq_simps and field_simps;
  3021   - dropped lemma eq_minus_self_iff which is a duplicate for
  3022     equal_neg_zero.
  3023 INCOMPATIBILITY.
  3024 
  3025 * Theory Finite_Set and List: some lemmas have been generalized from
  3026 sets to lattices:
  3027 
  3028   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3029   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3030   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3031   union_Union_fold_union        ~> sup_Sup_fold_sup
  3032   Inter_fold_inter              ~> Inf_fold_inf
  3033   Union_fold_union              ~> Sup_fold_sup
  3034   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3035   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3036   INTER_fold_inter              ~> INFI_fold_inf
  3037   UNION_fold_union              ~> SUPR_fold_sup
  3038 
  3039 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3040 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3041 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3042 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3043 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3044 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3045 
  3046 * Reorganized theory Multiset: swapped notation of pointwise and
  3047 multiset order:
  3048 
  3049   - pointwise ordering is instance of class order with standard syntax
  3050     <= and <;
  3051   - multiset ordering has syntax <=# and <#; partial order properties
  3052     are provided by means of interpretation with prefix
  3053     multiset_order;
  3054   - less duplication, less historical organization of sections,
  3055     conversion from associations lists to multisets, rudimentary code
  3056     generation;
  3057   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3058     if needed.
  3059 
  3060 Renamed:
  3061 
  3062   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3063   multi_count_ext  ~>  multiset_ext
  3064   diff_union_inverse2  ~>  diff_union_cancelR
  3065 
  3066 INCOMPATIBILITY.
  3067 
  3068 * Theory Permutation: replaced local "remove" by List.remove1.
  3069 
  3070 * Code generation: ML and OCaml code is decorated with signatures.
  3071 
  3072 * Theory List: added transpose.
  3073 
  3074 * Library/Nat_Bijection.thy is a collection of bijective functions
  3075 between nat and other types, which supersedes the older libraries
  3076 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3077 
  3078   Constants:
  3079   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3080   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3081   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3082   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3083   Countable.pair_encode           ~> prod_encode
  3084   NatIso.prod2nat                 ~> prod_encode
  3085   NatIso.nat2prod                 ~> prod_decode
  3086   NatIso.sum2nat                  ~> sum_encode
  3087   NatIso.nat2sum                  ~> sum_decode
  3088   NatIso.list2nat                 ~> list_encode
  3089   NatIso.nat2list                 ~> list_decode
  3090   NatIso.set2nat                  ~> set_encode
  3091   NatIso.nat2set                  ~> set_decode
  3092 
  3093   Lemmas:
  3094   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3095   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3096   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3097   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3098   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3099   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3100   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3101   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3102   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3103   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3104   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3105   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3106   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3107 
  3108 * Sledgehammer:
  3109   - Renamed ATP commands:
  3110     atp_info     ~> sledgehammer running_atps
  3111     atp_kill     ~> sledgehammer kill_atps
  3112     atp_messages ~> sledgehammer messages
  3113     atp_minimize ~> sledgehammer minimize
  3114     print_atps   ~> sledgehammer available_atps
  3115     INCOMPATIBILITY.
  3116   - Added user's manual ("isabelle doc sledgehammer").
  3117   - Added option syntax and "sledgehammer_params" to customize
  3118     Sledgehammer's behavior.  See the manual for details.
  3119   - Modified the Isar proof reconstruction code so that it produces
  3120     direct proofs rather than proofs by contradiction.  (This feature
  3121     is still experimental.)
  3122   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3123     full-typed mode.
  3124   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3125 
  3126 * Nitpick:
  3127   - Added and implemented "binary_ints" and "bits" options.
  3128   - Added "std" option and implemented support for nonstandard models.
  3129   - Added and implemented "finitize" option to improve the precision
  3130     of infinite datatypes based on a monotonicity analysis.
  3131   - Added support for quotient types.
  3132   - Added support for "specification" and "ax_specification"
  3133     constructs.
  3134   - Added support for local definitions (for "function" and
  3135     "termination" proofs).
  3136   - Added support for term postprocessors.
  3137   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3138   - Improved efficiency of "destroy_constrs" optimization.
  3139   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3140     record getters.
  3141   - Fixed soundness bug related to higher-order constructors.
  3142   - Fixed soundness bug when "full_descrs" is enabled.
  3143   - Improved precision of set constructs.
  3144   - Added "atoms" option.
  3145   - Added cache to speed up repeated Kodkod invocations on the same
  3146     problems.
  3147   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3148     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3149     "SAT4J_Light".  INCOMPATIBILITY.
  3150   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3151     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3152   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3153 
  3154 * Method "induct" now takes instantiations of the form t, where t is not
  3155   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3156   If this is not intended, t has to be enclosed in parentheses.
  3157   By default, the equalities generated by definitional instantiations
  3158   are pre-simplified, which may cause parameters of inductive cases
  3159   to disappear, or may even delete some of the inductive cases.
  3160   Use "induct (no_simp)" instead of "induct" to restore the old
  3161   behaviour. The (no_simp) option is also understood by the "cases"
  3162   and "nominal_induct" methods, which now perform pre-simplification, too.
  3163   INCOMPATIBILITY.
  3164 
  3165 
  3166 *** HOLCF ***
  3167 
  3168 * Variable names in lemmas generated by the domain package have
  3169 changed; the naming scheme is now consistent with the HOL datatype
  3170 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3171 
  3172 * The domain package no longer defines the function "foo_copy" for
  3173 recursive domain "foo".  The reach lemma is now stated directly in
  3174 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3175 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3176 
  3177 * Most definedness lemmas generated by the domain package (previously
  3178 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3179 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3180 Proofs that used definedness lemmas as intro rules may break,
  3181 potential INCOMPATIBILITY.
  3182 
  3183 * Induction and casedist rules generated by the domain package now
  3184 declare proper case_names (one called "bottom", and one named for each
  3185 constructor).  INCOMPATIBILITY.
  3186 
  3187 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3188 rules are generated for each domain, INCOMPATIBILITY.
  3189 
  3190   foo_bar.reach       ~> foo.reach  bar.reach
  3191   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3192 
  3193 * Some lemmas generated by the domain package have been renamed for
  3194 consistency with the datatype package, INCOMPATIBILITY.
  3195 
  3196   foo.ind        ~> foo.induct
  3197   foo.finite_ind ~> foo.finite_induct
  3198   foo.coind      ~> foo.coinduct
  3199   foo.casedist   ~> foo.exhaust
  3200   foo.exhaust    ~> foo.nchotomy
  3201 
  3202 * For consistency with other definition packages, the fixrec package
  3203 now generates qualified theorem names, INCOMPATIBILITY.
  3204 
  3205   foo_simps  ~> foo.simps
  3206   foo_unfold ~> foo.unfold
  3207   foo_induct ~> foo.induct
  3208 
  3209 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3210 method and internal fixrec proofs now use the default simpset instead.
  3211 INCOMPATIBILITY.
  3212 
  3213 * The "contlub" predicate has been removed.  Proof scripts should use
  3214 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3215 
  3216 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3217 
  3218 * The constants cpair, cfst, and csnd have been removed in favor of
  3219 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3220 
  3221 
  3222 *** ML ***
  3223 
  3224 * Antiquotations for basic formal entities:
  3225 
  3226     @{class NAME}         -- type class
  3227     @{class_syntax NAME}  -- syntax representation of the above
  3228 
  3229     @{type_name NAME}     -- logical type
  3230     @{type_abbrev NAME}   -- type abbreviation
  3231     @{nonterminal NAME}   -- type of concrete syntactic category
  3232     @{type_syntax NAME}   -- syntax representation of any of the above
  3233 
  3234     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3235     @{const_abbrev NAME}  -- abbreviated constant
  3236     @{const_syntax NAME}  -- syntax representation of any of the above
  3237 
  3238 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3239 syntax constant (cf. 'syntax' command).
  3240 
  3241 * Antiquotation @{make_string} inlines a function to print arbitrary
  3242 values similar to the ML toplevel.  The result is compiler dependent
  3243 and may fall back on "?" in certain situations.
  3244 
  3245 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3246 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3247 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3248 not work with the asynchronous Isar document model.
  3249 
  3250 * Configuration options now admit dynamic default values, depending on
  3251 the context or even global references.
  3252 
  3253 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3254 uses an efficient external library if available (for Poly/ML).
  3255 
  3256 * Renamed some important ML structures, while keeping the old names
  3257 for some time as aliases within the structure Legacy:
  3258 
  3259   OuterKeyword  ~>  Keyword
  3260   OuterLex      ~>  Token
  3261   OuterParse    ~>  Parse
  3262   OuterSyntax   ~>  Outer_Syntax
  3263   PrintMode     ~>  Print_Mode
  3264   SpecParse     ~>  Parse_Spec
  3265   ThyInfo       ~>  Thy_Info
  3266   ThyLoad       ~>  Thy_Load
  3267   ThyOutput     ~>  Thy_Output
  3268   TypeInfer     ~>  Type_Infer
  3269 
  3270 Note that "open Legacy" simplifies porting of sources, but forgetting
  3271 to remove it again will complicate porting again in the future.
  3272 
  3273 * Most operations that refer to a global context are named
  3274 accordingly, e.g. Simplifier.global_context or
  3275 ProofContext.init_global.  There are some situations where a global
  3276 context actually works, but under normal circumstances one needs to
  3277 pass the proper local context through the code!
  3278 
  3279 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3280 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3281 Pretty.pp argument to merge, which is absent in the standard
  3282 Theory_Data version.
  3283 
  3284 * Sorts.certify_sort and derived "cert" operations for types and terms
  3285 no longer minimize sorts.  Thus certification at the boundary of the
  3286 inference kernel becomes invariant under addition of class relations,
  3287 which is an important monotonicity principle.  Sorts are now minimized
  3288 in the syntax layer only, at the boundary between the end-user and the
  3289 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3290 explicitly in rare situations.
  3291 
  3292 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3293 emphasize that this is in no way a standard operation.
  3294 INCOMPATIBILITY.
  3295 
  3296 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3297 usual for resolution.  Rare INCOMPATIBILITY.
  3298 
  3299 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3300 to emphasize that these only work in a global situation (which is
  3301 quite rare).
  3302 
  3303 * Curried take and drop in library.ML; negative length is interpreted
  3304 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3305 
  3306 * Proof terms: type substitutions on proof constants now use canonical
  3307 order of type variables.  INCOMPATIBILITY for tools working with proof
  3308 terms.
  3309 
  3310 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3311 may no longer carry premises.  User-level specifications are
  3312 transformed accordingly by Thm.add_axiom/add_def.
  3313 
  3314 
  3315 *** System ***
  3316 
  3317 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3318 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3319 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3320 
  3321 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3322 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3323 respectively.
  3324 
  3325 * Old lib/scripts/polyml-platform is superseded by the
  3326 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3327 variant, even on a 64 bit machine.  The following example setting
  3328 prefers 64 bit if available:
  3329 
  3330   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3331 
  3332 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3333 Isabelle/Scala layer for advanced prover interaction and integration.
  3334 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3335 component.
  3336 
  3337 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3338 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3339 similar to the default assignment of the document preparation system
  3340 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3341 provides some operations for direct access to the font without asking
  3342 the user for manual installation.
  3343 
  3344 
  3345 
  3346 New in Isabelle2009-1 (December 2009)
  3347 -------------------------------------
  3348 
  3349 *** General ***
  3350 
  3351 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3352 one backslash should be used, even in ML sources.
  3353 
  3354 
  3355 *** Pure ***
  3356 
  3357 * Locale interpretation propagates mixins along the locale hierarchy.
  3358 The currently only available mixins are the equations used to map
  3359 local definitions to terms of the target domain of an interpretation.
  3360 
  3361 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3362 loc" to print all interpretations of locale "loc" in the theory.
  3363 Interpretations in proofs are not shown.
  3364 
  3365 * Thoroughly revised locales tutorial.  New section on conditional
  3366 interpretation.
  3367 
  3368 * On instantiation of classes, remaining undefined class parameters
  3369 are formally declared.  INCOMPATIBILITY.
  3370 
  3371 
  3372 *** Document preparation ***
  3373 
  3374 * New generalized style concept for printing terms: @{foo (style) ...}
  3375 instead of @{foo_style style ...}  (old form is still retained for
  3376 backward compatibility).  Styles can be also applied for
  3377 antiquotations prop, term_type and typeof.
  3378 
  3379 
  3380 *** HOL ***
  3381 
  3382 * New proof method "smt" for a combination of first-order logic with
  3383 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3384 fixed-size bitvectors; there is also basic support for higher-order
  3385 features (esp. lambda abstractions).  It is an incomplete decision
  3386 procedure based on external SMT solvers using the oracle mechanism;
  3387 for the SMT solver Z3, this method is proof-producing.  Certificates
  3388 are provided to avoid calling the external solvers solely for
  3389 re-checking proofs.  Due to a remote SMT service there is no need for
  3390 installing SMT solvers locally.  See src/HOL/SMT.
  3391 
  3392 * New commands to load and prove verification conditions generated by
  3393 the Boogie program verifier or derived systems (e.g. the Verifying C
  3394 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3395 
  3396 * New counterexample generator tool 'nitpick' based on the Kodkod
  3397 relational model finder.  See src/HOL/Tools/Nitpick and
  3398 src/HOL/Nitpick_Examples.
  3399 
  3400 * New commands 'code_pred' and 'values' to invoke the predicate
  3401 compiler and to enumerate values of inductive predicates.
  3402 
  3403 * A tabled implementation of the reflexive transitive closure.
  3404 
  3405 * New implementation of quickcheck uses generic code generator;
  3406 default generators are provided for all suitable HOL types, records
  3407 and datatypes.  Old quickcheck can be re-activated importing theory
  3408 Library/SML_Quickcheck.
  3409 
  3410 * New testing tool Mirabelle for automated proof tools.  Applies
  3411 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3412 every proof step in a theory.  To be used in batch mode via the
  3413 "mirabelle" utility.
  3414 
  3415 * New proof method "sos" (sum of squares) for nonlinear real
  3416 arithmetic (originally due to John Harison). It requires theory
  3417 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3418 works well in practice on quantifier-free real arithmetic with +, -,
  3419 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3420 inequalities between polynomials.  It makes use of external
  3421 semidefinite programming solvers.  Method "sos" generates a
  3422 certificate that can be pasted into the proof thus avoiding the need
  3423 to call an external tool every time the proof is checked.  See
  3424 src/HOL/Library/Sum_Of_Squares.
  3425 
  3426 * New method "linarith" invokes existing linear arithmetic decision
  3427 procedure only.
  3428 
  3429 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3430 
  3431 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3432 Causes full type information to be output to the ATPs.  This slows
  3433 ATPs down considerably but eliminates a source of unsound "proofs"
  3434 that fail later.
  3435 
  3436 * New method "metisFT": A version of metis that uses full type
  3437 information in order to avoid failures of proof reconstruction.
  3438 
  3439 * New evaluator "approximate" approximates an real valued term using
  3440 the same method as the approximation method.
  3441 
  3442 * Method "approximate" now supports arithmetic expressions as
  3443 boundaries of intervals and implements interval splitting and Taylor
  3444 series expansion.
  3445 
  3446 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3447 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3448 
  3449 * New theory SupInf of the supremum and infimum operators for sets of
  3450 reals.
  3451 
  3452 * New theory Probability, which contains a development of measure
  3453 theory, eventually leading to Lebesgue integration and probability.
  3454 
  3455 * Extended Multivariate Analysis to include derivation and Brouwer's
  3456 fixpoint theorem.
  3457 
  3458 * Reorganization of number theory, INCOMPATIBILITY:
  3459   - new number theory development for nat and int, in theories Divides
  3460     and GCD as well as in new session Number_Theory
  3461   - some constants and facts now suffixed with _nat and _int
  3462     accordingly
  3463   - former session NumberTheory now named Old_Number_Theory, including
  3464     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3465   - moved theory Pocklington from src/HOL/Library to
  3466     src/HOL/Old_Number_Theory
  3467 
  3468 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3469 lcm of finite and infinite sets. It is shown that they form a complete
  3470 lattice.
  3471 
  3472 * Class semiring_div requires superclass no_zero_divisors and proof of
  3473 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3474 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3475 generalized to class semiring_div, subsuming former theorems
  3476 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3477 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3478 INCOMPATIBILITY.
  3479 
  3480 * Refinements to lattice classes and sets:
  3481   - less default intro/elim rules in locale variant, more default
  3482     intro/elim rules in class variant: more uniformity
  3483   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3484     le_inf_iff
  3485   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3486     sup_aci)
  3487   - renamed ACI to inf_sup_aci
  3488   - new class "boolean_algebra"
  3489   - class "complete_lattice" moved to separate theory
  3490     "Complete_Lattice"; corresponding constants (and abbreviations)
  3491     renamed and with authentic syntax:
  3492     Set.Inf ~>    Complete_Lattice.Inf
  3493     Set.Sup ~>    Complete_Lattice.Sup
  3494     Set.INFI ~>   Complete_Lattice.INFI
  3495     Set.SUPR ~>   Complete_Lattice.SUPR
  3496     Set.Inter ~>  Complete_Lattice.Inter
  3497     Set.Union ~>  Complete_Lattice.Union
  3498     Set.INTER ~>  Complete_Lattice.INTER
  3499     Set.UNION ~>  Complete_Lattice.UNION
  3500   - authentic syntax for
  3501     Set.Pow
  3502     Set.image
  3503   - mere abbreviations:
  3504     Set.empty               (for bot)
  3505     Set.UNIV                (for top)
  3506     Set.inter               (for inf, formerly Set.Int)
  3507     Set.union               (for sup, formerly Set.Un)
  3508     Complete_Lattice.Inter  (for Inf)
  3509     Complete_Lattice.Union  (for Sup)
  3510     Complete_Lattice.INTER  (for INFI)
  3511     Complete_Lattice.UNION  (for SUPR)
  3512   - object-logic definitions as far as appropriate
  3513 
  3514 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3515 Un_subset_iff are explicitly deleted as default simp rules; then also
  3516 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3517 deleted to achieve the desired effect.
  3518 
  3519 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3520 rules by default any longer; the same applies to min_max.inf_absorb1
  3521 etc.  INCOMPATIBILITY.
  3522 
  3523 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3524 pred_set_conv by default.  INCOMPATIBILITY.
  3525 
  3526 * Power operations on relations and functions are now one dedicated
  3527 constant "compow" with infix syntax "^^".  Power operation on
  3528 multiplicative monoids retains syntax "^" and is now defined generic
  3529 in class power.  INCOMPATIBILITY.
  3530 
  3531 * Relation composition "R O S" now has a more standard argument order:
  3532 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3533 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3534 break, since the O_assoc rule was not rewritten like this.  Fix using
  3535 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3536 
  3537 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3538 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3539 INCOMPATIBILITY.
  3540 
  3541 * Most rules produced by inductive and datatype package have mandatory
  3542 prefixes.  INCOMPATIBILITY.
  3543 
  3544 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3545 the attribute of the same name.  Each of the theorems in the list
  3546 DERIV_intros assumes composition with an additional function and
  3547 matches a variable to the derivative, which has to be solved by the
  3548 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3549 of most elementary terms.  Former Maclauren.DERIV_tac and
  3550 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3551 INCOMPATIBILITY.
  3552 
  3553 * Code generator attributes follow the usual underscore convention:
  3554     code_unfold     replaces    code unfold
  3555     code_post       replaces    code post
  3556     etc.
  3557   INCOMPATIBILITY.
  3558 
  3559 * Renamed methods:
  3560     sizechange -> size_change
  3561     induct_scheme -> induction_schema
  3562   INCOMPATIBILITY.
  3563 
  3564 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3565 INCOMPATIBILITY, use "undefined" directly.
  3566 
  3567 * Renamed theorems:
  3568     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3569     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3570     Suc_plus1 -> Suc_eq_plus1
  3571     *anti_sym -> *antisym*
  3572     vector_less_eq_def -> vector_le_def
  3573   INCOMPATIBILITY.
  3574 
  3575 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3576 INCOMPATIBILITY.
  3577 
  3578 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3579 INCOMPATIBILITY.
  3580 
  3581 
  3582 *** HOLCF ***
  3583 
  3584 * Theory Representable defines a class "rep" of domains that are
  3585 representable (via an ep-pair) in the universal domain type "udom".
  3586 Instances are provided for all type constructors defined in HOLCF.
  3587 
  3588 * The 'new_domain' command is a purely definitional version of the
  3589 domain package, for representable domains.  Syntax is identical to the
  3590 old domain package.  The 'new_domain' package also supports indirect
  3591 recursion using previously-defined type constructors.  See
  3592 src/HOLCF/ex/New_Domain.thy for examples.
  3593 
  3594 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3595 on the left-hand side of an equation, and then performs
  3596 simplification.  Rewriting is done using rules declared with the
  3597 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3598 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3599 
  3600 * The pattern-match compiler in 'fixrec' can now handle constructors
  3601 with HOL function types.  Pattern-match combinators for the Pair
  3602 constructor are pre-configured.
  3603 
  3604 * The 'fixrec' package now produces better fixed-point induction rules
  3605 for mutually-recursive definitions:  Induction rules have conclusions
  3606 of the form "P foo bar" instead of "P <foo, bar>".
  3607 
  3608 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3609 been renamed to "below".  The name "below" now replaces "less" in many
  3610 theorem names.  (Legacy theorem names using "less" are still supported
  3611 as well.)
  3612 
  3613 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3614 patterns can be used to generate strictness rules, or to make
  3615 functions more strict (much like the bang-patterns supported by the
  3616 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3617 examples.
  3618 
  3619 
  3620 *** ML ***
  3621 
  3622 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3623 errors and run-time exceptions, including detailed source positions.
  3624 
  3625 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3626 identified entries, with some additional information such as source
  3627 position, logical grouping etc.
  3628 
  3629 * Theory and context data is now introduced by the simplified and
  3630 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3631 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3632 copy operation) is still available for some time.
  3633 
  3634 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3635 provides a high-level programming interface to synchronized state
  3636 variables with atomic update.  This works via pure function
  3637 application within a critical section -- its runtime should be as
  3638 short as possible; beware of deadlocks if critical code is nested,
  3639 either directly or indirectly via other synchronized variables!
  3640 
  3641 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3642 wraps raw ML references, explicitly indicating their non-thread-safe
  3643 behaviour.  The Isar toplevel keeps this structure open, to
  3644 accommodate Proof General as well as quick and dirty interactive
  3645 experiments with references.
  3646 
  3647 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3648 parallel tactical reasoning.
  3649 
  3650 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3651 are similar to SUBPROOF, but are slightly more flexible: only the
  3652 specified parts of the subgoal are imported into the context, and the
  3653 body tactic may introduce new subgoals and schematic variables.
  3654 
  3655 * Old tactical METAHYPS, which does not observe the proof context, has
  3656 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3657 or Subgoal.FOCUS etc.
  3658 
  3659 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3660 functors have their own ML name space there is no point to mark them
  3661 separately.)  Minor INCOMPATIBILITY.
  3662 
  3663 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3664 
  3665 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3666 INCOMPATIBILITY.
  3667 
  3668 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3669 mode, which avoids critical setmp.
  3670 
  3671 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3672 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3673 Attrib/Method.setup introduced in Isabelle2009.
  3674 
  3675 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3676 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3677 last resort.  INCOMPATIBILITY.
  3678 
  3679 * Display.pretty_thm now requires a proper context (cf. former
  3680 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3681 or even Display.pretty_thm_without_context as last resort.
  3682 INCOMPATIBILITY.
  3683 
  3684 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3685 Syntax.pretty_typ/term directly, preferably with proper context
  3686 instead of global theory.
  3687 
  3688 
  3689 *** System ***
  3690 
  3691 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3692 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3693 usedir option -q.
  3694 
  3695 * Support for additional "Isabelle components" via etc/components, see
  3696 also the system manual.
  3697 
  3698 * The isabelle makeall tool now operates on all components with
  3699 IsaMakefile, not just hardwired "logics".
  3700 
  3701 * Removed "compress" option from isabelle-process and isabelle usedir;
  3702 this is always enabled.
  3703 
  3704 * Discontinued support for Poly/ML 4.x versions.
  3705 
  3706 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3707 on a given logic image.  This requires the lighttpd webserver and is
  3708 currently supported on Linux only.
  3709 
  3710 
  3711 
  3712 New in Isabelle2009 (April 2009)
  3713 --------------------------------
  3714 
  3715 *** General ***
  3716 
  3717 * Simplified main Isabelle executables, with less surprises on
  3718 case-insensitive file-systems (such as Mac OS).
  3719 
  3720   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3721     "isatool."
  3722 
  3723   - The former "isabelle" alias for "isabelle-process" has been
  3724     removed (should rarely occur to regular users).
  3725 
  3726   - The former "isabelle-interface" and its alias "Isabelle" have been
  3727     removed (interfaces are now regular Isabelle tools).
  3728 
  3729 Within scripts and make files, the Isabelle environment variables
  3730 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3731 respectively.  (The latter are still available as legacy feature.)
  3732 
  3733 The old isabelle-interface wrapper could react in confusing ways if
  3734 the interface was uninstalled or changed otherwise.  Individual
  3735 interface tool configuration is now more explicit, see also the
  3736 Isabelle system manual.  In particular, Proof General is now available
  3737 via "isabelle emacs".
  3738 
  3739 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3740 purge installed copies of Isabelle executables and re-run "isabelle
  3741 install -p ...", or use symlinks.
  3742 
  3743 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3744 old ~/isabelle, which was slightly non-standard and apt to cause
  3745 surprises on case-insensitive file-systems (such as Mac OS).
  3746 
  3747 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3748 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3749 care is required when using older releases of Isabelle.  Note that
  3750 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3751 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3752 
  3753 * Proofs of fully specified statements are run in parallel on
  3754 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3755 a regular 4-core machine, if the initial heap space is made reasonably
  3756 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3757 
  3758 * The main reference manuals ("isar-ref", "implementation", and
  3759 "system") have been updated and extended.  Formally checked references
  3760 as hyperlinks are now available uniformly.
  3761 
  3762 
  3763 *** Pure ***
  3764 
  3765 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3766 respects.  The most important changes are listed below.  See the
  3767 Tutorial on Locales ("locales" manual) for details.
  3768 
  3769 - In locale expressions, instantiation replaces renaming.  Parameters
  3770 must be declared in a for clause.  To aid compatibility with previous
  3771 parameter inheritance, in locale declarations, parameters that are not
  3772 'touched' (instantiation position "_" or omitted) are implicitly added
  3773 with their syntax at the beginning of the for clause.
  3774 
  3775 - Syntax from abbreviations and definitions in locales is available in
  3776 locale expressions and context elements.  The latter is particularly
  3777 useful in locale declarations.
  3778 
  3779 - More flexible mechanisms to qualify names generated by locale
  3780 expressions.  Qualifiers (prefixes) may be specified in locale
  3781 expressions, and can be marked as mandatory (syntax: "name!:") or
  3782 optional (syntax "name?:").  The default depends for plain "name:"
  3783 depends on the situation where a locale expression is used: in
  3784 commands 'locale' and 'sublocale' prefixes are optional, in
  3785 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3786 implicit qualifiers derived from the parameter names of a locale are
  3787 no longer generated.
  3788 
  3789 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3790 instantiation clause in "interpretation" and "interpret" (square
  3791 brackets) is no longer available.  Use locale expressions.
  3792 
  3793 - When converting proof scripts, mandatory qualifiers in
  3794 'interpretation' and 'interpret' should be retained by default, even
  3795 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3796 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3797 in locale expressions range over a single locale instance only.
  3798 
  3799 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3800 In existing theorem specifications replace the includes element by the
  3801 respective context elements of the included locale, omitting those
  3802 that are already present in the theorem specification.  Multiple
  3803 assume elements of a locale should be replaced by a single one
  3804 involving the locale predicate.  In the proof body, declarations (most
  3805 notably theorems) may be regained by interpreting the respective
  3806 locales in the proof context as required (command "interpret").
  3807 
  3808 If using "includes" in replacement of a target solely because the
  3809 parameter types in the theorem are not as general as in the target,
  3810 consider declaring a new locale with additional type constraints on
  3811 the parameters (context element "constrains").
  3812 
  3813 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3814 
  3815 - Locale interpretation commands no longer attempt to simplify goal.
  3816 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3817 methods intro_locales and unfold_locales to clarify.
  3818 
  3819 - Locale interpretation commands no longer accept interpretation
  3820 attributes.  INCOMPATIBILITY.
  3821 
  3822 * Class declaration: so-called "base sort" must not be given in import
  3823 list any longer, but is inferred from the specification.  Particularly
  3824 in HOL, write
  3825 
  3826     class foo = ...
  3827 
  3828 instead of
  3829 
  3830     class foo = type + ...
  3831 
  3832 * Class target: global versions of theorems stemming do not carry a
  3833 parameter prefix any longer.  INCOMPATIBILITY.
  3834 
  3835 * Class 'instance' command no longer accepts attached definitions.
  3836 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3837 
  3838 * Recovered hiding of consts, which was accidentally broken in
  3839 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3840 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3841 
  3842 * Slightly more coherent Pure syntax, with updated documentation in
  3843 isar-ref manual.  Removed locales meta_term_syntax and
  3844 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3845 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3846 directly in regular applications.
  3847 
  3848 * There is a new syntactic category "float_const" for signed decimal
  3849 fractions (e.g. 123.45 or -123.45).
  3850 
  3851 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3852 interface with 'setup' command instead.
  3853 
  3854 * Command 'local_setup' is similar to 'setup', but operates on a local
  3855 theory context.
  3856 
  3857 * The 'axiomatization' command now only works within a global theory
  3858 context.  INCOMPATIBILITY.
  3859 
  3860 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3861 hypotheses.  Sorts required in the course of reasoning need to be
  3862 covered by the constraints in the initial statement, completed by the
  3863 type instance information of the background theory.  Non-trivial sort
  3864 hypotheses, which rarely occur in practice, may be specified via
  3865 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3866 
  3867   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3868 
  3869 The result contains an implicit sort hypotheses as before --
  3870 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3871 normalization.
  3872 
  3873 * Generalized Isar history, with support for linear undo, direct state
  3874 addressing etc.
  3875 
  3876 * Changed defaults for unify configuration options:
  3877 
  3878   unify_trace_bound = 50 (formerly 25)
  3879   unify_search_bound = 60 (formerly 30)
  3880 
  3881 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3882 
  3883   a) On theory merge, the last set of code equations for a particular
  3884      constant is taken (in accordance with the policy applied by other
  3885      parts of the code generator framework).
  3886 
  3887   b) Code equations stemming from explicit declarations (e.g. code
  3888      attribute) gain priority over default code equations stemming
  3889      from definition, primrec, fun etc.
  3890 
  3891 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3892 
  3893 * Unified theorem tables for both code generators.  Thus [code
  3894 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3895 
  3896 * Command 'find_consts' searches for constants based on type and name
  3897 patterns, e.g.
  3898 
  3899     find_consts "_ => bool"
  3900 
  3901 By default, matching is against subtypes, but it may be restricted to
  3902 the whole type.  Searching by name is possible.  Multiple queries are
  3903 conjunctive and queries may be negated by prefixing them with a
  3904 hyphen:
  3905 
  3906     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3907 
  3908 * New 'find_theorems' criterion "solves" matches theorems that
  3909 directly solve the current goal (modulo higher-order unification).
  3910 
  3911 * Auto solve feature for main theorem statements: whenever a new goal
  3912 is stated, "find_theorems solves" is called; any theorems that could
  3913 solve the lemma directly are listed as part of the goal state.
  3914 Cf. associated options in Proof General Isabelle settings menu,
  3915 enabled by default, with reasonable timeout for pathological cases of
  3916 higher-order unification.
  3917 
  3918 
  3919 *** Document preparation ***
  3920 
  3921 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3922 demanding keyword 'by' and supporting the full method expression
  3923 syntax just like the Isar command 'by'.
  3924 
  3925 
  3926 *** HOL ***
  3927 
  3928 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3929 points Main and Complex_Main remain as before.
  3930 
  3931 * Logic image HOL-Plain provides a minimal HOL with the most important
  3932 tools available (inductive, datatype, primrec, ...).  This facilitates
  3933 experimentation and tool development.  Note that user applications
  3934 (and library theories) should never refer to anything below theory
  3935 Main, as before.
  3936 
  3937 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3938 experimentation due to shorter build times.
  3939 
  3940 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3941 were previously part of former HOL-Complex.  Entry point Hyperreal
  3942 remains valid, but theories formerly using Complex_Main should now use
  3943 new entry point Hypercomplex.
  3944 
  3945 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3946 Posix processes.  Avoids potentially expensive forking of the ML
  3947 process.  New thread-based implementation also works on non-Unix
  3948 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3949 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3950 commands are covered in the isar-ref manual.
  3951 
  3952 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3953 sledgehammer without local ATP installation (Vampire etc.). Other
  3954 provers may be included via suitable ML wrappers, see also
  3955 src/HOL/ATP_Linkup.thy.
  3956 
  3957 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3958 menu.
  3959 
  3960 * The metis method no longer fails because the theorem is too trivial
  3961 (contains the empty clause).
  3962 
  3963 * The metis method now fails in the usual manner, rather than raising
  3964 an exception, if it determines that it cannot prove the theorem.
  3965 
  3966 * Method "coherent" implements a prover for coherent logic (see also
  3967 src/Tools/coherent.ML).
  3968 
  3969 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3970 "undefined" is the right choice to replace "arbitrary", though
  3971 logically there is no difference.  INCOMPATIBILITY.
  3972 
  3973 * Command "value" now integrates different evaluation mechanisms.  The
  3974 result of the first successful evaluation mechanism is printed.  In
  3975 square brackets a particular named evaluation mechanisms may be
  3976 specified (currently, [SML], [code] or [nbe]).  See further
  3977 src/HOL/ex/Eval_Examples.thy.
  3978 
  3979 * Normalization by evaluation now allows non-leftlinear equations.
  3980 Declare with attribute [code nbe].
  3981 
  3982 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3983 as the structured Isar versions "cases" and "induct", cf. the
  3984 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3985 are now presented as a list of individual projections
  3986 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3987 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3988 rare situations a different rule is selected --- notably nested tuple
  3989 elimination instead of former prod.exhaust: use explicit (case_tac t
  3990 rule: prod.exhaust) here.
  3991 
  3992 * Attributes "cases", "induct", "coinduct" support "del" option.
  3993 
  3994 * Removed fact "case_split_thm", which duplicates "case_split".
  3995 
  3996 * The option datatype has been moved to a new theory Option.  Renamed
  3997 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3998 
  3999 * New predicate "strict_mono" classifies strict functions on partial
  4000 orders.  With strict functions on linear orders, reasoning about
  4001 (in)equalities is facilitated by theorems "strict_mono_eq",
  4002 "strict_mono_less_eq" and "strict_mono_less".
  4003 
  4004 * Some set operations are now proper qualified constants with
  4005 authentic syntax.  INCOMPATIBILITY:
  4006 
  4007     op Int ~>   Set.Int
  4008     op Un ~>    Set.Un
  4009     INTER ~>    Set.INTER
  4010     UNION ~>    Set.UNION
  4011     Inter ~>    Set.Inter
  4012     Union ~>    Set.Union
  4013     {} ~>       Set.empty
  4014     UNIV ~>     Set.UNIV
  4015 
  4016 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4017 theory Set.
  4018 
  4019 * Auxiliary class "itself" has disappeared -- classes without any
  4020 parameter are treated as expected by the 'class' command.
  4021 
  4022 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4023 
  4024 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4025 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4026 
  4027 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4028 method "approximation".  It proves formulas on real values by using
  4029 interval arithmetic.  In the formulas are also the transcendental
  4030 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4031 allowed. For examples see
  4032 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4033 
  4034 * Theory "Reflection" now resides in HOL/Library.
  4035 
  4036 * Entry point to Word library now simply named "Word".
  4037 INCOMPATIBILITY.
  4038 
  4039 * Made source layout more coherent with logical distribution
  4040 structure:
  4041 
  4042     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4043     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4044     src/HOL/Library/GCD.thy ~> src/HOL/
  4045     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4046     src/HOL/Library/Parity.thy ~> src/HOL/
  4047     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4048     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4049     src/HOL/Real/Lubs.thy ~> src/HOL/
  4050     src/HOL/Real/PReal.thy ~> src/HOL/
  4051     src/HOL/Real/Rational.thy ~> src/HOL/
  4052     src/HOL/Real/RComplete.thy ~> src/HOL/
  4053     src/HOL/Real/RealDef.thy ~> src/HOL/
  4054     src/HOL/Real/RealPow.thy ~> src/HOL/
  4055     src/HOL/Real/Real.thy ~> src/HOL/
  4056     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4057     src/HOL/Complex/Complex.thy ~> src/HOL/
  4058     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4059     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4060     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4061     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4062     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4063     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4064     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4065     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4066     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4067     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4068     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4069     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4070     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4071     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4072     src/HOL/Real/Float ~> src/HOL/Library/
  4073     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4074     src/HOL/Real/RealVector.thy ~> src/HOL/
  4075 
  4076     src/HOL/arith_data.ML ~> src/HOL/Tools
  4077     src/HOL/hologic.ML ~> src/HOL/Tools
  4078     src/HOL/simpdata.ML ~> src/HOL/Tools
  4079     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4080     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4081     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4082     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4083     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4084     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4085     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4086 
  4087     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4088     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4089     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4090     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4091     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4092     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4093 
  4094 * If methods "eval" and "evaluation" encounter a structured proof
  4095 state with !!/==>, only the conclusion is evaluated to True (if
  4096 possible), avoiding strange error messages.
  4097 
  4098 * Method "sizechange" automates termination proofs using (a
  4099 modification of) the size-change principle.  Requires SAT solver.  See
  4100 src/HOL/ex/Termination.thy for examples.
  4101 
  4102 * Simplifier: simproc for let expressions now unfolds if bound
  4103 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4104 
  4105 * Method "arith": Linear arithmetic now ignores all inequalities when
  4106 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4107 
  4108 * New attribute "arith" for facts that should always be used
  4109 automatically by arithmetic. It is intended to be used locally in
  4110 proofs, e.g.
  4111 
  4112   assumes [arith]: "x > 0"
  4113 
  4114 Global usage is discouraged because of possible performance impact.
  4115 
  4116 * New classes "top" and "bot" with corresponding operations "top" and
  4117 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4118 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4119 
  4120 * Changed definition lemma "less_fun_def" in order to provide an
  4121 instance for preorders on functions; use lemma "less_le" instead.
  4122 INCOMPATIBILITY.
  4123 
  4124 * Theory Orderings: class "wellorder" moved here, with explicit
  4125 induction rule "less_induct" as assumption.  For instantiation of
  4126 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4127 INCOMPATIBILITY.
  4128 
  4129 * Theory Orderings: added class "preorder" as superclass of "order".
  4130 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4131 etc. slightly changed.  Some theorems named order_class.* now named
  4132 preorder_class.*.
  4133 
  4134 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4135 "diag" to "Id_on".
  4136 
  4137 * Theory Finite_Set: added a new fold combinator of type
  4138 
  4139   ('a => 'b => 'b) => 'b => 'a set => 'b
  4140 
  4141 Occasionally this is more convenient than the old fold combinator
  4142 which is now defined in terms of the new one and renamed to
  4143 fold_image.
  4144 
  4145 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4146 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4147 extended with further lemmas!).  At the moment both still exist but
  4148 the former will disappear at some point.
  4149 
  4150 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4151 class recpower.  Type-specific simp rules for various recpower types
  4152 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4153 
  4154 rat_power_0    -> power_0
  4155 rat_power_Suc  -> power_Suc
  4156 realpow_0      -> power_0
  4157 realpow_Suc    -> power_Suc
  4158 complexpow_0   -> power_0
  4159 complexpow_Suc -> power_Suc
  4160 power_poly_0   -> power_0
  4161 power_poly_Suc -> power_Suc
  4162 
  4163 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4164 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4165 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4166 of lemmas from Divides has been generalized from nat to class
  4167 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4168 renames resulting from duplicate elimination:
  4169 
  4170     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4171     zero_dvd_iff ~>         dvd_0_left_iff
  4172     dvd_0 ~>                dvd_0_right
  4173     DIVISION_BY_ZERO_DIV ~> div_by_0
  4174     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4175     mult_div ~>             div_mult_self2_is_id
  4176     mult_mod ~>             mod_mult_self2_is_0
  4177 
  4178 * Theory IntDiv: removed many lemmas that are instances of class-based
  4179 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4180 rename old lemmas as follows:
  4181 
  4182 dvd_diff               -> nat_dvd_diff
  4183 dvd_zminus_iff         -> dvd_minus_iff
  4184 mod_add1_eq            -> mod_add_eq
  4185 mod_mult1_eq           -> mod_mult_right_eq
  4186 mod_mult1_eq'          -> mod_mult_left_eq
  4187 mod_mult_distrib_mod   -> mod_mult_eq
  4188 nat_mod_add_left_eq    -> mod_add_left_eq
  4189 nat_mod_add_right_eq   -> mod_add_right_eq
  4190 nat_mod_div_trivial    -> mod_div_trivial
  4191 nat_mod_mod_trivial    -> mod_mod_trivial
  4192 zdiv_zadd_self1        -> div_add_self1
  4193 zdiv_zadd_self2        -> div_add_self2
  4194 zdiv_zmult_self1       -> div_mult_self2_is_id
  4195 zdiv_zmult_self2       -> div_mult_self1_is_id
  4196 zdvd_triv_left         -> dvd_triv_left
  4197 zdvd_triv_right        -> dvd_triv_right
  4198 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4199 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4200 zmod_zadd_left_eq      -> mod_add_left_eq
  4201 zmod_zadd_right_eq     -> mod_add_right_eq
  4202 zmod_zadd_self1        -> mod_add_self1
  4203 zmod_zadd_self2        -> mod_add_self2
  4204 zmod_zadd1_eq          -> mod_add_eq
  4205 zmod_zdiff1_eq         -> mod_diff_eq
  4206 zmod_zdvd_zmod         -> mod_mod_cancel
  4207 zmod_zmod_cancel       -> mod_mod_cancel
  4208 zmod_zmult_self1       -> mod_mult_self2_is_0
  4209 zmod_zmult_self2       -> mod_mult_self1_is_0
  4210 zmod_1                 -> mod_by_1
  4211 zdiv_1                 -> div_by_1
  4212 zdvd_abs1              -> abs_dvd_iff
  4213 zdvd_abs2              -> dvd_abs_iff
  4214 zdvd_refl              -> dvd_refl
  4215 zdvd_trans             -> dvd_trans
  4216 zdvd_zadd              -> dvd_add
  4217 zdvd_zdiff             -> dvd_diff
  4218 zdvd_zminus_iff        -> dvd_minus_iff
  4219 zdvd_zminus2_iff       -> minus_dvd_iff
  4220 zdvd_zmultD            -> dvd_mult_right
  4221 zdvd_zmultD2           -> dvd_mult_left
  4222 zdvd_zmult_mono        -> mult_dvd_mono
  4223 zdvd_0_right           -> dvd_0_right
  4224 zdvd_0_left            -> dvd_0_left_iff
  4225 zdvd_1_left            -> one_dvd
  4226 zminus_dvd_iff         -> minus_dvd_iff
  4227 
  4228 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4229 
  4230 * The real numbers offer decimal input syntax: 12.34 is translated
  4231 into 1234/10^2. This translation is not reversed upon output.
  4232 
  4233 * Theory Library/Polynomial defines an abstract type 'a poly of
  4234 univariate polynomials with coefficients of type 'a.  In addition to
  4235 the standard ring operations, it also supports div and mod.  Code
  4236 generation is also supported, using list-style constructors.
  4237 
  4238 * Theory Library/Inner_Product defines a class of real_inner for real
  4239 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4240 => real.  Class real_inner is a subclass of real_normed_vector from
  4241 theory RealVector.
  4242 
  4243 * Theory Library/Product_Vector provides instances for the product
  4244 type 'a * 'b of several classes from RealVector and Inner_Product.
  4245 Definitions of addition, subtraction, scalar multiplication, norms,
  4246 and inner products are included.
  4247 
  4248 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4249 addition to the field operations, numerals and case syntax are also
  4250 supported.
  4251 
  4252 * Theory Library/Diagonalize provides constructive version of Cantor's
  4253 first diagonalization argument.
  4254 
  4255 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4256 zlcm (for int); carried together from various gcd/lcm developements in
  4257 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4258 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4259 may recover tupled syntax as follows:
  4260 
  4261     hide (open) const gcd
  4262     abbreviation gcd where
  4263       "gcd == (%(a, b). GCD.gcd a b)"
  4264     notation (output)
  4265       GCD.gcd ("gcd '(_, _')")
  4266 
  4267 The same works for lcm, zgcd, zlcm.
  4268 
  4269 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4270 instantiations for algebraic structures.  Removed some duplicate
  4271 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4272 
  4273 * ML antiquotation @{code} takes a constant as argument and generates
  4274 corresponding code in background and inserts name of the corresponding
  4275 resulting ML value/function/datatype constructor binding in place.
  4276 All occurrences of @{code} with a single ML block are generated
  4277 simultaneously.  Provides a generic and safe interface for
  4278 instrumentalizing code generation.  See
  4279 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4280 In future you ought to refrain from ad-hoc compiling generated SML
  4281 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4282 cannot refer to constants for which user-defined serializations are
  4283 set.  Refer to the corresponding ML counterpart directly in that
  4284 cases.
  4285 
  4286 * Command 'rep_datatype': instead of theorem names the command now
  4287 takes a list of terms denoting the constructors of the type to be
  4288 represented as datatype.  The characteristic theorems have to be
  4289 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4290 have disappeared in favour of existing ones:
  4291 
  4292     unit_induct                 ~> unit.induct
  4293     prod_induct                 ~> prod.induct
  4294     sum_induct                  ~> sum.induct
  4295     Suc_Suc_eq                  ~> nat.inject
  4296     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4297 
  4298 
  4299 *** HOL-Algebra ***
  4300 
  4301 * New locales for orders and lattices where the equivalence relation
  4302 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4303 locales use a record structure with field eq for the equivalence.
  4304 
  4305 * New theory of factorial domains.
  4306 
  4307 * Units_l_inv and Units_r_inv are now simp rules by default.
  4308 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4309 and/or r_inv will now also require deletion of these lemmas.
  4310 
  4311 * Renamed the following theorems, INCOMPATIBILITY:
  4312 
  4313 UpperD ~> Upper_memD
  4314 LowerD ~> Lower_memD
  4315 least_carrier ~> least_closed
  4316 greatest_carrier ~> greatest_closed
  4317 greatest_Lower_above ~> greatest_Lower_below
  4318 one_zero ~> carrier_one_zero
  4319 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4320 
  4321 
  4322 *** HOL-Nominal ***
  4323 
  4324 * Nominal datatypes can now contain type-variables.
  4325 
  4326 * Commands 'nominal_inductive' and 'equivariance' work with local
  4327 theory targets.
  4328 
  4329 * Nominal primrec can now works with local theory targets and its
  4330 specification syntax now conforms to the general format as seen in
  4331 'inductive' etc.
  4332 
  4333 * Method "perm_simp" honours the standard simplifier attributes
  4334 (no_asm), (no_asm_use) etc.
  4335 
  4336 * The new predicate #* is defined like freshness, except that on the
  4337 left hand side can be a set or list of atoms.
  4338 
  4339 * Experimental command 'nominal_inductive2' derives strong induction
  4340 principles for inductive definitions.  In contrast to
  4341 'nominal_inductive', which can only deal with a fixed number of
  4342 binders, it can deal with arbitrary expressions standing for sets of
  4343 atoms to be avoided.  The only inductive definition we have at the
  4344 moment that needs this generalisation is the typing rule for Lets in
  4345 the algorithm W:
  4346 
  4347  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4348  -----------------------------------------------------------------
  4349          Gamma |- Let x be t1 in t2 : T2
  4350 
  4351 In this rule one wants to avoid all the binders that are introduced by
  4352 "close Gamma T1".  We are looking for other examples where this
  4353 feature might be useful.  Please let us know.
  4354 
  4355 
  4356 *** HOLCF ***
  4357 
  4358 * Reimplemented the simplification procedure for proving continuity
  4359 subgoals.  The new simproc is extensible; users can declare additional
  4360 continuity introduction rules with the attribute [cont2cont].
  4361 
  4362 * The continuity simproc now uses a different introduction rule for
  4363 solving continuity subgoals on terms with lambda abstractions.  In
  4364 some rare cases the new simproc may fail to solve subgoals that the
  4365 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4366 Potential INCOMPATIBILITY.
  4367 
  4368 * Command 'fixrec': specification syntax now conforms to the general
  4369 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4370 examples.  INCOMPATIBILITY.
  4371 
  4372 
  4373 *** ZF ***
  4374 
  4375 * Proof of Zorn's Lemma for partial orders.
  4376 
  4377 
  4378 *** ML ***
  4379 
  4380 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4381 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4382 depends on multithreading, so timouts will not work before Poly/ML
  4383 5.2.1!
  4384 
  4385 * High-level support for concurrent ML programming, see
  4386 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4387 particularly convenient to organize independent functional
  4388 computations.  The concept of "synchronized variables" provides a
  4389 higher-order interface for components with shared state, avoiding the
  4390 delicate details of mutexes and condition variables.  (Requires
  4391 Poly/ML 5.2.1 or later.)
  4392 
  4393 * ML bindings produced via Isar commands are stored within the Isar
  4394 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4395 become thread-safe and work with undo as expected (concerning
  4396 top-level bindings, not side-effects on global references).
  4397 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4398 compiler at runtime; really global bindings need to be given outside a
  4399 theory.  (Requires Poly/ML 5.2 or later.)
  4400 
  4401 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4402 context.  Top-level ML bindings are stored within the proof context in
  4403 a purely sequential fashion, disregarding the nested proof structure.
  4404 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4405 proof.  (Requires Poly/ML 5.2 or later.)
  4406 
  4407 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4408 and Method.setup, as well as Isar commands 'attribute_setup' and
  4409 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4410 existing code accordingly, or use plain 'setup' together with old
  4411 Method.add_method.
  4412 
  4413 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4414 to 'a -> thm, while results are always tagged with an authentic oracle
  4415 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4416 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4417 accordingly.  Note that extra performance may be gained by producing
  4418 the cterm carefully, avoiding slow Thm.cterm_of.
  4419 
  4420 * Simplified interface for defining document antiquotations via
  4421 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4422 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4423 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4424 examples.
  4425 
  4426 * More systematic treatment of long names, abstract name bindings, and
  4427 name space operations.  Basic operations on qualified names have been
  4428 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4429 Long_Name.append.  Old type bstring has been mostly replaced by
  4430 abstract type binding (see structure Binding), which supports precise
  4431 qualification by packages and local theory targets, as well as proper
  4432 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4433 bstring values into Binding.name, or better pass through abstract
  4434 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4435 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4436 
  4437 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4438 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4439 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4440 
  4441 * Disposed old type and term read functions (Sign.read_def_typ,
  4442 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4443 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4444 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4445 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4446 resort for legacy applications.
  4447 
  4448 * Disposed old declarations, tactics, tactic combinators that refer to
  4449 the simpset or claset of an implicit theory (such as Addsimps,
  4450 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4451 embedded ML text, or local_simpset_of with a proper context passed as
  4452 explicit runtime argument.
  4453 
  4454 * Rules and tactics that read instantiations (read_instantiate,
  4455 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4456 context, which is required for parsing and type-checking.  Moreover,
  4457 the variables are specified as plain indexnames, not string encodings
  4458 thereof.  INCOMPATIBILITY.
  4459 
  4460 * Generic Toplevel.add_hook interface allows to analyze the result of
  4461 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4462 for theorem dependency output of transactions resulting in a new
  4463 theory state.
  4464 
  4465 * ML antiquotations: block-structured compilation context indicated by
  4466 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4467 
  4468   @{binding name}                         - basic name binding
  4469   @{let ?pat = term}                      - term abbreviation (HO matching)
  4470   @{note name = fact}                     - fact abbreviation
  4471   @{thm fact}                             - singleton fact (with attributes)
  4472   @{thms fact}                            - general fact (with attributes)
  4473   @{lemma prop by method}                 - singleton goal
  4474   @{lemma prop by meth1 meth2}            - singleton goal
  4475   @{lemma prop1 ... propN by method}      - general goal
  4476   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4477   @{lemma (open) ...}                     - open derivation
  4478 
  4479 
  4480 *** System ***
  4481 
  4482 * The Isabelle "emacs" tool provides a specific interface to invoke
  4483 Proof General / Emacs, with more explicit failure if that is not
  4484 installed (the old isabelle-interface script silently falls back on
  4485 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4486 installation location of the Proof General distribution.
  4487 
  4488 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4489 the Isabelle process into a JVM/Scala application.  See
  4490 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4491 process wrapper has been discontinued.)
  4492 
  4493 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4494 
  4495 * Various status messages (with exact source position information) are
  4496 emitted, if proper markup print mode is enabled.  This allows
  4497 user-interface components to provide detailed feedback on internal
  4498 prover operations.
  4499 
  4500 
  4501 
  4502 New in Isabelle2008 (June 2008)
  4503 -------------------------------
  4504 
  4505 *** General ***
  4506 
  4507 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4508 and updated, with formally checked references as hyperlinks.
  4509 
  4510 * Theory loader: use_thy (and similar operations) no longer set the
  4511 implicit ML context, which was occasionally hard to predict and in
  4512 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4513 provides a proper context already.
  4514 
  4515 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4516 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4517 'use' within a theory file will do the job.
  4518 
  4519 * Name space merge now observes canonical order, i.e. the second space
  4520 is inserted into the first one, while existing entries in the first
  4521 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4522 swap theory imports.
  4523 
  4524 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4525 INCOMPATIBILITY in identifier syntax etc.
  4526 
  4527 * Outer syntax: string tokens no longer admit escaped white space,
  4528 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4529 white space without escapes.
  4530 
  4531 * Outer syntax: string tokens may contain arbitrary character codes
  4532 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4533 "foo_bar".
  4534 
  4535 
  4536 *** Pure ***
  4537 
  4538 * Context-dependent token translations.  Default setup reverts locally
  4539 fixed variables, and adds hilite markup for undeclared frees.
  4540 
  4541 * Unused theorems can be found using the new command 'unused_thms'.
  4542 There are three ways of invoking it:
  4543 
  4544 (1) unused_thms
  4545      Only finds unused theorems in the current theory.
  4546 
  4547 (2) unused_thms thy_1 ... thy_n -
  4548      Finds unused theorems in the current theory and all of its ancestors,
  4549      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4550 
  4551 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4552      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4553      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4554      their ancestors.
  4555 
  4556 In order to increase the readability of the list produced by
  4557 unused_thms, theorems that have been created by a particular instance
  4558 of a theory command such as 'inductive' or 'function' are considered
  4559 to belong to the same "group", meaning that if at least one theorem in
  4560 this group is used, the other theorems in the same group are no longer
  4561 reported as unused.  Moreover, if all theorems in the group are
  4562 unused, only one theorem in the group is displayed.
  4563 
  4564 Note that proof objects have to be switched on in order for
  4565 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4566 usually the case when using Proof General with the default settings).
  4567 
  4568 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4569 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4570 duplicate fact bindings, or even accidental fact duplications.  Note
  4571 that tools may maintain dynamically scoped facts systematically, using
  4572 PureThy.add_thms_dynamic.
  4573 
  4574 * Command 'hide' now allows to hide from "fact" name space as well.
  4575 
  4576 * Eliminated destructive theorem database, simpset, claset, and
  4577 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4578 update of theories within ML code.
  4579 
  4580 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4581 INCOMPATIBILITY, object-logics depending on former Pure require
  4582 additional setup PureThy.old_appl_syntax_setup; object-logics
  4583 depending on former CPure need to refer to Pure.
  4584 
  4585 * Commands 'use' and 'ML' are now purely functional, operating on
  4586 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4587 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4588 INCOMPATIBILITY.
  4589 
  4590 * Command 'setup': discontinued implicit version with ML reference.
  4591 
  4592 * Instantiation target allows for simultaneous specification of class
  4593 instance operations together with an instantiation proof.
  4594 Type-checking phase allows to refer to class operations uniformly.
  4595 See src/HOL/Complex/Complex.thy for an Isar example and
  4596 src/HOL/Library/Eval.thy for an ML example.
  4597 
  4598 * Indexing of literal facts: be more serious about including only
  4599 facts from the visible specification/proof context, but not the
  4600 background context (locale etc.).  Affects `prop` notation and method
  4601 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4602 situations.
  4603 
  4604 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4605 "(open)" option, which used to expose internal bound variables to the
  4606 proof text.
  4607 
  4608 * Isar statements: removed obsolete case "rule_context".
  4609 INCOMPATIBILITY, better use explicit fixes/assumes.
  4610 
  4611 * Locale proofs: default proof step now includes 'unfold_locales';
  4612 hence 'proof' without argument may be used to unfold locale
  4613 predicates.
  4614 
  4615 
  4616 *** Document preparation ***
  4617 
  4618 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4619 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4620 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4621 spot); renamed link color from "darkblue" to "linkcolor" (default
  4622 value unchanged, can be redefined via \definecolor); no longer sets
  4623 "a4paper" option (unnecessary or even intrusive).
  4624 
  4625 * Antiquotation @{lemma A method} proves proposition A by the given
  4626 method (either a method name or a method name plus (optional) method
  4627 arguments in parentheses) and prints A just like @{prop A}.
  4628 
  4629 
  4630 *** HOL ***
  4631 
  4632 * New primrec package.  Specification syntax conforms in style to
  4633 definition/function/....  No separate induction rule is provided.  The
  4634 "primrec" command distinguishes old-style and new-style specifications
  4635 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4636 When adjusting theories, beware: constants stemming from new-style
  4637 primrec specifications have authentic syntax.
  4638 
  4639 * Metis prover is now an order of magnitude faster, and also works
  4640 with multithreading.
  4641 
  4642 * Metis: the maximum number of clauses that can be produced from a
  4643 theorem is now given by the attribute max_clauses.  Theorems that
  4644 exceed this number are ignored, with a warning printed.
  4645 
  4646 * Sledgehammer no longer produces structured proofs by default. To
  4647 enable, declare [[sledgehammer_full = true]].  Attributes
  4648 reconstruction_modulus, reconstruction_sorts renamed
  4649 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4650 
  4651 * Method "induct_scheme" derives user-specified induction rules
  4652 from well-founded induction and completeness of patterns. This factors
  4653 out some operations that are done internally by the function package
  4654 and makes them available separately.  See
  4655 src/HOL/ex/Induction_Scheme.thy for examples.
  4656 
  4657 * More flexible generation of measure functions for termination
  4658 proofs: Measure functions can be declared by proving a rule of the
  4659 form "is_measure f" and giving it the [measure_function] attribute.
  4660 The "is_measure" predicate is logically meaningless (always true), and
  4661 just guides the heuristic.  To find suitable measure functions, the
  4662 termination prover sets up the goal "is_measure ?f" of the appropriate
  4663 type and generates all solutions by prolog-style backwards proof using
  4664 the declared rules.
  4665 
  4666 This setup also deals with rules like
  4667 
  4668   "is_measure f ==> is_measure (list_size f)"
  4669 
  4670 which accommodates nested datatypes that recurse through lists.
  4671 Similar rules are predeclared for products and option types.
  4672 
  4673 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4674 
  4675   INCOMPATIBILITIES:
  4676 
  4677   - Definitions of overloaded constants on sets have to be replaced by
  4678     definitions on => and bool.
  4679 
  4680   - Some definitions of overloaded operators on sets can now be proved
  4681     using the definitions of the operators on => and bool.  Therefore,
  4682     the following theorems have been renamed:
  4683 
  4684       subset_def   -> subset_eq
  4685       psubset_def  -> psubset_eq
  4686       set_diff_def -> set_diff_eq
  4687       Compl_def    -> Compl_eq
  4688       Sup_set_def  -> Sup_set_eq
  4689       Inf_set_def  -> Inf_set_eq
  4690       sup_set_def  -> sup_set_eq
  4691       inf_set_def  -> inf_set_eq
  4692 
  4693   - Due to the incompleteness of the HO unification algorithm, some
  4694     rules such as subst may require manual instantiation, if some of
  4695     the unknowns in the rule is a set.
  4696 
  4697   - Higher order unification and forward proofs:
  4698     The proof pattern
  4699 
  4700       have "P (S::'a set)" <...>
  4701       then have "EX S. P S" ..
  4702 
  4703     no longer works (due to the incompleteness of the HO unification
  4704     algorithm) and must be replaced by the pattern
  4705 
  4706       have "EX S. P S"
  4707       proof
  4708         show "P S" <...>
  4709       qed
  4710 
  4711   - Calculational reasoning with subst (or similar rules):
  4712     The proof pattern
  4713 
  4714       have "P (S::'a set)" <...>
  4715       also have "S = T" <...>
  4716       finally have "P T" .
  4717 
  4718     no longer works (for similar reasons as the previous example) and
  4719     must be replaced by something like
  4720 
  4721       have "P (S::'a set)" <...>
  4722       moreover have "S = T" <...>
  4723       ultimately have "P T" by simp
  4724 
  4725   - Tactics or packages written in ML code:
  4726     Code performing pattern matching on types via
  4727 
  4728       Type ("set", [T]) => ...
  4729 
  4730     must be rewritten. Moreover, functions like strip_type or
  4731     binder_types no longer return the right value when applied to a
  4732     type of the form
  4733 
  4734       T1 => ... => Tn => U => bool
  4735 
  4736     rather than
  4737 
  4738       T1 => ... => Tn => U set
  4739 
  4740 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4741 Wellfounded_Relations to theory Wellfounded.
  4742 
  4743 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4744 
  4745 * Class finite no longer treats UNIV as class parameter.  Use class
  4746 enum from theory Library/Enum instead to achieve a similar effect.
  4747 INCOMPATIBILITY.
  4748 
  4749 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4750 and "Cons".  INCOMPATIBILITY.
  4751 
  4752 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4753 Potential INCOMPATIBILITY.
  4754 
  4755 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4756 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4757 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4758 
  4759 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4760 the same name in theory Orderings:
  4761 
  4762   less_trans
  4763   less_linear
  4764   le_imp_less_or_eq
  4765   le_less_trans
  4766   less_le_trans
  4767   less_not_sym
  4768   less_asym
  4769 
  4770 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4771 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4772 and different variable names.
  4773 
  4774 * Library/Option_ord.thy: Canonical order on option type.
  4775 
  4776 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4777 finite maps.
  4778 
  4779 * Library/Countable.thy: Type class for countable types.
  4780 
  4781 * Theory Int: The representation of numerals has changed.  The infix
  4782 operator BIT and the bit datatype with constructors B0 and B1 have
  4783 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4784 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4785 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4786 accordingly.
  4787 
  4788 * Theory Nat: definition of <= and < on natural numbers no longer
  4789 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4790 le_def and less_def have disappeared.  Consider lemmas not_less
  4791 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4792 
  4793 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4794 (whose purpose mainly is for various fold_set functionals) have been
  4795 abandoned in favor of the existing algebraic classes
  4796 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4797 lower_semilattice (resp. upper_semilattice) and linorder.
  4798 INCOMPATIBILITY.
  4799 
  4800 * Theory Transitive_Closure: induct and cases rules now declare proper
  4801 case_names ("base" and "step").  INCOMPATIBILITY.
  4802 
  4803 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4804 lattices.  The form set-specific version is available as
  4805 Inductive.lfp_ordinal_induct_set.
  4806 
  4807 * Renamed theorems "power.simps" to "power_int.simps".
  4808 INCOMPATIBILITY.
  4809 
  4810 * Class semiring_div provides basic abstract properties of semirings
  4811 with division and modulo operations.  Subsumes former class dvd_mod.
  4812 
  4813 * Merged theories IntDef, Numeral and IntArith into unified theory
  4814 Int.  INCOMPATIBILITY.
  4815 
  4816 * Theory Library/Code_Index: type "index" now represents natural
  4817 numbers rather than integers.  INCOMPATIBILITY.
  4818 
  4819 * New class "uminus" with operation "uminus" (split of from class
  4820 "minus" which now only has operation "minus", binary).
  4821 INCOMPATIBILITY.
  4822 
  4823 * Constants "card", "internal_split", "option_map" now with authentic
  4824 syntax.  INCOMPATIBILITY.
  4825 
  4826 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4827 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4828 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4829 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4830 Sup_set_def, le_def, less_def, option_map_def now with object
  4831 equality.  INCOMPATIBILITY.
  4832 
  4833 * Records. Removed K_record, and replaced it by pure lambda term
  4834 %x. c. The simplifier setup is now more robust against eta expansion.
  4835 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4836 
  4837 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4838 
  4839 * Library/ListVector: new theory of arithmetic vector operations.
  4840 
  4841 * Library/Order_Relation: new theory of various orderings as sets of
  4842 pairs.  Defines preorders, partial orders, linear orders and
  4843 well-orders on sets and on types.
  4844 
  4845 
  4846 *** ZF ***
  4847 
  4848 * Renamed some theories to allow to loading both ZF and HOL in the
  4849 same session:
  4850 
  4851   Datatype  -> Datatype_ZF
  4852   Inductive -> Inductive_ZF
  4853   Int       -> Int_ZF
  4854   IntDiv    -> IntDiv_ZF
  4855   Nat       -> Nat_ZF
  4856   List      -> List_ZF
  4857   Main      -> Main_ZF
  4858 
  4859 INCOMPATIBILITY: ZF theories that import individual theories below
  4860 Main might need to be adapted.  Regular theory Main is still
  4861 available, as trivial extension of Main_ZF.
  4862 
  4863 
  4864 *** ML ***
  4865 
  4866 * ML within Isar: antiquotation @{const name} or @{const
  4867 name(typargs)} produces statically-checked Const term.
  4868 
  4869 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4870 (of the same name).  Removed obsolete print command.
  4871 
  4872 * Removed obsolete "use_legacy_bindings" function.
  4873 
  4874 * The ``print mode'' is now a thread-local value derived from a global
  4875 template (the former print_mode reference), thus access becomes
  4876 non-critical.  The global print_mode reference is for session
  4877 management only; user-code should use print_mode_value,
  4878 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4879 
  4880 * Functions system/system_out provide a robust way to invoke external
  4881 shell commands, with propagation of interrupts (requires Poly/ML
  4882 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4883 
  4884 
  4885 *** System ***
  4886 
  4887 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4888 in accordance with Proof General 3.7, which prefers GNU emacs.
  4889 
  4890 * isatool tty runs Isabelle process with plain tty interaction;
  4891 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4892 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4893 
  4894 * isatool browser now works with Cygwin as well, using general
  4895 "javapath" function defined in Isabelle process environment.
  4896 
  4897 * YXML notation provides a simple and efficient alternative to
  4898 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4899 isatool yxml as described in the Isabelle system manual.
  4900 
  4901 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4902 provides general wrapper for managing an Isabelle process in a robust
  4903 fashion, with ``cooked'' output from stdin/stderr.
  4904 
  4905 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4906 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4907 
  4908 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4909 way of changing the user's settings is via
  4910 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4911 script.
  4912 
  4913 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4914 cores of the underlying machine, which is a good starting point for
  4915 optimal performance tuning.  The corresponding usedir option -M allows
  4916 "max" as an alias for "0".  WARNING: does not work on certain versions
  4917 of Mac OS (with Poly/ML 5.1).
  4918 
  4919 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4920 adverse effect of Isabelle flooding interactive front-ends (notably
  4921 ProofGeneral / XEmacs).
  4922 
  4923 
  4924 
  4925 New in Isabelle2007 (November 2007)
  4926 -----------------------------------
  4927 
  4928 *** General ***
  4929 
  4930 * More uniform information about legacy features, notably a
  4931 warning/error of "Legacy feature: ...", depending on the state of the
  4932 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4933 legacy features will disappear eventually.
  4934 
  4935 * Theory syntax: the header format ``theory A = B + C:'' has been
  4936 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4937 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4938 
  4939 * Theory syntax: the old non-Isar theory file format has been
  4940 discontinued altogether.  Note that ML proof scripts may still be used
  4941 with Isar theories; migration is usually quite simple with the ML
  4942 function use_legacy_bindings.  INCOMPATIBILITY.
  4943 
  4944 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4945 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4946 quotes.
  4947 
  4948 * Theory loader: be more serious about observing the static theory
  4949 header specifications (including optional directories), but not the
  4950 accidental file locations of previously successful loads.  The strict
  4951 update policy of former update_thy is now already performed by
  4952 use_thy, so the former has been removed; use_thys updates several
  4953 theories simultaneously, just as 'imports' within a theory header
  4954 specification, but without merging the results.  Potential
  4955 INCOMPATIBILITY: may need to refine theory headers and commands
  4956 ROOT.ML which depend on load order.
  4957 
  4958 * Theory loader: optional support for content-based file
  4959 identification, instead of the traditional scheme of full physical
  4960 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4961 (cf. the system manual).  The new scheme allows to work with
  4962 non-finished theories in persistent session images, such that source
  4963 files may be moved later on without requiring reloads.
  4964 
  4965 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4966 file (with the same base name as the theory) are considered a legacy
  4967 feature, which will disappear eventually. Even now, the theory loader
  4968 no longer maintains dependencies on such files.
  4969 
  4970 * Syntax: the scope for resolving ambiguities via type-inference is
  4971 now limited to individual terms, instead of whole simultaneous
  4972 specifications as before. This greatly reduces the complexity of the
  4973 syntax module and improves flexibility by separating parsing and
  4974 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4975 'fixes' etc.) are required in rare situations.
  4976 
  4977 * Syntax: constants introduced by new-style packages ('definition',
  4978 'abbreviation' etc.) are passed through the syntax module in
  4979 ``authentic mode''. This means that associated mixfix annotations
  4980 really stick to such constants, independently of potential name space
  4981 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4982 trees are represented slightly differently, may need to adapt syntax
  4983 translations accordingly. Use CONST marker in 'translations' and
  4984 @{const_syntax} antiquotation in 'parse_translation' etc.
  4985 
  4986 * Legacy goal package: reduced interface to the bare minimum required
  4987 to keep existing proof scripts running.  Most other user-level
  4988 functions are now part of the OldGoals structure, which is *not* open
  4989 by default (consider isatool expandshort before open OldGoals).
  4990 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4991 these tend to cause confusion about the actual goal (!) context being
  4992 used here, which is not necessarily the same as the_context().
  4993 
  4994 * Command 'find_theorems': supports "*" wild-card in "name:"
  4995 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4996 support a specific search form (see ProofGeneral/CHANGES).
  4997 
  4998 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4999 by default, which means that "prems" (and also "fixed variables") are
  5000 suppressed from proof state output.  Note that the ProofGeneral
  5001 settings mechanism allows to change and save options persistently, but
  5002 older versions of Isabelle will fail to start up if a negative prems
  5003 limit is imposed.
  5004 
  5005 * Local theory targets may be specified by non-nested blocks of
  5006 ``context/locale/class ... begin'' followed by ``end''.  The body may
  5007 contain definitions, theorems etc., including any derived mechanism
  5008 that has been implemented on top of these primitives.  This concept
  5009 generalizes the existing ``theorem (in ...)'' towards more versatility
  5010 and scalability.
  5011 
  5012 * Proof General interface: proper undo of final 'end' command;
  5013 discontinued Isabelle/classic mode (ML proof scripts).
  5014 
  5015 
  5016 *** Document preparation ***
  5017 
  5018 * Added antiquotation @{theory name} which prints the given name,
  5019 after checking that it refers to a valid ancestor theory in the
  5020 current context.
  5021 
  5022 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5023 check the given source text as ML type/structure, printing verbatim.
  5024 
  5025 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5026 "c args == rhs" given in the current context.  (Any number of
  5027 arguments may be given on the LHS.)
  5028 
  5029 
  5030 *** Pure ***
  5031 
  5032 * The 'class' package offers a combination of axclass and locale to
  5033 achieve Haskell-like type classes in Isabelle.  Definitions and
  5034 theorems within a class context produce both relative results (with
  5035 implicit parameters according to the locale context), and polymorphic
  5036 constants with qualified polymorphism (according to the class
  5037 context).  Within the body context of a 'class' target, a separate
  5038 syntax layer ("user space type system") takes care of converting
  5039 between global polymorphic consts and internal locale representation.
  5040 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5041 "isatool doc classes" provides a tutorial.
  5042 
  5043 * Generic code generator framework allows to generate executable
  5044 code for ML and Haskell (including Isabelle classes).  A short usage
  5045 sketch:
  5046 
  5047     internal compilation:
  5048         export_code <list of constants (term syntax)> in SML
  5049     writing SML code to a file:
  5050         export_code <list of constants (term syntax)> in SML <filename>
  5051     writing OCaml code to a file:
  5052         export_code <list of constants (term syntax)> in OCaml <filename>
  5053     writing Haskell code to a bunch of files:
  5054         export_code <list of constants (term syntax)> in Haskell <filename>
  5055 
  5056     evaluating closed propositions to True/False using code generation:
  5057         method ``eval''
  5058 
  5059 Reasonable default setup of framework in HOL.
  5060 
  5061 Theorem attributs for selecting and transforming function equations theorems:
  5062 
  5063     [code fun]:        select a theorem as function equation for a specific constant
  5064     [code fun del]:    deselect a theorem as function equation for a specific constant
  5065     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5066     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5067 
  5068 User-defined serializations (target in {SML, OCaml, Haskell}):
  5069 
  5070     code_const <and-list of constants (term syntax)>
  5071       {(target) <and-list of const target syntax>}+
  5072 
  5073     code_type <and-list of type constructors>
  5074       {(target) <and-list of type target syntax>}+
  5075 
  5076     code_instance <and-list of instances>
  5077       {(target)}+
  5078         where instance ::= <type constructor> :: <class>
  5079 
  5080     code_class <and_list of classes>
  5081       {(target) <and-list of class target syntax>}+
  5082         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5083 
  5084 code_instance and code_class only are effective to target Haskell.
  5085 
  5086 For example usage see src/HOL/ex/Codegenerator.thy and
  5087 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5088 generation from Isabelle/HOL theories is available via "isatool doc
  5089 codegen".
  5090 
  5091 * Code generator: consts in 'consts_code' Isar commands are now
  5092 referred to by usual term syntax (including optional type
  5093 annotations).
  5094 
  5095 * Command 'no_translations' removes translation rules from theory
  5096 syntax.
  5097 
  5098 * Overloaded definitions are now actually checked for acyclic
  5099 dependencies.  The overloading scheme is slightly more general than
  5100 that of Haskell98, although Isabelle does not demand an exact
  5101 correspondence to type class and instance declarations.
  5102 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5103 exotic versions of overloading -- at the discretion of the user!
  5104 
  5105 Polymorphic constants are represented via type arguments, i.e. the
  5106 instantiation that matches an instance against the most general
  5107 declaration given in the signature.  For example, with the declaration
  5108 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5109 as c(nat).  Overloading is essentially simultaneous structural
  5110 recursion over such type arguments.  Incomplete specification patterns
  5111 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5112 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5113 Command 'print_theory' outputs the normalized system of recursive
  5114 equations, see section "definitions".
  5115 
  5116 * Configuration options are maintained within the theory or proof
  5117 context (with name and type bool/int/string), providing a very simple
  5118 interface to a poor-man's version of general context data.  Tools may
  5119 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5120 these values using Config.get etc.  Users may change options via an
  5121 associated attribute of the same name.  This form of context
  5122 declaration works particularly well with commands 'declare' or
  5123 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5124 very easy to avoid global references, which would not observe Isar
  5125 toplevel undo/redo and fail to work with multithreading.
  5126 
  5127 Various global ML references of Pure and HOL have been turned into
  5128 configuration options:
  5129 
  5130   Unify.search_bound		unify_search_bound
  5131   Unify.trace_bound		unify_trace_bound
  5132   Unify.trace_simp		unify_trace_simp
  5133   Unify.trace_types		unify_trace_types
  5134   Simplifier.simp_depth_limit	simp_depth_limit
  5135   Blast.depth_limit		blast_depth_limit
  5136   DatatypeProp.dtK		datatype_distinctness_limit
  5137   fast_arith_neq_limit  	fast_arith_neq_limit
  5138   fast_arith_split_limit	fast_arith_split_limit
  5139 
  5140 * Named collections of theorems may be easily installed as context
  5141 data using the functor NamedThmsFun (see also
  5142 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5143 attributes; there is also a toplevel print command.  This facility is
  5144 just a common case of general context data, which is the preferred way
  5145 for anything more complex than just a list of facts in canonical
  5146 order.
  5147 
  5148 * Isar: command 'declaration' augments a local theory by generic
  5149 declaration functions written in ML.  This enables arbitrary content
  5150 being added to the context, depending on a morphism that tells the
  5151 difference of the original declaration context wrt. the application
  5152 context encountered later on.
  5153 
  5154 * Isar: proper interfaces for simplification procedures.  Command
  5155 'simproc_setup' declares named simprocs (with match patterns, and body
  5156 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5157 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5158 
  5159 * Isar: an extra pair of brackets around attribute declarations
  5160 abbreviates a theorem reference involving an internal dummy fact,
  5161 which will be ignored later --- only the effect of the attribute on
  5162 the background context will persist.  This form of in-place
  5163 declarations is particularly useful with commands like 'declare' and
  5164 'using', for example ``have A using [[simproc a]] by simp''.
  5165 
  5166 * Isar: method "assumption" (and implicit closing of subproofs) now
  5167 takes simple non-atomic goal assumptions into account: after applying
  5168 an assumption as a rule the resulting subgoals are solved by atomic
  5169 assumption steps.  This is particularly useful to finish 'obtain'
  5170 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5171 without referring to the original premise "!!x. P x ==> thesis" in the
  5172 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5173 more permissive.
  5174 
  5175 * Isar: implicit use of prems from the Isar proof context is
  5176 considered a legacy feature.  Common applications like ``have A .''
  5177 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5178 referencing facts explicitly here improves readability and
  5179 maintainability of proof texts.
  5180 
  5181 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5182 the obtained context from the course of reasoning!  For example:
  5183 
  5184   assume "EX x y. A x & B y"   -- "any previous fact"
  5185   then guess x and y by clarify
  5186 
  5187 This technique is potentially adventurous, depending on the facts and
  5188 proof tools being involved here.
  5189 
  5190 * Isar: known facts from the proof context may be specified as literal
  5191 propositions, using ASCII back-quote syntax.  This works wherever
  5192 named facts used to be allowed so far, in proof commands, proof
  5193 methods, attributes etc.  Literal facts are retrieved from the context
  5194 according to unification of type and term parameters.  For example,
  5195 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5196 theorems in the current context, then these are valid literal facts:
  5197 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5198 
  5199 There is also a proof method "fact" which does the same composition
  5200 for explicit goal states, e.g. the following proof texts coincide with
  5201 certain special cases of literal facts:
  5202 
  5203   have "A" by fact                 ==  note `A`
  5204   have "A ==> B" by fact           ==  note `A ==> B`
  5205   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5206   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5207 
  5208 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5209 outer syntax.  Thus symbolic identifiers may be used without
  5210 additional white space in declarations like this: ``assume *: A''.
  5211 
  5212 * Isar: 'print_facts' prints all local facts of the current context,
  5213 both named and unnamed ones.
  5214 
  5215 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5216 
  5217   def x == "t" and y == "u"
  5218 
  5219 * Isar: added command 'unfolding', which is structurally similar to
  5220 'using', but affects both the goal state and facts by unfolding given
  5221 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5222 'unfolded' attribute may be replaced by first-class proof text.
  5223 
  5224 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5225 and command 'unfolding' now all support object-level equalities
  5226 (potentially conditional).  The underlying notion of rewrite rule is
  5227 analogous to the 'rule_format' attribute, but *not* that of the
  5228 Simplifier (which is usually more generous).
  5229 
  5230 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5231 premises of a theorem by n. Useful in conjunction with drule.
  5232 
  5233 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5234 method expression within a sandbox consisting of the first N
  5235 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5236 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5237 simplifies all new goals that emerge from applying rule foo to the
  5238 originally first one.
  5239 
  5240 * Isar: schematic goals are no longer restricted to higher-order
  5241 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5242 expected.
  5243 
  5244 * Isar: the conclusion of a long theorem statement is now either
  5245 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5246 (essentially a disjunction of cases with local parameters and
  5247 assumptions).  The latter allows to express general elimination rules
  5248 adequately; in this notation common elimination rules look like this:
  5249 
  5250   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5251     assumes "EX x. P x"
  5252     obtains x where "P x"
  5253 
  5254   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5255     assumes "A & B"
  5256     obtains A and B
  5257 
  5258   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5259     assumes "A | B"
  5260     obtains
  5261       A
  5262     | B
  5263 
  5264 The subsequent classical rules even refer to the formal "thesis"
  5265 explicitly:
  5266 
  5267   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5268     obtains "~ thesis"
  5269 
  5270   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5271     obtains "thesis ==> something"
  5272 
  5273 The actual proof of an 'obtains' statement is analogous to that of the
  5274 Isar proof element 'obtain', only that there may be several cases.
  5275 Optional case names may be specified in parentheses; these will be
  5276 available both in the present proof and as annotations in the
  5277 resulting rule, for later use with the 'cases' method (cf. attribute
  5278 case_names).
  5279 
  5280 * Isar: the assumptions of a long theorem statement are available as
  5281 "assms" fact in the proof context.  This is more appropriate than the
  5282 (historical) "prems", which refers to all assumptions of the current
  5283 context, including those from the target locale, proof body etc.
  5284 
  5285 * Isar: 'print_statement' prints theorems from the current theory or
  5286 proof context in long statement form, according to the syntax of a
  5287 top-level lemma.
  5288 
  5289 * Isar: 'obtain' takes an optional case name for the local context
  5290 introduction rule (default "that").
  5291 
  5292 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5293 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5294 to occur.
  5295 
  5296 * Pure: syntax "CONST name" produces a fully internalized constant
  5297 according to the current context.  This is particularly useful for
  5298 syntax translations that should refer to internal constant
  5299 representations independently of name spaces.
  5300 
  5301 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5302 instead of "FOO ". This allows multiple binder declarations to coexist
  5303 in the same context.  INCOMPATIBILITY.
  5304 
  5305 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5306 primitive that also works in a locale context (both for constants and
  5307 fixed variables). Type declaration and internal syntactic representation
  5308 of given constants retrieved from the context. Likewise, the
  5309 'no_notation' command allows to remove given syntax annotations from the
  5310 current context.
  5311 
  5312 * Isar/locales: new derived specification elements 'axiomatization',
  5313 'definition', 'abbreviation', which support type-inference, admit
  5314 object-level specifications (equality, equivalence).  See also the
  5315 isar-ref manual.  Examples:
  5316 
  5317   axiomatization
  5318     eq  (infix "===" 50) where
  5319     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5320 
  5321   definition "f x y = x + y + 1"
  5322   definition g where "g x = f x x"
  5323 
  5324   abbreviation
  5325     neq  (infix "=!=" 50) where
  5326     "x =!= y == ~ (x === y)"
  5327 
  5328 These specifications may be also used in a locale context.  Then the
  5329 constants being introduced depend on certain fixed parameters, and the
  5330 constant name is qualified by the locale base name.  An internal
  5331 abbreviation takes care for convenient input and output, making the
  5332 parameters implicit and using the original short name.  See also
  5333 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5334 entities from a monomorphic theory.
  5335 
  5336 Presently, abbreviations are only available 'in' a target locale, but
  5337 not inherited by general import expressions.  Also note that
  5338 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5339 'translations' in common applications.  The "no_abbrevs" print mode
  5340 prevents folding of abbreviations in term output.
  5341 
  5342 Concrete syntax is attached to specified constants in internal form,
  5343 independently of name spaces.  The parse tree representation is
  5344 slightly different -- use 'notation' instead of raw 'syntax', and
  5345 'translations' with explicit "CONST" markup to accommodate this.
  5346 
  5347 * Pure/Isar: unified syntax for new-style specification mechanisms
  5348 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5349 full type inference and dummy patterns ("_").  For example:
  5350 
  5351   definition "K x _ = x"
  5352 
  5353   inductive conj for A B
  5354   where "A ==> B ==> conj A B"
  5355 
  5356 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5357 the current context.  Print mode "no_abbrevs" prevents inversion of
  5358 abbreviations on output.
  5359 
  5360 * Isar/locales: improved parameter handling: use of locales "var" and
  5361 "struct" no longer necessary; - parameter renamings are no longer
  5362 required to be injective.  For example, this allows to define
  5363 endomorphisms as locale endom = homom mult mult h.
  5364 
  5365 * Isar/locales: changed the way locales with predicates are defined.
  5366 Instead of accumulating the specification, the imported expression is
  5367 now an interpretation.  INCOMPATIBILITY: different normal form of
  5368 locale expressions.  In particular, in interpretations of locales with
  5369 predicates, goals repesenting already interpreted fragments are not
  5370 removed automatically.  Use methods `intro_locales' and
  5371 `unfold_locales'; see below.
  5372 
  5373 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5374 provide backward reasoning on locales predicates.  The methods are
  5375 aware of interpretations and discharge corresponding goals.
  5376 `intro_locales' is less aggressive then `unfold_locales' and does not
  5377 unfold predicates to assumptions.
  5378 
  5379 * Isar/locales: the order in which locale fragments are accumulated
  5380 has changed.  This enables to override declarations from fragments due
  5381 to interpretations -- for example, unwanted simp rules.
  5382 
  5383 * Isar/locales: interpretation in theories and proof contexts has been
  5384 extended.  One may now specify (and prove) equations, which are
  5385 unfolded in interpreted theorems.  This is useful for replacing
  5386 defined concepts (constants depending on locale parameters) by
  5387 concepts already existing in the target context.  Example:
  5388 
  5389   interpretation partial_order ["op <= :: [int, int] => bool"]
  5390     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5391 
  5392 Typically, the constant `partial_order.less' is created by a
  5393 definition specification element in the context of locale
  5394 partial_order.
  5395 
  5396 * Method "induct": improved internal context management to support
  5397 local fixes and defines on-the-fly. Thus explicit meta-level
  5398 connectives !!  and ==> are rarely required anymore in inductive goals
  5399 (using object-logic connectives for this purpose has been long
  5400 obsolete anyway). Common proof patterns are explained in
  5401 src/HOL/Induct/Common_Patterns.thy, see also
  5402 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5403 examples.
  5404 
  5405 * Method "induct": improved handling of simultaneous goals. Instead of
  5406 introducing object-level conjunction, the statement is now split into
  5407 several conclusions, while the corresponding symbolic cases are nested
  5408 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5409 see src/HOL/Induct/Common_Patterns.thy, for example.
  5410 
  5411 * Method "induct": mutual induction rules are now specified as a list
  5412 of rule sharing the same induction cases. HOL packages usually provide
  5413 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5414 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5415 mutual induction rules differently, i.e. like this:
  5416 
  5417   (induct rule: foo_bar.inducts)
  5418   (induct set: foo bar)
  5419   (induct pred: foo bar)
  5420   (induct type: foo bar)
  5421 
  5422 The ML function ProjectRule.projections turns old-style rules into the
  5423 new format.
  5424 
  5425 * Method "coinduct": dual of induction, see
  5426 src/HOL/Library/Coinductive_List.thy for various examples.
  5427 
  5428 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5429 considered a legacy feature.
  5430 
  5431 * Attribute "symmetric" produces result with standardized schematic
  5432 variables (index 0).  Potential INCOMPATIBILITY.
  5433 
  5434 * Simplifier: by default the simplifier trace only shows top level
  5435 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5436 default. Thus there is less danger of being flooded by the trace. The
  5437 trace indicates where parts have been suppressed.
  5438 
  5439 * Provers/classical: removed obsolete classical version of elim_format
  5440 attribute; classical elim/dest rules are now treated uniformly when
  5441 manipulating the claset.
  5442 
  5443 * Provers/classical: stricter checks to ensure that supplied intro,
  5444 dest and elim rules are well-formed; dest and elim rules must have at
  5445 least one premise.
  5446 
  5447 * Provers/classical: attributes dest/elim/intro take an optional
  5448 weight argument for the rule (just as the Pure versions).  Weights are
  5449 ignored by automated tools, but determine the search order of single
  5450 rule steps.
  5451 
  5452 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5453 where the body does not mention the bound variable.  Note that dummy
  5454 patterns implicitly depend on their context of bounds, which makes
  5455 "{_. _}" match any set comprehension as expected.  Potential
  5456 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5457 constant "_idtdummy" in the binding position.
  5458 
  5459 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5460 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5461 for example "A -> B" => "Pi A (%_. B)".
  5462 
  5463 * Pure: 'class_deps' command visualizes the subclass relation, using
  5464 the graph browser tool.
  5465 
  5466 * Pure: 'print_theory' now suppresses certain internal declarations by
  5467 default; use '!' option for full details.
  5468 
  5469 
  5470 *** HOL ***
  5471 
  5472 * Method "metis" proves goals by applying the Metis general-purpose
  5473 resolution prover (see also http://gilith.com/software/metis/).
  5474 Examples are in the directory MetisExamples.  WARNING: the
  5475 Isabelle/HOL-Metis integration does not yet work properly with
  5476 multi-threading.
  5477 
  5478 * Command 'sledgehammer' invokes external automatic theorem provers as
  5479 background processes.  It generates calls to the "metis" method if
  5480 successful. These can be pasted into the proof.  Users do not have to
  5481 wait for the automatic provers to return.  WARNING: does not really
  5482 work with multi-threading.
  5483 
  5484 * New "auto_quickcheck" feature tests outermost goal statements for
  5485 potential counter-examples.  Controlled by ML references
  5486 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5487 5000 milliseconds).  Fails silently if statements is outside of
  5488 executable fragment, or any other codgenerator problem occurs.
  5489 
  5490 * New constant "undefined" with axiom "undefined x = undefined".
  5491 
  5492 * Added class "HOL.eq", allowing for code generation with polymorphic
  5493 equality.
  5494 
  5495 * Some renaming of class constants due to canonical name prefixing in
  5496 the new 'class' package:
  5497 
  5498     HOL.abs ~> HOL.abs_class.abs
  5499     HOL.divide ~> HOL.divide_class.divide
  5500     0 ~> HOL.zero_class.zero
  5501     1 ~> HOL.one_class.one
  5502     op + ~> HOL.plus_class.plus
  5503     op - ~> HOL.minus_class.minus
  5504     uminus ~> HOL.minus_class.uminus
  5505     op * ~> HOL.times_class.times
  5506     op < ~> HOL.ord_class.less
  5507     op <= > HOL.ord_class.less_eq
  5508     Nat.power ~> Power.power_class.power
  5509     Nat.size ~> Nat.size_class.size
  5510     Numeral.number_of ~> Numeral.number_class.number_of
  5511     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5512     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5513     Orderings.min ~> Orderings.ord_class.min
  5514     Orderings.max ~> Orderings.ord_class.max
  5515     Divides.op div ~> Divides.div_class.div
  5516     Divides.op mod ~> Divides.div_class.mod
  5517     Divides.op dvd ~> Divides.div_class.dvd
  5518 
  5519 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5520 
  5521 a) User-defined constants using any of the names "plus", "minus",
  5522 "times", "less" or "less_eq". The standard syntax translations for
  5523 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5524 names.
  5525 
  5526 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5527 INCOMPATIBILITY: use more specific names.
  5528 
  5529 c) Permutative equations (e.g. "a + b = b + a")
  5530 Since the change of names also changes the order of terms, permutative
  5531 rewrite rules may get applied in a different order. Experience shows
  5532 that this is rarely the case (only two adaptions in the whole Isabelle
  5533 distribution).  INCOMPATIBILITY: rewrite proofs
  5534 
  5535 d) ML code directly refering to constant names
  5536 This in general only affects hand-written proof tactics, simprocs and
  5537 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5538 Consider using @{const_name} antiquotation.
  5539 
  5540 * New class "default" with associated constant "default".
  5541 
  5542 * Function "sgn" is now overloaded and available on int, real, complex
  5543 (and other numeric types), using class "sgn".  Two possible defs of
  5544 sgn are given as equational assumptions in the classes sgn_if and
  5545 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5546 INCOMPATIBILITY.
  5547 
  5548 * Locale "partial_order" now unified with class "order" (cf. theory
  5549 Orderings), added parameter "less".  INCOMPATIBILITY.
  5550 
  5551 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5552 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5553 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5554 
  5555 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5556 changed order of proof goals in instance proofs.
  5557 
  5558 * The transitivity reasoner for partial and linear orders is set up
  5559 for classes "order" and "linorder".  Instances of the reasoner are available
  5560 in all contexts importing or interpreting the corresponding locales.
  5561 Method "order" invokes the reasoner separately; the reasoner
  5562 is also integrated with the Simplifier as a solver.  Diagnostic
  5563 command 'print_orders' shows the available instances of the reasoner
  5564 in the current context.
  5565 
  5566 * Localized monotonicity predicate in theory "Orderings"; integrated
  5567 lemmas max_of_mono and min_of_mono with this predicate.
  5568 INCOMPATIBILITY.
  5569 
  5570 * Formulation of theorem "dense" changed slightly due to integration
  5571 with new class dense_linear_order.
  5572 
  5573 * Uniform lattice theory development in HOL.
  5574 
  5575     constants "meet" and "join" now named "inf" and "sup"
  5576     constant "Meet" now named "Inf"
  5577 
  5578     classes "meet_semilorder" and "join_semilorder" now named
  5579       "lower_semilattice" and "upper_semilattice"
  5580     class "lorder" now named "lattice"
  5581     class "comp_lat" now named "complete_lattice"
  5582 
  5583     Instantiation of lattice classes allows explicit definitions
  5584     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5585 
  5586   INCOMPATIBILITY.  Theorem renames:
  5587 
  5588     meet_left_le            ~> inf_le1
  5589     meet_right_le           ~> inf_le2
  5590     join_left_le            ~> sup_ge1
  5591     join_right_le           ~> sup_ge2
  5592     meet_join_le            ~> inf_sup_ord
  5593     le_meetI                ~> le_infI
  5594     join_leI                ~> le_supI
  5595     le_meet                 ~> le_inf_iff
  5596     le_join                 ~> ge_sup_conv
  5597     meet_idempotent         ~> inf_idem
  5598     join_idempotent         ~> sup_idem
  5599     meet_comm               ~> inf_commute
  5600     join_comm               ~> sup_commute
  5601     meet_leI1               ~> le_infI1
  5602     meet_leI2               ~> le_infI2
  5603     le_joinI1               ~> le_supI1
  5604     le_joinI2               ~> le_supI2
  5605     meet_assoc              ~> inf_assoc
  5606     join_assoc              ~> sup_assoc
  5607     meet_left_comm          ~> inf_left_commute
  5608     meet_left_idempotent    ~> inf_left_idem
  5609     join_left_comm          ~> sup_left_commute
  5610     join_left_idempotent    ~> sup_left_idem
  5611     meet_aci                ~> inf_aci
  5612     join_aci                ~> sup_aci
  5613     le_def_meet             ~> le_iff_inf
  5614     le_def_join             ~> le_iff_sup
  5615     join_absorp2            ~> sup_absorb2
  5616     join_absorp1            ~> sup_absorb1
  5617     meet_absorp1            ~> inf_absorb1
  5618     meet_absorp2            ~> inf_absorb2
  5619     meet_join_absorp        ~> inf_sup_absorb
  5620     join_meet_absorp        ~> sup_inf_absorb
  5621     distrib_join_le         ~> distrib_sup_le
  5622     distrib_meet_le         ~> distrib_inf_le
  5623 
  5624     add_meet_distrib_left   ~> add_inf_distrib_left
  5625     add_join_distrib_left   ~> add_sup_distrib_left
  5626     is_join_neg_meet        ~> is_join_neg_inf
  5627     is_meet_neg_join        ~> is_meet_neg_sup
  5628     add_meet_distrib_right  ~> add_inf_distrib_right
  5629     add_join_distrib_right  ~> add_sup_distrib_right
  5630     add_meet_join_distribs  ~> add_sup_inf_distribs
  5631     join_eq_neg_meet        ~> sup_eq_neg_inf
  5632     meet_eq_neg_join        ~> inf_eq_neg_sup
  5633     add_eq_meet_join        ~> add_eq_inf_sup
  5634     meet_0_imp_0            ~> inf_0_imp_0
  5635     join_0_imp_0            ~> sup_0_imp_0
  5636     meet_0_eq_0             ~> inf_0_eq_0
  5637     join_0_eq_0             ~> sup_0_eq_0
  5638     neg_meet_eq_join        ~> neg_inf_eq_sup
  5639     neg_join_eq_meet        ~> neg_sup_eq_inf
  5640     join_eq_if              ~> sup_eq_if
  5641 
  5642     mono_meet               ~> mono_inf
  5643     mono_join               ~> mono_sup
  5644     meet_bool_eq            ~> inf_bool_eq
  5645     join_bool_eq            ~> sup_bool_eq
  5646     meet_fun_eq             ~> inf_fun_eq
  5647     join_fun_eq             ~> sup_fun_eq
  5648     meet_set_eq             ~> inf_set_eq
  5649     join_set_eq             ~> sup_set_eq
  5650     meet1_iff               ~> inf1_iff
  5651     meet2_iff               ~> inf2_iff
  5652     meet1I                  ~> inf1I
  5653     meet2I                  ~> inf2I
  5654     meet1D1                 ~> inf1D1
  5655     meet2D1                 ~> inf2D1
  5656     meet1D2                 ~> inf1D2
  5657     meet2D2                 ~> inf2D2
  5658     meet1E                  ~> inf1E
  5659     meet2E                  ~> inf2E
  5660     join1_iff               ~> sup1_iff
  5661     join2_iff               ~> sup2_iff
  5662     join1I1                 ~> sup1I1
  5663     join2I1                 ~> sup2I1
  5664     join1I1                 ~> sup1I1
  5665     join2I2                 ~> sup1I2
  5666     join1CI                 ~> sup1CI
  5667     join2CI                 ~> sup2CI
  5668     join1E                  ~> sup1E
  5669     join2E                  ~> sup2E
  5670 
  5671     is_meet_Meet            ~> is_meet_Inf
  5672     Meet_bool_def           ~> Inf_bool_def
  5673     Meet_fun_def            ~> Inf_fun_def
  5674     Meet_greatest           ~> Inf_greatest
  5675     Meet_lower              ~> Inf_lower
  5676     Meet_set_def            ~> Inf_set_def
  5677 
  5678     Sup_def                 ~> Sup_Inf
  5679     Sup_bool_eq             ~> Sup_bool_def
  5680     Sup_fun_eq              ~> Sup_fun_def
  5681     Sup_set_eq              ~> Sup_set_def
  5682 
  5683     listsp_meetI            ~> listsp_infI
  5684     listsp_meet_eq          ~> listsp_inf_eq
  5685 
  5686     meet_min                ~> inf_min
  5687     join_max                ~> sup_max
  5688 
  5689 * Added syntactic class "size"; overloaded constant "size" now has
  5690 type "'a::size ==> bool"
  5691 
  5692 * Internal reorganisation of `size' of datatypes: size theorems
  5693 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5694 simplification rules by default!); theorems "prod.size" now named
  5695 "*.size".
  5696 
  5697 * Class "div" now inherits from class "times" rather than "type".
  5698 INCOMPATIBILITY.
  5699 
  5700 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5701 Linorder etc.  have disappeared; operations defined in terms of
  5702 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5703 
  5704 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5705 
  5706 * HOL-Word: New extensive library and type for generic, fixed size
  5707 machine words, with arithmetic, bit-wise, shifting and rotating
  5708 operations, reflection into int, nat, and bool lists, automation for
  5709 linear arithmetic (by automatic reflection into nat or int), including
  5710 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5711 arithmetic type classes, supporting automatic simplification of
  5712 numerals on all operations.
  5713 
  5714 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5715 
  5716 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5717 
  5718 * Code generator library theories:
  5719   - Code_Integer represents HOL integers by big integer literals in target
  5720     languages.
  5721   - Code_Char represents HOL characters by character literals in target
  5722     languages.
  5723   - Code_Char_chr like Code_Char, but also offers treatment of character
  5724     codes; includes Code_Integer.
  5725   - Executable_Set allows to generate code for finite sets using lists.
  5726   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5727     denominator).
  5728   - Executable_Real implements a subset of real numbers, namly those
  5729     representable by rational numbers.
  5730   - Efficient_Nat implements natural numbers by integers, which in general will
  5731     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5732     includes Code_Integer.
  5733   - Code_Index provides an additional datatype index which is mapped to
  5734     target-language built-in integers.
  5735   - Code_Message provides an additional datatype message_string which is isomorphic to
  5736     strings; messages are mapped to target-language strings.
  5737 
  5738 * New package for inductive predicates
  5739 
  5740   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5741 
  5742     inductive
  5743       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5744       for z_1 :: U_1 and ... and z_n :: U_m
  5745     where
  5746       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5747     | ...
  5748 
  5749   with full support for type-inference, rather than
  5750 
  5751     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5752 
  5753     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5754     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5755 
  5756     inductive "s z_1 ... z_m"
  5757     intros
  5758       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5759       ...
  5760 
  5761   For backward compatibility, there is a wrapper allowing inductive
  5762   sets to be defined with the new package via
  5763 
  5764     inductive_set
  5765       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5766       for z_1 :: U_1 and ... and z_n :: U_m
  5767     where
  5768       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5769     | ...
  5770 
  5771   or
  5772 
  5773     inductive_set
  5774       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5775       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5776       for z_1 :: U_1 and ... and z_n :: U_m
  5777     where
  5778       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5779     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5780     | ...
  5781 
  5782   if the additional syntax "p ..." is required.
  5783 
  5784   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5785   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5786 
  5787   INCOMPATIBILITIES:
  5788 
  5789   - Since declaration and definition of inductive sets or predicates
  5790     is no longer separated, abbreviations involving the newly
  5791     introduced sets or predicates must be specified together with the
  5792     introduction rules after the 'where' keyword (see above), rather
  5793     than before the actual inductive definition.
  5794 
  5795   - The variables in induction and elimination rules are now
  5796     quantified in the order of their occurrence in the introduction
  5797     rules, rather than in alphabetical order. Since this may break
  5798     some proofs, these proofs either have to be repaired, e.g. by
  5799     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5800     statements of the form
  5801 
  5802       case (rule_i a_i_1 ... a_i_{k_i})
  5803 
  5804     or the old order of quantification has to be restored by explicitly adding
  5805     meta-level quantifiers in the introduction rules, i.e.
  5806 
  5807       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5808 
  5809   - The format of the elimination rules is now
  5810 
  5811       p z_1 ... z_m x_1 ... x_n ==>
  5812         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5813         ==> ... ==> P
  5814 
  5815     for predicates and
  5816 
  5817       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5818         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5819         ==> ... ==> P
  5820 
  5821     for sets rather than
  5822 
  5823       x : s z_1 ... z_m ==>
  5824         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5825         ==> ... ==> P
  5826 
  5827     This may require terms in goals to be expanded to n-tuples
  5828     (e.g. using case_tac or simplification with the split_paired_all
  5829     rule) before the above elimination rule is applicable.
  5830 
  5831   - The elimination or case analysis rules for (mutually) inductive
  5832     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5833     list of rules "p_1_..._p_k.elims" is no longer available.
  5834 
  5835 * New package "function"/"fun" for general recursive functions,
  5836 supporting mutual and nested recursion, definitions in local contexts,
  5837 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5838 for small examples, and the separate tutorial on the function
  5839 package. The old recdef "package" is still available as before, but
  5840 users are encouraged to use the new package.
  5841 
  5842 * Method "lexicographic_order" automatically synthesizes termination
  5843 relations as lexicographic combinations of size measures.
  5844 
  5845 * Case-expressions allow arbitrary constructor-patterns (including
  5846 "_") and take their order into account, like in functional
  5847 programming.  Internally, this is translated into nested
  5848 case-expressions; missing cases are added and mapped to the predefined
  5849 constant "undefined". In complicated cases printing may no longer show
  5850 the original input but the internal form. Lambda-abstractions allow
  5851 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5852 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5853 variable.
  5854 
  5855 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5856 is an abbreviation for "of_nat :: nat => int". The simplification
  5857 rules for "of_nat" have been changed to work like "int" did
  5858 previously.  Potential INCOMPATIBILITY:
  5859   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5860   - of_nat_diff and of_nat_mult are no longer default simp rules
  5861 
  5862 * Method "algebra" solves polynomial equations over (semi)rings using
  5863 Groebner bases. The (semi)ring structure is defined by locales and the
  5864 tool setup depends on that generic context. Installing the method for
  5865 a specific type involves instantiating the locale and possibly adding
  5866 declarations for computation on the coefficients.  The method is
  5867 already instantiated for natural numbers and for the axiomatic class
  5868 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5869 CALCULEMUS 2007 for the general principles underlying this
  5870 architecture of context-aware proof-tools.
  5871 
  5872 * Method "ferrack" implements quantifier elimination over
  5873 special-purpose dense linear orders using locales (analogous to
  5874 "algebra"). The method is already installed for class
  5875 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5876 rat, etc.
  5877 
  5878 * Former constant "List.op @" now named "List.append".  Use ML
  5879 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5880 circumvent possible incompatibilities when working on ML level.
  5881 
  5882 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5883 
  5884 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5885 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5886 
  5887 * New syntax for Haskell-like list comprehension (input only), eg.
  5888 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5889 
  5890 * The special syntax for function "filter" has changed from [x :
  5891 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5892 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5893 
  5894 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5895 defined on nat only, as an abbreviation for [a..<Suc b]
  5896 INCOMPATIBILITY.
  5897 
  5898 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5899 
  5900 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5901 
  5902 * New lemma collection field_simps (an extension of ring_simps) for
  5903 manipulating (in)equations involving division. Multiplies with all
  5904 denominators that can be proved to be non-zero (in equations) or
  5905 positive/negative (in inequations).
  5906 
  5907 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5908 have been improved and renamed to ring_simps, group_simps and
  5909 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5910 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5911 
  5912 * Theory Library/Commutative_Ring: switched from recdef to function
  5913 package; constants add, mul, pow now curried.  Infix syntax for
  5914 algebraic operations.
  5915 
  5916 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5917 INCOMPATIBILITY.
  5918 
  5919 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5920 INCOMPATIBILITY.
  5921 
  5922 * HOL/records: generalised field-update to take a function on the
  5923 field rather than the new value: r(|A := x|) is translated to A_update
  5924 (K x) r The K-combinator that is internally used is called K_record.
  5925 INCOMPATIBILITY: Usage of the plain update functions has to be
  5926 adapted.
  5927 
  5928 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5929 * x = 0, which are required for a semiring.  Richer structures do not
  5930 inherit from semiring_0 anymore, because this property is a theorem
  5931 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5932 there is more to prove, but this is mostly trivial.
  5933 
  5934 * Class "recpower" is generalized to arbitrary monoids, not just
  5935 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5936 commutativity or semiring properties additionally.
  5937 
  5938 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5939 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5940 better use 'abbreviation'.
  5941 
  5942 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5943 
  5944 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5945 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5946 significant changes for setting up numeral syntax for types:
  5947   - New constants Numeral.pred and Numeral.succ instead
  5948       of former Numeral.bin_pred and Numeral.bin_succ.
  5949   - Use integer operations instead of bin_add, bin_mult and so on.
  5950   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5951   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5952 
  5953 See src/HOL/Integ/IntArith.thy for an example setup.
  5954 
  5955 * Command 'normal_form' computes the normal form of a term that may
  5956 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5957 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5958 heavy-duty computations because the functions are compiled to ML
  5959 first.  Correspondingly, a method "normalization" is provided.  See
  5960 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5961 
  5962 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5963 25 like -->); output depends on the "iff" print_mode, the default is
  5964 "A = B" (with priority 50).
  5965 
  5966 * Relations less (<) and less_eq (<=) are also available on type bool.
  5967 Modified syntax to disallow nesting without explicit parentheses,
  5968 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5969 INCOMPATIBILITY.
  5970 
  5971 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5972 
  5973 * Relation composition operator "op O" now has precedence 75 and binds
  5974 stronger than union and intersection. INCOMPATIBILITY.
  5975 
  5976 * The old set interval syntax "{m..n(}" (and relatives) has been
  5977 removed.  Use "{m..<n}" (and relatives) instead.
  5978 
  5979 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5980 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5981 ``declare [[simproc del: neq]]''.
  5982 
  5983 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5984 True/False.
  5985 
  5986 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5987 
  5988 * Theorem setsum_mult renamed to setsum_right_distrib.
  5989 
  5990 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5991 ``rule'' method.
  5992 
  5993 * Reimplemented methods "sat" and "satx", with several improvements:
  5994 goals no longer need to be stated as "<prems> ==> False", equivalences
  5995 (i.e. "=" on type bool) are handled, variable names of the form
  5996 "lit_<n>" are no longer reserved, significant speedup.
  5997 
  5998 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5999 zChaff is still supported as well.
  6000 
  6001 * 'inductive' and 'datatype': provide projections of mutual rules,
  6002 bundled as foo_bar.inducts;
  6003 
  6004 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  6005 Library.
  6006 
  6007 * Library: moved theory Accessible_Part to main HOL.
  6008 
  6009 * Library: added theory Coinductive_List of potentially infinite lists
  6010 as greatest fixed-point.
  6011 
  6012 * Library: added theory AssocList which implements (finite) maps as
  6013 association lists.
  6014 
  6015 * Method "evaluation" solves goals (i.e. a boolean expression)
  6016 efficiently by compiling it to ML.  The goal is "proved" (via an
  6017 oracle) if it evaluates to True.
  6018 
  6019 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  6020 also when invoked by the simplifier.  This results in the Simplifier
  6021 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  6022 Configuration option fast_arith_split_limit=0 recovers the old
  6023 behavior.
  6024 
  6025 * Support for hex (0x20) and binary (0b1001) numerals.
  6026 
  6027 * New method: reify eqs (t), where eqs are equations for an
  6028 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  6029 parameter, computes a term s::'b and a list xs::'a list and proves the
  6030 theorem I xs s = t. This is also known as reification or quoting. The
  6031 resulting theorem is applied to the subgoal to substitute t with I xs
  6032 s.  If t is omitted, the subgoal itself is reified.
  6033 
  6034 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  6035 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  6036 where f is supposed to be a computable function (in the sense of code
  6037 generattion). The method uses reify to compute s and xs as above then
  6038 applies corr_thm and uses normalization by evaluation to "prove" f s =
  6039 r and finally gets the theorem t = r, which is again applied to the
  6040 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  6041 
  6042 * Reflection: Automatic reification now handels binding, an example is
  6043 available in src/HOL/ex/ReflectionEx.thy
  6044 
  6045 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  6046 command 'statespace' that is similar to 'record', but introduces an
  6047 abstract specification based on the locale infrastructure instead of
  6048 HOL types.  This leads to extra flexibility in composing state spaces,
  6049 in particular multiple inheritance and renaming of components.
  6050 
  6051 
  6052 *** HOL-Complex ***
  6053 
  6054 * Hyperreal: Functions root and sqrt are now defined on negative real
  6055 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  6056 Nonnegativity side conditions have been removed from many lemmas, so
  6057 that more subgoals may now be solved by simplification; potential
  6058 INCOMPATIBILITY.
  6059 
  6060 * Real: new type classes formalize real normed vector spaces and
  6061 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  6062 and norm :: 'a => real.
  6063 
  6064 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  6065 reals into other types. The overloaded constant Reals :: 'a set is now
  6066 defined as range of_real; potential INCOMPATIBILITY.
  6067 
  6068 * Real: proper support for ML code generation, including 'quickcheck'.
  6069 Reals are implemented as arbitrary precision rationals.
  6070 
  6071 * Hyperreal: Several constants that previously worked only for the
  6072 reals have been generalized, so they now work over arbitrary vector
  6073 spaces. Type annotations may need to be added in some cases; potential
  6074 INCOMPATIBILITY.
  6075 
  6076   Infinitesimal  :: ('a::real_normed_vector) star set
  6077   HFinite        :: ('a::real_normed_vector) star set
  6078   HInfinite      :: ('a::real_normed_vector) star set
  6079   approx         :: ('a::real_normed_vector) star => 'a star => bool
  6080   monad          :: ('a::real_normed_vector) star => 'a star set
  6081   galaxy         :: ('a::real_normed_vector) star => 'a star set
  6082   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  6083   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  6084   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  6085   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  6086   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  6087   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  6088   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6089   sgn            :: 'a::real_normed_vector => 'a
  6090   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6091 
  6092 * Complex: Some complex-specific constants are now abbreviations for
  6093 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6094 hnorm.  Other constants have been entirely removed in favor of the
  6095 polymorphic versions (INCOMPATIBILITY):
  6096 
  6097   approx        <-- capprox
  6098   HFinite       <-- CFinite
  6099   HInfinite     <-- CInfinite
  6100   Infinitesimal <-- CInfinitesimal
  6101   monad         <-- cmonad
  6102   galaxy        <-- cgalaxy
  6103   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  6104   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  6105   (ns)deriv     <-- (ns)cderiv
  6106 
  6107 
  6108 *** HOL-Algebra ***
  6109 
  6110 * Formalisation of ideals and the quotient construction over rings.
  6111 
  6112 * Order and lattice theory no longer based on records.
  6113 INCOMPATIBILITY.
  6114 
  6115 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  6116 greatest_closed.  INCOMPATIBILITY.
  6117 
  6118 * Method algebra is now set up via an attribute.  For examples see
  6119 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  6120 of algebraic structures.
  6121 
  6122 * Renamed theory CRing to Ring.
  6123 
  6124 
  6125 *** HOL-Nominal ***
  6126 
  6127 * Substantial, yet incomplete support for nominal datatypes (binding
  6128 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  6129 src/HOL/Nominal/Examples.  Prospective users should consult
  6130 http://isabelle.in.tum.de/nominal/
  6131 
  6132 
  6133 *** ML ***
  6134 
  6135 * ML basics: just one true type int, which coincides with IntInf.int
  6136 (even on SML/NJ).
  6137 
  6138 * ML within Isar: antiquotations allow to embed statically-checked
  6139 formal entities in the source, referring to the context available at
  6140 compile-time.  For example:
  6141 
  6142 ML {* @{sort "{zero,one}"} *}
  6143 ML {* @{typ "'a => 'b"} *}
  6144 ML {* @{term "%x. x"} *}
  6145 ML {* @{prop "x == y"} *}
  6146 ML {* @{ctyp "'a => 'b"} *}
  6147 ML {* @{cterm "%x. x"} *}
  6148 ML {* @{cprop "x == y"} *}
  6149 ML {* @{thm asm_rl} *}
  6150 ML {* @{thms asm_rl} *}
  6151 ML {* @{type_name c} *}
  6152 ML {* @{type_syntax c} *}
  6153 ML {* @{const_name c} *}
  6154 ML {* @{const_syntax c} *}
  6155 ML {* @{context} *}
  6156 ML {* @{theory} *}
  6157 ML {* @{theory Pure} *}
  6158 ML {* @{theory_ref} *}
  6159 ML {* @{theory_ref Pure} *}
  6160 ML {* @{simpset} *}
  6161 ML {* @{claset} *}
  6162 ML {* @{clasimpset} *}
  6163 
  6164 The same works for sources being ``used'' within an Isar context.
  6165 
  6166 * ML in Isar: improved error reporting; extra verbosity with
  6167 ML_Context.trace enabled.
  6168 
  6169 * Pure/General/table.ML: the join operations now works via exceptions
  6170 DUP/SAME instead of type option. This is simpler in simple cases, and
  6171 admits slightly more efficient complex applications.
  6172 
  6173 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6174 use Context.generic instead of just theory.
  6175 
  6176 * Pure: datatype Context.generic joins theory/Proof.context and
  6177 provides some facilities for code that works in either kind of
  6178 context, notably GenericDataFun for uniform theory and proof data.
  6179 
  6180 * Pure: simplified internal attribute type, which is now always
  6181 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6182 local (Proof.context) attributes have been discontinued, while
  6183 minimizing code duplication. Thm.rule_attribute and
  6184 Thm.declaration_attribute build canonical attributes; see also structure
  6185 Context for further operations on Context.generic, notably
  6186 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6187 declarations and definitions.
  6188 
  6189 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6190 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6191 init only required for impure data. INCOMPATIBILITY: empty really need
  6192 to be empty (no dependencies on theory content!)
  6193 
  6194 * Pure/kernel: consts certification ignores sort constraints given in
  6195 signature declarations. (This information is not relevant to the
  6196 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6197 potential INCOMPATIBILITY.
  6198 
  6199 * Pure: axiomatic type classes are now purely definitional, with
  6200 explicit proofs of class axioms and super class relations performed
  6201 internally. See Pure/axclass.ML for the main internal interfaces --
  6202 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6203 AxClass.axiomatize_class/classrel/arity supersede
  6204 Sign.add_classes/classrel/arities.
  6205 
  6206 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6207 global/local versions on theory vs. Proof.context have been
  6208 discontinued; Attrib.syntax and Method.syntax have been adapted
  6209 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6210 attributes, methods, etc.
  6211 
  6212 * Pure: several functions of signature "... -> theory -> theory * ..."
  6213 have been reoriented to "... -> theory -> ... * theory" in order to
  6214 allow natural usage in combination with the ||>, ||>>, |-> and
  6215 fold_map combinators.
  6216 
  6217 * Pure: official theorem names (closed derivations) and additional
  6218 comments (tags) are now strictly separate.  Name hints -- which are
  6219 maintained as tags -- may be attached any time without affecting the
  6220 derivation.
  6221 
  6222 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6223 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6224 achieve the old behaviour.
  6225 
  6226 * Pure: the "Goal" constant is now called "prop", supporting a
  6227 slightly more general idea of ``protecting'' meta-level rule
  6228 statements.
  6229 
  6230 * Pure: Logic.(un)varify only works in a global context, which is now
  6231 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6232 Logic.legacy_(un)varify as temporary workaround.
  6233 
  6234 * Pure: structure Name provides scalable operations for generating
  6235 internal variable names, notably Name.variants etc.  This replaces
  6236 some popular functions from term.ML:
  6237 
  6238   Term.variant		->  Name.variant
  6239   Term.variantlist	->  Name.variant_list