NEWS
author wenzelm
Tue Sep 25 22:36:06 2012 +0200 (2012-09-25 ago)
changeset 49566 66cbf8bb4693
parent 49532 6f7cc8e42716
child 49647 21ae8500d261
permissions -rw-r--r--
basic integration of graphview into document model;
added Graph_Dockable;
updated Isabelle/jEdit authors and dependencies etc.;
     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     . more plugin options and preferences, based on Isabelle/Scala;
    16     . uniform Java 7 platform on Linux, Mac OS X, Windows;
    17 
    18 * Command 'ML_file' evaluates ML text from a file directly within the
    19 theory, without any predeclaration via 'uses' in the theory header.
    20 
    21 * Old command 'use' command and corresponding keyword 'uses' in the
    22 theory header are legacy features and will be discontinued soon.
    23 Tools that load their additional source files may imitate the
    24 'ML_file' implementation, such that the system can take care of
    25 dependencies properly.
    26 
    27 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    28 is called fastforce / fast_force_tac already since Isabelle2011-1.
    29 
    30 * Updated and extended "isar-ref" manual, reduced remaining material
    31 in old "ref" manual.
    32 
    33 
    34 *** Pure ***
    35 
    36 * Code generation for Haskell: restrict unqualified imports from
    37 Haskell Prelude to a small set of fundamental operations.
    38 
    39 * Command "export_code": relative file names are interpreted
    40 relatively to master directory of current theory rather than
    41 the rather arbitrary current working directory.
    42 INCOMPATIBILITY.
    43 
    44 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    45 use regular rule composition via "OF" / "THEN", or explicit proof
    46 structure instead.  Note that Isabelle/ML provides a variety of
    47 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    48 with some care where this is really required.
    49 
    50 * Command 'typ' supports an additional variant with explicit sort
    51 constraint, to infer and check the most general type conforming to a
    52 given given sort.  Example (in HOL):
    53 
    54   typ "_ * _ * bool * unit" :: finite
    55 
    56 
    57 *** HOL ***
    58 
    59 * Class "comm_monoid_diff" formalised properties of bounded
    60 subtraction, with natural numbers and multisets as typical instances.
    61 
    62 * Theory "Library/Option_ord" provides instantiation of option type
    63 to lattice type classes.
    64 
    65 * New combinator "Option.these" with type "'a option set => 'a option".
    66 
    67 * Renamed theory Library/List_Prefix to Library/Sublist.
    68 INCOMPATIBILITY.  Related changes are:
    69 
    70   - Renamed constants:
    71 
    72       prefix ~> prefixeq
    73       strict_prefix ~> prefix
    74 
    75     Renamed lemmas accordingly, INCOMPATIBILITY.
    76 
    77   - Replaced constant "postfix" by "suffixeq" with swapped argument order
    78     (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix
    79     syntax "xs >>= ys"; use "suffixeq ys xs" instead.  Renamed lemmas
    80     accordingly.  INCOMPATIBILITY.
    81 
    82   - New constant "emb" for homeomorphic embedding on lists. New
    83     abbreviation "sub" for special case "emb (op =)".
    84 
    85   - Library/Sublist does no longer provide "order" and "bot" type class
    86     instances for the prefix order (merely corresponding locale
    87     interpretations). The type class instances are to be found in
    88     Library/Prefix_Order. INCOMPATIBILITY.
    89 
    90   - The sublist relation from Library/Sublist_Order is now based on
    91     "Sublist.sub". Replaced lemmas:
    92 
    93       le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff
    94       le_list_append_mono ~> Sublist.emb_append_mono
    95       le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2
    96       le_list_Cons_EX ~> Sublist.emb_ConsD
    97       le_list_drop_Cons2 ~> Sublist.sub_Cons2'
    98       le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq
    99       le_list_drop_Cons ~> Sublist.sub_Cons'
   100       le_list_drop_many ~> Sublist.sub_drop_many
   101       le_list_filter_left ~> Sublist.sub_filter_left
   102       le_list_rev_drop_many ~> Sublist.sub_rev_drop_many
   103       le_list_rev_take_iff ~> Sublist.sub_append
   104       le_list_same_length ~> Sublist.sub_same_length
   105       le_list_take_many_iff ~> Sublist.sub_append'
   106       less_eq_list.drop ~> less_eq_list_drop
   107       less_eq_list.induct ~> less_eq_list_induct
   108       not_le_list_length ~> Sublist.not_sub_length
   109 
   110     INCOMPATIBILITY.
   111 
   112 * HOL/BNF: New (co)datatype package based on bounded natural
   113 functors with support for mixed, nested recursion and interesting
   114 non-free datatypes.
   115 
   116 * HOL/Cardinals: Theories of ordinals and cardinals
   117 (supersedes the AFP entry "Ordinals_and_Cardinals").
   118 
   119 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
   120 execution for code generated towards Isabelle/ML.
   121 
   122 * Library/FinFun.thy: theory of almost everywhere constant functions
   123 (supersedes the AFP entry "Code Generation for Functions as Data").
   124 
   125 * Library/Phantom.thy: generic phantom type to make a type parameter
   126 appear in a constant's type. This alternative to adding TYPE('a) as
   127 another parameter avoids unnecessary closures in generated code.
   128 
   129 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   130 expressions.
   131 
   132 * Quickcheck:
   133 
   134   - added an optimisation for equality premises.
   135     It is switched on by default, and can be switched off by setting
   136     the configuration quickcheck_optimise_equality to false.
   137 
   138 * The SMT solver Z3 has now by default a restricted set of directly
   139 supported features. For the full set of features (div/mod, nonlinear
   140 arithmetic, datatypes/records) with potential proof reconstruction
   141 failures, enable the configuration option "z3_with_extensions".
   142 Minor INCOMPATIBILITY.
   143 
   144 * Sledgehammer:
   145 
   146   - Added MaSh relevance filter based on machine-learning; see the
   147     Sledgehammer manual for details.
   148   - Rationalized type encodings ("type_enc" option).
   149   - Renamed "kill_provers" subcommand to "kill"
   150   - Renamed options:
   151       max_relevant ~> max_facts
   152       relevance_thresholds ~> fact_thresholds
   153 
   154 
   155 *** Document preparation ***
   156 
   157 * Default for \<euro> is now based on eurosym package, instead of
   158 slightly exotic babel/greek.
   159 
   160 * Document variant NAME may use different LaTeX entry point
   161 document/root_NAME.tex if that file exists, instead of the common
   162 document/root.tex.
   163 
   164 * Simplified custom document/build script, instead of old-style
   165 document/IsaMakefile.  Minor INCOMPATIBILITY.
   166 
   167 
   168 *** ML ***
   169 
   170 * Renamed Position.str_of to Position.here to emphasize that this is a
   171 formal device to inline positions into message text, but not
   172 necessarily printing visible text.
   173 
   174 
   175 *** System ***
   176 
   177 * The ML system is configured as regular component, and no longer
   178 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   179 for home-made configurations.
   180 
   181 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   182 Minor INCOMPATIBILITY in command-line options.
   183 
   184 * Advanced support for Isabelle sessions and build management, see
   185 "system" manual for the chapter of that name, especially the "isabelle
   186 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   187 mkdir / make are rendered obsolete.
   188 
   189 * Discontinued obsolete "isabelle makeall".
   190 
   191 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   192 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   193 provides some traditional targets that invoke "isabelle build".  Note
   194 that this is inefficient!  Applications of Isabelle/HOL involving
   195 "isabelle make" should be upgraded to use "isabelle build" directly.
   196 
   197 * Discontinued obsolete Isabelle/build script, it is superseded by the
   198 regular isabelle build tool.  For example:
   199 
   200   isabelle build -s -b HOLCF
   201 
   202 * The "isabelle options" tool prints Isabelle system options, as
   203 required for "isabelle build", for example.
   204 
   205 * The "isabelle mkroot" tool prepares session root directories for use
   206 with "isabelle build", similar to former "isabelle mkdir" for
   207 "isabelle usedir".
   208 
   209 * The "isabelle components" tool helps to resolve add-on components
   210 that are not bundled, or referenced from a bare-bones repository
   211 version of Isabelle.
   212 
   213 * Discontinued support for Poly/ML 5.2.1, which was the last version
   214 without exception positions and advanced ML compiler/toplevel
   215 configuration.
   216 
   217 * Discontinued special treatment of Proof General -- no longer guess
   218 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   219 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   220 settings manually, or use a Proof General version that has been
   221 bundled as Isabelle component.
   222 
   223 
   224 New in Isabelle2012 (May 2012)
   225 ------------------------------
   226 
   227 *** General ***
   228 
   229 * Prover IDE (PIDE) improvements:
   230 
   231   - more robust Sledgehammer integration (as before the sledgehammer
   232     command-line needs to be typed into the source buffer)
   233   - markup for bound variables
   234   - markup for types of term variables (displayed as tooltips)
   235   - support for user-defined Isar commands within the running session
   236   - improved support for Unicode outside original 16bit range
   237     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   238 
   239 * Forward declaration of outer syntax keywords within the theory
   240 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   241 commands to be used in the same theory where defined.
   242 
   243 * Auxiliary contexts indicate block structure for specifications with
   244 additional parameters and assumptions.  Such unnamed contexts may be
   245 nested within other targets, like 'theory', 'locale', 'class',
   246 'instantiation' etc.  Results from the local context are generalized
   247 accordingly and applied to the enclosing target context.  Example:
   248 
   249   context
   250     fixes x y z :: 'a
   251     assumes xy: "x = y" and yz: "y = z"
   252   begin
   253 
   254   lemma my_trans: "x = z" using xy yz by simp
   255 
   256   end
   257 
   258   thm my_trans
   259 
   260 The most basic application is to factor-out context elements of
   261 several fixes/assumes/shows theorem statements, e.g. see
   262 ~~/src/HOL/Isar_Examples/Group_Context.thy
   263 
   264 Any other local theory specification element works within the "context
   265 ... begin ... end" block as well.
   266 
   267 * Bundled declarations associate attributed fact expressions with a
   268 given name in the context.  These may be later included in other
   269 contexts.  This allows to manage context extensions casually, without
   270 the logical dependencies of locales and locale interpretation.  See
   271 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   272 
   273 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   274 declaration, and results are standardized before being stored.  Thus
   275 old-style "standard" after instantiation or composition of facts
   276 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   277 indices of schematic variables.
   278 
   279 * Rule attributes in local theory declarations (e.g. locale or class)
   280 are now statically evaluated: the resulting theorem is stored instead
   281 of the original expression.  INCOMPATIBILITY in rare situations, where
   282 the historic accident of dynamic re-evaluation in interpretations
   283 etc. was exploited.
   284 
   285 * New tutorial "Programming and Proving in Isabelle/HOL"
   286 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   287 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   288 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   289 Logic" as the recommended beginners tutorial, but does not cover all
   290 of the material of that old tutorial.
   291 
   292 * Updated and extended reference manuals: "isar-ref",
   293 "implementation", "system"; reduced remaining material in old "ref"
   294 manual.
   295 
   296 
   297 *** Pure ***
   298 
   299 * Command 'definition' no longer exports the foundational "raw_def"
   300 into the user context.  Minor INCOMPATIBILITY, may use the regular
   301 "def" result with attribute "abs_def" to imitate the old version.
   302 
   303 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   304 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   305 expand it.  This also works for object-logic equality.  (Formerly
   306 undocumented feature.)
   307 
   308 * Sort constraints are now propagated in simultaneous statements, just
   309 like type constraints.  INCOMPATIBILITY in rare situations, where
   310 distinct sorts used to be assigned accidentally.  For example:
   311 
   312   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   313 
   314   lemma "P (x::'a)" and "Q (y::'a::bar)"
   315     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   316 
   317 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   318 tolerant against multiple unifiers, as long as the final result is
   319 unique.  (As before, rules are composed in canonical right-to-left
   320 order to accommodate newly introduced premises.)
   321 
   322 * Renamed some inner syntax categories:
   323 
   324     num ~> num_token
   325     xnum ~> xnum_token
   326     xstr ~> str_token
   327 
   328 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   329 "num_position" etc. are mainly used instead (which also include
   330 position information via constraints).
   331 
   332 * Simplified configuration options for syntax ambiguity: see
   333 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   334 manual.  Minor INCOMPATIBILITY.
   335 
   336 * Discontinued configuration option "syntax_positions": atomic terms
   337 in parse trees are always annotated by position constraints.
   338 
   339 * Old code generator for SML and its commands 'code_module',
   340 'code_library', 'consts_code', 'types_code' have been discontinued.
   341 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   342 
   343 * Redundant attribute "code_inline" has been discontinued. Use
   344 "code_unfold" instead.  INCOMPATIBILITY.
   345 
   346 * Dropped attribute "code_unfold_post" in favor of the its dual
   347 "code_abbrev", which yields a common pattern in definitions like
   348 
   349   definition [code_abbrev]: "f = t"
   350 
   351 INCOMPATIBILITY.
   352 
   353 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   354 instead.  INCOMPATIBILITY.
   355 
   356 * Discontinued old "prems" fact, which used to refer to the accidental
   357 collection of foundational premises in the context (already marked as
   358 legacy since Isabelle2011).
   359 
   360 
   361 *** HOL ***
   362 
   363 * Type 'a set is now a proper type constructor (just as before
   364 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   365 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   366 sets separate, it is often sufficient to rephrase some set S that has
   367 been accidentally used as predicates by "%x. x : S", and some
   368 predicate P that has been accidentally used as set by "{x. P x}".
   369 Corresponding proofs in a first step should be pruned from any
   370 tinkering with former theorems mem_def and Collect_def as far as
   371 possible.
   372 
   373 For developments which deliberately mix predicates and sets, a
   374 planning step is necessary to determine what should become a predicate
   375 and what a set.  It can be helpful to carry out that step in
   376 Isabelle2011-1 before jumping right into the current release.
   377 
   378 * Code generation by default implements sets as container type rather
   379 than predicates.  INCOMPATIBILITY.
   380 
   381 * New type synonym 'a rel = ('a * 'a) set
   382 
   383 * The representation of numerals has changed.  Datatype "num"
   384 represents strictly positive binary numerals, along with functions
   385 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   386 positive and negated numeric literals, respectively.  See also
   387 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   388 user theories may require adaptations as follows:
   389 
   390   - Theorems with number_ring or number_semiring constraints: These
   391     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   392 
   393   - Theories defining numeric types: Remove number, number_semiring,
   394     and number_ring instances. Defer all theorems about numerals until
   395     after classes one and semigroup_add have been instantiated.
   396 
   397   - Numeral-only simp rules: Replace each rule having a "number_of v"
   398     pattern with two copies, one for numeral and one for neg_numeral.
   399 
   400   - Theorems about subclasses of semiring_1 or ring_1: These classes
   401     automatically support numerals now, so more simp rules and
   402     simprocs may now apply within the proof.
   403 
   404   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   405     Redefine using other integer operations.
   406 
   407 * Transfer: New package intended to generalize the existing
   408 "descending" method and related theorem attributes from the Quotient
   409 package.  (Not all functionality is implemented yet, but future
   410 development will focus on Transfer as an eventual replacement for the
   411 corresponding parts of the Quotient package.)
   412 
   413   - transfer_rule attribute: Maintains a collection of transfer rules,
   414     which relate constants at two different types. Transfer rules may
   415     relate different type instances of the same polymorphic constant,
   416     or they may relate an operation on a raw type to a corresponding
   417     operation on an abstract type (quotient or subtype). For example:
   418 
   419     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   420     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   421 
   422   - transfer method: Replaces a subgoal on abstract types with an
   423     equivalent subgoal on the corresponding raw types. Constants are
   424     replaced with corresponding ones according to the transfer rules.
   425     Goals are generalized over all free variables by default; this is
   426     necessary for variables whose types change, but can be overridden
   427     for specific variables with e.g. "transfer fixing: x y z".  The
   428     variant transfer' method allows replacing a subgoal with one that
   429     is logically stronger (rather than equivalent).
   430 
   431   - relator_eq attribute: Collects identity laws for relators of
   432     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   433     transfer method uses these lemmas to infer transfer rules for
   434     non-polymorphic constants on the fly.
   435 
   436   - transfer_prover method: Assists with proving a transfer rule for a
   437     new constant, provided the constant is defined in terms of other
   438     constants that already have transfer rules. It should be applied
   439     after unfolding the constant definitions.
   440 
   441   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   442     from type nat to type int.
   443 
   444 * Lifting: New package intended to generalize the quotient_definition
   445 facility of the Quotient package; designed to work with Transfer.
   446 
   447   - lift_definition command: Defines operations on an abstract type in
   448     terms of a corresponding operation on a representation
   449     type.  Example syntax:
   450 
   451     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   452       is List.insert
   453 
   454     Users must discharge a respectfulness proof obligation when each
   455     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   456     the proof is discharged automatically.) The obligation is
   457     presented in a user-friendly, readable form; a respectfulness
   458     theorem in the standard format and a transfer rule are generated
   459     by the package.
   460 
   461   - Integration with code_abstype: For typedefs (e.g. subtypes
   462     corresponding to a datatype invariant, such as dlist),
   463     lift_definition generates a code certificate theorem and sets up
   464     code generation for each constant.
   465 
   466   - setup_lifting command: Sets up the Lifting package to work with a
   467     user-defined type. The user must provide either a quotient theorem
   468     or a type_definition theorem.  The package configures transfer
   469     rules for equality and quantifiers on the type, and sets up the
   470     lift_definition command to work with the type.
   471 
   472   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   473     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   474     Word/Word.thy and Library/Float.thy.
   475 
   476 * Quotient package:
   477 
   478   - The 'quotient_type' command now supports a 'morphisms' option with
   479     rep and abs functions, similar to typedef.
   480 
   481   - 'quotient_type' sets up new types to work with the Lifting and
   482     Transfer packages, as with 'setup_lifting'.
   483 
   484   - The 'quotient_definition' command now requires the user to prove a
   485     respectfulness property at the point where the constant is
   486     defined, similar to lift_definition; INCOMPATIBILITY.
   487 
   488   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   489     accordingly, INCOMPATIBILITY.
   490 
   491 * New diagnostic command 'find_unused_assms' to find potentially
   492 superfluous assumptions in theorems using Quickcheck.
   493 
   494 * Quickcheck:
   495 
   496   - Quickcheck returns variable assignments as counterexamples, which
   497     allows to reveal the underspecification of functions under test.
   498     For example, refuting "hd xs = x", it presents the variable
   499     assignment xs = [] and x = a1 as a counterexample, assuming that
   500     any property is false whenever "hd []" occurs in it.
   501 
   502     These counterexample are marked as potentially spurious, as
   503     Quickcheck also returns "xs = []" as a counterexample to the
   504     obvious theorem "hd xs = hd xs".
   505 
   506     After finding a potentially spurious counterexample, Quickcheck
   507     continues searching for genuine ones.
   508 
   509     By default, Quickcheck shows potentially spurious and genuine
   510     counterexamples. The option "genuine_only" sets quickcheck to only
   511     show genuine counterexamples.
   512 
   513   - The command 'quickcheck_generator' creates random and exhaustive
   514     value generators for a given type and operations.
   515 
   516     It generates values by using the operations as if they were
   517     constructors of that type.
   518 
   519   - Support for multisets.
   520 
   521   - Added "use_subtype" options.
   522 
   523   - Added "quickcheck_locale" configuration to specify how to process
   524     conjectures in a locale context.
   525 
   526 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   527 and affecting 'rat' and 'real'.
   528 
   529 * Sledgehammer:
   530   - Integrated more tightly with SPASS, as described in the ITP 2012
   531     paper "More SPASS with Isabelle".
   532   - Made it try "smt" as a fallback if "metis" fails or times out.
   533   - Added support for the following provers: Alt-Ergo (via Why3 and
   534     TFF1), iProver, iProver-Eq.
   535   - Sped up the minimizer.
   536   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   537   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   538   - Renamed "sound" option to "strict".
   539 
   540 * Metis: Added possibility to specify lambda translations scheme as a
   541 parenthesized argument (e.g., "by (metis (lifting) ...)").
   542 
   543 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   544 
   545 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   546 
   547 * New "case_product" attribute to generate a case rule doing multiple
   548 case distinctions at the same time.  E.g.
   549 
   550   list.exhaust [case_product nat.exhaust]
   551 
   552 produces a rule which can be used to perform case distinction on both
   553 a list and a nat.
   554 
   555 * New "eventually_elim" method as a generalized variant of the
   556 eventually_elim* rules.  Supports structured proofs.
   557 
   558 * Typedef with implicit set definition is considered legacy.  Use
   559 "typedef (open)" form instead, which will eventually become the
   560 default.
   561 
   562 * Record: code generation can be switched off manually with
   563 
   564   declare [[record_coden = false]]  -- "default true"
   565 
   566 * Datatype: type parameters allow explicit sort constraints.
   567 
   568 * Concrete syntax for case expressions includes constraints for source
   569 positions, and thus produces Prover IDE markup for its bindings.
   570 INCOMPATIBILITY for old-style syntax translations that augment the
   571 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   572 one_case.
   573 
   574 * Clarified attribute "mono_set": pure declaration without modifying
   575 the result of the fact expression.
   576 
   577 * More default pred/set conversions on a couple of relation operations
   578 and predicates.  Added powers of predicate relations.  Consolidation
   579 of some relation theorems:
   580 
   581   converse_def ~> converse_unfold
   582   rel_comp_def ~> relcomp_unfold
   583   symp_def ~> (modified, use symp_def and sym_def instead)
   584   transp_def ~> transp_trans
   585   Domain_def ~> Domain_unfold
   586   Range_def ~> Domain_converse [symmetric]
   587 
   588 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   589 
   590 See theory "Relation" for examples for making use of pred/set
   591 conversions by means of attributes "to_set" and "to_pred".
   592 
   593 INCOMPATIBILITY.
   594 
   595 * Renamed facts about the power operation on relations, i.e., relpow
   596 to match the constant's name:
   597 
   598   rel_pow_1 ~> relpow_1
   599   rel_pow_0_I ~> relpow_0_I
   600   rel_pow_Suc_I ~> relpow_Suc_I
   601   rel_pow_Suc_I2 ~> relpow_Suc_I2
   602   rel_pow_0_E ~> relpow_0_E
   603   rel_pow_Suc_E ~> relpow_Suc_E
   604   rel_pow_E ~> relpow_E
   605   rel_pow_Suc_D2 ~> relpow_Suc_D2
   606   rel_pow_Suc_E2 ~> relpow_Suc_E2
   607   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   608   rel_pow_E2 ~> relpow_E2
   609   rel_pow_add ~> relpow_add
   610   rel_pow_commute ~> relpow
   611   rel_pow_empty ~> relpow_empty:
   612   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   613   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   614   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   615   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   616   rel_pow_fun_conv ~> relpow_fun_conv
   617   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   618   rel_pow_finite_bounded ~> relpow_finite_bounded
   619   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   620   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   621   single_valued_rel_pow ~> single_valued_relpow
   622 
   623 INCOMPATIBILITY.
   624 
   625 * Theory Relation: Consolidated constant name for relation composition
   626 and corresponding theorem names:
   627 
   628   - Renamed constant rel_comp to relcomp.
   629 
   630   - Dropped abbreviation pred_comp. Use relcompp instead.
   631 
   632   - Renamed theorems:
   633 
   634     rel_compI ~> relcompI
   635     rel_compEpair ~> relcompEpair
   636     rel_compE ~> relcompE
   637     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   638     rel_comp_empty1 ~> relcomp_empty1
   639     rel_comp_mono ~> relcomp_mono
   640     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   641     rel_comp_distrib ~> relcomp_distrib
   642     rel_comp_distrib2 ~> relcomp_distrib2
   643     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   644     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   645     single_valued_rel_comp ~> single_valued_relcomp
   646     rel_comp_def ~> relcomp_unfold
   647     converse_rel_comp ~> converse_relcomp
   648     pred_compI ~> relcomppI
   649     pred_compE ~> relcomppE
   650     pred_comp_bot1 ~> relcompp_bot1
   651     pred_comp_bot2 ~> relcompp_bot2
   652     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   653     pred_comp_mono ~> relcompp_mono
   654     pred_comp_distrib ~> relcompp_distrib
   655     pred_comp_distrib2 ~> relcompp_distrib2
   656     converse_pred_comp ~> converse_relcompp
   657 
   658     finite_rel_comp ~> finite_relcomp
   659 
   660     set_rel_comp ~> set_relcomp
   661 
   662 INCOMPATIBILITY.
   663 
   664 * Theory Divides: Discontinued redundant theorems about div and mod.
   665 INCOMPATIBILITY, use the corresponding generic theorems instead.
   666 
   667   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   668   zdiv_self ~> div_self
   669   zmod_self ~> mod_self
   670   zdiv_zero ~> div_0
   671   zmod_zero ~> mod_0
   672   zdiv_zmod_equality ~> div_mod_equality2
   673   zdiv_zmod_equality2 ~> div_mod_equality
   674   zmod_zdiv_trivial ~> mod_div_trivial
   675   zdiv_zminus_zminus ~> div_minus_minus
   676   zmod_zminus_zminus ~> mod_minus_minus
   677   zdiv_zminus2 ~> div_minus_right
   678   zmod_zminus2 ~> mod_minus_right
   679   zdiv_minus1_right ~> div_minus1_right
   680   zmod_minus1_right ~> mod_minus1_right
   681   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   682   zmod_zmult1_eq ~> mod_mult_right_eq
   683   zpower_zmod ~> power_mod
   684   zdvd_zmod ~> dvd_mod
   685   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   686   mod_mult_distrib ~> mult_mod_left
   687   mod_mult_distrib2 ~> mult_mod_right
   688 
   689 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   690 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   691 
   692 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   693 
   694 * Consolidated theorem names concerning fold combinators:
   695 
   696   inf_INFI_fold_inf ~> inf_INF_fold_inf
   697   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   698   INFI_fold_inf ~> INF_fold_inf
   699   SUPR_fold_sup ~> SUP_fold_sup
   700   union_set ~> union_set_fold
   701   minus_set ~> minus_set_fold
   702   INFI_set_fold ~> INF_set_fold
   703   SUPR_set_fold ~> SUP_set_fold
   704   INF_code ~> INF_set_foldr
   705   SUP_code ~> SUP_set_foldr
   706   foldr.simps ~> foldr.simps (in point-free formulation)
   707   foldr_fold_rev ~> foldr_conv_fold
   708   foldl_fold ~> foldl_conv_fold
   709   foldr_foldr ~> foldr_conv_foldl
   710   foldl_foldr ~> foldl_conv_foldr
   711   fold_set_remdups ~> fold_set_fold_remdups
   712   fold_set ~> fold_set_fold
   713   fold1_set ~> fold1_set_fold
   714 
   715 INCOMPATIBILITY.
   716 
   717 * Dropped rarely useful theorems concerning fold combinators:
   718 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   719 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   720 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   721 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   722 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   723 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   724 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   725 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   726 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   727 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   728 
   729 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   730 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   731 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   732 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   733 lemmas over fold rather than foldr, or make use of lemmas
   734 fold_conv_foldr and fold_rev.
   735 
   736 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   737 through option types.
   738 
   739 * "Transitive_Closure.ntrancl": bounded transitive closure on
   740 relations.
   741 
   742 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   743 
   744 * Theory Int: Discontinued many legacy theorems specific to type int.
   745 INCOMPATIBILITY, use the corresponding generic theorems instead.
   746 
   747   zminus_zminus ~> minus_minus
   748   zminus_0 ~> minus_zero
   749   zminus_zadd_distrib ~> minus_add_distrib
   750   zadd_commute ~> add_commute
   751   zadd_assoc ~> add_assoc
   752   zadd_left_commute ~> add_left_commute
   753   zadd_ac ~> add_ac
   754   zmult_ac ~> mult_ac
   755   zadd_0 ~> add_0_left
   756   zadd_0_right ~> add_0_right
   757   zadd_zminus_inverse2 ~> left_minus
   758   zmult_zminus ~> mult_minus_left
   759   zmult_commute ~> mult_commute
   760   zmult_assoc ~> mult_assoc
   761   zadd_zmult_distrib ~> left_distrib
   762   zadd_zmult_distrib2 ~> right_distrib
   763   zdiff_zmult_distrib ~> left_diff_distrib
   764   zdiff_zmult_distrib2 ~> right_diff_distrib
   765   zmult_1 ~> mult_1_left
   766   zmult_1_right ~> mult_1_right
   767   zle_refl ~> order_refl
   768   zle_trans ~> order_trans
   769   zle_antisym ~> order_antisym
   770   zle_linear ~> linorder_linear
   771   zless_linear ~> linorder_less_linear
   772   zadd_left_mono ~> add_left_mono
   773   zadd_strict_right_mono ~> add_strict_right_mono
   774   zadd_zless_mono ~> add_less_le_mono
   775   int_0_less_1 ~> zero_less_one
   776   int_0_neq_1 ~> zero_neq_one
   777   zless_le ~> less_le
   778   zpower_zadd_distrib ~> power_add
   779   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   780   zero_le_zpower_abs ~> zero_le_power_abs
   781 
   782 * Theory Deriv: Renamed
   783 
   784   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   785 
   786 * Theory Library/Multiset: Improved code generation of multisets.
   787 
   788 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   789 are expressed via type classes again. The special syntax
   790 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   791 setsum_set, which is now subsumed by Big_Operators.setsum.
   792 INCOMPATIBILITY.
   793 
   794 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   795 use theory HOL/Library/Nat_Bijection instead.
   796 
   797 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   798 trees is now inside a type class context.  Names of affected
   799 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   800 theories working directly with raw red-black trees, adapt the names as
   801 follows:
   802 
   803   Operations:
   804   bulkload -> rbt_bulkload
   805   del_from_left -> rbt_del_from_left
   806   del_from_right -> rbt_del_from_right
   807   del -> rbt_del
   808   delete -> rbt_delete
   809   ins -> rbt_ins
   810   insert -> rbt_insert
   811   insertw -> rbt_insert_with
   812   insert_with_key -> rbt_insert_with_key
   813   map_entry -> rbt_map_entry
   814   lookup -> rbt_lookup
   815   sorted -> rbt_sorted
   816   tree_greater -> rbt_greater
   817   tree_less -> rbt_less
   818   tree_less_symbol -> rbt_less_symbol
   819   union -> rbt_union
   820   union_with -> rbt_union_with
   821   union_with_key -> rbt_union_with_key
   822 
   823   Lemmas:
   824   balance_left_sorted -> balance_left_rbt_sorted
   825   balance_left_tree_greater -> balance_left_rbt_greater
   826   balance_left_tree_less -> balance_left_rbt_less
   827   balance_right_sorted -> balance_right_rbt_sorted
   828   balance_right_tree_greater -> balance_right_rbt_greater
   829   balance_right_tree_less -> balance_right_rbt_less
   830   balance_sorted -> balance_rbt_sorted
   831   balance_tree_greater -> balance_rbt_greater
   832   balance_tree_less -> balance_rbt_less
   833   bulkload_is_rbt -> rbt_bulkload_is_rbt
   834   combine_sorted -> combine_rbt_sorted
   835   combine_tree_greater -> combine_rbt_greater
   836   combine_tree_less -> combine_rbt_less
   837   delete_in_tree -> rbt_delete_in_tree
   838   delete_is_rbt -> rbt_delete_is_rbt
   839   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   840   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   841   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   842   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   843   del_in_tree -> rbt_del_in_tree
   844   del_inv1_inv2 -> rbt_del_inv1_inv2
   845   del_sorted -> rbt_del_rbt_sorted
   846   del_tree_greater -> rbt_del_rbt_greater
   847   del_tree_less -> rbt_del_rbt_less
   848   dom_lookup_Branch -> dom_rbt_lookup_Branch
   849   entries_lookup -> entries_rbt_lookup
   850   finite_dom_lookup -> finite_dom_rbt_lookup
   851   insert_sorted -> rbt_insert_rbt_sorted
   852   insertw_is_rbt -> rbt_insertw_is_rbt
   853   insertwk_is_rbt -> rbt_insertwk_is_rbt
   854   insertwk_sorted -> rbt_insertwk_rbt_sorted
   855   insertw_sorted -> rbt_insertw_rbt_sorted
   856   ins_sorted -> ins_rbt_sorted
   857   ins_tree_greater -> ins_rbt_greater
   858   ins_tree_less -> ins_rbt_less
   859   is_rbt_sorted -> is_rbt_rbt_sorted
   860   lookup_balance -> rbt_lookup_balance
   861   lookup_bulkload -> rbt_lookup_rbt_bulkload
   862   lookup_delete -> rbt_lookup_rbt_delete
   863   lookup_Empty -> rbt_lookup_Empty
   864   lookup_from_in_tree -> rbt_lookup_from_in_tree
   865   lookup_in_tree -> rbt_lookup_in_tree
   866   lookup_ins -> rbt_lookup_ins
   867   lookup_insert -> rbt_lookup_rbt_insert
   868   lookup_insertw -> rbt_lookup_rbt_insertw
   869   lookup_insertwk -> rbt_lookup_rbt_insertwk
   870   lookup_keys -> rbt_lookup_keys
   871   lookup_map -> rbt_lookup_map
   872   lookup_map_entry -> rbt_lookup_rbt_map_entry
   873   lookup_tree_greater -> rbt_lookup_rbt_greater
   874   lookup_tree_less -> rbt_lookup_rbt_less
   875   lookup_union -> rbt_lookup_rbt_union
   876   map_entry_color_of -> rbt_map_entry_color_of
   877   map_entry_inv1 -> rbt_map_entry_inv1
   878   map_entry_inv2 -> rbt_map_entry_inv2
   879   map_entry_is_rbt -> rbt_map_entry_is_rbt
   880   map_entry_sorted -> rbt_map_entry_rbt_sorted
   881   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   882   map_entry_tree_less -> rbt_map_entry_rbt_less
   883   map_tree_greater -> map_rbt_greater
   884   map_tree_less -> map_rbt_less
   885   map_sorted -> map_rbt_sorted
   886   paint_sorted -> paint_rbt_sorted
   887   paint_lookup -> paint_rbt_lookup
   888   paint_tree_greater -> paint_rbt_greater
   889   paint_tree_less -> paint_rbt_less
   890   sorted_entries -> rbt_sorted_entries
   891   tree_greater_eq_trans -> rbt_greater_eq_trans
   892   tree_greater_nit -> rbt_greater_nit
   893   tree_greater_prop -> rbt_greater_prop
   894   tree_greater_simps -> rbt_greater_simps
   895   tree_greater_trans -> rbt_greater_trans
   896   tree_less_eq_trans -> rbt_less_eq_trans
   897   tree_less_nit -> rbt_less_nit
   898   tree_less_prop -> rbt_less_prop
   899   tree_less_simps -> rbt_less_simps
   900   tree_less_trans -> rbt_less_trans
   901   tree_ord_props -> rbt_ord_props
   902   union_Branch -> rbt_union_Branch
   903   union_is_rbt -> rbt_union_is_rbt
   904   unionw_is_rbt -> rbt_unionw_is_rbt
   905   unionwk_is_rbt -> rbt_unionwk_is_rbt
   906   unionwk_sorted -> rbt_unionwk_rbt_sorted
   907 
   908 * Theory HOL/Library/Float: Floating point numbers are now defined as
   909 a subset of the real numbers.  All operations are defined using the
   910 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   911 
   912   Changed Operations:
   913   float_abs -> abs
   914   float_nprt -> nprt
   915   float_pprt -> pprt
   916   pow2 -> use powr
   917   round_down -> float_round_down
   918   round_up -> float_round_up
   919   scale -> exponent
   920 
   921   Removed Operations:
   922   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   923 
   924   Renamed Lemmas:
   925   abs_float_def -> Float.compute_float_abs
   926   bitlen_ge0 -> bitlen_nonneg
   927   bitlen.simps -> Float.compute_bitlen
   928   float_components -> Float_mantissa_exponent
   929   float_divl.simps -> Float.compute_float_divl
   930   float_divr.simps -> Float.compute_float_divr
   931   float_eq_odd -> mult_powr_eq_mult_powr_iff
   932   float_power -> real_of_float_power
   933   lapprox_posrat_def -> Float.compute_lapprox_posrat
   934   lapprox_rat.simps -> Float.compute_lapprox_rat
   935   le_float_def' -> Float.compute_float_le
   936   le_float_def -> less_eq_float.rep_eq
   937   less_float_def' -> Float.compute_float_less
   938   less_float_def -> less_float.rep_eq
   939   normfloat_def -> Float.compute_normfloat
   940   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   941   normfloat -> normfloat_def
   942   normfloat_unique -> use normfloat_def
   943   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   944   one_float_def -> Float.compute_float_one
   945   plus_float_def -> Float.compute_float_plus
   946   rapprox_posrat_def -> Float.compute_rapprox_posrat
   947   rapprox_rat.simps -> Float.compute_rapprox_rat
   948   real_of_float_0 -> zero_float.rep_eq
   949   real_of_float_1 -> one_float.rep_eq
   950   real_of_float_abs -> abs_float.rep_eq
   951   real_of_float_add -> plus_float.rep_eq
   952   real_of_float_minus -> uminus_float.rep_eq
   953   real_of_float_mult -> times_float.rep_eq
   954   real_of_float_simp -> Float.rep_eq
   955   real_of_float_sub -> minus_float.rep_eq
   956   round_down.simps -> Float.compute_float_round_down
   957   round_up.simps -> Float.compute_float_round_up
   958   times_float_def -> Float.compute_float_times
   959   uminus_float_def -> Float.compute_float_uminus
   960   zero_float_def -> Float.compute_float_zero
   961 
   962   Lemmas not necessary anymore, use the transfer method:
   963   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   964   float_divr, float_le_simp, float_less1_mantissa_bound,
   965   float_less_simp, float_less_zero, float_le_zero,
   966   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   967   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   968   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   969   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   970   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   971   round_up, zero_le_float, zero_less_float
   972 
   973 * New theory HOL/Library/DAList provides an abstract type for
   974 association lists with distinct keys.
   975 
   976 * Session HOL/IMP: Added new theory of abstract interpretation of
   977 annotated commands.
   978 
   979 * Session HOL-Import: Re-implementation from scratch is faster,
   980 simpler, and more scalable.  Requires a proof bundle, which is
   981 available as an external component.  Discontinued old (and mostly
   982 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   983 
   984 * Session HOL-Word: Discontinued many redundant theorems specific to
   985 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   986 instead.
   987 
   988   word_sub_alt ~> word_sub_wi
   989   word_add_alt ~> word_add_def
   990   word_mult_alt ~> word_mult_def
   991   word_minus_alt ~> word_minus_def
   992   word_0_alt ~> word_0_wi
   993   word_1_alt ~> word_1_wi
   994   word_add_0 ~> add_0_left
   995   word_add_0_right ~> add_0_right
   996   word_mult_1 ~> mult_1_left
   997   word_mult_1_right ~> mult_1_right
   998   word_add_commute ~> add_commute
   999   word_add_assoc ~> add_assoc
  1000   word_add_left_commute ~> add_left_commute
  1001   word_mult_commute ~> mult_commute
  1002   word_mult_assoc ~> mult_assoc
  1003   word_mult_left_commute ~> mult_left_commute
  1004   word_left_distrib ~> left_distrib
  1005   word_right_distrib ~> right_distrib
  1006   word_left_minus ~> left_minus
  1007   word_diff_0_right ~> diff_0_right
  1008   word_diff_self ~> diff_self
  1009   word_sub_def ~> diff_minus
  1010   word_diff_minus ~> diff_minus
  1011   word_add_ac ~> add_ac
  1012   word_mult_ac ~> mult_ac
  1013   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1014   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1015   word_order_trans ~> order_trans
  1016   word_order_refl ~> order_refl
  1017   word_order_antisym ~> order_antisym
  1018   word_order_linear ~> linorder_linear
  1019   lenw1_zero_neq_one ~> zero_neq_one
  1020   word_number_of_eq ~> number_of_eq
  1021   word_of_int_add_hom ~> wi_hom_add
  1022   word_of_int_sub_hom ~> wi_hom_sub
  1023   word_of_int_mult_hom ~> wi_hom_mult
  1024   word_of_int_minus_hom ~> wi_hom_neg
  1025   word_of_int_succ_hom ~> wi_hom_succ
  1026   word_of_int_pred_hom ~> wi_hom_pred
  1027   word_of_int_0_hom ~> word_0_wi
  1028   word_of_int_1_hom ~> word_1_wi
  1029 
  1030 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1031 machine word equalities and inequalities into logical circuits,
  1032 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1033 multiplication, shifting by constants, bitwise operators and numeric
  1034 constants.  Requires fixed-length word types, not 'a word.  Solves
  1035 many standard word identities outright and converts more into first
  1036 order problems amenable to blast or similar.  See also examples in
  1037 HOL/Word/Examples/WordExamples.thy.
  1038 
  1039 * Session HOL-Probability: Introduced the type "'a measure" to
  1040 represent measures, this replaces the records 'a algebra and 'a
  1041 measure_space.  The locales based on subset_class now have two
  1042 locale-parameters the space \<Omega> and the set of measurable sets M.
  1043 The product of probability spaces uses now the same constant as the
  1044 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1045 measure".  Most constants are defined now outside of locales and gain
  1046 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1047 Measure space constructions for distributions and densities now got
  1048 their own constants distr and density.  Instead of using locales to
  1049 describe measure spaces with a finite space, the measure count_space
  1050 and point_measure is introduced.  INCOMPATIBILITY.
  1051 
  1052   Renamed constants:
  1053   measure -> emeasure
  1054   finite_measure.\<mu>' -> measure
  1055   product_algebra_generator -> prod_algebra
  1056   product_prob_space.emb -> prod_emb
  1057   product_prob_space.infprod_algebra -> PiM
  1058 
  1059   Removed locales:
  1060   completeable_measure_space
  1061   finite_measure_space
  1062   finite_prob_space
  1063   finite_product_finite_prob_space
  1064   finite_product_sigma_algebra
  1065   finite_sigma_algebra
  1066   measure_space
  1067   pair_finite_prob_space
  1068   pair_finite_sigma_algebra
  1069   pair_finite_space
  1070   pair_sigma_algebra
  1071   product_sigma_algebra
  1072 
  1073   Removed constants:
  1074   conditional_space
  1075   distribution -> use distr measure, or distributed predicate
  1076   image_space
  1077   joint_distribution -> use distr measure, or distributed predicate
  1078   pair_measure_generator
  1079   product_prob_space.infprod_algebra -> use PiM
  1080   subvimage
  1081 
  1082   Replacement theorems:
  1083   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1084   finite_measure.empty_measure -> measure_empty
  1085   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1086   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1087   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1088   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1089   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1090   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1091   finite_measure.positive_measure' -> measure_nonneg
  1092   finite_measure.real_measure -> finite_measure.emeasure_real
  1093   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1094   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1095   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1096   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1097   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1098   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1099   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1100   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1101   information_space.entropy_commute -> information_space.entropy_commute_simple
  1102   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1103   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1104   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1105   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1106   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1107   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1108   information_space.KL_ge_0 -> information_space.KL_nonneg
  1109   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1110   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1111   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1112   Int_stable_product_algebra_generator -> positive_integral
  1113   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1114   measure_space.additive -> emeasure_additive
  1115   measure_space.AE_iff_null_set -> AE_iff_null
  1116   measure_space.almost_everywhere_def -> eventually_ae_filter
  1117   measure_space.almost_everywhere_vimage -> AE_distrD
  1118   measure_space.continuity_from_above -> INF_emeasure_decseq
  1119   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1120   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1121   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1122   measure_space_density -> emeasure_density
  1123   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1124   measure_space.integrable_vimage -> integrable_distr
  1125   measure_space.integral_translated_density -> integral_density
  1126   measure_space.integral_vimage -> integral_distr
  1127   measure_space.measure_additive -> plus_emeasure
  1128   measure_space.measure_compl -> emeasure_compl
  1129   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1130   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1131   measure_space.measure_decseq -> decseq_emeasure
  1132   measure_space.measure_Diff -> emeasure_Diff
  1133   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1134   measure_space.measure_eq_0 -> emeasure_eq_0
  1135   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1136   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1137   measure_space.measure_incseq -> incseq_emeasure
  1138   measure_space.measure_insert -> emeasure_insert
  1139   measure_space.measure_mono -> emeasure_mono
  1140   measure_space.measure_not_negative -> emeasure_not_MInf
  1141   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1142   measure_space.measure_setsum -> setsum_emeasure
  1143   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1144   measure_space.measure_space_vimage -> emeasure_distr
  1145   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1146   measure_space.measure_subadditive -> subadditive
  1147   measure_space.measure_top -> emeasure_space
  1148   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1149   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1150   measure_space.positive_integral_translated_density -> positive_integral_density
  1151   measure_space.positive_integral_vimage -> positive_integral_distr
  1152   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1153   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1154   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1155   measure_space.real_measure_Diff -> measure_Diff
  1156   measure_space.real_measure_finite_Union -> measure_finite_Union
  1157   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1158   measure_space.real_measure_subadditive -> measure_subadditive
  1159   measure_space.real_measure_Union -> measure_Union
  1160   measure_space.real_measure_UNION -> measure_UNION
  1161   measure_space.simple_function_vimage -> simple_function_comp
  1162   measure_space.simple_integral_vimage -> simple_integral_distr
  1163   measure_space.simple_integral_vimage -> simple_integral_distr
  1164   measure_unique_Int_stable -> measure_eqI_generator_eq
  1165   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1166   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1167   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1168   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1169   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1170   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1171   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1172   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1173   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1174   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1175   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1176   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1177   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1178   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1179   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1180   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1181   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1182   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1183   prob_space.prob_space_vimage -> prob_space_distr
  1184   prob_space.random_variable_restrict -> measurable_restrict
  1185   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1186   product_algebraE -> prod_algebraE_all
  1187   product_algebra_generator_der -> prod_algebra_eq_finite
  1188   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1189   product_algebraI -> sets_PiM_I_finite
  1190   product_measure_exists -> product_sigma_finite.sigma_finite
  1191   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1192   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1193   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1194   product_prob_space.measurable_component -> measurable_component_singleton
  1195   product_prob_space.measurable_emb -> measurable_prod_emb
  1196   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1197   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1198   product_prob_space.measure_emb -> emeasure_prod_emb
  1199   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1200   product_sigma_algebra.product_algebra_into_space -> space_closed
  1201   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1202   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1203   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1204   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1205   sets_product_algebra -> sets_PiM
  1206   sigma_algebra.measurable_sigma -> measurable_measure_of
  1207   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1208   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1209   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1210   space_product_algebra -> space_PiM
  1211 
  1212 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1213 languages) into Isabelle/HOL.
  1214 
  1215 
  1216 *** FOL ***
  1217 
  1218 * New "case_product" attribute (see HOL).
  1219 
  1220 
  1221 *** ZF ***
  1222 
  1223 * Greater support for structured proofs involving induction or case
  1224 analysis.
  1225 
  1226 * Much greater use of mathematical symbols.
  1227 
  1228 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1229 
  1230 
  1231 *** ML ***
  1232 
  1233 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1234 from a minor keyword introduced via theory header declaration.
  1235 
  1236 * Antiquotation @{command_spec "name"} produces the
  1237 Outer_Syntax.command_spec from a major keyword introduced via theory
  1238 header declaration; it can be passed to Outer_Syntax.command etc.
  1239 
  1240 * Local_Theory.define no longer hard-wires default theorem name
  1241 "foo_def", but retains the binding as given.  If that is Binding.empty
  1242 / Attrib.empty_binding, the result is not registered as user-level
  1243 fact.  The Local_Theory.define_internal variant allows to specify a
  1244 non-empty name (used for the foundation in the background theory),
  1245 while omitting the fact binding in the user-context.  Potential
  1246 INCOMPATIBILITY for derived definitional packages: need to specify
  1247 naming policy for primitive definitions more explicitly.
  1248 
  1249 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1250 conformance with similar operations in structure Term and Logic.
  1251 
  1252 * Antiquotation @{attributes [...]} embeds attribute source
  1253 representation into the ML text, which is particularly useful with
  1254 declarations like Local_Theory.note.
  1255 
  1256 * Structure Proof_Context follows standard naming scheme.  Old
  1257 ProofContext has been discontinued.  INCOMPATIBILITY.
  1258 
  1259 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1260 change of semantics: update is applied to auxiliary local theory
  1261 context as well.
  1262 
  1263 * Modernized some old-style infix operations:
  1264 
  1265   addeqcongs    ~> Simplifier.add_eqcong
  1266   deleqcongs    ~> Simplifier.del_eqcong
  1267   addcongs      ~> Simplifier.add_cong
  1268   delcongs      ~> Simplifier.del_cong
  1269   setmksimps    ~> Simplifier.set_mksimps
  1270   setmkcong     ~> Simplifier.set_mkcong
  1271   setmksym      ~> Simplifier.set_mksym
  1272   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1273   settermless   ~> Simplifier.set_termless
  1274   setsubgoaler  ~> Simplifier.set_subgoaler
  1275   addsplits     ~> Splitter.add_split
  1276   delsplits     ~> Splitter.del_split
  1277 
  1278 
  1279 *** System ***
  1280 
  1281 * USER_HOME settings variable points to cross-platform user home
  1282 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1283 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1284 of former $HOME.  A different default for USER_HOME may be set
  1285 explicitly in shell environment, before Isabelle settings are
  1286 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1287 the generic user home was intended.
  1288 
  1289 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1290 notation, which is useful for the jEdit file browser, for example.
  1291 
  1292 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1293 (not just JRE).
  1294 
  1295 
  1296 
  1297 New in Isabelle2011-1 (October 2011)
  1298 ------------------------------------
  1299 
  1300 *** General ***
  1301 
  1302 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1303 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1304 
  1305   - Management of multiple theory files directly from the editor
  1306     buffer store -- bypassing the file-system (no requirement to save
  1307     files for checking).
  1308 
  1309   - Markup of formal entities within the text buffer, with semantic
  1310     highlighting, tooltips and hyperlinks to jump to defining source
  1311     positions.
  1312 
  1313   - Improved text rendering, with sub/superscripts in the source
  1314     buffer (including support for copy/paste wrt. output panel, HTML
  1315     theory output and other non-Isabelle text boxes).
  1316 
  1317   - Refined scheduling of proof checking and printing of results,
  1318     based on interactive editor view.  (Note: jEdit folding and
  1319     narrowing allows to restrict buffer perspectives explicitly.)
  1320 
  1321   - Reduced CPU performance requirements, usable on machines with few
  1322     cores.
  1323 
  1324   - Reduced memory requirements due to pruning of unused document
  1325     versions (garbage collection).
  1326 
  1327 See also ~~/src/Tools/jEdit/README.html for further information,
  1328 including some remaining limitations.
  1329 
  1330 * Theory loader: source files are exclusively located via the master
  1331 directory of each theory node (where the .thy file itself resides).
  1332 The global load path (such as src/HOL/Library) has been discontinued.
  1333 Note that the path element ~~ may be used to reference theories in the
  1334 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1335 INCOMPATIBILITY.
  1336 
  1337 * Theory loader: source files are identified by content via SHA1
  1338 digests.  Discontinued former path/modtime identification and optional
  1339 ISABELLE_FILE_IDENT plugin scripts.
  1340 
  1341 * Parallelization of nested Isar proofs is subject to
  1342 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1343 usedir option -Q.
  1344 
  1345 * Name space: former unsynchronized references are now proper
  1346 configuration options, with more conventional names:
  1347 
  1348   long_names   ~> names_long
  1349   short_names  ~> names_short
  1350   unique_names ~> names_unique
  1351 
  1352 Minor INCOMPATIBILITY, need to declare options in context like this:
  1353 
  1354   declare [[names_unique = false]]
  1355 
  1356 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1357 that the result needs to be unique, which means fact specifications
  1358 may have to be refined after enriching a proof context.
  1359 
  1360 * Attribute "case_names" has been refined: the assumptions in each case
  1361 can be named now by following the case name with [name1 name2 ...].
  1362 
  1363 * Isabelle/Isar reference manual has been updated and extended:
  1364   - "Synopsis" provides a catalog of main Isar language concepts.
  1365   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1366   - Updated material from classic "ref" manual, notably about
  1367     "Classical Reasoner".
  1368 
  1369 
  1370 *** HOL ***
  1371 
  1372 * Class bot and top require underlying partial order rather than
  1373 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1374 
  1375 * Class complete_lattice: generalized a couple of lemmas from sets;
  1376 generalized theorems INF_cong and SUP_cong.  New type classes for
  1377 complete boolean algebras and complete linear orders.  Lemmas
  1378 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1379 class complete_linorder.
  1380 
  1381 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1382 Sup_fun_def, Inf_apply, Sup_apply.
  1383 
  1384 Removed redundant lemmas (the right hand side gives hints how to
  1385 replace them for (metis ...), or (simp only: ...) proofs):
  1386 
  1387   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1388   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1389   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1390   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1391   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1392   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1393   Inter_def ~> INF_def, image_def
  1394   Union_def ~> SUP_def, image_def
  1395   INT_eq ~> INF_def, and image_def
  1396   UN_eq ~> SUP_def, and image_def
  1397   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1398 
  1399 More consistent and comprehensive names:
  1400 
  1401   INTER_eq_Inter_image ~> INF_def
  1402   UNION_eq_Union_image ~> SUP_def
  1403   INFI_def ~> INF_def
  1404   SUPR_def ~> SUP_def
  1405   INF_leI ~> INF_lower
  1406   INF_leI2 ~> INF_lower2
  1407   le_INFI ~> INF_greatest
  1408   le_SUPI ~> SUP_upper
  1409   le_SUPI2 ~> SUP_upper2
  1410   SUP_leI ~> SUP_least
  1411   INFI_bool_eq ~> INF_bool_eq
  1412   SUPR_bool_eq ~> SUP_bool_eq
  1413   INFI_apply ~> INF_apply
  1414   SUPR_apply ~> SUP_apply
  1415   INTER_def ~> INTER_eq
  1416   UNION_def ~> UNION_eq
  1417 
  1418 INCOMPATIBILITY.
  1419 
  1420 * Renamed theory Complete_Lattice to Complete_Lattices.
  1421 INCOMPATIBILITY.
  1422 
  1423 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1424 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1425 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1426 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1427 
  1428 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1429 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1430 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1431 INCOMPATIBILITY.
  1432 
  1433 * Added syntactic classes "inf" and "sup" for the respective
  1434 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1435 (mostly internal) locale predicates for some derived classes.
  1436 
  1437 * Theorem collections ball_simps and bex_simps do not contain theorems
  1438 referring to UNION any longer; these have been moved to collection
  1439 UN_ball_bex_simps.  INCOMPATIBILITY.
  1440 
  1441 * Theory Archimedean_Field: floor now is defined as parameter of a
  1442 separate type class floor_ceiling.
  1443 
  1444 * Theory Finite_Set: more coherent development of fold_set locales:
  1445 
  1446     locale fun_left_comm ~> locale comp_fun_commute
  1447     locale fun_left_comm_idem ~> locale comp_fun_idem
  1448 
  1449 Both use point-free characterization; interpretation proofs may need
  1450 adjustment.  INCOMPATIBILITY.
  1451 
  1452 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1453 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1454 
  1455 * Theory Complex_Main: The locale interpretations for the
  1456 bounded_linear and bounded_bilinear locales have been removed, in
  1457 order to reduce the number of duplicate lemmas. Users must use the
  1458 original names for distributivity theorems, potential INCOMPATIBILITY.
  1459 
  1460   divide.add ~> add_divide_distrib
  1461   divide.diff ~> diff_divide_distrib
  1462   divide.setsum ~> setsum_divide_distrib
  1463   mult.add_right ~> right_distrib
  1464   mult.diff_right ~> right_diff_distrib
  1465   mult_right.setsum ~> setsum_right_distrib
  1466   mult_left.diff ~> left_diff_distrib
  1467 
  1468 * Theory Complex_Main: Several redundant theorems have been removed or
  1469 replaced by more general versions. INCOMPATIBILITY.
  1470 
  1471   real_diff_def ~> minus_real_def
  1472   real_divide_def ~> divide_real_def
  1473   real_less_def ~> less_le
  1474   real_abs_def ~> abs_real_def
  1475   real_sgn_def ~> sgn_real_def
  1476   real_mult_commute ~> mult_commute
  1477   real_mult_assoc ~> mult_assoc
  1478   real_mult_1 ~> mult_1_left
  1479   real_add_mult_distrib ~> left_distrib
  1480   real_zero_not_eq_one ~> zero_neq_one
  1481   real_mult_inverse_left ~> left_inverse
  1482   INVERSE_ZERO ~> inverse_zero
  1483   real_le_refl ~> order_refl
  1484   real_le_antisym ~> order_antisym
  1485   real_le_trans ~> order_trans
  1486   real_le_linear ~> linear
  1487   real_le_eq_diff ~> le_iff_diff_le_0
  1488   real_add_left_mono ~> add_left_mono
  1489   real_mult_order ~> mult_pos_pos
  1490   real_mult_less_mono2 ~> mult_strict_left_mono
  1491   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1492   real_0_le_divide_iff ~> zero_le_divide_iff
  1493   realpow_two_disj ~> power2_eq_iff
  1494   real_squared_diff_one_factored ~> square_diff_one_factored
  1495   realpow_two_diff ~> square_diff_square_factored
  1496   reals_complete2 ~> complete_real
  1497   real_sum_squared_expand ~> power2_sum
  1498   exp_ln_eq ~> ln_unique
  1499   expi_add ~> exp_add
  1500   expi_zero ~> exp_zero
  1501   lemma_DERIV_subst ~> DERIV_cong
  1502   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1503   LIMSEQ_const ~> tendsto_const
  1504   LIMSEQ_norm ~> tendsto_norm
  1505   LIMSEQ_add ~> tendsto_add
  1506   LIMSEQ_minus ~> tendsto_minus
  1507   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1508   LIMSEQ_diff ~> tendsto_diff
  1509   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1510   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1511   LIMSEQ_mult ~> tendsto_mult
  1512   LIMSEQ_inverse ~> tendsto_inverse
  1513   LIMSEQ_divide ~> tendsto_divide
  1514   LIMSEQ_pow ~> tendsto_power
  1515   LIMSEQ_setsum ~> tendsto_setsum
  1516   LIMSEQ_setprod ~> tendsto_setprod
  1517   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1518   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1519   LIMSEQ_imp_rabs ~> tendsto_rabs
  1520   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1521   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1522   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1523   LIMSEQ_Complex ~> tendsto_Complex
  1524   LIM_ident ~> tendsto_ident_at
  1525   LIM_const ~> tendsto_const
  1526   LIM_add ~> tendsto_add
  1527   LIM_add_zero ~> tendsto_add_zero
  1528   LIM_minus ~> tendsto_minus
  1529   LIM_diff ~> tendsto_diff
  1530   LIM_norm ~> tendsto_norm
  1531   LIM_norm_zero ~> tendsto_norm_zero
  1532   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1533   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1534   LIM_rabs ~> tendsto_rabs
  1535   LIM_rabs_zero ~> tendsto_rabs_zero
  1536   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1537   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1538   LIM_compose ~> tendsto_compose
  1539   LIM_mult ~> tendsto_mult
  1540   LIM_scaleR ~> tendsto_scaleR
  1541   LIM_of_real ~> tendsto_of_real
  1542   LIM_power ~> tendsto_power
  1543   LIM_inverse ~> tendsto_inverse
  1544   LIM_sgn ~> tendsto_sgn
  1545   isCont_LIM_compose ~> isCont_tendsto_compose
  1546   bounded_linear.LIM ~> bounded_linear.tendsto
  1547   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1548   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1549   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1550   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1551   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1552   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1553 
  1554 * Theory Complex_Main: The definition of infinite series was
  1555 generalized.  Now it is defined on the type class {topological_space,
  1556 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1557 
  1558 * Theory Complex_Main: The complex exponential function "expi" is now
  1559 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1560 several polymorphic lemmas about "exp" are now applicable to "expi".
  1561 
  1562 * Code generation:
  1563 
  1564   - Theory Library/Code_Char_ord provides native ordering of
  1565     characters in the target language.
  1566 
  1567   - Commands code_module and code_library are legacy, use export_code
  1568     instead.
  1569 
  1570   - Method "evaluation" is legacy, use method "eval" instead.
  1571 
  1572   - Legacy evaluator "SML" is deactivated by default.  May be
  1573     reactivated by the following theory command:
  1574 
  1575       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1576 
  1577 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1578 
  1579 * New proof method "induction" that gives induction hypotheses the
  1580 name "IH", thus distinguishing them from further hypotheses that come
  1581 from rule induction.  The latter are still called "hyps".  Method
  1582 "induction" is a thin wrapper around "induct" and follows the same
  1583 syntax.
  1584 
  1585 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1586 still available as a legacy feature for some time.
  1587 
  1588 * Nitpick:
  1589   - Added "need" and "total_consts" options.
  1590   - Reintroduced "show_skolems" option by popular demand.
  1591   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1592     INCOMPATIBILITY.
  1593 
  1594 * Sledgehammer:
  1595   - Use quasi-sound (and efficient) translations by default.
  1596   - Added support for the following provers: E-ToFoF, LEO-II,
  1597     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1598   - Automatically preplay and minimize proofs before showing them if
  1599     this can be done within reasonable time.
  1600   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1601     INCOMPATIBILITY.
  1602   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1603     "max_mono_iters", and "max_new_mono_instances" options.
  1604   - Removed "explicit_apply" and "full_types" options as well as "Full
  1605     Types" Proof General menu item. INCOMPATIBILITY.
  1606 
  1607 * Metis:
  1608   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1609   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1610     INCOMPATIBILITY.
  1611 
  1612 * Command 'try':
  1613   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1614     "elim:" options. INCOMPATIBILITY.
  1615   - Introduced 'try' that not only runs 'try_methods' but also
  1616     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1617 
  1618 * Quickcheck:
  1619   - Added "eval" option to evaluate terms for the found counterexample
  1620     (currently only supported by the default (exhaustive) tester).
  1621   - Added post-processing of terms to obtain readable counterexamples
  1622     (currently only supported by the default (exhaustive) tester).
  1623   - New counterexample generator quickcheck[narrowing] enables
  1624     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1625     with its installation location defined in the Isabelle settings
  1626     environment as ISABELLE_GHC.
  1627   - Removed quickcheck tester "SML" based on the SML code generator
  1628     (formly in HOL/Library).
  1629 
  1630 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1631 use 'partial_function' instead.
  1632 
  1633 * Theory Library/Extended_Reals replaces now the positive extended
  1634 reals found in probability theory. This file is extended by
  1635 Multivariate_Analysis/Extended_Real_Limits.
  1636 
  1637 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1638 from where it must be imported explicitly if it is really required.
  1639 INCOMPATIBILITY.
  1640 
  1641 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1642 been moved here.  INCOMPATIBILITY.
  1643 
  1644 * Theory Library/Saturated provides type of numbers with saturated
  1645 arithmetic.
  1646 
  1647 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1648 product type 'a * 'b, and provides instance proofs for various order
  1649 and lattice type classes.
  1650 
  1651 * Theory Library/Countable now provides the "countable_datatype" proof
  1652 method for proving "countable" class instances for datatypes.
  1653 
  1654 * Theory Library/Cset_Monad allows do notation for computable sets
  1655 (cset) via the generic monad ad-hoc overloading facility.
  1656 
  1657 * Library: Theories of common data structures are split into theories
  1658 for implementation, an invariant-ensuring type, and connection to an
  1659 abstract type. INCOMPATIBILITY.
  1660 
  1661   - RBT is split into RBT and RBT_Mapping.
  1662   - AssocList is split and renamed into AList and AList_Mapping.
  1663   - DList is split into DList_Impl, DList, and DList_Cset.
  1664   - Cset is split into Cset and List_Cset.
  1665 
  1666 * Theory Library/Nat_Infinity has been renamed to
  1667 Library/Extended_Nat, with name changes of the following types and
  1668 constants:
  1669 
  1670   type inat   ~> type enat
  1671   Fin         ~> enat
  1672   Infty       ~> infinity (overloaded)
  1673   iSuc        ~> eSuc
  1674   the_Fin     ~> the_enat
  1675 
  1676 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1677 been renamed accordingly. INCOMPATIBILITY.
  1678 
  1679 * Session Multivariate_Analysis: The euclidean_space type class now
  1680 fixes a constant "Basis :: 'a set" consisting of the standard
  1681 orthonormal basis for the type. Users now have the option of
  1682 quantifying over this set instead of using the "basis" function, e.g.
  1683 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1684 
  1685 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1686 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1687 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1688 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1689 changed to match. Definition theorems for overloaded constants now use
  1690 the standard "foo_vec_def" naming scheme. A few other theorems have
  1691 been renamed as follows (INCOMPATIBILITY):
  1692 
  1693   Cart_eq          ~> vec_eq_iff
  1694   dist_nth_le_cart ~> dist_vec_nth_le
  1695   tendsto_vector   ~> vec_tendstoI
  1696   Cauchy_vector    ~> vec_CauchyI
  1697 
  1698 * Session Multivariate_Analysis: Several duplicate theorems have been
  1699 removed, and other theorems have been renamed or replaced with more
  1700 general versions. INCOMPATIBILITY.
  1701 
  1702   finite_choice ~> finite_set_choice
  1703   eventually_conjI ~> eventually_conj
  1704   eventually_and ~> eventually_conj_iff
  1705   eventually_false ~> eventually_False
  1706   setsum_norm ~> norm_setsum
  1707   Lim_sequentially ~> LIMSEQ_def
  1708   Lim_ident_at ~> LIM_ident
  1709   Lim_const ~> tendsto_const
  1710   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1711   Lim_neg ~> tendsto_minus
  1712   Lim_add ~> tendsto_add
  1713   Lim_sub ~> tendsto_diff
  1714   Lim_mul ~> tendsto_scaleR
  1715   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1716   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1717   Lim_linear ~> bounded_linear.tendsto
  1718   Lim_component ~> tendsto_euclidean_component
  1719   Lim_component_cart ~> tendsto_vec_nth
  1720   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1721   dot_lsum ~> inner_setsum_left
  1722   dot_rsum ~> inner_setsum_right
  1723   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1724   continuous_neg ~> continuous_minus
  1725   continuous_sub ~> continuous_diff
  1726   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1727   continuous_mul ~> continuous_scaleR
  1728   continuous_inv ~> continuous_inverse
  1729   continuous_at_within_inv ~> continuous_at_within_inverse
  1730   continuous_at_inv ~> continuous_at_inverse
  1731   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1732   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1733   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1734   continuous_on_neg ~> continuous_on_minus
  1735   continuous_on_sub ~> continuous_on_diff
  1736   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1737   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1738   continuous_on_mul ~> continuous_on_scaleR
  1739   continuous_on_mul_real ~> continuous_on_mult
  1740   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1741   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1742   continuous_on_inverse ~> continuous_on_inv
  1743   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1744   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1745   subset_interior ~> interior_mono
  1746   subset_closure ~> closure_mono
  1747   closure_univ ~> closure_UNIV
  1748   real_arch_lt ~> reals_Archimedean2
  1749   real_arch ~> reals_Archimedean3
  1750   real_abs_norm ~> abs_norm_cancel
  1751   real_abs_sub_norm ~> norm_triangle_ineq3
  1752   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1753 
  1754 * Session HOL-Probability:
  1755   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1756   - Infinite products of probability measures are now available.
  1757   - Sigma closure is independent, if the generator is independent
  1758   - Use extended reals instead of positive extended
  1759     reals. INCOMPATIBILITY.
  1760 
  1761 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1762 
  1763   expand_fun_below ~> fun_below_iff
  1764   below_fun_ext ~> fun_belowI
  1765   expand_cfun_eq ~> cfun_eq_iff
  1766   ext_cfun ~> cfun_eqI
  1767   expand_cfun_below ~> cfun_below_iff
  1768   below_cfun_ext ~> cfun_belowI
  1769   monofun_fun_fun ~> fun_belowD
  1770   monofun_fun_arg ~> monofunE
  1771   monofun_lub_fun ~> adm_monofun [THEN admD]
  1772   cont_lub_fun ~> adm_cont [THEN admD]
  1773   cont2cont_Rep_CFun ~> cont2cont_APP
  1774   cont_Rep_CFun_app ~> cont_APP_app
  1775   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1776   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1777   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1778   contlub_cfun ~> lub_APP [symmetric]
  1779   contlub_LAM ~> lub_LAM [symmetric]
  1780   thelubI ~> lub_eqI
  1781   UU_I ~> bottomI
  1782   lift_distinct1 ~> lift.distinct(1)
  1783   lift_distinct2 ~> lift.distinct(2)
  1784   Def_not_UU ~> lift.distinct(2)
  1785   Def_inject ~> lift.inject
  1786   below_UU_iff ~> below_bottom_iff
  1787   eq_UU_iff ~> eq_bottom_iff
  1788 
  1789 
  1790 *** Document preparation ***
  1791 
  1792 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1793 isar-ref manual, both for description and actual application of the
  1794 same.
  1795 
  1796 * Antiquotation @{value} evaluates the given term and presents its
  1797 result.
  1798 
  1799 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1800 variables x1, y23 into subscripted form x\<^isub>1,
  1801 y\<^isub>2\<^isub>3.
  1802 
  1803 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1804 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1805 
  1806 * Localized \isabellestyle switch can be used within blocks or groups
  1807 like this:
  1808 
  1809   \isabellestyle{it}  %preferred default
  1810   {\isabellestylett @{text "typewriter stuff"}}
  1811 
  1812 * Discontinued special treatment of hard tabulators.  Implicit
  1813 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1814 layouts.
  1815 
  1816 
  1817 *** ML ***
  1818 
  1819 * The inner syntax of sort/type/term/prop supports inlined YXML
  1820 representations within quoted string tokens.  By encoding logical
  1821 entities via Term_XML (in ML or Scala) concrete syntax can be
  1822 bypassed, which is particularly useful for producing bits of text
  1823 under external program control.
  1824 
  1825 * Antiquotations for ML and document preparation are managed as theory
  1826 data, which requires explicit setup.
  1827 
  1828 * Isabelle_Process.is_active allows tools to check if the official
  1829 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1830 (better known as Proof General).
  1831 
  1832 * Structure Proof_Context follows standard naming scheme.  Old
  1833 ProofContext is still available for some time as legacy alias.
  1834 
  1835 * Structure Timing provides various operations for timing; supersedes
  1836 former start_timing/end_timing etc.
  1837 
  1838 * Path.print is the official way to show file-system paths to users
  1839 (including quotes etc.).
  1840 
  1841 * Inner syntax: identifiers in parse trees of generic categories
  1842 "logic", "aprop", "idt" etc. carry position information (disguised as
  1843 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1844 translations that choke on unexpected type constraints.  Positions can
  1845 be stripped in ML translations via Syntax.strip_positions /
  1846 Syntax.strip_positions_ast, or via the syntax constant
  1847 "_strip_positions" within parse trees.  As last resort, positions can
  1848 be disabled via the configuration option Syntax.positions, which is
  1849 called "syntax_positions" in Isar attribute syntax.
  1850 
  1851 * Discontinued special status of various ML structures that contribute
  1852 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1853 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1854 refer directly to Ast.Constant, Lexicon.is_identifier,
  1855 Syntax_Trans.mk_binder_tr etc.
  1856 
  1857 * Typed print translation: discontinued show_sorts argument, which is
  1858 already available via context of "advanced" translation.
  1859 
  1860 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1861 goal states; body tactic needs to address all subgoals uniformly.
  1862 
  1863 * Slightly more special eq_list/eq_set, with shortcut involving
  1864 pointer equality (assumes that eq relation is reflexive).
  1865 
  1866 * Classical tactics use proper Proof.context instead of historic types
  1867 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1868 operate directly on Proof.context.  Raw type claset retains its use as
  1869 snapshot of the classical context, which can be recovered via
  1870 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1871 INCOMPATIBILITY, classical tactics and derived proof methods require
  1872 proper Proof.context.
  1873 
  1874 
  1875 *** System ***
  1876 
  1877 * Discontinued support for Poly/ML 5.2, which was the last version
  1878 without proper multithreading and TimeLimit implementation.
  1879 
  1880 * Discontinued old lib/scripts/polyml-platform, which has been
  1881 obsolete since Isabelle2009-2.
  1882 
  1883 * Various optional external tools are referenced more robustly and
  1884 uniformly by explicit Isabelle settings as follows:
  1885 
  1886   ISABELLE_CSDP   (formerly CSDP_EXE)
  1887   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1888   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1889   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1890   ISABELLE_YAP    (formerly EXEC_YAP)
  1891 
  1892 Note that automated detection from the file-system or search path has
  1893 been discontinued.  INCOMPATIBILITY.
  1894 
  1895 * Scala layer provides JVM method invocation service for static
  1896 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1897 example:
  1898 
  1899   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1900 
  1901 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1902 this allows to pass structured values between ML and Scala.
  1903 
  1904 * The IsabelleText fonts includes some further glyphs to support the
  1905 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1906 installed a local copy (which is normally *not* required) need to
  1907 delete or update it from ~~/lib/fonts/.
  1908 
  1909 
  1910 
  1911 New in Isabelle2011 (January 2011)
  1912 ----------------------------------
  1913 
  1914 *** General ***
  1915 
  1916 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1917 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1918 useful tooltips and hyperlinks produced from its static analysis.  The
  1919 bundled component provides an executable Isabelle tool that can be run
  1920 like this:
  1921 
  1922   Isabelle2011/bin/isabelle jedit
  1923 
  1924 * Significantly improved Isabelle/Isar implementation manual.
  1925 
  1926 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1927 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1928 while the default heap location within that directory lacks that extra
  1929 suffix.  This isolates multiple Isabelle installations from each
  1930 other, avoiding problems with old settings in new versions.
  1931 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1932 
  1933 * Source files are always encoded as UTF-8, instead of old-fashioned
  1934 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1935 the following package declarations:
  1936 
  1937   \usepackage[utf8]{inputenc}
  1938   \usepackage{textcomp}
  1939 
  1940 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1941 a Unicode character is treated as a single symbol, not a sequence of
  1942 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1943 contain symbols without further backslash escapes, Unicode can now be
  1944 used here as well.  Recall that Symbol.explode in ML provides a
  1945 consistent view on symbols, while raw explode (or String.explode)
  1946 merely give a byte-oriented representation.
  1947 
  1948 * Theory loader: source files are primarily located via the master
  1949 directory of each theory node (where the .thy file itself resides).
  1950 The global load path is still partially available as legacy feature.
  1951 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1952 explicit paths, relatively to the theory.
  1953 
  1954 * Special treatment of ML file names has been discontinued.
  1955 Historically, optional extensions .ML or .sml were added on demand --
  1956 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1957 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1958 
  1959 * Various options that affect pretty printing etc. are now properly
  1960 handled within the context via configuration options, instead of
  1961 unsynchronized references or print modes.  There are both ML Config.T
  1962 entities and Isar declaration attributes to access these.
  1963 
  1964   ML (Config.T)                 Isar (attribute)
  1965 
  1966   eta_contract                  eta_contract
  1967   show_brackets                 show_brackets
  1968   show_sorts                    show_sorts
  1969   show_types                    show_types
  1970   show_question_marks           show_question_marks
  1971   show_consts                   show_consts
  1972   show_abbrevs                  show_abbrevs
  1973 
  1974   Syntax.ast_trace              syntax_ast_trace
  1975   Syntax.ast_stat               syntax_ast_stat
  1976   Syntax.ambiguity_level        syntax_ambiguity_level
  1977 
  1978   Goal_Display.goals_limit      goals_limit
  1979   Goal_Display.show_main_goal   show_main_goal
  1980 
  1981   Method.rule_trace             rule_trace
  1982 
  1983   Thy_Output.display            thy_output_display
  1984   Thy_Output.quotes             thy_output_quotes
  1985   Thy_Output.indent             thy_output_indent
  1986   Thy_Output.source             thy_output_source
  1987   Thy_Output.break              thy_output_break
  1988 
  1989 Note that corresponding "..._default" references in ML may only be
  1990 changed globally at the ROOT session setup, but *not* within a theory.
  1991 The option "show_abbrevs" supersedes the former print mode
  1992 "no_abbrevs" with inverted meaning.
  1993 
  1994 * More systematic naming of some configuration options.
  1995 INCOMPATIBILITY.
  1996 
  1997   trace_simp  ~>  simp_trace
  1998   debug_simp  ~>  simp_debug
  1999 
  2000 * Support for real valued configuration options, using simplistic
  2001 floating-point notation that coincides with the inner syntax for
  2002 float_token.
  2003 
  2004 * Support for real valued preferences (with approximative PGIP type):
  2005 front-ends need to accept "pgint" values in float notation.
  2006 INCOMPATIBILITY.
  2007 
  2008 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2009 DejaVu Sans.
  2010 
  2011 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2012 
  2013 
  2014 *** Pure ***
  2015 
  2016 * Command 'type_synonym' (with single argument) replaces somewhat
  2017 outdated 'types', which is still available as legacy feature for some
  2018 time.
  2019 
  2020 * Command 'nonterminal' (with 'and' separated list of arguments)
  2021 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2022 
  2023 * Command 'notepad' replaces former 'example_proof' for
  2024 experimentation in Isar without any result.  INCOMPATIBILITY.
  2025 
  2026 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2027 lists of equations to map definitions in a locale to appropriate
  2028 entities in the context of the interpretation.  The 'interpretation'
  2029 command already provided this functionality.
  2030 
  2031 * Diagnostic command 'print_dependencies' prints the locale instances
  2032 that would be activated if the specified expression was interpreted in
  2033 the current context.  Variant "print_dependencies!" assumes a context
  2034 without interpretations.
  2035 
  2036 * Diagnostic command 'print_interps' prints interpretations in proofs
  2037 in addition to interpretations in theories.
  2038 
  2039 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2040 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2041 Sign.root_path and Sign.local_path may be applied directly where this
  2042 feature is still required for historical reasons.
  2043 
  2044 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2045 'definition' instead.
  2046 
  2047 * The "prems" fact, which refers to the accidental collection of
  2048 foundational premises in the context, is now explicitly marked as
  2049 legacy feature and will be discontinued soon.  Consider using "assms"
  2050 of the head statement or reference facts by explicit names.
  2051 
  2052 * Document antiquotations @{class} and @{type} print classes and type
  2053 constructors.
  2054 
  2055 * Document antiquotation @{file} checks file/directory entries within
  2056 the local file system.
  2057 
  2058 
  2059 *** HOL ***
  2060 
  2061 * Coercive subtyping: functions can be declared as coercions and type
  2062 inference will add them as necessary upon input of a term.  Theory
  2063 Complex_Main declares real :: nat => real and real :: int => real as
  2064 coercions. A coercion function f is declared like this:
  2065 
  2066   declare [[coercion f]]
  2067 
  2068 To lift coercions through type constructors (e.g. from nat => real to
  2069 nat list => real list), map functions can be declared, e.g.
  2070 
  2071   declare [[coercion_map map]]
  2072 
  2073 Currently coercion inference is activated only in theories including
  2074 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2075 the configuration option "coercion_enabled", e.g. it can be enabled in
  2076 other theories like this:
  2077 
  2078   declare [[coercion_enabled]]
  2079 
  2080 * Command 'partial_function' provides basic support for recursive
  2081 function definitions over complete partial orders.  Concrete instances
  2082 are provided for i) the option type, ii) tail recursion on arbitrary
  2083 types, and iii) the heap monad of Imperative_HOL.  See
  2084 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2085 for examples.
  2086 
  2087 * Function package: f.psimps rules are no longer implicitly declared
  2088 as [simp].  INCOMPATIBILITY.
  2089 
  2090 * Datatype package: theorems generated for executable equality (class
  2091 "eq") carry proper names and are treated as default code equations.
  2092 
  2093 * Inductive package: now offers command 'inductive_simps' to
  2094 automatically derive instantiated and simplified equations for
  2095 inductive predicates, similar to 'inductive_cases'.
  2096 
  2097 * Command 'enriched_type' allows to register properties of the
  2098 functorial structure of types.
  2099 
  2100 * Improved infrastructure for term evaluation using code generator
  2101 techniques, in particular static evaluation conversions.
  2102 
  2103 * Code generator: Scala (2.8 or higher) has been added to the target
  2104 languages.
  2105 
  2106 * Code generator: globbing constant expressions "*" and "Theory.*"
  2107 have been replaced by the more idiomatic "_" and "Theory._".
  2108 INCOMPATIBILITY.
  2109 
  2110 * Code generator: export_code without explicit file declaration prints
  2111 to standard output.  INCOMPATIBILITY.
  2112 
  2113 * Code generator: do not print function definitions for case
  2114 combinators any longer.
  2115 
  2116 * Code generator: simplification with rules determined with
  2117 src/Tools/Code/code_simp.ML and method "code_simp".
  2118 
  2119 * Code generator for records: more idiomatic representation of record
  2120 types.  Warning: records are not covered by ancient SML code
  2121 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2122 rep_datatype declaration helps to succeed then:
  2123 
  2124   record 'a foo = ...
  2125   ...
  2126   rep_datatype foo_ext ...
  2127 
  2128 * Records: logical foundation type for records does not carry a
  2129 '_type' suffix any longer (obsolete due to authentic syntax).
  2130 INCOMPATIBILITY.
  2131 
  2132 * Quickcheck now by default uses exhaustive testing instead of random
  2133 testing.  Random testing can be invoked by "quickcheck [random]",
  2134 exhaustive testing by "quickcheck [exhaustive]".
  2135 
  2136 * Quickcheck instantiates polymorphic types with small finite
  2137 datatypes by default. This enables a simple execution mechanism to
  2138 handle quantifiers and function equality over the finite datatypes.
  2139 
  2140 * Quickcheck random generator has been renamed from "code" to
  2141 "random".  INCOMPATIBILITY.
  2142 
  2143 * Quickcheck now has a configurable time limit which is set to 30
  2144 seconds by default. This can be changed by adding [timeout = n] to the
  2145 quickcheck command. The time limit for Auto Quickcheck is still set
  2146 independently.
  2147 
  2148 * Quickcheck in locales considers interpretations of that locale for
  2149 counter example search.
  2150 
  2151 * Sledgehammer:
  2152   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2153     method. See the Sledgehammer manual for details ("isabelle doc
  2154     sledgehammer").
  2155   - Renamed commands:
  2156     sledgehammer atp_info ~> sledgehammer running_provers
  2157     sledgehammer atp_kill ~> sledgehammer kill_provers
  2158     sledgehammer available_atps ~> sledgehammer available_provers
  2159     INCOMPATIBILITY.
  2160   - Renamed options:
  2161     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2162     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2163     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2164     (and "ms" and "min" are no longer supported)
  2165     INCOMPATIBILITY.
  2166 
  2167 * Nitpick:
  2168   - Renamed options:
  2169     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2170     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2171     INCOMPATIBILITY.
  2172   - Added support for partial quotient types.
  2173   - Added local versions of the "Nitpick.register_xxx" functions.
  2174   - Added "whack" option.
  2175   - Allow registration of quotient types as codatatypes.
  2176   - Improved "merge_type_vars" option to merge more types.
  2177   - Removed unsound "fast_descrs" option.
  2178   - Added custom symmetry breaking for datatypes, making it possible to reach
  2179     higher cardinalities.
  2180   - Prevent the expansion of too large definitions.
  2181 
  2182 * Proof methods "metis" and "meson" now have configuration options
  2183 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2184 to diagnose these tools. E.g.
  2185 
  2186     using [[metis_trace = true]]
  2187 
  2188 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2189 manually as command 'solve_direct'.
  2190 
  2191 * The default SMT solver Z3 must be enabled explicitly (due to
  2192 licensing issues) by setting the environment variable
  2193 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2194 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2195 changing the SMT solver is done via the configuration option
  2196 "smt_solver".
  2197 
  2198 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2199 i.e. "remote_cvc3" and "remote_z3".
  2200 
  2201 * Added basic SMT support for datatypes, records, and typedefs using
  2202 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2203 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2204 support for a similar behavior).  Minor INCOMPATIBILITY.
  2205 
  2206 * Changed SMT configuration options:
  2207   - Renamed:
  2208     z3_proofs ~> smt_oracle (with inverted meaning)
  2209     z3_trace_assms ~> smt_trace_used_facts
  2210     INCOMPATIBILITY.
  2211   - Added:
  2212     smt_verbose
  2213     smt_random_seed
  2214     smt_datatypes
  2215     smt_infer_triggers
  2216     smt_monomorph_limit
  2217     cvc3_options
  2218     remote_cvc3_options
  2219     remote_z3_options
  2220     yices_options
  2221 
  2222 * Boogie output files (.b2i files) need to be declared in the theory
  2223 header.
  2224 
  2225 * Simplification procedure "list_to_set_comprehension" rewrites list
  2226 comprehensions applied to List.set to set comprehensions.  Occasional
  2227 INCOMPATIBILITY, may be deactivated like this:
  2228 
  2229   declare [[simproc del: list_to_set_comprehension]]
  2230 
  2231 * Removed old version of primrec package.  INCOMPATIBILITY.
  2232 
  2233 * Removed simplifier congruence rule of "prod_case", as has for long
  2234 been the case with "split".  INCOMPATIBILITY.
  2235 
  2236 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2237 
  2238 * Removed [split_format ... and ... and ...] version of
  2239 [split_format].  Potential INCOMPATIBILITY.
  2240 
  2241 * Predicate "sorted" now defined inductively, with nice induction
  2242 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2243 
  2244 * Constant "contents" renamed to "the_elem", to free the generic name
  2245 contents for other uses.  INCOMPATIBILITY.
  2246 
  2247 * Renamed class eq and constant eq (for code generation) to class
  2248 equal and constant equal, plus renaming of related facts and various
  2249 tuning.  INCOMPATIBILITY.
  2250 
  2251 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2252 
  2253 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2254 INCOMPATIBILITY.
  2255 
  2256 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2257 avoid confusion with finite sets.  INCOMPATIBILITY.
  2258 
  2259 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2260 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2261 for congruent(2)).
  2262 
  2263 * Some previously unqualified names have been qualified:
  2264 
  2265   types
  2266     bool ~> HOL.bool
  2267     nat ~> Nat.nat
  2268 
  2269   constants
  2270     Trueprop ~> HOL.Trueprop
  2271     True ~> HOL.True
  2272     False ~> HOL.False
  2273     op & ~> HOL.conj
  2274     op | ~> HOL.disj
  2275     op --> ~> HOL.implies
  2276     op = ~> HOL.eq
  2277     Not ~> HOL.Not
  2278     The ~> HOL.The
  2279     All ~> HOL.All
  2280     Ex ~> HOL.Ex
  2281     Ex1 ~> HOL.Ex1
  2282     Let ~> HOL.Let
  2283     If ~> HOL.If
  2284     Ball ~> Set.Ball
  2285     Bex ~> Set.Bex
  2286     Suc ~> Nat.Suc
  2287     Pair ~> Product_Type.Pair
  2288     fst ~> Product_Type.fst
  2289     snd ~> Product_Type.snd
  2290     curry ~> Product_Type.curry
  2291     op : ~> Set.member
  2292     Collect ~> Set.Collect
  2293 
  2294 INCOMPATIBILITY.
  2295 
  2296 * More canonical naming convention for some fundamental definitions:
  2297 
  2298     bot_bool_eq ~> bot_bool_def
  2299     top_bool_eq ~> top_bool_def
  2300     inf_bool_eq ~> inf_bool_def
  2301     sup_bool_eq ~> sup_bool_def
  2302     bot_fun_eq  ~> bot_fun_def
  2303     top_fun_eq  ~> top_fun_def
  2304     inf_fun_eq  ~> inf_fun_def
  2305     sup_fun_eq  ~> sup_fun_def
  2306 
  2307 INCOMPATIBILITY.
  2308 
  2309 * More stylized fact names:
  2310 
  2311   expand_fun_eq ~> fun_eq_iff
  2312   expand_set_eq ~> set_eq_iff
  2313   set_ext       ~> set_eqI
  2314   nat_number    ~> eval_nat_numeral
  2315 
  2316 INCOMPATIBILITY.
  2317 
  2318 * Refactoring of code-generation specific operations in theory List:
  2319 
  2320   constants
  2321     null ~> List.null
  2322 
  2323   facts
  2324     mem_iff ~> member_def
  2325     null_empty ~> null_def
  2326 
  2327 INCOMPATIBILITY.  Note that these were not supposed to be used
  2328 regularly unless for striking reasons; their main purpose was code
  2329 generation.
  2330 
  2331 Various operations from the Haskell prelude are used for generating
  2332 Haskell code.
  2333 
  2334 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2335 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2336 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2337 
  2338 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2339 "*" and "+" respectively.  INCOMPATIBILITY.
  2340 
  2341 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2342 "Sum_Type.Plus" instead.
  2343 
  2344 * Constant "split" has been merged with constant "prod_case"; names of
  2345 ML functions, facts etc. involving split have been retained so far,
  2346 though.  INCOMPATIBILITY.
  2347 
  2348 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2349 instead.  INCOMPATIBILITY.
  2350 
  2351 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2352 INCOMPATIBILITY.
  2353 
  2354 * Former theory Library/Enum is now part of the HOL-Main image.
  2355 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2356 referred to by its qualified name.
  2357 
  2358   enum    ~>  Enum.enum
  2359   nlists  ~>  Enum.nlists
  2360   product ~>  Enum.product
  2361 
  2362 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2363 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2364 INCOMPATIBILITY.
  2365 
  2366 * Theory Library/SetsAndFunctions has been split into
  2367 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2368 for instance definitions for functions; various improvements.
  2369 INCOMPATIBILITY.
  2370 
  2371 * Theory Library/Multiset provides stable quicksort implementation of
  2372 sort_key.
  2373 
  2374 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2375 INCOMPATIBILITY.
  2376 
  2377 * Session Multivariate_Analysis: introduced a type class for euclidean
  2378 space.  Most theorems are now stated in terms of euclidean spaces
  2379 instead of finite cartesian products.
  2380 
  2381   types
  2382     real ^ 'n ~>  'a::real_vector
  2383               ~>  'a::euclidean_space
  2384               ~>  'a::ordered_euclidean_space
  2385         (depends on your needs)
  2386 
  2387   constants
  2388      _ $ _        ~> _ $$ _
  2389      \<chi> x. _  ~> \<chi>\<chi> x. _
  2390      CARD('n)     ~> DIM('a)
  2391 
  2392 Also note that the indices are now natural numbers and not from some
  2393 finite type. Finite cartesian products of euclidean spaces, products
  2394 of euclidean spaces the real and complex numbers are instantiated to
  2395 be euclidean_spaces.  INCOMPATIBILITY.
  2396 
  2397 * Session Probability: introduced pextreal as positive extended real
  2398 numbers.  Use pextreal as value for measures.  Introduce the
  2399 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2400 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2401 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2402 
  2403 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2404 INCOMPATIBILITY.
  2405 
  2406 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2407 prove verification conditions generated by the SPARK Ada program
  2408 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2409 
  2410 
  2411 *** HOL-Algebra ***
  2412 
  2413 * Theorems for additive ring operations (locale abelian_monoid and
  2414 descendants) are generated by interpretation from their multiplicative
  2415 counterparts.  Names (in particular theorem names) have the mandatory
  2416 qualifier 'add'.  Previous theorem names are redeclared for
  2417 compatibility.
  2418 
  2419 * Structure "int_ring" is now an abbreviation (previously a
  2420 definition).  This fits more natural with advanced interpretations.
  2421 
  2422 
  2423 *** HOLCF ***
  2424 
  2425 * The domain package now runs in definitional mode by default: The
  2426 former command 'new_domain' is now called 'domain'.  To use the domain
  2427 package in its original axiomatic mode, use 'domain (unsafe)'.
  2428 INCOMPATIBILITY.
  2429 
  2430 * The new class "domain" is now the default sort.  Class "predomain"
  2431 is an unpointed version of "domain". Theories can be updated by
  2432 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2433 
  2434   'a::type ~> 'a::countable
  2435   'a::cpo  ~> 'a::predomain
  2436   'a::pcpo ~> 'a::domain
  2437 
  2438 * The old type class "rep" has been superseded by class "domain".
  2439 Accordingly, users of the definitional package must remove any
  2440 "default_sort rep" declarations.  INCOMPATIBILITY.
  2441 
  2442 * The domain package (definitional mode) now supports unpointed
  2443 predomain argument types, as long as they are marked 'lazy'. (Strict
  2444 arguments must be in class "domain".) For example, the following
  2445 domain definition now works:
  2446 
  2447   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2448 
  2449 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2450 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2451 'a option, and 'a list.  Additionally, it configures fixrec and the
  2452 domain package to work with these types.  For example:
  2453 
  2454   fixrec isInl :: "('a + 'b) u -> tr"
  2455     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2456 
  2457   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2458 
  2459 * The "(permissive)" option of fixrec has been replaced with a
  2460 per-equation "(unchecked)" option. See
  2461 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2462 
  2463 * The "bifinite" class no longer fixes a constant "approx"; the class
  2464 now just asserts that such a function exists.  INCOMPATIBILITY.
  2465 
  2466 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2467 defines an embedding of type 'a defl into udom by default; instances
  2468 of "bifinite" and "domain" classes are available in
  2469 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2470 
  2471 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2472 
  2473 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2474 
  2475 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2476 
  2477 * The function "cprod_map" has been renamed to "prod_map".
  2478 INCOMPATIBILITY.
  2479 
  2480 * The monadic bind operator on each powerdomain has new binder syntax
  2481 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2482 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2483 
  2484 * The infix syntax for binary union on each powerdomain has changed
  2485 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2486 syntax.  INCOMPATIBILITY.
  2487 
  2488 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2489 still supported as an input translation.
  2490 
  2491 * Renamed some theorems (the original names are also still available).
  2492 
  2493   expand_fun_below   ~> fun_below_iff
  2494   below_fun_ext      ~> fun_belowI
  2495   expand_cfun_eq     ~> cfun_eq_iff
  2496   ext_cfun           ~> cfun_eqI
  2497   expand_cfun_below  ~> cfun_below_iff
  2498   below_cfun_ext     ~> cfun_belowI
  2499   cont2cont_Rep_CFun ~> cont2cont_APP
  2500 
  2501 * The Abs and Rep functions for various types have changed names.
  2502 Related theorem names have also changed to match. INCOMPATIBILITY.
  2503 
  2504   Rep_CFun  ~> Rep_cfun
  2505   Abs_CFun  ~> Abs_cfun
  2506   Rep_Sprod ~> Rep_sprod
  2507   Abs_Sprod ~> Abs_sprod
  2508   Rep_Ssum  ~> Rep_ssum
  2509   Abs_Ssum  ~> Abs_ssum
  2510 
  2511 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2512 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2513 
  2514 * Various changes to bisimulation/coinduction with domain package:
  2515 
  2516   - Definitions of "bisim" constants no longer mention definedness.
  2517   - With mutual recursion, "bisim" predicate is now curried.
  2518   - With mutual recursion, each type gets a separate coind theorem.
  2519   - Variable names in bisim_def and coinduct rules have changed.
  2520 
  2521 INCOMPATIBILITY.
  2522 
  2523 * Case combinators generated by the domain package for type "foo" are
  2524 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2525 
  2526 * Several theorems have been renamed to more accurately reflect the
  2527 names of constants and types involved.  INCOMPATIBILITY.
  2528 
  2529   thelub_const    ~> lub_const
  2530   lub_const       ~> is_lub_const
  2531   thelubI         ~> lub_eqI
  2532   is_lub_lub      ~> is_lubD2
  2533   lubI            ~> is_lub_lub
  2534   unique_lub      ~> is_lub_unique
  2535   is_ub_lub       ~> is_lub_rangeD1
  2536   lub_bin_chain   ~> is_lub_bin_chain
  2537   lub_fun         ~> is_lub_fun
  2538   thelub_fun      ~> lub_fun
  2539   thelub_cfun     ~> lub_cfun
  2540   thelub_Pair     ~> lub_Pair
  2541   lub_cprod       ~> is_lub_prod
  2542   thelub_cprod    ~> lub_prod
  2543   minimal_cprod   ~> minimal_prod
  2544   inst_cprod_pcpo ~> inst_prod_pcpo
  2545   UU_I            ~> bottomI
  2546   compact_UU      ~> compact_bottom
  2547   deflation_UU    ~> deflation_bottom
  2548   finite_deflation_UU ~> finite_deflation_bottom
  2549 
  2550 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2551 
  2552   sq_ord_less_eq_trans ~> below_eq_trans
  2553   sq_ord_eq_less_trans ~> eq_below_trans
  2554   refl_less            ~> below_refl
  2555   trans_less           ~> below_trans
  2556   antisym_less         ~> below_antisym
  2557   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2558   box_less             ~> box_below
  2559   rev_trans_less       ~> rev_below_trans
  2560   not_less2not_eq      ~> not_below2not_eq
  2561   less_UU_iff          ~> below_UU_iff
  2562   flat_less_iff        ~> flat_below_iff
  2563   adm_less             ~> adm_below
  2564   adm_not_less         ~> adm_not_below
  2565   adm_compact_not_less ~> adm_compact_not_below
  2566   less_fun_def         ~> below_fun_def
  2567   expand_fun_less      ~> fun_below_iff
  2568   less_fun_ext         ~> fun_belowI
  2569   less_discr_def       ~> below_discr_def
  2570   discr_less_eq        ~> discr_below_eq
  2571   less_unit_def        ~> below_unit_def
  2572   less_cprod_def       ~> below_prod_def
  2573   prod_lessI           ~> prod_belowI
  2574   Pair_less_iff        ~> Pair_below_iff
  2575   fst_less_iff         ~> fst_below_iff
  2576   snd_less_iff         ~> snd_below_iff
  2577   expand_cfun_less     ~> cfun_below_iff
  2578   less_cfun_ext        ~> cfun_belowI
  2579   injection_less       ~> injection_below
  2580   less_up_def          ~> below_up_def
  2581   not_Iup_less         ~> not_Iup_below
  2582   Iup_less             ~> Iup_below
  2583   up_less              ~> up_below
  2584   Def_inject_less_eq   ~> Def_below_Def
  2585   Def_less_is_eq       ~> Def_below_iff
  2586   spair_less_iff       ~> spair_below_iff
  2587   less_sprod           ~> below_sprod
  2588   spair_less           ~> spair_below
  2589   sfst_less_iff        ~> sfst_below_iff
  2590   ssnd_less_iff        ~> ssnd_below_iff
  2591   fix_least_less       ~> fix_least_below
  2592   dist_less_one        ~> dist_below_one
  2593   less_ONE             ~> below_ONE
  2594   ONE_less_iff         ~> ONE_below_iff
  2595   less_sinlD           ~> below_sinlD
  2596   less_sinrD           ~> below_sinrD
  2597 
  2598 
  2599 *** FOL and ZF ***
  2600 
  2601 * All constant names are now qualified internally and use proper
  2602 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2603 
  2604 
  2605 *** ML ***
  2606 
  2607 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2608 Fail if the argument is false.  Due to inlining the source position of
  2609 failed assertions is included in the error output.
  2610 
  2611 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2612 text is in practice always evaluated with a stable theory checkpoint.
  2613 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2614 
  2615 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2616 the current context, not any accidental theory loader state as before.
  2617 Potential INCOMPATIBILITY, subtle change in semantics.
  2618 
  2619 * Syntax.pretty_priority (default 0) configures the required priority
  2620 of pretty-printed output and thus affects insertion of parentheses.
  2621 
  2622 * Syntax.default_root (default "any") configures the inner syntax
  2623 category (nonterminal symbol) for parsing of terms.
  2624 
  2625 * Former exception Library.UnequalLengths now coincides with
  2626 ListPair.UnequalLengths.
  2627 
  2628 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2629 main functionality is provided by structure Simplifier.
  2630 
  2631 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2632 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2633 almost all situations.
  2634 
  2635 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2636 See implementation manual for further details on exceptions in
  2637 Isabelle/ML.
  2638 
  2639 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2640 meaning.
  2641 
  2642 * Renamed structure PureThy to Pure_Thy and moved most of its
  2643 operations to structure Global_Theory, to emphasize that this is
  2644 rarely-used global-only stuff.
  2645 
  2646 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2647 instead (or tracing for high-volume output).
  2648 
  2649 * Configuration option show_question_marks only affects regular pretty
  2650 printing of types and terms, not raw Term.string_of_vname.
  2651 
  2652 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2653 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2654 @{thms} for most purposes.
  2655 
  2656 * ML structure Unsynchronized is never opened, not even in Isar
  2657 interaction mode as before.  Old Unsynchronized.set etc. have been
  2658 discontinued -- use plain := instead.  This should be *rare* anyway,
  2659 since modern tools always work via official context data, notably
  2660 configuration options.
  2661 
  2662 * Parallel and asynchronous execution requires special care concerning
  2663 interrupts.  Structure Exn provides some convenience functions that
  2664 avoid working directly with raw Interrupt.  User code must not absorb
  2665 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2666 followed by re-raising of the original exception.  Another common
  2667 source of mistakes are "handle _" patterns, which make the meaning of
  2668 the program subject to physical effects of the environment.
  2669 
  2670 
  2671 
  2672 New in Isabelle2009-2 (June 2010)
  2673 ---------------------------------
  2674 
  2675 *** General ***
  2676 
  2677 * Authentic syntax for *all* logical entities (type classes, type
  2678 constructors, term constants): provides simple and robust
  2679 correspondence between formal entities and concrete syntax.  Within
  2680 the parse tree / AST representations, "constants" are decorated by
  2681 their category (class, type, const) and spelled out explicitly with
  2682 their full internal name.
  2683 
  2684 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2685 and translations (translation rules and translation functions in ML).
  2686 Some hints on upgrading:
  2687 
  2688   - Many existing uses of 'syntax' and 'translations' can be replaced
  2689     by more modern 'type_notation', 'notation' and 'abbreviation',
  2690     which are independent of this issue.
  2691 
  2692   - 'translations' require markup within the AST; the term syntax
  2693     provides the following special forms:
  2694 
  2695       CONST c   -- produces syntax version of constant c from context
  2696       XCONST c  -- literally c, checked as constant from context
  2697       c         -- literally c, if declared by 'syntax'
  2698 
  2699     Plain identifiers are treated as AST variables -- occasionally the
  2700     system indicates accidental variables via the error "rhs contains
  2701     extra variables".
  2702 
  2703     Type classes and type constructors are marked according to their
  2704     concrete syntax.  Some old translations rules need to be written
  2705     for the "type" category, using type constructor application
  2706     instead of pseudo-term application of the default category
  2707     "logic".
  2708 
  2709   - 'parse_translation' etc. in ML may use the following
  2710     antiquotations:
  2711 
  2712       @{class_syntax c}   -- type class c within parse tree / AST
  2713       @{term_syntax c}    -- type constructor c within parse tree / AST
  2714       @{const_syntax c}   -- ML version of "CONST c" above
  2715       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2716 
  2717   - Literal types within 'typed_print_translations', i.e. those *not*
  2718     represented as pseudo-terms are represented verbatim.  Use @{class
  2719     c} or @{type_name c} here instead of the above syntax
  2720     antiquotations.
  2721 
  2722 Note that old non-authentic syntax was based on unqualified base
  2723 names, so all of the above "constant" names would coincide.  Recall
  2724 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2725 diagnose syntax problems.
  2726 
  2727 * Type constructors admit general mixfix syntax, not just infix.
  2728 
  2729 * Concrete syntax may be attached to local entities without a proof
  2730 body, too.  This works via regular mixfix annotations for 'fix',
  2731 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2732 similar to the 'notation' command in theory specifications.
  2733 
  2734 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2735 need to specify constant name and syntax separately.  Internal ML
  2736 datatype constructors have been renamed from InfixName to Infix etc.
  2737 Minor INCOMPATIBILITY.
  2738 
  2739 * Schematic theorem statements need to be explicitly markup as such,
  2740 via commands 'schematic_lemma', 'schematic_theorem',
  2741 'schematic_corollary'.  Thus the relevance of the proof is made
  2742 syntactically clear, which impacts performance in a parallel or
  2743 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2744 
  2745 * Use of cumulative prems via "!" in some proof methods has been
  2746 discontinued (old legacy feature).
  2747 
  2748 * References 'trace_simp' and 'debug_simp' have been replaced by
  2749 configuration options stored in the context. Enabling tracing (the
  2750 case of debugging is similar) in proofs works via
  2751 
  2752   using [[trace_simp = true]]
  2753 
  2754 Tracing is then active for all invocations of the simplifier in
  2755 subsequent goal refinement steps. Tracing may also still be enabled or
  2756 disabled via the ProofGeneral settings menu.
  2757 
  2758 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2759 'hide_fact' replace the former 'hide' KIND command.  Minor
  2760 INCOMPATIBILITY.
  2761 
  2762 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2763 more efficient than combinations with -q1 or -q2.
  2764 
  2765 
  2766 *** Pure ***
  2767 
  2768 * Proofterms record type-class reasoning explicitly, using the
  2769 "unconstrain" operation internally.  This eliminates all sort
  2770 constraints from a theorem and proof, introducing explicit
  2771 OFCLASS-premises.  On the proof term level, this operation is
  2772 automatically applied at theorem boundaries, such that closed proofs
  2773 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2774 inspect proof terms.
  2775 
  2776 * Local theory specifications may depend on extra type variables that
  2777 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2778 are added internally.  For example:
  2779 
  2780   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2781 
  2782 * Predicates of locales introduced by classes carry a mandatory
  2783 "class" prefix.  INCOMPATIBILITY.
  2784 
  2785 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2786 
  2787 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2788 'class' instead.
  2789 
  2790 * Command 'code_reflect' allows to incorporate generated ML code into
  2791 runtime environment; replaces immature code_datatype antiquotation.
  2792 INCOMPATIBILITY.
  2793 
  2794 * Code generator: simple concept for abstract datatypes obeying
  2795 invariants.
  2796 
  2797 * Code generator: details of internal data cache have no impact on the
  2798 user space functionality any longer.
  2799 
  2800 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2801 subgoals.  This is more appropriate for interpretations with 'where'.
  2802 INCOMPATIBILITY.
  2803 
  2804 * Command 'example_proof' opens an empty proof body.  This allows to
  2805 experiment with Isar, without producing any persistent result.
  2806 
  2807 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2808 within a local theory context, with explicit checking of the
  2809 constructors involved (in contrast to the raw 'syntax' versions).
  2810 
  2811 * Commands 'types' and 'typedecl' now work within a local theory
  2812 context -- without introducing dependencies on parameters or
  2813 assumptions, which is not possible in Isabelle/Pure.
  2814 
  2815 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2816 within a local theory context.  Minor INCOMPATIBILITY.
  2817 
  2818 
  2819 *** HOL ***
  2820 
  2821 * Command 'typedef' now works within a local theory context -- without
  2822 introducing dependencies on parameters or assumptions, which is not
  2823 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2824 contain multiple interpretations of local typedefs (with different
  2825 non-emptiness proofs), even in a global theory context.
  2826 
  2827 * New package for quotient types.  Commands 'quotient_type' and
  2828 'quotient_definition' may be used for defining types and constants by
  2829 quotient constructions.  An example is the type of integers created by
  2830 quotienting pairs of natural numbers:
  2831 
  2832   fun
  2833     intrel :: "(nat * nat) => (nat * nat) => bool"
  2834   where
  2835     "intrel (x, y) (u, v) = (x + v = u + y)"
  2836 
  2837   quotient_type int = "nat * nat" / intrel
  2838     by (auto simp add: equivp_def expand_fun_eq)
  2839 
  2840   quotient_definition
  2841     "0::int" is "(0::nat, 0::nat)"
  2842 
  2843 The method "lifting" can be used to lift of theorems from the
  2844 underlying "raw" type to the quotient type.  The example
  2845 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2846 construction and provides a reasoning infrastructure for finite sets.
  2847 
  2848 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2849 clash with new theory Quotient in Main HOL.
  2850 
  2851 * Moved the SMT binding into the main HOL session, eliminating
  2852 separate HOL-SMT session.
  2853 
  2854 * List membership infix mem operation is only an input abbreviation.
  2855 INCOMPATIBILITY.
  2856 
  2857 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2858 for future developements; former Library/Word.thy is still present in
  2859 the AFP entry RSAPPS.
  2860 
  2861 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2862 longer shadowed.  INCOMPATIBILITY.
  2863 
  2864 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2865 INCOMPATIBILITY.
  2866 
  2867 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2868 INCOMPATIBILITY.
  2869 
  2870 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2871 instead.  INCOMPATIBILITY.
  2872 
  2873 * Dropped several real-specific versions of lemmas about floor and
  2874 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2875 instead.  INCOMPATIBILITY.
  2876 
  2877   floor_number_of_eq         ~> floor_number_of
  2878   le_floor_eq_number_of      ~> number_of_le_floor
  2879   le_floor_eq_zero           ~> zero_le_floor
  2880   le_floor_eq_one            ~> one_le_floor
  2881   floor_less_eq_number_of    ~> floor_less_number_of
  2882   floor_less_eq_zero         ~> floor_less_zero
  2883   floor_less_eq_one          ~> floor_less_one
  2884   less_floor_eq_number_of    ~> number_of_less_floor
  2885   less_floor_eq_zero         ~> zero_less_floor
  2886   less_floor_eq_one          ~> one_less_floor
  2887   floor_le_eq_number_of      ~> floor_le_number_of
  2888   floor_le_eq_zero           ~> floor_le_zero
  2889   floor_le_eq_one            ~> floor_le_one
  2890   floor_subtract_number_of   ~> floor_diff_number_of
  2891   floor_subtract_one         ~> floor_diff_one
  2892   ceiling_number_of_eq       ~> ceiling_number_of
  2893   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2894   ceiling_le_zero_eq         ~> ceiling_le_zero
  2895   ceiling_le_eq_one          ~> ceiling_le_one
  2896   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2897   less_ceiling_eq_zero       ~> zero_less_ceiling
  2898   less_ceiling_eq_one        ~> one_less_ceiling
  2899   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2900   ceiling_less_eq_zero       ~> ceiling_less_zero
  2901   ceiling_less_eq_one        ~> ceiling_less_one
  2902   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2903   le_ceiling_eq_zero         ~> zero_le_ceiling
  2904   le_ceiling_eq_one          ~> one_le_ceiling
  2905   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2906   ceiling_subtract_one       ~> ceiling_diff_one
  2907 
  2908 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2909 application of the various fold combinators on finite sets.
  2910 
  2911 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2912 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2913 implementation.  INCOMPATIBILTY.
  2914 
  2915 * Theory Library/Coinductive_List has been removed -- superseded by
  2916 AFP/thys/Coinductive.
  2917 
  2918 * Theory PReal, including the type "preal" and related operations, has
  2919 been removed.  INCOMPATIBILITY.
  2920 
  2921 * Real: new development using Cauchy Sequences.
  2922 
  2923 * Split off theory "Big_Operators" containing setsum, setprod,
  2924 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2925 
  2926 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2927 "Int" etc.  INCOMPATIBILITY.
  2928 
  2929 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2930 
  2931 * New set of rules "ac_simps" provides combined assoc / commute
  2932 rewrites for all interpretations of the appropriate generic locales.
  2933 
  2934 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2935 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2936 appropriate and more consistent names suitable for name prefixes
  2937 within the HOL theories.  INCOMPATIBILITY.
  2938 
  2939 * Some generic constants have been put to appropriate theories:
  2940   - less_eq, less: Orderings
  2941   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2942   - inverse, divide: Rings
  2943 INCOMPATIBILITY.
  2944 
  2945 * More consistent naming of type classes involving orderings (and
  2946 lattices):
  2947 
  2948     lower_semilattice                   ~> semilattice_inf
  2949     upper_semilattice                   ~> semilattice_sup
  2950 
  2951     dense_linear_order                  ~> dense_linorder
  2952 
  2953     pordered_ab_group_add               ~> ordered_ab_group_add
  2954     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2955     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2956     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2957     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2958     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2959     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2960     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2961     pordered_comm_ring                  ~> ordered_comm_ring
  2962     pordered_comm_semiring              ~> ordered_comm_semiring
  2963     pordered_ring                       ~> ordered_ring
  2964     pordered_ring_abs                   ~> ordered_ring_abs
  2965     pordered_semiring                   ~> ordered_semiring
  2966 
  2967     ordered_ab_group_add                ~> linordered_ab_group_add
  2968     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2969     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2970     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2971     ordered_field                       ~> linordered_field
  2972     ordered_field_no_lb                 ~> linordered_field_no_lb
  2973     ordered_field_no_ub                 ~> linordered_field_no_ub
  2974     ordered_field_dense_linear_order    ~> dense_linordered_field
  2975     ordered_idom                        ~> linordered_idom
  2976     ordered_ring                        ~> linordered_ring
  2977     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2978     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2979     ordered_ring_strict                 ~> linordered_ring_strict
  2980     ordered_semidom                     ~> linordered_semidom
  2981     ordered_semiring                    ~> linordered_semiring
  2982     ordered_semiring_1                  ~> linordered_semiring_1
  2983     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2984     ordered_semiring_strict             ~> linordered_semiring_strict
  2985 
  2986   The following slightly odd type classes have been moved to a
  2987   separate theory Library/Lattice_Algebras:
  2988 
  2989     lordered_ab_group_add               ~> lattice_ab_group_add
  2990     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2991     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2992     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2993     lordered_ring                       ~> lattice_ring
  2994 
  2995 INCOMPATIBILITY.
  2996 
  2997 * Refined field classes:
  2998   - classes division_ring_inverse_zero, field_inverse_zero,
  2999     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3000     subsumes former division_by_zero class;
  3001   - numerous lemmas have been ported from field to division_ring.
  3002 INCOMPATIBILITY.
  3003 
  3004 * Refined algebra theorem collections:
  3005   - dropped theorem group group_simps, use algebra_simps instead;
  3006   - dropped theorem group ring_simps, use field_simps instead;
  3007   - proper theorem collection field_simps subsumes former theorem
  3008     groups field_eq_simps and field_simps;
  3009   - dropped lemma eq_minus_self_iff which is a duplicate for
  3010     equal_neg_zero.
  3011 INCOMPATIBILITY.
  3012 
  3013 * Theory Finite_Set and List: some lemmas have been generalized from
  3014 sets to lattices:
  3015 
  3016   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3017   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3018   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3019   union_Union_fold_union        ~> sup_Sup_fold_sup
  3020   Inter_fold_inter              ~> Inf_fold_inf
  3021   Union_fold_union              ~> Sup_fold_sup
  3022   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3023   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3024   INTER_fold_inter              ~> INFI_fold_inf
  3025   UNION_fold_union              ~> SUPR_fold_sup
  3026 
  3027 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3028 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3029 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3030 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3031 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3032 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3033 
  3034 * Reorganized theory Multiset: swapped notation of pointwise and
  3035 multiset order:
  3036 
  3037   - pointwise ordering is instance of class order with standard syntax
  3038     <= and <;
  3039   - multiset ordering has syntax <=# and <#; partial order properties
  3040     are provided by means of interpretation with prefix
  3041     multiset_order;
  3042   - less duplication, less historical organization of sections,
  3043     conversion from associations lists to multisets, rudimentary code
  3044     generation;
  3045   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3046     if needed.
  3047 
  3048 Renamed:
  3049 
  3050   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3051   multi_count_ext  ~>  multiset_ext
  3052   diff_union_inverse2  ~>  diff_union_cancelR
  3053 
  3054 INCOMPATIBILITY.
  3055 
  3056 * Theory Permutation: replaced local "remove" by List.remove1.
  3057 
  3058 * Code generation: ML and OCaml code is decorated with signatures.
  3059 
  3060 * Theory List: added transpose.
  3061 
  3062 * Library/Nat_Bijection.thy is a collection of bijective functions
  3063 between nat and other types, which supersedes the older libraries
  3064 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3065 
  3066   Constants:
  3067   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3068   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3069   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3070   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3071   Countable.pair_encode           ~> prod_encode
  3072   NatIso.prod2nat                 ~> prod_encode
  3073   NatIso.nat2prod                 ~> prod_decode
  3074   NatIso.sum2nat                  ~> sum_encode
  3075   NatIso.nat2sum                  ~> sum_decode
  3076   NatIso.list2nat                 ~> list_encode
  3077   NatIso.nat2list                 ~> list_decode
  3078   NatIso.set2nat                  ~> set_encode
  3079   NatIso.nat2set                  ~> set_decode
  3080 
  3081   Lemmas:
  3082   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3083   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3084   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3085   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3086   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3087   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3088   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3089   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3090   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3091   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3092   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3093   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3094   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3095 
  3096 * Sledgehammer:
  3097   - Renamed ATP commands:
  3098     atp_info     ~> sledgehammer running_atps
  3099     atp_kill     ~> sledgehammer kill_atps
  3100     atp_messages ~> sledgehammer messages
  3101     atp_minimize ~> sledgehammer minimize
  3102     print_atps   ~> sledgehammer available_atps
  3103     INCOMPATIBILITY.
  3104   - Added user's manual ("isabelle doc sledgehammer").
  3105   - Added option syntax and "sledgehammer_params" to customize
  3106     Sledgehammer's behavior.  See the manual for details.
  3107   - Modified the Isar proof reconstruction code so that it produces
  3108     direct proofs rather than proofs by contradiction.  (This feature
  3109     is still experimental.)
  3110   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3111     full-typed mode.
  3112   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3113 
  3114 * Nitpick:
  3115   - Added and implemented "binary_ints" and "bits" options.
  3116   - Added "std" option and implemented support for nonstandard models.
  3117   - Added and implemented "finitize" option to improve the precision
  3118     of infinite datatypes based on a monotonicity analysis.
  3119   - Added support for quotient types.
  3120   - Added support for "specification" and "ax_specification"
  3121     constructs.
  3122   - Added support for local definitions (for "function" and
  3123     "termination" proofs).
  3124   - Added support for term postprocessors.
  3125   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3126   - Improved efficiency of "destroy_constrs" optimization.
  3127   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3128     record getters.
  3129   - Fixed soundness bug related to higher-order constructors.
  3130   - Fixed soundness bug when "full_descrs" is enabled.
  3131   - Improved precision of set constructs.
  3132   - Added "atoms" option.
  3133   - Added cache to speed up repeated Kodkod invocations on the same
  3134     problems.
  3135   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3136     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3137     "SAT4J_Light".  INCOMPATIBILITY.
  3138   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3139     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3140   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3141 
  3142 * Method "induct" now takes instantiations of the form t, where t is not
  3143   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3144   If this is not intended, t has to be enclosed in parentheses.
  3145   By default, the equalities generated by definitional instantiations
  3146   are pre-simplified, which may cause parameters of inductive cases
  3147   to disappear, or may even delete some of the inductive cases.
  3148   Use "induct (no_simp)" instead of "induct" to restore the old
  3149   behaviour. The (no_simp) option is also understood by the "cases"
  3150   and "nominal_induct" methods, which now perform pre-simplification, too.
  3151   INCOMPATIBILITY.
  3152 
  3153 
  3154 *** HOLCF ***
  3155 
  3156 * Variable names in lemmas generated by the domain package have
  3157 changed; the naming scheme is now consistent with the HOL datatype
  3158 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3159 
  3160 * The domain package no longer defines the function "foo_copy" for
  3161 recursive domain "foo".  The reach lemma is now stated directly in
  3162 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3163 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3164 
  3165 * Most definedness lemmas generated by the domain package (previously
  3166 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3167 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3168 Proofs that used definedness lemmas as intro rules may break,
  3169 potential INCOMPATIBILITY.
  3170 
  3171 * Induction and casedist rules generated by the domain package now
  3172 declare proper case_names (one called "bottom", and one named for each
  3173 constructor).  INCOMPATIBILITY.
  3174 
  3175 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3176 rules are generated for each domain, INCOMPATIBILITY.
  3177 
  3178   foo_bar.reach       ~> foo.reach  bar.reach
  3179   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3180 
  3181 * Some lemmas generated by the domain package have been renamed for
  3182 consistency with the datatype package, INCOMPATIBILITY.
  3183 
  3184   foo.ind        ~> foo.induct
  3185   foo.finite_ind ~> foo.finite_induct
  3186   foo.coind      ~> foo.coinduct
  3187   foo.casedist   ~> foo.exhaust
  3188   foo.exhaust    ~> foo.nchotomy
  3189 
  3190 * For consistency with other definition packages, the fixrec package
  3191 now generates qualified theorem names, INCOMPATIBILITY.
  3192 
  3193   foo_simps  ~> foo.simps
  3194   foo_unfold ~> foo.unfold
  3195   foo_induct ~> foo.induct
  3196 
  3197 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3198 method and internal fixrec proofs now use the default simpset instead.
  3199 INCOMPATIBILITY.
  3200 
  3201 * The "contlub" predicate has been removed.  Proof scripts should use
  3202 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3203 
  3204 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3205 
  3206 * The constants cpair, cfst, and csnd have been removed in favor of
  3207 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3208 
  3209 
  3210 *** ML ***
  3211 
  3212 * Antiquotations for basic formal entities:
  3213 
  3214     @{class NAME}         -- type class
  3215     @{class_syntax NAME}  -- syntax representation of the above
  3216 
  3217     @{type_name NAME}     -- logical type
  3218     @{type_abbrev NAME}   -- type abbreviation
  3219     @{nonterminal NAME}   -- type of concrete syntactic category
  3220     @{type_syntax NAME}   -- syntax representation of any of the above
  3221 
  3222     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3223     @{const_abbrev NAME}  -- abbreviated constant
  3224     @{const_syntax NAME}  -- syntax representation of any of the above
  3225 
  3226 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3227 syntax constant (cf. 'syntax' command).
  3228 
  3229 * Antiquotation @{make_string} inlines a function to print arbitrary
  3230 values similar to the ML toplevel.  The result is compiler dependent
  3231 and may fall back on "?" in certain situations.
  3232 
  3233 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3234 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3235 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3236 not work with the asynchronous Isar document model.
  3237 
  3238 * Configuration options now admit dynamic default values, depending on
  3239 the context or even global references.
  3240 
  3241 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3242 uses an efficient external library if available (for Poly/ML).
  3243 
  3244 * Renamed some important ML structures, while keeping the old names
  3245 for some time as aliases within the structure Legacy:
  3246 
  3247   OuterKeyword  ~>  Keyword
  3248   OuterLex      ~>  Token
  3249   OuterParse    ~>  Parse
  3250   OuterSyntax   ~>  Outer_Syntax
  3251   PrintMode     ~>  Print_Mode
  3252   SpecParse     ~>  Parse_Spec
  3253   ThyInfo       ~>  Thy_Info
  3254   ThyLoad       ~>  Thy_Load
  3255   ThyOutput     ~>  Thy_Output
  3256   TypeInfer     ~>  Type_Infer
  3257 
  3258 Note that "open Legacy" simplifies porting of sources, but forgetting
  3259 to remove it again will complicate porting again in the future.
  3260 
  3261 * Most operations that refer to a global context are named
  3262 accordingly, e.g. Simplifier.global_context or
  3263 ProofContext.init_global.  There are some situations where a global
  3264 context actually works, but under normal circumstances one needs to
  3265 pass the proper local context through the code!
  3266 
  3267 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3268 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3269 Pretty.pp argument to merge, which is absent in the standard
  3270 Theory_Data version.
  3271 
  3272 * Sorts.certify_sort and derived "cert" operations for types and terms
  3273 no longer minimize sorts.  Thus certification at the boundary of the
  3274 inference kernel becomes invariant under addition of class relations,
  3275 which is an important monotonicity principle.  Sorts are now minimized
  3276 in the syntax layer only, at the boundary between the end-user and the
  3277 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3278 explicitly in rare situations.
  3279 
  3280 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3281 emphasize that this is in no way a standard operation.
  3282 INCOMPATIBILITY.
  3283 
  3284 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3285 usual for resolution.  Rare INCOMPATIBILITY.
  3286 
  3287 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3288 to emphasize that these only work in a global situation (which is
  3289 quite rare).
  3290 
  3291 * Curried take and drop in library.ML; negative length is interpreted
  3292 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3293 
  3294 * Proof terms: type substitutions on proof constants now use canonical
  3295 order of type variables.  INCOMPATIBILITY for tools working with proof
  3296 terms.
  3297 
  3298 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3299 may no longer carry premises.  User-level specifications are
  3300 transformed accordingly by Thm.add_axiom/add_def.
  3301 
  3302 
  3303 *** System ***
  3304 
  3305 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3306 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3307 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3308 
  3309 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3310 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3311 respectively.
  3312 
  3313 * Old lib/scripts/polyml-platform is superseded by the
  3314 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3315 variant, even on a 64 bit machine.  The following example setting
  3316 prefers 64 bit if available:
  3317 
  3318   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3319 
  3320 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3321 Isabelle/Scala layer for advanced prover interaction and integration.
  3322 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3323 component.
  3324 
  3325 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3326 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3327 similar to the default assignment of the document preparation system
  3328 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3329 provides some operations for direct access to the font without asking
  3330 the user for manual installation.
  3331 
  3332 
  3333 
  3334 New in Isabelle2009-1 (December 2009)
  3335 -------------------------------------
  3336 
  3337 *** General ***
  3338 
  3339 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3340 one backslash should be used, even in ML sources.
  3341 
  3342 
  3343 *** Pure ***
  3344 
  3345 * Locale interpretation propagates mixins along the locale hierarchy.
  3346 The currently only available mixins are the equations used to map
  3347 local definitions to terms of the target domain of an interpretation.
  3348 
  3349 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3350 loc" to print all interpretations of locale "loc" in the theory.
  3351 Interpretations in proofs are not shown.
  3352 
  3353 * Thoroughly revised locales tutorial.  New section on conditional
  3354 interpretation.
  3355 
  3356 * On instantiation of classes, remaining undefined class parameters
  3357 are formally declared.  INCOMPATIBILITY.
  3358 
  3359 
  3360 *** Document preparation ***
  3361 
  3362 * New generalized style concept for printing terms: @{foo (style) ...}
  3363 instead of @{foo_style style ...}  (old form is still retained for
  3364 backward compatibility).  Styles can be also applied for
  3365 antiquotations prop, term_type and typeof.
  3366 
  3367 
  3368 *** HOL ***
  3369 
  3370 * New proof method "smt" for a combination of first-order logic with
  3371 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3372 fixed-size bitvectors; there is also basic support for higher-order
  3373 features (esp. lambda abstractions).  It is an incomplete decision
  3374 procedure based on external SMT solvers using the oracle mechanism;
  3375 for the SMT solver Z3, this method is proof-producing.  Certificates
  3376 are provided to avoid calling the external solvers solely for
  3377 re-checking proofs.  Due to a remote SMT service there is no need for
  3378 installing SMT solvers locally.  See src/HOL/SMT.
  3379 
  3380 * New commands to load and prove verification conditions generated by
  3381 the Boogie program verifier or derived systems (e.g. the Verifying C
  3382 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3383 
  3384 * New counterexample generator tool 'nitpick' based on the Kodkod
  3385 relational model finder.  See src/HOL/Tools/Nitpick and
  3386 src/HOL/Nitpick_Examples.
  3387 
  3388 * New commands 'code_pred' and 'values' to invoke the predicate
  3389 compiler and to enumerate values of inductive predicates.
  3390 
  3391 * A tabled implementation of the reflexive transitive closure.
  3392 
  3393 * New implementation of quickcheck uses generic code generator;
  3394 default generators are provided for all suitable HOL types, records
  3395 and datatypes.  Old quickcheck can be re-activated importing theory
  3396 Library/SML_Quickcheck.
  3397 
  3398 * New testing tool Mirabelle for automated proof tools.  Applies
  3399 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3400 every proof step in a theory.  To be used in batch mode via the
  3401 "mirabelle" utility.
  3402 
  3403 * New proof method "sos" (sum of squares) for nonlinear real
  3404 arithmetic (originally due to John Harison). It requires theory
  3405 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3406 works well in practice on quantifier-free real arithmetic with +, -,
  3407 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3408 inequalities between polynomials.  It makes use of external
  3409 semidefinite programming solvers.  Method "sos" generates a
  3410 certificate that can be pasted into the proof thus avoiding the need
  3411 to call an external tool every time the proof is checked.  See
  3412 src/HOL/Library/Sum_Of_Squares.
  3413 
  3414 * New method "linarith" invokes existing linear arithmetic decision
  3415 procedure only.
  3416 
  3417 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3418 
  3419 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3420 Causes full type information to be output to the ATPs.  This slows
  3421 ATPs down considerably but eliminates a source of unsound "proofs"
  3422 that fail later.
  3423 
  3424 * New method "metisFT": A version of metis that uses full type
  3425 information in order to avoid failures of proof reconstruction.
  3426 
  3427 * New evaluator "approximate" approximates an real valued term using
  3428 the same method as the approximation method.
  3429 
  3430 * Method "approximate" now supports arithmetic expressions as
  3431 boundaries of intervals and implements interval splitting and Taylor
  3432 series expansion.
  3433 
  3434 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3435 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3436 
  3437 * New theory SupInf of the supremum and infimum operators for sets of
  3438 reals.
  3439 
  3440 * New theory Probability, which contains a development of measure
  3441 theory, eventually leading to Lebesgue integration and probability.
  3442 
  3443 * Extended Multivariate Analysis to include derivation and Brouwer's
  3444 fixpoint theorem.
  3445 
  3446 * Reorganization of number theory, INCOMPATIBILITY:
  3447   - new number theory development for nat and int, in theories Divides
  3448     and GCD as well as in new session Number_Theory
  3449   - some constants and facts now suffixed with _nat and _int
  3450     accordingly
  3451   - former session NumberTheory now named Old_Number_Theory, including
  3452     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3453   - moved theory Pocklington from src/HOL/Library to
  3454     src/HOL/Old_Number_Theory
  3455 
  3456 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3457 lcm of finite and infinite sets. It is shown that they form a complete
  3458 lattice.
  3459 
  3460 * Class semiring_div requires superclass no_zero_divisors and proof of
  3461 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3462 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3463 generalized to class semiring_div, subsuming former theorems
  3464 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3465 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3466 INCOMPATIBILITY.
  3467 
  3468 * Refinements to lattice classes and sets:
  3469   - less default intro/elim rules in locale variant, more default
  3470     intro/elim rules in class variant: more uniformity
  3471   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3472     le_inf_iff
  3473   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3474     sup_aci)
  3475   - renamed ACI to inf_sup_aci
  3476   - new class "boolean_algebra"
  3477   - class "complete_lattice" moved to separate theory
  3478     "Complete_Lattice"; corresponding constants (and abbreviations)
  3479     renamed and with authentic syntax:
  3480     Set.Inf ~>    Complete_Lattice.Inf
  3481     Set.Sup ~>    Complete_Lattice.Sup
  3482     Set.INFI ~>   Complete_Lattice.INFI
  3483     Set.SUPR ~>   Complete_Lattice.SUPR
  3484     Set.Inter ~>  Complete_Lattice.Inter
  3485     Set.Union ~>  Complete_Lattice.Union
  3486     Set.INTER ~>  Complete_Lattice.INTER
  3487     Set.UNION ~>  Complete_Lattice.UNION
  3488   - authentic syntax for
  3489     Set.Pow
  3490     Set.image
  3491   - mere abbreviations:
  3492     Set.empty               (for bot)
  3493     Set.UNIV                (for top)
  3494     Set.inter               (for inf, formerly Set.Int)
  3495     Set.union               (for sup, formerly Set.Un)
  3496     Complete_Lattice.Inter  (for Inf)
  3497     Complete_Lattice.Union  (for Sup)
  3498     Complete_Lattice.INTER  (for INFI)
  3499     Complete_Lattice.UNION  (for SUPR)
  3500   - object-logic definitions as far as appropriate
  3501 
  3502 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3503 Un_subset_iff are explicitly deleted as default simp rules; then also
  3504 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3505 deleted to achieve the desired effect.
  3506 
  3507 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3508 rules by default any longer; the same applies to min_max.inf_absorb1
  3509 etc.  INCOMPATIBILITY.
  3510 
  3511 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3512 pred_set_conv by default.  INCOMPATIBILITY.
  3513 
  3514 * Power operations on relations and functions are now one dedicated
  3515 constant "compow" with infix syntax "^^".  Power operation on
  3516 multiplicative monoids retains syntax "^" and is now defined generic
  3517 in class power.  INCOMPATIBILITY.
  3518 
  3519 * Relation composition "R O S" now has a more standard argument order:
  3520 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3521 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3522 break, since the O_assoc rule was not rewritten like this.  Fix using
  3523 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3524 
  3525 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3526 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3527 INCOMPATIBILITY.
  3528 
  3529 * Most rules produced by inductive and datatype package have mandatory
  3530 prefixes.  INCOMPATIBILITY.
  3531 
  3532 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3533 the attribute of the same name.  Each of the theorems in the list
  3534 DERIV_intros assumes composition with an additional function and
  3535 matches a variable to the derivative, which has to be solved by the
  3536 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3537 of most elementary terms.  Former Maclauren.DERIV_tac and
  3538 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3539 INCOMPATIBILITY.
  3540 
  3541 * Code generator attributes follow the usual underscore convention:
  3542     code_unfold     replaces    code unfold
  3543     code_post       replaces    code post
  3544     etc.
  3545   INCOMPATIBILITY.
  3546 
  3547 * Renamed methods:
  3548     sizechange -> size_change
  3549     induct_scheme -> induction_schema
  3550   INCOMPATIBILITY.
  3551 
  3552 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3553 INCOMPATIBILITY, use "undefined" directly.
  3554 
  3555 * Renamed theorems:
  3556     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3557     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3558     Suc_plus1 -> Suc_eq_plus1
  3559     *anti_sym -> *antisym*
  3560     vector_less_eq_def -> vector_le_def
  3561   INCOMPATIBILITY.
  3562 
  3563 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3564 INCOMPATIBILITY.
  3565 
  3566 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3567 INCOMPATIBILITY.
  3568 
  3569 
  3570 *** HOLCF ***
  3571 
  3572 * Theory Representable defines a class "rep" of domains that are
  3573 representable (via an ep-pair) in the universal domain type "udom".
  3574 Instances are provided for all type constructors defined in HOLCF.
  3575 
  3576 * The 'new_domain' command is a purely definitional version of the
  3577 domain package, for representable domains.  Syntax is identical to the
  3578 old domain package.  The 'new_domain' package also supports indirect
  3579 recursion using previously-defined type constructors.  See
  3580 src/HOLCF/ex/New_Domain.thy for examples.
  3581 
  3582 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3583 on the left-hand side of an equation, and then performs
  3584 simplification.  Rewriting is done using rules declared with the
  3585 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3586 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3587 
  3588 * The pattern-match compiler in 'fixrec' can now handle constructors
  3589 with HOL function types.  Pattern-match combinators for the Pair
  3590 constructor are pre-configured.
  3591 
  3592 * The 'fixrec' package now produces better fixed-point induction rules
  3593 for mutually-recursive definitions:  Induction rules have conclusions
  3594 of the form "P foo bar" instead of "P <foo, bar>".
  3595 
  3596 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3597 been renamed to "below".  The name "below" now replaces "less" in many
  3598 theorem names.  (Legacy theorem names using "less" are still supported
  3599 as well.)
  3600 
  3601 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3602 patterns can be used to generate strictness rules, or to make
  3603 functions more strict (much like the bang-patterns supported by the
  3604 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3605 examples.
  3606 
  3607 
  3608 *** ML ***
  3609 
  3610 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3611 errors and run-time exceptions, including detailed source positions.
  3612 
  3613 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3614 identified entries, with some additional information such as source
  3615 position, logical grouping etc.
  3616 
  3617 * Theory and context data is now introduced by the simplified and
  3618 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3619 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3620 copy operation) is still available for some time.
  3621 
  3622 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3623 provides a high-level programming interface to synchronized state
  3624 variables with atomic update.  This works via pure function
  3625 application within a critical section -- its runtime should be as
  3626 short as possible; beware of deadlocks if critical code is nested,
  3627 either directly or indirectly via other synchronized variables!
  3628 
  3629 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3630 wraps raw ML references, explicitly indicating their non-thread-safe
  3631 behaviour.  The Isar toplevel keeps this structure open, to
  3632 accommodate Proof General as well as quick and dirty interactive
  3633 experiments with references.
  3634 
  3635 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3636 parallel tactical reasoning.
  3637 
  3638 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3639 are similar to SUBPROOF, but are slightly more flexible: only the
  3640 specified parts of the subgoal are imported into the context, and the
  3641 body tactic may introduce new subgoals and schematic variables.
  3642 
  3643 * Old tactical METAHYPS, which does not observe the proof context, has
  3644 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3645 or Subgoal.FOCUS etc.
  3646 
  3647 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3648 functors have their own ML name space there is no point to mark them
  3649 separately.)  Minor INCOMPATIBILITY.
  3650 
  3651 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3652 
  3653 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3654 INCOMPATIBILITY.
  3655 
  3656 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3657 mode, which avoids critical setmp.
  3658 
  3659 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3660 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3661 Attrib/Method.setup introduced in Isabelle2009.
  3662 
  3663 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3664 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3665 last resort.  INCOMPATIBILITY.
  3666 
  3667 * Display.pretty_thm now requires a proper context (cf. former
  3668 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3669 or even Display.pretty_thm_without_context as last resort.
  3670 INCOMPATIBILITY.
  3671 
  3672 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3673 Syntax.pretty_typ/term directly, preferably with proper context
  3674 instead of global theory.
  3675 
  3676 
  3677 *** System ***
  3678 
  3679 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3680 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3681 usedir option -q.
  3682 
  3683 * Support for additional "Isabelle components" via etc/components, see
  3684 also the system manual.
  3685 
  3686 * The isabelle makeall tool now operates on all components with
  3687 IsaMakefile, not just hardwired "logics".
  3688 
  3689 * Removed "compress" option from isabelle-process and isabelle usedir;
  3690 this is always enabled.
  3691 
  3692 * Discontinued support for Poly/ML 4.x versions.
  3693 
  3694 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3695 on a given logic image.  This requires the lighttpd webserver and is
  3696 currently supported on Linux only.
  3697 
  3698 
  3699 
  3700 New in Isabelle2009 (April 2009)
  3701 --------------------------------
  3702 
  3703 *** General ***
  3704 
  3705 * Simplified main Isabelle executables, with less surprises on
  3706 case-insensitive file-systems (such as Mac OS).
  3707 
  3708   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3709     "isatool."
  3710 
  3711   - The former "isabelle" alias for "isabelle-process" has been
  3712     removed (should rarely occur to regular users).
  3713 
  3714   - The former "isabelle-interface" and its alias "Isabelle" have been
  3715     removed (interfaces are now regular Isabelle tools).
  3716 
  3717 Within scripts and make files, the Isabelle environment variables
  3718 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3719 respectively.  (The latter are still available as legacy feature.)
  3720 
  3721 The old isabelle-interface wrapper could react in confusing ways if
  3722 the interface was uninstalled or changed otherwise.  Individual
  3723 interface tool configuration is now more explicit, see also the
  3724 Isabelle system manual.  In particular, Proof General is now available
  3725 via "isabelle emacs".
  3726 
  3727 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3728 purge installed copies of Isabelle executables and re-run "isabelle
  3729 install -p ...", or use symlinks.
  3730 
  3731 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3732 old ~/isabelle, which was slightly non-standard and apt to cause
  3733 surprises on case-insensitive file-systems (such as Mac OS).
  3734 
  3735 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3736 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3737 care is required when using older releases of Isabelle.  Note that
  3738 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3739 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3740 
  3741 * Proofs of fully specified statements are run in parallel on
  3742 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3743 a regular 4-core machine, if the initial heap space is made reasonably
  3744 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3745 
  3746 * The main reference manuals ("isar-ref", "implementation", and
  3747 "system") have been updated and extended.  Formally checked references
  3748 as hyperlinks are now available uniformly.
  3749 
  3750 
  3751 *** Pure ***
  3752 
  3753 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3754 respects.  The most important changes are listed below.  See the
  3755 Tutorial on Locales ("locales" manual) for details.
  3756 
  3757 - In locale expressions, instantiation replaces renaming.  Parameters
  3758 must be declared in a for clause.  To aid compatibility with previous
  3759 parameter inheritance, in locale declarations, parameters that are not
  3760 'touched' (instantiation position "_" or omitted) are implicitly added
  3761 with their syntax at the beginning of the for clause.
  3762 
  3763 - Syntax from abbreviations and definitions in locales is available in
  3764 locale expressions and context elements.  The latter is particularly
  3765 useful in locale declarations.
  3766 
  3767 - More flexible mechanisms to qualify names generated by locale
  3768 expressions.  Qualifiers (prefixes) may be specified in locale
  3769 expressions, and can be marked as mandatory (syntax: "name!:") or
  3770 optional (syntax "name?:").  The default depends for plain "name:"
  3771 depends on the situation where a locale expression is used: in
  3772 commands 'locale' and 'sublocale' prefixes are optional, in
  3773 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3774 implicit qualifiers derived from the parameter names of a locale are
  3775 no longer generated.
  3776 
  3777 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3778 instantiation clause in "interpretation" and "interpret" (square
  3779 brackets) is no longer available.  Use locale expressions.
  3780 
  3781 - When converting proof scripts, mandatory qualifiers in
  3782 'interpretation' and 'interpret' should be retained by default, even
  3783 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3784 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3785 in locale expressions range over a single locale instance only.
  3786 
  3787 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3788 In existing theorem specifications replace the includes element by the
  3789 respective context elements of the included locale, omitting those
  3790 that are already present in the theorem specification.  Multiple
  3791 assume elements of a locale should be replaced by a single one
  3792 involving the locale predicate.  In the proof body, declarations (most
  3793 notably theorems) may be regained by interpreting the respective
  3794 locales in the proof context as required (command "interpret").
  3795 
  3796 If using "includes" in replacement of a target solely because the
  3797 parameter types in the theorem are not as general as in the target,
  3798 consider declaring a new locale with additional type constraints on
  3799 the parameters (context element "constrains").
  3800 
  3801 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3802 
  3803 - Locale interpretation commands no longer attempt to simplify goal.
  3804 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3805 methods intro_locales and unfold_locales to clarify.
  3806 
  3807 - Locale interpretation commands no longer accept interpretation
  3808 attributes.  INCOMPATIBILITY.
  3809 
  3810 * Class declaration: so-called "base sort" must not be given in import
  3811 list any longer, but is inferred from the specification.  Particularly
  3812 in HOL, write
  3813 
  3814     class foo = ...
  3815 
  3816 instead of
  3817 
  3818     class foo = type + ...
  3819 
  3820 * Class target: global versions of theorems stemming do not carry a
  3821 parameter prefix any longer.  INCOMPATIBILITY.
  3822 
  3823 * Class 'instance' command no longer accepts attached definitions.
  3824 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3825 
  3826 * Recovered hiding of consts, which was accidentally broken in
  3827 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3828 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3829 
  3830 * Slightly more coherent Pure syntax, with updated documentation in
  3831 isar-ref manual.  Removed locales meta_term_syntax and
  3832 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3833 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3834 directly in regular applications.
  3835 
  3836 * There is a new syntactic category "float_const" for signed decimal
  3837 fractions (e.g. 123.45 or -123.45).
  3838 
  3839 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3840 interface with 'setup' command instead.
  3841 
  3842 * Command 'local_setup' is similar to 'setup', but operates on a local
  3843 theory context.
  3844 
  3845 * The 'axiomatization' command now only works within a global theory
  3846 context.  INCOMPATIBILITY.
  3847 
  3848 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3849 hypotheses.  Sorts required in the course of reasoning need to be
  3850 covered by the constraints in the initial statement, completed by the
  3851 type instance information of the background theory.  Non-trivial sort
  3852 hypotheses, which rarely occur in practice, may be specified via
  3853 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3854 
  3855   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3856 
  3857 The result contains an implicit sort hypotheses as before --
  3858 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3859 normalization.
  3860 
  3861 * Generalized Isar history, with support for linear undo, direct state
  3862 addressing etc.
  3863 
  3864 * Changed defaults for unify configuration options:
  3865 
  3866   unify_trace_bound = 50 (formerly 25)
  3867   unify_search_bound = 60 (formerly 30)
  3868 
  3869 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3870 
  3871   a) On theory merge, the last set of code equations for a particular
  3872      constant is taken (in accordance with the policy applied by other
  3873      parts of the code generator framework).
  3874 
  3875   b) Code equations stemming from explicit declarations (e.g. code
  3876      attribute) gain priority over default code equations stemming
  3877      from definition, primrec, fun etc.
  3878 
  3879 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3880 
  3881 * Unified theorem tables for both code generators.  Thus [code
  3882 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3883 
  3884 * Command 'find_consts' searches for constants based on type and name
  3885 patterns, e.g.
  3886 
  3887     find_consts "_ => bool"
  3888 
  3889 By default, matching is against subtypes, but it may be restricted to
  3890 the whole type.  Searching by name is possible.  Multiple queries are
  3891 conjunctive and queries may be negated by prefixing them with a
  3892 hyphen:
  3893 
  3894     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3895 
  3896 * New 'find_theorems' criterion "solves" matches theorems that
  3897 directly solve the current goal (modulo higher-order unification).
  3898 
  3899 * Auto solve feature for main theorem statements: whenever a new goal
  3900 is stated, "find_theorems solves" is called; any theorems that could
  3901 solve the lemma directly are listed as part of the goal state.
  3902 Cf. associated options in Proof General Isabelle settings menu,
  3903 enabled by default, with reasonable timeout for pathological cases of
  3904 higher-order unification.
  3905 
  3906 
  3907 *** Document preparation ***
  3908 
  3909 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3910 demanding keyword 'by' and supporting the full method expression
  3911 syntax just like the Isar command 'by'.
  3912 
  3913 
  3914 *** HOL ***
  3915 
  3916 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3917 points Main and Complex_Main remain as before.
  3918 
  3919 * Logic image HOL-Plain provides a minimal HOL with the most important
  3920 tools available (inductive, datatype, primrec, ...).  This facilitates
  3921 experimentation and tool development.  Note that user applications
  3922 (and library theories) should never refer to anything below theory
  3923 Main, as before.
  3924 
  3925 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3926 experimentation due to shorter build times.
  3927 
  3928 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3929 were previously part of former HOL-Complex.  Entry point Hyperreal
  3930 remains valid, but theories formerly using Complex_Main should now use
  3931 new entry point Hypercomplex.
  3932 
  3933 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3934 Posix processes.  Avoids potentially expensive forking of the ML
  3935 process.  New thread-based implementation also works on non-Unix
  3936 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3937 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3938 commands are covered in the isar-ref manual.
  3939 
  3940 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3941 sledgehammer without local ATP installation (Vampire etc.). Other
  3942 provers may be included via suitable ML wrappers, see also
  3943 src/HOL/ATP_Linkup.thy.
  3944 
  3945 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3946 menu.
  3947 
  3948 * The metis method no longer fails because the theorem is too trivial
  3949 (contains the empty clause).
  3950 
  3951 * The metis method now fails in the usual manner, rather than raising
  3952 an exception, if it determines that it cannot prove the theorem.
  3953 
  3954 * Method "coherent" implements a prover for coherent logic (see also
  3955 src/Tools/coherent.ML).
  3956 
  3957 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3958 "undefined" is the right choice to replace "arbitrary", though
  3959 logically there is no difference.  INCOMPATIBILITY.
  3960 
  3961 * Command "value" now integrates different evaluation mechanisms.  The
  3962 result of the first successful evaluation mechanism is printed.  In
  3963 square brackets a particular named evaluation mechanisms may be
  3964 specified (currently, [SML], [code] or [nbe]).  See further
  3965 src/HOL/ex/Eval_Examples.thy.
  3966 
  3967 * Normalization by evaluation now allows non-leftlinear equations.
  3968 Declare with attribute [code nbe].
  3969 
  3970 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3971 as the structured Isar versions "cases" and "induct", cf. the
  3972 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3973 are now presented as a list of individual projections
  3974 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3975 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3976 rare situations a different rule is selected --- notably nested tuple
  3977 elimination instead of former prod.exhaust: use explicit (case_tac t
  3978 rule: prod.exhaust) here.
  3979 
  3980 * Attributes "cases", "induct", "coinduct" support "del" option.
  3981 
  3982 * Removed fact "case_split_thm", which duplicates "case_split".
  3983 
  3984 * The option datatype has been moved to a new theory Option.  Renamed
  3985 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3986 
  3987 * New predicate "strict_mono" classifies strict functions on partial
  3988 orders.  With strict functions on linear orders, reasoning about
  3989 (in)equalities is facilitated by theorems "strict_mono_eq",
  3990 "strict_mono_less_eq" and "strict_mono_less".
  3991 
  3992 * Some set operations are now proper qualified constants with
  3993 authentic syntax.  INCOMPATIBILITY:
  3994 
  3995     op Int ~>   Set.Int
  3996     op Un ~>    Set.Un
  3997     INTER ~>    Set.INTER
  3998     UNION ~>    Set.UNION
  3999     Inter ~>    Set.Inter
  4000     Union ~>    Set.Union
  4001     {} ~>       Set.empty
  4002     UNIV ~>     Set.UNIV
  4003 
  4004 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4005 theory Set.
  4006 
  4007 * Auxiliary class "itself" has disappeared -- classes without any
  4008 parameter are treated as expected by the 'class' command.
  4009 
  4010 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4011 
  4012 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4013 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4014 
  4015 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4016 method "approximation".  It proves formulas on real values by using
  4017 interval arithmetic.  In the formulas are also the transcendental
  4018 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4019 allowed. For examples see
  4020 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4021 
  4022 * Theory "Reflection" now resides in HOL/Library.
  4023 
  4024 * Entry point to Word library now simply named "Word".
  4025 INCOMPATIBILITY.
  4026 
  4027 * Made source layout more coherent with logical distribution
  4028 structure:
  4029 
  4030     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4031     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4032     src/HOL/Library/GCD.thy ~> src/HOL/
  4033     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4034     src/HOL/Library/Parity.thy ~> src/HOL/
  4035     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4036     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4037     src/HOL/Real/Lubs.thy ~> src/HOL/
  4038     src/HOL/Real/PReal.thy ~> src/HOL/
  4039     src/HOL/Real/Rational.thy ~> src/HOL/
  4040     src/HOL/Real/RComplete.thy ~> src/HOL/
  4041     src/HOL/Real/RealDef.thy ~> src/HOL/
  4042     src/HOL/Real/RealPow.thy ~> src/HOL/
  4043     src/HOL/Real/Real.thy ~> src/HOL/
  4044     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4045     src/HOL/Complex/Complex.thy ~> src/HOL/
  4046     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4047     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4048     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4049     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4050     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4051     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4052     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4053     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4054     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4055     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4056     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4057     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4058     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4059     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4060     src/HOL/Real/Float ~> src/HOL/Library/
  4061     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4062     src/HOL/Real/RealVector.thy ~> src/HOL/
  4063 
  4064     src/HOL/arith_data.ML ~> src/HOL/Tools
  4065     src/HOL/hologic.ML ~> src/HOL/Tools
  4066     src/HOL/simpdata.ML ~> src/HOL/Tools
  4067     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4068     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4069     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4070     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4071     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4072     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4073     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4074 
  4075     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4076     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4077     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4078     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4079     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4080     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4081 
  4082 * If methods "eval" and "evaluation" encounter a structured proof
  4083 state with !!/==>, only the conclusion is evaluated to True (if
  4084 possible), avoiding strange error messages.
  4085 
  4086 * Method "sizechange" automates termination proofs using (a
  4087 modification of) the size-change principle.  Requires SAT solver.  See
  4088 src/HOL/ex/Termination.thy for examples.
  4089 
  4090 * Simplifier: simproc for let expressions now unfolds if bound
  4091 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4092 
  4093 * Method "arith": Linear arithmetic now ignores all inequalities when
  4094 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4095 
  4096 * New attribute "arith" for facts that should always be used
  4097 automatically by arithmetic. It is intended to be used locally in
  4098 proofs, e.g.
  4099 
  4100   assumes [arith]: "x > 0"
  4101 
  4102 Global usage is discouraged because of possible performance impact.
  4103 
  4104 * New classes "top" and "bot" with corresponding operations "top" and
  4105 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4106 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4107 
  4108 * Changed definition lemma "less_fun_def" in order to provide an
  4109 instance for preorders on functions; use lemma "less_le" instead.
  4110 INCOMPATIBILITY.
  4111 
  4112 * Theory Orderings: class "wellorder" moved here, with explicit
  4113 induction rule "less_induct" as assumption.  For instantiation of
  4114 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4115 INCOMPATIBILITY.
  4116 
  4117 * Theory Orderings: added class "preorder" as superclass of "order".
  4118 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4119 etc. slightly changed.  Some theorems named order_class.* now named
  4120 preorder_class.*.
  4121 
  4122 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4123 "diag" to "Id_on".
  4124 
  4125 * Theory Finite_Set: added a new fold combinator of type
  4126 
  4127   ('a => 'b => 'b) => 'b => 'a set => 'b
  4128 
  4129 Occasionally this is more convenient than the old fold combinator
  4130 which is now defined in terms of the new one and renamed to
  4131 fold_image.
  4132 
  4133 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4134 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4135 extended with further lemmas!).  At the moment both still exist but
  4136 the former will disappear at some point.
  4137 
  4138 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4139 class recpower.  Type-specific simp rules for various recpower types
  4140 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4141 
  4142 rat_power_0    -> power_0
  4143 rat_power_Suc  -> power_Suc
  4144 realpow_0      -> power_0
  4145 realpow_Suc    -> power_Suc
  4146 complexpow_0   -> power_0
  4147 complexpow_Suc -> power_Suc
  4148 power_poly_0   -> power_0
  4149 power_poly_Suc -> power_Suc
  4150 
  4151 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4152 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4153 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4154 of lemmas from Divides has been generalized from nat to class
  4155 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4156 renames resulting from duplicate elimination:
  4157 
  4158     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4159     zero_dvd_iff ~>         dvd_0_left_iff
  4160     dvd_0 ~>                dvd_0_right
  4161     DIVISION_BY_ZERO_DIV ~> div_by_0
  4162     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4163     mult_div ~>             div_mult_self2_is_id
  4164     mult_mod ~>             mod_mult_self2_is_0
  4165 
  4166 * Theory IntDiv: removed many lemmas that are instances of class-based
  4167 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4168 rename old lemmas as follows:
  4169 
  4170 dvd_diff               -> nat_dvd_diff
  4171 dvd_zminus_iff         -> dvd_minus_iff
  4172 mod_add1_eq            -> mod_add_eq
  4173 mod_mult1_eq           -> mod_mult_right_eq
  4174 mod_mult1_eq'          -> mod_mult_left_eq
  4175 mod_mult_distrib_mod   -> mod_mult_eq
  4176 nat_mod_add_left_eq    -> mod_add_left_eq
  4177 nat_mod_add_right_eq   -> mod_add_right_eq
  4178 nat_mod_div_trivial    -> mod_div_trivial
  4179 nat_mod_mod_trivial    -> mod_mod_trivial
  4180 zdiv_zadd_self1        -> div_add_self1
  4181 zdiv_zadd_self2        -> div_add_self2
  4182 zdiv_zmult_self1       -> div_mult_self2_is_id
  4183 zdiv_zmult_self2       -> div_mult_self1_is_id
  4184 zdvd_triv_left         -> dvd_triv_left
  4185 zdvd_triv_right        -> dvd_triv_right
  4186 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4187 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4188 zmod_zadd_left_eq      -> mod_add_left_eq
  4189 zmod_zadd_right_eq     -> mod_add_right_eq
  4190 zmod_zadd_self1        -> mod_add_self1
  4191 zmod_zadd_self2        -> mod_add_self2
  4192 zmod_zadd1_eq          -> mod_add_eq
  4193 zmod_zdiff1_eq         -> mod_diff_eq
  4194 zmod_zdvd_zmod         -> mod_mod_cancel
  4195 zmod_zmod_cancel       -> mod_mod_cancel
  4196 zmod_zmult_self1       -> mod_mult_self2_is_0
  4197 zmod_zmult_self2       -> mod_mult_self1_is_0
  4198 zmod_1                 -> mod_by_1
  4199 zdiv_1                 -> div_by_1
  4200 zdvd_abs1              -> abs_dvd_iff
  4201 zdvd_abs2              -> dvd_abs_iff
  4202 zdvd_refl              -> dvd_refl
  4203 zdvd_trans             -> dvd_trans
  4204 zdvd_zadd              -> dvd_add
  4205 zdvd_zdiff             -> dvd_diff
  4206 zdvd_zminus_iff        -> dvd_minus_iff
  4207 zdvd_zminus2_iff       -> minus_dvd_iff
  4208 zdvd_zmultD            -> dvd_mult_right
  4209 zdvd_zmultD2           -> dvd_mult_left
  4210 zdvd_zmult_mono        -> mult_dvd_mono
  4211 zdvd_0_right           -> dvd_0_right
  4212 zdvd_0_left            -> dvd_0_left_iff
  4213 zdvd_1_left            -> one_dvd
  4214 zminus_dvd_iff         -> minus_dvd_iff
  4215 
  4216 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4217 
  4218 * The real numbers offer decimal input syntax: 12.34 is translated
  4219 into 1234/10^2. This translation is not reversed upon output.
  4220 
  4221 * Theory Library/Polynomial defines an abstract type 'a poly of
  4222 univariate polynomials with coefficients of type 'a.  In addition to
  4223 the standard ring operations, it also supports div and mod.  Code
  4224 generation is also supported, using list-style constructors.
  4225 
  4226 * Theory Library/Inner_Product defines a class of real_inner for real
  4227 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4228 => real.  Class real_inner is a subclass of real_normed_vector from
  4229 theory RealVector.
  4230 
  4231 * Theory Library/Product_Vector provides instances for the product
  4232 type 'a * 'b of several classes from RealVector and Inner_Product.
  4233 Definitions of addition, subtraction, scalar multiplication, norms,
  4234 and inner products are included.
  4235 
  4236 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4237 addition to the field operations, numerals and case syntax are also
  4238 supported.
  4239 
  4240 * Theory Library/Diagonalize provides constructive version of Cantor's
  4241 first diagonalization argument.
  4242 
  4243 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4244 zlcm (for int); carried together from various gcd/lcm developements in
  4245 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4246 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4247 may recover tupled syntax as follows:
  4248 
  4249     hide (open) const gcd
  4250     abbreviation gcd where
  4251       "gcd == (%(a, b). GCD.gcd a b)"
  4252     notation (output)
  4253       GCD.gcd ("gcd '(_, _')")
  4254 
  4255 The same works for lcm, zgcd, zlcm.
  4256 
  4257 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4258 instantiations for algebraic structures.  Removed some duplicate
  4259 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4260 
  4261 * ML antiquotation @{code} takes a constant as argument and generates
  4262 corresponding code in background and inserts name of the corresponding
  4263 resulting ML value/function/datatype constructor binding in place.
  4264 All occurrences of @{code} with a single ML block are generated
  4265 simultaneously.  Provides a generic and safe interface for
  4266 instrumentalizing code generation.  See
  4267 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4268 In future you ought to refrain from ad-hoc compiling generated SML
  4269 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4270 cannot refer to constants for which user-defined serializations are
  4271 set.  Refer to the corresponding ML counterpart directly in that
  4272 cases.
  4273 
  4274 * Command 'rep_datatype': instead of theorem names the command now
  4275 takes a list of terms denoting the constructors of the type to be
  4276 represented as datatype.  The characteristic theorems have to be
  4277 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4278 have disappeared in favour of existing ones:
  4279 
  4280     unit_induct                 ~> unit.induct
  4281     prod_induct                 ~> prod.induct
  4282     sum_induct                  ~> sum.induct
  4283     Suc_Suc_eq                  ~> nat.inject
  4284     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4285 
  4286 
  4287 *** HOL-Algebra ***
  4288 
  4289 * New locales for orders and lattices where the equivalence relation
  4290 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4291 locales use a record structure with field eq for the equivalence.
  4292 
  4293 * New theory of factorial domains.
  4294 
  4295 * Units_l_inv and Units_r_inv are now simp rules by default.
  4296 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4297 and/or r_inv will now also require deletion of these lemmas.
  4298 
  4299 * Renamed the following theorems, INCOMPATIBILITY:
  4300 
  4301 UpperD ~> Upper_memD
  4302 LowerD ~> Lower_memD
  4303 least_carrier ~> least_closed
  4304 greatest_carrier ~> greatest_closed
  4305 greatest_Lower_above ~> greatest_Lower_below
  4306 one_zero ~> carrier_one_zero
  4307 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4308 
  4309 
  4310 *** HOL-Nominal ***
  4311 
  4312 * Nominal datatypes can now contain type-variables.
  4313 
  4314 * Commands 'nominal_inductive' and 'equivariance' work with local
  4315 theory targets.
  4316 
  4317 * Nominal primrec can now works with local theory targets and its
  4318 specification syntax now conforms to the general format as seen in
  4319 'inductive' etc.
  4320 
  4321 * Method "perm_simp" honours the standard simplifier attributes
  4322 (no_asm), (no_asm_use) etc.
  4323 
  4324 * The new predicate #* is defined like freshness, except that on the
  4325 left hand side can be a set or list of atoms.
  4326 
  4327 * Experimental command 'nominal_inductive2' derives strong induction
  4328 principles for inductive definitions.  In contrast to
  4329 'nominal_inductive', which can only deal with a fixed number of
  4330 binders, it can deal with arbitrary expressions standing for sets of
  4331 atoms to be avoided.  The only inductive definition we have at the
  4332 moment that needs this generalisation is the typing rule for Lets in
  4333 the algorithm W:
  4334 
  4335  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4336  -----------------------------------------------------------------
  4337          Gamma |- Let x be t1 in t2 : T2
  4338 
  4339 In this rule one wants to avoid all the binders that are introduced by
  4340 "close Gamma T1".  We are looking for other examples where this
  4341 feature might be useful.  Please let us know.
  4342 
  4343 
  4344 *** HOLCF ***
  4345 
  4346 * Reimplemented the simplification procedure for proving continuity
  4347 subgoals.  The new simproc is extensible; users can declare additional
  4348 continuity introduction rules with the attribute [cont2cont].
  4349 
  4350 * The continuity simproc now uses a different introduction rule for
  4351 solving continuity subgoals on terms with lambda abstractions.  In
  4352 some rare cases the new simproc may fail to solve subgoals that the
  4353 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4354 Potential INCOMPATIBILITY.
  4355 
  4356 * Command 'fixrec': specification syntax now conforms to the general
  4357 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4358 examples.  INCOMPATIBILITY.
  4359 
  4360 
  4361 *** ZF ***
  4362 
  4363 * Proof of Zorn's Lemma for partial orders.
  4364 
  4365 
  4366 *** ML ***
  4367 
  4368 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4369 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4370 depends on multithreading, so timouts will not work before Poly/ML
  4371 5.2.1!
  4372 
  4373 * High-level support for concurrent ML programming, see
  4374 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4375 particularly convenient to organize independent functional
  4376 computations.  The concept of "synchronized variables" provides a
  4377 higher-order interface for components with shared state, avoiding the
  4378 delicate details of mutexes and condition variables.  (Requires
  4379 Poly/ML 5.2.1 or later.)
  4380 
  4381 * ML bindings produced via Isar commands are stored within the Isar
  4382 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4383 become thread-safe and work with undo as expected (concerning
  4384 top-level bindings, not side-effects on global references).
  4385 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4386 compiler at runtime; really global bindings need to be given outside a
  4387 theory.  (Requires Poly/ML 5.2 or later.)
  4388 
  4389 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4390 context.  Top-level ML bindings are stored within the proof context in
  4391 a purely sequential fashion, disregarding the nested proof structure.
  4392 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4393 proof.  (Requires Poly/ML 5.2 or later.)
  4394 
  4395 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4396 and Method.setup, as well as Isar commands 'attribute_setup' and
  4397 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4398 existing code accordingly, or use plain 'setup' together with old
  4399 Method.add_method.
  4400 
  4401 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4402 to 'a -> thm, while results are always tagged with an authentic oracle
  4403 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4404 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4405 accordingly.  Note that extra performance may be gained by producing
  4406 the cterm carefully, avoiding slow Thm.cterm_of.
  4407 
  4408 * Simplified interface for defining document antiquotations via
  4409 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4410 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4411 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4412 examples.
  4413 
  4414 * More systematic treatment of long names, abstract name bindings, and
  4415 name space operations.  Basic operations on qualified names have been
  4416 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4417 Long_Name.append.  Old type bstring has been mostly replaced by
  4418 abstract type binding (see structure Binding), which supports precise
  4419 qualification by packages and local theory targets, as well as proper
  4420 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4421 bstring values into Binding.name, or better pass through abstract
  4422 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4423 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4424 
  4425 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4426 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4427 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4428 
  4429 * Disposed old type and term read functions (Sign.read_def_typ,
  4430 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4431 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4432 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4433 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4434 resort for legacy applications.
  4435 
  4436 * Disposed old declarations, tactics, tactic combinators that refer to
  4437 the simpset or claset of an implicit theory (such as Addsimps,
  4438 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4439 embedded ML text, or local_simpset_of with a proper context passed as
  4440 explicit runtime argument.
  4441 
  4442 * Rules and tactics that read instantiations (read_instantiate,
  4443 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4444 context, which is required for parsing and type-checking.  Moreover,
  4445 the variables are specified as plain indexnames, not string encodings
  4446 thereof.  INCOMPATIBILITY.
  4447 
  4448 * Generic Toplevel.add_hook interface allows to analyze the result of
  4449 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4450 for theorem dependency output of transactions resulting in a new
  4451 theory state.
  4452 
  4453 * ML antiquotations: block-structured compilation context indicated by
  4454 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4455 
  4456   @{binding name}                         - basic name binding
  4457   @{let ?pat = term}                      - term abbreviation (HO matching)
  4458   @{note name = fact}                     - fact abbreviation
  4459   @{thm fact}                             - singleton fact (with attributes)
  4460   @{thms fact}                            - general fact (with attributes)
  4461   @{lemma prop by method}                 - singleton goal
  4462   @{lemma prop by meth1 meth2}            - singleton goal
  4463   @{lemma prop1 ... propN by method}      - general goal
  4464   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4465   @{lemma (open) ...}                     - open derivation
  4466 
  4467 
  4468 *** System ***
  4469 
  4470 * The Isabelle "emacs" tool provides a specific interface to invoke
  4471 Proof General / Emacs, with more explicit failure if that is not
  4472 installed (the old isabelle-interface script silently falls back on
  4473 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4474 installation location of the Proof General distribution.
  4475 
  4476 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4477 the Isabelle process into a JVM/Scala application.  See
  4478 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4479 process wrapper has been discontinued.)
  4480 
  4481 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4482 
  4483 * Various status messages (with exact source position information) are
  4484 emitted, if proper markup print mode is enabled.  This allows
  4485 user-interface components to provide detailed feedback on internal
  4486 prover operations.
  4487 
  4488 
  4489 
  4490 New in Isabelle2008 (June 2008)
  4491 -------------------------------
  4492 
  4493 *** General ***
  4494 
  4495 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4496 and updated, with formally checked references as hyperlinks.
  4497 
  4498 * Theory loader: use_thy (and similar operations) no longer set the
  4499 implicit ML context, which was occasionally hard to predict and in
  4500 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4501 provides a proper context already.
  4502 
  4503 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4504 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4505 'use' within a theory file will do the job.
  4506 
  4507 * Name space merge now observes canonical order, i.e. the second space
  4508 is inserted into the first one, while existing entries in the first
  4509 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4510 swap theory imports.
  4511 
  4512 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4513 INCOMPATIBILITY in identifier syntax etc.
  4514 
  4515 * Outer syntax: string tokens no longer admit escaped white space,
  4516 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4517 white space without escapes.
  4518 
  4519 * Outer syntax: string tokens may contain arbitrary character codes
  4520 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4521 "foo_bar".
  4522 
  4523 
  4524 *** Pure ***
  4525 
  4526 * Context-dependent token translations.  Default setup reverts locally
  4527 fixed variables, and adds hilite markup for undeclared frees.
  4528 
  4529 * Unused theorems can be found using the new command 'unused_thms'.
  4530 There are three ways of invoking it:
  4531 
  4532 (1) unused_thms
  4533      Only finds unused theorems in the current theory.
  4534 
  4535 (2) unused_thms thy_1 ... thy_n -
  4536      Finds unused theorems in the current theory and all of its ancestors,
  4537      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4538 
  4539 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4540      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4541      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4542      their ancestors.
  4543 
  4544 In order to increase the readability of the list produced by
  4545 unused_thms, theorems that have been created by a particular instance
  4546 of a theory command such as 'inductive' or 'function' are considered
  4547 to belong to the same "group", meaning that if at least one theorem in
  4548 this group is used, the other theorems in the same group are no longer
  4549 reported as unused.  Moreover, if all theorems in the group are
  4550 unused, only one theorem in the group is displayed.
  4551 
  4552 Note that proof objects have to be switched on in order for
  4553 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4554 usually the case when using Proof General with the default settings).
  4555 
  4556 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4557 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4558 duplicate fact bindings, or even accidental fact duplications.  Note
  4559 that tools may maintain dynamically scoped facts systematically, using
  4560 PureThy.add_thms_dynamic.
  4561 
  4562 * Command 'hide' now allows to hide from "fact" name space as well.
  4563 
  4564 * Eliminated destructive theorem database, simpset, claset, and
  4565 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4566 update of theories within ML code.
  4567 
  4568 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4569 INCOMPATIBILITY, object-logics depending on former Pure require
  4570 additional setup PureThy.old_appl_syntax_setup; object-logics
  4571 depending on former CPure need to refer to Pure.
  4572 
  4573 * Commands 'use' and 'ML' are now purely functional, operating on
  4574 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4575 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4576 INCOMPATIBILITY.
  4577 
  4578 * Command 'setup': discontinued implicit version with ML reference.
  4579 
  4580 * Instantiation target allows for simultaneous specification of class
  4581 instance operations together with an instantiation proof.
  4582 Type-checking phase allows to refer to class operations uniformly.
  4583 See src/HOL/Complex/Complex.thy for an Isar example and
  4584 src/HOL/Library/Eval.thy for an ML example.
  4585 
  4586 * Indexing of literal facts: be more serious about including only
  4587 facts from the visible specification/proof context, but not the
  4588 background context (locale etc.).  Affects `prop` notation and method
  4589 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4590 situations.
  4591 
  4592 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4593 "(open)" option, which used to expose internal bound variables to the
  4594 proof text.
  4595 
  4596 * Isar statements: removed obsolete case "rule_context".
  4597 INCOMPATIBILITY, better use explicit fixes/assumes.
  4598 
  4599 * Locale proofs: default proof step now includes 'unfold_locales';
  4600 hence 'proof' without argument may be used to unfold locale
  4601 predicates.
  4602 
  4603 
  4604 *** Document preparation ***
  4605 
  4606 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4607 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4608 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4609 spot); renamed link color from "darkblue" to "linkcolor" (default
  4610 value unchanged, can be redefined via \definecolor); no longer sets
  4611 "a4paper" option (unnecessary or even intrusive).
  4612 
  4613 * Antiquotation @{lemma A method} proves proposition A by the given
  4614 method (either a method name or a method name plus (optional) method
  4615 arguments in parentheses) and prints A just like @{prop A}.
  4616 
  4617 
  4618 *** HOL ***
  4619 
  4620 * New primrec package.  Specification syntax conforms in style to
  4621 definition/function/....  No separate induction rule is provided.  The
  4622 "primrec" command distinguishes old-style and new-style specifications
  4623 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4624 When adjusting theories, beware: constants stemming from new-style
  4625 primrec specifications have authentic syntax.
  4626 
  4627 * Metis prover is now an order of magnitude faster, and also works
  4628 with multithreading.
  4629 
  4630 * Metis: the maximum number of clauses that can be produced from a
  4631 theorem is now given by the attribute max_clauses.  Theorems that
  4632 exceed this number are ignored, with a warning printed.
  4633 
  4634 * Sledgehammer no longer produces structured proofs by default. To
  4635 enable, declare [[sledgehammer_full = true]].  Attributes
  4636 reconstruction_modulus, reconstruction_sorts renamed
  4637 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4638 
  4639 * Method "induct_scheme" derives user-specified induction rules
  4640 from well-founded induction and completeness of patterns. This factors
  4641 out some operations that are done internally by the function package
  4642 and makes them available separately.  See
  4643 src/HOL/ex/Induction_Scheme.thy for examples.
  4644 
  4645 * More flexible generation of measure functions for termination
  4646 proofs: Measure functions can be declared by proving a rule of the
  4647 form "is_measure f" and giving it the [measure_function] attribute.
  4648 The "is_measure" predicate is logically meaningless (always true), and
  4649 just guides the heuristic.  To find suitable measure functions, the
  4650 termination prover sets up the goal "is_measure ?f" of the appropriate
  4651 type and generates all solutions by prolog-style backwards proof using
  4652 the declared rules.
  4653 
  4654 This setup also deals with rules like
  4655 
  4656   "is_measure f ==> is_measure (list_size f)"
  4657 
  4658 which accommodates nested datatypes that recurse through lists.
  4659 Similar rules are predeclared for products and option types.
  4660 
  4661 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4662 
  4663   INCOMPATIBILITIES:
  4664 
  4665   - Definitions of overloaded constants on sets have to be replaced by
  4666     definitions on => and bool.
  4667 
  4668   - Some definitions of overloaded operators on sets can now be proved
  4669     using the definitions of the operators on => and bool.  Therefore,
  4670     the following theorems have been renamed:
  4671 
  4672       subset_def   -> subset_eq
  4673       psubset_def  -> psubset_eq
  4674       set_diff_def -> set_diff_eq
  4675       Compl_def    -> Compl_eq
  4676       Sup_set_def  -> Sup_set_eq
  4677       Inf_set_def  -> Inf_set_eq
  4678       sup_set_def  -> sup_set_eq
  4679       inf_set_def  -> inf_set_eq
  4680 
  4681   - Due to the incompleteness of the HO unification algorithm, some
  4682     rules such as subst may require manual instantiation, if some of
  4683     the unknowns in the rule is a set.
  4684 
  4685   - Higher order unification and forward proofs:
  4686     The proof pattern
  4687 
  4688       have "P (S::'a set)" <...>
  4689       then have "EX S. P S" ..
  4690 
  4691     no longer works (due to the incompleteness of the HO unification
  4692     algorithm) and must be replaced by the pattern
  4693 
  4694       have "EX S. P S"
  4695       proof
  4696         show "P S" <...>
  4697       qed
  4698 
  4699   - Calculational reasoning with subst (or similar rules):
  4700     The proof pattern
  4701 
  4702       have "P (S::'a set)" <...>
  4703       also have "S = T" <...>
  4704       finally have "P T" .
  4705 
  4706     no longer works (for similar reasons as the previous example) and
  4707     must be replaced by something like
  4708 
  4709       have "P (S::'a set)" <...>
  4710       moreover have "S = T" <...>
  4711       ultimately have "P T" by simp
  4712 
  4713   - Tactics or packages written in ML code:
  4714     Code performing pattern matching on types via
  4715 
  4716       Type ("set", [T]) => ...
  4717 
  4718     must be rewritten. Moreover, functions like strip_type or
  4719     binder_types no longer return the right value when applied to a
  4720     type of the form
  4721 
  4722       T1 => ... => Tn => U => bool
  4723 
  4724     rather than
  4725 
  4726       T1 => ... => Tn => U set
  4727 
  4728 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4729 Wellfounded_Relations to theory Wellfounded.
  4730 
  4731 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4732 
  4733 * Class finite no longer treats UNIV as class parameter.  Use class
  4734 enum from theory Library/Enum instead to achieve a similar effect.
  4735 INCOMPATIBILITY.
  4736 
  4737 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4738 and "Cons".  INCOMPATIBILITY.
  4739 
  4740 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4741 Potential INCOMPATIBILITY.
  4742 
  4743 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4744 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4745 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4746 
  4747 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4748 the same name in theory Orderings:
  4749 
  4750   less_trans
  4751   less_linear
  4752   le_imp_less_or_eq
  4753   le_less_trans
  4754   less_le_trans
  4755   less_not_sym
  4756   less_asym
  4757 
  4758 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4759 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4760 and different variable names.
  4761 
  4762 * Library/Option_ord.thy: Canonical order on option type.
  4763 
  4764 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4765 finite maps.
  4766 
  4767 * Library/Countable.thy: Type class for countable types.
  4768 
  4769 * Theory Int: The representation of numerals has changed.  The infix
  4770 operator BIT and the bit datatype with constructors B0 and B1 have
  4771 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4772 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4773 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4774 accordingly.
  4775 
  4776 * Theory Nat: definition of <= and < on natural numbers no longer
  4777 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4778 le_def and less_def have disappeared.  Consider lemmas not_less
  4779 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4780 
  4781 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4782 (whose purpose mainly is for various fold_set functionals) have been
  4783 abandoned in favor of the existing algebraic classes
  4784 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4785 lower_semilattice (resp. upper_semilattice) and linorder.
  4786 INCOMPATIBILITY.
  4787 
  4788 * Theory Transitive_Closure: induct and cases rules now declare proper
  4789 case_names ("base" and "step").  INCOMPATIBILITY.
  4790 
  4791 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4792 lattices.  The form set-specific version is available as
  4793 Inductive.lfp_ordinal_induct_set.
  4794 
  4795 * Renamed theorems "power.simps" to "power_int.simps".
  4796 INCOMPATIBILITY.
  4797 
  4798 * Class semiring_div provides basic abstract properties of semirings
  4799 with division and modulo operations.  Subsumes former class dvd_mod.
  4800 
  4801 * Merged theories IntDef, Numeral and IntArith into unified theory
  4802 Int.  INCOMPATIBILITY.
  4803 
  4804 * Theory Library/Code_Index: type "index" now represents natural
  4805 numbers rather than integers.  INCOMPATIBILITY.
  4806 
  4807 * New class "uminus" with operation "uminus" (split of from class
  4808 "minus" which now only has operation "minus", binary).
  4809 INCOMPATIBILITY.
  4810 
  4811 * Constants "card", "internal_split", "option_map" now with authentic
  4812 syntax.  INCOMPATIBILITY.
  4813 
  4814 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4815 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4816 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4817 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4818 Sup_set_def, le_def, less_def, option_map_def now with object
  4819 equality.  INCOMPATIBILITY.
  4820 
  4821 * Records. Removed K_record, and replaced it by pure lambda term
  4822 %x. c. The simplifier setup is now more robust against eta expansion.
  4823 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4824 
  4825 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4826 
  4827 * Library/ListVector: new theory of arithmetic vector operations.
  4828 
  4829 * Library/Order_Relation: new theory of various orderings as sets of
  4830 pairs.  Defines preorders, partial orders, linear orders and
  4831 well-orders on sets and on types.
  4832 
  4833 
  4834 *** ZF ***
  4835 
  4836 * Renamed some theories to allow to loading both ZF and HOL in the
  4837 same session:
  4838 
  4839   Datatype  -> Datatype_ZF
  4840   Inductive -> Inductive_ZF
  4841   Int       -> Int_ZF
  4842   IntDiv    -> IntDiv_ZF
  4843   Nat       -> Nat_ZF
  4844   List      -> List_ZF
  4845   Main      -> Main_ZF
  4846 
  4847 INCOMPATIBILITY: ZF theories that import individual theories below
  4848 Main might need to be adapted.  Regular theory Main is still
  4849 available, as trivial extension of Main_ZF.
  4850 
  4851 
  4852 *** ML ***
  4853 
  4854 * ML within Isar: antiquotation @{const name} or @{const
  4855 name(typargs)} produces statically-checked Const term.
  4856 
  4857 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4858 (of the same name).  Removed obsolete print command.
  4859 
  4860 * Removed obsolete "use_legacy_bindings" function.
  4861 
  4862 * The ``print mode'' is now a thread-local value derived from a global
  4863 template (the former print_mode reference), thus access becomes
  4864 non-critical.  The global print_mode reference is for session
  4865 management only; user-code should use print_mode_value,
  4866 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4867 
  4868 * Functions system/system_out provide a robust way to invoke external
  4869 shell commands, with propagation of interrupts (requires Poly/ML
  4870 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4871 
  4872 
  4873 *** System ***
  4874 
  4875 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4876 in accordance with Proof General 3.7, which prefers GNU emacs.
  4877 
  4878 * isatool tty runs Isabelle process with plain tty interaction;
  4879 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4880 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4881 
  4882 * isatool browser now works with Cygwin as well, using general
  4883 "javapath" function defined in Isabelle process environment.
  4884 
  4885 * YXML notation provides a simple and efficient alternative to
  4886 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4887 isatool yxml as described in the Isabelle system manual.
  4888 
  4889 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4890 provides general wrapper for managing an Isabelle process in a robust
  4891 fashion, with ``cooked'' output from stdin/stderr.
  4892 
  4893 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4894 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4895 
  4896 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4897 way of changing the user's settings is via
  4898 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4899 script.
  4900 
  4901 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4902 cores of the underlying machine, which is a good starting point for
  4903 optimal performance tuning.  The corresponding usedir option -M allows
  4904 "max" as an alias for "0".  WARNING: does not work on certain versions
  4905 of Mac OS (with Poly/ML 5.1).
  4906 
  4907 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4908 adverse effect of Isabelle flooding interactive front-ends (notably
  4909 ProofGeneral / XEmacs).
  4910 
  4911 
  4912 
  4913 New in Isabelle2007 (November 2007)
  4914 -----------------------------------
  4915 
  4916 *** General ***
  4917 
  4918 * More uniform information about legacy features, notably a
  4919 warning/error of "Legacy feature: ...", depending on the state of the
  4920 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4921 legacy features will disappear eventually.
  4922 
  4923 * Theory syntax: the header format ``theory A = B + C:'' has been
  4924 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4925 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4926 
  4927 * Theory syntax: the old non-Isar theory file format has been
  4928 discontinued altogether.  Note that ML proof scripts may still be used
  4929 with Isar theories; migration is usually quite simple with the ML
  4930 function use_legacy_bindings.  INCOMPATIBILITY.
  4931 
  4932 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4933 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4934 quotes.
  4935 
  4936 * Theory loader: be more serious about observing the static theory
  4937 header specifications (including optional directories), but not the
  4938 accidental file locations of previously successful loads.  The strict
  4939 update policy of former update_thy is now already performed by
  4940 use_thy, so the former has been removed; use_thys updates several
  4941 theories simultaneously, just as 'imports' within a theory header
  4942 specification, but without merging the results.  Potential
  4943 INCOMPATIBILITY: may need to refine theory headers and commands
  4944 ROOT.ML which depend on load order.
  4945 
  4946 * Theory loader: optional support for content-based file
  4947 identification, instead of the traditional scheme of full physical
  4948 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4949 (cf. the system manual).  The new scheme allows to work with
  4950 non-finished theories in persistent session images, such that source
  4951 files may be moved later on without requiring reloads.
  4952 
  4953 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4954 file (with the same base name as the theory) are considered a legacy
  4955 feature, which will disappear eventually. Even now, the theory loader
  4956 no longer maintains dependencies on such files.
  4957 
  4958 * Syntax: the scope for resolving ambiguities via type-inference is
  4959 now limited to individual terms, instead of whole simultaneous
  4960 specifications as before. This greatly reduces the complexity of the
  4961 syntax module and improves flexibility by separating parsing and
  4962 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4963 'fixes' etc.) are required in rare situations.
  4964 
  4965 * Syntax: constants introduced by new-style packages ('definition',
  4966 'abbreviation' etc.) are passed through the syntax module in
  4967 ``authentic mode''. This means that associated mixfix annotations
  4968 really stick to such constants, independently of potential name space
  4969 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4970 trees are represented slightly differently, may need to adapt syntax
  4971 translations accordingly. Use CONST marker in 'translations' and
  4972 @{const_syntax} antiquotation in 'parse_translation' etc.
  4973 
  4974 * Legacy goal package: reduced interface to the bare minimum required
  4975 to keep existing proof scripts running.  Most other user-level
  4976 functions are now part of the OldGoals structure, which is *not* open
  4977 by default (consider isatool expandshort before open OldGoals).
  4978 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4979 these tend to cause confusion about the actual goal (!) context being
  4980 used here, which is not necessarily the same as the_context().
  4981 
  4982 * Command 'find_theorems': supports "*" wild-card in "name:"
  4983 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4984 support a specific search form (see ProofGeneral/CHANGES).
  4985 
  4986 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4987 by default, which means that "prems" (and also "fixed variables") are
  4988 suppressed from proof state output.  Note that the ProofGeneral
  4989 settings mechanism allows to change and save options persistently, but
  4990 older versions of Isabelle will fail to start up if a negative prems
  4991 limit is imposed.
  4992 
  4993 * Local theory targets may be specified by non-nested blocks of
  4994 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4995 contain definitions, theorems etc., including any derived mechanism
  4996 that has been implemented on top of these primitives.  This concept
  4997 generalizes the existing ``theorem (in ...)'' towards more versatility
  4998 and scalability.
  4999 
  5000 * Proof General interface: proper undo of final 'end' command;
  5001 discontinued Isabelle/classic mode (ML proof scripts).
  5002 
  5003 
  5004 *** Document preparation ***
  5005 
  5006 * Added antiquotation @{theory name} which prints the given name,
  5007 after checking that it refers to a valid ancestor theory in the
  5008 current context.
  5009 
  5010 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5011 check the given source text as ML type/structure, printing verbatim.
  5012 
  5013 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5014 "c args == rhs" given in the current context.  (Any number of
  5015 arguments may be given on the LHS.)
  5016 
  5017 
  5018 *** Pure ***
  5019 
  5020 * The 'class' package offers a combination of axclass and locale to
  5021 achieve Haskell-like type classes in Isabelle.  Definitions and
  5022 theorems within a class context produce both relative results (with
  5023 implicit parameters according to the locale context), and polymorphic
  5024 constants with qualified polymorphism (according to the class
  5025 context).  Within the body context of a 'class' target, a separate
  5026 syntax layer ("user space type system") takes care of converting
  5027 between global polymorphic consts and internal locale representation.
  5028 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5029 "isatool doc classes" provides a tutorial.
  5030 
  5031 * Generic code generator framework allows to generate executable
  5032 code for ML and Haskell (including Isabelle classes).  A short usage
  5033 sketch:
  5034 
  5035     internal compilation:
  5036         export_code <list of constants (term syntax)> in SML
  5037     writing SML code to a file:
  5038         export_code <list of constants (term syntax)> in SML <filename>
  5039     writing OCaml code to a file:
  5040         export_code <list of constants (term syntax)> in OCaml <filename>
  5041     writing Haskell code to a bunch of files:
  5042         export_code <list of constants (term syntax)> in Haskell <filename>
  5043 
  5044     evaluating closed propositions to True/False using code generation:
  5045         method ``eval''
  5046 
  5047 Reasonable default setup of framework in HOL.
  5048 
  5049 Theorem attributs for selecting and transforming function equations theorems:
  5050 
  5051     [code fun]:        select a theorem as function equation for a specific constant
  5052     [code fun del]:    deselect a theorem as function equation for a specific constant
  5053     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5054     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5055 
  5056 User-defined serializations (target in {SML, OCaml, Haskell}):
  5057 
  5058     code_const <and-list of constants (term syntax)>
  5059       {(target) <and-list of const target syntax>}+
  5060 
  5061     code_type <and-list of type constructors>
  5062       {(target) <and-list of type target syntax>}+
  5063 
  5064     code_instance <and-list of instances>
  5065       {(target)}+
  5066         where instance ::= <type constructor> :: <class>
  5067 
  5068     code_class <and_list of classes>
  5069       {(target) <and-list of class target syntax>}+
  5070         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5071 
  5072 code_instance and code_class only are effective to target Haskell.
  5073 
  5074 For example usage see src/HOL/ex/Codegenerator.thy and
  5075 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5076 generation from Isabelle/HOL theories is available via "isatool doc
  5077 codegen".
  5078 
  5079 * Code generator: consts in 'consts_code' Isar commands are now
  5080 referred to by usual term syntax (including optional type
  5081 annotations).
  5082 
  5083 * Command 'no_translations' removes translation rules from theory
  5084 syntax.
  5085 
  5086 * Overloaded definitions are now actually checked for acyclic
  5087 dependencies.  The overloading scheme is slightly more general than
  5088 that of Haskell98, although Isabelle does not demand an exact
  5089 correspondence to type class and instance declarations.
  5090 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5091 exotic versions of overloading -- at the discretion of the user!
  5092 
  5093 Polymorphic constants are represented via type arguments, i.e. the
  5094 instantiation that matches an instance against the most general
  5095 declaration given in the signature.  For example, with the declaration
  5096 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5097 as c(nat).  Overloading is essentially simultaneous structural
  5098 recursion over such type arguments.  Incomplete specification patterns
  5099 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5100 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5101 Command 'print_theory' outputs the normalized system of recursive
  5102 equations, see section "definitions".
  5103 
  5104 * Configuration options are maintained within the theory or proof
  5105 context (with name and type bool/int/string), providing a very simple
  5106 interface to a poor-man's version of general context data.  Tools may
  5107 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5108 these values using Config.get etc.  Users may change options via an
  5109 associated attribute of the same name.  This form of context
  5110 declaration works particularly well with commands 'declare' or
  5111 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5112 very easy to avoid global references, which would not observe Isar
  5113 toplevel undo/redo and fail to work with multithreading.
  5114 
  5115 Various global ML references of Pure and HOL have been turned into
  5116 configuration options:
  5117 
  5118   Unify.search_bound		unify_search_bound
  5119   Unify.trace_bound		unify_trace_bound
  5120   Unify.trace_simp		unify_trace_simp
  5121   Unify.trace_types		unify_trace_types
  5122   Simplifier.simp_depth_limit	simp_depth_limit
  5123   Blast.depth_limit		blast_depth_limit
  5124   DatatypeProp.dtK		datatype_distinctness_limit
  5125   fast_arith_neq_limit  	fast_arith_neq_limit
  5126   fast_arith_split_limit	fast_arith_split_limit
  5127 
  5128 * Named collections of theorems may be easily installed as context
  5129 data using the functor NamedThmsFun (see also
  5130 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5131 attributes; there is also a toplevel print command.  This facility is
  5132 just a common case of general context data, which is the preferred way
  5133 for anything more complex than just a list of facts in canonical
  5134 order.
  5135 
  5136 * Isar: command 'declaration' augments a local theory by generic
  5137 declaration functions written in ML.  This enables arbitrary content
  5138 being added to the context, depending on a morphism that tells the
  5139 difference of the original declaration context wrt. the application
  5140 context encountered later on.
  5141 
  5142 * Isar: proper interfaces for simplification procedures.  Command
  5143 'simproc_setup' declares named simprocs (with match patterns, and body
  5144 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5145 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5146 
  5147 * Isar: an extra pair of brackets around attribute declarations
  5148 abbreviates a theorem reference involving an internal dummy fact,
  5149 which will be ignored later --- only the effect of the attribute on
  5150 the background context will persist.  This form of in-place
  5151 declarations is particularly useful with commands like 'declare' and
  5152 'using', for example ``have A using [[simproc a]] by simp''.
  5153 
  5154 * Isar: method "assumption" (and implicit closing of subproofs) now
  5155 takes simple non-atomic goal assumptions into account: after applying
  5156 an assumption as a rule the resulting subgoals are solved by atomic
  5157 assumption steps.  This is particularly useful to finish 'obtain'
  5158 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5159 without referring to the original premise "!!x. P x ==> thesis" in the
  5160 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5161 more permissive.
  5162 
  5163 * Isar: implicit use of prems from the Isar proof context is
  5164 considered a legacy feature.  Common applications like ``have A .''
  5165 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5166 referencing facts explicitly here improves readability and
  5167 maintainability of proof texts.
  5168 
  5169 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5170 the obtained context from the course of reasoning!  For example:
  5171 
  5172   assume "EX x y. A x & B y"   -- "any previous fact"
  5173   then guess x and y by clarify
  5174 
  5175 This technique is potentially adventurous, depending on the facts and
  5176 proof tools being involved here.
  5177 
  5178 * Isar: known facts from the proof context may be specified as literal
  5179 propositions, using ASCII back-quote syntax.  This works wherever
  5180 named facts used to be allowed so far, in proof commands, proof
  5181 methods, attributes etc.  Literal facts are retrieved from the context
  5182 according to unification of type and term parameters.  For example,
  5183 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5184 theorems in the current context, then these are valid literal facts:
  5185 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5186 
  5187 There is also a proof method "fact" which does the same composition
  5188 for explicit goal states, e.g. the following proof texts coincide with
  5189 certain special cases of literal facts:
  5190 
  5191   have "A" by fact                 ==  note `A`
  5192   have "A ==> B" by fact           ==  note `A ==> B`
  5193   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5194   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5195 
  5196 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5197 outer syntax.  Thus symbolic identifiers may be used without
  5198 additional white space in declarations like this: ``assume *: A''.
  5199 
  5200 * Isar: 'print_facts' prints all local facts of the current context,
  5201 both named and unnamed ones.
  5202 
  5203 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5204 
  5205   def x == "t" and y == "u"
  5206 
  5207 * Isar: added command 'unfolding', which is structurally similar to
  5208 'using', but affects both the goal state and facts by unfolding given
  5209 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5210 'unfolded' attribute may be replaced by first-class proof text.
  5211 
  5212 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5213 and command 'unfolding' now all support object-level equalities
  5214 (potentially conditional).  The underlying notion of rewrite rule is
  5215 analogous to the 'rule_format' attribute, but *not* that of the
  5216 Simplifier (which is usually more generous).
  5217 
  5218 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5219 premises of a theorem by n. Useful in conjunction with drule.
  5220 
  5221 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5222 method expression within a sandbox consisting of the first N
  5223 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5224 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5225 simplifies all new goals that emerge from applying rule foo to the
  5226 originally first one.
  5227 
  5228 * Isar: schematic goals are no longer restricted to higher-order
  5229 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5230 expected.
  5231 
  5232 * Isar: the conclusion of a long theorem statement is now either
  5233 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5234 (essentially a disjunction of cases with local parameters and
  5235 assumptions).  The latter allows to express general elimination rules
  5236 adequately; in this notation common elimination rules look like this:
  5237 
  5238   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5239     assumes "EX x. P x"
  5240     obtains x where "P x"
  5241 
  5242   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5243     assumes "A & B"
  5244     obtains A and B
  5245 
  5246   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5247     assumes "A | B"
  5248     obtains
  5249       A
  5250     | B
  5251 
  5252 The subsequent classical rules even refer to the formal "thesis"
  5253 explicitly:
  5254 
  5255   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5256     obtains "~ thesis"
  5257 
  5258   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5259     obtains "thesis ==> something"
  5260 
  5261 The actual proof of an 'obtains' statement is analogous to that of the
  5262 Isar proof element 'obtain', only that there may be several cases.
  5263 Optional case names may be specified in parentheses; these will be
  5264 available both in the present proof and as annotations in the
  5265 resulting rule, for later use with the 'cases' method (cf. attribute
  5266 case_names).
  5267 
  5268 * Isar: the assumptions of a long theorem statement are available as
  5269 "assms" fact in the proof context.  This is more appropriate than the
  5270 (historical) "prems", which refers to all assumptions of the current
  5271 context, including those from the target locale, proof body etc.
  5272 
  5273 * Isar: 'print_statement' prints theorems from the current theory or
  5274 proof context in long statement form, according to the syntax of a
  5275 top-level lemma.
  5276 
  5277 * Isar: 'obtain' takes an optional case name for the local context
  5278 introduction rule (default "that").
  5279 
  5280 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5281 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5282 to occur.
  5283 
  5284 * Pure: syntax "CONST name" produces a fully internalized constant
  5285 according to the current context.  This is particularly useful for
  5286 syntax translations that should refer to internal constant
  5287 representations independently of name spaces.
  5288 
  5289 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5290 instead of "FOO ". This allows multiple binder declarations to coexist
  5291 in the same context.  INCOMPATIBILITY.
  5292 
  5293 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5294 primitive that also works in a locale context (both for constants and
  5295 fixed variables). Type declaration and internal syntactic representation
  5296 of given constants retrieved from the context. Likewise, the
  5297 'no_notation' command allows to remove given syntax annotations from the
  5298 current context.
  5299 
  5300 * Isar/locales: new derived specification elements 'axiomatization',
  5301 'definition', 'abbreviation', which support type-inference, admit
  5302 object-level specifications (equality, equivalence).  See also the
  5303 isar-ref manual.  Examples:
  5304 
  5305   axiomatization
  5306     eq  (infix "===" 50) where
  5307     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5308 
  5309   definition "f x y = x + y + 1"
  5310   definition g where "g x = f x x"
  5311 
  5312   abbreviation
  5313     neq  (infix "=!=" 50) where
  5314     "x =!= y == ~ (x === y)"
  5315 
  5316 These specifications may be also used in a locale context.  Then the
  5317 constants being introduced depend on certain fixed parameters, and the
  5318 constant name is qualified by the locale base name.  An internal
  5319 abbreviation takes care for convenient input and output, making the
  5320 parameters implicit and using the original short name.  See also
  5321 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5322 entities from a monomorphic theory.
  5323 
  5324 Presently, abbreviations are only available 'in' a target locale, but
  5325 not inherited by general import expressions.  Also note that
  5326 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5327 'translations' in common applications.  The "no_abbrevs" print mode
  5328 prevents folding of abbreviations in term output.
  5329 
  5330 Concrete syntax is attached to specified constants in internal form,
  5331 independently of name spaces.  The parse tree representation is
  5332 slightly different -- use 'notation' instead of raw 'syntax', and
  5333 'translations' with explicit "CONST" markup to accommodate this.
  5334 
  5335 * Pure/Isar: unified syntax for new-style specification mechanisms
  5336 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5337 full type inference and dummy patterns ("_").  For example:
  5338 
  5339   definition "K x _ = x"
  5340 
  5341   inductive conj for A B
  5342   where "A ==> B ==> conj A B"
  5343 
  5344 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5345 the current context.  Print mode "no_abbrevs" prevents inversion of
  5346 abbreviations on output.
  5347 
  5348 * Isar/locales: improved parameter handling: use of locales "var" and
  5349 "struct" no longer necessary; - parameter renamings are no longer
  5350 required to be injective.  For example, this allows to define
  5351 endomorphisms as locale endom = homom mult mult h.
  5352 
  5353 * Isar/locales: changed the way locales with predicates are defined.
  5354 Instead of accumulating the specification, the imported expression is
  5355 now an interpretation.  INCOMPATIBILITY: different normal form of
  5356 locale expressions.  In particular, in interpretations of locales with
  5357 predicates, goals repesenting already interpreted fragments are not
  5358 removed automatically.  Use methods `intro_locales' and
  5359 `unfold_locales'; see below.
  5360 
  5361 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5362 provide backward reasoning on locales predicates.  The methods are
  5363 aware of interpretations and discharge corresponding goals.
  5364 `intro_locales' is less aggressive then `unfold_locales' and does not
  5365 unfold predicates to assumptions.
  5366 
  5367 * Isar/locales: the order in which locale fragments are accumulated
  5368 has changed.  This enables to override declarations from fragments due
  5369 to interpretations -- for example, unwanted simp rules.
  5370 
  5371 * Isar/locales: interpretation in theories and proof contexts has been
  5372 extended.  One may now specify (and prove) equations, which are
  5373 unfolded in interpreted theorems.  This is useful for replacing
  5374 defined concepts (constants depending on locale parameters) by
  5375 concepts already existing in the target context.  Example:
  5376 
  5377   interpretation partial_order ["op <= :: [int, int] => bool"]
  5378     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5379 
  5380 Typically, the constant `partial_order.less' is created by a
  5381 definition specification element in the context of locale
  5382 partial_order.
  5383 
  5384 * Method "induct": improved internal context management to support
  5385 local fixes and defines on-the-fly. Thus explicit meta-level
  5386 connectives !!  and ==> are rarely required anymore in inductive goals
  5387 (using object-logic connectives for this purpose has been long
  5388 obsolete anyway). Common proof patterns are explained in
  5389 src/HOL/Induct/Common_Patterns.thy, see also
  5390 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5391 examples.
  5392 
  5393 * Method "induct": improved handling of simultaneous goals. Instead of
  5394 introducing object-level conjunction, the statement is now split into
  5395 several conclusions, while the corresponding symbolic cases are nested
  5396 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5397 see src/HOL/Induct/Common_Patterns.thy, for example.
  5398 
  5399 * Method "induct": mutual induction rules are now specified as a list
  5400 of rule sharing the same induction cases. HOL packages usually provide
  5401 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5402 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5403 mutual induction rules differently, i.e. like this:
  5404 
  5405   (induct rule: foo_bar.inducts)
  5406   (induct set: foo bar)
  5407   (induct pred: foo bar)
  5408   (induct type: foo bar)
  5409 
  5410 The ML function ProjectRule.projections turns old-style rules into the
  5411 new format.
  5412 
  5413 * Method "coinduct": dual of induction, see
  5414 src/HOL/Library/Coinductive_List.thy for various examples.
  5415 
  5416 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5417 considered a legacy feature.
  5418 
  5419 * Attribute "symmetric" produces result with standardized schematic
  5420 variables (index 0).  Potential INCOMPATIBILITY.
  5421 
  5422 * Simplifier: by default the simplifier trace only shows top level
  5423 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5424 default. Thus there is less danger of being flooded by the trace. The
  5425 trace indicates where parts have been suppressed.
  5426 
  5427 * Provers/classical: removed obsolete classical version of elim_format
  5428 attribute; classical elim/dest rules are now treated uniformly when
  5429 manipulating the claset.
  5430 
  5431 * Provers/classical: stricter checks to ensure that supplied intro,
  5432 dest and elim rules are well-formed; dest and elim rules must have at
  5433 least one premise.
  5434 
  5435 * Provers/classical: attributes dest/elim/intro take an optional
  5436 weight argument for the rule (just as the Pure versions).  Weights are
  5437 ignored by automated tools, but determine the search order of single
  5438 rule steps.
  5439 
  5440 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5441 where the body does not mention the bound variable.  Note that dummy
  5442 patterns implicitly depend on their context of bounds, which makes
  5443 "{_. _}" match any set comprehension as expected.  Potential
  5444 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5445 constant "_idtdummy" in the binding position.
  5446 
  5447 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5448 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5449 for example "A -> B" => "Pi A (%_. B)".
  5450 
  5451 * Pure: 'class_deps' command visualizes the subclass relation, using
  5452 the graph browser tool.
  5453 
  5454 * Pure: 'print_theory' now suppresses certain internal declarations by
  5455 default; use '!' option for full details.
  5456 
  5457 
  5458 *** HOL ***
  5459 
  5460 * Method "metis" proves goals by applying the Metis general-purpose
  5461 resolution prover (see also http://gilith.com/software/metis/).
  5462 Examples are in the directory MetisExamples.  WARNING: the
  5463 Isabelle/HOL-Metis integration does not yet work properly with
  5464 multi-threading.
  5465 
  5466 * Command 'sledgehammer' invokes external automatic theorem provers as
  5467 background processes.  It generates calls to the "metis" method if
  5468 successful. These can be pasted into the proof.  Users do not have to
  5469 wait for the automatic provers to return.  WARNING: does not really
  5470 work with multi-threading.
  5471 
  5472 * New "auto_quickcheck" feature tests outermost goal statements for
  5473 potential counter-examples.  Controlled by ML references
  5474 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5475 5000 milliseconds).  Fails silently if statements is outside of
  5476 executable fragment, or any other codgenerator problem occurs.
  5477 
  5478 * New constant "undefined" with axiom "undefined x = undefined".
  5479 
  5480 * Added class "HOL.eq", allowing for code generation with polymorphic
  5481 equality.
  5482 
  5483 * Some renaming of class constants due to canonical name prefixing in
  5484 the new 'class' package:
  5485 
  5486     HOL.abs ~> HOL.abs_class.abs
  5487     HOL.divide ~> HOL.divide_class.divide
  5488     0 ~> HOL.zero_class.zero
  5489     1 ~> HOL.one_class.one
  5490     op + ~> HOL.plus_class.plus
  5491     op - ~> HOL.minus_class.minus
  5492     uminus ~> HOL.minus_class.uminus
  5493     op * ~> HOL.times_class.times
  5494     op < ~> HOL.ord_class.less
  5495     op <= > HOL.ord_class.less_eq
  5496     Nat.power ~> Power.power_class.power
  5497     Nat.size ~> Nat.size_class.size
  5498     Numeral.number_of ~> Numeral.number_class.number_of
  5499     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5500     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5501     Orderings.min ~> Orderings.ord_class.min
  5502     Orderings.max ~> Orderings.ord_class.max
  5503     Divides.op div ~> Divides.div_class.div
  5504     Divides.op mod ~> Divides.div_class.mod
  5505     Divides.op dvd ~> Divides.div_class.dvd
  5506 
  5507 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5508 
  5509 a) User-defined constants using any of the names "plus", "minus",
  5510 "times", "less" or "less_eq". The standard syntax translations for
  5511 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5512 names.
  5513 
  5514 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5515 INCOMPATIBILITY: use more specific names.
  5516 
  5517 c) Permutative equations (e.g. "a + b = b + a")
  5518 Since the change of names also changes the order of terms, permutative
  5519 rewrite rules may get applied in a different order. Experience shows
  5520 that this is rarely the case (only two adaptions in the whole Isabelle
  5521 distribution).  INCOMPATIBILITY: rewrite proofs
  5522 
  5523 d) ML code directly refering to constant names
  5524 This in general only affects hand-written proof tactics, simprocs and
  5525 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5526 Consider using @{const_name} antiquotation.
  5527 
  5528 * New class "default" with associated constant "default".
  5529 
  5530 * Function "sgn" is now overloaded and available on int, real, complex
  5531 (and other numeric types), using class "sgn".  Two possible defs of
  5532 sgn are given as equational assumptions in the classes sgn_if and
  5533 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5534 INCOMPATIBILITY.
  5535 
  5536 * Locale "partial_order" now unified with class "order" (cf. theory
  5537 Orderings), added parameter "less".  INCOMPATIBILITY.
  5538 
  5539 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5540 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5541 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5542 
  5543 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5544 changed order of proof goals in instance proofs.
  5545 
  5546 * The transitivity reasoner for partial and linear orders is set up
  5547 for classes "order" and "linorder".  Instances of the reasoner are available
  5548 in all contexts importing or interpreting the corresponding locales.
  5549 Method "order" invokes the reasoner separately; the reasoner
  5550 is also integrated with the Simplifier as a solver.  Diagnostic
  5551 command 'print_orders' shows the available instances of the reasoner
  5552 in the current context.
  5553 
  5554 * Localized monotonicity predicate in theory "Orderings"; integrated
  5555 lemmas max_of_mono and min_of_mono with this predicate.
  5556 INCOMPATIBILITY.
  5557 
  5558 * Formulation of theorem "dense" changed slightly due to integration
  5559 with new class dense_linear_order.
  5560 
  5561 * Uniform lattice theory development in HOL.
  5562 
  5563     constants "meet" and "join" now named "inf" and "sup"
  5564     constant "Meet" now named "Inf"
  5565 
  5566     classes "meet_semilorder" and "join_semilorder" now named
  5567       "lower_semilattice" and "upper_semilattice"
  5568     class "lorder" now named "lattice"
  5569     class "comp_lat" now named "complete_lattice"
  5570 
  5571     Instantiation of lattice classes allows explicit definitions
  5572     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5573 
  5574   INCOMPATIBILITY.  Theorem renames:
  5575 
  5576     meet_left_le            ~> inf_le1
  5577     meet_right_le           ~> inf_le2
  5578     join_left_le            ~> sup_ge1
  5579     join_right_le           ~> sup_ge2
  5580     meet_join_le            ~> inf_sup_ord
  5581     le_meetI                ~> le_infI
  5582     join_leI                ~> le_supI
  5583     le_meet                 ~> le_inf_iff
  5584     le_join                 ~> ge_sup_conv
  5585     meet_idempotent         ~> inf_idem
  5586     join_idempotent         ~> sup_idem
  5587     meet_comm               ~> inf_commute
  5588     join_comm               ~> sup_commute
  5589     meet_leI1               ~> le_infI1
  5590     meet_leI2               ~> le_infI2
  5591     le_joinI1               ~> le_supI1
  5592     le_joinI2               ~> le_supI2
  5593     meet_assoc              ~> inf_assoc
  5594     join_assoc              ~> sup_assoc
  5595     meet_left_comm          ~> inf_left_commute
  5596     meet_left_idempotent    ~> inf_left_idem
  5597     join_left_comm          ~> sup_left_commute
  5598     join_left_idempotent    ~> sup_left_idem
  5599     meet_aci                ~> inf_aci
  5600     join_aci                ~> sup_aci
  5601     le_def_meet             ~> le_iff_inf
  5602     le_def_join             ~> le_iff_sup
  5603     join_absorp2            ~> sup_absorb2
  5604     join_absorp1            ~> sup_absorb1
  5605     meet_absorp1            ~> inf_absorb1
  5606     meet_absorp2            ~> inf_absorb2
  5607     meet_join_absorp        ~> inf_sup_absorb
  5608     join_meet_absorp        ~> sup_inf_absorb
  5609     distrib_join_le         ~> distrib_sup_le
  5610     distrib_meet_le         ~> distrib_inf_le
  5611 
  5612     add_meet_distrib_left   ~> add_inf_distrib_left
  5613     add_join_distrib_left   ~> add_sup_distrib_left
  5614     is_join_neg_meet        ~> is_join_neg_inf
  5615     is_meet_neg_join        ~> is_meet_neg_sup
  5616     add_meet_distrib_right  ~> add_inf_distrib_right
  5617     add_join_distrib_right  ~> add_sup_distrib_right
  5618     add_meet_join_distribs  ~> add_sup_inf_distribs
  5619     join_eq_neg_meet        ~> sup_eq_neg_inf
  5620     meet_eq_neg_join        ~> inf_eq_neg_sup
  5621     add_eq_meet_join        ~> add_eq_inf_sup
  5622     meet_0_imp_0            ~> inf_0_imp_0
  5623     join_0_imp_0            ~> sup_0_imp_0
  5624     meet_0_eq_0             ~> inf_0_eq_0
  5625     join_0_eq_0             ~> sup_0_eq_0
  5626     neg_meet_eq_join        ~> neg_inf_eq_sup
  5627     neg_join_eq_meet        ~> neg_sup_eq_inf
  5628     join_eq_if              ~> sup_eq_if
  5629 
  5630     mono_meet               ~> mono_inf
  5631     mono_join               ~> mono_sup
  5632     meet_bool_eq            ~> inf_bool_eq
  5633     join_bool_eq            ~> sup_bool_eq
  5634     meet_fun_eq             ~> inf_fun_eq
  5635     join_fun_eq             ~> sup_fun_eq
  5636     meet_set_eq             ~> inf_set_eq
  5637     join_set_eq             ~> sup_set_eq
  5638     meet1_iff               ~> inf1_iff
  5639     meet2_iff               ~> inf2_iff
  5640     meet1I                  ~> inf1I
  5641     meet2I                  ~> inf2I
  5642     meet1D1                 ~> inf1D1
  5643     meet2D1                 ~> inf2D1
  5644     meet1D2                 ~> inf1D2
  5645     meet2D2                 ~> inf2D2
  5646     meet1E                  ~> inf1E
  5647     meet2E                  ~> inf2E
  5648     join1_iff               ~> sup1_iff
  5649     join2_iff               ~> sup2_iff
  5650     join1I1                 ~> sup1I1
  5651     join2I1                 ~> sup2I1
  5652     join1I1                 ~> sup1I1
  5653     join2I2                 ~> sup1I2
  5654     join1CI                 ~> sup1CI
  5655     join2CI                 ~> sup2CI
  5656     join1E                  ~> sup1E
  5657     join2E                  ~> sup2E
  5658 
  5659     is_meet_Meet            ~> is_meet_Inf
  5660     Meet_bool_def           ~> Inf_bool_def
  5661     Meet_fun_def            ~> Inf_fun_def
  5662     Meet_greatest           ~> Inf_greatest
  5663     Meet_lower              ~> Inf_lower
  5664     Meet_set_def            ~> Inf_set_def
  5665 
  5666     Sup_def                 ~> Sup_Inf
  5667     Sup_bool_eq             ~> Sup_bool_def
  5668     Sup_fun_eq              ~> Sup_fun_def
  5669     Sup_set_eq              ~> Sup_set_def
  5670 
  5671     listsp_meetI            ~> listsp_infI
  5672     listsp_meet_eq          ~> listsp_inf_eq
  5673 
  5674     meet_min                ~> inf_min
  5675     join_max                ~> sup_max
  5676 
  5677 * Added syntactic class "size"; overloaded constant "size" now has
  5678 type "'a::size ==> bool"
  5679 
  5680 * Internal reorganisation of `size' of datatypes: size theorems
  5681 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5682 simplification rules by default!); theorems "prod.size" now named
  5683 "*.size".
  5684 
  5685 * Class "div" now inherits from class "times" rather than "type".
  5686 INCOMPATIBILITY.
  5687 
  5688 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5689 Linorder etc.  have disappeared; operations defined in terms of
  5690 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5691 
  5692 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5693 
  5694 * HOL-Word: New extensive library and type for generic, fixed size
  5695 machine words, with arithmetic, bit-wise, shifting and rotating
  5696 operations, reflection into int, nat, and bool lists, automation for
  5697 linear arithmetic (by automatic reflection into nat or int), including
  5698 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5699 arithmetic type classes, supporting automatic simplification of
  5700 numerals on all operations.
  5701 
  5702 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5703 
  5704 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5705 
  5706 * Code generator library theories:
  5707   - Code_Integer represents HOL integers by big integer literals in target
  5708     languages.
  5709   - Code_Char represents HOL characters by character literals in target
  5710     languages.
  5711   - Code_Char_chr like Code_Char, but also offers treatment of character
  5712     codes; includes Code_Integer.
  5713   - Executable_Set allows to generate code for finite sets using lists.
  5714   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5715     denominator).
  5716   - Executable_Real implements a subset of real numbers, namly those
  5717     representable by rational numbers.
  5718   - Efficient_Nat implements natural numbers by integers, which in general will
  5719     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5720     includes Code_Integer.
  5721   - Code_Index provides an additional datatype index which is mapped to
  5722     target-language built-in integers.
  5723   - Code_Message provides an additional datatype message_string which is isomorphic to
  5724     strings; messages are mapped to target-language strings.
  5725 
  5726 * New package for inductive predicates
  5727 
  5728   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5729 
  5730     inductive
  5731       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5732       for z_1 :: U_1 and ... and z_n :: U_m
  5733     where
  5734       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5735     | ...
  5736 
  5737   with full support for type-inference, rather than
  5738 
  5739     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5740 
  5741     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5742     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5743 
  5744     inductive "s z_1 ... z_m"
  5745     intros
  5746       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5747       ...
  5748 
  5749   For backward compatibility, there is a wrapper allowing inductive
  5750   sets to be defined with the new package via
  5751 
  5752     inductive_set
  5753       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5754       for z_1 :: U_1 and ... and z_n :: U_m
  5755     where
  5756       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5757     | ...
  5758 
  5759   or
  5760 
  5761     inductive_set
  5762       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5763       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5764       for z_1 :: U_1 and ... and z_n :: U_m
  5765     where
  5766       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5767     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5768     | ...
  5769 
  5770   if the additional syntax "p ..." is required.
  5771 
  5772   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5773   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5774 
  5775   INCOMPATIBILITIES:
  5776 
  5777   - Since declaration and definition of inductive sets or predicates
  5778     is no longer separated, abbreviations involving the newly
  5779     introduced sets or predicates must be specified together with the
  5780     introduction rules after the 'where' keyword (see above), rather
  5781     than before the actual inductive definition.
  5782 
  5783   - The variables in induction and elimination rules are now
  5784     quantified in the order of their occurrence in the introduction
  5785     rules, rather than in alphabetical order. Since this may break
  5786     some proofs, these proofs either have to be repaired, e.g. by
  5787     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5788     statements of the form
  5789 
  5790       case (rule_i a_i_1 ... a_i_{k_i})
  5791 
  5792     or the old order of quantification has to be restored by explicitly adding
  5793     meta-level quantifiers in the introduction rules, i.e.
  5794 
  5795       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5796 
  5797   - The format of the elimination rules is now
  5798 
  5799       p z_1 ... z_m x_1 ... x_n ==>
  5800         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5801         ==> ... ==> P
  5802 
  5803     for predicates and
  5804 
  5805       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5806         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5807         ==> ... ==> P
  5808 
  5809     for sets rather than
  5810 
  5811       x : s z_1 ... z_m ==>
  5812         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5813         ==> ... ==> P
  5814 
  5815     This may require terms in goals to be expanded to n-tuples
  5816     (e.g. using case_tac or simplification with the split_paired_all
  5817     rule) before the above elimination rule is applicable.
  5818 
  5819   - The elimination or case analysis rules for (mutually) inductive
  5820     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5821     list of rules "p_1_..._p_k.elims" is no longer available.
  5822 
  5823 * New package "function"/"fun" for general recursive functions,
  5824 supporting mutual and nested recursion, definitions in local contexts,
  5825 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5826 for small examples, and the separate tutorial on the function
  5827 package. The old recdef "package" is still available as before, but
  5828 users are encouraged to use the new package.
  5829 
  5830 * Method "lexicographic_order" automatically synthesizes termination
  5831 relations as lexicographic combinations of size measures.
  5832 
  5833 * Case-expressions allow arbitrary constructor-patterns (including
  5834 "_") and take their order into account, like in functional
  5835 programming.  Internally, this is translated into nested
  5836 case-expressions; missing cases are added and mapped to the predefined
  5837 constant "undefined". In complicated cases printing may no longer show
  5838 the original input but the internal form. Lambda-abstractions allow
  5839 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5840 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5841 variable.
  5842 
  5843 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5844 is an abbreviation for "of_nat :: nat => int". The simplification
  5845 rules for "of_nat" have been changed to work like "int" did
  5846 previously.  Potential INCOMPATIBILITY:
  5847   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5848   - of_nat_diff and of_nat_mult are no longer default simp rules
  5849 
  5850 * Method "algebra" solves polynomial equations over (semi)rings using
  5851 Groebner bases. The (semi)ring structure is defined by locales and the
  5852 tool setup depends on that generic context. Installing the method for
  5853 a specific type involves instantiating the locale and possibly adding
  5854 declarations for computation on the coefficients.  The method is
  5855 already instantiated for natural numbers and for the axiomatic class
  5856 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5857 CALCULEMUS 2007 for the general principles underlying this
  5858 architecture of context-aware proof-tools.
  5859 
  5860 * Method "ferrack" implements quantifier elimination over
  5861 special-purpose dense linear orders using locales (analogous to
  5862 "algebra"). The method is already installed for class
  5863 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5864 rat, etc.
  5865 
  5866 * Former constant "List.op @" now named "List.append".  Use ML
  5867 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5868 circumvent possible incompatibilities when working on ML level.
  5869 
  5870 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5871 
  5872 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5873 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5874 
  5875 * New syntax for Haskell-like list comprehension (input only), eg.
  5876 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5877 
  5878 * The special syntax for function "filter" has changed from [x :
  5879 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5880 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5881 
  5882 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5883 defined on nat only, as an abbreviation for [a..<Suc b]
  5884 INCOMPATIBILITY.
  5885 
  5886 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5887 
  5888 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5889 
  5890 * New lemma collection field_simps (an extension of ring_simps) for
  5891 manipulating (in)equations involving division. Multiplies with all
  5892 denominators that can be proved to be non-zero (in equations) or
  5893 positive/negative (in inequations).
  5894 
  5895 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5896 have been improved and renamed to ring_simps, group_simps and
  5897 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5898 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5899 
  5900 * Theory Library/Commutative_Ring: switched from recdef to function
  5901 package; constants add, mul, pow now curried.  Infix syntax for
  5902 algebraic operations.
  5903 
  5904 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5905 INCOMPATIBILITY.
  5906 
  5907 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5908 INCOMPATIBILITY.
  5909 
  5910 * HOL/records: generalised field-update to take a function on the
  5911 field rather than the new value: r(|A := x|) is translated to A_update
  5912 (K x) r The K-combinator that is internally used is called K_record.
  5913 INCOMPATIBILITY: Usage of the plain update functions has to be
  5914 adapted.
  5915 
  5916 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5917 * x = 0, which are required for a semiring.  Richer structures do not
  5918 inherit from semiring_0 anymore, because this property is a theorem
  5919 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5920 there is more to prove, but this is mostly trivial.
  5921 
  5922 * Class "recpower" is generalized to arbitrary monoids, not just
  5923 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5924 commutativity or semiring properties additionally.
  5925 
  5926 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5927 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5928 better use 'abbreviation'.
  5929 
  5930 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5931 
  5932 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5933 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5934 significant changes for setting up numeral syntax for types:
  5935   - New constants Numeral.pred and Numeral.succ instead
  5936       of former Numeral.bin_pred and Numeral.bin_succ.
  5937   - Use integer operations instead of bin_add, bin_mult and so on.
  5938   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5939   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5940 
  5941 See src/HOL/Integ/IntArith.thy for an example setup.
  5942 
  5943 * Command 'normal_form' computes the normal form of a term that may
  5944 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5945 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5946 heavy-duty computations because the functions are compiled to ML
  5947 first.  Correspondingly, a method "normalization" is provided.  See
  5948 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5949 
  5950 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5951 25 like -->); output depends on the "iff" print_mode, the default is
  5952 "A = B" (with priority 50).
  5953 
  5954 * Relations less (<) and less_eq (<=) are also available on type bool.
  5955 Modified syntax to disallow nesting without explicit parentheses,
  5956 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5957 INCOMPATIBILITY.
  5958 
  5959 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5960 
  5961 * Relation composition operator "op O" now has precedence 75 and binds
  5962 stronger than union and intersection. INCOMPATIBILITY.
  5963 
  5964 * The old set interval syntax "{m..n(}" (and relatives) has been
  5965 removed.  Use "{m..<n}" (and relatives) instead.
  5966 
  5967 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5968 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5969 ``declare [[simproc del: neq]]''.
  5970 
  5971 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5972 True/False.
  5973 
  5974 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5975 
  5976 * Theorem setsum_mult renamed to setsum_right_distrib.
  5977 
  5978 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5979 ``rule'' method.
  5980 
  5981 * Reimplemented methods "sat" and "satx", with several improvements:
  5982 goals no longer need to be stated as "<prems> ==> False", equivalences
  5983 (i.e. "=" on type bool) are handled, variable names of the form
  5984 "lit_<n>" are no longer reserved, significant speedup.
  5985 
  5986 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5987 zChaff is still supported as well.
  5988 
  5989 * 'inductive' and 'datatype': provide projections of mutual rules,
  5990 bundled as foo_bar.inducts;
  5991 
  5992 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5993 Library.
  5994 
  5995 * Library: moved theory Accessible_Part to main HOL.
  5996 
  5997 * Library: added theory Coinductive_List of potentially infinite lists
  5998 as greatest fixed-point.
  5999 
  6000 * Library: added theory AssocList which implements (finite) maps as
  6001 association lists.
  6002 
  6003 * Method "evaluation" solves goals (i.e. a boolean expression)
  6004 efficiently by compiling it to ML.  The goal is "proved" (via an
  6005 oracle) if it evaluates to True.
  6006 
  6007 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  6008 also when invoked by the simplifier.  This results in the Simplifier
  6009 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  6010 Configuration option fast_arith_split_limit=0 recovers the old
  6011 behavior.
  6012 
  6013 * Support for hex (0x20) and binary (0b1001) numerals.
  6014 
  6015 * New method: reify eqs (t), where eqs are equations for an
  6016 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  6017 parameter, computes a term s::'b and a list xs::'a list and proves the
  6018 theorem I xs s = t. This is also known as reification or quoting. The
  6019 resulting theorem is applied to the subgoal to substitute t with I xs
  6020 s.  If t is omitted, the subgoal itself is reified.
  6021 
  6022 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  6023 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  6024 where f is supposed to be a computable function (in the sense of code
  6025 generattion). The method uses reify to compute s and xs as above then
  6026 applies corr_thm and uses normalization by evaluation to "prove" f s =
  6027 r and finally gets the theorem t = r, which is again applied to the
  6028 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  6029 
  6030 * Reflection: Automatic reification now handels binding, an example is
  6031 available in src/HOL/ex/ReflectionEx.thy
  6032 
  6033 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  6034 command 'statespace' that is similar to 'record', but introduces an
  6035 abstract specification based on the locale infrastructure instead of
  6036 HOL types.  This leads to extra flexibility in composing state spaces,
  6037 in particular multiple inheritance and renaming of components.
  6038 
  6039 
  6040 *** HOL-Complex ***
  6041 
  6042 * Hyperreal: Functions root and sqrt are now defined on negative real
  6043 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  6044 Nonnegativity side conditions have been removed from many lemmas, so
  6045 that more subgoals may now be solved by simplification; potential
  6046 INCOMPATIBILITY.
  6047 
  6048 * Real: new type classes formalize real normed vector spaces and
  6049 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  6050 and norm :: 'a => real.
  6051 
  6052 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  6053 reals into other types. The overloaded constant Reals :: 'a set is now
  6054 defined as range of_real; potential INCOMPATIBILITY.
  6055 
  6056 * Real: proper support for ML code generation, including 'quickcheck'.
  6057 Reals are implemented as arbitrary precision rationals.
  6058 
  6059 * Hyperreal: Several constants that previously worked only for the
  6060 reals have been generalized, so they now work over arbitrary vector
  6061 spaces. Type annotations may need to be added in some cases; potential
  6062 INCOMPATIBILITY.
  6063 
  6064   Infinitesimal  :: ('a::real_normed_vector) star set
  6065   HFinite        :: ('a::real_normed_vector) star set
  6066   HInfinite      :: ('a::real_normed_vector) star set
  6067   approx         :: ('a::real_normed_vector) star => 'a star => bool
  6068   monad          :: ('a::real_normed_vector) star => 'a star set
  6069   galaxy         :: ('a::real_normed_vector) star => 'a star set
  6070   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  6071   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  6072   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  6073   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  6074   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  6075   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  6076   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6077   sgn            :: 'a::real_normed_vector => 'a
  6078   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6079 
  6080 * Complex: Some complex-specific constants are now abbreviations for
  6081 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6082 hnorm.  Other constants have been entirely removed in favor of the
  6083 polymorphic versions (INCOMPATIBILITY):
  6084 
  6085   approx        <-- capprox
  6086   HFinite       <-- CFinite
  6087   HInfinite     <-- CInfinite
  6088   Infinitesimal <-- CInfinitesimal
  6089   monad         <-- cmonad
  6090   galaxy        <-- cgalaxy
  6091   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  6092   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  6093   (ns)deriv     <-- (ns)cderiv
  6094 
  6095 
  6096 *** HOL-Algebra ***
  6097 
  6098 * Formalisation of ideals and the quotient construction over rings.
  6099 
  6100 * Order and lattice theory no longer based on records.
  6101 INCOMPATIBILITY.
  6102 
  6103 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  6104 greatest_closed.  INCOMPATIBILITY.
  6105 
  6106 * Method algebra is now set up via an attribute.  For examples see
  6107 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  6108 of algebraic structures.
  6109 
  6110 * Renamed theory CRing to Ring.
  6111 
  6112 
  6113 *** HOL-Nominal ***
  6114 
  6115 * Substantial, yet incomplete support for nominal datatypes (binding
  6116 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  6117 src/HOL/Nominal/Examples.  Prospective users should consult
  6118 http://isabelle.in.tum.de/nominal/
  6119 
  6120 
  6121 *** ML ***
  6122 
  6123 * ML basics: just one true type int, which coincides with IntInf.int
  6124 (even on SML/NJ).
  6125 
  6126 * ML within Isar: antiquotations allow to embed statically-checked
  6127 formal entities in the source, referring to the context available at
  6128 compile-time.  For example:
  6129 
  6130 ML {* @{sort "{zero,one}"} *}
  6131 ML {* @{typ "'a => 'b"} *}
  6132 ML {* @{term "%x. x"} *}
  6133 ML {* @{prop "x == y"} *}
  6134 ML {* @{ctyp "'a => 'b"} *}
  6135 ML {* @{cterm "%x. x"} *}
  6136 ML {* @{cprop "x == y"} *}
  6137 ML {* @{thm asm_rl} *}
  6138 ML {* @{thms asm_rl} *}
  6139 ML {* @{type_name c} *}
  6140 ML {* @{type_syntax c} *}
  6141 ML {* @{const_name c} *}
  6142 ML {* @{const_syntax c} *}
  6143 ML {* @{context} *}
  6144 ML {* @{theory} *}
  6145 ML {* @{theory Pure} *}
  6146 ML {* @{theory_ref} *}
  6147 ML {* @{theory_ref Pure} *}
  6148 ML {* @{simpset} *}
  6149 ML {* @{claset} *}
  6150 ML {* @{clasimpset} *}
  6151 
  6152 The same works for sources being ``used'' within an Isar context.
  6153 
  6154 * ML in Isar: improved error reporting; extra verbosity with
  6155 ML_Context.trace enabled.
  6156 
  6157 * Pure/General/table.ML: the join operations now works via exceptions
  6158 DUP/SAME instead of type option. This is simpler in simple cases, and
  6159 admits slightly more efficient complex applications.
  6160 
  6161 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6162 use Context.generic instead of just theory.
  6163 
  6164 * Pure: datatype Context.generic joins theory/Proof.context and
  6165 provides some facilities for code that works in either kind of
  6166 context, notably GenericDataFun for uniform theory and proof data.
  6167 
  6168 * Pure: simplified internal attribute type, which is now always
  6169 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6170 local (Proof.context) attributes have been discontinued, while
  6171 minimizing code duplication. Thm.rule_attribute and
  6172 Thm.declaration_attribute build canonical attributes; see also structure
  6173 Context for further operations on Context.generic, notably
  6174 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6175 declarations and definitions.
  6176 
  6177 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6178 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6179 init only required for impure data. INCOMPATIBILITY: empty really need
  6180 to be empty (no dependencies on theory content!)
  6181 
  6182 * Pure/kernel: consts certification ignores sort constraints given in
  6183 signature declarations. (This information is not relevant to the
  6184 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6185 potential INCOMPATIBILITY.
  6186 
  6187 * Pure: axiomatic type classes are now purely definitional, with
  6188 explicit proofs of class axioms and super class relations performed
  6189 internally. See Pure/axclass.ML for the main internal interfaces --
  6190 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6191 AxClass.axiomatize_class/classrel/arity supersede
  6192 Sign.add_classes/classrel/arities.
  6193 
  6194 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6195 global/local versions on theory vs. Proof.context have been
  6196 discontinued; Attrib.syntax and Method.syntax have been adapted
  6197 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6198 attributes, methods, etc.
  6199 
  6200 * Pure: several functions of signature "... -> theory -> theory * ..."
  6201 have been reoriented to "... -> theory -> ... * theory" in order to
  6202 allow natural usage in combination with the ||>, ||>>, |-> and
  6203 fold_map combinators.
  6204 
  6205 * Pure: official theorem names (closed derivations) and additional
  6206 comments (tags) are now strictly separate.  Name hints -- which are
  6207 maintained as tags -- may be attached any time without affecting the
  6208 derivation.
  6209 
  6210 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6211 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6212 achieve the old behaviour.
  6213 
  6214 * Pure: the "Goal" constant is now called "prop", supporting a
  6215 slightly more general idea of ``protecting'' meta-level rule
  6216 statements.
  6217 
  6218 * Pure: Logic.(un)varify only works in a global context, which is now
  6219 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6220 Logic.legacy_(un)varify as temporary workaround.
  6221 
  6222 * Pure: structure Name provides scalable operations for generating
  6223 internal variable names, notably Name.variants etc.  This replaces
  6224 some popular functions from term.ML:
  6225 
  6226   Term.variant		->  Name.variant
  6227   Term.variantlist	->  Name.variant_list
  6228   Term.invent_names	->  Name.invent_list
  6229 
  6230 Note that low-level renaming rarely occurs in new code -- operations
  6231 from structure Variable are used instead (see below).
  6232 
  6233 * Pure: structure Variable provides fundamental operations for proper
  6234 treatment of fixed/schematic variables in a context.  For example,
  6235 Variable.import introduces fixes for schematics of given facts and
  6236 Variable.export reverses the effect (up to renaming) -- this replaces
  6237 various freeze_thaw operations.
  6238 
  6239 * Pure: structure Goal provides simple interfaces for