NEWS
author wenzelm
Tue Nov 20 15:18:11 2012 +0100 (2012-11-20)
changeset 50132 180d086c30dd
parent 50126 3dec88149176
child 50138 ca989d793b34
permissions -rw-r--r--
simplified command line of "isabelle install";
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Theorem status about oracles and unfinished/failed future proofs is
    10 no longer printed by default, since it is incompatible with
    11 incremental / parallel checking of the persistent document model.  ML
    12 function Thm.peek_status may be used to inspect a snapshot of the
    13 ongoing evaluation process.  Note that in batch mode --- notably
    14 isabelle build --- the system ensures that future proofs of all
    15 accessible theorems in the theory context are finished (as before).
    16 
    17 * Configuration option show_markup controls direct inlining of markup
    18 into the printed representation of formal entities --- notably type
    19 and sort constraints.  This enables Prover IDE users to retrieve that
    20 information via tooltips in the output window, for example.
    21 
    22 * Command 'ML_file' evaluates ML text from a file directly within the
    23 theory, without any predeclaration via 'uses' in the theory header.
    24 
    25 * Old command 'use' command and corresponding keyword 'uses' in the
    26 theory header are legacy features and will be discontinued soon.
    27 Tools that load their additional source files may imitate the
    28 'ML_file' implementation, such that the system can take care of
    29 dependencies properly.
    30 
    31 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    32 is called fastforce / fast_force_tac already since Isabelle2011-1.
    33 
    34 * Updated and extended "isar-ref" and "implementation" manual, reduced
    35 remaining material in old "ref" manual.
    36 
    37 * Improved support for auxiliary contexts indicate block structure for
    38 specifications: nesting of "context fixes ... context assumes ..."
    39 and "class ... context ...".
    40 
    41 * More informative error messages for Isar proof commands involving
    42 lazy enumerations (method applications etc.).
    43 
    44 
    45 *** Prover IDE -- Isabelle/Scala/jEdit ***
    46 
    47 * Parallel terminal proofs ('by') are enabled by default, likewise
    48 proofs that are built into packages like 'datatype', 'function'.  This
    49 allows to "run ahead" checking the theory specifications on the
    50 surface, while the prover is still crunching on internal
    51 justifications.  Unfinished / cancelled proofs are restarted as
    52 required to complete full proof checking eventually.
    53 
    54 * Improved output panel with tooltips, hyperlinks etc. based on the
    55 same Rich_Text_Area as regular Isabelle/jEdit buffers.  Activation of
    56 tooltips leads to some window that supports the same recursively,
    57 which can lead to stacks of tooltips as the semantic document content
    58 is explored.  ESCAPE closes the whole stack, individual windows may be
    59 closed separately, or detached to become independent jEdit dockables.
    60 
    61 * More robust incremental parsing of outer syntax (partial comments,
    62 malformed symbols).  Changing the balance of open/close quotes and
    63 comment delimiters works more conveniently with unfinished situations
    64 that frequently occur in user interaction.
    65 
    66 * More efficient painting and improved reactivity when editing large
    67 files.  More scalable management of formal document content.
    68 
    69 * Smarter handling of tracing messages: output window informs about
    70 accumulated messages; prover transactions are limited to emit maximum
    71 amount of output, before being canceled (cf. system option
    72 "editor_tracing_limit").  This avoids swamping the front-end with
    73 potentially infinite message streams.
    74 
    75 * More plugin options and preferences, based on Isabelle/Scala.  The
    76 jEdit plugin option panel provides access to some Isabelle/Scala
    77 options, including tuning parameters for editor reactivity and color
    78 schemes.
    79 
    80 * Uniform Java 7 platform on Linux, Mac OS X, Windows: recent updates
    81 from Oracle provide better multi-platform experience.  This version is
    82 now bundled exclusively with Isabelle.
    83 
    84 
    85 *** Pure ***
    86 
    87 * Code generation for Haskell: restrict unqualified imports from
    88 Haskell Prelude to a small set of fundamental operations.
    89 
    90 * Command "export_code": relative file names are interpreted
    91 relatively to master directory of current theory rather than
    92 the rather arbitrary current working directory.
    93 INCOMPATIBILITY.
    94 
    95 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    96 use regular rule composition via "OF" / "THEN", or explicit proof
    97 structure instead.  Note that Isabelle/ML provides a variety of
    98 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    99 with some care where this is really required.
   100 
   101 * Command 'typ' supports an additional variant with explicit sort
   102 constraint, to infer and check the most general type conforming to a
   103 given given sort.  Example (in HOL):
   104 
   105   typ "_ * _ * bool * unit" :: finite
   106 
   107 
   108 *** HOL ***
   109 
   110 * Removed constant "chars".  Prefer "Enum.enum" on type "char"
   111 directly.  INCOMPATIBILITY.
   112 
   113 * Moved operation product, sublists and n_lists from Enum.thy
   114 to List.thy.  INCOMPATIBILITY.
   115 
   116 * Simplified 'typedef' specifications: historical options for implicit
   117 set definition and alternative name have been discontinued.  The
   118 former behavior of "typedef (open) t = A" is now the default, but
   119 written just "typedef t = A".  INCOMPATIBILITY, need to adapt theories
   120 accordingly.
   121 
   122 * Theory "Library/Multiset":
   123 
   124   - Renamed constants
   125       fold_mset ~> Multiset.fold  -- for coherence with other fold combinators
   126 
   127   - Renamed facts
   128       fold_mset_commute ~> fold_mset_comm  -- for coherence with fold_comm
   129 
   130 INCOMPATIBILITY.
   131 
   132 * Theorem UN_o generalized to SUP_comp.  INCOMPATIBILITY.
   133 
   134 * Class "comm_monoid_diff" formalises properties of bounded
   135 subtraction, with natural numbers and multisets as typical instances.
   136 
   137 * Theory "Library/Option_ord" provides instantiation of option type
   138 to lattice type classes.
   139 
   140 * New combinator "Option.these" with type "'a option set => 'a set".
   141 
   142 * Renamed theory Library/List_Prefix to Library/Sublist.
   143 INCOMPATIBILITY.  Related changes are:
   144 
   145   - Renamed constants:
   146 
   147       prefix ~> prefixeq
   148       strict_prefix ~> prefix
   149 
   150     Renamed lemmas accordingly, INCOMPATIBILITY.
   151 
   152   - Replaced constant "postfix" by "suffixeq" with swapped argument order
   153     (i.e., "postfix xs ys" is now "suffixeq ys xs") and dropped old infix
   154     syntax "xs >>= ys"; use "suffixeq ys xs" instead.  Renamed lemmas
   155     accordingly.  INCOMPATIBILITY.
   156 
   157   - New constant "emb" for homeomorphic embedding on lists. New
   158     abbreviation "sub" for special case "emb (op =)".
   159 
   160   - Library/Sublist does no longer provide "order" and "bot" type class
   161     instances for the prefix order (merely corresponding locale
   162     interpretations). The type class instances are to be found in
   163     Library/Prefix_Order. INCOMPATIBILITY.
   164 
   165   - The sublist relation from Library/Sublist_Order is now based on
   166     "Sublist.sub". Replaced lemmas:
   167 
   168       le_list_append_le_same_iff ~> Sublist.sub_append_le_same_iff
   169       le_list_append_mono ~> Sublist.emb_append_mono
   170       le_list_below_empty ~> Sublist.emb_Nil, Sublist.emb_Nil2
   171       le_list_Cons_EX ~> Sublist.emb_ConsD
   172       le_list_drop_Cons2 ~> Sublist.sub_Cons2'
   173       le_list_drop_Cons_neq ~> Sublist.sub_Cons2_neq
   174       le_list_drop_Cons ~> Sublist.sub_Cons'
   175       le_list_drop_many ~> Sublist.sub_drop_many
   176       le_list_filter_left ~> Sublist.sub_filter_left
   177       le_list_rev_drop_many ~> Sublist.sub_rev_drop_many
   178       le_list_rev_take_iff ~> Sublist.sub_append
   179       le_list_same_length ~> Sublist.sub_same_length
   180       le_list_take_many_iff ~> Sublist.sub_append'
   181       less_eq_list.drop ~> less_eq_list_drop
   182       less_eq_list.induct ~> less_eq_list_induct
   183       not_le_list_length ~> Sublist.not_sub_length
   184 
   185     INCOMPATIBILITY.
   186 
   187 * HOL/Rings: renamed lemmas
   188 
   189 left_distrib ~> distrib_right
   190 right_distrib ~> distrib_left
   191 
   192 in class semiring.  INCOMPATIBILITY.
   193 
   194 * HOL/BNF: New (co)datatype package based on bounded natural
   195 functors with support for mixed, nested recursion and interesting
   196 non-free datatypes.
   197 
   198 * HOL/Cardinals: Theories of ordinals and cardinals
   199 (supersedes the AFP entry "Ordinals_and_Cardinals").
   200 
   201 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
   202 execution for code generated towards Isabelle/ML.
   203 
   204 * Library/FinFun.thy: theory of almost everywhere constant functions
   205 (supersedes the AFP entry "Code Generation for Functions as Data").
   206 
   207 * Library/Phantom.thy: generic phantom type to make a type parameter
   208 appear in a constant's type. This alternative to adding TYPE('a) as
   209 another parameter avoids unnecessary closures in generated code.
   210 
   211 * Library/RBT_Impl.thy: efficient construction of red-black trees 
   212 from sorted associative lists. Merging two trees with rbt_union may
   213 return a structurally different tree than before. MINOR INCOMPATIBILITY.
   214 
   215 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
   216 expressions.
   217 
   218 * Preprocessing of the code generator rewrites set comprehensions into
   219 pointfree expressions.
   220 
   221 * Quickcheck:
   222 
   223   - added an optimisation for equality premises.
   224     It is switched on by default, and can be switched off by setting
   225     the configuration quickcheck_optimise_equality to false.
   226 
   227 * The SMT solver Z3 has now by default a restricted set of directly
   228 supported features. For the full set of features (div/mod, nonlinear
   229 arithmetic, datatypes/records) with potential proof reconstruction
   230 failures, enable the configuration option "z3_with_extensions".
   231 Minor INCOMPATIBILITY.
   232 
   233 * Sledgehammer:
   234 
   235   - Added MaSh relevance filter based on machine-learning; see the
   236     Sledgehammer manual for details.
   237   - Rationalized type encodings ("type_enc" option).
   238   - Renamed "kill_provers" subcommand to "kill"
   239   - Renamed options:
   240       isar_proof ~> isar_proofs
   241       isar_shrink_factor ~> isar_shrink
   242       max_relevant ~> max_facts
   243       relevance_thresholds ~> fact_thresholds
   244 
   245 
   246 *** Document preparation ***
   247 
   248 * Default for \<euro> is now based on eurosym package, instead of
   249 slightly exotic babel/greek.
   250 
   251 * Document variant NAME may use different LaTeX entry point
   252 document/root_NAME.tex if that file exists, instead of the common
   253 document/root.tex.
   254 
   255 * Simplified custom document/build script, instead of old-style
   256 document/IsaMakefile.  Minor INCOMPATIBILITY.
   257 
   258 
   259 *** ML ***
   260 
   261 * Type Seq.results and related operations support embedded error
   262 messages within lazy enumerations, and thus allow to provide
   263 informative errors in the absence of any usable results.
   264 
   265 * Renamed Position.str_of to Position.here to emphasize that this is a
   266 formal device to inline positions into message text, but not
   267 necessarily printing visible text.
   268 
   269 
   270 *** System ***
   271 
   272 * The default limit for maximum number of worker threads is now 8,
   273 instead of 4.
   274 
   275 * The ML system is configured as regular component, and no longer
   276 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   277 for home-made configurations.
   278 
   279 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   280 Minor INCOMPATIBILITY in command-line options.
   281 
   282 * Advanced support for Isabelle sessions and build management, see
   283 "system" manual for the chapter of that name, especially the "isabelle
   284 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   285 mkdir / make are rendered obsolete.
   286 
   287 * Discontinued obsolete "isabelle makeall".
   288 
   289 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   290 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   291 provides some traditional targets that invoke "isabelle build".  Note
   292 that this is inefficient!  Applications of Isabelle/HOL involving
   293 "isabelle make" should be upgraded to use "isabelle build" directly.
   294 
   295 * Discontinued obsolete Isabelle/build script, it is superseded by the
   296 regular isabelle build tool.  For example:
   297 
   298   isabelle build -s -b HOLCF
   299 
   300 * The "isabelle options" tool prints Isabelle system options, as
   301 required for "isabelle build", for example.
   302 
   303 * The "isabelle mkroot" tool prepares session root directories for use
   304 with "isabelle build", similar to former "isabelle mkdir" for
   305 "isabelle usedir".
   306 
   307 * The "isabelle components" tool helps to resolve add-on components
   308 that are not bundled, or referenced from a bare-bones repository
   309 version of Isabelle.
   310 
   311 * The "isabelle install" tool has now a simpler command-line.  Minor
   312 INCOMPATIBILITY.
   313 
   314 * Discontinued support for Poly/ML 5.2.1, which was the last version
   315 without exception positions and advanced ML compiler/toplevel
   316 configuration.
   317 
   318 * Discontinued special treatment of Proof General -- no longer guess
   319 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   320 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   321 settings manually, or use a Proof General version that has been
   322 bundled as Isabelle component.
   323 
   324 
   325 New in Isabelle2012 (May 2012)
   326 ------------------------------
   327 
   328 *** General ***
   329 
   330 * Prover IDE (PIDE) improvements:
   331 
   332   - more robust Sledgehammer integration (as before the sledgehammer
   333     command-line needs to be typed into the source buffer)
   334   - markup for bound variables
   335   - markup for types of term variables (displayed as tooltips)
   336   - support for user-defined Isar commands within the running session
   337   - improved support for Unicode outside original 16bit range
   338     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   339 
   340 * Forward declaration of outer syntax keywords within the theory
   341 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   342 commands to be used in the same theory where defined.
   343 
   344 * Auxiliary contexts indicate block structure for specifications with
   345 additional parameters and assumptions.  Such unnamed contexts may be
   346 nested within other targets, like 'theory', 'locale', 'class',
   347 'instantiation' etc.  Results from the local context are generalized
   348 accordingly and applied to the enclosing target context.  Example:
   349 
   350   context
   351     fixes x y z :: 'a
   352     assumes xy: "x = y" and yz: "y = z"
   353   begin
   354 
   355   lemma my_trans: "x = z" using xy yz by simp
   356 
   357   end
   358 
   359   thm my_trans
   360 
   361 The most basic application is to factor-out context elements of
   362 several fixes/assumes/shows theorem statements, e.g. see
   363 ~~/src/HOL/Isar_Examples/Group_Context.thy
   364 
   365 Any other local theory specification element works within the "context
   366 ... begin ... end" block as well.
   367 
   368 * Bundled declarations associate attributed fact expressions with a
   369 given name in the context.  These may be later included in other
   370 contexts.  This allows to manage context extensions casually, without
   371 the logical dependencies of locales and locale interpretation.  See
   372 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   373 
   374 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   375 declaration, and results are standardized before being stored.  Thus
   376 old-style "standard" after instantiation or composition of facts
   377 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   378 indices of schematic variables.
   379 
   380 * Rule attributes in local theory declarations (e.g. locale or class)
   381 are now statically evaluated: the resulting theorem is stored instead
   382 of the original expression.  INCOMPATIBILITY in rare situations, where
   383 the historic accident of dynamic re-evaluation in interpretations
   384 etc. was exploited.
   385 
   386 * New tutorial "Programming and Proving in Isabelle/HOL"
   387 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   388 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   389 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   390 Logic" as the recommended beginners tutorial, but does not cover all
   391 of the material of that old tutorial.
   392 
   393 * Updated and extended reference manuals: "isar-ref",
   394 "implementation", "system"; reduced remaining material in old "ref"
   395 manual.
   396 
   397 
   398 *** Pure ***
   399 
   400 * Command 'definition' no longer exports the foundational "raw_def"
   401 into the user context.  Minor INCOMPATIBILITY, may use the regular
   402 "def" result with attribute "abs_def" to imitate the old version.
   403 
   404 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   405 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   406 expand it.  This also works for object-logic equality.  (Formerly
   407 undocumented feature.)
   408 
   409 * Sort constraints are now propagated in simultaneous statements, just
   410 like type constraints.  INCOMPATIBILITY in rare situations, where
   411 distinct sorts used to be assigned accidentally.  For example:
   412 
   413   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   414 
   415   lemma "P (x::'a)" and "Q (y::'a::bar)"
   416     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   417 
   418 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   419 tolerant against multiple unifiers, as long as the final result is
   420 unique.  (As before, rules are composed in canonical right-to-left
   421 order to accommodate newly introduced premises.)
   422 
   423 * Renamed some inner syntax categories:
   424 
   425     num ~> num_token
   426     xnum ~> xnum_token
   427     xstr ~> str_token
   428 
   429 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   430 "num_position" etc. are mainly used instead (which also include
   431 position information via constraints).
   432 
   433 * Simplified configuration options for syntax ambiguity: see
   434 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   435 manual.  Minor INCOMPATIBILITY.
   436 
   437 * Discontinued configuration option "syntax_positions": atomic terms
   438 in parse trees are always annotated by position constraints.
   439 
   440 * Old code generator for SML and its commands 'code_module',
   441 'code_library', 'consts_code', 'types_code' have been discontinued.
   442 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   443 
   444 * Redundant attribute "code_inline" has been discontinued. Use
   445 "code_unfold" instead.  INCOMPATIBILITY.
   446 
   447 * Dropped attribute "code_unfold_post" in favor of the its dual
   448 "code_abbrev", which yields a common pattern in definitions like
   449 
   450   definition [code_abbrev]: "f = t"
   451 
   452 INCOMPATIBILITY.
   453 
   454 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   455 instead.  INCOMPATIBILITY.
   456 
   457 * Discontinued old "prems" fact, which used to refer to the accidental
   458 collection of foundational premises in the context (already marked as
   459 legacy since Isabelle2011).
   460 
   461 
   462 *** HOL ***
   463 
   464 * Type 'a set is now a proper type constructor (just as before
   465 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   466 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   467 sets separate, it is often sufficient to rephrase some set S that has
   468 been accidentally used as predicates by "%x. x : S", and some
   469 predicate P that has been accidentally used as set by "{x. P x}".
   470 Corresponding proofs in a first step should be pruned from any
   471 tinkering with former theorems mem_def and Collect_def as far as
   472 possible.
   473 
   474 For developments which deliberately mix predicates and sets, a
   475 planning step is necessary to determine what should become a predicate
   476 and what a set.  It can be helpful to carry out that step in
   477 Isabelle2011-1 before jumping right into the current release.
   478 
   479 * Code generation by default implements sets as container type rather
   480 than predicates.  INCOMPATIBILITY.
   481 
   482 * New type synonym 'a rel = ('a * 'a) set
   483 
   484 * The representation of numerals has changed.  Datatype "num"
   485 represents strictly positive binary numerals, along with functions
   486 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   487 positive and negated numeric literals, respectively.  See also
   488 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   489 user theories may require adaptations as follows:
   490 
   491   - Theorems with number_ring or number_semiring constraints: These
   492     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   493 
   494   - Theories defining numeric types: Remove number, number_semiring,
   495     and number_ring instances. Defer all theorems about numerals until
   496     after classes one and semigroup_add have been instantiated.
   497 
   498   - Numeral-only simp rules: Replace each rule having a "number_of v"
   499     pattern with two copies, one for numeral and one for neg_numeral.
   500 
   501   - Theorems about subclasses of semiring_1 or ring_1: These classes
   502     automatically support numerals now, so more simp rules and
   503     simprocs may now apply within the proof.
   504 
   505   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   506     Redefine using other integer operations.
   507 
   508 * Transfer: New package intended to generalize the existing
   509 "descending" method and related theorem attributes from the Quotient
   510 package.  (Not all functionality is implemented yet, but future
   511 development will focus on Transfer as an eventual replacement for the
   512 corresponding parts of the Quotient package.)
   513 
   514   - transfer_rule attribute: Maintains a collection of transfer rules,
   515     which relate constants at two different types. Transfer rules may
   516     relate different type instances of the same polymorphic constant,
   517     or they may relate an operation on a raw type to a corresponding
   518     operation on an abstract type (quotient or subtype). For example:
   519 
   520     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   521     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   522 
   523   - transfer method: Replaces a subgoal on abstract types with an
   524     equivalent subgoal on the corresponding raw types. Constants are
   525     replaced with corresponding ones according to the transfer rules.
   526     Goals are generalized over all free variables by default; this is
   527     necessary for variables whose types change, but can be overridden
   528     for specific variables with e.g. "transfer fixing: x y z".  The
   529     variant transfer' method allows replacing a subgoal with one that
   530     is logically stronger (rather than equivalent).
   531 
   532   - relator_eq attribute: Collects identity laws for relators of
   533     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   534     transfer method uses these lemmas to infer transfer rules for
   535     non-polymorphic constants on the fly.
   536 
   537   - transfer_prover method: Assists with proving a transfer rule for a
   538     new constant, provided the constant is defined in terms of other
   539     constants that already have transfer rules. It should be applied
   540     after unfolding the constant definitions.
   541 
   542   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   543     from type nat to type int.
   544 
   545 * Lifting: New package intended to generalize the quotient_definition
   546 facility of the Quotient package; designed to work with Transfer.
   547 
   548   - lift_definition command: Defines operations on an abstract type in
   549     terms of a corresponding operation on a representation
   550     type.  Example syntax:
   551 
   552     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   553       is List.insert
   554 
   555     Users must discharge a respectfulness proof obligation when each
   556     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   557     the proof is discharged automatically.) The obligation is
   558     presented in a user-friendly, readable form; a respectfulness
   559     theorem in the standard format and a transfer rule are generated
   560     by the package.
   561 
   562   - Integration with code_abstype: For typedefs (e.g. subtypes
   563     corresponding to a datatype invariant, such as dlist),
   564     lift_definition generates a code certificate theorem and sets up
   565     code generation for each constant.
   566 
   567   - setup_lifting command: Sets up the Lifting package to work with a
   568     user-defined type. The user must provide either a quotient theorem
   569     or a type_definition theorem.  The package configures transfer
   570     rules for equality and quantifiers on the type, and sets up the
   571     lift_definition command to work with the type.
   572 
   573   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   574     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   575     Word/Word.thy and Library/Float.thy.
   576 
   577 * Quotient package:
   578 
   579   - The 'quotient_type' command now supports a 'morphisms' option with
   580     rep and abs functions, similar to typedef.
   581 
   582   - 'quotient_type' sets up new types to work with the Lifting and
   583     Transfer packages, as with 'setup_lifting'.
   584 
   585   - The 'quotient_definition' command now requires the user to prove a
   586     respectfulness property at the point where the constant is
   587     defined, similar to lift_definition; INCOMPATIBILITY.
   588 
   589   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   590     accordingly, INCOMPATIBILITY.
   591 
   592 * New diagnostic command 'find_unused_assms' to find potentially
   593 superfluous assumptions in theorems using Quickcheck.
   594 
   595 * Quickcheck:
   596 
   597   - Quickcheck returns variable assignments as counterexamples, which
   598     allows to reveal the underspecification of functions under test.
   599     For example, refuting "hd xs = x", it presents the variable
   600     assignment xs = [] and x = a1 as a counterexample, assuming that
   601     any property is false whenever "hd []" occurs in it.
   602 
   603     These counterexample are marked as potentially spurious, as
   604     Quickcheck also returns "xs = []" as a counterexample to the
   605     obvious theorem "hd xs = hd xs".
   606 
   607     After finding a potentially spurious counterexample, Quickcheck
   608     continues searching for genuine ones.
   609 
   610     By default, Quickcheck shows potentially spurious and genuine
   611     counterexamples. The option "genuine_only" sets quickcheck to only
   612     show genuine counterexamples.
   613 
   614   - The command 'quickcheck_generator' creates random and exhaustive
   615     value generators for a given type and operations.
   616 
   617     It generates values by using the operations as if they were
   618     constructors of that type.
   619 
   620   - Support for multisets.
   621 
   622   - Added "use_subtype" options.
   623 
   624   - Added "quickcheck_locale" configuration to specify how to process
   625     conjectures in a locale context.
   626 
   627 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   628 and affecting 'rat' and 'real'.
   629 
   630 * Sledgehammer:
   631   - Integrated more tightly with SPASS, as described in the ITP 2012
   632     paper "More SPASS with Isabelle".
   633   - Made it try "smt" as a fallback if "metis" fails or times out.
   634   - Added support for the following provers: Alt-Ergo (via Why3 and
   635     TFF1), iProver, iProver-Eq.
   636   - Sped up the minimizer.
   637   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   638   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   639   - Renamed "sound" option to "strict".
   640 
   641 * Metis: Added possibility to specify lambda translations scheme as a
   642 parenthesized argument (e.g., "by (metis (lifting) ...)").
   643 
   644 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   645 
   646 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   647 
   648 * New "case_product" attribute to generate a case rule doing multiple
   649 case distinctions at the same time.  E.g.
   650 
   651   list.exhaust [case_product nat.exhaust]
   652 
   653 produces a rule which can be used to perform case distinction on both
   654 a list and a nat.
   655 
   656 * New "eventually_elim" method as a generalized variant of the
   657 eventually_elim* rules.  Supports structured proofs.
   658 
   659 * Typedef with implicit set definition is considered legacy.  Use
   660 "typedef (open)" form instead, which will eventually become the
   661 default.
   662 
   663 * Record: code generation can be switched off manually with
   664 
   665   declare [[record_coden = false]]  -- "default true"
   666 
   667 * Datatype: type parameters allow explicit sort constraints.
   668 
   669 * Concrete syntax for case expressions includes constraints for source
   670 positions, and thus produces Prover IDE markup for its bindings.
   671 INCOMPATIBILITY for old-style syntax translations that augment the
   672 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   673 one_case.
   674 
   675 * Clarified attribute "mono_set": pure declaration without modifying
   676 the result of the fact expression.
   677 
   678 * More default pred/set conversions on a couple of relation operations
   679 and predicates.  Added powers of predicate relations.  Consolidation
   680 of some relation theorems:
   681 
   682   converse_def ~> converse_unfold
   683   rel_comp_def ~> relcomp_unfold
   684   symp_def ~> (modified, use symp_def and sym_def instead)
   685   transp_def ~> transp_trans
   686   Domain_def ~> Domain_unfold
   687   Range_def ~> Domain_converse [symmetric]
   688 
   689 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   690 
   691 See theory "Relation" for examples for making use of pred/set
   692 conversions by means of attributes "to_set" and "to_pred".
   693 
   694 INCOMPATIBILITY.
   695 
   696 * Renamed facts about the power operation on relations, i.e., relpow
   697 to match the constant's name:
   698 
   699   rel_pow_1 ~> relpow_1
   700   rel_pow_0_I ~> relpow_0_I
   701   rel_pow_Suc_I ~> relpow_Suc_I
   702   rel_pow_Suc_I2 ~> relpow_Suc_I2
   703   rel_pow_0_E ~> relpow_0_E
   704   rel_pow_Suc_E ~> relpow_Suc_E
   705   rel_pow_E ~> relpow_E
   706   rel_pow_Suc_D2 ~> relpow_Suc_D2
   707   rel_pow_Suc_E2 ~> relpow_Suc_E2
   708   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   709   rel_pow_E2 ~> relpow_E2
   710   rel_pow_add ~> relpow_add
   711   rel_pow_commute ~> relpow
   712   rel_pow_empty ~> relpow_empty:
   713   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   714   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   715   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   716   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   717   rel_pow_fun_conv ~> relpow_fun_conv
   718   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   719   rel_pow_finite_bounded ~> relpow_finite_bounded
   720   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   721   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   722   single_valued_rel_pow ~> single_valued_relpow
   723 
   724 INCOMPATIBILITY.
   725 
   726 * Theory Relation: Consolidated constant name for relation composition
   727 and corresponding theorem names:
   728 
   729   - Renamed constant rel_comp to relcomp.
   730 
   731   - Dropped abbreviation pred_comp. Use relcompp instead.
   732 
   733   - Renamed theorems:
   734 
   735     rel_compI ~> relcompI
   736     rel_compEpair ~> relcompEpair
   737     rel_compE ~> relcompE
   738     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   739     rel_comp_empty1 ~> relcomp_empty1
   740     rel_comp_mono ~> relcomp_mono
   741     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   742     rel_comp_distrib ~> relcomp_distrib
   743     rel_comp_distrib2 ~> relcomp_distrib2
   744     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   745     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   746     single_valued_rel_comp ~> single_valued_relcomp
   747     rel_comp_def ~> relcomp_unfold
   748     converse_rel_comp ~> converse_relcomp
   749     pred_compI ~> relcomppI
   750     pred_compE ~> relcomppE
   751     pred_comp_bot1 ~> relcompp_bot1
   752     pred_comp_bot2 ~> relcompp_bot2
   753     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   754     pred_comp_mono ~> relcompp_mono
   755     pred_comp_distrib ~> relcompp_distrib
   756     pred_comp_distrib2 ~> relcompp_distrib2
   757     converse_pred_comp ~> converse_relcompp
   758 
   759     finite_rel_comp ~> finite_relcomp
   760 
   761     set_rel_comp ~> set_relcomp
   762 
   763 INCOMPATIBILITY.
   764 
   765 * Theory Divides: Discontinued redundant theorems about div and mod.
   766 INCOMPATIBILITY, use the corresponding generic theorems instead.
   767 
   768   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   769   zdiv_self ~> div_self
   770   zmod_self ~> mod_self
   771   zdiv_zero ~> div_0
   772   zmod_zero ~> mod_0
   773   zdiv_zmod_equality ~> div_mod_equality2
   774   zdiv_zmod_equality2 ~> div_mod_equality
   775   zmod_zdiv_trivial ~> mod_div_trivial
   776   zdiv_zminus_zminus ~> div_minus_minus
   777   zmod_zminus_zminus ~> mod_minus_minus
   778   zdiv_zminus2 ~> div_minus_right
   779   zmod_zminus2 ~> mod_minus_right
   780   zdiv_minus1_right ~> div_minus1_right
   781   zmod_minus1_right ~> mod_minus1_right
   782   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   783   zmod_zmult1_eq ~> mod_mult_right_eq
   784   zpower_zmod ~> power_mod
   785   zdvd_zmod ~> dvd_mod
   786   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   787   mod_mult_distrib ~> mult_mod_left
   788   mod_mult_distrib2 ~> mult_mod_right
   789 
   790 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   791 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   792 
   793 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   794 
   795 * Consolidated theorem names concerning fold combinators:
   796 
   797   inf_INFI_fold_inf ~> inf_INF_fold_inf
   798   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   799   INFI_fold_inf ~> INF_fold_inf
   800   SUPR_fold_sup ~> SUP_fold_sup
   801   union_set ~> union_set_fold
   802   minus_set ~> minus_set_fold
   803   INFI_set_fold ~> INF_set_fold
   804   SUPR_set_fold ~> SUP_set_fold
   805   INF_code ~> INF_set_foldr
   806   SUP_code ~> SUP_set_foldr
   807   foldr.simps ~> foldr.simps (in point-free formulation)
   808   foldr_fold_rev ~> foldr_conv_fold
   809   foldl_fold ~> foldl_conv_fold
   810   foldr_foldr ~> foldr_conv_foldl
   811   foldl_foldr ~> foldl_conv_foldr
   812   fold_set_remdups ~> fold_set_fold_remdups
   813   fold_set ~> fold_set_fold
   814   fold1_set ~> fold1_set_fold
   815 
   816 INCOMPATIBILITY.
   817 
   818 * Dropped rarely useful theorems concerning fold combinators:
   819 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   820 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   821 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   822 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   823 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   824 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   825 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   826 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   827 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   828 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   829 
   830 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   831 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   832 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   833 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   834 lemmas over fold rather than foldr, or make use of lemmas
   835 fold_conv_foldr and fold_rev.
   836 
   837 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   838 through option types.
   839 
   840 * "Transitive_Closure.ntrancl": bounded transitive closure on
   841 relations.
   842 
   843 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   844 
   845 * Theory Int: Discontinued many legacy theorems specific to type int.
   846 INCOMPATIBILITY, use the corresponding generic theorems instead.
   847 
   848   zminus_zminus ~> minus_minus
   849   zminus_0 ~> minus_zero
   850   zminus_zadd_distrib ~> minus_add_distrib
   851   zadd_commute ~> add_commute
   852   zadd_assoc ~> add_assoc
   853   zadd_left_commute ~> add_left_commute
   854   zadd_ac ~> add_ac
   855   zmult_ac ~> mult_ac
   856   zadd_0 ~> add_0_left
   857   zadd_0_right ~> add_0_right
   858   zadd_zminus_inverse2 ~> left_minus
   859   zmult_zminus ~> mult_minus_left
   860   zmult_commute ~> mult_commute
   861   zmult_assoc ~> mult_assoc
   862   zadd_zmult_distrib ~> left_distrib
   863   zadd_zmult_distrib2 ~> right_distrib
   864   zdiff_zmult_distrib ~> left_diff_distrib
   865   zdiff_zmult_distrib2 ~> right_diff_distrib
   866   zmult_1 ~> mult_1_left
   867   zmult_1_right ~> mult_1_right
   868   zle_refl ~> order_refl
   869   zle_trans ~> order_trans
   870   zle_antisym ~> order_antisym
   871   zle_linear ~> linorder_linear
   872   zless_linear ~> linorder_less_linear
   873   zadd_left_mono ~> add_left_mono
   874   zadd_strict_right_mono ~> add_strict_right_mono
   875   zadd_zless_mono ~> add_less_le_mono
   876   int_0_less_1 ~> zero_less_one
   877   int_0_neq_1 ~> zero_neq_one
   878   zless_le ~> less_le
   879   zpower_zadd_distrib ~> power_add
   880   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   881   zero_le_zpower_abs ~> zero_le_power_abs
   882 
   883 * Theory Deriv: Renamed
   884 
   885   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   886 
   887 * Theory Library/Multiset: Improved code generation of multisets.
   888 
   889 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   890 are expressed via type classes again. The special syntax
   891 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   892 setsum_set, which is now subsumed by Big_Operators.setsum.
   893 INCOMPATIBILITY.
   894 
   895 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   896 use theory HOL/Library/Nat_Bijection instead.
   897 
   898 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   899 trees is now inside a type class context.  Names of affected
   900 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   901 theories working directly with raw red-black trees, adapt the names as
   902 follows:
   903 
   904   Operations:
   905   bulkload -> rbt_bulkload
   906   del_from_left -> rbt_del_from_left
   907   del_from_right -> rbt_del_from_right
   908   del -> rbt_del
   909   delete -> rbt_delete
   910   ins -> rbt_ins
   911   insert -> rbt_insert
   912   insertw -> rbt_insert_with
   913   insert_with_key -> rbt_insert_with_key
   914   map_entry -> rbt_map_entry
   915   lookup -> rbt_lookup
   916   sorted -> rbt_sorted
   917   tree_greater -> rbt_greater
   918   tree_less -> rbt_less
   919   tree_less_symbol -> rbt_less_symbol
   920   union -> rbt_union
   921   union_with -> rbt_union_with
   922   union_with_key -> rbt_union_with_key
   923 
   924   Lemmas:
   925   balance_left_sorted -> balance_left_rbt_sorted
   926   balance_left_tree_greater -> balance_left_rbt_greater
   927   balance_left_tree_less -> balance_left_rbt_less
   928   balance_right_sorted -> balance_right_rbt_sorted
   929   balance_right_tree_greater -> balance_right_rbt_greater
   930   balance_right_tree_less -> balance_right_rbt_less
   931   balance_sorted -> balance_rbt_sorted
   932   balance_tree_greater -> balance_rbt_greater
   933   balance_tree_less -> balance_rbt_less
   934   bulkload_is_rbt -> rbt_bulkload_is_rbt
   935   combine_sorted -> combine_rbt_sorted
   936   combine_tree_greater -> combine_rbt_greater
   937   combine_tree_less -> combine_rbt_less
   938   delete_in_tree -> rbt_delete_in_tree
   939   delete_is_rbt -> rbt_delete_is_rbt
   940   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   941   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   942   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   943   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   944   del_in_tree -> rbt_del_in_tree
   945   del_inv1_inv2 -> rbt_del_inv1_inv2
   946   del_sorted -> rbt_del_rbt_sorted
   947   del_tree_greater -> rbt_del_rbt_greater
   948   del_tree_less -> rbt_del_rbt_less
   949   dom_lookup_Branch -> dom_rbt_lookup_Branch
   950   entries_lookup -> entries_rbt_lookup
   951   finite_dom_lookup -> finite_dom_rbt_lookup
   952   insert_sorted -> rbt_insert_rbt_sorted
   953   insertw_is_rbt -> rbt_insertw_is_rbt
   954   insertwk_is_rbt -> rbt_insertwk_is_rbt
   955   insertwk_sorted -> rbt_insertwk_rbt_sorted
   956   insertw_sorted -> rbt_insertw_rbt_sorted
   957   ins_sorted -> ins_rbt_sorted
   958   ins_tree_greater -> ins_rbt_greater
   959   ins_tree_less -> ins_rbt_less
   960   is_rbt_sorted -> is_rbt_rbt_sorted
   961   lookup_balance -> rbt_lookup_balance
   962   lookup_bulkload -> rbt_lookup_rbt_bulkload
   963   lookup_delete -> rbt_lookup_rbt_delete
   964   lookup_Empty -> rbt_lookup_Empty
   965   lookup_from_in_tree -> rbt_lookup_from_in_tree
   966   lookup_in_tree -> rbt_lookup_in_tree
   967   lookup_ins -> rbt_lookup_ins
   968   lookup_insert -> rbt_lookup_rbt_insert
   969   lookup_insertw -> rbt_lookup_rbt_insertw
   970   lookup_insertwk -> rbt_lookup_rbt_insertwk
   971   lookup_keys -> rbt_lookup_keys
   972   lookup_map -> rbt_lookup_map
   973   lookup_map_entry -> rbt_lookup_rbt_map_entry
   974   lookup_tree_greater -> rbt_lookup_rbt_greater
   975   lookup_tree_less -> rbt_lookup_rbt_less
   976   lookup_union -> rbt_lookup_rbt_union
   977   map_entry_color_of -> rbt_map_entry_color_of
   978   map_entry_inv1 -> rbt_map_entry_inv1
   979   map_entry_inv2 -> rbt_map_entry_inv2
   980   map_entry_is_rbt -> rbt_map_entry_is_rbt
   981   map_entry_sorted -> rbt_map_entry_rbt_sorted
   982   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   983   map_entry_tree_less -> rbt_map_entry_rbt_less
   984   map_tree_greater -> map_rbt_greater
   985   map_tree_less -> map_rbt_less
   986   map_sorted -> map_rbt_sorted
   987   paint_sorted -> paint_rbt_sorted
   988   paint_lookup -> paint_rbt_lookup
   989   paint_tree_greater -> paint_rbt_greater
   990   paint_tree_less -> paint_rbt_less
   991   sorted_entries -> rbt_sorted_entries
   992   tree_greater_eq_trans -> rbt_greater_eq_trans
   993   tree_greater_nit -> rbt_greater_nit
   994   tree_greater_prop -> rbt_greater_prop
   995   tree_greater_simps -> rbt_greater_simps
   996   tree_greater_trans -> rbt_greater_trans
   997   tree_less_eq_trans -> rbt_less_eq_trans
   998   tree_less_nit -> rbt_less_nit
   999   tree_less_prop -> rbt_less_prop
  1000   tree_less_simps -> rbt_less_simps
  1001   tree_less_trans -> rbt_less_trans
  1002   tree_ord_props -> rbt_ord_props
  1003   union_Branch -> rbt_union_Branch
  1004   union_is_rbt -> rbt_union_is_rbt
  1005   unionw_is_rbt -> rbt_unionw_is_rbt
  1006   unionwk_is_rbt -> rbt_unionwk_is_rbt
  1007   unionwk_sorted -> rbt_unionwk_rbt_sorted
  1008 
  1009 * Theory HOL/Library/Float: Floating point numbers are now defined as
  1010 a subset of the real numbers.  All operations are defined using the
  1011 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
  1012 
  1013   Changed Operations:
  1014   float_abs -> abs
  1015   float_nprt -> nprt
  1016   float_pprt -> pprt
  1017   pow2 -> use powr
  1018   round_down -> float_round_down
  1019   round_up -> float_round_up
  1020   scale -> exponent
  1021 
  1022   Removed Operations:
  1023   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
  1024 
  1025   Renamed Lemmas:
  1026   abs_float_def -> Float.compute_float_abs
  1027   bitlen_ge0 -> bitlen_nonneg
  1028   bitlen.simps -> Float.compute_bitlen
  1029   float_components -> Float_mantissa_exponent
  1030   float_divl.simps -> Float.compute_float_divl
  1031   float_divr.simps -> Float.compute_float_divr
  1032   float_eq_odd -> mult_powr_eq_mult_powr_iff
  1033   float_power -> real_of_float_power
  1034   lapprox_posrat_def -> Float.compute_lapprox_posrat
  1035   lapprox_rat.simps -> Float.compute_lapprox_rat
  1036   le_float_def' -> Float.compute_float_le
  1037   le_float_def -> less_eq_float.rep_eq
  1038   less_float_def' -> Float.compute_float_less
  1039   less_float_def -> less_float.rep_eq
  1040   normfloat_def -> Float.compute_normfloat
  1041   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
  1042   normfloat -> normfloat_def
  1043   normfloat_unique -> use normfloat_def
  1044   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
  1045   one_float_def -> Float.compute_float_one
  1046   plus_float_def -> Float.compute_float_plus
  1047   rapprox_posrat_def -> Float.compute_rapprox_posrat
  1048   rapprox_rat.simps -> Float.compute_rapprox_rat
  1049   real_of_float_0 -> zero_float.rep_eq
  1050   real_of_float_1 -> one_float.rep_eq
  1051   real_of_float_abs -> abs_float.rep_eq
  1052   real_of_float_add -> plus_float.rep_eq
  1053   real_of_float_minus -> uminus_float.rep_eq
  1054   real_of_float_mult -> times_float.rep_eq
  1055   real_of_float_simp -> Float.rep_eq
  1056   real_of_float_sub -> minus_float.rep_eq
  1057   round_down.simps -> Float.compute_float_round_down
  1058   round_up.simps -> Float.compute_float_round_up
  1059   times_float_def -> Float.compute_float_times
  1060   uminus_float_def -> Float.compute_float_uminus
  1061   zero_float_def -> Float.compute_float_zero
  1062 
  1063   Lemmas not necessary anymore, use the transfer method:
  1064   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
  1065   float_divr, float_le_simp, float_less1_mantissa_bound,
  1066   float_less_simp, float_less_zero, float_le_zero,
  1067   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
  1068   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
  1069   lapprox_rat_bottom, normalized_float, rapprox_posrat,
  1070   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
  1071   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
  1072   round_up, zero_le_float, zero_less_float
  1073 
  1074 * New theory HOL/Library/DAList provides an abstract type for
  1075 association lists with distinct keys.
  1076 
  1077 * Session HOL/IMP: Added new theory of abstract interpretation of
  1078 annotated commands.
  1079 
  1080 * Session HOL-Import: Re-implementation from scratch is faster,
  1081 simpler, and more scalable.  Requires a proof bundle, which is
  1082 available as an external component.  Discontinued old (and mostly
  1083 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
  1084 
  1085 * Session HOL-Word: Discontinued many redundant theorems specific to
  1086 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
  1087 instead.
  1088 
  1089   word_sub_alt ~> word_sub_wi
  1090   word_add_alt ~> word_add_def
  1091   word_mult_alt ~> word_mult_def
  1092   word_minus_alt ~> word_minus_def
  1093   word_0_alt ~> word_0_wi
  1094   word_1_alt ~> word_1_wi
  1095   word_add_0 ~> add_0_left
  1096   word_add_0_right ~> add_0_right
  1097   word_mult_1 ~> mult_1_left
  1098   word_mult_1_right ~> mult_1_right
  1099   word_add_commute ~> add_commute
  1100   word_add_assoc ~> add_assoc
  1101   word_add_left_commute ~> add_left_commute
  1102   word_mult_commute ~> mult_commute
  1103   word_mult_assoc ~> mult_assoc
  1104   word_mult_left_commute ~> mult_left_commute
  1105   word_left_distrib ~> left_distrib
  1106   word_right_distrib ~> right_distrib
  1107   word_left_minus ~> left_minus
  1108   word_diff_0_right ~> diff_0_right
  1109   word_diff_self ~> diff_self
  1110   word_sub_def ~> diff_minus
  1111   word_diff_minus ~> diff_minus
  1112   word_add_ac ~> add_ac
  1113   word_mult_ac ~> mult_ac
  1114   word_plus_ac0 ~> add_0_left add_0_right add_ac
  1115   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
  1116   word_order_trans ~> order_trans
  1117   word_order_refl ~> order_refl
  1118   word_order_antisym ~> order_antisym
  1119   word_order_linear ~> linorder_linear
  1120   lenw1_zero_neq_one ~> zero_neq_one
  1121   word_number_of_eq ~> number_of_eq
  1122   word_of_int_add_hom ~> wi_hom_add
  1123   word_of_int_sub_hom ~> wi_hom_sub
  1124   word_of_int_mult_hom ~> wi_hom_mult
  1125   word_of_int_minus_hom ~> wi_hom_neg
  1126   word_of_int_succ_hom ~> wi_hom_succ
  1127   word_of_int_pred_hom ~> wi_hom_pred
  1128   word_of_int_0_hom ~> word_0_wi
  1129   word_of_int_1_hom ~> word_1_wi
  1130 
  1131 * Session HOL-Word: New proof method "word_bitwise" for splitting
  1132 machine word equalities and inequalities into logical circuits,
  1133 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
  1134 multiplication, shifting by constants, bitwise operators and numeric
  1135 constants.  Requires fixed-length word types, not 'a word.  Solves
  1136 many standard word identities outright and converts more into first
  1137 order problems amenable to blast or similar.  See also examples in
  1138 HOL/Word/Examples/WordExamples.thy.
  1139 
  1140 * Session HOL-Probability: Introduced the type "'a measure" to
  1141 represent measures, this replaces the records 'a algebra and 'a
  1142 measure_space.  The locales based on subset_class now have two
  1143 locale-parameters the space \<Omega> and the set of measurable sets M.
  1144 The product of probability spaces uses now the same constant as the
  1145 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
  1146 measure".  Most constants are defined now outside of locales and gain
  1147 an additional parameter, like null_sets, almost_eventually or \<mu>'.
  1148 Measure space constructions for distributions and densities now got
  1149 their own constants distr and density.  Instead of using locales to
  1150 describe measure spaces with a finite space, the measure count_space
  1151 and point_measure is introduced.  INCOMPATIBILITY.
  1152 
  1153   Renamed constants:
  1154   measure -> emeasure
  1155   finite_measure.\<mu>' -> measure
  1156   product_algebra_generator -> prod_algebra
  1157   product_prob_space.emb -> prod_emb
  1158   product_prob_space.infprod_algebra -> PiM
  1159 
  1160   Removed locales:
  1161   completeable_measure_space
  1162   finite_measure_space
  1163   finite_prob_space
  1164   finite_product_finite_prob_space
  1165   finite_product_sigma_algebra
  1166   finite_sigma_algebra
  1167   measure_space
  1168   pair_finite_prob_space
  1169   pair_finite_sigma_algebra
  1170   pair_finite_space
  1171   pair_sigma_algebra
  1172   product_sigma_algebra
  1173 
  1174   Removed constants:
  1175   conditional_space
  1176   distribution -> use distr measure, or distributed predicate
  1177   image_space
  1178   joint_distribution -> use distr measure, or distributed predicate
  1179   pair_measure_generator
  1180   product_prob_space.infprod_algebra -> use PiM
  1181   subvimage
  1182 
  1183   Replacement theorems:
  1184   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1185   finite_measure.empty_measure -> measure_empty
  1186   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1187   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1188   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1189   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1190   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1191   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1192   finite_measure.positive_measure' -> measure_nonneg
  1193   finite_measure.real_measure -> finite_measure.emeasure_real
  1194   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1195   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1196   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1197   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1198   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1199   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1200   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1201   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1202   information_space.entropy_commute -> information_space.entropy_commute_simple
  1203   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1204   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1205   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1206   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1207   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1208   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1209   information_space.KL_ge_0 -> information_space.KL_nonneg
  1210   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1211   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1212   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1213   Int_stable_product_algebra_generator -> positive_integral
  1214   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1215   measure_space.additive -> emeasure_additive
  1216   measure_space.AE_iff_null_set -> AE_iff_null
  1217   measure_space.almost_everywhere_def -> eventually_ae_filter
  1218   measure_space.almost_everywhere_vimage -> AE_distrD
  1219   measure_space.continuity_from_above -> INF_emeasure_decseq
  1220   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1221   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1222   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1223   measure_space_density -> emeasure_density
  1224   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1225   measure_space.integrable_vimage -> integrable_distr
  1226   measure_space.integral_translated_density -> integral_density
  1227   measure_space.integral_vimage -> integral_distr
  1228   measure_space.measure_additive -> plus_emeasure
  1229   measure_space.measure_compl -> emeasure_compl
  1230   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1231   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1232   measure_space.measure_decseq -> decseq_emeasure
  1233   measure_space.measure_Diff -> emeasure_Diff
  1234   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1235   measure_space.measure_eq_0 -> emeasure_eq_0
  1236   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1237   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1238   measure_space.measure_incseq -> incseq_emeasure
  1239   measure_space.measure_insert -> emeasure_insert
  1240   measure_space.measure_mono -> emeasure_mono
  1241   measure_space.measure_not_negative -> emeasure_not_MInf
  1242   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1243   measure_space.measure_setsum -> setsum_emeasure
  1244   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1245   measure_space.measure_space_vimage -> emeasure_distr
  1246   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1247   measure_space.measure_subadditive -> subadditive
  1248   measure_space.measure_top -> emeasure_space
  1249   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1250   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1251   measure_space.positive_integral_translated_density -> positive_integral_density
  1252   measure_space.positive_integral_vimage -> positive_integral_distr
  1253   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1254   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1255   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1256   measure_space.real_measure_Diff -> measure_Diff
  1257   measure_space.real_measure_finite_Union -> measure_finite_Union
  1258   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1259   measure_space.real_measure_subadditive -> measure_subadditive
  1260   measure_space.real_measure_Union -> measure_Union
  1261   measure_space.real_measure_UNION -> measure_UNION
  1262   measure_space.simple_function_vimage -> simple_function_comp
  1263   measure_space.simple_integral_vimage -> simple_integral_distr
  1264   measure_space.simple_integral_vimage -> simple_integral_distr
  1265   measure_unique_Int_stable -> measure_eqI_generator_eq
  1266   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1267   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1268   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1269   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1270   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1271   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1272   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1273   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1274   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1275   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1276   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1277   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1278   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1279   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1280   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1281   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1282   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1283   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1284   prob_space.prob_space_vimage -> prob_space_distr
  1285   prob_space.random_variable_restrict -> measurable_restrict
  1286   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1287   product_algebraE -> prod_algebraE_all
  1288   product_algebra_generator_der -> prod_algebra_eq_finite
  1289   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1290   product_algebraI -> sets_PiM_I_finite
  1291   product_measure_exists -> product_sigma_finite.sigma_finite
  1292   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1293   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1294   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1295   product_prob_space.measurable_component -> measurable_component_singleton
  1296   product_prob_space.measurable_emb -> measurable_prod_emb
  1297   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1298   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1299   product_prob_space.measure_emb -> emeasure_prod_emb
  1300   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1301   product_sigma_algebra.product_algebra_into_space -> space_closed
  1302   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1303   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1304   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1305   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1306   sets_product_algebra -> sets_PiM
  1307   sigma_algebra.measurable_sigma -> measurable_measure_of
  1308   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1309   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1310   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1311   space_product_algebra -> space_PiM
  1312 
  1313 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1314 languages) into Isabelle/HOL.
  1315 
  1316 
  1317 *** FOL ***
  1318 
  1319 * New "case_product" attribute (see HOL).
  1320 
  1321 
  1322 *** ZF ***
  1323 
  1324 * Greater support for structured proofs involving induction or case
  1325 analysis.
  1326 
  1327 * Much greater use of mathematical symbols.
  1328 
  1329 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1330 
  1331 
  1332 *** ML ***
  1333 
  1334 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1335 from a minor keyword introduced via theory header declaration.
  1336 
  1337 * Antiquotation @{command_spec "name"} produces the
  1338 Outer_Syntax.command_spec from a major keyword introduced via theory
  1339 header declaration; it can be passed to Outer_Syntax.command etc.
  1340 
  1341 * Local_Theory.define no longer hard-wires default theorem name
  1342 "foo_def", but retains the binding as given.  If that is Binding.empty
  1343 / Attrib.empty_binding, the result is not registered as user-level
  1344 fact.  The Local_Theory.define_internal variant allows to specify a
  1345 non-empty name (used for the foundation in the background theory),
  1346 while omitting the fact binding in the user-context.  Potential
  1347 INCOMPATIBILITY for derived definitional packages: need to specify
  1348 naming policy for primitive definitions more explicitly.
  1349 
  1350 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1351 conformance with similar operations in structure Term and Logic.
  1352 
  1353 * Antiquotation @{attributes [...]} embeds attribute source
  1354 representation into the ML text, which is particularly useful with
  1355 declarations like Local_Theory.note.
  1356 
  1357 * Structure Proof_Context follows standard naming scheme.  Old
  1358 ProofContext has been discontinued.  INCOMPATIBILITY.
  1359 
  1360 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1361 change of semantics: update is applied to auxiliary local theory
  1362 context as well.
  1363 
  1364 * Modernized some old-style infix operations:
  1365 
  1366   addeqcongs    ~> Simplifier.add_eqcong
  1367   deleqcongs    ~> Simplifier.del_eqcong
  1368   addcongs      ~> Simplifier.add_cong
  1369   delcongs      ~> Simplifier.del_cong
  1370   setmksimps    ~> Simplifier.set_mksimps
  1371   setmkcong     ~> Simplifier.set_mkcong
  1372   setmksym      ~> Simplifier.set_mksym
  1373   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1374   settermless   ~> Simplifier.set_termless
  1375   setsubgoaler  ~> Simplifier.set_subgoaler
  1376   addsplits     ~> Splitter.add_split
  1377   delsplits     ~> Splitter.del_split
  1378 
  1379 
  1380 *** System ***
  1381 
  1382 * USER_HOME settings variable points to cross-platform user home
  1383 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1384 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1385 of former $HOME.  A different default for USER_HOME may be set
  1386 explicitly in shell environment, before Isabelle settings are
  1387 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1388 the generic user home was intended.
  1389 
  1390 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1391 notation, which is useful for the jEdit file browser, for example.
  1392 
  1393 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1394 (not just JRE).
  1395 
  1396 
  1397 
  1398 New in Isabelle2011-1 (October 2011)
  1399 ------------------------------------
  1400 
  1401 *** General ***
  1402 
  1403 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1404 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1405 
  1406   - Management of multiple theory files directly from the editor
  1407     buffer store -- bypassing the file-system (no requirement to save
  1408     files for checking).
  1409 
  1410   - Markup of formal entities within the text buffer, with semantic
  1411     highlighting, tooltips and hyperlinks to jump to defining source
  1412     positions.
  1413 
  1414   - Improved text rendering, with sub/superscripts in the source
  1415     buffer (including support for copy/paste wrt. output panel, HTML
  1416     theory output and other non-Isabelle text boxes).
  1417 
  1418   - Refined scheduling of proof checking and printing of results,
  1419     based on interactive editor view.  (Note: jEdit folding and
  1420     narrowing allows to restrict buffer perspectives explicitly.)
  1421 
  1422   - Reduced CPU performance requirements, usable on machines with few
  1423     cores.
  1424 
  1425   - Reduced memory requirements due to pruning of unused document
  1426     versions (garbage collection).
  1427 
  1428 See also ~~/src/Tools/jEdit/README.html for further information,
  1429 including some remaining limitations.
  1430 
  1431 * Theory loader: source files are exclusively located via the master
  1432 directory of each theory node (where the .thy file itself resides).
  1433 The global load path (such as src/HOL/Library) has been discontinued.
  1434 Note that the path element ~~ may be used to reference theories in the
  1435 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1436 INCOMPATIBILITY.
  1437 
  1438 * Theory loader: source files are identified by content via SHA1
  1439 digests.  Discontinued former path/modtime identification and optional
  1440 ISABELLE_FILE_IDENT plugin scripts.
  1441 
  1442 * Parallelization of nested Isar proofs is subject to
  1443 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1444 usedir option -Q.
  1445 
  1446 * Name space: former unsynchronized references are now proper
  1447 configuration options, with more conventional names:
  1448 
  1449   long_names   ~> names_long
  1450   short_names  ~> names_short
  1451   unique_names ~> names_unique
  1452 
  1453 Minor INCOMPATIBILITY, need to declare options in context like this:
  1454 
  1455   declare [[names_unique = false]]
  1456 
  1457 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1458 that the result needs to be unique, which means fact specifications
  1459 may have to be refined after enriching a proof context.
  1460 
  1461 * Attribute "case_names" has been refined: the assumptions in each case
  1462 can be named now by following the case name with [name1 name2 ...].
  1463 
  1464 * Isabelle/Isar reference manual has been updated and extended:
  1465   - "Synopsis" provides a catalog of main Isar language concepts.
  1466   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1467   - Updated material from classic "ref" manual, notably about
  1468     "Classical Reasoner".
  1469 
  1470 
  1471 *** HOL ***
  1472 
  1473 * Class bot and top require underlying partial order rather than
  1474 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1475 
  1476 * Class complete_lattice: generalized a couple of lemmas from sets;
  1477 generalized theorems INF_cong and SUP_cong.  New type classes for
  1478 complete boolean algebras and complete linear orders.  Lemmas
  1479 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1480 class complete_linorder.
  1481 
  1482 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1483 Sup_fun_def, Inf_apply, Sup_apply.
  1484 
  1485 Removed redundant lemmas (the right hand side gives hints how to
  1486 replace them for (metis ...), or (simp only: ...) proofs):
  1487 
  1488   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1489   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1490   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1491   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1492   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1493   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1494   Inter_def ~> INF_def, image_def
  1495   Union_def ~> SUP_def, image_def
  1496   INT_eq ~> INF_def, and image_def
  1497   UN_eq ~> SUP_def, and image_def
  1498   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1499 
  1500 More consistent and comprehensive names:
  1501 
  1502   INTER_eq_Inter_image ~> INF_def
  1503   UNION_eq_Union_image ~> SUP_def
  1504   INFI_def ~> INF_def
  1505   SUPR_def ~> SUP_def
  1506   INF_leI ~> INF_lower
  1507   INF_leI2 ~> INF_lower2
  1508   le_INFI ~> INF_greatest
  1509   le_SUPI ~> SUP_upper
  1510   le_SUPI2 ~> SUP_upper2
  1511   SUP_leI ~> SUP_least
  1512   INFI_bool_eq ~> INF_bool_eq
  1513   SUPR_bool_eq ~> SUP_bool_eq
  1514   INFI_apply ~> INF_apply
  1515   SUPR_apply ~> SUP_apply
  1516   INTER_def ~> INTER_eq
  1517   UNION_def ~> UNION_eq
  1518 
  1519 INCOMPATIBILITY.
  1520 
  1521 * Renamed theory Complete_Lattice to Complete_Lattices.
  1522 INCOMPATIBILITY.
  1523 
  1524 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1525 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1526 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1527 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1528 
  1529 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1530 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1531 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1532 INCOMPATIBILITY.
  1533 
  1534 * Added syntactic classes "inf" and "sup" for the respective
  1535 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1536 (mostly internal) locale predicates for some derived classes.
  1537 
  1538 * Theorem collections ball_simps and bex_simps do not contain theorems
  1539 referring to UNION any longer; these have been moved to collection
  1540 UN_ball_bex_simps.  INCOMPATIBILITY.
  1541 
  1542 * Theory Archimedean_Field: floor now is defined as parameter of a
  1543 separate type class floor_ceiling.
  1544 
  1545 * Theory Finite_Set: more coherent development of fold_set locales:
  1546 
  1547     locale fun_left_comm ~> locale comp_fun_commute
  1548     locale fun_left_comm_idem ~> locale comp_fun_idem
  1549 
  1550 Both use point-free characterization; interpretation proofs may need
  1551 adjustment.  INCOMPATIBILITY.
  1552 
  1553 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1554 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1555 
  1556 * Theory Complex_Main: The locale interpretations for the
  1557 bounded_linear and bounded_bilinear locales have been removed, in
  1558 order to reduce the number of duplicate lemmas. Users must use the
  1559 original names for distributivity theorems, potential INCOMPATIBILITY.
  1560 
  1561   divide.add ~> add_divide_distrib
  1562   divide.diff ~> diff_divide_distrib
  1563   divide.setsum ~> setsum_divide_distrib
  1564   mult.add_right ~> right_distrib
  1565   mult.diff_right ~> right_diff_distrib
  1566   mult_right.setsum ~> setsum_right_distrib
  1567   mult_left.diff ~> left_diff_distrib
  1568 
  1569 * Theory Complex_Main: Several redundant theorems have been removed or
  1570 replaced by more general versions. INCOMPATIBILITY.
  1571 
  1572   real_diff_def ~> minus_real_def
  1573   real_divide_def ~> divide_real_def
  1574   real_less_def ~> less_le
  1575   real_abs_def ~> abs_real_def
  1576   real_sgn_def ~> sgn_real_def
  1577   real_mult_commute ~> mult_commute
  1578   real_mult_assoc ~> mult_assoc
  1579   real_mult_1 ~> mult_1_left
  1580   real_add_mult_distrib ~> left_distrib
  1581   real_zero_not_eq_one ~> zero_neq_one
  1582   real_mult_inverse_left ~> left_inverse
  1583   INVERSE_ZERO ~> inverse_zero
  1584   real_le_refl ~> order_refl
  1585   real_le_antisym ~> order_antisym
  1586   real_le_trans ~> order_trans
  1587   real_le_linear ~> linear
  1588   real_le_eq_diff ~> le_iff_diff_le_0
  1589   real_add_left_mono ~> add_left_mono
  1590   real_mult_order ~> mult_pos_pos
  1591   real_mult_less_mono2 ~> mult_strict_left_mono
  1592   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1593   real_0_le_divide_iff ~> zero_le_divide_iff
  1594   realpow_two_disj ~> power2_eq_iff
  1595   real_squared_diff_one_factored ~> square_diff_one_factored
  1596   realpow_two_diff ~> square_diff_square_factored
  1597   reals_complete2 ~> complete_real
  1598   real_sum_squared_expand ~> power2_sum
  1599   exp_ln_eq ~> ln_unique
  1600   expi_add ~> exp_add
  1601   expi_zero ~> exp_zero
  1602   lemma_DERIV_subst ~> DERIV_cong
  1603   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1604   LIMSEQ_const ~> tendsto_const
  1605   LIMSEQ_norm ~> tendsto_norm
  1606   LIMSEQ_add ~> tendsto_add
  1607   LIMSEQ_minus ~> tendsto_minus
  1608   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1609   LIMSEQ_diff ~> tendsto_diff
  1610   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1611   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1612   LIMSEQ_mult ~> tendsto_mult
  1613   LIMSEQ_inverse ~> tendsto_inverse
  1614   LIMSEQ_divide ~> tendsto_divide
  1615   LIMSEQ_pow ~> tendsto_power
  1616   LIMSEQ_setsum ~> tendsto_setsum
  1617   LIMSEQ_setprod ~> tendsto_setprod
  1618   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1619   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1620   LIMSEQ_imp_rabs ~> tendsto_rabs
  1621   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1622   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1623   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1624   LIMSEQ_Complex ~> tendsto_Complex
  1625   LIM_ident ~> tendsto_ident_at
  1626   LIM_const ~> tendsto_const
  1627   LIM_add ~> tendsto_add
  1628   LIM_add_zero ~> tendsto_add_zero
  1629   LIM_minus ~> tendsto_minus
  1630   LIM_diff ~> tendsto_diff
  1631   LIM_norm ~> tendsto_norm
  1632   LIM_norm_zero ~> tendsto_norm_zero
  1633   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1634   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1635   LIM_rabs ~> tendsto_rabs
  1636   LIM_rabs_zero ~> tendsto_rabs_zero
  1637   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1638   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1639   LIM_compose ~> tendsto_compose
  1640   LIM_mult ~> tendsto_mult
  1641   LIM_scaleR ~> tendsto_scaleR
  1642   LIM_of_real ~> tendsto_of_real
  1643   LIM_power ~> tendsto_power
  1644   LIM_inverse ~> tendsto_inverse
  1645   LIM_sgn ~> tendsto_sgn
  1646   isCont_LIM_compose ~> isCont_tendsto_compose
  1647   bounded_linear.LIM ~> bounded_linear.tendsto
  1648   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1649   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1650   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1651   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1652   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1653   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1654 
  1655 * Theory Complex_Main: The definition of infinite series was
  1656 generalized.  Now it is defined on the type class {topological_space,
  1657 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1658 
  1659 * Theory Complex_Main: The complex exponential function "expi" is now
  1660 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1661 several polymorphic lemmas about "exp" are now applicable to "expi".
  1662 
  1663 * Code generation:
  1664 
  1665   - Theory Library/Code_Char_ord provides native ordering of
  1666     characters in the target language.
  1667 
  1668   - Commands code_module and code_library are legacy, use export_code
  1669     instead.
  1670 
  1671   - Method "evaluation" is legacy, use method "eval" instead.
  1672 
  1673   - Legacy evaluator "SML" is deactivated by default.  May be
  1674     reactivated by the following theory command:
  1675 
  1676       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1677 
  1678 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1679 
  1680 * New proof method "induction" that gives induction hypotheses the
  1681 name "IH", thus distinguishing them from further hypotheses that come
  1682 from rule induction.  The latter are still called "hyps".  Method
  1683 "induction" is a thin wrapper around "induct" and follows the same
  1684 syntax.
  1685 
  1686 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1687 still available as a legacy feature for some time.
  1688 
  1689 * Nitpick:
  1690   - Added "need" and "total_consts" options.
  1691   - Reintroduced "show_skolems" option by popular demand.
  1692   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1693     INCOMPATIBILITY.
  1694 
  1695 * Sledgehammer:
  1696   - Use quasi-sound (and efficient) translations by default.
  1697   - Added support for the following provers: E-ToFoF, LEO-II,
  1698     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1699   - Automatically preplay and minimize proofs before showing them if
  1700     this can be done within reasonable time.
  1701   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1702     INCOMPATIBILITY.
  1703   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1704     "max_mono_iters", and "max_new_mono_instances" options.
  1705   - Removed "explicit_apply" and "full_types" options as well as "Full
  1706     Types" Proof General menu item. INCOMPATIBILITY.
  1707 
  1708 * Metis:
  1709   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1710   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1711     INCOMPATIBILITY.
  1712 
  1713 * Command 'try':
  1714   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1715     "elim:" options. INCOMPATIBILITY.
  1716   - Introduced 'try' that not only runs 'try_methods' but also
  1717     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1718 
  1719 * Quickcheck:
  1720   - Added "eval" option to evaluate terms for the found counterexample
  1721     (currently only supported by the default (exhaustive) tester).
  1722   - Added post-processing of terms to obtain readable counterexamples
  1723     (currently only supported by the default (exhaustive) tester).
  1724   - New counterexample generator quickcheck[narrowing] enables
  1725     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1726     with its installation location defined in the Isabelle settings
  1727     environment as ISABELLE_GHC.
  1728   - Removed quickcheck tester "SML" based on the SML code generator
  1729     (formly in HOL/Library).
  1730 
  1731 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1732 use 'partial_function' instead.
  1733 
  1734 * Theory Library/Extended_Reals replaces now the positive extended
  1735 reals found in probability theory. This file is extended by
  1736 Multivariate_Analysis/Extended_Real_Limits.
  1737 
  1738 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1739 from where it must be imported explicitly if it is really required.
  1740 INCOMPATIBILITY.
  1741 
  1742 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1743 been moved here.  INCOMPATIBILITY.
  1744 
  1745 * Theory Library/Saturated provides type of numbers with saturated
  1746 arithmetic.
  1747 
  1748 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1749 product type 'a * 'b, and provides instance proofs for various order
  1750 and lattice type classes.
  1751 
  1752 * Theory Library/Countable now provides the "countable_datatype" proof
  1753 method for proving "countable" class instances for datatypes.
  1754 
  1755 * Theory Library/Cset_Monad allows do notation for computable sets
  1756 (cset) via the generic monad ad-hoc overloading facility.
  1757 
  1758 * Library: Theories of common data structures are split into theories
  1759 for implementation, an invariant-ensuring type, and connection to an
  1760 abstract type. INCOMPATIBILITY.
  1761 
  1762   - RBT is split into RBT and RBT_Mapping.
  1763   - AssocList is split and renamed into AList and AList_Mapping.
  1764   - DList is split into DList_Impl, DList, and DList_Cset.
  1765   - Cset is split into Cset and List_Cset.
  1766 
  1767 * Theory Library/Nat_Infinity has been renamed to
  1768 Library/Extended_Nat, with name changes of the following types and
  1769 constants:
  1770 
  1771   type inat   ~> type enat
  1772   Fin         ~> enat
  1773   Infty       ~> infinity (overloaded)
  1774   iSuc        ~> eSuc
  1775   the_Fin     ~> the_enat
  1776 
  1777 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1778 been renamed accordingly. INCOMPATIBILITY.
  1779 
  1780 * Session Multivariate_Analysis: The euclidean_space type class now
  1781 fixes a constant "Basis :: 'a set" consisting of the standard
  1782 orthonormal basis for the type. Users now have the option of
  1783 quantifying over this set instead of using the "basis" function, e.g.
  1784 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1785 
  1786 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1787 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1788 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1789 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1790 changed to match. Definition theorems for overloaded constants now use
  1791 the standard "foo_vec_def" naming scheme. A few other theorems have
  1792 been renamed as follows (INCOMPATIBILITY):
  1793 
  1794   Cart_eq          ~> vec_eq_iff
  1795   dist_nth_le_cart ~> dist_vec_nth_le
  1796   tendsto_vector   ~> vec_tendstoI
  1797   Cauchy_vector    ~> vec_CauchyI
  1798 
  1799 * Session Multivariate_Analysis: Several duplicate theorems have been
  1800 removed, and other theorems have been renamed or replaced with more
  1801 general versions. INCOMPATIBILITY.
  1802 
  1803   finite_choice ~> finite_set_choice
  1804   eventually_conjI ~> eventually_conj
  1805   eventually_and ~> eventually_conj_iff
  1806   eventually_false ~> eventually_False
  1807   setsum_norm ~> norm_setsum
  1808   Lim_sequentially ~> LIMSEQ_def
  1809   Lim_ident_at ~> LIM_ident
  1810   Lim_const ~> tendsto_const
  1811   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1812   Lim_neg ~> tendsto_minus
  1813   Lim_add ~> tendsto_add
  1814   Lim_sub ~> tendsto_diff
  1815   Lim_mul ~> tendsto_scaleR
  1816   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1817   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1818   Lim_linear ~> bounded_linear.tendsto
  1819   Lim_component ~> tendsto_euclidean_component
  1820   Lim_component_cart ~> tendsto_vec_nth
  1821   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1822   dot_lsum ~> inner_setsum_left
  1823   dot_rsum ~> inner_setsum_right
  1824   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1825   continuous_neg ~> continuous_minus
  1826   continuous_sub ~> continuous_diff
  1827   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1828   continuous_mul ~> continuous_scaleR
  1829   continuous_inv ~> continuous_inverse
  1830   continuous_at_within_inv ~> continuous_at_within_inverse
  1831   continuous_at_inv ~> continuous_at_inverse
  1832   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1833   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1834   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1835   continuous_on_neg ~> continuous_on_minus
  1836   continuous_on_sub ~> continuous_on_diff
  1837   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1838   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1839   continuous_on_mul ~> continuous_on_scaleR
  1840   continuous_on_mul_real ~> continuous_on_mult
  1841   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1842   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1843   continuous_on_inverse ~> continuous_on_inv
  1844   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1845   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1846   subset_interior ~> interior_mono
  1847   subset_closure ~> closure_mono
  1848   closure_univ ~> closure_UNIV
  1849   real_arch_lt ~> reals_Archimedean2
  1850   real_arch ~> reals_Archimedean3
  1851   real_abs_norm ~> abs_norm_cancel
  1852   real_abs_sub_norm ~> norm_triangle_ineq3
  1853   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1854 
  1855 * Session HOL-Probability:
  1856   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1857   - Infinite products of probability measures are now available.
  1858   - Sigma closure is independent, if the generator is independent
  1859   - Use extended reals instead of positive extended
  1860     reals. INCOMPATIBILITY.
  1861 
  1862 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1863 
  1864   expand_fun_below ~> fun_below_iff
  1865   below_fun_ext ~> fun_belowI
  1866   expand_cfun_eq ~> cfun_eq_iff
  1867   ext_cfun ~> cfun_eqI
  1868   expand_cfun_below ~> cfun_below_iff
  1869   below_cfun_ext ~> cfun_belowI
  1870   monofun_fun_fun ~> fun_belowD
  1871   monofun_fun_arg ~> monofunE
  1872   monofun_lub_fun ~> adm_monofun [THEN admD]
  1873   cont_lub_fun ~> adm_cont [THEN admD]
  1874   cont2cont_Rep_CFun ~> cont2cont_APP
  1875   cont_Rep_CFun_app ~> cont_APP_app
  1876   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1877   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1878   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1879   contlub_cfun ~> lub_APP [symmetric]
  1880   contlub_LAM ~> lub_LAM [symmetric]
  1881   thelubI ~> lub_eqI
  1882   UU_I ~> bottomI
  1883   lift_distinct1 ~> lift.distinct(1)
  1884   lift_distinct2 ~> lift.distinct(2)
  1885   Def_not_UU ~> lift.distinct(2)
  1886   Def_inject ~> lift.inject
  1887   below_UU_iff ~> below_bottom_iff
  1888   eq_UU_iff ~> eq_bottom_iff
  1889 
  1890 
  1891 *** Document preparation ***
  1892 
  1893 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1894 isar-ref manual, both for description and actual application of the
  1895 same.
  1896 
  1897 * Antiquotation @{value} evaluates the given term and presents its
  1898 result.
  1899 
  1900 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1901 variables x1, y23 into subscripted form x\<^isub>1,
  1902 y\<^isub>2\<^isub>3.
  1903 
  1904 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1905 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1906 
  1907 * Localized \isabellestyle switch can be used within blocks or groups
  1908 like this:
  1909 
  1910   \isabellestyle{it}  %preferred default
  1911   {\isabellestylett @{text "typewriter stuff"}}
  1912 
  1913 * Discontinued special treatment of hard tabulators.  Implicit
  1914 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1915 layouts.
  1916 
  1917 
  1918 *** ML ***
  1919 
  1920 * The inner syntax of sort/type/term/prop supports inlined YXML
  1921 representations within quoted string tokens.  By encoding logical
  1922 entities via Term_XML (in ML or Scala) concrete syntax can be
  1923 bypassed, which is particularly useful for producing bits of text
  1924 under external program control.
  1925 
  1926 * Antiquotations for ML and document preparation are managed as theory
  1927 data, which requires explicit setup.
  1928 
  1929 * Isabelle_Process.is_active allows tools to check if the official
  1930 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1931 (better known as Proof General).
  1932 
  1933 * Structure Proof_Context follows standard naming scheme.  Old
  1934 ProofContext is still available for some time as legacy alias.
  1935 
  1936 * Structure Timing provides various operations for timing; supersedes
  1937 former start_timing/end_timing etc.
  1938 
  1939 * Path.print is the official way to show file-system paths to users
  1940 (including quotes etc.).
  1941 
  1942 * Inner syntax: identifiers in parse trees of generic categories
  1943 "logic", "aprop", "idt" etc. carry position information (disguised as
  1944 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1945 translations that choke on unexpected type constraints.  Positions can
  1946 be stripped in ML translations via Syntax.strip_positions /
  1947 Syntax.strip_positions_ast, or via the syntax constant
  1948 "_strip_positions" within parse trees.  As last resort, positions can
  1949 be disabled via the configuration option Syntax.positions, which is
  1950 called "syntax_positions" in Isar attribute syntax.
  1951 
  1952 * Discontinued special status of various ML structures that contribute
  1953 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1954 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1955 refer directly to Ast.Constant, Lexicon.is_identifier,
  1956 Syntax_Trans.mk_binder_tr etc.
  1957 
  1958 * Typed print translation: discontinued show_sorts argument, which is
  1959 already available via context of "advanced" translation.
  1960 
  1961 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1962 goal states; body tactic needs to address all subgoals uniformly.
  1963 
  1964 * Slightly more special eq_list/eq_set, with shortcut involving
  1965 pointer equality (assumes that eq relation is reflexive).
  1966 
  1967 * Classical tactics use proper Proof.context instead of historic types
  1968 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1969 operate directly on Proof.context.  Raw type claset retains its use as
  1970 snapshot of the classical context, which can be recovered via
  1971 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1972 INCOMPATIBILITY, classical tactics and derived proof methods require
  1973 proper Proof.context.
  1974 
  1975 
  1976 *** System ***
  1977 
  1978 * Discontinued support for Poly/ML 5.2, which was the last version
  1979 without proper multithreading and TimeLimit implementation.
  1980 
  1981 * Discontinued old lib/scripts/polyml-platform, which has been
  1982 obsolete since Isabelle2009-2.
  1983 
  1984 * Various optional external tools are referenced more robustly and
  1985 uniformly by explicit Isabelle settings as follows:
  1986 
  1987   ISABELLE_CSDP   (formerly CSDP_EXE)
  1988   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1989   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1990   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1991   ISABELLE_YAP    (formerly EXEC_YAP)
  1992 
  1993 Note that automated detection from the file-system or search path has
  1994 been discontinued.  INCOMPATIBILITY.
  1995 
  1996 * Scala layer provides JVM method invocation service for static
  1997 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1998 example:
  1999 
  2000   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  2001 
  2002 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  2003 this allows to pass structured values between ML and Scala.
  2004 
  2005 * The IsabelleText fonts includes some further glyphs to support the
  2006 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  2007 installed a local copy (which is normally *not* required) need to
  2008 delete or update it from ~~/lib/fonts/.
  2009 
  2010 
  2011 
  2012 New in Isabelle2011 (January 2011)
  2013 ----------------------------------
  2014 
  2015 *** General ***
  2016 
  2017 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  2018 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  2019 useful tooltips and hyperlinks produced from its static analysis.  The
  2020 bundled component provides an executable Isabelle tool that can be run
  2021 like this:
  2022 
  2023   Isabelle2011/bin/isabelle jedit
  2024 
  2025 * Significantly improved Isabelle/Isar implementation manual.
  2026 
  2027 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  2028 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  2029 while the default heap location within that directory lacks that extra
  2030 suffix.  This isolates multiple Isabelle installations from each
  2031 other, avoiding problems with old settings in new versions.
  2032 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  2033 
  2034 * Source files are always encoded as UTF-8, instead of old-fashioned
  2035 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  2036 the following package declarations:
  2037 
  2038   \usepackage[utf8]{inputenc}
  2039   \usepackage{textcomp}
  2040 
  2041 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  2042 a Unicode character is treated as a single symbol, not a sequence of
  2043 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  2044 contain symbols without further backslash escapes, Unicode can now be
  2045 used here as well.  Recall that Symbol.explode in ML provides a
  2046 consistent view on symbols, while raw explode (or String.explode)
  2047 merely give a byte-oriented representation.
  2048 
  2049 * Theory loader: source files are primarily located via the master
  2050 directory of each theory node (where the .thy file itself resides).
  2051 The global load path is still partially available as legacy feature.
  2052 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  2053 explicit paths, relatively to the theory.
  2054 
  2055 * Special treatment of ML file names has been discontinued.
  2056 Historically, optional extensions .ML or .sml were added on demand --
  2057 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  2058 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  2059 
  2060 * Various options that affect pretty printing etc. are now properly
  2061 handled within the context via configuration options, instead of
  2062 unsynchronized references or print modes.  There are both ML Config.T
  2063 entities and Isar declaration attributes to access these.
  2064 
  2065   ML (Config.T)                 Isar (attribute)
  2066 
  2067   eta_contract                  eta_contract
  2068   show_brackets                 show_brackets
  2069   show_sorts                    show_sorts
  2070   show_types                    show_types
  2071   show_question_marks           show_question_marks
  2072   show_consts                   show_consts
  2073   show_abbrevs                  show_abbrevs
  2074 
  2075   Syntax.ast_trace              syntax_ast_trace
  2076   Syntax.ast_stat               syntax_ast_stat
  2077   Syntax.ambiguity_level        syntax_ambiguity_level
  2078 
  2079   Goal_Display.goals_limit      goals_limit
  2080   Goal_Display.show_main_goal   show_main_goal
  2081 
  2082   Method.rule_trace             rule_trace
  2083 
  2084   Thy_Output.display            thy_output_display
  2085   Thy_Output.quotes             thy_output_quotes
  2086   Thy_Output.indent             thy_output_indent
  2087   Thy_Output.source             thy_output_source
  2088   Thy_Output.break              thy_output_break
  2089 
  2090 Note that corresponding "..._default" references in ML may only be
  2091 changed globally at the ROOT session setup, but *not* within a theory.
  2092 The option "show_abbrevs" supersedes the former print mode
  2093 "no_abbrevs" with inverted meaning.
  2094 
  2095 * More systematic naming of some configuration options.
  2096 INCOMPATIBILITY.
  2097 
  2098   trace_simp  ~>  simp_trace
  2099   debug_simp  ~>  simp_debug
  2100 
  2101 * Support for real valued configuration options, using simplistic
  2102 floating-point notation that coincides with the inner syntax for
  2103 float_token.
  2104 
  2105 * Support for real valued preferences (with approximative PGIP type):
  2106 front-ends need to accept "pgint" values in float notation.
  2107 INCOMPATIBILITY.
  2108 
  2109 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  2110 DejaVu Sans.
  2111 
  2112 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  2113 
  2114 
  2115 *** Pure ***
  2116 
  2117 * Command 'type_synonym' (with single argument) replaces somewhat
  2118 outdated 'types', which is still available as legacy feature for some
  2119 time.
  2120 
  2121 * Command 'nonterminal' (with 'and' separated list of arguments)
  2122 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  2123 
  2124 * Command 'notepad' replaces former 'example_proof' for
  2125 experimentation in Isar without any result.  INCOMPATIBILITY.
  2126 
  2127 * Locale interpretation commands 'interpret' and 'sublocale' accept
  2128 lists of equations to map definitions in a locale to appropriate
  2129 entities in the context of the interpretation.  The 'interpretation'
  2130 command already provided this functionality.
  2131 
  2132 * Diagnostic command 'print_dependencies' prints the locale instances
  2133 that would be activated if the specified expression was interpreted in
  2134 the current context.  Variant "print_dependencies!" assumes a context
  2135 without interpretations.
  2136 
  2137 * Diagnostic command 'print_interps' prints interpretations in proofs
  2138 in addition to interpretations in theories.
  2139 
  2140 * Discontinued obsolete 'global' and 'local' commands to manipulate
  2141 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  2142 Sign.root_path and Sign.local_path may be applied directly where this
  2143 feature is still required for historical reasons.
  2144 
  2145 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  2146 'definition' instead.
  2147 
  2148 * The "prems" fact, which refers to the accidental collection of
  2149 foundational premises in the context, is now explicitly marked as
  2150 legacy feature and will be discontinued soon.  Consider using "assms"
  2151 of the head statement or reference facts by explicit names.
  2152 
  2153 * Document antiquotations @{class} and @{type} print classes and type
  2154 constructors.
  2155 
  2156 * Document antiquotation @{file} checks file/directory entries within
  2157 the local file system.
  2158 
  2159 
  2160 *** HOL ***
  2161 
  2162 * Coercive subtyping: functions can be declared as coercions and type
  2163 inference will add them as necessary upon input of a term.  Theory
  2164 Complex_Main declares real :: nat => real and real :: int => real as
  2165 coercions. A coercion function f is declared like this:
  2166 
  2167   declare [[coercion f]]
  2168 
  2169 To lift coercions through type constructors (e.g. from nat => real to
  2170 nat list => real list), map functions can be declared, e.g.
  2171 
  2172   declare [[coercion_map map]]
  2173 
  2174 Currently coercion inference is activated only in theories including
  2175 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  2176 the configuration option "coercion_enabled", e.g. it can be enabled in
  2177 other theories like this:
  2178 
  2179   declare [[coercion_enabled]]
  2180 
  2181 * Command 'partial_function' provides basic support for recursive
  2182 function definitions over complete partial orders.  Concrete instances
  2183 are provided for i) the option type, ii) tail recursion on arbitrary
  2184 types, and iii) the heap monad of Imperative_HOL.  See
  2185 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2186 for examples.
  2187 
  2188 * Function package: f.psimps rules are no longer implicitly declared
  2189 as [simp].  INCOMPATIBILITY.
  2190 
  2191 * Datatype package: theorems generated for executable equality (class
  2192 "eq") carry proper names and are treated as default code equations.
  2193 
  2194 * Inductive package: now offers command 'inductive_simps' to
  2195 automatically derive instantiated and simplified equations for
  2196 inductive predicates, similar to 'inductive_cases'.
  2197 
  2198 * Command 'enriched_type' allows to register properties of the
  2199 functorial structure of types.
  2200 
  2201 * Improved infrastructure for term evaluation using code generator
  2202 techniques, in particular static evaluation conversions.
  2203 
  2204 * Code generator: Scala (2.8 or higher) has been added to the target
  2205 languages.
  2206 
  2207 * Code generator: globbing constant expressions "*" and "Theory.*"
  2208 have been replaced by the more idiomatic "_" and "Theory._".
  2209 INCOMPATIBILITY.
  2210 
  2211 * Code generator: export_code without explicit file declaration prints
  2212 to standard output.  INCOMPATIBILITY.
  2213 
  2214 * Code generator: do not print function definitions for case
  2215 combinators any longer.
  2216 
  2217 * Code generator: simplification with rules determined with
  2218 src/Tools/Code/code_simp.ML and method "code_simp".
  2219 
  2220 * Code generator for records: more idiomatic representation of record
  2221 types.  Warning: records are not covered by ancient SML code
  2222 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2223 rep_datatype declaration helps to succeed then:
  2224 
  2225   record 'a foo = ...
  2226   ...
  2227   rep_datatype foo_ext ...
  2228 
  2229 * Records: logical foundation type for records does not carry a
  2230 '_type' suffix any longer (obsolete due to authentic syntax).
  2231 INCOMPATIBILITY.
  2232 
  2233 * Quickcheck now by default uses exhaustive testing instead of random
  2234 testing.  Random testing can be invoked by "quickcheck [random]",
  2235 exhaustive testing by "quickcheck [exhaustive]".
  2236 
  2237 * Quickcheck instantiates polymorphic types with small finite
  2238 datatypes by default. This enables a simple execution mechanism to
  2239 handle quantifiers and function equality over the finite datatypes.
  2240 
  2241 * Quickcheck random generator has been renamed from "code" to
  2242 "random".  INCOMPATIBILITY.
  2243 
  2244 * Quickcheck now has a configurable time limit which is set to 30
  2245 seconds by default. This can be changed by adding [timeout = n] to the
  2246 quickcheck command. The time limit for Auto Quickcheck is still set
  2247 independently.
  2248 
  2249 * Quickcheck in locales considers interpretations of that locale for
  2250 counter example search.
  2251 
  2252 * Sledgehammer:
  2253   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2254     method. See the Sledgehammer manual for details ("isabelle doc
  2255     sledgehammer").
  2256   - Renamed commands:
  2257     sledgehammer atp_info ~> sledgehammer running_provers
  2258     sledgehammer atp_kill ~> sledgehammer kill_provers
  2259     sledgehammer available_atps ~> sledgehammer available_provers
  2260     INCOMPATIBILITY.
  2261   - Renamed options:
  2262     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2263     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2264     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2265     (and "ms" and "min" are no longer supported)
  2266     INCOMPATIBILITY.
  2267 
  2268 * Nitpick:
  2269   - Renamed options:
  2270     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2271     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2272     INCOMPATIBILITY.
  2273   - Added support for partial quotient types.
  2274   - Added local versions of the "Nitpick.register_xxx" functions.
  2275   - Added "whack" option.
  2276   - Allow registration of quotient types as codatatypes.
  2277   - Improved "merge_type_vars" option to merge more types.
  2278   - Removed unsound "fast_descrs" option.
  2279   - Added custom symmetry breaking for datatypes, making it possible to reach
  2280     higher cardinalities.
  2281   - Prevent the expansion of too large definitions.
  2282 
  2283 * Proof methods "metis" and "meson" now have configuration options
  2284 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2285 to diagnose these tools. E.g.
  2286 
  2287     using [[metis_trace = true]]
  2288 
  2289 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2290 manually as command 'solve_direct'.
  2291 
  2292 * The default SMT solver Z3 must be enabled explicitly (due to
  2293 licensing issues) by setting the environment variable
  2294 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2295 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2296 changing the SMT solver is done via the configuration option
  2297 "smt_solver".
  2298 
  2299 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2300 i.e. "remote_cvc3" and "remote_z3".
  2301 
  2302 * Added basic SMT support for datatypes, records, and typedefs using
  2303 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2304 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2305 support for a similar behavior).  Minor INCOMPATIBILITY.
  2306 
  2307 * Changed SMT configuration options:
  2308   - Renamed:
  2309     z3_proofs ~> smt_oracle (with inverted meaning)
  2310     z3_trace_assms ~> smt_trace_used_facts
  2311     INCOMPATIBILITY.
  2312   - Added:
  2313     smt_verbose
  2314     smt_random_seed
  2315     smt_datatypes
  2316     smt_infer_triggers
  2317     smt_monomorph_limit
  2318     cvc3_options
  2319     remote_cvc3_options
  2320     remote_z3_options
  2321     yices_options
  2322 
  2323 * Boogie output files (.b2i files) need to be declared in the theory
  2324 header.
  2325 
  2326 * Simplification procedure "list_to_set_comprehension" rewrites list
  2327 comprehensions applied to List.set to set comprehensions.  Occasional
  2328 INCOMPATIBILITY, may be deactivated like this:
  2329 
  2330   declare [[simproc del: list_to_set_comprehension]]
  2331 
  2332 * Removed old version of primrec package.  INCOMPATIBILITY.
  2333 
  2334 * Removed simplifier congruence rule of "prod_case", as has for long
  2335 been the case with "split".  INCOMPATIBILITY.
  2336 
  2337 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2338 
  2339 * Removed [split_format ... and ... and ...] version of
  2340 [split_format].  Potential INCOMPATIBILITY.
  2341 
  2342 * Predicate "sorted" now defined inductively, with nice induction
  2343 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2344 
  2345 * Constant "contents" renamed to "the_elem", to free the generic name
  2346 contents for other uses.  INCOMPATIBILITY.
  2347 
  2348 * Renamed class eq and constant eq (for code generation) to class
  2349 equal and constant equal, plus renaming of related facts and various
  2350 tuning.  INCOMPATIBILITY.
  2351 
  2352 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2353 
  2354 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2355 INCOMPATIBILITY.
  2356 
  2357 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2358 avoid confusion with finite sets.  INCOMPATIBILITY.
  2359 
  2360 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2361 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2362 for congruent(2)).
  2363 
  2364 * Some previously unqualified names have been qualified:
  2365 
  2366   types
  2367     bool ~> HOL.bool
  2368     nat ~> Nat.nat
  2369 
  2370   constants
  2371     Trueprop ~> HOL.Trueprop
  2372     True ~> HOL.True
  2373     False ~> HOL.False
  2374     op & ~> HOL.conj
  2375     op | ~> HOL.disj
  2376     op --> ~> HOL.implies
  2377     op = ~> HOL.eq
  2378     Not ~> HOL.Not
  2379     The ~> HOL.The
  2380     All ~> HOL.All
  2381     Ex ~> HOL.Ex
  2382     Ex1 ~> HOL.Ex1
  2383     Let ~> HOL.Let
  2384     If ~> HOL.If
  2385     Ball ~> Set.Ball
  2386     Bex ~> Set.Bex
  2387     Suc ~> Nat.Suc
  2388     Pair ~> Product_Type.Pair
  2389     fst ~> Product_Type.fst
  2390     snd ~> Product_Type.snd
  2391     curry ~> Product_Type.curry
  2392     op : ~> Set.member
  2393     Collect ~> Set.Collect
  2394 
  2395 INCOMPATIBILITY.
  2396 
  2397 * More canonical naming convention for some fundamental definitions:
  2398 
  2399     bot_bool_eq ~> bot_bool_def
  2400     top_bool_eq ~> top_bool_def
  2401     inf_bool_eq ~> inf_bool_def
  2402     sup_bool_eq ~> sup_bool_def
  2403     bot_fun_eq  ~> bot_fun_def
  2404     top_fun_eq  ~> top_fun_def
  2405     inf_fun_eq  ~> inf_fun_def
  2406     sup_fun_eq  ~> sup_fun_def
  2407 
  2408 INCOMPATIBILITY.
  2409 
  2410 * More stylized fact names:
  2411 
  2412   expand_fun_eq ~> fun_eq_iff
  2413   expand_set_eq ~> set_eq_iff
  2414   set_ext       ~> set_eqI
  2415   nat_number    ~> eval_nat_numeral
  2416 
  2417 INCOMPATIBILITY.
  2418 
  2419 * Refactoring of code-generation specific operations in theory List:
  2420 
  2421   constants
  2422     null ~> List.null
  2423 
  2424   facts
  2425     mem_iff ~> member_def
  2426     null_empty ~> null_def
  2427 
  2428 INCOMPATIBILITY.  Note that these were not supposed to be used
  2429 regularly unless for striking reasons; their main purpose was code
  2430 generation.
  2431 
  2432 Various operations from the Haskell prelude are used for generating
  2433 Haskell code.
  2434 
  2435 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2436 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2437 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2438 
  2439 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2440 "*" and "+" respectively.  INCOMPATIBILITY.
  2441 
  2442 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2443 "Sum_Type.Plus" instead.
  2444 
  2445 * Constant "split" has been merged with constant "prod_case"; names of
  2446 ML functions, facts etc. involving split have been retained so far,
  2447 though.  INCOMPATIBILITY.
  2448 
  2449 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2450 instead.  INCOMPATIBILITY.
  2451 
  2452 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2453 INCOMPATIBILITY.
  2454 
  2455 * Former theory Library/Enum is now part of the HOL-Main image.
  2456 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2457 referred to by its qualified name.
  2458 
  2459   enum    ~>  Enum.enum
  2460   nlists  ~>  Enum.nlists
  2461   product ~>  Enum.product
  2462 
  2463 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2464 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2465 INCOMPATIBILITY.
  2466 
  2467 * Theory Library/SetsAndFunctions has been split into
  2468 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2469 for instance definitions for functions; various improvements.
  2470 INCOMPATIBILITY.
  2471 
  2472 * Theory Library/Multiset provides stable quicksort implementation of
  2473 sort_key.
  2474 
  2475 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2476 INCOMPATIBILITY.
  2477 
  2478 * Session Multivariate_Analysis: introduced a type class for euclidean
  2479 space.  Most theorems are now stated in terms of euclidean spaces
  2480 instead of finite cartesian products.
  2481 
  2482   types
  2483     real ^ 'n ~>  'a::real_vector
  2484               ~>  'a::euclidean_space
  2485               ~>  'a::ordered_euclidean_space
  2486         (depends on your needs)
  2487 
  2488   constants
  2489      _ $ _        ~> _ $$ _
  2490      \<chi> x. _  ~> \<chi>\<chi> x. _
  2491      CARD('n)     ~> DIM('a)
  2492 
  2493 Also note that the indices are now natural numbers and not from some
  2494 finite type. Finite cartesian products of euclidean spaces, products
  2495 of euclidean spaces the real and complex numbers are instantiated to
  2496 be euclidean_spaces.  INCOMPATIBILITY.
  2497 
  2498 * Session Probability: introduced pextreal as positive extended real
  2499 numbers.  Use pextreal as value for measures.  Introduce the
  2500 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2501 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2502 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2503 
  2504 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2505 INCOMPATIBILITY.
  2506 
  2507 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2508 prove verification conditions generated by the SPARK Ada program
  2509 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2510 
  2511 
  2512 *** HOL-Algebra ***
  2513 
  2514 * Theorems for additive ring operations (locale abelian_monoid and
  2515 descendants) are generated by interpretation from their multiplicative
  2516 counterparts.  Names (in particular theorem names) have the mandatory
  2517 qualifier 'add'.  Previous theorem names are redeclared for
  2518 compatibility.
  2519 
  2520 * Structure "int_ring" is now an abbreviation (previously a
  2521 definition).  This fits more natural with advanced interpretations.
  2522 
  2523 
  2524 *** HOLCF ***
  2525 
  2526 * The domain package now runs in definitional mode by default: The
  2527 former command 'new_domain' is now called 'domain'.  To use the domain
  2528 package in its original axiomatic mode, use 'domain (unsafe)'.
  2529 INCOMPATIBILITY.
  2530 
  2531 * The new class "domain" is now the default sort.  Class "predomain"
  2532 is an unpointed version of "domain". Theories can be updated by
  2533 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2534 
  2535   'a::type ~> 'a::countable
  2536   'a::cpo  ~> 'a::predomain
  2537   'a::pcpo ~> 'a::domain
  2538 
  2539 * The old type class "rep" has been superseded by class "domain".
  2540 Accordingly, users of the definitional package must remove any
  2541 "default_sort rep" declarations.  INCOMPATIBILITY.
  2542 
  2543 * The domain package (definitional mode) now supports unpointed
  2544 predomain argument types, as long as they are marked 'lazy'. (Strict
  2545 arguments must be in class "domain".) For example, the following
  2546 domain definition now works:
  2547 
  2548   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2549 
  2550 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2551 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2552 'a option, and 'a list.  Additionally, it configures fixrec and the
  2553 domain package to work with these types.  For example:
  2554 
  2555   fixrec isInl :: "('a + 'b) u -> tr"
  2556     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2557 
  2558   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2559 
  2560 * The "(permissive)" option of fixrec has been replaced with a
  2561 per-equation "(unchecked)" option. See
  2562 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2563 
  2564 * The "bifinite" class no longer fixes a constant "approx"; the class
  2565 now just asserts that such a function exists.  INCOMPATIBILITY.
  2566 
  2567 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2568 defines an embedding of type 'a defl into udom by default; instances
  2569 of "bifinite" and "domain" classes are available in
  2570 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2571 
  2572 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2573 
  2574 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2575 
  2576 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2577 
  2578 * The function "cprod_map" has been renamed to "prod_map".
  2579 INCOMPATIBILITY.
  2580 
  2581 * The monadic bind operator on each powerdomain has new binder syntax
  2582 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2583 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2584 
  2585 * The infix syntax for binary union on each powerdomain has changed
  2586 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2587 syntax.  INCOMPATIBILITY.
  2588 
  2589 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2590 still supported as an input translation.
  2591 
  2592 * Renamed some theorems (the original names are also still available).
  2593 
  2594   expand_fun_below   ~> fun_below_iff
  2595   below_fun_ext      ~> fun_belowI
  2596   expand_cfun_eq     ~> cfun_eq_iff
  2597   ext_cfun           ~> cfun_eqI
  2598   expand_cfun_below  ~> cfun_below_iff
  2599   below_cfun_ext     ~> cfun_belowI
  2600   cont2cont_Rep_CFun ~> cont2cont_APP
  2601 
  2602 * The Abs and Rep functions for various types have changed names.
  2603 Related theorem names have also changed to match. INCOMPATIBILITY.
  2604 
  2605   Rep_CFun  ~> Rep_cfun
  2606   Abs_CFun  ~> Abs_cfun
  2607   Rep_Sprod ~> Rep_sprod
  2608   Abs_Sprod ~> Abs_sprod
  2609   Rep_Ssum  ~> Rep_ssum
  2610   Abs_Ssum  ~> Abs_ssum
  2611 
  2612 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2613 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2614 
  2615 * Various changes to bisimulation/coinduction with domain package:
  2616 
  2617   - Definitions of "bisim" constants no longer mention definedness.
  2618   - With mutual recursion, "bisim" predicate is now curried.
  2619   - With mutual recursion, each type gets a separate coind theorem.
  2620   - Variable names in bisim_def and coinduct rules have changed.
  2621 
  2622 INCOMPATIBILITY.
  2623 
  2624 * Case combinators generated by the domain package for type "foo" are
  2625 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2626 
  2627 * Several theorems have been renamed to more accurately reflect the
  2628 names of constants and types involved.  INCOMPATIBILITY.
  2629 
  2630   thelub_const    ~> lub_const
  2631   lub_const       ~> is_lub_const
  2632   thelubI         ~> lub_eqI
  2633   is_lub_lub      ~> is_lubD2
  2634   lubI            ~> is_lub_lub
  2635   unique_lub      ~> is_lub_unique
  2636   is_ub_lub       ~> is_lub_rangeD1
  2637   lub_bin_chain   ~> is_lub_bin_chain
  2638   lub_fun         ~> is_lub_fun
  2639   thelub_fun      ~> lub_fun
  2640   thelub_cfun     ~> lub_cfun
  2641   thelub_Pair     ~> lub_Pair
  2642   lub_cprod       ~> is_lub_prod
  2643   thelub_cprod    ~> lub_prod
  2644   minimal_cprod   ~> minimal_prod
  2645   inst_cprod_pcpo ~> inst_prod_pcpo
  2646   UU_I            ~> bottomI
  2647   compact_UU      ~> compact_bottom
  2648   deflation_UU    ~> deflation_bottom
  2649   finite_deflation_UU ~> finite_deflation_bottom
  2650 
  2651 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2652 
  2653   sq_ord_less_eq_trans ~> below_eq_trans
  2654   sq_ord_eq_less_trans ~> eq_below_trans
  2655   refl_less            ~> below_refl
  2656   trans_less           ~> below_trans
  2657   antisym_less         ~> below_antisym
  2658   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2659   box_less             ~> box_below
  2660   rev_trans_less       ~> rev_below_trans
  2661   not_less2not_eq      ~> not_below2not_eq
  2662   less_UU_iff          ~> below_UU_iff
  2663   flat_less_iff        ~> flat_below_iff
  2664   adm_less             ~> adm_below
  2665   adm_not_less         ~> adm_not_below
  2666   adm_compact_not_less ~> adm_compact_not_below
  2667   less_fun_def         ~> below_fun_def
  2668   expand_fun_less      ~> fun_below_iff
  2669   less_fun_ext         ~> fun_belowI
  2670   less_discr_def       ~> below_discr_def
  2671   discr_less_eq        ~> discr_below_eq
  2672   less_unit_def        ~> below_unit_def
  2673   less_cprod_def       ~> below_prod_def
  2674   prod_lessI           ~> prod_belowI
  2675   Pair_less_iff        ~> Pair_below_iff
  2676   fst_less_iff         ~> fst_below_iff
  2677   snd_less_iff         ~> snd_below_iff
  2678   expand_cfun_less     ~> cfun_below_iff
  2679   less_cfun_ext        ~> cfun_belowI
  2680   injection_less       ~> injection_below
  2681   less_up_def          ~> below_up_def
  2682   not_Iup_less         ~> not_Iup_below
  2683   Iup_less             ~> Iup_below
  2684   up_less              ~> up_below
  2685   Def_inject_less_eq   ~> Def_below_Def
  2686   Def_less_is_eq       ~> Def_below_iff
  2687   spair_less_iff       ~> spair_below_iff
  2688   less_sprod           ~> below_sprod
  2689   spair_less           ~> spair_below
  2690   sfst_less_iff        ~> sfst_below_iff
  2691   ssnd_less_iff        ~> ssnd_below_iff
  2692   fix_least_less       ~> fix_least_below
  2693   dist_less_one        ~> dist_below_one
  2694   less_ONE             ~> below_ONE
  2695   ONE_less_iff         ~> ONE_below_iff
  2696   less_sinlD           ~> below_sinlD
  2697   less_sinrD           ~> below_sinrD
  2698 
  2699 
  2700 *** FOL and ZF ***
  2701 
  2702 * All constant names are now qualified internally and use proper
  2703 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2704 
  2705 
  2706 *** ML ***
  2707 
  2708 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2709 Fail if the argument is false.  Due to inlining the source position of
  2710 failed assertions is included in the error output.
  2711 
  2712 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2713 text is in practice always evaluated with a stable theory checkpoint.
  2714 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2715 
  2716 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2717 the current context, not any accidental theory loader state as before.
  2718 Potential INCOMPATIBILITY, subtle change in semantics.
  2719 
  2720 * Syntax.pretty_priority (default 0) configures the required priority
  2721 of pretty-printed output and thus affects insertion of parentheses.
  2722 
  2723 * Syntax.default_root (default "any") configures the inner syntax
  2724 category (nonterminal symbol) for parsing of terms.
  2725 
  2726 * Former exception Library.UnequalLengths now coincides with
  2727 ListPair.UnequalLengths.
  2728 
  2729 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2730 main functionality is provided by structure Simplifier.
  2731 
  2732 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2733 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2734 almost all situations.
  2735 
  2736 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2737 See implementation manual for further details on exceptions in
  2738 Isabelle/ML.
  2739 
  2740 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2741 meaning.
  2742 
  2743 * Renamed structure PureThy to Pure_Thy and moved most of its
  2744 operations to structure Global_Theory, to emphasize that this is
  2745 rarely-used global-only stuff.
  2746 
  2747 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2748 instead (or tracing for high-volume output).
  2749 
  2750 * Configuration option show_question_marks only affects regular pretty
  2751 printing of types and terms, not raw Term.string_of_vname.
  2752 
  2753 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2754 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2755 @{thms} for most purposes.
  2756 
  2757 * ML structure Unsynchronized is never opened, not even in Isar
  2758 interaction mode as before.  Old Unsynchronized.set etc. have been
  2759 discontinued -- use plain := instead.  This should be *rare* anyway,
  2760 since modern tools always work via official context data, notably
  2761 configuration options.
  2762 
  2763 * Parallel and asynchronous execution requires special care concerning
  2764 interrupts.  Structure Exn provides some convenience functions that
  2765 avoid working directly with raw Interrupt.  User code must not absorb
  2766 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2767 followed by re-raising of the original exception.  Another common
  2768 source of mistakes are "handle _" patterns, which make the meaning of
  2769 the program subject to physical effects of the environment.
  2770 
  2771 
  2772 
  2773 New in Isabelle2009-2 (June 2010)
  2774 ---------------------------------
  2775 
  2776 *** General ***
  2777 
  2778 * Authentic syntax for *all* logical entities (type classes, type
  2779 constructors, term constants): provides simple and robust
  2780 correspondence between formal entities and concrete syntax.  Within
  2781 the parse tree / AST representations, "constants" are decorated by
  2782 their category (class, type, const) and spelled out explicitly with
  2783 their full internal name.
  2784 
  2785 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2786 and translations (translation rules and translation functions in ML).
  2787 Some hints on upgrading:
  2788 
  2789   - Many existing uses of 'syntax' and 'translations' can be replaced
  2790     by more modern 'type_notation', 'notation' and 'abbreviation',
  2791     which are independent of this issue.
  2792 
  2793   - 'translations' require markup within the AST; the term syntax
  2794     provides the following special forms:
  2795 
  2796       CONST c   -- produces syntax version of constant c from context
  2797       XCONST c  -- literally c, checked as constant from context
  2798       c         -- literally c, if declared by 'syntax'
  2799 
  2800     Plain identifiers are treated as AST variables -- occasionally the
  2801     system indicates accidental variables via the error "rhs contains
  2802     extra variables".
  2803 
  2804     Type classes and type constructors are marked according to their
  2805     concrete syntax.  Some old translations rules need to be written
  2806     for the "type" category, using type constructor application
  2807     instead of pseudo-term application of the default category
  2808     "logic".
  2809 
  2810   - 'parse_translation' etc. in ML may use the following
  2811     antiquotations:
  2812 
  2813       @{class_syntax c}   -- type class c within parse tree / AST
  2814       @{term_syntax c}    -- type constructor c within parse tree / AST
  2815       @{const_syntax c}   -- ML version of "CONST c" above
  2816       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2817 
  2818   - Literal types within 'typed_print_translations', i.e. those *not*
  2819     represented as pseudo-terms are represented verbatim.  Use @{class
  2820     c} or @{type_name c} here instead of the above syntax
  2821     antiquotations.
  2822 
  2823 Note that old non-authentic syntax was based on unqualified base
  2824 names, so all of the above "constant" names would coincide.  Recall
  2825 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2826 diagnose syntax problems.
  2827 
  2828 * Type constructors admit general mixfix syntax, not just infix.
  2829 
  2830 * Concrete syntax may be attached to local entities without a proof
  2831 body, too.  This works via regular mixfix annotations for 'fix',
  2832 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2833 similar to the 'notation' command in theory specifications.
  2834 
  2835 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2836 need to specify constant name and syntax separately.  Internal ML
  2837 datatype constructors have been renamed from InfixName to Infix etc.
  2838 Minor INCOMPATIBILITY.
  2839 
  2840 * Schematic theorem statements need to be explicitly markup as such,
  2841 via commands 'schematic_lemma', 'schematic_theorem',
  2842 'schematic_corollary'.  Thus the relevance of the proof is made
  2843 syntactically clear, which impacts performance in a parallel or
  2844 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2845 
  2846 * Use of cumulative prems via "!" in some proof methods has been
  2847 discontinued (old legacy feature).
  2848 
  2849 * References 'trace_simp' and 'debug_simp' have been replaced by
  2850 configuration options stored in the context. Enabling tracing (the
  2851 case of debugging is similar) in proofs works via
  2852 
  2853   using [[trace_simp = true]]
  2854 
  2855 Tracing is then active for all invocations of the simplifier in
  2856 subsequent goal refinement steps. Tracing may also still be enabled or
  2857 disabled via the ProofGeneral settings menu.
  2858 
  2859 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2860 'hide_fact' replace the former 'hide' KIND command.  Minor
  2861 INCOMPATIBILITY.
  2862 
  2863 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2864 more efficient than combinations with -q1 or -q2.
  2865 
  2866 
  2867 *** Pure ***
  2868 
  2869 * Proofterms record type-class reasoning explicitly, using the
  2870 "unconstrain" operation internally.  This eliminates all sort
  2871 constraints from a theorem and proof, introducing explicit
  2872 OFCLASS-premises.  On the proof term level, this operation is
  2873 automatically applied at theorem boundaries, such that closed proofs
  2874 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2875 inspect proof terms.
  2876 
  2877 * Local theory specifications may depend on extra type variables that
  2878 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2879 are added internally.  For example:
  2880 
  2881   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2882 
  2883 * Predicates of locales introduced by classes carry a mandatory
  2884 "class" prefix.  INCOMPATIBILITY.
  2885 
  2886 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2887 
  2888 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2889 'class' instead.
  2890 
  2891 * Command 'code_reflect' allows to incorporate generated ML code into
  2892 runtime environment; replaces immature code_datatype antiquotation.
  2893 INCOMPATIBILITY.
  2894 
  2895 * Code generator: simple concept for abstract datatypes obeying
  2896 invariants.
  2897 
  2898 * Code generator: details of internal data cache have no impact on the
  2899 user space functionality any longer.
  2900 
  2901 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2902 subgoals.  This is more appropriate for interpretations with 'where'.
  2903 INCOMPATIBILITY.
  2904 
  2905 * Command 'example_proof' opens an empty proof body.  This allows to
  2906 experiment with Isar, without producing any persistent result.
  2907 
  2908 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2909 within a local theory context, with explicit checking of the
  2910 constructors involved (in contrast to the raw 'syntax' versions).
  2911 
  2912 * Commands 'types' and 'typedecl' now work within a local theory
  2913 context -- without introducing dependencies on parameters or
  2914 assumptions, which is not possible in Isabelle/Pure.
  2915 
  2916 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2917 within a local theory context.  Minor INCOMPATIBILITY.
  2918 
  2919 
  2920 *** HOL ***
  2921 
  2922 * Command 'typedef' now works within a local theory context -- without
  2923 introducing dependencies on parameters or assumptions, which is not
  2924 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2925 contain multiple interpretations of local typedefs (with different
  2926 non-emptiness proofs), even in a global theory context.
  2927 
  2928 * New package for quotient types.  Commands 'quotient_type' and
  2929 'quotient_definition' may be used for defining types and constants by
  2930 quotient constructions.  An example is the type of integers created by
  2931 quotienting pairs of natural numbers:
  2932 
  2933   fun
  2934     intrel :: "(nat * nat) => (nat * nat) => bool"
  2935   where
  2936     "intrel (x, y) (u, v) = (x + v = u + y)"
  2937 
  2938   quotient_type int = "nat * nat" / intrel
  2939     by (auto simp add: equivp_def expand_fun_eq)
  2940 
  2941   quotient_definition
  2942     "0::int" is "(0::nat, 0::nat)"
  2943 
  2944 The method "lifting" can be used to lift of theorems from the
  2945 underlying "raw" type to the quotient type.  The example
  2946 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2947 construction and provides a reasoning infrastructure for finite sets.
  2948 
  2949 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2950 clash with new theory Quotient in Main HOL.
  2951 
  2952 * Moved the SMT binding into the main HOL session, eliminating
  2953 separate HOL-SMT session.
  2954 
  2955 * List membership infix mem operation is only an input abbreviation.
  2956 INCOMPATIBILITY.
  2957 
  2958 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2959 for future developements; former Library/Word.thy is still present in
  2960 the AFP entry RSAPPS.
  2961 
  2962 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2963 longer shadowed.  INCOMPATIBILITY.
  2964 
  2965 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2966 INCOMPATIBILITY.
  2967 
  2968 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2969 INCOMPATIBILITY.
  2970 
  2971 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2972 instead.  INCOMPATIBILITY.
  2973 
  2974 * Dropped several real-specific versions of lemmas about floor and
  2975 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2976 instead.  INCOMPATIBILITY.
  2977 
  2978   floor_number_of_eq         ~> floor_number_of
  2979   le_floor_eq_number_of      ~> number_of_le_floor
  2980   le_floor_eq_zero           ~> zero_le_floor
  2981   le_floor_eq_one            ~> one_le_floor
  2982   floor_less_eq_number_of    ~> floor_less_number_of
  2983   floor_less_eq_zero         ~> floor_less_zero
  2984   floor_less_eq_one          ~> floor_less_one
  2985   less_floor_eq_number_of    ~> number_of_less_floor
  2986   less_floor_eq_zero         ~> zero_less_floor
  2987   less_floor_eq_one          ~> one_less_floor
  2988   floor_le_eq_number_of      ~> floor_le_number_of
  2989   floor_le_eq_zero           ~> floor_le_zero
  2990   floor_le_eq_one            ~> floor_le_one
  2991   floor_subtract_number_of   ~> floor_diff_number_of
  2992   floor_subtract_one         ~> floor_diff_one
  2993   ceiling_number_of_eq       ~> ceiling_number_of
  2994   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2995   ceiling_le_zero_eq         ~> ceiling_le_zero
  2996   ceiling_le_eq_one          ~> ceiling_le_one
  2997   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2998   less_ceiling_eq_zero       ~> zero_less_ceiling
  2999   less_ceiling_eq_one        ~> one_less_ceiling
  3000   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  3001   ceiling_less_eq_zero       ~> ceiling_less_zero
  3002   ceiling_less_eq_one        ~> ceiling_less_one
  3003   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  3004   le_ceiling_eq_zero         ~> zero_le_ceiling
  3005   le_ceiling_eq_one          ~> one_le_ceiling
  3006   ceiling_subtract_number_of ~> ceiling_diff_number_of
  3007   ceiling_subtract_one       ~> ceiling_diff_one
  3008 
  3009 * Theory "Finite_Set": various folding_XXX locales facilitate the
  3010 application of the various fold combinators on finite sets.
  3011 
  3012 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  3013 provides abstract red-black tree type which is backed by "RBT_Impl" as
  3014 implementation.  INCOMPATIBILTY.
  3015 
  3016 * Theory Library/Coinductive_List has been removed -- superseded by
  3017 AFP/thys/Coinductive.
  3018 
  3019 * Theory PReal, including the type "preal" and related operations, has
  3020 been removed.  INCOMPATIBILITY.
  3021 
  3022 * Real: new development using Cauchy Sequences.
  3023 
  3024 * Split off theory "Big_Operators" containing setsum, setprod,
  3025 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  3026 
  3027 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  3028 "Int" etc.  INCOMPATIBILITY.
  3029 
  3030 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  3031 
  3032 * New set of rules "ac_simps" provides combined assoc / commute
  3033 rewrites for all interpretations of the appropriate generic locales.
  3034 
  3035 * Renamed theory "OrderedGroup" to "Groups" and split theory
  3036 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  3037 appropriate and more consistent names suitable for name prefixes
  3038 within the HOL theories.  INCOMPATIBILITY.
  3039 
  3040 * Some generic constants have been put to appropriate theories:
  3041   - less_eq, less: Orderings
  3042   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  3043   - inverse, divide: Rings
  3044 INCOMPATIBILITY.
  3045 
  3046 * More consistent naming of type classes involving orderings (and
  3047 lattices):
  3048 
  3049     lower_semilattice                   ~> semilattice_inf
  3050     upper_semilattice                   ~> semilattice_sup
  3051 
  3052     dense_linear_order                  ~> dense_linorder
  3053 
  3054     pordered_ab_group_add               ~> ordered_ab_group_add
  3055     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  3056     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  3057     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  3058     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  3059     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  3060     pordered_cancel_semiring            ~> ordered_cancel_semiring
  3061     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  3062     pordered_comm_ring                  ~> ordered_comm_ring
  3063     pordered_comm_semiring              ~> ordered_comm_semiring
  3064     pordered_ring                       ~> ordered_ring
  3065     pordered_ring_abs                   ~> ordered_ring_abs
  3066     pordered_semiring                   ~> ordered_semiring
  3067 
  3068     ordered_ab_group_add                ~> linordered_ab_group_add
  3069     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  3070     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  3071     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  3072     ordered_field                       ~> linordered_field
  3073     ordered_field_no_lb                 ~> linordered_field_no_lb
  3074     ordered_field_no_ub                 ~> linordered_field_no_ub
  3075     ordered_field_dense_linear_order    ~> dense_linordered_field
  3076     ordered_idom                        ~> linordered_idom
  3077     ordered_ring                        ~> linordered_ring
  3078     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  3079     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  3080     ordered_ring_strict                 ~> linordered_ring_strict
  3081     ordered_semidom                     ~> linordered_semidom
  3082     ordered_semiring                    ~> linordered_semiring
  3083     ordered_semiring_1                  ~> linordered_semiring_1
  3084     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  3085     ordered_semiring_strict             ~> linordered_semiring_strict
  3086 
  3087   The following slightly odd type classes have been moved to a
  3088   separate theory Library/Lattice_Algebras:
  3089 
  3090     lordered_ab_group_add               ~> lattice_ab_group_add
  3091     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  3092     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  3093     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  3094     lordered_ring                       ~> lattice_ring
  3095 
  3096 INCOMPATIBILITY.
  3097 
  3098 * Refined field classes:
  3099   - classes division_ring_inverse_zero, field_inverse_zero,
  3100     linordered_field_inverse_zero include rule inverse 0 = 0 --
  3101     subsumes former division_by_zero class;
  3102   - numerous lemmas have been ported from field to division_ring.
  3103 INCOMPATIBILITY.
  3104 
  3105 * Refined algebra theorem collections:
  3106   - dropped theorem group group_simps, use algebra_simps instead;
  3107   - dropped theorem group ring_simps, use field_simps instead;
  3108   - proper theorem collection field_simps subsumes former theorem
  3109     groups field_eq_simps and field_simps;
  3110   - dropped lemma eq_minus_self_iff which is a duplicate for
  3111     equal_neg_zero.
  3112 INCOMPATIBILITY.
  3113 
  3114 * Theory Finite_Set and List: some lemmas have been generalized from
  3115 sets to lattices:
  3116 
  3117   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  3118   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  3119   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  3120   union_Union_fold_union        ~> sup_Sup_fold_sup
  3121   Inter_fold_inter              ~> Inf_fold_inf
  3122   Union_fold_union              ~> Sup_fold_sup
  3123   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  3124   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  3125   INTER_fold_inter              ~> INFI_fold_inf
  3126   UNION_fold_union              ~> SUPR_fold_sup
  3127 
  3128 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  3129 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  3130 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  3131 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  3132 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  3133 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  3134 
  3135 * Reorganized theory Multiset: swapped notation of pointwise and
  3136 multiset order:
  3137 
  3138   - pointwise ordering is instance of class order with standard syntax
  3139     <= and <;
  3140   - multiset ordering has syntax <=# and <#; partial order properties
  3141     are provided by means of interpretation with prefix
  3142     multiset_order;
  3143   - less duplication, less historical organization of sections,
  3144     conversion from associations lists to multisets, rudimentary code
  3145     generation;
  3146   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  3147     if needed.
  3148 
  3149 Renamed:
  3150 
  3151   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  3152   multi_count_ext  ~>  multiset_ext
  3153   diff_union_inverse2  ~>  diff_union_cancelR
  3154 
  3155 INCOMPATIBILITY.
  3156 
  3157 * Theory Permutation: replaced local "remove" by List.remove1.
  3158 
  3159 * Code generation: ML and OCaml code is decorated with signatures.
  3160 
  3161 * Theory List: added transpose.
  3162 
  3163 * Library/Nat_Bijection.thy is a collection of bijective functions
  3164 between nat and other types, which supersedes the older libraries
  3165 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  3166 
  3167   Constants:
  3168   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  3169   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  3170   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  3171   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  3172   Countable.pair_encode           ~> prod_encode
  3173   NatIso.prod2nat                 ~> prod_encode
  3174   NatIso.nat2prod                 ~> prod_decode
  3175   NatIso.sum2nat                  ~> sum_encode
  3176   NatIso.nat2sum                  ~> sum_decode
  3177   NatIso.list2nat                 ~> list_encode
  3178   NatIso.nat2list                 ~> list_decode
  3179   NatIso.set2nat                  ~> set_encode
  3180   NatIso.nat2set                  ~> set_decode
  3181 
  3182   Lemmas:
  3183   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3184   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3185   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3186   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3187   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3188   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3189   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3190   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3191   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3192   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3193   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3194   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3195   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3196 
  3197 * Sledgehammer:
  3198   - Renamed ATP commands:
  3199     atp_info     ~> sledgehammer running_atps
  3200     atp_kill     ~> sledgehammer kill_atps
  3201     atp_messages ~> sledgehammer messages
  3202     atp_minimize ~> sledgehammer minimize
  3203     print_atps   ~> sledgehammer available_atps
  3204     INCOMPATIBILITY.
  3205   - Added user's manual ("isabelle doc sledgehammer").
  3206   - Added option syntax and "sledgehammer_params" to customize
  3207     Sledgehammer's behavior.  See the manual for details.
  3208   - Modified the Isar proof reconstruction code so that it produces
  3209     direct proofs rather than proofs by contradiction.  (This feature
  3210     is still experimental.)
  3211   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3212     full-typed mode.
  3213   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3214 
  3215 * Nitpick:
  3216   - Added and implemented "binary_ints" and "bits" options.
  3217   - Added "std" option and implemented support for nonstandard models.
  3218   - Added and implemented "finitize" option to improve the precision
  3219     of infinite datatypes based on a monotonicity analysis.
  3220   - Added support for quotient types.
  3221   - Added support for "specification" and "ax_specification"
  3222     constructs.
  3223   - Added support for local definitions (for "function" and
  3224     "termination" proofs).
  3225   - Added support for term postprocessors.
  3226   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3227   - Improved efficiency of "destroy_constrs" optimization.
  3228   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3229     record getters.
  3230   - Fixed soundness bug related to higher-order constructors.
  3231   - Fixed soundness bug when "full_descrs" is enabled.
  3232   - Improved precision of set constructs.
  3233   - Added "atoms" option.
  3234   - Added cache to speed up repeated Kodkod invocations on the same
  3235     problems.
  3236   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3237     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3238     "SAT4J_Light".  INCOMPATIBILITY.
  3239   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3240     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3241   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3242 
  3243 * Method "induct" now takes instantiations of the form t, where t is not
  3244   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3245   If this is not intended, t has to be enclosed in parentheses.
  3246   By default, the equalities generated by definitional instantiations
  3247   are pre-simplified, which may cause parameters of inductive cases
  3248   to disappear, or may even delete some of the inductive cases.
  3249   Use "induct (no_simp)" instead of "induct" to restore the old
  3250   behaviour. The (no_simp) option is also understood by the "cases"
  3251   and "nominal_induct" methods, which now perform pre-simplification, too.
  3252   INCOMPATIBILITY.
  3253 
  3254 
  3255 *** HOLCF ***
  3256 
  3257 * Variable names in lemmas generated by the domain package have
  3258 changed; the naming scheme is now consistent with the HOL datatype
  3259 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3260 
  3261 * The domain package no longer defines the function "foo_copy" for
  3262 recursive domain "foo".  The reach lemma is now stated directly in
  3263 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3264 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3265 
  3266 * Most definedness lemmas generated by the domain package (previously
  3267 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3268 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3269 Proofs that used definedness lemmas as intro rules may break,
  3270 potential INCOMPATIBILITY.
  3271 
  3272 * Induction and casedist rules generated by the domain package now
  3273 declare proper case_names (one called "bottom", and one named for each
  3274 constructor).  INCOMPATIBILITY.
  3275 
  3276 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3277 rules are generated for each domain, INCOMPATIBILITY.
  3278 
  3279   foo_bar.reach       ~> foo.reach  bar.reach
  3280   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3281 
  3282 * Some lemmas generated by the domain package have been renamed for
  3283 consistency with the datatype package, INCOMPATIBILITY.
  3284 
  3285   foo.ind        ~> foo.induct
  3286   foo.finite_ind ~> foo.finite_induct
  3287   foo.coind      ~> foo.coinduct
  3288   foo.casedist   ~> foo.exhaust
  3289   foo.exhaust    ~> foo.nchotomy
  3290 
  3291 * For consistency with other definition packages, the fixrec package
  3292 now generates qualified theorem names, INCOMPATIBILITY.
  3293 
  3294   foo_simps  ~> foo.simps
  3295   foo_unfold ~> foo.unfold
  3296   foo_induct ~> foo.induct
  3297 
  3298 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3299 method and internal fixrec proofs now use the default simpset instead.
  3300 INCOMPATIBILITY.
  3301 
  3302 * The "contlub" predicate has been removed.  Proof scripts should use
  3303 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3304 
  3305 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3306 
  3307 * The constants cpair, cfst, and csnd have been removed in favor of
  3308 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3309 
  3310 
  3311 *** ML ***
  3312 
  3313 * Antiquotations for basic formal entities:
  3314 
  3315     @{class NAME}         -- type class
  3316     @{class_syntax NAME}  -- syntax representation of the above
  3317 
  3318     @{type_name NAME}     -- logical type
  3319     @{type_abbrev NAME}   -- type abbreviation
  3320     @{nonterminal NAME}   -- type of concrete syntactic category
  3321     @{type_syntax NAME}   -- syntax representation of any of the above
  3322 
  3323     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3324     @{const_abbrev NAME}  -- abbreviated constant
  3325     @{const_syntax NAME}  -- syntax representation of any of the above
  3326 
  3327 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3328 syntax constant (cf. 'syntax' command).
  3329 
  3330 * Antiquotation @{make_string} inlines a function to print arbitrary
  3331 values similar to the ML toplevel.  The result is compiler dependent
  3332 and may fall back on "?" in certain situations.
  3333 
  3334 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3335 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3336 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3337 not work with the asynchronous Isar document model.
  3338 
  3339 * Configuration options now admit dynamic default values, depending on
  3340 the context or even global references.
  3341 
  3342 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3343 uses an efficient external library if available (for Poly/ML).
  3344 
  3345 * Renamed some important ML structures, while keeping the old names
  3346 for some time as aliases within the structure Legacy:
  3347 
  3348   OuterKeyword  ~>  Keyword
  3349   OuterLex      ~>  Token
  3350   OuterParse    ~>  Parse
  3351   OuterSyntax   ~>  Outer_Syntax
  3352   PrintMode     ~>  Print_Mode
  3353   SpecParse     ~>  Parse_Spec
  3354   ThyInfo       ~>  Thy_Info
  3355   ThyLoad       ~>  Thy_Load
  3356   ThyOutput     ~>  Thy_Output
  3357   TypeInfer     ~>  Type_Infer
  3358 
  3359 Note that "open Legacy" simplifies porting of sources, but forgetting
  3360 to remove it again will complicate porting again in the future.
  3361 
  3362 * Most operations that refer to a global context are named
  3363 accordingly, e.g. Simplifier.global_context or
  3364 ProofContext.init_global.  There are some situations where a global
  3365 context actually works, but under normal circumstances one needs to
  3366 pass the proper local context through the code!
  3367 
  3368 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3369 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3370 Pretty.pp argument to merge, which is absent in the standard
  3371 Theory_Data version.
  3372 
  3373 * Sorts.certify_sort and derived "cert" operations for types and terms
  3374 no longer minimize sorts.  Thus certification at the boundary of the
  3375 inference kernel becomes invariant under addition of class relations,
  3376 which is an important monotonicity principle.  Sorts are now minimized
  3377 in the syntax layer only, at the boundary between the end-user and the
  3378 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3379 explicitly in rare situations.
  3380 
  3381 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3382 emphasize that this is in no way a standard operation.
  3383 INCOMPATIBILITY.
  3384 
  3385 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3386 usual for resolution.  Rare INCOMPATIBILITY.
  3387 
  3388 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3389 to emphasize that these only work in a global situation (which is
  3390 quite rare).
  3391 
  3392 * Curried take and drop in library.ML; negative length is interpreted
  3393 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3394 
  3395 * Proof terms: type substitutions on proof constants now use canonical
  3396 order of type variables.  INCOMPATIBILITY for tools working with proof
  3397 terms.
  3398 
  3399 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3400 may no longer carry premises.  User-level specifications are
  3401 transformed accordingly by Thm.add_axiom/add_def.
  3402 
  3403 
  3404 *** System ***
  3405 
  3406 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3407 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3408 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3409 
  3410 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3411 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3412 respectively.
  3413 
  3414 * Old lib/scripts/polyml-platform is superseded by the
  3415 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3416 variant, even on a 64 bit machine.  The following example setting
  3417 prefers 64 bit if available:
  3418 
  3419   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3420 
  3421 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3422 Isabelle/Scala layer for advanced prover interaction and integration.
  3423 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3424 component.
  3425 
  3426 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3427 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3428 similar to the default assignment of the document preparation system
  3429 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3430 provides some operations for direct access to the font without asking
  3431 the user for manual installation.
  3432 
  3433 
  3434 
  3435 New in Isabelle2009-1 (December 2009)
  3436 -------------------------------------
  3437 
  3438 *** General ***
  3439 
  3440 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3441 one backslash should be used, even in ML sources.
  3442 
  3443 
  3444 *** Pure ***
  3445 
  3446 * Locale interpretation propagates mixins along the locale hierarchy.
  3447 The currently only available mixins are the equations used to map
  3448 local definitions to terms of the target domain of an interpretation.
  3449 
  3450 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3451 loc" to print all interpretations of locale "loc" in the theory.
  3452 Interpretations in proofs are not shown.
  3453 
  3454 * Thoroughly revised locales tutorial.  New section on conditional
  3455 interpretation.
  3456 
  3457 * On instantiation of classes, remaining undefined class parameters
  3458 are formally declared.  INCOMPATIBILITY.
  3459 
  3460 
  3461 *** Document preparation ***
  3462 
  3463 * New generalized style concept for printing terms: @{foo (style) ...}
  3464 instead of @{foo_style style ...}  (old form is still retained for
  3465 backward compatibility).  Styles can be also applied for
  3466 antiquotations prop, term_type and typeof.
  3467 
  3468 
  3469 *** HOL ***
  3470 
  3471 * New proof method "smt" for a combination of first-order logic with
  3472 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3473 fixed-size bitvectors; there is also basic support for higher-order
  3474 features (esp. lambda abstractions).  It is an incomplete decision
  3475 procedure based on external SMT solvers using the oracle mechanism;
  3476 for the SMT solver Z3, this method is proof-producing.  Certificates
  3477 are provided to avoid calling the external solvers solely for
  3478 re-checking proofs.  Due to a remote SMT service there is no need for
  3479 installing SMT solvers locally.  See src/HOL/SMT.
  3480 
  3481 * New commands to load and prove verification conditions generated by
  3482 the Boogie program verifier or derived systems (e.g. the Verifying C
  3483 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3484 
  3485 * New counterexample generator tool 'nitpick' based on the Kodkod
  3486 relational model finder.  See src/HOL/Tools/Nitpick and
  3487 src/HOL/Nitpick_Examples.
  3488 
  3489 * New commands 'code_pred' and 'values' to invoke the predicate
  3490 compiler and to enumerate values of inductive predicates.
  3491 
  3492 * A tabled implementation of the reflexive transitive closure.
  3493 
  3494 * New implementation of quickcheck uses generic code generator;
  3495 default generators are provided for all suitable HOL types, records
  3496 and datatypes.  Old quickcheck can be re-activated importing theory
  3497 Library/SML_Quickcheck.
  3498 
  3499 * New testing tool Mirabelle for automated proof tools.  Applies
  3500 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3501 every proof step in a theory.  To be used in batch mode via the
  3502 "mirabelle" utility.
  3503 
  3504 * New proof method "sos" (sum of squares) for nonlinear real
  3505 arithmetic (originally due to John Harison). It requires theory
  3506 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3507 works well in practice on quantifier-free real arithmetic with +, -,
  3508 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3509 inequalities between polynomials.  It makes use of external
  3510 semidefinite programming solvers.  Method "sos" generates a
  3511 certificate that can be pasted into the proof thus avoiding the need
  3512 to call an external tool every time the proof is checked.  See
  3513 src/HOL/Library/Sum_Of_Squares.
  3514 
  3515 * New method "linarith" invokes existing linear arithmetic decision
  3516 procedure only.
  3517 
  3518 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3519 
  3520 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3521 Causes full type information to be output to the ATPs.  This slows
  3522 ATPs down considerably but eliminates a source of unsound "proofs"
  3523 that fail later.
  3524 
  3525 * New method "metisFT": A version of metis that uses full type
  3526 information in order to avoid failures of proof reconstruction.
  3527 
  3528 * New evaluator "approximate" approximates an real valued term using
  3529 the same method as the approximation method.
  3530 
  3531 * Method "approximate" now supports arithmetic expressions as
  3532 boundaries of intervals and implements interval splitting and Taylor
  3533 series expansion.
  3534 
  3535 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3536 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3537 
  3538 * New theory SupInf of the supremum and infimum operators for sets of
  3539 reals.
  3540 
  3541 * New theory Probability, which contains a development of measure
  3542 theory, eventually leading to Lebesgue integration and probability.
  3543 
  3544 * Extended Multivariate Analysis to include derivation and Brouwer's
  3545 fixpoint theorem.
  3546 
  3547 * Reorganization of number theory, INCOMPATIBILITY:
  3548   - new number theory development for nat and int, in theories Divides
  3549     and GCD as well as in new session Number_Theory
  3550   - some constants and facts now suffixed with _nat and _int
  3551     accordingly
  3552   - former session NumberTheory now named Old_Number_Theory, including
  3553     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3554   - moved theory Pocklington from src/HOL/Library to
  3555     src/HOL/Old_Number_Theory
  3556 
  3557 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3558 lcm of finite and infinite sets. It is shown that they form a complete
  3559 lattice.
  3560 
  3561 * Class semiring_div requires superclass no_zero_divisors and proof of
  3562 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3563 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3564 generalized to class semiring_div, subsuming former theorems
  3565 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3566 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3567 INCOMPATIBILITY.
  3568 
  3569 * Refinements to lattice classes and sets:
  3570   - less default intro/elim rules in locale variant, more default
  3571     intro/elim rules in class variant: more uniformity
  3572   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3573     le_inf_iff
  3574   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3575     sup_aci)
  3576   - renamed ACI to inf_sup_aci
  3577   - new class "boolean_algebra"
  3578   - class "complete_lattice" moved to separate theory
  3579     "Complete_Lattice"; corresponding constants (and abbreviations)
  3580     renamed and with authentic syntax:
  3581     Set.Inf ~>    Complete_Lattice.Inf
  3582     Set.Sup ~>    Complete_Lattice.Sup
  3583     Set.INFI ~>   Complete_Lattice.INFI
  3584     Set.SUPR ~>   Complete_Lattice.SUPR
  3585     Set.Inter ~>  Complete_Lattice.Inter
  3586     Set.Union ~>  Complete_Lattice.Union
  3587     Set.INTER ~>  Complete_Lattice.INTER
  3588     Set.UNION ~>  Complete_Lattice.UNION
  3589   - authentic syntax for
  3590     Set.Pow
  3591     Set.image
  3592   - mere abbreviations:
  3593     Set.empty               (for bot)
  3594     Set.UNIV                (for top)
  3595     Set.inter               (for inf, formerly Set.Int)
  3596     Set.union               (for sup, formerly Set.Un)
  3597     Complete_Lattice.Inter  (for Inf)
  3598     Complete_Lattice.Union  (for Sup)
  3599     Complete_Lattice.INTER  (for INFI)
  3600     Complete_Lattice.UNION  (for SUPR)
  3601   - object-logic definitions as far as appropriate
  3602 
  3603 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3604 Un_subset_iff are explicitly deleted as default simp rules; then also
  3605 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3606 deleted to achieve the desired effect.
  3607 
  3608 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3609 rules by default any longer; the same applies to min_max.inf_absorb1
  3610 etc.  INCOMPATIBILITY.
  3611 
  3612 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3613 pred_set_conv by default.  INCOMPATIBILITY.
  3614 
  3615 * Power operations on relations and functions are now one dedicated
  3616 constant "compow" with infix syntax "^^".  Power operation on
  3617 multiplicative monoids retains syntax "^" and is now defined generic
  3618 in class power.  INCOMPATIBILITY.
  3619 
  3620 * Relation composition "R O S" now has a more standard argument order:
  3621 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3622 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3623 break, since the O_assoc rule was not rewritten like this.  Fix using
  3624 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3625 
  3626 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3627 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3628 INCOMPATIBILITY.
  3629 
  3630 * Most rules produced by inductive and datatype package have mandatory
  3631 prefixes.  INCOMPATIBILITY.
  3632 
  3633 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3634 the attribute of the same name.  Each of the theorems in the list
  3635 DERIV_intros assumes composition with an additional function and
  3636 matches a variable to the derivative, which has to be solved by the
  3637 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3638 of most elementary terms.  Former Maclauren.DERIV_tac and
  3639 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3640 INCOMPATIBILITY.
  3641 
  3642 * Code generator attributes follow the usual underscore convention:
  3643     code_unfold     replaces    code unfold
  3644     code_post       replaces    code post
  3645     etc.
  3646   INCOMPATIBILITY.
  3647 
  3648 * Renamed methods:
  3649     sizechange -> size_change
  3650     induct_scheme -> induction_schema
  3651   INCOMPATIBILITY.
  3652 
  3653 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3654 INCOMPATIBILITY, use "undefined" directly.
  3655 
  3656 * Renamed theorems:
  3657     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3658     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3659     Suc_plus1 -> Suc_eq_plus1
  3660     *anti_sym -> *antisym*
  3661     vector_less_eq_def -> vector_le_def
  3662   INCOMPATIBILITY.
  3663 
  3664 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3665 INCOMPATIBILITY.
  3666 
  3667 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3668 INCOMPATIBILITY.
  3669 
  3670 
  3671 *** HOLCF ***
  3672 
  3673 * Theory Representable defines a class "rep" of domains that are
  3674 representable (via an ep-pair) in the universal domain type "udom".
  3675 Instances are provided for all type constructors defined in HOLCF.
  3676 
  3677 * The 'new_domain' command is a purely definitional version of the
  3678 domain package, for representable domains.  Syntax is identical to the
  3679 old domain package.  The 'new_domain' package also supports indirect
  3680 recursion using previously-defined type constructors.  See
  3681 src/HOLCF/ex/New_Domain.thy for examples.
  3682 
  3683 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3684 on the left-hand side of an equation, and then performs
  3685 simplification.  Rewriting is done using rules declared with the
  3686 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3687 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3688 
  3689 * The pattern-match compiler in 'fixrec' can now handle constructors
  3690 with HOL function types.  Pattern-match combinators for the Pair
  3691 constructor are pre-configured.
  3692 
  3693 * The 'fixrec' package now produces better fixed-point induction rules
  3694 for mutually-recursive definitions:  Induction rules have conclusions
  3695 of the form "P foo bar" instead of "P <foo, bar>".
  3696 
  3697 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3698 been renamed to "below".  The name "below" now replaces "less" in many
  3699 theorem names.  (Legacy theorem names using "less" are still supported
  3700 as well.)
  3701 
  3702 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3703 patterns can be used to generate strictness rules, or to make
  3704 functions more strict (much like the bang-patterns supported by the
  3705 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3706 examples.
  3707 
  3708 
  3709 *** ML ***
  3710 
  3711 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3712 errors and run-time exceptions, including detailed source positions.
  3713 
  3714 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3715 identified entries, with some additional information such as source
  3716 position, logical grouping etc.
  3717 
  3718 * Theory and context data is now introduced by the simplified and
  3719 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3720 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3721 copy operation) is still available for some time.
  3722 
  3723 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3724 provides a high-level programming interface to synchronized state
  3725 variables with atomic update.  This works via pure function
  3726 application within a critical section -- its runtime should be as
  3727 short as possible; beware of deadlocks if critical code is nested,
  3728 either directly or indirectly via other synchronized variables!
  3729 
  3730 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3731 wraps raw ML references, explicitly indicating their non-thread-safe
  3732 behaviour.  The Isar toplevel keeps this structure open, to
  3733 accommodate Proof General as well as quick and dirty interactive
  3734 experiments with references.
  3735 
  3736 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3737 parallel tactical reasoning.
  3738 
  3739 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3740 are similar to SUBPROOF, but are slightly more flexible: only the
  3741 specified parts of the subgoal are imported into the context, and the
  3742 body tactic may introduce new subgoals and schematic variables.
  3743 
  3744 * Old tactical METAHYPS, which does not observe the proof context, has
  3745 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3746 or Subgoal.FOCUS etc.
  3747 
  3748 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3749 functors have their own ML name space there is no point to mark them
  3750 separately.)  Minor INCOMPATIBILITY.
  3751 
  3752 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3753 
  3754 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3755 INCOMPATIBILITY.
  3756 
  3757 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3758 mode, which avoids critical setmp.
  3759 
  3760 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3761 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3762 Attrib/Method.setup introduced in Isabelle2009.
  3763 
  3764 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3765 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3766 last resort.  INCOMPATIBILITY.
  3767 
  3768 * Display.pretty_thm now requires a proper context (cf. former
  3769 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3770 or even Display.pretty_thm_without_context as last resort.
  3771 INCOMPATIBILITY.
  3772 
  3773 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3774 Syntax.pretty_typ/term directly, preferably with proper context
  3775 instead of global theory.
  3776 
  3777 
  3778 *** System ***
  3779 
  3780 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3781 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3782 usedir option -q.
  3783 
  3784 * Support for additional "Isabelle components" via etc/components, see
  3785 also the system manual.
  3786 
  3787 * The isabelle makeall tool now operates on all components with
  3788 IsaMakefile, not just hardwired "logics".
  3789 
  3790 * Removed "compress" option from isabelle-process and isabelle usedir;
  3791 this is always enabled.
  3792 
  3793 * Discontinued support for Poly/ML 4.x versions.
  3794 
  3795 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3796 on a given logic image.  This requires the lighttpd webserver and is
  3797 currently supported on Linux only.
  3798 
  3799 
  3800 
  3801 New in Isabelle2009 (April 2009)
  3802 --------------------------------
  3803 
  3804 *** General ***
  3805 
  3806 * Simplified main Isabelle executables, with less surprises on
  3807 case-insensitive file-systems (such as Mac OS).
  3808 
  3809   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3810     "isatool."
  3811 
  3812   - The former "isabelle" alias for "isabelle-process" has been
  3813     removed (should rarely occur to regular users).
  3814 
  3815   - The former "isabelle-interface" and its alias "Isabelle" have been
  3816     removed (interfaces are now regular Isabelle tools).
  3817 
  3818 Within scripts and make files, the Isabelle environment variables
  3819 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3820 respectively.  (The latter are still available as legacy feature.)
  3821 
  3822 The old isabelle-interface wrapper could react in confusing ways if
  3823 the interface was uninstalled or changed otherwise.  Individual
  3824 interface tool configuration is now more explicit, see also the
  3825 Isabelle system manual.  In particular, Proof General is now available
  3826 via "isabelle emacs".
  3827 
  3828 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3829 purge installed copies of Isabelle executables and re-run "isabelle
  3830 install -p ...", or use symlinks.
  3831 
  3832 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3833 old ~/isabelle, which was slightly non-standard and apt to cause
  3834 surprises on case-insensitive file-systems (such as Mac OS).
  3835 
  3836 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3837 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3838 care is required when using older releases of Isabelle.  Note that
  3839 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3840 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3841 
  3842 * Proofs of fully specified statements are run in parallel on
  3843 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3844 a regular 4-core machine, if the initial heap space is made reasonably
  3845 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3846 
  3847 * The main reference manuals ("isar-ref", "implementation", and
  3848 "system") have been updated and extended.  Formally checked references
  3849 as hyperlinks are now available uniformly.
  3850 
  3851 
  3852 *** Pure ***
  3853 
  3854 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3855 respects.  The most important changes are listed below.  See the
  3856 Tutorial on Locales ("locales" manual) for details.
  3857 
  3858 - In locale expressions, instantiation replaces renaming.  Parameters
  3859 must be declared in a for clause.  To aid compatibility with previous
  3860 parameter inheritance, in locale declarations, parameters that are not
  3861 'touched' (instantiation position "_" or omitted) are implicitly added
  3862 with their syntax at the beginning of the for clause.
  3863 
  3864 - Syntax from abbreviations and definitions in locales is available in
  3865 locale expressions and context elements.  The latter is particularly
  3866 useful in locale declarations.
  3867 
  3868 - More flexible mechanisms to qualify names generated by locale
  3869 expressions.  Qualifiers (prefixes) may be specified in locale
  3870 expressions, and can be marked as mandatory (syntax: "name!:") or
  3871 optional (syntax "name?:").  The default depends for plain "name:"
  3872 depends on the situation where a locale expression is used: in
  3873 commands 'locale' and 'sublocale' prefixes are optional, in
  3874 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3875 implicit qualifiers derived from the parameter names of a locale are
  3876 no longer generated.
  3877 
  3878 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3879 instantiation clause in "interpretation" and "interpret" (square
  3880 brackets) is no longer available.  Use locale expressions.
  3881 
  3882 - When converting proof scripts, mandatory qualifiers in
  3883 'interpretation' and 'interpret' should be retained by default, even
  3884 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3885 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3886 in locale expressions range over a single locale instance only.
  3887 
  3888 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3889 In existing theorem specifications replace the includes element by the
  3890 respective context elements of the included locale, omitting those
  3891 that are already present in the theorem specification.  Multiple
  3892 assume elements of a locale should be replaced by a single one
  3893 involving the locale predicate.  In the proof body, declarations (most
  3894 notably theorems) may be regained by interpreting the respective
  3895 locales in the proof context as required (command "interpret").
  3896 
  3897 If using "includes" in replacement of a target solely because the
  3898 parameter types in the theorem are not as general as in the target,
  3899 consider declaring a new locale with additional type constraints on
  3900 the parameters (context element "constrains").
  3901 
  3902 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3903 
  3904 - Locale interpretation commands no longer attempt to simplify goal.
  3905 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3906 methods intro_locales and unfold_locales to clarify.
  3907 
  3908 - Locale interpretation commands no longer accept interpretation
  3909 attributes.  INCOMPATIBILITY.
  3910 
  3911 * Class declaration: so-called "base sort" must not be given in import
  3912 list any longer, but is inferred from the specification.  Particularly
  3913 in HOL, write
  3914 
  3915     class foo = ...
  3916 
  3917 instead of
  3918 
  3919     class foo = type + ...
  3920 
  3921 * Class target: global versions of theorems stemming do not carry a
  3922 parameter prefix any longer.  INCOMPATIBILITY.
  3923 
  3924 * Class 'instance' command no longer accepts attached definitions.
  3925 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3926 
  3927 * Recovered hiding of consts, which was accidentally broken in
  3928 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3929 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3930 
  3931 * Slightly more coherent Pure syntax, with updated documentation in
  3932 isar-ref manual.  Removed locales meta_term_syntax and
  3933 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3934 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3935 directly in regular applications.
  3936 
  3937 * There is a new syntactic category "float_const" for signed decimal
  3938 fractions (e.g. 123.45 or -123.45).
  3939 
  3940 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3941 interface with 'setup' command instead.
  3942 
  3943 * Command 'local_setup' is similar to 'setup', but operates on a local
  3944 theory context.
  3945 
  3946 * The 'axiomatization' command now only works within a global theory
  3947 context.  INCOMPATIBILITY.
  3948 
  3949 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3950 hypotheses.  Sorts required in the course of reasoning need to be
  3951 covered by the constraints in the initial statement, completed by the
  3952 type instance information of the background theory.  Non-trivial sort
  3953 hypotheses, which rarely occur in practice, may be specified via
  3954 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3955 
  3956   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3957 
  3958 The result contains an implicit sort hypotheses as before --
  3959 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3960 normalization.
  3961 
  3962 * Generalized Isar history, with support for linear undo, direct state
  3963 addressing etc.
  3964 
  3965 * Changed defaults for unify configuration options:
  3966 
  3967   unify_trace_bound = 50 (formerly 25)
  3968   unify_search_bound = 60 (formerly 30)
  3969 
  3970 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3971 
  3972   a) On theory merge, the last set of code equations for a particular
  3973      constant is taken (in accordance with the policy applied by other
  3974      parts of the code generator framework).
  3975 
  3976   b) Code equations stemming from explicit declarations (e.g. code
  3977      attribute) gain priority over default code equations stemming
  3978      from definition, primrec, fun etc.
  3979 
  3980 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3981 
  3982 * Unified theorem tables for both code generators.  Thus [code
  3983 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3984 
  3985 * Command 'find_consts' searches for constants based on type and name
  3986 patterns, e.g.
  3987 
  3988     find_consts "_ => bool"
  3989 
  3990 By default, matching is against subtypes, but it may be restricted to
  3991 the whole type.  Searching by name is possible.  Multiple queries are
  3992 conjunctive and queries may be negated by prefixing them with a
  3993 hyphen:
  3994 
  3995     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3996 
  3997 * New 'find_theorems' criterion "solves" matches theorems that
  3998 directly solve the current goal (modulo higher-order unification).
  3999 
  4000 * Auto solve feature for main theorem statements: whenever a new goal
  4001 is stated, "find_theorems solves" is called; any theorems that could
  4002 solve the lemma directly are listed as part of the goal state.
  4003 Cf. associated options in Proof General Isabelle settings menu,
  4004 enabled by default, with reasonable timeout for pathological cases of
  4005 higher-order unification.
  4006 
  4007 
  4008 *** Document preparation ***
  4009 
  4010 * Antiquotation @{lemma} now imitates a regular terminal proof,
  4011 demanding keyword 'by' and supporting the full method expression
  4012 syntax just like the Isar command 'by'.
  4013 
  4014 
  4015 *** HOL ***
  4016 
  4017 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  4018 points Main and Complex_Main remain as before.
  4019 
  4020 * Logic image HOL-Plain provides a minimal HOL with the most important
  4021 tools available (inductive, datatype, primrec, ...).  This facilitates
  4022 experimentation and tool development.  Note that user applications
  4023 (and library theories) should never refer to anything below theory
  4024 Main, as before.
  4025 
  4026 * Logic image HOL-Main stops at theory Main, and thus facilitates
  4027 experimentation due to shorter build times.
  4028 
  4029 * Logic image HOL-NSA contains theories of nonstandard analysis which
  4030 were previously part of former HOL-Complex.  Entry point Hyperreal
  4031 remains valid, but theories formerly using Complex_Main should now use
  4032 new entry point Hypercomplex.
  4033 
  4034 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  4035 Posix processes.  Avoids potentially expensive forking of the ML
  4036 process.  New thread-based implementation also works on non-Unix
  4037 platforms (Cygwin).  Provers are no longer hardwired, but defined
  4038 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  4039 commands are covered in the isar-ref manual.
  4040 
  4041 * Wrapper scripts for remote SystemOnTPTP service allows to use
  4042 sledgehammer without local ATP installation (Vampire etc.). Other
  4043 provers may be included via suitable ML wrappers, see also
  4044 src/HOL/ATP_Linkup.thy.
  4045 
  4046 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  4047 menu.
  4048 
  4049 * The metis method no longer fails because the theorem is too trivial
  4050 (contains the empty clause).
  4051 
  4052 * The metis method now fails in the usual manner, rather than raising
  4053 an exception, if it determines that it cannot prove the theorem.
  4054 
  4055 * Method "coherent" implements a prover for coherent logic (see also
  4056 src/Tools/coherent.ML).
  4057 
  4058 * Constants "undefined" and "default" replace "arbitrary".  Usually
  4059 "undefined" is the right choice to replace "arbitrary", though
  4060 logically there is no difference.  INCOMPATIBILITY.
  4061 
  4062 * Command "value" now integrates different evaluation mechanisms.  The
  4063 result of the first successful evaluation mechanism is printed.  In
  4064 square brackets a particular named evaluation mechanisms may be
  4065 specified (currently, [SML], [code] or [nbe]).  See further
  4066 src/HOL/ex/Eval_Examples.thy.
  4067 
  4068 * Normalization by evaluation now allows non-leftlinear equations.
  4069 Declare with attribute [code nbe].
  4070 
  4071 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  4072 as the structured Isar versions "cases" and "induct", cf. the
  4073 corresponding "cases" and "induct" attributes.  Mutual induction rules
  4074 are now presented as a list of individual projections
  4075 (e.g. foo_bar.inducts for types foo and bar); the old format with
  4076 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  4077 rare situations a different rule is selected --- notably nested tuple
  4078 elimination instead of former prod.exhaust: use explicit (case_tac t
  4079 rule: prod.exhaust) here.
  4080 
  4081 * Attributes "cases", "induct", "coinduct" support "del" option.
  4082 
  4083 * Removed fact "case_split_thm", which duplicates "case_split".
  4084 
  4085 * The option datatype has been moved to a new theory Option.  Renamed
  4086 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  4087 
  4088 * New predicate "strict_mono" classifies strict functions on partial
  4089 orders.  With strict functions on linear orders, reasoning about
  4090 (in)equalities is facilitated by theorems "strict_mono_eq",
  4091 "strict_mono_less_eq" and "strict_mono_less".
  4092 
  4093 * Some set operations are now proper qualified constants with
  4094 authentic syntax.  INCOMPATIBILITY:
  4095 
  4096     op Int ~>   Set.Int
  4097     op Un ~>    Set.Un
  4098     INTER ~>    Set.INTER
  4099     UNION ~>    Set.UNION
  4100     Inter ~>    Set.Inter
  4101     Union ~>    Set.Union
  4102     {} ~>       Set.empty
  4103     UNIV ~>     Set.UNIV
  4104 
  4105 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  4106 theory Set.
  4107 
  4108 * Auxiliary class "itself" has disappeared -- classes without any
  4109 parameter are treated as expected by the 'class' command.
  4110 
  4111 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  4112 
  4113 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  4114 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  4115 
  4116 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  4117 method "approximation".  It proves formulas on real values by using
  4118 interval arithmetic.  In the formulas are also the transcendental
  4119 functions sin, cos, tan, atan, ln, exp and the constant pi are
  4120 allowed. For examples see
  4121 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  4122 
  4123 * Theory "Reflection" now resides in HOL/Library.
  4124 
  4125 * Entry point to Word library now simply named "Word".
  4126 INCOMPATIBILITY.
  4127 
  4128 * Made source layout more coherent with logical distribution
  4129 structure:
  4130 
  4131     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  4132     src/HOL/Library/Code_Message.thy ~> src/HOL/
  4133     src/HOL/Library/GCD.thy ~> src/HOL/
  4134     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  4135     src/HOL/Library/Parity.thy ~> src/HOL/
  4136     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  4137     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  4138     src/HOL/Real/Lubs.thy ~> src/HOL/
  4139     src/HOL/Real/PReal.thy ~> src/HOL/
  4140     src/HOL/Real/Rational.thy ~> src/HOL/
  4141     src/HOL/Real/RComplete.thy ~> src/HOL/
  4142     src/HOL/Real/RealDef.thy ~> src/HOL/
  4143     src/HOL/Real/RealPow.thy ~> src/HOL/
  4144     src/HOL/Real/Real.thy ~> src/HOL/
  4145     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  4146     src/HOL/Complex/Complex.thy ~> src/HOL/
  4147     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  4148     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  4149     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  4150     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  4151     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  4152     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  4153     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  4154     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  4155     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  4156     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  4157     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  4158     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  4159     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  4160     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  4161     src/HOL/Real/Float ~> src/HOL/Library/
  4162     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  4163     src/HOL/Real/RealVector.thy ~> src/HOL/
  4164 
  4165     src/HOL/arith_data.ML ~> src/HOL/Tools
  4166     src/HOL/hologic.ML ~> src/HOL/Tools
  4167     src/HOL/simpdata.ML ~> src/HOL/Tools
  4168     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  4169     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  4170     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  4171     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  4172     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  4173     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  4174     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  4175 
  4176     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  4177     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4178     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4179     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4180     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4181     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4182 
  4183 * If methods "eval" and "evaluation" encounter a structured proof
  4184 state with !!/==>, only the conclusion is evaluated to True (if
  4185 possible), avoiding strange error messages.
  4186 
  4187 * Method "sizechange" automates termination proofs using (a
  4188 modification of) the size-change principle.  Requires SAT solver.  See
  4189 src/HOL/ex/Termination.thy for examples.
  4190 
  4191 * Simplifier: simproc for let expressions now unfolds if bound
  4192 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4193 
  4194 * Method "arith": Linear arithmetic now ignores all inequalities when
  4195 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4196 
  4197 * New attribute "arith" for facts that should always be used
  4198 automatically by arithmetic. It is intended to be used locally in
  4199 proofs, e.g.
  4200 
  4201   assumes [arith]: "x > 0"
  4202 
  4203 Global usage is discouraged because of possible performance impact.
  4204 
  4205 * New classes "top" and "bot" with corresponding operations "top" and
  4206 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4207 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4208 
  4209 * Changed definition lemma "less_fun_def" in order to provide an
  4210 instance for preorders on functions; use lemma "less_le" instead.
  4211 INCOMPATIBILITY.
  4212 
  4213 * Theory Orderings: class "wellorder" moved here, with explicit
  4214 induction rule "less_induct" as assumption.  For instantiation of
  4215 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4216 INCOMPATIBILITY.
  4217 
  4218 * Theory Orderings: added class "preorder" as superclass of "order".
  4219 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4220 etc. slightly changed.  Some theorems named order_class.* now named
  4221 preorder_class.*.
  4222 
  4223 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4224 "diag" to "Id_on".
  4225 
  4226 * Theory Finite_Set: added a new fold combinator of type
  4227 
  4228   ('a => 'b => 'b) => 'b => 'a set => 'b
  4229 
  4230 Occasionally this is more convenient than the old fold combinator
  4231 which is now defined in terms of the new one and renamed to
  4232 fold_image.
  4233 
  4234 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4235 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4236 extended with further lemmas!).  At the moment both still exist but
  4237 the former will disappear at some point.
  4238 
  4239 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4240 class recpower.  Type-specific simp rules for various recpower types
  4241 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4242 
  4243 rat_power_0    -> power_0
  4244 rat_power_Suc  -> power_Suc
  4245 realpow_0      -> power_0
  4246 realpow_Suc    -> power_Suc
  4247 complexpow_0   -> power_0
  4248 complexpow_Suc -> power_Suc
  4249 power_poly_0   -> power_0
  4250 power_poly_Suc -> power_Suc
  4251 
  4252 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4253 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4254 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4255 of lemmas from Divides has been generalized from nat to class
  4256 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4257 renames resulting from duplicate elimination:
  4258 
  4259     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4260     zero_dvd_iff ~>         dvd_0_left_iff
  4261     dvd_0 ~>                dvd_0_right
  4262     DIVISION_BY_ZERO_DIV ~> div_by_0
  4263     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4264     mult_div ~>             div_mult_self2_is_id
  4265     mult_mod ~>             mod_mult_self2_is_0
  4266 
  4267 * Theory IntDiv: removed many lemmas that are instances of class-based
  4268 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4269 rename old lemmas as follows:
  4270 
  4271 dvd_diff               -> nat_dvd_diff
  4272 dvd_zminus_iff         -> dvd_minus_iff
  4273 mod_add1_eq            -> mod_add_eq
  4274 mod_mult1_eq           -> mod_mult_right_eq
  4275 mod_mult1_eq'          -> mod_mult_left_eq
  4276 mod_mult_distrib_mod   -> mod_mult_eq
  4277 nat_mod_add_left_eq    -> mod_add_left_eq
  4278 nat_mod_add_right_eq   -> mod_add_right_eq
  4279 nat_mod_div_trivial    -> mod_div_trivial
  4280 nat_mod_mod_trivial    -> mod_mod_trivial
  4281 zdiv_zadd_self1        -> div_add_self1
  4282 zdiv_zadd_self2        -> div_add_self2
  4283 zdiv_zmult_self1       -> div_mult_self2_is_id
  4284 zdiv_zmult_self2       -> div_mult_self1_is_id
  4285 zdvd_triv_left         -> dvd_triv_left
  4286 zdvd_triv_right        -> dvd_triv_right
  4287 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4288 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4289 zmod_zadd_left_eq      -> mod_add_left_eq
  4290 zmod_zadd_right_eq     -> mod_add_right_eq
  4291 zmod_zadd_self1        -> mod_add_self1
  4292 zmod_zadd_self2        -> mod_add_self2
  4293 zmod_zadd1_eq          -> mod_add_eq
  4294 zmod_zdiff1_eq         -> mod_diff_eq
  4295 zmod_zdvd_zmod         -> mod_mod_cancel
  4296 zmod_zmod_cancel       -> mod_mod_cancel
  4297 zmod_zmult_self1       -> mod_mult_self2_is_0
  4298 zmod_zmult_self2       -> mod_mult_self1_is_0
  4299 zmod_1                 -> mod_by_1
  4300 zdiv_1                 -> div_by_1
  4301 zdvd_abs1              -> abs_dvd_iff
  4302 zdvd_abs2              -> dvd_abs_iff
  4303 zdvd_refl              -> dvd_refl
  4304 zdvd_trans             -> dvd_trans
  4305 zdvd_zadd              -> dvd_add
  4306 zdvd_zdiff             -> dvd_diff
  4307 zdvd_zminus_iff        -> dvd_minus_iff
  4308 zdvd_zminus2_iff       -> minus_dvd_iff
  4309 zdvd_zmultD            -> dvd_mult_right
  4310 zdvd_zmultD2           -> dvd_mult_left
  4311 zdvd_zmult_mono        -> mult_dvd_mono
  4312 zdvd_0_right           -> dvd_0_right
  4313 zdvd_0_left            -> dvd_0_left_iff
  4314 zdvd_1_left            -> one_dvd
  4315 zminus_dvd_iff         -> minus_dvd_iff
  4316 
  4317 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4318 
  4319 * The real numbers offer decimal input syntax: 12.34 is translated
  4320 into 1234/10^2. This translation is not reversed upon output.
  4321 
  4322 * Theory Library/Polynomial defines an abstract type 'a poly of
  4323 univariate polynomials with coefficients of type 'a.  In addition to
  4324 the standard ring operations, it also supports div and mod.  Code
  4325 generation is also supported, using list-style constructors.
  4326 
  4327 * Theory Library/Inner_Product defines a class of real_inner for real
  4328 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4329 => real.  Class real_inner is a subclass of real_normed_vector from
  4330 theory RealVector.
  4331 
  4332 * Theory Library/Product_Vector provides instances for the product
  4333 type 'a * 'b of several classes from RealVector and Inner_Product.
  4334 Definitions of addition, subtraction, scalar multiplication, norms,
  4335 and inner products are included.
  4336 
  4337 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4338 addition to the field operations, numerals and case syntax are also
  4339 supported.
  4340 
  4341 * Theory Library/Diagonalize provides constructive version of Cantor's
  4342 first diagonalization argument.
  4343 
  4344 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4345 zlcm (for int); carried together from various gcd/lcm developements in
  4346 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4347 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4348 may recover tupled syntax as follows:
  4349 
  4350     hide (open) const gcd
  4351     abbreviation gcd where
  4352       "gcd == (%(a, b). GCD.gcd a b)"
  4353     notation (output)
  4354       GCD.gcd ("gcd '(_, _')")
  4355 
  4356 The same works for lcm, zgcd, zlcm.
  4357 
  4358 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4359 instantiations for algebraic structures.  Removed some duplicate
  4360 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4361 
  4362 * ML antiquotation @{code} takes a constant as argument and generates
  4363 corresponding code in background and inserts name of the corresponding
  4364 resulting ML value/function/datatype constructor binding in place.
  4365 All occurrences of @{code} with a single ML block are generated
  4366 simultaneously.  Provides a generic and safe interface for
  4367 instrumentalizing code generation.  See
  4368 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4369 In future you ought to refrain from ad-hoc compiling generated SML
  4370 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4371 cannot refer to constants for which user-defined serializations are
  4372 set.  Refer to the corresponding ML counterpart directly in that
  4373 cases.
  4374 
  4375 * Command 'rep_datatype': instead of theorem names the command now
  4376 takes a list of terms denoting the constructors of the type to be
  4377 represented as datatype.  The characteristic theorems have to be
  4378 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4379 have disappeared in favour of existing ones:
  4380 
  4381     unit_induct                 ~> unit.induct
  4382     prod_induct                 ~> prod.induct
  4383     sum_induct                  ~> sum.induct
  4384     Suc_Suc_eq                  ~> nat.inject
  4385     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4386 
  4387 
  4388 *** HOL-Algebra ***
  4389 
  4390 * New locales for orders and lattices where the equivalence relation
  4391 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4392 locales use a record structure with field eq for the equivalence.
  4393 
  4394 * New theory of factorial domains.
  4395 
  4396 * Units_l_inv and Units_r_inv are now simp rules by default.
  4397 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4398 and/or r_inv will now also require deletion of these lemmas.
  4399 
  4400 * Renamed the following theorems, INCOMPATIBILITY:
  4401 
  4402 UpperD ~> Upper_memD
  4403 LowerD ~> Lower_memD
  4404 least_carrier ~> least_closed
  4405 greatest_carrier ~> greatest_closed
  4406 greatest_Lower_above ~> greatest_Lower_below
  4407 one_zero ~> carrier_one_zero
  4408 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4409 
  4410 
  4411 *** HOL-Nominal ***
  4412 
  4413 * Nominal datatypes can now contain type-variables.
  4414 
  4415 * Commands 'nominal_inductive' and 'equivariance' work with local
  4416 theory targets.
  4417 
  4418 * Nominal primrec can now works with local theory targets and its
  4419 specification syntax now conforms to the general format as seen in
  4420 'inductive' etc.
  4421 
  4422 * Method "perm_simp" honours the standard simplifier attributes
  4423 (no_asm), (no_asm_use) etc.
  4424 
  4425 * The new predicate #* is defined like freshness, except that on the
  4426 left hand side can be a set or list of atoms.
  4427 
  4428 * Experimental command 'nominal_inductive2' derives strong induction
  4429 principles for inductive definitions.  In contrast to
  4430 'nominal_inductive', which can only deal with a fixed number of
  4431 binders, it can deal with arbitrary expressions standing for sets of
  4432 atoms to be avoided.  The only inductive definition we have at the
  4433 moment that needs this generalisation is the typing rule for Lets in
  4434 the algorithm W:
  4435 
  4436  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4437  -----------------------------------------------------------------
  4438          Gamma |- Let x be t1 in t2 : T2
  4439 
  4440 In this rule one wants to avoid all the binders that are introduced by
  4441 "close Gamma T1".  We are looking for other examples where this
  4442 feature might be useful.  Please let us know.
  4443 
  4444 
  4445 *** HOLCF ***
  4446 
  4447 * Reimplemented the simplification procedure for proving continuity
  4448 subgoals.  The new simproc is extensible; users can declare additional
  4449 continuity introduction rules with the attribute [cont2cont].
  4450 
  4451 * The continuity simproc now uses a different introduction rule for
  4452 solving continuity subgoals on terms with lambda abstractions.  In
  4453 some rare cases the new simproc may fail to solve subgoals that the
  4454 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4455 Potential INCOMPATIBILITY.
  4456 
  4457 * Command 'fixrec': specification syntax now conforms to the general
  4458 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4459 examples.  INCOMPATIBILITY.
  4460 
  4461 
  4462 *** ZF ***
  4463 
  4464 * Proof of Zorn's Lemma for partial orders.
  4465 
  4466 
  4467 *** ML ***
  4468 
  4469 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4470 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4471 depends on multithreading, so timouts will not work before Poly/ML
  4472 5.2.1!
  4473 
  4474 * High-level support for concurrent ML programming, see
  4475 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4476 particularly convenient to organize independent functional
  4477 computations.  The concept of "synchronized variables" provides a
  4478 higher-order interface for components with shared state, avoiding the
  4479 delicate details of mutexes and condition variables.  (Requires
  4480 Poly/ML 5.2.1 or later.)
  4481 
  4482 * ML bindings produced via Isar commands are stored within the Isar
  4483 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4484 become thread-safe and work with undo as expected (concerning
  4485 top-level bindings, not side-effects on global references).
  4486 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4487 compiler at runtime; really global bindings need to be given outside a
  4488 theory.  (Requires Poly/ML 5.2 or later.)
  4489 
  4490 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4491 context.  Top-level ML bindings are stored within the proof context in
  4492 a purely sequential fashion, disregarding the nested proof structure.
  4493 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4494 proof.  (Requires Poly/ML 5.2 or later.)
  4495 
  4496 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4497 and Method.setup, as well as Isar commands 'attribute_setup' and
  4498 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4499 existing code accordingly, or use plain 'setup' together with old
  4500 Method.add_method.
  4501 
  4502 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4503 to 'a -> thm, while results are always tagged with an authentic oracle
  4504 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4505 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4506 accordingly.  Note that extra performance may be gained by producing
  4507 the cterm carefully, avoiding slow Thm.cterm_of.
  4508 
  4509 * Simplified interface for defining document antiquotations via
  4510 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4511 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4512 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4513 examples.
  4514 
  4515 * More systematic treatment of long names, abstract name bindings, and
  4516 name space operations.  Basic operations on qualified names have been
  4517 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4518 Long_Name.append.  Old type bstring has been mostly replaced by
  4519 abstract type binding (see structure Binding), which supports precise
  4520 qualification by packages and local theory targets, as well as proper
  4521 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4522 bstring values into Binding.name, or better pass through abstract
  4523 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4524 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4525 
  4526 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4527 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4528 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4529 
  4530 * Disposed old type and term read functions (Sign.read_def_typ,
  4531 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4532 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4533 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4534 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4535 resort for legacy applications.
  4536 
  4537 * Disposed old declarations, tactics, tactic combinators that refer to
  4538 the simpset or claset of an implicit theory (such as Addsimps,
  4539 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4540 embedded ML text, or local_simpset_of with a proper context passed as
  4541 explicit runtime argument.
  4542 
  4543 * Rules and tactics that read instantiations (read_instantiate,
  4544 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4545 context, which is required for parsing and type-checking.  Moreover,
  4546 the variables are specified as plain indexnames, not string encodings
  4547 thereof.  INCOMPATIBILITY.
  4548 
  4549 * Generic Toplevel.add_hook interface allows to analyze the result of
  4550 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4551 for theorem dependency output of transactions resulting in a new
  4552 theory state.
  4553 
  4554 * ML antiquotations: block-structured compilation context indicated by
  4555 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4556 
  4557   @{binding name}                         - basic name binding
  4558   @{let ?pat = term}                      - term abbreviation (HO matching)
  4559   @{note name = fact}                     - fact abbreviation
  4560   @{thm fact}                             - singleton fact (with attributes)
  4561   @{thms fact}                            - general fact (with attributes)
  4562   @{lemma prop by method}                 - singleton goal
  4563   @{lemma prop by meth1 meth2}            - singleton goal
  4564   @{lemma prop1 ... propN by method}      - general goal
  4565   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4566   @{lemma (open) ...}                     - open derivation
  4567 
  4568 
  4569 *** System ***
  4570 
  4571 * The Isabelle "emacs" tool provides a specific interface to invoke
  4572 Proof General / Emacs, with more explicit failure if that is not
  4573 installed (the old isabelle-interface script silently falls back on
  4574 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4575 installation location of the Proof General distribution.
  4576 
  4577 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4578 the Isabelle process into a JVM/Scala application.  See
  4579 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4580 process wrapper has been discontinued.)
  4581 
  4582 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4583 
  4584 * Various status messages (with exact source position information) are
  4585 emitted, if proper markup print mode is enabled.  This allows
  4586 user-interface components to provide detailed feedback on internal
  4587 prover operations.
  4588 
  4589 
  4590 
  4591 New in Isabelle2008 (June 2008)
  4592 -------------------------------
  4593 
  4594 *** General ***
  4595 
  4596 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4597 and updated, with formally checked references as hyperlinks.
  4598 
  4599 * Theory loader: use_thy (and similar operations) no longer set the
  4600 implicit ML context, which was occasionally hard to predict and in
  4601 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4602 provides a proper context already.
  4603 
  4604 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4605 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4606 'use' within a theory file will do the job.
  4607 
  4608 * Name space merge now observes canonical order, i.e. the second space
  4609 is inserted into the first one, while existing entries in the first
  4610 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4611 swap theory imports.
  4612 
  4613 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4614 INCOMPATIBILITY in identifier syntax etc.
  4615 
  4616 * Outer syntax: string tokens no longer admit escaped white space,
  4617 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4618 white space without escapes.
  4619 
  4620 * Outer syntax: string tokens may contain arbitrary character codes
  4621 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4622 "foo_bar".
  4623 
  4624 
  4625 *** Pure ***
  4626 
  4627 * Context-dependent token translations.  Default setup reverts locally
  4628 fixed variables, and adds hilite markup for undeclared frees.
  4629 
  4630 * Unused theorems can be found using the new command 'unused_thms'.
  4631 There are three ways of invoking it:
  4632 
  4633 (1) unused_thms
  4634      Only finds unused theorems in the current theory.
  4635 
  4636 (2) unused_thms thy_1 ... thy_n -
  4637      Finds unused theorems in the current theory and all of its ancestors,
  4638      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4639 
  4640 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4641      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4642      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4643      their ancestors.
  4644 
  4645 In order to increase the readability of the list produced by
  4646 unused_thms, theorems that have been created by a particular instance
  4647 of a theory command such as 'inductive' or 'function' are considered
  4648 to belong to the same "group", meaning that if at least one theorem in
  4649 this group is used, the other theorems in the same group are no longer
  4650 reported as unused.  Moreover, if all theorems in the group are
  4651 unused, only one theorem in the group is displayed.
  4652 
  4653 Note that proof objects have to be switched on in order for
  4654 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4655 usually the case when using Proof General with the default settings).
  4656 
  4657 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4658 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4659 duplicate fact bindings, or even accidental fact duplications.  Note
  4660 that tools may maintain dynamically scoped facts systematically, using
  4661 PureThy.add_thms_dynamic.
  4662 
  4663 * Command 'hide' now allows to hide from "fact" name space as well.
  4664 
  4665 * Eliminated destructive theorem database, simpset, claset, and
  4666 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4667 update of theories within ML code.
  4668 
  4669 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4670 INCOMPATIBILITY, object-logics depending on former Pure require
  4671 additional setup PureThy.old_appl_syntax_setup; object-logics
  4672 depending on former CPure need to refer to Pure.
  4673 
  4674 * Commands 'use' and 'ML' are now purely functional, operating on
  4675 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4676 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4677 INCOMPATIBILITY.
  4678 
  4679 * Command 'setup': discontinued implicit version with ML reference.
  4680 
  4681 * Instantiation target allows for simultaneous specification of class
  4682 instance operations together with an instantiation proof.
  4683 Type-checking phase allows to refer to class operations uniformly.
  4684 See src/HOL/Complex/Complex.thy for an Isar example and
  4685 src/HOL/Library/Eval.thy for an ML example.
  4686 
  4687 * Indexing of literal facts: be more serious about including only
  4688 facts from the visible specification/proof context, but not the
  4689 background context (locale etc.).  Affects `prop` notation and method
  4690 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4691 situations.
  4692 
  4693 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4694 "(open)" option, which used to expose internal bound variables to the
  4695 proof text.
  4696 
  4697 * Isar statements: removed obsolete case "rule_context".
  4698 INCOMPATIBILITY, better use explicit fixes/assumes.
  4699 
  4700 * Locale proofs: default proof step now includes 'unfold_locales';
  4701 hence 'proof' without argument may be used to unfold locale
  4702 predicates.
  4703 
  4704 
  4705 *** Document preparation ***
  4706 
  4707 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4708 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4709 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4710 spot); renamed link color from "darkblue" to "linkcolor" (default
  4711 value unchanged, can be redefined via \definecolor); no longer sets
  4712 "a4paper" option (unnecessary or even intrusive).
  4713 
  4714 * Antiquotation @{lemma A method} proves proposition A by the given
  4715 method (either a method name or a method name plus (optional) method
  4716 arguments in parentheses) and prints A just like @{prop A}.
  4717 
  4718 
  4719 *** HOL ***
  4720 
  4721 * New primrec package.  Specification syntax conforms in style to
  4722 definition/function/....  No separate induction rule is provided.  The
  4723 "primrec" command distinguishes old-style and new-style specifications
  4724 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4725 When adjusting theories, beware: constants stemming from new-style
  4726 primrec specifications have authentic syntax.
  4727 
  4728 * Metis prover is now an order of magnitude faster, and also works
  4729 with multithreading.
  4730 
  4731 * Metis: the maximum number of clauses that can be produced from a
  4732 theorem is now given by the attribute max_clauses.  Theorems that
  4733 exceed this number are ignored, with a warning printed.
  4734 
  4735 * Sledgehammer no longer produces structured proofs by default. To
  4736 enable, declare [[sledgehammer_full = true]].  Attributes
  4737 reconstruction_modulus, reconstruction_sorts renamed
  4738 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4739 
  4740 * Method "induct_scheme" derives user-specified induction rules
  4741 from well-founded induction and completeness of patterns. This factors
  4742 out some operations that are done internally by the function package
  4743 and makes them available separately.  See
  4744 src/HOL/ex/Induction_Scheme.thy for examples.
  4745 
  4746 * More flexible generation of measure functions for termination
  4747 proofs: Measure functions can be declared by proving a rule of the
  4748 form "is_measure f" and giving it the [measure_function] attribute.
  4749 The "is_measure" predicate is logically meaningless (always true), and
  4750 just guides the heuristic.  To find suitable measure functions, the
  4751 termination prover sets up the goal "is_measure ?f" of the appropriate
  4752 type and generates all solutions by prolog-style backwards proof using
  4753 the declared rules.
  4754 
  4755 This setup also deals with rules like
  4756 
  4757   "is_measure f ==> is_measure (list_size f)"
  4758 
  4759 which accommodates nested datatypes that recurse through lists.
  4760 Similar rules are predeclared for products and option types.
  4761 
  4762 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4763 
  4764   INCOMPATIBILITIES:
  4765 
  4766   - Definitions of overloaded constants on sets have to be replaced by
  4767     definitions on => and bool.
  4768 
  4769   - Some definitions of overloaded operators on sets can now be proved
  4770     using the definitions of the operators on => and bool.  Therefore,
  4771     the following theorems have been renamed:
  4772 
  4773       subset_def   -> subset_eq
  4774       psubset_def  -> psubset_eq
  4775       set_diff_def -> set_diff_eq
  4776       Compl_def    -> Compl_eq
  4777       Sup_set_def  -> Sup_set_eq
  4778       Inf_set_def  -> Inf_set_eq
  4779       sup_set_def  -> sup_set_eq
  4780       inf_set_def  -> inf_set_eq
  4781 
  4782   - Due to the incompleteness of the HO unification algorithm, some
  4783     rules such as subst may require manual instantiation, if some of
  4784     the unknowns in the rule is a set.
  4785 
  4786   - Higher order unification and forward proofs:
  4787     The proof pattern
  4788 
  4789       have "P (S::'a set)" <...>
  4790       then have "EX S. P S" ..
  4791 
  4792     no longer works (due to the incompleteness of the HO unification
  4793     algorithm) and must be replaced by the pattern
  4794 
  4795       have "EX S. P S"
  4796       proof
  4797         show "P S" <...>
  4798       qed
  4799 
  4800   - Calculational reasoning with subst (or similar rules):
  4801     The proof pattern
  4802 
  4803       have "P (S::'a set)" <...>
  4804       also have "S = T" <...>
  4805       finally have "P T" .
  4806 
  4807     no longer works (for similar reasons as the previous example) and
  4808     must be replaced by something like
  4809 
  4810       have "P (S::'a set)" <...>
  4811       moreover have "S = T" <...>
  4812       ultimately have "P T" by simp
  4813 
  4814   - Tactics or packages written in ML code:
  4815     Code performing pattern matching on types via
  4816 
  4817       Type ("set", [T]) => ...
  4818 
  4819     must be rewritten. Moreover, functions like strip_type or
  4820     binder_types no longer return the right value when applied to a
  4821     type of the form
  4822 
  4823       T1 => ... => Tn => U => bool
  4824 
  4825     rather than
  4826 
  4827       T1 => ... => Tn => U set
  4828 
  4829 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4830 Wellfounded_Relations to theory Wellfounded.
  4831 
  4832 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4833 
  4834 * Class finite no longer treats UNIV as class parameter.  Use class
  4835 enum from theory Library/Enum instead to achieve a similar effect.
  4836 INCOMPATIBILITY.
  4837 
  4838 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4839 and "Cons".  INCOMPATIBILITY.
  4840 
  4841 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4842 Potential INCOMPATIBILITY.
  4843 
  4844 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4845 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4846 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4847 
  4848 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4849 the same name in theory Orderings:
  4850 
  4851   less_trans
  4852   less_linear
  4853   le_imp_less_or_eq
  4854   le_less_trans
  4855   less_le_trans
  4856   less_not_sym
  4857   less_asym
  4858 
  4859 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4860 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4861 and different variable names.
  4862 
  4863 * Library/Option_ord.thy: Canonical order on option type.
  4864 
  4865 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4866 finite maps.
  4867 
  4868 * Library/Countable.thy: Type class for countable types.
  4869 
  4870 * Theory Int: The representation of numerals has changed.  The infix
  4871 operator BIT and the bit datatype with constructors B0 and B1 have
  4872 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4873 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4874 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4875 accordingly.
  4876 
  4877 * Theory Nat: definition of <= and < on natural numbers no longer
  4878 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4879 le_def and less_def have disappeared.  Consider lemmas not_less
  4880 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4881 
  4882 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4883 (whose purpose mainly is for various fold_set functionals) have been
  4884 abandoned in favor of the existing algebraic classes
  4885 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4886 lower_semilattice (resp. upper_semilattice) and linorder.
  4887 INCOMPATIBILITY.
  4888 
  4889 * Theory Transitive_Closure: induct and cases rules now declare proper
  4890 case_names ("base" and "step").  INCOMPATIBILITY.
  4891 
  4892 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4893 lattices.  The form set-specific version is available as
  4894 Inductive.lfp_ordinal_induct_set.
  4895 
  4896 * Renamed theorems "power.simps" to "power_int.simps".
  4897 INCOMPATIBILITY.
  4898 
  4899 * Class semiring_div provides basic abstract properties of semirings
  4900 with division and modulo operations.  Subsumes former class dvd_mod.
  4901 
  4902 * Merged theories IntDef, Numeral and IntArith into unified theory
  4903 Int.  INCOMPATIBILITY.
  4904 
  4905 * Theory Library/Code_Index: type "index" now represents natural
  4906 numbers rather than integers.  INCOMPATIBILITY.
  4907 
  4908 * New class "uminus" with operation "uminus" (split of from class
  4909 "minus" which now only has operation "minus", binary).
  4910 INCOMPATIBILITY.
  4911 
  4912 * Constants "card", "internal_split", "option_map" now with authentic
  4913 syntax.  INCOMPATIBILITY.
  4914 
  4915 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4916 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4917 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4918 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4919 Sup_set_def, le_def, less_def, option_map_def now with object
  4920 equality.  INCOMPATIBILITY.
  4921 
  4922 * Records. Removed K_record, and replaced it by pure lambda term
  4923 %x. c. The simplifier setup is now more robust against eta expansion.
  4924 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4925 
  4926 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4927 
  4928 * Library/ListVector: new theory of arithmetic vector operations.
  4929 
  4930 * Library/Order_Relation: new theory of various orderings as sets of
  4931 pairs.  Defines preorders, partial orders, linear orders and
  4932 well-orders on sets and on types.
  4933 
  4934 
  4935 *** ZF ***
  4936 
  4937 * Renamed some theories to allow to loading both ZF and HOL in the
  4938 same session:
  4939 
  4940   Datatype  -> Datatype_ZF
  4941   Inductive -> Inductive_ZF
  4942   Int       -> Int_ZF
  4943   IntDiv    -> IntDiv_ZF
  4944   Nat       -> Nat_ZF
  4945   List      -> List_ZF
  4946   Main      -> Main_ZF
  4947 
  4948 INCOMPATIBILITY: ZF theories that import individual theories below
  4949 Main might need to be adapted.  Regular theory Main is still
  4950 available, as trivial extension of Main_ZF.
  4951 
  4952 
  4953 *** ML ***
  4954 
  4955 * ML within Isar: antiquotation @{const name} or @{const
  4956 name(typargs)} produces statically-checked Const term.
  4957 
  4958 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4959 (of the same name).  Removed obsolete print command.
  4960 
  4961 * Removed obsolete "use_legacy_bindings" function.
  4962 
  4963 * The ``print mode'' is now a thread-local value derived from a global
  4964 template (the former print_mode reference), thus access becomes
  4965 non-critical.  The global print_mode reference is for session
  4966 management only; user-code should use print_mode_value,
  4967 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4968 
  4969 * Functions system/system_out provide a robust way to invoke external
  4970 shell commands, with propagation of interrupts (requires Poly/ML
  4971 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4972 
  4973 
  4974 *** System ***
  4975 
  4976 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4977 in accordance with Proof General 3.7, which prefers GNU emacs.
  4978 
  4979 * isatool tty runs Isabelle process with plain tty interaction;
  4980 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4981 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4982 
  4983 * isatool browser now works with Cygwin as well, using general
  4984 "javapath" function defined in Isabelle process environment.
  4985 
  4986 * YXML notation provides a simple and efficient alternative to
  4987 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4988 isatool yxml as described in the Isabelle system manual.
  4989 
  4990 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4991 provides general wrapper for managing an Isabelle process in a robust
  4992 fashion, with ``cooked'' output from stdin/stderr.
  4993 
  4994 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4995 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4996 
  4997 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4998 way of changing the user's settings is via
  4999 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  5000 script.
  5001 
  5002 * Multithreading.max_threads := 0 refers to the number of actual CPU
  5003 cores of the underlying machine, which is a good starting point for
  5004 optimal performance tuning.  The corresponding usedir option -M allows
  5005 "max" as an alias for "0".  WARNING: does not work on certain versions
  5006 of Mac OS (with Poly/ML 5.1).
  5007 
  5008 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  5009 adverse effect of Isabelle flooding interactive front-ends (notably
  5010 ProofGeneral / XEmacs).
  5011 
  5012 
  5013 
  5014 New in Isabelle2007 (November 2007)
  5015 -----------------------------------
  5016 
  5017 *** General ***
  5018 
  5019 * More uniform information about legacy features, notably a
  5020 warning/error of "Legacy feature: ...", depending on the state of the
  5021 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  5022 legacy features will disappear eventually.
  5023 
  5024 * Theory syntax: the header format ``theory A = B + C:'' has been
  5025 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  5026 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  5027 
  5028 * Theory syntax: the old non-Isar theory file format has been
  5029 discontinued altogether.  Note that ML proof scripts may still be used
  5030 with Isar theories; migration is usually quite simple with the ML
  5031 function use_legacy_bindings.  INCOMPATIBILITY.
  5032 
  5033 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  5034 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  5035 quotes.
  5036 
  5037 * Theory loader: be more serious about observing the static theory
  5038 header specifications (including optional directories), but not the
  5039 accidental file locations of previously successful loads.  The strict
  5040 update policy of former update_thy is now already performed by
  5041 use_thy, so the former has been removed; use_thys updates several
  5042 theories simultaneously, just as 'imports' within a theory header
  5043 specification, but without merging the results.  Potential
  5044 INCOMPATIBILITY: may need to refine theory headers and commands
  5045 ROOT.ML which depend on load order.
  5046 
  5047 * Theory loader: optional support for content-based file
  5048 identification, instead of the traditional scheme of full physical
  5049 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  5050 (cf. the system manual).  The new scheme allows to work with
  5051 non-finished theories in persistent session images, such that source
  5052 files may be moved later on without requiring reloads.
  5053 
  5054 * Theory loader: old-style ML proof scripts being *attached* to a thy
  5055 file (with the same base name as the theory) are considered a legacy
  5056 feature, which will disappear eventually. Even now, the theory loader
  5057 no longer maintains dependencies on such files.
  5058 
  5059 * Syntax: the scope for resolving ambiguities via type-inference is
  5060 now limited to individual terms, instead of whole simultaneous
  5061 specifications as before. This greatly reduces the complexity of the
  5062 syntax module and improves flexibility by separating parsing and
  5063 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  5064 'fixes' etc.) are required in rare situations.
  5065 
  5066 * Syntax: constants introduced by new-style packages ('definition',
  5067 'abbreviation' etc.) are passed through the syntax module in
  5068 ``authentic mode''. This means that associated mixfix annotations
  5069 really stick to such constants, independently of potential name space
  5070 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  5071 trees are represented slightly differently, may need to adapt syntax
  5072 translations accordingly. Use CONST marker in 'translations' and
  5073 @{const_syntax} antiquotation in 'parse_translation' etc.
  5074 
  5075 * Legacy goal package: reduced interface to the bare minimum required
  5076 to keep existing proof scripts running.  Most other user-level
  5077 functions are now part of the OldGoals structure, which is *not* open
  5078 by default (consider isatool expandshort before open OldGoals).
  5079 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  5080 these tend to cause confusion about the actual goal (!) context being
  5081 used here, which is not necessarily the same as the_context().
  5082 
  5083 * Command 'find_theorems': supports "*" wild-card in "name:"
  5084 criterion; "with_dups" option.  Certain ProofGeneral versions might
  5085 support a specific search form (see ProofGeneral/CHANGES).
  5086 
  5087 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  5088 by default, which means that "prems" (and also "fixed variables") are
  5089 suppressed from proof state output.  Note that the ProofGeneral
  5090 settings mechanism allows to change and save options persistently, but
  5091 older versions of Isabelle will fail to start up if a negative prems
  5092 limit is imposed.
  5093 
  5094 * Local theory targets may be specified by non-nested blocks of
  5095 ``context/locale/class ... begin'' followed by ``end''.  The body may
  5096 contain definitions, theorems etc., including any derived mechanism
  5097 that has been implemented on top of these primitives.  This concept
  5098 generalizes the existing ``theorem (in ...)'' towards more versatility
  5099 and scalability.
  5100 
  5101 * Proof General interface: proper undo of final 'end' command;
  5102 discontinued Isabelle/classic mode (ML proof scripts).
  5103 
  5104 
  5105 *** Document preparation ***
  5106 
  5107 * Added antiquotation @{theory name} which prints the given name,
  5108 after checking that it refers to a valid ancestor theory in the
  5109 current context.
  5110 
  5111 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  5112 check the given source text as ML type/structure, printing verbatim.
  5113 
  5114 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  5115 "c args == rhs" given in the current context.  (Any number of
  5116 arguments may be given on the LHS.)
  5117 
  5118 
  5119 *** Pure ***
  5120 
  5121 * The 'class' package offers a combination of axclass and locale to
  5122 achieve Haskell-like type classes in Isabelle.  Definitions and
  5123 theorems within a class context produce both relative results (with
  5124 implicit parameters according to the locale context), and polymorphic
  5125 constants with qualified polymorphism (according to the class
  5126 context).  Within the body context of a 'class' target, a separate
  5127 syntax layer ("user space type system") takes care of converting
  5128 between global polymorphic consts and internal locale representation.
  5129 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  5130 "isatool doc classes" provides a tutorial.
  5131 
  5132 * Generic code generator framework allows to generate executable
  5133 code for ML and Haskell (including Isabelle classes).  A short usage
  5134 sketch:
  5135 
  5136     internal compilation:
  5137         export_code <list of constants (term syntax)> in SML
  5138     writing SML code to a file:
  5139         export_code <list of constants (term syntax)> in SML <filename>
  5140     writing OCaml code to a file:
  5141         export_code <list of constants (term syntax)> in OCaml <filename>
  5142     writing Haskell code to a bunch of files:
  5143         export_code <list of constants (term syntax)> in Haskell <filename>
  5144 
  5145     evaluating closed propositions to True/False using code generation:
  5146         method ``eval''
  5147 
  5148 Reasonable default setup of framework in HOL.
  5149 
  5150 Theorem attributs for selecting and transforming function equations theorems:
  5151 
  5152     [code fun]:        select a theorem as function equation for a specific constant
  5153     [code fun del]:    deselect a theorem as function equation for a specific constant
  5154     [code inline]:     select an equation theorem for unfolding (inlining) in place
  5155     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  5156 
  5157 User-defined serializations (target in {SML, OCaml, Haskell}):
  5158 
  5159     code_const <and-list of constants (term syntax)>
  5160       {(target) <and-list of const target syntax>}+
  5161 
  5162     code_type <and-list of type constructors>
  5163       {(target) <and-list of type target syntax>}+
  5164 
  5165     code_instance <and-list of instances>
  5166       {(target)}+
  5167         where instance ::= <type constructor> :: <class>
  5168 
  5169     code_class <and_list of classes>
  5170       {(target) <and-list of class target syntax>}+
  5171         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  5172 
  5173 code_instance and code_class only are effective to target Haskell.
  5174 
  5175 For example usage see src/HOL/ex/Codegenerator.thy and
  5176 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  5177 generation from Isabelle/HOL theories is available via "isatool doc
  5178 codegen".
  5179 
  5180 * Code generator: consts in 'consts_code' Isar commands are now
  5181 referred to by usual term syntax (including optional type
  5182 annotations).
  5183 
  5184 * Command 'no_translations' removes translation rules from theory
  5185 syntax.
  5186 
  5187 * Overloaded definitions are now actually checked for acyclic
  5188 dependencies.  The overloading scheme is slightly more general than
  5189 that of Haskell98, although Isabelle does not demand an exact
  5190 correspondence to type class and instance declarations.
  5191 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5192 exotic versions of overloading -- at the discretion of the user!
  5193 
  5194 Polymorphic constants are represented via type arguments, i.e. the
  5195 instantiation that matches an instance against the most general
  5196 declaration given in the signature.  For example, with the declaration
  5197 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5198 as c(nat).  Overloading is essentially simultaneous structural
  5199 recursion over such type arguments.  Incomplete specification patterns
  5200 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5201 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5202 Command 'print_theory' outputs the normalized system of recursive
  5203 equations, see section "definitions".
  5204 
  5205 * Configuration options are maintained within the theory or proof
  5206 context (with name and type bool/int/string), providing a very simple
  5207 interface to a poor-man's version of general context data.  Tools may
  5208 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5209 these values using Config.get etc.  Users may change options via an
  5210 associated attribute of the same name.  This form of context
  5211 declaration works particularly well with commands 'declare' or
  5212 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5213 very easy to avoid global references, which would not observe Isar
  5214 toplevel undo/redo and fail to work with multithreading.
  5215 
  5216 Various global ML references of Pure and HOL have been turned into
  5217 configuration options:
  5218 
  5219   Unify.search_bound		unify_search_bound
  5220   Unify.trace_bound		unify_trace_bound
  5221   Unify.trace_simp		unify_trace_simp
  5222   Unify.trace_types		unify_trace_types
  5223   Simplifier.simp_depth_limit	simp_depth_limit
  5224   Blast.depth_limit		blast_depth_limit
  5225   DatatypeProp.dtK		datatype_distinctness_limit
  5226   fast_arith_neq_limit  	fast_arith_neq_limit
  5227   fast_arith_split_limit	fast_arith_split_limit
  5228 
  5229 * Named collections of theorems may be easily installed as context
  5230 data using the functor NamedThmsFun (see also
  5231 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5232 attributes; there is also a toplevel print command.  This facility is
  5233 just a common case of general context data, which is the preferred way
  5234 for anything more complex than just a list of facts in canonical
  5235 order.
  5236 
  5237 * Isar: command 'declaration' augments a local theory by generic
  5238 declaration functions written in ML.  This enables arbitrary content
  5239 being added to the context, depending on a morphism that tells the
  5240 difference of the original declaration context wrt. the application
  5241 context encountered later on.
  5242 
  5243 * Isar: proper interfaces for simplification procedures.  Command
  5244 'simproc_setup' declares named simprocs (with match patterns, and body
  5245 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5246 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5247 
  5248 * Isar: an extra pair of brackets around attribute declarations
  5249 abbreviates a theorem reference involving an internal dummy fact,
  5250 which will be ignored later --- only the effect of the attribute on
  5251 the background context will persist.  This form of in-place
  5252 declarations is particularly useful with commands like 'declare' and
  5253 'using', for example ``have A using [[simproc a]] by simp''.
  5254 
  5255 * Isar: method "assumption" (and implicit closing of subproofs) now
  5256 takes simple non-atomic goal assumptions into account: after applying
  5257 an assumption as a rule the resulting subgoals are solved by atomic
  5258 assumption steps.  This is particularly useful to finish 'obtain'
  5259 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5260 without referring to the original premise "!!x. P x ==> thesis" in the
  5261 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5262 more permissive.
  5263 
  5264 * Isar: implicit use of prems from the Isar proof context is
  5265 considered a legacy feature.  Common applications like ``have A .''
  5266 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5267 referencing facts explicitly here improves readability and
  5268 maintainability of proof texts.
  5269 
  5270 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5271 the obtained context from the course of reasoning!  For example:
  5272 
  5273   assume "EX x y. A x & B y"   -- "any previous fact"
  5274   then guess x and y by clarify
  5275 
  5276 This technique is potentially adventurous, depending on the facts and
  5277 proof tools being involved here.
  5278 
  5279 * Isar: known facts from the proof context may be specified as literal
  5280 propositions, using ASCII back-quote syntax.  This works wherever
  5281 named facts used to be allowed so far, in proof commands, proof
  5282 methods, attributes etc.  Literal facts are retrieved from the context
  5283 according to unification of type and term parameters.  For example,
  5284 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5285 theorems in the current context, then these are valid literal facts:
  5286 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5287 
  5288 There is also a proof method "fact" which does the same composition
  5289 for explicit goal states, e.g. the following proof texts coincide with
  5290 certain special cases of literal facts:
  5291 
  5292   have "A" by fact                 ==  note `A`
  5293   have "A ==> B" by fact           ==  note `A ==> B`
  5294   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5295   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5296 
  5297 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5298 outer syntax.  Thus symbolic identifiers may be used without
  5299 additional white space in declarations like this: ``assume *: A''.
  5300 
  5301 * Isar: 'print_facts' prints all local facts of the current context,
  5302 both named and unnamed ones.
  5303 
  5304 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5305 
  5306   def x == "t" and y == "u"
  5307 
  5308 * Isar: added command 'unfolding', which is structurally similar to
  5309 'using', but affects both the goal state and facts by unfolding given
  5310 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5311 'unfolded' attribute may be replaced by first-class proof text.
  5312 
  5313 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5314 and command 'unfolding' now all support object-level equalities
  5315 (potentially conditional).  The underlying notion of rewrite rule is
  5316 analogous to the 'rule_format' attribute, but *not* that of the
  5317 Simplifier (which is usually more generous).
  5318 
  5319 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5320 premises of a theorem by n. Useful in conjunction with drule.
  5321 
  5322 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5323 method expression within a sandbox consisting of the first N
  5324 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5325 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5326 simplifies all new goals that emerge from applying rule foo to the
  5327 originally first one.
  5328 
  5329 * Isar: schematic goals are no longer restricted to higher-order
  5330 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5331 expected.
  5332 
  5333 * Isar: the conclusion of a long theorem statement is now either
  5334 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5335 (essentially a disjunction of cases with local parameters and
  5336 assumptions).  The latter allows to express general elimination rules
  5337 adequately; in this notation common elimination rules look like this:
  5338 
  5339   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5340     assumes "EX x. P x"
  5341     obtains x where "P x"
  5342 
  5343   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5344     assumes "A & B"
  5345     obtains A and B
  5346 
  5347   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5348     assumes "A | B"
  5349     obtains
  5350       A
  5351     | B
  5352 
  5353 The subsequent classical rules even refer to the formal "thesis"
  5354 explicitly:
  5355 
  5356   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5357     obtains "~ thesis"
  5358 
  5359   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5360     obtains "thesis ==> something"
  5361 
  5362 The actual proof of an 'obtains' statement is analogous to that of the
  5363 Isar proof element 'obtain', only that there may be several cases.
  5364 Optional case names may be specified in parentheses; these will be
  5365 available both in the present proof and as annotations in the
  5366 resulting rule, for later use with the 'cases' method (cf. attribute
  5367 case_names).
  5368 
  5369 * Isar: the assumptions of a long theorem statement are available as
  5370 "assms" fact in the proof context.  This is more appropriate than the
  5371 (historical) "prems", which refers to all assumptions of the current
  5372 context, including those from the target locale, proof body etc.
  5373 
  5374 * Isar: 'print_statement' prints theorems from the current theory or
  5375 proof context in long statement form, according to the syntax of a
  5376 top-level lemma.
  5377 
  5378 * Isar: 'obtain' takes an optional case name for the local context
  5379 introduction rule (default "that").
  5380 
  5381 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5382 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5383 to occur.
  5384 
  5385 * Pure: syntax "CONST name" produces a fully internalized constant
  5386 according to the current context.  This is particularly useful for
  5387 syntax translations that should refer to internal constant
  5388 representations independently of name spaces.
  5389 
  5390 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5391 instead of "FOO ". This allows multiple binder declarations to coexist
  5392 in the same context.  INCOMPATIBILITY.
  5393 
  5394 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5395 primitive that also works in a locale context (both for constants and
  5396 fixed variables). Type declaration and internal syntactic representation
  5397 of given constants retrieved from the context. Likewise, the
  5398 'no_notation' command allows to remove given syntax annotations from the
  5399 current context.
  5400 
  5401 * Isar/locales: new derived specification elements 'axiomatization',
  5402 'definition', 'abbreviation', which support type-inference, admit
  5403 object-level specifications (equality, equivalence).  See also the
  5404 isar-ref manual.  Examples:
  5405 
  5406   axiomatization
  5407     eq  (infix "===" 50) where
  5408     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5409 
  5410   definition "f x y = x + y + 1"
  5411   definition g where "g x = f x x"
  5412 
  5413   abbreviation
  5414     neq  (infix "=!=" 50) where
  5415     "x =!= y == ~ (x === y)"
  5416 
  5417 These specifications may be also used in a locale context.  Then the
  5418 constants being introduced depend on certain fixed parameters, and the
  5419 constant name is qualified by the locale base name.  An internal
  5420 abbreviation takes care for convenient input and output, making the
  5421 parameters implicit and using the original short name.  See also
  5422 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5423 entities from a monomorphic theory.
  5424 
  5425 Presently, abbreviations are only available 'in' a target locale, but
  5426 not inherited by general import expressions.  Also note that
  5427 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5428 'translations' in common applications.  The "no_abbrevs" print mode
  5429 prevents folding of abbreviations in term output.
  5430 
  5431 Concrete syntax is attached to specified constants in internal form,
  5432 independently of name spaces.  The parse tree representation is
  5433 slightly different -- use 'notation' instead of raw 'syntax', and
  5434 'translations' with explicit "CONST" markup to accommodate this.
  5435 
  5436 * Pure/Isar: unified syntax for new-style specification mechanisms
  5437 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5438 full type inference and dummy patterns ("_").  For example:
  5439 
  5440   definition "K x _ = x"
  5441 
  5442   inductive conj for A B
  5443   where "A ==> B ==> conj A B"
  5444 
  5445 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5446 the current context.  Print mode "no_abbrevs" prevents inversion of
  5447 abbreviations on output.
  5448 
  5449 * Isar/locales: improved parameter handling: use of locales "var" and
  5450 "struct" no longer necessary; - parameter renamings are no longer
  5451 required to be injective.  For example, this allows to define
  5452 endomorphisms as locale endom = homom mult mult h.
  5453 
  5454 * Isar/locales: changed the way locales with predicates are defined.
  5455 Instead of accumulating the specification, the imported expression is
  5456 now an interpretation.  INCOMPATIBILITY: different normal form of
  5457 locale expressions.  In particular, in interpretations of locales with
  5458 predicates, goals repesenting already interpreted fragments are not
  5459 removed automatically.  Use methods `intro_locales' and
  5460 `unfold_locales'; see below.
  5461 
  5462 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5463 provide backward reasoning on locales predicates.  The methods are
  5464 aware of interpretations and discharge corresponding goals.
  5465 `intro_locales' is less aggressive then `unfold_locales' and does not
  5466 unfold predicates to assumptions.
  5467 
  5468 * Isar/locales: the order in which locale fragments are accumulated
  5469 has changed.  This enables to override declarations from fragments due
  5470 to interpretations -- for example, unwanted simp rules.
  5471 
  5472 * Isar/locales: interpretation in theories and proof contexts has been
  5473 extended.  One may now specify (and prove) equations, which are
  5474 unfolded in interpreted theorems.  This is useful for replacing
  5475 defined concepts (constants depending on locale parameters) by
  5476 concepts already existing in the target context.  Example:
  5477 
  5478   interpretation partial_order ["op <= :: [int, int] => bool"]
  5479     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5480 
  5481 Typically, the constant `partial_order.less' is created by a
  5482 definition specification element in the context of locale
  5483 partial_order.
  5484 
  5485 * Method "induct": improved internal context management to support
  5486 local fixes and defines on-the-fly. Thus explicit meta-level
  5487 connectives !!  and ==> are rarely required anymore in inductive goals
  5488 (using object-logic connectives for this purpose has been long
  5489 obsolete anyway). Common proof patterns are explained in
  5490 src/HOL/Induct/Common_Patterns.thy, see also
  5491 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5492 examples.
  5493 
  5494 * Method "induct": improved handling of simultaneous goals. Instead of
  5495 introducing object-level conjunction, the statement is now split into
  5496 several conclusions, while the corresponding symbolic cases are nested
  5497 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5498 see src/HOL/Induct/Common_Patterns.thy, for example.
  5499 
  5500 * Method "induct": mutual induction rules are now specified as a list
  5501 of rule sharing the same induction cases. HOL packages usually provide
  5502 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5503 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5504 mutual induction rules differently, i.e. like this:
  5505 
  5506   (induct rule: foo_bar.inducts)
  5507   (induct set: foo bar)
  5508   (induct pred: foo bar)
  5509   (induct type: foo bar)
  5510 
  5511 The ML function ProjectRule.projections turns old-style rules into the
  5512 new format.
  5513 
  5514 * Method "coinduct": dual of induction, see
  5515 src/HOL/Library/Coinductive_List.thy for various examples.
  5516 
  5517 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5518 considered a legacy feature.
  5519 
  5520 * Attribute "symmetric" produces result with standardized schematic
  5521 variables (index 0).  Potential INCOMPATIBILITY.
  5522 
  5523 * Simplifier: by default the simplifier trace only shows top level
  5524 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5525 default. Thus there is less danger of being flooded by the trace. The
  5526 trace indicates where parts have been suppressed.
  5527 
  5528 * Provers/classical: removed obsolete classical version of elim_format
  5529 attribute; classical elim/dest rules are now treated uniformly when
  5530 manipulating the claset.
  5531 
  5532 * Provers/classical: stricter checks to ensure that supplied intro,
  5533 dest and elim rules are well-formed; dest and elim rules must have at
  5534 least one premise.
  5535 
  5536 * Provers/classical: attributes dest/elim/intro take an optional
  5537 weight argument for the rule (just as the Pure versions).  Weights are
  5538 ignored by automated tools, but determine the search order of single
  5539 rule steps.
  5540 
  5541 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5542 where the body does not mention the bound variable.  Note that dummy
  5543 patterns implicitly depend on their context of bounds, which makes
  5544 "{_. _}" match any set comprehension as expected.  Potential
  5545 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5546 constant "_idtdummy" in the binding position.
  5547 
  5548 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5549 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5550 for example "A -> B" => "Pi A (%_. B)".
  5551 
  5552 * Pure: 'class_deps' command visualizes the subclass relation, using
  5553 the graph browser tool.
  5554 
  5555 * Pure: 'print_theory' now suppresses certain internal declarations by
  5556 default; use '!' option for full details.
  5557 
  5558 
  5559 *** HOL ***
  5560 
  5561 * Method "metis" proves goals by applying the Metis general-purpose
  5562 resolution prover (see also http://gilith.com/software/metis/).
  5563 Examples are in the directory MetisExamples.  WARNING: the
  5564 Isabelle/HOL-Metis integration does not yet work properly with
  5565 multi-threading.
  5566 
  5567 * Command 'sledgehammer' invokes external automatic theorem provers as
  5568 background processes.  It generates calls to the "metis" method if
  5569 successful. These can be pasted into the proof.  Users do not have to
  5570 wait for the automatic provers to return.  WARNING: does not really
  5571 work with multi-threading.
  5572 
  5573 * New "auto_quickcheck" feature tests outermost goal statements for
  5574 potential counter-examples.  Controlled by ML references
  5575 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5576 5000 milliseconds).  Fails silently if statements is outside of
  5577 executable fragment, or any other codgenerator problem occurs.
  5578 
  5579 * New constant "undefined" with axiom "undefined x = undefined".
  5580 
  5581 * Added class "HOL.eq", allowing for code generation with polymorphic
  5582 equality.
  5583 
  5584 * Some renaming of class constants due to canonical name prefixing in
  5585 the new 'class' package:
  5586 
  5587     HOL.abs ~> HOL.abs_class.abs
  5588     HOL.divide ~> HOL.divide_class.divide
  5589     0 ~> HOL.zero_class.zero
  5590     1 ~> HOL.one_class.one
  5591     op + ~> HOL.plus_class.plus
  5592     op - ~> HOL.minus_class.minus
  5593     uminus ~> HOL.minus_class.uminus
  5594     op * ~> HOL.times_class.times
  5595     op < ~> HOL.ord_class.less
  5596     op <= > HOL.ord_class.less_eq
  5597     Nat.power ~> Power.power_class.power
  5598     Nat.size ~> Nat.size_class.size
  5599     Numeral.number_of ~> Numeral.number_class.number_of
  5600     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5601     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5602     Orderings.min ~> Orderings.ord_class.min
  5603     Orderings.max ~> Orderings.ord_class.max
  5604     Divides.op div ~> Divides.div_class.div
  5605     Divides.op mod ~> Divides.div_class.mod
  5606     Divides.op dvd ~> Divides.div_class.dvd
  5607 
  5608 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5609 
  5610 a) User-defined constants using any of the names "plus", "minus",
  5611 "times", "less" or "less_eq". The standard syntax translations for
  5612 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5613 names.
  5614 
  5615 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5616 INCOMPATIBILITY: use more specific names.
  5617 
  5618 c) Permutative equations (e.g. "a + b = b + a")
  5619 Since the change of names also changes the order of terms, permutative
  5620 rewrite rules may get applied in a different order. Experience shows
  5621 that this is rarely the case (only two adaptions in the whole Isabelle
  5622 distribution).  INCOMPATIBILITY: rewrite proofs
  5623 
  5624 d) ML code directly refering to constant names
  5625 This in general only affects hand-written proof tactics, simprocs and
  5626 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5627 Consider using @{const_name} antiquotation.
  5628 
  5629 * New class "default" with associated constant "default".
  5630 
  5631 * Function "sgn" is now overloaded and available on int, real, complex
  5632 (and other numeric types), using class "sgn".  Two possible defs of
  5633 sgn are given as equational assumptions in the classes sgn_if and
  5634 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5635 INCOMPATIBILITY.
  5636 
  5637 * Locale "partial_order" now unified with class "order" (cf. theory
  5638 Orderings), added parameter "less".  INCOMPATIBILITY.
  5639 
  5640 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5641 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5642 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5643 
  5644 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5645 changed order of proof goals in instance proofs.
  5646 
  5647 * The transitivity reasoner for partial and linear orders is set up
  5648 for classes "order" and "linorder".  Instances of the reasoner are available
  5649 in all contexts importing or interpreting the corresponding locales.
  5650 Method "order" invokes the reasoner separately; the reasoner
  5651 is also integrated with the Simplifier as a solver.  Diagnostic
  5652 command 'print_orders' shows the available instances of the reasoner
  5653 in the current context.
  5654 
  5655 * Localized monotonicity predicate in theory "Orderings"; integrated
  5656 lemmas max_of_mono and min_of_mono with this predicate.
  5657 INCOMPATIBILITY.
  5658 
  5659 * Formulation of theorem "dense" changed slightly due to integration
  5660 with new class dense_linear_order.
  5661 
  5662 * Uniform lattice theory development in HOL.
  5663 
  5664     constants "meet" and "join" now named "inf" and "sup"
  5665     constant "Meet" now named "Inf"
  5666 
  5667     classes "meet_semilorder" and "join_semilorder" now named
  5668       "lower_semilattice" and "upper_semilattice"
  5669     class "lorder" now named "lattice"
  5670     class "comp_lat" now named "complete_lattice"
  5671 
  5672     Instantiation of lattice classes allows explicit definitions
  5673     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5674 
  5675   INCOMPATIBILITY.  Theorem renames:
  5676 
  5677     meet_left_le            ~> inf_le1
  5678     meet_right_le           ~> inf_le2
  5679     join_left_le            ~> sup_ge1
  5680     join_right_le           ~> sup_ge2
  5681     meet_join_le            ~> inf_sup_ord
  5682     le_meetI                ~> le_infI
  5683     join_leI                ~> le_supI
  5684     le_meet                 ~> le_inf_iff
  5685     le_join                 ~> ge_sup_conv
  5686     meet_idempotent         ~> inf_idem
  5687     join_idempotent         ~> sup_idem
  5688     meet_comm               ~> inf_commute
  5689     join_comm               ~> sup_commute
  5690     meet_leI1               ~> le_infI1
  5691     meet_leI2               ~> le_infI2
  5692     le_joinI1               ~> le_supI1
  5693     le_joinI2               ~> le_supI2
  5694     meet_assoc              ~> inf_assoc
  5695     join_assoc              ~> sup_assoc
  5696     meet_left_comm          ~> inf_left_commute
  5697     meet_left_idempotent    ~> inf_left_idem
  5698     join_left_comm          ~> sup_left_commute
  5699     join_left_idempotent    ~> sup_left_idem
  5700     meet_aci                ~> inf_aci
  5701     join_aci                ~> sup_aci
  5702     le_def_meet             ~> le_iff_inf
  5703     le_def_join             ~> le_iff_sup
  5704     join_absorp2            ~> sup_absorb2
  5705     join_absorp1            ~> sup_absorb1
  5706     meet_absorp1            ~> inf_absorb1
  5707     meet_absorp2            ~> inf_absorb2
  5708     meet_join_absorp        ~> inf_sup_absorb
  5709     join_meet_absorp        ~> sup_inf_absorb
  5710     distrib_join_le         ~> distrib_sup_le
  5711     distrib_meet_le         ~> distrib_inf_le
  5712 
  5713     add_meet_distrib_left   ~> add_inf_distrib_left
  5714     add_join_distrib_left   ~> add_sup_distrib_left
  5715     is_join_neg_meet        ~> is_join_neg_inf
  5716     is_meet_neg_join        ~> is_meet_neg_sup
  5717     add_meet_distrib_right  ~> add_inf_distrib_right
  5718     add_join_distrib_right  ~> add_sup_distrib_right
  5719     add_meet_join_distribs  ~> add_sup_inf_distribs
  5720     join_eq_neg_meet        ~> sup_eq_neg_inf
  5721     meet_eq_neg_join        ~> inf_eq_neg_sup
  5722     add_eq_meet_join        ~> add_eq_inf_sup
  5723     meet_0_imp_0            ~> inf_0_imp_0
  5724     join_0_imp_0            ~> sup_0_imp_0
  5725     meet_0_eq_0             ~> inf_0_eq_0
  5726     join_0_eq_0             ~> sup_0_eq_0
  5727     neg_meet_eq_join        ~> neg_inf_eq_sup
  5728     neg_join_eq_meet        ~> neg_sup_eq_inf
  5729     join_eq_if              ~> sup_eq_if
  5730 
  5731     mono_meet               ~> mono_inf
  5732     mono_join               ~> mono_sup
  5733     meet_bool_eq            ~> inf_bool_eq
  5734     join_bool_eq            ~> sup_bool_eq
  5735     meet_fun_eq             ~> inf_fun_eq
  5736     join_fun_eq             ~> sup_fun_eq
  5737     meet_set_eq             ~> inf_set_eq
  5738     join_set_eq             ~> sup_set_eq
  5739     meet1_iff               ~> inf1_iff
  5740     meet2_iff               ~> inf2_iff
  5741     meet1I                  ~> inf1I
  5742     meet2I                  ~> inf2I
  5743     meet1D1                 ~> inf1D1
  5744     meet2D1                 ~> inf2D1
  5745     meet1D2                 ~> inf1D2
  5746     meet2D2                 ~> inf2D2
  5747     meet1E                  ~> inf1E
  5748     meet2E                  ~> inf2E
  5749     join1_iff               ~> sup1_iff
  5750     join2_iff               ~> sup2_iff
  5751     join1I1                 ~> sup1I1
  5752     join2I1                 ~> sup2I1
  5753     join1I1                 ~> sup1I1
  5754     join2I2                 ~> sup1I2
  5755     join1CI                 ~> sup1CI
  5756     join2CI                 ~> sup2CI
  5757     join1E                  ~> sup1E
  5758     join2E                  ~> sup2E
  5759 
  5760     is_meet_Meet            ~> is_meet_Inf
  5761     Meet_bool_def           ~> Inf_bool_def
  5762     Meet_fun_def            ~> Inf_fun_def
  5763     Meet_greatest           ~> Inf_greatest
  5764     Meet_lower              ~> Inf_lower
  5765     Meet_set_def            ~> Inf_set_def
  5766 
  5767     Sup_def                 ~> Sup_Inf
  5768     Sup_bool_eq             ~> Sup_bool_def
  5769     Sup_fun_eq              ~> Sup_fun_def
  5770     Sup_set_eq              ~> Sup_set_def
  5771 
  5772     listsp_meetI            ~> listsp_infI
  5773     listsp_meet_eq          ~> listsp_inf_eq
  5774 
  5775     meet_min                ~> inf_min
  5776     join_max                ~> sup_max
  5777 
  5778 * Added syntactic class "size"; overloaded constant "size" now has
  5779 type "'a::size ==> bool"
  5780 
  5781 * Internal reorganisation of `size' of datatypes: size theorems
  5782 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5783 simplification rules by default!); theorems "prod.size" now named
  5784 "*.size".
  5785 
  5786 * Class "div" now inherits from class "times" rather than "type".
  5787 INCOMPATIBILITY.
  5788 
  5789 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5790 Linorder etc.  have disappeared; operations defined in terms of
  5791 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5792 
  5793 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5794 
  5795 * HOL-Word: New extensive library and type for generic, fixed size
  5796 machine words, with arithmetic, bit-wise, shifting and rotating
  5797 operations, reflection into int, nat, and bool lists, automation for
  5798 linear arithmetic (by automatic reflection into nat or int), including
  5799 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5800 arithmetic type classes, supporting automatic simplification of
  5801 numerals on all operations.
  5802 
  5803 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5804 
  5805 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5806 
  5807 * Code generator library theories:
  5808   - Code_Integer represents HOL integers by big integer literals in target
  5809     languages.
  5810   - Code_Char represents HOL characters by character literals in target
  5811     languages.
  5812   - Code_Char_chr like Code_Char, but also offers treatment of character
  5813     codes; includes Code_Integer.
  5814   - Executable_Set allows to generate code for finite sets using lists.
  5815   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5816     denominator).
  5817   - Executable_Real implements a subset of real numbers, namly those
  5818     representable by rational numbers.
  5819   - Efficient_Nat implements natural numbers by integers, which in general will
  5820     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5821     includes Code_Integer.
  5822   - Code_Index provides an additional datatype index which is mapped to
  5823     target-language built-in integers.
  5824   - Code_Message provides an additional datatype message_string which is isomorphic to
  5825     strings; messages are mapped to target-language strings.
  5826 
  5827 * New package for inductive predicates
  5828 
  5829   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5830 
  5831     inductive
  5832       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5833       for z_1 :: U_1 and ... and z_n :: U_m
  5834     where
  5835       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5836     | ...
  5837 
  5838   with full support for type-inference, rather than
  5839 
  5840     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5841 
  5842     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5843     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5844 
  5845     inductive "s z_1 ... z_m"
  5846     intros
  5847       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5848       ...
  5849 
  5850   For backward compatibility, there is a wrapper allowing inductive
  5851   sets to be defined with the new package via
  5852 
  5853     inductive_set
  5854       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5855       for z_1 :: U_1 and ... and z_n :: U_m
  5856     where
  5857       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5858     | ...
  5859 
  5860   or
  5861 
  5862     inductive_set
  5863       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5864       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5865       for z_1 :: U_1 and ... and z_n :: U_m
  5866     where
  5867       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5868     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5869     | ...
  5870 
  5871   if the additional syntax "p ..." is required.
  5872 
  5873   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5874   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5875 
  5876   INCOMPATIBILITIES:
  5877 
  5878   - Since declaration and definition of inductive sets or predicates
  5879     is no longer separated, abbreviations involving the newly
  5880     introduced sets or predicates must be specified together with the
  5881     introduction rules after the 'where' keyword (see above), rather
  5882     than before the actual inductive definition.
  5883 
  5884   - The variables in induction and elimination rules are now
  5885     quantified in the order of their occurrence in the introduction
  5886     rules, rather than in alphabetical order. Since this may break
  5887     some proofs, these proofs either have to be repaired, e.g. by
  5888     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5889     statements of the form
  5890 
  5891       case (rule_i a_i_1 ... a_i_{k_i})
  5892 
  5893     or the old order of quantification has to be restored by explicitly adding
  5894     meta-level quantifiers in the introduction rules, i.e.
  5895 
  5896       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5897 
  5898   - The format of the elimination rules is now
  5899 
  5900       p z_1 ... z_m x_1 ... x_n ==>
  5901         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5902         ==> ... ==> P
  5903 
  5904     for predicates and
  5905 
  5906       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5907         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5908         ==> ... ==> P
  5909 
  5910     for sets rather than
  5911 
  5912       x : s z_1 ... z_m ==>
  5913         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5914         ==> ... ==> P
  5915 
  5916     This may require terms in goals to be expanded to n-tuples
  5917     (e.g. using case_tac or simplification with the split_paired_all
  5918     rule) before the above elimination rule is applicable.
  5919 
  5920   - The elimination or case analysis rules for (mutually) inductive
  5921     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5922     list of rules "p_1_..._p_k.elims" is no longer available.
  5923 
  5924 * New package "function"/"fun" for general recursive functions,
  5925 supporting mutual and nested recursion, definitions in local contexts,
  5926 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5927 for small examples, and the separate tutorial on the function
  5928 package. The old recdef "package" is still available as before, but
  5929 users are encouraged to use the new package.
  5930 
  5931 * Method "lexicographic_order" automatically synthesizes termination
  5932 relations as lexicographic combinations of size measures.
  5933 
  5934 * Case-expressions allow arbitrary constructor-patterns (including
  5935 "_") and take their order into account, like in functional
  5936 programming.  Internally, this is translated into nested
  5937 case-expressions; missing cases are added and mapped to the predefined
  5938 constant "undefined". In complicated cases printing may no longer show
  5939 the original input but the internal form. Lambda-abstractions allow
  5940 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5941 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5942 variable.
  5943 
  5944 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5945 is an abbreviation for "of_nat :: nat => int". The simplification
  5946 rules for "of_nat" have been changed to work like "int" did
  5947 previously.  Potential INCOMPATIBILITY:
  5948   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5949   - of_nat_diff and of_nat_mult are no longer default simp rules
  5950 
  5951 * Method "algebra" solves polynomial equations over (semi)rings using
  5952 Groebner bases. The (semi)ring structure is defined by locales and the
  5953 tool setup depends on that generic context. Installing the method for
  5954 a specific type involves instantiating the locale and possibly adding
  5955 declarations for computation on the coefficients.  The method is
  5956 already instantiated for natural numbers and for the axiomatic class
  5957 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5958 CALCULEMUS 2007 for the general principles underlying this
  5959 architecture of context-aware proof-tools.
  5960 
  5961 * Method "ferrack" implements quantifier elimination over
  5962 special-purpose dense linear orders using locales (analogous to
  5963 "algebra"). The method is already installed for class
  5964 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5965 rat, etc.
  5966 
  5967 * Former constant "List.op @" now named "List.append".  Use ML
  5968 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5969 circumvent possible incompatibilities when working on ML level.
  5970 
  5971 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5972 
  5973 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5974 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5975 
  5976 * New syntax for Haskell-like list comprehension (input only), eg.
  5977 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5978 
  5979 * The special syntax for function "filter" has changed from [x :
  5980 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5981 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5982 
  5983 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5984 defined on nat only, as an abbreviation for [a..<Suc b]
  5985 INCOMPATIBILITY.
  5986 
  5987 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5988 
  5989 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5990 
  5991 * New lemma collection field_simps (an extension of ring_simps) for
  5992 manipulating (in)equations involving division. Multiplies with all
  5993 denominators that can be proved to be non-zero (in equations) or
  5994 positive/negative (in inequations).
  5995 
  5996 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5997 have been improved and renamed to ring_simps, group_simps and
  5998 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5999 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  6000 
  6001 * Theory Library/Commutative_Ring: switched from recdef to function
  6002 package; constants add, mul, pow now curried.  Infix syntax for
  6003 algebraic operations.
  6004 
  6005 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  6006 INCOMPATIBILITY.
  6007 
  6008 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  6009 INCOMPATIBILITY.
  6010 
  6011 * HOL/records: generalised field-update to take a function on the
  6012 field rather than the new value: r(|A := x|) is translated to A_update
  6013 (K x) r The K-combinator that is internally used is called K_record.
  6014 INCOMPATIBILITY: Usage of the plain update functions has to be
  6015 adapted.
  6016 
  6017 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  6018 * x = 0, which are required for a semiring.  Richer structures do not
  6019 inherit from semiring_0 anymore, because this property is a theorem
  6020 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  6021 there is more to prove, but this is mostly trivial.
  6022 
  6023 * Class "recpower" is generalized to arbitrary monoids, not just
  6024 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  6025 commutativity or semiring properties additionally.
  6026 
  6027 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  6028 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  6029 better use 'abbreviation'.
  6030 
  6031 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  6032 
  6033 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  6034 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  6035 significant changes for setting up numeral syntax for types:
  6036   - New constants Numeral.pred and Numeral.succ instead
  6037       of former Numeral.bin_pred and Numeral.bin_succ.
  6038   - Use integer operations instead of bin_add, bin_mult and so on.
  6039   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  6040   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  6041 
  6042 See src/HOL/Integ/IntArith.thy for an example setup.
  6043 
  6044 * Command 'normal_form' computes the normal form of a term that may
  6045 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  6046 produces ``[b, c, a]'' (without proof).  This command is suitable for
  6047 heavy-duty computations because the functions are compiled to ML
  6048 first.  Correspondingly, a method "normalization" is provided.  See
  6049 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  6050 
  6051 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  6052 25 like -->); output depends on the "iff" print_mode, the default is
  6053 "A = B" (with priority 50).
  6054 
  6055 * Relations less (<) and less_eq (<=) are also available on type bool.
  6056 Modified syntax to disallow nesting without explicit parentheses,
  6057 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  6058 INCOMPATIBILITY.
  6059 
  6060 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  6061 
  6062 * Relation composition operator "op O" now has precedence 75 and binds
  6063 stronger than union and intersection. INCOMPATIBILITY.
  6064 
  6065 * The old set interval syntax "{m..n(}" (and relatives) has been
  6066 removed.  Use "{m..<n}" (and relatives) instead.
  6067 
  6068 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  6069 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  6070 ``declare [[simproc del: neq]]''.
  6071 
  6072 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  6073 True/False.
  6074 
  6075 * Theorem Cons_eq_map_conv no longer declared as "simp".
  6076 
  6077 * Theorem setsum_mult renamed to setsum_right_distrib.
  6078 
  6079 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  6080 ``rule'' method.
  6081 
  6082 * Reimplemented methods "sat" and "satx", with several improvements:
  6083 goals no longer need to be stated as "<prems> ==> False", equivalences
  6084 (i.e. "=" on type bool) are handled, variable names of the form
  6085 "lit_<n>" are no longer reserved, significant speedup.
  6086 
  6087 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  6088 zChaff is still supported as well.
  6089 
  6090 * 'inductive' and 'datatype': provide projections of mutual rules,
  6091 bundled as foo_bar.inducts;
  6092 
  6093 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  6094 Library.
  6095 
  6096 * Library: moved theory Accessible_Part to main HOL.
  6097 
  6098 * Library: added theory Coinductive_List of potentially infinite lists
  6099 as greatest fixed-point.
  6100 
  6101 * Library: added theory AssocList which implements (finite) maps as
  6102 association lists.
  6103 
  6104 * Method "evaluation" solves goals (i.e. a boolean expression)
  6105 efficiently by compiling it to ML.  The goal is "proved" (via an
  6106 oracle) if it evaluates to True.
  6107 
  6108 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  6109 also when invoked by the simplifier.  This results in the Simplifier
  6110 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  6111 Configuration option fast_arith_split_limit=0 recovers the old
  6112 behavior.
  6113 
  6114 * Support for hex (0x20) and binary (0b1001) numerals.
  6115 
  6116 * New method: reify eqs (t), where eqs are equations for an
  6117 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  6118 parameter, computes a term s::'b and a list xs::'a list and proves the
  6119 theorem I xs s = t. This is also known as reification or quoting. The
  6120 resulting theorem is applied to the subgoal to substitute t with I xs
  6121 s.  If t is omitted, the subgoal itself is reified.
  6122 
  6123 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  6124 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  6125 where f is supposed to be a computable function (in the sense of code
  6126 generattion). The method uses reify to compute s and xs as above then
  6127 applies corr_thm and uses normalization by evaluation to "prove" f s =
  6128 r and finally gets the theorem t = r, which is again applied to the
  6129 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  6130 
  6131 * Reflection: Automatic reification now handels binding, an example is
  6132 available in src/HOL/ex/ReflectionEx.thy
  6133 
  6134 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  6135 command 'statespace' that is similar to 'record', but introduces an
  6136 abstract specification based on the locale infrastructure instead of
  6137 HOL types.  This leads to extra flexibility in composing state spaces,
  6138 in particular multiple inheritance and renaming of components.
  6139 
  6140 
  6141 *** HOL-Complex ***
  6142 
  6143 * Hyperreal: Functions root and sqrt are now defined on negative real
  6144 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  6145 Nonnegativity side conditions have been removed from many lemmas, so
  6146 that more subgoals may now be solved by simplification; potential
  6147 INCOMPATIBILITY.
  6148 
  6149 * Real: new type classes formalize real normed vector spaces and
  6150 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  6151 and norm :: 'a => real.
  6152 
  6153 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  6154 reals into other types. The overloaded constant Reals :: 'a set is now
  6155 defined as range of_real; potential INCOMPATIBILITY.
  6156 
  6157 * Real: proper support for ML code generation, including 'quickcheck'.
  6158 Reals are implemented as arbitrary precision rationals.
  6159 
  6160 * Hyperreal: Several constants that previously worked only for the
  6161 reals have been generalized, so they now work over arbitrary vector
  6162 spaces. Type annotations may need to be added in some cases; potential
  6163 INCOMPATIBILITY.
  6164 
  6165   Infinitesimal  :: ('a::real_normed_vector) star set
  6166   HFinite        :: ('a::real_normed_vector) star set
  6167   HInfinite      :: ('a::real_normed_vector) star set
  6168   approx         :: ('a::real_normed_vector) star => 'a star => bool
  6169   monad          :: ('a::real_normed_vector) star => 'a star set
  6170   galaxy         :: ('a::real_normed_vector) star => 'a star set
  6171   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  6172   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  6173   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  6174   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  6175   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  6176   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  6177   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6178   sgn            :: 'a::real_normed_vector => 'a
  6179   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6180 
  6181 * Complex: Some complex-specific constants are now abbreviations for
  6182 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6183 hnorm.  Other constants have been entirely removed in favor of the
  6184 polymorphic versions (INCOMPATIBILITY):
  6185 
  6186   approx        <-- capprox
  6187   HFinite       <-- CFinite
  6188   HInfinite     <-- CInfinite
  6189   Infinitesimal <-- CInfinitesimal
  6190   monad         <-- cmonad
  6191   galaxy        <-- cgalaxy
  6192   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  6193   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  6194   (ns)deriv     <-- (ns)cderiv
  6195 
  6196 
  6197 *** HOL-Algebra ***
  6198 
  6199 * Formalisation of ideals and the quotient construction over rings.
  6200 
  6201 * Order and lattice theory no longer based on records.
  6202 INCOMPATIBILITY.
  6203 
  6204 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  6205 greatest_closed.  INCOMPATIBILITY.
  6206 
  6207 * Method algebra is now set up via an attribute.  For examples see
  6208 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  6209 of algebraic structures.
  6210 
  6211 * Renamed theory CRing to Ring.
  6212 
  6213 
  6214 *** HOL-Nominal ***
  6215 
  6216 * Substantial, yet incomplete support for nominal datatypes (binding
  6217 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  6218 src/HOL/Nominal/Examples.  Prospective users should consult
  6219 http://isabelle.in.tum.de/nominal/
  6220 
  6221 
  6222 *** ML ***
  6223 
  6224 * ML basics: just one true type int, which coincides with IntInf.int
  6225 (even on SML/NJ).
  6226 
  6227 * ML within Isar: antiquotations allow to embed statically-checked
  6228 formal entities in the source, referring to the context available at
  6229 compile-time.  For example:
  6230 
  6231 ML {* @{sort "{zero,one}"} *}
  6232 ML {* @{typ "'a => 'b"} *}
  6233 ML {* @{term "%x. x"} *}
  6234 ML {* @{prop "x == y"} *}
  6235 ML {* @{ctyp "'a => 'b"} *}
  6236 ML {* @{cterm "%x. x"} *}
  6237 ML {* @{cprop "x == y"} *}
  6238 ML {* @{thm asm_rl} *}
  6239 ML {* @{thms asm_rl} *}
  6240 ML {* @{type_name c} *}